Theory and practice of water and wastewater treatment [Second edition.] 9781119312369, 1119312361

Table of contents :
Theory and Practice of Water and Wastewater Treatment
Contents
Acknowledgments
Preface
Abbreviations and Acronyms Used in the Text
About the Companion Website
Section I: Chemistry
Chapter 1: Basic Chemistry
1.1 Definitions
1.2 The Expression of Concentration
1.3 Ions and Molecules in Water
1.3.1 Oxidation Number
1.4 Balancing Reactions
1.5 Oxidation-Reduction Reactions
1.6 Equilibrium
1.7 Conductivity and Ionic Strength
1.7.1 Conductance
1.7.2 Ionic Strength
1.8 Chemical Kinetics
1.8.1 Other Formulations
Consecutive or Series
Parallel
Retardant
Autocatalytic
Catalysis
1.8.2 The Effect of Temperature on Rate of Reaction
1.9 Gas Laws
1.10 Gas Solubility: Henry's Law
1.11 Solubility Product
1.12 Complexes
1.13 Nuclear Chemistry
1.13.1 Radioactivity Units
Questions and Problems
References
Chapter 2: The Thermodynamic Basis for Equilibrium
2.1 Thermodynamic Relations
2.1.1 Free Energy
Expression of Concentration in Equilibrium Expressions
2.1.2 Enthalpy and Temperature Effects on the Equilibrium Constant
2.2 Redox Potentials
2.2.1 Cell or Couple Potential
2.2.2 Oxidation-Reduction Potential and System Potential
2.3 Corrosion
2.3.1 Microbial Corrosion
2.3.2 Corrosion Prevention from External Environmental Factors
Galvanic Cathodic Protection
Electrolytic (or Impressed Current) Cathodic Protection
Questions and Problems
References
Chapter 3: Acid–Base Chemistry
3.1 pH
3.2 Acids and Bases
3.2.1 Conjugate Acids and Bases
3.3 Equivalents and Normality
3.4 Solution of Multiequilibria Systems
3.5 Buffers
3.5.1 Dilution of a Buffered Solution
3.5.2 The Most Effective pH for a Buffer
3.6 Acid–Base Titrations
3.6.1 Titration of Strong Acids and Bases
3.6.2 Titration of Weak Acids and Bases
3.6.3 Indicating the Endpoint of an Acid–Base Titration
3.7 Natural Buffering of Waters from Carbon Dioxide and Related Compounds
3.7.1 Acidity and Alkalinity
Questions and Problems
References
Chapter 4: Organic and Biochemistry
4.1 Carbon
4.2 Properties of Organic Compounds
4.3 Functional Groups
4.4 Types of Organic Compounds
4.4.1 Aliphatic Compounds
Aldehydes and Ketones
Alcohols, Esters, and Ethers
4.4.2 Nitrogen-containing Compounds
4.5 Aromatic Compounds
4.5.1 Compounds of Sulfur
4.6 Naturally Occurring Organic Compounds
4.6.1 Carbohydrates
4.6.2 Proteins
4.6.3 Fats and Oils
4.7 Biochemistry
4.8 Glycolysis
4.9 The Tricarboxylic Acid Cycle
4.10 Enzyme Kinetics
Questions and Problems
References
Chapter 5: Analyses and Constituents in Water
5.1 Titration
5.1.1 Complex and Precipitate Formation Titrations
5.1.2 Redox Titrations and Potentiometric Analyses
5.1.3 Indicators for Potentiometric Analysis
5.2 Colorimetric Analyses
5.2.1 The Beer–Lambert Laws for Light Transmittance
5.3 Physical Analyses
5.3.1 Solids
5.3.2 Turbidity and Color
5.4 Determination of Organic Matter
5.4.1 Chemical Oxygen Demand
General Reaction for COD
Interferences with the COD Test
5.4.2. Biochemical Oxygen Demand
Effects of Temperature on BOD Exertion
Carbonaceous and Nitrogenous BOD
Laboratory Methods for Determining BOD
Limitations of the BOD Test for Biological Wastewater Treatment Process Design
Analysis of a BOD Progression
5.4.3. Total Organic Carbon
Questions and Problems
References
Section II: Microorganisms in Water and Water Quality
Chapter 6: Microbiology
6.1 Groups of Microorganisms and the Phylogenetic Tree
6.2 Bacteria and Archaea
6.2.1 Classification of Bacteria
Taxonomy
Metabolic Requirements
Oxygen Requirements
Temperature
Salt and Sugar Concentrations
pH
6.3 Eukaryotes
6.3.1 Algae
6.3.2 Fungi
6.3.3 Protists
6.4 Other Microorganisms
6.4.1 Viruses and Phages
6.4.2 Rotifers
6.4.3 Worms
6.5 Determining the Growth of Microorganisms
6.5.1 Growth of Pure Cultures
6.5.2 Growth of Mixed Cultures
6.5.3 Viability and Mass in Growing Cultures
6.5.4 Enumeration of Microorganisms
Plate Counts
Practical Considerations in Determining Mean Values
6.5.5 Microbial Genomics and Molecular Microbiology Tools
Phylogenetic Microbial Community Composition Analysis
Functional Analysis
Questions and Problems
References
Chapter 7: Water, Wastes, and Disease
7.1 Agents of Disease
7.1.1 Bacterial Pathogens
7.1.2 Viral Pathogens
7.1.3 Protozoan Pathogens
7.1.4 Helminths
7.1.5 Insect and Animal Vectors of Disease
7.2 Indicator, Test, and Model Microorganisms
7.3 Indicators of Fecal Contamination
7.4 Indicator Microorganisms
7.4.1 Coliforms: Total, Thermotolerant, and E. coli
7.4.2 Enterococci
7.5 Surrogates
7.6 Survival of Microorganisms in the Aquatic Environment
7.7 Minimum Infective Dose
Questions and Problems
References
Chapter 8: Water Constituents and Quality Standards
8.1 Toxicity of Elements and Compounds
8.2 Contaminants in Water
8.2.1 Emerging Contaminants
8.2.2 Common Contaminants
Aluminum
Nitrate
Fluoride
Detergents
8.2.3 Carcinogens
8.2.4 Radioactive Constituents
8.3 Taste and Odor
8.4 Bases for Standards
8.4.1 Risk Assessment for Microbial Infection
8.4.2 Determination of Carcinogenicity
8.4.3 Toxicity Determination
8.4.4 Environmental Water Quality Standards
8.5 Standards for Drinking Water
8.5.1 International Drinking Water Standards
8.5.2 US Safe Drinking Water Act
8.5.3 Canadian Water Quality Guidelines
8.6 Comparison of Drinking Water Standards
8.6.1 Microbiological Parameters
WHO Guidelines for Microbiological Quality
United States Standards for Microbiological Quality
Canadian Guidelines for Microbiological Quality
8.6.2 Chemical and Physical Qualities
8.6.3 Aesthetic Quality
8.6.4 Radiological Constituents
8.6.5 Other Water Standards
8.7 Water Consumption
8.8 Canadian Federal Wastewater Quality Guidelines
8.9 Wastewater Characteristics
Greywater
8.10 Wastewater Production
Questions and Problems
References
Section III: Water and Wastewater Treatment
Chapter 9: Water and Wastewater Treatment Operations
9.1 Water Treatment Operations
Microbial Contaminants
Reservoirs
9.1.1 Home Water Treatment Units
9.2 Wastewater Treatment Unit Operations
9.3 Hydraulic Design of Water and Wastewater Treatment Plants
Flow in Pressurized Pipes
Flow in Open Channels
Other Losses
Questions and Problems
References
Chapter 10: Mass Balances and Hydraulic Flow Regimes
10.1 Setup of Mass Balances
10.1.1 Mixing Characteristics of Basins
10.1.2 Mass Balances for PF Reactors
Method I
Method II
Method III
10.1.3 Mass Balances and Reaction for CM Basins
10.1.4 Batch Processes
10.2 Flow Analysis of CM and PF Reactors
10.2.1 Tracer Analysis of Complete Mixed Reactors
10.2.2 Tracer Analysis of Plug Flow
10.2.3 Complete Mixed Reactors in Series
10.2.4 Other Flow Irregularities: Dead Volume and Short-circuiting
10.2.5 Typical Flow Characteristics of Basins
10.2.6 Measurement of Dispersion
10.3 Detention Time in Vessels
10.3.1 Average Detention Time
10.3.2 The Effects of Flow Recycle on Detention Time
10.3.3 The Effects of Recycle on Mixing
10.4 Flow and Quality Equalization
10.5 System Material Balances
Questions and Problems
References
Section IV: Physical–Chemical Treatment Processes
Chapter 11: Screening and Sedimentation
11.1 Screens and Bar Racks
11.1.1 Screens for Water Treatment Plants
11.1.2 Screens at Wastewater Treatment Plants
11.1.3 Microstrainers
11.2 Sedimentation
11.2.1 Particle Settling Velocity
11.3 Grit Chambers
11.3.1 Horizontal Flow Grit Chambers
Channel with Varying Cross Section
Design Notes for a Parabolic Grit Chamber
11.3.2 Aerated Grit Chambers
11.3.3 Square Tank Degritter
11.3.4 Vortex Grit Removal Devices
Grit Washing
11.4 Type I Sedimentation
11.4.1 Theory
11.5 Type II Sedimentation
11.5.1 Laboratory Determination of Settling Velocity Distribution
11.5.2 Type II Sedimentation Data Analysis
11.5.3 Alternative Method for Calculating Total Removal
11.5.4 Sizing the Basin
11.6 Tube and Lamella Clarifiers
11.7 Weir–Launder Design
11.8 Clarifier Design for Water and Primary Wastewater Treatment
11.8.1 Design Ranges for Typical Clarifiers for Water and Wastewater Treatment
11.8.2 Chemically Enhanced Primary Treatment
11.8.3 Depth in Sedimentation Basins
11.9 Inlet Hydraulics for Sedimentation Basins
11.9.1 Flow Distributions
11.9.2 Inlet Baffling
Questions and Problems
References
Chapter 12: Mass Transfer and Aeration
12.1 Fick’s Law
12.2 Gas Transfer
12.2.1 Calculating the Mass Transfer Coefficient
12.2.2 The Effects of pH on Mass Transfer
12.3 Aeration in Water and Wastewater Treatment
12.3.1 Hazards Associated with Oxygen, Carbon Monoxide, and Hydrogen Sulfide
12.4 Design of Aeration Systems
12.4.1 Gravity Aerators
12.4.2 Spray Aerators
12.4.3 Diffused Aerators
Questions and Problems
References
Chapter 13: Coagulation and Flocculation
13.1 Coagulation
Recovery of Alum and Iron Coagulants
13.2 Mixing and Power Dissipation
13.3 Mixers
13.3.1 Mechanical Mixers
13.3.2 Pneumatic Mixers
13.3.3 Hydraulic Mixers
Venturi Sections and Hydraulic Jumps
13.4 Flocculators
13.4.1 Paddle Flocculators
13.4.2 Vertical-Shaft Turbine Flocculators
13.4.3 Pipes
13.4.4 Baffled Channels
13.4.5 Upflow Solids Contact Clarifier
13.4.6 Alabama Flocculator
13.4.7 Spiral Flow Tanks
13.4.8 Pebble Bed Flocculators
13.4.9 Ballasted Flocculation
Questions and Problems
References
Chapter 14: Filtration
14.1 Slow Sand Filters and Rapid Filters
14.2 Filtering Materials
14.2.1 Grain Size and Distribution
14.3 Headloss in Filters
14.3.1 Grain Size Distribution and Headloss
14.4 Backwashing Filters
14.4.1. Total Head Requirements for Backwashing
Losses in the Expanded Media
14.4.2. Backwash Velocity
Method 1
Method 2
Headloss and Expansion in a Stratified Bed
14.5 Support Media and Underdrains in Rapid Filters
Other Design Features of Filters
Auxiliary Wash and Air Scour Systems
14.6 Filter Beds for Water and Wastewater Treatment
14.7 Air Binding of Filters
14.8 Rapid Filtration Alternatives
14.8.1 Single-medium and Multimedia Filters
14.8.2 Constant- and Declining-rate Filtration
14.8.3 Direct Filtration
14.9 Pressure Filters
14.10 Slow Sand Filters
14.10.1 Slow Sand Filters for Tertiary Wastewater Treatment
14.11 Biological Filtration for Water Treatment
Questions and Problems
References
Chapter 15: Physical–Chemical Treatment for Dissolved Constituents
15.1 Water Softening
15.2 Lime–Soda Softening
15.2.1 Treatment Methods for Lime–Soda Hardness Removal
15.2.2 Bar Graphs
Lime Recovery and Sludge Reduction
15.3 Corrosion Prevention in Water Supply Systems
15.3.1 The Langelier Index Misconception
15.4 Iron and Manganese Removal
15.4.1 Greensand
15.4.2 Aeration
15.4.3 Sequestering Iron and Manganese
15.4.4 Biological Removal of Iron and Manganese
15.5 Phosphorus Removal from Wastewater by Chemical Precipitation
15.5.1 Removal of Phosphorus by Chemically Reactive Species
15.6 Removal of Arsenic and Metals
15.6.1 Metals Removal
15.6.2 Arsenic Removal
15.7 Advanced Oxidation Processes
15.8 Ion Exchange
15.8.1 Activated Alumina
15.8.2 Ammonia and Nitrate Removal by Ion Exchange
15.9 Fluoridation and Defluoridation
15.10 Membrane Processes
15.10.1 Assessment of Water Suitability for Membrane Treatment
15.10.2 Concentrate Disposal
15.10.3 Membranes for Water Treatment
Microfiltration and Ultrafiltration Systems
Nanofiltration and Reverse Osmosis Treatment
Electrodialysis
15.11 Activated Carbon Adsorption
15.11.1 Activated Carbon – Preparation and Characteristics
15.11.2 Adsorption Isotherms
15.11.3 Granular Activated Carbon Adsorbers
15.12 Design of Fixed-bed Adsorbers
15.12.1 Rate Formulation for Adsorption
15.12.2 Theory of Fixed-bed Adsorber Systems
The Capacity Utilized in the Adsorption Zone
Competitive Adsorption
15.12.3 Bed-depth Service Time Method
15.12.4 Rapid Small-Scale Column Tests
15.12.5 Granular Activated Carbon Reactors in Series
15.12.6 Design of a Suspended Media PAC or GAC Continuous Flow Reactor
Questions and Problems
References
Chapter 16: Disinfection
16.1 Kinetics of Disinfection
16.2 Chlorination
16.2.1 Chemistry of Chlorine
16.2.2 Measurement of Free and Residual Chlorine
16.2.3 Chlorine Decay
16.2.4 Drinking Water Disinfection by Chlorine
16.2.5 Wastewater Disinfection by Chlorine
16.2.6 Design of Contacting Systems for Chlorine
16.2.7 Disinfection as the Sole Treatment of Surface Water
16.2.8 Other Applications of Chlorine
16.2.9 Dechlorination
16.3 Chloramines
16.4 Chlorine Dioxide
16.4.1 Chlorine Dioxide Doses as a Primary Disinfectant
16.4.2 Chlorine Dioxide for Pre-disinfection or for Residual Disinfection
16.4.3 Generation of Chlorine Dioxide
16.5 Peracids: Peracetic Acid (PAA) and Performic Acid (PFA)
16.5.1 Peracetic Acid
Kinetics of Disinfection Using PAA
Measuring PAA Residuals
Applications for Wastewater Disinfection
Chemical Disinfection Process Control
16.5.2 Performic Acid
16.6 Ozone
16.6.1 Determining the Appropriate Ozone Dose
16.6.2 Ozone Generation
16.6.3 Ozone Dissolution Systems
16.6.4 Ozone Contactor Basins
16.6.5 Ozone Chemistry: Mass Transfer Coefficients and Radicals Production
16.6.6 Ozone for Wastewater Disinfection
16.6.7 Ozone for Destruction of Micropollutants
16.7 Ultraviolet Radiation
16.7.1 Mechanism of UV Disinfection
16.7.2 Repair of UV Damage
Photo Repair
Dark Repair
16.7.3 Interferences
16.7.4 Generation of Ultraviolet Light and Ultraviolet Reactors
16.7.5 Disinfection Kinetics
16.7.6 Disinfection Doses (or Fluences)
16.7.7 Determination of UV Fluence
16.7.8 Ultraviolet Reactors
16.8 Point-of-use Disinfectants: Solar Disinfection (SODIS), with or without Photoreactants such as TiO2
16.9 Disinfection Byproducts
16.9.1 Chlorine
16.9.2 Chloramines
16.9.3 Chlorine Dioxide
16.9.4 Peracids
16.9.5 Ozone
16.9.6 Ultraviolet
16.9.7 Comparative Risks
16.10 Disinfection to Combat Invasive Species
Questions and Problems
References
Section V: Biological Wastewater Treatment
Chapter 17: Aerobic Biological Treatment: Biotreatment Processes
17.1 Microorganisms in Aerobic Biological Treatment
17.2 The Activated Sludge Process
17.3 Substrate Removal and Growth of Microorganisms
17.3.1 Substrate Removal
Temperature Dependence of Rate Coefficients
BOD, COD, and TOC Removal
17.3.2 Growth of Microorganisms and Biological Sludge Production
Sludge Composition and Nutrient Requirements
17.4 Activated Sludge Configurations
17.4.1 Definition of Symbols for the Activated Sludge Process Models
17.4.2 Reactor
17.4.3 System Effluent and Waste Sludge Line
17.4.4 Clarifier
17.5 Process Analysis
17.5.1 Physical Concentration of Solids in the Bioreactor
17.5.2 Solids Retention Time
17.5.3 Sludge Volume Index
17.5.4 CM Reactor Without Recycle
Substrate Balance
Biomass Balance
17.5.5 CM Reactor with Recycle
Biomass Balance
17.5.6 Application of the Basic Model in the Historical Context
Frailties of the Historical Models
17.5.7 Matrix Representation of the Basic (Soluble Substrate) Model
17.5.8 The Rate of Recycle
17.5.9 Food-to-Microorganism Ratio and SRT
17.6 Advanced Model for Carbon Removal
17.6.1 Total Effluent COD from the Process
17.6.2 Removal of Influent Particulate Organic Matter
17.6.3 Estimation of Parameters and Calibration of the Advanced Model
17.6.4 Calibration of Models to Existing Data
17.7 Sludge Production in Activated Sludge Systems
17.8 Plug Flow Activated Sludge Treatment
17.9 Variations of the Activated Sludge Process
17.9.1 Sequencing Batch Reactors
17.9.2 Extended Aeration
17.10 Other Activated Sludge Process Variations
17.10.1 Pure Oxygen Activated Sludge Process
17.10.2 Powdered Activated Carbon Activated Sludge Process
Design Parameters and Operating Conditions for Activated Sludge Processes
17.11 Design of Activated Sludge Processes for Nitrogen and Phosphorus Removal
17.11.1 Nitrogen Transformations
Nitrogen Removal–Denitrification
17.11.2 Advanced Denitrification Processes
SHARON Process
Anammox Process
Other Processes
17.11.3 Enhanced Phosphorus Uptake
Fermentation of Primary or Activated Sludge
Phostrip and Bardenpho Bio-P Processes
17.12 Operating Characteristics of Activated Sludge Processes
17.12.1 SRT and Characteristics of Waste Activated Sludge
17.13 Granular Activated Sludge and Membrane Processes
17.13.1 Granular Activated Sludge Processes
17.13.2 Membrane Activated Sludge Processes
Design of Submerged Membrane Reactors
17.14 Fixed-Film Activated Sludge Processes
17.14.1 Integrated Fixed-Film Activated Sludge and Moving Bed Bioreactor Processes
Design of MBBRs
17.14.2 Biologically Activated Filters
Design of Biological Active Filters
17.14.3 Rotating Biological Contact Units
17.15 Fixed-Film Trickling Filter Processes
17.15.1 Trickling Filters
Sludge Production from Trickling Filters
Air Supply in Trickling Filters
Operation of Trickling Filters
17.15.2 Hydraulic Design of Distributors for Trickling Filters
17.16 Oxygen Uptake in Activated Sludge Processes
17.17 Metals Removal in Activated Sludge Processes
17.18 Aerobic Sludge Digestion
17.18.1 Model for Aerobic Sludge Digestion
Oxygen Uptake in Aerobic Digestion
Rate Constants and Sludge Degradability
17.18.2 Thermophilic Aerobic Digestion
Pre-treatment for Aerobic Sludge Digestion
17.18.3 Indicator Microorganism Reduction in Aerobic Digestion
Questions and Problems
References
Chapter 18: Aerobic Biological Treatment: Other Process Operations
18.1 Aeration in Biological Wastewater Treatment
18.1.1 Aeration Devices in Wastewater Treatment
Diffused Aerators
Surface and Other Aerators
18.2 Post-aeration Systems for Wastewater Treatment
18.2.1 Diffused Aeration Systems
18.2.2 Cascades
18.2.3 Weirs
18.3 Type III Sedimentation: Zone Settling
18.3.1 Design of a Basin for Type III Sedimentation
Gravity Flux
Underflow Flux
18.3.2 Secondary Clarifier Design
18.3.3 Modeling for Secondary Clarifier and Operation
18.3.4 Membrane Separation of Solids
Lamella Clarifiers
18.4 Sludge Settling Problems and Foaming
18.4.1 Microorganisms
18.4.2 Selectors and Process Operating Conditions
Questions and Problems
References
Chapter 19: Anaerobic Wastewater Treatment
History
19.1 Anaerobic Metabolism
19.1.1 Hydrolysis
19.1.2 Acid Formation: Acidogenesis and Acetogenesis
19.1.3 Methanogenesis
19.1.4 Other Metabolic Pathways
19.1.5 Environmental Variables
Oxidation–Reduction Potential
Temperature
pH
Mixing
Ammonia and Sulfide Control
Nutrient Requirements
19.2 Process Fundamentals
19.2.1 Solids Yield and Retention Time
19.2.2 Biogas Potential
Biochemical Methane Potential and Anaerobic Toxicity Assay
Methane Production in Anaerobic Treatment
Dissolved Methane
Biogas Utilization
19.3 Process Analysis
19.3.1 Definition of Symbols for the Anaerobic Models
19.3.2 General Model for an Anaerobic Process
Anaerobic Reactor Receiving Only Particulate Substrate
Anaerobic Reactor Receiving Only Soluble Substrate
The Traditional Digester Sizing Equation for Anaerobic Sludge Digesters
19.3.3 Advanced Model for an Anaerobic Process
Substrate Removal and Biomass Accumulation
Temperature Effects on Rate Coefficients
19.4 Misconceptions and Barriers about Anaerobic Treatment
19.5 Anaerobic Treatment Processes
19.5.1 Conventional Anaerobic Treatment
19.5.2 Contact Process
19.5.3 Upflow Anaerobic Sludge Blanket Reactor
19.5.4 Fixed-Film Reactors
Upflow Fixed-Film Reactors
Downflow Fixed-Film Reactors
Fluidized Bed Reactors
19.5.5 Two-Phase Anaerobic Digestion
19.5.6 Thermophilic Digestion
19.5.7 Membrane Anaerobic Treatment
19.5.8 Pre-treatment of Sludge for Anaerobic Digestion of Biosolids
19.6 Anaerobic Digestion of Municipal Solid Waste
19.7 Process Stability and Monitoring
19.7.1 Chemical Precipitation Problems in Anaerobic Digesters
19.7.2 Recovery of Nutrients through Struvite Harvesting
19.7.3 Sludge Production
19.7.4 Anaerobic Treatment of Low-Strength Wastes
19.8 Comparison of Anaerobic and Aerobic Treatment Processes
19.8.1 Pollutant Removal Efficiency
19.8.2 Number and Size of Operations
19.8.3 Energy and Chemical Inputs
19.8.4 Heat Exchanger
19.9 Energy Assessment of Anaerobic and Aerobic Treatment
Anaerobic Versus Aerobic Treatment
Calculation of the Energy Potential of a Waste
19.10 Pathogen Reduction in Anaerobic Processes
Questions and Problems
References
Chapter 20: Treatment in Ponds and Land Systems
20.1 Overview of Stabilization Ponds
20.1.1 Pond Operation
20.1.2 Pond Effluent Quality
20.2 Pond Types
20.3 Design of Pond Systems
20.3.1 Design of Ponds in the Far North
20.3.2 Models for Facultative Ponds
20.3.3 Nitrogen and Phosphorus Removal
20.3.4 Heat Balance for Ponds
20.4 Removal of Suspended Solids from Pond Effluents
20.5 Indicator Microorganism Die-off in Ponds
20.6 Aerated Lagoons
20.7 Treatment of Wastewater in Land Systems
20.7.1 Land Treatment of Wastewater
Measurement of Hydraulic Conductivity
Wastewater Constituents Influencing Land Treatment
20.7.2 Slow Rate Land Application Systems
20.7.3 Soil Aquifer Treatment
20.7.4 Overland Flow Systems
Questions and Problems
References
Section VI: Final Disposal and Impact Analysis
Chapter 21: Sludge Processing and Land Application
21.1 Sludge Characteristics and Conditioning
Sludge Density
Sludge Viscosity
21.2 Sludge Generation and Treatment Processes
21.3 Sludge Conditioning
21.4 Sludge Thickening
21.4.1 Gravity Thickening
21.4.2 Flotation Thickening
21.5 Mechanical Sludge Dewatering
21.5.1 Centrifugation
21.5.2 Vacuum Dewatering
21.5.3 Plate Pressure Filters
21.6 Land Application of Sludge
Questions and Problems
References
Chapter 22: Effluent Disposal in Natural Waters
22.1 Pollutants in Natural Waters
22.1.1 Water Quality Indices
Fish Survival and Temperature
Nutrient Loadings to Lakes
22.2 Loading Equations for Streams
22.2.1 Pollutant Decay in Streams
22.2.2 Conservative Substance
Point Source
Distributed Source
22.2.3 Substances That Are Transformed by One Reaction
Point Source
Distributed Source
22.3 Dissolved Oxygen Variation in a Stream
22.3.1 Nitrification in Natural Waters
22.3.2 Factors Affecting the Dissolved Oxygen Sag Curve
22.3.3 The Reaeration Rate Coefficient
22.3.4 Reaeration at Dams
22.4 Combined Sewer Overflows Abatement
Questions and Problems
References
Chapter 23: Life Cycle Analysis
23.1 Historical Development of LCA
23.2 Why Use LCA; What Are the Objectives; What Are Its Benefits and What Does It Not Do?
23.3 ISO Standards 14040 and 14044
23.4 Definitions of Terms in ISO 14040 and 14044
23.5 Principles Established by ISO 14040
23.6 Key Components of the ISO Standards
23.6.1 Goal and Scope
23.6.2 System Boundaries
Life Cycle Inventory Analysis
23.6.3 Life Cycle Impact Assessment
Selection of Impact Categories, Category Indicators, and Characterization Models
Assignment of LCI Results to the Selected Impact Categories (Classification)
Calculation of Category Indicator Results (Characterization)
Characterization Factors, Midpoints, and Endpoints
Optional Elements of the LCIA
23.6.4 Limitations of LCIA
23.6.5 Interpretation
23.7 Software and Databases
23.8 Examples of Case Studies of LCA in Water and Wastewater Treatment Projects
Questions and Problems
References
Appendix A
A.1 Normal Distribution
A.2 Integrating Factor for Linear Differential Equations of the First Order
References
Author Index
Subject Index
End User License Agreement

Citation preview

Periodic Table of the Elements IA

1 2 3 4 5 6 7

1 H 1.007 3 Li 6.933 11 Na 22.99 19 K 39.1 37 Rb 85.47 55 Cs 132.9 87 Fr (223)

IIA 4 Be 9.012 12 Mg 24.30 20 Ca 40.08 38 Sr 87.62 56 Ba 137.3 88 Ra (226)

1 H 1.0079

IIIB 21 Sc 44.98 39 Y 88.91 57-71

89-103

IVB 22 Ti 47.87 40 Zr 91.22 72 Hf 178.5 104 Rf (261)

VB 23 V 50.94 41 Nb 92.91 73 Ta 180.9 105 Ha (262)

Atomic number Atomic mass

VIB 24 Cr 52.00 42 Mo 95.95 74 W 183.8 106 Sg (263)

VIIB 25 Mn 54.94 43 Tc (97) 75 Re 186.2 107 Ns (262)

VIII 26 Fe 55.85 44 Ru 101.1 76 Os 190.2 108 Hs

27 Co 58.93 45 Rh 102.9 77 Ir 192.2 109 Mt 192.2

28 Ni 58.69 46 Pd 106.4 78 Pt 195.1 110 Ds (223)

IB 29 Cu 63.55 47 Ag 107.9 79 Au 197.9 111 Rg (226)

IIB 30 Zn 65.38 48 Cd 112.4 80 Hg 200.6 112 Cn (103)

IIIA 5 B 10.8 13 Al 26.98 31 Ga 69.72 49 In 114.8 81 TI 204.3 113 Uut

IVA 6 C 12.09 14 Si 28.08 32 Ge 72.63 50 Sn 118.7 82 Pb 207.2 114 Fl

VA 7 N 14.00 15 P 30.97 33 As 74.92 51 Sb 121.8 83 Bi 209.0 115 Uup

VIA 8 O 15.99 16 S 32.06 34 Se 78.97 52 Te 127.6 84 Po (209) 116 Lv

VIIA 9 F 19.00 17 Cl 35.45 35 Br 79.90 53 I 126.9 85 At (210) 117 Uus

Alkali Alkaline metals earth metals

Lanthanide series Actinide series

57 La 138.9 89 Ac (227)

58 Ce 140.1 90 Th 232.0

59 Pr 140.9 91 Pa 231.9

60 Nd 144.24 92 U 238.0

61 Pm (145) 93 Np (237)

62 Sm 150.4 94 Pu (244)

63 Eu 152.9 95 Am (243)

64 Gd 157.3 96 Cm (247)

65 Tb 158.9 97 Bk (247)

66 Dy 162.5 98 Cf (251)

67 Ho 164.9 99 Es (252)

68 Er 167.3 100 Fm (257)

69 Tm 168.9 101 Md (258)

70 Yb 173.0 102 No (259)

71 Lu 175.0 103 Lr (262)

0 2 He 4.003 10 Ne 20.18 18 Ar 39.95 36 Kr 83.80 54 Xe 131.3 86 Rn (222) 118 Uuo

FFIRS

07/11/2018

13:50:59

Page ii

FFIRS

07/11/2018

13:50:59

Page iii

Theory and Practice of Water and Wastewater Treatment Second Edition

Ronald Droste Department of Civil Engineering University of Ottawa Ontario, Canada

Ronald Gehr Department of Civil Engineering and Applied Mechanics McGill University Montreal, Canada

FFIRS

07/11/2018

13:50:59

Page iv

This edition first published 2019.  2019 by John Wiley & Sons, Inc. Edition History John Wiley & Sons, Inc. (1e, 1997) All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by law. Advice on how to obtain permission to reuse material from this title is available at http://www.wiley.com/go/permissions. The right of Ronald L. Droste and Ronald Gehr to be identified as the authors of this work has been asserted in accordance with law. Registered Office John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USA Editorial Office 111 River Street, Hoboken, NJ07030, USA For details of our global editorial offices, customer services, and more information about Wiley products visit us at www.wiley.com. Wiley also publishes its books in a variety of electronic formats and by print-on-demand. Some content that appears in standard print versions of this book may not be available in other formats. Limit of Liability/Disclaimer of Warranty In view of ongoing research, equipment modifications, changes in governmental regulations, and the constant flow of information relating to the use of experimental reagents, equipment, and devices, the reader is urged to review and evaluate the information provided in the package insert or instructions for each chemical, piece of equipment, reagent, or device for, among other things, any changes in the instructions or indication of usage and for added warnings and precautions. While the publisher and authors have used their best efforts in preparing this work, they make no representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives, written sales materials or promotional statements for this work. The fact that an organization, website, or product is referred to in this work as a citation and/or potential source of further information does not mean that the publisher and authors endorse the information or services the organization, website, or product may provide or recommendations it may make. This work is sold with the understanding that the publisher is not engaged in rendering professional services. The advice and strategies contained herein may not be suitable for your situation. You should consult with a specialist where appropriate. Further, readers should be aware that websites listed in this work may have changed or disappeared between when this work was written and when it is read. Neither the publisher nor authors shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages. Library of Congress Cataloging-in-Publication Data Names: Droste, Ronald L., author. | Gehr, Ronald L., author. Title: Theory and practice of water and wastewater treatment / by Ronald L Droste and Ronald L Gehr. Description: Hoboken, NJ, USA : Wiley, 2019. | Includes bibliographical references and index. | Identifiers: LCCN 2018012644 (print) | LCCN 2018013671 (ebook) | ISBN 9781119312376 (pdf) | ISBN 9781119312383 (epub) | ISBN 9781119312369 (cloth) Subjects: LCSH: Water–Purification. | Sewage–Purification. Classification: LCC TD430 (ebook) | LCC TD430 .D78 2019 (print) | DDC 628.1/62–dc23 LC record available at https://lccn.loc.gov/2018012644 Cover image: Courtesy of Ronald Gehr Cover design by Wiley Set in 10/12 pt WarnockPro-Regular by Thomson Digital, Noida, India 10 9 8 7 6

5 4 3 2

1

FTOC

07/16/2018

13:22:8

Page v

v

Contents Acknowledgments XXI Preface XXIII Abbreviations and Acronyms Used in the Text About the Companion Website XXXIII

Section I: 1 1.1 1.2 1.3 1.3.1 1.4 1.5 1.6 1.7 1.7.1 1.7.2 1.8 1.8.1

1.8.2 1.9 1.10 1.11 1.12 1.13 1.13.1

Chemistry

XXV

1

3 Definitions 3 The Expression of Concentration 4 Ions and Molecules in Water 5 Oxidation Number 5 Balancing Reactions 9 Oxidation–Reduction Reactions 10 Equilibrium 12 Conductivity and Ionic Strength 13 Conductance 14 Ionic Strength 14 Chemical Kinetics 15 Other Formulations 16 Consecutive or Series 16 Parallel 17 Retardant 17 Autocatalytic 17 Catalysis 18 The Effect of Temperature on Rate of Reaction Gas Laws 19 Gas Solubility: Henry’s Law 20 Solubility Product 23 Complexes 25 Nuclear Chemistry 27 Radioactivity Units 27 Questions and Problems 30 References 33

Basic Chemistry

19

FTOC

07/16/2018

vi

13:22:8

Page vi

Contents

2.1.2 2.2 2.2.1 2.2.2 2.3 2.3.1 2.3.2

35 Thermodynamic Relations 35 Free Energy 35 Expression of Concentration in Equilibrium Expressions 39 Enthalpy and Temperature Effects on the Equilibrium Constant 42 Redox Potentials 43 Cell or Couple Potential 46 Oxidation–Reduction Potential and System Potential 48 Corrosion 49 Microbial Corrosion 51 Corrosion Prevention from External Environmental Factors 52 Galvanic Cathodic Protection 52 Electrolytic (or Impressed Current) Cathodic Protection 53 Questions and Problems 53 References 55

3

Acid–Base Chemistry

2

2.1 2.1.1

3.1 3.2 3.2.1 3.3 3.4 3.5 3.5.1 3.5.2 3.6 3.6.1 3.6.2 3.6.3 3.7 3.7.1

4

4.1 4.2 4.3 4.4 4.4.1

4.4.2 4.5 4.5.1 4.6 4.6.1 4.6.2

The Thermodynamic Basis for Equilibrium

57 pH 57 Acids and Bases 58 Conjugate Acids and Bases 61 Equivalents and Normality 61 Solution of Multiequilibria Systems 62 Buffers 63 Dilution of a Buffered Solution 65 The Most Effective pH for a Buffer 65 Acid–Base Titrations 66 Titration of Strong Acids and Bases 66 Titration of Weak Acids and Bases 68 Indicating the Endpoint of an Acid–Base Titration 71 Natural Buffering of Waters from Carbon Dioxide and Related

Compounds 73 Acidity and Alkalinity 74 Questions and Problems 76 References 78 81 Carbon 81 Properties of Organic Compounds 81 Functional Groups 82 Types of Organic Compounds 83 Aliphatic Compounds 83 Aldehydes and Ketones 83 Alcohols, Esters, and Ethers 83 Nitrogen-containing Compounds 83 Aromatic Compounds 84 Compounds of Sulfur 85 Naturally Occurring Organic Compounds Carbohydrates 85 Proteins 86

Organic and Biochemistry

85

FTOC

07/16/2018

13:22:8

Page vii

Contents

4.6.3 4.7 4.8 4.9 4.10

Fats and Oils 86 Biochemistry 86 Glycolysis 87 The Tricarboxylic Acid Cycle Enzyme Kinetics 89 Questions and Problems 91 References 93

5 5.1 5.1.1 5.1.2 5.1.3 5.2 5.2.1 5.3 5.3.1 5.3.2 5.4 5.4.1

95 Titration 95 Complex and Precipitate Formation Titrations 95 Redox Titrations and Potentiometric Analyses 96 Indicators for Potentiometric Analysis 98 Colorimetric Analyses 99 The Beer–Lambert Laws for Light Transmittance 99 Physical Analyses 99 Solids 99 Turbidity and Color 101 Determination of Organic Matter 102 Chemical Oxygen Demand 103 General Reaction for COD 104 Interferences with the COD Test 105 Biochemical Oxygen Demand 105 Effects of Temperature on BOD Exertion 108 Carbonaceous and Nitrogenous BOD 109 Laboratory Methods for Determining BOD 110 Limitations of the BOD Test for Biological Wastewater Treatment

Process Design 110 Analysis of a BOD Progression 111 Total Organic Carbon 113 Questions and Problems 113 References 118

5.4.2

5.4.3

Analyses and Constituents in Water

Section II: 6

6.1 6.2 6.2.1

6.3 6.3.1 6.3.2

88

Microorganisms in Water and Water Quality

121 Groups of Microorganisms and the Phylogenetic Tree Bacteria and Archaea 121 Classification of Bacteria 124 Taxonomy 124 Metabolic Requirements 125 Oxygen Requirements 125 Temperature 126 Salt and Sugar Concentrations 127 pH 127 Eukaryotes 127 Algae 128 Fungi 129

119

Microbiology

121

vii

FTOC

07/16/2018

viii

13:22:8

Page viii

Contents

6.3.3 6.4 6.4.1 6.4.2 6.4.3 6.5 6.5.1 6.5.2 6.5.3 6.5.4

6.5.5

Protists 129 Other Microorganisms 130 Viruses and Phages 130 Rotifers 131 Worms 131 Determining the Growth of Microorganisms 132 Growth of Pure Cultures 132 Growth of Mixed Cultures 135 Viability and Mass in Growing Cultures 136 Enumeration of Microorganisms 136 Plate Counts 136 Practical Considerations in Determining Mean Values 140 Microbial Genomics and Molecular Microbiology Tools 141 Phylogenetic Microbial Community Composition Analysis 141 Functional Analysis 142 Questions and Problems 143 References 145 147 Agents of Disease 147 Bacterial Pathogens 147 Viral Pathogens 149 Protozoan Pathogens 150 Helminths 150 Insect and Animal Vectors of Disease 153 Indicator, Test, and Model Microorganisms 153 Indicators of Fecal Contamination 155 Indicator Microorganisms 156 Coliforms: Total, Thermotolerant, and E. coli 156 Enterococci 157 Surrogates 157 Survival of Microorganisms in the Aquatic Environment Minimum Infective Dose 162 Questions and Problems 163 References 164

7 7.1 7.1.1 7.1.2 7.1.3 7.1.4 7.1.5 7.2 7.3 7.4 7.4.1 7.4.2 7.5 7.6 7.7

Water, Wastes, and Disease

8

Water Constituents and Quality Standards

8.1 8.2 8.2.1 8.2.2

Toxicity of Elements and Compounds Contaminants in Water 170 Emerging Contaminants 171 Common Contaminants 173 Aluminum 173 Nitrate 173 Fluoride 173 Detergents 174 Carcinogens 174 Radioactive Constituents 175 Taste and Odor 176 Bases for Standards 178

8.2.3 8.2.4 8.3 8.4

167 167

159

FTOC

07/16/2018

13:22:9

Page ix

Contents

8.4.1 8.4.2 8.4.3 8.4.4 8.5 8.5.1 8.5.2 8.5.3 8.6 8.6.1

8.6.2 8.6.3 8.6.4 8.6.5 8.7 8.8 8.9 8.10

Risk Assessment for Microbial Infection 179 Determination of Carcinogenicity 180 Toxicity Determination 182 Environmental Water Quality Standards 184 Standards for Drinking Water 184 International Drinking Water Standards 185 US Safe Drinking Water Act 185 Canadian Water Quality Guidelines 186 Comparison of Drinking Water Standards 187 Microbiological Parameters 187 WHO Guidelines for Microbiological Quality 187 United States Standards for Microbiological Quality 187 Canadian Guidelines for Microbiological Quality 188 Chemical and Physical Qualities 188 Aesthetic Quality 188 Radiological Constituents 188 Other Water Standards 192 Water Consumption 192 Canadian Federal Wastewater Quality Guidelines 195 Wastewater Characteristics 195 Greywater 196 Wastewater Production 197 Questions and Problems 198 References 200 Section III: Water and Wastewater Treatment

9

9.1

9.1.1 9.2 9.3

10 10.1 10.1.1 10.1.2

10.1.3 10.1.4

205

207 Water Treatment Operations 207 Microbial Contaminants 212 Reservoirs 213 Home Water Treatment Units 216 Wastewater Treatment Unit Operations 216 Hydraulic Design of Water and Wastewater Treatment

Plants 225 Flow in Pressurized Pipes 225 Flow in Open Channels 226 Other Losses 227 Questions and Problems 230 References 232

Water and Wastewater Treatment Operations

Mass Balances and Hydraulic Flow Regimes

Setup of Mass Balances 235 Mixing Characteristics of Basins 236 Mass Balances for PF Reactors 237 Method I 238 Method II 239 Method III 239 Mass Balances and Reaction for CM Basins Batch Processes 244

235

242

ix

FTOC

07/16/2018

x

13:22:9

Page x

Contents

10.2 10.2.1 10.2.2 10.2.3 10.2.4 10.2.5 10.2.6 10.3 10.3.1 10.3.2 10.3.3 10.4 10.5

Flow Analysis of CM and PF Reactors 245 Tracer Analysis of Complete Mixed Reactors 245 Tracer Analysis of Plug Flow 247 Complete Mixed Reactors in Series 247 Other Flow Irregularities: Dead Volume and Short-circuiting 248 Typical Flow Characteristics of Basins 249 Measurement of Dispersion 250 Detention Time in Vessels 250 Average Detention Time 251 The Effects of Flow Recycle on Detention Time 251 The Effects of Recycle on Mixing 253 Flow and Quality Equalization 253 System Material Balances 256 Questions and Problems 266 References 271

Section IV: Physical–Chemical Treatment Processes 11

11.1 11.1.1 11.1.2 11.1.3 11.2 11.2.1 11.3 11.3.1

11.3.2 11.3.3 11.3.4 11.4 11.4.1 11.5 11.5.1 11.5.2 11.5.3 11.5.4 11.6 11.7 11.8 11.8.1 11.8.2 11.8.3 11.9 11.9.1

273

275 Screens and Bar Racks 275 Screens for Water Treatment Plants 276 Screens at Wastewater Treatment Plants 277 Microstrainers 277 Sedimentation 278 Particle Settling Velocity 279 Grit Chambers 281 Horizontal Flow Grit Chambers 282 Channel with Varying Cross Section 283 Design Notes for a Parabolic Grit Chamber 284 Aerated Grit Chambers 290 Square Tank Degritter 292 Vortex Grit Removal Devices 293 Grit Washing 294 Type I Sedimentation 294 Theory 294 Type II Sedimentation 297 Laboratory Determination of Settling Velocity Distribution 298 Type II Sedimentation Data Analysis 298 Alternative Method for Calculating Total Removal 302 Sizing the Basin 303 Tube and Lamella Clarifiers 303 Weir–Launder Design 309 Clarifier Design for Water and Primary Wastewater Treatment 313 Design Ranges for Typical Clarifiers for Water and Wastewater Treatment 313 Chemically Enhanced Primary Treatment 315 Depth in Sedimentation Basins 318 Inlet Hydraulics for Sedimentation Basins 319 Flow Distributions 319 Screening and Sedimentation

FTOC

07/16/2018

13:22:9

Page xi

Contents

11.9.2

Inlet Baffling 322 Questions and Problems References 328

323

331 Fick’s Law 331 Gas Transfer 332 Calculating the Mass Transfer Coefficient 335 The Effects of pH on Mass Transfer 336 Aeration in Water and Wastewater Treatment 336 Hazards Associated with Oxygen, Carbon monoxide, and Hydrogen sulfide Design of Aeration Systems 339 Gravity Aerators 339 Spray Aerators 341 Diffused Aerators 344 Questions and Problems 346 References 348

12 12.1 12.2 12.2.1 12.2.2 12.3 12.3.1 12.4 12.4.1 12.4.2 12.4.3

Mass Transfer and Aeration

13

351 Coagulation 351 Recovery of Alum and Iron Coagulants 355 Mixing and Power Dissipation 356 Mixers 358 Mechanical Mixers 359 Pneumatic Mixers 362 Hydraulic Mixers 363 Venturi Sections and Hydraulic Jumps 363 Flocculators 368 Paddle Flocculators 369 Vertical-Shaft Turbine Flocculators 375 Pipes 376 Baffled Channels 376 Upflow Solids Contact Clarifier 377 Alabama Flocculator 377 Spiral Flow Tanks 378 Pebble Bed Flocculators 379 Ballasted Flocculation 380 Questions and Problems 382 References 384

13.1 13.2 13.3 13.3.1 13.3.2 13.3.3 13.4 13.4.1 13.4.2 13.4.3 13.4.4 13.4.5 13.4.6 13.4.7 13.4.8 13.4.9

14

14.1 14.2 14.2.1 14.3 14.3.1 14.4 14.4.1

Coagulation and Flocculation

387 Slow Sand Filters and Rapid Filters 388 Filtering Materials 389 Grain Size and Distribution 389 Headloss in Filters 394 Grain Size Distribution and Headloss 397 Backwashing Filters 398 Total Head Requirements for Backwashing 400 Losses in the Expanded Media 400

Filtration

338

xi

FTOC

07/16/2018

xii

13:22:9

Page xii

Contents

14.4.2

Backwash Velocity 401 Method 1 401 Method 2 402 Headloss and Expansion in a Stratified Bed 405 14.5 Support Media and Underdrains in Rapid Filters 409 Other Design Features of Filters 411 Auxiliary Wash and Air Scour Systems 411 14.6 Filter Beds for Water and Wastewater Treatment 412 14.7 Air Binding of Filters 415 14.8 Rapid Filtration Alternatives 417 14.8.1 Single-medium and Multimedia Filters 417 14.8.2 Constant- and Declining-rate Filtration 417 14.8.3 Direct Filtration 418 14.9 Pressure Filters 419 14.10 Slow Sand Filters 419 14.10.1 Slow Sand Filters for Tertiary Wastewater Treatment 421 14.11 Biological Filtration for Water Treatment 421 Questions and Problems 424 References 427 431 Water Softening 431 Lime–Soda Softening 433 Treatment Methods for Lime–Soda Hardness Removal 434 Bar Graphs 439 Lime Recovery and Sludge Reduction 441 15.3 Corrosion Prevention in Water Supply Systems 441 15.3.1 The Langelier Index Misconception 443 15.4 Iron and Manganese Removal 447 15.4.1 Greensand 448 15.4.2 Aeration 449 15.4.3 Sequestering Iron and Manganese 449 15.4.4 Biological Removal of Iron and Manganese 449 15.5 Phosphorus Removal from Wastewater by Chemical

Precipitation 450 15.5.1 Removal of Phosphorus by Chemically Reactive Species 452 15.6 Removal of Arsenic and Metals 453 15.6.1 Metals Removal 453 15.6.2 Arsenic Removal 454 15.7 Advanced Oxidation Processes 455 15.8 Ion Exchange 456 15.8.1 Activated Alumina 457 15.8.2 Ammonia and Nitrate Removal by Ion Exchange 458 15.9 Fluoridation and Defluoridation 458 15.10 Membrane Processes 460 15.10.1 Assessment of Water Suitability for Membrane Treatment 466 15.10.2 Concentrate Disposal 468 15.10.3 Membranes for Water Treatment 468 Microfiltration and Ultrafiltration Systems 468

15

15.1 15.2 15.2.1 15.2.2

Physical–Chemical Treatment for Dissolved Constituents

FTOC

07/16/2018

13:22:9

Page xiii

Contents

15.11 15.11.1 15.11.2 15.11.3 15.12 15.12.1 15.12.2

15.12.3 15.12.4 15.12.5 15.12.6

16 16.1 16.2 16.2.1 16.2.2 16.2.3 16.2.4 16.2.5 16.2.6 16.2.7 16.2.8 16.2.9 16.3 16.4 16.4.1 16.4.2

16.4.3 16.5 16.5.1

16.5.2 16.6 16.6.1 16.6.2 16.6.3 16.6.4

Nanofiltration and Reverse Osmosis Treatment 469 Electrodialysis 472 Activated Carbon Adsorption 472 Activated Carbon – Preparation and Characteristics 473 Adsorption Isotherms 474 Granular Activated Carbon Adsorbers 477 Design of Fixed-bed Adsorbers 478 Rate Formulation for Adsorption 479 Theory of Fixed-bed Adsorber Systems 480 The Capacity Utilized in the Adsorption Zone 481 Competitive Adsorption 490 Bed-depth Service Time Method 490 Rapid Small-Scale Column Tests 494 Granular Activated Carbon Reactors in Series 498 Design of a Suspended Media PAC or GAC Continuous

Flow Reactor 498 Questions and Problems 499 References 503 509 Kinetics of Disinfection 510 Chlorination 512 Chemistry of Chlorine 512 Measurement of Free and Residual Chlorine 516 Chlorine Decay 517 Drinking Water Disinfection by Chlorine 518 Wastewater Disinfection by Chlorine 519 Design of Contacting Systems for Chlorine 521 Disinfection as the Sole Treatment of Surface Water 521 Other Applications of Chlorine 522 Dechlorination 522 Chloramines 523 Chlorine Dioxide 524 Chlorine Dioxide Doses as a Primary Disinfectant 525 Chlorine Dioxide for Pre-disinfection or for Residual

Disinfection 525 Generation of Chlorine Dioxide 526 Peracids: Peracetic Acid (PAA) and Performic Acid (PFA) 527 Peracetic Acid 527 Kinetics of Disinfection Using PAA 528 Measuring PAA Residuals 529 Applications for Wastewater Disinfection 530 Chemical Disinfection Process Control 530 Performic Acid 531 Ozone 531 Determining the Appropriate Ozone Dose 532 Ozone Generation 533 Ozone Dissolution Systems 534 Ozone Contactor Basins 535 Disinfection

xiii

FTOC

07/16/2018

xiv

13:22:9

Page xiv

Contents

16.6.5 16.6.6 16.6.7 16.7 16.7.1 16.7.2

16.7.3 16.7.4 16.7.5 16.7.6 16.7.7 16.7.8 16.8 16.9 16.9.1 16.9.2 16.9.3 16.9.4 16.9.5 16.9.6 16.9.7 16.10

Ozone Chemistry: Mass Transfer Coefficients and Radicals

Production 536 Ozone for Wastewater Disinfection 537 Ozone for Destruction of Micropollutants 538 Ultraviolet Radiation 538 Mechanism of UV Disinfection 538 Repair of UV Damage 539 Photo Repair 539 Dark Repair 540 Interferences 540 Generation of Ultraviolet Light and Ultraviolet Reactors 541 Disinfection Kinetics 541 Disinfection Doses (or Fluences) 542 Determination of UV Fluence 542 Ultraviolet Reactors 545 Point-of-use Disinfectants: Solar Disinfection (SODIS), with or

without Photoreactants such as TiO2 547 Disinfection Byproducts 548 Chlorine 549 Chloramines 549 Chlorine Dioxide 550 Peracids 550 Ozone 550 Ultraviolet 551 Comparative Risks 551 Disinfection to Combat Invasive Species 551 Questions and Problems 553 References 556

Section V: Biological Wastewater Treatment 17

17.1 17.2 17.3 17.3.1

17.3.2 17.4 17.4.1 17.4.2 17.4.3 17.4.4 17.5 17.5.1 17.5.2

565

567 Microorganisms in Aerobic Biological Treatment 567 The Activated Sludge Process 568 Substrate Removal and Growth of Microorganisms 569 Substrate Removal 569 Temperature Dependence of Rate Coefficients 571 BOD, COD, and TOC Removal 571 Growth of Microorganisms and Biological Sludge Production 572 Sludge Composition and Nutrient Requirements 573 Activated Sludge Configurations 574 Definition of Symbols for the Activated Sludge Process Models 575 Reactor 577 System Effluent and Waste Sludge Line 577 Clarifier 577 Process Analysis 578 Physical Concentration of Solids in the Bioreactor 578 Solids Retention Time 580

Aerobic Biological Treatment: Biotreatment Processes

FTOC

07/16/2018

13:22:9

Page xv

Contents

17.5.3 17.5.4

17.5.5 17.5.6 17.5.7 17.5.8 17.5.9 17.6 17.6.1 17.6.2 17.6.3 17.6.4 17.7 17.8 17.9 17.9.1 17.9.2 17.10 17.10.1 17.10.2 17.11 17.11.1 17.11.2

17.11.3

17.12 17.12.1 17.13 17.13.1 17.13.2 17.14 17.14.1 17.14.2 17.14.3 17.15 17.15.1

Sludge Volume Index 580 CM Reactor Without Recycle 582 Substrate Balance 582 Biomass Balance 583 CM Reactor with Recycle 585 Biomass Balance 585 Application of the Basic Model in the Historical Context 586 Frailties of the Historical Models 590 Matrix Representation of the Basic (Soluble Substrate) Model 591 The Rate of Recycle 593 Food-to-Microorganism Ratio and SRT 594 Advanced Model for Carbon Removal 596 Total Effluent COD from the Process 599 Removal of Influent Particulate Organic Matter 599 Estimation of Parameters and Calibration of the Advanced Model 600 Calibration of Models to Existing Data 602 Sludge Production in Activated Sludge Systems 604 Plug Flow Activated Sludge Treatment 607 Variations of the Activated Sludge Process 609 Sequencing Batch Reactors 609 Extended Aeration 612 Other Activated Sludge Process Variations 613 Pure Oxygen Activated Sludge Process 615 Powdered Activated Carbon Activated Sludge Process 615 Design Parameters and Operating Conditions for Activated Sludge Processes 615 Design of Activated Sludge Processes for Nitrogen and Phosphorus Removal 616 Nitrogen Transformations 616 Nitrogen Removal–Denitrification 621 Advanced Denitrification Processes 626 SHARON Process 626 Anammox Process 627 Other Processes 628 Enhanced Phosphorus Uptake 628 Fermentation of Primary or Activated Sludge 630 Phostrip and Bardenpho Bio-P Processes 632 Operating Characteristics of Activated Sludge Processes 632 SRT and Characteristics of Waste Activated Sludge 632 Granular Activated Sludge and Membrane Processes 634 Granular Activated Sludge Processes 634 Membrane Activated Sludge Processes 635 Design of Submerged Membrane Reactors 637 Fixed-Film Activated Sludge Processes 639 Integrated Fixed-Film Activated Sludge and Moving Bed Bioreactor Processes 639 Design of MBBRs 641 Biologically Activated Filters 645 Design of Biological Active Filters 647 Rotating Biological Contact Units 648 Fixed-Film Trickling Filter Processes 650 Trickling Filters 650

xv

FTOC

07/16/2018

xvi

13:22:9

Page xvi

Contents

17.15.2 17.16 17.17 17.18 17.18.1

17.18.2 17.18.3

Sludge Production from Trickling Filters 656 Air Supply in Trickling Filters 656 Operation of Trickling Filters 660 Hydraulic Design of Distributors for Trickling Filters 660 Oxygen Uptake in Activated Sludge Processes 663 Metals Removal in Activated Sludge Processes 664 Aerobic Sludge Digestion 664 Model for Aerobic Sludge Digestion 665 Oxygen Uptake in Aerobic Digestion 668 Rate Constants and Sludge Degradability 668 Thermophilic Aerobic Digestion 669 Pre-treatment for Aerobic Sludge Digestion 672 Indicator Microorganism Reduction in Aerobic Digestion 672 Questions and Problems 673 References 680

18

Aerobic Biological Treatment: Other Process Operations

18.1 18.1.1

Aeration in Biological Wastewater Treatment 689 Aeration Devices in Wastewater Treatment 692 Diffused Aerators 692 Surface and Other Aerators 692 Post-aeration Systems for Wastewater Treatment 697 Diffused Aeration Systems 697 Cascades 699 Weirs 699 Type III Sedimentation: Zone Settling 700 Design of a Basin for Type III Sedimentation 703 Gravity Flux 703 Underflow Flux 704 Secondary Clarifier Design 708 Modeling for Secondary Clarifier and Operation 709 Membrane Separation of Solids 711 Lamella Clarifiers 712 Sludge Settling Problems and Foaming 712 Microorganisms 712 Selectors and Process Operating Conditions 713 Questions and Problems 715 References 718

18.2 18.2.1 18.2.2 18.2.3 18.3 18.3.1

18.3.2 18.3.3 18.3.4 18.4 18.4.1 18.4.2

19

19.1 19.1.1 19.1.2 19.1.3 19.1.4 19.1.5

721 History 721 Anaerobic Metabolism 722 Hydrolysis 722 Acid Formation: Acidogenesis and Acetogenesis Methanogenesis 724 Other Metabolic Pathways 725 Environmental Variables 725 Oxidation–Reduction Potential 725 Temperature 725

Anaerobic Wastewater Treatment

723

689

FTOC

07/16/2018

13:22:9

Page xvii

Contents

19.2 19.2.1 19.2.2

19.3 19.3.1 19.3.2

19.3.3

19.4 19.5 19.5.1 19.5.2 19.5.3 19.5.4

19.5.5 19.5.6 19.5.7 19.5.8 19.6 19.7 19.7.1 19.7.2 19.7.3 19.7.4 19.8 19.8.1 19.8.2 19.8.3 19.8.4 19.9

19.10

pH 725 Mixing 726 Ammonia and Sulfide Control 726 Nutrient Requirements 727 Process Fundamentals 727 Solids Yield and Retention Time 727 Biogas Potential 729 Biochemical Methane Potential and Anaerobic Toxicity Assay 729 Methane Production in Anaerobic Treatment 730 Dissolved Methane 731 Biogas Utilization 732 Process Analysis 732 Definition of Symbols for the Anaerobic Models 733 General Model for an Anaerobic Process 734 Anaerobic Reactor Receiving Only Particulate Substrate 734 Anaerobic Reactor Receiving Only Soluble Substrate 737 The Traditional Digester Sizing Equation for Anaerobic Sludge Digesters Advanced Model for an Anaerobic Process 740 Substrate Removal and Biomass Accumulation 741 Temperature Effects on Rate Coefficients 747 Misconceptions and Barriers about Anaerobic Treatment 747 Anaerobic Treatment Processes 750 Conventional Anaerobic Treatment 750 Contact Process 753 Upflow Anaerobic Sludge Blanket Reactor 754 Fixed-Film Reactors 756 Upflow Fixed-Film Reactors 757 Downflow Fixed-Film Reactors 758 Fluidized Bed Reactors 759 Two-Phase Anaerobic Digestion 759 Thermophilic Digestion 760 Membrane Anaerobic Treatment 760 Pre-treatment of Sludge for Anaerobic Digestion of Biosolids 760 Anaerobic Digestion of Municipal Solid Waste 762 Process Stability and Monitoring 763 Chemical Precipitation Problems in Anaerobic Digesters 764 Recovery of Nutrients through Struvite Harvesting 764 Sludge Production 766 Anaerobic Treatment of Low-Strength Wastes 766 Comparison of Anaerobic and Aerobic Treatment Processes 767 Pollutant Removal Efficiency 768 Number and Size of Operations 768 Energy and Chemical Inputs 769 Heat Exchanger 770 Energy Assessment of Anaerobic and Aerobic Treatment 774 Anaerobic Versus Aerobic Treatment 776 Calculation of the Energy Potential of a Waste 777 Pathogen Reduction in Anaerobic Processes 777 Questions and Problems 778 References 781

737

xvii

FTOC

07/16/2018

xviii

13:22:9

Page xviii

Contents

20 20.1 20.1.1 20.1.2 20.2 20.3 20.3.1 20.3.2 20.3.3 20.3.4 20.4 20.5 20.6 20.7 20.7.1

20.7.2 20.7.3 20.7.4

789 Overview of Stabilization Ponds 789 Pond Operation 790 Pond Effluent Quality 791 Pond Types 792 Design of Pond Systems 795 Design of Ponds in the Far North 796 Models for Facultative Ponds 798 Nitrogen and Phosphorus Removal 798 Heat Balance for Ponds 799 Removal of Suspended Solids from Pond Effluents 800 Indicator Microorganism Die-off in Ponds 801 Aerated Lagoons 802 Treatment of Wastewater in Land Systems 804 Land Treatment of Wastewater 804 Measurement of Hydraulic Conductivity 805 Wastewater Constituents Influencing Land Treatment 807 Slow Rate Land Application Systems 807 Soil Aquifer Treatment 814 Overland Flow Systems 815 Questions and Problems 817 References 819 Treatment in Ponds and Land Systems

Section VI: Final Disposal and Impact Analysis 21

21.1

21.2 21.3 21.4 21.4.1 21.4.2 21.5 21.5.1 21.5.2 21.5.3 21.6

22

22.1 22.1.1

22.2

823

825 Sludge Characteristics and Conditioning 825 Sludge Density 825 Sludge Viscosity 827 Sludge Generation and Treatment Processes 828 Sludge Conditioning 833 Sludge Thickening 836 Gravity Thickening 836 Flotation Thickening 837 Mechanical Sludge Dewatering 839 Centrifugation 840 Vacuum Dewatering 843 Plate Pressure Filters 846 Land Application of Sludge 847 Questions and Problems 854 References 856 Sludge Processing and Land Application

859 Pollutants in Natural Waters 859 Water Quality Indices 859 Fish Survival and Temperature 862 Nutrient Loadings to Lakes 864 Loading Equations for Streams 865

Effluent Disposal in Natural Waters

FTOC

07/16/2018

13:22:9

Page xix

Contents

22.2.1 22.2.2

22.2.3

22.3 22.3.1 22.3.2 22.3.3 22.3.4 22.4

23

23.1 23.2 23.3 23.4 23.5 23.6 23.6.1 23.6.2 23.6.3

23.6.4 23.6.5 23.7 23.8

Pollutant Decay in Streams 865 Conservative Substance 866 Point Source 866 Distributed Source 866 Substances That Are Transformed by One Reaction 866 Point Source 866 Distributed Source 867 Dissolved Oxygen Variation in a Stream 870 Nitrification in Natural Waters 873 Factors Affecting the Dissolved Oxygen Sag Curve 874 The Reaeration Rate Coefficient 877 Reaeration at Dams 878 Combined Sewer Overflows Abatement 878 Questions and Problems 881 References 883 887 Historical Development of LCA 888 Why Use LCA; What Are the Objectives; What Are Its Benefits and What Does It

Not Do? 888 ISO Standards 14040 and 14044 889 Definitions of Terms in ISO 14040 and 14044 889 Principles Established by ISO 14040 890 Key Components of the ISO Standards 891 Goal and Scope 892 System Boundaries 892 Life Cycle Inventory Analysis 893 Life Cycle Impact Assessment 894 Selection of Impact Categories, Category Indicators, and Characterization

Models 894 Assignment of LCI Results to the Selected Impact Categories (Classification) 895 Calculation of Category Indicator Results (Characterization) 895 Characterization Factors, Midpoints and Endpoints 896 Optional Elements of the LCIA 897 Limitations of LCIA 898 Interpretation 898 Software and Databases 899 Examples of Case Studies of LCA in Water and Wastewater Treatment

Projects 899 Questions and Problems 906 References 909 Life Cycle Analysis

Appendix A

913

Author Index

927

Subject Index

937

xix

FTOC

07/16/2018

13:22:9

Page xx

ACKNOWLEGMENT

07/12/2018

12:50:49

Page xxi

xxi

Acknowledgments This book is dedicated to Marie Durocher and Mary Ruth Gehr, who have shown endless support. The authors are deeply indebted to previous instructors, colleagues, and students who have contributed to this work in numerous ways. Among the many who have made useful suggestions, we wish to express our gratitude for extraordinary efforts to the following people: Dr. J Dudley, Dr. D Frigon, Ms. NM Gaudet, Mr. S Humphries, Mr. Z Lukic, Mr R Marshall, Dr. RM Narbaitz, Dr. S Pavlostathis, Mr. R Reardon, Ms. E Penkarski-Rodon, Dr. Banu Örmeci, and Kinjal Mehulkumar Shah.

ACKNOWLEGMENT

07/12/2018

12:50:49

Page xxii

PREFACE

05/12/2018

2:33:1

Page xxiii

xxiii

Preface We had a colleague, a mechanical engineer, who taught a first-year course in mechanics. As his first exercise, he posed the question “What is the greatest engineering achievement?” to his classes. The students’ answers ranged from rockets, to computers to automobiles. His answer was the availability of safe drinking water (which distinguishes a developed country from a developing country) in adequate amounts and removal of wastewater. We, of course, agree and hope to make an incremental advancement through this monograph. The nature of this text has not changed from the first to the second edition, but the technological developments in the 21 years since the first edition was published have been astounding. The amount of information available now, in the age of web searching and other electronic resources, is overwhelming. A simple literature search with carefully chosen keywords usually brings up many more than 100 articles. For those familiar with the first edition, the major upgrades have been in Chapter 6 (Microbiology), Chapter 15 (Membrane and Other Technologies to Remove Undesirable Dissolved Substances), Chapter 16 (Disinfection), Chapters 17–19 (A host of New Biotreatment Technologies from Membranes to Hybrid Suspended Growth Fixed Film Processes) and a new chapter on life cycle analysis. This book has been written as a text based on the typical undergraduate background of a civil engineering student. Necessary background information for design and assessment of treatment processes is included in the first sections. These sections do include some review of basic concepts from other courses and should not be considered as replacing the need for these courses. However, these concepts have been focused on the needs of the water quality environmental engineer. Basic theory of most treatment processes is presented in later sections. The variety of treat­ ment processes is increasing and the issues surrounding their application are often subtle. An introductory coverage of most of the processes is given with more emphasis on common applications. Processes are grouped according to theoretical principles rather than their occurrence in water or wastewater treatment operations. It has been the experience of the authors that many books on the subject simply present formulae without any detailed develop­ ment: this book addresses that problem. The reader is made aware of theoretical and empirical developments. This book has a more intense focus on the mechanics of the processes used to treat water or wastewater; design of water distribution and wastewater collection systems are applied hydrau­ lics exercises that are dealt with in other courses. The theory presented in this book is at the level of an undergraduate course. In the past, much of this basic theory was left for the graduate level. As need and room in the undergraduate curriculum for environmental engineering grows, the opportunity now exists to bring undergraduates to a level of theoretical comprehension equivalent to that gained in other disciplines.

PREFACE

05/12/2018

xxiv

2:33:1

Page xxiv

Preface

The material in the book can be organized in numerous ways depending on the backgrounds of students and the breadth and depth of the environmental engineering curriculum. As a note to the reader, the theory presented is the first principles for design and operation of treatment processes. These principles govern the major phenomena occurring in a process but, as with all models, they are idealizations. Field studies are always required to confirm or refine the basic theory results into processes that will meet the objectives. Typical design ranges are given for most processes. There are many design handbooks available that note many practical considerations and experiences for consideration in design and process analysis.

ABBREVIATIONS

07/16/2018

13:13:21

Page xxv

xxv

Abbreviations and Acronyms Used in the Text ABES Acy acu ADP AGI AIDS Alk AO A/O AMP AOB AOTR APHA aq AS ASCE ASME asu ATA ATAD atm ATP AWWA BAF BAT BDST BMP BNR BOD BODu bp Bq Btu cap CAS c/c

Associação Brasileria de Engenharia Sanitária e Ambiental acidity apparent color unit adenosine diphosphate acute gastrointestinal illness acquired immunodeficiency syndrome alkalinity aesthetic objective anaerobic/oxic adenosine monophosphate ammonia oxidizing bacteria actual oxygen transfer rate American Public Health Association aqueous activated sludge American Society of Civil Engineers American Society of Mechanical Engineers areal standard unit anaerobic toxicity assay autothermal aerobic digestion atmosphere adenosine triphosphate American Water Works Association biological activated filter best available technology bed-depth-service time biomethane potential biological nutrient removal biochemical oxygen demand ultimate BOD base pair; boiling point Becquerel British thermal unit capita conventional activated sludge center to center

ABBREVIATIONS

xxvi

07/16/2018

13:13:21

Page xxvi

Abbreviations and Acronyms Used in the Text

CCME CDC cf CFD CFU CH Ci CM CML COD CSCE CSO CSTR CTA CWQI 1-D Da DALY DBP DC DDT deg DF DNA DND/DNB DO DOI DPD DSFF DST EBCT EC EC50 ed. EDTA EIA EMF EPA eq ES EU FA FAO FBD FC ft-c FD FE FISH

Canadian Council of Ministers of the Environment Center for Disease Control and Prevention compare with computational fluid dynamics colony forming unit carbonate hardness curie complete mixed Centrum voor Milieukunde – Universiteit Leiden chemical oxygen demand Canadian Society of Civil Engineers combined sewer overflow completely stirred tank reactor cell transformation assay Canadian Water Quality Index one-dimensional Dalton Disability-Adjusted Life Year disinfection byproduct direct current dichloro-diphenyl-trichloroethane degree dilution factor deoxyribonucleic acid do not drink/do not boil dissolved oxygen digital object identifier N,N-diethyl-p-phenylenediamine downflow stationary fixed film defined substrate technology empty bed contact time Escherichia coli see LC50 edition; editor ethylene diamine tetraacetic acid environmental impact assessment electromotive force Environmental Protection Agency equivalent effective size European Union formic acid Food and Agriculture Organization fine bubble diffuser fecal coliforms foot-candles filtration and disinfection free energy fluorescence in situ hybridization

ABBREVIATIONS

07/16/2018

13:13:21

Page xxvii

Abbreviations and Acronyms Used in the Text

F:M Fr FS FSS ft g GAC gal GAO GC GLUMRB GTP HF HIV h ha HF hp HP HPC HPLC HRT IAWPRC IAWQ i.d. IFAS I/I IJC IMO in. IRC ISS IWA IUPAC KN L l lb lm LC50 LCA LCI LCIA LE LHS LI ln log LUCA

food:microorganism ratio Froude number fecal streptococci; flat sheet; flat screen fixed suspended solids foot (feet) gaseous; gram granular activated carbon gallon (US) glycogen accumulating organisms gas chromatography Great Lakes Upper Mississippi River Board guanosine triphosphate hollow fiber human immunodeficiency viruses hour hectare hollow fiber horsepower hydrogen peroxide heterotrophic plate count high-performance liquid chromatography hydraulic retention time International Association on Water Pollution Research and Control International Association on Water Quality internal diameter integrated fixed film activated sludge inflow and infiltration International Joint Commission International Maritime Organization inch International Reference Centre inert suspended solids International Water Association International Union of Pure and Applied Chemistry Kjeldahl nitrogen length; liter liquid pound lumen lethal concentration for 50% of the organisms life cycle analysis (assessment) life cycle inventory analysis life cycle impact assessment Ludzack–Ettinger left-hand side Langelier index natural (base e) or Naperian logarithm base 10 logarithm last universal common ancestor

xxvii

ABBREVIATIONS

xxviii

07/16/2018

13:13:21

Page xxviii

Abbreviations and Acronyms Used in the Text

m M MAC MAF max MBBR MBR MCL MCLG meq MF MFI Mgal/d mi MIB min ML MLE MLSS MLTSS MLVSS mo mp MPN MSW MUG MW MWCO N NAD NADP NCH ND NDP NDMA NF NOB NOD NOEC NOM NTA NTU o.d. OF OG OI OMOECC ONPG ORP

molal; meter mass; molar maximum acceptable concentration Ministry of Agriculture and Food maximum moving bed bioreactor membrane bioreactor maximum contaminant level maximum contaminant level goal milli-equivalent membrane filter; microfiltration modified fouling index million gallons (US) per day mile methylisoborneol minute; minimum mixed liquor modified Ludzack–Ettinger mixed liquor suspended solids mixed liquor total suspended solids mixed liquor volatile suspended solids month melting point most probable number municipal solid waste methylumbelliferyl-β-d-glucuronide molecular weight molecular weight cutoff Newton; normal nicotinamide adenine dinucleotide nicotinamide adenine dinucleotide phosphate noncarbonate hardness nondetectable net driving pressure N-nitrosodimethylamine nanofiltration nitrite oxidizing bacteria nitrogenous oxygen demand no observed effect concentration natural organic matter nitriloacetic acid nephelometric turbidity unit outside diameter overland flow land treatment operational guideline odor index Ontario Ministry of the Environment and Climate Change o-nitrophenol β-d-galactopyranoside oxidation-reduction potential

ABBREVIATIONS

07/16/2018

13:13:21

Page xxix

Abbreviations and Acronyms Used in the Text

OTE OUR P/A PAA PAC PACT PAO PCB PCP PCR PF PFA PFC PFP PFU PHA PHB PID PolyP POU ppb ppm psi PVC rad RAS RBC Re redox rem rev RHS RI rms RNA RO rpm rps rRNA RSSCT RTC s SAD SAR SAT SBR SCC SDWA SDI

oxygen transfer efficiency oxygen uptake rate presence-absence peracetic acid powdered activated carbon powdered activated carbon activated sludge process phosphate accumulating organisms polychlorinated biphenyl pentachlorophenol polymerase chain reaction plug flow performic acid perfluorinated compound plaque forming particle plaque forming unit polyhydroxyalkanoates poly-β-hydroxybutyric acid proportional–integral–derivative polyphosphate point of use parts per billion parts per million pounds per square inch polyvinylchloride Roentgen absorption dose return activated sludge rotating biological contactor Reynold’s number oxidation–reduction Roentgen-equivalent-man revolution right hand side rapid infiltration land treatment root mean square ribonucleic acid reverse osmosis revolutions per minute rotations per second ribosomal ribonucleic acid rapid small-scale column test real-time control solid; second specific aeration demand sodium adsorption ratio soil aquifer treatment sequencing batch reactor Standards Council of Canada Safe Drinking Water Act salt density index

xxix

ABBREVIATIONS

xxx

07/16/2018

13:13:21

Page xxx

Abbreviations and Acronyms Used in the Text

s.g. SHE SI SOC SOTR spp SR SRT SS SSVI STP Sv SVI SWTR T T&O TBOD TC TCA TCMP TDS TH THM TKN TN TMP TNTC TOC TOD TON TOX TP TS TSS TT TU TVA UASB UCT UF UK UNICEF US USEPA UV VFA VOC VS

specific gravity standard hydrogen electrode; Syrian Hamster Embryo Système International; saturation index synthetic organic chemical standard oxygen transfer rate species slow rate land treatment solids retention time suspended solids stirred sludge volume index standard temperature and pressure Sievert sludge volume index Surface Water Treatment Rule time taste and odor total biochemical oxygen demand total coliforms tricarboxylic acid 2-chloro-6-(trichloro methyl) pyridine total dissolved solids total hardness trihalomethane total Kjeldahl nitrogen total nitrogen transmembrane pressure too numerous to count total organic carbon transferred ozone dose threshold odor number total organic halides total phosphorus total solids total suspended solids treatment technique turbidity unit Tennessee Valley Authority upflow anaerobic sludge blanket University of Cape Town ultrafiltration United Kingdom United Nations International Children’s Emergency Fund United States United States Environmental Protection Agency ultraviolet volatile fatty acid volatile organic chemical volatile solids

ABBREVIATIONS

07/16/2018

13:13:21

Page xxxi

Abbreviations and Acronyms Used in the Text

VSS W WAS WEF WHO wk WPCF WQI YLD YLL yr

volatile suspended solids watt waste activated sludge Water Environment Federation World Health Organization week Water Pollution Control Federation water quality index years lost due to disability years of life lost year

xxxi

ABBREVIATIONS

07/16/2018

13:13:21

Page xxxii

GPS-XTM(Lite) provided with the book is a limited version of the software. It allows students to work with several wastewater treatment plant exercises to learn the principles of wastewater treatment plant modeling. The software is provided under an academic license agreement and its use for commercial purposes is prohibited. To download the software, go to www.hydromantis.com/ GPS-X-Limited/download.html

FABOUT

07/03/2018

13:40:7

Page xxxiii

xxxiii

About the Companion Website This book is accompanied by a companion website: www.wiley.com/go/droste/water The website includes



Solutions Manual

FABOUT

07/03/2018

13:40:7

Page xxxiv

SECTION1

05/09/2018

17:10:43

Page 1

1

Section I Chemistry

SECTION1

05/09/2018

17:10:43

Page 2

3

1 Basic Chemistry This chapter gives a review of chemistry definitions and fundamentals to refresh and supplement background obtained in first courses. Our view is toward the applied chemistry of water and wastewater and their treatment.

1.1 Definitions A fundamental substance that cannot be further decomposed by ordinary chemical means is an element. The smallest unit of an element is an atom. The names, symbols, atomic numbers, and atomic weights of the elements are provided for convenience in a periodic chart and a table located inside the front cover of the book. The symbols are not always the first letter(s) of the element names because they are derived from the Latin, Greek, or German names of the elements. The atomic number is the number of protons in the atom, whereas the atomic weight reflects the number of protons and neutrons contained in the atom. Atoms with the same number of protons but different atomic weights are isotopes of an element. Atomic weights of the elements are referred to the weight of the 12C isotope of carbon, which was assigned a value of exactly 12 by the International Union of Pure and Applied Chemistry (IUPAC).1 The atomic weight also reflects the relative occurrence of isotopes of a given element. Isotopes contain the same number of protons and therefore electrons, but differ in the number of neutrons. The precision of the atomic weight depends upon the natural variation in occurrence of the isotopes. The atomic weights are all constant within three significant digits, which is suitable for engineering calculations. The gram atomic weight refers to the quantity of an element in grams equal to the atomic weight of the element. One gram atomic weight of an element or a compound contains one Avogadro number (6.023 × 1023) of each of the atoms contained in the chemical formula. The gram molecular weight refers to the molecular weight (MW) of a compound in grams.

1 Abbreviations and acronyms used in the text are defined at the end of the preface.

Theory and Practice of Water and Wastewater Treatment, Second Edition. Ronald L. Droste and Ronald L. Gehr.

 2019 John Wiley & Sons Inc. Published 2019 by John Wiley & Sons Inc.

Companion website: www.wiley.com/go/droste/water

4

Theory and Practice of Water and Wastewater Treatment

1.2 The Expression of Concentration Concentration of ions in a solution is expressed on the following bases: molar, molal, mole fraction, or mass concentration. Molar. Molar concentration is the number of moles per volume of solution indicated by M. A 1 M solution contains one mole per liter. Molal. Molal concentration expresses the number of moles per 1000 g of solvent (water). A molal concentration is indicated by m. Density and volume of a solution are influenced by the characteristics and quantities of dissolved substances and the temperature of the solution; thus the volume of 1000 g of water containing dissolved or suspended substances will not exactly occupy 1 L. (Also there was a slight error in the density determination of pure water.) However, these differences are insignificant, and molar and molal concentrations are practically equal for water solutions of interest to environmental engineers. Mole fraction. The equation defining the mole fraction xi of the ith substance is xi ˆ

ni Σnj

(1.1)

where n is the number of moles and Σnj is the total number of moles of all substances, including the solvent, in the solution (for k substances, j = 1 to k). Expression of concentration on a mole fraction basis is most useful for thermodynamic analysis. Mass concentration. Concentration is often expressed in terms of parts per million parts (ppm) or mg L 1. Sometimes parts per thousand (ppt) or parts per billion (ppb) are also used. The concentration of a solute X in a solvent Y in ppm is ppm ˆ

mass of substance ; mass of solution

e:g:

x g of X ˆ x ppm 10 g of …Y ‡ X† 6

Because 1 kg of solution with water as the solvent has a volume of approximately 1 L, 1 ppm  1 mg L

1

Square brackets, [ ], around a species indicate concentration of the species (e.g., [Cl2] indicates concentration of Cl2). The concentration may be expressed in any of the above forms in an equation. The reader is cautioned to take note of the definition of concentration used in each equation. Tables 1.1–1.3 list various substances found in water. Many important elements exist in various combinations with other elements as discussed in the following sections. The concen­ trations of many compounds are often reported in terms of the key element in the compound. This is indicated by writing the chemical formula for the compound hyphen (-) the key element. For instance, the concentration of ammonia (NH3) can be written as [NH3] or [NH3-N]. The former indicates the concentration of ammonia as ammonia and the latter indicates the concentration of ammonia as nitrogen. The “-N” is also written “as N,” that is [NH3 as N]. The transformation between the two concentration expressions is based on the gram molecular weight of the entity and the gram atomic weight of the element. For ammonia, the gram molecular weight is 17 g, and the gram atomic weight of nitrogen is 14 g. Therefore, 1 mg L 1 of NH3 is equivalent to 14/17 or 0.82 mg L 1 of NH3-N, and vice versa. If only the chemical formula is indicated, then the concentration is taken as being of the entity itself. Sometimes the concentration of one species is expressed in terms of another chemical entity based on chemical reactions. Expressing the concentration in terms of one element or entity is

1 Basic Chemistry

convenient when the element is undergoing transformation from one species to another or when many substances have the same chemical effects. Example 1.1 Expression of Concentration A water contains 3.10 mg L 1 of H2S, 1.40 mg L 1 of HS , and 26.50 mg L 1 of SO24 . Express the concentrations of each species as S and as SO24 . Base the SO42 equivalency on the number of sulfur atoms in the entity. The elemental weight2 of sulfur is 32.1, and the gram molecular weight of SO24 is 96.1. The table in this example expresses the concentrations in the various forms. As SO24

As S []

[]

Substance

mg L − 1

M

H2S

3.10

9.09 × 10

5

32.1/34.1

1.40

4.23 × 10

5

26.50

2.76 × 10

4

HS SO24

[]

[]

mg L − 1

Factora)

H2S-S

2.92

96.1/34.1

H2 S-SO24

8.74

32.1/33.1

HS -S

1.36

96.1/33.1

HS -SO24

4.06

32.1/96.1

SO24

96.1/96.1

SO24

Factora)

-S

Total

8.85 13.13

mg L − 1

-SO24

Total

26.50 39.30

a) The factor is the MW of S or SO24 divided by the MW of the substance of interest

As indicated in the table, the total concentrations of all species as S or SO24 are 13.13 and 39.30 mg L 1, respectively.

1.3 Ions and Molecules in Water Refer to Tables 1.1 and 1.2 and other tables throughout this book, which list information on common elements, ions, or radicals and compounds of concern to environmental engineers. Whether an ion or a radical (charged entity) or a compound actually exists is often a topic for advanced chemistry. Overall stability of an ion or a compound depends on a combination of many factors, some of which have been touched on above and others such as steric constraints that have not been mentioned. The polarity and associated hydrogen-bond properties of the water molecule dictate to a large extent its chemical behavior. Water has a higher melting point, boiling point, heat of vaporiza­ tion, and heat of fusion than other comparable hydrides (a hydride is a compound containing H and another element) such as H2S, NH3 and other common liquids. Because of its polarity, water is able to dissolve ionic and polar compounds. These properties are very important for life processes and environmental phenomena. 1.3.1 Oxidation Number The oxidation number of an atom in a compound depends on the number of electrons that are associated with it. The oxidation number for an atom can vary depending on the compound in 2 The periodic table is given at the front of the book; the atomic weights of the elements are given after the index at the end of the book.

5

6

Theory and Practice of Water and Wastewater Treatment

Table 1.1 Basic information on common elements. Element

Symbol

Atomic weight

Common valencea)

Equivalent weightb)

Aluminum

Al

27.0

3+

9.0

+

Arsenic

As

74.9

3

Barium

Ba

137.3

2+

25.0 68.7

+

Boron

B

10.8

3

Bromine

Br

79.9

1

79.9

Cadmium

Cd

112.4

2+

56.2

Calcium

Ca

40.1

2+

Carbon

C

12.0

4 ,4

Chlorine

Cl

35.5

1

Chromium

Cr

Copper

Cu

Fluorine Gold

52.0

Other valence states

3.6

20.0 +

3.0, 3.0 35.5

+

+

+

+

3 ,6

17.3, 8.7

63.5

2 ,1

F

19.0

1

19.0

Au

197.0

3+

65.7

Hydrogen

H

1.0

1+

1.0

Iodine

I

126.9

1

2+

31.8, 63.5 1+

126.9 +

+

Iron

Fe

55.8

2 ,3

Lead

Pb

207.2

2+

103.6

Magnesium

Mg

24.3

2+

12.2

Manganese

Mn

54.9

2+

27.5

4+, 7+

Mercury

Hg

200.6

2+

100.3

1+

Nickel

Ni

58.7

2

+

Nitrogen

N

14.0

3 ,5

Oxygen

O

16.0

2

Phosphorus Potassium

P K

31.0 39.1

5 1

1+, 2+, 3+, 4+

79.0

6

28.1

4+

13.1 6.5

+

Silver

Ag

107.9

1

Sodium

Na

23.0

1+ 2 ,6

Sn

118.7

2

Zinc

Zn

65.4

2+

4

107.9 23.0 +

+

Tin

3 , 3+

39.1

+

Si

a)

6.0

+

Se

32.1

4.7, 2.8 8.0

Silicon

S

4+

29.4 +

+

Selenium

Sulfur

27.9, 18.6

16.0, 5.4

4+

59.4

4+

32.7 3+

Valence is the charge on ionic substances such as Al or Cl or the charge associated with the atom in a covalent grouping such as N(5+) in NO3 or N(3 ) in NH3. The single N5+ or N3 ions do not exist naturally. Also see Section 1.3.1. b) Equivalent weight equals atomic weight divided by valence. The equivalent weights given in this column are based on the most common valence. Depending on the element and the circumstances of the reaction, equivalent weight may be for an acid–base or an oxidation–reduction reaction.

1 Basic Chemistry

Table 1.2 Common ions encountered in water and water analyses. Name

Formula

Molecular weight

Valence

Equivalent weighta)

Acetate

C2 H3 O2

59.0

1

59.0

Ammonium

NH‡4

18.0

1

Borate

BO33

58.8

3

19.6

Bicarbonate

HCO3

61.0

1

61.0

Carbonate

CO23

60.0

2

30.0

Chlorate

ClO3

83.5

1

83.5

Chlorite

ClO2

67.5

1

67.5

Hypochlorite

OCl

51.5

1

51.5

Chromate

CrO24

116.0

2

58.0

Dichromate

Cr2 O27

216.0

2

108.0

Hydroxide

OH

17.0

1

17.0

Nitrate

NO3

62.0

1

62.0

Nitrite

NO2

46.0

1

46.0

Orthophosphate

PO34

95.0

3

31.7

Orthophosphate, monohydrogen

HPO24

96.0

2

48.0

Orthophosphate, dihydrogen

H2 PO4

97.0

1

97.0

Permanganate

MnO4

118.9

1

118.9

Bisulfate

HSO4

97.0

1

97.0

Sulfate

SO24

96.0

2

48.0

Bisulfide

HS

33.1

1

33.1

Sulfite

SO23

80.0

2

40.0

Bisulfite

HSO3

81.0

1

81.0

a)

+

18.0

Equivalent weights are given as in Table 1.1 based on weight of the ion divided by its valence. In redox reactions, the equivalent weight will depend on the element being oxidized or reduced.

which it appears. In compounds, the electrons are associated with the atoms according to their electronegativity; electrons are assigned to the more electronegative element. Electronegativity of atoms generally increases from left to right along the periodic table and from bottom to top. The rules for assigning oxidation numbers are as follows: 1) The algebraic sum of the oxidation numbers of all atoms in a neutral compound is zero; otherwise, the sum must equal the charge on the ionic species. 2) The oxidation number of all elements in the free state is zero. When an atom is bonded to another chemically identical element, this bond makes no contribution to the oxidation state of either atom. 3) Alkali metals (those in the first column of the periodic table) assume a + 1 oxidation number in their compounds; the alkaline earth elements (those in the second column) assume an oxidation number of +2. 4) The oxidation state of oxygen is 2 except in peroxide, in which case it is 1.

7

8

Theory and Practice of Water and Wastewater Treatment

Table 1.3 Half-reactions. Reaction No.

Half-reaction

1

1 2 Br2 …aq† 4+

‡ e ˆ Br

105.2

+ e = Ce

2

Ce

3

2 1 4 CO3

4

2 1 4 CO3

5

1 2 Cl2 …aq†

6 7 8

ΔG° (kJ mol − 1)a)

3+

‡

7 ‡ 8H

‡e ˆ

‡

‡H ‡e ˆ

‡

1 8 CH3 COO

1 24 C6 H12 O6

‡

1 4 H2 O

1 4 H2 O

‡ e ˆ Cl

E° (V)a)

1.09

139.6

1.72

7.24

0.075

1.47

0.0015

134.7

1.396

ClO2 …aq† ‡ e ˆ ClO2

92.05

0.954

ClO3 ‡ 2H‡ ‡ e ˆ ClO2 ‡ H2 O

111.0

1.15

‡

164.8

1.708

‡

134.1

1.39

128.9

1.36

32.81

0.340

‡ H ‡ e ˆ 12 Cl ‡ 12 H2 O

1 2 OCl

‡ H ‡ e ˆ 18 Cl ‡ 12 H2 O

9

1 8 ClO4

10

2 1 6 Cr2 O7

11

2‡ 1 2 Cu

12

‡ e ˆ 12 Fe(s)

13

2‡ 1 2 Fe 3+

Fe

14

3‡ 1 3 Fe

‡ e ˆ 13 Fe(s)

15

H‡ ‡ e ˆ 12 H2 (g)

‡ 73 H‡ ‡ e ˆ 13 Cr3‡ ‡ 76 H2 O

‡ e ˆ 12 Cu…s†

42.55

+ e = Fe2+

74.40 3.57 0.00

+

0.441 0.771 0.0037 0.00

16

1 2H2O2

+ H + e = H2O

170.8

1.77

17

2‡ 1 2 Hg

‡ e ˆ 12 Hg

82.43

0.854

18

1 2 I2 …g†

‡e ˆI

51.91

0.538

19

1 2 MnO2 …c†

118.7

1.230

20

1 5 MnO4

145.9

1.512

21

1 6 NO2

86.56

0.897

22

1 8 NO3

85.01

0.881

23

1 3 NO2

146.6

1.519

24

1 5 NO3

120.1

1.245

25

1 4 O2 …aq†

122.5

1.272

26

1 2 O3 …g†

199.7

2.07

27

2‡ 1 2 Pb

28

2 1 6 SO4

29

2 1 8 SO4

30

2 1 4 SO4

‡ 54 H ‡ e ˆ

31

2 1 2 SO4

‡ H‡ ‡ e ˆ 12 SO23 ‡ 12 H2 O

3.89

0.040

32

2‡ 1 2 Zn

‡ e ˆ 12 Zn…s†

73.62

0.763

a)

‡ 2H‡ ‡ e ˆ 12 Mn2‡ ‡ H2 O

‡ 85 H‡ ‡ e ˆ 15 Mn2‡ ‡ 45 H2 O

‡ 43 H‡ ‡ e ˆ 16 NH‡4 ‡ 13 H2 O ‡

5 ‡ 4H

‡

4 ‡ 3H

‡

6 ‡ 5H

‡e ˆ

‡ 1 8 NH4

‡e ˆ

1 6 N2 …g†

‡e ˆ

1 10 N2 …g†

‡

‡H ‡e ˆ ‡

‡H ‡e ˆ

‡e ˆ ‡

1 ‡ 3H

‡

5 ‡ 4H

‡

3 8 H2 O

‡

2 3 H2 O

‡

3 5 H2 O

1 2 H2 O

1 2 O2 …g†

‡

1 2 H2 O

1 2 Pb…s†

12.16

‡e ˆ

1 6 S…s†

‡e ˆ

1 8 H2 S…aq†

‡

‡

2 1 8 S2 O3

At a temperature of 25 °C and 1 atm pressure.

2 3 H2 O

‡

1 2 H2 O

‡ 58 H2 O

0.126

34.48

0.357

29.27

0.303

27.70

0.287

1 Basic Chemistry

5) Hydrogen has an oxidation number of +1 except when bonded to metallic hydrides such as NaH, in which case it has an oxidation number of 1. From the general rule of assigning electrons to the more electronegative element, N in NCl3 would have an oxidation number of +3, and each Cl atom would have an oxidation number of 1. However, in NH3 because of the general rule expressed through rule 5, N would have an oxidation number of 3, and each H has an oxidation number of +1. There are some other rules that apply in special circumstances that will generally not be of concern in this text. Example 1.2 Oxidation Numbers What are the oxidation numbers of each atom in the following compounds? (a) Na2HPO4 (b) For Na2HPO4, by rule 3 each Na has an oxidation number of +1; by rule 5 each H has an oxidation number of +1; and by rule 4 each O has an oxidation number of 2. The molecule has no net charge; therefore, from rule 1, the oxidation number on P is +5. For the second compound, propene, the oxidation number on each H is +1 from rule 5. By rule 1, the 3 carbon atoms must have oxidation numbers that sum to 6. From rule 2, the carbon– carbon bonds make no contributions to the oxidation numbers. There are 2 Hs associated with the left C atom, 1 with the center C atom, and 3 with the right C atom, and the oxidation numbers of the C atoms are, from left to right, 2, 1, and 3.

1.4 Balancing Reactions Two fundamental rules must be followed when balancing chemical reactions: 1) Mass must be conserved. 2) Charge must be conserved. The first law states that the number of atoms of an element on the left-hand side of the equation must be equal to the number of atoms of the element on the right-hand side of the equation for each substance. This is the law of mass action. The atoms may have changed the atoms with which they are associated, but they cannot be created or destroyed. Electrons do not have any significant mass associated with them, and they may appear on either side of the equation and not on the other. They are given up or taken up by an ion, a radical, or a compound. The second rule states that there cannot be a charge difference between either side of the reaction after adding up all charged species on each side of the reaction. Electrons, of course, do have a charge associated with them, and they must be incorporated into the charge balance. Consider the reaction: aA ‡ bB ‡ ∙ ∙ ∙ ⇆ cC ‡ dD ‡ ∙ ∙ ∙

(1.2)

Arrows going in either direction indicate that the reaction is reversible. All reactions are reversible but, for practical purposes, a significant number of reactions only proceed in one direction or conditions are provided that force the reaction to proceed in one direction. If the reaction is going to the right, then the substances on the left-hand side of the equation are the reactants, and those on the right-hand side are the products.

9

10

Theory and Practice of Water and Wastewater Treatment

The lower case letters a, b, c, and d represent the stoichiometric combination values needed to satisfy the equation in accordance with the rules stated above. Do these numbers represent valence or oxidation numbers? Or are they related to valence or oxidation numbers? The stoichiometric numbers can be algebraically manipulated in any fashion considering the arrows to be equivalent to an equal sign. The following example demonstrates the additivity of chemical reactions. R1 ‡ R2 ⇆ P1 ‡ P2

P1 ‡ R3 ⇆ R2 ‡ P3

R1 ‡ R3 ⇆ P2 ‡ P3

Reactions must be written to describe processes that are actually feasible, that is, the ions or compounds participating in the reaction must be true substances. Example 1.3 Balancing Reactions Balance the following reactions: 1) MnO2 + HCl ⇆ MnCl2 + Cl2 + H2O 2) MnO4 + H2O + e ⇆ MnO2 + OH Both reactions represent true phenomena. For reaction 1, it is observed that at least 4 Cl atoms must appear on the right-hand side. Multiplying HCl by 4 results in 4 H atoms on the left-hand side, which forces us to multiply H2O by 2. A check on the number of atoms of each element on either side of the equation will show that the equation is balanced; thus the total masses on either side of the equation will be equal. No charged species participate in the reaction as written. The final balanced reaction is MnO2 ‡ 4HCl ⇆ MnCl2 ‡ Cl2 ‡ 2H2 O For reaction 2, there is an excess of oxygen and hydrogen on the left-hand side. Fixing the number of moles of MnO4 involved at 1 determines the number of moles of MnO2 to be 1. Trial and error will show that OH must be multiplied by 4 and H2O must be multiplied by 2 to balance the equation with respect to mass. MnO4 ‡ 2H2 O ‡ e ⇆ MnO2 ‡ 4OH Checking the charge balance, it is found that there is an excess of two negative charges on the right-hand side of the equation. The electrons (which do not influence the mass) must be increased to 3 to balance the equation. MnO4 ‡ 2H2 O ‡ 3e ⇆ MnO2 ‡ 4OH Electrons do not explicitly appear on the right-hand side of the equation; they have been incorporated into Mn in MnO2. Check the oxidation numbers on Mn on either side of the equation.

1.5 Oxidation–Reduction Reactions Oxidation–reduction (redox) reactions involve transfer of electrons. A simple redox reaction has been examined in Example 1.3. In a redox reaction, one entity is giving up electrons and another entity is taking them up. The overall reaction can be formulated from two half-reactions, one producing electrons and the other receiving them; electrons will not explicitly appear in the overall balanced reaction. The substance giving up electrons is being oxidized, so it is the reducing agent; the substance receiving electrons is being reduced, so it is the oxidizing agent. A

1 Basic Chemistry

simple mnemonic device to discern which substance is the oxidizing agent is to note that reduction means a reduction in oxidation number; i.e., molecular oxygen is an oxidizing agent that receives electrons and has its oxidation number reduced from 0 to 1 or, more commonly, 2 in typical reactions. Half-reactions are normally written as a reduction, a convention that will be followed in this book. Therefore, when combining two half-reactions, one of them must be reversed. Table 1.3 lists common half-reactions of significance for water and wastewater treatment. For convenience in combining them, these reactions are balanced and normalized to transfer one electron. Simply subtracting one from another will then produce a balanced reaction. Multiplying a half-reaction in this table by any number and combining it with the appropriate sign with any other reaction will produce an overall reaction that is balanced with respect to mass. Will the overall reaction be balanced with respect to charge? As seen from Table 1.3, there is more than one possible product (oxidation state) for some substances (e.g., Mn). The particular species formed in a given reaction depend on the reaction conditions such as [H+] (pH, defined below) and other factors. In many cases, it is a matter of experience to know which product dominates. A variety of oxidation states may exist at any time and change as time proceeds. Normalizing the half-reactions to one electron also conveniently gives the electron equiv­ alence and equivalent weight of the species for a redox reaction. For instance, for reaction 13 in Table 1.3, the equivalence of Fe3+ or Fe2+ is 1 mol and the equivalent weight is 55.85 g. However, if the Fe3+ is reacting according to reaction 14, the electron equivalence is 0.33 mol, and the equivalent weight is 18.6 g. It can be observed from Table 1.3 that many redox reactions are influenced by pH, which is discussed in more detail in Chapter 3. The definition of pH is pH ˆ log H‡ +

(1.3)

From Eq. (1.3), [H ] = 10 . To determine whether a low or high pH is desirable to promote formation of the products, the overall reaction is formulated. If H+ ions appear on the left-hand side, a low pH (high [H+]) will promote the formation of the products; if H+ appears on the right-hand side, a high pH is necessary. Side reactions may also influence the pH, and they must be considered when making the overall assessment of the effect of [H+]. For the case in which a half-reaction is not available in Table 1.3, the core reaction participants are written, and the following steps are performed to balance the reaction. pH

1) Write the core reactants and products on opposite sides of the equal sign. Balance the core reaction with respect to the atom that is changing oxidation state. 2) Balance the reaction with respect to oxygen by adding H2O where needed. 3) Balance H by adding H+ ions. 4) Balance the charge by adjusting the number of electrons. 5) Add any extraneous ions that do not participate in the redox reaction. Example 1.4 Balancing Redox Reactions Chlorine is to be used to oxidize sulfide to sulfate. What is the overall balanced reaction? From Table 1.3, select the appropriate half-reactions. For Cl2, the half-reaction is 1 Cl2 ‡ e ˆ Cl 2

11

12

Theory and Practice of Water and Wastewater Treatment

For sulfide–sulfate, the half-reaction is 1 2 5 1 1 SO ‡ H‡ ‡ e ˆ H2 S ‡ H2 O 8 4 4 8 2 One of these reactions must be reversed upon combining the two half-reactions. It has been specified that Cl2 is the oxidant (Eo values, discussed later, indicate that this will be the case); therefore, this reaction remains as written. 1 1 1 1 5 Cl2 ‡ H2 S ‡ H2 O ˆ SO42 ‡ H‡ ‡ Cl 2 8 2 8 4 Electrons do not appear in the balanced overall reaction. Chlorine has taken up electrons and sulfur has given up electrons; sulfur’s oxidation number has changed from 2 to +6.

Example 1.5 Redox Half-Reaction Write the balanced half-reaction for reduction of formate (HCOO ) to organic matter, which will be represented as CH2O. Step 1

HCOO ‡ e ˆ CH2 O

Carbon is changing oxidation state (what is its change in oxidation number?) in this halfreaction. The stoichiometry remains 1:1 because there is only one carbon atom on either side of the equation. If a carbon compound that contained two carbon atoms were present on the right-hand side of the equation, then it would have to be multiplied by ½. Step 2 HCOO ‡ e ˆ CH2 O ‡ H2 O Step 3 Step 4

HCOO ‡ 3H‡ ‡ e ˆ CH2 O ‡ H2 O HCOO ‡ 3H‡ ‡ 2e ˆ CH2 O ‡ H2 O

Step 5 If sodium formate has been used, the Na+ may be incorporated on each side of the equation. Na‡ ‡ HCOO ‡ 3H‡ ‡ 2e ˆ CH2 O ‡ H2 O ‡ Na‡

1.6 Equilibrium Elementary reactions may be monomolecular, bimolecular, or rarely trimolecular. Complex reactions are composed of elementary reactions. Equilibrium is the state where the concentra­ tions of all species are constant. In fact, equilibrium is a dynamic state where the rate of product formation is exactly equal to the rate of reactant formation. This is a statement of the principle of microscopic reversibility (Laidler 1965). Although some chemical reactions practically go to completion, there is always at least an infinitesimal amount of both reactants and products present. In the chemical reaction [Eq. (1.2)], aA ‡ bB ‡ ∙ ∙ ∙ ⇆ cC ‡ dD ‡ ∙ ∙ ∙ the arrows going in either direction indicate equilibrium.

1 Basic Chemistry

The equilibrium expression for Reaction (1.2) is Kˆ

‰C C Šc ‰ C D Šd ∙ ∙ ∙ ‰C A Ša ‰C B Šb ∙ ∙ ∙

(1.4)

where [CA] is the concentration of substance A, etc. and K is the equilibrium constant. The equilibrium constant corresponding to a chemical equation is always formed by placing the species on the right-hand side of the equation in the numerator, and the species on the left-hand side in the denominator. The equilibrium expression corresponds to the manner in which the equation has been written. Normally, the starting species are placed on the lefthand side and are referred to as reactants. The right-hand species are the products. This does not necessarily mean that the species on the right-hand side will have the highest concen­ trations at equilibrium. In analytical methods or treatment situations, it is normally desirable to have a substance completely or nearly completely transformed to a particular product. The principle of microscopic reversibility allows this to be achieved. From the equilibrium expression, it is seen that increasing or decreasing the concentration of any species changes the ratio of the products at equilibrium. This fact is used to displace the equilibrium concentrations in the desired direction. For instance in Eq. (1.2), if substance A is being transformed to substance C, to promote the transformation of A, substance B will be added to the water. This will decrease the concentration of A at equilibrium. Any species except A on the left-hand side of the reaction can be added to achieve enhanced transformation of A. Alternately, decreasing the concentration of a species on the right-hand side of Eq. (1.2) by chemical reaction or other means is equivalent to increasing the concentration of a species on the lefthand side. The equilibrium relation is used to determine the amount of agent required to achieve the transformation. The time to reach equilibrium varies with the type of reaction (e.g., acid–base or redox), the species involved, and environmental conditions such as the concentrations of all ions and molecules, temperature, and pH. The equilibrium constant is a function of temperature (see Section 2.1.2). The units for expressing concentration in an equilibrium expression vary depending on the phase of the substance. Ions or molecules dissolved in water are expressed as moles per liter. Gases in equilibrium with a solution have their concentrations expressed in atmospheres (atm). The concentration of the solvent, water, is assumed to be 1.0 regardless of whether water is being produced or consumed by the reaction within the water solution. A pure solid (precipitate) or liquid in equilibrium with a water solution is assumed to have a concentration of 1.0. Later sections and chapters discuss some of these special cases. The equilibrium constant can be determined by carefully measuring the concentrations of all species with time. When the concentrations are stable, their values are put into the equilibrium expression to determine the equilibrium constant. Section 2.1 provides the thermodynamic basis for the equilibrium constant.

1.7 Conductivity and Ionic Strength Equilibrium relations depend on concentrations of participants in a reaction. The effective concentration of a substance depends on its concentration, its charge, and the electrical characteristics of the solution. The major influence on the electrical characteristics of the solution is the sum total of all charged species.

13

14

Theory and Practice of Water and Wastewater Treatment

1.7.1 Conductance Concentrations and charges of ions in a solution determine the ability of the solution to carry a current or the conductivity of the solution. Resistance is measured by placing two electrodes in a solution and impressing a voltage drop across the electrodes. In accordance with Ohm’s law, the current that flows depends on the resistance of the solution. E ˆ iR

(1.5)

where E is the voltage drop across the electrodes, i is the current in amperes, and R is the resistance in ohms. Conductance, G, is the reciprocal of resistance and is measured in ohm 1, mho (“ohm” backwards) or the SI unit siemens (S) (after the industrialist Werner von Siemens). Conductance of a solution is measured between two electrodes; it is proportional to the area of the electrodes (usually 1 cm2) and inversely proportional to the distance between them (usually 1 cm), hence Gˆk

A ℓ

(1.6)

where k is the conductivity (or specific conductance) of the solution (1 Ω 1 cm 1 or ℧ cm 1 or mS m 1). Note that 1 mS m 1 = 10 μ℧ cm 1; 1 μS cm 1 = 1 μ℧ cm 1, A is the area of the electrodes (cm2), and ℓ is the distance between electrodes (cm). Conductivity is related to the sum of charge carriers, which are the positive and negative ions in a solution. The equivalent weight of an ion related to its charge-carrying ability is simply its formula weight divided by the charge on the species. Ideally, one equivalent weight of any substance has the same conductance; however, the conductances of all compounds or ions are not the same because of size and charge interaction effects. Conductivity is a rough but useful tool to gauge the concentration of total dissolved solids (TDS) in a sample. Standard Methods for the Examination of Water and Wastewater (APHA et al. 2012) notes that TDS (in mg L 1) may be estimated by multiplying conductivity (in μ℧ cm 1) by a factor ranging from 0.55 to 0.9. 1.7.2 Ionic Strength The ionic strength, I, of a solution is related to the sum total of all charged species in the solution. Iˆ

1 2

C i z2i

(1.7)

where Ci is molar concentration of the ith ion and zi is the charge on the ith ion. The following correlation was determined by Langelier (1936) for a number of waters: I ˆ 2:5  10 5 …TDS†

(1.8)

where TDS is in mg L 1. Another correlation between conductivity (k) and ionic strength was derived by Russell (1976). I ˆ 1:6  10 5 k

(1.9)

where k is in μ℧ cm . The above correlations are useful for making quick estimates of the TDS content of a water. It must be kept in mind that the correlations are only approximate because nonionic species do not contribute to ionic strength and individual ionic species have different weights. 1

1 Basic Chemistry

Also, the effective concentration or activity of a substance with respect to equilibrium calculations is related to the ionic strength of the solution. The effective concentration is determined by multiplying the actual concentration (e.g., in mg L 1, or in M) by an activity coefficient (Section 2.1.1). For most of the calculations in this book, the activity coefficient will be assumed to be 1.0.

1.8 Chemical Kinetics Rate of a reaction is generally dependent on the concentration of one or more species involved in the reaction. Other influences discussed below, such as temperature and catalysis, can accelerate a reaction. When all the intermediate steps of a reaction are known, it is often possible to formulate theoretically the rate model exactly and verify it with experimental data; otherwise, the rate model is simply the empirical model that best correlates the experimental data. A general reaction rate model is dC A ˆ ± kC aA C bB ∙ ∙ ∙ C nN dt

(1.10)

where CA is the concentration of substance A, etc., k is the rate constant, and t is time. The exponents in the above equation may have any value (not necessarily an integer). The sum of the exponents gives the order of the reaction. The reaction is of order a with respect to substance A, order b with respect to B, and so on. The rate constant applies at the experimental conditions. If k is positive, the reaction describes production; a negative k describes removal. When all the intermediate steps are not known or when experimental data prove that inter­ mediates are not important influences on the overall rate of reaction, the following simplified expressions are often used. 1) Zero-Order Formulation dC A ˆ ±k dt

(1.11)

In this case, the rate of change of the concentration of substance A is not significantly influenced by the amounts of any other substances present. 2) First-Order Formulation dC A ˆ ± kC A (1.12) dt This equation is one of the most frequently used formulations in environmental engineering to fit data from complex reactions. In such reactions, CA is often a nonspecific measure of many substances. For instance, organic compounds can be measured by their carbon content. A logarithmic formulation for concentration variation with time indicates that the reaction rate depends on more readily reacted species followed by increasingly difficult to react substances that implicitly partake in the overall reaction. 3) Second-Order Formulation Any of the following expressions describe second-order reactions. dC A ˆ ± kC 2A dt

(1.13a)

15

16

Theory and Practice of Water and Wastewater Treatment

dC A ˆ ± kC aA C bB dt

(1.13b)

dC A ˆ ± kC 2B dt

(1.13c)

In Eq. (1.13b), exponents on CA and CB must sum to 2. When only data on a single substance are used to formulate the rate expression, examination of arithmetic, semilogarithmic, or log–log C–t plots will determine if an nth order expression is appropriate. Zero- and first-order reactions yield straight lines on arithmetic and semi­ logarithmic C–t plots, respectively. All other order reactions yield a straight line on a log– log C–t plot. The general equation for an nth-order reaction based on a single substance, where n is not equal to 0 or 1 is derived from generalizing Eq. (1.13c) to dC ˆ ± kC n dt

(1.14)

and integrating it. C C0

t

dC ˆ ±k Cn

dt 0

where the subscript 0 indicates the initial value. If C0 is not zero, the integrated equation becomes unwieldy; therefore, the following transfor­ mation is used. XˆC

C0

which changes the lower and upper limits on the integral to 0 and C The result of the integration is 1 1

n 1

1

n

X …1

n† C C 0 0

ˆ ± kt jt0

…C

C 0 †1

ˆ ± kt

n

C0, respectively.

(1.15)

or taking logarithms of the above equation log…C

C0† ˆ

1 1

n

log‰ ± …1

n†kŠ ‡ logt

(1.16)

Regression should be performed to find the best estimates of the slope and intercept of the line which determine n and k. 1.8.1 Other Formulations Other formulations have been empirically fit to data. Some of these are given below. Also consult Section 4.10 for development and use of saturation expressions (Monod and Michaelis–Menten equations) which are commonly used in environmental engineering. Consecutive or Series

Reactant A forms product B, which in turn forms a new product C. A common example here is carbonate chemistry, with carbonic acid reacting to form bicarbonate followed by carbonate. Radioactive decay is another case, which is illustrated in Example 1.11.

1 Basic Chemistry

Parallel

We will consider only competitive parallel reactions, wherein a single reactant will yield two different products, B and C, and each reaction has its own rate constant. An example would be an organic waste decomposing by two different pathways. If the paths are independent, dC A ˆ k 1 C A ‡ k 2 C A ˆ …k 1 ‡ k 2 †C A dt C A

ln ˆ …k 1 ‡ k 2 †t C A0

and CB CC

C B0 k 1 ˆ C C0 k 2

Retardant

The rate of reaction of some reactions is observed to increase or decrease with time beyond the change in reaction rate as a result of change in concentration of the reactant. This is known as a retardant reaction. The rate constant is adjusted in the following manner: dC A k ˆ± CA dt 1 ‡ αt

(1.17)

where α is a characteristic reaction constant. Effectively, the reaction rate constant is decreased with time. Autocatalytic

Some reactions accelerate spontaneously as time proceeds. The velocity of the reaction is dependent upon a catalytic effect of the products. Consider the reaction A ‡ P→P ‡ P

(1.18)

where A is the substance being converted and P is the product. For an elementary reaction the kinetic formulation is dC A ˆ k 1CA ‡ k2CACP dt

(1.19)

At any time the total concentration of a reactant and a product, CT is C T ˆ C A ‡ C P ˆ C A0 ‡ C P0 ˆ constant

(1.20)

Substituting for CP in Eq. (1.19) and using Eq. (1.20), dC A ˆ k 1 C A ‡ k 2 C A …C T dt

CA†

(1.21)

If CA = C and CP0 = 0, dC ˆ k 1 C ‡ k 2 C …C T0 dt

C†

There are other empirical models that can be used in any particular situation. The choice of a rate model depends on the model that most generally applies, giving the best statistical fit to the data.

17

18

Theory and Practice of Water and Wastewater Treatment

When the stoichiometry of a reaction is known, the rate of change of one substance may be easily related to the rate of change of another. Consider the general reaction: aA ‡ bB ‡ ∙ ∙ ∙ ⇆ cC ‡ dD ‡ ∙ ∙ ∙

(1.22)

Dividing the equation by a, b c d A ‡ B ‡ ∙∙∙ ⇆ C ‡ D ‡ ∙∙∙ a a a

(1.23)

The equation states that 1 mol of A reacts with b/a moles of B to produce c/a moles of C and d/a moles of D. The change in the number of moles of A, ΔNA is ΔN A ˆ N A

(1.24)

N A0

If ΔNA = 1 then ΔNB = b/a, ΔNC = c/a and ΔND = d/a. The above equations may then be rearranged to ΔN B ΔN C ΔN D ˆ ˆ b c d

ΔN A ˆ a

(1.25)

Differentiating Eq. (1.25) with respect to time yields 1 dN A ˆ a dt

1 dN B 1 dN C 1 dN D ˆ ˆ b dt c dt d dt

(1.26)

At equilibrium or steady-state conditions, the above derivatives are equal to 0. Catalysis

A catalyst is an agent that accelerates a reaction, but it is not changed in composition as a result of the reaction. Dissolved metals can often serve as catalysts. The effect of adding excess hydrogen or hydroxyl ions (i.e., changing the pH) can also be analyzed in some cases as a catalytic reaction. Enzymes are biological molecules that speed reactions many hundreds or even thousands of times. Catalysts do not cause reactions that are thermodynamically impossible to happen; they only bring about the final equilibrium state more rapidly. Example 1.6 Change in Concentration in a Reaction A decay reaction has been found to proceed according to a retardant model with α = 0.052 h 1 and rate constant k = 0.095 h 1. What is the percentage decrease in concentration of the reactant after 5 hours? dC ˆ dt

k C 1 ‡ αt

C C0

ln

dC ˆ k C

C ˆ C0

t 0

dt 1 ‡ αt

k ln…1 ‡ αt † α

ln C

C C0

ˆ

C ˆ …1 ‡ αt† C0

Substituting for k, α and t: C ˆ 1 ‡ 0:052 h C0

1

…5 h†

0:095 h 0:052 h

1 1

k ln…1 ‡ αt† α

ˆ 0:656 or 65:6%

The decrease in concentration is 34.4% after 5 hours.

k α

t 0

1 Basic Chemistry

1.8.2 The Effect of Temperature on Rate of Reaction An increase in temperature raises the energy level of the molecules and also produces a slight increase in the rate of collisions of molecules. Because of the elevated energy levels of the molecules, the number of collisions in which the energy threshold for the reaction is exceeded is increased. From empirical observation, Arrhenius found that the rate constant for a reaction depends on temperature as follows: ln k ˆ a

b T

where a and b are constants and T is temperature in °K. This equation can be related to the Boltzman equation with the result: k ˆ Aexp

Ea RT

(1.27)

where A is a constant related to the frequency of collisions, Ea is the activation energy for reaction, and R is the universal gas constant. Equation (1.27) can be simplified to k T 2 ˆ k T 1 θ…T 2

T 1†

(1.28)

where kTi is the rate constant at temperature Ti (temperature can be expressed in °C or °K) and θ is a constant. Note that θ in Eq. (1.28) is not truly a constant, but over small temperature ranges (ideally E ⚬B

If the concentration of each species is unity, then the species with the higher E° will be the oxidant (substance A for the above two reactions). 2.2.1 Cell or Couple Potential In the overall reaction, the half-reaction with the higher potential will go as written, the other half-reaction will be reversed. The Nernst equations for the above two half-reactions are E A ˆ E ⚬A

0:0591log

‰A Š ‰A Š

E B ˆ E ⚬B

0:0591log

‰B Š ‰BŠ

If EA > EB, the cell or couple potential (Ec) is the difference between EA and EB. It is the driving force for the redox reaction. However, it does not provide any information on the rate of reaction, only on the direction in which the reaction will naturally proceed. For the A and B couples, the

2 The Thermodynamic Basis for Equilibrium

overall reaction is A ‡e →A B

E ⚬A E ⚬B

e →B

A ‡ B →A ‡ B Ec ˆ EA

E ⚬A

E B ˆ E A⚬

E ⚬B

E ⚬B

…0:0591†log

‰A Š‰BŠ ‰AŠ‰B Š

If the number of electrons involved in a half-reaction is not equal to 1, this will influence the stoichiometry of the overall reaction, but not the potential associated with a half-reaction. Consider the reaction C ‡ 4e → 2D2 ‡ X E ⚬CD The Nernst equation for this half-reaction is 2

D2 ‰XŠ 0:0591 log 4 ‰ CŠ

E CD ˆ E ⚬CD

This equation would not change if the reaction were written as 100C ‡ 400e → 200D2 ‡ 100X

E ⚬CD

The Nernst equation for this reaction is 200

D2 ‰XŠ100 0:0591 ⚬ log ˆ E CD 100 400 ‰CŠ

E CD ˆ E ⚬CD

2

D2 ‰XŠ 0:0591 log 4 ‰ CŠ

Combining the C–D2 half-reaction with the A–A half-reaction (assuming EA > ECD), 4A ‡ 2D2 ‡ X → 4A ‡ C Ec ˆ EA

E CD ˆ E ⚬A

E ⚬CD

0:0591 ‰A Š‰CŠ log 2 4 ‰AŠ4 D2 ‰XŠ

from which it can be shown that the individual Nernst equations that make up Ecell have not changed in the overall equation. The cathode (electrode where reduction occurs) will have a higher potential than the anode (where oxidation occurs). The cell potential for a reaction to proceed as written will be positive. Therefore, E c ˆ E cathode

E anode

(2.18)

Note that in the case of an electrochemical cell, Ec refers to the potential difference between the couples that exist in separate well-defined cells. In general, as different waters mix or substances are added to a water, half-reaction couples are set up, and there will be potential differences between the couples. The potential difference between couples is Ec regardless of whether these couples occur in separate cells or not. An electrochemical cell allows the potential difference to be measured accurately. This information is useful for defining potential differences in an actual solution when concentrations are known and for determining in which direction the reaction will proceed.

47

48

Theory and Practice of Water and Wastewater Treatment

2.2.2 Oxidation–Reduction Potential and System Potential When equilibrium exists, Ec = 0. In the ideal case, the potentials of all possible half-reactions are equal. This potential is known as the system potential (Esys). E sys ˆ E 1 ˆ E 2 ˆ ∙ ∙ ∙ ˆ E i ˆ ∙ ∙ ∙ ˆ E n

…E c ˆ 0 for all possible couples†

(2.19)

Note: To use Eq. (2.19) with the Nernst equation, each reaction must be written as a reduction, which is consistent with Ec = 0. When equilibrium does not exist, the system potential is not well defined, but it will be somewhere in between the potentials of the couples that exist in the system. The oxidation–reduction potential (ORP) is the system potential of a solution. It is measured by an inert electrode that transmits potential. The electrode is usually platinum. As noted before, platinum is a good choice for the electrode because it does not significantly influence the potential that it transmits and it is sensitive to many species. To complete the external circuit, a reference electrode is wired to a voltmeter and inserted into the solution. Reference electrodes maintain a constant potential and do not significantly participate in reactions of interest in the solution. Their only purpose is to make electrical contact with the solution. The choice and makeup of a reference electrode vary depending on the situation in which it will be used. Effectively, the reference electrode can be considered to be one half-cell of an electrochemical cell. The other half-cell of the electrochemical cell is the solution with the other electrode immersed in it. The half-cells are separated and make electrical contact by a membrane or other equivalent means in the reference electrode. For a solution at equilibrium with its potential measured by an ideal electrode, Eq. (2.19) will apply. The platinum electrode will transmit Esys. Defining the reference electrode potential as Eref and considering it to be the anode, the potential reading on the voltmeter will be E meter ˆ E c ˆ E sys

E ref

(2.20)

The ORP is usually reported with respect to the hydrogen scale, i.e., the result is normalized with respect to using the SHE as the reference electrode. Because natural waters are unlikely to be in a state of complete equilibrium, the ORP is a gross indicator of the state of oxidation–reduction in the system. Besides this, the potential reading is influenced by how well the electrode “sees” species in solution among other factors. Never­ theless, this measurement is still a useful tool in many circumstances. The general form of the Nernst equation for the ORP for a half-reaction is ORP ˆ E sys ˆ E ⚬

0:0591 ‰reduced speciesŠ log n ‰oxidized speciesŠ

(2.21)

The charge balance always sums to neutrality, but it is seen from Eq. (2.21) that as the concentrations of species able to donate electrons (reduced species) increase, the ORP decreases and vice versa. Thus as oxygen or other oxidants are added to a solution, the ORP rises. Water with a pH of 7.00 and temperature of 25 °C in equilibrium with oxygen at a partial pressure of 0.21 atm has an ORP of 800 mV on the hydrogen scale. At high ORP, there will be a predominance of ions such as SO24 , NO3 , and Fe3+ as opposed to their reduced forms of S2 , NH‡4 , and Fe2+, respectively. As free (dissolved) oxygen is removed from water by chemical or biological reactions, the state of the water changes from an aerobic to an anoxic state, wherein the concentrations of oxidized species are predominant compared to their reduced counterparts, but there is no significant dissolved oxygen concentration. Many microorganisms are able to use SO24 , NO3 , or other oxidized species as electron acceptors. As the reaction progresses and

2 The Thermodynamic Basis for Equilibrium

oxidized species are consumed and the reduced species predominate, the water reaches an anaerobic state. In anaerobic reactors, ORPs in the range of 300 to 400 mV exist. Example 2.6 Use of the Nernst Equation A platinum electrode and reference electrode were used to measure the ORP of a water containing NH‡4 and NO3 as the only possible species involved in redox reactions. The meter reading was 0.01 V. The reference electrode was known to have a potential of 0.25 V (relative to the SHE), and it can be considered to be the anode. The pH of the water was 8.21. If the water was in equilibrium, what is the ratio of [NH‡4 ] to [NO3 ]? Was there any free dissolved oxygen in the water? The platinum electrode senses the potential of the solution. The potential reading on the meter is Ec. From Eq. (2.20), E sys ˆ E meter ‡ E ref ˆ 0:01 V ‡ 0:25 V ˆ 0:26 V From Table 1.3, the half-reaction for NH‡4 –NO3 is 1 5 1 3 NO3 ‡ H‡ ‡ e ˆ NH‡4 ‡ H2 O 8 4 8 8

E° ˆ 0:880

and the corresponding Nernst equation set equal to Esys is E sys ˆ 0:26 ˆ 0:880

NH‡4 0:0591 log 8 NO3 ‰H‡ Š10

The ratio of ammonium to nitrate is log

NH‡4 ˆ NO3

8…0:26 0:88† 0:0591

10pH ˆ 84:1

82:1 ˆ 2:0

The ratio is found by taking the antilog of both sides: NH‡4 ˆ 100 NO3 To find the concentration of dissolved oxygen that would be in equilibrium at this potential, use the half-reaction for dissolved oxygen: 1 1 O2 ‡ H‡ ‡ e ˆ H2 O E° ˆ 1:27 4 2 0:26 ˆ 1:27

0:0591 1 log 4 ‰O2 Š‰H‡ Š4

Solving the above equation for [O2], the equilibrium [O2] is found to be 10 Anaerobic conditions exist.

35.6

or effectively 0.

2.3 Corrosion The principles of electrochemistry are the underlying reasons for corrosion. When a metal has a potential that is anodic with respect to its environment, it will tend to corrode or dissolve. Corrosion is a very complex phenomenon as will be seen from this brief discussion. Many factors

49

50

Theory and Practice of Water and Wastewater Treatment

Figure 2.3 Galvanic corrosion.

mediate to accelerate or retard the rate of corrosion and the final determination of remedial measures may be largely a matter of trial and error. Two dissimilar metals in contact with each other and exposed to a conductive solution form an electrochemical cell. The metals themselves form the external circuit and the ions in solution form an internal circuit. A potential difference will exist between the metals promoting the flow of electrons. The less resistant metal (M1) becomes the anode and corrodes. The other metal (M2) becomes the cathode. M1 → Mn‡ 1 ‡ ne Mn‡ 2 ‡ ne → M2 An example of this type of situation is shown in Figure 2.3. It is not necessary to have Cu2+ ions in solution; any electron acceptor is suitable. This type of corrosion is called galvanic corrosion. The potential difference between the two metals accelerates the removal of electrons from the more active metal. Metals have been organized by their electrochemical activity. Table 2.2 has standard EMF values (E° values) for various metals. The values in Table 2.2 are not necessarily equal to those in Table 1.3 because the EMF values are more reflective of the effective potential of a metal. The more active the metal (the lower its position in the table), the more likely it is to be oxidized (lose electrons) or corrode. The noble elements at the top of the table tend to be cathodic with respect to the surrounding environment, and thus not readily oxidizing. From examination of the Nernst equation, it is seen that a lower ORP promotes the dissolution of metals; i.e., the ionized metal concentration increases. In general for corrosion to occur, dissimilar metals are not required. Direct attack by oxidizing agents will cause corrosion. Furthermore, differences in the composition of a metal or imper­ fections in the metal may cause potential differences within it. Likewise, differences in the con­ centrations of species in solution at two different locations may set up a potential difference. Oxygen is present in most waters, and it causes corrosion because of its oxidizing power. The complexity of the corrosion process is shown by the following equations that illustrate how oxygen can be involved in both the cathodic and anodic reactions for the corrosion of iron. 4Fe ‡ 2O2 ‡ 8H2 O → 4Fe…OH†3 ‡ 4H‡ ‡ 4e

…anodic reaction†

‡

O2 ‡ 4H ‡ 4e → 2H2 O …cathodic reaction† Another route of corrosion is through high hydrogen ion concentrations that may cause direct acid attack as illustrated with the following reaction. Fe ‡ 2H‡ → Fe2‡ ‡ H2 The H2 may react with O2 to form water. Pipes are subject to corrosion from both the inside and the outside. On the outside, external direct current (DC) could enter and travel along a pipe (the cathodic area) and leave the pipe at another area (the anodic area). Sources of DC include electrified transit systems, cathodic protection systems for other structures, DC welding equipment, and crane operations. This is known as electrolytic corrosion, that is similar to galvanic corrosion.

2 The Thermodynamic Basis for Equilibrium

Soil characteristics influence the rate of external corrosion in a number of ways. Electro­ chemical differences in the soil along a pipe can create a galvanic cell. The resistance of the soil affects the current-carrying ability of the soil. The soil can also influence the buildup or removal of corrosion products on the structure being corroded. These products will produce an insulating barrier against further corrosion of the structure, which is one example of polarization phenomena. Polarization is the reduction in potential difference from the ideal potential difference. This lowers the rate of corrosion. O’Day (1989) has discussed other factors causing external corrosion of pipes. Grounding of electrical systems to household water supply plumbing systems may increase corrosion and, therefore, should not be done. Not only would the stray-current-induced corrosion increase the system deterioration but lead and other toxic metals in solder and pipes would also appear at elevated levels in the drinking water (Lee et al. 1989). 2.3.1 Microbial Corrosion Many bacteria, algae, and fungi may cause corrosion directly or indirectly. At locations of bacteria colonies, pH and dissolved oxygen concentrations are different from the bulk phase concentrations and concentrations at other locations on a structure or pipe. Anaerobic colonies also cause differences in O2 concentration from the bulk phase because they manufacture a protective biofilm that shields them from oxygen. Thus concentration cells are set up that cause electron flow and corrosion. Bacteria that derive their energy from the oxidation of inorganic compounds are primary agents of corrosion. Sulfate-reducing bacteria aid in the corrosion of sewers. These anaerobes derive energy from the following reaction: SO24 ‡ 8Hads → S2 ‡ 4H2 O where Hads is hydrogen adsorbed to the metal surface. The removal of elemental hydrogen that has adsorbed to a metal surface accelerates the corrosion of the surface. The adsorbed hydrogen retards the transfer of electrons to an oxidizing agent by acting as an insulator. Furthermore, in iron pipes, the sulfide produced in this reaction reacts with Fe2+ to produce FeS. FeS is more efficient than iron for promotion of the reaction: 2H‡ ‡ 2e → H2 This further accelerates corrosion. In sewers, the evolution of hydrogen sulfide gas produced by sulfate-reducing bacteria is taken to advantage by aerobic sulfide-oxidizing bacteria adhering to the crown of the pipes. The reaction is H2 S ‡ 2O2 → SO42 ‡ 2H‡ → H2 SO4 H2SO4, being a much stronger acid than H2S, enhances acid attack. These bacteria can also oxidize elemental sulfur. Other bacteria produce strong organic acids as byproducts. Another group of bacteria that is significant in pipe corrosion is the iron-oxidizing bacteria. The energy-producing reaction for these bacteria is 2Fe ‡ 1½O2 → Fe2 O3 The colonies form iron crust deposits known as “tubercles.” Corrosion is accelerated at the location of these tubercles.

51

52

Theory and Practice of Water and Wastewater Treatment

2.3.2 Corrosion Prevention from External Environmental Factors There are a number of measures that can be taken to reduce the rate of corrosion. Uniform environmental conditions. Uniform environmental conditions create uniform poten­ tials, minimizing the existence of concentration cells. Therefore, mixing a solution is desirable if possible. Pipe laid underground or a structure below ground is exposed to a soil solution that has a potential relative to the structure. Uniform backfill will spread the attack over a larger surface area instead of concentrating it, which would have resulted in an earlier failure. Cathodic protection. This method consists of rendering the structure to be protected as a cathode in relation to its surrounding environment. The potential of the structure is raised such that it becomes less likely to donate electrons. This occurs at the expense of either electrical or chemical energy. Another substance will be sacrificed. There are two methods of cathodic protection: galvanic and electrolytic, which are discussed next. Galvanic Cathodic Protection

In this method a sacrificial anode, made of a metal lower in standard potential than the metal to be protected, is connected to the structure (Figure 2.4). The anode is consumed and must be periodically replaced. The sacrificial anodes are usually made of Zn or Mg. The lifetime of a sacrificial anode depends on the current that flows from it. For a current, i, flowing through an electrochemical cell, the mass deposited in the cathodic cell or mass consumed in the anodic cell is proportional to time, current and equivalent weight of the metal. In equation form: m ∝ itM eq where m is the mass deposited or consumed, i is the current in C s 1, t is time, and Meq is the mass per equivalent (i.e., equivalent weight of the entity). The proportionality constant is the inverse of the Faraday constant, F (96 500 C/eq). itM eq (2.22) mˆ F Solving Eq. (2.22) for time and incorporating an efficiency factor, E, to account for extraneous losses, EFm tˆ (2.23) iM eq The efficiency factors are approximately 0.90 and 0.50 for Zn and Mg, respectively (Bosich 1970).

Figure 2.4 Galvanic cathodic protection.

2 The Thermodynamic Basis for Equilibrium

Table 2.3 Soil corrosivity rating based on resistivity. Soil Resistivity (Ω-cm)

Corrosivity

0–2 000

Very corrosive

2 000–5 000

Corrosive

5 000–10 000

Moderately corrosive

10 000–30 000

Mildly corrosive

>30 000

Unlikely

Source: Adapted from Hubbell Inc. (2003) and O’Day (1989).

Figure 2.5 Electrolytic cathodic protection.

Studies have shown that current in the range of 10–215 mA m 2 of pipe surface is needed to protect a pipe. For normal conditions, a range of 10–30 mA m 2 is typical (Bosich 1970). The amount of current that can flow from the anode depends on the soil resistance, which can be measured; the resistivity depends on moisture content and concentration of ionic salts. Qualitative estimates of soil corrosivity as a function of resistance are given in Table 2.3. Sea water has a typical resistivity of 200, clay 4000 and sand 200 000 (Nasserdine et al. 2013). There are also special electrodes to measure soil potential. Electrolytic (or Impressed Current) Cathodic Protection

Electrolytic cathodic protection uses a rectifier, which is essentially an electron pump, to bring electrons to the structure (Figure 2.5). The anode may or may not be reactive. Nonreactive graphite (carbon), which is an excellent conductor, is commonly used for the anode. In this case, ions surrounding the anode will be oxidized. The driving force for the reaction is an external power supply.

Questions and Problems 2.1 What is the release or gain of free energy at standard conditions for each mole of water that is evaporated?

53

54

Theory and Practice of Water and Wastewater Treatment

2.2 Using standard free energies, what is the equilibrium constant for the dissociation of acetic acid (CH3COOH ⇆ H+ + CH3COO ) at standard conditions? 2.3 Using the half-reactions in Table 1.3, what are the equations of the overall balanced reaction, the standard free energy change associated with the reaction, and the equili­ brium constant for the oxidation of ferrous iron to ferric iron with chlorine? 2.4 Using information in Tables 1.3 and 2.1 determine the standard free energy of formation for the perchlorate ion (ClO4 ). 2.5 What is the ratio of the equilibrium constant (Ksp) for the formation of calcite (CaCO3) at temperatures of 20 and 80 °C? 2.6 Explain the function of a semipermeable membrane in an electrochemical cell. 2.7 What is E° for the reaction ZnS ⇆ Zn2+ + S2 ? 2.8 Considering only the half-reaction involving MnO4 and Mn2+ given below, at what system potential will the concentration of Mn2+ ions begin to exceed the concentration of MnO4 ions if (a) the pH is 5.0 and the temperature is 25 °C and (b) the pH is 4.5 and the temperature is 5 °C? MnO4 ‡ 8H‡ ‡ 5e ˆ Mn2‡ ‡ 4H2 O E ⚬ ˆ 1:512 V 2.9

a The following substances are added to 1 L of water: 1.0 g of solid copper, 1.0 g of solid zinc, 100 mg of ZnSO4, and 100 mg of CuSO4. The temperature is 25 °C. Assuming that only zinc and copper will react and that all dissolved copper is in Cu(II) form, what are the concentrations of dissolved zinc and copper at equilibrium? What is Esys at equilibrium? Use Table 1.3. b What is the initial potential difference between the couples?

2.10 Determine the equilibrium constant for the formation of Mg(OH)2(s) from standard free energy changes at 25 °C. 2.11 Would it be possible to measure ORP of a sample with the ORP electrode immersed in the sample and the reference electrode (a) placed in the air or (b) placed in a separate beaker containing distilled water? Explain your answers. 2.12 In Example 2.6, what is the ORP with respect to the hydrogen scale? 2.13 What is the potential (ORP) at which the concentration of NH‡4 is equal to the con­ centration of NO3 ions? The pH of the solution is 6.58 and its temperature is 25 °C. What would be the reading on a voltmeter for water at these conditions if the reference electrode had a potential of 0.24 V and it was the anode? 2.14 (a) Explain why water containing a higher concentration of dissolved ions promotes corrosion reactions more readily than water with a low concentration of dissolved ions. (b) Does the same reasoning apply to water that contains a high concentration of glucose, which is a nonionizing compound?

2 The Thermodynamic Basis for Equilibrium

2.15 Based on potential considerations, which metal would be more likely to corrode if tin were in contact with steel? 2.16 List and discuss all the factors that would influence the corrosion of a water distribution pipe laid underground. 2.17 How do microorganisms influence corrosion? 2.18 How does a uniform backfill retard the rate of corrosion? 2.19 Does galvanic or electrolytic cathodic protection protect the inside, the outside, or both surfaces of a pipe against corrosion? 2.20 What is the cost break-even ratio in the price per kilogram of Mg and Zn, respectively, for the same current flow and design lifetime of a sacrificial anode to be used in a galvanic cathodic protection system? 2.21 What is the annual rate of consumption of a sacrificial zinc anode used to protect a 100 m length of pipe with an o.d. of 315 mm in a soil where the average current is 17 mA m 2?

References Bosich, J.F. (1970). Corrosion Prevention for Practicing Engineers. New York: Barnes & Noble. Hubbell Inc. (2003). Step 7 – Corrosion Guide. http://www.vickars.com/screwpile_manual/ PDF_Files/Step7-Corrosion%20Guide.pdf (accessed February 2017). Lee, R.G., Becker, W.C., and Collins, D.W. (1989). Lead at the tap: sources and control. J. Am. Water Works Assn. 81 (7): 52–62. http://www.jstor.org/stable/41292764. Nasserdine, M., Rick, J., and Nasserdine, G. (2013). Soil resistivity data computations; single and two-layer soil resistivity structure and its implication on earthing design. World Acad. Sci., Eng. Technol. 7 (1): 35–40. O’Day, D.K. (1989). External corrosion in distribution systems. J. Am. Water Works Assn. 81 (10): 45–52. http://www.jstor.org/stable/41292477. Speight, J. (ed.) (2005). Lange’s Handbook of Chemistry, 16e. New York: McGraw-Hill.

55

57

3 Acid–Base Chemistry The concentration of free hydrogen ions is one of the major controlling factors of the state of a water because H+ ions are directly or indirectly involved in a large number of reactions. In most situations, the concentration of H+ or OH is small or insignificant compared to the concen­ trations of other species, but this does not mitigate the influence of these ions as controlling variables of the state of the water. The underlying influence of these ions is due to the concentration of water itself, which is the master variable.

3.1 pH Some water molecules will dissociate according to the following reaction1: H2 O ⇆ H‡ ‡ OH

(3.1)

The expression for the equilibrium constant for this reaction is K eq ˆ K ´w ˆ

‰H‡ Š‰OH Š ‰H2 OŠ

(3.2)

However, the concentration of water itself does not change significantly because of this reaction or by the participation of H+ and OH in other reactions in usual circumstances. As pointed out in Section 1.6, the concentration of water is taken to be unity and thermodynamic equations such as equilibrium expressions incorporate this definition. Therefore, the equili­ brium expression for the dissociation of water is K w ˆ H‡ ‰OH Š

(3.3)

The values of Kw at different temperatures are given in Table 3.1. The value at 25 °C will be used in many examples for convenience. The magnitude of [H+] can vary over a wide range, and a logarithmic scale is most convenient for describing its concentration. The “p” notation defined as pX ˆ log10 X

p ˆ log10

(3.4)

is used. The p is a mathematical operator that can be applied to any coefficient or expression.

1 A more accurate description of this dissociation is 2H2O ⇆ H3O+ + OH . The H+ ion does not exist alone; the proton is hydrated. Theory and Practice of Water and Wastewater Treatment, Second Edition. Ronald L. Droste and Ronald L. Gehr.

 2019 John Wiley & Sons Inc. Published 2019 by John Wiley & Sons Inc.

Companion website: www.wiley.com/go/droste/water

58

Theory and Practice of Water and Wastewater Treatment

Table 3.1 Kw at different temperatures. Temperature (°C)

Kw

0

1.154 × 10

15

10

2.964 × 10

15

20

6.871 × 10

15

25

1.012 × 10

14

30

1.459 × 10

14

50

5.309 × 10

14

Source: After Marshall and Franck (1981) and Speight (2005).

At 25 °C, pH ‡ pOH ˆ 14

(3.5)

An equation describing the variation of the equilibrium constant for water with temperature is (Hanson and Cleasby 1990): pK w ˆ 6:0875 ‡ 4470:99=T ‡ 0:017 06T where T is in °K.

3.2 Acids and Bases Acids and bases are substances that influence the pH of a solution. One of the earlier definitions of an acid is the Arrhenius theory of ionization. Acid :

HA ⇆ H‡ ‡ A

Base :

BOH ⇆ B‡ ‡ OH

Salt :

HA ‡ BOH ⇆ BA ‡ H2 O

An acid is a substance that ionizes producing H+ and an anion; a base produces OH and a cation upon ionization. Salts are produced by the reaction of an acid and a base; water is necessarily a byproduct. This concept is limited to ionic solutions only, but it will suffice for many applications in this book. A more comprehensive definition is the Brönsted and Lowry concept, developed in 1923. An acid is a proton donor. A base is a proton acceptor. This concept extends to substances that do not explicitly contain hydrogen ions. The dissociation (equilibrium) constants of a number of acids and bases that are important in environmental engineering are given in Table 3.2. Strong acids (e.g., HCl) dissociate completely. There are essentially no HCl entities in solution, only the component ions exist. The same is true for strong bases such as NaOH. Only when the solution has an extremely low pH in the case of an acid, or an extremely high pH in the case of a base, would the assumption of complete dissociation be negated when the strongly dissociating substance is added to solution. Strong acids have pK values less than 3.

3 Acid–Base Chemistry

Table 3.2 Dissociation constants at 25 °C. Substance

Equilibrium equation

K

Acetic acid

CH3COOH ⇆ H+ + CH3COO

1.754 × 10

Ammonia

NH3 + H2O ⇆ NH‡4 + OH

1.8 × 10

+

H3BO3 ⇆ H + H2 BO3

Boric acid

+

5

5

pK

Significance

4.76

Organic wastes, anaerobic digestion

4.74

Disinfection

5.808 × 10

10

9.24

Nitrogen analysis

Butyric acid

C3H7COOH ⇆ H + C3H7COO

1.524 × 10

5

4.817

Anaerobic digestion

Carbonic acid

H2 CO∗3 ⇆ H+ + HCO3

4.446 × 10

7

6.35

Numerous applications

4.688 × 10

11

10.32

—

 6.2

Analyses

+

HCO3 ⇆ H +

CO23

Hydrochloric acid

HCl ⇆ H+ + Cl

Strong

Hydrocyanic acid

HCN ⇆ H+ + CN

3.98 × 10

10

9.40

Toxicity

+

5.60 × 10

4

3.25

Fluoridation

6.97

Odor, corrosion, anaerobic digestion,

Hydrofluoric acid Hydrosulfuric acid (hydrogen sulfide) Hypochlorous acid

HF ⇆ H + F +

H2S ⇆ H + HS

1.072 × 1 0

HS ⇆ H+ + S2

1.26 × 10

13

12.90

Toxicity

3.31 × 10

8

7.48

Disinfection

4

3.34

Nitrification

+

HOCl ⇆ H + OCl +

Nitric acid

HNO3 ⇆ H + NO3

23.44

Nitrous acid

HNO2 ⇆ H+ + NO2

4.60 × 10

+

7

1.37

Nitrification, analyses

Perchloric acid

HClO4 ⇆ H + ClO4

Strong

 1.6

Analyses

Phenol

C6H5OH ⇆ H+ + C6H5O

1.023 × 10

10

9.99

Tastes, odors

+

7.112 × 10

3

2.148

Buffer, nutrient

6.338 × 10

8

7.198

—

4.786 × 10

13

12.32

—

Phosphoric acid

H3PO4 ⇆ H + H2 PO4 +

H2 PO4 ⇆ H +

HPO24

HPO24 ⇆ H+ + PO34 Potassium hydroxide

+

KOH ⇆ K + OH

Strong (base)

+

Analyses

Propionic acid

C2H5COOH ⇆ H + C2H5COO

1.259 × 10

Sodium hydroxide

NaOH ⇆ Na+ + OH

Strong (base)

—

Analyses, neutralization

Sulfuric acid

H2SO4 ⇆ H+ + HSO4

Strong

 3.85

Coagulation, pH

HSO4 ⇆ H+ + SO24

1.023 × 10

2

1.99

control, analyses

1.288 × 10

2

1.89

Dechlorination

6.237 × 10

8

7.20

—

Sulfurous acid

+

H2SO3 ⇆ H + HSO3 +

HSO3 ⇆ H +

SO23

5

4.874

Anaerobic digestion

Weak acids and bases are only partly dissociated in solution. The equilibrium expression provides the ratio of dissociated ions to undissociated compound. The H+ (or OH ) concentra­ tion is the governing factor in dictating the ratio of the salt ion concentration (the salt ion is the ion or molecule associated with H+ or OH ) to the undissociated compound concentration, if equilibrium exists. Consider acetic acid, which is a weak acid: CH3 COOH ⇆ CH3 COO ‡ H‡

K ˆ 1:8  10

5

…pK ˆ 4:74†

(3.6)

The equilibrium expression is Kˆ

‰CH3 COO Š‰H‡ Š ‰CH3 COOHŠ

or

K ‰CH3 COO Š ˆ ‡ ‰CH3 COOHŠ ‰H Š

(3.7)

59

60

Theory and Practice of Water and Wastewater Treatment

At pH = 4.74, there will be equal amounts (or concentrations) of CH3COOH and CH3COO . The amount of CH3COO in solution relative to the amount of CH3COOH becomes very large as [H+] decreases or vice versa. The tendency for acetic acid to give up a hydrogen ion is much greater as the concentration of hydrogen ions is decreased. For a base, the situation is reversed. A salt is formed by the reaction of an acid and a base according to the following reaction: HA ‡ BOH ⇆ H2 O ‡ BA → H2 O ‡ B‡ ‡ A When a salt is added to water, it ionizes completely and the product ions hydrolyze to the extent predicted by the equilibrium expressions associated with the ions. Some salt may precipitate depending on the solubility product discussed in Section 1.11. Example 3.1 Acid − Base Equilibrium An example problem using equilibrium expressions from Table 3.2 is as follows. How much sodium propionate, C2H5COONa (NaPr), must be added to a liter of water containing 0.800 × 10 3 mol of HCl to attain a pH of 4.75? From Table 3.2, HCl is a strong acid and HPr is a weak acid. Because HCl is a strong acid, it completely dissociates and the initial pH and pOH of the water are H‡ ˆ 8:00  10

4

M and pH ˆ log H‡ ˆ 3:10

pOH ˆ log‰OH Š ˆ 14:00

3:10 ˆ 10:90 and ‰OH Š ˆ 1:25  10

11

M

The final pH is 4.75, which corresponds to [H+] and [OH ] of 1.78 × 10 5 and 5.62 × 10 10 M, respectively. Now consider the amount of NaPr that must be added. NaPr is a salt and HPr is an acid. The equilibrium expression that applies is K ˆ 1:3  10

5

ˆ

‰H‡ Š‰Pr Š ‰HPrŠ

(i)

The chemical equations that apply are NaPr → Na‡ ‡ Pr H‡ ‡ Pr ⇆ HPr H‡ ‡ OH ⇆ H2 O

(ii) (iii) (iv)

The difference in the initial and final [H+] is Δ H‡ ˆ 8:00  10

1:78  10

4

5

ˆ 7:82  10

4

M

The difference between the final and initial [OH ] is Δ‰OH Š ˆ 5:62  10

10

1:25  10

11

ˆ 5:49  10

10

M

The H+ can be removed only by forming H2O or HPr, but from the initial and final OH concentrations, OH must be produced. The only source of OH ions is from the dissociation of additional water molecules according to Eq. (iv); therefore, this equation actually proceeds from right to left. The OH production is less than 10 9 M, and it is accompanied by the production of an equal amount of H+ ions. The amount of H+ removed is equal to the initial amount of H+ plus the production of H+ from dissociation minus the final amount of H+. The contribution of H+ from the dissociation of water is negligible. Using Eq. (iii), the amount of HPr formed is 7.82 × 10 4 M. The amount of H+ produced from H2O dissociation is negligible compared to H+ used to form the weak acid HPr.

3 Acid–Base Chemistry

From the equilibrium expression (i), the ratio of [Pr ] to [HPr] is

1:3  10 5

‰Pr Š ˆ ˆ 0:73 ‰HPrŠ 1:78  10 5

Using this ratio the concentration of Pr is

‰Pr Š ˆ 0:73‰HPrŠ ˆ 0:73 7:82  10

4

ˆ 5:71  10

4

M

The total amount of NaPr to be added consists of the NaPr consumed by H+ to produce HPr and the NaPr dissociated to produce Pr . ‰NaPrŠ added ˆ 7:82  10

4

‡ 5:71  10

4

ˆ 1:35  10

3

M

3.2.1 Conjugate Acids and Bases The loss of H+ by an acid, HA, gives rise to a potential H+ acceptor, A , that is, A is a conjugate base of HA. Similarly, the loss of OH by the base BOH gives rise to a potential OH acceptor, B+, which is indirectly a proton donor and therefore is the conjugate acid of BOH. HA ⇆ H‡ ‡ A Acid Conjugate base

KA

BOH ⇆ B‡ ‡ OH Base Conjugate acid

KB

For a conjugate acid–base pair it is readily shown that K AK B ˆ K w

(3.8)

3.3 Equivalents and Normality In an acid or a base context, the equivalent weight of a compound is that weight that contains one gram-atom of available H+ or its chemical equivalent. Equivalent weights are based on gram atoms; milli-equivalent weights (meq) are conveniently based on milligram atoms. The equiv­ alent weight is often equal to the molecular weight of a compound or ion divided by its valence number. A one normal (1 N) solution contains one equivalent weight of compound. Example 3.2 Equivalent Weight Determination What are the equivalent weights of (a) Ca(HCO3)2 (b) NaH2PO4? a) Ca(HCO3)2 dissociates as follows: Ca…HCO3 †2 → Ca2‡ ‡ 2HCO3 Each HCO3 has the potential to give up 1 H+ or, alternatively, Ca2+ has the potential to accept 2OH s. The equivalent weight of this compound is its MW (162 g) divided by 2, which makes its equivalent weight 81 g. b) NaH2PO4 dissociates as follows: NaH2 PO4 → Na‡ ‡ H2 PO4

61

62

Theory and Practice of Water and Wastewater Treatment

The definition of normality for NaH2PO4 is ambiguous. Following the definition of normality as a measure of the available H+, each H2 PO4 has 2H+s, and the equivalent weight of the compound would be its MW (120 g) divided by 2. However, each H2 PO4 could take up 1H+, which would yield an equivalent weight of 120 g. Using molarity avoids confusion in labeling bottles. The normality of the compound depends on whether it is participating in an acid or a base reaction.

Some compounds do not donate or take up H+ or OH ions directly. The equivalent weight of these compounds is determined after their reaction with water as shown in the following example. Example 3.3 Equivalent Weight Determination What is the equivalent weight of CO2? The hydrolysis reaction for CO2 is CO2 ‡ H2 O → H2 CO3 Each H2CO3 molecule has the potential to donate 2H+ ions: H2 CO3 → 2H‡ ‡ CO23 Therefore, the equivalent weight of CO2 is its MW (44 g) divided by 2, which results in an equivalent weight of 22 g.

The equivalent weight of a compound or ion depends on the circumstances of the reaction of interest. For instance, for a multiprotic acid H3X that is participating in a reaction (e.g., a titration), where the endpoint is fixed when the acid has only given up two hydrogen ions, the equivalent weight of H3X is its gram molecular weight divided by 2. However, when equivalent weights are tabulated, they are normally based on the total potential of the substance to gain or produce H+ ions.

3.4 Solution of Multiequilibria Systems The solution of multiequilibria problems follows this procedure: 1) Identify all the important species participating in the reaction. 2) Write all the known equilibrium relations involving the species identified in step 1. These include acid–base and redox equilibria as well as equilibria between phases such as Henry’s law and Ksp relations. Do not forget the equilibrium relation for the dissociation of water. 3) Write the mass balance relations that apply.

4) Write the charge balance equation for the system.

5) Make any simplifying assumptions.

6) Ensure that the number of equations is equal to the number of unknowns. If more equations

are needed, return to steps 2 or 3. 7) Solve the system of equations algebraically or with computer methods for systems of equations. If computer methods are used, check that a realistic solution has been found (e.g., the pH is between 0 and 14). 8) Check the assumptions made in step 5.

3 Acid–Base Chemistry

Regarding step 3, if the total amount of sulfur ([ST]) in a water is known, then including species listed in Tables 1.2, 1.3, and 3.2 in a mass balance: ‰ST Š ˆ S2

‡ ‰HS Š ‡ ‰H2 SŠ ‡ SO42

‡ SO23

‡ HSO4 ‡ ‰H2 SO4 Š ‡ ‰SŠ ‡ 2 S2 O32

‡ HSO3 ‡ ‰H2 SO3 Š

This mass balance does not include possible complexes of sulfur that could occur among other possible sulfur species. The solution to a multiequilibria problem can become quite involved if all possible entities are considered. The key to solving multiequilibria problems is step 5, where simplifying assumptions are made. Many of the species in the mass balance can be ignored because they are present in minor amounts under normal conditions. For instance, the concentrations of HSO4 and H2SO4 are negligible in all but highly acidic solutions. Typically, these substances would not be included in the mass balance expression. The assumption that concentrations of HSO4 and H2SO4 are negligible is a routine assump­ tion. Similarly, the assumption that strong bases are completely dissociated is normal. Other assumptions that are reasonable for the starting points are as follows: Weak acids are not dissociated to any significant extent in acidic solutions. Weak bases are not dissociated to any extent in basic solutions. A judgment is made on the final pH, and the commensurate assumptions are made. Of course, species at low concentrations are included in the appropriate equilibrium expressions. The starting assumptions should be adequate to reduce the solution to solving one or two equations. The information obtained from this solution is systematically used to solve the remaining equations, which ultimately provides verification of the assumptions or disproves them. If the assumptions have been incorrect, the first solution of the equations will indicate the direction in which adjustments must be made and an iterative approach can be used to solve the problem. The alternative to simplifying assumptions is to write and solve all of the equations algebraically. This process is lengthy and involved, but it will lead to the correct answer. Experience is often required to eliminate certain entities from consideration. In this text, problems will focus on the major possible species as shown in tables in the text or specified in the problems.

3.5 Buffers Buffers are important in analyses and in natural waters. Buffers resist pH change. This has obvious importance for life forms that usually can survive only within a narrow pH range. In analyses, desired reactions can be forced to a specified degree of completion or the rate of reaction may be increased by governing the pH in an appropriate range. Buffers rely upon the combined effects from a weak acid or a weak base, and the salt of the weak acid or base. This is called the common ion effect. Consider a solution of acetic acid, which is a weak acid, and its salt, sodium acetate. The reactions in water are CH3 COOH ⇆ CH3 COO ‡ H‡ CH3 COONa → CH3 COO ‡ Na

KA ‡

(3.9) (3.10)

The salt is highly soluble and dissociates completely. The equilibrium expression is KA ˆ

‰CH3 COO Š‰H‡ Š ‰CH3 COOHŠ

(3.11)

63

64

Theory and Practice of Water and Wastewater Treatment

which can be rearranged to pH ˆ pK A ‡ log

‰CH3 COO Š ‰CH3 COOHŠ

(3.12)

or in general, pH ˆ pK A ‡ log

‰saltŠ ‰acidŠ

or

pH ˆ pK A

p

‰saltŠ ‰acidŠ

(3.13)

Equation (3.13) is known as the Henderson–Hasselbach equation. If all species are in equilibrium, the ratio of [salt]:[acid] for any acid or base is readily determined from this equation at a given pH. When acid is added to the above solution, the acetate ion will combine with H+ to form weakly dissociated acetic acid until essentially all of the acetate ion is exhausted. When base is added to the solution, the OH consumes H+ ions that are replaced by H+ ions donated by dissociation of the acetic acid until essentially all of the acetic acid is dissociated. Example 3.4 Buffer An analyst needs a solution with a pH that is fairly constant around a value of 4.75. Absorption of CO2 and lab fumes and the reagents and impurities in the reagents added to the buffer solution will cause deviation from the desired pH. Noting that the pKa of acetic acid (HAc) is 4.756, the analyst decides to make a solution containing acetate and acetic acid. Using Eq. (3.13), the ratio of salt to acid in the solution should be 1.0 to obtain the desired pH. The analyst chooses to make the solution by adding 0.1 mol of HAc and 0.1 mol of NaAc to 1 L of water. The pH is calculated as follows: CH3 COOH → CH3 COO ‡ H‡ 0:1 x x x CH3 COONa → CH3 COO ‡ Na‡ 0:1 0:1 ˆ 0 0:1 0:1 CH3COO is produced by both reactions. The equilibrium expression is KA ˆ

‰CH3 COO Š‰H‡ Š …0:1 ‡ x†x ˆ 0:1 x ‰CH3 COOHŠ

To solve the above expression, the analyst recognizes that HAc is a weak acid and assumes that x is small compared to 0.1. The equation reduces to KA ˆ

0:1x ˆ 1:754  10 0:1

5

and x is readily determined to be 1.754 × 10 5 (i.e., the pH is very close to 4.75). The assumption that x was small compared to 0.1 was correct. Is the solution a buffer? Addition of 0.05 mol of a strong acid or strong base to 1 L of an unbuffered water would result in a pH of 1.30 or 12.70, respectively. Let us find the pH of our buffered solution after addition of the same quantities of either an acid or a base. When 0.05 mol of HCl is added, the H+ ions will react with Ac to form weakly dissociated HAc: HCl ‡ Ac → HAc ‡ Cl

3 Acid–Base Chemistry

Again, using the assumption that the contribution of Ac and H+ from dissociation of HAc is small, the resulting concentrations of species involved are ‰Ac Š ˆ 0:1 0:05 ˆ 0:05 M

‰HAcŠ ˆ 0:1 ‡ 0:05 ˆ 0:15 M

The pH is found from Eq. (3.13): pH ˆ pK A ‡ log

‰saltŠ ‰acidŠ

or pH ˆ 4:75 ‡ log

‰0:05Š ˆ 4:27 ‰0:15Š

A check will prove the assumption to be correct. When 0.05 mol of NaOH is added, the OH ions will react with the HAc. NaOH ‡ HAc → Ac ‡ Na‡ ‡ H2 O Using the assumption that the contributions of Ac and H+ from dissociation of HAc are small, the resulting concentrations of species involved are ‰Ac Š ˆ 0:1 ‡ 0:05 ˆ 0:15 M

‰HAcŠ ˆ 0:1 0:05 ˆ 0:05 M

The pH is pH ˆ pK A ‡ log

‰saltŠ ‰acidŠ

or pH ˆ 4:75 ‡ log

‰0:15Š ˆ 5:23 ‰0:05Š

A check will again prove the assumption to be correct. The pH change of the buffered solution is much less dramatic than the pH change of an unbuffered solution. The effective buffering capacity of the solution is approximately equal to the amounts of HAc and Ac in the initial solution.

If a weak acid or base with a pK value exactly at the desired pH cannot be found, the ratios of the buffering agent acid and salt are adjusted to provide the desired pH in the buffered solution. 3.5.1 Dilution of a Buffered Solution Addition of water has no effect on the pH of a buffered solution unless the dilution is large. This can be proven by using the Henderson–Hasselbach equation. Consider a solution that contains m mol of salt and n mol of acid in a volume V. The pH is calculated from pH ˆ pK A ‡ log

m=V m ‰saltŠ ˆ pK A ‡ log ˆ pK A ‡ log n=V n ‰acidŠ

This equation proves that dilution has no effect on the pH of a buffered solution until the dilution is large enough to cause the contribution of H+ from the dissociation of water to be significant. At this point, the buffering capacity is negligible, and the resulting equations are of little practical significance. 3.5.2 The Most Effective pH for a Buffer The buffering capacity of a solution is quantitatively determined by measuring the pH change upon addition of a given amount of strong acid or base.

65

66

Theory and Practice of Water and Wastewater Treatment

In general terms, consider a buffer solution made from the addition of a strong base, BOH, to a solution containing a weak acid, HA. The equations to be used are BOH ‡ HA → B‡ ‡ A ‡ H2 O

(3.14)

‡

HA → H ‡ A

(3.15)

The amounts of HA and A are determined assuming that the contribution of A from the dissociation of HA is negligible as demonstrated in Example 3.4. In this case, the salt ion (A ) is formed by Reaction (3.14); there is no direct addition of a salt containing A . If the starting concentration of HA in terms of equivalents in a volume, V, is n/V and the equivalents concentration of BOH added is m/V, the equilibrium expression is calculated as follows: KA ˆ

‰H‡ Š‰A Š ‰H‡ Šm ˆ …n m† ‰HAŠ

H‡ ˆ K A

…n

m† m

or

(3.16) pH ˆ pK A ‡ log

m n

m

(3.17)

The rate of change of pH with respect to addition of BOH is 2:303

d pH 1 1 ˆ ‡ dm n m m

(3.18)

The question is when is the above expression a minimum? It is found by taking the derivative and setting it equal to zero. d2 pH n…2m n† ˆ ˆ0 dm2 m2 …n m†2

(3.19)

This implies that m = n/2. It is not possible for n to be zero because there was HA in the starting solution. Substituting this into the pH expression [Eq. (3.17)], it is found that the solution is most highly buffered when pH ˆ pK a

and

‰saltŠ ˆ ‰acidŠ

3.6 Acid–Base Titrations A titration is a volumetric form of analysis in which a measured amount of solution of known concentration (the titrant) is added to a sample (see Chapter 5 for more discussion of titrations). When an acid or a base is added to a solution, all species with a capability of donating or accepting H+ or OH ions react to attain a new state of equilibrium. Acid–base reactions are usually rapid, occurring in a matter of seconds. The amount of pH change after the addition of titrant depends on the buffering compounds in solution. 3.6.1 Titration of Strong Acids and Bases The presence of significant amounts of a strong acid in a water will cause the pH to be low. To examine the change in pH with the addition of a strong base consider the following example. Consider a 25 mL volume of a 0.10 M HCl that is titrated with 0.10 M NaOH. Because the reactants are strong, they are completely dissociated and the titration reaction simply is H‡ ‡ OH → H2 O

3 Acid–Base Chemistry

Table 3.3 pH variation in a strong acid strong base titration. Volume of NaOH added (mL)

[H+] (M)

pH

0.00

1.00 × 10

1

1.00

10.00

4.27 × 10

2

1.37

20.00

1.11 × 10

2

1.95

24.00

2.04 × 10

3

2.69

24.90

2.00 × 10

4

3.70

24.99

2.0 × 10

5

4.70

25.00

1.00 × 10

7

7.00

25.01 25.10

5.0 × 10

10

9.30

5.0 × 10

11

10.30

26.00

5.0 × 10

12

11.30

30.00

1.0 × 10

12

12.00

40.00

4.3 × 10

13

12.37

The variation of pH after the additions of various amounts of titrant is tabulated in Table 3.3. The data in Table 3.3 are plotted in Figure 3.1. The curve for the titration of a strong base with a strong acid is also shown on the figure. It is the reverse mirror image of the first curve. It is observed that strong acids and bases effectively buffer the pH at low and high values, respectively. The pH fluctuation around a value of 7 is dramatic. The equivalence point is the pH at which the equivalents of titrant added are equal to the initial equivalents in solution. For this titration the equivalence point (endpoint for stopping the titration) is a pH of 7. The curves in Figure 3.1 are typical for all titrations involving strong monoprotic (single hydrogen) acids and strong bases.

Figure 3.1 Titration of strong acids and bases.

67

68

Theory and Practice of Water and Wastewater Treatment

Table 3.4 pH variation in a weak acid strong base titration. Volume of NaOH added (mL)

[H+] (M)

pH

0.00

2.00 × 10

3

2.70

10.00

2.69 × 10

5

4.57

20.00

5.01 × 10

6

5.35

24.00

7.50 × 10

7

6.12

24.90

7.9 × 10

8

7.10

25.00

2.0 × 10

9

25.10

5.0 × 10

11

10.30

26.00

5.1 × 10

12

11.29

30.00

1.1 × 10

12

11.96

40.00

4.3 × 10

13

12.37

8.70

3.6.2 Titration of Weak Acids and Bases To illustrate the variation of pH in a solution that contains a weak acid when titrated with a strong base, consider a 25 mL volume of 0.10 M acetic acid (HAc) titrated with 0.10 M NaOH. The reaction is HAc ‡ NaOH → H2 O ‡ Na‡ ‡ Ac In addition, until the HAc has been consumed, the dissociation of the remaining HAc must be considered. HAc ⇆ H‡ ‡ Ac

K eq ˆ 1:8  10

5

The data for this titration are given in Table 3.4. See Example 3.5 for the calculation of the solution pH after addition of titrant. The data in Table 3.4 are plotted in Figure 3.2 along with the data for a 0.10 M solution of NH3 (a weak base) titrated with 0.10 M HCl.

Figure 3.2 Titration of weak acids and bases.

3 Acid–Base Chemistry

The following observations are made from Figure 3.2 compared with Figure 3.1. The equiv­ alence points for these titrations have shifted from neutrality. Also the range of pH variation near the equivalence point has decreased compared to the strong acid–strong base situation. For the HAc titration with NaOH, the titration curve is the nearest to being flat at a pH of ~4.75, which is the pH of maximum buffering capacity as previously illustrated in Section 3.5.2. As noted earlier, the correct solution of any system can be made by using the equilibria, mass balance, and charge balance equations without any simplifying assumptions. However, in many cases, it is possible to make simplifying assumptions and reduce the effort in solving the problem. The important point is to check whether the assumptions have been valid. This approach is used in Example 3.5 for pH variation in a titration. Example 3.5 pH Determination in a Titration Find the pH of the solution, when (a) 20 and (b) 25 mL of 0.10 N NaOH have been added to 25 mL of 0.10 N acetic acid. Note that the latter pH is the equivalence (endpoint) pH for the titration. The temperature is 25 °C. a) NaOH is a strong base and dissociates completely. The OH– ion then reacts with HAc to form Ac– as follows. NaOH ‡ HAc → Na‡ ‡ Ac ‡ H2 O The volume of the solution is 20 + 25 = 45 mL. The equivalents of acid originally present are 25 mL  0:10 N 

1 eq L 1N

1



1L ˆ 2:50  10 1000 mL

3

eq

3

eq

The equivalents of base added are 20 mL  0:10 N 

1 eq L 1N

1



1L ˆ 2:00  10 1000 mL

The reaction and equilibrium relation to be satisfied are HAc ⇆ H‡ ‡ Ac

K HAc ˆ

‰H‡ Š‰Ac Š ˆ 1:754  10 ‰HAcŠ

5

Because HAc is a weak acid, there are 2.50 × 10 3 2.00 × 10 3 = 5.00 × 10 4 eq of HAc and 2.00 × 10 3 eq of Ac in the 45 mL solution. The concentrations of these substances are ‰HAcŠ ˆ

5:00  10 4 eq 1000 mL  1 mol=eq  ˆ 1:11  10 45 mL 1L

2

M

‰Ac Š ˆ

1000 mL

2:00  10 3 eq  1 mol=eq  ˆ 4:44  10 45 mL 1L

2

M

Substituting these concentrations into the Henderson–Hasselbach (equilibrium) expression the solution pH is pH ˆ pK HAc ‡ log

4:44  10 ‰Ac Š ˆ 4:756 ‡ log ‰HAcŠ 1:11  10

2 2

ˆ 5:35

The assumption on the weakly acidic behavior of HAc can be verified. b) The equations noted for part (a) apply. In this case, the volume of the solution is 25 + 25 = 50 mL. The equivalents of base added is the same as the starting amount of HAc and equal to 2.5 × 10 3 eq. It is true that the amount of Ac is nearly equal to 2.5 × 10 3 eq, but the amount of HAc in solution is not zero. Another equation is required to solve the system. Recall the

69

70

Theory and Practice of Water and Wastewater Treatment

charge conservation (or charge balance) principle. The statement of charge conservation for this system is H‡ ‡ Na‡ ˆ ‰OH Š ‡ ‰Ac Š [Na+] is known. Assume 2:5  10 3 eq  1 mol=eq  1000 mL L 50 mL

‰Ac Š ˆ

1

ˆ 0:050 M

[OH ] can be put in terms of [H+] by use of the equilibrium relation for the dissociation of water: H‡ ‡ ‰OH Š ⇆ H2 O

K w ˆ 1  10

14

Substituting this into the charge balance equation and rearranging, Kw 2 ‰Ac Š H‡ ‡ H‡ Na‡ H‡ ‡ Na‡ ˆ ‡ ‡ ‰Ac Š ) ‰H Š

Kw ˆ 0

Because [Ac ] = [Na+] = 0.050 M, the quadratic reduces to [H+]2 = Kw = 10 14 with a solution of [H ] = 1 × 10 7 or pH = 7.00. To check the assumption, use the equilibrium relation for HAc +

‰HAcŠ ˆ

10 7 …0:050† ‰H‡ Š‰Ac Š ˆ ˆ 2:85  10 K HAc 1:754  10 5

4

M

The assumption that [Ac ] = 0.050 M (or that there were 2.5 × 10 3 eq of Ac in solution) is reasonable because 0.050 2.85 × 10 4 = 0.0497, which is an error of approximately 0.6%. However, a further check will show that the error with respect to the hydrogen ion concentration is much larger. The amount of HAc consumed is HAi

HAf ˆ …25 mL† 0:10 eq L

1L 1000 mL

…1 mol=eq†

1

…50 mL† 2:85  10

4

mol L

1

1L 1000 mL 1L 1000 mL

…50 mL† 2:85  10 ˆ 2:472  10

3

4

mol L

1

mol

If 2.50 × 10 3 mol of NaOH have been added, then the excess OH is 2.80 × 10 5 mol that are present in the final 50 mL volume, which is equivalent to [OH ] = 5.60 × 10 4 M or a pOH of 3.25. This corresponds to a pH of 10.75, which is significantly different from 7.00. Our assumption has effectively dictated that the only source of H+ or OH ions is from the dissociation of water itself, which results in the small concentration of 10 7 M for H+. Small errors in the amount of HAc that dissociates can dramatically affect the H+ concentration in this circumstance. A pOH of 3.25 or pH of 10.75 is not the final answer. For this special situation, we must solve the system without any simplifying assumptions. The total amount of acetate, AcT is AcT ˆ ‰HAcŠ ‡ ‰Ac Š ˆ 0:050 M Substituting this into the equilibrium relation for HAc and solving for [Ac ], K HAc ˆ

‰H‡ Š‰Ac Š AcT ‰Ac Š

‰Ac Š ˆ

AcT K HAc ‰H‡ Š ‡ K HAc

Replacing [Ac ] in the charge balance with this relation: AcT K HAc Kw ‡ ‰H‡ Š ‰H‡ Š ‡ K HAc

) ‰H‡ Š3 ‡ ‰H‡ Š2 …‰Na‡ Š ‡ K HAc † ‡ ‰H‡ Š…‰Na‡ ŠK HAc

‰H‡ Š ‡ ‰Na‡ Š ˆ

AcT K HAc

K w†

K w K HAc ˆ 0

3 Acid–Base Chemistry

Substituting the known values into the above equation and solving it for the only feasible [H+] yields a pH of 8.73. From the equilibrium expression and mass balance for acetate, [Ac ] = 0.0500 M and [HAc] = 5.31 × 10 6 M. The amount of undissociated HAc is significant (and therefore, the unreacted OH from the NaOH is also significant) with respect to the dissociation of water. Part (a) should also be checked to ensure that the charge balance relation is satisfied.

Figure 3.3 is a comparison of titration curves for acids with varying dissociation constants. The pH change near the equivalence point decreases as the dissociation constant for the acid decreases. For a dissociation constant below a value of about 10 9, an acid–base titration is not a useful analytical tool to measure the acid concentration. 3.6.3 Indicating the Endpoint of an Acid–Base Titration There are a couple of methods to determine the equivalence point (endpoint) pH in a titration. The solution should be stirred during the titration. The pH can be directly monitored during the titration, which provides the most accurate point for stopping the titration. An entirely suitable method for determining the stopping point is the use of a visual indicator. Indicators are convenient and are commonly utilized in routine work. An indicator produces a visible change in color as the pH changes; therefore, it is sensitive to the presence of H+ ions. An equation describing the reaction of an indicator with H+ ions is HIn ⇆ H‡ ‡ In Acid color Base color

(3.20)

The predominance of either the acidic or basic species determines the color of the solution. The equilibrium expression for the above reaction is K In ˆ

‰H‡ Š‰In Š ‰HInŠ

(3.21)

Typically a five-to-tenfold ratio of concentration of one form to the other is required for the color change to be apparent to visual observation. 1 ‰In Š  Acid color 10

‰HInŠ

(3.22a)

‰In Š  10 Base color ‰HInŠ

(3.22b)

Figure 3.3 Titration curves for acids of varying strength.

71

72

Theory and Practice of Water and Wastewater Treatment

Table 3.5 Acid–base indicators. Indicator

pH transition range

pKa)

Bromophenol blue

3.0 (yellow)–4.6 (purple)

4.10

Bromocresol green

3.8 (yellow)–5.4 (blue)

4.90

Bromothymol blue

6.0 (yellow)–7.6 (blue)

7.30

Chlorophenol red

4.8 (yellow)–6.4 (purple)

6.25

Cresol red

0.2 (red)–1.8 (yellow)

—

7.0 (yellow)–8.8 (purple)

8.46

2,6-Dinitrophenol

1.7 (colorless)–4.7 (yellow)

3.70

Litmus

5 (red)–8 (blue)

6.5

Metacresol purple

1.2 (red)–2.8 (yellow)

1.51b)

7.4 (yellow)–9.0 (purple)

8.32

Methyl orange

3.1 (red)–4.4 (yellow)

3.46

Methyl red

4.4 (red)–6.2 (yellow)

5.00

Phenolphthalein

8.2 (colorless)–9.8 (purple)

9

Phenol red

6.4 (yellow)–8.2 (red)

8.00

Thymol blue

1.2 (red)–2.8 (yellow)

1.65

8.0 (yellow)–9.6 (blue)

9.20

a) All values at 20 °C and ionic strength of 0.

b) At an ionic strength of 0.1.

Source: Adapted from Bányai (1972) and Chemguide (2017).

Applying the Henderson–Hasselbach equation [Eq. (3.13)] to Eqs. (3.22a) and (3.22b) determines the effective pH range for color change of the indicator. pH  pH  1

1

logK In logK In

Acid color Base color

and the effective pH range for the indicator is pH range ˆ pK In  1

(3.23)

Acid–base indicators are generally weak organic acids or bases; commonly used acid–base indicators in water analyses are given in Table 3.5. Example 3.6 Influence of an Indicator on a Titration Indicator solutions are typically made using 0.5–1 g of indicator per liter. Determine the equivalent CaCO3 concentration of an acid–base indicator in a solution to be titrated. The indicator has a molecular weight of 250 g and one drop of indicator solution is to be added to a sample volume of 25 mL for the titration. A drop contains approximately 0.05 mL. The indicator has an equivalent weight of 250 g. The concentration of indicator in the indicator solution ([I]I) is ‰IŠI ˆ 1 g L

1

1 mol 250 g

ˆ 4:00  10 3 M

or 4:00  10

3

eq L

1

3 Acid–Base Chemistry

The concentration of indicator equivalents in the sample solution ([I]s) is ‰IŠs ˆ

…0:05 mL† 4:00  10 25 mL

3

eq L

1

ˆ 8:00  10

6

eq L

1

The equivalent weight of CaCO3 is 100 g/2 = 50 g ‰CaCO3 Š ˆ 8:00  10

6

eq L

1

…50 g=eq† 103 mg g

1

ˆ 0:40 mg L

1

This amount of indicator is equivalent only to a very small amount of CaCO3 in the sample and can be ignored in usual circumstances.

3.7 Natural Buffering of Waters from Carbon Dioxide and Related Compounds The most significant buffering system in natural waters and wastewaters is due to carbon dioxide and its related species: carbonic acid (H2CO3), bicarbonate (HCO3 ), and carbonate (CO23 ). Carbon dioxide and its related species exhibit weak acid behavior, which provides buffering capacity. These species are present in relatively large amounts in waters because of biological activity that produces CO2 and because of the high degree of solubility of CO2 as a result of its hydrolysis. The equilibria among these are described by the following equations: CO2 …g† ⇆ CO2 …aq†

KH

(3.24)

CO2 ‡ H2 O ⇆ H2 CO3 K 0

(3.25)

H2 CO3 ⇆ HCO3 ‡ H‡ K ´1

(3.26a)

HCO3 ⇆ CO23 ‡ H‡

K 2

(3.27)

The reaction in Eq. (3.25) is quite sensitive to the ionic strength and temperature2 of the water. Normally, the concentrations of CO2 and H2CO3 are considered together with the following mass balance, H2 CO∗3 ˆ ‰CO2 Š ‡ ‰H2 CO3 Š

(3.28)

and the following equation is used to describe the first dissociation of carbonic acid. H2 CO∗3 ⇆ HCO3 ‡ H‡

K 1

(3.26b)

Note that the equilibrium constants for carbonic acid in Table 3.2 are reported for H2 CO∗3 . Example 3.7 gives the derivation of the equilibrium constant for the dissociation of H2 CO∗3 . The H2 CO∗3 concentration is the free carbon dioxide concentration of a water.

2 The temperature variation of pK1 and pK2 from 273 to 363 K are described by pK1 = 356.3094 + 0.060 919 64 T 21 834.37/T (126.8339)logT + 1 684 915/T2 and pK2 = 107.8871 + 0.032 528 49 T 5151.79/T (38.925 61) logT + 563 713.9/T2 (Plummer and Busenberg 1982).

73

74

Theory and Practice of Water and Wastewater Treatment

Example 3.7 Equilibrium Constant for Dissociation of H2 CO∗3 Determine the dissociation constant for H2 CO∗3 in terms of the equilibrium constants for CO2 and H2CO3. The equilibrium expressions for Eqs. (3.25), (3.26a), and (3.26b) are given in Eqs. (i), (ii), and (iii), respectively. K0 ˆ

‰H2 CO3 Š ‰CO2 Š

(i)

K ´1 ˆ

HCO3 ‰H‡ Š ‰H2 CO3 Š

(ii)

K1 ˆ

HCO3 ‰H‡ Š H2 CO∗3

(iii)

We can use Eq. (3.28) to relate H2 CO∗3 to [H2CO3] and [CO2] and then use the equilibrium expressions. K1 ˆ

HCO3 ‰H‡ Š ‰H2 CO3 Š ‡ ‰CO2 Š

Eliminating [CO2] in the above equation with Eq. (i), K1 ˆ

HCO3 ‰H‡ Š ˆ ‰H2 CO3 Š ‡ ‰CO2 Š

HCO3 ‰H‡ Š HCO3 ‰H‡ Š ˆ ‰H2 CO3 Š 1 ‰H2 CO3 Š ‡ ‰H2 CO3 Š 1 ‡ K0 K0

ˆ

K 1´ 1‡

1 K0

ˆ

K 0 K 1´ 1 ‡ K0

Note that Henry’s law for CO2 dissolution can also be reformulated in terms of H2 CO∗3 such that CO2 …g† ⇆ H2 CO∗3

and

´ H2 CO3∗ ˆ K H CO2 …g†

3.7.1 Acidity and Alkalinity Alkalinity (Alk) is defined as the ability of a water to neutralize acids; it is a measure of buffering capacity against a pH drop. In addition to inorganic carbon forms, [OH ] may be significant. The definition of alkalinity, assuming that only inorganic carbon is significant, is ‰AlkŠ ˆ HCO3 ‡ CO32

‡ ‰OH Š

H‡

(3.29)

In this equation all concentrations are expressed in eq L 1. In most circumstances, [H+] is insignificant and the equation simplifies to ‰AlkŠ ˆ HCO3 ‡ CO32

‡ ‰OH Š

If other ions that can take up H+ ions are present in significant concentrations, they must be included on the right-hand side of the equation. Examples of other ions that can contribute to alkalinity are: SiO…OH†3 , PO34 , HPO24 , and H2 PO4 . Figure 3.4 gives the distribution of inorganic carbon among its three forms as a function of pH. The curves for the carbon species maintain their relative spacing regardless of the total concentration of inorganic carbon in solution. Figure 3.5, which is partly derived from Figure 3.4,

3 Acid–Base Chemistry

Figure 3.4 Inorganic carbon species as a function of pH (total concentration of inorganic carbon = 0.005 M).

shows the forms of alkalinity that are significant in various pH ranges. These figures are based on the equilibrium constants for the dissociations of carbonic acid. If the pH of a water is above 10, the presence of strong base is indicated and [OH ] or hydroxyl alkalinity becomes significant. Acidity (Acy) is the opposite of alkalinity: it is the capacity to neutralize bases. Hydrogen ion itself can neutralize bases, and assuming that only CO2-related species are significant, it is defined as follows: Acy ˆ H2 CO∗3 ‡ HCO3 ‡ H‡

‰OH Š

(3.30) 1

In this equation, all concentrations are expressed in eq L . In most circumstances, [OH ] is insignificant and the equation simplifies to Acy ˆ H2 CO3∗ ‡ HCO3 ‡ H‡ If the presence of other ions that can take up OH is significant, they must be included on the right-hand side of the equation. Examples of other possible acidity species are HOCl, HPO24 , H2 PO4 , and H3PO4. It is observed from Figure 3.5 that below a pH of 4.6, the concentration of H+ ions is significant, which indicates a significant concentration of mineral or strong acids. Bicarbonate ion appears in both alkalinity and acidity expressions because it can neutralize acids or bases.

Figure 3.5 pH ranges over which forms of alkalinity and acidity are significant.

75

76

Theory and Practice of Water and Wastewater Treatment

Regardless of the components of alkalinity and acidity that are present in a water, the carbon dioxide titration endpoints of pH 4.6 (carbonic acid) and 8.3 (bicarbonate) are often used in the determination of alkalinity and acidity in situations where carbon dioxide species are significant and even in circumstances where they are present in small amounts compared to other alkalinityor acidity-causing agents. Functional rules based on the results of the titration are developed in these circumstances. For instance, the wine industry uses the bicarbonate endpoint to determine acidity even though the primary acids present are tartaric, malic, and citric acids, which have dissociation constants much different from CO2 species. Example 3.8 Alkalinity Determination A water with a pH of 8.00 contains HPO24 ,H2 PO4 , HCO3 , and NH3 at 7.6, 2.4, 52, and 1.5 mg L 1, respectively. What is its alkalinity in eq L 1 and as CaCO3? The concentrations of each of these species are ˆ

7:6 mg L 1 96 000 mg mol

H2 PO4 ˆ

2:4 mg L 1 97 000 mg mol

HPO24

HCO3 ˆ ‰NH3 Š ˆ

52 mg L 1 61 000 mg mol

1:5 mg L 1 17 000 mg mol

ˆ 7:92  10

5

1

M

ˆ 2:47  10

5

1

M

1

ˆ 8:52  10

ˆ 8:82  10

1

5

4

M

M

A check of equilibrium relations will show that the concentrations of the other possible alkalinity ions are insignificant. Except for HPO24 all species have the capability of taking up 1H+ ion. The total alkalinity of the water is ‰AlkŠ ˆ 2 eq mol

7:92  10 5 M

1

‡ 1 eq mol ˆ 1:12  10

3

1

2:47  10 5 M ‡ 8:52  10 4 M ‡ 8:82  10 5 M

eq L

1

CaCO3 has 50 g/eq. The alkalinity expressed as CaCO3 is ‰AlkŠ ˆ 1:12  10

3

eq L

1

…50 g=eq† 1000 mg g

1

ˆ 56 mg L

1

as CaCO3

The contributions of ions other than HCO3 are small, which is typical.

Questions and Problems 3.1 What is the molar concentration of water in pure water? Calculate the equilibrium constant at 25 °C for the dissociation of water when the concentration of water is incorporated into the denominator of the equilibrium expression. 3.2 Show by actual calculations whether the pH of pure water will increase or decrease with an increase in temperature using the data for Kw at 0, 25, and 50 °C given in the text.

3 Acid–Base Chemistry

3.3 Is a negative pH theoretically possible? 3.4 The pH of a solution is 5.92. What are the pOH, the hydrogen ion concentration, and the hydroxyl ion concentration at 25 °C? 3.5 What is the equilibrium constant and pKeq for the following reaction?

NH‡4 ⇆ NH3 ‡ H‡ 3.6 At a pH of 7.79, what are the ratios of the following species: H3PO4 and H2 PO4 ; NH3 and NH‡4 ; and HOCl and OCl ? 3.7 The hydrogen ion concentration in a dilute solution of sulfuric acid is 2 × 10 the pH value? What is the pOH value at 25 °C?

5

M. What is

3.8 What is the hydrogen ion concentration and pH of a 0.200 M solution of acetic acid at 25 °C? 3.9

a How many mL of a 0.10 N solution of NaOH must be added to a 25 mL solution of 0.10 N HAc to obtain a pH of 9.75? What is the concentration of HAc at this pH? b Answer the same question with the following changes: [HAc] = 0.010 N in 25 mL and the normality of the titrant is 0.020 N. The final pH is 11.00.

3.10 What are the conjugate acids or bases of the following: HCl, SO24 , Na+, NH3, KOH, HOCl, and HF? 3.11 The dissociation constants for a monoprotic-acid, its conjugate base and water are Ka, Kb, and Kw, respectively. Prove that KAKB = Kw for a conjugate acid–base pair. 3.12 What is the equivalent weight of NH3 for an acid–base reaction? 3.13 What is the equivalent weight of Na2HPO4 in an acid reaction (i.e., H+ is being added)? In a basic reaction? 3.14

a Explain why a weak acid or base without its salt is not a good buffer. b Explain why a strong acid or strong base buffers a solution at a low or high pH, respectively.

3.15 In the buffer example (Example 3.4), what is the contribution of H+ from the dissociation of water in the initial solution containing only HAc and NaAc? 3.16

a What is the hydrogen ion concentration in 500 mL of 0.100 M solution of acetic acid at 25 °C, if the solution contains additional 2.00 g of acetate ions added in the form of sodium acetate (i.e., 2.78 g NaAc were added). b What will be the hydrogen ion concentration if 4 mmol of NaOH are introduced into this buffered solution? c What is the pH in each case?

3.17 In what ratio must H3PO4 and NaOH be added to a liter of water to buffer it at a pH of 4.00? (Hint: HPO24

and PO34

will be negligible.)

77

78

Theory and Practice of Water and Wastewater Treatment

3.18 What is the ratio of propionic acid and sodium propionate that must be added to a water to initially buffer it at a pH of 5.63? 3.19 Show by calculations that the equivalence pH for the conditions given in Table 3.3 is 7.0. 3.20 Plot the data from Table 3.3 for the range of volumes of NaOH added between 24.0 and 26.0 mL. Use the full page of normal graph paper for the plot. Draw the line of best fit through the points. How much error would occur in the titration if an indicator that changed color at a pH of 8.3 was used as the stopping point? 3.21 Prepare a table similar to Table 3.4 for the titration of 25 mL of 0.10 M NH3 with 0.10 M HCl. 3.22 Would Cl , H2S, HS , and S2 contribute to the alkalinity or acidity of water? 3.23 Write the balanced reaction for the oxidation of NH‡4 to NO3 . How much alkalinity as CaCO3 is consumed by this reaction? 3.24 Four liters of water are equilibrated with a gas mixture containing carbon dioxide at a partial pressure of 0.3 atm. Use a Henry’s law constant for H2 CO∗3 solubility of 2.0 g (L atm) 1. a How many grams of CO2 are dissolved in the water? What is the pH of this water? (Neglect the secondary dissociation of carbonic acid.) b Assume that HCO3 dissolves first without dissociating, achieving equilibrium with the atmosphere, and then the solution is closed to the atmosphere. H2 CO∗3 then dissociates (neglect the secondary dissociation). Find the pH of this solution and compare it to the result obtained in (a)? 3.25 What is the equilibrium constant for the reaction below in terms of K0, K ´1 , and K2 as defined by Eqs. (3.25), (3.26a), and (3.27), respectively? H2 CO∗3 ⇆ CO23 ‡ 2H‡ 3.26 Can alkalinity be expressed as HCl? 3.27 Why is bicarbonate ion a component of both alkalinity and acidity? 3.28 What are the values of the following ratios at pHs of 4.6 and 8.3: H2 CO∗3 : HCO3 and HCO3 : CO23 ? Do these results agree with Figure 3.4? 3.29 What are the alkalinity and acidity of water that contains 79 mg L of 7.3? Assume a temperature of 25 °C.

1

of HCO3 and has a pH

References Bányai, É. (1972). Acid–base indicators. In: Indicators (ed. E. Bishop), 65–176. Toronto: Pergamon Press. Chemguide (2017). Acid-base indicators. http://chemguide.co.uk/physical/acidbaseeqia/indicators. html (accessed February 2017).

3 Acid–Base Chemistry

Hanson, A.T. and Cleasby, J.L. (1990). The effects of temperature on turbulent flocculation: fluid dynamics and chemistry. J. Am. Water Works Assoc. 82 (11): 56–73. http://www.jstor.org/stable/ 41293075. Marshall, W.L. and Franck, E.U. (1981). Ion product of water substance, 0–1000 °C, 1–10,000 bars. New international formulation and its background. J. Phys. Chem. Ref. Data 10 (2): 295–304. doi: 10.1063/1.555643. Plummer, L.N. and Busenberg, E. (1982). The solubilities of calcite, aragonite and veterite in CO2– H2O solutions between 0 and 90°C and an evaluation of the aqueous model for the system CaCO3–CO2–H2O. Geochim. Cosmochim. Acta 46 (6): 1011–1040. doi: 10.1016/0016-7037(82) 90056-4. Speight, J. (ed.) (2005). Lange’s Handbook of Chemistry, 16e. New York: McGraw-Hill.

79

81

4 Organic and Biochemistry Organic chemistry is the study of compounds of carbon. Living systems form organic molecules, but there is also an abundance of synthetic organics. Modern industry manufactures something on the order of 10 000 new compounds per year, most of which are organic. More than 1 100 000 organic compounds have been prepared. Determining their behavior in the environment and methods for treatment of these compounds presents a formidable task for chemists, biologists, ecologists, and environmental engineers. In this chapter, the major classes of organic compounds are described. A brief overview of metabolism is also presented to gain some appreciation of the means of processing foodstuffs and the complexity of metabolism.

4.1 Carbon Carbon is a unique atom, with properties that account for the limitless compounds it can form. The bonds between carbon atoms are stable, which allows chains to be formed. Atoms near carbon in the periodic table can also bond to carbon to form chains. Each carbon atom can form four bonds. The bonds are covalent in nature; that is, electrons are shared between the participating atoms. If the atoms bonded to carbon have an electronegativity that is significantly different from carbon, then the bond will have a polar character, with the electrons being shared unequally between the atoms. The degree of polarity in an organic molecule determines its solubility in the polar solvent, water. Because of the four bonds and chaining ability of carbon, many different configurations are possible for the same group of atoms, which is known as isomerism. The principal atoms that form compounds with carbon are hydrogen and oxygen. Nitrogen, phosphorus, and sulfur are the most common minor elements found in naturally occurring organic molecules. Carbon dioxide and its related species are inorganic ions or compounds.

4.2 Properties of Organic Compounds There are more known compounds of carbon than any other element except hydrogen. Sawyer et al. (2003) have summarized the major characteristics of organic compounds compared to inorganic compounds. A brief explanation is added to most of their points. 1) Organic compounds are usually combustible. A large amount of energy is released as an organic compound is oxidized to CO2, H2O, and other oxides. The energy release propagates the reaction. Theory and Practice of Water and Wastewater Treatment, Second Edition. Ronald L. Droste and Ronald L. Gehr.

 2019 John Wiley & Sons Inc. Published 2019 by John Wiley & Sons Inc.

Companion website: www.wiley.com/go/droste/water

82

Theory and Practice of Water and Wastewater Treatment

2) Organic compounds usually have lower melting and boiling points. Covalent bonding within organic molecules reduces their bonding or attraction to identical molecules or other molecules. 3) Organic compounds are usually less soluble in water. The nonpolar nature of carbon–carbon bonds reduces solubility of these compounds in the polar solvent, water. 4) Several isomers exist for a given formula. The isomers are a result of different geometries for a compound or result from changing the bonding location of atoms or groups of atoms. 5) Reactions of organic compounds are usually molecular rather than ionic, which slows their rate of reaction compared to ionic compounds. 6) The valency of carbon allows it to form multiple bonds and carbon has the ability to form strong bonds with itself. This can result in high-molecular-weight compounds, often well over 1000. 7) Most organic compounds can serve as a source of food for bacteria and other micro­ organisms. Organic compounds are energy rich (carbon is in a reduced state) and contain the raw materials for cell fabrication. In general, as the size of an organic molecule increases, its melting and boiling points begin to rise. Hydrocarbons are organic compounds that only contain hydrogen and carbon. There are three major types of organic compounds: aliphatic, aromatic, and heterocyclic groups. Aliphatic com­ pounds are chained and branched compounds. Examples of isomers for a five-carbon compound are illustrated in Figure 4.1. Because carbon atoms can be rotated about their bonds, a bend in the compound does not produce a new isomer. Only changes in association of atoms produce an isomer. As atoms are substituted for H, more isomers are possible, as illustrated in Figure 4.2.

Figure 4.1 The two possible isomers of pentane.

Figure 4.2 Isomers of dichloroethane.

4.3 Functional Groups

Functional groups strongly influence the behavior of organic compounds. Three major functional groups are the alcohol, OH; carboxyl, COOH; and, carbonyl, >C O, groups.

4 Organic and Biochemistry

The alcohol group is similar in behavior to water; that is, it is neither strongly acidic nor basic, but the bond is polar. The carboxyl group is acidic in nature. The electronegativity of the two oxygen atoms pulls electrons away from the hydrogen atom, which then dissociates as H+. The carbonyl group is an extremely important group in modern synthetic organic chemistry. Because of the high electronegativity difference between carbon and oxygen, the free bond in the carbonyl group provides a highly reactive site.

4.4 Types of Organic Compounds A brief survey of some of the major types of organic compounds is presented. 4.4.1 Aliphatic Compounds Aliphatic compounds can be broken down into alkanes, which contain only single carbon– carbon bonds; alkenes, which contain at least one double bond between two carbon atoms; and alkynes, which contain at least one triple bond between a carbon pair. Compounds that contain double or triple carbon bonds are unsaturated compounds; single carbon–carbon bonded compounds are saturated. Double or triple carbon bonds are major reaction sites. Many synthetic reactions break these multiple bonds and add halogens or other atoms to the carbon. Aldehydes and Ketones

All aldehydes contain the carbonyl group with a hydrogen atom bonded to the carbon atom in the carbonyl group. Their general formula is

. Ketones have the general formula

.

Alcohols, Esters, and Ethers

Alcohols have the functional group, OH, and general formula ROH; ethers are compounds that have the characteristic formula R O R´ . They can be regarded as substitution products of water. Esters are compounds formed by the reaction of acids and alcohols, and they correspond to salts in inorganic chemistry. The general formula of an ester is R COO R´ . A typical reaction for the hydrolysis formation of an ester is as follows: RCO

OH ‡ H

OR´ → H2 O ‡ RCOOR´

4.4.2 Nitrogen-containing Compounds Amines are derivatives of ammonia, NH3. Primary, secondary, and tertiary amines have one, two, or three of the hydrogen atoms, respectively, substituted with alkyl groups. Tertiary amines combine with alkyl halides to form quaternary ammonium salts. These salts have bactericidal properties that make them useful disinfecting agents. An amide is another functional group consisting of a primary amine–carbonyl combination: (C O) NH2. It is a reactive group of interest to synthetic organic chemists. Amino acids have at least one amine ( NH2) and one carboxylic group per molecule. Because of the presence of the amino group and the carboxylic acid group, these molecules have both acidic and basic properties. The general structural formula for an amino acid is shown in Figure 4.3. The hydrolysis of two amino acids, which involves the reaction of the carboxylic group on one acid with the amine group on the other with a molecule of water leaving, results in the formation

83

84

Theory and Practice of Water and Wastewater Treatment

Figure 4.3 Amino acid.

Figure 4.4 An isomer of cysteine.

of a peptide bond. Amino acids can thus be chained together in an infinite number of ways to form proteins. The 20 or so amino acids that are essential to life have both carboxylic and amino groups bonded to the same carbon atom. The formula for glutamic acid, which is an amino acid, is

Some amino acids, such as cysteine and methionine, contain sulfur. One of the isomers of cysteine is shown in Figure 4.4.

4.5 Aromatic Compounds Aromatic compounds contain the benzene ring (Figure 4.5). Many of the compounds containing the benzene ring exhibit a characteristic odor, which is why this class of compounds is termed aromatic. Benzene is a compound that exhibits resonance, which is a very stable configuration because of the sharing of the electrons in the bonds among the six participating carbon atoms. These compounds are difficult to degrade because of this property. Resonance in the benzene ring is often indicated by drawing a circle within the ring. A carbon atom in the ring may have an OH group attached to it, in which case the compound is referred to as a phenol; an OOH group, which makes the ring an acid; or a CHO or (CO) R group, which makes the compound an aldehyde or a ketone, respectively. It is possible to attach carbon chains to benzene rings, and polyring aromatic hydrocarbons exist. There are many naturally occurring aromatics as well as synthetic aromatics.

Figure 4.5 Benzene.

4 Organic and Biochemistry

Figure 4.6 Pentachlorophenol (PCP).

Figure 4.7 Sulfuric-acid-based organic compounds. (a) Diphenyl sulfone. (b) Methane sulfonic acid.

Chlorinated benzene derivatives are often quite toxic. Pentachlorophenol (PCP), shown in Figure 4.6, was one of the most widely used biocides in the United States; it was mainly used as a wood preservative, but it is currently a restricted use pesticide and is no longer available to the general public. Pentachlorophenol is extremely toxic to humans primarily because of its high number of chlorine atoms. 4.5.1 Compounds of Sulfur Similarly to ammonia, sulfuric acid may have either one or both H and OH atoms substituted with organic chains to form various compounds. Two examples of these compounds are diphenyl sulfone and methane sulfonic acid, with the structures shown in Figure 4.7. A more interesting group of compounds is the derivatives of H2S. These compounds are also known as mercaptans. Like hydrogen sulfide, mercaptans have disagreeable odors. The odors from many of these compounds are noticeable at very low concentrations. An example of a mercaptan is diethyl sulfide, which has the formula C2H5SC2H5.

4.6 Naturally Occurring Organic Compounds All living organisms from the smallest to the largest require organic compounds for their metabolism. There are three major classes of naturally occurring organic compounds: carbohy­ drates, proteins, and fats and oils. 4.6.1 Carbohydrates Carbohydrates contain C, H, and O, exclusively. Their general formula is (CH2O)n. The simplest carbohydrates or sugars are monosaccharides, of which glucose, C6H12O6, (Figure 4.8) is a typi­ cal example. Disaccharides contain two monosaccharides bonded together (common household

Figure 4.8 Glucose.

85

86

Theory and Practice of Water and Wastewater Treatment

sugar, sucrose, is a disaccharide of glucose and fructose) and polysaccharides are chains of monosaccharides. Two main classes of polysaccharides are energy storage compounds and structural compounds. Energy storage compounds are readily degraded. They are used by the cell in energy-demanding situations. Starch is an example of an energy storage carbohydrate. Structural compounds serve to maintain the integrity of the cell and as such they are resistant to degradation. Cellulose is a good example of a structural carbohydrate. 4.6.2 Proteins Proteins are complex molecules with no general formula that contain C, H, O, N, P, and S along with trace occurrences of other elements. N, P, and S are contained in lesser amounts than C, H, and O on a molar basis. The building blocks of proteins are amino acids (see Figure 4.3); chains shorter than 30 amino acids are commonly referred to as peptides. There are 20 specific amino acids that form the building blocks of proteins; other amino acids can perform other functions. Amino acids have acidic and basic properties and may be polar or nonpolar, among other characteristics. The properties of amino acids and their sequence in a protein give the protein four recognized levels of structure (primary, secondary, etc.), which is critical to their functioning. The biological functions of proteins are multivariate. The main classes of proteins are enzymes, structural compounds, and transport compounds. Enzymes catalyze reactions, accelerating them 10–100 000 times or more. Structural elements keep the cell intact. Transport proteins serve to move compounds within the cell as well as between the cell and its environment. In metabolism, proteins are broken down into amino acids, which can be resynthesized into proteins, but the pathways for each process are not the reverse of the other. 4.6.3 Fats and Oils Fats and oils have the general formula CnH2n+1COOH. Lighter weight compounds tend to be oily and higher weight compounds more wax-like. Many of these compounds are sparingly soluble in water, which can be a treatment problem because microorganisms lack the ability to secrete emulsifying agents to keep the fatty acids in suspension. Many are odoriferous. Formulas for two typical fatty acids found in municipal wastewaters are given in Figure 4.9. Fatty acids play an important role in metabolism as energy-rich fuels with more energy per unit weight than carbohydrates. Lipids are biological fats that are sparingly soluble in water; thus, they are extracted with nonpolar organic solvents such as ether, benzene, or chloroform. Many lipids are esters of fatty acids and various alcohols. Short-chain fatty acids, acetic, propionic, and butyric, among others are volatile (acetic acid is the primary scent of vinegar), thus known as volatile fatty acids. They are important inter­ mediates in anaerobic wastewater treatment.

4.7 Biochemistry Given the diversity of species, there is a remarkable similarity among the metabolic machineries of all life forms, from the smallest to the largest. Metabolism is the sum total of all chemical

Figure 4.9 Typical fatty acids.

4 Organic and Biochemistry

reactions within the organism. These reactions proceed in accordance with thermodynamic principles. All heterotrophic organisms ultimately obtain their energy from oxidation–reduction reactions. Metabolism is broken down into processes to obtain energy and processes that use this energy to synthesize new cell components and perform other functions. Dissimilation or catabolism is the breakdown of energy-rich (reduced) compounds to obtain energy. Assimilation, anabolism, or synthesis is the fabrication of complex molecules from simpler ones. Aerobic organisms use oxygen as the ultimate electron acceptor in catabolism. Respiration is oxidation of fuels with molecular oxygen. Anaerobic organisms do not use oxygen and essentially break organic molecules into a component that is oxidized and another component that becomes reduced while the organism derives some energy from the process. The extraction of chemical energy from substrates in the absence of molecular oxygen is known as fermentation. Metabolic processes occur through a series of steps depending on the presence of enzymes and the substance to be degraded or fabricated. The series of steps is known as a metabolic pathway; these pathways are rather complex. To obtain an appreciation of metabolic pathways, two of the major pathways are presented summarily here. There are many others. Before the pathways are examined, two important energy carrier groups or classes must be described. The energy obtained from oxidation is stored in energy carriers that release energy in synthesis reactions. Adenosine triphosphate (ATP) is the primary energy carrier in a cell. As adenosine monophosphate (AMP) adds successively one (forming adenosine diphosphate, ADP) and another inorganic phosphate group, the bonds holding these phosphate groups become more energy rich. ADP–ATP transformations are involved in oxidation–reduction reactions, the energy-rich ATP being formed in oxidation reactions (catabolism) and the lower energy ADP being formed with the release of inorganic phosphorus in reduction (synthesis) reactions. ATP is representative of a wide variety of energy carriers. There is no electron transfer involved in the formation of ATP, ADP, or AMP. The ATP content of microorganisms is variable (Levin et al. 1975), but for both aerobic and anaerobic microorganisms that are found in sewage treatment processes, ATP content of viable cells is in the range of 1–2 μg ATP/mg cell (Weddle and Jenkins 1971; Chung and Neethling 1990). Another important class of compounds involved as energy carriers in metabolism is dehydrogenase enzymes. Nicotinamide adenine dinucleotide (NAD) is a common electron carrier representative of those that are associated with enzymes of this class. The oxidized form of NAD is symbolized as NAD and the reduced form is symbolized as NADH2. Also, NADPH2 (nicotinamide adenine dinucleotide phosphate) and FADH2 (flavin adenine dinucleotide) are reduced forms of dehydrogenase enzymes. Their roles are essentially the same as NADH2. The pairs NAD/NADH2, NADP/NADPH2, and FAD/FADH2 have essentially the same roles in metabolism. These compounds participate in reactions by being reversibly reduced and oxidized in transferring electrons from the substrate to oxygen or the ultimate electron acceptor in a generally long series of steps. The concentrations of ATP and dehydrogenase enzymes can be used as a measure of activity of cultures.

4.8 Glycolysis Glycolysis is a fermentation process whereby glucose is broken down into lactic acid. The ability to use this pathway (also known as the Embden–Meyerhof–Parnas pathway) exists in, of course, anaerobic organisms and many aerobic organisms, which may use this pathway in preparation for aerobic completion of the oxidation of glucose. The pathway is shown in Figure 4.10.

87

88

Theory and Practice of Water and Wastewater Treatment

Figure 4.10 Diagram of simplified glycolysis pathway.

Glycolysis is catalyzed by the action of 11 enzymes. It is seen from the figure that the first step, in which phosphate is added to glucose (phosphorylation), requires an initial expenditure of energy by the use of one ATP molecule. The net release of ATP from the process is two ATP molecules per molecule of glucose. As noted on the figure, other monosaccharides can also enter the cycle. Other fermentations proceed in a similar manner, although the final products vary. For instance, ethanol is a major product of organisms used in the brewery industry. The change occurs in the last step, where pyruvate is converted into acetaldehyde, which is then converted into ethanol. The free energy change associated with the conversion of glucose to two lactate molecules is 47.0 kcal mol 1.

4.9 The Tricarboxylic Acid Cycle Glycolysis releases only a small amount of the chemical energy contained within glucose. Complete oxidation of glucose to CO2 and H2O magnifies the yield of energy considerably. The theoretical free energy change for complete oxidation of one molecule of glucose to CO2 and H2O is 686.0 kcal mol 1. The TCA cycle (Figure 4.11) is sometimes referred to as the Krebs cycle after the man who first postulated its existence. It is almost universally present in aerobic organisms and is a primary pathway. In catabolism, the cycle proceeds clockwise as it is represented in Figure 4.11.

Figure 4.11 Diagram of simplified TCA cycle.

4 Organic and Biochemistry

In an aerobic organism, the NADH2 formed in glycolysis can ultimately transfer its electrons to oxygen and form six molecules of ATP in addition to the two molecules of ATP formed from the original glucose molecule. Thus, glycolysis will yield a net gain of eight ATP molecules in aerobic organisms. The pyruvate formed in glycolysis enters the TCA cycle. As shown in the figure, GTP (guanosine triphosphate) is formed in the TCA cycle. This compound is similar to ATP. Also, NADPH2 and FADH2 are formed. Their roles are essentially the same as NADH2. Because 1 mol of glucose yields 2 mol of pyruvate and 6 mol of ATP and 1 mol of pyruvate yields 15 mol of ATP, the aerobic bio-oxidation of 1 mol of glucose to carbon dioxide and water yields 36 mol of ATP. About 73% of the theoretical energy yield of oxidation of glucose is captured in ATP. There are 10 enzymes involved in the TCA cycle and other enzymes are involved in the ultimate transport of electrons to oxygen.

4.10 Enzyme Kinetics Enzymes catalyze reactions by forming an intermediate compound with the reactants (sub­ strate). The intermediate compound is metastable and normally breaks down to yield the unchanged enzyme and the products of the overall reaction. As with all chemical reactions, each step is reversible (the principle of microscopic reversibility). The reaction sequence can be represented as follows. (4.1) where E is enzyme, ES is the intermediate enzyme-substrate complex, ki are rate constants, P is product, and S is substrate. If the reaction is proceeding at a uniform rate, the rates of formation and destruction of the intermediate are equal, although the concentrations of reactants and products are continu­ ously changing. This situation often occurs, particularly when the enzyme concentration is low compared to the concentrations of the reactants. Assuming that normal first-order reactions with respect to each entity apply, the following equation describes the balance for [ES]. d‰ESŠ ˆ k 2 ‰EŠ‰SŠ dt

k 1 ‰ESŠ ‡ k 3 ‰EŠ‰PŠ

k 4 ‰ESŠ ˆ 0

(4.2)

where t is time. The rate of formation of ES from product and enzyme is normally negligible, i.e., k3  0. The overall velocity (or rate) of the reaction depends on the rate of product formation: vˆ

d ‰ PŠ ˆ k 4 ‰ESŠ dt

(4.3)

where v is the overall velocity of the reaction. The following mass balance also applies at any time. ‰EŠ0 ˆ ‰EŠ ‡ ‰ESŠ

(4.4)

where [E]0 is the initial concentration of enzyme. It is difficult to measure the amount of ES present for use in Eq. (4.4); therefore, Eqs. (4.2) and (4.4) are used to develop a formula in terms of more readily assessed parameters (of [S] or [P]

89

90

Theory and Practice of Water and Wastewater Treatment

and t). The following expression for [ES] can be derived. ‰ESŠ ˆ

k 2 ‰EŠ0 ‰ SŠ ˆ k 2 ‰SŠ ‡ k 1 ‡ k 4

‰ EŠ 0 ‰ SŠ k1 ‡ k4 ‰SŠ ‡ k2

(4.5)

Determination of the individual rate constants involved is difficult, and an overall constant, K, known as the Michaelis–Menten constant (they made the original development) or half-velocity constant (see Problem 16), is defined as Kˆ

k1 ‡ k4 k2

(4.6)

Substituting Eqs. (4.5) and (4.6) into Eq. (4.3) yields vˆ

k 4 ‰ EŠ 0 ‰ SŠ K ‡ ‰SŠ

(4.7)

The maximum rate of reaction (vmax) will occur when [ES] = [E]0. Therefore, v ˆ k 4 ‰EŠ0 ˆ vmax

(4.8)

Substituting Eq. (4.8) into Eq. (4.7) results in vˆ

vmax ‰SŠ K ‡ ‰SŠ

(4.9)

The hyperbolic equation Eq. (4.9) is known as the Michaelis–Menten equation. It was originally developed by them to describe enzyme kinetics but has been found to apply to many other types of reactions. The equation is plotted in Figure 4.12. It is observed from the plot that when substrate concentration is high, the velocity of reaction is zero order; when it is low the velocity is first order. In between, the reaction velocity has an order between zero and one. The rate of product formation is related to the rate of substrate removal and the stoichiometry of the overall reaction. The constant relating moles (or mass) of substrate is absorbed into vmax. If mass, m, of substrate is required to form one mass unit of product, d‰PŠ v´max ‰SŠ d ‰SŠ ˆ ˆ m dt K ‡ ‰S Š dt

or

v´max =m ‰SŠ ˆ K ‡ ‰SŠ

d ‰ SŠ ˆ dt

vmax ‰SŠ K ‡ ‰SŠ

(4.10)

where [P] and [S] are in mass or molar concentration units and vmax = v´ max/m. Example 4.1 Application of the Michaelis–Menten Equation What is the time for 90% completion of a reaction described by the Michaelis–Menten equation when the initial substrate concentration is 32 mg L 1, vmax = 4.3 mg (L h) 1, and K = 1.5 mg L 1? The concentration at 90% completion will be 0.10(32 mg L 1) = 3.2 mg L 1. Integrating Eq. (4.10), ‰SŠ

K

tˆ

‰SŠ0

d‰SŠ ‡ ‰SŠ

‰SŠ ‰SŠ0

K ln ‰SŠ=‰SŠ0 ‡ ‰SŠ vmax

t

d‰SŠ ˆ vmax

dt ) K ln

0

‰SŠ0

ˆ

1:5 mg L

1

‰S Š ‡ ‰SŠ ‰SŠ0

‰SŠ0 ˆ vmax t

ln …0:10† ‡ 3:2 mg L 4:3 mg …L-h†

1

1

32 mg L

1

ˆ 7:5 h

4 Organic and Biochemistry

Figure 4.12 The Michaelis–Menten equation.

Figure 4.13 Eadie–Hofstee plot.

The two constants in the equation, K and vmax, may be readily determined when reaction velocity and substrate concentration data are available. A common method for determining the constants is to invert the equation and plot 1/v against 1/[S]. This plot is known as a Lineweaver– Burk plot. 1 1 K 1 ˆ ‡ v vmax vmax ‰SŠ

(4.11)

The slope and intercept define the constants. A Lineweaver–Burk plot weights data at lower substrate concentrations more than data taken at high substrate concentrations. Further­ more, lower substrate concentrations are determined with less accuracy than high substrate concentrations. There are other methods of algebraically manipulating Eq. (4.9) to obtain vmax and K from the measured data. Rearranging Eq. (4.9) into the form of Eq. (4.12) and plotting v against v/[S] results in an Eadie–Hofstee plot (Figure 4.13). The Eadie–Hofstee plot magnifies departures from linearity that may not be apparent in a double reciprocal (Lineweaver–Burk) plot. vˆ K

v ‡ vmax ‰SŠ

(4.12)

Inhibitory substances limit the action of enzymes by binding with a substrate or the enzyme. It can be demonstrated that inhibitory substances effectively increase the half-velocity constant. Toxic substances destroy the enzyme. It can be seen from Eqs. (4.7) and (4.8) that a decrease in [E]0 decreases the maximum velocity of the reaction.

Questions and Problems 4.1 Compare the amount of oxygen required to oxidize the alkanes of methane (CH4) and ethane (C2H6) completely in terms of g O2/g of compound and g O2/g C. 4.2 Define isomerism and illustrate with structural formulas for the compounds with molecular formula C6H14.

91

92

Theory and Practice of Water and Wastewater Treatment

4.3 Is the following compound another isomer of dichloroethane in addition to those depicted in Figure 4.2?

4.4 What are the major functional groups and how do they affect organic molecules? What are the general chemical formulas for alcohols, aldehydes, ketones, esters, and ethers? 4.5 In general, how do organic compounds differ from inorganic compounds? 4.6 Why does the carboxyl group behave as an acid? 4.7 Write the chemical reaction for the formation of an ether from ethanol (C2H5OH) and methanol (CH3OH). 4.8 What are the three major classes of naturally occurring organics and general character­ istics of compounds in each group? 4.9 What is necessary to saturate oleic acid (Figure 4.9)? Write the formula of the saturated compound. 4.10 The generic chemical formulas for carbohydrates and fatty acids are (CH2O)n and CnH2n+1COOH, respectively. Consult a reference on organic chemistry and list three or four carbohydrates and three or four fatty acids (not found in this chapter) and their chemical formulas to prove that these formulas are generally correct. 4.11 Why can aerobic microorganisms extract more energy than anaerobic microorganisms from metabolism of a given compound? 4.12 Write the chemical equations for the oxidation of glucose to pyruvate (C3 H3 O3 ) and for the oxidation of pyruvate to CO2 and H2O with oxygen as the oxidant. Combine these two equations to find the overall equation for the oxidation of glucose with oxygen. 4.13 What are the characteristics of an aromatic compound? 4.14 Write the structural formula for phenol. Why is it difficult to degrade? Is the compound acidic or basic? 4.15 What are roles of ATP and NAD in metabolism? 4.16 In the Michaelis–Menten equation, what is the reaction velocity in terms of vmax when K = [S]? 4.17 For the Michaelis–Menten equation, how long will it take for 95% conversion of a substrate with an initial concentration of 280 mg L 1, if vmax = 1.45 mg (L-h) 1 and K = 165 mg L 1?

4 Organic and Biochemistry

4.18 For vmax = 2.08 mg (L h) 1 and K = 45 mg L 1, plot the reaction velocity as a function of substrate concentration up to [S] = 500 mg L 1. At what substrate concentration does the Michaelis–Menten reaction velocity differ by 10% from the first-order and zero-order reaction velocities that apply when the substrate concentrations are low and high, respectively? 4.19 Graphically compare reaction velocity as a function of substrate concentration for a vmax of 5.0 mg (L d) 1 and K values of 10, 100, and 1000 mg L 1. Make the plots for substrate concentrations up to 5000 mg L 1. At what substrate concentrations, does the reaction become essentially first order (90% of the first-order velocity) for each K value? 4.20 What are vmax and K for the following data? Use both Lineweaver–Burk and Eadie– Hofstee methods to determine the constants. S × 102 (mol L 1) v × 10 (mol (L-min) ) 4

1

1.0

0.80

0.60

0.40

0.20

6.4

5.8

4.79

3.80

2.19

4.21 Could an Eadie–Hofstee or Lineweaver–Burk approach be used to evaluate the coef­ ficients for a reaction model of v ˆ kS 2 =…K ‡ S 2 †? 4.22 In the Michaelis-Menten equation would you expect vmax only or both vmax and K to be affected by a temperature change?

References Chung, Y.C. and Neethling, J.B. (1990). Viability of anaerobic digester sludge. J. Environ. Eng. 116 (2): 330–342. doi: 10.1061/(ASCE)0733-9372(1990)116:2(330). Levin, G.V., Schrot, J.R., and Hess, W.C. (1975). Methodology for application of adenosine triphosphate determination in wastewater treatment. Environ. Sci. Technol. 9 (10): 961–965. doi: 10.1021/es60108a011. Sawyer, C.N., McCarty, P.L., and Parkin, G.F. (2003). Chemistry for Environmental Engineering and Science, 5e. New York: McGraw-Hill. Weddle, C.L. and Jenkins, D. (1971). The viability and activity of activated sludge. Water Res. 5 (8): 621–640. doi: 10.1016/0043-1354(71)90117-5.

93

95

5 Analyses and Constituents in Water Analysis of a water for its constituents and characteristics allows us to apply and exploit chemical phenomena as well as other procedures to evaluate the state of the water vis-à-vis its health aspects, use as a resource, and role in the ecosystem. This chapter covers additional theory and techniques that have not been discussed in previous chapters to arrive at these objectives. The cookbook for environmental engineers is Standard Methods for the Examination of Water and Wastewater (APHA et al. 2012) (hereafter referred to as Standard Methods). Detailed procedures for most of the analytical procedures (physical, chemical, biological, and others) of interest to the environmental engineer are contained in this reference, which is compiled in consultation with experts around the world. This book and the online version are revised regularly to remain current. Brief background information and qualifying remarks are given with each procedure. As well, the precision and accuracy of the procedures are given based on the results from many qualified laboratories. Use of this reference ensures that consistent techniques are applied and that the analyst is aware of the validity and shortcomings of their results and those reported by anyone using the prescribed techniques.

5.1 Titration Acid–base titrations have been discussed in Chapter 3. Complex and redox titrations rely on similar principles. 5.1.1 Complex and Precipitate Formation Titrations The formation of precipitates or complexes removes the free ion from solution. An indicator sensitive to the free ion signals its presence or absence and allows for easy determination of the species. The agent causing the formation of the complex or precipitate must have a stronger affinity for the species of interest than other competing complexing agents. An examination of solubility product constants (Section 1.11) shows which reactions may be feasible for this type of analysis. A common analysis for chloride (argentometric method) uses silver nitrate as a titration agent. White silver chloride precipitates. In this case, the indicator used is the chromate ion, which forms a reddish brown precipitate (Ag2CrO4). Silver chloride is less soluble than silver chromate (see Problem 1); therefore, the chloride is precipitated first. After the chloride is removed, further addition of silver forms the highly visible silver chromate precipitate. Because there is some consumption of titrant by the indicator, a blank containing

Theory and Practice of Water and Wastewater Treatment, Second Edition. Ronald L. Droste and Ronald L. Gehr.

 2019 John Wiley & Sons Inc. Published 2019 by John Wiley & Sons Inc.

Companion website: www.wiley.com/go/droste/water

96

Theory and Practice of Water and Wastewater Treatment

Figure 5.1 Ethylenediaminetetraacetic acid.

only the indicator is titrated and the volume of titrant used for the blank is subtracted from the volume of titrant used for the sample. Complex formation titrations can be used to determine the concentration of either a metal or a ligand such as chloride or cyanide. The most common complex formation reagent is ethyl­ enediaminetetraacetic acid (EDTA) (Figure 5.1). This organic ligand is able to form up to six bonds with a metal ion. Ligands that form multiple bonds with a metal are known as chelating agents (from the Greek chela, meaning claw). EDTA combines with metal ions on a 1 : 1 ratio regardless of the charge on the metal ion. The volumetric calcium and total hardness procedures in APHA et al. (2012) use EDTA as the titrant. The indicators used in these titrations are metallochromic indicators, which are themselves chelating agents that form a colored complex with a metal ion. EDTA is a stronger complexing agent than the indicator. The first additions of EDTA will sequester metal ions that are not complexed with the indicator. As the concentration of uncomplexed metals decreases with further addition of EDTA, the indicator begins to release metal ions. When EDTA has sequestered the metal ions from the indicator, the indicator changes color, signaling the endpoint of the titration. 5.1.2 Redox Titrations and Potentiometric Analyses The principles for a redox titration are similar to an acid–base titration except that the potential of the solution is changing with the addition of titrant, which is an oxidant or reductant. The solution potential changes after the addition of titrant and reaction of it with the species in solution. A plot of system potential versus amount of titrant added will produce curves similar to those in Figures 3.1–3.3. There will be a large change in system potential around the equivalence point. The analysis of a redox titration will be presented by an example. Recall that a 1 N solution has the potential to give up or take up 1 electron equivalent. Example 5.1 Potential Variation during Titration and Equivalence Potential Consider the titration of Ce4+ with Fe2+. The initial concentrations in a volume of 50 mL are [Ce4+] = 2.70 × 10 3 M and [Ce3+] = 1.50 × 10 4 M. The solution is being titrated with 0.020 N Fe2+. Using data from Table 1.3, the half-reactions and their potentials are Ce4‡ ‡ e ˆ Ce3‡ Fe

3‡

‡ e ˆ Fe

2‡

E ° ˆ 1:72 °

E ˆ 0:771

(i) (ii)

The Nernst equations for these reactions are, respectively, E Ce ˆ E oCe

0:059log

Ce3‡ Ce4‡

(iii)

E Fe ˆ E oFe

0:059log

Fe2‡ Fe3‡

(iv)

5 Analyses and Constituents in Water

The overall chemical equation is Fe2‡ ‡ Ce4‡ → Fe3‡ ‡ Ce3‡

(v)

This reaction reaches equilibrium almost immediately upon addition of titrant. As Fe2+ is added to a solution containing Ce4+, the system potential decreases. Before the equivalence point is reached, the concentration of Fe2+ in solution becomes exceedingly small. However, the concentrations of Ce4+ and Ce3+ are significant. If the initial amount of Ce4+ in solution is known, the system potential can be determined using Eqs. (iii) and (v). After the equivalence point is passed, the concentration of Fe2+ becomes significant, and Eqs. (iv) and (v) can be used to determine the system potential. For instance, after 5.0 mL of titrant has been added (do not forget to adjust the total volume of the solution): Ce4‡ ˆ 6:36  10

4

Ce3‡ ˆ 1:50  10 From Eq. (iii): E Ce ˆ 1:72

M 4

Fe3‡ ˆ 1:82  10

3

M

…0:050=0:055† ‡ 1:82  10 1:96  10 0:059 log 6:36  10

3

ˆ 1:96  10

3

M

3

4 ˆ

1:69 V

Because ECe = EFe = Esys, Eq. (iv) can be used to find [Fe2+].

Fe2‡ ˆ Fe3‡ 10

E oFe E sys 0:059

ˆ 1:82  10

3

10

0:771 1:69 0:059

ˆ 4:83  10

19

M

The assumption that [Fe2+] is small is correct. At equivalence, the concentration of Fe2+ is not exceedingly small because there is not an excess of Ce4+ ions. To avoid an iterative solution of the mass balance and potential equations, a slightly different approach is taken. The system potential can be determined by adding Eqs. (iii) and (iv) and considering the mass relations that exist at the equivalence point. o 2E eq ˆ E oCe ‡ E Fe

0:059log

Ce3‡ Fe2‡ Ce4‡ Fe3‡

(vi)

where Eeq is the system potential at the equivalence point. The volume of titrant added at equivalence is determined from N t V t ˆ N Ce V 0 : where Nt is the normality of the titrant, NCe is the normality of the solution, and Vt and V0 are the volume of titrant and initial volume of the solution, respectively. V t ˆ N Ce V 0 =N t ˆ 2:70  10

3

eq=L …0:050 L†=…0:020 eq=L† ˆ 0:006 75 L ˆ 6:75 mL

From the stoichiometry of the reaction and at the equivalence point, the amount of Fe2+ added must be equal to the amount of Ce4+ that was initially present. Fe2‡ ‡ Fe3‡ V eq ˆ Ce4‡ 0 V 0 where Veq is the volume of the solution at equivalence. Also at equivalence, the number of equivalents of Fe2+ that are present will be equal to the number of unreacted equivalents of Ce4+. Fe2‡ V eq ˆ Ce4‡ V eq

(vii)

97

98

Theory and Practice of Water and Wastewater Treatment

From the overall reaction, the equivalents of Ce3+ produced are equal to the equivalents of Fe3+ that are present. [Ce3+] is referred to the volume of the solution at equivalence. Fe3‡ V eq ˆ Δ Ce3‡ V eq

(viii)

The total concentration of Ce3+ at the equivalence point depends on the initial concentration and the amount of Ce3+ produced. Ce3‡ V eq ˆ Ce3‡

0

V 0 ‡ Δ Ce3‡ V eq

(ix)

Substituting the above relations into Eq. (vi) and reducing the equation,

o 2E eq ˆ E oCe ‡ E Fe

0:059log

Ce3‡ Fe2‡ o ˆ E oCe ‡ E Fe Ce4‡ Fe3‡

Ce3‡ 0 V 0 ‡ Fe3‡ V eq V eq 0:059log Fe3‡ (x)

4+

2+

3+

3+

It is also the case that [Ce ] and [Fe ] are very small and [Ce ] and [Fe ] comprise most of the Ce and Fe species present at the equivalence point. Therefore, it will be assumed that Fe3‡ ˆ 0:02 mol L

…0:006 75 L†=…0:056 75 L† ˆ 0:002 379 M:

1

Substituting the values into the above equation: 1:5  10 2E eq ˆ 1:72 ‡ 0:771

0:059log

E eq ˆ 2:49=2 ˆ 1:25 V:

4

…0:050† ‡ …0:002 379†…0:056 75† 0:056 75 ˆ 2:49 V 0:002 379

Check to see that the assumptions are correct. Use Eqs. (iii) and (iv) to verify that [Ce4+] and [Fe2+] are negligible compared to their counterparts.

5.1.3 Indicators for Potentiometric Analysis The potential of the solution being titrated can be monitored directly or redox indicators may be used to indicate the endpoint of a redox titration. A redox indicator will exhibit a color change in the solution at the potential inflection range for the titration. A general redox equation for the indicator is In ‡ ne ⇆ Inn Oxidized color Reduced color

(5.1)

with a corresponding Nernst equation of E In ˆ E oIn

0:059 ‰Inn Š log n ‰InŠ

(5.2)

As in the case of an acid–base indicator, typically a concentration ratio of 1 : 10 (log = 1) for the oxidized and reduced species is required to produce a visible color change. Following the same approach for an acid–base indicator (Section 3.6.3), the EIn range can be determined to be 0:059 E In range ˆ E oIn  (5.3) n Bishop (1972) and Ottaway (1972) give thorough discussions of redox indicators.

5 Analyses and Constituents in Water

5.2 Colorimetric Analyses Colorimetric analysis depends on the formation of a colored species by the substance in question. Many different substances can be analyzed by this method. Principles of light transmittance need to be understood to relate light intensity to the amount of the colorproducing substance. 5.2.1 The Beer–Lambert Laws for Light Transmittance Two basic laws govern the phenomenon of light transmittance: Lambert’s and Beer’s laws. The intensity of light is reduced by absorption by the medium. In 1760, Lambert published his law stating that each layer of medium reduces the passage of light proportionately. This results in a first-order law: Tˆ

I ˆ 10 I0

kl

(5.4a)

where I is intensity of light at any distance from the surface, I0 is intensity of light at the surface, k is absorptivity, l is length of absorbing medium, and T is transmittance. Absorbance, A, is defined as A ˆ log …I=I 0 † ˆ log T

(5.4b)

The intensity of light is also reduced proportionately by the concentration of the absorbing species in the medium, as demonstrated by Beer in 1852, hence: Tˆ

I ˆ 10 I0

k´C

(5.5)

where C is concentration and k ´ is molar absorptivity. The above laws may be combined to give the Beer–Lambert law. Tˆ

I ˆ 10 I0

k ´´ lC

(5.6)

where k´´ is an absorption coefficient.

5.3 Physical Analyses The characterization of solids is one of the most common assessments of water quality. Turbidity is associated with suspended solids (SS) concentrations. 5.3.1 Solids Solids in water fall into one of the following categories. 1) Dissolved 2) Colloidal 3) Suspended. Dissolved solids are considered truly in solution and pass through a standard glass-fiber filter. The solution consisting of the dissolved components and water is homogeneous, forming a single phase.

99

100

Theory and Practice of Water and Wastewater Treatment

Figure 5.2 Colloid formed from soap molecules.

Figure 5.3 Size ranges of solids.

Colloidal solids are uniformly dispersed in solution, but they form a solid phase that is distinct from the water phase. Colloidal solutions are termed sols. Colloidal particles are charged entities that derive their stability primarily from their charge characteristics. An example is the soap molecule shown in Figure 5.2. Soap is the sodium salt of a fatty acid derived from animal fat. One end of the molecule is ionic and the other end of the hydrocarbon chain is nonpolar. In a water solution, the soap molecules dissociate. The nonpolar ends of the molecule group together with the polar ends extending into the polar water solution. A negatively charged particle surrounded by positively charged ions results (see the inset in Figure 5.2). This type of colloid is known as a micelle. SS are also a separate phase from the solution. Some SS are classified as settleable solids. Settleable solids are determined by placing a sample in a cylinder and measuring the amount of solids that have settled after a set amount of time. The time is usually 30 minutes or 1 hour. The size of solids increases from dissolved solids to SS (Figure 5.3). Water can be removed from a sample by heating it to a temperature near 103 °C, which is slightly above the boiling point of water and low enough to prevent organic compounds from burning. Volatile solids are primarily organic solids that will burn in the presence of oxygen. A high temperature of 550 °C will cause complete combustion of the organic matter, but some inorganic solids may also be lost at these temperatures, e.g., Ca…HCO3 †2 …NH4 †2 CO3

Δ

! CaO ‡ 2CO2 …g† ‡ H2 O…g†

Δ

! 2NH3 …g† ‡ CO2 …g† ‡ H2 O…g†

5 Analyses and Constituents in Water

The amount of inorganic solids lost through these reactions is normally insignificant. The solids that remain after combustion will be inert or fixed inorganic solids. Figure 5.4 characterizes the makeup of solids in a water sample. 5.3.2 Turbidity and Color Turbidity and color are important aesthetic parameters for drinking waters. They also influence the biotic community in ecosystems. Turbidity (or cloudiness) is a result of the scattering of light by SS. A rough correlation exists between SS concentration and turbidity. Raleigh’s law (Lord Raleigh (John Strutt), late nine­ teenth century) describes the scattering of white light by suspended particles: Is ∝

V2 n λ4

(5.7)

where Is is the intensity of scattered light, n is the number of particles, V is the volume of particles, and λ is the wavelength of light. From Eq. (5.7), it is observed that size and concentration of particles influence the measure­ ment of turbidity. A natural water or wastewater will contain many different sized particles at different concentrations; the relation between SS concentration and turbidity can be highly variable. However, particularly in water treatment operations where SS concentrations are low, turbidity can be a very useful parameter. The precise terminology for measuring the amount of scattered light is “nephelometry”. Turbidity measurements are reported in nephelometric turbidity units (NTU). Turbidity of a drinking water is always an aesthetic concern. The nature of solids causing turbidity may have other health ramifications. Turbidity in natural waters reduces light transmittance and affects the health of aquatic flora and fauna.

Figure 5.4 Relations among types of solids.

101

102

Theory and Practice of Water and Wastewater Treatment

Color is imparted to a water by dissolved constituents that absorb white light at specific wavelengths. The color of natural waters is often due to humic and fulvic acids. They approximate the copper-orange color of a platinum–cobalt solution produced by dissolving specified amounts of potassium chloroplatinate, K2PtCl6, and cobaltous chloride, CoCl2, in water. The sample of water is visually compared against standards prepared with these agents or against standardized colored disks in a comparator. Spectrophotometric methods may also be used for greater precision. Industrial wastes may contain significant concentrations of compounds that produce any color in the rainbow. In these cases, a series of standards of any compound causing the color may be prepared or a spectrophotometer may be used to assess the amount of color using the tristimulus procedure (APHA et al. 2012).

5.4 Determination of Organic Matter Several ways of quantifying organic matter are available, depending on the nature of the sample and the purpose of the measurement. Volatile solids is a crude measure of organic matter. Dissolved organic matter is most often assessed indirectly in terms of the oxygen required to completely oxidize the organic matter to CO2, H2O, and other oxidized species. Besides expressing the organics in terms of a common denominator, this is practical because an important consequence of the presence of organic matter is consumption of oxygen. The dissolved oxygen (DO) concentration in waters is vital for the maintenance of higher life forms. In natural bodies of water, DO is depleted by the aerobic biological decomposition of organic matter. The theoretical oxygen demand of a compound is calculated by writing a balanced reaction for the compound with oxygen to produce CO2, H2O, and oxidized inorganic components. If the chemical composition of the organic compounds and their concentrations are known, the theoretical oxygen demand can be accurately calculated, but this is an impossible task for most natural waters and wastewaters because of the number of substances that are present. In those cases, analytical techniques can be used, which lump the compounds to yield a single concentration. Another important method of expressing organic matter is in terms of its carbon content. Carbon is the primary constituent of organic matter. Total organic carbon (TOC) can be determined instrumentally. The reactions to determine the amount of oxygen or carbon are based on the average formula weight of all organic compounds considering their respective concentrations in the water. This results in a hybrid organic compound of formula CnHaObNc, where n is the number of moles of organic carbon, and a, b, and c are, respectively, the number of moles of organically bound hydrogen, oxygen, and nitrogen in a liter of water. The coefficients, n, a, b, and c are usually not whole numbers. Example 5.2 Oxygen Demand and Total Organic Carbon Concentrations What are the theoretical oxygen demand and total organic carbon concentration of a water that contains the following components? glucose …C6 H12 O6 †; 150 mg L

1

benzene …C6 H6 †; 15 mg L

The equations for total oxidation of the two constituents are C6 H12 O6 ‡ 6O2 → 6CO2 ‡ 6H2 O

180 192 264 108

C6 H6 ‡ 7:5O2 → 6CO2 ‡ 3H2 O

78 240 264 64

1

5 Analyses and Constituents in Water

From the equations, it is determined that the theoretical oxygen demand for glucose is (192/ 180) = 1.07 mg O2/mg of glucose and the theoretical oxygen demand of benzene is (240/78) = 3.08 mg O2/mg of benzene. The total theoretical oxygen demand of the solution is 1:07  150 mg L

1

‡ 3:08  15 mg L

1

ˆ 206:7 mg O2 =L

The carbon content of glucose is 72 mg C/180 mg glucose = 0.40 mg C/mg glucose, and the carbon content of benzene is 72 mg C/78 mg benzene = 0.92 mg C/mg benzene. The TOC concentration of this solution is 0:40  150 mg L

1

‡ 0:92  15 mg L

1

ˆ 73:8 mg C=L

What is the formula weight of the organic matter in this solution? The concentrations in mg L 1 of each atom in the organic matter are 72 72 ‡ 15  ˆ 73:8 mg L 180 78 12 6 H : 150  ‡ 15  ˆ 11:2 mg L 180 78 96 O : 150  ˆ 80:0 mg L 1 180 C : 150 

1

1

Converting the above into molar concentrations: 1 ˆ 6:15  10 3 M 12 000

1

H : 11:2  ˆ 1:12  10 2 M 1000

1

O : 80:0  ˆ 5:00  10 3 M 16 000 C : 73:8 

The formula for the organic matter is C6.15H11.2O5.0 at a concentration of 1.00 mM. Obviously, this formula does not represent the structure of the organic matter but only the relative amounts of the elements contained in it.

5.4.1 Chemical Oxygen Demand Chemical oxygen demand (COD) is the amount of oxygen required to stabilize organic matter. It is determined by using a strong oxidant. Theoretically, any strong oxidant could be used. Ideally, the oxidant should be able to oxidize any organic compound, it should not present disposal problems, and it should be inexpensive. The first criterion is the most important. Considering these criteria, the oxidant of choice is dichromate (Cr2 O27 ). It has a high equivalent weight, is inexpensive, and participates in a well-defined reaction. The analysis is conducted by adding sulfuric acid (a very strong acid) to the sample along with silver, which is a catalyst, and a known amount of dichromate. The strong acid helps to digest (break down) complex molecules. Mercuric sulfate is also added to complex chlorides that interfere with the test. To accelerate the reaction and achieve the maximum effectiveness of the oxidant, the sample with added reagents is heated in a flask, which is connected to a condensing tube to minimize the loss of volatile organics. Alternatively, closed reflux tubes use smaller volumes of hazardous materials and mitigate disposal problems.

103

104

Theory and Practice of Water and Wastewater Treatment

General Reaction for COD

In the COD reaction, dichromate (Cr2 O27 ) oxidizes organic matter to end products of CO2 and H2O while becoming reduced to Cr3+. The oxidant contains oxygen; also H+ ions will be consumed in the oxidation (see the half-reaction for Cr2 O27 ). From Table 1.3, in a redox reaction the equivalent weight of dichromate is 36.0 g and the equivalent weight of oxygen is 8 g, which are used to calculate the oxygen demand. The general reaction takes the following form: Cn Ha Ob ‡ cCr2 O27 ‡ f H‡ → dCr3‡ ‡ gCO2 ‡ eH2 O

(5.8)

where c, d, f, g, and e are the stoichiometric coefficients required to balance the reaction. Element

LHS

RHS

C:

n

n

H:

a+f

2e

O:

b + 7c

2n + e

Cr:

2c

d

From examination of Eq. (5.8), it is obvious that g = n on the right-hand side (RHS). Using this and making mass balance relations between the left-hand side (LHS) and RHS components, the number of unknowns can be further reduced. From the information in the mass balance table, the equation can be reduced to Cn Ha Ob ‡ cCr2 O27 ‡ …2b ‡ 14c

4n

a†H‡ → 2cCr3‡ ‡ nCO2 ‡ …b ‡ 7c

2n†H2 O

(5.9)

A further reduction in unknowns can be made by considering the charge balance equation. 2c ‡ 2b ‡ 14c

4n

a ˆ …2†…3†c or 6c ˆ 2b ‡ 4n ‡ a

This equation is readily substituted into the coefficients for H+ and H2O. The resulting equation is Cn Ha Ob ‡ cCr2 O27 ‡ 8cH‡ → 2cCr3‡ ‡ nCO2 ‡

a ‡ 8c H2 O 2

(5.10)

The same final result in Eq. (5.10) could have been obtained from an electron balance. The oxidation numbers on C in CnHaOb and CO2 are a 2b Cn Ha Ob : ; CO2 : 4 n The total number of electrons lost by carbon in the oxidation is 4n ‡ a 2b: The number of electrons gained by each mole of dichromate is 6; therefore, the total number of electrons gained by dichromate is 6c. The number of electrons gained is equal to the number of electrons lost, therefore: 6c ˆ 4n ‡ a

2b

This equation can be used to modify the coefficients for H+ and H2O in Eq. (5.9) in the same manner as above. Equation (5.10) will ultimately be obtained. The charge balance equation and electron balance equation are not independent relations; therefore, only one of them provides new information.

5 Analyses and Constituents in Water

Example 5.3 COD Calculation A 10 mL sample diluted to 25 mL with distilled water was digested according to the standard procedure. It was determined that 3.12 × 10 4 mol of dichromate (DC) were consumed by the sample. What is the COD of the sample? Dilution with distilled water does not affect the amount of organic matter present. From the half-reactions (Table 1.3) for DC and oxygen, 1 mol of DC has 6 electron equivalents and 1 mol of O2 has 4 equivalents. The equivalent weight of oxygen is 32.0 g/4 eq = 8.0 g/eq. The COD of the sample is COD ˆ

3:12  10 4 mol DC 6 eq 10 mL mol DC

8 g O2 eq

1000 mL L

1000 mg g

ˆ 1498 mg O2 =L

COD is normally reported in terms of mass O2/L, not in terms of moles DC/L or eq/L.

Interferences with the COD Test

Ammonia can interfere to a small extent with the COD determination if significant amounts of chloride are present and high concentrations of dichromate are used (Kim 1989; APHA et al. 2012). Kim (1989) found that a false COD, equivalent to about 4% of the ammonia, was exerted using a 0.25 N dichromate solution when equimolar amounts of ammonium and chloride were present, even when a large excess of mercuric sulfate was present to complex the chloride. When a 0.025 N solution of dichromate was used, the ammonia was not oxidized. If nitrogen is contained in significant amounts in the organic matter, it will be released as NH‡4 and be subject to minor oxidation if high concentrations of dichromate are used and chloride concentrations are significant. Hydrogen peroxide (H2O2) may be present in industrial wastewaters and will interfere with COD determinations by reducing the amount of dichromate consumption. The measured COD must be corrected by subtracting 0.25[H2O2] (Talinli and Anderson 1992). [H2O2] is in mg L 1 and determined independently. Functionally, the results of a COD determination include whatever substances are oxidized by dichromate under the test conditions. The results may be greater or less than the true theoretical oxygen demand of all organics present in the sample. Not all organic compounds are subject to oxidation by dichromate under the conditions of this test. Aromatic compounds and pyridine (a common biomolecule) are refractory to this analysis. In addition, inorganic reducing agents will result in an inorganic chemical oxygen demand (or false positive COD). Some inorganic reducing agents, Cl and NO2 , can be removed or prevented from oxidation by dichromate (APHA et al. 2012). Furthermore, the test does not distinguish between biodegradable and nonbiodegradable organics. The test is relatively rapid; normally, 2–3 hours of digestion suffice to complete the oxidation. In theory, any strong oxidant will suffice. The limitations of the test regarding inorganic reducing agents and biodegradability of the organic matter will apply to any oxidant. The cost and broad-spectrum oxidizing capabilities of dichromate have favored it over other oxidants. The test is the best practical measure of the oxygen demand of a water.

5.4.2. Biochemical Oxygen Demand Biochemical oxygen demand (BOD) is defined as the amount of oxygen required for the biological decomposition of organic matter under aerobic conditions at a standardized

105

106

Theory and Practice of Water and Wastewater Treatment

temperature and time of incubation. The amount of oxygen required after defined incubation times will depend upon the concentration and nature (biodegradability) of organic matter, temperature, and concentration and type of bacteria (“seed” is often added). See Chapter 19 for a description of an anaerobic “BOD” procedure. The BOD test has many limitations, but the standard procedure is now well established (APHA et al. 2012). It was conceived as a way to mimic, in a bottle, the natural aerobic decomposition process (biodegradation) occurring in a stream. Assuming that the rate of oxidation of organic matter at any instant of time is proportional to the amount of oxidizable matter present, i.e., a first-order reaction, dL ∝ dt

L

or

dL ˆ kL dt

(5.11)

where k is a rate constant and L is the concentration of organic matter expressed as O2. The BOD test is performed in the laboratory by diluting the wastewater sample with water containing sufficient amounts of DO and nutrients and measuring the depletion in DO after a fixed time of incubation at a fixed temperature. Figure 5.5 illustrates transformations as the BOD is satisfied. The amount of BOD exerted, from Eq. (5.11), is dependent on time and the amount of degradable organic matter. The ultimate BOD (sometimes, written as BODu), La, is for most practical purposes a constant, although it has been found by some investigators that La increases with temperature; i.e., increasing temperature promotes the degradation of more-difficult-to­ degrade substances. The ultimate BOD is usually taken to be equal to the COD in the absence of a measurement of the ultimate BOD. A BOD progression curve is shown in Figure 5.6. The amount of BOD exerted at any time is equal to the difference between the BOD existing at the initial time and the BOD remaining at any time, L. x ˆ La

L ˆ La 1

10

k´t

ˆ La 1

e

kt

(5.12)

where x is BOD exerted at any time, t and k´ is a rate constant when log10 units are used. The BOD determination is normally made for a duration of 5 d (“BOD5”). Figure 5.7a,b compare the progression curves and ultimate BOD for samples that have the same BOD5 or the same ultimate BOD, respectively.

Figure 5.5 BOD transformations.

5 Analyses and Constituents in Water

Figure 5.6 BOD progression.

Figure 5.7 (a,b) Comparison of ultimate and 5-d BODs for different rate constants.

BOD exertion is a very complex process. It is dependent on the nature of the seed micro­ organisms and their acclimation to the organic materials in the waste. Environmental factors such as pH, temperature, or the presence of nutrients and other metabolites all influence the rate of BOD exertion. In the BOD procedure, measures are taken to provide controlled environ­ mental conditions to minimize the effects of adverse environmental conditions and to stan­ dardize test conditions. The presence of inhibitory substances can highly influence the results of a BOD determination. Toxic or inhibitory substances can delay the rate of BOD exertion, yielding falsely low BOD5 values, or may totally prevent any BOD exertion. BOD exertion is the sum total of a number of processes qualitatively depicted in Figure 5.8. Organic matter present in the sample is metabolized primarily by bacteria in the early stages. Some of the organics are oxidized, which causes oxygen uptake, and the remainder are transformed into new bacterial cells. As the supply of external substrate becomes scarce, the major source of organics is the bacteria themselves. Some bacteria die through starvation, releasing organic material, which serves as food for the remaining bacteria. If protozoans are

Figure 5.8 Phenomena in a BOD bottle. Adapted from Gaudy (1972).

107

108

Theory and Practice of Water and Wastewater Treatment

present in the seed, they can also contribute to the removal of organic matter, but bacteria are much more efficient in competing for the substrate. However, protozoans will increase because of predation on the increasing population of bacteria. As the food source for the bacteria diminishes to small amounts, protozoans will eventually become dominant. At this stage, bacteria and protozoans continue to feed on living and dead cells. The original organic matter is recycled through a number of organisms, with oxygen being consumed as the organic matter is metabolized for each biomass transformation. There is a succession of microorganisms and an overall decrease in the amount of organic matter in the bottle as the cycle progresses. Oxygen will be utilized for metabolism until it is exhausted or the amount of organic matter decreases to negligible amounts. The overall rate of BOD exertion or oxygen uptake exhibits two phases. The readily metabolized substrate components in the sample will be removed rather rapidly with a concomitant consumption of oxygen. After organic matter is synthesized into bacteria and protozoan cells, the process of organic matter metabolism decreases because the transformed organic matter is not as easily metabolized. The first-order BOD model described above fits a single rate constant over the whole oxygen uptake curve. The rate constant is a weighted average of the rate of oxygen uptake throughout the cycle. Different mixtures of organic components and seed populations can shift the duration and maximum values of individual microbial groups. The procedure for laboratory determination of BOD is described below in the “Laboratory Determination of BOD” section. It usually involves dilution of the wastewater sample, which may reduce the concentrations of toxic or inhibitory substances below the threshold at which they significantly affect microbial activity. This can cause complications in a BOD progression exercise where different dilutions are used. Because of the complexities involved, BOD determinations are sometimes difficult to reproduce. BOD determinations should be interpreted with care. A BOD determination for a period of t days is not meaningful as a measure of the total biodegradable potential of a sample unless the rate constant is known. A BOD progression analysis as below must be performed to determine k. A commonly assumed value for k is 0.1 d 1 (log base 10) at 20 °C, which results in a BOD5:ultimate BOD ratio (x5:La) of 2 : 3. Effects of Temperature on BOD Exertion

Temperature exerts an influence on the rate of oxidation of organic matter; this effect on the rate constant can be adequately described as k T ˆ k 20 θ…T

20†

(5.13)

where kT is the rate constant at a temperature T (°C) and θ is a constant, often called the Arrhenius temperature correction constant. In the absence of laboratory studies on the particular waste being tested, the value of θ is commonly taken to be 1.047. Example 5.4 BOD Exertion An experimenter set a sample for a BOD determination in an incubator at 20 °C. The sample had an ultimate BOD of 330 mg L 1 and the rate constant for this waste was 0.13 d 1 (base 10) at 20 °C. On the beginning of day 3, another person adjusted the temperature of the incubator to 25 °C for another analysis. The incubator temperature changed to the new temperature within 20 minutes. What were the true 5-d BOD of the sample and the 5-d BOD that was determined? Assume θ = 1.047.

5 Analyses and Constituents in Water

The true x5 of the sample was x5 ˆ La 1

5k

10

ˆ 330 mg L

1

1

10

…5 d†…0:13 d

1

† ˆ 0:776 330 mg L

1

ˆ 256 mg L

1

At T = 25 °C, k 25 ˆ k 20 θ…25

20†

ˆ 0:13 d

…1:047†5 ˆ 0:13 d

1

1

…1:26† ˆ 0:16 d

1

After 2 days the concentration of organic matter in the sample is. L2 ˆ La 10

2k 20

ˆ 330 mg L

1

10

…2†…0:13†

ˆ 181 mg L

1

The amount of BOD exerted was 330 181 = 149 mg L 1. The time for the temperature change is small compared to the time scale of the determination and will be ignored. The BOD exerted over days 3–5 is x3−5 ˆ L2 1

10

3k 25

ˆ 181 mg L

1

1

10

…3 d†…0:16 d

1

† ˆ 0:669 181 mg L

1

ˆ 121 mg L

1

The 5-d BOD determined was x2 ‡ x3

5

ˆ 149 ‡ 121 ˆ 270 mg L 1 :

Carbonaceous and Nitrogenous BOD

The primary products of oxidation of organic matter will be CO2, H2O, and NH3. The first-stage or carbonaceous BOD consists of oxidizing organic matter to these products. The bacteria involved in this stage are known as heterotrophs (see Chapter 6). Oxidation of ammonia (in the sample itself, or produced during oxidation of organic matter) to nitrate (nitrification) con­ stitutes a second-stage BOD that occurs simultaneously with the oxidation of carbonaceous BOD (Figure 5.9). Following the concepts introduced in Example 5.1 for oxidation of organic carbon compounds, nitrification can be modeled by the following chemical reactions: 3 Nitrosomonas ! NO2 ‡ H‡ ‡ H2 O NH3 ‡ O2 2 1 Nitrobacter NO2 ‡ O2 ! NO3 2 NH3 ‡ 2O2 → NO3 ‡ H‡ ‡ H2 O

(5.14a) (5.14b) (5.14c)

Two separate genera of bacteria (Nitrosomonas, also known as ammonia oxidizing bacteria and Nitrobacter – nitrite-oxidizing bacteria) are involved in the conversion of ammonia into nitrate as indicated in Eqs. (5.14a) and (5.14b). The reaction in Eq. (5.14b) depends on the rate at which the reaction in Eq. (5.14a) proceeds. Nitrite is generally not thermodynamically stable in natural water environments, and it is readily transformed into nitrate.

Figure 5.9 Evolution of carbonaceous and nitrogenous BOD.

109

110

Theory and Practice of Water and Wastewater Treatment

The organisms responsible for nitrification (known as autotrophs) do not oxidize carbon. They obtain energy from reactions (5.14a) and (5.14b) and use carbon dioxide as a carbon source for cell synthesis. The nitrogenous (or second-stage) BOD, also known as nitroge­ nous oxygen demand (NOD), is readily calculated from the above equations to be 3.76 mg O2/mg NH3. The oxidation of ammonia normally does not become significant until 7 or 8 days from the start of the test because of the slower metabolism of the autotrophic bacteria. This is one reason for limiting a carbonaceous BOD determination to 5 days. Nitrification can be suppressed by the addition of inhibitors followed by the addition of a seed that does not contain nitrifying bacteria. The recommended inhibitor is 2-chloro-6-(trichloro methyl) pyridine (TCMP) (APHA et al. 2012). Nitrogenous BOD kinetics can be modeled by a first-order reaction with the appropriate lag time. Laboratory Methods for Determining BOD

The standard procedure (APHA et al. 2012), which is widely accepted, is to use 60 mL, or preferably 300 mL, airtight glass bottles filled to overflowing with diluted, aerated, and seeded samples, and incubate them in the dark at 20 °C for 5 days. The DO is measured by an electrochemical probe under stirred conditions at time zero and after 5 days. The samples must be diluted to allow for the fact that only 9.2 mg L 1 of DO is available at 20 °C. The dilution solution contains a phosphate buffer to maintain neutral pH, trace elements, nitrification inhibitor, and seed (ideally, raw domestic wastewater, nondisinfected effluent from a biological wastewater treatment plant, or a commercial preparation). The method is quite tedious, takes up significant bench space (as several dilutions need to be tested together with controls), is poorly reproducible (±20%), and the 5-days lag time is unsuitable for real-time monitoring or even informing the public of discharges to water bodies that may affect them. Jouanneau et al. (2014) have reviewed modifications to the standard BOD5 procedure, which have been designed to overcome the abovementioned problems. The following are commercially available modifications: continuous monitoring of DO with an optical probe or a spectro­ photometer, which are noninvasive; manometric procedures to measure DO depletion indirectly; fully automated semirobotic equipment, which eliminates all human intervention once the BOD bottles have been filled. Other developments still at the research stage include biosensors based on bioluminescent bacteria, biosensors with entrapped bacteria, microbial fuel cells, and redox mediators. All these give predictions of BOD5 after a test time as short as 10 minutes; the utility of these shorter-time-lag values must be weighed against the prediction uncertainties. Limitations of the BOD Test for Biological Wastewater Treatment Process Design

The BOD test is a suitable indicator of biologically removable organic matter in a water sample, but it is subject to a number of limitations. It is highly dependent on the care taken to maintain an optimal environment with acclimatized microorganisms, and a progression (see below) should have been run to obtain meaningful results. The test will not reflect rates of organic matter removal in a biological wastewater treatment process, where environmental conditions are significantly different from conditions in a BOD bottle. Wastewater treatment reactors have much higher concentrations of microorganisms, which is the main difference, but temperature and water quality fluctuations will also impact the removal rates. Effluents from wastewater treatment processes are often disinfected prior to discharge to the receiving water to reduce pathogen content. Contrary to expectations, some disinfectants may actually increase the BOD. Absi et al. (1993) observed this effect for ozone, El-Rehaili (1995) for chlorine, and Wagner et al. (2002) for peracetic acid (PAA). The only feasible explanation for the

5 Analyses and Constituents in Water

increase is that any oxidation of organics by chlorine or ozone is offset by these chemicals, which enhance the biodegradability of some refractory compounds, but for PAA, the impact is complicated by the presence of H2O2 and acetic acid in commercial PAA disinfectant solutions. Analysis of a BOD Progression

The first-order equation is only an approximation to the overall rate of organic matter removal in the BOD bottle as discussed above. The two parameters that must be assessed in a BOD determination are the rate constant and the ultimate BOD. A BOD progression is required to estimate both of these parameters. In a BOD progression analysis, a wastewater sample is separated into a number of aliquots for which BOD determinations will be performed for times ranging from near 0 to 20 or more days. The upper time limit depends on the time for the ultimate BOD to be exerted. More dilution will be applied to aliquots incubated for longer periods of time. Various statistical approaches are applied to analyze the data according to Eq. (5.11) or (5.12). Two of the methods are discussed next. The method of moments described by Moore et al. (1950) is another valid approach. Thomas’s Graphical Method

In finding k and La, from x–t data, difficulty arises because of the exponential term in Eq. (5.12). There are various ways to handle this problem, one of which is to replace the exponential term with a nonexponential function that closely approximates the original function. Thomas (1950) recognized that the (1 e kt) term is similar to the function1 kt 1 ‡

1 kt 6

3

The similarity is seen by series expansion of each expression. 1

e

kt

kt 1 ‡

1 kt 6

1 1 …kt † ‡ …kt †2 2 6

ˆ kt 1 3

ˆ kt 1

1 …kt †3 ‡ ∙ ∙ ∙ 24

1 1 …kt † ‡ …kt †2 2 6

1 …kt †3 ‡ ∙ ∙ ∙ 21:6

(5.15a)

(5.15b)

The expansions of Eqs. (5.15a) and (5.15b) are identical for the first three terms and the difference in the fourth term is small. Therefore, x can be approximated by x ˆ La kt 1 ‡

3

1 kt 6

(5.16)

This equation can be rearranged into a linear equation by solving for t/x, t 1 1 1 ‡ kt ˆ x kLa 6

3

1 The first step in the regression of nonlinear equations is often to find a linear equation that has a series expansion similar to the original equation. This equation is then solved to estimate the parameters.

111

112

Theory and Practice of Water and Wastewater Treatment

Taking the cube root and rearranging, t x

⅓

1 kLa

ˆ

⅓

‡

…k †⅔ t 6La⅓

(5.17)

This equation has a slope, S, and intercept, I, which depend only on the two unknowns. Sˆ

…k †⅔ 6La⅓

Iˆ

1 kLa

(5.18a) ⅓

(5.18b)

Solving these two equations, the unknowns are found to be 6S …k in base e† I 1 La ˆ 3 kI

kˆ

(5.19a) (5.19b)

The procedure for the Thomas method is to do the following steps: 1) Construct a table with columns for ti, xi, ti/xi and (ti/xi)1/3. 2) Plot (t/x)1/3 versus t and draw the best fitting curve. 3) Obtain S and I from the curve and use Eqs. (5.19a) and (5.19b) to solve for k and La. The method is approximate. Experimental values of x > 0.9La should not be used in fitting the line because deviations from the exact equation become significant when about 90% of the BOD has been exerted. Least Squares Regression

Regression analyses can be applied to x–t data by differentiating Eq. (5.12) and applying a logarithmic transformation to produce a linear equation. dx ˆ kLa e dt

kt

(5.20)

The logarithmic transformation of Eq. (5.20) is ln

dx dt

ˆ ln…kLa †

kt

(5.21)

which has the general form: y = mt + b. The standard least squares technique2 (see any elementary statistics book, e.g., Kennedy and Neville 1986) is applied to minimize the sum of squares of the residuals, which results when the parameters k and La are chosen and applied in Eq. (5.21) with the measured data. R ˆ ln where

dx dt

ln…kLa † ‡ kt i

(5.22)

i

dx is the estimate of the derivative at time ti. and R is the residual. dt i

2 The equation to be regressed is nonlinear. Taking the logarithm of the equation may change the variance structure of the data. This may distort the weights given to portions of the data and thus lend some error to parameter estimation. However, in many cases, the procedure gives good estimates of the parameters.

5 Analyses and Constituents in Water

The central difference approximation is used to find the derivatives: dx xi‡1  dt i t i‡1

xi ti

1 1

For the last derivative in (x–t) data pairs, use dx dt

 n

xn tn

xn tn

1 1

which applies at t ˆ

tn ‡ tn 2

1

The slope and the intercept of the best-fit line determine k and La. 5.4.3. Total Organic Carbon Total organic carbon is the organic carbon content of a sample. It is determined with a dedicated instrument, of which several are available commercially. Inorganic carbon (CO2, HCO3 , and CO23 ) must first be purged from the sample under acidic conditions; alternatively, it may be assessed through an acidity analysis. A sample is injected into the instrument that uses a catalyst (usually platinum) and heat while supplying oxygen to convert organic C into CO2. An alternative processes uses high-intensity ultraviolet light, but this is suitable only for already high-quality waters with low TOC. The amount of CO2 produced is measured for the known volume of sample. See Example 5.4 for the calculation of TOC when the sample composition is known. Drawbacks associated with this analysis are the expense of the instrument and that the average oxidation state of the organic carbon is not determined. The oxidation number on carbon may range from 4 to +4 (CH4 to CO2). Also the technique does not distinguish between biologically degradable and nondegradable substances. These limitations make TOC the less frequently reported parameter compared to BOD and COD. On the positive side, almost all organic carbon is measured by this technique, which is rapid and reliable. The ratios of BOD5 and COD to each other and the ratio of either of these parameters to TOC are not constant for wastewaters. The relative biodegradability and state of oxidation of organics in the waste influence these ratios. Ratios of BOD5:TOC and COD:TOC ranged from 1.28 to 2.53 and from 1.75 to 6.65, respectively, for a number of chemical wastes (Eckenfelder 2000). In a biological wastewater treatment process, these ratios change as the waste becomes degraded by microorganisms. BOD5 becomes very low in an efficient process. COD will decrease because of oxidation of organics, but it will be higher than BOD because of the production of some substances that are difficult to degrade. Total organic carbon will decrease but not by as much as the other two parameters; therefore, the ratios of BOD5 or COD to TOC will decrease with biological treatment.

Questions and Problems 5.1 The solubility products for Ag2CrO4 and AgCl are 2 × 10 12 and 3 × 10 10, respectively. Chromate ion (CrO24 ) is added at about 300 mg L 1 in the standard procedure for Cl determination. For this concentration of chromate, how much Cl can be in solution? (Hint: find the amount of Ag+ in solution.) If 0.7 mg of Ag2CrO4 must form in a 100 mL sample to give a noticeable brownish red color to the analyst, how much AgNO3 must be added to a 100 mL sample of water containing a Cl concentration of 28 mg L 1? 5.2 Identify the components of the EDTA molecule that are responsible for its name.

113

114

Theory and Practice of Water and Wastewater Treatment

5.3

a Using data in Table 1.3, what is the equivalence potential of 100 mL of a solution containing 1.00 × 10 3 mol of Ce4+ titrated with a 0.010 N solution of Fe2+? The initial concentration of Ce3+ is negligible. b What is the equivalence potential of 100 mL of a solution containing 1.00 × 10 3 mol of Ce4+ and 1.00 × 10 4 mol of Ce3+ titrated with a 0.010 N solution of Fe2+? c Will the difference in equivalence potentials determined in (a) and (b) have a significant influence on the results of the titration if the same indicator is used? (A few calculations may help prove your point.)

5.4 In Example 5.1, does an equation similar to Eq. (vi) apply at any time during the titration? 5.5 Derive Eq. (5.3). 5.6 Dr. Drastic wishes to give the impression at a magic show that he can change water into wine (or at least show that he has the potential for doing this). He has prepared a solution with resorufin (a derivative of resazurin), an indicator that has an E° of 0.051 V at a pH of 7.00. Resorufin is an indicator that is colorless in its reduced state and pink in its oxidized state. The reduced form gives up 2 electrons when it is oxidized. The redox equation for resorufin (R) at a constant pH can be represented as E ° ˆ 0:051 V …at pH ˆ 7:00† R ‡ 2e ˆ R2 Pink Colorless He has prepared a 1 L solution from distilled water and added a buffering agent to maintain the pH at 7.00. Initially, the buffered solution contained 9.00 mg L 1 of oxygen and no sulfate. Sodium sulfite was added to the solution at 800 mg L 1 to remove the oxygen. A small amount of resorufin (insignificant compared to other species) was added to the sulfited solution. A 1 L solution of perchloric acid (HClO4) will be mixed with 1 L of sulfite solution to produce a pink solution. HClO4 is a very strong acid that ionizes to produce ClO4 (perchlorate) and the perchloric acid solution is also buffered at a pH of 7.00. Perchlorate is a strong oxidizing agent. There was no chloride ion initially in the solution to which the perchlorate was added. Ignoring any addition of oxygen, what is the minimum concentration of perchlorate in the second solution required to obtain a definite pink color (a ratio of 10 is required)? The temperature will be 25 °C the night of the show. 5.7 Light at a certain wavelength was beamed through a sample of distilled water and its intensity decreased by 10%. In a sample of water in a similar cell, the intensity of light transmitted was 82% of the intensity generated by the light source. What is the absorbance resulting from SS and other matter in the sample? 5.8 An analyst found that the transmittance of a sample that contained 3.00 mg L 1 of an absorbent substance is 0.847 in a cell with a path length of 2.00 cm. What is the transmittance of the same substance at a concentration of 2.00 mg L 1 in a cell with a path length of 1.00 cm? 5.9 How would a spectrophotometric analysis be affected by (a) a voltage fluctuation and (b) SS in the sample? 5.10 Characterize the different forms of solids that can occur in a water sample.

5 Analyses and Constituents in Water

5.11 How can some inorganic solids be lost when a total SS–volatile suspended solids analysis is performed? Write the chemical equations given in the text illustrating this phenome­ non. Can you write any other chemical equations for loss of inorganic components at high temperatures besides those given in this chapter? 5.12

a In the COD determination, a solution containing dichromate is titrated with ferrous ion. Prepare a table listing the concentrations of Fe2+, Fe3+, Cr2 O27 , and Cr3+ with the following additions of 3.0 M Fe(NH4)2(SO4)2 to the solution described below: 10.0, 30.0, 50.0, 80.0, 95.0, 100.0, and 105.0 mL. The solution initially contains 0.05 mol of potassium dichromate in 100 mL of water. The pH of the solution is initially 1.0 and may be assumed to remain at this value throughout the titration. Include in your table a column for the system potential after each addition of titrant and prepare a plot of system potential versus amount of titrant added. Use appropriate assumptions about the concentrations of the various ions before, after, and at equivalence. If the concentration of an ion or ions is negligible, do not iterate to find its exact concentra­ tion. At equivalence, use the fact that the amount of ferrous iron added must be equivalent to the initial amount of dichromate present. Also, the dichromate has been almost totally converted into Cr3+ at this point. b Calculate the concentration of Fe2+ after the addition of 10.0 mL of 3.0 M Fe (NH4)2(SO4)2 and equilibrium is established. If a second aliquot of 20.0 mL of titrant is added instantaneously to this solution, calculate the cell potential (EDC EFe) that exists after the addition of titrant but before any reaction has occurred. c The reduction of Cr2 O27 consumes a significant amount of H+ ions that could have a significant effect on the potential. Consult the latest edition of Standard Methods to find out how the analysis is conducted and explain how this problem is handled.

5.13 In the COD procedure in Standard Methods (APHA et al. 2012), 25.0 mL of 0.0417 M K2Cr2O7 and 75 mL of concentrated H2SO4 are added to a 50 mL sample. Concentrated H2SO4 has a specific gravity of 1.835 and contains 97% H2SO4. If the sample is distilled water, how much hydrogen ion will be consumed if the solution is titrated with ferrous iron? Do you expect a significant pH change when a dichromate-sample solution is titrated with ferrous ion? 5.14 Does a pH change influence the COD content of a sample? 5.15 If a water contained an agent that could complex significant amounts of Fe2+, would a lower or higher COD be reported than the actual value? 5.16 What is the COD of a water that contains 100 mg L

1

of phenol (C6H5OH)?

5.17 General formulas for carbohydrate, protein, and fats are CH2O, C16H24O5N4, and C8H16O, respectively. If a wastewater contains carbohydrates, proteins, and fats in ratios of 50 : 40 : 10 on a COD basis (N does not contribute to the COD), what are the concentrations of carbohydrates, proteins, and fats if the total COD concentration is 1000 mg L 1? 5.18

a What is the COD of a water in terms of n, a, and b for a water that contains organic matter that is represented by the formula CnHaOb? Assume that there are no other substances that consume oxygen.

115

116

Theory and Practice of Water and Wastewater Treatment

b Determine c in terms of n, a, b, and z for the following reaction: Cn Ha Ob Nz ‡ cCr2 O72 ‡ f H‡ → dCr3‡ ‡ gCO2 ‡ eH2 O ‡ hNH3 5.19 Quadrivalent cerium [Ce(iv)] can be used in the COD test although results are generally not as satisfactory as with dichromate. Find the general reaction [equivalent to Eq. (5.10)] for oxidation of an organic compound using Ce(IV), which is reduced to Ce(III) in the reaction. 5.20 What result would be obtained for COD of a sample if it contained organic matter with a COD of 568 mg L 1 resulting from organic carbon and an organic and ammonia nitrogen content of 28 mg L 1 as N. The digestion was carried out with 0.25 N dichromate and the chloride concentration of the sample was known to be between 200 and 300 mg L 1. 5.21 What effects do inorganic reducing agents have on COD and COD determination? 5.22 Comment on the statement: “Some wastes have a nonbiodegradable BOD.” 5.23 Describe and discuss some of the reasons for variability in a BOD determination. 5.24 Once the stopper is placed in a BOD bottle, the system is closed. Does the total amount of organic matter (microorganisms and nonliving organic matter) remain constant? 5.25 If a rate constant has a value of 0.092 d of min 1?

1

for base 10, what is its value for base e with units

5.26 The 5-d BOD of a sewage sample is 200 mg L 1. What is its first-stage ultimate BOD at 20 °C? k20 (base e) = 0.37 d 1. 5.27 Construct a BOD progression curve for a sample that has a first-stage rate constant 0.37 d 1 (base 2) and a nitrogenous first-order rate constant of 0.18 d 1 (base e). Nitrification is not significant until after a period of 7.5 days has elapsed. The ultimate first-stage BOD is 335 mg L 1 and the ultimate nitrogenous oxygen demand is 90 mg L 1. What is the total BOD exerted at 10 days? 5.28 The BOD of a sewage sample incubated for 2 days at 30 °C has been found to be 140 mg L 1. What will be the 5 days 20 °C BOD? k20 (base e) = 0.37 d 1. Assume θ is 1.047 for Problems 29–31. 5.29 If the 3 days, 20 °C BOD of a sample is 200 mg L 1, what will be its 7 days, 25 °C BOD? k20 (base e) = 0.37 d 1. 5.30 If the BOD of a wastewater sample shows a first-stage BOD value of 188 mg L 1 and a reaction velocity constant, k (base e), of 0.52 d 1 at 15 °C, what is its expected 5-d BOD at 20 °C? 5.31 What are the ultimate BODs of 100 mg of each of the following substances: (i) nitrite, (ii) nitrate, (iii) ammonia, and (iv) ammonium ion? What are the oxygen demands of

5 Analyses and Constituents in Water

ammonia and ammonium ion, respectively, when they are expressed as N? Is it more practical to express ammonia as ammonia, and ammonium as ammonium, or to express both as N when referring to their oxygen demand (explain)? 5.32 Assuming that the cumulative respiration of algae over 5 days is 0.25 mg O2/mg algae, what are the algae concentrations in the sample that would cause a DO decrease of (a) 2.5 mg L 1 and (b) 2.75 mg L 1 due to algal respiration in a sample with a BOD5 of 20 mg L 1 from other organic matter? The dilution factor is 10 in each case. 5.33 While conducting BOD tests on a wastewater sample, the following measurements were taken. Calculate the BOD of the sample. Sample

Dilution (%) 1

DO decrease (mg L )

1

2

3

1

2

3

2.7

4.9

7.2

5.34 What are the permissible dilution limits to attain a DO decrease between 2.0 and 7.0 mg L 1 for a sample with a BOD5 of 1500 mg L 1? 5.35 A BOD determination was made on an industrial waste that required addition of a seed culture. The seed culture had a BOD of 320 mg L 1 and 1.00 mL of seed was added to the standard 300 mL BOD bottle, which contained 8 mL of the wastewater. A DO decrease of 6.25 mg L 1 was measured in the BOD bottle after 5 days. What was the BOD5 of the wastewater? 5.36 A BOD determination was performed using 10 mL of sample in a 300 mL bottle. It was discovered that the DO meter was improperly calibrated for the initial DO determination; therefore, a DO determination was made 24 hours after the sample was set. The DO was 7.20 mg L 1. The DO after 5 days of incubation at 20 °C was measured at 3.05 mg L 1. The rate constant for this waste is known to be 0.25 d 1 (base e). What was the BOD5 of this sample? 5.37 (a) Is it possible to have a sample with an ultimate BOD greater than the COD? (b) Is it possible for BOD to decrease without a decrease in COD? (c) Is it possible for COD to decrease without a reduction in BOD? 5.38 Discuss the practicality of accelerating the BOD test by adding a large amount of seed organisms. 5.39 The BOD data in the following table were obtained for a wastewater sample. Estimate the rate constant and ultimate BOD of the waste using (a) the Thomas method and (b) the regression method. Time (d)

1

2

3

4

5

6

7

8

9

10

BOD (mg L 1)

190

306

360

438

500

490

510

500

524

542

117

118

Theory and Practice of Water and Wastewater Treatment

5.40 The energy yield from the oxidation of ammonia to nitrite yields more energy than the oxidation of nitrite to nitrate. (a) Why would you expect this to be true? (b) Would you expect a higher concentration of Nitrosomonas or of Nitrobacter in a reactor where nitrification is proceeding? 5.41 What is the relative improvement in accuracy of a BOD determination as the number of sample aliquots increases from one to two to three? 5.42

a What is the 5-d BOD of a sample that has a concentration of 159 mg L 1 of organic matter that can be represented as C4.2H6.1N0.8O2? The rate constant is 0.14 d 1 (base 10) and nitrification was not significant until day 6. b What are the ultimate carbonaceous and nitrogenous BODs, COD and TOC of this waste?

References Absi, F., Gamache, F., Gehr, R. et al. (1993). Pilot plant investigation of ozone disinfection of physico-chemically treated municipal wastewater. Proceedings of the 11th World Ozone Congress, San Francisco, S-7-33-42. APHA, AWWA, and WEF (2012). Standard Methods for the Examination of Water and Wastewater, 22 (eds E.W. Rice, R.B. Baird, A.D. Eaton and L.S. Clesceri). Washington, DC: American Public Health Association. Bishop, E. (1972). Oxidation–reduction indicators of high formal potential. In: Indicators (ed. E. Bishop), 531–684. Toronto: Pergamon Press. Eckenfelder, W.W. Jr. (2000). Industrial Water Pollution Control, 3e. Boston: McGraw-Hill. El-Rehaili, A.M. (1995). Response of BOD, COD and TOC of secondary effluents to chlorination. Water Res. 29 (6): 1571–1577. doi: 10.1016/0043-1354(94)00234-X. Gaudy, A.F. Jr. (1972). Biochemical oxygen demand. In: Water Pollution Microbiology, vol. 1 (ed. R. Mitchell), 305–332. Toronto: John Wiley & Sons. Jouanneau, S., Recoules, L., Durand, M.J. et al. (2014). Methods for assessing biochemical oxygen demand (BOD): a review. Water Res. 49: 62–82. doi: 10.1016/j.watres.2013.10.066. Kennedy, J.B. and Neville, A.M. (1986). Basic Statistical Methods for Engineers and Scientists, 3e. New York: Harper & Row. Kim, B.R. (1989). Effect of ammonia on COD analysis. J. Water Pollut. Control Fed. 61 (5): 614–617. doi: 10.2307/25043656. Moore, E.W., Thomas, H.A. Jr., and Snow, W.B. (1950). Simplified method for analysis of BOD data. Sewage Ind. Wastes 22 (10): 1343–1353. http://www.jstor.org/stable/25031429. Ottaway, J.M. (1972). Oxidation–reduction indicators of E° < 0.76 Volt. In: Indicators (ed. E. Bishop), 469–530. Toronto: Pergamon Press. Talinli, I. and Anderson, G.K. (1992). Interference of hydrogen peroxide on the standard COD test. Water Res. 26 (1): 107–110. doi: 10.1016/0043-1354(92)90118-N. Thomas, H.A. Jr. (1950). Graphical determination of BOD curve constants. Water Sewage Works 97: 123–124. Wagner, M., Brumelis, D., and Gehr, R. (2002). Disinfection of wastewater by H2O2 or PAA. Development of procedures for measurement of residual disinfectant, and application to a physico-chemically treated municipal effluent. Water Environ. Res. 74 (1): 33–50. doi: 10.2175/ 106143002X139730.

SECTION11

05/08/2018

15:14:8

Page 119

119

Section II Microorganisms in Water and Water Quality

SECTION11

05/08/2018

15:14:8

Page 120

121

6 Microbiology Microorganisms are significant in water and wastewater because of their roles in disease transmission, and they are the primary agents of biological treatment. They are the most diverse group of living organisms on earth and occupy important niches in the ecosystem. Within the microbial realm, there are species able to oxidize iron, and others are able to survive and proliferate in extreme environments ranging from, for instance, extremely low to high pH or temperature. Their simplicity and minimal survival requirements allow them to exist in diverse situations.

6.1 Groups of Microorganisms and the Phylogenetic Tree The current grouping of cellular-based organisms (which excludes viruses and bacteriophages, as they are not cellular) is shown diagrammatically as the evolutionary tree of life, or “phylogeny” (Figure 6.1). It represents the evolution of organisms reconstructed from variations in the DNA sequence. The genes that were first used for this analysis and that are still the principal genes used today are those coding for ribosomal ribonucleic acids (rRNA; specifically the 16S/18S rRNA1) because all cellular-based organisms express these RNA as parts of the protein-synthesizing structure called the ribosome; however, viruses and phages do not contain ribosomes, therefore, they cannot synthesize their own proteins. As shown in Figure 6.1, three distinct lineages – called “domains” – have emerged from the “last universal common ancestor” or LUCA: they are Bacteria, Archaea, and Eukarya. Based solely on their cell structure, however, both Bacteria and Archaea are prokaryotes (i.e., they have no defined nucleus and simpler cells) whereas Eukarya are eukaryotes, i.e., they have a defined membrane inside the cell enclosing nuclear DNA, and other components not found in prokaryotes. Archaea as a separate domain was defined relatively recently by Carl Woese in 1977.

6.2 Bacteria and Archaea Because of their similarity in cellular structures and lifestyles, bacteria and archaea will be covered in the same section, although many archaea are evolutionarily closer to eukaryotes as can be observed in the phylogenetic tree. 1 “S” is the Svedberg unit, named after Swedish chemist Theodor Svedberg (1884–1971), as a unit of size; 16S in prokaryotes and 18S in eukaryotes. Theory and Practice of Water and Wastewater Treatment, Second Edition. Ronald L. Droste and Ronald L. Gehr.

 2019 John Wiley & Sons Inc. Published 2019 by John Wiley & Sons Inc.

Companion website: www.wiley.com/go/droste/water

122

Theory and Practice of Water and Wastewater Treatment

Figure 6.1 Evolutionary tree of life. Source: Black and Black (2015). Reproduced with permission of John Wiley & Sons.

The simplest wholly contained life systems are bacteria. As noted, they are characterized by the lack of a nuclear membrane and their machinery of metabolism is typically not contained in organelles. They reproduce by simple fission. A typical bacterial cell is shown in Figure 6.2. The bacterial cell is enclosed within a cell wall that is a polymeric layer about 100 nm thick, conferring structural rigidity to the cell. The cell wall maintains the integrity of the cell, holding it

Figure 6.2 Typical bacterial cell.

6 Microbiology

together against osmotic pressure gradients that occur because of concentration differences between the cell contents and the surrounding liquid. The cytoplasmic membrane is immediately inside the cell wall. Though much thinner than the cell wall, it is a highly complex structure, containing a phospholipid bilayer with hydrophobic ends of molecules facing each other and hydrophilic ends facing the aqueous environment of the cell. In this manner, it controls the passage of nutrients and other compounds into and out of the cell. The cytoplasm is a solution that contains the biomolecules essential for metabolism. The primary component of prokaryotic cells, as for all living systems, is water, which is approximately 75% of the cell by weight. The fundamental molecules included in the cytoplasm are the macromolecules proteins, lipids, and polysaccharides, as well as organic and inorganic metabo­ lites. In prokaryotes, the cytoplasm also contains deoxyribonucleic acid (DNA) as the cell’s “chromosome” repository of instructions for all cellular characteristics, and ribonucleic acids (RNAs) as blueprints of genes coding for specific proteins and other accessories for protein synthesis. Finally, ribosomes are major structures found in the cytoplasm; they consist of rRNA and proteins that catalyze the synthesis of proteins based on the instructions of specific RNA molecules. The following features are not common to all bacteria. There are three types of filamentous appendages, all composed essentially of proteins. The flagellum (or flagella, if more than one) is the longest. It can be attached at the end or along the sides of the cell. It is the primary means for cell motility. It rotates almost as a corkscrew to propel the cell and can reach speeds of 300 rps (rotations per second) (measured as Hz) or even higher. Pili (singular “pilus”) are shorter appendages that may have several functions. One type of pilus is involved in bacterial conjugation (transmission of genetic material from one bacterial cell to another), as well as the adhesion of pathogenic bacteria to specific host tissues preceding invasion (e.g., Streptococcus pyogenes). Other pili can be receptors for bacteriophages (bacterial viruses) and can act as bacterial nanowires in microbial fuel cells. Yet others, known as Type IV pili, are involved in twitching motility (e.g., some Pseudomonas spp.) Finally, fimbriae are short hair-like structures that enable cells to stick to surfaces or to form pellicles (thin sheets of cells on a liquid surface). They are responsible for biofilm formation that can be put to beneficial use in water and wastewater processes or can assist pathogens in their disease process (e.g., Salmonella spp. and Streptococcus pneumoniae) that then make them extremely resistant to destruction by the host’s immune system. Granules, aggregates, or other inclusions are observed in many bacteria. The nature of these inclusions differs in different microorganisms. Poly-β-hydroxybutyric acid (PHB), or more generically, the molecular family of poly-β-hydroxyalkanoates (PHA) is a carbon- and energycontaining storage polymer. When carbon is in excess in the cell’s environment, these polymers will accumulate; they will be broken down for synthesis or energy during periods of starvation. Glycogen, which is a polymer of glucose, is another type of carbon- and energy-containing polymer. Inorganic phosphate (PO43 ) is stored as polyphosphates in granules called volutin; this will be used for nucleic acid, phospholipids, and adenosine triphosphate (ATP) biosynthesis when needed. Similarly, globules of elemental sulfur (S0) may be stored, then oxidized to SO42 when necessary. Magnetosomes (Fe3O4 – magnetite, and in some cases Fe3S4 – greigite) orient a cell according to the earth’s magnetic field. Their function is currently unknown, but speculation points to assisting the cell to locate food. Finally, gas vesicles in the cell assist with flotation. The cell may be surrounded by extracellular material that is usually composed of polysac­ charides, polypeptides, or polysaccharide–protein complexes. When the layer is compact and covalently attached to the cell wall, it is known as a capsule. It forms a tight matrix that excludes small particles and can be seen under a light microscope when it is stained with India ink. When it

123

124

Theory and Practice of Water and Wastewater Treatment

is diffuse and not attached to the cell wall, it is known as a slime layer. This layer will not exclude particles and is more difficult to see. Neither layer is part of the cell wall, and as such they do not provide additional structural integrity. Instead, they are involved with forming biofilms, together with the fimbria as mentioned earlier, or protect against aggression such as phagocytosis by immune cells. Some bacteria are able to form spores (endospores or free spores) in adverse circumstances. A spore is a nonvegetative cell that bears little resemblance to the living cell. The protoplasm is covered by many coatings, making it heat-resistant and resistant to lack of moisture. Upon finding itself in an agreeable environment, the spore reverts to a vegetative cell. This trait is common in two important genera of bacteria, Bacillus and Clostridium. Spores are remarkably heat-resistant, which has implications for sterilization. The structure of a spore is much more complex than the vegetative cell, especially in the kinds of layered structures (proteins, peptidoglycans) outside the core wall. Fundamental cell shapes are circular (coccus), rod (bacillus), curved (vibrio), spiral (spirillum), and filamentous (Figure 6.3). Bacteria of a species may be singular, chained, or clumped together. Individual cells range in size from 0.125 μm diameter to 1.35 × 30 μm. Further details on the components of bacterial cells may be found in microbiology textbooks such as those by Madigan et al. (2015) and Willey et al. (2017). 6.2.1 Classification of Bacteria This section will focus on classification of bacteria; however, many – if not most – of these methods could apply to all other life forms. There are many means of classifying bacteria, including physical characteristics (taxonomy), energy sources, nutrient sources, environmental conditions, biochemical capabilities, and genetic analysis. We need to define two key concepts here, as they will be referred to within several of these classification schemes. Catabolism is the breakdown of larger molecules (usually organic) to yield energy and building blocks for the synthesis of protein and other cellular materials, and ultimately cell growth. Anabolism is the opposite; thus, it is the use of available energy and nutrients for growth and reproduction. Taxonomy

Taxonomy was the earliest scheme used to classify all organisms, including bacteria. It is now based on the phylogenetic tree, supplemented with physiological characteristics including biochemical responses. The scheme gave rise to the classical designation “Genus species” that is still used regardless of any other method of classification, although it is typically supplemented with “strain.” The standard reference for bacterial taxonomy (strictly, all

Figure 6.3 Bacteria shapes.

6 Microbiology

prokaryotes, hence Archaea and Bacteria domains) is Bergey’s Manual of Systematics of Archaea and Bacteria (n.d.). It is continually updated on the web and has been integrated with the List of prokaryotic names with standing in nomenclature (LPSN, n.d.). Metabolic Requirements

Bacteria and all life forms require: Energy. Photo or chemo. Reductant. Litho or organo. Carbon. Auto or hetero. The same substrate may serve more than one function. The two sources of energy are the following: (i) Sunlight. The organisms using sunlight for energy are called phototrophs (from the Greek photo – light; troph – feed). Some may function in environments with oxygen, others without. (ii) Chemicals. Organisms using this source are called chemotrophs, and they are also found in oxygen-containing and oxygen-free environments. The chemicals may be organic – hence chemoorganotrophs, or inorganic, hence – chemolithotrophs (Greek lithos – stone). Note that only prokaryotes have this ability; eukaryotes cannot use inorganic compounds for energy. The sources of reducing equivalents (electrons) can be organic compounds (organotrophs) or inorganic compounds (lithotrophs). These sources can be used by both phototrophs and chemotrophs. Both photolithotrophs and chemolithotrophs are referred to as autotrophs (Greek auto – self); chemoorganotrophs are also classified as heterotrophs (Greek hetero – many). Carbon (C) is considered as one of the seven essential nutrients for all organisms. The others are H, N, O, P, S, and Se, but carbon is required in the highest percentage by mass. Heterotrophs use organic carbon (i.e., a reduced form of carbon) as their carbon source, whereas autotrophs are content with inorganic carbon (CO2) for their carbon needs. Some bacteria are able to extract all their essential nutrients from a basic inorganic salts medium. The nutrients are in a reduced form; for instance, nitrogen is supplied as ammonia. Other bacteria require certain preformed growth factors ranging from amino acids to vitamins. Oxygen Requirements

Chemotrophs extract their energy through oxidation and reduction (redox) reactions, which involve electron donors (which are oxidized as part of metabolism) and electron acceptors (which are reduced during metabolism). Redox couples always involve a reduced substance (the electron donor) whose E ´0 (oxidation–reduction potential, ORP, under standard conditions) is more negative (measured as volts) than the oxidized substance (the electron acceptor). Table 1.3 contains a short list of ORPs as a part of basic chemistry; microbiology textbooks such as that by Madigan et al. (2015) show a “redox tower” that lists those redox couples more relevant to biochemical reactions. Within the microbial cell, the coenzyme pair NAD+/NADH (nicotinamide adenine dinucleotide ion/acid) are redox mediators that can carry electrons (2e ) and protons (2H+) at the same time. They increase the variety of feasible redox reactions by allowing chemically dissimilar electron donors and acceptors to interact. Fermentation refers to anaerobic catabolism, during which an organic compound is both an electron donor and an electron acceptor. Energy release is comparatively small. Respiration is catabolism under aerobic or anaerobic/anoxic conditions, in which an electron donor is oxidized with O2 or an O2 substitute (such as NO3 ) as the terminal electron acceptor. Energy release is much higher.

125

126

Theory and Practice of Water and Wastewater Treatment

The ORP for survival and optimum growth of bacteria, spans a wide range. Obligate aerobes require a full aerobic (presence of free oxygen) environment, even outside water. Habitats include the skin and dust; an example is Micrococcus luteus. Microaerophilic organisms need only low concentrations of dissolved oxygen (slightly above 0 mg L 1). Habitats include fresh water bodies; an example is Spirillum volutans. Facultative organisms can adjust their metabolism according to the presence or absence of oxygen; typical environments include the animal intestine and the wellknown example is Escherichia coli. Aerotolerant anaerobes can grow but cannot respire in the presence of oxygen; an example is S. pyogenes, which can be found in the upper respiratory tract. Finally, obligate anaerobes are inhibited or killed by oxygen; they can be found in sewage sludges and lake sediments; an example is the archaea Methanobacterium formicicum. Aerobic metabolism yields the most energy per unit of foodstuff processed. Briefly, ATP is the prime energy “currency” in all cells. It is continually broken down to drive anabolic reactions, then resynthesized during catabolism. Different biochemical pathways yield different ratios of ATP (molecules/molecule of energy-yielding substrate [food]). Taking glucose as an example, glycolysis, also known as the “Embden–Meyerhof–Parnas” pathway (see Figure 4.10), yields two molecules of ATP and of NADH per molecule of glucose oxidized to pyruvate. This is a universal process, for both aerobic and anaerobic metabolisms; however, the absence of a terminal electron acceptor means that NADH cannot be reoxidized by the electron transport chain to produce additional ATP, and pyruvate needs to be fermented to more reduced end-products such as lactic acid, acetic acid, or ethanol to achieve NADH re-oxidation. Consequently, anaerobic fermentation essentially stops at this point. If terminal electron acceptors are present, pyruvate can be completely oxidized to CO2 in the tricarboxylic acid cycle, also known as the citric acid cycle or “Krebs cycle” (see Figure 4.11), which produces additionally the equivalent of 2 ATP and 9.3 NADH per glucose. The electron transport chain reoxidizes NADH by transferring electrons to oxygen or to other terminal electron acceptors, and the amount of ATP produced from this process is dependent on the strength (i.e., relative position on the redox tower) of the electron acceptor used by the organism. Thus, comparing all these processes, 2 ATP would be produced from the fermentation of a glucose molecule, while respiring organisms could produce a lot more with a maximum for aerobic organisms of a theoretical 38 ATP/glucose molecule (assuming 3 ATP/NADH). Temperature

Microorganisms are grouped into four temperature ranges, as shown in Table 6.1, depending on their optimum growth temperature, though they are still able to grow, but more slowly, over a temperature range. The nomenclature uses the suffix -phile (Greek for loving). The lowest temperature at which a living microorganism was found, Psychromonas spp., is –12 °C, and the highest, at 122 °C, is the archaeon Methanopyrus kandleri. Table 6.1 Microorganism groupings based on optimum growth temperature. Type

Optimum growth temperatures (°C)

Psychrophile

4

Temperature range (°C)

4 to 12

Example

Polaromonas vacuolata

Mesophile

28–39

8–46

Escherichia coli

Thermophile

60

42–66

Geobacillus stearothermophilus

Hyperthermophile a

88

65–96

Thermococcus celer

Hyperthermophile b

106

90–112

Pyrolobus furnarii

Source: After Madigan et al. (2015).

6 Microbiology

Salt and Sugar Concentrations

Dissolved substances produce an osmotic pressure on cells as well as contribute to other effects. Organisms that grow best in the presence of high concentrations of solutes are osmophilic. Microorganisms that tolerate high concentrations of salt are known as halophiles or halotolerant – for example, Aliivibrio fischeri, which grows at an optimum salt (NaCl) concentration of 6%. Extremely halophilic (halo-obligatory) microorganisms require salt concentrations over 12% for growth; Halobacterium salinarum cannot grow below 11% NaCl, and it prefers above 18% NaCl. (Note that sea water salt concentration is ~3.5%.) If a microorganism is tolerant of a high concentration of sugar as opposed to a salt, it is termed saccharophylic. pH

Most living organisms, including microorganisms, are “comfortable” at neutral pH (7.0), but can survive well over a relatively narrow range above or below that. Acidophiles can function at much lower pH values, the archaeon Picrophilus oshimae growing at an optimum pH of 0.7 and temperature of 60 °C. Alkaliphiles, as expected, prefer pH values from 7 to 11, the most extreme example being Bacillus firmus.

6.3 Eukaryotes Eukaryotes is the third domain of life. Eukaryotic cells are about an order of magnitude larger than bacteria, although some algae such as seaweed are macroscopic in size, growing to over 30 m in length. Few anaerobic eukaryotes are found because of their higher levels of complexity and size. The major characteristic of this group compared to prokaryotes is the existence of a nucleus containing most of the genetic materials of the cell (Figure 6.4). The area outside the nuclear membrane is the cytoplasm. The endoplasmic reticulum is the site of production of membrane lipids and proteins. Eukaryotic cells also contain mitochondria (mitochondrion, singular) that is the organelle that produces ATP from the oxidation of NADH. According to the endosymbiotic

Figure 6.4 Typical eukaryotic cell.

127

128

Theory and Practice of Water and Wastewater Treatment

hypothesis, this organelle was originally a bacterial cell associated with a eukaryotic cell. Interestingly, mitochondria still have bacterial DNA remnants, including their 16S rRNA genes. Photo-synthesizers (e.g., algae) contain chloroplasts, which convert photons of sunlight into usable energy (ATP and NADH). Other organelles that may be found in eukaryotic cells include Golgi complexes, which store and transport material, and the lysosome, which is a membraneenclosed compartment containing digestive enzymes that hydrolyze proteins, fats, and poly­ saccharides. The lysosome also assists in hydrolyzing damaged cellular material inside the cell and recycling it for new biosynthesis. Microfilaments are structural components that help maintain cell shape and positioning of organelles in the cell, and microtubules are additionally involved with moving chromosomes during mitosis (cell division) and in the movement of organelles inside the cell. Vacuoles that contain a variety of storage products such as lipids (food) and water may be present, and appendages including flagella and cilia may also be found. Flagella in eukaryotes assist with motility, but they move in a whip-like fashion instead of rotating as in prokaryotes. While animal cells do not have cell walls, several other eukaryotes do, including plants (cellulose) and fungi (chitin). 6.3.1 Algae Current thinking is that algae developed from a primary endosymbiotic association between cells of Eukarya and cells from the cyanobacterial lineage of Bacteria (i.e., phototrophic bacteria) to form a single line. A subsequent split occurred resulting in two separate lineages, which are now called red and green algae (Madigan et al. 2015). Both types contain chlorophyll and carry out oxygenic photosynthesis. Red algae (rhodophytes) have chloroplasts containing chlorophyll a and phycobiliproteins, which are light-harvesting pigments. The reddish color of these algae is due to the phycoerythrin pigment that masks the green color of chlorophyll a. Most red algae are multicellular, even forming seaweeds. They are a source of agar (the basis of many microbial growth media) and carrageenan (used as a thickener and emulsifier in dairy and soy food products). One order, Corallinales, has a very high CaCO3 content and assists in coral reef formation. Green algae (chlorophytes) contain both chlorophyll a and b, but no phycobiliproteins. Phylogenetically, they are closely related to plants. They exist in a wide range of morphologies – unicellular, aggregates, colonies, filamentous (still unicellular), and multicellular. Some have flagella. They have a complex life cycle, including both sexual and asexual reproductive stages. One species, Botryococcus braunii, produces oil globules and is believed to be responsible for some of the earth’s oil deposits. Different species of green algae have been found in almost all earth environments, including fresh water, sea water, snow, moist soil, on rocks or inside them (thus responsible for weathering), hot dry deserts or cold dry deserts (such as Antarctica), and as symbionts in lichens. Algae play a role in some wastewater treatment processes, particularly stabilization ponds. Many algae species are harmless, but in water treatment, algae are a nuisance. Algae have been wrongly associated with some taste and odor problems, notably those due to the chemicals 2­ methyl isoborneol (MIB) and geosmin, but many of these are in fact due to compounds (produced by Cyanobacteria), which are not algae. Nevertheless, large blooms of algae, which are a characteristic of eutrophied inland waters resulting from natural aging or anthropogenic pollution, particularly nutrient enrichment, will eventually die and settle to the bottom of lakes. This situation can lead to anaerobic conditions developing which do produce odorous gases. Algae have also been implicated in the production of neurotoxins (affecting the nervous system), hepatotoxins (the liver), and dermotoxins (the skin), but again, the culprits are Cyanobacteria, and these toxins are now correctly termed cyanotoxins to reflect their origins.

6 Microbiology

6.3.2 Fungi There are three major growth forms of fungi: molds, mushrooms, and yeasts. They form a phylogenetic cluster that is on a separate branch from all other protists and are the microbial group most closely related to animals (Madigan et al. 2015). Fungi are generally filamentous and have a true cell wall. Individual filaments are known as hyphae that may have no cross-walls or may be divided at irregular intervals by cross-walls. Fungi typically grow by an extension of the tips of hyphal filaments. Many hyphae will, in turn, grow together to form a tuft called a mycelium. From here, aerial hyphae reach into the air, and spores – called conidia – form at their tips. These are asexual spores that disperse in the wind to colonize new habitats. In some cases, much larger reproductive structures – called fruiting bodies – are formed; these include the mushrooms (part of the group known as Basidiomycetes). Yeasts are nonfilamentous, single-celled fungi that reproduce by a process known as budding. The cells are typically much larger than bacteria. A small bubble is produced on the mother cell that grows to about the same size as the mother cell, then a cross-wall is formed, and the new cell separates. Most fungi are aerobic, and all are chemoorganoheterotrophs with simple nutritional require­ ments; some, including the yeasts, are facultative aerobes or even obligate anaerobes. They feed by excreting extracellular enzymes that digest larger molecules, such as proteins and polysaccharides, which can then enter the cell and the regular anabolic pathways for energy and cell synthesis. Fungi can tolerate a lower pH and lower water activity2 than bacteria, and their nitrogen and phosphorus requirements are also lower. All these characteristics make them excellent decomposers of dead animal and plant materials, hence key participants in composting and landfill operations. They can also function as parasites of plants or animals, taking up nutrients from living cells. In this role, they are dangerous and often resistant to medical interventions. They are valuable for treating some industrial wastewaters, but filamentous forms are difficult to remove by sedimentation and significant growth of fungi in wastewater treatment plants can lead to poor effluent quality. 6.3.3 Protists The greatest diversity in the eukaryotic world lies in the protists (Madigan et al. 2015). They include phototrophic and nonphototrophic microbial eukaryotes. Protist groups of importance in the water and wastewater environment are as follows: Euglenozoans. These are unicellular, free-living, or parasitic, and flagellated. Euglenids (e.g., Euglena spp.) are, unusually, chemotrophic or phototrophic, hence, they can live in light and dark environments. In addition, most have two flagella, and the flagella allow them to move to the most advantageous environment. They inhabit fresh water and marine environments and are nonpathogenic. They feed on bacterial cells by phagocytosis, i.e., they engulf the unfortunate bacterium with their flexible cytoplasmic membranes to bring it into the cell where it is digested. Alveolates. These have alveoli, which are cytoplasmic sacs just under the cytoplasmic membrane, thought to maintain osmotic balance in some species, and for dinoflagellates (see below), to function as protective plates. There are three kinds of Alveolates, based in part on the types of strands extending outward from their cells: Ciliates use cilia for motility and to obtain food by moving particles into a type of mouth; e.g., Paramecium spp. Many Paramecium spp. (as well as other protists) are hosts for endo­ symbiotic prokaryotes and eukaryotes. 2 Water activity or aw is the partial vapor pressure of water in a substance divided by the standard state partial vapor pressure of water. It is a measure of the energy status of the water; below a certain water activity level – typically 0.85 or less – a microorganism will not be able to grow in that environment.

129

130

Theory and Practice of Water and Wastewater Treatment

Dinoflagellates (dinos is Greek for whirling) have flagella that encircle the cell, causing it to spin. They can photosynthesize through secondary endosymbiosis. Some interact symbi­ otically with marine animals to form coral reefs. Others are toxic, e.g., Gonyaulax cells form “red tides” that are associated with fish poisoning as well as human poisoning through consumption of contaminated shellfish. Stramenopiles (stramen is Greek for straw, pilus for hair) include chemoorganoheterotrophic and phototrophic micro- and macroorganisms. Their flagella have short hair-like exten­ sions, hence the name. Major groups are the following: Diatoms, which are unicellular, are a major component of the planktonic community in marine and fresh waters and have a silica cell wall. Oomycetes are water molds. They are not important in aquatic environments or water and wastewater treatment operations, but are significant plant pathogens. Chrysophytes (golden algae), which are primarily unicellular, and Phaeophytes (brown algae), which are multicellular, are respectively, freshwater and marine algae. An example of the former is Prymnesium parvum, responsible for fish kills, whereas the giant kelp Macrocystis seaweed is a well-known example of the latter. Amoebozoa is a large group of terrestrial and aquatic protists that use lobe-shaped pseudopodia for movement and feeding. There are two major groups: Gymnamoebas are free-living and inhabit aquatic and soil environments. They feed by phagocytosis on bacteria, other protists, and organic particles. Example: Amoeba spp. Entamoebas are animal parasites. A particularly nasty example is Entamoeba histolytica, which forms cysts and can be transmitted from person-to-person via contaminated water, food, and eating utensils. The result of an infection is dysentery and bloody diarrhea. Protista play a role in the performance of aerobic biological treatment processes, especially activated sludge (AS). They feed on particulates in the influent wastewater, as well as on bacteria growing during the AS process. Some protista groups can be identified directly with a microscope by a skilled observer. Being higher on the food chain, their presence and identifica­ tion provides a rapid, inexpensive tool for monitoring process performance. Curds and Cockburn (1970) carried out a comprehensive study of AS plants. Simplifying the complex taxonomy required to identify protozoa to a level usable by wastewater treatment operators is a significant problem (Kinner and Curds 1987).

6.4 Other Microorganisms Other types of microorganisms that fall outside of the above classes are also found in natural waters and wastewaters and have health and/or engineering significance. 6.4.1 Viruses and Phages Viruses are noncellular entities that contain proteins and nucleic acids and are unable to reproduce or metabolize on their own. A protein coat (capsid) surrounds the nucleic acid molecule (genome). There are 219 virus species and at least 2000 known virus types. The smallest adenovirus is 18 nm and needs a helper virus to live (Gerba et al. 2017). The largest virus particle was until this century thought to be the smallpox virus (200 nm long); however, giant viruses were first reported in 2003 (La Scola et al. 2003). They can carry more genes than some bacteria and are larger than many microorganisms. They can even harbor virophages (e.g., the virophage Ma infects the giant virus Cafeteria roenbergensis virus CroV; Moelling 2013). The largest virus discovered to date is the Pandoravirus (Philippe et al. 2013), approaching 1000 nm (1 μm) in size.

6 Microbiology

A modern view of viruses is that they are not parasites in most cases, but equivalent to commensals in a well-balanced system, and most do not cause any diseases (Moelling 2013). Viruses, bacteria, archaea, and eukarya – all inhabit the human gut at all times. Since they generally have small genomes, and since mutation rate (of any organism, not only viruses) is inversely proportional to the length of the genome, it is believed that viruses were among the earliest ancestors of life on earth; actually, they may have preceded LUCA of cellular life. Phages (phagein is Greek for devour) are a type of virus that infect organisms other than animals; indeed viruses could be considered a class or subgroup of phages. Thus Bacteriophages are a type of virus that use bacteria as hosts, and Virophages use other viruses for that purpose. Bacteriophages are nonpathogenic to humans. They are predominantly found in oceans (1012 mL 1) and in the soil, also in our guts, on the skin, and in plants. Most bacteria (~80%) are infected with phages, and ~20% of all bacteria in the oceans are lysed through the activity of phages daily (Moelling 2013). Although used in the former Soviet Union over 100 years ago to treat bacterial infections, bacteriophages have recently been “rediscovered” as tools to destroy pathogenic bacteria when traditional antibiotics have failed. They are also very useful as measures of success for water and wastewater treatment processes, especially disinfection, in which case they are added to the sample in a laboratory test, or inflow to a pilot plant. Examples of bacteriophages of the common indicator microorganism, E. coli include MS2 (single-stranded RNA), ΦX174 (single-stranded DNA), and T2 (double-stranded DNA). 6.4.2 Rotifers Rotifers (Latin for wheel-bearer) are aerobic, multicellular organisms at the first stage of development above single-celled organisms. They have well-defined organelles (equivalent to organs that have specific functions in animals). They have cilia, used for locomotion, and some are located around their mouth to create flow currents to bring in food. These organelles and moving cilia can be clearly seen under a reasonably high-powered light microscope. Their taxonomy is currently unclear, but Barnes et al. (2001) place them in the phylum Rotifera. The major classes are Monogononta and Bdelloidea. Rotifers are naturally found in marine and fresh waters in relatively high numbers, as well as in the “mixed liquor” of AS aeration basins that are operating effectively. Moreira et al. (2016) have described the benefits of using rotifers as indicators of toxicity of contaminants in the aquatic environment: small size, high fecundity, and short life cycle that translate to easy and convenient laboratory procedures, ready availability as eggs resistant to decay, and easy to transport. 6.4.3 Worms Worms are multicellular animals that can be viewed at the interface of microbiology and biology, as many worms are far from being microscopic. In fact, they range in size from microscopic (e.g., phylum Platyhelminthes) to almost 60 m in length (Lineus longissimus). Even the term “worms” is outdated (though still widely used colloquially), as some “worms” are actually insects or fungi. Physically, these organisms have an elongated body and move with an undulating motion. They are found in several sewage treatment processes, especially trickling filters, slow sand filters, and sludges, but their greater significance in the context of the water environment is that many are pathogenic. The following classes are important in the context of water and wastewater treatment processes, as well as from the health aspect if they are transmitted by fresh, marine, or polluted waters and wastewaters. “Helminths” is the term generally used for parasitic worms. Flatworms (Platyhelminthes) are indeed “flat” but can vary widely in length from millimeters to meters. They can be found in fresh

131

132

Theory and Practice of Water and Wastewater Treatment

water, seawater, and other animals. Included in this group are tapeworms (Cestoda) that are parasites living in the intestines of humans and other vertebrates, and flukes, also known as trematodes, which are likewise parasites, living in the internal organs of vertebrates. The phylum Nematoda, containing the tubular-bodied nematodes (or “roundworms”) are the most ubiquitous of animals on earth (including the oceans), accounting for about 80% of all individual animals. Their ubiquity may in part be due to disagreements among biologists as to their exact placing on the phylogenetic tree, as they are closely related to other phyla. They inhabit all environments, including ocean sediments, fresh water, soils, and the human gut, in which case they are parasites. The most common human pathogens include ascarids (Ascaris), filarias, hookworms, pinworms (Enterobius), and whipworms (Trichuris trichiura). Trichinella spiralis, commonly known as the “trichina worm,” occurs in rats, pigs, and humans and is responsible for the disease trichinosis.

6.5 Determining the Growth of Microorganisms Microorganisms are able to grow at very high rates under suitable conditions because of their relatively simple structures and growth requirements. In fact, microbial metabolism can be considerably faster than the rate of chemical reactions. For instance, Buisman et al. (1990) found that biological sulfide oxidation was 6–75 times higher than the noncatalyzed chemical oxidation of sulfide. A particular environment will favor some species over others. The rate of growth is influenced by chemical and physical variables. Many of these have been described in Section 6.2.1 with reference to bacteria, but the same principles hold for all other organisms. The presence or absence of oxygen is a primary growth determinant. Temperature and pH are primary control variables. The availability of substrates (food; nutrients) and other minerals and compounds all influence the rate of metabolism and growth. As expected, toxins inhibit metabolism and growth. The influence of pH, temperature, and other environmental conditions on the rate of metabolism and growth for a single species or a group of microorganisms can be systematically evaluated by holding conditions constant and varying one of the parameters. Generally, a bellshaped response curve will be obtained.

6.5.1 Growth of Pure Cultures Most of the prokaryotes and unicellular algae, fission yeasts, and most protista reproduce by binary fission or budding. This leads to exponential growth of a pure culture under ideal conditions. Ideal conditions consist of an excess availability of substrates and other growth requirements as well as a lack of inhibitory substances. Consider a single microorganism in a vessel (batch reactor) containing an ideal medium for the species. The generation time, g, is the time for a cell to divide into two daughter cells. The number of cells would increase as follows: 1

g

!2

g

!4

g

!8

g

! ∙∙∙

The number of microorganisms present at any time is

N ˆ N 0 2t=g ˆ N 0 ekt

(6.1)

where k is a rate constant and k ˆ 1g ln…2†, N is the number of microorganisms, t is time, and N0 is the number of microorganisms at t = 0.

6 Microbiology

The generation time may not be constant for each cycle or for all cells in the population; therefore, g represents the average generation time for the culture. The rate of increase of microorganisms is dN ˆ kN dt

(6.2)

The rate of substrate removal is proportional to the rate of microorganism growth dS ∝ dt

dN dt

or

dS ˆ dt

1 dN Y N dt

(6.3)

where S is the substrate concentration (mass/volume) and YN is a yield factor (number of microorganisms formed/mass of substrate removed). Eventually, substrate (i.e., one principal nutrient such as carbon or phosphorus) or a trace element will reach a limiting supply for the number of microorganisms present. At this time, the growth rate will begin to decrease. It is also possible that a buildup of toxic byproducts of metabolism (including hydrogen ions leading to low pH) is responsible for growth retardation. Assuming that the substrate is the limiting factor, the rate of microbial growth becomes dependent on the amount of substrate present as well as the number of microorganisms in the vessel. dN ˆ k ´ SN dt

(6.4)

where k´ is a rate constant [=k/K in Eq. (6.5)]. After this stage, any combination of limiting substance or toxic by-product buildup will cause the number of viable microorganisms to decrease. A complete growth cycle for a pure culture (or “seed”) of microorganisms started in a vessel containing an ideal medium is shown in Figure 6.5. The growth phases are indicated on the diagram. This is a general cycle, and not all growth (or death) phases may be manifested in every situation. The first phase is an acclimation (or lag) phase during which the culture fabricates the metabolic machinery (enzymes) to handle the substrate. During this phase, the growth rate may be quite low or even nonexistent, depending on the history of the seed. This phase is followed by the log or exponential growth phase. The only factor limiting growth during this phase is the number of microorganisms present. At the point where a nutrient becomes limiting or where the concentration of a toxic by-product becomes significant, the declining growth phase begins. There is still a net increase in the number of viable micro­ organisms. After a period of time, a stationary phase is reached. There is no change in the number of viable microorganisms during this phase.

Figure 6.5 Growth of a pure or mixed culture in an ideal medium.

133

134

Theory and Practice of Water and Wastewater Treatment

Following the stationary phase, the supply of substrate dwindles past the critical amount required to sustain the numbers present, alternatively toxins or a combination of toxins and lack of substrate cause numbers to decline. This decline often accelerates into a log death phase. Note that at the end of the cycle, the number of viable organisms will be very low, but not zero. The death phase and following endogenous phase are phases where cells predate on other viable cells or debris resulting from the death of cells. All phases may be variable in length or even nonexistent depending on the species, the medium, and the growth history. Monod (1949) has given a comprehensive discussion of microbial growth. From experience, he formulated an empirical expression that describes growth and substrate removal during the log growth and substrate limiting phases. dS ˆ dt

k S X YX K ‡ S

(6.5)

where k is the maximum specific rate of substrate removal (mass of substrate removed/mass of microorganisms/time), K is the half-velocity constant (mass/volume), X is the concentration of microorganisms (mass/volume), and YX is a yield factor (mass of microorganisms formed/mass of substrate removed). Equation (6.5) is formulated in terms of concentrations. The concentration of microorganisms may be expressed in terms of numbers or mass per unit volume. The mass concentration is more commonly used because of the ease of its determination. It is recognized that individual cells within the culture may vary in mass and that mass of the cells may change with time. In addition, nonviable cells may also be included in the mass. The constant K is also called the half-velocity constant. This hyperbolic equation is of the same form as the Michaelis–Menten equation [Eq. (4.9), also see Figure 4.12] developed from theory for enzyme kinetics. The analogy between enzyme kinetics and microbial growth is reasonable because, indeed, it is the microbial enzymes that regulate growth; thus, it inspired Monod’s description. It can be shown that Eq. (6.5) reduces to Eqs. (6.3) and (6.4) at high and low values of substrate concentration, respectively, using the yield factor YN, given in Eq. (6.3). The yield factor is precisely defined as the average yield of microorganisms produced from the removal of a unit of substrate under ideal growth conditions. Die-off of microorganisms will be negligible in these circumstances, and YN is known as the true yield constant. Thus at high values of substrate concentration, growth will be independent of the substrate concentration (S), whereas at low values, growth will be first-order with respect to S. Example 6.1 Growth of a Pure Culture The data in the table below were obtained for the growth of a pure culture of E. coli in nutrient broth at a temperature of 37 °C. Determine the generation time of E. coli and the duration of the lag and log growth phases. Comment on the suitability of turbidity as a measure of viable microorganisms. The bacteria count data were plotted on a semilog scale, and the turbidity data were plotted on an arithmetic scale in the figure below. Time (h)

0

1

2

3

4

5

6

7

12

24

Bacteria 4.30 × 104 3.50 × 104 8.90 × 104 5.20 × 105 2.60 × 106 1.60 × 107 7.10 × 107 1.20 × 108 2.40 × 108 5.90 × 108 (No./mL) Turbidity (TU)

0

0

0

0

0.4

0.8

7.5

20.4

22.6

45.1

6 Microbiology

From the plot, the lag phase lasted approximately 1.5 hours. The log growth phase occurred between 1.5 and 6 hours for a duration of nearly 4.5 hours. Regression of the count data between 2 and 6 hours using Eq. (6.1) yielded a slope of 1.68 h 1. The generation time was gˆ

ln 2 0:693 ˆ ˆ 0:413 h ˆ 24:8 min k 1:68 h 1

When the turbidity curve is visually compared to the bacterial numbers curve, turbidity is a reasonable measure of viable mass during the growth phase, but turbidity is not an accurate measure of bacterial numbers when suspended debris accumulates in the culture after death begins to occur to a significant extent. The curves indicate that there was a significant amount of die-off between 12 and 24 hours, although the culture was primarily in a stationary phase over this period.

The yield factor may also be expressed with units of chemical oxygen demand (COD) of biomass formed per COD of substrate transformed or removed. The yield factor with the above units is simply multiplied by the COD per unit of biomass. The yield factor expressed on a COD or biochemical oxygen demand (BOD) basis is always less than 1. For every unit of substrate removed, a portion of the substrate will be synthesized into biomass, whereas the remainder will be used (oxidized) to provide energy for cell synthesis and other life functions. 6.5.2 Growth of Mixed Cultures The growth of mixed cultures in batch reactors is quite similar to the growth of pure cultures. Mixed cultures, because of the variety of microbial species contained within them, are better able to survive and grow in any given situation. The probability of a few of the species in the mixed culture being adapted to the medium at inoculation is higher because of the diversity of organisms present. There will likely be a continuous succession of dominant species in a mixed culture as substrate availability and environmental conditions change within the medium. The viability of the culture as a whole is extended in the log death phase because the waste products of species and dead cells from starved species become substrate for other species. This phase is known as the endogenous phase

135

136

Theory and Practice of Water and Wastewater Treatment

(ancient Greek endo, “inside” and -genous, “born from”) for mixed cultures. Predation can play a significant role in all phases, and especially in the last one, where protozoa and phages can proliferate. Equations (6.2)–(6.5) apply to the growth of mixed cultures as well as to pure cultures. 6.5.3 Viability and Mass in Growing Cultures The equations in the previous sections have been formulated in terms of numbers or mass of viable microorganisms. Evaluation of these quantities in environmental engineered systems is not straightforward, and special considerations should be included in the technical approaches. First, the relationship between the cell numbers and the mass of viable organisms is dependent on the growth rate and on the composition of the nutrient sources consumed by the population. Consequently, in describing growing systems such as wastewater or subsurface treatment systems, cell numbers are almost never used in mass balances. The application of cell numbers for engineering design is typically restricted to nongrowing systems such as disinfection. Second, it is now well established that volatile suspended solids (VSS; see Section 5.3.1) contains living biomass as well as nonbiodegradable and nondegradable organic materials. This recognition that came initially from modeling exercises, forced the development of other laboratory techniques to estimate the proportion of live biomass in the mixture of organic material. Some of the new molecular techniques (Section 6.5.5) could provide these estimates for defined microbial populations. More generally, the concentration of ATP has been used to approximate viable fractions of VSS. Commercially available kits are available for on-site use by wastewater treatment plant operators. 6.5.4 Enumeration of Microorganisms Enumeration techniques should also effectively identify the microorganisms of interest, or at minimum, separate them from other microorganisms within the same sample in order to perform an accurate count. Because of the microscopic size of microorganisms, it is necessary and convenient to develop measurement methods that produce more readily observed results. There is no chemical analysis that can definitively identify a complex entity such as a microorganism. However, chemical agents can be associated with a culture either through their production of the chemical (for example, a specific enzyme, or a particular gas) or through requirements by the bacteria of specific nutrients, and these can be important tools in devising separation and enumeration techniques. Most techniques in an industrial setting (such as a water or wastewater treatment plant) rely on the rapid growth of microorganisms, where one microorganism can produce a visible colony or cause identifiable chemical reactions within a short period. Modern laboratory techniques, such as polymerase chain reactions (PCRs), use molecular (DNA- and RNA-based) methods and other microbial approaches to be able to more precisely identify the microorganism under study. They can also detect the presence of microbes that will not show up when using plate counting or similar techniques, which rely on metabolism and growth. Note that when performing microbial enumeration in the laboratory, it is always assumed that the laboratory procedures are conducted under sterile conditions, with sterilized materials (glassware, membrane filters (MFs), and media) and using aseptic techniques. Plate Counts Direct Plate Counts

There are multiple applications of plate counts in the overall field of microbiology, but for water and wastewater analysis, the heterotrophic plate count (HPC, formerly known as the standard

6 Microbiology

plate count) is most common. As the name implies, it is designed to measure the total heterotrophic bacterial population in water. With this objective, it is necessary to design a medium that will support the growth of as wide an array of bacterial heterotrophs as possible in a reasonable period. This requirement is a particularly severe constraint because of the diversity of species, their individual growth needs (temperature, nutrients, growth factors, and others) that have been discussed previously, and their competitive abilities in laboratory media. Therefore, no medium and other environmental conditions will suffice for all microorganisms. In fact, only a small fraction of species can typically be retrieved by cultivation from any environment, and it is believed that globally less than 1% of the species have ever been cultured. However, there are some general-purpose media that do satisfy the metabolic needs of the total heterotrophic microbial population, and one of these would be employed in the method. Incubation temperatures are either 20 or 35 °C. The first step in the HPC procedure is to thoroughly disperse the aggregates in the original samples and then prepare a dilution series as necessary. A small volume, 0.1–1 mL, is withdrawn from the sample and is either added to a petri dish with the medium that contains melted agar (44–46 °C) of the specific rich HPC medium that has previously been sterilized or is spread on the surface of the hardened medium. The covered petri dish is placed in an incubator. Most of the microorganisms that can grow will produce a visible colony in 24–48 hours, although in some instances, incubation times as long as 4 days may be required to achieve the best results. The colonies are counted to compute the concentration based on the volume of sample that has been added. Further information on this and other enumeration methods can be found in Standard Methods (APHA et al. 2012). The HPC is used as a monitor of water quality in water treatment operations. The test can provide a general measure of microorganism activity in natural waters and wastewater treatment operations. Membrane Filter Techniques

MF analyses have developed into a formidable tool for the identification and enumeration of viable microorganisms. The procedure as outlined in Standard Methods (APHA et al. 2012) consists of passing a known volume of sample through a filter that contains openings that are smaller than the microorganisms being analyzed. A vacuum is required to accomplish the filtration. The filter is then placed in a petri dish containing a selective medium (which encourages growth of the desired species, while discouraging the growth of others) and a differential medium (which contains a visible indicator for detecting the target species). The dish is incubated at the appropriate temperature. Single bacteria will give rise to visible colonies during incubation. Colonies exhibiting the characteristic reaction are counted (Figure 6.6). This direct count is assumed to be the number of microorganisms from the group that was present in the volume of sample that was filtered. The technique assumes that each colony is derived from a single cell, and it is essential that the sample be well shaken to break up clumps and distribute the microorganisms on the filter. The analyses are usually performed in duplicate or triplicate to improve the accuracy of the result and check its consistency. It is usually necessary to perform serial dilutions of the sample, as no firm estimate of the final count is available a priori. These dilutions would be designed to bracket the final colony count, ideally 20–80 but not more than 200 colony forming units (CFU)/100 mL. Effective selective and differential media have been developed for a number of microorganisms of importance to water supply and sanitation engineers. MF techniques have been tested against most probable number (MPN) techniques to ensure their reliability. The MF technique is rapid and convenient. If a high concentration of suspended solids (SS) exists, filtration time may be too lengthy, and the MPN technique is the only alternative.

137

138

Theory and Practice of Water and Wastewater Treatment

Figure 6.6 Counting colonies on a membrane filter.

Most Probable Number Techniques

MPN techniques (Greenwood and Yule, 1917) consist of taking a series of aliquots from a sample containing the culture and placing the aliquots in several separate test tubes containing a culture medium to form a set. Typically, 10× serial dilutions using test tubes are performed to create the aliquots and inoculate different sets. This means that starting with the original sample in the first set of test tubes, each successive set of test tubes receives one tenth of its volume from the previous test tube. If an aliquot contains one or more of the microorganisms for which the test is designed, a positive reaction will occur in the test tube. The number of positive tubes in a set is related to the density of microorganisms in the original dilution, and the cell count in the original sample can be back-calculated. The MPNs are based on standard statistical analysis. It is common to examine three sets of sample dilutions, replicated five times. The first set consists of five 10 mL aliquots from the sample, the second consists of five 1 mL aliquots, and the third set consists of five 0.1 mL aliquots. The Poisson distribution describes the probability of r occurrences of an event when the mean frequency of occurrence is μ. The expression (Kennedy and Neville 1986) is pr ˆ

e μ μr ˆp r!

(6.6)

where pr (or p) is the probability of r occurrences of the event (finding bacteria in a sample). A binomial distribution (Kennedy and Neville 1986) describes the probability of an event succeeding z times in n trials. The expression is pz ˆ

n! pz …1 z!…n z†!

p†…n

z†

(6.7)

If the concentration of bacteria in the sample is C (No./mL) and the volume of a sample aliquot is Vi (mL), the mean number of bacteria in the volume is μ ˆ V iC

(6.8)

As noted above, only one bacterium is needed to provide a positive tube; therefore, Eq. (6.8) is substituted into Eq. (6.6) with r set equal to 0, resulting in Eq. (6.9a), which gives the

6 Microbiology

probability of a tube not being positive. Because all probabilities sum to 1, the probability of finding one or more bacteria in a sample and thus a positive tube when the mean number is μ is given by Eq. (6.9b). pˆe 1

V iC

pˆ1

(6.9a) e

V iC

(6.9b)

Using the above equations, the probability of finding a, b, and c positive tubes in three sets of tubes with n1, n2, and n3 tubes in each set where sample volumes of V1, V2, and V3 are used, respectively, is pa;b;c ˆ

a!…n1  1

n1 !n2 !n3 ! a†!b!…n2 b†!c!…n3 e

V 2C b

e

c†!

V 2 C …n2 b†

1

1

e e

V 1C a

V 3C c

e

e

V 1 C …n1 a†

V 3 C …n3 c†

(6.10)

where pa,b,c is the probability of finding a, b, and c positive tubes. From an analysis, a, b, and c will be determined for volumes V1, V2, and V3 with n1, n2, and n3 sets of tubes, respectively. The problem is to estimate the bacterial density that is most likely to result in these numbers of positive tubes. A given concentration of bacteria results in a given probability. Statistical techniques of maximum likelihood estimation are used to transform the data, weight it, and assume an underlying distribution of the small number of samples. McCrady (1915) put forward the basic technique, which has been refined by many authors (see, for example, Finney 1978). The MPN is the bacterial density that produces the highest probability with a given set of constants. A procedure that gives an approximate answer without considering weighting factors and other corrective measures is to find the value of C that produces the maximum value of Eq. (6.10) by solving the following equation: dpa;b;c ˆ0 dC

(6.11)

Alternatively, one may plot pa,b,c for a range of values of C, and the C giving the highest value of pa,b,c will be the MPN. Standard deviations, confidence limits, and other measures of spread of the MPN have been tabulated. See Table A.1 (or Standard Methods, APHA et al. 2012) for MPN values for selected aliquot combinations. It may be necessary to subject the positive tubes from the preliminary analysis to one or more tests or enrichment media. This additional information finally confirms the presence of the suspected species or group in the positive tubes. The MPN technique is often the method against which other bacteriological analyses are compared to assess their accuracy. Enzymatic Defined Substrate Technologies

These technologies use proprietary media that select for particular bacteria and simultaneously emit a characteristic color, if the bacteria grow, to enable detection as well as quantification. The test can be set up as for the MPN technique to yield a MPN value. One example, ColilertTM, simultaneously detects or quantifies both total coliforms and E. coli, with results in 24 hours. Similar systems are available for enterococci, Pseudomonas aeruginosa, HPC, and Legionella pneumophila. The advantage is the simplicity of the method, as the reagents and glassware are all in kit form, ready to be used, and neither filtration nor dilution are needed. Disadvantages are the costs of the kits and equipment.

139

140

Theory and Practice of Water and Wastewater Treatment

Online Methods

All the abovementioned techniques for detection and enumeration involve significant time delays, ~24 hours or more. This is a clear problem for water supply managers, who should alert consumers without delay if indicator organisms are present. Online methods for determining the presence of these organisms in real time or at least within only a few hours is obviously desirable, but reliability is critical, as false positives are almost as problematic as false negatives. One type of method uses the presence of specific enzymes for detection. An example is the ColiMinder® measurement system, which uses fluorescence spectroscopic detection of E. colispecific enzymatic activity (Koschelnik et al. 2015). Sensor development and other online techniques (such as optical methods with high-resolution, 3-D, imaging, and image recognition (Højris et al. 2016), as well as flow cytometry, are rapidly evolving and are continually under development. Miles et al. (2011) evaluated the ability of four different types of online sensors, which are commercially available, to detect E. coli in tap water. They reported that the sensitivity of all four sensors, from a drinking-water perspective, was poor. Nevertheless using such sensors for extended periods might provide utilities with baseline data and would give more timely alerts if intrusions occur, or of equipment failure. Practical Considerations in Determining Mean Values

Arithmetic and geometric means are used in reporting coliform count values. The geometric mean is commonly used because counts are often log-normally distributed. Outliers (i.e., single very large or very small values) will hardly affect the geometric mean. The median is also useful in describing data. These measures are defined as follows: Σyi n p n Geometric mean ˆ y1  y2  ∙ ∙ ∙  yn

Arithmetic mean ˆ

(6.12) (6.13)

where n is the total number of observations and yi is an individual observation. Median: if n is odd, use the middle value if n is even, use the average of n2 and n2 ‡ 1 values. The example below discusses some problems in describing bacteria concentration data. Example 6.2 Interpretation of Microorganism Assay Data Weekly fecal coliform counts (CFU/100 mL) are reported below in ascending order. Counts : 0; < 25; 60; 85; 168; 234; 330; TNTC where TNTC means too numerous to count. How should these data be reported for median, arithmetic mean, and geometric mean? Some problems encountered in answering these questions are the following:

 If 0 is considered, the geometric mean will be 0 regardless of other data.

 What value does one use for the 1

Gelatin capsule

Type 12

17 PFU

1

Oral suspension

Respiratory

Enteric

Norovirus Echovirus

3

TCID50

a) As determined by virus shedding and/or increase in antibody titer.

b) 50% tissue culture infective dose.

c) Focus forming units.

d) Plaque-forming particles.

Source: Yezli, S and JA Otter (2011), “Minimum Infective Dose of the Major Human Respiratory and Enteric Viruses

Transmitted Through Food and the Environment,” Food Environ Virol, 3, 1, 1–30”. Reproduced with permission of

Springer.

attempts have been made to provide guidance to health practitioners and engineers responsible for protecting public health. These doses are based on animal models, epidemiology, or in some cases, healthy human volunteers, and must be interpreted with caution. Yezli and Otter (2011) provided information in Table 7.12 for infectious doses for the most common respiratory and enteric viruses.

Questions and Problems 7.1 List three pathogens and their associated diseases from the bacteria, protozoan, and virus groups. 7.2 Describe the life cycle of the helminth responsible for bilharzia. 7.3 Explain why construction of the Aswan Dam caused a rise in the incidence of schistosomiasis. 7.4 Consult local authorities to find incidents of waterborne diseases in your community over the past 10 years. What were the agents and reasons for the outbreaks? 7.5 Is there a significant risk of HIV (AIDS) infection from water?

163

164

Theory and Practice of Water and Wastewater Treatment

7.6 Describe indicator, test, and model microorganisms. 7.7 What are the characteristics of an ideal indicator microorganism for fecal contamination? Discuss each group of indicator microorganisms with respect to characteristic 1. 7.8 What is the role of a surrogate as an indicator of water quality? 7.9 What is a coliphage? 7.10 Why is salt incorporated at higher than normal concentrations in media for enterococci? 7.11 If the indicator count decreased from 1.12 × 106/100 mL to 4.83 × 104/100 mL in 36 hours, what is the t90 in this situation? 7.12 How many days would it take coliform bacteria in the groundwater stream at 10 °C to decrease 90%, 99%, 99.9%, and 99.99% from the initial concentration (see Table 7.10)? 7.13 On a day with average cloud cover, the average sunlight intensity at a location was 2.12 MJ m 2 h 1. The sun was above the horizon for 14.4 hours. If kS is 0.12 m2 MJ 1, what is the average die-off rate constant during this period? 7.14 If a rapid test for Salmonella species enumeration were developed, would it be a suitable indicator microorganism? 7.15 Would FC or E. coli be a reliable indicator for the helminth, A. lumbricoides (that causes ascariasis)? 7.16 Samples from a stream had an initial ratio of enterococci to FC of 9.0 that decreased to a ratio of 1.0 after 3.20 day travel time in the stream. The t90 for FC was 58 hours. What was the die-off coefficient for enterococci? 7.17 If the average die-off coefficient for an indicator microorganism was 0.50 h 1 during daylight hours and 0.25 h 1 during darkness, what is the average die-off coefficient over a 24-hour period where daylight and darkness periods each equal 12 hours?

References Anderson, E.J. and Weber, S.G. (2004). Rotavirus infection in adults. Lancet Infect. Dis. 4 (2): 91–99. doi: 10.1016/S1473-3099(04)00928-4. APHA, AWWA and WEF (2012). Standard Methods for the Examination of Water and Wastewater, 22e (ed. E.W. Rice, R.B. Baird, A.D. Eaton and L.S. Clesceri). Washington, DC: American Public Health Association. Beer, K.D., Gargano, J.W., Roberts, V.A. et al. (2015). Surveillance for Waterborne Disease Outbreaks Associated with Drinking Water – United States, 2011–2012. Morbidity & Mortality Weekly Rep. CDC (USA), 64, 31, 842–848. Blaustein, R.A., Pachepsky, Y., Hill, R.L. et al. (2013). Escherichia coli survival in waters: temperature dependence. Water Res. 47 (2): 569–578. doi: 10.1016/j.watres.2012.10.027.

7 Water, Wastes, and Disease

Boehm, W.A. (2006). Waterborne Pathogens. AWWA Manual M48, 2e. Denver, CO: American Water Works Association. Boehm, A.B. and Sassoubre, L.M. (2014). Enterococci as indicators of environmental fecal contamination. In: Enterococci: From Commensals to Leading Causes of Drug Resistant Infection [Internet] (ed. M.S. Gilmore, D.B. Clewell, Y. Ike, et al.). Boston: Massachusetts Eye and Ear Infirmary. Cabral, J.P.S. (2010). Water microbiology: bacterial pathogens and water. Int. J. Environ. Res. Public Health 7 (10): 3657–3703. doi: 10.3390/ijerph7103657. Falkinham, J.O. III (2015). Common features of opportunistic premise plumbing pathogens. Int. J. Environ. Res. Public Health 12 (5): 4533–4545. doi: 10.3390/ijerph120504533. Feachem, R.G., Bradley, D.J., Garelick, H., and Mara, D.D. (1981). Health Aspects of Excreta and Sullage Management – A State-of-the-Art Review. Washington, DC: The World Bank. Fox, J.C., Fitzgerald, P.R., and Lue-Hing, C. (1981). Sewage Microorganisms: A Color Atlas. Chelsea, MI: Lewis Publishers. Geldreich, E.E. (1978). Bacterial populations and indicator concepts in feces, sewage, stormwater and solid wastes. In: Indicators of Viruses in Water and Food (ed. G. Berg), 51–97. Ann Arbor, MI: Ann Arbor Science Publishers. Guillot, E. and Loret, J.F. (2010). Waterborne Pathogens: Review for the Drinking Water Industry. London: IWA Publishing. Health Canada (2013). Guidance on the use of the microbiological drinking water quality guidelines. http://www.hc-sc.gc.ca/ewh-semt/alt_formats/pdf/pubs/water-eau/micro/micro-eng. pdf (accessed March 2017). IAWPRC Study Group (1991). Bacteriophages as model viruses in water quality control. Water Res. 25 (5): 529–545. doi: 10.1016/0043-1354(91)90126-B. Johnson, R.W., E.R., Blatchley, I.I.I., and Mason, D.R. (1994). HIV and the bloodborne pathogen regulation: implications for the wastewater industry. Water Environ. Res. 66 (5): 684–691. doi: 10.2175/WER.66.5.4. Kadir, K. and Nelson, K.L. (2014). Sunlight mediated inactivation mechanisms of Enterococcus faecalis and Escherichia coli in clear water versus waste stabilization pond water. Water Res. 50 (1): 307–317. doi: 10.1016/j.watres.2013.10.046. Lopman, B.A., Hall, A.J., Curns, A.T., and Parashar, U.D. (2011). Increasing rates of gastroenteritis hospital discharges in US adults and the contribution of norovirus, 1996–2007. Clin. Infect. Dis. 52 (4): 466–474. doi: 10.1093/cid/ciq163. Mitchell, R. and Chamberlin, C. (1978). Survival of indicator microorganisms. In: Indicators of Viruses in Water and Food (ed. G. Berg), 15–37. Ann Arbor, MI: Ann Arbor Science Publishers. Munro, P.M. and Colwell, R.R. (1996). Fate of Vibrio cholerae in seawater microcosms. Water Res. 30 (1): 47–50. doi: 10.1016/0043-1354(95)00137-A. Riggs, J.L. (1989). AIDS transmission in drinking water: no threat. J. Am. Water Works Assoc. 81 (9): 69–70. http://www.jstor.org/stable/41292599. Porter, R. (ed.) (2017). Merck Manual Professional Version (Known as the Merck Manual in the US and Canada and the MSD Manual in the rest of the world). Kenilworth, NJ: Merck Sharp & Dohme Corp., a subsidiary of Merck & Co, Inc. https://www.merckmanuals.com/professional (accessed March 2017). Sinton, L.W., Davies-Colley, R.J., and Bell, R.G. (1994). Inactivation of enterococci and fecal coliforms from sewage and meatworks effluents in seawater chambers. Appl. Environ. Microbiol. 60 (6): 2040, PMCID: PMC201599–2048. Sinton, L.W., Hall, C.H., Lynch, P.A., and Davies-Colley, R.J. (2002). Sunlight inactivation of fecal indicator bacteria and bacteriophages from waste stabilization pond effluent in fresh and saline waters. Appl. Environ. Microbiol. 68 (3): 1122–1131. doi: 10.1128/AEM.68.3.1122-1131.2002.

165

166

Theory and Practice of Water and Wastewater Treatment

USEPA (2014). Revised total coliform rule assessments and corrective actions guidance manual. https://www.epa.gov/sites/production/files/2015-10/documents/revised_total_coliform_ rule_assessments_and_corrective_actions_guidance_manual.pdf (accessed March 2017). WHO (2016). Health statistics and information systems. http://www.who.int/healthinfo/ global_burden_disease/metrics_daly/en (accessed 14 March 2018). Yezli, S. and Otter, J.A. (2011). Minimum infective dose of the major human respiratory and enteric viruses transmitted through food and the environment. Food Environ. Virol. 3 (1): 1–30. doi: 10.1007/s12560-011-9056-7.

167

8 Water Constituents and Quality Standards Constituents in water supplies that are of primary concern are those with health implications. The majority of these constituents are dissolved compounds, but small particulates such as microbes are also included. Constituents that affect health are identified, and in most cases, much effort is required to establish acceptable concentrations in water. Municipal water supply is intended to promote the general well-being of the population; industrial and other users of water have their own quality requirements for optimal benefits. Wastewater must be similarly evaluated for its effects on the ecosystem. Any substance that is not naturally present in waters is suspect for adverse impacts in the environment. Significant change, particularly elevation, in the concentration of naturally occurring substances can also be problematic. The best approach is reduction or recovery and reuse of any components that are significantly changed from their naturally occurring levels. This chapter presents a cursory overview of constituents and contaminants in natural waters and water for use by humans. Regulations governing allowable or desirable concentrations or exposures are an evolutionary process, changing as more research becomes available. Fortu­ nately, the Internet has made regulations and supporting documents readily available. Although information on standards presented in this chapter are current at publication, the reader is advised to consult the web for the latest regulations or guidelines. Not only do values change but also the web addresses change as regulatory bodies change; therefore, a careful search is sometimes required.

8.1 Toxicity of Elements and Compounds Out of the more than 90 elements that are naturally found on the earth, only 22 are essential to all living systems (Nelson et al. 2008). Ten are considered “bulk” elements (gram quantities per day required in the human diet) and twelve are “trace” (milligrams per day). Therefore, there has been selectivity in the evolution of living systems. Besides the laws of chemistry and thermodynamics, which govern biochemical phenomena, evolution is a process of inheritance as well as change. The elements that form the backbone of organic biomolecules are present in the highest amounts. Other elements are present in varying amounts. Table 8.1 lists the 22 elements found in all living systems. Elements not listed in Table 8.1 have varying effects on living systems, ranging from innocuous to extremely toxic. Natural defense mechanisms have evolved in organisms, allowing them to cope with harmful substances to a limited concentration. Some elements that are harmless or even essential to living systems may cause harm to ecosystems when their concentrations rise

Theory and Practice of Water and Wastewater Treatment, Second Edition. Ronald L. Droste and Ronald L. Gehr.

 2019 John Wiley & Sons Inc. Published 2019 by John Wiley & Sons Inc.

Companion website: www.wiley.com/go/droste/water

168

Theory and Practice of Water and Wastewater Treatment

Table 8.1 Elements found in living systems. Groups of elements

Remark

C, Ca, Cl, H, N, Na, K, O, P, S

The major constituents of organic molecules – the “bulk” elements

Co, Cr, Cu, Fe, I, Mg, Mo, Mn, Ni, Se, V, Zn

Elements required at trace levels

beyond natural levels. A pollutant is often defined as a substance that is “out of place” in the environment. Imbalances in a few elements or compounds can result in total disruption of an ecosystem. Anthropogenic sources disperse substances into the environment, threatening ecosystems that have evolved over long periods of time. Mercury is an element that illustrates the variability in toxicity response of some substances. Mercury has been used for centuries in various applications such as in tooth amalgams, as a fungicide, and in many industrial uses. Elemental mercury is relatively harmless unless it is vaporized and inhaled directly into the lungs. The toxic forms of mercury for ingestion are methyl mercury (CH3Hg+ and CH3
Hg
CH3) and inorganic salts, particularly mercuric chloride (HgCl2). Methyl mercury can be formed by bacteria in sediments and acidic waters. Inorganic ionized mercury is acutely toxic (Kim et al. 2016). Elemental mercury has a relatively short residence time in the body, but methyl mercury compounds stay in the body 10 times longer than metallic mercury and they cause malfunction of the brain, nervous system, kidneys, and liver as well as birth defects. Methyl mercury is cumulative in the food chain. Phenyl mercury compounds (C6H5Hg+ and C6H5
Hg
C6H5) are moderately toxic with short retention times in the body but these compounds are rapidly transformed in the environment to release inorganic mercury. Other heavy metals besides mercury can be methylated biologically or abiotically. Formation of organometal compounds affects their accumulation and toxicity to humans and other life forms. Ferguson (1990) has discussed chemistry and environmental impacts of heavy metals. The variety of synthetic organic compounds is enormous. The bank of data describing their toxicity and environmental effects is growing continuously, and only a few of the classes or specific compounds are commented on in this book. A good reference on environmental data for organic chemicals has been compiled by Verschueren (2009). A survey of harmful effects of some elements and compounds is presented in Table 8.2; the survey is by no means exhaustive. Table 8.2 A sampler of toxic elements and compounds. Substance

Remarks

Antimony, Sb

Little research on Sb. Accumulates in liver and is detrimental to the heart in humans. Can be accumulated by marine organisms

Arsenic, As

Acute or chronic toxicity to humans. Long-term exposure causes dermal changes. Toxic to all life. Naturally occurring in some locations and a byproduct of smelting ores and used in other industries

Barium, Ba

Ingested Ba salts are highly toxic to humans. Usually found in trace amounts in natural waters, but surface water concentrations are sometimes as high as 0.340 mg L 1. May be toxic to plants if present above trace amounts

Beryllium, Be

Extremely toxic to all life. Usually naturally present in concentrations less than 0.0001 mg L 1 in surface waters. Oxides and hydroxides are insoluble within normal pH ranges

8 Water Constituents and Quality Standards

Table 8.2 (Continued ) Substance

Remarks

Boron, B

No evidence of accumulation in humans. Large amounts may produce digestive difficulties and nerve disorders

Bromine, Br

Free bromine, Br2, is a strong oxidant not found naturally. Bromide salts are harmless. Bromate (BrO3 ) is a known animal carcinogen. Some organic bromides are carcinogenic

Cadmium, Cd

Cumulative, highly toxic in humans and livestock. Affects all life. Protects other metals against oxidation; also used in other industries

Chlorine, Cl

Cl2 is a strong oxidant not found naturally, but it is manufactured and commonly used as a disinfectant. Chloride salts are harmless. Some compounds produced from reactions of chlorine with organic matter are carcinogenic

Chromium, Cr

Natural Cr is rare. Cr(VI) is the toxic form to humans. Cr(III) is slowly oxidized to Cr(VI) in waters. Toxic to plants. Varying tolerance to Cr salts in aquatic life. Used in various industries

Cobalt, Co

Low toxicity to humans; essential in trace amounts

Copper, Cu

Essential to humans in small daily amounts (2.0 mg). Upper limits are not determined but water is very distasteful at 1–5 mg L 1 Cu. Essential to all life but is toxic at differing levels to plants and aquatic life

Cyanide, CN

Cyanide renders tissues incapable of oxygen exchange. It is not cumulative and it is biodegradable in streams. CN behaves like halides

Fluorine, F

Fluoride has been shown to reduce dental caries. Above concentration guidelines, there is no further reduction in caries but mottling increases. Natural F concentrations are generally low but wide fluctuations occur

Lead, Pb

Cumulative in humans and livestock. Human absorption of ingested lead is small; single large doses are not a problem

Lithium, Li

Higher concentrations cause phytotoxicity

Mercury, Hg

Toxic to all forms of life. Mercury is very slowly excreted from the human body. Methyl mercury is 50 times more toxic than inorganic mercury

Molybdenum, Mo

Essential micronutrient. Does not accumulate and humans can tolerate large quantities. Plants accumulate Mo in their foliage but at normal concentrations in water it is not harmful. Some grazing animals, e.g., cattle, sheep and swine, exhibit sensitivity to this metal

Nickel, Ni

Low oral toxicity to humans. Toxic to plants and marine life

Nitrogen, N Ammonia, NH3

Nontoxic to humans at natural levels. Fish cannot tolerate large quantities of ammonia

Nitrate, NO3

Toxic to infants at high concentrations

Nitrite, NO2

More toxic than nitrate but less chemically stable than nitrate and generally found in low concentrations

Organic N

No health effects per se

Phenol

Taste and odor from these compounds are more significant than their toxicity. They exhibit direct toxicity to fish

Selenium, Se

Cumulative poison in humans and animals. Moderately toxic to plants

Silver, Ag

Cumulative in human tissue resulting in blue–gray discoloration of skin (argyria). Toxic to aquatic organisms (continued )

169

170

Theory and Practice of Water and Wastewater Treatment

Table 8.2 (Continued ) Substance

Remarks

Sodium, Na

Harmful to some persons with cardiac problems. Destructive to soils

Strontium, Sr

The stable isotope is a widespread minor component of igneous rocks. It is similar to calcium and of no health significance. Industrial sources of radioactive strontium are insignificant

Sulfide, S2

Undissociated hydrogen sulfide (H2S) is the toxic entity. Also toxic to aquatic life

Thallium, Tl

Cumulative poison with sublethal effects such as hair loss and hypertension. Thallium has been shown to inhibit photosynthesis and acts as a neuropoison to fish and aquatic invertebrates

Tungsten, W

Highly insoluble in water and little information on health effects

Uranium, U

Uranium and its salts are quite toxic to humans and have also been reported to be toxic to aquatic organisms

Vanadium, V

Low concentrations are not toxic to humans. Toxicity has been demonstrated in plants

Zinc, Zn

Relatively nontoxic to humans and animals. Essential nutrient for life. Only at high concentrations has it been found toxic to plants. However, zinc is acutely and chronically toxic to aquatic organisms

Source: Adapted from Inland Waters Directorate (1979) and other sources.

8.2 Contaminants in Water Contaminants enter a water supply from natural and anthropogenic sources. Direct discharge of effluents from population centers and industries are point sources. Stormwater runoff, which may be a point (at a sewer outfall) or nonpoint source (e.g., runoff from a field), washes residues of products falling onto roads or applied to the land. Dust and other air pollutants are transported by air and deposited via various forms of precipitation into receiving waters. Contamination of water by human and animal wastewaters is the major source of waterborne diseases. The application of certain agents such as chlorine and aluminum (discussed below) in water treatment processes is associated with low-risk contamination. Some harmful disinfection byproducts result from the application of chlorine and other disinfectants. This has led to extensive research and change in disinfection practice in the last decade. These issues are further discussed in Chapter 16. Corrosion of materials in the distribution system or household plumbing and appurtenances may be a significant source of contaminants in drinking water. Iron is a common contaminant, imparting an orange tinge to the water, but there is no health impact. Copper is also leached from pipes at concentrations that are too low to affect most people. Schock and Neff (1988) found that new chrome-plated brass faucets were a significant source of Pb, Zn, and Cu contamination of drinking water, particularly when water had been standing in them for some time. Lead concentrations in standing (8-hour standing period) water samples exceeded US maximum contaminant levels (MCLs) of 0.05 mg L 1 for more than 50% of the samples from some sites. The most problematic element is lead, which can appear in domestic water from lead pipes, fittings made of some types of brass alloys, and soldered pipe joints. The situation is most critical with waters of low alkalinity and low pH (the theory for which is described in Section 3.7.1) and that have not been stabilized at the water treatment plant. A particularly troubling occurrence of

8 Water Constituents and Quality Standards

this problem occurred in Flint, Michigan, from 2012 to 2015 (Masten et al. 2016). Most jurisdictions have now banned lead from all these plumbing components including solder. Addition of nutrients to freshwater lakes from agricultural runoff and treated or untreated domestic sewage is making blooms of algae more common. Cyanobacteria (blue-green algae), particularly Anabaena species which are neurotoxic, and Microcystis which is hepatotoxic (toxic to the liver), can be particularly problematic in surface drinking water sources in the summer. Besides being toxic, these organisms cause distinct tastes and odors. Although not regulated in the United States and elsewhere due to a lack of robust and reliable analytical methods (especially for the toxins), many states have issued their own guidelines to trigger do not drink/do not boil (DND/DNB) advisories. For example, maximum concentrations of microcystin range between 0.1 and 1 μg L 1 (or ppb) (Henrie et al. 2017). Copper (usually applied as copper sulfate) is a control measure for these microorganisms and algae, in general, but they pass through the normal treatment operations of flocculation, sedimentation, and rapid sand filtration. Specialized activated carbon treatment is required to remove the toxicity associated with these bacteria (Falconer et al. 1989). Agricultural runoff is also the source of many pesticides used in modern agricultural practice. According to the National Summary of Impaired Waters, which is prepared regularly by the US Environmental Protection Agency (USEPA 2017d), in this case using data reported by the individual states from 2008 to 2016, pesticides were 13th on a list of 34 causes of impairment of natural waters. Pesticides were responsible for 2.6% of all causes of impairment (1848 cases). Pesticides are persistent in the environment and many are not removed to any significant extent in typical wastewater treatment operations incorporating sedimentation and biological treatment. For instance, Köck-Schulmeyer et al. (2013) determined the concentration of 22 target pesticides in wastewater influent and effluent samples from three treatment plants in Catalonia (NE-Spain) and found that most were poorly removed. The pesticides most frequently detected and at the highest concentration were diazinon, diuron, atrazine, simazine, and malathion. Similarly, Arvai et al. (2014) reported that atrazine and pyrene were barely removed from wastewater undergoing activated sludge treatment, with effluents being discharged into the Great Lakes. Runoff from solid waste disposal sites is another significant source of water pollution. Runoff leaches soluble substances and suspends particulates at these sites and transports them to streams. Emissions to air cause air pollution, and many substances are returned to land and water by rainfall. Acid rain is one of the most familiar of these phenomena. The oxides of sulfur and nitrogen, produced by combustion of fuels, hydrolyze to cause a reduction in pH of rain. 8.2.1 Emerging Contaminants Several groups or classes of compounds have only recently been recognized as potentially problem-causing; they fall under the category of “emerging contaminants,” or “contaminants of concern.” They are typically present at very low concentrations (μg L 1 or even lower); hence, they are also referred to as “micropollutants.” Insufficient evidence of harm exists for them to be regulated to date; however, the “precautionary principle” suggests that the authorities should follow the recent literature or in some cases conduct or sponsor the research directly. Figure 8.1 shows five of the most widely identified types and examples of each; it is by no means exhaustive. Endocrine disruptors (“gender benders” are among these) are chemicals that interfere with an organism’s endocrine system. The endocrine system produces hormones that regulate, among other functions, reproduction and immunity. Chemicals that are believed, or in some cases are known, to be endocrine disruptors include pharmaceuticals (especially estrogens), pesticides including DDT, polychlorinated biphenyls (PCB), and bisphenol A. Such chemicals affect not only humans but also many higher aquatic organisms including fish and amphibians. Some

171

172

Theory and Practice of Water and Wastewater Treatment

Figure 8.1 Types of emerging contaminants in the environment. Sources: Henrie et al. (2017), Tijani et al. (2016), and Troester et al. (2016).

pesticides have been banned precisely because they have been clearly linked to toxicity and other forms of environmental disruption, nevertheless many “newer” pesticides once thought to be benign may not be. Personal care products include disinfectants, soaps (including antimicrobial soaps), creams, nail polish, toothpastes, and many others that will be washed off during bathing and/or disposed of in domestic wastewater. They may contain microbeads, oils, and solvents that could be toxic to aquatic organisms. Pharmaceuticals can be considered as a subset of personal care products, but more specifically consist of prescription drugs such as antibiotics, immunotherapy, and chemotherapy medications, which may pass through the body only partially deactivated, or may be disposed of as above. Nanoparticles are particles between 1 and 100 nanometers (nm) in size. They are human-made and can be organic, inorganic, or a combination of both. They can be any shape – spherical, cubic, rod-shaped, planar – and they can be assembled into larger structures. They have entered the ambit of environmental engineering as tools for remediation as well as detection; however, in the context of water contaminants and toxicity, they are unique as the natural environment has no “experience” in dealing with them. Hence, they may penetrate cellular defenses and may accumulate without engendering an immune response. Perfluorinated compounds (PFCs) contain carbon–fluorine bonds as their basis; these are then typically functionalized with carbon, nitrogen, sulfur, and/or other halogen molecules to yield highly resistant materials. Teflon is a well-known example, but PFCs entering the aquatic environment include perfluorooctanesulfonic acid (PFOS), which is a foam used for fire-fighting. The availability of information on emerging contaminants and an understanding of their effects on human health and the environment is growing rapidly. Excellent and detailed review

8 Water Constituents and Quality Standards

articles with information current in 2016 are available by Troester et al. (2016) and Wilkinson et al. (2016). 8.2.2 Common Contaminants Additional information on a few contaminants is provided in the following sections. Aluminum

Aluminum is the third most common element in the earth’s crust and its presence in natural waters is ubiquitous; it is also present in food plants (Reiber et al. 1995). Aluminum is not known to be beneficial to human biology. Alum is the most common coagulant used in water treatment operations (it has been used since ancient times) but less than 1% of the aluminum consumed by an individual comes from drinking water. At the acid pH (1.5–2.0) in the human stomach, Al from organic-Al and other Al-bearing compounds are likely solubilized, eliminating the association of Al with its source. High intake of aluminum has been associated with neuropathological disorders, one of which is Alzheimer’s disease (Crapper et al. 1973; Kopeloff et al. 1942), although the etiologic link between aluminum and Alzheimer’s disease is tenuous (Perl and Moalem 2006). Aluminum is regulated as an aesthetic concern, not for health reasons. Nitrate

The application of fertilizers can result in elevated concentrations of nitrate, which, as noted in Table 8.2, can be toxic to infants. The disease caused by nitrate and nitrite is methemoglobinemia. Nitrates ingested through food or water are converted into nitrites by bacteria in the digestive system of infants. Nitrite reacts with hemoglobin, preventing it from carrying oxygen, thus the common name “blue baby disease” for white-skinned people. This condition is generally confined to infants up to the age of 6 months. The standard for nitrate (10 mg L 1 as N) is based on a US survey in 1951 in which no observed cases of methemoglobinemia occurred when nitrate-N was less than 10 mg L 1. The ubiquitous use of fertilizers commonly causes groundwaters in agricultural areas to exceed the nitrate standard (Bouchard et al. 1992). Certain regions have geologic deposits of nitrate salts that can significantly affect groundwater supplies. European drinking waters also commonly exceed the nitrate standard. Fluoride

Fluoride is a naturally occurring element in many water supplies around the world. Concentra­ tions range from insignificant to over 50 mg L 1. Surface waters generally do not contain fluoride in excess of 0.3 mg L 1. Low concentrations of fluoride are beneficial to the formation of teeth that resist dental caries. The fluoride ion replaces hydroxide ion in the apatite that forms the enamel in teeth. High concentrations lead to mottling (brownish discoloration) of teeth or dental fluorosis. Fluoride concentrations above 1.5 mg L 1 must be reduced to prevent fluorosis. Fluoride is added to deficient waters to produce a concentration near 1 mg L 1 in the finished water. The fluoride content of water is regulated to be lower in regions where water consumption is higher. Studies have shown that the level of fluoride ingestion from water with this concentration provides optimal dental protection with no risk of mottling. Opposition to fluoridation remains, and there have been some cogent arguments made against fluoridation (Camp and Meserve 1974); many European countries do not fluoridate their water supplies. Modern dentistry and home dental hygiene reduce the need for publicly enforced fluoridation, but the benefits of fluoridation are still substantial. There have been numerous studies on the

173

174

Theory and Practice of Water and Wastewater Treatment

physiological effects of ingestion of fluoride. The overwhelming research evidence indicates that fluoride near 1 mg L 1 with normal water consumption is safe with a good margin of safety. Dental associations in Canada and the United States widely encourage the addition of fluoride to water when it is lacking, as it is the only method of ensuring that all segments of the population, regardless of income or level of education, receive the benefits. As of 2007, 45.1% of the Canadian population had access to naturally or artificially fluoridated water at or near recommended levels (Rabb-Waytowich 2009). The Centers for Disease Control and Prevention (CDC) in the United States reported that in 2014, 66.3% of the US population received fluoridated water (CDC 2014). The decision to fluoridate a water is generally made at the local level. Detergents

Historically, soaps were made from animal fats. These soaps are precipitated by multivalent cations in water. The composition of synthetic detergents has undergone a continuous evolution by manufacturers since they were introduced in the 1940s, in order to maintain the performance of a laundering agent that is unaffected by the mineral content of the water and does not cause water quality problems. At least, any such agents, if harmful to the receiving waters, should be readily treatable at a wastewater treatment facility. Alkylbenzenesulfonates and phosphate builders were major constituents of detergents until the 1960s. The former were not readily biodegradable, and rivers were known to become covered with foam caused by this agent, whereas the latter accelerated eutrophication of lakes. To remedy the problem with refractory alkylbenzenesulfonates, a biodegradable linear alkyklbenzenesulfonate was substi­ tuted for the nonlinear alkylbenzenesulfonates. Detergents remain a significant source of phosphorus in wastewaters and receiving waters. Polyphosphate is used as a complexing agent in some detergents and carbonate is used to remove Ca2+ by precipitation in others. Regulation of phosphate in detergents can result in significant reductions of phosphorus in streams and lakes. A statewide ban on phosphorus-containing detergents in Virginia reduced the total contribution of phosphorus from Richmond, VA, to the James River by a factor greater than 5 (Hoffman and Bishop 1994). In the United States, some states have implemented regulations banning phosphorus content of dishwasher detergents (Cohen and Keiser 2016). In 2010, Canada has passed regulation limiting P content to no more than 0.5% by mass of household cleaners and detergents (Holeton et al. 2011). The European Union has passed legislation to limit phosphorus content of laundry and dishwasher detergents to 0.5 g P per washing process for laundry detergents and 0.3 g of P in a standard dosage for dishwasher detergents (European Report 2012). 8.2.3 Carcinogens Cancer-causing agents are a major concern of society. Increasing life spans and increased production of these agents among other factors contribute to increased mortality rates from cancer. There are many factors that contribute to the potency of an agent causing cancer. Naturally occurring carcinogens exist in foods that are staples in diets throughout the world. There do not appear to be any threshold levels for carcinogenic agent concentrations. The risk of contracting the disease is directly proportional to the exposure to the agent; however, there is wide variability in the potency of agents. The World Health Organization (WHO) sets guideline values for carcinogens in drinking water at a level where the lifetime exposure will increase the risk by one additional cancer per 100 000 (a risk factor of 10 5 over a lifetime of 70 years) for the population ingesting the drinking water (WHO 2011). The USEPA has a similar goal of setting limits that reduce lifetime cancer risk below 1 in 104 (Post et al. 2012).

8 Water Constituents and Quality Standards

8.2.4 Radioactive Constituents The major health effects of radiation are an increase in the occurrences of cancer and genetic defects. Radiation induces changes in the cell that lead to cancer. Some radioactive elements such as uranium can have a chemically toxic effect on organs. High doses of radiation can cause death in a short period of time. Radiation units in terms of disintegrations per second have been defined in Section 1.13.1. The biological effects of radiation depend on the ionizing effects of radiation in cells. The roentgen (named after W. Roentgen, who discovered X-rays) is the unit of gamma or X-radiation intensity. The roentgen is defined as the quantity of gamma or X radiation that will produce ions carrying one electrostatic unit of electricity of either sign in 0.001 293 g (1 cm3) of dry air at 0 °C and 760 mm (1 atm) pressure. This is equivalent to 1.61 × 1012 ion pairs per gram of air and corresponds to the absorption of 83.8 ergs of energy. The sievert (Sv) is another measure of radiation dose equal to approximately 8.38 roentgens. The rad (roentgen absorption dose) is another measure of the dose of radiation delivered by radioactivity. One rad equals 100 ergs of energy deposited per unit mass of any absorbing material. A rad is not necessarily a measure of the damage resulting to humans. Rads are modified for humans by multiplication with a factor called the relative biological effectiveness factor that gauges how effective the deposited material is in damaging human tissue. The resulting measure is the roentgen equivalent-man (rem). Recommended maximum exposures for the general public and for workers in the radiation industry are 0.5 and 5 rem yr 1, respectively. The issue of biological effectiveness of a given dose of radiation is complicated by the accumulation of some isotopes in the body; for instance, iodine is accumulated in the thyroid gland. The penetrating power of radiation is also an important factor. Alpha particles are relatively massive and travel only a short distance in air and cannot generally penetrate the dead layer of skin of a human. Thus, alpha emitters are only dangerous if they are ingested. Then, a particle is able to do great damage inside the body over the short pathway in which its energy is deposited. Beta particles are lighter than alpha particles and are generally considered to have about 100 times the penetrating power of alpha particles (although there is considerable variation from this value). The most penetrating forms of radiation are neutrons, which are rare, and gamma rays. Gamma rays are 10 000 times more penetrating than alpha particles (Lowry and Lowry 1988). Table 8.3 contains US data on the annual effective dose equivalents from radiation sources. Radon-222 and its progeny are the major radiation hazards. Radon and its sister radionuclide Table 8.3 Annual effective dose equivalent to people in the United States from various sources. Source

Dose equivalent (mSv yr − 1)

Radon and thoron (airborne)

1

Other natural sourcesa)

0.005–0.01

Diagnostic X rays

0.39

Nuclear medicine

0.14

Occupational

0.009

Miscellaneous environmental exposure

0.006

Consumer products

0.005

a) The uranium and thorium series radionuclides are the major sources. Source: Adapted from Lowry and Lowry (1988) and Carny (2003).

175

176

Theory and Practice of Water and Wastewater Treatment

thoron (radon-220) are ubiquitous, with their average outdoor levels at 10 Bq m 3 of air (Carny 2003). Both radon nuclides are volatile and their presence in drinking water is therefore not significant (Lowry and Lowry 1988), but in areas with igneous rocks and soil containing uranium or thorium, it may be released into the home through normal water use (with groundwater as the raw water source) as well as through cracks in basement walls. It therefore presents a serious health threat, especially to the lungs, and is the second leading cause of lung cancer after smoking.

8.3 Taste and Odor There are many natural sources of taste- and odor-causing compounds in addition to synthetic chemicals. Taste and odor problems with a water supply elicit rapid response from consumers. Odor emanation from wastewater treatment works is also a cause of complaints. The most well-known natural odoriferous compound is hydrogen sulfide (H2S), commonly found in well water. Anaerobic conditions generate H2S, which is readily soluble in water. H2S is one of the few taste- and odor-causing compounds addressed in water quality standards. Even though H2S is toxic at elevated levels, taste and odor problems occur at much lower concentra­ tions. Many other taste- and odor-causing compounds have been identified, but they are generally not addressed beyond the general requirement that water be inoffensive to consumers. The problem is complicated by many factors. First, there is wide variability in the sensitivity of individuals to different compounds and the threshold concentrations for detecting the com­ pounds. Mineral content and synergisms among water components produce different taste and odor reactions. Many taste and odor compounds are noticeable at concentrations less than 1 ppb. For instance, geosmin, an earthy-smelling byproduct of cyanobacteria and Actinomycetes, is detectable at concentrations of 0.2 ppb. The sources of taste and odor compounds are many. Decaying organic matter releases tasteand odor-causing compounds into waters. Phenols are found naturally in fossil fuels, but they are also produced in many industries. Phenols are detectable at low concentrations and they are also mildly toxic. Chlorine substitution on phenols increases both unpleasant taste and odor, as well as toxicity effects. The earliest microorganisms identified with tastes and odors were algae. The infestation of zebra mussels in some lakes has been indirectly linked to taste and odor problems. Densities have reached over 250 000 mussels m 2 in bed zones in the Great Lakes (Vogel et al. 1997); numerous smaller lakes have also been affected and the mussels continue to spread. High densities of zebra mussels are associated with elevated growth of benthic green algae that do not themselves produce the taste and odor problems; however, their die-off supports growth of cyanobacteria or Actinomycetes that do. Actinomycetes are mold-like filamentous bacteria, widely distributed in the environment, that produce many offensive tastes and odors described as woody, haylike, marshy, musty, manurial, potato bin, and bitter. Such taste- and odor-causing microbes are generally not harmful to the health of people drinking water containing viable cells; therefore, their allowable concentrations are not directly proscribed. On the other hand, cyanobacteria do produce toxins that are harmful – to the extent of causing death – to domestic and farm animals; therefore, measures are being taken more widely to limit their presence (Chorus 2005). Table 8.4 lists some chemicals with high odor recognition. The odor potential of a substance depends not only on the concentration at which the odor is noticed but also on the ability of the agent to volatilize. Thus, Verschueren (2009) used the odor index (OI) proposed by Hellman and Small (1974) as well as the 100% recognition level to characterize odor potential. The equation

8 Water Constituents and Quality Standards

for OI is OI ˆ

vapor pressure …ppm† 100%odor recognition threshold …ppm†

(8.1)

where ppm is on a volume basis and 1 atm = 1 000 000 ppm. The 100% odor recognition level is the concentration at which 100% of the odor panel members perceive the odor. Compounds listed in Table 8.4 are those that have an odor potential in the medium (OI = 100 000–1 000 000) to high range (OI > 1 000 000). Table 8.4 Odor potential of selected chemical compounds. Chemical

Odor indexa)

Formula

Molecular weight

100% recognition

Ethylbutyrate

C3H7COOC2H5

116

1 982 000

7 ppb

Ethyldecanoate

C9H19COOC2H5

200

—

0.17 ppb

53 300 000

35 ppb

940 000

0.2 ppb

Ethylesters

Mercaptans Methylmercaptan

CH3SH

48

Phenylmercaptan

C6H5SH

110

Sulfides Hydrogen sulfide

H2S

34

17 000 000

1 ppm

Methylsulfide

(CH3)2S

62

2 760 000

0.1 ppm

Butyrates Methylbutyrate

CH3

CH2

CH2

COOCH3

102

19 000 000

2 ppb

Ethylbutyrate

CH3

CH2

CH2

COOC2H5

116

1 982 000

7 ppb

Amines Ammonia

NH3

17

167 300

55 ppm

Methylamine

CH3NH2

31

940 000

3 ppm

Ethylamine

CH3CH2NH2

45

1 445 000

0.8 ppm

Trimethylamine

(CH3)3N

59

493 500

4 ppm

56

43 480 000

0.07 ppm

70

376 000 000

2 ppb

Alkenes 1-Butene

CH3CH2CH

CH2

1-Pentene

CH3CH2CH2CH

CH2

Ethers Ethylether

CH3CH2OCH2CH3

Isopropylether

(CH3)2CHOCH(CH3)2

74

1 939 000

0.3 ppm

100

3 227 000

0.06 ppm

Aldehydes Formaldehyde

HCHO

30

5 000 000

1 ppm

Acetaldehyde

CH3CHO

44

4 300 000

0.3 ppm

Propionic acid

CH3CH2COOH

74

112 300

40 ppb

Acetic acid

CH3COOH

60

15 000

2 ppm

Acids

a) See text for definition of OI. OIs were calculated at 20 °C.

Source: From Verschueren (2009). Reproduced with permission of John Wiley & Sons.

177

178

Theory and Practice of Water and Wastewater Treatment

8.4 Bases for Standards Standards drive the need and degree of water and wastewater treatment. The promulgation and enforcement of standards that ensure reasonable protection of public health and environmental quality are required. In many countries, water quality criteria are incorporated into laws that specify substantial penalties for their violation. Standards for drinking water quality are generally established on the following bases: 1) Established or ongoing practices. Practices that have been used for many years without noticeable harmful effects are used as a basis to set water quality criteria. Hidden dangers can exist in merely adopting historical practice without scientific analysis. Chlorine is an example of a substance used without any noticeable harmful effects until careful analysis has shown that it forms carcinogenic compounds. 2) Animal models. Historically, this has been the primary method for assessing chemicals that are toxic to humans. It is most desirable to use animals with a physiological response as similar as possible to humans, e.g., rodents, to find exposure levels that produce harmful effects. These studies tend to be expensive. Tests such as the Ames test (Ames et al. 1975) described in Section 8.4.2, which indicates the carcinogenicity of a compound, use bacteria to assess whether a chemical agent causes mutations. Tissue cultures, which use cell lines originating from mammalian cells (human or animal), may in some cases provide reliable indications of toxicity levels to humans without having to sacrifice live animals (Stacey 2012). All these tests are increasingly being used as an alternative to animal testing; nevertheless, there always remain questions on translating animal or other organism toxicity effects to humans. 3) Human exposure. Epidemiology or studies that show direct causal evidence between exposure to an agent in water and disease or toxic effects in humans are the most reliable source of information. Experimenting with humans directly is undertaken only in unusual circumstances; however, there are numerous occurrences of long-term or sudden contami­ nation of water supplies that have led to mild or disastrous results. Some benefit can be derived from these unfortunate situations by careful assessment of exposure levels and resulting health deterioration in the affected population. 4) Statistical comparisons. Statistical analysis of data from two populations can confirm differences in the occurrence of disease and in conjunction with other studies, the possible causes of the disease. The interpretation of data from these studies, being a well-developed branch of epidemiology, must be done with caution because of the impossibility of isolating two naturally occurring populations that differ only in one aspect of their behavior or exposure to a substance. Confidence limits should be given with the statistical study to gauge the statistical certainty of the result. A complicating factor that applies to many studies is extrapolating acute exposure (high­ dose effects) to chronic exposure at low levels. Once a threshold level (usually at 50% of the tested population) for a causative agent is established from the best available information, this level is divided by a factor of 100 or more to provide a suitable safety factor to arrive at a standard. Risk analysis is increasingly being applied. The objective is to regulate all harmful substances at the same level of risk and other substances at levels that are inoffensive or aesthetically pleasing to the majority of consumers. Drinking water is not the major route for exposure to many chemical substances, but it may be the only route that can be controlled. Period and degree of exposure of a population from all sources (air, water, and food) are taken into account. Technological and economic constraints may mitigate the immediate attain­ ment of a given standard.

8 Water Constituents and Quality Standards

8.4.1 Risk Assessment for Microbial Infection Risk assessment is an important consideration, along with costs and technical limitations for setting standards. Risk is a measure of the probability of an adverse outcome from an event. For a microbial agent causing infection, the probability, p, of an exposure causing the infection is a function of a dose–response relationship. The density or concentration of an infectious agent in a medium (water, air, or food) is C (in number of microorganisms per unit volume). Two major dose–response relationships have been found to describe infectivity (Haas 1996). A Poisson distribution (see Section 6.5.4) describes the probability of infection from a single exposure. pˆ1

CV k

exp

(8.2)

where p is probability of an occurrence, k is the number of microorganisms that must, on average, be ingested to initiate an infection, and V is the volume of medium ingested. The β-Poisson model is derived from the exponential model [Eq. (8.2)]. It depends on two parameters that describe the dose–response data. α

CV β

pˆ1

(8.3)

where α and β are parameters. The β-Poisson model is derived from the Poisson model [Eq. (8.2)] when there is a heteroge­ neous microbial survival probability distribution. The parameter α reflects the degree of heterogeneity, and as it increases, Eq. (8.3) reverts to Eq. (8.2). At low doses and risks, Eq. (8.3) can be linearized to pˆ

αV C β

(8.4)

where α is a parameter. When the risk from an independent exposure is independent of other exposures, the overall risk from cumulative exposures is Pˆ1

n

∏ 1

(8.5)

pi

iˆ1

where i is the event index for n exposures and P is the overall risk of infection. The average is defined as the per exposure risk, that would, if it existed during each exposure, produce the same overall risk. From Eq. (8.5), P pP ˆ 1

1

pP

n

(8.6)

where pP is the risk per exposure and PpP is the overall risk of infection for multiple exposures when pP has the same value. Equation (8.6) can be inverted to find pP ˆ 1

1

P pP

1=n

(8.7)

and the average risk can be found as a function of the individual risk from Eqs. (8.5) and (8.7). pP ˆ 1

1=n

n

∏ 1 iˆ1

pi

(8.8)

179

180

Theory and Practice of Water and Wastewater Treatment

As shown by Haas (1996), the arithmetic mean (as opposed to the geometric mean) is the appropriate parameter characterizing the risk when either the Poisson model applies or the dose–response relationship is linear at low doses. When the Poisson exponential dose–response is followed, Eq. (8.2) is substituted into Eq. (8.8) to find the average concentration to use in developing a risk estimate. 1

n

CPV k

exp

ˆ ∏ exp iˆ1

CiV k

1=n

(8.9)

which can be simplified to CP ˆ

1 n

n

(8.10)

Ci iˆ1

If the individual risks in Eq. (8.8) are 10 μm. Concentration is based on fibers L 1.

c) It is generally accepted that only zero-, di-, tri-, and hexavalent oxidation states of chromium have biological

importance. The trivalent form is essential in micro amounts (50–200 μg d 1) for humans (Anon. 1989). Hexavalent chromium is the toxic form of chromium causing a variety of ill effects. d) USEPA (2017b) specifies a limit of 2 mg L 1 for aesthetic purposes. e) Treatment technology specified.

Radon is not included in radiological standards for water because it is volatile. Its main sources as an air pollutant are rocks and soils containing radium-226, as well as groundwaters. Ingestion of radon from food or water is generally considered to be an insignificant health risk; inhalation poses the cancer risk (Health Canada 2009). Radon is not included in WHO or US regulations for similar reasons. Tables 8.8–8.11 not only illustrate different stages of regulation development but also different philosophies of regulation. Some constituents that are controlled by one country have been

189

190

Theory and Practice of Water and Wastewater Treatment

Table 8.9 Organic constituents of health significancea) Concentration (mg L − 1)

Constituent

Guideline WHO

Canadab)

US standards MCLG

MCL

Acrylamide

0.000 5

0

TTc)

Alachlor

0.02

0

0.002

Aldrin and dieldrin

0.000 03

Atrazine

0.1

0.005

0.003

0.003

Benzene

0.01

0.005

0

0.005

Carbaryl (sevin)

0.09

Carbofuran

0.007

0.09

0.04

0.04

Carbon tetrachloride

0.004

0.002

0

0.005

Chlordane

0.000 2

0

0.002

0.1

0.1

Chlorobenzene Chloroform

0.3

0.07

Chrysene 2,4,D (2,4-dichlorophenoxyacetic acid)

0.03

DDT (dichloro-diphenyl-trichloroethane)

0.001

0

0.000 2

0.07

0.07

0

0.005

0.007

0.007

0.14

0.7

0.7

0.28

0.7

0.7

0.1

Diazinon

0.02

Dicamba (dianat)

0.12

Dichloroacetic acid

0.05

Dichloromethane

0.02

2,4-Dichlorophenol

0.05 0.9

Dinoseb Ethylbenzene

0.3

Glyphosate Haloacetic acids γ-HCH (lindane; γ-hexachlorocyclohexane)

0.08 0.002

0.000 2

0

0.000 4

0.04

0.04

0.06

0

0.001

0.19

0.5

0.5

0

0.000 5

0.1

0.1

0

3 × 10

1

1

Heptachlor Malathion

0.19

Methoxychlor

0.02

NTA (nitrilotriacetic acid)

0.2

Paraquat Pentachlorophenol

0.40 0.01

0.009

Picloram Polychlorinated biphenyls Styrene (vinylbenzene)

0.02

2,3,7,8-TCDD (dioxin) Toluene

0.06 0.000 2

0.7

0.06

8

8 Water Constituents and Quality Standards

Table 8.9 (Continued ) Concentration (mg L − 1)

Constituent

Guideline Canadab)

WHO

Trichloroacetic acid Trihalomethanes (THMs)

0.2 d)

US standards MCLG

MCL

0.3

0.06

0.1

Vinyl chloride

0.000 3

0.002

0

0.002

Xylene

0.5

0.09

10

10

a) See notes at beginning of Section 8.6.

b) MAC.

c) Treatment technology specified.

d) See WHO (2011) for an allocation formula for THMs. The United States has standards for individual THMs.

Table 8.10 Aesthetic parametersa) Concentration (mg L − 1)

Constituent

USb),c)

Guideline WHO

Canada

MCL

Aluminum

0.1–0.2

0.1

0.05–0.2

Chloride

250

250

250

Color

15 (units)

15 (units)

15 (units)

Copper

2

1.0

1.0

Corrosivity

Noncorrosive

Foaming agents

0.5

Hydrogen sulfide

0.05–0.1

0.05

Iron

0.3

0.3

0.3

Manganese

0.1

0.05

0.05

pH

Preferably < 8.0

7.0–10.5

6.5–8.5

50

200

Silver Sodium

0.1

Solids, total dissolved

1000

500

500

Sulfate

250

500

250

Taste and odor

Inoffensive

Inoffensive

3 TONd)

Temperature

Acceptable

15 °C

Turbidity (NTU)

5

TTe)

TT

5

5

Zinc a) b) c) d) e)

See notes at beginning of Section 8.6. From USEPA (2017b). Except for copper, these are secondary contaminants. Only odor is specified as three threshold odor numbers (TON). Treatment technology specified.

191

192

Theory and Practice of Water and Wastewater Treatment

Table 8.11 Radiological parametersa) Concentration (Bq L − 1)

Constituent

US standardsb)

Guideline WHO

Canada

Cesium-137

10

10

Iodine-131

10

6

Radium-226

1

0.5

Radium-228

0.1

Radium-226 + 228 Strontium-90

10

Tritium

10 000

Uranium

0.03 mg L

MCLG

MCL

0

20 pCi L

0

30 μg L

1

5 7000 1

0.02 mg L

1

1 1

Gross alpha activity

0.5

0

15 pCi L

Gross beta/photon activity

1

0

4 mrem yr

1

a) See notes at beginning of Section 8.6. b) Except for tritium and strontium-90, USEPA bases its values on the concentration of radionuclide causing 4 mrem yr 1 total body or organ dose-equivalents (USEPA 2017c).

eliminated as a hazard by the other. Differences in guidelines also reflect differing approaches to risk assessment. 8.6.5 Other Water Standards In addition to drinking water standards, guidelines have been developed for many other uses of water. For example, in Canada, federal guidelines for the following water uses can be found in CCME (2017): Water Quality Guidelines and Aesthetics Water Quality Guidelines for the Protection of Aquatic Life Water Quality Guidelines for the Protection of Agriculture Uses The WHO and USEPA offer documents of a similar nature on their websites See Chapter 22 for surface water standards and other information on contaminants in surface waters.

8.7 Water Consumption The quantity of water consumed is variable depending on water supply, climate, and cultural habits of the population. Persich (2016) analyzed the data of Frontinus, water commissioner of Rome, who published a report in CE 97. At nearly 946 ML d 1 capacity, but recorded delivery of ~580 ML d 1, he arrived at an average consumption of 946 L d 1 per population-equivalent for its one million or so inhabitants, which included almost 50% leakage. Amazingly, these figures are not far different from some large cities in North America and elsewhere, though the objective should be far lower consumption.

8 Water Constituents and Quality Standards

Table 8.12 Commercial and institutional water demand. Source

Average daily water usea)

Domestic usageb)

270–450 L/cap-d

Shopping centers

2.5–5.0 L m

Hospitals

900–1800 L/bed

Schools

70–140 L/student

2

(based on total floor area)

Travel trailer parks Without individual hookups

340 L/site

With individual hookups

800 L/site

Campgrounds

225–570 L/campsite

Mobile home parks

1000 L/unit

Motels

150–200 L/bed

Hotels

225 L/bed

Industrial areas Light industrial area

35 m3/ha-d

Heavy industry

55 m3/ha-d

a) These are design values for Ontario.

b) Use lower values with metering.

Source: From OMOECC (2008a). Reproduced with permission of Queen’s Printer for Ontario, 2008.

Rural households tend to use less water than urban residences because city residents have more water-using appliances and practices than do rural residents. Water demands can vary from 180 to 1500 L/cap-d (OMOECC 2008a). In Ontario, average daily water demands in the range of 270–450 L/cap-d are recommended. Table 8.12 gives unit water demands for commer­ cial and institutional establishments in Ontario. Similar and more detailed tables are published in specialized texts dealing with water distribution, and wastewater and stormwater collection, notably that by Shammas and Wang (2011) for the United States. Demands can be further broken down by types of industries (food, car manufacturing, etc.), types and numbers of fixtures in the building, and regions that have different climates. Northern communities in Canada are remote with varying accessibility to water supplies. Design water consumption for various types of water delivery and collection systems are given in Table 8.13 for these communities. In piped, pressurized water supplies, water bleeder systems to prevent the pipes from freezing can significantly increase the rate of water consumption. Flushing the toilet is the major indoor use of water in the United States at around 28% of total indoor use (Metcalf and Eddy:AECOM 2014). Older toilets consume 18–25 L/flush; fortunately, they are becoming rare. Low-volume flush toilets are as efficient or more efficient than the larger volume units that they replace. Ultra-low-volume toilets using 3 L/flush have been designed and tested in residential districts (Anderson and Siegrist 1989). They performed as satisfactorily as conventional toilets (using more than 13 L/flush). As well, many appliances are using less water. Conservation coupled with water-saving devices is reducing water consumption for households and industries. Providing cash incentives for residents to purchase water-saving devices can be cost–effective to stave off expensive capacity upgrades for water and wastewater treatment plants.

193

194

Theory and Practice of Water and Wastewater Treatment

Table 8.13 Residential water use for northern Canadian communities. Source

Water use (L/cap-da))

Range (L/cap-d)

Nonpressure water system with bucket

10

5–25

Trucked water delivery Nonpressure system with bucket toilets

10–25

5–50

Nonpressure system, holding tank with wastewater pumpout collection

40

20–70

Pressure water system, normal flush toilet and wastewater pumpout collection

90

40–250

Piped water supply and vacuum or pressure sewers

145

60–250

Piped water supply and gravity sewers

225

100–400

a) Design values.

Source: Adapted from Smith and Knoll (1986).

Peaking factors (PFs) are multiplication factors used to estimate high and low flows with respect to the average flow; PFs for water distribution are provided in Table 8.14. Population increases and urbanization are placing severe demands on scarce water resources throughout the world. Reclamation of wastewater distributed through dual distribution systems is being implemented, particularly in new developments, where water is scarce (Okun 1997); the Romans were the first to apply the concept (Herschel 1973). Nonpotable reuse extends to irrigation, cooling, and commercial uses such as vehicle washing to water for flushing toilets. This last-mentioned use must be approached with extreme caution, as recycled water may still contain pathogens that would affect health, as well as organics that could lead to biofilm buildup and pipe clogging. NSF International published the first American standard for commercial and residential onsite water reuse treatment systems (NSF/ANSI 350) in 2011. In the same year, NSF published a separate standard (NSF/ANSI 350-1-2011), which applies to greywater treatment systems for subsurface discharge only. A detailed and authoritative resource on water reuse is that by Metcalf and Eddy:AECOM (2007).

Table 8.14 Peaking factors for water distribution. Population

Minimum rate factor (minimum hour)

Maximum day factor

Peak hour factor

500–1000

0.40

2.75

4.13

1001–2000

0.45

2.50

3.75

2001–3000

0.45

2.25

3.38

3001–10 000

0.50

2.00

3.00

10 001–25 000

0.60

1.90

2.85

25 001–50 000

0.65

1.80

2.70

50 001–75 000

0.65

1.75

2.62

75 001–150 000

0.70

1.65

2.48

>150 000

0.80

1.50

2.25

Source: From OMOECC (2008a). Reproduced with permission of Queen’s Printer for Ontario, 2008.

8 Water Constituents and Quality Standards

The quantity of water actually drunk is a very small percentage of the total amount of water used. The total amount of tap water consumed by drinking either tap water itself or in tap-water­ based beverages such as coffee, tea, or rehydrated soup or milk in Canada averaged 1.34 L/cap-d (Health and Welfare Canada 1981). There was little variation in consumption with sex, but consumption varied with age, which reflects differences in body weight. This information is used in formulating water quality guidelines. As might be expected, consumers who were less satisfied with the taste of their water tended to drink a greater percentage of water-based beverages, probably to disguise the taste of the water. There was not a significant change in total water consumption as satisfaction with the water taste changed.

8.8 Canadian Federal Wastewater Quality Guidelines Extracts from Canadian wastewater effluent guidelines are given in Table 8.15. Provinces often adopt these guidelines in their own regulations or guidelines. Industry-specific guidelines also exist for a number of industries. Effluent concentrations are often based on the amount of raw material processed, although stream assimilative capacity may override guidelines based on the amount of production.

8.9 Wastewater Characteristics Domestic wastewater quality varies widely from community to community depending on drinking water quality, use and conservation practices, cultural attributes of the population, population density, industries present, and treatment applied at industry locations, as well as other factors such as aging infrastructure and infiltration in the sewer system. The major source of organics in domestic wastewater is human excreta. For the United States, Metcalf and Eddy:AECOM (2014) reported daily typical BOD5 contributions to sewage per adult from excreta ranging from 50–120 g/cap-d, which is higher than most other countries due to the reality that approximately 37% of adults in the United States are obese. Feachem et al. (1981) arrived at 39–42 g/cap-d, which may be an unrealistically low and narrow range; certainly, all of the 12 other countries reported in the former source have maximum values that are higher. Both sources give daily adult urine production (which is a rich source of nitrogen) at 0.8–1.3 kg (or 0.8–1.3 L). The ammonia and urea (which is readily hydrolyzed to ammonia) contents of urine are typically around 300 and 20 000 mg L 1, respectively. Organic kitchen wastes and industrial wastes will increase the

Table 8.15 Canadian federal wastewater quality guidelines. Substance

Limit (mg L − 1)

BOD5

25

TSS

25

Nonionized ammonia-N

1.25

Cl2 residual

0.02

Source: Adapted from Environment Canada (2017).

195

196

Theory and Practice of Water and Wastewater Treatment

Table 8.16 Typical composition (mg L 1) of municipal wastewater containing minor industrial wastewatera) Parameter

High

Medium

Lowb)

COD total

1200 (1015)

750 (510)

500 (340)

COD soluble

480

300

200

COD suspended

720

450

300

BOD5

560 (400)

350 (200)

230 (135)

Nondegradable COD

180

TSS

600 (390)

400 (195)

250 (130)

VSS

480 (305)

320 (150)

200 (100)

Volatile fatty acids (as acetate)

80

30

10

Total N

100 (70)

60 (35)

30 (25)

NH3-N

75 (40)

45 (20)

20 (15)

Total P

25 (10)

15 (5)

6 (4)

Ortho-P

15

10

4

Organic P

10 (6)

5 (3)

2 (2)

Chloride

600

400

200

Fecal coliforms (no./100 mL)

(105–106)

(104–106)

(103–105)

a) Values in parentheses are for the US.

b) Strength is generally inversely proportional to flow rate. For US values “High” corresponds to a flowrate of 190 L/

cap-d, “Medium” to 380 L/cap-d, and “Low” to 570 L/cap-d (Metcalf and Eddy:AECOM 2014). Source: Adapted from Henze and Comeau (2008) and Metcalf and Eddy:AECOM (2014).

per-capita contributions of organics and other substances, which may also explain the higher values for the United States, where as many as 25% of homes have kitchen-waste food grinders. Table 8.16 provides information on the typical composition of domestic wastewaters. Besides water consumption rates, age of the sewer system, which affects infiltration/inflow, and the extent of combined sewers affect the concentration of wastewater constituents arriving at the treatment plant. Nemerow and Dasgupta (1991) provided information on the quality of wastewater produced from various industries. Raunkjær et al. (1994) analyzed the protein, carbohydrate, and lipid fractions in domestic wastewater and reviewed results from other studies. The ranges of average results for these components expressed as a percentage of total COD from all studies were 8–28, 6–18, and 12–31 for protein, carbohydrates, and lipids, respectively.

Greywater

Greywater is defined as all wastewater produced from a household excluding toilet wastes. There is potential for reuse of greywater, reducing demands for domestic water consumption; however, fecal coliforms and other indicator microorganisms can be found in greywater in significant numbers. Rose et al. (1991) found fecal coliforms ranging from 104 to 107/100 mL; Casanova et al. (2001) found a similar range. Other data on the quality of greywater are given in Table 8.17.

8 Water Constituents and Quality Standards

Table 8.17 Characteristics of greywater. Constituent

Units

pH

Range

Average

Average in tap water

5–7.5

6.54

6.6

149–198

158

131

Alkalinity

mg L

1

Ammonia nitrogen

mg L

1

0.15–3.2

0.74

0

Nitrate

mg L

1

0–4.9

0.98

1.0

Total nitrogen

mg L

1

0.6–5.2

1.7

1.0

Chloride

mg L

1

3.1–12

9.0

10

TSS

mg L

1

15–112

35

0

Hardness

mg L

1

112–152

144

142

Phosphate

mg L

1

4–35

9.3

3.1

Sulfate

mg L

1

12–88

Turbidity

NTU

76.3

0.8

20–140

0

Source: Adapted from Rose et al. (1991) and Casanova et al. (2001).

8.10 Wastewater Production Flows for sewage treatment plants are based on the design population and commercial and industrial activity. Historical data should be gathered to find existing flow information. The plant must be able to handle all flows anticipated in the design period. In Ontario, design periods for sewage treatment plants have generally been for 20 years as opposed to sewer systems, which are designed for ultimate flows from each tributary area (OMOECC 2008b). Ontario’s experience is that design peak flows are two to three times design average flow rates. Table 8.12 can be used for typical design sewage flows from various sources. Actual flows and design criteria will vary from region to region, and infiltration must be added to design wastewater flows. Small communities generate lower flows than larger cities. In northern Canadian communities, nearly all of the water consumed is returned as wastewater, and the design values given in Table 8.13 apply for wastewater flows. Design PFs are usually specified by the client (municipality, province, or state). They will vary according to season, climate, but most importantly, population. In the absence of measured data, one or more of the following equations may be used according to local conditions and practices. The most commonly used PF is the ratio of maximum daily flow in the year to the average daily flow over that year. Shammas and Wang (2011) collected the following widely used equations: PF ˆ

Qmax day Qavg day

(8.14a)

5:0 P0:2

(8.14b)

2:69 Q0:121 avg day

(8.14c)

Babbit: PF ˆ Federov: PF ˆ

197

198

Theory and Practice of Water and Wastewater Treatment

Table 8.18 Unit process design basis. Process

Design basis

Sewage pumping stations

Design peak instantaneous flow

Screening

Design peak instantaneous flow

Grit removal

Design peak hourly flow, peak hourly grit loading

Primary sedimentation

Design peak daily flow

Aeration (without nitrification)

Average daily BOD5 loading (based on design average daily flow)

Aeration (with nitrification)

Average daily BOD5 loading (based on design average daily flow), peak daily TKN loading (based on design peak daily flow)

Secondary sedimentation

Design peak hourly flow, peak daily solids loading

Sludge return

For activated sludge processes, 50–200% of design average daily flow rate

Disinfection

Design peak hourly flow

Effluent filtration

Design peak hourly flow

Outfall sewer

Design peak instantaneous flow

Sludge treatment (digestion and dewatering)

Maximum monthly mass loading and flow rates

Source: From OMOECC (2008b). Reproduced with permission of Queen’s Printer for Ontario 2008.

Gifft: PF ˆ

5:0 P1=6

(8.14d)

Harmon and Ten-States standards: p 18 ‡ P p PF ˆ 4:0 ‡ P

(8.14e)

where P is the population in thousands and Qavg day is the average daily flow (L s 1). For other flow factors, Shammas and Wang (2011) cited the following ratios: Qmin day ˆ 0:2P1=6 Qavg day

(8.15a)

Qmax day ˆ 25P Qmin day

(8.15b)

1=3

Flow rates and other factors dictate the design capacity of wastewater treatment processes. Table 8.18 provides the design bases for wastewater treatment processes in Ontario.

Questions and Problems 8.1 What are the federal and provincial (state) drinking water guidelines for your city? 8.2 Would breaking a mercury thermometer in your mouth and ingesting some of the mercury (but no glass) pose a serious threat to your health?

8 Water Constituents and Quality Standards

8.3 What is methemoglobinemia? 8.4 Distinguish between a carcinogen and a mutagen. 8.5 The vapor pressure of phenol is 3.2 × 10 4 atm at 20 °C. The concentration at which there was 100% odor recognition by a panel was 20 ppm (volumetric basis). Find the OI. Also find the concentration on a mass basis of phenol in air at 20 °C at which there was 100% odor recognition. 8.6 (a) Describe the features of the Ames test. (b) Search the literature for information on other similar tests and write a detailed description of the analyses. 8.7 What is the cost of water in your community? Is it broken down into drinking water, sewage, and sewer costs? 8.8 What is the purpose of the Microtox® assay? Describe the test. 8.9 It is inevitable that research on the health effects of various agents will suggest standards that are below detectability limits of current instruments when safety factors are applied or risk analysis is performed. The defensibility of regulations when standards are below detectability limits is difficult if not impossible in court. Discuss your opinions on the regulation of substances in this situation. 8.10 At a public meeting on fluoridation of drinking water in Ottawa, ON, a disputant of the practice argued that hydrofluoric acid, which is so strong that it is used to etch glass, is commonly added to water as the fluoride agent (and related effects would occur in humans). Discuss the merits of this argument. 8.11 Research and summarize the health effects of one of the following: (a) aluminum, (b) fluoride, (c) nitrates, (d) trihalomethanes. 8.12 Derive Eq. (8.7) from Eq. (8.6). 8.13 What are the design guidelines for water and wastewater treatment works in your community? 8.14 What is greywater? 8.15 Keep track of water use and wastewater production by yourself and residents of your dwelling for 1 week and report the results. Note the amount of water used for each activity (drinking, laundry, lawn watering, food preparation/cooking, toilet flushing, bathing/ personal hygiene, and car washing are suggested categories). 8.16 Find the detectability limits of aluminum for various analytical techniques. 8.17 Compare the PF derived from the Babbit, Gifft, and Harmon and Ten States standards for populations of 10 000 and 500 000.

199

200

Theory and Practice of Water and Wastewater Treatment

References Ames, B.N., McCann, J., and Yamasaki, E. (1975). Methods for detecting carcinogens and mutagens with the Salmonella/mammalian-microsome test. Mutat. Res. 31 (6): 347–363. doi: 10.1016/0165-1161(76)90196-5. Anderson, D.L. and Siegrist, R. (1989). The performance of ultra-low-volume flush toilets in Phoenix. J. Am. Water Works Assoc. 81 (3): 52–57. http://www.jstor.org/stable/41292400 (accessed 14 March 2018). Anon (1989). Journal of the International Register of Potentially Toxic Chemicals Devoted to Information on Hazardous Chemicals, United Nations Environment Programme, 9, 2, 19–21. Arvai, A., Klecka, G., Jasim, S. et al. (2014). Protecting our Great Lakes: assessing the effectiveness of wastewater treatments for the removal of chemicals of emerging concern. Water Quality Res J. Can. 49 (1): 23–31. doi: 10.2166/wqrjc.2013.104. Balls, M., Combes, R.D., and Bhogal, N. (eds) (2012). New Technologies for Toxicity Testing, 1–13. New York: Springer-Verlag. doi: 10.1007/978-1-4614-3055-1. Beer, K.D., Gargano, J.W., Roberts, V.A. et al. (2015). Surveillance of waterborne disease outbreaks associated with drinking water – United States, 2011–2012. Am. J. Transp. 15 (12): 3280–3267. doi: 10.1111/ajt.13602. Blum, D.J.W. and Speece, R.E. (1991). A database of chemical toxicity to environmental bacteria and its use in interspecies comparisons and correlations. J. Water Pollut. Control Fed. 63 (3): 198–207. doi: 10.2307/25043983. Bouchard, D.C., Williams, M.K., and Surampalli, R.Y. (1992). Nitrate contamination of groundwater: sources and potential health effects. J. Am. Water Works Assoc. 84 (9): 85–90. http://www.jstor.org/stable/41293852. Camp, T.R. and Meserve, R.L. (1974). Water and Its Impurities, 2e. Stroudsburg, PA: Dowden, Hutchinson & Ross. Carny, P. (2003). Radioactivity in the air. In: Radionuclide Concentrations in Food and the Environment (ed. M. Pöschl and L.M.L. Nollet), 37–57. Boca Raton, FL: Taylor and Francis. Casanova, L.M., Gerba, C.P., and Karpiscak, M. (2001). Chemical and microbial characterization of household graywater. J. Environ. Sci. Health. 30 (4): 395–401. doi: 10.1081/ESE-100103471. CCME (2012). Scientific Criteria Document for the Development of the Canadian Water Quality Guidelines for the Protection of Aquatic Life: Nitrate Ion. Winnipeg, MB: Canadian Council of Ministers of the Environment. CCME (2017). Canadian Council of Ministers of the Environment. http://st-ts.ccme.ca/en/index. html (accessed 13 March 2018). CDC (2014). “Fluoridation Statistics,” Center for disease control. www.cdc.gov/fluoridation/ statistics/2014stats.htm (accessed 13 March 2018). Chang, J.C., Taylor, P.B., and Leach, F.R. (1981). Use of the Microtox assay system for environmental samples. Bull. Environ. Contam. Toxicol. 26 (2): 150–156. doi: 10.1007/ BF01622069. Chorus, I. (2005). Water safety plans: a better regulatory approach to prevent human exposure to harmful cyanobacteria. In: Harmful Cyanobacteria (ed. J. Huisman, H.C.P. Matthijs and P.M. Visser), 201–227. Springer. Clancy, J.L. and Hansen, J. (1999). Uses of protozoan monitoring data. J. Am. Water Works Assoc. 91 (5): 51–65. http://www.jstor.org/stable/41296546 (accessed 13 March 2018). Cohen, A. and Keiser, D.A. (2016). The effectiveness of overlapping pollution regulation: evidence from the ban on phosphate in dishwasher detergent. Presented at Agricultural & Applied Economics Association Annual Meeting, Boston.

8 Water Constituents and Quality Standards

Cortruvo, J.A. (2017). 2017 WHO guidelines for drinking water quality: first addendum to the fourth edition. J. Am. Water Works Assoc. 109 (7): 44–51. doi: 10.5942/jawwa.2017.109.0087. Corvi, R. and Vanparys, P. (ed.) (2012). International prevalidation study on cell transformation assays. Mutat. Res./Genet. Toxicol. Environ. Mutagen. 744 (1): 1–116. doi: 10.1016/j. mrgentox.2012.02.004. Crapper, D.R., Krishnan, S.S., and Dalton, A.J. (1973). Brain aluminum in Alzheimer’s disease and experimental neurofibrillary degeneration. Science 180 (4085): 511–513. http://www.jstor.org/ stable/1736751. Craun, G.F. (1986). Statistics of waterborne outbreaks in the US (1920–80). In: Waterborne Disease in the United States (ed. G.F. Craun). Boca Raton, FL: CRC Press. Dunn, G., Bakker, K., and Harris, L. (2014). Drinking water quality guidelines across Canadian provinces and territories: jurisdictional variation in the context of decentralized water governance. Int. J. Environ. Res. Public Health 11 (5): 4634–4651. doi: 10.3390/ijerph110504634. Environment Canada (2017). Wastewater systems effluent regulations. http://laws-lois.justice.gc. ca/eng/regulations/SOR-2012-139 (accessed January 2017). EPA-HQ-OW-2008-0878-0298 (2013). National Primary Drinking Water Regulations: Revisions to the Total Coliform Rule, 40 CFR Parts 141 and 142. European Report (2012). Final adoption of household detergent regulation, Europe Information Service SA, 308680. Falconer, I.R., Runnegar, M.T.C., Buckley, T. et al. (1989). Using activated carbon to remove toxicity from drinking water containing cyanobacterial blooms. J. Am. Water Works Assoc. 81 (2): 102–105. http://www.jstor.org/stable/41292380. Feachem, R.G., Bradley, D.J., Garelick, H., and Mara, D.D. (1981). Health Aspects of Excreta and Sullage Management – A State-of-The-Art Review. Washington, DC: The World Bank. Ferguson, J.E. (1990). The Heavy Elements: Chemistry, Environmental Impact and Health Effects. Toronto: Pergamon Press. Frontinus, S.J. (1973). The Water Supply of the City of Rome, A.D. 97 (trans. F. Herschel). Boston: New England Water Works Association. Haas, C.N. (1996). How to average microbial densities to characterize risk. Water Res. 30 (4): 1036–1038. doi: 10.1016/0043-1354(95)00228-6. Health and Welfare Canada (1981). Tap Water Consumption in Canada, Report No. 82-EHD-80, Department of National Health and Welfare, Ottawa, Canada. Health Canada (2009). Guidelines for Canadian Drinking Water Quality—Guideline Technical Document- Radiological Parameters. Ottawa, ON: Health Canada https://www.canada.ca/en/ health-canada/services/publications/healthy-living/guidelines-canadian-drinking-water-quality­ guideline-technical-document-radiological-parameters.html (accessed March 2017). Health Canada (2017). Guidelines for Canadian Drinking Water Quality. Ottawa, ON: Health Canada www.hc-sc.gc.ca/ewh-semt/pubs/water-eau/sum_guide-res_recom/index-eng.php (accessed March 2017). Hellman, T.M. and Small, F.H. (1974). Characterization of odor properties of 101 petrochemicals using sensory methods. J. Air Pollut. Control Assoc. 24 (10): 979–982. doi: 10.1080/ 00022470.1974.10470005. Henrie, T., Plummer, S., and Roberson, J.A. (2017). Occurrence and state approaches for addressing cyanotoxins in US drinking water. J. Am. Water Works Assoc. 109 (2): 40–47. doi: 10.5942/jawwa.2017.109.0022. Henze, M. and Comeau, Y. (2008). Wastewater characterization. In: Biological Wastewater Treatment: Principles, Modelling and Design (ed. M. Henze, M.C.M. van Loosdrecht, G.A. Ekama and D. Brdjanovic), 33–52. London: IWA Publishing.

201

202

Theory and Practice of Water and Wastewater Treatment

Hlavsa, M.C., Roberts, V.A., and Anderson, A.R. et al. (2011). Surveillance for waterborne disease outbreaks and other health events associated with recreational water – United States, 2007–2008. https://www.cdc.gov/mmwr/preview/mmwrhtml/ss6012a1.htm?s_cid=ss6012a1_w (accessed 14 March 2018). Hoffman, F.A. and Bishop, J.W. (1994). Impacts of a phosphate detergent ban on concentrations of phosphorus in the James River, Virginia. Water Res. 28 (5): 1239–1240. doi: 10.1016/0043-1354 (94)90212-7. Holeton, C., Chambers, P.A., and Grace, L. (2011). Wastewater release and its impacts on Canadian waters. Can. J. Fish Aquat. Sci. 68: 10, 1836–1859. doi: 10.1139/f2011-096. Inland Waters Directorate (1979). Water Quality Sourcebook, a Guide to Water Quality Parameters. Ottawa, Canada: Environment Canada, Inland Waters Directorate. Isom, B.G. (1990). Aquatic toxicity testing. In: Toxicity Reduction in Industrial Effluents (ed. P.W. Lankford and W.W. Eckenfelder Jr.), 18–34. New York: Van Nostrand Reinhold. Kim, K., Kabir, E., and Jahan, S.A. (2016). A review on the distribution of hg in the environment and its human health impacts. J. Hazard. Mater. 306: 376–385. doi: 10.1016/j. jhazmat.2015.11.031. Köck-Schulmeyer, M., Villagrasa, M., López de Alda, M. et al. (2013). Occurrence and behavior of pesticides in wastewater treatment plants and their environmental impact. Sci. Total Environ. 458–460: 466–476. doi: 10.1016/j.scitotenv.2013.04.010. Koorse, S.J. (1990). MCL noncompliance: is the laboratory at fault. J. Am. Water Works Assoc. 82 (2): 53–58. http://www.jstor.org/stable/41292623. Kopeloff, L.M., Barrera, S.E., and Kopeloff, N. (1942). Recurrent conclusive seizures in animals produced by immunologic and chemical means. Am. J. Psychiatry 98 (6): 881–902. doi: 10.1176/ ajp.98.6.881. Lowry, J.D. and Lowry, S.B. (1988). Radionuclides in drinking water. J. Am. Water Works Assoc. 80 (7): 50–64. http://www.jstor.org/stable/41292220. Madia, F., Worth, A., and Corvi, R. (2016). Analysis of Carcinogenicity Testing for Regulatory Purposes in the European Union. Joint Research Centre, European Commission. doi: 10.2788/ 547846. Masten, S.J., Davies, S.H., and McElmurry, S.P. (2016). Flint water crisis: what happened and why? J. Am. Water Works Assoc. 108 (12): 22–34. doi: 10.5942/jawwa.2016.108.0195. Meier, J.R. and Daniel, F.B. (1990). The role of short-term tests in evaluating health effects associated with drinking water. J. Am. Water Works Assoc. 82 (11): 48–56. http://www.jstor.org/ stable/41293049. Metcalf and Eddy:AECOM (2007). Water Reuse: Issues, Technologies, and Applications (ed. T. Asano, F.L. Burton, H.L. Leverenz, et al.). New York: McGraw-Hill. Metcalf and Eddy:AECOM (2014). Wastewater Engineering: Treatment and Resource Recovery (ed. G. Tchobanoglous, H.D. Stensel, R. Tsuchihashi and F.L. Burton). New York: McGraw-Hill. Nelson, D.L., Lehninger, A.L., and Cox, M.M. (2008). Principles of Biochemistry, 5e. New York: WH Freeman Publishers. Nemerow, N.L. and Dasgupta, A. (1991). Industrial and Hazardous Waste Treatment. New York: Van Nostrand Reinhold. NSF/ANSI Standard 350-1 (2011). On-Site Residential and Commercial Graywater Treatment Systems for Subsurface Discharge. NSF International. Ohanian, E.V. (1992). New approaches in setting drinking water standards. J. Am. Coll. Toxicol. 11 (3): 321–324. doi: 10.3109/10915819209141869. Okun, D.A. (1997). Distributing reclaimed water through dual systems. J. Am. Water Works Assoc. 89 (11): 52–64. http://www.jstor.org/stable/41296070.

8 Water Constituents and Quality Standards

OMOECC (2008a). Design guidelines for drinking water systems, Ontario Ministry of the Environment and Climate Change. https://www.ontario.ca/document/design-guidelines­ drinking-water-systems. (accessed January 2017). OMOECC (2008b). Design guidelines for sewage works, Ontario Ministry of the Environment and Climate Change. https://www.ontario.ca/document/design-guidelines-sewage-works/ (accessed January 2017). Perl, D.P. and Moalem, S. (2006). Aluminum and Alzheimer’s disease, a personal perspective after 25 years. J. Alzheimer’s Dis. 9 (3): 291–300. Persich, B. (2016). Great Caesar’s ghost: water supply management, 97 AD and now. J. Am. Water Works Assoc. 108 (10): 60–72. doi: 10.5942/jawwa.2016.108.0161. Pons, W., Young, I., Truong, J. et al. (2015). A systematic review of waterborne disease outbreaks associated with small non-community drinking water systems in Canada and the United States. PLoS One 10: 10. doi: 10.1371/journal.pone.0141646. Post, G.B., Atherholt, T.B., and Cohn, P.D. (2012). Health and aesthetic aspects of drinking water. In: Water Quality & Treatment: A Handbook on Drinking Water, 6e (ed. J.K. Edzwald). McGraw-Hill, Toronto: American Water Works Association. Rabb-Waytowich, D. (2009). Water fluoridation in Canada: past and present. J. Can. Dent. Assoc. 75 (6): 451–454. www.cda-adc.ca/jcda/vol-75/issue-6/451.html. Raunkjær, K., Hvitved-Jacobsen, T., and Nielsen, P.H. (1994). Measurement of pools of protein, carbohydrate and lipid in domestic wastewater. Water Res. 28 (2): 251–262. doi: 10.1016/0043­ 1354(94)90261-5. Reiber, S., Kukull, W., and Standish-Lee, P. (1995). Drinking water aluminum and bioavailability. J. Am. Water Works Assoc. 87 (5): 86–100. http://www.jstor.org/stable/41295104. Rose, J.B., Sun, G.S., Gerba, C.P., and Sinclair, N.A. (1991). Microbial quality and persistence of enteric pathogens in graywater from various sources. Water Res. 25 (1): 37–42. doi: 10.1016/ 0043-1354(91)90096-9. Sayre, I.M. (1988). International standards for drinking water. J. Am. Water Works Assoc. 80 (1): 53–60. http://www.jstor.org/stable/41290894. Schock, M.R. and Neff, C.H. (1988). Trace metal contamination from brass fittings. J. Am. Water Works Assoc. 80 (11): 47–56. http://www.jstor.org/stable/41292061. Shammas, N.K. and Wang, L.K. (2011). Water Supply and Wastewater Removal. [Fair, Geyer, and Okun’s Water and Wastewater Engineering], 3e. Hoboken, NJ: John Wiley & Sons. Smith, D.W. and Knoll, H. (ed.) (1986). Cold Climate Utilities Manual. Montreal, QC: Canadian Society for Civil Engineering. Stacey, G. (2012). Current developments in cell culture technology. In: New Technologies for Toxicity Testing (ed. M. Balls, R.D. Combes and N. Bhogal), 1–13. New York, NY: SpringerVerlag. doi: 10.1007/978-1-4614-3055-1. Tijani, J.O., Fatoba, O.O., Babajide, O.O., and Petrik, L.F. (2016). Pharmaceuticals, endocrine disruptors, personal care products, nanomaterials and perfluorinated pollutants: a review. Environ. Chem. Lett. 14 (1): 27–49. doi: 10.1007/s10311-015-0537-z. Troester, M., Brauch, H.-J., and Hofmann, T. (2016). Vulnerability of drinking water supplies to engineered nanoparticles. Water Res. 96: 255–279. doi: 10.1016/j.watres.2016.03.038. US Government (2000). The Safe Drinking Water Act as Amended by The Safe Drinking Water Act of 1996, Public Law 104–182, August 6, 1996, US Government Printing Office, Washington, DC. USEPA (2005). Guidelines for Carcinogen Risk Assessment (EPA/630/P-03/001B). Washington, DC: USEPA. USEPA (2017a). National primary drinking water regulations. https://www.epa.gov/ground-water­ and-drinking-water/table-regulated-drinking-water-contaminants (accessed January 2017).

203

204

Theory and Practice of Water and Wastewater Treatment

USEPA (2017b). Secondary drinking water standards: guidance for nuisance chemicals. https:// www.epa.gov/dwstandardsregulations/secondary-drinking-water-standards-guidance-nuisance­ chemicals#table-of-secondary (accessed January 2017). USEPA (2017c). Title 40 Part 141-National primary drinking water regulations implementation. www.ecfr.gov/cgi-bin/retrieveECFR?gp=1&SID=9e6bacad9eb5c6e10d713713355ae618&ty =HTML&h=L&mc=true&n=pt40.25.141&r=PART#se40.25.141_170 (accessed February 2017). USEPA (2017d). National Summary of Impaired Waters and TMDL Information. iaspub.epa.gov/ waters10/attains_nation_cy.control?p_report_type=T#status_of_data (accessed January 2017). Verschueren, K. (2009). Handbook of Environmental Data on Organic Chemicals, 5e, vol. 1. New York: John Wiley & Sons. Vogel, J., Royal, E., and Snoeyink, V. (1997). Zebra mussels linked to taste, odor problems. Opflow 23 (7): 1, 4–5. WHO (2011). Guidelines for Drinking-Water Quality, Health Criteria, 4e. Geneva: World Health Organization. Wilkinson, J.L., Hooda, P.S., Barker, J. et al. (2016). Ecotoxic pharmaceuticals, personal care products, and other emerging contaminants: a review of environmental, receptor-mediated, developmental, and epigenetic toxicity with discussion of proposed toxicity to humans. Crit. Rev. Environ. Sci. Technol. 46 (4): 336–381. doi: 10.1080/10643389.2015.1096876.

SECTION111

05/09/2018

9:18:13

Page 205

205

Section III Water and Wastewater Treatment

SECTION111

05/09/2018

9:18:13

Page 206

207

9 Water and Wastewater Treatment Operations Water and wastewater treatment plants are designed on a unit operations concept in which one operation is optimized to accomplish one task, although more than one problem substance may be remedied in the operation. Each operation generally has ramifications on other downstream treatment processes, and tradeoffs between increasing the efficiency of one process or another depend on water characteristics and costs of each operation. A brief description of the processes used for water and wastewater treatment is given here with an explanation of where the processes may occur in the treatment stream. There are a number of manufacturers that supply units designed for one process to prefabricated package plants that incorporate a number of individual unit operations. Package plants may be suitable for smaller installations; larger installations are usually custom designed. Basins in a treatment operation may be built in parallel or in series. Either option may be dictated by the efficiency of the process and other considerations. Multiple units also provide the capability to treat water or wastewater, while one unit is out of service because of breakdown or routine maintenance. In the case of water treatment, a multiple-barrier approach (different unit processes used together or in sequence) is encouraged, in particular, for the reduction of or inactivation of pathogens (WHO 2011). In smaller plants, where the design of multiple units is not economical or feasible, bypasses are designed to accommodate shutdown of a unit.

9.1 Water Treatment Operations The choice of treatment operations depends on the quality and variability of the raw water source and the treatment objectives, which may vary for industrial as opposed to municipal needs. A thorough survey of the quality and quantity of all possible sources is the first and most important step for designing a water supply process. Compromising the water survey can prove to be very costly in the long run through payment for more complex and expensive treatment operations. Water treatment operations must be designed to handle the extremes in raw water quality variation to provide an acceptable product water at all times. The unit operations that may be incorporated into a water treatment plant are listed alpha­ betically with some comments on their location in the treatment train: Activated Carbon. The activated carbon particles may be granular or powdered; the treat­ ment capabilities of each of them vary to a degree. Activated carbon is a broad-scale adsorbent of dissolved substances. Dissolved, colloidal, and particulate substances are attracted and attached to the surface of the carbon particles. It is used to remove taste- and odor-causing compounds as

Theory and Practice of Water and Wastewater Treatment, Second Edition. Ronald L. Droste and Ronald L. Gehr.

 2019 John Wiley & Sons Inc. Published 2019 by John Wiley & Sons Inc.

Companion website: www.wiley.com/go/droste/water

208

Theory and Practice of Water and Wastewater Treatment

well as toxic organic chemicals. Precipitation and other chemical reactions also occur on the carbon surface. Activated carbon may be used for dechlorination. A variety of activated carbon adsorbers can be designed, including batch and continuous flow units. The adsorption capacity of the carbon is eventually exhausted. The carbon is regenerated by heating the carbon, which burns and volatilizes the substances accumulated on it. Instead of heat, strong acids or bases or other solvent solutions can be used to regenerate the carbon. Smaller operations normally do not regenerate the carbon onsite. Activated carbon can also be used as one of the media in a filtration unit. Advanced Oxidation. Highly reactive oxidizing agents such as ozone, hydrogen peroxide, chlorine, and permanganate are used for a variety of purposes. Taste and odor removal, oxidation of iron and manganese to states where they will precipitate, oxidation of natural organic matter, and oxidation of emerging contaminants are common applications. Aeration and Air Stripping. Aerators expose water to air to remove volatile dissolved components that are in excess of their saturation concentration. Some toxic organics are volatile. Taste- and odor-causing compounds may be removed to satisfactory levels. Groundwaters may have a high CO2 content that will be stripped to tolerable levels in aeration. Volatile organic chemicals (VOCs) will be removed in these processes. The transfer of gases from the atmosphere to the water will also be effected. The addition of dissolved oxygen will enhance the oxidation of iron, manganese, and other metals to higher and more insoluble oxidation states. These precipitates will be removed in sedimentation basins and filtration units. Aeration is one of the first treatment operations applied to a water. It can be designed as an aesthetically pleasing spray aerator open to public view. Ballasted Flocculation. A ballasted flocculation process is an enhanced physical–chemical three-stage solids removal process. The first stage is a rapid mixer where a chemical coagulating agent is added to raw water to coagulate solids. This is followed by a flocculation tank where polymer and microsand are added. Under gentle mixing conditions, the flocs attach to the microsand with bonding assisted by the polymer. This is followed by clarification. The sandsludge is collected and sent to a hydro-cyclone to separate the sand from the sludge; the microsand is returned to the flocculation basin. The process has a higher rate of throughput and is more efficient than a conventional coagulation–flocculation–sedimentation sequence. Biological Treatment. Biological treatment is used for ammonia nitrogen conversion to nitrate, nitrogen removal and removal of organics in water. It can be accomplished in filters, fluidized beds, or packed beds containing granular activated carbon. Ammonia nitrogen (NH3–N) makes a water biologically unstable, and high concentrations of it may be converted to nitrate (NO3 ). Biological removal of nitrogen is accomplished by converting nitrates to nitrogen gas (N2) (denitrification). The nitrogen gas is stripped from the water. It is an alternative to ion exchange. Biological removal of organic matter lessens the amount of disinfection by-products formed when disinfection agents are added at the end of the treatment plant and renders water less hospitable for growth or regrowth of microorganisms in the distribution system. The water is more biologically stable. Chemical Feed Mixers. Many processes rely on the addition of chemical agents.1 Mixers are designed to disperse the chemicals rapidly and thoroughly throughout the water. Coagulant Recovery. The sludge generated with alum and iron coagulant salts may be regenerated by addition of acid. Recalcination is used to recover calcium oxide (CaO) from 1 Some chemical agents used in water and wastewater treatment are given in the appendix along with their bulk densities.

9 Water and Wastewater Treatment Operations

calcium carbonate (CaCO3) sludge by heating the sludge to drive off carbon dioxide (CO2). Calcium oxide is then slaked by adding water to convert it to lime [Ca(OH)2]. Coagulation. Coagulation is the process of adding chemical reagents in a mixing device to destabilize colloidal particles and allow them to agglomerate or flocculate with other suspended particles to form larger, more readily settled particles. Coagulating agents used are typically iron and aluminum salts along with lime, and commercially made (sometimes natural) polymers. Coagulation reactions are fast and occur in a rapid mixing device. It is essential that the coagulant be dispersed throughout the water to contact and react with the target substances. Dechlorination. Dechlorination is the removal of excess chlorine. Activated carbon and sulfur dioxide (SO2) are two common dechlorinating agents. Disinfection. Disinfection is the removal or inactivation of pathogenic microorganisms (not necessarily sterilization). Chemical agents, commonly chlorine (Cl2), chlorine dioxide (ClO2), chloramines (NH3 xClx), or ozone may be used, or the water may be exposed to ultraviolet (UV) light. Chloramines are formed by reacting ammonia with chlorine. The disinfection tank or device (such as a UV chamber) maintains the water in contact with the dose of disinfectant for a time long enough to assure the required log reductions in the indicator bacteria. It is exceedingly rare to find raw water that would not require disinfection. North American practice advocates the addition of a small amount of chlorine (and possibly ammonia) to form chloramines in the treated water, which maintains a small disinfectant residual in the distribution system when other disinfectants are used as the primary disinfectant. Disinfection (or possibly fluoridation) is the last treatment applied to water. Dissolved Air Flotation (DAF). In this process, air is injected at high pressure into water in one tank and then released in a second tank where the water is exposed to the atmosphere. The reduction of pressure causes the release of fine bubbles of air to which solids attach as the bubbles rise. The thickened sludge that accumulates at the surface of the second tank is skimmed to remove the solids. DAF is an alternative to sedimentation. Electrodialysis. See reverse osmosis. Filtration. Filtration accomplishes polishing of a water and is required for almost every water. Filtration follows sedimentation if the latter is provided. Water moves through tanks that contain sand and/or other types of media. Fine solids that did not settle out in a sedimentation basin will be entrapped in the filter. There also will be significant removal of bacteria in a filter but not enough (more than 1-log reduction of bacteria is usually required) to provide a safe water. Larger microorganisms, such as protista, are completely removed in a properly operated filter. Two filtration types are in common use. Slow sand filters have only sand media. They are cleaned by scraping off the top layer of media on a periodic basis as the filter clogs. Rapid filters are sand filters or multimedia filters that have a layer of anthracite, sand, and possibly other media. Loading rates of rapid filters are much higher than slow sand filters. Rapid filters are cleaned by backwashing – reversing the flow of water through the media and pumping at a rate sufficient to expand the media. Backwashing is necessary every 1–4 days, depending on influent water quality. The influent to rapid filters generally must have a coagulating agent (filter aid) added to it at some upstream location. Flow through rapid and slow sand filters is due to gravity. Pressure filters, where water is forced through the filter by applied pressure in a completely enclosed unit, are used in some installations. Roughing filters that contain coarse media may be used to prefilter water with a very high suspended solids (SS) content. Raw water that is of high quality may require filtration only to remove the small quantities of SS that are present. Otherwise, rapid filters are preceded by coagulation, flocculation, and sedimentation. Precoat filtration uses a permeable cloth that will support a relatively thin layer of fine filtration medium, usually diatomaceous earth. Water is

209

210

Theory and Practice of Water and Wastewater Treatment

passed through the filter until clogging causes headloss to reach a cutoff value. The filtration medium is usually discarded after a run. This process is a surface filtration as opposed to rapid or slow sand filtration where filtration occurs through the depth of the filter. Flocculation. Flocculators provide gentle agitation of a water that has been coagulated to promote particle contact and formation of larger particles. Hydraulic or mechanically driven flocculators may be designed. Flocculators follow the rapid mixing coagulation tank and precede sedimentation and filtration units. Fluoridation. Fluoride is added to waters to reduce the incidence of dental caries in the population. It is added by a chemical feeder. Pumping and flowing through the distribution pipes will ensure that the fluoride ion is thoroughly dispersed in the water. Fluoride is added at the end of the treatment train, where its concentration will not be affected by coagulation, precipitation, sedimentation, and adsorption processes. When fluoride concentrations in the raw water exceed recommended limits, ion exchange with activated alumina is generally used to remove the excess fluoride. Hardness Removal. See softening. Ion Exchange. Ion exchange can be used to remove hardness ions (multivalent cations), nitrates, or other inorganic constituents. Ion exchangers contain large molecular weight organic substances (resins) that have the ability to exchange one ion for another. For instance, there are resins that will exchange hydrogen ions for calcium ions or other positively charged ions. Most ion exchangers are synthetic, commercial products. Naturally, occurring ion exchangers (crystalline aluminosilicates) are known as zeolites. Eventually, the ion exchanger’s capacity is exhausted and concentrated solutions of acid, base, or salt are used to regenerate the ion exchanger. Membrane Treatment. Also see reverse osmosis and ultrafiltration. Membrane treatment can be used for removal of both soluble and particulate matter. Applications can include softening, reduction of microbes, and SS removal. Ozone (O3) as noted above is widely used as a disinfectant. Ozone is highly reactive; therefore, it leaves no residual disinfection power (or tastes) in water as it moves through the distribution system. Because of its high reactivity, it is also commonly used in advanced oxidation processes. Pipes, Channels, and Other Conduits. Flow in pipes and channels should have a minimum velocity of 0.3 m s 1 to avoid deposition of SS. Recommended velocities at various locations in the plant depend on the state of the water. Prechlorination or Pre-oxidation. Instead of using aeration, high concentrations of nuisance metals, such as Fe and Mn and other metals that may be toxic, can be oxidized to more insoluble states by chlorine or other oxidants. The precipitates will be removed by downstream processes of coagulation, flocculation, sedimentation, and filtration. Sulfides (S2 ) are also oxidized to sulfates (SO24 ). This treatment is also used to retard microbial growth throughout the plant. This operation is, as the name implies, at the front-end of the treatment plant. Elemental chlorine should not be used at this point because it will have a greater tendency to form chloro­ organics, some of which are carcinogenic. Other chlorine derivatives such as chloramines or other oxidation agents such as permanganate are preferable. Recarbonation. In water-softening plants, it is often necessary to add an excess of lime to facilitate removal of hardness ions. After sedimentation of the CaCO3 and Mg(OH)2 precipitates, the pH of the water is high because of the excess lime. Carbon dioxide is added to neutralize the excess OH . Reverse Osmosis (Membrane) and Electrodialysis. These technologies are used for the removal of high concentrations of dissolved solids. Reverse osmosis essentially “filters” dissolved solids from the water by forcing the water through a membrane by applying pressure in excess of the osmotic pressure of the dissolved components in the solution. Electrodialysis uses electric

9 Water and Wastewater Treatment Operations

current that induces ions to migrate through a membrane. These technologies are primarily used to desalinate brackish waters or to remove ions such as fluoride. They are also used for softening. Suspended solids must be removed to a low level before water is subjected to reverse osmosis to prevent fouling of the membrane. Screens and Bar Racks. Screens and bar racks are used to remove coarse debris from a raw water source. This debris may damage pumps or deposit in channels, causing clogging. They are located at the intake for the water source and at the beginning of the treatment plant. Screenings are collected and disposed at a landfill site. Sedimentation. Exposing the water to relatively quiescent conditions will allow settleable solids to be removed by the action of the force of gravity. The sludge accumulated in these tanks may be disposed of in landfills or into sewers for ultimate removal in the wastewater treatment plant, avoiding discharge of the sludge to a receiving water body. Sedimentation that has not been preceded by coagulation and flocculation is known as plain sedimentation. Raw waters that contain a high sediment load may be settled in a presedi­ mentation basin to remove the readily settled particles. Then a chemical assist may be provided through the addition of coagulant followed by flocculation and larger sedimentation basins to remove slower settling particulates. Conventional clarifiers or sedimentation basins are usually rectangular or circular in shape. Clarifier performance is related to its surface area. The effective surface area in a given volume can be markedly increased by inserting inclined plates spaced at regular intervals to make lamella clarifiers or similarly installing inclined tubes to make tube clarifiers, both of which improve settling efficiency. Waters, particularly groundwaters that have a low concentration of SS, may not require sedimentation. They can be directly filtered. Sludge (Residuals) Concentration and Dewatering. The purpose of sludge concentration units is to reduce the volume of sludge for ultimate disposal in a landfill. Vacuum dewatering, filter presses, belt filters, and centrifugation are commonly used to separate the solids from the water. Chemical agents are added to the sludge going into the dewatering unit to improve the efficiency of the device. Sludges are also dewatered in drying beds that may be solar or freeze– thaw types. Sludge Thickening. Thickening is a settling process where the sludge concentrates in a basin by the action of gravity. Thickeners are rectangular or circular. The supernatant from a gravity thickener (or subnatant from a DAF) contains significant concentrations of colloidal and suspended matter, and it is returned to the head of the plant or discharged to a receiving water. Thickened sludge is usually further dewatered before disposal. Softening. Softening is the removal of multivalent cations (hardness) in the water. As noted above, ion exchange can be used to accomplish this task. Lime–soda ash [Ca(OH)2-Na2CO3] softening involves the addition of these agents in a particular sequence, followed by flocculation, sedimentation, and filtration. The alkalinity added by lime and soda ash causes CaCO3 and Mg (OH)2 to precipitate and be removed in the sedimentation and filtration units. Softening will be practiced in conjunction with coagulation. It is possible to recover the lime. Increasingly, membrane processes with pore sizes in the nanofiltration range are being used for softening. Chemicals are saved at a cost of energy. Super-chlorination. Chlorine is a strong oxidizing agent that can destroy some taste and odor (T&O) compounds. Also, in plants where filtration is not present, and there is a significant risk of protozoan pathogens, elevated levels of chlorination beyond those normally required to disinfect water will be required. After the chlorine is added, the water is held in a contact basin for the required contact time. Super-chlorination will have to be followed by dechlorination to reduce

211

212

Theory and Practice of Water and Wastewater Treatment

residual chlorine and the tastes and odors arising from chlorine itself. The dechlorinating agent is usually sulfite, although activated carbon can also be used for dechlorination. Ultrafiltration. In ultrafiltration or nanofiltration, pressure is used to drive a liquid through a membrane. The process is used to separate higher molecular weight compounds and/or microorganisms from water. Microfiltration uses membranes with larger-sized openings to filter small particles, bacteria, and protozoans from a water. For smaller treatment plants, these filtration processes may be used in place of granular media filtration. Water Stabilization. Water stabilization refers to the adjustment of the pH, alkalinity, and calcium content of a water such that there is a slight tendency for the finished water to precipitate CaCO3. The practice is based on the notion that a thin deposit of CaCO3 in the distribution system pipes will form a protective barrier to retard corrosion of the pipes by agents in the finished water; however, research indicates that the practice is not a reliable indicator of corrosion potential (McNeill and Edwards 2001). If softening is practiced, the process must be adjusted to produce the desired concentrations. It may be necessary to add acid or alkaline agents to achieve the desired balance among the three components. Obviously, not all of the above treatment processes will be required for raw water. There is usually more than one process or a combination of processes that can be used for the same purpose. Also the efficiency of a single unit operation may be significantly influenced by upstream operations. It is even possible for treatment to produce negative results. For example, Himberg et al. (1989) found in a few instances that a conventional treatment process as depicted in Figure 9.1 produced negative removals of blue-green algae toxins presumably because of cell lysis and release of toxins to solution in an otherwise proper functioning treatment train. Typical flow charts for water treatment operations are shown in Figures 9.1–9.3; many variations are possible in these configurations. There are many possible substitutions and configuration changes possible. Water treatment processes are classified according to their degree of contaminant removal in Table 9.1. Microbial Contaminants USEPA (2017b) specifies concentration-time (Ct) values (see Chapter 16) at different tempera­ tures for 99.9% (103 or log 3) inactivation of Giardia lamblia cysts for chlorine, ozone, chlorine dioxide, and chloramines that are the most commonly used disinfectants. Ultraviolet disinfection is also permitted; specifications on the process are given in USEPA (2006). G. lamblia cysts require a higher dose of disinfectant than most pathogens. The Ct values may be assumed to achieve greater than 99.99% (104) inactivation of viruses (USEPA 2017b). Likewise WHO (2011)

Figure 9.1 Rapid sand filtration water treatment plant.

9 Water and Wastewater Treatment Operations

Figure 9.2 Lime–soda softening water treatment plant.

specifies Ct values for chlorine, ozone, and chlorine dioxide and UV doses for varying degrees of inactivation of viruses, bacteria, and protozoa. WHO (2011) provides ranges of inactivation of viruses, bacteria, and protozoa for other (non­ chemical) water treatment processes; the ranges are wide. According to the USEPA regulations, the treatment technique must provide at least 99.9% removal or inactivation of G. lamblia cysts (USEPA 2017b). Filtration is generally the most effective treatment process after disinfection for microbe inactivation or removal. Reservoirs Reservoirs are often required to provide a steady supply of water over periods of drought. Some improvement in the quality of river water can also be obtained by building an impoundment from which the water will be drawn. Quiescent conditions in a reservoir will allow SS to settle. Turbidity and some organic matter will be removed. The holding time in a reservoir will cause the devitalization and death of waterborne pathogens. Sunlight will photooxidize some colorcausing compounds. Some equalization of quantity and quality fluctuations will also be achieved. Softening may occur naturally in small lakes or reservoirs. Hot summer days promote high growth rates of algae, which remove carbon dioxide from the water. As a consequence, the pH of the water rises and the remaining dissolved inorganic carbon species are converted into carbonates. The low solubility product of Ca2+ and CO23 will precipitate CaCO3. However, this advantage is not permanent because when the pH decreases, the precipitated CaCO3 can be redissolved if the reservoir water is below saturation.

Figure 9.3 Water treatment plant for Fe–Mn removal in groundwater.

213

Table 9.1 Best water treatment processes for contaminant removal.

B

B

B

Chlorine

B

Activated alumina

A

Granular activated carbon

A

B

Electrodialysis

A

B

Oxidation/ filtration

Nanofiltration

B B

Reverse osmosis

Ion exchange

Adsorption

Lime softening

Precoat filtration

Coagulation sedimentation, DAF, filtration

Contaminant categories

Aeration and stripping

Membrane Processes

Primary inorganic contaminants Antimony

B

Arsenica)

Bb)

Barium Berylium

B

B

B

Cadmium

B

B

B

B

A

A

Chromium

B

Bc)

B

B

A

A

B

B

B

A

B

B

Cyanide Mercury (inorganic)

B

Nickel

B

B

B

B

Nitrite

B

B

Bd)

B

B A A

B A

B Β

Thallium

B

A

Nitrate Selenium

B

B

B

Α

Β

A

Primary organic contaminants Pesticides/herbicidese)

A

A

SOCse)

A

A

VOCse)

B

B

B

Radionuclides Beta particle and photon activity

B

Gross alpha particle activity (excluding radon and uranium)

B B

Radium-226, -228 Uranium

B

B

B

B

B

B

B

Α

Α

Α

B

Secondary and other contaminants Α

Chlorides Color

A

Copper

A

E Α

Dissolved organic carbon Fluoride

A

A

A

A

A

B

Α

A

Α

A

Β

A

Hepatotoxins and neurotoxins from blue-green algaef)

A

Hardness Iron

AOg)

Lead

E

Manganese

AO

Radon

A

A

AO

A

A

AO

A

A

E

A E

A

A

Sulfate

A

Sulfide (hydrogen sulfide)

A

Taste and odor

A

A

A

A A A

Total dissolved solids

A

Total organic carbon Zinc

Α

E

E A

A

A

A E

A

A

a) For arsenic(V). Pre-oxidation may be required to convert arsenic(III) to arsenic(V).

b) To obtain high removals, iron:arsenic ratio must be at least 20 : 1.

c) For chromium(III) only.

d) For selenium(IV) only.

e) Consult the references below for specific compounds.

f) From Himberg et al. (1989).

g) “AO” indicates appropriate in conjunction with an oxidation process as designated by Elder and Budd (2012)

Source: From USEPA (2017a), Elder and Budd (2012) and Hamann et al. (1990). A “B” indicates best available technology as designated by USEPA (2017a); an “A” indicates appropriate

technology or “AO” (appropriate in conjunction with oxidation) designated by Elder and Budd (2012); an “E” indicates 90–100% removal by Hamann et al. (1990).

216

Theory and Practice of Water and Wastewater Treatment

Reservoir impoundment may cause changes in water quality because of T&O compound production. Vegetative growth will occur in shallow bodies of water, causing changes in water quality. Growth of bacteria on the reservoir bottom, as well as algal growth, may be responsible for these T&O problems. The benefits of building an impoundment in a river only for quality improvement must be weighed against the disadvantages. The costs of impoundment are generally high. Preliminary treatment gained in the reservoir is a permanent gain. 9.1.1 Home Water Treatment Units A number of point-of-use (applied at the water tap) or point-of-entry (applied at water supply point to the residence) treatment units, including water distillation units, activated carbon filters, reverse osmosis systems, and ion exchangers, have appeared on the market in recent years. Myths about water quality coupled with exaggerated claims from many manufacturers of these devices can lead to erroneous conclusions by the consumer. Distillation units produce water of high purity. The other units treat water as described above depending on the treatment process incorporated into the unit. The onus is on the user to maintain and operate these units properly according to manufacturer’s instructions. Chlorine or its derivatives will be removed in carbon filters and the filter medium provides support surfaces for bacteria to colonize. Bacteria can populate filters, particularly those containing activated carbon, to high levels resulting in slug discharges (Geldreich et al. 1985). There is a risk that adventitious, opportunistic pathogens may proliferate. Many activated carbon filters are impregnated with silver, which is toxic to microorganisms, to retard microbial growth. When these devices are properly designed and used they can supplement a central treatment system (Regunathan et al. 1983).

9.2 Wastewater Treatment Unit Operations Wastewaters are normally treated by a combination of physical–chemical and biological operations. However, it is possible to treat wastewaters solely with physical–chemical methods. Again, operations are listed alphabetically for easy reference. Activated Carbon Treatment. Activated carbon may be added to biological treatment processes or used as a treatment by itself. Activated carbon is used to adsorb toxic substances such as metals and pesticides that may be present in a wastewater. Both powdered and granularactivated carbon are used; the former is more common for seasonal applications, the latter for year-round use. Advanced Oxidation. Advanced oxidation processes, generally combinations of ozone, peroxide, UV, and others are used for removal of low concentrations of biologically recalcitrant substances that are not readily mineralized. Aerobic Biological Treatment. Many types of processes are possible for the aerobic biological removal of dissolved and suspended organics. In the process, air (oxygen) is supplied to microorganisms that are in contact with the wastewater. The microorganisms metabolize the organic material into carbon dioxide and other endproducts and new biomass. The putrescibility and soluble oxygen demand are reduced to a small amount. There is a variety of configurations and operational modes. Some processes such as trickling filters or biodisks (rotating biological contactors) use a solid support medium to provide surfaces on which bacteria grow and accumulate. In a trickling filter, wastewater is sprayed over a rock or synthetic media bed. Wastewater is applied on a continuous or an intermittent basis.

9 Water and Wastewater Treatment Operations

Bacteria attach and grow on the media. Natural air currents supply oxygen. Biodisks work on a similar principle. A series of disks is partially submerged in a tank of wastewater. The disks are rotated to expose the biomass-covered disks to the atmosphere, where oxygen transfer to the biomass occurs. Providing immobilized media for microorganism attachment and growth is known as a fixed-film process. Some processes use submerged fixed or freely suspended media to provide surfaces that retain biomass and improve treatment. The integrated film-activated sludge process incorporates fixed media into the bioreactor. A moving bed bioreactor has media near the specific gravity of water in the bioreactor. The media are prevented from exiting the reactor by screens. The common activated sludge process is a suspended growth process where the micro­ organisms are mixed with the wastewater. Air is pumped into the basin to supply oxygen and mixing. Pure oxygen may be supplied as an alternative to air but supplemental mixing is required. There are many variations of the activated sludge process. Processes may also be designed to oxidize nitrogen or remove nitrogen from the wastewater. It is also possible to design a process for increased removal of phosphorus. All aerobic biological treatment processes convert some of the dissolved organics into biological solids that must normally be removed in a sedimentation basin following the biological treatment unit. Aerobic biological treatment is often preceded by sedimentation to physically remove suspended organics that would be degraded in the biological treatment process at more expense. Air Strippers. Volatile compounds or gases present at supersaturated concentrations are amenable to air stripping. Air stripping units are devices that enhance the mass transfer between liquid and the atmosphere. This is accomplished by increasing the surface area of the liquid to maximize its exposure to the atmosphere. There are many types of stripping devices, ranging from highly efficient countercurrent flow to diffused aeration systems. In countercurrent flow devices, the wastewater is injected at the top of a column that contains media such as ceramic berl saddles, and air is injected at the bottom of the column. Diffused aerators inject air into the liquid through porous plates located near the bottom of a tank. Ammonia Stripping. Ammonia is a volatile component that may be removed by exposing the wastewater to the atmosphere. A high pH favors the formation of ammonia as opposed to ammonium ion. Air strippers are used to remove the non-ionized ammonia. Anaerobic Biological Treatment. High-strength wastewaters are amenable to treatment by an anaerobic process. Industrial and agricultural wastewaters are often highly concentrated. Anaerobic treatment occurs in enclosed reactors to prevent access of oxygen. The anaerobic microorganisms that are in contact with the wastewater convert the dissolved and suspended organics into biomass and methane. Significant quantities of methane are produced, and it may be used as fuel. As in aerobic biological treatment operations, some processes incorporate solid support media for the microorganisms (fixed-film processes), whereas others keep the micro­ organisms in suspension. Fixed-film processes include anaerobic filters and fluidized beds. The suspended growth process is also known as an anaerobic contact process. A unique process in anaerobic treatment is the upflow anaerobic sludge blanket reactor, wherein dense granules of microorganisms are retained in the reactor by inertia. It is often not necessary to presettle the influent to an anaerobic reactor. The biological solids produced in an anaerobic process must be settled in a clarifier that may follow the process or be incorporated into the anaerobic reactor. Fixed-film and suspended growth processes are among the other options used in anaerobic treatment. Anaerobic digesters are often used to treat solids produced in an aerobic treatment process and in the primary clarifier, considerably reducing the volume of solids finally to be removed

217

218

Theory and Practice of Water and Wastewater Treatment

from the plant. A suspended growth process is usually the only feasible process for a highly concentrated waste dominated by particulate matter. Anammox. The anaerobic ammonium oxidation (anammox) process relies on a group of bacteria that are strictly anaerobic and use nitrate to oxidize ammonia to produce nitrogen gas. Ballasted Flocculation. Similar to the water treatment application, ballasted flocculation significantly enhances the performance of the primary sedimentation process. See the descrip­ tion of ballasted flocculation for water treatment. Biological Nutrient Removal (BNR). BNR is a biotreatment process able to accomplish nitrogen and phosphorus removal. In addition to an aerobic reactor, there will be anaerobic and anoxic (no dissolved oxygen but nitrate is present) stages in the treatment train. Nitrogen is removed through the oxidation of ammonia to nitrate (nitrification) in an aerobic stage and denitrification in the anoxic stage where nitrate is used as an electron acceptor and converted to nitrogen gas by metabolism of organic carbon. There will be an anaerobic stage to provide conditions necessary for enhanced biological uptake of phosphorus by the bacteria. Bio-P Process. A biological process designed for enhanced phosphorus removal is a bio-P process. This process will require at least one anaerobic stage along with an aerobic stage. Chemical Feed Mixers. Physical–chemical plants rely heavily on the addition of chemical agents.2 Chemical agents are also used in biological treatment plants. Mixers are designed to disperse the chemicals rapidly and thoroughly throughout the water. Chemically Enhanced Primary Treatment (CEPT). Addition of higher amounts of precipi­ tating agents can significantly improve primary clarifier performance and the size of the clarifier can be reduced. Coagulant Recovery. Physical–chemical plants use methods for coagulant recovery described under water treatment. Coagulation. This process has the same objectives as noted for water treatment, and it is used in physical–chemical wastewater treatment processes. Also, chemicals may be added to precipitate a particular constituent such as phosphorus. Alum is commonly applied to the effluent from an aeration basin for biological treatment to ensure that phosphorus is precipitated in the final clarifier. Cogeneration. Methane produced in an anaerobic treatment or digestion process is similar to natural gas, although it does contain some impurities. Cogeneration is the beneficial use of the methane in a heat or an electricity generator. Comminutors. Comminutors are devices that macerate rags, sticks, paper, and other gross solid debris in the wastewater that can clog channels or otherwise obstruct flow through the plant. These devices are placed downstream of the grit chamber. The cuttings from the comminutor are returned directly to the wastewater. Disinfection. Clarified effluent from the plant may be disinfected to reduce pathogen concentrations prior to discharge to the receiving body of water. Chlorine is usually the disinfectant applied, although UV units are being used at greater frequency. A plug flow contact chamber is used to maintain the wastewater in contact with the disinfectant for the specified time. DAF is used to separate suspended matter from a wastewater. It is primarily used to thicken biological or chemical sludge suspensions. See the description of the process given above. Equalization. When flow quantity and quality variations are significant, holding basins are incorporated near the front of the treatment train to allow wastewater to be input to the plant at a more uniform rate and quality. More uniform flows and concentrations reduce the variability of 2 Some chemical agents used in water and wastewater treatment are given in the appendix along with their bulk densities.

9 Water and Wastewater Treatment Operations

treatment and allow a more compact wastewater treatment plant to be designed with higher utilization of all units. The equalization basin is mixed and aeration may be applied to avoid septic conditions. The amount of biological treatment provided by aerating the wastewater will not be significant. Some stripping of volatile compounds will occur. Fermenter. Biological nutrient removal (see above) requires an anaerobic stage where volatile fatty acids and other readily degradable substrates are required for the phosphate accumulating organisms. Sludge from a primary clarifier may be sent to a fermenter, which is a short detention time, anaerobic reactor where some solids will be fermented to volatile fatty acids. Filtration. Filtration is often used in physical–chemical wastewater treatment processes to remove fine colloidal or solid particles that cannot be settled in a sedimentation basin. Filtration is also used finally to polish effluent from clarifiers that discharge settled effluent from a biological treatment process. It is also used to remove algae in effluent from a stabilization pond process. Flocculation. This process has the same objectives as noted for water treatment. It is not commonly used in treatment plants that use biological processes, but it is an important operation in physical–chemical treatment plants. Grit Chambers. After the wastewater is screened, grit is removed. Grit chambers are sedimentation basins designed to remove nonputrescible matter such as silt and sand that cause abrasive wear on channels and pumps and can accumulate and clog channels. Because the materials removed in a grit chamber are essentially nonbiodegradable, they are collected and disposed of in a landfill site. Modern grit chambers are usually aerated chambers with an offset aerator that induces a small degree of water circulation to keep lighter organic particles in suspension while the heavier grit materials settle. Vortex grit removal devices, where the flow pattern creates a centrifugal force on particulates, are also common. Incineration. Incineration is used for the final stabilization of sludge. An incinerator is a furnace for high-temperature combustion of the sludge. Without emission control, incinerators produce a significant amount of atmospheric contamination. Lagoons. Lagoons are mechanically aerated ponds that provide aerobic biological treatment. Lamella Clarifiers. As for water treatment, sedimentation basins with slanted sheets spaced to provide flow channels can be used for sedimentation. The sheets will have to be cleaned regularly, usually with high-pressure wash. Membrane Technologies. Membrane bioprocesses use a membrane to separate solids from the liquid, replacing a secondary clarifier. Membranes are also used for the selective removal of various agents. Neutralization. Industrial wastes with extremely high or low pH are neutralized by acid or base addition. If biological treatment is to be used, a pH near neutrality is required. Oxidation Ditches. An oxidation ditch is an oval channel with a mechanical aeration device installed to provide aerobic biological treatment. Pipes, Channels, and Other Conduits. Flow in pipes and channels should have a minimum velocity of 0.6–0.76 m s 1 at design average flow to avoid deposition of solids (WEF and ASCE 2010). Some facilities use minimum velocities as low as 0.3 m s 1, but the higher minimum is preferred by more operators. Primary Treatment. Refers to sedimentation (primary sedimentation) and grit removal only. Screens and Bar Racks. Screens and bar racks are used to remove coarse debris from wastewater for the same reasons as in a water treatment plant. This debris may damage pumps or may settle in channels causing clogging. Bar racks are located at the intakes to wet wells in pumping stations in the sewer system as well as at the intake to the wastewater treatment plant. At the wastewater treatment, plant screens will also be used after the bar racks. Materials

219

220

Theory and Practice of Water and Wastewater Treatment

removed on the screens and racks are relatively nonbiodegradable. They are collected and disposed of in a landfill site. Secondary Treatment. Refers to a treatment scheme that includes biological treatment for carbon removal. Sedimentation. Primary clarifiers are designed to remove settleable solids from a wastewater before biologically treating it for dissolved organics. Removal of biodegradable solids by sedimentation is much less expensive than treating them aerobically in an aeration basin. Secondary clarifiers follow biological treatment units and are designed to remove biomass formed during biological treatment and other solids present in the influent to the biological treatment unit. In addition to clarification of the wastewater, a secondary clarifier for a biological treatment process is designed to perform some thickening of the biological sludge that accumulates in it. Sedimentation basins in physical–chemical treatment systems will follow coagulation and flocculation units. Sludge Concentration and Dewatering. The purpose of sludge concentration units is to reduce the volume of sludge for ultimate disposal in a landfill, incinerator, or on land. Vacuum dewatering, filter presses, belt filters, and centrifugation are commonly used to separate the solids from the liquid. Chemical agents are added to the sludge going into the dewatering unit to improve the efficiency of the device. Sludges are also dewatered in drying beds. Sludge Digestion. Sludge removed from primary clarifiers, as well as sludge generated during aerobic biological treatment (often referred to as “biosolids”), may be reduced in quantity and volume by microbial action in aerobic or anaerobic sludge digesters. Aerobic digesters are often used at smaller installations. Larger installations use anaerobic digesters that normally operate in a mesophilic (30–40 °C) or thermophilic (50–55 °C) temperature range. Sludge Thickening. Gravity thickening is a settling process where the sludge concentrates in a basin by the action of gravity. The supernatant from a thickener contains significant concentra­ tions of colloidal and suspended matter, and it is returned to the primary clarifier to pass through the treatment process. Dissolved air flotation may also be used for sludge thickening. Stabilization Ponds. Stabilization ponds are a series of ponds that settle and biologically treat a wastewater. They are not normally used with other treatment processes except possibly a final filter or screen to remove effluent SS. Mechanical aeration is normally not provided in stabilization ponds. In the final ponds in a series of stabilization ponds, fish may be grown for harvest, which is referred to as aquaculture. Aquaculture is not common in North America, but it is practiced in other parts of the world. Struvite Harvesting. Soluble nitrogen and phosphorus are products of biotreatment. Their concentrations are highest in an anaerobic digester where the nitrogen will be in the form of ammonia. Struvite is Mg(NH4)(PO4). Recovering struvite from liquid from a sludge dewatering operation not only recovers valuable nutrients but also considerably reduces the potential for struvite formation, which is a problematic scale in conduits and reactors. Ultrafiltration. Ultrafiltration or nanofiltration and the related process of microfiltration have been described under water treatment operations. They can be used for filtration and the recovery of compounds in a wastewater. Typical wastewater treatment plant configurations for various situations are shown in Figures 9.4–9.8. A physical–chemical treatment plant for wastewater is shown in Figure 9.6. The major difference between physical–chemical treatment plants and biological treatment plants is the processes used to remove dissolved organic matter. The quantities and qualities of sludge produced by a physical–chemical treatment will be different than for a plant incorporat­ ing biological treatment because of the coagulation agents added and the absence of biological oxidation of organics. Indeed, physical–chemical wastewater treatment plants may remove only

9 Water and Wastewater Treatment Operations

Figure 9.4 Activated sludge plant for domestic wastewater treatment.

SS and dissolved phosphorus, through the actions of coagulation and precipitation, respectively. Not all wastes are amenable to biological treatment. A multitude of changes could be made to the treatment plant schemes shown in Figures 9.4 to 9.7. As in water treatment, the wastewater treatment plant is an integrated work where the performance of any individual process is linked to the processes that precede it. The effectiveness of various treatment processes in removing classes of contaminants is given in Table 9.2. A wastewater treatment plant that uses only sedimentation as the major treatment operation is often referred to as a primary treatment plant. Provision of biological treatment is termed secondary treatment. Adding physical–chemical treatment then constitutes advanced waste­ water treatment or tertiary treatment. Biological treatment processes designed for nutrient removal are also referred to as advanced treatment processes. The physical–chemical processes of filtration and carbon adsorption can be added to the biological treatment process depicted in

Figure 9.5 A biological nutrient removal wastewater treatment plant.

221

222

Theory and Practice of Water and Wastewater Treatment

Figure 9.6 Physical–chemical plant for wastewater treatment.

Figure 9.4 to produce a very highly purified wastewater. Such wastewaters can be used to augment drinking water supplies. Land application of raw sewage is another alternative that has been used throughout the world but is generally not accepted in North America because of concern for pathogens. Heavy metals will be problematic for land application of industrial wastes. Irrigation, groundwater recharge, and fertilization (nutrient recycle) are some benefits of land application of sewage. Land application of primary or secondary treated sewage is also practiced and encouraged where feasible. Raw or partially treated sewage may also be discharged to wetlands. Landfills are the most common means of disposal of sludge produced from treatment operations, but a significant quantity of sludges generated from domestic wastes can be applied to soils.

Figure 9.7 An anaerobic wastewater treatment plant for a food or beverage wastewater.

9 Water and Wastewater Treatment Operations

Figure 9.8 Secondary biological wastewater treatment plant at Peterborough, ON. (1) Raw sewage pumping station; (2) grit tanks; (3) screens building; (4) primary clarifiers; (5) secondary (activated sludge) plant one; (6) secondary (activated sludge) plant two; (7) UV disinfection; (8) anaerobic digesters. Source: Municipality of Peterborough. Reproduced with permission of the Municipality of Peterborough.

Table 9.2 Contaminant removal in wastewater treatment processes. Contaminant

Process

Comment

Aerobic biotreatment

Numerous processes. They oxidize some to CO2 and synthesize some to biosolids

Anaerobic biotreatment

Numerous processes. They oxidize some to CO2, reduce some to CH4 and synthesize some to biosolids

Aerobic biotreatment

They oxidize some to CO2, synthesize some to biosolids

Anaerobic biotreatment

They oxidize some to CO2, reduce some to CH4 and synthesize some to biosolids

Advanced oxidation

Requires a strong oxidizing agent such as ozone or hydrogen peroxide and also UV

Organic carbon Dissolved degradable organic C

Particulate degradable organic C (also see “Suspended solids”)

Recalcitrant dissolved organic C including pesticides, microconstituentsa)

Carbon adsorption Biotreatment

Gross debris

Bar racks, screens

Inert large, dense particulates

Grit removal devices

Suspended solids

Sedimentation

Bioprocesses can remove some of these contaminants to varying degrees by sorption to solids or metabolism

Primary sedimentation

The first settling basin in a plant. Removes all types of solids to a degree

CEPT

Primary sedimentation with improved performance due to increased chemical addition. Removes all types of solids to a high degree

(continued )

223

224

Theory and Practice of Water and Wastewater Treatment

Table 9.2 (Continued ) Contaminant

Biosolids

Process

Comment

Coagulation–flocculation– sedimentation

Removes all types of solids to a degree

Ballasted flocculation

Enhanced coagulation–flocculation– sedimentation process, using microsand and higher chemical doses to enhance settling. Removes all types of solids to a higher degree than conventional coag–floc–sed sequence

Secondary sedimentation or DAF

Removes solids exiting from a biotreatment process or preceding chemical precipitate formation processes

Membrane

Replaces sedimentation basin for separation of solids from liquid. Nearly 100% solids removal

Filtration

Used as a polishing process to remove low concentrations of SS

Anaerobic digestion

Metabolizes degradable particulate C to CO2 and CH4. There is some anaerobic biomass synthesis, but it is very low compared to reduction of solids

Aerobic digestion

Metabolizes degradable particulate C to CO2. Oxidation of solids is very high compared to re­ synthesis

Nitrogen NH3 b)

Generated in some aerobic and most anaerobic bioprocesses

NH3

Air stripping

Requires high pH > 9

NH3 → NO3

Nitrification in aerobic treatment

Requires biosolids to remain in bioreactors for a time longer than 6–8 days

NH3 ‡ NO3

Anammox

Requires higher concentrations of nitrogen species

NO3

Denitrification bioprocess

Requires degradable organic carbon

NH‡4

Struvite harvester

Nitrogen reductions will be relatively low

Phosphorus

Heavy metals

Pathogens a)

Bio-P process

Anaerobic-aerobic reactors required

Chemical precipitation

Al or Fe salts added to primary or secondary clarifiers

Struvite harvester

Applies to highly concentrated steams. P removal >90%

Coagulant-precipitation

Can be implemented at primary or secondary sedimentation

Oxidation–coagulation– precipitation

Can be implemented at primary or secondary sedimentation

Disinfection

UV, chlorine, PAA, PFA or ozone

Microconstituents refers to a broad class of contaminants that are toxic at very low concentrations, but some agents have the power of hormones. b) Ammonia refers to [NH3] + [NH‡4 ]. At the pH of most wastewaters, [NH3] is insignificant.

9 Water and Wastewater Treatment Operations

9.3 Hydraulic Design of Water and Wastewater Treatment Plants The hydraulic design of water and wastewater treatment plants is a challenging exercise. Plants are designed to flow as much as possible by gravity, minimizing the number of pumps in the plant. Near the front-end of the treatment works, enough head is supplied by pumping if necessary to drive the water through the treatment operations that follow. Efficient hydraulic design provides treatment over a wide range of flows at a minimum energy expenditure. Inefficient hydraulic design results in higher energy consumption and costs for the plant throughout its service time. The designer must be cognizant of possible future modifications and expansions and their hydraulic requirements. To provide the necessary flexibility to handle the flow variation within treatment and physical constraints and other requirements, a number of alternatives should be examined. The plant must be designed to function under all hydraulic conditions from low to high flows. Because two or more units are usually designed for any treatment operation, the extreme highflow condition for a treatment will be when one of the basins is out of service and the remaining basins must accommodate high flow. The hydraulic conditions should always be checked for the low, average, and high flows and also for when one basin is out of service. Backwater conditions are generally avoided; the plant must not flood at any time and surcharged sewers cannot be tolerated for wastewater treatment plants. The critical design flow depends on the treatment unit. Practical considerations are applied. At extreme conditions, guidelines are often violated. For instance, at low flow, velocities in channels may not be high enough to prevent deposition of solids in channels. Designing to achieve a scouring velocity at all flows is impractical because higher energy losses will be incurred and abrasive scour may occur at higher flows. As long as flow velocities that will scour channels or pipes occur for a significant time during a day, lowvelocity conditions can be tolerated. Extreme conditions and suboptimal treatment conditions that accompany them occur with lower frequency. Minimal treatment objectives must be established, then flow and load variation throughout the design period of the plant must be examined to find the plant capacity. Additional units that are idle at times may be required to provide the flexibility to handle flow and load variation at the minimum operating cost within the treatment requirements. Another restriction that applies to hydraulic design of plants is the general necessity to use standard sizes. Devices are supplied in standard sizes to minimize fabrication costs, improve quality control, and facilitate construction. Custom sizes are used infrequently because of the additional expense. Therefore, hydraulic computations are adjusted to the nearest standardized size that provides a more conservative condition. Basic hydraulic formulas commonly used in treatment plants are given in this section. Hydraulic design information on specific operations is also found in the later chapters. In addition, there are sources such as Benefield et al. (1984), Kawamura (2000), Qasim (1999), Sanks (1978), and Schulz and Okun (1984) that provide more detailed information on the hydraulic design of treatment units. Handbooks on hydraulics and manufacturers also provide information on energy losses and hydraulic performance for various devices.

Flow in Pressurized Pipes Flow in pressurized conduits is described by the Bernoulli equation: p v2 p v21 ‡ z 1 ‡ 1 ˆ 2 ‡ z 2 ‡ 2 ‡ hL 2g ρg 2g ρg

(9.1)

225

226

Theory and Practice of Water and Wastewater Treatment

where subscripts 1 and 2 refer to the upstream and downstream locations, respectively, g is the acceleration of gravity, hL is the headloss from location 1 to location 2, p is pressure, v is velocity, z is elevation from a fixed datum, and ρ is the density of the fluid. Headloss in pipe flow is calculated with the Darcy–Weisbach equation: hL ˆ f

L v2 D 2g

(9.2)

where D is pipe diameter, f is a friction factor, and L is length of pipe. The headloss is a function of the velocity head. When irregularly shaped conduits are used, the hydraulic radius [Eq. (9.3)] is substituted for D in Eq. (9.2). Rh ˆ

Af P

(9.3)

where Af is the cross-sectional area of flow, P is the wetted perimeter, and Rh is the hydraulic radius.The friction factor is a function of the Reynolds number. Re ˆ

ρvD μ

or

Re ˆ

4ρvRh μ

(9.4)

where Re is the Reynolds number, μ is the viscosity. The Moody diagram (Figure A.1) describes the variation of f with Re. Flow in Open Channels The energy equation for flow in channels exposed to the atmosphere is usually written with the channel bottom taken as the datum. The total head is the depth of flow plus the velocity head, which is referred to as the specific energy. y1 ‡

v21 v2 ˆ y2 ‡ 2 ‡ h L 2g 2g

(9.5)

where y is the depth of flow. The Darcy–Weisbach equation can be used to find headloss in open channel flow situations, but there are a number of other relations that have been developed. In uniform flow, the water surface is parallel to the channel bottom. Manning’s equation [Eq. (9.6)] is the equation most commonly applied to calculate energy losses in open channel flow. 1 2=3 v ˆ Rh S1=2 (9.6) n where n is a roughness coefficient and S is the slope of the energy line (and the slope of the channel), S = hL/L. Channels are often designed with flat slopes that are easier to construct and install. Varied flows may require a solution of gradually or rapidly varied flow equations, which are covered in some of the references given earlier and in texts on open-channel hydraulics. Manning’s equation is applied discretely in these solutions. Several flow control points will exist in a treatment plant that control the hydraulic profile upstream or downstream from the control point. A known depth (stage)-discharge condition exists at a flow control point. The Froude (Fr) number [Eq. (9.7)] dictates whether gravity flow is sub- or supercritical. At Fr < 1, flow is subcritical and when Fr > 1 flow is supercritical. Subcritical flows and depths are controlled by downstream control points; supercritical flows and depths are controlled by upstream control points. Most flows through treatment plants are subcritical. Any transition between sub- and supercritical flow or vice versa produces a control point at which

9 Water and Wastewater Treatment Operations

critical flow occurs and Fr = 1. The critical flow condition provides the depth–discharge relation. v Fr ˆ p (9.7) gy Weirs, discussed below, with free fall over them are the most common control devices. Valves can also be used to achieve flow control. Other Losses Besides friction losses along the walls of a conduit, any change in direction or velocity causes increased headloss. Obstructions such as valves for flow control or orifices for flow measurement result in headloss. An obstruction may be installed for the sole purpose of causing headloss to equalize flows. Losses from these appurtenances are termed minor losses in fluid mechanics texts, but losses associated with these devices are often the major losses in the short conduits installed in treatment works. The losses to be considered for any unit are the following: 1) 2) 3) 4)

Friction losses through the unit Entrance losses Exit losses Losses through appurtenances.

The losses are usually proportional to the velocity head. Table 9.3 lists the headloss factors for both pressurized and open-channel conduits. The higher values are used for design purposes because of fouling through microbial growth, deposition, or corrosion. These losses are summed for all conduits and basins to find the total head requirement for the plant. Other miscellaneous losses such as the additional depth required to ensure free fall over a weir must also be included. Additional head allowance is also provided for future expansions. Wall friction losses in treatment basins are usually negligible because of the relatively low flow velocities through basins, and the water surface in a basin without a significant number of baffles is assumed to be horizontal. Other flow devices that are regularly used in treatment works are weirs, orifices, and baffles. Weirs are used to control the flow from basins and to measure flow. Maintenance of free flow conditions from a weir will require that the water surface in the receiving channel be lowered. The water surface is typically lowered 5–10 cm below the crest of the weir. Weirs have the disadvantage that solids may settle ahead of the weir, and a higher headloss is associated with maintenance of free fall conditions. The equations for weirs are based on idealized flow over the weir and an empirically determined discharge coefficient that accounts for aeration of the nappe, contraction, and surface tension effects. For a sharp-crested weir with free fall that spans the entire width of a channel (known as a suppressed weir), Q ˆ Cd

2 p L 2g H 3=2 3

(9.8)

where Cd is the discharge coefficient, H is the depth of water above the weir, L is the length of the weir, and Q is the volumetric flow rate. Evett and Liu (1987) give the following equation for Cd that is applicable for SI or US units. C d ˆ 0:611 ‡ 0:075

H Hw

where Hw is the height of the weir.

(9.9)

227

228

Theory and Practice of Water and Wastewater Treatment

Table 9.3 Headlosses in appurtenances. Ka)

Appurtenance (alphabetically)

Butterfly valves

Ka)

Obstruction in pipes (in terms of pipe velocities) pipe to obstruction area ratio

Fully open Angle closed

Appurtenance (alphabetically)

θ = 10°

0.3

1.1

0.21

0.46

1.4

1.15

θ = 20°

1.38

1.6

2.40

θ = 30°

3.6

2.0

5.55

θ = 40°

10

3.0

15.0

θ = 50°

31

4.0

27.3

θ = 60°

94

Check (reflux) valves

5.0

42.0

6.0

57.0

Ball type (fully open)

2.5–3.5

7.0

72.5

Horizontal lift type

8–12

10.0

121.0

Swing check

0.6–2.3

Swing check (fully open)

2.5

Contraction sudden

Orifice meters (in terms of pipe velocities) orifice to pipe diameter ratio

0.25 (1 : 4)

4.8

0.33 (1 : 3)

2.5

4 : 1 (in terms of velocity 0.42 at small end)

0.50 (1 : 2)

1.0

2:1

0.33

0.67 (2 : 3)

0.4

4:3

0.19

0.75 (3 : 4)

0.24

Also see Reducers

Outlet losses

Diaphragm valve

Bell-mouthed outlet

0.1 v21 =2g

v22 =2g

Sharp-cornered outlet

1.0 v21 =2g

v22 =2g

Fully open

2.3

¾ open

2.6

½ open

4.3

air (free discharge)

¼ open

21.0

Plug globe or stop valve

Elbow 90° Flanged-regular

0.21–0.30

Pipe into still water or

1.0

Fully open

4.0

¾ open

4.6

Flanged-long radius

0.18–0.20

½ open

6.4

Intersection of two cylinders (welded pipe–not rounded)

1.25–1.8

¼ open

780

Screwed-short radius

0.9

Reducers

Screwed-medium radius 0.75

Ordinary (in terms of velocity at small end)

0.25

Screwed-long radius

Bell mouthed

0.10

Standard

0.04

Bushing or coupling

0.05–2.0

0.60

Elbow 45° Flanged-regular

0.20–0.30

9 Water and Wastewater Treatment Operations

Table 9.3 (Continued ) Appurtenance (alphabetically)

Flanged-long radius

Ka)

Appurtenance (alphabetically)

0.18–0.20

Return bend (2 Nos. 90°)

Ka)

Screwed regular

0.30–0.42

Flanged–regular

0.38

Elbow 22.5° (flanged)

0.10–0.12

Flanged–long radius

0.25

Screwed ends

2.2

Enlargement sudden 1 : 4 (in terms of velocity 0.92 at small end)

Siphon

1:2

0.56

Sluice gates

3:4

0.19

Also see increasers Entrance losses Bell mouthed

0.04

Pipe flush with tank

0.5

2.8

Contraction in conduit

0.5

Same as conduit width without top submergence

0.2

Submerged port in 30 cm wall

0.8

Tees Standard–bifurcating

1.5–1.8

Pipe projecting into tank 0.83–1.0 (Borda entrance)

Standard 90° turn

1.80

Slightly rounded

0.23

Standard–run of tee

0.60

Strainer and foot valve

2.50

Reducing–run of tee

Gate valves

2 : 1 (based on velocity at smaller end) 0.90

Fully open

0.19

¾ closed

1.15

½ closed

5.6

Headloss occurs mostly in and downstream of throat, but losses are given in terms of velocity at inlet end to assist in design

¼ closed

24.0

Long tube type throat-to-inlet diameter ratio

Also see sluice gates Increasers v21 =2g

v22 =2g

, 0.25 where v1 is velocity at small end Miter bends Deflection angle, θ

4:1

0.75

Venturi meters

0.33 (1 : 3)

1.0–1.2

0.50 (1 : 2)

0.44–0.52

0.67 (2 : 3)

0.25–0.30

0.75 (3 : 4)

0.20–0.23

Long tube type throat-to-inlet diameter ratio

5°

0.016–0.024

0.33 (1 : 3)

2.43

10°

0.034–0.044

0.50 (1 : 2)

0.72

15°

0.042–0.062

0.67 (2 : 3)

0.32

22.5°

0.066–0.154

0.75 (3 : 4)

0.24

30°

0.130–0.165

45°

0.236–0.320

60°

0.471–0.684

90°

1.129–1.265

Y branches (regular)

a) The headloss is given by hL = Kv2/2g unless otherwise indicated. Source: Adapted from Amirtharajah (1978) and Kawamura (2000).

0.50–0.75

229

230

Theory and Practice of Water and Wastewater Treatment

A contracted weir is a weir that does not span the entire length of the channel. It is also referred to as a notched weir. For a rectangular, sharp-crested notched weir (Evett and Liu 1987): p Q ˆ …1:69†L1:02 2g H 1:47 (9.10) where Q is in m3 s 1 for L and H in m. V-notched weirs are also used, particularly for sedimentation basins. They are discussed in Section 11.6. The above equations do not apply to submerged weirs. Slight improvement in the accuracy of headloss calculations for weirs can be gained by using the formulations in American Society of Mechanical Engineers (ASME 1971), but Eqs. (9.8)– (9.10) are simpler and suitable for water or wastewater treatment plant problems. Application of Bernoulli’s equation results in the equation for a freely discharging orifice. p (9.11) Q ˆ C d a 2gh where a is the area of the orifice and h is the piezometric head above the center of the orifice. The discharge coefficient varies with the hydraulic smoothness of the orifice. For a sharpedged orifice, Cd = 0.62. When the water level drops below the top of the orifice, the orifice behaves as a weir. When the orifice is submerged, the discharge through it is governed by the difference in piezometric head across the orifice. p (9.12) Q ˆ C d a 2gΔh where Δh = h1 h2. The discharge coefficient for sharp-edged submerged orifices is also 0.62. Baffles are often used in basins to dissipate energy of incoming flows and disperse the jet. For a rectangular baffle (reaction baffle), the drag force from the baffle is F D ˆ ½ρC D Ab v2 where Ab is the area of the baffle, CD is the drag coefficient, FD is the drag force, and v is the approach velocity to the baffle. Converting the above equation to energy loss per unit weight of water, hL ˆ C D

v2 Ab 2g A

(9.13)

where A is the cross-sectional area of the channel. Rouse (1959) gives the value of CD as 1.2.

Questions and Problems 9.1 What unit operations are incorporated into the local water and wastewater treatment plants? 9.2 Describe the changes in water quality after each unit operation in (a) Figure 9.1 and (b) Figure 9.2. 9.3 Which unit operations would be most likely required for treating a groundwater as opposed to a surface water?

9 Water and Wastewater Treatment Operations

9.4 What is the risk associated with point-of-use water treatment devices? 9.5 Would you expect water produced from a water treatment plant to contain more dissolved inorganic or more dissolved organic components? 9.6 Describe the general purpose of each unit operation in the treatment schemes in (a) Figure 9.4 and (b) Figure 9.6. 9.7 List all water and wastewater treatment operations that would produce or remove SS. Describe the removal or solids transformation in each process. 9.8

a Describe the purposes of (i) equalization basins, (ii) grit chambers, and (iii) stabilization ponds in wastewater treatment. b Describe the purposes of (i) flocculation, (ii) softening, and (iii) disinfection in water treatment.

9.9 You have been assigned the task of making a preliminary choice on unit operations to be incorporated into a water treatment plant. The data for the raw water are given below. Draw the flow diagram of the units that you will incorporate into your design. In brief statements, give reasons for choosing these units. Use the WHO standards in Chapter 8 as the basis for determining whether a substance requires treatment. River data

pH

7.7

Alkalinity

180 mg L

1

as CaCO3

Total hardness

300 mg L

1

as CaCO3

Calcium hardness

250 mg L

1

as CaCO3

Iron

0.3 mg L

1

Manganese

0.3 mg L

1

Turbidity Water temperature

Coliform count

300 NTU Mean

18 °C

Low

10 °C

High

25 °C 3000 MPN/100 mL

Assume other substances are within acceptable ranges. 9.10 Would you consider a biological or a physical–chemical treatment process better for treating the wastewater produced from a potato processing plant? Why? 9.11 What are the losses in the following system at a temperature of 15 °C? A 6.0 cm diameter pipe with entrance flush with a tank wall and below the surface of water in the tank conveys water horizontally 2.0 m, then turns 90° in a flanged regular elbow and travels vertically down for 0.5 m. Then the pipe turns horizontal through another 90° flanged regular elbow. The horizontal length is 1.5 m, and the pipe enters another large basin where the water velocity may be neglected. The pipe entrance to the tank is flush with the

231

232

Theory and Practice of Water and Wastewater Treatment

tank wall. There is a fully open gate valve in the line and the flow velocity is 1.2 m s 1. Assume that the roughness height of the pipe is 0.000 08 m. 9.12 What is the headloss and energy loss for a reaction baffle in a rectangular sedimentation basin? The baffle area is 25% of the cross-sectional area of the basin. The flow entering the basin is 4500 m3 d 1 and the average horizontal flow velocity is 20 m h 1.

References Amirtharajah, A. (1978). Design of granular-media filter units. In: Water Treatment Plant Design (ed. R.L. Sanks), 675–737. Ann Arbor, MI: Ann Arbor Science. American Society of Mechanical Engineers (1971). Fluid Meters: Their Theory and Application. New York: ASME. Benefield, L.D., Judkins, J.F. Jr., and Parr, A.D. (1984). Treatment Plant Hydraulics for Environmental Engineers. Englewood Cliffs, NJ: Prentice-Hall. Elder, D. and Budd, G.C. (2012). Overview of Water Treatment Processes. In: Water Quality & Treatment: A Handbook on Drinking Water, 6 (ed. American Water Works Association and J. Edzwald). New York: McGraw-Hill. Evett, J.B. and Liu, C. (1987). Fundamentals of Fluid Mechanics. Toronto: McGraw-Hill. Geldreich, E.E., Taylor, R.H., Blannon, J.C., and Reasoner, D.J. (1985). Bacterial Colonization of Point-of-Use Water Treatment Devices. J. Am. Water Works Assn. 77 (2): 72–80. Hamann, C.L., McEwen, J.B., and Myers, A.G. (1990). Guide to selection of water treatment processes. In: Water Quality and Treatment, 4 (ed. F.W. Pontius and American Water Works Association), 157–187. Toronto: McGraw-Hill. Himberg, K., Keijola, A.M., Hiisvirta, L. et al. (1989). The effect of water treatment processes on the removal of hepatotoxins from Microcystis and Oscillatoria cyanobacteria: a laboratory study. Water Res. 23 (8): 979–984. doi: 10.1016/0043-1354(89)90171-1. Kawamura, S. (2000). Integrated Design of Water Treatment Facilities, 2e. New York: WileyInterscience. McNeill, L.S. and Edwards, M. (2001). Iron pipe corrosion in distribution systems. J. Am. Water Works Assn. 93 (7): 88–100. http://www.jstor.org/stable/41297605. Qasim, S.R. (1999). Wastewater Treatment Plants: Planning, Design and Operation, 2e. Lancaster: Technomic Publishing Co. Inc. Regunathan, P., Beauman, W.H., and Kreusch, E.G. (1983). Efficiency of point-of-use treatment devices. J. Am. Water Works Assn. 75 (1): 42–50. jstor.org/stable/41272875. Rouse, H. (ed.) (1959). Engineering Hydraulics. Toronto: John Wiley & Sons. Sanks, R.L. (ed.) (1978). Water Treatment Plant Design. Ann Arbor, MI: Ann Arbor Science. Schulz, C.R. and Okun, D.A. (1984). Surface Water Treatment for Communities in Developing Countries. Toronto: Wiley. USEPA (2017a). Title 40 Part 142-National primary drinking water regulations implementation. http://www.ecfr.gov/cgi-bin/text-idx?SID=dd85f34ddcfc3f57c608bad71a4cca50&mc=true&node =pt40.25.142&rgn=div5 (accessed January 2017). USEPA (2017b). Title 40 Part 141-National primary drinking water regulations implementation. http://www.ecfr.gov/cgi-bin/retrieveECFR?gp=1&SID=9e6bacad9eb5c6e10d713713355ae618 &ty=HTML&h=L&mc=true&n=pt40.25.141&r=PART#se40.25.141_170 (accessed February 2017).

9 Water and Wastewater Treatment Operations

USEPA (2006). Ultraviolet Disinfection Guidance Manual for the Final Long Term 2 Enhanced Surface Water Treatment Rule. EPA. WEF and ASCE (2010). Design of Wastewater Treatment Plants, 5e, vol. 1. New York: WEF Press, Alexandra and McGraw-Hill. WHO (2011). Guidelines for Drinking-Water Quality, 4e. Geneva: World Health Organization.

233

235

10 Mass Balances and Hydraulic Flow Regimes The concepts in this chapter apply to any water or wastewater unit operation. Mass balances relate influent flow rate and concentrations to effluent flow rate and concentrations by accounting for removal or transformation phenomena. The pattern of flow in a reactor is influenced by reactor geometry and mixing devices, which in turn influence the kinetics of a unit operation.

10.1 Setup of Mass Balances There are a number of different notations used in textbooks and the literature for describing mass balances. Each method, of course, ultimately leads to the same differential equation. A few different approaches are discussed here along with other considerations in formulating the correct equation. In particular, the assumptions that underlie the development of an equation are important. Assumptions are normally made to simplify a complex situation; the assumptions must be examined to determine whether they are reasonable for the circumstances under consideration. Mass balances are applied to rivers, lakes, or treatment basins, where the problem is to find the concentration of a substance at a location or its rate of change in a section. Varying degrees of complexity are reached as more parameters are introduced and when the parameters vary with time (i.e., when nonsteady-state conditions exist). Often it is not necessary to consider variation at all times. Only the worst-case or perhaps the worst- and best-case scenarios need to be considered to arrive at the design constraints (costs and material requirements). The steady-state condition in the majority of circumstances will render an intractable equation analytically solvable, greatly reducing the costs (normally associated with computer runs) and time involved in analyzing the problem. If a reactor is completely mixed (CM) (sections below discuss CM and plug flow (PF) hydraulic regimes in detail), i.e., uniform conditions exist throughout the reactor, the whole volume is most conveniently considered. If PF conditions exist (water flows through the reactor as a series of plugs), an elemental volume must be considered, because each elemental volume is different with respect to the mass and concentration of reacting (or physically changing) substances contained in it. In the latter case, the average contribution of the whole reactor volume to change in a substance’s concentration is obtained by integrating the elemental mass balance over the whole volume. A substance is transported into a given volume of a reactor (which may be a vessel or a section of a river or lake) by bulk flow (convection or advection) or other hydraulic phenomena such as dispersion. Bulk flow is merely the displacement of liquid in a volume with incoming fluid. Theory and Practice of Water and Wastewater Treatment, Second Edition. Ronald L. Droste and Ronald L. Gehr.

 2019 John Wiley & Sons Inc. Published 2019 by John Wiley & Sons Inc.

Companion website: www.wiley.com/go/droste/water

236

Theory and Practice of Water and Wastewater Treatment

Figure 10.1 Qualitative process description.

Dispersion is a function of turbulence where random fluctuations in the movement of fluid transport dissolved and suspended matter in random directions. Diffusion, which is transport through the random motion of molecules, may also be important. Transport out of the volume is normally due to the same mechanisms as transport into the volume. A case in which in and out transport mechanisms differ would occur if the substance is transformed into a gas in the reactor, for instance, partial oxidation of large molecules into smaller volatile compounds. The general situation is qualitatively described by Figure 10.1. Reaction of the substance in question may be production or destruction caused by chemical, biochemical, or physical (such as temperature) phenomena. During the time in which the substance resides in the volume, it is exposed to the reaction mechanism, producing an alteration in its concentration. The general word statement describing the situation in Figure 10.1 is In

Out ‡ Generation ˆ Accumulation

(10.1)

In Eq. (10.1), In Out refers to the net transport of a substance into the reactor, Generation refers to the net amount of production or destruction by reaction or physical processes, and Accumulation is the amount left over. Physical removal such as sedimentation is a transport process, but it is usually viewed as a generation term. Mention should be made here of time and its meaning in steady state, which is commonly assumed to exist in the first analysis, and time of reaction, which refers to rate of change of a substance with time. The assumption of steady state means that all quantities are constant in value at each point in the system. This does not imply that they are equal throughout the system, only that a value at a given point does not change with time. On the other hand, two points in a PF system are separated by a distance of finite travel time, and the substance will decay (or be produced) in the period of time it takes to travel from one point to another. Once Eq. (10.1) is written, the important processes must be identified and mathematically formulated. In environmental engineering, the rate of reaction, r, is often formulated as shown below. Hyperbolic expressions such as the Monod or Michaelis–Menten [Eq. (4.10)] formula­ tions are also commonly used. r ˆ kC n

(10.2)

where C is concentration (usually mg L 1), k is a rate coefficient, and n is a constant (not necessarily with an integer value). The units on r are always mass/volume-time regardless of the value of n. The units on k are adjusted to make the equation dimensionally consistent. Many reactions in environmental engineering are reasonably described with a first-order model, i.e., n is equal to 1. A positive sign on the right-hand side corresponds to production; a negative sign refers to destruction. 10.1.1 Mixing Characteristics of Basins The degree of mixing that exists in a basin affects reaction kinetics in a fundamental manner. The addition of a reagent to water often requires intense mixing to achieve contact between the

10 Mass Balances and Hydraulic Flow Regimes

Figure 10.2 Basin characteristics for PF and CM basins.

reagent and the target substance in the water before the reagent is dissipated. On the other hand, some processes such as sedimentation rely on the existence of quiescent conditions for maximum efficiency. The two extremes in hydraulic mixing regime are CM (sometimes referred to as a completely stirred tank reactor, CSTR) and PF. An influent particle of water to a CM basin is instantaneously dispersed throughout the entire contents of the basin. Uniform conditions exist throughout the basin; i.e., there is no spatial variation in any component. The concentrations of components in the effluent are the same as their concentrations in the basin. In an ideal basin, it would not make any difference where the inlet and outlet points were located; practically locating them next to each other provides enhanced opportunity for short-circuiting, as discussed below. Impellers (Figure 10.2), pumping, and gas or liquid recirculation are means of increasing the degree of mixing. The opposite end of the spectrum is PF. In a PF hydraulic regime, the water flows through the reactor as a series of plugs, and there is no transfer of contents between plugs. Vertical and lateral (into the paper) mixing are consistent with this definition, but longitudinal (left to right) mixing is not. Inlet and outlet locations should be at opposite ends of a PF basin. PF conditions are achieved by designing long narrow reactors or by placing baffles in a reactor (Figure 10.2). Baffles provide a physical barrier to mixing. 10.1.2 Mass Balances for PF Reactors In a PF situation, the mass balance must be taken over an elemental volume. The typical situation is shown in Figure 10.3, where Q is the volumetric flow rate, A is the cross-sectional area, and Δx is the length of the element. For the basic model, area and volumetric flow rate will be assumed to be constant, and change in the y and z directions is assumed to be zero or insignificant, i.e., one-dimensional change exists. The following two relations are used in all derivations: ΔV ˆ AΔx

(10.3a)

Q A

(10.3b)

vˆ

where A is cross-sectional area of the element, Q is volumetric flow rate, V is volume, and v is velocity.

Figure 10.3 Elemental volume for PF.

237

238

Theory and Practice of Water and Wastewater Treatment

Method I

Assuming a steady volumetric flow rate, the following mass balance applies to the element over a time period, Δt = (t2 t1): C 1 …t 1 † ‡ C 1 …t 2 † Δt 2

Q

Q

C 2 …t 1 † ‡ C 2 …t 2 † C 1 …t 2 † ‡ C 2 …t 2 † Δt ‡ rΔV Δt ˆ 2 2

C 1 …t 1 † ‡ C 2 …t 1 † ΔV 2 (10.4a)

In the above equation, Ci refers to the concentration at interface i. If a decay reaction of order n is assumed, the expression for r is rˆ k

C 1 …t 1 † ‡ C 1 …t 2 † ‡ C 2 …t 1 † ‡ C 2 …t 2 † 4

n

Dividing by Δt and letting Δt → 0 changes terms in the equation in the following manner. C i …t 1 † ‡ C i …t 2 † → Ci 2 C 1 …t 2 † ‡ C 2 …t 2 † C 1 …t 1 † ‡ C 2 …t 1 † @C 2 2 ΔV → ΔV Δt @t C1 ‡ C2 2

Making these substitutions into the reaction expression and Eq. (10.4a),

where C ˆ

QC 1

QC 2 ‡ rΔV ˆ

QC 1

QC 2

@C ΔV @t

or n

kC ΔV ˆ

@C ΔV @t

(10.4b)

The left-hand side of the equation describes the spatial variation and the right-hand side describes the time variation. Equation (10.4b) is commonly written as the immediate starting equation for the mass balance. Assuming steady state and a first-order decay (r = kC), substituting for these quantities and ΔV, and rearranging Eq. (10.4b): Q…C 1

C2†

kCAΔx ˆ 0

(10.4c)

Now, C1

C 2 ˆ ΔC

Substituting the above into Eq. (10.4c) and dividing by Δx, Q

ΔC ˆ kCA Δx

(10.4d)

Letting Δx → 0 and dividing by A, Q @C @C ˆ kC or v ˆ kC A @x @x

(10.4e)

In the above two equations, @C/@x can be changed to dC/dx since C is only a function of x.

10 Mass Balances and Hydraulic Flow Regimes

Because v = dx/dt, by application of the chain rule Eq. (10.4e) can be changed to dx dC dC ˆ ˆ kC dt dx dt

(10.4f)

In the above equation time is synonymous with distance. Steady state conditions exist. In the literature, there are other equivalent methods to arrive at the governing differential equation. A couple of the methods will be commented on here. All procedures when correctly applied must arrive at the same equation. The choice of the approach is arbitrary; however, sometimes one is more convenient than the others. Method II

In this method, the concentrations at interfaces 1 and 2 are formulated as C and C ‡

@C

@x

Δx,

respectively. Placing these in the mass balance: Q C‡

QC

@C @C Δx ‡r ΔV ˆ ΔV @x @t

(10.5a)

(why is C, not C, used in the partial derivative term in the expression for concentration at interface 2?) Assuming steady state, substituting for r and ΔV, expanding the equation, and simplifying: Q

@C Δx @x

kCA Δx ˆ 0

(10.5b)

Following the procedures outlined above, Eqs. (10.4e) and (10.4f) would be derived. Method III

In the final method, the notation to describe the value of a substance z at x + Δx is zjx‡Δx Formulating the mass balance, QC jx

QC jx‡Δx ‡ r ΔV ˆ

@C ΔV @t

(10.6a)

Applying the steady state condition and substituting for ΔV, QC jx

QC jx‡Δx ‡ rA Δx ˆ 0

(10.6b)

Substituting for r, noting that Q is constant, and dividing by Δx: Q

C jx

C jx‡Δx ˆ kCA Δx

(10.6c)

As Δx → dx, the expression within the brackets becomes ( @C/@x). In this case, C is only a function of x, and it becomes ( dC/dx) as before. After this step, the procedure is the same as outlined from Eqs. (10.4d) to (10.4f). Note that if Q were variable over the distance, the left-hand side of Eq. (10.6c) would become: QC jx

QC jx‡Δx Δx

Δx → dx

→

@…QC† @Q ˆ C @x @x

Q

@C @x

Now consider variation in area, which would be more realistic for a river. The definition sketch in Figure 10.3 applies except that the area of interface 1 is different from the area of interface 2. If

239

240

Theory and Practice of Water and Wastewater Treatment

Q is constant and input of the substance to the system is constant, do steady-state conditions exist? Using the approach in Method II, the formulation of the mass balance is the same as Eq. (10.5a). @C @C ΔV (10.7) QC Q C ‡ Δx ‡ r ΔV ˆ @x @t Because A1 is different from A2 the following expression must be substituted for the elemental volume. A1 ‡ A2 Δx ΔV ˆ AΔx ˆ 2 The assumption of steady state is valid for the conditions just described; velocity varies from point to point, but it (along with the other parameters) is constant at a given point. Applying this condition and substituting for ΔV and r: @C Δx kCA Δx ˆ 0 (10.8) Q @x Letting Δx → dx, CA becomes CA, the concentration in the plane multiplied by the area of the plane. Q

@C ˆ kCA @x

(10.9)

Dividing by A and noting that area is a function of x, A(x): 1 dC Q ˆ kC A…x† dx

(10.10)

This equation can be integrated as follows: C

dC ˆ ∫ C0 C

k x A…x†dx Q ∫0

(10.11)

But, Q dx ˆ v …x † ˆ A…x† dt This equation can be solved for dt, dt ˆ

A…x† dx dx ˆ Q v…x†

Substituting this relation into Eq. (10.11), C

t dC ˆ k dt ∫0 ∫ C0 C

(10.12)

In this case time and distance are not linearly related. For an example, examine a channel that has a constantly increasing area, A(x), given by A…x† ˆ A0 …1 ‡ ix† where A0 and i are constants. Using Eq. (10.11), C

dC ˆ ∫ C0 C

k x A0 …1 ‡ ix†dx Q ∫0

(10.13)

10 Mass Balances and Hydraulic Flow Regimes

which integrates to: ln

ix2 kA0 x‡ Q 2

C ˆ C0

(10.14)

Also, from Eq. (10.12), ln

C ˆ kt C0

(10.15)

and v… x † ˆ

Q Q dx ˆ ˆ A…x† A0 …1 ‡ ix† dt

To find the x t

∫0

dt ˆ

tˆ

(10.16)

t relation,

A0 x …1 ‡ ix†dx Q ∫0

(10.17)

A0 x2 x‡i Q 2

(10.18)

Substituting Eq. (10.18) into Eq. (10.14), ln

C ˆ kt C0

(10.19)

which is the same equation [Eq. (10.15)] derived before. Example 10.1 Plug Flow Reactor Volume Determine the volume of a PF reactor required to give a treatment efficiency of 95% for a substance that decays according to half-order kinetics with a rate constant of 0.05 (mg L 1)½-h. The flow rate is steady at 300 L h 1, and the influent concentration is 150 mg L 1. The rate expression is r ˆ kC ½ The required effluent concentration is C ˆ 0:05 150 mg L

1

ˆ 7:5 mg L

1

Substituting the relation for r into Eq. (10.5a), the mass balance for steady state conditions over an elemental volume becomes QC

Q C‡

@C Δx @x

kC

1

=2 ΔV ˆ0

This reduces to the integral C

dC ˆ ∫ C 0 C 1=2 V ˆ2

k V Q 1=2 dV → V ˆ 2 C0 Q ∫0 k 300 L h

0:05 mg L

1

1 1=2

h

1

150 mg L

C 1=2 1 1=2

7:5 mg L

1 1=2

ˆ 114  103 L ˆ 114 m3

241

242

Theory and Practice of Water and Wastewater Treatment

Figure 10.4 CM reactor.

10.1.3 Mass Balances and Reaction for CM Basins An elemental volume approach is fundamental and can be formulated for a reactor with any characteristics. However, recognizing the uniformity in all elemental volumes in a CM reactor, mass balances for these reactors are more easily formulated. The only difference in making a mass balance for a CM reactor (Figure 10.4) compared to a PF reactor is to make the mass balance over the entire reactor volume (replace ΔV with V in the equations). The concentration of a substance in the reactor is the same as its concentration in the effluent. Setting up a mass balance:

In−Out ‡ Generation ˆ Accumulation

QC 0

QC ‡ rV ˆ V

dC dt

(10.20)

Assuming steady-state conditions and substituting a first-order decay reaction for r,

QC 0

QC

kCV ˆ 0

Solving for the effluent concentration, C0 V Cˆ θd ˆ 1 ‡ kθd Q

(10.21)

where θd is the liquid detention time (hydraulic retention time, HRT) in the vessel. The dilution rate (D) is the inverse of the detention time. D ˆ 1=θd ˆ Q=V

(10.22)

A series of CM basins is used in some situations as shown in Figure 10.5. In this case, the effluent from one reactor becomes the influent for the next reactor. Using the symbols in Figure 10.5, the equations describing concentration of S for the first two reactors at steady-state conditions are the following: QC 0

QC 1

k1C1V 1 ˆ 0

(10.23a)

QC 1

QC 2

k2C2V 2 ˆ 0

(10.23b)

where ki is the rate coefficient for the ith reactor. Similar equations can be written for the other reactors. For the ith reactor: Ci ˆ

Ci 1 1 ‡ k i θdi

(10.24)

Figure 10.5 CM reactors in series.

10 Mass Balances and Hydraulic Flow Regimes

The influent (C0) and effluent (Ce) concentrations for an n reactor system are related by Ce Cn C1 C2 Cn 1 1 1 ˆ ˆ   ∙∙∙  ˆ   ∙∙∙  C n 1 1 ‡ k 1 θd1 1 ‡ k 2 θd2 1 ‡ k n θdn C0 C0 C0 C1

(10.25)

If the reactors are identical (Vi/Q = θd) and the rate coefficient does not change throughout the system (ki = k), Ce ˆ

C0 ˆ …1 ‡ kθd †n

C0 1‡k

V nQ

(10.26)

n

where V is the total volume in the system. As the number of reactors in the system increases, the system response approaches the response obtained from a PF system. The total reactor volume required to obtain a given degree of treatment will dramatically decrease for a system containing a series of CM reactors when the reaction kinetic expression is of order greater than zero. Example 10.2 Complete Mixed Reactor Volume a) Determine the volume of a CM reactor required to give a treatment efficiency of 95% for a substance that decays according to half-order kinetics with a rate constant of 0.05 (mg L 1)½-h. The flow rate is steady at 300 L h 1, and the influent concentration is 150 mg L 1. b) Determine the volumes of two identical CM reactors in series to provide the same degree of treatment for the conditions given in (a). Compare the results with those of Example 10.1. For part (a): The required effluent concentration is 7.5 mg L rate expression is

1

as given in Example 10.1 and the

r ˆ kC ½ Substituting the relation for r into Eq. (10.20), the mass balance for steady-state conditions over the volume is QC 0

QC

kC ½ V ˆ 0

Solving this for V and substituting the given information, V ˆ

QC 0

QC ˆ ½ kC

300 L h 0:05 mg L

1

150

1 1=2

h

7:5 mg L 1

1

7:5 mg L

1

½

ˆ 3:12  105 L ˆ 312 m3

For part (b), using the approach outlined above, the mass balance expressions are QC 0

QC 1

kC ½ 1 V 1 ˆ 0

QC 1

QC 2

kC ½ 2 V 2 ˆ 0

V1 ˆ V2 ˆ V Solving the second mass balance for C1, C 1 ˆ C 2 ‡ kC ½ 2

V Q

243

244

Theory and Practice of Water and Wastewater Treatment

and substituting this into the first mass balance: QC 0

1=2

QC 2

300 L h

1

1=2

k C 2 ‡ kC 2

kC 2 V 150

0:05 mg L

7:5 mg L

1 1=2

h

1

1

V Q

½

V ˆ0

0:05 mg L

7:5mg L

1

‡

1 ½

h

1

0:05 mg L

7:5 mg L 1 1=2

h

1

300 L h

1 ½

V

7:5mg L 1

1 1=2

V

1=2

V ˆ0

Reducing the above equation and eliminating the units, 42 750−0:137 V

0:05…7:5 ‡ 0:000 456 V †½ V ˆ 0

where V is in L. This equation can be solved by a nonlinear equation numerical method or by trial and error. V for a single reactor is found to be V ˆ 88:5  103 L ˆ 88:5 m3 The total volume in this system is 2(88.5 m3) = 177 m3.

A comparison table for the results is

Volume (m3) Comparison for the three systems PF

Single CM reactor

Two CM reactors in series

114

312

177

Examples 10.1 and 10.2 illustrate the kinetic behavior of CM compared to intermediate mixed and PF systems. When the same reaction model (except for zero-order reactions) applies regardless of the mixing regime, a PF system is always the most efficient. As the total reactor volume in a CM system is divided into more compartments, less total volume is required. Fundamentally, when a reaction process is in any way dependent on the concentration of a reactant, there will be a concentration gradient in a PF reactor, but a complete or partially mixed reactor will dilute higher concentration elemental volumes with lower concentration elemental volumes. Thus, ±kC (or similar expressions) will be lower in portions of the complete or partially mixed reactor compared to a PF reactor.

10.1.4 Batch Processes In a batch process, a reactor is filled and any required reagents are added to the reactor. If the reagents are added along with the influent, the reaction begins as the reactor is filling. Once the reactor is full, addition of influent is terminated. After this time, the reactor contents are held for a specified time and then the reactor is drained. The process is repeated on a cyclic basis. The cycle time for a process with only these three phases is tC ˆ tf ‡ tr ‡ td

(10.27)

10 Mass Balances and Hydraulic Flow Regimes

where tC is total cycle time, tf is filling time, and td is drain time. The fill and draw times are defined by tf ˆ

Vf

td ˆ

Vf

Vi Qf Vi Qd

(10.28a) (10.28b)

where Vf is the volume after the reactor is filled, Vi is the volume at the beginning of the fill period (or at the end of the draw period), Qf is the volumetric flow rate during the fill period, and Qd is the volumetric flow rate to drain the reactor. To handle a continuous flow of water, more than one batch reactor is needed, and the cycle phases are staggered throughout the series of reactors. Regardless of the degree of mixing when the reactor is full, all of the reactor contents are subject to the reaction conditions for the holding time until the reactor is drained or a reaction condition such as mixing is terminated. The reaction period can be divided into various phases where, for instance, mixing is applied and then suspended, or chemicals are added, at various times. A batch process is similar to a PF reactor in that each volume of water resides in the reactor for the reaction time. Depending on the reactor operation, there may be reaction occurring during either or both of the fill and drain periods. Even if reaction proceeds during both the fill­ and-drain periods, a batch process will not be as efficient as a PF process because the volume of the reactor is, on average, only half-full during the fill and drain times.

10.2 Flow Analysis of CM and PF Reactors The actual flow characteristics through a basin rarely achieve the desired ideal regime, but basins are designed to approach one regime or the other depending on the process. A basin with an intermediate degree of mixing has dispersed flow (non-ideal flow). As illustrated in Exam­ ples 10.1 and 10.2, the degree of mixing exerts a fundamental influence on the volume or time required for a given amount of reaction. Tracers (dyes, electrolytes, or radioactive isotopes) are used to characterize the degree of mixing. The tracer must be conservative. A conservative tracer is soluble and does not participate in any reaction, precipitate, volatilize, and it is not adsorbed or absorbed by the reactor or its contents. The tracer molecules are assumed to be moved about in the same manner as the water molecules, and therefore, their flow pattern will mimic the liquid flow pattern. One could consider that the tracer coats the water molecules to which it is added. If a conservative tracer is added at the inlet of a basin as one short impulse, the tracer will appear at the outlet over a period of time following a curve that reflects the degree of mixing as shown in Figure 10.6. The tracer curves for CM and PF flows are derived in the following sections. 10.2.1 Tracer Analysis of Complete Mixed Reactors For a CM reactor, Eq. (10.20) applies at all times. If the equation is applied at the instant after the slug (impulse) input of tracer has entered the basin, the influent concentration, CI is unchanging with a value of 0. The initial concentration of tracer in the basin can be determined from the known mass, M of tracer input into the basin of volume, V. M C0 ˆ V

245

246

Theory and Practice of Water and Wastewater Treatment

Figure 10.6 Effluent tracer concentrations for varying degrees of mixing.

Therefore, taking the initial time at the instant immediately after the tracer has entered the basin, Eq. (10.20) becomes QC ˆ V

dC dt

(10.29)

which is integrated between the following limits to determine the equation for the reactor and exit tracer concentration: C

Q t dt V ∫0

dC ˆ ∫ C0 C

(10.30)

The resulting equation is C …t † ˆ C 0 e

t=θd

(10.31)

Example 10.3 provides the analysis of a step feed of tracer. Example 10.3 Tracer Exit Curve What is the exit tracer concentration curve for a CM reactor of volume V that receives a steady flow containing a concentration of tracer, CI, beginning at time 0? The initial concentrations of tracer in the reactor and influent were 0.

The mass balance expression is QC ˆ V

QC I

dC dt

Rearranging the equation, noting Q/V = 1/θd and integrating, C

dC

∫ 0 CI

C

ˆ

Q t CI C dt ) ln V ∫0 CI

ˆ

t θd

10 Mass Balances and Hydraulic Flow Regimes

C ˆ CI 1

e

t=θd

The tracer exit curve is shown in the figure in this example. Ultimately, the effluent tracer concentration equals the influent concentration because there is no transformation of the tracer.

As shown in the example, the exit curve in nondimensional form is C ˆ 1 CI

e

t=θd

(10.32)

The residence time distribution curve for a step feed is known as an F-distribution. It is an alternative approach to assessing the mixing behavior of a basin. 10.2.2 Tracer Analysis of Plug Flow The definition sketch used for the CM reactor in Figure 10.4 will apply to the fundamental analysis of PF except that there is no mixer in a PF basin. A steady-flow situation exists. Because there is no mixing allowed between adjacent planes in a PF reactor, a slug of conservative tracer will appear in the effluent at the time of exit of the plane containing the slug of tracer. If the time of tracer injection is t = 0, then the tracer concentration is zero everywhere except at the plane x = vt; that is, it is a plane moving with velocity, v, through the vessel, which is ideal PF. If the flow rate varies with time, the location of the plane containing the tracer is at a distance x ˆ vt (v is the average velocity from time 0 to time t) from the entrance to the reactor. There is no kinetic advantage gained by providing a series of PF reactors as for CM reactors because of the nature of flow in a PF reactor. However, PF reactors in series can allow the staged introduction of chemicals or changing the physical conditions in each reactor to optimize treatment. 10.2.3 Complete Mixed Reactors in Series A series of CM reactors approaches the flow regime of a PF reactor. A tracer analysis will demonstrate that the tracer output curve from the final reactor approaches the output curve for a PF reactor of the same volume. Consider a CM reactor with volume, V, into which a conservative tracer of mass, M, is injected at time, t = 0. As shown above, the tracer output curve is Cˆ

M e V

Qt=V

or

CV ˆe M

t=θd

Now divide the reactor into two CM reactors in series, each with a volume of V/2. The same amount of tracer is injected in a slug into the first reactor. The resulting mass balance on tracer for the second reactor is Q

2M e V

2Qt=V

QC ˆ

V dC 2 dt

(10.33)

Rearranging the differential equation,

dC 2Q 22 M ‡ C ˆ Q 2 e dt V V

2Qt=V

This equation may be solved using an integrating factor to find

QM22 2Qt=V Cˆ te V2

(10.34)

(10.35)

247

248

Theory and Practice of Water and Wastewater Treatment

Figure 10.7 Tracer exit curves for 1, 2, 3, and 4 CM reactors in series for a slug input.

Following a similar procedure for an n-reactor series, the general expression for tracer exiting the final reactor is Cˆ

Qn 1 Mnn t n 1 e Vn …n 1†!

nQt=V

(10.36)

Plots for n = 1, 2, 3, and 4 (n is the number of reactors in series) are shown in Figure 10.7. The differential equations that result for treatment in PF reactors when reaction models are not first-order, usually requires a numerical solution. Often, software programs solve this problem by dividing the total reactor volume into a series of 10–12 CM reactors in series. Algebraic solutions can then be applied to each CM reactor to simplify the overall solution to determine effluent concentrations. Total computation time is considerably reduced and errors are minimized. 10.2.4 Other Flow Irregularities: Dead Volume and Short-circuiting Flow that passes through the reactor in a very short time relative to the average detention time is short-circuited. Naturally in a CM reactor, a large quantity of the flow passes through the reactor in a time less than the average detention time. Flow that leaves the reactor before the detention time because of the hydraulic mixing regime is not short-circuited. However, true shortcircuiting can exist in a CM reactor. Short-circuited flow is the excess flow that leaves the reactor in a relatively short period of time beyond the flow that leaves dictated by the hydraulic mixing. The tracer curve for a CM basin with some short-circuiting is illustrated in Figure 10.8. Short-circuiting can be caused by improper baffling, improper location of influent and effluent pipes, or more often by a density difference between the influent and reactor contents. (If ideal CM conditions existed, could the influent and effluent points be located next to each other?) A conceptualization of short-circuiting (QSC) is illustrated in Figure 10.9a. In reality, the shortcircuited flow occurs within the reactor and the viewer, of course, is only able to observe the total flow entering or exiting the reactor. However, short-circuited flow manifests its behavior according to the conceptualization. Based on this conceptualization, time-dependent phenomena will be significantly affected by short-circuited flow with essentially no treatment of the portion of influent that is shortcircuited. The average detention time of the total flow in the basin remains the same and the remainder of the flow (the portion that is not short-circuited) is treated for a longer period of time than the average detention time. Short-circuited flow can be analyzed by only considering the

10 Mass Balances and Hydraulic Flow Regimes

Figure 10.8 Tracer output for short-circuited flow in a CM basin.

Figure 10.9 Short-circuiting (a) and dead volume (b).

´

effective flow Q = Q QSC into the basin and mixing the untreated flow, QSC, with the treated portion. Dead volume is a portion of the reactor volume that is unused. Reactor geometry, obstructions, inadequate mixing and density currents can cause flow to bypass a portion of the reactor. The effective volume of the reactor is decreased from its physical volume, resulting in a decrease in ´ the treatment time. Dead volume (VD) is illustrated in Figure 10.9b. Use a volume V = V VD as the reactor volume in this case. 10.2.5 Typical Flow Characteristics of Basins The degree of mixing in a basin is assessed by the use of tracers as noted above. In the first analysis, tracer tests determine the flow regime and the appropriate model (CM or PF). When intermediate degrees of mixing occur, assuming that the basin is CM will always provide the most conservative estimate of basin performance (see Example 10.2), although when the flow regime approaches PF the factor of safety may be excessive. Conceptualizing the basin as a series of CM compartments can be used for a more accurate assessment of intermediate mixing situations as presented in Section 10.2.3. This common approach implies certain assumptions about how mixing occurs, but the approach is reasonable when water is the liquid (Lawler and Singer 1993). Levenspiel (1999) gives a more thorough analysis and discussion of these issues. Lawler and Singer (1993) give an excellent application of the concepts that are beyond the scope of this presentation. The following indices based on the tracer exit curve for a slug input of tracer give a gross characterization of the flow situation. Flow that is truly short-circuited and flow that leaves early because of mixing cannot be distinguished by these measures. Therefore, short-circuited flow will refer to both mechanisms below.

249

250

Theory and Practice of Water and Wastewater Treatment

The following definitions apply: ti θd tP tg t10 t90

time of first appearance of the tracer detention time time of the mode time of the mean time of passage of 10% of the tracer time of passage of 90% of the tracer.

1) ti/θd. This is a measure of the degree of mixing. Smaller values indicate more mixing: 0 for ideal mixing and 1.0 for ideal PF. For a typical sedimentation basin, it usually ranges between 0.2 and 0.3. 2) tP/θd. This indicates the average shortcircuiting, dead space, and effective tank volume. For a typical sedimentation basin, it ranges from 0 to 0.8. 3) t90/t10. This provides a measure of the dispersion index. It is 1.0 for an ideal PF basin and 21.9 for an ideal mixed (CM) basin. 4) 1 tP/tg. This is referred to as the index of short-circuiting. It is 0 for an ideal PF basin, and 1.0 for an ideal CM basin. 5) (t90 tP)/(tP t10). If this index is equal to 1.0, then the dispersion curve is symmetrical about the time tP/θd. The spread of the concentration curve depends on the reactor dispersion number, which is given by D/vL, where D is the longitudinal or axial dispersion coefficient, v is the mean displacement velocity, and L is the length of the reactor. 10.2.6 Measurement of Dispersion A better description of the mixing regime and other flow irregularities is obtained by considering the whole tracer curve. See Levenspiel (1999) for a more extensive discussion. It is essential that the tracer used to assess mixing, dead space, and short-circuiting be conservative and not influence the flow pattern in a significant way by causing, for example, density currents. Common dyes that have been used are rhodamine B or fluorescein. Radioactive tracers such as tritium are another option. Sodium chloride can also be employed, but using high concentrations of this salt can result in a significant density difference between the water in the vessel and sodium chloride containing influent. Sodium chloride has the advantage of being readily measured with a conductivity meter. Dextran blue (Jiminez et al. 1988) was found to be better than a number of tracers for submerged biological filters because of its high molecular weight, which minimized adsorption and diffusion in the biomass and packing. Lithium is commonly used as a tracer in anaerobic reactors. At high concentrations, lithium can inhibit activity in an anaerobic reactor (Anderson et al. 1991). Maximum lithium concentrations of 10–20 mg L 1 in a CM reactor are recommended; when PF conditions are suspected, lithium concentration delivered from a pulse input should not exceed 1.0 g L 1 to avoid inhibition.

10.3 Detention Time in Vessels Perhaps surprisingly, the mixing patterns in vessels do not affect the average retention time of a particle of fluid in a vessel, as the developments in the following sections will demonstrate.

10 Mass Balances and Hydraulic Flow Regimes

Figure 10.10 General tracer exit curve.

10.3.1 Average Detention Time The average retention time of a particle of tracer in a vessel is determined by weighting the exiting tracer masses by the time each elemental mass has spent in the reactor. For the general tracer exit curve shown in Figure 10.10, each elemental mass, mi, stayed a time of ti. The average detention time is θd ˆ

Σmi t i Σmi

(10.37)

mi ˆ QC i Δt θd ˆ

Σt i QC i Δt i Q ˆ tC …t † dt ΣQC i Δt i M∫

(10.38)

where M is the total mass of tracer injected. By applying the above equation to the tracer exit curves for CM and PF conditions, the average retention time of a particle of tracer in each case is seen to be V/Q. If an actual tracer exit curve does not produce this result, then dead space exists in the basin. The average retention of a particle of water is the same as the average retention time of a particle of conservative tracer. 10.3.2 The Effects of Flow Recycle on Detention Time Many treatment operations incorporate the recycle of a portion of effluent flow into the basin to improve treatment. The activated sludge process for example returns a portion of the underflow from a clarifier that follows the reactor. This situation will be examined for its effects on the average detention time of a particle of water in the reactor. Figure 10.11 shows the flow pattern of the system. Water exits the primary basin and passes through a second basin, from which a portion of the water is recycled to the influent pipe of the first basin. The recycled flow and the influent are well mixed at their confluence. Although the recycled flow was taken from the second basin, as shown in this figure, the same analysis would apply if the recycled flow has been taken directly from the effluent from the first basin. For an ideal CM basin, the recycled flow can be returned at any point to the basin without consequence. Practically, addition and withdrawal points should be separated by as great a distance as possible to prevent short circuiting. For a basin with a PF hydraulic flow regime the recycled flow would normally be mixed with the fresh influent before the combined flow entered the reactor.

Figure 10.11 System with recycled flow.

251

252

Theory and Practice of Water and Wastewater Treatment

In Figure 10.11, the ratio of recycled flow, Qr, to influent flow is r. The volume of the primary basin is V. rˆ

Qr Q

(10.39)

To calculate the average residence time of water in the first basin, an elemental volume of water, Δq will be followed. Regardless of the mixing regime in the basin, the average residence time of the elemental volume during one pass, tp, will be tp ˆ

V Q ‡ rQ

(10.40)

A portion of the effluent from the primary basin is recycled. The recycled fraction of the elemental volume is rQ Δq Q ‡ rQ

(10.41)

This recycled water again stays in the basin for an average time of tp. A fraction of this fraction will be recycled: rQ rQ Δq Q ‡ rQ Q ‡ rQ

ˆ

rQ Q ‡ rQ

2

Δq

(10.42)

This process is repeated an infinite number of times, with the exponent increasing with each pass. The total retention time for a unit elemental volume (tT) can now be formulated as tT ˆ

V rQ ‡ Q ‡ rQ Q ‡ rQ

ˆ

V 1 Q 1‡r

1‡

V rQ ‡ Q ‡ rQ Q ‡ rQ

r r ‡ 1‡r 1‡r

2

‡ ∙∙∙

2

V rQ ‡ ∙∙∙ ‡ Q ‡ rQ Q ‡ rQ

r 1‡r

i

V ‡ ∙∙∙ Q ‡ rQ

i

‡ ∙∙∙

(10.43)

This series is recognized to be a geometric series with a general term, r xˆ 1‡r tT ˆ

V 1 Q 1‡r

1 nˆ1

xn

1

ˆ

V 1 Q 1‡r

1 1

x

ˆ

V 1 Q 1‡r

r‡1 1‡r r

ˆ

V ˆ θd Q (10.44)

This proves that the average retention in the basin is not affected by the amount of flow recycled (theoretically r may range from 0 to 1). Recycle is an internal phenomenon within the system. In this case, the system consists of the two basins. If an enclosed line is drawn around the two basins, the input rate to the system is Q which is equal to the output rate. The volume of the first basin is V, and the HRT in the basin is simply V/Q. The same is true for the second basin. A volume of influent to the system stays in the second basin an average time calculated by dividing the basin’s volume by the volumetric flow rate into the system. This analysis should avoid confusion over the retention time for an individual pass (tp) through a basin and the overall average retention time (θd) in the basin. This distinction will have ramifications on the kinetics of treatment processes to be discussed in later chapters.

10 Mass Balances and Hydraulic Flow Regimes

Figure 10.12 Tracer exit curve for PF basin with recycle.

Recycling flow into a CM basin for the purpose of increasing fluid contact time is a fallacy. Likewise average fluid residence time in a PF basin with recycled flow is not changed with respect to the nonrecycle flow case, but the age distribution of the effluent is adversely affected as illustrated in the following section. As well, when removal of a constituent is the objective of the treatment, recycle of the effluent dilutes the influent that generally deteriorates efficiency of treatment. There are other reasons for recycling flow from a following basin into a preceding basin. One reason may be the possibility of short-circuited flow in the primary basin. Recycling flow from the secondary basin would increase exposure of the fluid to reaction in the first basin. In a biological treatment scheme, recycle of effluent in which biomass has been concentrated improves effective contact time in a reactor because the agents (biomass) responsible for reaction are increased in concentration in the reactor as explained in Chapters 17–19. One must understand the basic mechanisms at work in each reactor to evaluate disadvantages and advantages of recycle in each situation. 10.3.3 The Effects of Recycle on Mixing In a CM basin, recycling some of the effluent does not influence the degree of mixing because mixing is already at the extreme upper limit. It is sometimes not appreciated that recycle will increase the overall amount of mixing in a PF basin even though the flow regime remains PF during an individual pass of a plug through the basin. Recycle causes a portion of the effluent to be mixed with the influent. A slug input of tracer at a concentration of C0 over a small time dt into a PF reactor without recycle will have a spike of all tracer at a concentration of C0 over a time, dt, exiting at a time of θd (or V/Q). For a PF reactor with recycle from the effluent with an r value of 1, the tracer exit curve shown in Figure 10.12 would result for the same input of tracer at C0 over a time, dt. In this case, one-half of the effluent flow is returned to the influent to the basin and the remaining portion exits along with the corresponding amount of tracer. Therefore, tracer appears at time intervals of 0.5θd at concentrations that are 50% of the previous value. The tracer exit curve takes on characteristics of the curve expected from a CM basin. An infinite amount of recycle would produce a CM exit curve; a CM basin, in fact, has an infinite amount of internal recycle.

10.4 Flow and Quality Equalization Variation in flow rate and concentration adds variability to the treatment operation. More operator skill and careful monitoring and control are required to produce an acceptable

253

254

Theory and Practice of Water and Wastewater Treatment

treatment. Some of the treatment basins will have to be designed to accommodate the larger flows or additional basins will have to be provided, even though these flow rates will only exist for a part of the day, unless the excess flow is to be bypassed. Wastewater treatment operations are subject to variable flows and quality and providing an equalization basin will smooth variation in flow rate and quality, with resultant steady operation of the treatment operation. The equal­ ization basin may be in-line or off-line. An off-line basin would have flows above a predetermined rate diverted to it. The contents of the equalization basin would be returned to the process train when flows are below the target flowrate. Water treatment plants are normally run at a more constant flow rate because of the storage reservoirs provided in the distribution system. An equalization basin should not be allowed to run dry which would damage pumps; basin volume can be designed to prevent this or other pump control measures can be provided. For wastewater treatment plants, it may be possible to use excess volume in large interceptor sewers entering the plant for equalization storage. The USEPA (1974) recommends nightly or weekly drawdown of the interceptor system to flush out solids deposited during previous storage periods. In wastewater treatment operations, when organic loadings vary by more than 4 : 1 based on 4 hours composite samples, equalization is recommended (Eckenfelder and Ford 1970). An equalization basin affords the opportunity to pre-aerate the sewage before primary sedimenta­ tion, which results in improved settling of suspended solids (SS). To maintain aerobic conditions in a sewage equalization basin, air should be supplied in the range of 9–15 L air m 3 of basin volume (USEPA 1974). A time-varying wastewater flow rate is shown in Figure 10.13. The total volume, V, of water delivered over a 24-hour period is 24

V ˆ

∫0

Q…t †dt

(10.45)

and the average flow rate Q, over this period is Qˆ

1 24 Q…t †dt 24 ∫ 0

Figure 10.13 Variable flow rate over a 24-hour period.

(10.46)

10 Mass Balances and Hydraulic Flow Regimes

The volume of a basin required to store water to allow water to be delivered at a constant flow rate is computed by considering the volume of water delivered when the flow rate is higher than the average flow rate. In Figure 10.13a, the area above the average flow rate line and the corresponding areas below this line have been labeled V3, V1, and V2, respectively. It is clear that V3 ˆ V1 ‡ V2 The volume required for an equalization basin is V3. Using a volume balance on the equalization basin, starting with an empty basin at time t1, Out ˆ Accumulation

In t2

∫ t1

Q dt

Q…t 2

t1† ˆ V 3

From time t2 to 24 hours, 24

∫ t2

Q dt

t 2 † ˆ V 2

Q…24

and for time 0 to t1 t1

∫0

Q dt

Q…t 1

0† ˆ V 1

A graphical approach is illustrated in Figure 10.13b. From this figure, the equalization basin volume is V2 V1. A more typical sewage flow variation has the shape of the curve shown in Figure 10.14a. There are no differences to the approach outlined above if the double humped portion of the curve lies above the average flow rate line. However, if the valley extends below the average flow rate line, as shown in Figure 10.14b, the basin is designed for a volume of V 5;

if V 4 < V 2

or V5 ‡ V4

V 2;

if V 4 > V 2

Figure 10.15 illustrates a flow situation where the cumulative volume is above and below the cumulative volume discharged at the average flow rate. The graphical solution is similar to the above graphical solution. Again, from this figure, the volume of the equalization basin is V2 V1.

Figure 10.14 Diurnal variation of flow rate.

255

256

Theory and Practice of Water and Wastewater Treatment

Figure 10.15 Graphical solution for equalization volume.

The quantity variation will be rectified by the approach outlined above, but the quality variation of flow exiting from the equalization basin will not be totally smoothed. Quality or loading depends on both quantity and concentration. Off-line equalization results in less smoothing of concentration than in-line equalization. The following equation would be used to calculate the average loading rate. QC ˆ

1 24 Q…t †C …t † dt 24 ∫ 0

(10.47)

To design a basin to produce a constant loading would require continuous monitoring of the constituent’s concentration in the influent. Mixing characteristics in the equalization basin would have to be known and a variable flow rate from the basin would be used to produce a constant loading. Variable concentrations of different components will make it impossible to achieve a uniform loading for all components. However, there will be some mixing in the basin and loading variation is normally dampened considerably by pumping at a constant flow rate. There will also be some equalization of flow rate and quality as the water passes through basins in the treatment plant because of mixing and the hydraulic characteristics of the channels. Each basin should be evaluated to determine if the equalization is significant (see Problem 30).

10.5 System Material Balances Individual operations are assembled together to form the complete treatment train. Chemical addition, sludge withdrawal, flow recycle, and chemical and biological transformations cause the liquid and solids flows to differ at various locations within the treatment plant. Sometimes split treatment schemes are used for one part of the treatment. In split treatment, a portion of the water is treated to a degree higher than the desired level, while the remainder bypasses the treatment. The two streams are combined to form the product water from the unit operation. To characterize the system, the performance of each individual operation, including each of their inputs and outputs, must be known. The principles of mass balances discussed earlier apply to each basin. There is usually more than one basin or reactor for each unit operation. Some operations receive input from more than one other unit operation. For instance, a solids digester in a wastewater treatment operation typically has influent from both the primary and secondary clarifiers. The influent is a function of the sludge concentrations in each of the clarifiers. The

10 Mass Balances and Hydraulic Flow Regimes

number of unit operations and the number of cross-connections among them, as well as the variability of the influent quality and flow, increase the challenge of finally sizing the number of basins and characterizing the overall system inputs and outputs. The approach to making the system balances is to lay out the complete plant in a logical sequence as done in Figures 9.1–9.6. To facilitate making mass balances, a color scheme can be used or liquid flow lines can be made solid and solids flow lines can be dotted. Chemical feed lines can be dashed. The process performance characteristics of each unit operation (e.g., chemical concentration in the process, quantity of solids generated or destroyed) are indicated along with other a priori known characteristics such as the concentrations of commercially available agents. The influent flow and its quality characteristics are listed at the front of the plant. Then mass balances are made for each individual operation. All inputs and outputs are isolated. The inputs and transformations in an operation determine its outputs. This procedure is applied to each unit operation of the plant. The process is not necessarily straightforward because of return flows from later treatment operations to operations near the front of the treatment train. A trial and error approach may be required to achieve the correct balance. For each individual unit operation, and for the system as a whole, the sum of liquid inputs must equal the sum of liquid outputs. Each basin or device is isolated and the symbolic representation of Eq. (10.1) is written for the unit. A box is drawn around the unit and all lines that intersect the box are included in the materials balance expression. Lines going into the box appear on the input side of the materials balance expression and those going out of the box appear on the output side. In addition, any generation or destruction of a constituent by chemical or other means must be included in the materials balance. Mass balances for any constituent must be fully accounted for in the input–generation–output balances. When flow enters a Tee joint, the concentration of dissolved solids or SS does not change through the joint. To perform a materials balance exercise, it is not necessary to know the detention times or other design features of the units, which will be covered in later chapters. The process is best illustrated by example. In the example, the necessary performance characteristics of various processes will be supplied. However, it is assumed that principles covered in preceding chapters (particularly on chemical reaction stoichiometry) are to be applied as necessary. As the first exercise in designing a plant, typical performance of individual operations is assumed and modified by any data that are available. Later chapters cover the design of individual operations, after which this exercise can be repeated taking into account the specific design details to achieve the expected performance of each operation. Example 10.4 Mass Balance The problem is to find the flow rates and solids concentrations at all locations within an activated sludge wastewater treatment plant as illustrated in Figure 9.4. A centrifuge will be used to dewater the digested sludge from the anaerobic digester instead of a vacuum filter. Influent data and unit operation performance information are given below. The process has been redrawn in Figure 10.16 with symbols for flow and solids concentrations throughout the plant. The definition of symbols in Figure 10.16 is as follows: The list is arranged according to the sequence through the treatment train: S0 Q0 X0 QSc

influent BOD5 concentration influent flow rate influent SS concentration (excluding screenings and grit) flow rate after the screens

257

258

Theory and Practice of Water and Wastewater Treatment

Figure 10.16 Materials balance flow diagram for an activated sludge wastewater treatment plant. The symbols are defined in Example 10.4.

QScw XScw Qg Qgw Xgw Qp0 Xp0 Qp Xp Sp Qup Xup QA XA SA QS XS QuS XuS r QFe XFe QCl Qf Qw Qt Xt

volumetric removal rate of screenings concentration (density) of screenings flow rate after the grit chamber volumetric removal rate of grit concentration (density) of grit collected in the grit chamber total flow rate into the primary sedimentation basin SS concentration of influent to the primary sedimentation basin flow rate of clarified supernatant from the primary sedimentation basin SS concentration of clarified supernatant from the primary sedimentation basin BOD5 concentration in the effluent from the primary sedimentation basin flow rate of underflow from the primary sedimentation basin SS concentration in the underflow from the primary sedimentation basin flow rate of effluent from the aeration (activated sludge) basin SS concentration in the effluent from the aeration basin BOD5 concentration in the effluent from the aeration basin flow rate of clarified supernatant from the secondary sedimentation basin SS concentration of clarified supernatant from the secondary sedimentation basin flow rate of underflow from the secondary sedimentation basin SS concentration in the underflow from the secondary sedimentation basin ratio of recycled flow from underflow of secondary sedimentation basin to flow from primary sedimentation basin flow rate of chemical feed for ferric chloride SS concentration in the ferric chloride chemical feed gas flow rate of chemical feed for chlorine flow rate from the disinfection basin flow rate of waste-activated sludge from the underflow of the secondary sedimentation basin flow rate of underflow from the sludge thickener SS concentration in the underflow from the sludge thickener

10 Mass Balances and Hydraulic Flow Regimes

QtS XtS Qd Xd Qpl Xpl Qck Xck Qct Xct

flow rate of supernatant from the sludge thickener SS concentration in the supernatant from the sludge thickener flow rate from the underflow of the anaerobic digester SS concentration in the underflow from the anaerobic digester flow rate of chemical feed for conditioning polymer for the centrifuge SS concentration in the polymer chemical feed volumetric removal rate of cake from the centrifuge SS concentration in the cake from the centrifuge flow rate of centrate from the centrifuge SS concentration in the centrate from the centrifuge

The following information is given: Q0 ˆ 437 ML d 1 ;

S 0 ˆ 320 mg L 1 ;

X 0 ˆ 330 mg L

1

S0 and X0 do not include material that will be removed by the screens or in the grit chamber. Screens:

The quantity of screenings collected is 0.005 m3/1000 m3.

Grit chamber:

The quantity of grit collected is 0.008 m3/1000 m3.

Primary sedimentation The SS removal is 55%. basin: The BOD5 removal is 35%. The underflow concentration of SS is 3.8%. Activated sludge aeration basin:

Effluent soluble BOD5 is 5 mg L 1. The net SS yield based on influent BOD5 and soluble effluent BOD5 is 0.65 mg SS produced/mg BOD5 removed. The concentration of SS in the aeration basin is 2000 mg L 1.

Secondary sedimentation basin:

Ferric iron is being added to the influent to the clarifier at 5 mg L 1 as Fe. The coagulant is FeCl3. The feed solution of FeCl3 contains FeCl3 at 20% by weight. The clarified overflow (effluent) from this basin contains 10 mg L 1 of SS. The underflow SS concentration is 0.75%.

Thickener:

The thickener captures 85% of the solids (i.e., 85% of influent solids appears in the underflow). The underflow contains SS at 4.5%.

Disinfection:

Chlorine gas is being added at a dose of 1.5 mg L 1.

Anaerobic digester:

SS reduction of 55% is achieved.

Centrifuge:

9 kg of polymer are being added to each tonne of solids. The feed solution of

polymer contains polymer at 80 g L 1. The unit captures 97.5% of the solids. The cake SS content is 32%.

For convenience it will be assumed that the specific gravity (s.g.) of SS is 1.00. The error associated with this assumption is small. (Chapter 21 discusses the equations necessary to correct this problem.) It is also assumed that there is no soluble BOD5 generation in the anaerobic digester. Likewise, it is assumed that the thickener supernatant and centrifuge centrate do not contain soluble BOD. The only points where BOD5 transformations occur are after the primary

259

260

Theory and Practice of Water and Wastewater Treatment

clarifier and the aeration basin. After consultation of later chapters, corrections for these assumptions and other refinements on the BOD and SS transformations can be made and incorporated into the material balance exercise, but the procedure for determining flows and solids concentrations is as presented here. The solids concentration in the aeration basin is assumed to be uniform (i.e., CM conditions exist). Liquid, solids, and substrate balances can be written at each node in the system. For example, the box in Figure 10.16 isolates the liquid flow streams for the primary clarifier and will be used below to construct the flow balance for this unit. Proceeding first with the balances for liquid flows: Screens Q0 ˆ QSc ‡ QScw

(i)

Grit chamber QSc ˆ Qg ‡ Qgw

(ii)

Primary sedimentation basin Qg ‡ Qct ‡ QtS ˆ Qp ‡ Qup

(iii)

Qg ‡ Qct ‡ QtS ˆ Qp0

(iv)

Activated sludge aeration basin Qp ‡ rQp ˆ QA

(v)

Secondary sedimentation basin QA ‡ QFe ˆ QS ‡ QuS

(vi)

QuS ˆ Qw ‡ rQp

(vii)

Disinfection basin QS ‡ QCl ˆ Qf QCl is a gas flow rate and will not affect the liquid flow rate. Therefore, this equation simplifies to: QS ˆ Q f

(viii)

Thickener Qup ‡ Qw ˆ QtS ‡ Qt

(ix)

Anaerobic digester Qt ˆ Qd

(x)

10 Mass Balances and Hydraulic Flow Regimes

Centrifuge Qd ‡ Qpl ˆ Qck ‡ Qct

(xi)

In addition to these liquid balances for each individual unit, an overall liquid balance can be made for the process. This equation is not an independent relation from the above equations; it can be constructed from the individual unit balances. There are 22 variables in the above equations. QCl has been eliminated and only one of the variables, Q0, has been specified. We need some more equations! The solids and substrate balance equations are now formulated.

Screens The volumetric collection rate of screenings has been given; therefore, screenings collected will be calculated on a volumetric basis with XScw = 0.005 m3/1000 m3. Note that screenings are not included in the influent solids concentration. Q0 X Scw ˆ QScw

(xii)

Grit Chamber As for screenings, grit is not included in the influent solids concentration, and the concentra­ tion of grit removed is Xgw = 0.008 m3/1000 m3 on a volumetric basis. QSc X gw ˆ Qgw

(xiii)

Primary Sedimentation Basin Balances must be made for both solids and soluble substances in each of the basins.

Solids Balances There has been no change in X0 in the first two operations. Qg X 0 ‡ Qct X ct ‡ QtS X tS ˆ Qp X p ‡ Qup X up

(xiv)

Qg X 0 ‡ Qct X ct ‡ QtS X tS ˆ Qp0 X p0

(xv)

The removal ratio for the primary clarifier has been specified; it will be represented by Rp. Qp X p ˆ Rp Qp0 X p0

(xvi)

Substrate Balance There has been no change in the BOD5 concentration in the first two operations. Also by assumption, there is no BOD5 in Qct or QtS. In the information supplied, the reduction factor for BOD5 in the primary clarifier is fpB0D = 0.35. The BOD5 in the underflow from the primary clarifier is also ignored. Therefore, the substrate balance is Qg S 0 ˆ Qp S p ‡ f pB0D Qg S 0 The last term in Eq. (xvii) describes the removal of BOD5 in the primary clarifier.

(xvii)

261

262

Theory and Practice of Water and Wastewater Treatment

Activated Sludge Aeration Basin Solids Balance There is generation of solids in the aeration basin, which will be termed ΔX in the balance. Qp X p ‡ rQp X uS ‡ ΔX ˆ QA X A

(xviii)

Substrate Balance The substrate balance for the aeration basin includes a term (ΔS) for removal of BOD5. Qp S p ‡ rQp SA ˆ QA S A ‡ ΔS

(xix)

To determine ΔX in Eq. (xviii), the net yield (designated Y below) was supplied in the given information. The net yield is based on the removal of BOD5 in the aeration basin. ΔX ˆ Y ΔS

(xx)

Secondary Sedimentation Basin The dosing rate of FeCl3 is defined as DFe. …QFe X Fe †=QA ˆ DFe

(xxi)

The Fe3+ will be transformed by chemical reaction to Fe(OH)3. X ´Fe is used to designate Fe(OH)3. QA X A ‡ QFe X ´Fe ˆ QS X S ‡ QuS X uS

(xxii)

There is no change in solids concentration where QuS splits into Qp and Qw.

Disinfection Basin There is no change in SS concentration in the disinfection basin because chlorine is a soluble gas. A very small amount of BOD will be oxidized by chlorine, but it will assumed to be negligible.

Thickener The capture ratio for the thickener will be defined as Ct. Qup X up ‡ Qw X uS ˆ QtS X tS ‡ Qt Xt

(xxiii)

Qt X t ˆ Ct Qw X uS ‡ Qup X up

(xxiv)

Anaerobic Digester Solids reduction in the anaerobic digester has been specified. A factor, fAD, will be used to describe the amount of solids reduction in the digester. Qt X t ˆ Qd X d ‡ f AD Qt X t

(xxv)

Centrifuge The dosing rate for polymer is defined as Dpl. The capture ratio for the centrifuge is defined as Cc. Qpl X pl =…Qd X d † ˆ Dpl

(xxvi)

Qd X d ‡ Qpl X pl ˆ Qck X ck ‡ Qct X ct

(xxvii)

Qck X ck ˆ Cc Qd X d ‡ Qpl X pl

(xxviii)

10 Mass Balances and Hydraulic Flow Regimes

There are 21 new variables (including ΔX and ΔS) in the above 17 equations [Eqs. (xii)–(xxviii)]. However, 13 of these variables are specified in the information supplied. There are 28 unknowns and an equal number of equations; therefore, the system is determined. As the equations are solved, keeping units with the parameters will provide a check to ensure that the calculations are correct. Also, it may be helpful to isolate two or more units in the system to solve for some of the variables. The solids removals in the screens and grit chamber are straightforward. From Eq. (xii): QScw ˆ 437  103 m3 d

1

 0:005 m3 =1000 m3 ˆ 2:19 m3 d

1

Using Eq. (i), QSc = 437 × 103 m3 d 1 2.19 m3 d 1 = 437 × 103 m3 d 1 From Eq. (xiii), Qgw = (437 × 103 m3 d 1) × (0.008 m3/1000 m3) = 3.50 m3 d Using Eq. (ii), Qg = 437 × 103 m3 d 1 3.50 m3 d 1 = 437 × 103 m3 d 1

1

The equations can be solved algebraically or by making a few reasonable assumptions and iteratively solving the equations. The latter approach is most practical. The main liquid stream has its solids removed in the primary and secondary clarifiers. The substrate in the influent is removed in the aeration basin. These phenomena will be used to derive the first estimates of unknown flows. In the primary clarifier, 55% of SS are removed. The first estimate of the underflow from this clarifier can be based on the influent solids concentration. From Eqs. (xv) and (xvi) and ignoring Qct and QtS in Eq. (xv), SS removed in primary clarifier ˆ 0:55 437  103 m3 d ˆ 79 315 kg d

1

330 mg L

1

78 315 kg d 1 1 m3 0:038 kg L 1 1000 L

3

1 kg 106 mg

1

The solids content of the underflow from the clarifier is 3.8% or 38 g L for the solids. Qup can be directly calculated by Qup ˆ

1000 L m

ˆ 2087 m3 d

1

based on a s.g. of 1.00

1

Now Qp can be calculated from Eq. (iii): Qp ˆ Qg

Qup ˆ 437  103 m3 d

1

2087 m3 d

1

ˆ 434:9  103 m3 d

1

The effluent solids content in Qp from the primary clarifier is Xp ˆ 1

Rp X 0 ˆ …1

0:55† 330 mg L

1

ˆ 149 mg L

1

The effluent substrate content in Qp from the primary clarifier is found from Eq. (xvii), Sp ˆ

1

f p Qg S 0 Qp

ˆ

…1

0:35† 437  103 m3 d 434:9  10

3

m3

1

320 mg L

d

1

1

ˆ 209 mg L

1

In the equations for the aeration basin, the unknowns are r, ΔX, QA, and ΔS. The only source of BOD5 is the influent to the aeration basin and the final BOD5 concentration from the aeration basin is given as 5 mg L 1. Therefore, using Eqs. (v) and (xix), ΔS ˆ Qp Sp

S A ˆ 434:9  103 m3 d

1

208

5 mg L

1

103 L m

3

1 kg 106 mg

ˆ 88 719 kg d

1

263

264

Theory and Practice of Water and Wastewater Treatment

Now Eq. (xx) can be used to find ΔX. Then Eqs. (xviii) and (v) can be solved together to find r and QA. For ΔX, ΔX ˆ Y ΔS ˆ 0:65 88 719 kg d

1

ˆ 57 667 kg d

1

Combining Eqs. (xviii) and (v), Qp X p ‡ rQp X uS ‡ ΔX ˆ …1 ‡ r†Qp X A The underflow from the secondary clarifier, XuS has been given as 7500 mg L aeration basin SS concentration (XA) has been given as 2000 mg L 1. rˆ

1

(0.75%) and the

Qp X p X A ‡ ΔX Qp …X A X uS † 434:9  103 m3 d

1

149

2000 mg L

1

1 kg ‡ 57 667 kg d 106 mg 1 kg …1000 L m 3 † 106 mg

1000 L m

ˆ 434:9  103 m3 d

1

2000

7500 mg L

1

3

1

ˆ 0:31

From Eq. (v), QA ˆ …1 ‡ r†Qp ˆ …1 ‡ 0:31† 434:9  103 m3 d

1

ˆ 570:8  103 m3 d

1

The concentration of Fe3+ in the influent to the secondary clarifier is 5 mg L 1. The amount of FeCl3 added will be 14.5 mg L 1. The concentration of FeCl3 in the feed stream is 0.20 kg L 1. QFe can be determined from Eq. (xxi). QFe ˆ

14:5 mg L 1 570:8  103 m3 d DFe QA ˆ X Fe 0:20 kg L 1 106 mg kg 1

1

ˆ 41:4 m3 d

1

FeCl3 will react with alkalinity to form Fe(OH)3. The Fe(OH)3 formed is X ´Fe in Eq. (xxii), which can be considered to be part of the influent solids to the secondary clarifier. For every 14.5 mg L 1 of FeCl3, 9.6 mg L 1 of Fe(OH)3 will be formed. Equations (vi) and (xxii) can be combined to solve for QS and QuS. XS has been specified as 10 mg L 1. ´ ˆ …QA ‡ QFe QA X A ‡ QFe X Fe

QuS †X S ‡ QuS X uS

which can be rearranged to solve for QuS: ´ X S † ‡ QFe X Fe XS X uS X S

570:8  103 m3 d 1 2000 10 mg L 1 ‡ 41:4 m3 d ˆ 7500 10 mg L 1

QuS ˆ

QA …X A

1

9:6

1

10 mg L

ˆ 151:7  103 m3 d

1

From Eqs. (vi) and (vii), QS ˆ QA ‡ QFe Qw ˆ QuS

QuS ˆ 570:8  103 ‡ 41:4

rQp ˆ 151:7  103

151:7  103 m3 d

…0:31† 434:9  103

m3 d

1

1

ˆ 419:1  103 m3 d

ˆ 16:9  103 m3 d

1

1

The effluent flow from the treatment plant, Qf, is equal to QS from Eq. (viii) and is 419.1 × 103 m3 d 1. Because the dose of Cl2 is 1.5 mg L 1, the flow rate of Cl2 gas can be calculated.

10 Mass Balances and Hydraulic Flow Regimes

Applying Eq. (xxiv), QtXt may be found. Qt X t ˆ C t Qw X uS ‡ Qup X up

ˆ 0:85 16:9  103 m3 d 103 L 1 m3

1 kg 106 mg

1

7500 mg L

‡ 2087 m3 d

1

ˆ 175:1  103 kg d

175:1  103 kg d

1

1

(4.5%).

3

1 m 103 L

1

ˆ 3891 m3 d

1

0:045 kg L

38 000 mg L

1

The thickener underflow is determined using Xt = 0.045 kg L Q Xt Qt ˆ t Xt

1

1

Qt is also the flow rate from the anaerobic digester according to Eq. (x). The supernatant flow from the thickener is [Eq. (ix)], QtS ˆ Qup ‡ Qw

Qt ˆ 2087 ‡ 16:9  103

ˆ 15:1  103 m3 d

1

3891 m3 d

1

The concentration of solids in the supernatant is [Eq. (xxiii)]: X tS ˆ

Qup X up ‡ Qw X uS QtS

Qt X t

ˆ ‡ 16:9  103 m3 d

2087 m3 d 1

1

1

7500 mg L

38 000 mg L

3891 m3 d

15:1  10 m3 d 3

1

1

45 000 mg L

1

1

ˆ 2050 mg L

1

The anaerobic digester’s output loading is [Eq. (xxv)]: Qd X d ˆ 1

f AD Qt X t ˆ …1

0:55† 175:1  103 kg d

ˆ 78 795 kg d

1

1

The concentration of solids in the effluent from the digester is X d ˆ Qd X d =Qd ˆ 78 795 kg d

1

= 3891 m3 d

1

ˆ 20:3 kg m

3

ˆ 20:3 g L

1

The polymer is applied at 9 kg/tonne of solids (Dpl). The application rate of polymer (Apl) to be added to the flow rate of 392 m3 d 1 from the digester containing 79 380 kg of solids is Apl ˆ 78 795 kg d

1 tonne 1000 kg

1

9 kg tonne

ˆ 709 kg d

1

The polymer feed flow rate is [Eq. (xxvi)], Dp Q d X d ˆ Qpl ˆ X pl

…9 kg=tonne† 3891 m3 d 80 g L

1

1

20:3 kg m

103 L m

3

1 tonne 103 kg

1 kg 103 g

3

ˆ 8:9 m3 d

Equation (xxviii) is used to find the centrifuge cake output loading rate, QckXck. Qck X ck ˆ C c Qd X d ‡ Qpl X pl

ˆ 0:975 3891 m3 d 1000 L m

3

1

1 kg 1000 g

20:3 g L

1

‡ 8:9 m3 d

ˆ 77 707 kg d

1

1

80 g L

1

1

265

266

Theory and Practice of Water and Wastewater Treatment

The volumetric rate of cake production is Qck ˆ

Qck X ck 77 707 kg d 1 1 m3 ˆ X ck 0:32 kg L 1 103 L

ˆ 243 m3 d

1

Using Eq. (xi), Qct can be determined as Qct ˆ Qd ‡ Qpl

Qck ˆ …3891 ‡ 8:9

243† m3 d

1

ˆ 3657 m3 d

1

Equation (xxvii) can be used to find the concentration of solids in Qct. X ct ˆ

Qd X d ‡ Qpl X pl Qct

ˆ

3891 m3 d

1

Qck X ck 20:3 kg m

3

‡ 8:9 m3 d

1

80 kg m

3657 m3 d

ˆ 0:53 kg m

3

ˆ 530 mg L

3

243 m3 d

1

320 kg m

3

1

1

Flow (m3 d − 1)

SS (mg L − 1)

Flow (m3 d − 1)

SS (mg L − 1)

Flow (m3 d − 1)

Q0 = 437 × 103

X0 = 330

Qp = 435 × 103

Xp = 149

QS = 419 × 103

QScw = 2.19

XScwa)

QtS = 15.3 × 10

XtS = 2036

Qw = 16.9 × 10

Qct = 3679

Xct = 580

Qt = 3891

Xt = 45b)

Xup = 38b)

Qd = 3891

Xd = 20.3b)

XA = 2000

Qck = 243

Xck = 320b)

XuS = 7500

QFe = 41.4

XFe = 200b)

= 0.005

QSc = 437 × 103 Qgw = 3.50 Qg = 437 × 10

Xgwa) = 0.008

Qup = 2087 QA = 571 × 10

3

Qp0 = 437 × 103

3

Xp0 = 330

3

QuS = 152 × 103

SS (mg L − 1)

XS = 10

3

a) In m3/1000 m3. b) In g L 1.

The first estimates of all quantities are now completed. At least another iteration should be performed using the values obtained for thickener and centrifuge supernatant return flows and their respective SS concentrations to obtain better accuracy. This is left as an exercise for the student (see Problem 33). Examine the flow rates and solids concentrations around the system that are given in the table above. It is seen that the underflow from the primary clarifier is small compared to the main liquid stream (the secondary clarifier has recycle flow which significantly augments its underflow). In the first iteration, it is often a good assumption to ignore the adjustments in the main liquid flow resulting from reductions such as these.

Questions and Problems 10.1 Would mixing only in the vertical and lateral directions negate PF conditions? Would it negate CM conditions? Explain. 10.2 Why or why not can mass balances be made over the whole basin volume for PF and CM basins? 10.3 What are the differences among bulk flow, advection, diffusion, and dispersion? Would SS be subject to all phenomena?

10 Mass Balances and Hydraulic Flow Regimes

10.4 What volume of a PF reactor is required to remove 90% of a substance that decays according to first-order kinetics with a rate constant of 0.05 d 1? The flow rate is 395 m3 d 1. 10.5 Compare the total reactor volume required for 90% removal of a substance that decays according to first-order kinetics with a rate constant of 0.05 d 1 in a CM system that contains a single reactor against a system that contains three CM reactors of equal volume in series. 10.6 a What are the relative volume requirements for a CM reactor compared to a PF reactor if first-order kinetics apply and treatment efficiencies of 50%, 75%, 90%, 95%, and 99% are required? The same rate constant and flow rate apply to each system. b Perform the same exercise if second-order kinetics apply. 10.7 What are the characteristics of an ideal tracer? 10.8 Using the results from the exit concentration curve for a slug input of dye into a CM reactor, prove that the average residence (detention) time of a particle of dye and thus of a particle of water, in a CM unit is V/Q for a reactor with a volume, V, receiving a steady input flow of Q. 10.9 An experimenter wishes to conduct steady-state studies on a continuous flow CM reactor. It is generally thought that data can be safely taken after at least 95% of the initial material that has been in the reactor at startup is washed out. For how many detention times must this experimenter wait until data should be taken? 10.10 A reactor is operated in the following manner: (i) the reactor fills with influent; (ii) a chemical agent is added and mixing is applied to the basin; and (iii) after the reaction is completed, mixing is terminated and the basin is emptied to begin another cycle. Would this reaction be modeled with CM or PF kinetics? 10.11 What is the expression describing the concentration of tracer in the effluent from a CM reactor that receives a steady concentration, CI, of conservative tracer in a flow, Q for a finite period of time t0 to t1? The reactor initially had no tracer in it. 10.12 Why can recycle flow be delivered to any point in a CM basin? Would the same principle apply to a PF basin? 10.13 Prove that the ratio t90/t10 should be 21.9 for an ideal CM basin receiving a slug input of tracer. 10.14 A CM reactor receives influent containing 10.0 mg L 1 of tracer for 2 hours then tracer addition is terminated, but the flow remains steady. The reactor had a concentration of tracer of 1.0 mg L 1 when tracer addition at 10.0 mg L 1 commenced. The volume of the reactor is 10 L, and the influent flow rate is 2 L h 1. What is the concentration of tracer in the reactor 1 hour after tracer addition is terminated? 10.15 In Example 10.3, it is shown that the curve describing concentration of tracer in the effluent from a CM reactor receiving a constant concentration of tracer in the influent

267

268

Theory and Practice of Water and Wastewater Treatment

starts at a concentration of zero. However, when a slug input of tracer is input into a CM reactor, the initial concentration of tracer in the reactor and effluent is C0 [see Eq. (10.31)]. In both instances, the concentration of tracer in the reactor before tracer addition was zero. Explain this seemingly contradictory situation. 10.16 Stormwater runoff is often treated in flow-through ponds that approach CM conditions when the runoff event is occurring. Some treatment occurs during the runoff event, but the bulk of treatment occurs by retention of the stormwater in the pond until the next event occurs. A pond with a volume of 15 000 m3 receives a runoff volume of 10 000 m3 delivered uniformly over the 4 hours duration of the event. What fraction of the influent 10 000 m3 volume exits the pond during the event and what is the fraction that remains until the next event occurs? 10.17 If you have studied convolution in hydrology (or elsewhere), apply it to determine the effluent tracer concentration change with time from a CM basin receiving pulse inputs with a concentration of 100 μg L 1 of tracer lasting 1 minute and followed by a 4 minutes interval until the next pulse. The flow is 550 m3 d 1 and the detention time in the basin is 6 hours. Assume the pulses are impulses occurring over 5 minutes (is this reasonable?). Use discrete convolution to find the effluent tracer concentration immediately after the sixth pulse. 10.18 If short-circuited flow is defined as the flow that stays in the basin for less than 5% of the average detention time, how much flow is “naturally” short-circuited in an ideal CM basin? 10.19 What portion of a CM reactor is occupied by water that has stayed in the reactor for more than four detention times? 10.20 What is the average retention time of a particle of water in a PF reactor with a portion of the reactor effluent directly returned to the front of the basin? Draw the exit tracer curves for a PF basin with a volume of V and an influent flow of Q that have effluent recycle with r values of 0.5 and 2. Plot mass of tracer exiting versus t/θd. 10.21 Compare the effluent tracer concentration from the final reactor of three CM reactors each with volume V/3 in series with the effluent tracer concentration from one CM reactor of volume V. A slug input of tracer was used and the same mass of tracer was injected into each system. For both systems plot C/C0 against t/θd, where C0 and θd apply to the single-reactor system. Travel time between reactors is negligible. 10.22 As more CM reactors in series are provided, the system becomes more “PF.” For a substance that decays according to first-order kinetics, what is the equivalent PF rate constant that would be used to describe treatment in a series of three CM reactors in terms of Q, V (the total volume in the system), and the rate constant observed for the CM system? 10.23 What is the effluent concentration of a substance that enters the first reactor, in a series of three CM reactors, at a concentration of 560 mg L 1? Each reactor has a volume of 80 m3 and the flow rate is 150 m3 h 1. The substance decays according to second-order kinetics with a rate constant of 0.063 L (mg h) 1.

10 Mass Balances and Hydraulic Flow Regimes

10.24 An industry uses a special cleaning compound to wash some equipment every 8 hours. The cleaning compound contains a water-soluble substance that is inhibitory to their biological treatment process above a threshold concentration. The biological treatment process has a CM reactor without recycle. The reactor volume is 50 m3 and the influent flow rate is steady at 10 m3 h 1. The average amount of the inhibitory compound used in each washing and discharged to the waste stream is 75 g. Assuming that the inhibitory compound is delivered in a slug, what are the minimum and maximum concentrations of this compound in the biological reactor after a large number of cycles? 10.25 In Eq. (10.4f), how is time related to distance? 10.26 A CM basin with a volume of 475 m3 receives a flow of 2000 m3 d 1. The basin has a dead volume of 42 m3. Find the effluent tracer curve if a mass of 5 g of tracer is added instantaneously to the influent. Also plot the effluent tracer curve for the reactor without dead volume. 10.27 Does moving the location of dead volume change its effect on the performance of (a) a CM reactor (b) a PF reactor? 10.28 a For the flow rate data given in the table below, what is the minimum volume required for an equalization basin to produce a constant flow rate? What is the average hydraulic residence time in the equalization basin? Time (h) 3

1

Q (m s )

0000

0100

0200

0300

0400

0500

0600

0700

0800

0.11

0.15

0.20

0.26

0.30

0.34

0.36

0.37

0.37

Time (h)

0900

1000

1100

1200

1300

1400

1500

1600

1700

Q (m3 s 1)

0.35

0.31

0.26

0.25

0.26

0.29

0.34

0.40

0.46

Time (h)

1800

1900

2000

2100

2200

2300

2400

0.49

0.49

0.46

0.31

0.26

0.20

0.11

3

1

Q (m s )

b If the concentration of a constituent in the influent to the equalization basin is constant over the 24-hour period, will the load of the constituent from the basin be constant? If the concentration of the constituent in the basin influent varies over the 24-hour period, will the basin produce a constant loading rate? 10.29 A basin of volume V receives a steady flow Q, but the concentration of a nonremovable component in the influent varies according to C ˆ C 1 ‡ C 2 sin ωt …C 1 ; C 2 ; and ω are constants† What is the expression for effluent concentration of this component if the basin is (a) CM; (b) PF? 10.30 Flow into a basin follows the equation Q ˆ 7500 ‡ 3900 sin…0:2618t† where Q is in m3 d

1

and t is in hours.

269

270

Theory and Practice of Water and Wastewater Treatment

a Find the discharge curve over a 24-hour period using the above equation if the inflow is into a rectangular basin with a length:width ratio of 3 : 1 and overflow straight edge weir that has a length that is twice the width of the basin. The plan area of the basin is 166 m2 and the discharge coefficient for the weir is 0.62. Compare the ratio of the maximum and minimum values of the inflow to the ratio of the maximum and minimum values of the outflow. (Hint: Both parts a and b will require a numerical solution.) b Perform the same exercise for a circular basin with the same plan area and an overflow straight edge weir extending around the periphery of the basin. 10.31 An engineer wishes to evaluate the mean detention time of stormwater in a stormwater collection pond. The engineer has written a versatile computer program to convert a rainfall record over time into runoff, which is routed into a pond. The treatment algorithm for the pond allows the engineer to simulate varying degrees of mixing in the pond and any type of reaction. The pond to be evaluated is always full and water exits the pond by means of an overflow weir. Flow into the pond is highly variable over any period of time because of the intermittent, variable nature of rainfall. To evaluate the residence time distribution (and mean detention time) of stormwater runoff in the pond the engineer chooses to run the program with a 20 year historical rainfall record that will provide ample runoff flow variation. The engineer will input into the runoff a constant concentration of a substance that will be decayed by a first-order reaction in the pond. The average effluent concentration, Ci, of the substance in each 1000 m3 of runoff will be calculated by the program. Then using the equation Ci ˆ C0e

kt

(where k is known and constant, C0 is the influent concentration, and Ci is the concentration of the ith pond effluent 1000 m3 volume), the time that each volume has stayed in the pond will be calculated by tˆ

1 Ci ln k C0

Statistical analyses of the times will provide the mean, standard deviation, and cumula­ tive residence time distribution of runoff volume. What is the error in the engineer’s reasoning? What could be done to obtain the correct result? It is assumed that the pond is not mixed between events. Does it matter if the pond is CM or PF during a runoff event? 10.32 A system with three treatment units has an influent flow, Q, of 3800 m3 d 1. The volumes of the first and second basins are 950 and 500 m3, respectively. Flow from the

10 Mass Balances and Hydraulic Flow Regimes

second basin is recycled at a rate of 0.50Q to the first basin. Also 20% of the flow coming into the second basin is sent to the third basin. Seventy percent of the flow from the third basin is returned to the first basin. See the sketch above. What are the detention times in the first and second basins based on the total flow entering them? What volume is required in the third basin to provide a detention time of 15 days based on the total flow entering it? 10.33 Perform another iteration in Example 10.4 to find all the quantities asked for in the example. 10.34 Assuming that the solids in the underflow from the secondary sedimentation basin split in proportion to the flow, what is the concentration of Fe in the aeration basin for Example 10.4? Use the flows in the table at the end of the example. Assume that the Fe concentrations in the supernatant from the thickener and centrate from the centrifuge are insignificant and that all Fe forms precipitate and settle in the secondary clarifier. All Fe in the influent to the process is removed in the primary clarifier. 10.35 Perform a mass balance for a water treatment plant with the treatment sequence indicated in Figure 9.1. Ignore screenings removed from the bar rack and traveling screen. Use the following information. Influent flow – 235 ML d 1, influent SS – 5.0 mg L 1. Assume that the s.g. of all chemical and other solids is 1.00. Alum is the coagulant added to achieve a concentration of 22 mg L 1 Al2(SO4)3. The concentration of the alum feed is 10% Al2(SO4)3
14H2O. There is sufficient alkalinity in the water to react with the alum and form the precipitate Al(OH)3, which does not hydrate. The sedimentation basin removes 80% of the SS. The underflow from the sedimentation basin contains SS at 0.55%. The remainder of the suspended particles are removed in the rapid sand filter. The filter backwash uses 2% of the product water from the filter. The filter backwash water is returned ahead of the mixing unit. Assume that the backwash flow rate is continuous. Chlorine is added at 1.0 mg L 1 to the effluent before discharge to the clear well. Fluoride is added as NaF from a dry feeder to achieve a concentration of 1.0 mg F L 1 in the effluent before discharge to the clear well. The vacuum filter achieves a cake content of 35% solids by weight. The vacuum captures 95% of the influent solids. A polymer is added to the influent sludge at a dose rate of 5 kg polymer/tonne of solids. The concentration of polymer in the polymer feed is 90 g L 1. Determine the flows and SS concentrations before and after each unit. Also determine the mass feed rates of chlorine and sodium fluoride to be added on a daily basis.

References Anderson, G.K., Campos, C.M.M., Chernicharo, C.A.L., and Smith, L.C. (1991). Evaluation of the inhibitory effects of lithium when used as a tracer for anaerobic digesters. Water Res. 25 (7): 755–760. doi: 10.1016/0043-1354(91)90154-I.

271

272

Theory and Practice of Water and Wastewater Treatment

Eckenfelder, W.W. Jr. and Ford, D.L. (1970). Water Pollution Control. New York: Pemberton Press. Jiminez, B., Noyola, A., Capdeville, B. et al. (1988). Dextran blue colorant as a reliable tracer in submerged filters. Water Res. 22 (10): 1253–1257. doi: 10.1016/0043-1354(88)90112-1. Lawler, D.F. and Singer, P.C. (1993). Analyzing disinfection kinetics and reactor design: a conceptual approach versus the SWTR. J. Am. Water Works Assoc. 85 (11): 67–76. http://www. jstor.org/stable/41294432. Levenspiel, O. (1999). Chemical Reaction Engineering, 3e. New York: Wiley. USEPA (1974). Process Design Manual for Upgrading Existing Wastewater Treatment Plants. Cincinnati, OH: Center for Environmental Research Information.

SECTION1V

05/09/2018

13:22:9

Page 273

273

Section IV Physical–Chemical Treatment Processes

SECTION1V

05/09/2018

13:22:9

Page 274

275

11 Screening and Sedimentation Screening and sedimentation are inexpensive physical processes that are widely incorporated into treatment operations for water, wastewater, and stormwater runoff. The basic laws of physics and fluid mechanics govern the processes.

11.1 Screens and Bar Racks Screens (Figure 11.1) and bar racks are located at intakes from rivers, lakes, and reservoirs for water treatment plants or at the wet well (Figure 11.2) into which the main trunk sewer discharges for a wastewater treatment plant. They are also located before pumps in stormwater and wastewater pumping stations. They are almost always provided at these locations. These devices remove coarse debris (such as rags, solids, and sticks), which may damage pumps or clog downstream pipes and channels. The WEF and ASCE manual (2010) categorizes screens and racks according to their openings as follows: Trash racks (or bar racks) and by-pass screens have openings greater than 36 mm, coarse screens between 6 and 36 mm, fine screens 0.5–6 mm, and microscreens (also called microstainers) 10 μm to 0.5 mm. To prevent the settling of coarse matter, the velocity in the approach channel to the screens should not be less than 0.4 m s 1 at low flow and at least 0.76 m s 1 at peak flow (WEF and ASCE 2010). The ratio of the depth to width in the approach channel ranges from 1 to 2. Coarse screens may be installed on an incline to facilitate the removal of debris. The headloss through the screens is a function of the flow velocity and the openings in the screens. A sketch of the water profile through a screen is given in Figure 11.1. Bernoulli’s equation [Eq. (11.1)] is used to analyze the headloss. h1 ‡

v2 v2 ˆ h2 ‡ sc ‡ losses 2g 2g

(11.1)

where g is the acceleration of gravity, h1 is the upstream depth of flow, h2 is the downstream depth of flow, v is the upstream velocity, and vsc is the velocity of flow through the screen. The losses are usually incorporated into a coefficient, Cd: Δh ˆ h1

h2 ˆ

1 v2 2gC 2d sc

v2

(11.2)

where Cd is a discharge coefficient (typical value = 0.84).

Theory and Practice of Water and Wastewater Treatment, Second Edition. Ronald L. Droste and Ronald L. Gehr.

 2019 John Wiley & Sons Inc. Published 2019 by John Wiley & Sons Inc.

Companion website: www.wiley.com/go/droste/water

276

Theory and Practice of Water and Wastewater Treatment

Figure 11.1 Water profile through a screen.

Alternatively, an orifice equation [Eq. (11.3)] is often applied to the velocity through the screen. Δh ˆ

v2sc 1 Q 2 ˆ 2g C A 2gC d d

2

(11.3)

where A is the area of the openings and Q is the volumetric flow rate. The value of Cd in Eq. (11.3) will be supplied by the manufacturer or can be obtained from experimentation. The design value for the area is the open area of the screens for mechanically cleaned screens. The value of the discharge coefficient supplied by a manufacturer may take into account the open area of the screens. The open area of a screen may be considerably reduced by the space taken by the mesh. If the screens are to be manually cleaned, the open area should be taken as 50% of the open area (the half-clogged condition). The headloss is estimated at the maximum flow condition. Screens or racks may be cleaned by hand or automatically. Screenings are generally nonputrescible, but at a wastewater treatment plant, screens with openings less than 12.5 mm will collect some putrescible matter. Screenings are generally hauled away to an incinerator or landfill disposal site, but Wid and Horan (2016) were able to digest screenings in a mesophilic anaerobic digester that produced 355 m3 of methane/kg volatile solids in the screenings. 11.1.1 Screens for Water Treatment Plants When raw water is being withdrawn from the surface of a river, coarse screens (75 mm or larger) are installed to prevent the intake of small logs or other floating debris. For a submerged intake from a reservoir or lake, smaller coarse screens can be used. Screens at these intakes are, in general, not mechanically cleaned. Screens at the treatment plant that have smaller openings usually follow these coarse screens. The treatment plant screens may be mechanically or manually cleaned depending on the size of the operation. Quantities of screenings collected at water treatment installations are highly variable depend­ ing on the opening of the screens and the raw water source. Screenings may be washed back into

Figure 11.2 Bar rack. Source: Hubert Technology.

Reproduced with permission of Huber Technology.

11 Screening and Sedimentation

the water source. For example, the primary water treatment installation for the city of Ottawa, Ontario, which draws water from the Ottawa River, has mechanically cleaned traveling screens with openings of 1.3 mm and the screenings are washed back into the Ottawa River. A second water treatment facility in Ottawa has manually cleaned 1.3 mm opening screens that collect an average volume of 0.29 L/1000 m3 on an annual basis. Quantities collected during the spring freshet are much higher than at other times during the year. 11.1.2 Screens at Wastewater Treatment Plants Rags, sanitary products, plastics, sticks (particularly from combined systems), vegetable cuttings, and paper are among the coarse materials found in wastewater arriving at the treatment plant. Trash racks were used in older plants and continue to be used in plants treating wastewater from combined sewers (WEF and ASCE 2010). Coarse screens are the most common stand­ alone screens, but modern plants generally follow these by fine screens or even microscreens, especially if the plant uses membrane bioreactors for subsequent biological treatment. Table 11.1 provides information on screenings collected as a function of opening size. There are many factors that affect the quantities of screenings. The per capita production of screenings in the United States, United Kingdom, and France was calculated to be up to 6, 1.4–9.6, and 6–23 g/cap, respectively (Cadavid-Rodriguez and Horan 2012). Combined sewer areas can produce several times the amount of screenings collected from separate sewered areas during storm flows. The peak daily collection can vary as high as 20 : 1 on an hourly basis. 11.1.3 Microstrainers Microstrainers (Figure 11.3) have been used to reduce suspended solids (SS) for raw waters that contain high concentrations of algae or to further reduce SS in effluent from secondary clarifiers following biological wastewater treatment. The microstrainer is made of a very fine fabric or Table 11.1 Screenings characteristics. Opening (mm)

Moisture content (%)

Specific weight (kg m − 3)

2.0

Volume of screenings (L/1000 m3) Range

Typical

43–89

57

6.0

60–90

700–1100

51–100

67

9.0

80–90

900–1100

37–85

60

12.5

60–90

700–1100

37–74

50

25.0

50–80

600–1000

15–56

39

37.5

50–80

600–1000

7–56

34

50.0

50–80

600–1000

4–56

32

80–90

640–1100

3.5–35

640–1100

3.5–84

Separate sewer system 50–50

Combined sewer system 50–50

80–90

Source: Adapted from WEF and ASCE (2010) and Metcalf and Eddy:AECOM (2014).

277

278

Theory and Practice of Water and Wastewater Treatment

Figure 11.3 Microstrainer. Source: Hubert Technology. Reproduced with permission of Huber Technology.

screen wound around a drum. The drum is typically 75% submerged and rotated with water commonly flowing from the inside to the outside of the drum. In some strainers, flow moves from the perimeter to the center. The solids deposit is removed by water jets that can be activated by a pressure differential across the screen. The water jets are directed at the exposed drum surface and collected in a channel under the top of the drum. Openings in microstrainers vary from 20 to 60 μm. SS will be removed, but bacteria will not be removed to any significant extent. To minimize slime growth that will cause high headloss, ultraviolet light may be applied to the strainer. The headloss performance of a microstrainer is evaluated by a semi-empirical equation. It is observed that the headloss is directly proportional to the flow rate, degree of clogging, and time, and inversely proportional to the surface area (A) of the strainer. These parameters are incorporated into a first-order relation: Q dhL ˆ k hL dt A

(11.4)

where k is a characteristic loss coefficient and t is time. The above equation integrates to: Q

hL ˆ hL0 ek At

(11.5)

where hL0 is the headloss of the clean strainer. The loss coefficient for the strainer should be experimentally determined. Typical design parameters for solids removal from secondary effluents are given in Table 11.2. USEPA (1975) surveyed a number of microstrainers treating secondary effluent with solids concentrations in the range of 6–65 mg L 1 and found average removals from 43% to 85%. Microstrainers also find application in the treatment of stormwater runoff and the polishing of effluents (removal of algae) from stabilization pond systems.

11.2 Sedimentation Sedimentation is the physical separation of SS from a water by the action of gravity. It is a common operation for water treatment and found in almost all wastewater treatment plants. It is less costly than many other treatment operations.

11 Screening and Sedimentation

Table 11.2 Microstrainer design parameters. Item

Typical value

Screen mesh

20–35 μm

Submergence

75% of height (66% of area)

Hydraulic loading

180–360 m3/m2-h of submerged drum surface area

Headloss through screen (hL)

7.5–15 cm

Max. hL

30–45 cma)

Peripheral drum speed

4.5 m min

1

at 7.5 cm hL

40–45 m min

1

at 15 cm hL

Typical drum diameter

2.5–5 m

Washwater flow

2% of throughput at 345 kPa 5% of throughput at 100 kPa

a) Typical designs provide an overflow to bypass part of the flow when hL exceeds 15–20 cm. Source: Adapted from USEPA (1975) and Metcalf and Eddy (2003).

11.2.1 Particle Settling Velocity Before a basin to settle particles is designed, settling velocities of the particles must be known. The physical characteristics of a particle determine its settling velocity. Consider a particle falling in a body of fluid with the following assumptions: 1) 2) 3) 4) 5)

The particle is discrete and its size and shape do not change.

Infinite size vessel.

Viscous fluid.

Single particle.

Quiescent fluid.

The forces acting on the particle (Figure 11.4) are the effective gravitational force and the drag force, FD, caused by fluid resistance. The effective gravitational force (downward) is the difference between the gravitational force, Fg, and the buoyant force, Fb: Fn ˆ Fg

F b ˆ ρp

ρ gV p

(11.6)

where Fn is the net downward force, ρp is the density of the particle, ρ is the density of fluid (water in this case), and Vp is the volume of the particle. The drag force (FD) can be found from dimensional analysis to be F D ˆ ½ρC D Ap v2

(11.7)

where Ap is the cross-sectional area of the particle, CD is the drag coefficient, and v is the settling velocity of the particle.

Figure 11.4 Particle force balance.

279

280

Theory and Practice of Water and Wastewater Treatment

The force balance applying to the particle while it is accelerating is m~ a ˆ~ F g ‡~ F b ‡~ FD

(11.8)

where a is the rate of acceleration of the particle, m is the mass of the particle, and the arrows represent vector quantities. Removing the vector notation from this equation and substituting for the forces in the vertical direction result in the governing differential equation. ρ p V p a ˆ ρp V p

dv ˆ dt

ρp

ρ gV p ‡ 1

2 ρC D Ap v

2

(11.9)

The settling velocity increases in a very short time from 0 to a constant ultimate settling velocity (see Problem 11). Taking a force balance after the ultimate settling velocity has been reached (a = 0): ρp

ρ gV p ˆ ½ρC D Ap v2

(11.10)

Solving for v: vˆ

2g V p ρp ρ ρ C D Ap

(11.11)

For a spherical particle of diameter, d, vˆ

4 gd ρp ρ ρ 3 CD

(11.12)

The drag coefficient, CD, is not constant, but varies with the Reynolds number for the particle, Re, and with the shape of the particle. ρvd (11.13) μ For spherical particles, the following relations apply: 24 Re < 1 : C D ˆ (11.14a) Re Re < 1 is the laminar range also known as the Stokes’ range. The next range is the transition between laminar and turbulent settling. 24 3 18:5 ‡ 0:5 ‡ 0:34 or C D  0:6 (11.14b) 1 < Re < 103 : C D ˆ Re Re Re For fully developed turbulent settling: Re ˆ

Re > 103 :

C D ˆ 0:34−0:40

(11.14c)

CD varies with the effective resistance area per unit volume of the particle as shown in Figure 11.5.

Figure 11.5 Variation of CD with particle geometry.

11 Screening and Sedimentation

11.3 Grit Chambers Grit chambers are sedimentation basins placed at the front of wastewater treatment plants to remove sand, egg shells, coffee grounds, and other nonputrescible materials that may clog channels or cause abrasive wear of pumps and other devices. Because the grit material is nonputrescible, no further treatment is required before disposal at the ultimate site. The grit is collected in containers or directly in truck beds and hauled away at required intervals. Grit is defined as sand, gravel, or other mineral matter that has a nominal diameter of 0.15–0.20 mm or larger. Actually, grit will also include smaller mineral particles that may settle as well as organic matter such as rags, coffee grounds, vegetable cuttings, ash, clinker, wood pieces, and tea leaves. Even though some of the grit components such as coffee grounds and vegetable cuttings are organic, they are essentially nonbiodegradable over time spans for grit collection and disposal, but the grit-forming combination is frequently responsible for the strongest odors at treatment plants. The quality and quantity of grit in the sewage determine the design factors and choice of grit removal method. The amount of grit collected is a function of the removal device, its operation, and the quantity of grit in the sewage; therefore, the amount varies over a wide range. Table 11.3 gives typical values for grit quantities. As noted in the table, the quantities of grit are higher during storm events (compare separate versus combined system quantities). Grit solids content varies from 35 to 80% and volatile content from 1 to 55% (USEPA 1979). Grit that is washed should achieve a solids content of 70–80% with a minimum of putrescible matter. The bulk density of grit is from 1450 to 1750 kg m 3. Generally, grit chambers are designed to remove all particles with a nominal diameter of 0.20 mm (particles retained on a 65-mesh screen) or larger and with a specific gravity (s.g.) of 2.65 (sand) (Camp 1942). The settling velocity of these particles at 10 °C is in the order of 0.02 m s 1 based on curves of sewage grit settling velocities given by Camp (1942). Using Eq. (11.12), the settling velocity of a particle with these characteristics can be calculated to be near 0.02 m s 1, but the angularity of typical grit particles causes a small deviation from the calculated value. Also not all “grit” particles have an s.g. of 2.65 (see Figure 11.6). Sometimes, grit removal devices are designed to remove 0.15 mm sand particles (retained on a 100-mesh screen) with a settling velocity of 0.015 cm s 1 taken from Camp’s curve. It is not desirable to remove organic matter in grit chambers because no further treatment of the grit is necessary or provided; however, it is unavoidable unless the grit is washed separately after collection. The chamber must be designed to scour the lighter organic particles while the heavier grit particles remain settled. Different types of devices can accomplish this: (i) constant velocity horizontal flow channels; (ii) rectangular grit chambers with a grit washing device; and (iii) aerated grit chambers.

Table 11.3 Estimated grit quantities. Type of system

Average quantity of grit (typical range) m3/1000 m3

Separate

0.004–0.037

1.5 to 3 : 1

Combined

0.004–0.18

3 to 15 : 1

Source: Adapted from USEPA (1979, 1987).

Ratios of maximum day to average day

281

282

Theory and Practice of Water and Wastewater Treatment

Figure 11.6 Field measured settling velocities of grit. Source: Hydro International. Reproduced with permission of Hydro International.

11.3.1 Horizontal Flow Grit Chambers Horizontal flow grit chambers are open channels with detention times sufficient to allow design particles to settle and are designed to maintain a constant velocity sufficient to scour organics. A weir at the end of the channel, or the channel geometry itself, provides velocity control. These types of grit removal devices are more commonly found in older installations; they remain an appropriate hydraulic design (fewer mechanical parts) for developing countries. They may be manually cleaned or have mechanical sludge (grit) scrapers installed in them. The Camp–Shields equation (Camp 1942) is used to estimate the velocity of scour necessary to resuspend settled organics. vs ˆ

8kgd ρp ρ ρ f

(11.15)

where d is the nominal diameter of the particle, f is the Darcy–Weisbach friction factor, k is an empirically determined constant, and vs is the velocity of scour. The above equation is based on a tractive force analysis. The constant k is related to the “stickiness” of the organic material. The values of f and k were found by Camp (1942) to be 0.03 and 0.04–0.06, respectively, for domestic sewage in grit chambers. The higher value of k is most appropriate for design. It is common practice to maintain a horizontal flow velocity between 0.15 and 0.30 m s 1 to keep organic particles in suspension. A velocity of 0.23 m s 1 will scour 0.2-mm-diameter quartz particles or organic particles with a specific gravity of 1.05 up to a nominal diameter of 6.5 mm. The problem in designing a horizontal flow chamber is to maintain the constant velocity in the channel, although varying quantities of flow are passed through the chamber over a 24-hours

11 Screening and Sedimentation

period. If the channel is rectangular and discharges over a rectangular weir, the discharge relation based on Bernoulli’s equation is p 3=2 Q ˆ C d A 2gH ˆ CwH (11.16) p where A is the cross-sectional area of the weir, C is equal to C d 2g, Cd is the discharge coefficient, H is the depth of flow above the weir, and w is the width of the weir. To illustrate the general problem, it will be assumed that the channel width and depth are equal to w and H, respectively; any variations in w and H between the weir and the channel can easily be incorporated into a specific solution, but they will not change the nature of the problem of velocity variation. The horizontal velocity, vh, is related to the discharge rate and channel geometry by vh ˆ

Q Q Q ˆ ˆ CH ½ ˆ C A wH Cw

⅓

(11.17)

If the ratio of the maximum to the minimum volumetric flow (Qmax:Qmin) is 5 : 1, which is common, the ratio of the corresponding maximum and minimum horizontal velocities (vhmax: vhmin) is Qmax vhmax wH max Qmax H min ˆ ˆ Qmin Qmin H max vhmin wH min ⅔ Q min

vhmax Qmax Cw ˆ ˆ ⅔ vhmin Qmin Q max Cw

Qmax Qmin

⅓

ˆ 5⅓ ˆ 1:71

where Hmax and Hmin correspond to Qmax and Qmin, respectively. Therefore, the shape of either the channel or the weir has to be modified to maintain a satisfactory flow-through velocity. Channel with Varying Cross Section

A channel with a varying cross-section geometry is an alternative to modifying the weir. The channel may discharge into a Parshall flume (Figure 11.7) to measure the discharge accurately. In this case, there is a rectangular control section at the end of the channel, but the channel width varies as a function of depth. Figure 11.8 is a definition sketch.

Figure 11.7 Parshall flume.

283

284

Theory and Practice of Water and Wastewater Treatment

Figure 11.8 Grit chamber with varying geometry.

Let wt be the width of the rectangular throat (control) section in the Parshall flume at the downstream end of the channel; wt is constant. The flow through the throat is p (11.18) Q ˆ C d wt H t 2gH t where Ht is the depth in throat. Equation (11.17) must be satisfied. From Eq. (11.17), which applies to discharge in the channel, dQ ˆ vh dA ˆ vh w dy

(11.19)

Differentiating Eq. (11.18),

3p ½ 2gwt y dy dQ ˆ C d (11.20) 2 A function for w must be found using these relations. Setting Eq. (11.19) equal to Eq. (11.20), 3p ½ vh w dy ˆ C d 2gwt y dy (11.21) 2 and solving for w, p 3C d 2g ½ (11.22) wˆ wt y 2vh The equation for the channel geometry describes a parabolic section. In this case, because y is dQ synonymous with H, it is found that dQ dy ˆ dH ˆ vh w from Eqs. (11.20) and (11.21). Design Notes for a Parabolic Grit Chamber

In actual practice, the parabolic shape is approximated with a trapezoid to reduce construction costs. As above, one channel and a bypass or two or more channels should be installed. When the number of channels is determined, the maximum, average, and minimum flows in a channel can be calculated. When one channel is out of service, the flow to this channel will be diverted to the other channels, resulting in an emergency flow, Qemerg. The emergency flow in a channel should be based on the maximum flow into the set of grit chambers with one channel out of service. These flows, Qemerg, Qmax, Qave, and Qmin, are used to design the shape and length of a channel. The procedure to determine the trapezoidal shape of the channel is first to choose the horizontal velocity in the chamber. The maximum horizontal velocity is usually 0.30 m s 1 as noted above. Using Eqs. (11.18) and (11.22), the discharge is related to the width, x, of the channel at a depth y by Eq. (11.23a). The equation is then rearranged to find the depth as a function of the other parameters [Eq. (11.23b)]. 2 (11.23a) Q ˆ xvh y 3 3 Q yˆ (11.23b) 2 xvh because Q ˆ vh A: where A is the cross-sectional area of flow.

11 Screening and Sedimentation

For a parabola, from Eq. (11.23a): A ˆ 23 xy. Qmax is used for Q and a value for xmax is chosen to solve Eq. (11.23b) for ymax. The width, xmax, should be in the range of 1–1.25 ymax to keep the slope of the trapezoid sidewalls sufficiently steep to minimize debris accumulation on them. The depth and width of the downstream control section can now be determined. Critical flow conditions exist in the throat of the Parshall flume. The specific energy relation that applies is y‡

v2h v2 v2 ˆ dc ‡ c ‡ 0:1 c 2g 2g 2g

(11.24)

where dc is the depth of flow in the throat (equivalent to Ht) and vc is the velocity of flow in the throat. The factor 0.1 in Eq. (11.24) is a reasonable value (10%) to gauge the headloss for a wellrounded smooth approach to the throat. At the critical depth, the Froude number (Fr) is equal to 1. The critical depth occurs at the throat entrance. vc ˆ1 (11.25a) Fr ˆ p gd c and dc ˆ

v2c g

(11.25b)

Substituting for dc in Eq. (11.24), y‡

v2h v2c v2c v2 v2 ˆ ‡ ‡ 0:1 c ˆ 3:1 c 2g g 2g 2g 2g

(11.26)

and this equation may be solved for vc. v2 2g y‡ h vc ˆ 3:1 2g

=2

1

(11.27)

The depth, dc, may now be found from Eq. (11.25a), and the following equations are used to find the width of the throat, wt. Q ˆ wt dc vc Q wt ˆ vc d c At ˆ

(11.28a) (11.28b)

where At is the cross-sectional area of flow in the throat. Now that the throat width is determined, finding the width of the channel at the other discharge rates is facilitated. The area of the throat is related to vc by the following equations. wt d c ˆ At ˆ

v2c wt g

(11.29a)

and vc ˆ

Q At

(11.29b)

285

286

Theory and Practice of Water and Wastewater Treatment

and 2

Q At

At ˆ

wt (11.30a)

g

and At ˆ

3

Q2 wt g

(11.30b)

Equation (11.30b) is used to find the throat area at the other flows. Equations (11.29a) and (11.29b) are used to find dc and vc, respectively, and y is then found from Eq. (11.24). Equation (11.23a) is finally used to find the width of the trapezoid at Qave and Qmin. There should always be two channels or one channel with a bypass installed to allow for a channel being taken out of service for repair and maintenance. A freeboard of 30–45 cm should be designed for the channel. For best flow distribution, the inlet pipe to the chamber should be located on the centerline. The flow velocity will not be constant across the channel cross section; the velocity will be greater at the bottom of the channel. Water depths in horizontal flow chambers are typically in the range of 0.6–1.5 m (WEF and ASCE 2010). Lengths range from 3 to 25 m. The length of the chamber is increased beyond the theoretical value by a factor of 10–50% to account for nonidealities in the flow pattern and settling of the particles. Example 11.1 Parabolic Grit Chamber Design Design a horizontal flow grit chamber with a varying cross-sectional area for a plant with an average flow of 5.00 × 104 m3 d 1, minimum flow of 2.50 × 104 m3 d 1, and maximum flow of 12.0 × 104 m3 d 1. Use three channels in the grit chamber that should remove 65-mesh particles. Each channel should handle a flow of 6.00 × 104 m3 d 1 to allow maintenance of one channel. The flow-through velocity (vh) will be 0.30 m s 1. Size the grit storage area for an accumulation of 35 m3/106 m3 sewage to be removed manually once per week. Qmax ˆ 12:0  104 m3 d

1

=3 ˆ 4:00  104 m3 d

1

ˆ 0:463 m3 s

1

Qave ˆ 5:00  104 m3 d

1

=3 ˆ 1:67  104 m3 d

1

ˆ 0:193 m3 s

Qmin ˆ 2:50  104 m3 d

1

=3 ˆ 8:33  103 m3 d

1

ˆ 0:0965 m3 s

Qemerg ˆ 6:00  104 m3 d

1

ˆ 0:694 m3 s

1 1

1

The design settling velocity, vs, of a 65-mesh particle is 1.38 m min 1 or 0.023 m s 1. Use a fixed-width control section with a smooth approach. Assume losses equal 10% of velocity head in the throat.

Governing Equations In the channel, Q = vA. In the rectangular throat control section, Q = CwtH3/2.

11 Screening and Sedimentation

From Eq. (11.22), the width in the channel varies according to wˆ

3 Cwt ½ 3 Cwt ½ H y ˆ 2 v 2 v

Q ˆ Cwt H

3=2

w ˆ T …top width†

2 2

Q ˆ vwH ˆ vTH 3 3

Qmax The top width T should be equal to 1–1.25Hmax. Also the width will be affected by the Parshall flume geometry that is chosen. Using a value of 1.20Hmax (at Qmax), Hmax is determined from 2 Q ˆ vTH 3

2 Q ˆ v…1:20H†H 3

3Q ˆ 2:40v

Hˆ

3…0:463 m3 s 1 † ˆ 1:39 m 2:40…0:30 m s 1 †

T ˆ …1:20†…1:39 m† ˆ 1:67 m

The top width will be fixed at 1.65 m. Upstream energy ˆ downstream energy ‡ losses H‡

v2 v2 v2 ˆ dc ‡ c ‡ 0:1 c 2g 2g 2g

H‡

v2 v2 ˆ 3:1 c 2g 2g

2 9:81 m s vc ˆ 3:1

vc ˆ 2

dc ˆ

v2

c

g

2g v2 H‡ 3:1 2g 2

0:30 m s 1 1:39 m‡ 2…9:81 m s 2 †

½ ˆ 2:97 m s

1

2

dc ˆ

2:97 m s 1 v2c ˆ ˆ 0:90 m g 9:81 m s 2

The area of the throat is the area at the critical flow condition that occurs within it at the specified volumetric flow rate. At ˆ

Q 0:463 m3 s 1 ˆ ˆ 0:156 m2 2:97 m s 1 vc

wt ˆ

0:156 m2 ˆ 0:17 m 0:90 m

vc ˆ

Q Ac

Qave

At ˆ

3

At ˆ dc wt

dc ˆ

vc2 g

At ˆ

…0:193†2 …0:17† ˆ 0:086 m2 9:81

3

Q2 wt g

287

288

Theory and Practice of Water and Wastewater Treatment

At 0:086 ˆ ˆ 0:51 m wt 0:17

dc ˆ

H ˆ 3:1 Tˆ

v2c 2g

vc ˆ

v2 …2:24†2 ˆ 3:1 2g 2…9:81†

0:193 ˆ 2:24 m s 0:086

1

…0:30†2 ˆ 0:79 m 2…9:81†

3 …0:193† ˆ 1:22 m 2 …0:79†…0:30†

Qmin 3

At ˆ

…0:0965†2 …0:17† ˆ 0:054 m2 9:81

0:054 ˆ 0:32 m 0:17

dc ˆ

H ˆ 3:1 Tˆ

…1:79†2 2…9:81†

vc ˆ

0:0965 ˆ 1:79 m s 0:054

1

…0:30†2 ˆ 0:50 m 2…9:81†

3 …0:0965† ˆ 0:96 m 2 …0:30†…0:501†

Qemerg 3

At ˆ

…0:694†2 …0:17† ˆ 0:21 m2 9:81

0:21 0:694 ˆ 1:23 m vc ˆ ˆ 3:31 m s 0:17 0:21

dc ˆ

H ˆ 3:1 Tˆ

…3:31†2 2…9:81†

1

…0:30†2 ˆ 1:73 m 2…9:81†

3 …0:694† ˆ 2:01 m 2 …0:30†…1:73†

Because T is larger than T assumed (1.65 m), either T must be adjusted or the velocity will vary from 0.30 m s 1. Because emergency flow situations occur rarely, allow the velocity to vary. To find the velocity at emergency flow, vˆ p

3 Q 2 HT

3:1v2c

3:1v2c

H‡ 2gH ˆ

2gH ˆ

…3:1†…3:31†2 H 2 34:0H 2 vˆ

p v2 v2 ˆ 3:1 c ) v ˆ 3:1v2c 2g 2g

2gH

3 Q 2 HT 3Q 2T

2

1 H

2

3 …0:694† 2 …1:65†

2…9:81†H 3 ˆ

19:6H 3 ˆ 0:398 ) H ˆ 1:73 m

3 …0:694† ˆ 0:36 m s 2 …1:73†…1:65†

2gH 3 ˆ

3:1v2c H 2

1

2

3Q 2T

2

11 Screening and Sedimentation

The velocity variation (20%) is tolerable.

Design of channel for grit chamber example.

Length of the Tank Since vh is constant, depth in the grit chamber and the time for sedimentation will increase with flow rate. The maximum flow condition should be used to establish the length of the tank. At Qmax, H = 1.39 m. Lˆ

H 1:39 m vh ˆ vs 0:023 m s

1

 0:30 m s

1

ˆ 18:1 m

For Qemerg, Lˆ

1:73 m 0:023 m s

1

 0:36 m s

1

ˆ 27:1 m

Use L = 27.0 m to provide a 50% safety factor at the maximum flow condition. Field studies should be performed to establish the required safety factor.

Grit Accumulation Base the design on the average flow for one channel. For 1 week: Qt = (1.67 × 104 m3 d 1)(7 d) = 1.17 × 105 m3. 35 m3 The grit accumulation is 6 3 1:17  105 m3 ˆ 4:10 m3 . 10 m 4:10 m3 The area required for grit storage is = 0.152 m2. 27:0 m Check the width at H = 0 for the required depth–width relation to accommodate the minimum and average flows. Qave : H ˆ 0:79 m;

w ˆ 1:22 m;

0:5w ˆ 0:61 m

Qmin : H ˆ 0:50 m;

w ˆ 0:96 m;

0:5w ˆ 0:48 m

289

290

Theory and Practice of Water and Wastewater Treatment

The equation of the line describing the slope of the lower portion of the grit chamber is H ˆ 2:23…0:5†w

0:57

At H ˆ 0; w ˆ 0:51 m

Using this width, the depth required for grit storage is 0:152 m2 ˆ 0:30 m 0:51 m If in the future a mechanical rake is to be installed (an expensive option for this small installation), manufacturer’s specifications should be consulted. It may be necessary to adjust the grit accumulation dimensions to fit standard equipment. Mechanical grit removal devices will result in lower accumulations of grit. Small changes in the channel geometry will not have a significant effect on the flow velocity. In any case, the channel width should be maintained as near as possible to the widths calculated for Qmin, Qave, and Qmax. The flows will be in this range most of the time. The velocity in the grit storage area is assumed to be zero because this zone has a wall at either end. See the sketch for the final design. A freeboard of 15 cm was added to the depth of the channel.

11.3.2 Aerated Grit Chambers A common approach currently used in medium-to-large plants for grit removal is aerated grit chambers (Figure 11.9). Aerated grit chambers have two main advantages over other grit removal processes. They are able to be fine-tuned to variable circumstances for optimum performance and they freshen the wastewater by the introduction of air. The latter reduces odors and improves primary clarification. In aerated grit chambers, a spiral flow pattern is induced in the sewage as it moves through the tank, by supplying air from a diffuser located on one side of the tank. The tank inlet and outlet are positioned to direct the flow perpendicular to the spiral roll pattern. The roll velocity is sufficient

Figure 11.9 Aerated grit chamber. Source: Hubert Technology. Reproduced with permission of Huber Technology.

11 Screening and Sedimentation

Table 11.4 Aerated grit chamber design information. Item

Range

Comment

Depth (m)

3.7 5

Varies widely

L:W ratio

3 : 1 to 8 : 1

W:D ratio

Dimensions

1 5

Typical W = 2D

Detention time at peak hour flow (min)

3 10

10–15 min at average flow; minimum detention time 2 3 min

Mean longitudinal velocity (m s 1)

0.20 0.30

Air supply (medium to coarse bubble diffusers) m3 min

1

1

of tank length

0.3 0.75

Distance of aerators from bottom (Ha) (m)

0.6 1.0

Alternatively, 70% of total depth

Traverse roll velocity (m s 1)

0.60 0.75

Measured at about 150 mm below the free surface

m

Typical: 0.45

Source: Compiled from WEF and ASCE (2010), Sawicki (2004), and Metcalf and Eddy:AECOM (2014).

to maintain lighter organic particles in suspension while allowing heavier grit particles to settle. The air supply must be adjustable to provide the “optimum” roll velocity for different conditions. Adjustment of the air supply is easily performed, which makes the device flexible. The WEF and ASCE (2010) design manual states that a minimum hydraulic detention time of 3 minutes at peak hourly flow rates will provide sufficient coarse grit removal. Baffle and diffuser locations and air flow rate are more important design parameters than detention time. Other design values for an aerated grit chamber are given in Table 11.4 (refer to Figure 11.10 for a definition sketch). To meet or approach the design criteria in this table, large flows (>1 m3 s 1) are required. The flow-through velocity criterion can be relaxed. Lower velocities will improve performance. The inclusion of a baffle as shown in Figure 11.10 contributes to a significant improvement in performance (Sawicki 2004). The scour or transverse velocity immediately above the grit collection zone should be sufficient to maintain grit (sand) particles with diameters less than

Figure 11.10 Definition sketch for aerated grit chamber.

291

292

Theory and Practice of Water and Wastewater Treatment

0.1 mm in suspension but allow particles of diameter greater than 0.20 mm to settle. Defining the flow-through velocity as v (Q/WD) and the maximum velocity at the bottom of the circulation zone as ubmax, uB ˆ

u2bmax ‡ v2

where uB is the magnitude of the resultant velocity at the bottom of the circulation zone. The range for uB is (Sawicki 2004). 0:076 < uB < 0:150 m s

1

Sawicki (2004) has also provided an in-depth analysis of flow circulation to determine uB. 11.3.3 Square Tank Degritter An early design for grit removal was a square, constant level, short-detention-time tank (Figure 11.11). These tanks are also known as detritus tanks. The corner zones are sloped into the central circular grit collection zone. A horizontal arm with flow deflectors rotates above the circular grit collection zone to maintain circulation within the tank. The influent enters through baffles along one side of the tank and the effluent exits over a weir on the opposite side. Because there is no flow control, the operation of these tanks is simple. These basins are designed to maintain a detention time of 1 minute or less at the design flow rate (WEF and ASCE 2010). The maximum day flow rate may be used as the design flow rate. The area of the tank is based on the grit particle size desired to be removed. A particle with a diameter of 0.83 mm (mesh size of 20) and s.g. of 2.65 at 15.5 °C can be calculated from Eq. (11.12) to have a settling velocity of 5 m min 1 equivalent to a theoretical overflow rate of 7120 m3/m2/d; for a particle diameter of 0.15 mm (mesh size of 100) and the same conditions, the settling velocity will be found to be equivalent to an overflow rate of 1320 m3/m2/d. A safety factor of 2 is generally applied to the theoretical surface overflow rate to allow for inlet and outlet turbulence in addition to the short-circuiting that will occur in the basin. Metcalf and Eddy:AECOM (2014) note that a common criterion for design is to remove all of the 70-mesh (0.21 mm diameter) particles, but some are designed more conservatively to remove all the 100-mesh (0.15 mm) particles. Given the surface overflow rates and detention time criteria, the tanks will be shallow. An additional depth of 150–250 mm is added to accommodate a raking mechanism. Hydraulic retention time of the liquid within these tanks varies directly with the flow rate. This is a disadvantage because, particularly at low flow rates, significant quantities of organic material

Figure 11.11 Square tank (detritus) degritter. Source: Walker Process Equipment. Reproduced with permission of Walker Process Equipment.

11 Screening and Sedimentation

293

Figure 11.12 Vortex grit removal device. Source: Hydro International. Reproduced with permission of Hydro International.

will be removed with the grit. It is necessary to use a grit washing device on the material collected from the detritus tank. The typical design criteria ensure that most of the grit is removed at all flow rates. 11.3.4 Vortex Grit Removal Devices There are various devices on the market that are cylindrical and introduce flow around the periphery to induce a circular, vortex flow pattern (Figure 11.12). Centrifugal force causes higher density particles to move to the outside of the tank and descend downward to a grit outflow outlet. Design information for these devices is available from the manufacturer. Aerated grit chambers and vortex grit removal devices are more commonly being installed. Field performances of various types of degritters are given in Table 11.5.

Table 11.5 Relative performance of grit removal devices. Technology

Design flow when tested (%)

Design removal efficiency at 100% of flow

Observed total % removal 150 μm and up

Observed total % removal 106 μm and up

Mechanically induced vortex

27–90

95% removal of 270 μm, s.g. 2.65

43–52

43–50

Aerated grit chamber

66–100

Unknown

35–70

32–67

65% removal of 150 μm, s.g. 2.65

Detritus tank

66

150 μm and larger, s.g. 2.65

66–71

57–68

Stacked traya)

100

95% removal of 75 μm, s.g. 2.65

91–92.5

89–90

Structured flow vortex

66–100

95% removal of 106 μm, s.g. 2.65

90–95

87–93

a) A lamella type vortex device.

Source: Hydro International. Reproduced with permission of Hydro International.

294

Theory and Practice of Water and Wastewater Treatment

Grit Washing

In some treatment facilities, a grit chamber may not be installed and grit is accumulated with sludge removed in the primary clarifiers. A cyclone degritter, similar to a vortex grit remover, is used to separate grit from the sludge in this instance and when grit is collected in detritus tanks. Sometimes, it is necessary to wash grit collected in other types of grit chambers. A cyclone degritter uses centrifugal force to separate the heavier particles of grit from the lighter ones. The cyclone degritter feed enters the unit tangentially and creates a vortex wherein the heavier particles are transported to the walls of the device. The unit is inclined and the solids travel downward while the liquid and lighter particles exit at the top. A cyclone separator requires a steady feed of grit at pressures between 34 and 140 kPa (WEF and ASCE 2010). Manufacturers supply design information. Other types of grit washing devices are available from manufacturers. Washed grit is nonputrescible and does not have a significant insect or rodent attraction.

11.4 Type I Sedimentation Type I sedimentation refers to discrete particle settling. The design of an ideal settling basin is based on the removal of all particles that have a settling velocity greater than a specified settling velocity. The work of Hazen (1904) and Camp (1945) provides the basis of sedimentation theory and basin design. 11.4.1 Theory A definition sketch for an ideal, horizontal flow, rectangular section basin is given in Figure 11.13. In this figure, H is the effective depth of the settling zone and vf is the longitudinal velocity of the water. The width of the basin is B. The settling velocities v1 and v0 apply to two different particles entering at the top of the basin. The settling velocity v2 applies to a particle entering the settling zone at a height, h, above the sludge zone. There will be dissipation of energy (turbulence) near the entrance as the flow profile through the basin is established. There is assumed to be no settling in the inlet zone. A similar phenomenon occurs at the exit side as the flow streamlines turn upward, and no settling is assumed to occur in the outlet zone. Sludge accumulates in the sludge zone, which is not part of the effective settling zone.

Figure 11.13 An ideal horizontal flow sedimentation basin.

11 Screening and Sedimentation

Other important assumptions are as follows: 1) There is uniform dispersion of water and suspended particles in the inlet zone. Therefore, the SS concentration is the same at all depths in the inlet zone. 2) Continuous flow at a constant rate (steady flow) exists. 3) Once a particle enters the sludge zone, it remains there (i.e., there is no resuspension of settled particles). 4) The flow-through period is equal to the detention time; i.e., there is no dead space or shortcircuiting in the volume above the sludge zone. 5) PF conditions exist. 6) Settling is ideal discrete particle sedimentation. 7) Particles move forward with the same velocity as the liquid. 8) There is no liquid movement in the sludge zone. The design volume must be related to the influent flow rate and the particle settling velocity. The particle that takes the longest time to remove will be one that enters at the top of the effective settling zone. The design settling velocity is v0, which is the settling velocity of the particle that settles through the total effective depth of the tank in the theoretical detention (hydraulic retention) time, θd. The flow-through velocity is vf. V Q

Q

vf ˆ BH θd ˆ

(11.31) (11.32)

Because the particle must travel the length and depth of the basin in the time θd, v0 θ d ˆ H

(11.33a)

vf θ d ˆ L

(11.33b)

Substituting Eq. (11.33b) into Eq. (11.33a) and using Eq. (11.32): L H ˆ v f v0

and

v0 ˆ vf

H L

or

v0 ˆ

QH Q ˆ BHL BL

(11.34)

The surface overflow rate, Q/As (As is the surface area of the basin), is defined by

v0 ˆ

Q Q

ˆ BL As

(11.35)

This shows that the sedimentation basin design is independent of the depth and depends only on the surface overflow or loading rate (Q/As) for particles with a specified settling velocity v0, provided that all the assumptions are met. From this, it follows that the sedimentation efficiency is also theoretically independent of detention time in the basin. This fact is not a mathematical curiosity. Consider a basin with the flow introduced at the bottom of the basin and uniformly distributed across the plan (surface) area, resulting in an upflow velocity v0. Any particle with a settling velocity greater than v0 will be removed (settled) after being introduced into the basin regardless of the residence time of the water in the basin. Similarly, any particle with a settling velocity less than v0 will eventually exit with the effluent overflowing from the basin. Horizontal (or radial) flow and upflow are the two possible operational modes of a settling basin. In either case, all particles with settling velocities greater than v0 will be removed. In the horizontal flow mode, some particles with settling velocities less than v0 will also be removed if they enter the basin at a depth less than H. Assumption 1 above is critical to analysis of the total

295

296

Theory and Practice of Water and Wastewater Treatment

Figure 11.14 Settling velocity curve for a suspension, where p is the weight fraction of particles with settling velocity less than stated velocity.

removal. Assume that a particle with settling velocity, v, which is less than v0, will travel a vertical distance h in time θd. h ˆ vθd

(11.36)

All particles with settling velocity v that enter at a depth h or lower will be removed. The criterion for removal of particles with this settling velocity is h v  (11.37) H v0 Because all particles with settling velocity v are uniformly distributed throughout the inlet depth (assumption 1), the fraction removal, r, of particles with this settling velocity is h v ˆ (11.38) H v0 The settling velocity distribution for a suspension of different sizes (or densities) can be determined from a column settling test as described in Section 11.5. The results of the test provide data to construct a plot as shown in Figure 11.14, which is a cumulative settling velocity frequency distribution. For simplicity, consider in this analysis that all particles have the same density. Let pi be the fraction of particles having the same size (and weight). The total removal R (fraction by weight) of all particles is the sum of the fractional removals of each fraction of particles, pi. Applying Eq. (11.38) to each fractional interval Δp, rˆ

Rˆ1 Rˆ1

p0 ‡ Σr i v0 ‡ v1 p0 ‡ p0 2v0

p1 ‡

v1 ‡ v2 p1 2v0

p2 ‡ ∙ ∙ ∙ ‡

vi ‡ vi‡1 pi 2v0

pi‡1 ‡ ∙ ∙ ∙ (11.39a)

which in the limit is po

Rˆ 1

p0

1 v dp ‡ v0 ∫

(11.39b)

0

where p0 is the fraction of particles by weight with a settling velocity equal to or less than v0. A polynomial fit can be applied to data used to construct the curve in Figure 11.14 to use Eq. (11.39b). For circular tanks with an inlet in the middle and radial flow under the same assumptions as above, it can also be shown that Q h v v0 ˆ ; r ˆ ˆ H v0 As and the same overall removal expressions [Eqs. (11.39a) and (11.39b)] apply (see Problem 8).

11 Screening and Sedimentation

Equation (11.35) also holds for vertical flow tanks. Particles with a settling velocity less than the upflow velocity are entrained in the upward flow and are washed out of the system. The s.g. of inorganic particles is near 2.65 and the s.g.s of organic particles are normally in the range of 1.001–1.01. For mixed suspensions, the overall removal will be the weighted sum of the removals of each fraction. Under conditions of a typical settling test, it may be assumed that the organic matter removal is about 5%. The total suspended solids–volatile suspended solids (TSS– VSS) analysis may be used to discriminate between inorganic and organic matter. When inorganic solids predominate, Eq. (11.12) can be used to calculate the nominal diameters of the fractions from the settling velocity distribution. Theory dictates that increasing the surface area of a settling basin will improve its performance. Lamella and tube clarifiers, discussed in Section 11.6, exploit concepts from basic theory, resulting in the design of clarifiers with very high loading rates.

11.5 Type II Sedimentation Under quiescent conditions, suspended particles in many waters exhibit a natural tendency to agglomerate, or the addition of chemical agents promotes this tendency. This phenomenon is known as flocculent or type II sedimentation. Analysis of type II sedimentation proceeds from the principles of type I sedimentation. As particles settle and coalesce with other particles, the sizes of particles and their settling velocities will increase. The trajectory traced by a settling particle will be curvilinear (Fig­ ure 11.15) because of the increase in its settling velocity as other particles attach to it. The instantaneous settling velocity is the tangent to the curve. The average settling velocity for the particle in Figure 11.15 is vˆ

H t1

(11.40)

The average settling velocity distribution for the suspension is continually changing with time as shown in Figure 11.16.

Figure 11.15 Settling trajectory in type II sedimentation.

Figure 11.16 Settling velocity distribution at various times.

297

298

Theory and Practice of Water and Wastewater Treatment

Figure 11.17 Settling column.

To design a basin for flocculent settling, the average settling velocity distribution variation with time must be found to calculate the total removal as time (or volume of the basin) increases. At some point, an incremental increase in the volume of the basin (which increases the detention time) will not produce a significant increase in the amount of solids removed. There is no theoretical means of predicting the amount of flocculation and settling velocity distribution variation for a suspension. A laboratory analysis as described in the following section is required. 11.5.1 Laboratory Determination of Settling Velocity Distribution The water to be analyzed must have the same coagulants and other agents added that will be applied in the field situation. The suspension is mixed and added to a column that has approximately the same depth as the anticipated settling basin. Because type II sedimentation is time–depth dependent, more representative settling curves are obtained when the column depth is near the prototype basin depth. The column (Figure 11.17) is normally made of clear plastic so that one may visually observe the process. Sampling ports are uniformly spaced along the length of the column. The bottom port will provide samples that are indicative of the compaction of the settled sludge. The effective settling depth is the depth above the bottom port. The column internal diameter (i.d.) should be at least 14 cm to avoid bridging of the suspension and other wall effects. After the initial sample is taken, samples are taken from each port at uniform time intervals, noting the time and port number. The volume of samples removed from the column causes the water surface elevation to descend, which should be accounted for in processing the data. 11.5.2 Type II Sedimentation Data Analysis The analysis of data gathered as outlined in the previous section is best presented by example. The objective of the analysis is to obtain a plot of the settling trajectories for various fractions of the SS. Then the total removals at any time may be estimated. For a column with a total depth of 240 cm and sampling ports spaced at 60 cm intervals, the data in Table 11.6 have been obtained. The effective depth of the sedimentation basin under consideration is 1.8 m. The initial concentration of SS was 430 mg L 1. The first step is to convert the concentrations into percentage removals at each depth. r ti;d …%† ˆ

X0

X ti;d  100 X0

11 Screening and Sedimentation

where r is removal, X is concentration of SS, and the subscript ti,d indicates time i and depth, d, respectively. The results of these calculations are listed in Table 11.7. The desired plot of the settling trajectories of various fractions of the SS (see Figure 11.19) can be obtained by constructing a depth–time plot with percentage solids removed as a parameter. An intermediate step described in the following paragraphs improves the interpolation that is required (Ramalho 1977). A plot of the percentage solids removed at each depth versus time is constructed using the data in Table 11.7. This is done in Figure 11.18. A smooth curve is drawn between the data points for each depth. Figure 11.18 can be used to easily estimate the time required to attain a specified removal at a given depth. The times to attain a given removal at each depth are found by extending a horizontal line from the removal to the curves and dropping vertical lines at the intersections. These times are then tabulated for the removals at each depth (Table 11.8). Table 11.6 Raw data. Concentration (mg L − 1)

Time (min) 60 cm

120 cm

180 cm

5

357

387

396

10

310

346

366

20

252

299

316

30

198

254

288

40

163

230

252

50

144

196

232

60

116

179

204

75

108

143

181

Table 11.7 Percentage solids removed. Time (min)

Solids removed (%) 60 cm

120 cm

180 cm

5

17.0

10.0

7.9

10

28.0

19.5

14.9

20

41.5

30.5

26.0

30

54.0

40.9

33.0

40

62.0

46.5

41.4

50

66.5

54.4

46.0

60

73.0

58.6

52.5

75

75.0

66.7

57.9

299

300

Theory and Practice of Water and Wastewater Treatment

Figure 11.18 Percentage SS removed at each depth.

Using the data in Table 11.8, isoconcentration (or isoremoval) lines are now constructed on a depth versus time plot (Figure 11.19). The curves in Figure 11.19 trace the settling trajectories of particulate fractions of the suspension. The actual particle makeup of each fraction is continuously changing as the particles coalesce and these curves represent the gross phenomena. Now for any time, a p–v plot similar to Figure 11.14 can be made and the overall removal for the suspension can be determined in the same manner outlined for discrete particle settling. The data from Figure 11.19 can be used directly to estimate the total removal. To find the total removal at any chosen time, a vertical line is projected upward. It is most convenient to choose times that are at the end of an isoconcentration line. (Why?) For instance, when a time of 39 minutes is chosen for the above data, it is seen that 40% of the particles are completely removed; i.e., 40% of the particles had an average settling velocity greater than or equal to 180 cm ˆ 4:62 cm min 39 min

1

in the first 39 minutes of settling.

Table 11.8 Interpolated percentage solids removed. t (min)

Percentage SS removed 60 cm

120 cm

180 cm

5

1.2

2.5

3.7

10

2.5

5.0

6.5

20

6.7

11.0

14.5

30

1.7

19.0

25.0

40

18.0

30.0

39.0

50

27.0

44.0

56.5

60

38.5

61.5

77.5

70

55.0

87.5

—

75

75.0

—

—

11 Screening and Sedimentation

Figure 11.19 Isoconcentration curves.

The other fractions are partially removed. To estimate the removal of these fractions, median lines should be drawn between the isoconcentration curves. Where possible the median lines should be located based on the average depth between two isoconcentration lines at a given time. For example, at a time of 57 minutes the 50 and 60% isoconcentration lines intersect the vertical line at depths of 180 and 106 cm, respectively. The median line between these isoconcentration lines should pass through the point (57 minutes, 143 cm). The fractional removal of the 40–50% fraction (10% of the particles) is calculated by reading the depth at the intersection of the vertical and the median isoconcentration lines for this fraction (130 cm). This is the average depth that this fraction reached in 39 minutes. In a manner analogous to the discrete particle settling analysis, the average settling velocity of a fraction compared to the design settling velocity will dictate the percentage removal of the fraction. d i

vi t d d i

ˆ ˆ vo D D td

(11.41)

where di is the average depth reached by the ith fraction in time td, D is the total effective settling depth, and vi is the average settling velocity of the ith fraction. The fractional removal, ri, of the fraction, Δpi, is ri ˆ

di Δp D i

The fractional removals for the data above in a time of 39 minutes are 130 …50 180

40† ˆ 7:2

78 …60 180 48 …70 180 30 …75 180

50† ˆ 4:3 60† ˆ 2:7 70† ˆ 0:8

(11.42)

301

302

Theory and Practice of Water and Wastewater Treatment

The same fractional removals would be obtained using US units. The removal of the fraction between 75% and 100% is small and is ignored. This is a conservative approach when designing a basin. The upper value of 100% removal would probably exist only at the surface plane of the volume. An upper limit lower than 100% would be dictated by the presence of colloidal particles that are practically unsettleable. The total removal, R, is in this case R ˆ r 0 ‡ Σr i ˆ 40 ‡ 7:2 ‡ 4:3 ‡ 2:7 ‡ 0:8 ˆ 55:0%

(11.43)

This procedure is repeated for different times and the overall removals at each selected time are tabulated to construct a graph as shown in Figure 11.21. 11.5.3 Alternative Method for Calculating Total Removal As noted, the method just given is analogous to the procedure for discrete particle settling. An equivalent method to find the total removal is to examine the amount of removal in each section of the column (Figure 11.20) and sum them. The initial suspended mass, M0, in the column is M 0 ˆ X 0 AD

(11.44)

where A is the cross-sectional area of the column and C0 is the initial concentration of suspended particles. Referring to Figure 11.20, the suspended mass, Mt, at any time is Mt ˆ X 1 V 1 ‡ X 2 V 2 ‡ X 3 V 3 ‡ X 4 V 4 ˆ A…X 1 Δd 1 ‡ X 2 Δd2 ‡ X 3 Δd 3 ‡ X 4 Δd 4 † The percentage removal on any isoconcentration line in Figure 11.19 is ri ˆ

X0 Xi  100 X0

Figure 11.20 Solids concentrations in the column.

Figure 11.21 Overall removal versus detention time.

(11.45)

11 Screening and Sedimentation

A vertical line projected upward from a chosen time in Figure 11.19 intersects isoconcentra­ tion lines that determine each column section length, Δdi. The average removal in a section of column is the average of the isoconcentrations that define Δdi. Assuming the section numbering to start from the lower depth, the total removal, R, in percent, is Rˆ

Δd 1 r 1 ‡ r 2 Δd 2 r 2 ‡ r 3 Δdi r i ‡ r i‡1 ‡ ‡ ∙∙∙ ‡ ‡ ∙∙∙ D 2 D 2 D 2

(11.46)

Equation (11.46) is applied for different times and the results for overall removal at various times can be tabulated and plotted as shown in Figure 11.21. 11.5.4 Sizing the Basin A graph of total removal (R) versus time will provide a design curve. A typical graph is shown in Figure 11.21. The removal curve will eventually become nearly horizontal as time increases. The design point is in the region where the marginal increase in removal is less than the marginal increase in time, which is equivalent to the size of the clarifier. Costs of the clarifier compared to costs associated with other solids removal processes will determine the optimum design point. The design detention time from the laboratory column for the effective settling depth is equivalent to a design settling velocity of D/θd, which is also equal to the design surface loading rate (Q/As). To translate the laboratory data to the field, where nonideal flow conditions exist and sedimentation does not occur under completely quiescent conditions, safety factors of 1.25–1.75 are applied to the detention time and the surface overflow rate. 1) Multiply the design θd based on the column performance by 1.25–1.75. 2) Divide the design Q/As based on the column performance by 1.25–1.75. Application of the safety factors will result in an increase in the surface area of the settling basin. These safety factors are somewhat arbitrary; a computational fluid dynamics (CFD) analysis will provide a much better estimate of the safety factor required.

11.6 Tube and Lamella Clarifiers Theoretically, the efficiency of a clarifier is independent of depth as discussed previously. Fundamentally, this is because the liquid upflow velocity in the basin must be less than the velocity of the slowest settling particle that is to be removed. The pioneer environmental engineer, TR Camp (1945), attempted to exploit this concept by inserting a number of subfloors into horizontal sedimentation basins to increase the surface area. Practically, sludge removal (each floor would need a scraping device) was a problem and the idea became dormant until the 1960s, with the development of tube or lamella settlers, which are an interesting and effective application of the concept. Tube settlers are plastic (polyvinylchloride, PVC) modules with uniformly spaced, inclined channels (Figure 11.22). Lamella settlers (also known as inclined plate clarifiers) have uniformly spaced, inclined panels (Figure 11.23). Lamella settlers can be made with plastic, rawhide, or other available resilient materials. Both types of clarifiers solve the problem of sludge removal. The resultant velocity on a particle from the upward flow of water and the vertically downward settling velocity of the particle direct the particle to the bottom wall of a tube or toward the lamella. The particle then slides down the surface and exits at the bottom to be collected in a

303

304

Theory and Practice of Water and Wastewater Treatment

Figure 11.22 Tube clarifier. Source: Reproduced with permission of ENEXIO Water Technologies.

Figure 11.23 Lamella clarifier.

sludge chamber (see the inset in Figure 11.22). Lamella clarifiers are more common than tube settlers. The theory of these clarifiers is discussed by Yao (1970). For inclined discrete particle settling, a relation between the angle of inclination, θ, settling velocity, length, and diameter of a tube is needed for design. Refer to the definition sketch in Figure 11.24 for a particle settling in a tube. The following definitions apply: V – local fluid velocity FDv – drag force in the vertical direction FDl – drag force in the direction of fluid flow Fb – buoyant vertical force Fg – gravity force θ – angle of inclination of a tube or lamella. The drag force along the direction of fluid flow causes movement of the particle in that direction. The net downward force causes a vertically down velocity. The movement of the particle is the vector sum of these velocities. v ˆ vf ‡ vs

(11.47)

11 Screening and Sedimentation

where v is the resultant particle velocity, vf is the velocity due to fluid movement, and vs is the downward settling velocity. Figure 11.25a depicts particle movement for an upward flow tube or lamella clarifier; Figure 11.25b depicts particle movement in a crossflow lamella (also see Figure 11.23). The particles collect on the tube wall or lamella, and the net downward force causes them to slide downward as shown in Figure 11.25c. It is assumed that once a particle reaches a wall, it will move downward. To find the relation of the length of a tube or channel, angle of inclination, and distance between tube walls or lamellae, refer to Figure 11.26 for a fluid upward flow, solids downward movement, or countercurrent flow situation. To simplify the development, choose the x-axis parallel to the inclination of the tube. The following definitions apply: d – perpendicular distance between tube walls or lamellae L – length of the tube or lamella. The analysis will be based on the average upward flow velocity of a particle [a more refined assessment would be based on the Hagen–Poiseuille law that provides the flow velocity distribution in a flow channel (Yao 1970)]. The fluid velocity will be assumed to impart a particle upward velocity component of vf. The net upward flow velocity of a particle is vu ˆ vf

vs sin θ

(11.48a)

where vu is the net upward particle velocity. The travel velocity to the wall where particle removal occurs is vw ˆ vs cos θ

(11.48b)

where vw is the particle velocity toward a wall.

Figure 11.24 Forces on a particle settling in a tube settler.

Figure 11.25 Particle movement in tube or lamella clarifiers.

305

306

Theory and Practice of Water and Wastewater Treatment

Figure 11.26 Velocity relations in an upflow tube or lamella clarifier.

The critical or limiting case is a particle that begins its trajectory at the top of a wall. In the time, t, to travel the length of the channel, L, it must reach the wall, or travel the distance, d. vu t ˆ L

(11.49a)

vw t ˆ d

(11.49b)

Taking the ratio of Eqs. (11.49a) and (11.49b) provides the design relation, Eq. (11.50): L vf vs sin θ ˆ d vs cos θ

or rearranging

vs 

vf d L cos θ ‡ d sin θ

(11.50)

The critical particle settling velocity can be estimated from principles given in Section 11.2.1. Flow in a tube or between lamellae will be laminar. The average upward velocity depends on the number of channels. vf ˆ

Q Ndw

(11.51)

where N is the number of channels and w is the width of a channel. In the case of lamellae with a crossflow arrangement, the design relation is vs 

vh d L cos θ

(11.52)

where vh is the horizontal flow velocity. The increased surface area (wL cos θ for each channel) available in tube or lamella clarifiers allows surface loading rates based on the plan area that are two or more times higher than loading rates for conventional clarifiers, with the same or better performance (C.A. Richter, personal communication). The total projected surface area for a lamella clarifier containing n lamellae or n + 1 tubes is Ap ˆ nwL cos θ

(11.53)

where Ap is the projected plan area within the clarifier. Depending on the spacing of the tube walls or lamellae, projected areas can be increased by up to a factor of 10 with an angle of inclination of 55°. The Reynolds number in the tubes or between lamella plates is very low compared to the Reynolds number for a conventional clarifier where it will be in the turbulent range. Turbulence

11 Screening and Sedimentation

effectively decreases the settling velocity of particles and causes resuspension of settled particles by scour. Existing clarifiers can be upgraded to higher loading rates by the installation of a tube module or lamellae. Lamellae can be installed in a concentric array in circular clarifiers. Both tubes and lamellae can be installed in horizontal flow sedimentation basins and upflow solids contact clarifiers. There are a number of manufacturers that supply ready-made tube modules. The concept has found its most common application in water treatment installations, but they have been used successfully in wastewater treatment plants for primary and secondary clarifi­ cation. Small tubes or close spacing of lamellae are subject to clogging with the high SS content in wastewater treatment plants. Tube clarifiers are not generally used in wastewater treatment. The lamella surfaces in a clarifier provide a medium on which microorganisms can establish a fixed film growth, necessitating more frequent cleaning with a high-pressure wash. Water treatment plant installations will also need periodic cleaning with a high-pressure wash to remove accumulated particles and biological growth. Lamella clarifiers also have found application in treatment of stormwater and combined sewer overflows. Nominal spacing between plates is 50 mm and inclined lengths range from 1 to 2 m (Metcalf and Eddy:AECOM 2014). The settlers are installed in two different ways. Steeply inclined settlers at angles of 45–60° are stand-alone units (Figure 11.27). The sludge blanket should not extend into the lamellae or tubes. The other alternative is to reduce the angle of inclination to a small value, but the angle must be high enough that solids are not washed out with the clarified effluent. Angles as small as 5° have been used. Solids will not travel significantly downward in a clarifier at this angle of inclination. But these clarifiers are designed in conjunction with rapid sand filters (Chapter 14) such that the discharge of water from backwashing the filter is directed through the clarifier to cleanse the accumulated solids from its walls. Normal backwash quantities and velocities are adequate to scour the tubes. The first portion of the backwash is not directed into the tube settler because it is laden with solids removed from the filter; the intermediate portion is used to wash the tube module; and the latter portion of the backwash water is used to fill the tube settler. Conventional inlets can be used for lamella clarifiers when they are configured as indicated in Figure 11.23. However, in tube settlers, front-end inlets produce flow patterns that do not distribute the flow uniformly to the clarifier module (Figure 11.28). If lamellae are oriented perpendicular to the inlet flow, the same problem will occur; a series of inlet ports that introduce flow parallel to the lamella is one solution. Another improved inlet arrangement used in Brazil is shown in Figure 11.29. Flow is evenly discharged along the length of the clarifier by a series of ports. Sludge is withdrawn through tubes (minimum 38 mm diameter) equidistantly spaced along the bottom of the clarifier. Di Bernardo (1993) has given alternative sludge removal designs for tube settlers. There are other designs in the literature that address this problem.

Figure 11.27 Tube settler configuration in water treatment.

307

308

Theory and Practice of Water and Wastewater Treatment

Figure 11.28 Flow distribution from a conventional inlet in a tube or lamella settler. Source: Di Bernardo (1993). Reproduced with permission of Editora Loibe Ltda.

Figure 11.29 Influent distribution for upflow tube or lamella settler. Source: Di Bernardo (1993). Reproduced with permission of Editora Loibe Ltda.

Example 11.2 Critical Settling Velocity in a Lamella Clarifier Find the critical settling velocity for a crossflow lamella clarifier (see Figure 11.23) that has lamellae spaced at 6 cm (perpendicular distance between the lamellae) intervals. The lamellae are inclined at an angle of 60°. The volumetric flow rate into the clarifier is 1200 m3 h 1 and the clarifier crosssectional area receiving the influent flow is 8 m2. The vertical depth of the lamellae is 2.0 m. The length of the clarifier is 20 m.

11 Screening and Sedimentation

How advantageous is the lamella clarifier over a conventional clarifier of the same dimensions? Equation (11.52) provides the critical settling velocity. vs ˆ

vh d L cos θ

From the problem, d = 6.0 cm, L = 20 m and θ = 60°.

The horizontal velocity is v h ˆ

vh d ˆ vs ˆ L cos θ

1200 m3 h 8 m2

1

= 150 m h 1.

1m 100 cm …20 m†…cos 60†

150 m h

1

…6 cm†

ˆ 0:90 m h

1

The surface area of the clarifier is (4 m)(20 m) = 80 m2. The surface overflow rate of a conventional clarifier would be Q 1200 m3 h ˆ 80 m2 As

1

ˆ 15:0 m h

1

The lamella clarifier is theoretically (15.0/0.9) = 16.7 times more advantageous than the conventional clarifier of the same volume.

11.7 Weir–Launder Design Rising water in a sedimentation basin flows over a weir into a channel or launder that conveys the collected water to the exit channel or pipe. Weirs are located as far as possible from the basin inlet. Weir loading rates are specified to prevent strong updrafts that would carry solids out of the basin. Weir loading rates are specified later on for clarifiers in water and wastewater treatment. As the depth of the basin increases, higher weir loading rates have less influence on the performance of the clarifier. For basins that are not covered, the weir is frequently of the V-notch type (Figure 11.30) to minimize wind effects. Also, straight-edge weirs that are not perfectly level do not have uniform flow over the entire weir. This will cause uneven flow patterns in the sedimentation basin and deterioration of its performance. Submerged orifices are also sometimes designed to discharge effluent from a clarifier. V-notch weirs must have a depth that permits discharge of the peak flow through the basin. Spacing of the notches is in the range of 150–300 mm center to center.

Figure 11.30 V-notch weir.

309

310

Theory and Practice of Water and Wastewater Treatment

Figure 11.31 Definition sketch for flow in a launder.

The discharge through a V-notch weir is given by Vennard and Street (1982) as follows: Qˆ

p 8 θ C d 2g tan H 5=2 15 2 w

(11.54)

where Cd is the discharge coefficient, Hw is the depth of water over the weir, and θ is the angle of the V-notch. The discharge coefficient is around 0.62. The depth–width relation for a V-notch (triangular) weir is w θ ˆ tan 2H w 2

(11.55)

where w is the width of the weir at any height Hw. Flow in the launder is spatially varied flow; refer to the definition sketch in Figure 11.31. These flumes are usually built with no slope. The momentum principle is used for analysis. The following symbols are used in the development: A cross-sectional area F force H water depth at upstream end of launder P pressure Q total volumetric discharge rate y depth of flow in the launder at any x ρ density of water

b launder width

g acceleration of gravity

L launder length

q discharge per unit length

x distance from upstream end of launder

v flow velocity

Friction (Ff in Figure 11.32) will be ignored in the development of the governing equation. It will be accounted for after the equation is derived. This approach is an approximate but reasonable solution of the problem. Steady state conditions are also assumed. Benefield et al. (1984) give a program to perform the numerical integration of the flow equations for this lateral spillway channel and arrive at a more exact solution.

Figure 11.32 Elemental volume in a launder.

11 Screening and Sedimentation

In a circular basin, the flow into the launder is uniform around the circumference. Further­ more, the flow in the launder is symmetrical for the two halves of the launder. At the point opposite from the launder discharge outlet, the flow splits and travels in opposite directions. Therefore, the development is for half of the channel with a length of L/2. To develop the equation governing flow in either half of the channel, examine an elemental volume (Fig­ ure 11.32) indicating parameters to be used in a momentum analysis. The momentum principle is d…m~ v† ~ F ˆ dt

(11.56)

and is applied in the x direction. The following relations are used: Q ˆ qx vˆ

dQ ˆ q dx

Q Q ˆ A by

dv ˆ

A ˆ by dQ A

dA ˆ b dy

Q dA A2

The resultant pressure force is

p2 A2 ˆ ρgby

F ˆ p1 A1 ˆ

1 ρgby2 2

ˆ

y 2

ρgb…y ‡ dy†

…y ‡ dy† 2

1 ρgb y2 ‡ 2y dy ‡ …dy†2 2

(11.57)

ρgby dy

Evaluating the right-hand side of the momentum equation [Eq. (11.56)] for the horizontal direction: Δ…mv† ˆ ρ…Q ‡ dQ†…v ‡ dv† Δt

ρQv

ˆ ρQv ‡ ρQ dv ‡ ρv dQ ‡ ρ dQ dv

ρQv

(11.58)

ˆ ρQ dv ‡ ρv dQ

Equating Eqs. (11.57) and (11.58), ρQ dv ‡ ρv dQ ˆ ρgby dy Substituting for dv in the above equation, Q dQ A

Q2 Q 2 dA ‡ A dQ ˆ gb dy A

Substituting for A, dA, Q, and dQ results in 2qx q dx by

q2 x2 dy ˆ gby dy by2

The above equation can be simplified to 2x

dx dy

x2 b 2 y2 ˆ g 2 y q

(11.59a)

311

312

Theory and Practice of Water and Wastewater Treatment

or x2 b 2 y2 ˆ g 2 y q

d x2 dy

(11.59b)

This equation can be integrated to x2 ˆ Cy

gb2 y3 2q2

(11.60)

The boundary condition is x ˆ 0;

yˆH

The boundary condition is used to find the integration constant, C. gb2 H 2 2q2

Cˆ

Substituting for C, the equation for H becomes Hˆ

y2 ‡

2q2 x2 gb2 y

0:5

(11.61)

To determine H, it must be recognized that a free fall condition exists at the end of the launder, which means that critical flow exists there. For critical flow, the depth is yc and the Froude number (Fr) is equal to 1. v Fr ˆ 1 ˆ p gyc

or

vˆ

p

gyc

(11.62)

The velocity at the end of the launder is vˆ

Q qL ˆ A 2byc

Substituting this into Eq. (11.62) and solving for yc yields yc ˆ

…qL†2 4b2 g

1=3

(11.63)

Equation (11.63) is used to find yc, which occurs at x = L/2. This information may be used in Eq. (11.61) to find H. Headloss can be estimated from the Chezy or Manning equations. Using Manning’s equation, 1 v ˆ R⅔ S ½ n

(11.64)

by where n is a roughness coefficient, R is the hydraulic radius = , and S is the slope of the b ‡ 2y energy line. The value of n is 0.011–0.015 for smooth concrete; n values for other materials can be obtained from standard references (Chow 1959). The value of n should be slightly increased to account for curvature in the lateral and turbulence from the falling water. For smooth concrete, use

11 Screening and Sedimentation

n = 0.014. At any distance x, v = qx/by. Substituting this into the above equation and solving for S, Sˆ

qxn by

b ‡ 2y yb

⅔ 2

ˆ

4 …qx†2 n2 =3 10=3 …b ‡ 2y† …by†

(11.65)

Equation (11.61) is used to determine the value of y at any x. Equation (11.65) is applied to find the energy slope at the midpoint of the flume, SMP, and at the end (depth of flow is yc), Sc. The energy slope at the beginning of the flume is 0. The headloss (hL) in each half of the flume is estimated from hL ˆ

0 ‡ SMP L S MP ‡ S c L ‡ 2 4 2 4

(11.66)

The headloss is added to the value of H determined from Eq. (11.61). The depth of the launder is increased by 5–10 cm to ensure free fall from the weir. Freeboard depth is also added to prevent splashing and to account for nonuniform flow variation over the weir because of the wind or flow distribution in the clarifier and uncertainty in the estimation of the flow. A freeboard of 10–20% of the depth calculated from the headloss and H is used. The total depth of the launder is the sum of freeboard, H, headloss, and free fall allowance.

11.8 Clarifier Design for Water and Primary Wastewater Treatment There are many varieties of clarifier designs. 11.8.1 Design Ranges for Typical Clarifiers for Water and Wastewater Treatment The design ranges for clarification basins used in water treatment plants are highly variable depending on the quality of raw water and type of floc formed, which is naturally dependent on the coagulant used and the operation of the flocculation process (see Chapter 13). Clarifier surface areas are either rectangular (Figure 11.33) or circular; square surface areas are less popular. Lamellae may be incorporated in either type. Solids contact clarifiers incorporate coagulation, flocculation, and clarification into a single unit and are usually circular. Ranges for design variables for different configurations are given in Table 11.9. Handbooks or other references should be consulted to find the narrower ranges for different types of sediments

Figure 11.33 Rectangular clarifier.

313

314

Theory and Practice of Water and Wastewater Treatment

Table 11.9 Clarifiers in water treatment. Item

Value

Rectangular and circular clarifiers 2.4 4.9

Depth (m) Overflow rate (m3/m2-d)

20 70

3

Weir loading rate (m /m-d)

< 315

Maximum length of rectangular basin (m)

70 75

length:width of rectangular basin

3 : 1 to 5 : 1

Circular basin maximum diameter (m)

38

Upflow solids contact clarifiers, incl. pulsed bed with lamella or tube settlers 2.5 3

Depth (m) 3

2

Overflow rate (m /m -d)

58–290

Inclined tube or lamella clarifiers Inclined length (m)

1 2

Angle of inclination (°)

7 60 (55–60° is most common)

Tube diameter or plate spacing (cm)

5–10 3

2

Overflow rates based on actual plan area (m /m -d)

2 10 times rate for conventional clarifiersa)

Tube overflow rate

58–175

Lamella overflow rate

120–530

Depth (m)

6 7

a) With chemical addition overflow rates based on plan area may be increased by 10–50%. Source: Compiled from AWWA and ASCE (2012), AWWA (1999), and Dowbiggin and Breese (2013).

and coagulants. The majority of SS removal will occur in the sedimentation basin in a water treatment process. Dissolved air flotation is also used for clarification of flocculated waters in water treatment plants. See Section 21.4.2 for a description of this process. Clarifiers used in wastewater treatment may be rectangular, square, or circular although rectangular clarifiers are more common for primary clarifiers and circular for secondary clarifiers. Design guidelines for primary clarifiers are given in Table 11.10; secondary clarifiers are discussed in Section 18.3.2. Variation among design codes is large. Primary clarifiers are designed more conservatively if sedimentation is the only treatment and if activated sludge is being returned to the primary clarifiers. Rectangular tanks are generally designed with the same criteria as circular tanks. Detention times in primary clarifiers range from 1.5–2 hours. Rectangular clarifiers have lengths in the range of 15–90 m and widths from 3–24 m; circular clarifier diameters range from 3 to 60 m (Metcalf and Eddy:AECOM 2014). Lamella clarifiers for primary treatment will require approximately weekly cleaning (WEF and ASCE 2010). Properly designed and operated primary clarifiers should remove 50–70% of the influent TSS. Sludge removal usually occurs over 2–4-hour intervals. Longer detention times of the sludge will promote septic conditions resulting from some fermentation of the settled sludge. Effluent soluble COD will increase and odors may be produced. Gas production may buoy up some settled sludge.

11 Screening and Sedimentation

Table 11.10 Primary clarifiers in wastewater treatmenta),b). Item

Value

Overflow rate (m3/m2-d) For average dry weather flow rate

32 49

For peak flow condition

49 122

Sidewater depth (m)

2.1 5

Length:width ratio for rectangular

1.5 : 1 to 15 : 1

3

Weir loading rate (m /m-d)

125 500

Inclined tube or lamella clarifiers Plate spacing (mm)

40 120

a)

Criteria are based on the maximum ranges specified by a number of firms and agencies reported in WEF and ASCE (2010). b) Generally for average flow conditions. Source: Compiled from WEF (2005), WEF and ASCE (2010), and Metcalf and Eddy:AECOM (2014).

Table 11.11 Primary clarifier surface loading rates. Clear water zone, Z

Surface loading rate (m h − 1)

m

Average flow

Peak flow

1.83 < Z < 3.05

0.091 Z2

0.182 Z2

3.05 < Z < 4.57

0.278 Z

0.556 Z

Source: Adapted from Albertson (1992).

Albertson (1992) examined primary clarifier performance from a large number of installations and determined the empirical correlations in Table 11.11 for design surface overflow rates as a function of the clarifier clear water depth. 11.8.2 Chemically Enhanced Primary Treatment The performance of primary clarifiers in wastewater treatment can be enhanced by the addition of chemicals to improve treatment and/or cope with increased wastewater flows. Ferric chloride or alum is typically applied at doses 1000

Al2(SO4)3a)

30

>1000

FeCl3

1

>1000

Fe2(SO4)3a)

30

>1000

a)

The most common coagulating agents.

pH range of 4.5–5.5, whereas aluminum salts are most effective around a pH range of 5.5–6.3. These pH values should be attained after the coagulant is added. If necessary, the pH may be adjusted with acid or alkalinity. Residual aluminum or iron will be minimized when the pH is in the optimum range. Table 13.2 Chemistry of aluminum, iron salts, and lime coagulation. 1 Alum Al2 …SO4 †3
18H2 O ‡ 6H2 O → 2Al…OH†3 …s† ‡ 6H‡ ‡ 3SO24 ‡ 18H2 O With an increase in H+, pH is depressed and no more Al(OH)3 is formed. If natural alkalinity is present, then HCO3 ‡ H‡ ⇆ H2 CO3 ⇆ CO2 ‡ H2 O Al2(SO4)3
18H2O + 3Ca(HCO3)2 → 2Al(OH)3 (s) + 3CaSO4 + 6CO2 + 18H2O If natural alkalinity is insufficient, then lime or caustic soda can be added. Al2(SO4)3
18H2O + 3Ca(OH)2 → 2Al(OH)3(s) + 3CaSO4 + 18H2O 2

Sodium aluminate

3

Ferrous sulfate FeSO4
7H2O + Ca(OH)2 → Fe(OH)2 + CaSO4 + 7H2O 4Fe(OH)2 + O2 + 2H2O → 4Fe(OH)3(s)

4

Ferric chloride

Na2Al2O4 + Ca(HCO3)2 + 2H2O → 2Al(OH)3(s) + CaCO3(s) + Na2CO3

2FeCl3 + 3Ca(HCO3)2 → 2Fe(OH)3(s) + 3CaCl2 + 6CO2 The above reaction takes place if natural alkalinity is present in a sufficient amount. Otherwise if lime is added, 2FeCl3 + 3Ca(OH)2 → 2Fe(OH)3(s) + 3CaCl2 5

Ferric sulfate Fe2(SO4)3 + 3Ca(HCO3)2 → 2Fe(OH)3(s) + 3CaSO4 + 6CO2

Fe2(SO4)3 + 3Ca(OH)2 → 2Fe(OH)3(s) + 3CaSO4

13 Coagulation and Flocculation

Note that alum and some iron salts are supplied in hydrated states. Dry alum is hydrated with 14.3–18 water molecules [Al2(SO4)3
14.3H2O or Al2(SO4)3
18H2O]. Table A.3 gives concentra­ tions of commercially available coagulating agents. Temperature exerts an effect on the efficiency of coagulation and flocculation. Metal salts given in Table 13.2 form hydroxide precipitates, which logically leads to the supposition that pOH is an important factor in the chemistry of the process. Temperature affects the equilibrium constant of water, as discussed in Section 3.1; therefore, maintaining the same pH at all temperatures will result in varying pOHs. For both iron and alum salts, Hanson and Cleasby (1990) found that a constant pOH over the temperature range of 5–20 °C produced the best coagulation–flocculation results. The association of aluminum with Alzheimer’s disease was discussed in Chapter 8; the etiologic link between aluminum and Alzheimer’s disease is tenuous. Use of aluminum salts as coagulants may increase the concentration of Al in product water. Letterman and Driscoll (1988) surveyed 91 plants in North America that use alum and found that 75% of the plants produced water with 210 μg L 1 or less total Al; the 50th percentile value was 90 μg L 1. The results of their survey were in good agreement with that of an earlier survey (Miller et al. 1984). High raw water Al concentrations were associated with higher residual Al concentrations. Effective removal of particulate matter by filtration minimizes residual Al concentrations. Letterman and Driscoll’s study found that lime used for pH adjustment after filtration may be an important source of residual Al in product water. Residual Al depends on the final pH, temperature, ionic strength, and formation of complexes. A final pH between 6 and 7 is the range in which Al solubility will be lowest (Letterman and Viacoumi 2011). Residual iron solubility for ferric ion is very low (98% of all metals [except Cr(VI)] in the presence of iron hydroxide. Synthetic coagulating agents are widely available. Cationic, anionic, and nonionic polymers have all been found to provide excellent results in different situations. These agents are usually more costly than alum or iron salts but much smaller dosages are required. Figure 13.1 gives general formulas of synthetic polymers. Nonionic polymers are almost exclusively polyacrylamides with a molar mass between 1 and 30 million. Anionic polyelec­ trolytes have masses of several millions. They contain groups that permit absorption and negatively ionized groups (carboxyl or sulfuric groups), which extend the polymer. Soda ash (Na2CO3) has been used to partially hydrolyze the polyacrylamide groups in the figure. Cationic polymers have molar masses ranging from hundreds of thousands to several millions. Their chains have amine, imine ( NH group attached to a carbon atom), or quaternary ammonium groups that produce the positive charge.

353

354

Theory and Practice of Water and Wastewater Treatment

Figure 13.1 Typical formulas of coagulant polymers.

Activated silica is another commonly used coagulation agent. Sodium silicate (Na2SiO3) is “activated” by acidification. The chemistry of activated silica is fairly complex, but Stumm et al. (1967) provided the basic explanation of the process. When a concentrated solution of sodium silicate is acidified, it becomes oversaturated with respect to the precipitation of SiO2. The precipitation process begins with the formation of polysilicate polymers that contain Si O Si linkages. These polysilicates react further by cross-linking and aggregation to form negatively charged silica sols. The solution contains anionic polysilicates and other silicate species, which are effective coagulants. Over time, SiO2 will precipitate and the solution will lose much of its coagulating strength. Although the acidified solution is thermodynamically unstable, it can remain active for periods up to a few weeks. Ozone has been found to be an effective coagulant aid. Low dosages ( 0.2, log …Θ=6† ˆ 0:565 43 ‡ 1:093 48 log Ref ‡ 0:179 79 …log Ref †2

0:003 92 …log Ref †4

1:5 …log ψ†2 (14.39b)

Excellent agreement was achieved with Eqs. (14.39a) and (14.39b) for a wide variety of media at different expanded porosities above approximately 10% expansion of the bed (J.L. Cleasby, personal communication). Dharmarajah and Cleasby (1986) report that the correlations are good up to ee = 0.85 and for media with a s.g. 0.2, only Eq. (14.39b) was used to calculate Θ.

b) Le is the sum of the expanded depths of the two layers.

c) The expanded porosities are less than the rest porosity and the depth of the layer should be adjusted to its

unexpanded depth.

A backwash velocity of 32 m h 1 will achieve an overall bed expansion of 10%. Note that the two layers are expanded to different degrees. The minimum fluidization velocity does not achieve any expansion of the anthracite layer; however, the backwash rate of 26.3 m h 1 (1.3vmf for the anthracite) is near the velocity predicted by Eqs. (14.38) and (14.39a) to achieve expansion of this layer. On the other hand, the minimum fluidization velocity for the sand layer is near the minimum fluidization velocity calculated with Dhar­ marajah and Cleasby’s approach. With the 1.3 adjustment factor (resulting in a velocity of 27.3 m h 1), the expansion of the sand layer is about 10% which should be sufficient to expand the largest particles. The results in the table are reasonably close to the observations recorded in Table 14.6. Bed fluidization is a complex hydrodynamic phenomenon dependent on a number of fluid and medium characteristics as well as the manner in which the flow is introduced into the bed. A more accurate estimate of the expansion could be obtained by applying the equations to individual layers within each medium.

14.5 Support Media and Underdrains in Rapid Filters Historically, the media in a filter were supported by graded-gravel layers that prevent the media from reaching and clogging the water collection underdrains. Figure 14.6 shows two commonly used gravel layer gradations. Headloss through the gravel layers is strictly a function of the filtration velocity. The gravel layers do not clog as filtration progresses. The Ergun equation [Eq. (14.16)] can be used to calculate the headloss through each layer in a manner similar to the procedure in Example 14.2. Porosities in gravel vary from 0.18 to 0.35 for coarse to fine gravel (Reed et al. 1988). Support media are more widely used where biological fouling is of concern, e.g., biologically active filtration in water treatment or wastewater filtration (S. Humphries, personal communication). Underdrain systems serve both collection of the filtered water and uniform distribution of the backwash water. Modern underdrain systems eliminate the need for gravel support layers. There

409

410

Theory and Practice of Water and Wastewater Treatment

Figure 14.6 Supporting gravel layers for sand filters. Source: From Fair et al. (1968). Used with permission of John Wiley & Sons.

are a variety of underdrain designs. Figure 14.7 shows one type of underdrain made from plastic or vitrified clay. These blocks are laid across the filter bottom to form continuous channels that feed into larger collection pipes. Some companies supply strainer underdrains (Figure 14.8) that also do not require a supporting gravel layer. Backwash water is also delivered through the underdrains. Backwash velocities will not be sufficient to expand the supporting gravel layers when they are used. The backwash velocity should be slowly increased over a period of at least 30 seconds to avoid disturbing the supporting gravel layers. Headloss through the gravel layers during backwash is also calculated with the Ergun equation. Headloss through the underdrain openings during filtration or backwash is calculated with an orifice equation. The flow is distributed through all of the orifices to calculate the velocity through each orifice. hL ˆ C d

v2 2g

(14.43)

where Cd is the discharge coefficient for the orifice, hL is the total headloss through the underdrains, and v is the velocity through an individual orifice.

Figure 14.7 Underdrain block for filters. Source: TechValidate Survey. Reproduced with permission of Xylem Inc.

14 Filtration

Figure 14.8 Strainers used in false-bottom underdrains without gravel. Source: Orthos Liquid Systems. Reproduced with permission of Orthos Liquid Systems.

The orifices drain into channels that feed into larger collection channels. Flow in all underdrains is pressurized, and the Darcy–Weisbach equation is suitable for calculating headloss in these channels. Headlosses through valves, bends, and other appurtenances in the channels and pipes from the filter are proportional to the velocity head through the device. Pipe lateral systems can also be used for underdrains as shown in Figure 14.9. Nozzles are installed on laterals to provide sufficient headloss to increase waster distribution. Other Design Features of Filters

Conduits in filters are designed for velocities near the following ranges: Influent pipe to filters Effluent pipe carrying filtered water Drains carrying spent backwash water Wash water line (influent) Filter to waste drain

0.6–1.8 m s 0.9–1.8 m s 1.2–2.4 m s 2.4–3.7 m s 3.7–4.8 m s

1 1 1 1 1

Auxiliary Wash and Air Scour Systems

The primary scouring mechanism to cleanse the media during backwashing is hydrodynamic shear (Amirtharajah 1978a); abrasive scour caused by particle collisions is not significant. Auxiliary wash systems are incorporated into filter systems to promote particle collisions and improve backwashing performance significantly reducing backwash water volumes. Two types of auxiliary wash systems are used: surface wash and air scour. The latter is also commonly used in Europe.

Figure 14.9 Filter underdrain systems. Source: From Fair al. (1968). Used with permission of John Wiley & Sons.

411

412

Theory and Practice of Water and Wastewater Treatment

A surface wash system supplies jets of water from nozzles located 2.5–5 cm above the fixed bed surface. The distribution device may be a rotating arm or fixed laterals. The nozzles are directed 15–45° below the horizontal. Operating pressures are typically in the range of 350–520 kPa (Cleasby 1990). The nozzle orifice sizes are 2–3 mm. The surface wash is initiated 1–2 minutes before the backwash flow is started and continued until 2–3 minutes before the backwash flow is terminated. Surface wash systems affect only the upper layers of the expanded bed. Air scour systems deliver air across the entire area of the filter. The air is introduced at the bottom of the filter, but above gravel support layers when they are used, or the air can be supplied with the water in modern nozzle support systems. The air causes particle contact to occur throughout the entire depth of media. These systems are more effective than surface wash systems. Air scouring is most effective when the water is flowing between 25% and 50% of the minimum fluidization velocity that causes water cavities within the media to form and collapse, which is known as collapse pulsing (Amirtharajah 1993). Amirtharajah (1993) provides details of the operating procedure. The performance of biological filters for water treatment (Section 14.11), where scour and removal of biomass during backwash is a consideration, was not impaired with air scour at sub-fluidization water flow with collapse pulsing (Ahmad et al. 1998).

14.6 Filter Beds for Water and Wastewater Treatment The design of a rapid filter system for water treatment depends on the treatment objectives and the pre-treatment that has been applied to the filter influent. Design information for filter beds for various applications is given in Table 14.8. A rule of thumb commonly used by engineers for filters is to have a L/d ratio (L is depth and d is ES of media) between 1000 and 2000 (Crittenden et al. 2012). The minimum number of individual filter beds is two. For midsized receiving flow between 0.4 and 1.1 × 105 m3 d 1, Kawamura (1999) recommends four beds. When only two beds are installed, a single bed must be capable of meeting water demands during periods of shutdown of either filter for maintenance and backwashing. Kawamura (2000) also recommends a minimum of four beds for larger plants. The practical maximum area of an individual bed is approximately 90 m2, although for large plants two filter beds sharing a central gullet can have each have an area up to 225 m2 (Kawamura 1999). The gullet will backwash only one filter at a time in the latter case. Initial headloss through the media, gravel, and underdrain effluent system in clean filters is in the range of 0.6–1.2 m. Terminal headlosses range from 1.2 to 3 m, depending on filter characteristics and costs of providing additional depth in the filters. Wash water collection launders will be submerged during filtration. Filtration of wastewater is becoming a more common practice to enhance SS removal. More stringent wastewater treatment standards promote the practice. Suspended solids will be removed to 2–8 mg L 1 and indicator microorganisms will have removals of 0–1 log in a properly designed and operated filter (Metcalf and Eddy:AECOM 2014). Typical design information for wastewater treatment filter beds is given in Table 14.9. For wastewater, monomedium filters are more common than dual-media filters. The maximum area of an individual filter for wastewater treatment is the same as for a water treatment filter (Culp et al. 1978). For intermittent filtration of effluent from stabilization ponds, there are two basic configura­ tions: single-stage intermittent filters or intermittent filters in series (USEPA 1983). Single-stage intermittent filters use sand medium with a small ES in the range of 0.20–0.30 mm. Uniformity coefficients are high, ranging from 5 to 10 for a number of installations, although Truax and

14 Filtration

Table 14.8 Design features of filter beds for water treatmenta). Effective size (ES) (mm)

Depth (L) (m)

Filtration rate (m h − 1)

After coagulation and flocculation Sand alone 0.45–0.55 a)

Brazil

0.6–0.7

Cleasby (1990)

L/(ES) > 1000

7.5 (max)b)

Kawamura (1999)

0.45–0.55 (media size range: 0.41–2.00 mm)

0.6–0.8

5–7.5

Di Bernardo (2002)

1.0–1.2 mm (media size range: 0.84–1.68 mm)

0.8–1.2

7.5–12.5

Di Bernardo (2002)

0.9–1.1

0.6–0.9

Dual media Add top-layer anthracite (0.1 to 0.7 of bed)

L/(ES) > 1000

Cleasby (1990) Tobiason et al. (2012) 10–12.5 (max)b)

Kawamura (1999)

Triple media Add 0.1 m garnet

0.2–0.3

0.7–1.0

Cleasby (1990)

L/(ES) > 1250

Kawamura (1999)

1.2–1.4

L = 1.8–2.0L/ (ES) > 1300

Kawamura (1999)

 1.5

L/(ES) > 1500

Kawamura (1999)

Coarse media filters with simultaneous air/water wash For coagulated and clarified water

0.9–1.1

0.9–1.2

Tobiason et al. (2012)

For direct filtration: may add 0.15–0.30 m bottom sand layer

1.4–1.6

1–2

Tobiason et al. (2012)

patm. When the effluent weir is located at the height of the surface of the media, available head within the media is a monotonically decreasing

Figure 14.11 Available head development with different weir locations. Weir located at (a) bottom of filter, (b) top of media, (c) an intermediate depth.

14 Filtration

function of depth below the media surface (Figure 14.11b). The head at the bottom of the media is equal to the headlosses in the gravel, underdrain system, and other appurtenances in the filter effluent line. No available head is gained below the surface of the media. This solution solves the problem, but as Monk (1984) advises, the safety factor involved is excessive. Increasing the elevation of the weir will require an increase in depth of water over the media to achieve the same throughput of water. Some economies can be gained by lab studies on the headloss performance of the filter under probable operating conditions, and the weir can be lowered and still provide a reasonable safety factor against the development of negative head. The available head curve for this situation is shown in Figure 14.11c. It is a combination of curves applying to the cases where the weir is located at the bottom (Figure 14.11a) or top (Figure 14.11b) of the filter.

14.8 Rapid Filtration Alternatives There are some features that apply to all rapid filtration operations. It is essential that the influent to a rapid filter be coagulated. There will be a slight deterioration in filtrate quality after backwashing. Sudden increases in the rate of filtration also diminish the filter effluent quality. The bed must not be allowed to dewater. 14.8.1 Single-medium and Multimedia Filters Most of the solids removal in single-medium (sand) filters occurs in the top layers of the filter. The full depth of the medium is not effectively used and headloss increases rapidly. Replacing the upper depth of sand with coarser anthracite medium definitely retards the rate of headloss development and increases the length of a filter run. A dual-media filter does not necessarily improve the quality of the filtrate (Cleasby 1990), but there is no deterioration of the filtrate over the longer run times. The extension of the dual-media filter is the triple-media (multimedia) filter, which uses garnet or ilmenite media of finer size than sand in the bottom layer of the filter. As a result of a finer grain size being included in the filter, the initial clean bed headloss is larger for triple-media filters compared to single- or dual-media filters, but triple-media filters outperform single-medium filters for headloss development (Cleasby 1990). The incorporation of finer-sized media is expected to improve effluent quality. Cleasby (1990), in a review of multimedia filter studies, was unable to conclude that triple-media filters resulted in superior effluent quality compared to dual-media filters. Some studies were poorly designed, and there was a lack of studies comparing dual- and triple-media filters. A comparison of dual-media and mixed media (anthracite, sand, and two size ranges of garnet) performance at a full-scale plant installation showed no significant differences in performance (Barnett et al. 1992). Granular activated carbon (GAC) can be used as the top medium for taste and odor control and to adsorb organic compounds. GAC has a lower density than anthracite, which affects backwashing requirements. 14.8.2 Constant- and Declining-rate Filtration The rate of water throughput in a filter is a function of the headloss through the filter system and the driving head of water over the filter. Q ˆ f …hD

hc

hu

hf

hv †

(14.48)

417

418

Theory and Practice of Water and Wastewater Treatment

where hD is the driving head (depth of water over the filter), hc is the clean bed headloss, hu is the headloss through the gravel and underdrain system, hf is the friction headloss, resulting from solids accumulation in the filter, and hv is the headloss through valves and other appurtenances in the effluent pipe from the filter. A constant-rate filter produces water at a rate that is more or less constant. The flow rate may be influent- or effluent-controlled. If the supply of water to the filter is constant, the headloss in the filter (hf) will increase as the run progresses to cause the water depth over the filter to increase to maintain the throughput. If effluent control is used, a constant water level over the filter can be maintained by progressively opening a valve on the effluent line as hf increases during the run to maintain hf + hv constant. Either alternative can be used, or a constant feed rate with some variation in depth of the water over the filter and some valve control can also be used. For a filter operated in the declining-rate mode, the water level above the filter is held more constant throughout the run, resulting in a declining rate of water throughput. Therefore, declining-rate filtration has a lower total head requirement than constant-rate filtration that does not rely on valve control of the effluent. Declining-rate filters can be designed in a bank of filters in which a filter is backwashed by the effluent from the other filters without any requirement for a pump. For this design, a minimum of four filters is required. The filters are fed by a common header and the same water level exists over all filters. The underdrain systems from each filter feed into a common collection pipe. The filter most recently backwashed has the highest filtration rate, whereas the dirtiest filter has the slowest filtration rate. Amirtharajah (1978b) has given a detailed design example of a declining-rate filtration system. Effluent quality is better from a declining-rate filter compared to a constant-rate filter, although there is some debate over this issue. Neither mode of operation results in rapid changes in filtration rate when the filters are properly controlled. Constant-rate filters will be operated until the desired terminal headloss is reached without any special instrumentation. In a bank of declining-rate filters, instrumentation must be provided to determine the flow rate through each filter. The headloss is the same for each filter at all times.

14.8.3 Direct Filtration Direct filtration refers to omitting sedimentation from the treatment train. Coagulation is still essential for the filter influent in this circumstance. Less coagulant is generally used for water to be directly filtered to promote the formation of smaller, filterable floc rather than large, readily settled floc. A separate flocculation unit may precede the filter or flocculation may occur in the filter. Direct filtration is used for good quality surface water because of the cost savings associated with elimination of sedimentation. The quality of water defined as the “perfect candidate” for direct filtration is as follows (AWWA Committee 1980): Color, 0 in Figure 15.1). The combined flow may be further treated for calcium reduction as shown in Figure 15.1. Economies are gained by treating smaller volumes of water. In addition, magnesium concentrations in the product water can be controlled in a split-treatment process. Chemical costs are also reduced in a split-treatment process. The bypassed flow can be treated with coagulation agents as required for removal of SS in the settling basin. A split-treatment process is normally applied to water when a low amount of NCH and magnesium hardness reduction is desired (Crittenden et al. 2012). In a split-treatment process all lime based on the total flow is added to the stream entering the first stage to cause Reactions (15.1)–(15.5). There is normally a large excess of lime in the first basin and magnesium

15 Physical–Chemical Treatment for Dissolved Constituents

Figure 15.1 Split treatment flow diagram for hardness removal (H0, initial total hardness; Hf, final total hardness). Source: Adapted from Cleasby and Dillingham (1966). Reproduced with permission of the American Society of Civil Engineers.

concentrations below 0.2 meq L 1 down to practically zero can be attained in the settled effluent from this stage. Recarbonation of the effluent from the first stage is almost never required because there is sufficient alkalinity and carbon dioxide in the bypassed flow to react with the excess lime. There will be no precipitation of magnesium after the first stage because the pH will be lowered below solubility product values for Mg(OH)2 precipitation. Mixing the treated and bypassed streams results in more carbonate formation and subsequent precipitation of calcium carbonate. If further reduction of NCH is required, the influent to the second stage can be supplemented with Na2CO3 to reduce the noncarbonate Ca2+ according to Reaction (15.6). Another settling basin is required to settle the precipitate formed by these reactions. Finally, CO2 is added to cause Reaction (15.8) to bring the pH to the desired final value. The final desired magnesium concentration dictates the fraction of flow that is bypassed in a split-treatment process. Water heater fouling problems are avoided when magnesium concen­ trations are 40 mg L 1 as CaCO3 (0.8 meq L 1) or lower (Larson et al. 1959). Because no magnesium is removed in the second stage, …1

x†Q Mg

1

‡ xQ Mg

0

ˆ Q Mg

f

or x ˆ

Mg Mg

f

Mg

1

0

Mg

1

(15.9)

where x is bypassed fraction of influent flow, Q; [Mg]0, [Mg]1, and [Mg]f are the influent, first stage, and final magnesium concentrations, respectively. The maximum fraction that can be bypassed occurs when [Mg]1 = 0. Recycling a portion of the sludge from the clarifier promotes reaction and settling in the process. Doses higher than stoichiometric requirements may be required to achieve desired results. A split recarbonation process is used to remove magnesium from waters that have high amounts of NCH. There is no bypass (x = 0 in Figure 15.1) in a split recarbonation process. Excess lime is required to precipitate magnesium, and recarbonation is required to neutralize the

435

436

Theory and Practice of Water and Wastewater Treatment

excess lime. If recarbonation is split into two stages where only enough CO2 is applied in the first stage to react with the excess lime and produce carbonate, at the expense of an additional settling basin more calcium may be precipitated. The second-stage recarbonation will be used for final pH adjustment. A single-stage process will produce water with more alkalinity and higher hardness concentrations. Carbon dioxide can be supplied in bulk containers or generated by burning a hydrocarbon such as gas or oil or burning coal or coke. Polyphosphates are synthetic phosphate compounds that sequester hardness ions, keeping them in solution. Polyphosphates are often added to treated waters to prevent deposit buildup on pipes. Example 15.1 Lime and Soda Determination for Hardness Removal Find the lime and soda requirements to treat the water with characteristics given below to a final hardness of (a) 100 mg L 1 as CaCO3 and (b) the practical limit in a single-stage process. In a single-stage process, no CO2 is added in the hardness precipitation reactor. Assume the excess lime requirement to remove magnesium is 0.75 meq L 1. The pertinent data are as follows: mg L − 1 as CaCO3 pH

Alkalinity

Total hardness

Calcium hardness

7.7

180

215

160

It is assumed that the alkalinity is due totally to inorganic carbon species. The concentrations of iron and manganese are not specified and therefore are assumed to be insignificant. The concentration of free CO2 (H2 CO∗3 ) will be checked to determine whether it is significant. The dissociation constant for H2 CO∗3 (Table 3.2) is 4.5 × 10 7. [H+] = 10 pH = 10 7.7 = 2.00 × 10 8 M. HCO3 ˆ 180 mg CaCO3 L

122 mg HCO3 100 mg CaCO3

1

1 mol 61 000 mg

ˆ 3:60  10

3

M

1

as CaCO3

From the equilibrium expression: H2 CO3∗ ˆ

2:00  10 8 3:60  10 ‰H‡ Š HCO3 ˆ K 4:5  10 7

The concentration of H2 CO∗3 in meq L H2 CO3∗ ˆ 1:60  10

4

Mg2‡ ˆ 55 mg CaCO3 L Ca2‡ ˆ 160 mg CaCO3 L Because there is 1 meq mmol 1

1

1

1

ˆ 1:60  10

4

M

is

M 2000 meq mol

Mg hardness ˆ Total hardness

HCO3 ˆ 3:60 meq L

1

3

1

ˆ 0:32 meq L

Ca hardness ˆ 215

1

160 mg L

1 meq 50 mg CaCO3

ˆ 1:10 meq L

1 meq 50 mg CaCO3

ˆ 3:20 meq L

for HCO3 ,

1

1

1

ˆ 55 mg L

15 Physical–Chemical Treatment for Dissolved Constituents

The Ca carbonate hardness is 3.20 meq L 1 and the Mg carbonate hardness is 0.40 meq L 1. There is 0.70 meq L 1 of Mg NCH. A residual hardness of 100 mg L 1 as CaCO3 is equivalent to 2 meq L 1. a) The amount of total hardness to be removed (ΔTH) is ΔTH ˆ 3:20 ‡ 1:10

2:00 ˆ 2:30 meq L

1

It will only be necessary to remove calcium hardness. The lime requirement is equal to the CO2 content and the amount of calcium to be removed, which is increased by 0.6 meq L 1 to compensate for the residual HCO3 . Ca…OH†2 ˆ ‰CO2 Š ‡ Ca…HCO3 †2 ‡ 0:60 meq L

1

ˆ 0:32 ‡ 2:30 ‡ 0:60 ˆ 3:22 meq L

1

The final concentrations (meq L 1) of ions in the water are [Ca2+] = 0.90; [Mg2+] = 1.10; [CO23 ] = 0.60; [HCO3 ] = 0.70. b) In this case, the lime requirement using Eqs. (15.1)–(15.6) is Ca…OH†2 ˆ ‰CO2 Š ‡ Ca…HCO3 †2 ‡ 2  Mg…HCO3 †2 ‡ MgX ‡ excess ˆ 0:32 ‡ 3:20 ‡ 2  0:40 ‡ 0:70 ‡ 0:75 ˆ 5:77 meq L

1

If only this amount of lime were added, the amounts of Ca2+ and CO23 formed would be CO23

ˆ 0:32 ‡ 6:40 ‡ 0:80 ˆ 7:52 meq L

Ca2‡ ˆ 3:20 ‡ 5:77 ˆ 8:97 meq L

1

1

The CO23 is limiting, and the residual Ca2+ and CO32 concentrations would be

CO23

ˆ 0:60 meq L

Ca2‡ ˆ 8:97

1

6:93 ˆ 2:04 meq L

1

The residuals for Mg2+ and OH would be Mg2‡ ˆ 1:10 ‰OH Š ˆ 5:78

…0:40 ‡ 0:70† ‡ 0:20 ˆ 0:20 meq L

1

…0:32 ‡ 3:20 ‡ 0:40 ‡ 0:90† ˆ 0:96 meq L

1

The soda ash requirement is ‰Na2 CO3 Š ˆ MgX ‡ excess lime ˆ 0:70 ‡ 0:75 ˆ 1:45 meq L

1

The final composition (meq L 1) of the water leaving the settling basin is [Ca2+] = 0.60; [Mg2+] = 0.20; [CO23 ] = 0.60; [OH ] = 0.95; [Na+] = 1.45. Acid addition will be required for OH neutralization and pH adjustment of the effluent.

Example 15.2 Hardness Removal in a Split-treatment Process Determine the bypassed fraction and doses of lime and soda ash to be applied at each stage in a splittreatment process to a groundwater that contains the following softening-related constituents: ‰CO2 Š ˆ 0:40 meq L 1 ; alkalinity ˆ 5:80 meq L

Ca2‡ ˆ 3:60 meq L 1 ;

Mg2‡ ˆ 2:00 meq L 1 ;

1

The product water should have a magnesium concentration of 50 mg L 1 as CaCO3 (1 meq L 1) and a total hardness concentration of 100 mg L 1 as CaCO3 (2 meq L 1) or lower. Perform the analysis when the concentration of magnesium from the first-stage process is 0.2 meq L 1. Assume the excess lime required to precipitate Mg(OH)2 is 0.8 meq L 1.

437

438

Theory and Practice of Water and Wastewater Treatment

From Eq. (15.9), xˆ

Mg Mg

f 0

Mg Mg

1

ˆ

1

1:0 2:0

0:2 ˆ 0:444 0:2

Because the alkalinity exceeds the sum of the concentrations of the hardness ions, the raw water calcium carbonate hardness (CH) is 3.60 meq L 1 and the magnesium CH is 2.00 meq L 1. There is no need to add Na2CO3 because there is no NCH. The final concentration of calcium is 2:00

1:00 ˆ 1:00 meq L

1

There are two constraints in this problem: 1) To precipitate Mg(OH)2, lime must be added in excess of the carbon dioxide and alkalinity to result in free hydroxyl ions in the first stage. 2) There must be at least 0.8 meq L 1 excess OH to reduce Mg2+ to 0.2 meq L 1. The total equivalents of lime required is the amount required to remove CO2, CaCH, and MgCH plus excess. CO2. From Eq. (15.1), 0.40 eq L 1 CaCH. From Eq. (15.2), 3.60 1.00 = 2.60 meq L 1 MgCH. From Eq. (15.4), 2(2.00 1.00) = 2.00 meq L 1 Excess for Mg removal. The usual excess is 0.6 meq L 1. The lime requirement is 0.40 + 2.60 + 2.00 + 0.6 = 5.60 meq L 1 based on the total flow. The fraction of flow entering the first stage is 1 0.444 = 0.556. The dose of lime (LD) to be added to the flow entering the first stage is 0:556 LD ˆ 5:60 meq L

1

or LD ˆ 5:60 meq L

1

=0:556 ˆ 10:07 meq L

1

Because the CO2 and HCO3 concentrations sum to 6.20 meq L 1, there is 4.07 meq L which satisfies constraint 1. After addition of lime, but before reaction, ‰OH Š ˆ 10:07 meq L

Ca2‡ ˆ 10:07 ‡ 3:60 ˆ 13:67 meq L 1 ;

1

of OH ,

1

After reaction of OH with CO2 and HCO3 , CO23

ˆ 0:40 ‡ 2…5:80† ˆ 12:00 meq L

1

CaCO3 is precipitated. CO23 is limiting and its concentration is 0.60 meq L formation. Ca2‡ ˆ 13:67

…12:00

0:60† ˆ 2:27 meq L

1

after precipitate

1

After reaction of Mg2+ with OH , Mg2‡

1

ˆ 0:20 meq L

‰OH Š ˆ 10:07

0:40

1

5:80

…2:00

0:20† ˆ 2:07 meq L

1

This is considerably in excess of the 0.8 meq L 1 constraint. It is probable that more than 1.8 meq L 1 of Mg2+ will be removed, but the calculations will be carried on assuming [Mg2+]1 = 0.20 meq L 1.

15 Physical–Chemical Treatment for Dissolved Constituents

After mixing the first stage and bypassed flows, the following concentrations result: 0:556…2:07† ‡ 0 ˆ 1:15 meq L 1 1 0:556…2:27† ‡ 0:444…3:60† ˆ ˆ 2:86 meq L 1

‰OH Š ˆ Ca2‡

HCO3 ˆ 0:444…5:80† ˆ 2:58 meq L

1

Mg2‡ ˆ 0:556…0:2† ‡ 0:444…2:00† ˆ 1:00 meq L

1

‰CO2 Š ˆ 0:444…0:40† ˆ 0:18 meq L 1

CO23

1

ˆ 0:556…0:60† ˆ 0:33 meq L

1

OH will react with CO2 and HCO3 to produce CO23

ˆ 0:33 ‡ 0:18 ‡ 2…1:15

0:18† ˆ 2:45 meq L

1

CO23 is limiting again. The final concentrations of all species (meq L 1) in the water are [Ca2+] = 1.01; [CO23 ] = 0.60; [Mg2+] = 1.00; [OH ]  0. Because of operation of the process wherein all of the lime is added at the first stage, a sufficient excess for Mg(OH)2 removal results, and it is not necessary to add the Mg(OH)2 excess into the lime dose. If this water were to be treated in a conventional process, the lime requirement would be Ca…OH†2 ˆ 0:40 ‡ 5:80 ‡ 2…2:00

1:00† ‡ 0:80 ˆ 9:00 meq L

1

to be added to the whole flow to meet the effluent magnesium objective. Calcium would be removed to 0.6 meq L 1, which is better than the requirement, but constraint 1 necessitates this condition. The lime savings is 3.4 meq L 1. Also, the effluent from the conventional process would contain 0.8 meq L 1 of OH , requiring CO2 or other neutralizing agent.

15.2.2 Bar Graphs The solution of softening problems is conveniently assisted by using bar graphs, which show the concentrations of the species involved at various stages during the process. On a bar graph, all concentrations are expressed in meq L 1. The first step in a water-softening analysis is to measure the concentrations of the major cations and anions. After the concentrations are converted to meq L 1, the sums of the cation and anion concentrations should be approximately equal; if not, there has been an error in analysis or computations, or a significant ion has not been accounted for. The use of bar graphs is illustrated by an example. Consider the water with the concentrations of major ions listed in Table 15.4. Usually, only the ions listed in Table 15.4 are significant in natural waters. The bar graph corresponding to the raw water is shown in Figure 15.2a. Carbon dioxide is given as the first component. It is not a cation or an anion. The cations are arranged in the order of Ca2+, Sr2+, and Mg2+, which is followed by other monovalent cations in any order. The anions are arranged with OH , CO23 , HCO3 , followed by SO24 and other monovalent anions in any order. The hypothetical concentrations for Ca(HCO3)2, CaSO4, MgSO4, and NaCl can be determined easily and are given on the graph for this water. The bar graph provides an easy check on the total cation and anion charge balance. The sum of the positive ions should be within 0.1–0.2 meq L 1 of sum of the negative ions. It is desired to attain hardness removal to the practical limits of 30 mg L 1 of CaCO3 (0.6 meq L 1) and 10 mg L 1 of MgSO4 (0.2 meq L 1), using a split recarbonation method.

439

440

Theory and Practice of Water and Wastewater Treatment

Table 15.4 Water components. Constituent

Concentration (mg L − 1)

Equivalent weight

meq L − 1

CO2

9.5

22

0.43

Ca2+

98

20

4.90

2+

27

12.2

2.20

Na+

6.5

23

0.28

HCO3

281

61

4.60

SO24

120

48

2.50

Cl

6.1

35.5

0.18

Mg

Figure 15.2 State of the water at stages during treatment. (All numerical values are concentrations in meq L 1.)

15 Physical–Chemical Treatment for Dissolved Constituents

The lime requirement is equal to the CO2, Ca(HCO3)2, and Mg2+ concentrations plus the excess lime (only when Mg removal is desired). The lime requirement is determined from an examination of Eqs. (15.1)–(15.6) and the composition of the water. In this case, it is Ca…OH†2 ˆ CO2 ‡ Ca…HCO3 †2 ‡ Mg2‡ ‡ excess ˆ 0:43 ‡ 4:60 ‡ 2:20 ‡ 1:25 ˆ 8:48 meq L

1

The excess lime was chosen to be 1.25 meq L 1, which is somewhat higher than the more typical value of 0.7 meq L 1. The excess lime requirement will be dictated by trial and error for the water and treatment times in each situation. This amount of lime will remove the CO2, and Ca2+ or CO23 will be removed down to the level of 0.60 meq L 1, depending on which of these ions is limiting. Likewise, Mg2+ is removed only to a level of 0.2 meq L 1. The state of the water after addition of the lime is shown in Figure 15.2b. Figure 15.2c shows the state of the water after reaction of OH with CO2 and HCO3 . Note that the total alkalinity has not changed, but the forms of alkalinity have changed. After precipitation of the remaining CaCO3 and Mg(OH)2, the water is in the state shown in Figure 15.2d. Note that the cation and anion orders given previously are maintained throughout all of the bar graphs. The water in Figure 15.2d is the effluent from the first stage of the process. Now CO2 is added to convert the excess OH (1.45 meq L 1 from excess lime and Mg2+ that was not precipitated) to CO23 , and Na2CO3 is added before the second mixing (reaction) vessel to precipitate the remaining Ca2+ as CaCO3 in the second settling basin. The Na2CO3 requirement is Na2 CO3 ˆ 4:35

1:45

0:60 ˆ 2:30 meq L

1

Figure 15.2e shows the water after recarbonation and addition of Na2CO3. After final precipitation of CaCO3 the water is in the state shown in Figure 15.2f. Recarbonation or acid addition to achieve the desired pH for stabilization of the water may now be performed. Lime Recovery and Sludge Reduction

Softening sludges, generated in basins that are separate from primary sedimentation basins, contain only CaCO3 and Mg(OH)2. Softening processes may be operated to exclusively remove CaCO3 in one basin. Lime is readily recovered from calcium carbonate sludge by heating it to drive off CO2. CaCO3

Δ

! CaO ‡ CO2

(15.10)

Lime recovery reduces the amount of lime required as well as decreases the quantity of sludge. For waters with a high alkalinity content, each mole of calcium removed produces 2 mol of calcium carbonate as shown in Reaction (15.2). Not all sludge will need to be treated for lime recovery to sustain the process.

15.3 Corrosion Prevention in Water Supply Systems Water leaving the treatment plant should not be corrosive to the pipes in the distribution system or in households. Pipes in the water supply system are made from a variety of materials. Pipes in the distribution system may be made from cast iron, ductile iron, or plastic (polyvinyl chloride). Pipes in households are usually made from copper, although plastic piping is becoming more common. Fittings and fixtures can be made from brass, stainless steel, and other materials. Lead can be found in joint solder in older households.

441

442

Theory and Practice of Water and Wastewater Treatment

Corrosion tends to accelerate at pipe deformities, impurities, or nonuniformities in the pipe, and joints where improper welding, or soldering techniques (in households) were used. Thus failure can occur in very young pipes primarily due to these reasons. It is a daunting task to minimize corrosion of the many different metal pipes in an entire water supply line, particularly given the complex nature of corrosion of any metal. Inert coatings, such as plastic, can protect a pipe from corrosion. Various constituents in water influence the rate of corrosion; the more significant chemical factors that influence the rate of corrosion are given in Table 15.5. Water treatment plant operators should consider these factors in controlling the treatment process to provide safe water that will be less corrosive in the distribution system. Beyond the adjustment of chemical factors, there are a number of measures that can be taken to retard the rate of corrosion. Protective coatings. Materials that are subject to corrosion should be coated to provide a barrier to electron transfer. The surfaces of metals that will be cathodic should be coated to minimize opportunity for the circuit to be completed. Corrosion-resistant alloys. These hybrid metals have been developed by metallurgists and chemists to strongly hold metal ions within the crystalline structure of the metal. Thus, they are very stable and resistant to oxidation but are naturally more costly than simpler materials. Table 15.5 Effects of water constituents on corrosion. Factor

Effect

Alkalinity (buffering capacity; see Chapter 3)

Alkalinities may help form protective coating; helps control pH change. Low to moderate alkalinity reduces corrosion of most materials. High alkalinities increase corrosion of copper and lead

Ammonia

Ammonia may increase the solubility of some metals such as copper and lead through the formation of complexes

Chloride and sulfate

High levels of chloride or sulfate increase corrosion of iron, copper and lead

Chlorine residual

Chlorine residual increases metallic corrosion, particularly for copper, iron and steel

Copper

Copper causes pitting in galvanized pipe

Dissolved oxygen

Dissolved oxygen (DO) increases the rate of many corrosion reactions

Hydrogen sulfide

Hydrogen sulfide increases corrosion rates

Orthophosphates and polyphosphates

These agents form protective films with some metals. Orthophosphate can be used for both copper and lead control; polyphosphates are effective only for copper corrosion control

Natural color, organic matter

Organic matter may decrease corrosion by coating pipe surfaces. Some organics can complex metals and accelerate corrosion

Silicates

Silicates are another inhibitor that are most effective in passivating ironcontaining pipes

pH

Low pH may increase corrosion rate; high pH may protect pipes and decrease corrosion rates or could cause release of zinc from brass fittings

Total dissolved solids (TDS)

High TDS increases conductivity and corrosion rate

Source: From Schock (1990), http://www.corrosion-doctors.org/Household/water-treatment.htm (accessed April 2016), http://www.thewatertreatments.com/corrosion/corrosion-control-methods (accessed April 2016), Lytle et al. (2012), and Edwards et al. (1996, 2002).

15 Physical–Chemical Treatment for Dissolved Constituents

Inhibitors. Inhibitors are substances that react in water to form a compound that lays down a protective barrier to electron transfer. Anodic inhibitors migrate toward anodic sites and are oxidized, forming a film that is relatively impenetrable. An example is chromate ion, which has an overall reaction as follows: 2CrO24

6e → Cr2 O3

The intermediate reactions involved are not definitely known, but the oxide does form a layer of insulation. Cathodic inhibitors stop the flow of H+ or O2 to the cathodic site. Aluminum is fairly low on the EMF scale (Table 2.2), yet it does not appear to corrode. This is because it reacts with oxygen to form a thin film of aluminum oxide (effectively, a corrosion product) that protects the aluminum from further oxidation. Other metals form surface films of varying composition that passivate the metal from further corrosion, for example, green copper oxide. Orthophosphates, polyphosphates, and silicates (Table 15.5) are inhibitors. Pure noble metals. The noble metals are high in the EMF series and tend to be cathodic with respect to the surrounding environment. However, they are expensive, and most of them are not rigid enough for structural support. In any case, they will have some imperfections in molecular structure that will tend to cause corrosion. Copper is one exception that makes it the most common material for residential pipes. The USEPA implemented a lead and copper rule in 1991 (USEPA 1991) for action to be taken when lead and copper concentrations exceed guidelines. Copper is an essential element but excessive amounts can lead to some health and aesthetic problems. In Ottawa, ON the failure rate of household lines is roughly 90 per 230 000 services (most of which are copper) (City of Ottawa, 2016). Corrosion is complex, and the prescription of remedies to retard corrosion are not well understood with many seemingly contradictory results in the literature that illustrate the illusory nature of the problem. The best recommendation is to measure corrosion in a system through coupon or other studies before and after any preventative measure has been implemented (ASTM 2014). A coupon is a thin strip of metal inserted into a pipe for a few months or more to measure its change in weight and observe other characteristics. Measurement of metals, such as iron, copper, and lead, in water at various locations also provides corrosion information. 15.3.1 The Langelier Index Misconception Preceding modern coatings, it was thought that adjusting water to conditions that would create and preserve a thin film of calcium carbonate in piping would provide a barrier to corrosion. The pH, [Ca2+], and alkalinity content of water are adjusted to the calcium carbonate saturation equilibrium value at the temperature and ionic strength of the water. Langelier (1936) performed the development given below to determine the saturation pH required for the alkalinity and calcium content of the water. Water that has a pH below the saturation pH is known as an aggressive water (purportedly allowing corrosion to occur unimpeded); water with a pH at or above the saturation pH was described as conditioned or stable water. Based on the logic of a protective coating, the Langelier Index (LI) or Langelier Saturation Index (LSI) became entrenched as a norm for water plant managers to provide wholesome water to their communities. Numerous studies have now discredited the LI as an indicator of susceptibility of any pipe material to corrosion (McNeill and Edwards 2001; Edwards et al. 1996). In fact, for copper pipes, the LI can be contraindicative of the corrosiveness of a water to this material. Nevertheless, the LI remains in use by many water utilities as an indicator of

443

444

Theory and Practice of Water and Wastewater Treatment

corrosivity of a water. The LI development is presented here to inform the reader (and opportunity to practice some chemistry). The chemical reactions involved are the solubility product relation for calcium carbonate and a carbonate equilibrium expression. Bicarbonate alkalinity will be the predominant form of alkalinity at the pH of treated water. The two equations can be added to find a single equation relating [Ca2+], pH([H+]), and alkalinity as follows: Ca2‡ ‡ CO23 ⇆ CaCO3 …s†

1=K sp

(15.11)

K2

(15.12)

HCO3 ⇆ H‡ ‡ CO23

Ca2‡ ‡ HCO3 ⇆ H‡ ‡ CaCO3 …s†

(15.13)

To accurately determine the pH required to precipitate CaCO3, activity coefficients must be incorporated into the equilibrium expressions [see Eq. (2.2)]. The equilibrium expressions for the above equations are For Eq. (15.11): Ksp = γ Ca2‡ [Ca2+] γ CO23 [CO23 ] For Eq. (15.12): K 2 ˆ

γ CO23 CO23 γ H‡ ‰H‡ Š γ HCO3 HCO3

Therefore, the equilibrium expression for Eq. (15.13) is Kˆ

2‡ K sp γ Ca2‡ Ca γ HCO3 HCO3 ˆ K2 γ H‡ ‰H‡ Š

(15.14)

A pH meter actually measures the activity of the hydrogen ion. pHmeter ˆ log γ H‡ H‡ Defining the pH corresponding to the saturation condition as pHs, from Eq. (15.14), pHs ˆ log γ H‡ H‡ ˆ log K ˆ log K sp

log K 2

log γ Ca2‡ Ca2‡

log Ca

2‡

log γ HCO3 HCO3

log HCO3

log γ Ca2‡ γ HCO3

Using the “p” notation, the definition of alkalinity given by Eq. (3.29), and assuming that the carbonate ion concentration is negligible, pHs ˆ pK 2

pK sp

log Ca2‡ ‡ S

log Alkalinity

(15.15)

where S is a salinity correction factor equal to log γ Ca2‡ γ HCO3 . (Note that [Ca ] and [Alkalinity] are in mol L 1 and alkalinity is assumed to be [HCO3 ].) The factor S adjusts the equation for the true activities of the ions in the equilibrium expressions. Equation (15.15) is valid in the pH range of 6.0–8.5 in which most treated waters fall (AWWA Joint Task Group 1990). This equation is based on the assumption that bicarbonate ion is the only significant alkalinity ion. Equation (15.15) should be applied at the temperature of the treated water. The equilibrium constants, Ksp and K2, and activity coefficients are functions of temperature. Table 15.6 gives values of S and the equilibrium constants over a broad temperature range. Note that pKsp values in Table 15.5 are based on the formation of calcite, which is one crystalline form of CaCO3. Other isomorphs of CaCO3 that may form are aragonite and veterite. The solubility products for these other forms are slightly different from that for calcite and their formation changes pHs. This is a complicating factor in any pHs calculation. However, the AWWA Joint Task Group (1990) points out that the most common form of CaCO3 in fresh waters is indeed calcite. 2+

15 Physical–Chemical Treatment for Dissolved Constituents

Table 15.6 Equilibrium constants and salinity factors for saturation index. Temperature (°C)

pK2a)

Sc)

pKspa),b)

TDS (mg L − 1) 50

150

400

1000

1500

5

10.55

8.39

0.0825

0.137

0.210

0.300

0.345

10

10.49

8.41

0.0832

0.138

0.211

0.303

0.348

15

10.43

8.43

0.0838

0.139

0.213

0.305

0.351

20

10.38

8.45

0.0845

0.140

0.215

0.308

0.354

25

10.33

8.48

0.0854

0.142

0.217

0.311

0.358

30

10.29

8.51

0.0861

0.143

0.219

0.314

0.362

35

10.25

8.54

0.0869

0.144

0.221

0.318

0.366

40

10.22

8.58

0.0879

0.146

0.224

0.322

0.370

45

10.20

8.62

0.0888

0.148

0.226

0.325

0.375

50

10.17

8.66

0.0898

0.149

0.229

0.329

0.379

60

10.14

8.76

0.0919

0.153

0.235

0.337

0.389

70

10.13

8.87

0.0941

0.157

0.241

0.346

0.400

80

10.13

8.99

0.0965

0.161

0.247

0.356

0.411

90

10.14

9.12

0.0990

0.165

0.254

0.366

0.423

a) From Plummer and Busenberg (1982).

b) For calcite as recommended by AWWA Joint Task Group (1990).

c) Based on Eq. (1.8) for ionic strength and the activity expression given by AWWA Joint Task Group (1990).

Source: Adapted from AWWA Joint Task Group (1990).

The parameters involved are all readily assessed by simple lab procedures. It is also possible to determine pHs for a water by keeping the water in contact with pure CaCO3 overnight, which is sufficient time to establish equilibrium conditions. Then the pH, which will be pHs, should be measured. If the pH of the water is less than pHs, there will be no deposition of CaCO3. If the pH of the water is greater than pHs, then CaCO3 can be deposited in pipes. The value of pH pHs (pH is the actual pH of the water being tested) is called the LI or saturation index LI ˆ pH

pHs

(15.16)

The common practice is to adjust waters to a LI value of +0.2 by addition of the appropriate alkalinity or acidity agent. Lime changes both calcium and alkalinity concentrations. Other agents that can be used to adjust water to the desired conditions are Na2CO3, CO2, and strong acids or bases such as HCl and NaOH. These agents will change the ratio of H2 CO∗3 and HCO3 from which the new pH of the water can be calculated with Eq. (3.26b). The LI is only a qualitative indication of the amount of potential CaCO3 deposition. A larger value of the index does not necessarily mean that more CaCO3 will deposit and, at the extreme case of a pH greater than pK2 for carbonic acid, an under-saturated solution will yield a positive index value. Snoeyink and Jenkins (1980) and Rossum and Merrill (1983) offer a comprehensive discussion of the index, pointing out other deficiencies.

445

446

Theory and Practice of Water and Wastewater Treatment

There are a number of other indexes of a similar nature to the LI (Rossum and Merrill 1983; AWWA Joint Task Group 1990). These indices relating to calcium carbonate precipitation potential, like the LI, should not be used as indicators of corrosion potential (McNeill and Edwards 2001). The modified Caldwell–Lawrence approach (Caldwell and Lawrence 1953; Merrill 1978), which is beyond the scope of this text, offers another approach to quantitatively estimate the extent of potential precipitation and hardness removal. Other indexes or methods to quantify the deposition potential of CaCO3 require computer code or extensive calculations. These indexes usually consider more chemical equilibria, particularly the formation of com­ plexes, and so produce more accurate results. Corrosion results in the addition of metals to water and wastewater. The bulk of metals in a wastewater is removed with the sludge in a wastewater treatment process. Another benefit of corrosion control is a reduction in the metal content of sludges, which improves the reusability of the sludge from the wastewater treatment plant (Kuchenrither et al. 1992). Example 15.3 Saturation Index The calcium concentration of a water is 42 mg L 1, and the alkalinity concentration is 60 mg L 1 as CaCO3. TDS have been measured at 120 mg L 1. The pH of the water is 7.73, and its temperature is 12 °C. Find the SI of this water. Equation (15.15) is required to calculate pHs. The values for pK2, pKsp, and S are obtained from Table 15.6. At the given values of temperature and TDS, it will be necessary to interpolate for each parameter. The values for pK2, pKsp, and S are the following: pK 2 ˆ 10:49 ‡

…12 …15

10† …10:43 10†

pK sp ˆ 8:41 ‡

…12 …15

10† …8:43 10†

S 12;50 ˆ 0:0832 ‡ S 12;150 ˆ

…12 …15

S 12;120 ˆ S 12;50 ‡

…12 …15

8:41† ˆ 8:42

10† …0:0838 10†

10† …0:139 10† …120 …150

10:49† ˆ 10:47

0:0832† ˆ 0:083 44

0:138† ˆ 0:1384

50† S 12;150 50†

S 12;50 ˆ 0:08344 ‡

…70† …0:1384 …100†

0:08344† ˆ 0:122

Converting the calcium and alkalinity concentrations to M: Ca2‡ ˆ 42 mg L

1

1 mol 40 g

Alkalinity ˆ 60 mg CaCO3 L

1

1g 1000 mg 61 g HCO3 50 g CaCO3

ˆ 1:05  10 1 mol 61 g HCO3

3

M 1g 1000 mg

ˆ 1:20  10

3

M

Substituting these values into Eq. (15.15), pHs ˆ 10:47

8:42

log 1:05  10

3

‡ 0:122

log 1:20  10

3

ˆ 8:07

and SI ˆ 7:73

8:07 ˆ 0:34

The water is undersaturated, and it will be necessary to add a strong base or Ca2+ [Ca(OH)2 accomplishes both] to raise the SI.

15 Physical–Chemical Treatment for Dissolved Constituents

15.4 Iron and Manganese Removal Iron and manganese are discussed together because they commonly occur together in raw waters, and the problems resulting from them and their methods of treatment are similar. Iron and manganese are minerals that cause staining of plumbing fixtures and laundered clothes as well as producing distinct tastes and odors in a drinking water. Iron and manganese also contribute to the hardness of a water. These are aesthetic problems; there is no health risk associated with excessive amounts of iron and manganese. The WHO (2011) standard for acceptable concentrations of manganese is 0.4 mg L 1; there is no standard for iron. The solubilities of iron and manganese are primarily controlled by their oxidation state. The redox reactions are the following: Fe3‡ ‡ e → Fe2‡ Mn

4‡

‡ 2e → Mn

(15.17a) 2‡

(15.17b)

The higher oxidation states of both iron and manganese are soluble to an insignificant degree in waters in normal pH ranges. At these oxidation states, the precipitates of Fe2O3 and MnO2 form. The iron oxide (rust) has a reddish-brown color and the manganese dioxide has a brown to brownish-black appearance. In the presence of DO, iron and manganese are oxidized to their insoluble oxidation states of Fe(III) and Mn(IV) according to redox reactions (15.17a) and (15.17b). Reducing (anaerobic) conditions cause the dissolution of these minerals; hence, groundwaters are particularly susceptible to excess concentrations of them. Also, in the bottom sediments of streams and lakes, anaerobic conditions exist and precipitated organic matter may release iron and manga­ nese. Redox reactions are slower than acid–base reactions. Environmental conditions such as pH, temperature, complexing agents, and other factors influence the rate of a redox reaction. Sometimes iron and manganese are found in aerated surface waters because the rate of conversion of soluble iron and manganese to their stable, insoluble forms is slower than their rate of production or input from the sources. The treatments for iron and manganese removal are based on accelerating their rate of oxidation. Manganese is more difficult to remove than iron; its oxidation kinetics are slower. Copper sulfate catalyzes the oxidation of manganese. Contact processes form MnO2, which also catalyzes the oxidation of Mn(II). Iron and manganese precipitates are colloidal in nature; therefore, they settle slowly. Rapid sand filtration as well as membrane ultrafiltration (UF) with a molecular weight cutoff (MWCO) at 30 000 Dalton (Da) or lower can remove colloidal as well as larger Fe and Mn precipitates (Carlson et al. 1997). A Dalton is one-twelfth the mass of carbon­ 12, about the mass of a proton or a neutron. A 0.1 μm opening is roughly equivalent to 100 kDa. Addition of oxidizing agents is a commonly used method for iron and manganese removal. Chlorine or permanganate are the chemical agents most often used. Other oxidizing agents such as chlorine dioxide and ozone are also effective. Chloramines will not convert ferrous iron to ferric iron or manganese to the +4 state. The overall chemical redox reactions are obtained by combining the appropriate half-reactions from Table 1.3. The redox reactions using per­ manganate as the oxidant are given in Eqs. (15.18) and (15.19a). 3Mn2‡ ‡ 2MnO4 ‡ 4OH → 5MnO2 …s† ‡ 2H2 O 3Fe

2‡

‡

‡ MnO4 ‡ 4H → 3Fe

3‡

‡ MnO2 ‡ 2H2 O

(15.18) (15.19a)

The Fe3+ reacts with alkalinity in water to form Fe(OH)3: Fe3‡ ‡ 3OH → Fe…OH†3 …s†

(15.19b)

447

448

Theory and Practice of Water and Wastewater Treatment

The overall reaction for iron removal is 3Fe2‡ ‡ MnO4 ‡ 5OH ‡ 2H2 O → 3Fe…OH†3 …s† ‡ MnO2

(15.20)

From the reactions, it is seen that the oxidation of iron with permanganate is favored by a low pH and the oxidation of manganese is favored by a high pH. However, a high pH favors the formation of Fe(OH)3 and the overall reaction for iron removal is favored by a high pH. When potassium permanganate is used as the oxidant, the formation of pink water is possible if doses are too high. The amount of permanganate required may be less than the stoichiometric amount because of the catalytic action of MnO2 (pyrolusite) on the oxidation of both soluble iron and manganese. Also, applying permanganate based on the total amount of Fe or Mn in the water to be treated can be counterproductive as the amount of oxidant is in excess of the requirement and adds Mn to the water (Carlson et al. 1997). The amount of soluble Fe and Mn can be determined by fractionation of the water. Fractionation of water into soluble, colloidal, and larger Fe and Mn precipitates is achieved by first passing water through a 0.2 μm filter to remove particulates, then passing it through a 30 000 MWCO filter to separate colloidal from dissolved forms (Carlson et al. 1997). The reactions for oxidation of iron and manganese with oxygen, chlorine, and chlorine dioxide are as follows: 4Fe2‡ ‡ O2 ‡ 10H2 O → 4Fe…OH†3 ‡ 8H‡

(15.21)

2Fe2‡ ‡ Cl2 ‡ 6H2 O → 2Fe…OH†3 ‡ 6H‡ ‡ 2Cl Fe

2‡

2Mn

‡ ClO2 ‡ 3H2 O → Fe…OH†3 ‡ ClO2 ‡ 3H 2‡

(15.22) ‡

(15.23)

‡

‡ O2 ‡ 2H2 O → 2MnO2 ‡ 4H

(15.24)

Mn2‡ ‡ Cl2 ‡ 2H2 O → MnO2 ‡ 4H‡ ‡ 2Cl Mn

2‡

‡ 2ClO2 ‡ 2H2 O → MnO2 ‡ 2ClO2 ‡ 4H

(15.25) ‡

(15.26)

Reducing agents other than iron and manganese and the presence of natural organic matter (NOM) will increase the demand for oxidant beyond the stoichiometric requirements given in the above equations. Chemical oxidizing agents are usually added near the beginning of a treatment process to allow them to react with iron and manganese in subsequent unit operations. It may be necessary to provide a holding tank for the reaction, depending on which operations follow the application of the agent. Coagulation, flocculation, and sedimentation will remove some of the iron and manganese precipitates formed from addition of oxygen or chemical agents. Filtration is almost always necessary to remove the fine iron and manganese precipitates to a satisfactory degree, regardless of whether sedimentation is used.

15.4.1 Greensand Greensand (also known as glauconite) is a granular, naturally occurring, mineral with a greenish color. This zeolite has a composition of (K,Na,Ca)1.2–2-(Fe3+,Al,Fe,Mg)4Si7–7.6Al1–1.4O2(OH)4? nH2O. Iron and manganese can be removed by greensand medium that has a manganese coating. Adsorbed manganese catalyzes the oxidation of Fe(II) and Mn(II). The mechanism of removal is a combination of sorption and oxidation. Greensand is regenerated with potassium per­ manganate. The regenerant removes the sorbed, oxidized iron and manganese and restores the removal capability of the greensand. The reactions for oxidation and regeneration are the

15 Physical–Chemical Treatment for Dissolved Constituents

following: Fe2‡ →Z Mn2‡

Z

MnO2 ‡

Z

Mn2 O3 ‡ KMnO4 → Z

Mn2 O3 ‡

Fe3‡ Mn3‡ Mn4‡

MnO2

(15.27) (15.28)

where Z is the greensand. Calcium and magnesium are not removed in a greensand filter. However, in addition to iron and manganese, greensand can remove hydrogen sulfide, phenols, and radium-226 (Brinck et al. 1978; Sayell and Davis 1975). Manganese greensand beds are usually placed in pressure filters operated in the downflow mode. It is recommended that the pressure differential across the bed not exceed 83 kPa (Sayell and Davis 1975). When iron or sulfide is the primary contaminant to be removed, potassium permanganate and other oxidants are continuously added to the influent to the greensand filter. Backwashing is required only infrequently; backwashing frequency will depend on the influent manganese concentration. The manganese dioxide coating of the greensand will grow when manganese concentrations are significant because some of the Mn(II) will be oxidized and retained by the greensand. Plugging of the greensand medium can be largely eliminated by placing a rapid sand filter ahead of the greensand filter and feeding the potassium permanganate upstream of the rapid filter to remove some of the Fe and Mn precipitates and other SS in the filter. Waters containing 18 mg L 1 Fe(II) or more can be easily treated in a greensand filter. Proper process operation produces effluents with iron concentrations not exceeding 0.1 mg L 1 and manganese concentrations less than 0.01 mg L 1 (Sayell and Davis 1975). 15.4.2 Aeration Aeration is another alternative for the removal of iron and manganese. The aeration configura­ tion depicted in Figure 12.3d is effective for oxidation of Fe(II). The reaction is given by Eq. (15.21). Manganese is more difficult to oxidize than iron. Addition of coke coated with oxides to the trays will effectively catalyze the oxidation of Mn(II) to its insoluble forms. Sedimentation and filtration will remove the iron and manganese precipitates formed in the aeration device. 15.4.3 Sequestering Iron and Manganese Instead of removing iron and manganese, sequestering agents may be added to a water. They bind the iron and manganese into polymeric colloidal complexes that prevent them from forming color or turbidity within typical detention times in the distribution system. Sodium silicates, phosphates, or polyphosphates are commonly used sequestering agents. Chlorine must be added simultaneously with sodium silicate to oxidize iron and manganese (Robinson et al. 1992). Iron is most effectively sequestered with silicate; manganese is less susceptible. Silicate doses in the range of 5–25 mg L 1 as SiO2 are normal. 15.4.4 Biological Removal of Iron and Manganese All of the processes previously discussed for iron and manganese removal are physical–chemical processes. Biological iron and manganese removal is an alternative process that is quite complex, and only a brief description is given here (see Mouchet 1992, for more details). The process uses

449

450

Theory and Practice of Water and Wastewater Treatment

filters that support autotrophic iron- and manganese-oxidizing bacteria. Although some bacteria can oxidize both species, others can oxidize only one of them, and the optimal conditions for iron removal are different from those for manganese removal. Two separate filters are usually required if both iron and manganese must be removed. The influent must be adjusted to redox potential and pH conditions that promote the activity of the respective groups of bacteria. Dissolved oxygen content of the influent is critical. Conditions that cause physical–chemical removal of iron or manganese will decrease the activity of iron and manganese bacteria and ultimately impair process performance. Nitrate has been demonstrated to promote the removal of manganese (Vandenabeele et al. 1995). A water must be fully nitrified before biological manganese removal can take place. Higher filtration rates in the range of 25–40 m h 1 are used in iron removal filters; typical filtration rates in manganese filters are 10–40 m h 1. The higher filtration rates in iron or manganese filters are achieved with coarse sand media with an effective size (d10) of 0.95–1.35 mm or higher. Backwashing frequency depends on the raw water quality, but the quantity of backwash water should be well below 1% of the product water. There are many bio-Fe or -Mn operations in France alone, and many other installations throughout Europe and elsewhere. The process greatly reduces the use of chemical agents and is claimed to reduce capital and operating costs (Mouchet 1992). Sludge produced from the process does not pose treatability problems.

15.5 Phosphorus Removal from Wastewater by Chemical Precipitation Phosphorus is usually the limiting nutrient for eutrophication in inland receiving waters for wastewater treatment plant effluents; therefore, phosphorus concentrations in effluents are controlled. Allowable phosphorus concentrations in wastewater treatment plant effluents are decreasing. Ontario has a limit of 1 mg P L 1, and Ohio has a limit of 0.3 mg P L 1. Biological wastewater treatment processes remove some phosphorus in the biological sludge that is formed in these processes, but the amount of P removal is normally inadequate for typical wastewaters. Agents used to precipitate dissolved phosphorus are salts of the metals calcium (lime; Ca(OH)2), iron (chloride or sulfate salts of Fe2+ and Fe3+), or aluminum (either alum, Al2(SO4)3
18H2O or sodium aluminate, Na2Al2O4). Pickle liquor, which is a waste product of the steel industry that contains ferrous iron in either a sulfuric or hydrochloric acid solution, may also be used. All of these metals form sparingly soluble salts with dissolved forms of phosphate. Besides forming precipitates with dissolved phosphorus, these agents also aid the coagulation–flocculation of SS, which enhances their settling. Any improvement in the removal of biological solids or other inorganic solids that contain phosphorus further contributes to total phosphorus (TP) reduction. Orthophosphate (dissolved inorganic phosphate) equilibrium reactions are given in Table 3.2. In the pH range of wastewaters (around 7), the dominant forms of orthophosphate are H2 PO4 and HPO24 . As the pH increases, the equilibrium moves toward formation of PO34 . Condensed phosphates (polyphosphates) are also present in fresh wastewaters, but microorganisms convert polyphosphates into orthophosphates in the sewers and biological treatment processes. The chemistry of phosphate precipitate formation is complex because of complexes formed between phosphate and metals and between metals and other ligands in the wastewater. Side reactions of the metals with alkalinity to form hydroxide precipitates are another factor to be considered. The common precipitates formed by the metals given above are given in Table 15.7 along with the optimal pH range for phosphate precipitation.

15 Physical–Chemical Treatment for Dissolved Constituents

Table 15.7 Phosphate precipitates. log Ksp

Metal

Precipitates

Al3+

Al3+ + PO34 ⇆ AlPO4

pH range

Comment

21

6–8.5

3Ca2+ + 2PO34 ⇆ Ca3(PO4)2

33

10

Produces lowest residual P concentrations. The alkalinity of the water determines the dose because of the formation of CaCO3

Ca2+ + HPO24 ⇆ CaHPO4

7.3

9.5

Residual P in the range of 1–2 mg L 1

32

6–8.5

There will be some oxidation of Fe2+ to Fe3+

26

6–8.5

Alx(OH)y(PO4)3, Al(OH)3 Ca

2+

Various calcium phosphates, e.g., Ca3(PO4)2, Ca5(OH)(PO4)3 CaCO3 Fe2+

3Fe2+ + 2PO34 ⇆ Fe3(PO4)2, Fex(OH)y(PO4)3, Fe(OH)2, Fe(OH)3

Fe3+

Fe3+ + PO34 ⇆ FePO4 Fex(OH)y(PO4)z, Fe(OH)3

Source: From Jenkins and Hermanowicz (1991), Stumm and Morgan (1996), and other sources.

The amount of phosphorus removal at different doses and pH values can be assessed in jar tests (Section 13.1). The dose of coagulant required is a function of pH. A typical curve for residual phosphorus versus dose is given in Figure 15.3. There is a clear trend in the figure, but scatter of the data is due to many factors. At a given dose of phosphorus-precipitating agent, the concentration of TP remaining in solution typically follows a V-shaped curve as illustrated in Figure 15.4. The amount of metal needed to remove phosphorus exceeds the stoichiometric amount at the low levels at which phosphorus is present in wastewater (4–8 mg P L 1), but effluent concen­ trations are commonly specified to be 1 mg L 1 or less to minimize environmental effects. Considering the general relation for a solubility product for species listed in Table 15.7, Hx PO34

x

ˆ

n

K sp ‰Mea‡ Š

(15.29)

b

where [Mea+] is the concentration of metal and n, a, and b depend on the metal.

Figure 15.3 Residual soluble P as a function of dose. Source: Adapted from USEPA (1987c).

451

452

Theory and Practice of Water and Wastewater Treatment

Figure 15.4 Typical TP–pH variation.

The inverse relation between phosphate species concentration and metal ion concentration dictates larger doses of the metal agent as smaller residual concentrations of phosphorus are required. There will also be a higher concentration of complexes formed at the higher metal doses. Lime can achieve the lowest residual phosphate concentration because of the formation of apatites. Apatites (calcium-, hydroxide-, and phosphate-containing compounds), such as Ca5(OH)(PO4)3 are highly insoluble; however, a high pH must be attained to realize their formation. Because calcium forms CaCO3 near pH 9.5 and pHs higher than 9.5 are required for significant apatite formation, the alkalinity of the water is the governing factor for the required dose of lime. Very high doses of lime are required, which makes this choice costly compared to other metal agents, although the lowest residual P concentrations can be achieved. For iron and aluminum salts, without jar tests, the general guideline for the metal dose to achieve total soluble residual P concentrations of 1–2 mg P L 1 is 1 mole of metal per mole of phosphorus (USEPA 1987a; Jenkins and Hermanowicz 1991). As the metal:P molar ratio is increased to 1.5 and 2, residual P concentrations decrease to 0.5 and 0.3 mg L 1, respectively. Szabo et al. (2008) found that in the pH range of 5.0–7.0, which is the common pH range for biological processes, aluminum and ferric salts were similar in their ability to precipitate phosphorus. Mixing intensity at the point of dosage was the most important factor influencing the kinetics of precipitate formation. Addition of metal coagulants to wastewater can have a significant impact on the amount of sludge produced. The common addition points for metals in biological treatment plants are before the primary clarifier or before the secondary clarifier (see Figure 9.4). In the physical– chemical plant depicted in Figure 9.6, phosphorus removal would occur in the primary clarifier. In a biological treatment plant, metal salt addition to the primary clarifier can result in sludge mass increases of 50–100% from this clarifier and an overall increase of 60–70% in the sludge mass produced by the plant. If the metal agent is added before the secondary clarifier, the increases in sludge mass are 35–45% from the clarifier and 10–25% for the whole plant (USEPA 1987b). These salts will also have an impact on sludge dewaterability. Chemical costs and sludge handling costs must be considered together for optimization. 15.5.1 Removal of Phosphorus by Chemically Reactive Species Phosphorus can be effectively removed by filtering wastewater through a medium that will chemically react with phosphate. Filter media include iron-rich and calcium-rich slag, calcium silicates, oil shale ash, zeolites, and biological materials such as shells, marine plant fibers, and water hyacinth straw (Pratt et al. 2012). Pratt et al. note that these media are usually inexpensive and available in various locales, making them effective alternative phosphorus removal options. They reviewed studies where various types of slag were used with 50–90% phosphorus removal; the majority of studies had removal above 80%. Media life is dictated by the phosphorus reaction/ adsorption sites on the media which dictates the media replacement rate.

15 Physical–Chemical Treatment for Dissolved Constituents

15.6 Removal of Arsenic and Metals Heavy metals are classic contaminants. Arsenic was a later addition to the list. These substances are captured but not transformed in the treatment processes; thus, the ultimate disposal must be evaluated to ensure that they do not cycle through the water system at harmful levels. 15.6.1 Metals Removal Many industries produce significant concentrations of metals in their effluents. Heavy metals are generally toxic, but their toxicity to different organisms varies widely. The hardness of water affects the availability of metals for biological uptake with a lower availability at higher hardness levels. Precipitation is the common treatment to remove metals. Examination of solubility products in Table 1.5 shows that metal hydroxides yield low concentrations of soluble metals. The highest pH is not necessarily the optimum pH for metal precipitation with hydroxide. Complex formation with water molecules, hydroxides, and other molecules called ligands keeps metals in solution, and this complicating phenomenon influences the pH at which the lowest concentration of the metal will be achieved. Complexation affects the optimum pH for precipitation of a metal with any anion that participates in acid–base reactions. Log-C versus pH diagrams, where C is the concentration of the metal, can show the contribution of each complex to the total dissolved concentration (Benjamin 2015). Software to simplify this task is now widely used; examples are MINEQL+ (http://www.mineql.com/) and Visual MINTEQ (https://vminteq.lwr.kth.se/). Lime is the most common precipitation agent; it or any other caustic agent is added to attain the pH of minimum solubility. Sulfide, carbonate, and phosphate are other anions that have lowsolubility products with metals. Doses significantly higher than the stoichiometric amount are usually required. The kinetics of precipitation are slow in the final stages and excess ions provide more contact opportunity, accelerating the rate of reaction. Also, precipitation agents are not specific to one metal, and other nontoxic metals will be precipitated along with the target species. In choosing a precipitation agent, consideration must be given to the presence of other metals or cations such as iron and silicon that may coprecipitate with the target metals. Excess sludge generation can be minimized by a judicious choice of the precipitation agent. The theoretical stoichiometry for precipitation reactions with the agents above are the following: a Ca…OH†2 Me…OH†a ‡ Caa=2 X 2 a NaOH Me…OH†a ‡ Naa X a 2 a‡ S ‡Z MeSa=2 ‡ ZX → (15.30) Me ‡ X ‡ 2

Me…CO3 †a=2 ‡ Naa X Na2 CO3

a Me…PO4 †a=3 ‡ ZX PO34 ‡ Z 3 where X is the stoichiometric amount of any anion associated with Mea+ and Z is the stoichiometric amount of any cation associated with S2 or PO34 . Adsorption and ion exchange, discussed in the following sections, are other means of metal removal. Activated carbon, oxides of aluminum, iron, and other metals, as well as silica, resins, clays, and synthetic or natural zeolites have varying affinities for metals. Microorganisms can be used to enhance the removal of metals. A Citrobacter species is able to promote the precipitation of metals to lower concentrations than dictated by the solubility

453

454

Theory and Practice of Water and Wastewater Treatment

product (Roig et al. 1997). In an immobilized, nongrowth state, the species produces large amounts of a phosphatase enzyme that cleaves inorganic phosphorus from an organophosphate compound. The enzyme is produced on the cell surface. The high concentration of inorganic phosphorus at their cell surface precipitates metal phosphates that bind to the cell wall at points where crystalline nucleation begins. The enzyme has a working pH range of 4.5–8.5 and temperature range of 2–45 °C. 15.6.2 Arsenic Removal Arsenic is a widely distributed metalloid element in the earth’s crust; it is toxic and a carcinogen. Bangladesh is the most notorious region for arsenic contamination; worldwide, more than 100 million people may be at risk from arsenic contaminated waters (Heinke 2009a). Arsenic is usually found in water in the As(III) and As(V) oxidation states (Heinke and Hutchinson 2009). The acidic nature of the common arsenic forms makes it highly soluble in water (Table 15.8). As(III) mostly exists in low-oxygen (reducing) groundwaters and hydro­ thermal waters; As(V) is more common in oxidizing groundwaters and surface waters. The four oxidation states of arsenic (+5, +3, 0, and 4) make it a versatile element that exists in many mineral formulations. It commonly substitutes for sulfur and phosphorus in metals (Heinke and Hutchinson 2009), but it is also found in many organic compounds. As(III) is more toxic than As(V) (Heinke 2009b). Sodium arsenite, a mixture of NaAsO2 (sodium meta-arsenite) and Na3AsO3 (sodium orthoarsenite) is a pesticide; arsenic also has various industrial uses including incorporation into the feed of chickens for human consumption (Heinke and Hutchinson 2009). For treatment, As(III) is usually oxidized to As(V), which is more readily removed by sorption, anion exchange, or precipitation/co-precipitation. Oxygen is not a very effective oxidizing agent, but chlorine and its derivatives, as well as ozone and permanganate, are (Heinke 2009b). Calcium, copper, iron, and zinc among other metals form relatively insoluble species with arsenate (Heinke and Hutchinson 2009). It also co-precipitates with or is sorbed onto calcium, iron, and manganese precipitates. Iron, aluminum, and manganese hydroxides are good adsorbents for arsenic compounds (Heinke and Hutchinson 2009). Sorption and co­ precipitation with iron hydroxides are some of the more common means of removing arsenic (Heinke 2009b). There are a number of treatment methods for arsenic removal; some are listed in Table 15.9. Conventional coagulation with alum or iron salts followed by filtration is used in a number of plants (USEPA 2000). In smaller plants, ion (anion) exchange (see Section 15.8) and activated alumina sorption/ion exchange are more common treatment methods. Ion exchange is not effective for removing As(III) because below a pH of 9, the species is largely undissociated and therefore uncharged. As(III) should be oxidized to As(V) before activated alumina treatment (USEPA 2000). Brine solutions are used to regenerate resins.

Table 15.8 Arsenic (III) and (V). As valencea)

pK1

pK2

Arsenic acid

H3AsO4

V

9.17

14.10

Arsenous acid

H3AsO3

III

2.20

7.01

a)

Arsine, H3As is the other nonzero oxidation state.

pK3

Salt

11.80

Sodium arsenite: Na3AsO3

Sodium arsenate: Na3AsO4

15 Physical–Chemical Treatment for Dissolved Constituents

Table 15.9 Some treatment methods for arsenic removal. As form

As(III) without pre-oxidation

Methods

Maximum removal capacity (mg As g − 1)

Iron hydroxide-coated sand filtration Manganese hydroxide-coated sand filtration Zirconium anion exchange resins

As(V)

Aluminum hydroxide co-precipitation followed by filtration Iron impregnated activated carbon Iron hydroxide-coated sand filtration Ion exchange Manganese hydroxide-coated sand filtration

As

Activated alumina

0.2–121a); 0.25–0.30 (or 156–219 g m 3 alumina)b)

Activated carbon

0.04–20a)

Greensand Lime softening Nanofiltration or reverse osmosis a) Data from laboratory studies.

b) Data from small plants.

Source: Adapted from Heinke (2009b) and USEPA (2000).

15.7 Advanced Oxidation Processes Advanced oxidation processes (AOPs) are used in both water and wastewater treatment. Recalcitrant, nonbiodegradable organics require stronger oxidizing agents than oxygen or chlorine to break them down to harmless endproducts. In general, these are radicals, which need to be created in situ, usually by a combination of processes. A radical is an atom or group of atoms with unpaired electrons. In a covalent bond, two electrons are shared. In a radical, there is at least one bond that only has one electron, that is, an unpaired electron in the bond. (Hydrogen without an electron is also a free radical.) Radicals are highly reactive; the lifetime of a hydroxyl radical is less than a second. The hydroxyl radical is a hydroxide ion without an electron, symbolized by OH*. Photolysis is a light-catalyzed reaction. An example of generation of the hydroxyl radical is through the photolysis (using ultraviolet (UV) irradiation) of hydrogen peroxide: H2 O2 ‡ hν → 2OH∗

(15.31a)

where hν represents the photon energy from UV. The hydroxyl radical then oxidizes any available donor (hopefully, the recalcitrant compound) by scavenging an electron from it: OH∗ ‡ e → OH

(15.31b)

Thus, hydrogen peroxide and UV irradiation are used together to provide one AOP alternative. This AOP is commonly used for removal of pesticides, geosmin, and other taste and odor compounds.

455

456

Theory and Practice of Water and Wastewater Treatment

Table 15.10 Advanced oxidation processes for water and wastewater applications. Processes

Applicationsa)

References

Titanium dioxideb) (TiO2) and UV

Dyes, landfill leachates, pharmaceuticals

Hassan et al. (2016), Garcia-Segura and Brillas (2017)

TiO2, UV, and O3

Phenols

Suzuki et al. (2015)

Persulfate and UV

Contaminated soils, groundwater, leachate

Deng and Zhao (2015), Tsitonaki et al. (2010)

Fenton reaction (Fe2+ and H2O2 at low pH)

Organic contaminants

Pignatello et al. (2006)

Ultrasound and O3

1,4 Dioxane in drinking water

Dietrich et al. (2017)

a) Each process may have many applications; examples from the recent literature are listed.

b) TiO2 is known as a semiconductor. When TiO2 particles are exposed to UV, positive holes with oxidative powers are

created in the valence band, and negative electrons with reductive powers with a reductive capacity at the conduction band. Other semiconductors, not as efficient as TiO2 for this purpose, include, CeO2, In2O3, MoO2, SiO2, V2O5, WO3, and ZnO (Chan et al. 2011). The semiconductors may also be doped (coated with another element such as Ag, Ce, Co, Fe, N, etc.) to enhance reactivity.

Reaction of the hydroxyl radical with organics proceeds according to a second-order reaction. r ˆ k‰OH∗ Š‰RŠ

(15.32)

where r is the rate of reaction, k is a rate constant, and R is an organic compound. Crittenden et al. (2012) provide rate constants for a number of compounds reacting according to Eq. (15.32). Reactions with organics usually proceed through more than one stage of intermediate products. Bicarbonate and carbonate species scavenge OH* radicals, reducing reaction rates with the target species; therefore, pH will be also be an important factor affecting reaction rate. UV irradiation plus ozone, being another AOP option, is effective at decomposing humic acids and other organic compounds. In one study, the destruction of dissolved organic substances with UV and ozone proceeded at least 10 times faster than in the presence of ozone alone (Kusakabe et al. 1990). Several other AOP processes, applications, and key references are summarized in Table 15.10.

15.8 Ion Exchange Ion exchange is the exchange of ions in a solution for other ions on a medium. Ion exchange is primarily used for hardness removal in water treatment by industries to minimize scaling. In wastewater treatment, it can be used for the removal of toxic metals or recovery of precious metals. There are many natural substances that are able to exchange ions. Chemists have developed many synthetic media that are highly specific for target ions and are very efficient. Synthetic resins are most commonly used because of their better performance. A resin is a three-dimensional hydrocarbon network to which functional groups (Section 4.3) are attached. These groups contain the exchangeable ions and are soluble in water. The exchange reaction is nR B‡ ‡ An‡ ⇆ Rn An‡ ‡ nB‡

(15.33) n+

where Rn indicates that n functional groups are coordinated to A .

15 Physical–Chemical Treatment for Dissolved Constituents

The equilibrium relation for the above equation is defined as a quotient, Q, or a selectivity quotient, Qs which is similar to an equilibrium constant. Qˆ

‰B‡ Š Rn An‡ ‰R B‡ Š An‡

Qs ˆ

(15.34a)

‰B‡ Šn Rn An‡ n ‰R B‡ Š An‡

(15.34b)

where [Rn An‡ ] and [R B+] are expressed either as (mol X) per gram of resin or as (mol X) per volume of resin. It is essential that the effective concentrations of Rn An‡ and R B+ be used in the above expressions; that is, activity coefficients for the operating conditions must be assessed because the concentration of the chemical species on the resin are no longer dilute. Ion exchange resins are designed to remove ions of a certain class. In general, there will be exchange of more than one ion species. The selectivity characteristics of resins that are available from manufacturers, and the presence and concentration of dissolved solids in the water, dictate the most favorable choice of a resin to achieve removal of the target species. Resins are subject to fouling. Iron and manganese can be problematic in some circumstances producing precipitates that coat resin particles irreversibly. Resins are good solid coagulants; therefore, SS concentra­ tion of the influent should be low (Sanks 1978). Influent turbidities less than 2 NTU are recommended. The total exchange capacities for a number of commercial resins ranged from 2.5 to 4.9 meq g 1 of resin on a dry basis (Flick 1991). Resins have different swelling characteristics and the total exchange capacity on a volumetric basis ranged from 1.0 to 4.0 meq mL 1. The saturation capacity of a resin can be measured simply by means of a titration. The procedure is as follows: 1) Regenerate the resin with acid (base). 2) Rinse the resin with distilled water or water that does not contain any ions that are exchangeable with the resin. This water may be prepared by passing it through an active resin. The rinse will remove any excess regenerating agent in the resin column. 3) Measure the resin volume. 4) Titrate the resin with base (acid). The exchange capacity of the resin is calculated from exchange capacity ˆ

N tV t volume of resin

(15.35)

where Nt is the normality of the titrant and Vt is the volume of titrant used. The residence time of water in ion exchangers and other packed-bed reactors is often characterized in terms of the empty bed contact time (EBCT), which is simply the depth of media divided by the superficial velocity or the total bed volume divided by the volumetric flow rate (V/Q). The theory and design of ion exchangers are the same as for adsorbers, which are described in the next section. 15.8.1 Activated Alumina Activated alumina is a granular natural zeolite that is preferentially selective for fluoride, arsenic, selenium, phosphate, and silica (Clifford 1990; Rubel and Woosely 1979; Trussell et al. 1980).

457

458

Theory and Practice of Water and Wastewater Treatment

The most common application of activated alumina has been for defluoridation. Trussell et al. (1980) reported the order of preference of activated alumina at pH 6.5 for these substances as follows: OH > H2 PO4 > F > H2 AsO4 > HSeO3 The pH is critical for the performance of activated alumina. A pH of 5.5 is generally reported as optimal for removal of fluoride, As(V), and Se(IV) (Rubel and Woosely 1979; Trussell et al. 1980). If the pH is greater than 9.5, activated alumina acts as a cation exchanger. An EBCT of at least 5 minutes with activated alumina particle sizes between 28 and 48 mesh is most commonly employed (Rubel and Woosely 1979). Sodium hydroxide at concentrations between 0.5% and 5% is used to regenerate activated alumina.

15.8.2 Ammonia and Nitrate Removal by Ion Exchange Wastewaters high in ammonia can be treated with ion exchangers using the natural zeolite clinophlolite (also called clinoptilolite), which exhibits unusual selectivity for ammonium ion in the presence of calcium, magnesium, and sodium ions (Eckenfelder and Argaman 1991; Sims and Hindin 1978). Other forms of inorganic nitrogen are not removed by the zeolite. The capacity of the exchanger is approximately 2 meq g 1 of zeolite. Regeneration of the zeolite can be accomplished with neutral sodium chloride or alkaline reagents such as calcium or sodium hydroxide. The bed may become unstable and decompose with high-pH regenerants, although high-pH regeneration may require as little as one-third of the regenerant (Sims and Hindin 1978). A 2% NaCl solution required 25–30 bed volumes for regeneration. Ammonia can be biologically removed or stripped from the regenerant. For drinking water treatment, nitrate removal by anion exchange is an option. In conventional strong base anion, nitrate-selective, resins, trimethylamines are the functional groups. Sulfate is a strong competitor to nitrate in these conventional resins. Replacing the trimethylamines with triethyl- or tributyl-amines greatly enhances the selectivity of the resin for nitrate (Liu and Clifford 1996) and is preferred for waters with sulfate concentrations in excess of 140 mg L 1. Regeneration of either conventional or nitrate-enhanced selective resins is done with sodium chloride solutions. For a complete regeneration of triethyl-amine and tributyl-amine resins, the ratios of equivalents of chloride to equivalents of NO3 –N (the “brine use factor”) were 21 and 17, respectively, compared to 27 for the trimethylamine resin, for a complete regeneration of resins treating high sulfate (>140 mg L 1) waters (Liu and Clifford 1996). Denitrification of the spent brine water can significantly reduce salt consumption and disposal costs of spent brine water. Partial regeneration of the resins, allowing for leakage of nitrate, but maintaining effluent concentrations below 10 mg L 1 NO3 –N, also reduces salt consumption.

15.9 Fluoridation and Defluoridation The fluoride objective for finished waters is around 1 mg F L 1 for most communities in North America (Section 8.6.2). For water deficient in fluoride, fluoride is added after filtration or after activated carbon treatment if this treatment is used. The common fluoridation agents are sodium fluoride, sodium silicofluoride (Na2SiF6), and fluorosilicic acid (H2SiF6). Fluorspar (fluorite) is a natural mineral form of primarily CaF2 but may include other minerals, particularly CaCO3 that influence the solubility of fluoride when fluorspar is dissolved. Fluorspar was suggested as an economical alternative to other commercial agents for fluoridation (Peng et al. 1996).

15 Physical–Chemical Treatment for Dissolved Constituents

Natural fluoride concentrations in water supplies (both surface and groundwaters) exceed desirable levels at various locations throughout the world. Groundwater concentrations are generally higher than surface water concentrations. In some cases, the concentrations can be quite high; for instance, in India, some water supplies contain F as high as 19 mg L 1 (Bulusu et al. 1979); in Africa, concentrations over 50 mg L 1 have been observed. Fluoride content in natural drinking water sources in Canada ranges from less than 0.01–4.5 mg F L 1 (Department of National Health and Welfare 1980). Defluoridation is costly. Methods for fluoride removal include chemical precipitation, electro­ dialysis, ion exchange, or ion exchange/adsorption processes. Activated alumina discussed above has been the most common agent used for defluoridation. Bone char, activated carbon, tricalcium phosphate, lime, magnesium oxide, and other anion exchangers also have been used or studied (Bulusu et al. 1979). Bone char, which is naturally high in calcium, is pulverized bone that has been carbonized by heating in the absence of air or it may be burnt in the presence of air. It has been commonly used, but it is more expensive than activated alumina. Its capacity is around 1000–1500 mg F L 1 of bone char (Bulusu et al. 1979). It is regenerated with caustic solution. Variable results and generally poor performance of activated carbon have been reported. Alum and lime are used to precipitate fluoride in a defluoridation technique developed in India and referred to as the “Nalgonda” technique (Nawlakhe and Paramasivam 1993). The amount of alum required increases with the amount of fluoride in the water. Alkalinity must be adjusted with lime (which is added before the alum) to ensure that the alum is precipitated or, when excess alkalinity is present, more alum must be added to achieve the desired removal. Table 15.11 provides doses found from many years of experience (Nawlakhe and Paramasivam 1993). Alum is required in excess and an excess of alkalinity must also be present to ensure that the alum is precipitated. There are some inconsistencies in excess alum and alkalinity trends; pilot studies should be performed with the water. The chemicals are added in a rapid mixing chamber followed by flocculation, sedimentation, and filtration. The reaction using alum and sodium

Table 15.11 Approximate alum dose (mg Al L 1) to obtain 1 mg L

1

fluoride levels.

Water alkalinity (mg CaCO3 L − 1)

Raw water (mg F − L − 1) 125

200

400

600

1000

1a)

2b)

3c)

1

2

3

1

2

3

1

2

3

1

2

3

2

12

3.5

61

18

9.5

103

25

16.5

264

33

24.5

419

42

33.5

769

3

18

0.9

30

24

6.9

72

33

15.9

222

42

24.9

372

62

44.9

661

32

6.4

30

4

38

12.4

197

49

23.4

336

76

50.4

586

5

49

14.9

138

58

23.9

288

82

47.9

555

6

58

15.4

91

8 10

76

33.4

191

98

55.4

469

91

31.3

113

116

56.3

374

137

60.3

263

Al dose [cols (1)] in mg Al L 1 calculated from Nawlakhe and Paramasivam (1993). See Nawlakhe and Paramasivam (1993) and Bulusu et al. (1979) for additional data. b) Cols 2: Excess Al (mg Al L 1) included in the dose; calculated from Eq. (15.36). c) Cols 3: Excess alkalinity (as mg CaCO3 L 1). For the Al dose; calculated from Eq. (15.36). a)

459

460

Theory and Practice of Water and Wastewater Treatment

carbonate as the alkalinity agent is the following: 3Al2 …SO4 †3
18H2 O ‡ NaF ‡ 8:5Na2 CO3 → 5Al…OH†3 ? Al…OH†2 F ‡ 9Na2 SO4 ‡ 8:5CO2 ‡ 45:5H2 O

(15.36)

The Nalgonda technique is simple, with significant cost savings compared to other techniques (Bulusu et al. 1979; Nawlakhe and Paramasivam 1993). Culp and Stoltenberg (1958) found that lime–alum addition for fluoride removal would be more expensive than activated alumina at one location, but there were several advantages of the former compared to the latter considering the characteristics of the water to be treated. The simplicity of the lime–alum technique was also noted as a primary advantage. The Nalgonda technique can be applied for any size of treatment operation, but it is practical for small villages in developing countries, where treatment to international standards may not be economical. Sludge production from the Nalgonda process is relatively high.

15.10 Membrane Processes Membrane treatment processes are used to separate dissolved and colloidal constituents from water. In membrane treatment, water or components in water are driven through a membrane under the driving force of a pressure, electrical potential, or concentration gradient. Necessarily, particulates are also trapped in the fine openings of membranes. Significant advances have been made in the design of membranes for selectivity and efficiency over the past three decades. Membrane technology in water treatment is used in the desalination of brackish waters; however, membrane treatment is also used for filtration, removal of microorganisms, hardness, volatile organics and other soluble organics, and biological treatment. Membrane technologies continue to improve and applications of it increase. Singh (2011) noted that between 2000 and 2010, the cost of seawater desalination decreased by 50%. In 2010 about 60 million m3 were treated daily by membrane processes and annual growth of 10% was projected (Schrotter and Schrotter 2010). An important application in wastewater treatment is the recovery of precious metals in certain industries; filtration is another application. Membranes can be used to remove suspended and colloidal matter in both water and wastewater treatment. Specific ion probes are applications of membranes in which certain ions or molecules, such as oxygen in the case of DO probes, are selectively allowed to migrate through the membranes. A semipermeable membrane is selective to the species it passes. The size of the openings in the membrane is a major determinant of species that can pass because the openings present a physical barrier to any substances that are larger than the openings. Other chemical character­ istics of the membrane also influence the substances that it can pass. Charge or polarity, shape, and size are characteristics of species that influence their ability to migrate through a membrane. Membranes are usually made from organic polymers, including cellulose acetate, polysulfone, polyamide, polyurea, and polycarbonate. Membranes are classified in four ranges, depending on their pore size: microfiltration (MF), UF, NF, and reverse osmosis (RO). The size ranges of particles removed by various membrane processes are shown in Figure 15.5. Larger opening membranes, MF, remove particles in the order of a micron (10 6 m) in size. MF can remove Giardia, coliforms, Cryptosporidium, and particles that shield pathogens from disinfectants (Yoo et al. 1995). Ultrafiltration and NF membranes are usually rated according to the smallest molecular weight of substances removed by the membrane. The size of the pore openings is in the order of a nanometer (10 9 m).

15 Physical–Chemical Treatment for Dissolved Constituents

Figure 15.5 Particle size ranges removed by membrane processes and other filtration processes. From Pearce (2011), Stumm and Morgan (1996), and Metcalf and Eddy:AECOM (2014).

NF is used for softening of water (removal of calcium and magnesium). RO membranes are able to remove ions such as sodium and chloride in addition to calcium, magnesium, and other ions. The removal capabilities of membranes are sometimes given in terms of MWCO, which is specified in Daltons. Membrane separation is a mass transfer phenomenon. Underlying principles for mass transfer have been presented in Chapter 12, where a concentration gradient is the driving force. A fairly direct application of these principles is separation of metals or measurement of dissolved oxygen. In these instances, metals are removed from solution by plating out; for oxygen, it is consumed on one side of the membrane. A concentration gradient is created across the membrane. A selective membrane prevents or minimizes the passage of interfering substances to the side where consumption is occurring. Mass transfer through a membrane proceeds in both directions. Membrane characteristics govern the direction in which the solutes and solvent are traveling. The ultimate state of any system is an equilibrium condition where the concentrations of all species are equal. With membrane selection technologies energy in one form or another must be applied to counteract and reverse natural tendencies. RO is a good illustration of the principles besides being a common water treatment application. When a semipermeable membrane separates two solutions with different concentrations of solute, the fluid or solvent passes through the membrane to equalize the concentrations of all species on either side of the membrane. The passage of solvent, as opposed to solutes, is osmosis. The chemical imbalance on either side of the membrane causes water to flow against a pressure gradient as illustrated in Figure 15.6a. Ultimately, the forces generated by the chemical imbalance are opposed by the physical pressure force and a static condition occurs. The pressure that exists in this state is the osmotic pressure, Π in Figure 15.6a. If a pressure force is applied to the more highly concentrated solution, the flow of liquid can be reversed as shown in Figure 15.6b, which is RO. As water is removed from the right-hand cell, the

461

462

Theory and Practice of Water and Wastewater Treatment

Figure 15.6 (a) Osmosis. Π is the osmotic pressure. (b) RO.

concentration difference increases as well as the osmotic pressure. In full-scale operations, a supply of brackish water may be fed to the cell to keep osmotic pressures at values near the lowest value. The mass transfer of water through a membrane is proportional to the area of the membrane, the net pressure difference above the osmotic pressure, and the distance over which the pressure difference occurs. Δm A ∝ …Δp Δt Δx

Π†

or

J w ˆ K w …Δp

Π†

(15.37)

where A is the area of the membrane, Kw is a resistance coefficient, Jw is the flux of water (volume/ area-time), m is mass of water passed, Δp is the applied pressure, Π is the osmotic pressure, t is time, and x is the thickness of the membrane. The resistance coefficient is a function of membrane characteristics, solutes in the water, temperature, fouling, and other effects. The concentrations of solutes in the bulk phases of the feed and permeate are Cbf and Cbp, respectively. The osmotic pressure depends on the concentrations and the boundary layers at the membrane surface. In MF and UF membranes, osmotic pressure will not be significant. As noted earlier, some solute will pass through the membrane. The flux of solute depends on the concentration gradient and a resistance parameter. J s ˆ K s …C F

CP†

(15.38)

where CF and CP are the concentrations of solute in the feed and permeate, respectively, Ks is the resistance parameter for solute passage, and Js is the flux of solute. From Eqs. (15.38) and (15.37) it can be determined that CP ˆ

Js Jw

(15.39)

The recovery of water, R, is simply the ratios of product (permeate) to feed water. Rˆ

QP QF

(15.40)

Another parameter describing the performance of a membrane is the solute rejection coefficient, which is a measure of the ability of a membrane to reject the passage of a species, i. Rji ˆ

C iF C iP C iF

(15.41)

where CiF and CiP are the concentrations of species i in the feed and permeate, respectively, Rji is the rejection coefficient for species i.

15 Physical–Chemical Treatment for Dissolved Constituents

Figure 15.7 Confined flow in a pipe.

Another fundamental approach to describe the flux of liquid through membranes uses the Hagen–Poiseuille equation, developed for flow in pipes, to model flow through membrane pores that are considered as cylindrical pores (Stephenson et al. 2000). The force balance in the direction of flow on a cylindrical element in a tube (Figure 15.7) is p2πr dr

p‡

dp dl 2πr dr dl

ρg2πr dl dr sinθ ‡ τ2πr dl

τ2πr dl

d …2πrτ†drdl ˆ 0 dr (15.42)

where g is the acceleration due to gravity, l is the length of the tube, θ is the angle of inclination of the tube, p is the pressure, r is the radial distance from the center of the tube, ρ is the density of water, W is the weight, dW = ρg2πr dl dr sin θ, and τ is the viscous force. The equation can be simplified to dp dl2πr dr dl

ρg2πr dl dr sinθ

2π

d …rτ†dr dl ˆ 0 dr

(15.43)

But recognize that sin θ ˆ

dh dl

where h is energy head. Substituting this into the equation and rearranging it, d…p ‡ ρgh† ˆ dl

1d …rτ† r dr

The equation is easily integrated since l and r are independent. d…p ‡ ρgh† r dr ˆ d…rτ† ∫ ∫ dl r 2 d…p ‡ ρgh† ˆ rτ ‡ C 1 2 dl

(15.44)

Newton’s law of viscosity applies along with the boundary conditions. τˆ μ

dv dr

where μ is viscosity and v is velocity.

(15.45)

463

464

Theory and Practice of Water and Wastewater Treatment

The boundary conditions are r ˆ 0; dv=dr ˆ 0 because of symmetry; and r ˆ d=2 ; v ˆ 0. The resulting equation for velocity is …d=2†2 r 2 d…p ‡ ρgh† 4μ dl

vˆ

(15.46)

where d is the diameter of the capillary tube. Also the average velocity in the tube (vave) is vave ˆ

d 2 d…p ‡ ρgh† 32μ dl

(15.47)

The Hagen–Poiseuille equation describing flow (Qi) through the ith tube is

d2 d…p ‡ ρgh† Qi ˆ vave ˆ 32μ dl Ai

(15.48)

where Ai is the cross-sectional area of a tube. The Hagen–Poiseuille equation applies to the flow through a single pore in a membrane. The flow through a membrane is related to measurable parameters for the membrane that are its void volume fraction, e, and the hydraulic radius of the membrane pores, Rh, in a manner similar to the development for the Carman–Kozeny equation (Section 14.3). Rh ˆ eˆ

area volume …available for flow†  wetted perimeter total surface area of the pores

pore volume membrane volume

1

eˆ

volume of pores

membrane volume

(15.49) (15.50)

Equation (14.12), reproduced below, applies with the above definitions.

enV p

e Vp e ˆ Rh ˆ 1 e ˆ nAp 1 e Ap …1 e†S p

(15.51)

where n is the total number of pores, Vp is the volume of a pore, Ap is the surface area of a pore, and Sp is specific area of a pore. The change in head will be negligible compared to the pressure gradient. The total flow, Q, through the membrane is nQi, which is referred to the total membrane surface area, A. Also Rh = d/4. Using this information and substituting Eq. (15.51) for d into Eq. (15.48): Q ˆ A

nAi

e2 …1 e†2 S2p dp 2μA

(15.52)

dl

The thickness of the membrane is equal to the length of a pore, l, and the transmembrane pressure (TMP) difference is Δp. Recognizing

nAi nAi l ˆ ˆe A Al

(15.53)

The flux of water, Jw, through the membrane is described by Jw ˆ

Q e3 Δp ˆ 2 2 A 2μ…1 e† Sp l

or J ˆ

Δp μRm

(15.54)

15 Physical–Chemical Treatment for Dissolved Constituents

where Rm is the resistance factor, (length) 1; Rm ˆ K m …1 e†2 S2p l=e3 and Km is a constant related to the geometry of the pores. The accumulation of matter at the membrane interface offers additional resistance to flow through the membrane. The additional resistance (cake resistance), Rc, is modeled by the Carman–Kozeny equation [Eq. (14.13)] and is added to the membrane resistance. Jw ˆ

Δp μ…Rm ‡ Rc †

(15.55)

where Rc ˆ

K c …1

ec †2 S c2 lc e3c

with all parameters related to the characteristics of the accumulated matter, (length) 1. Note that Δp is the pressure drop above osmotic pressure when osmotic pressure is significant. There are a number of other formulations that can be developed from various theoretical considerations and operating conditions (Stephenson et al. 2000). All equations must be calibrated to the specific water and operating conditions. Temperature has an effect on viscosity; thus, fluxes must be corrected for temperatures different from a standard temperature used to report a design flux. A temperature-dynamic viscosity relation valid from 0 to 370 °C is (Al-Shemmeri 2012): μT ˆ 2:414 x 10

5

10247:8=…T −140†

(15.56)

where μ has units of N-s m 2 and T is temperature in °K. Note that water flux is inversely proportional to viscosity. The lowest temperature will be the limiting condition. Another operating parameter used for membrane systems is the permeability defined by JP ˆ

J Δp

(15.57)

Two types of elements for membranes are shown in Figure 15.8. The parallel plate arrange­ ment is also known as tangential flow. The elements are bundled together in modules. The permeate is collected in separate channels from the retentate or concentrate. Another configu­ ration is a spiral wound module, which consists of sheets of membranes and spacers wound around a central permeate collection tube.

Figure 15.8 (a) Parallel plate and (b) hollow fiber membrane elements. Top: outside-in; bottom: inside-out.

465

466

Theory and Practice of Water and Wastewater Treatment

Fouling of a membrane increases resistance to flow and reduces the flux of water through a membrane. Prophylactic measures for fouling have been mentioned above. Backwashing or chemical treatment may be applied to remove foulants. Colmatage (French for “clogging”) refers to flux reduction by foulants that may be reversibly removed by hydraulic or chemical treatments. Irreversible fouling of membranes is a more serious problem. Oxidizing agents such as chlorine or ozone may attack membranes and change their structure. Other particles may be irreversibly sorbed to membranes. Raw water characteristics and water quality objectives must be considered along with pre-treatment technologies and membrane type in the system design. Hollow fiber membrane systems can incorporate automatic backwashing every 20–60 minutes in water treatment applications. Normally, over 90% of fouling is removed by this backwashing (Narbaitz, 2017). Spiral wound membranes (Figure 15.9) are most common for RO treatment. They do not permit hydraulic backwashing. Operating pressures decrease as the pore opening size of the membrane increases. Operating pressures for RO systems treating seawater (35 000 mg L 1 TDS) are in the range of 5520– 8270 kPa; for brackish water (5000 mg L 1 TDS), operating pressures range between 860 and 4340 kPa (Bergman et al. 2013). For seawater, RO recoveries range from 30% to 45%; for brackish water recoveries rise to 65–85% (Bergman et al. 2013). NF plants operate at 340–1030 kPa with recoveries in the range of 80–90% for waters with TDS of 500 mg L 1 (Bergman et al. 2013). 15.10.1 Assessment of Water Suitability for Membrane Treatment There are two empirical measures in common use that have been developed for determining the fouling tendency and therefore suitability of a water for membrane treatments, particularly RO and NF. The assessments are the salt density index (SDI) and modified fouling index (MFI). The SDI (ASTM 2001a) is a static test, where water is passed through a 0.45 μm filter having a 47 mm diameter under a pressure of 207 kPag. The first stage begins with a clean filter to collect 500 mL and record the time, ti, to collect this volume. The second stage is to continue filtering under pressure until a time of 5, 10, or 15 minutes has elapsed from the beginning of the first stage filtration; 15 minutes is usually chosen as the interval time unless the filter clogging does not allow the third stage to collect 500 mL. In the third stage, another 500 mL is collected, and the

Figure 15.9 Cutaway view of a spiral membrane element. Source: Adapted from USAID (1980). Courtesy of US Agency for International Development, Washington, DC.

15 Physical–Chemical Treatment for Dissolved Constituents

time for this to occur is recorded as tf. The SDI is calculated as follows: SDI ˆ

t i =t f †

100…1 t

(15.58)

where t is the time from the beginning of the first filtration stage until beginning collection of the second 500 mL volume, and all times are in minutes. The test is not reflective of a continuously running process, and the variability that would be encountered. Also, the test does not use the same material or pore size as the field unit. The test is not conducted in crossflow mode and does not measure chemical fouling. The MFI (Schippers and Verdouw 1980) uses the same apparatus and filtering pressure as the SDI test; however, it is a dynamic test where the volume filtered is recorded every 30 seconds over a 15 minute filtration period. The data are plotted according to the discretized version of Eq. (15.55): J wA ˆ

dV Δp A ˆ dt μ Rf ‡ Rc

(15.59)

where Rf and Rc are the resistances of the filter and cake, respectively, (length) 1. As noted above, Rc depends on the length of the flow path through the cake, lc. This length depends on the volume of the cake, Vc, and the area of the filter, A. lc ˆ k c V c =A

(15.60a)

where kc is a factor related to the tortuous flow path through the cake. The volume of cake depends on the capture of particles in the volume of water filtered, V. Vc ∝V;

V c ˆ k pc V

(15.60b)

where kpc is a factor related to particle capture in the water being filtered. The cake resistance can be formulated from Eqs. (15.60a) and (15.60b) with a factor that lumps the constants together and also accounts for other factors contributing to resistance. Rc ˆ FV =A

(15.60c)

where F is a factor accounting for all resistance factors. Substituting Eq. (15.60c) into Eq. (15.59): dV Δp ˆ dt μ

A Rf ‡

FV A

(15.61)

The variables may be separated and then the equation integrated to μRf V μFV 2 ‡ ΔpA 2ΔpA2

(15.62a)

t 1 μRf μFV ˆ ˆ ‡ V Q ΔpA 2ΔpA2

(15.62b)

tˆ or

Lumping terms together, 1 ˆ a ‡ MFI…V † Q

(15.63)

467

468

Theory and Practice of Water and Wastewater Treatment

where a=

μRf

MFI =

ΔpA , μF

sL

2ΔpA2 ,

1

(L is liters)

sL

2

(L is liters)

Note that Q is the total volume (in liters) filtered over the time, t. A typical plot of data taken for Eq. (15.63) is shown in Figure 15.10. The MFI is the slope of the straight-line portion of the curve. The intercept, a, in Eq. (15.63), is found from the extension of the straight-line portion of the curve. The criticisms of the SDI test apply similarly to the MFI test. For RO applications, manufacturers recommend a maximum value of 4–5 min 1 for the SDI. Duranceau and Taylor (2011) state that MFI values less than 2 and 10 s L 2 are suitable for RO and NF, respectively. 15.10.2 Concentrate Disposal RO and NF produce concentrates that are highly concentrated in dissolved salts, ranging from two to six times their concentrations in the feed water, and other agents used for membrane cleaning (Schrotter and Schrotter 2010). Sulfates and chlorides from acids used for cleaning, and antiscaling agents such as phosphonates, may be present. Where possible, concentrates can be returned to the source water (either marine or freshwater) if toxicity is not a problem. Mixing conditions in the receiving water dictate the allowable size of, and concentrations in, the zone of high-salt concentrations and toxicity. The concentrate is sometimes returned to the sewer system, ultimately going to the wastewater treatment plant, but there will be no treatment of the dissolved salts, and the higher-salt concentrations may be deleterious to biotreatment. Finally, the same receiving water toxicity regulations will govern the discharging of the concentrate in this manner. Deep well disposal may be an option. Digging deep wells is expensive. There must be no opportunity for the discharged concentrate to migrate to and contaminate groundwater supplies. Evaporation ponds are an option in warm climates. Land area requirements are high. Land disposal can be used for salt-tolerant crops. The salts in the concentrate will be subject to surface wash and groundwater infiltration. 15.10.3 Membranes for Water Treatment The applications of membrane technologies in water treatment are many: SS removal, softening, DBP removal, pathogen reduction or removal, and removal of NOM as well as removal of dissolved salts. Membrane systems are available in modules that are conveniently removed for major cleaning, or package systems. Microfiltration and Ultrafiltration Systems

Hollow fiber membranes (Figure 15.11) with internal diameters 0.8–1.2 μm, have a very high surface to volume ratio. They are the most commonly used membranes in MF and UF water

Figure 15.10 Plot to determine MFI.

15 Physical–Chemical Treatment for Dissolved Constituents

Figure 15.11 (a) A hollow fiber membrane module. (b) Lifting a hollow fiber membrane module for cleaning. Source: Courtesy of R. Narbaitz.

treatment applications. Pore sizes of MF systems are one to two orders of magnitude larger than UF membranes. MF membranes may achieve economies and improve overall performance of a system with downstream RO or UF membranes. Pre-treatment with coagulation and sedimen­ tation will produce a higher-quality water. Waters with low turbidity, less than 5 NTU, and having no other contaminants such as iron, manganese, or high NOM, can be fed directly to a MF or UF process (Duranceau and Taylor 2011). Many MF and UF membranes use dead-end filtration where the end of the HF is plugged. Influent water is forced through the membrane, and there is no retentate. The accumulated material is removed by backwashing. These systems have a recovery near 95% after accounting for the product water used in backwashing (Narbaitz, 2017). Besides filtration of particulates, UF membranes can completely disinfect water with the removal of viruses. Ultrafiltration used for disinfection of water has advantages of being a chemical-free process, thus no production of DBPs that consistently produces high-quality water regardless of fluctuations in influent water quality. Nevertheless, chlorine is often used at low doses after UF to protect consumers against membrane breaches. The design procedure for a MF or UF membrane system is to select a membrane type with its flux, Jw [Eqs. (15.37) or (15.55)], and permeability, JP [Eq. (15.57)], characteristics. (In Eq. (15.37), the osmotic pressure will be negligible.) Flux should be corrected to the lowest-operating temperature. These parameters define the membrane area required for a given flow and TMP range. The number of membrane modules is determined based on the membrane area per module. Typical operation and other parameters for MF and UF membranes in water treatment are given in Table 15.12. Nanofiltration and Reverse Osmosis Treatment

NF is used for the removal of divalent ions (hardness and other problematic ions) and NOM (precursors of DBPs), whereas RO is used for the removal of monovalent ions, primarily sodium chloride and other salts in seawater (desalination). The MWCO of RO membranes ranges down to 10; NF separates particles in the range of 0.0005–0.001 μm. There are theoretical and practical considerations beyond the basic design equations for UF or RO systems. Both theory and design are rather complicated; only an overview of the design process will be given here. The approach is taken from Crittenden et al. (2012) based on ASTM (2001b) and AWWA (2007). In NF or RO systems, concentration polarization phenomena, which increase the resistance to mass transfer, are significant. Concentration polarization (Figure 15.12) refers to the buildup of

469

470

Theory and Practice of Water and Wastewater Treatment

Table 15.12 Operational parameters for MF and UF membrane systems. Parameters

Values

Maximum TMP to ensure stable low fouling

0.7 bar

Typical burst and collapse pressure

10–15 bar

Typical operating TMP

0.2–1.0 bar

Inside feed, i.d.

0.8–0.9 μm

Inside feed, o.d.

1.3–1.4 μm

Permeate flux, pressurized

30–170 L (m2 h)

1

Permeate flux, submerged

18–100 L (m2 h)

1

Recovery

>95%

Backwash

a)

1–4 times h

1

Chemical wash (backwash)

Several times per day to once per week

Backwash cycle duration

1–3 min

a) Air may be combined with backwash to enhance cleaning.

Source: From Pearce (2011), Crittenden et al. (2012), and Duranceau and Taylor

(2011).

Figure 15.12 Concentration polarization.

ions on either side of the membrane. As water is removed from the solution in the right-hand cell, the concentration of ions at the membrane surface increases and movement of these ions from the surface occurs by mixing and diffusion. There will be some solute carried with the water through the membrane. These ions must diffuse away or be hydraulically transported from the left-hand membrane surface. Concentration polarization is one cause of membrane fouling. Another is high concentrations of dissolved ions on the feed side of the membrane, which causes solubility products to be exceeded and deposition of the precipitates on the membrane. Sequestering agents can be added to the feed water to retain precipitation species in solution. Other measures such as pre­ treatment or pH control can be used to prevent precipitation. Acidification to a pH of 4–6 prevents the formation of carbonate and precipitation of CaCO3 (AWWA Membrane Com­ mittee 1992). Basic material balances are the beginning equations. The flow balance is QF ˆ QC ‡ QP

(15.64)

15 Physical–Chemical Treatment for Dissolved Constituents

where QC is volumetric feed flowrate of concentrate. The mass balance is

QF C iF ˆ QC C iC ‡ QP C iP

(15.65)

where CiC is concentration of species i in the concentrate. The expressions for rejection [Eq. (15.41)] and recovery [Eq. (15.40)] can be substituted into the flow and mass balances to find C iC ˆ C iF

1

1 Rji R 1 R

(15.66)

Osmotic pressure will be significant in a NF or RO system. In a dilute system, osmotic pressure is approximated by an equation similar to the universal gas equation of state. ЈΦ

Σni RT  ΦC TDS RT V

(15.67)

where Φ is a nonidealty correction factor, ni is the number of moles of solute i, R is the universal gas constant, T is temperature in °K, and CTDS is the TDS concentration. As indicated, all solutes are included in the summation. In the case of an ionizing species, e.g., NaCl, ni is the molar concentration of each ion. In other words, 1 mol of NaCl would have a molar factor of 2 in Eq. (15.67). Osmotic nonidealty factors, Φ, are available for a number of solutes in Robinson and Stokes (1959). Temperature dependence of the water flux has been defined above. The actual pressure drop or the net driving pressure (NDP) consists of differences between feed and permeate pressures, headloss in the feed channel, and changes in osmotic pressure. ΔΠ ˆ pFC;ave −pP − ΠFC;ave −ΠP

NDP ˆ Δp

(15.68)

where pFC,ave is average pressure in the feed-concentrate channel. pFC,ave = ½ (pF + pC), bar. pP is permeate pressure, bar; ΠFC,ave is average osmotic pressure in the feed-concentrate channel, bar; and ΠP is permeate osmotic pressure, bar. ΠFC,ave depends on the average concentration of TDS in the feed-concentrate channel by one of the following equations. C FC;TDS;ave ˆ ½ C F;TDS ‡ C C;TDS C FC;TDS;ave ˆ

C F;TDS 1 ln R 1 R

(15.69a) (15.69b)

where CFC,TDS,ave is average TDS concentration in the feed-concentrate stream, CF,TDS and CC,TDS are TDS concentrations in the feed and concentrate channels, respectively. Equation (15.69a) or (15.69b) is substituted into Eq. (15.67) to find the osmotic pressure in the feed-concentrate channel. A membrane manufacturer would report the performance of a NF or RO system at standard conditions where the temperature, TDS concentration, recovery, rejection, and pressures are specified. Performance parameters must be adjusted to field conditions. Adjustment factors can be based on measured conditions. In the following equations [Eqs. (15.70a)–(15.74)], the subscript (S) refers to the standard condition and the subscript (M) refers to the measured or field condition. For water flux, J w…S† ˆ J w…M†

μT…M† NDP…S† μT…S† NDP…M†

(15.70a)

471

472

Theory and Practice of Water and Wastewater Treatment

or μT…M† NDP…S† μT…S† NDP…M†

(15.70b)

μT…M† C FC;TDS…S† μT…S† C FC;TDS…M†

(15.71)

QP…S† ˆ QP…M† For solute flux, J s…S† ˆ J s…M†

where μT(M) and μT(S) are the dynamic viscosities at the measured field and standard tempera­ tures, respectively. Salt passage (SP) is the ratio of salt concentrations in the permeate and feed: SP ˆ

CP ˆ1 CF

(15.72)

Rj

In Eq. (15.72), CP and CF may be TDS or an individual solute such as sodium chloride. Substituting Eqs. (15.39), (15.70a), and (15.72) into Eq. (15.71) yields SP…S† ˆ SP…M†

NDP…M† NDP…S†

C FC…S† C FC…M†

C F…M† C F… S †

(15.73)

which can be rearranged to R j… S † ˆ 1

1

Rj…M†

NDP…M† NDP…S†

C FC…S† C FC…M†

C F…M† C F… S †

(15.74)

Concentration polarization and precipitation potential of various salts are important factors that also influence the design of NF and RO systems. Refer Crittenden et al. (2012) and Duranceau and Taylor (2011) for more details on these factors and other design considerations. Electrodialysis

In electrodialysis, an electrical potential is used to provide the driving force. A voltage is impressed across an anode and a cathode. Membranes that are selective to cations or anions are inserted in a series of stacks in the water between the electrodes. The ions migrate to the electrode of opposite charge, producing a brine, and a desalinated product water. To avoid fouling and buildup of ions on membrane surfaces, common operational practice is to reverse the charges on the electrodes; this process is called electrodialysis reversal (EDR). Reversing the charges must be accompanied by manipulating the flows between the stacks so that the brine pathway becomes the desalinated water pathway, and vice versa. EDR is particularly useful for the treatment of metal finishing wastewaters (Dalla Costa et al. 2002), but has also been used in other applications for the removal of ionic species, such as F .

15.11 Activated Carbon Adsorption Carbon has been known since the middle ages to be able to remove dissolved substances from liquids. Understanding of the process has advanced considerably, but there is still much to learn. Activated carbons are carbonaceous materials derived from a variety of materials. They are very porous and are able to adsorb dissolved substances onto their porous, fissured surfaces which have large internal surface areas of around 1000 m2 g 1. The activated carbon is the adsorbent and substances being adsorbed are adsorbates. Other materials can be used as adsorbents, but

15 Physical–Chemical Treatment for Dissolved Constituents

carbon is the choice for water treatment because it is able to remove a broad range of adsorbates some of which are targeted and others, such as NOM, are not. Nontargeted adsorbates compete with targeted species for adsorption sites. Activated carbon is prepared in a manner that results in a large surface area by creating fissures within the particles themselves. Eventually, after a certain volume of water has been treated by the activated carbon, it becomes saturated with adsorbates, and the carbon must be regenerated. Regeneration can be performed by chemical or thermal means, which remove or destroy the adsorbed compounds. Synthetic carbonaceous adsorbents are available that have high adsorp­ tion capacities, regenerate more efficiently, and have better mechanical stability than those prepared with conventional techniques (Hand et al. 1994). Activated carbon is most often used to remove taste and odor-causing compounds in drinking water treatment, but it is also effective at removing low concentrations of organic contaminants, particularly those that are recalcitrant to biological treatment or even advanced oxidation. It can remove many inorganic contaminants such as radon-222, mercury, and other toxic metals (Faust and Aly 1987). Significantly, trihalomethanes are not adsorbed to a large extent by activated carbon. Both granular-activated carbon (GAC) and powdered activated carbon (PAC) are efficacious treatments for taste and odor removal in drinking water (Suffet et al. 1996). PAC is particularly useful for seasonal or emergency applications such as treating the toxins from Microcystis outbreaks that have plagued Lake Erie communities (Egan 2017). Treatment with PAC consists of the preparation of a slurry that is continuously added at the end of the coagulation stage to let flocs form first. Then a flocculation stage provides contact time for contaminant removal to occur, and finally the PAC particles are removed by settling with the flocs in the clarifier. Implementation of such a scheme requires only minor modifications to conventional water treatment systems with minimal capital cost. Despite the faster kinetics of PAC compared to GAC, the contact time in the system is not sufficient to saturate the carbon. PAC in the sludge cannot be regenerated. If a high degree of contaminant removal is required, PAC will not likely be the most economical choice due to the large amount of carbon required. Dechlorination is another use of activated carbon. Chlorine and chloramines react with the carbon to form chloride and carbon dioxide products. Activated carbon can serve as one of the media in multimedia rapid (sand) filtration, or it can replace the sand completely, simultaneously functioning as a filter and an adsorber. Activated carbon is not often used in domestic wastewater treatment, but for industrial wastewater treatment, it can remove toxic agents to render the wastewater suitable for down­ stream biological treatment. PAC can be added to biological treatment processes to further treat their effluents. 15.11.1 Activated Carbon – Preparation and Characteristics Activated carbon can be prepared from almost any carbonaceous material (e.g., wood, coal, lignite, and coconut shells) by heating it with or without addition of dehydrating chemicals in the absence of air to liberate carbon from its associated atoms. This step is carbonization. Activation of the carbon occurs by passing mildly oxidative hot gases (carbon dioxide or steam) through the carbon at temperatures between 315 and 925 °C. This causes the formation of tiny fissures or pores. Some noncarbonaceous materials and volatile organics will also be removed in this process. Table 15.13 provides typical characteristics of activated carbons. Adsorption is a surface-area phenomenon. Commercially available GAC has most of its particles ranging in size from 40 mesh (0.425 mm) to 8 mesh (2.36 mm), whereas 80% of PAC particles are smaller than 325 mesh (0.025 mm). Because the majority of the surface area in activated carbon is within internal pores, there is essentially no difference in the surface area per

473

474

Theory and Practice of Water and Wastewater Treatment

Table 15.13 Activated carbon characteristics. GAC

PAC

700–1600

800–1800

Apparent dry density (kg m )

220–650

200–750

Bed density (kg m 3)

∼90% of apparent dry density

Surface area (m2 g 1) 3

Average particle diameter (mm)

0.6–3.0

Effective size (mm)

0.6–0.9

0.01–0.03

Porosity

0.4–0.8

0.4–0.8

Pore volume (mL g 1)

0.85–0.95

2.2–2.5

Ash (%)

10–20

∼10–20

Source: From Chowdhury et al. (2013), Metcalf and Eddy: AECOM (2014), Flick (1991), and Narbaitz (1985).

unit mass, and thus adsorption capacity, for GAC and PAC (Narbaitz 1985). On the other hand, the smaller particle diameters of PAC allow it to reach equilibrium more quickly than GAC. 15.11.2 Adsorption Isotherms Many factors affect the amount of adsorption: chemical properties of the adsorbate, activated carbon properties, and liquid phase characteristics such as pH and temperature. Oxygen concentration was found to affect the adsorption capacity for chlorophenol of activated carbons made from bituminous coal and lignite coal, but it did not affect adsorption for a wood base carbon (Sorial et al. 1993). The molecular size, structure, and polarity are important character­ istics of the adsorbate. The activated carbon adsorption properties are similar for PAC and GAC, as noted earlier. Competitive adsorption is another issue; waters contain many substances that will adsorb and affect the adsorption capacities of each other. This makes it important to maintain laboratory conditions similar to field conditions when gathering adsorption data. An adsorption isotherm describes the relation between the amount or concentration of adsorbate that accumulates on the adsorbent, and the equilibrium concentration of dissolved adsorbate. An adsorption isotherm is an expression of the principle of microscopic reversibility, although adsorption can be irreversible. The most common method for gathering isotherm data is the bottle point technique. Accurately measured concentrations of adsorbate are placed in 10 or more bottles, each containing different amounts of adsorbent. The bottles are mixed using shakers or other means to provide good mixing and contact of the carbon and the solution until a constant liquid phase concentration of adsorbate is achieved. Adsorption is a rather slow process, and it is critical that sufficient time is allowed to reach equilibrium. The necessary time will primarily depend on the adsorbent particle size and the solute being studied. The rate of adsorption is inversely proportional to the particle diameter squared, thus equilibrium will be achieved much faster for PAC. NOM adsorbs more slowly than synthetic organic compounds. The best approach is to first conduct a set of batch kinetic tests to determine the minimum necessary equilibration time. For the isotherms of SOCs, using a one-week contact period should suffice. For isotherms of NOM using PAC, a two-week contact period should suffice. If GAC is used instead, a one- to two-month contact period may be required to achieve equilibrium (Randtke and Snoeyink 1983). The data from the bottles are then analyzed according to the

15 Physical–Chemical Treatment for Dissolved Constituents

following equation: V …C 0

C† ˆ M qe −qi

(15.75)

where C is the equilibrium concentration in solution, C0 is the initial concentration of adsorbate in solution, M is the mass of carbon in the bottle, qe is the final (equilibrium) solid phase concentration of the contaminant on the carbon, mass or moles of contaminant/mass of carbon, qi is the initial solid phase concentration of the contaminant, and V is the volume of the sample. The tests are usually conducted with virgin carbon (qi = 0). With this condition, the above equation can be rearranged to form: …C 0

C†V (15.76) M These equilibrium data are then formulated into an adsorption isotherm model. The first adsorption isotherm was derived by Langmuir (1918), who from theoretical principles developed an equation similar to the Michaelis–Menten equation. Langmuir theorized that only a single adsorption layer could exist. This is a reasonable assumption for gases because the chemical forces bonding the gas to the adsorbent probably do not extend beyond a single layer. At equilibrium, a fraction of the surface of the adsorbent will be covered with adsorbate and the rate of adsorption will equal the rate of desorption. Only molecules striking the bare surface have an opportunity to be retained because the forces of attraction do not extend over multilayers of the adsorbed component. Likewise, the rate of removal of adsorbed molecules from the surface is proportional to the surface area covered by them. Stating these principles in equation form: Define λ as the fraction of surface area covered by adsorbate at equilibrium. Then 1 λ is the fraction of bare surface at equilibrium. Because rate of adsorption is equal to rate of desorption, qe ˆ

kC…1

λ† ˆ k ´ λ

(15.77)

where C is the concentration of the substance being adsorbed (if the adsorbate were a gas the C would be replaced by the partial pressure of the gas) and k and k´ are the rate constants for adsorption and desorption, respectively. The fraction of covered surface area can be determined from Eq. (15.77). λˆ

kC kC ‡ k ´

(15.78)

Define qe as the number of moles or mass of adsorbate per unit weight of adsorbent at concentration C of solute when equilibrium exists. The amount of adsorption is directly proportional to the covered surface area. qe ∝ λ or qe ˆ k ´´ λ

(15.79)

where k´´ is a proportionality constant. Using Eq. (15.79) with Eq. (15.78) and rearranging it, qe ˆ

k ´´ KC k ´´ kC ´ ˆ kC ‡ k KC ‡ 1

(15.80)

where K = k/k´ . The maximum amount of solute that can be adsorbed is defined as Q0. The maximum adsorption occurs when the surface is fully covered. From Eq. (15.79), Q0 ˆ k ´´ …1† ˆ k ´´

475

476

Theory and Practice of Water and Wastewater Treatment

Substituting this into Eq. (15.80), we obtain the Langmuir isotherm: qe ˆ

Q0 KC KC ‡ 1

(15.81)

Another commonly used adsorption isotherm is the Freundlich isotherm (Freundlich 1926), which is an empirical equation. qe ˆ K F C 1=n

(15.82)

where KF and n are constants. A restriction of the Freundlich isotherm is that it should only be used to model equilibrium data in the concentration range for which it was developed (Chowdhury et al. 2013). The Brunauer, Emmett, Teller (BET) isotherm was developed from theory as an extension to the Langmuir isotherm for the case where multilayer adsorption was occurring (Brunauer et al. 1938). It is qe ˆ

…C s

BCQo …B C† 1 ‡

1†C

(15.83)

Cs

where B and Q0 are constants and Cs is the saturation concentration of the solute in water. The three adsorption isotherms result in significantly different functional relations. In fact, there are many other adsorption isotherms (Foo and Hameed 2010). The choice of the best relation will depend on which one best describes the data. The Freundlich isotherm most often proves to be the best relation. The isotherms are equilibrium relationships that do not indicate the rate of reaction. Removal of dissolved NOM which is comprised of humic acids, and many other compounds in surface waters is a common application of activated carbon adsorption. From 5% to 15% of NOM is nonadsorbable (Chowdhury et al. 2013). When nonadsorbable matter is present, adsorption isotherms may appear as shown in Figure 15.13 or they may change to a concave shape because of the different rates of adsorption of different components of NOM. In water treatment operations, the long run times allow for significant biological growth. Biological activity in activated carbon filters may remove 10–15% of organic matter in surface waters; with pre-ozonation, removals may increase to 25% (Chowdhury et al. 2013).

Figure 15.13 Isotherms with and without a nonadsorbable component.

15 Physical–Chemical Treatment for Dissolved Constituents

Figure 15.14 Continuous flow adsorbers. (a) Fixed bed. (b) Counter­ current flow moving bed. (c) Fluidized bed.

15.11.3 Granular Activated Carbon Adsorbers Batch treatment can be accomplished by simply having water remain in contact with carbon for a period of time. In addition to batch units, there are three types of continuous flow adsorbers illustrated in Figure 15.14. In a fixed bed adsorber, the flow is introduced at the top or bottom of the bed. The bed is replaced upon exhaustion. Typical design information for fixed-bed GAC adsorbers is given in Table 15.14. A minimum number of two filters is recommended. In a countercurrent operation, the adsorbent material is continuously input into the unit in the opposite direction of the fluid movement. The third option is a fluidized bed unit, in which the liquid is introduced at the bottom of the unit and its velocity is sufficient to expand the medium particles. An activated carbon system incorporating onsite regeneration is shown in Figure 15.15. From an examination of the isotherm equations, the countercurrent flow configuration will achieve the most efficient use of the carbon because the rate of adsorption depends on the degree of unsaturation of the carbon. The other two methods will result in lower-effluent concentra­ tions of adsorbate.

Figure 15.15 Activated carbon treatment system. Source: After USEPA (1973).

477

478

Theory and Practice of Water and Wastewater Treatment

Table 15.14 Design characteristics of fixed bed GAC adsorbers. Q/A (m h − 1)

Depth (m)

Aspect ratio

Filter adsorber

4.9–14.7

1.2–1.8

2 : 1 to 4 : 1

Post filter adsorber

9.8–19.6

3–3.7

2 : 1 to 4 : 1

Source: Adapted from Chowdhury et al. (2013).

15.12 Design of Fixed-bed Adsorbers As adsorbate-bearing water is introduced into the top of a fixed-bed adsorber containing fresh medium, the top layer of the medium begins to remove adsorbate until it becomes saturated at a level corresponding to the influent concentration. Successive layers of medium will have lower accumulations of adsorbate because of the diminishing concentration of adsorbate in the liquid. At some point, the liquid adsorbate concentration will decrease to essentially zero, and medium below this point will not have any accumulation of adsorbate. The zone where the concentration in the liquid decreases from the influent value to a very small value is the adsorption zone or mass transfer zone, and this zone moves downward through the column as time progresses. The situation is depicted in Figure 15.16 along with a graph of effluent concentration as a function of volume passed through the bed. Concentration of the adsorbate exhibits an S-shaped curve in the adsorption zone with ends asymptotically approaching zero and the influent concentration C0. This curve is known as a breakthrough curve. The breakthrough concentration, Cb, is selected based on the effluent criterion with a safety factor. The volume of water treated at breakthrough is Vb. The volume of

Figure 15.16 Adsorption zone progression in a fixed bed adsorber.

15 Physical–Chemical Treatment for Dissolved Constituents

water passed through the bed at exhaustion is Ve. The exhaustion concentration, Ce, is 0.90–0.99 of C0 depending on the shape of the curve. Also any nonadsorbable constituents are not considered for the breakthrough curve; nonadsorbable constituents will pass through the column, immediately appearing in the effluent. When the breakthrough curve exhibits a gradual rise to the asymptotic concentration (C0), Ce is chosen near 0.90C0; when the curve does not exhibit a significant amount of tailing Ce is chosen near 0.99C0. The classic S-shaped curve applies to a single solute. For NOM with many different components with different rates of adsorption, the curve will have a slightly concave shape. Also nonadsorbable NOM components will immediately appear in the effluent. EBCT is a primary design variable. It is the detention time in a bed without medium. Conventional water treatment plant adsorbers have EBCTs in the range of 7–20 minutes. Depending on the treatment requirement, NOM or taste and odor removal, for instance, concentration of the target agent and other factors, bed life may be from 6 months to 4 or more years. The volumetric throughput rate influences the shape of the curve as shown in Figure 15.17. As volumetric throughput and velocity through the bed decrease, the depth of the adsorption zone decreases because there is more time for adsorption in each layer. Filter adsorbers will be backwashed as rapid sand filters. Postfilter adsorber columns will also accumulate particles over the lengthy run times. To reduce headloss, filters are backwashed from time-to-time. Only water is typically used (no air scour) to not disturb the adsorption zone. 15.12.1 Rate Formulation for Adsorption The rate formulation for adsorption is based on an analogy to Fick’s first law. The driving force for adsorption is due to a liquid adsorbate concentration that is higher than the concentration that would be in equilibrium with the amount of mass adsorbed on the adsorbent. Jˆ

1 dm C C∗ ˆ k´ ˆ k …C ∗ a dt x

C†

(15.84)

where a is the adsorption area, C is the concentration in the liquid, C* is the concentration that exists when the liquid is in equilibrium with the adsorbent, k = k´ /x k´ is the resistance to mass transfer, m is the mass of adsorbate, J is the mass flux, t is the time, and x is a nominal distance. All resistances resulting from the tortuous nature of flow and diffusion are lumped into the coefficients k´ or k. Rearranging Eq. (15.84) and dividing by V, the volume, rˆ

ka ∗ …C V

C† ˆ kα…C ∗

C†

(15.85)

where r is the rate of mass removal, mass per volume-time and α = a/V.

Figure 15.17 Effects of flow-through velocity on the breakthrough curve.

479

480

Theory and Practice of Water and Wastewater Treatment

15.12.2 Theory of Fixed-bed Adsorber Systems The design and theory of fixed-bed adsorption systems center on establishing the shape of the breakthrough curve and its velocity through the bed. Pilot studies are required to establish the shape of the curve which is dependent on the kinetics of mass transfer. The breakthrough curve is depicted in Figure 15.18. Above the breakthrough curve, the capacity of the adsorbent is exhausted. Below the curve, the adsorbent is essentially virgin. The depiction of the curve is a conceptualization of the phenomenon. The amount of mass adsorbed in the adsorption zone varies from the ultimate capacity of the adsorbent at the input concentration to no mass adsorbed at the leading edge of the adsorption zone. For practical purposes, the adsorption zone is defined as the zone where the liquid concentration changes from C0 to Cb. The capacity of the adsorbent varies in the characteristic S manner given in Figure 15.18. The adsorption capacity of any layer in the adsorption zone is distributed throughout the layer. Uniform flow patterns will produce a uniform adsorption capacity at each location across a layer of medium. The medium must be replaced when the effluent concentration begins to rise, i.e., when the leading edge of the adsorption zone has reached the end of the column. The capacity of the bed above the adsorption zone is readily established based on the ultimate capacity of the adsorbent. The capacity of the adsorbent utilized in the adsorption zone must be established to be added to the capacity of the saturated adsorbent above the zone to determine the total capacity of the bed at the influent conditions and bed flow through velocity. Moving with the adsorption zone, taking down as the positive direction and setting up a mass balance for a point within the zone (Figure 15.19), QC

Q C‡

@C @C Δx ‡ rΔV ˆ p ΔV @x @t

(15.86)

where p is the porosity of the bed. The cross-sectional area of the column is A. Substituting AΔx for ΔV and Eq. (15.85) for r and assuming steady state, this equation reduces to Q dC dC ˆv ˆ kα…C ∗ A dx dx

C†

where v is the nominal flow velocity through the bed.

Figure 15.18 Breakthrough curve.

Figure 15.19 Elemental volume.

(15.87)

15 Physical–Chemical Treatment for Dissolved Constituents

The development is most conveniently performed in terms of mass-related parameters. The mass flux, Fm, of liquid through the column is given by Fm ˆ

ρl Q A

(15.88)

where ρl is the density of the liquid. Substituting this into Eq. (15.88) and taking the up direction as positive: Fm

dC Q dC ˆ kρ1 ˆ k d α…C dx A dx

C∗†

(15.89)

where kd = kρl. Defining the mass of adsorbent in the bed as m, m ˆ ρp V ˆ ρp AD and dm ˆ ρp A dD

(15.90)

where D is the depth of the bed and ρp is the packed density of the adsorbent. Substituting Eq. (15.90) into Eq. (15.89) results in the differential equation describing adsorption in the adsorption zone. ρp F m A

dC ˆ k d α…C dm

C∗†

(15.91)

The Capacity Utilized in the Adsorption Zone

As noted above, the adsorption zone is defined as that part of the bed where the concentration in the liquid varies from C0 to Cb. The adsorption zone is assumed to have a constant depth, δ. δˆ

Ve

Vb A

(15.92)

where Ve is the volume of liquid passed through the bed at exhaustion and Vb is the volume of liquid passed at breakthrough. The time for the adsorption zone to travel the full depth of the column is te ˆ

Ve Q

(15.93)

where te is the time to bed exhaustion. The time for the adsorption zone to move a distance, δ, tδ, is tδ ˆ

δ v

(15.94)

There will be a finite time, tf, for the adsorption zone to form. The time to exhaustion will be composed of the time for the zone to form plus the time for the zone to travel the depth of the bed, D. te ˆ tf ‡ tD

(15.95)

where tD is the time for the zone to travel the depth of the bed. Noting that vtD = D and vtδ = δ, the ratio of tδ and tD is determined from Eqs. (15.94) and (15.95): δ tδ tδ ˆ ˆ tD D te tf

(15.96)

481

482

Theory and Practice of Water and Wastewater Treatment

Figure 15.20 Adsorption in an incremental volume.

The total mass adsorbed in the adsorption zone is the sum of the masses adsorbed in each incremental layer in the zone. From Figure 15.20, the concentration adsorbed in an incremental volume, ΔV, is C0 C. If the concentration in the liquid is C0, there is no adsorption in the incremental volume; if it is 0, then all mass in the liquid is adsorbed. The incremental mass adsorbed is Δm ˆ …C 0

C†ΔV

The adsorption zone volume is Ve zone, mT, is mT ˆ C 0 …V e

Vb. The total mass that may be adsorbed in the adsorption

V b†

Defining the fractional capacity of the adsorption zone as f, it is calculated from Ve

…C C†dV ∫V b 0 f ˆ C 0 …V e V b †

(15.97)

(C is the function plotted on the breakthrough curve.) The extreme values for f are 0 and 1. Figure 15.21 shows two adsorption zones with small and large depths. It is seen from Figure 15.21a that in an adsorption zone with a small depth, the capacity of the zone is nearly fully utilized. In the limiting case of Figure 15.21a, the adsorption zone has no depth and the concentration gradient is a vertical line. The average concentration of adsorbate in the liquid within the zone is zero and the fractional capacity of the zone is 1.

Figure 15.21 (a, b) Fractional capacity in the adsorption zone.

15 Physical–Chemical Treatment for Dissolved Constituents

Figure 15.21b approaches the other extreme. As the zone stretches, less capacity in the zone is utilized. In the limiting case, the average concentration of adsorbate in the liquid within the zone is C0 and the fractional capacity utilized is 0. From the definition of f, when f = 0, the concentration of the adsorbate in the adsorption zone is 0, and the adsorption zone has a large available adsorption capacity. Conversely when f is 1, the adsorption zone has a low available capacity. The integral in Eq. (15.97) may be rearranged to more clearly illustrate these principles. V e

…C C† dV Ve

∫V b 0 ˆ 1 f ˆ ∫V b C 0 …V e V b †

C V Vb d C0 Ve Vb

(15.98)

Substituting C = 0 and C = C0 into Eq. (15.98) will result in f = 1 and 0, respectively. When a large amount of adsorption has occurred in the adsorption zone, f = 0, and the liquid concentration in the zone, C  C0. Because the adsorbent in the zone is almost saturated, the time to form the zone (tf) will be approximately the time for the zone to travel its own depth, δ. When the zone is compact, the formation time is nearly zero. Therefore, a reasonable (time of formation) – f relation is tf (15.99a) f ˆ1 tδ or

t f ˆ …1

f †t δ

(15.99b)

Either Eq. (15.99a) or (15.99b) may be substituted into Eq. (15.96) to find δ tδ ˆ D t e ‡ t δ …f

1†

Using t e ˆ V e =Q and t δ ˆ …V e

V b †=Q

δ Ve Vb ˆ D V b ‡ f …V e V b †

(15.100)

Define M1 = the mass of adsorbate that will accumulate at complete saturation mass of adsorbate adsorbed at the influent concentration unit mass of adsorbent ˆ ρp ADq1

q1 ˆ M1

(15.101a)

and M1 ˆ ρp Dq1 A

(15.101b)

At the breakpoint, the primary adsorption zone is just beginning to exit the column. The fraction of the saturation mass contained in the depth, δ, is 1 f. Defining Mb as the amount of solute that has accumulated in the column at the time of breakthrough: Mb ˆ ρp q1 …D A ˆ ρp q1 …D

δ† ‡ ρp q1 δ…1 δ† ‡ δ…1

f†

f†

(15.102)

483

484

Theory and Practice of Water and Wastewater Treatment

The percentage saturation of the whole column at breakthrough, Sb, is Sb ˆ

ρp q1 …D Mb =A  100 ˆ M1 =A

δ† ‡ ρp q1 δ…1 ρp q1 D

f†

 100 ˆ

δf

D D

 100

(15.103)

With this information, the volume of the column required to remove the mass of influent solute at a mass loading rate of QC0 over the time te can be determined. To establish the shape of the breakthrough curve, the isotherm for the adsorbate (Figure 15.22) and the kinetic expression [Eq. (15.91)] must be used. From the isotherm plot in Figure 15.22, the liquid concentration C0 is in equilibrium with q0. Now q0 = q1, the ultimate or saturation capacity for the influent characteristics. Define the superficial mass rate of saturation, ms, as ms ˆ

mass of adsorbent saturated time-area of the column

then F m C 0 ˆ ms q 1 (A check on the word definitions and dimensions of each term in this equation should be performed to verify it.) Likewise, at any point C in Figure 15.22, the corresponding equilibrium adsorbate concentra­ tion on the adsorbent (mass basis) is q, and F m C ˆ ms q

(15.104)

Equation (15.104) is known as the operating line. The slope of the operating line is ms/Fm, and it is constant. If virgin carbon is used and the effluent concentration is initially negligible, the operating line passes through the origin and through the point (q1, C0). As liquid moves through the adsorption zone, some adsorbate is removed in a layer, which results in an increase in q for the layer. The liquid then moves to the next layer, which has a q value that is less than the value that would be in equilibrium with the current liquid adsorbate concentration. Consider an elemental volume with a concentration C ´1 that is passing a layer of adsorbent at concentration q1. This is depicted by the C ´1 –q1 lines in Figure 15.22. The layer would be in equilibrium with a liquid concentration of C1. Mass transfer occurs from the liquid to the adsorbent since C ´1 is greater than C1. The driving force for mass transfer from the liquid to the adsorbent is C ´1 C1. The liquid concentration decreases toward C1, and the adsorbent concentration increases toward q∗1 that would be in equilibrium with C ´1 . The equilibrium liquid and adsorbent concentrations are between C ´1 and C1, and q∗1 and q1, respectively. Equilibrium between the liquid and adsorbent phases is probably not reached because the elemental volume

Figure 15.22 Adsorption isotherm.

15 Physical–Chemical Treatment for Dissolved Constituents

is moved to the next layer by the following elemental volume of liquid. Equation (15.89) describes the rate of mass transfer that is given next, using the definition in Eq. (15.88). ρl

Q dC dC ˆ Fm ˆ k d α…C A dx dx

C∗†

Rearranging this equation and integrating it across the primary adsorption zone, Ce k dα δ dC k d αδ dx ˆ ˆ ∫ Cb C C ∗ Fm F m ∫0

(15.105)

The relation between δ and the volume of liquid passed through the bed is given by Eq. (15.92). Similarly, for a point at a distance x below the beginning of the primary adsorption zone: xˆ

Vb

Vx

(15.106)

A

where Vx is the volume passed when a depth x of the primary adsorption zone has exited. The concentration of solute at the point x is C. Taking the ratio of the above equation to Eq. (15.92), C

x Vx ˆ δ Ve

dC V b ∫ Cb C C ∗ ˆ Ce Vb dC ∫ Cb C C ∗

(15.107)

The breakthrough curve can now be plotted by numerical or graphical integration of Eq. (15.107). A relation between C C* and C is required. The procedure to find this relation is as follows: 1) Determine the best fit isotherm and plot the isotherm relation.

2) Draw the operating line.

3) Referring to Figure 15.23, pivot about the operating line to find the liquid concentration C1

and the corresponding concentration, q1, of the solute on the adsorbent layer. Liquid at concentration C1 is passing the adsorbent layer at q1. Now use the isotherm curve to find the concentration in the liquid C ∗1 that would be in equilibrium with adsorbent at q1. 4) Repeat Step 3 and construct the curve shown in Figure 15.24. 5) For graphical integration, discretize Eq. (15.105): C1

dC ˆ ∫ Cb C C ∗ Ce

dC ˆ ∫ Cb C C ∗

3 iˆ1

ΔC i C C∗

n iˆ1

…C

ˆ A1 ‡ A2 ‡ A3 i

ΔC i ˆA C∗†i

Figure 15.23 Operating line and isotherm.

Figure 15.24 Concentration difference curve.

485

486

Theory and Practice of Water and Wastewater Treatment

Figure 15.25 Integration for fractional capacity.

Now x Vx ˆ δ Ve

V b x1 A1 ‡ A2 ‡ A3 ; ˆ Vb δ A

6) The values of C/C0 are also found from the concentration difference plot; i.e., for the point x1, the concentration ratio is C1/C0. These values are used to construct the breakthrough curve and find f. For the breakthrough curve, plot C=C 0 versus …V x V b †=…V e V b † as shown in Figure 15.25. This curve is numerically or graphically integrated to find f. Example 15.4 Design of a Fixed-bed Adsorber The data in Columns 1–3 in the accompanying table were found for 1,1,2-trichloroethane (TCE) adsorption on activated carbon from equilibrium bottle studies (Narbaitz 1985). Virgin carbon was used in the bottle studies, and the water was obtained from the field site that contained background concentrations of TOC. Design a fixed-bed adsorber to remove 1,1,2-TCE to 0.01 mg L 1 when the influent concentration is 4.0 mg L 1 and the flow is 1250 m3 d 1. The bulk density of the carbon in the bed is 0.6 g cm 3 and the bed height is 3.5 m. Water is to be fed to the bed at a rate of 0.15 m3 (m2 min) 1. Laboratory studies with water from the site have shown that kdα is 13.5 g (cm3 min) 1. The fresh carbon contained no 1,1,2-TCE or TOC and the initial effluent contains negligible TOC; therefore, it can be assumed that the operating line passes through the origin. The breakthrough concentration will be set at 80% of the effluent criterion to provide a safety factor. The lab studies have shown that the breakthrough curve does not exhibit significant tailing, so Ce will be set at 0.98C0 = 3.92 mg L 1. The first step in the solution is to establish the adsorption isotherm. Equation (15.76) was used to find the qe values in Column 4 of the table. To evaluate the fitness of the Langmuir isotherm, a regression analysis of 1/qe against 1/C was performed. The Freundlich equation [Eq. (15.83)] was linearized by taking logarithms of each side of the equation and regressing ln qe against ln C. Cs data were unavailable; therefore, the BET isotherm was not appropriate. The results of the regression analyses were the following: Langmuir

Intercept: 0.2487 = 1/Q0 ) Q0 = 4.021 mg g 1 Slope: 0.001 429 = 1/(Q0K) ) K = 173.9 L mg 1 The correlation coefficient (R2) is R2 = 0.977. Freundlich Intercept: 3.260 = ln KF ) KF = 24.66 Slope: 0.5919 = 1/n ) n = 1.689 R2 = 0.997.

15 Physical–Chemical Treatment for Dissolved Constituents (1) C0

(2) C

(3) M/V

(4) qe

0.430

0.000 23

2.72

0.158

0.430

0.000 39

1.71

0.251

0.430

0.002 45

0.632

0.600

0.430

0.038 7

0.097 2

3.52

0.430

0.109

0.044 8

6.08

0.522

0.0160

0.208

2.44

0.522

0.007 40

0.343

1.50

4.21

0.009 55

2.67

1.57

4.21

0.041 2

1.091

3.82

4.21

0.119

0.602

6.80

4.21

0.206

0.405

9.89

5.78

0.140

0.666

8.46

9.09

0.242

0.766

11.6

9.09

0.374

0.596

14.2

9.09

1.182

0.310

25.5

9.09

1.778

0.233

31.4

The two isotherms were plotted. From observation of the plots and the R2 values, the Freundlich isotherm best described the data. The Langmuir isotherm describes only data near the origin well. The working equation is qe ˆ 24:66C 0:5919

(i)

Now the isotherm relation [Eq. (i)] is plotted and the operating line which passes through the origin and the point (C0, q1) is drawn [Eq. (15.104)]. The plot is shown in Figure 15.26. The curves in Figure 15.26 are now used to construct a graph as shown in Figure 15.24. Data from the two curves in Figure 15.26 are tabulated below for the C C* versus C plot (Figure 15.27). The starting value of q corresponds to C = Cb. The equation describing the operating line is C = 0.0678q and Eq. (i) describes the isotherm (C*). These relations were used to construct the table. These computations and others in the table are readily performed with a spreadsheet. A smaller interval size leads to more accurate results.

487

488

Theory and Practice of Water and Wastewater Treatment

Figure 15.26 Operating line plot.

Graphical integration of Eq. (15.106) using the curve in Figure 15.27 was performed. The data for Ai and C/C0 are tabulated with the C and C* data and used to construct the plot of the breakthrough curve shown in Figure 15.28.

Figure 15.27 Concentration difference curve for Example 15.4.

Figure 15.28 Breakthrough curve for Example 15.4.

15 Physical–Chemical Treatment for Dissolved Constituents

The equation for Ai is Ai ˆ …C i

Ci 1†

C i∗ ‡ C i 2

Ci

C ∗i

1

1

q

1.72

5

10

15

20

25

30

35

40

C

0.117

0.339

0.678

1.017

1.356

1.695

2.034

2.373

2.712

C*

0.0101

0.0615

0.198

0.393

0.639

0.931

1.267

1.644

2.060

0.278

0.480

0.624

0.717

0.764

0.767

0.729

0.652

Ai

0.043

0.128

0.187

0.227

0.251

0.259

0.253

0.234

ΣAi/At

0.021

0.083

0.173

0.282

0.404

0.529

0.651

0.764

C/C0

0.085

0.169

0.254

0.339

0.424

0.508

0.593

0.678

1.16

0.895

0.614

0.506

0.458

0.443

0.453

0.491

At =

2.072

C

0.106

C*

ΔC/(C

C*)

q

45

50

55

59.2

C

3.051

3.390

3.729

4.014

2.514

3.003

3.528

3.995

0.537

0.387

0.201

0.019

Ai

0.202

0.157

0.100

0.031

ΣAi/At

0.861

0.937

0.985

1.000

0.763

0.847

0.932

1.000

0.570

0.734

1.15

0.052a)

C* C

C*

C/C0 ΔC/(C

C*)

ΣΔC/(C

C*) =

7.53

a) 0.052 is based on q = 57.8, corresponding to Ce = 3.92 mg L 1.

At = Ai The breakthrough curve does not exhibit tailing. The fractional capacity in the adsorption zone is calculated from j 1

j n

f ˆ jˆ1

Ai 1

At

Ai 1

Cj

At

Cj 2

1

ˆ 0:49

From Figure 15.26 or Eq. (i), the saturated concentration in the adsorbate (at C = 4.0 mg L 1) is q1 = 59.2 mg g 1. Now Eq. (15.105) must be applied. The table above can be used to find the value of the integral, which is approximated as ΣΔC/(C C*). By starting at Cb and carrying the integration to Ce = 3.92 mg L 1, Ce k d αδ dC ˆ ˆ 7:53 ∫ Cb C C ∗ Fm

The mass flow rate is Fm ˆ ρ

Q ˆ 1:00  103 kg m A

3

0:15 m min

1

ˆ 150 kg m2 min

1

489

490

Theory and Practice of Water and Wastewater Treatment

δ can now be found: δˆ

2 …7:53†F m …7:53† 150 kg m min ˆ kdα 13:5 g …cm3 min† 1

1

1000 g kg

1m 100 cm

1

3

ˆ 0:0837 m ˆ 83:7 mm

In this example, the height of the adsorption zone is very small compared to the height of the bed. The fractional saturation of the carbon at the time of breakthrough is Fractional saturation ˆ

δf

D D

ˆ

3:5 m

…0:0837 m†…0:49† ˆ 0:99 3:5 m

At breakthrough the carbon contains …0:99†…0:0592 g=g† 0:6 g cm

3

F m C 0 ˆ 150 kg m2 min

1

ˆ 6:00  10

4

1 kg 1000 g 4:0 mg L

kg m2 min

100 cm m 1

1 kg 106 mg

1 3

…3:5 m† ˆ 123 kg TCE m

1000 L m

3

2

1 m3 103 kg

1

The run time is tˆ

123 kg TCE m 2 6:00  10 4 kg …m2 min†

1

ˆ 2:05  105 min ˆ 142 d

The volume of water passed through the column at breakthrough is V b ˆ 150 kg m2 min

1

2:05  105 min

1 m3 103 kg

ˆ 3:08  104 m3 =m2

Competitive Adsorption

Competitive adsorption alters the breakthrough as noted above.

 

The degree to which the adsorption capacity of a toxic compound is reduced by competing solute(s) is a function of the adsorptivity of the compounds involved and their initial concentrations. In contaminated water applications, the highest concentration of organic compounds is generally the NOM, so the NOM reduces the adsorption capacity of the target toxic compound(s). In isotherm tests, the reduction is often in the order of 70% (i.e., 30% of that without competition).

15.12.3 Bed-depth Service Time Method The design methodology presented in the previous section assumes that the breakthrough curve does not change its shape as the adsorption wave moves through the column, and it is assumed that equilibrium is established between the feed and exhausted carbon. Significant biological activity and complex adsorption isotherms will negate these conditions. A practical generalized method for design of continuous flow carbon adsorption systems based on laboratory or pilot plant data is the bed-depth-service time (BDST) method (Hutchins 1973). The method is not restricted to favorable isotherms or nonbiodegradable wastewaters (Benedek 1974). Break­ through curves are obtained at several different column lengths that span the lengths to be used in the eventual design.

15 Physical–Chemical Treatment for Dissolved Constituents

The design must meet two constraints: the minimum contact time must be provided, and carbon must be supplied at a rate equal to the exhaustion rate. The actual minimum exhaustion rate and theoretical minimum exhaustion rates may be quite different. Data from the BDST technique will provide the information to determine actual minimum exhaustion rates and give the designer information to optimize contact time and exhaustion rate combinations to give the minimum cost of the operation. Wastewater of a relatively constant concentration is applied to columns of different depth at the same loading rate for each column and the effluent concentrations are monitored over time. The fractional effluent concentration data are plotted as shown in Figure 15.29. These curves are the breakthrough curves for the columns. If the curves do not begin at C/C0 = 0, some of the material is nonadsorbable. The curves in Figure 15.29 indicate that the fraction of nonadsorbable matter is approximately 0.10–0.15. A horizontal line is drawn at the specified breakthrough concentration as indicated on the plot. The intersection of the breakthrough curves and the effluent limiting concentration (chosen with a suitable safety factor) defines the time between regenerations for a bed, tb. The service time is less than the regeneration time because it includes the filling time of the bed, which is given here expressed in terms of the available volume of the bed. ts ˆ tb

eAD Q

(15.108)

where ts is the service time of the bed [similar to tD in Eq. (15.95)]. Normally ts will be equal to tb, but it should be checked with the above equation. The overall rate of carbon exhaustion for a bed depth in Figure 15.29 is calculated from Re ˆ

ρp AD tb

(15.109)

where Re is the rate of carbon exhaustion. Equation (15.109) is applied to each bed. The EBCT is also calculated for each bed. td ˆ

D VB ˆ Q Q=A

(15.110)

where td is EBCT and VB is the volume of the bed (carbon + void volume).

Figure 15.29 Breakthrough curves at different bed depths.

491

492

Theory and Practice of Water and Wastewater Treatment

Figure 15.30 BDST curve.

The BDST curve is obtained by plotting bed depth versus service time (Figure 15.30). A regression is performed to find the curve of best fit through the points. The point where the curve passes the ordinate defines the minimum bed depth, Dmin (or detention time), required to achieve the effluent criterion. The minimum contact time is determined from Figure 15.30. t d;min ˆ

Dmin Q=A

(15.111)

where Dmin is the minimum depth determined by the intersection of the curve with the ordinate on Figure 15.30 and td,min is the minimum EBCT. The carbon exhaustion rate data are plotted against the EBCT data to form an operating line (Figure 15.31) for the process (Erskine and Schuliger 1971). The designer can choose any point on the operating line for a proper design. An economic evaluation of the costs associated with larger bed depths and rates of carbon regeneration at each depth can be performed to find the least-cost design. A more comprehensive analysis can be performed by applying different surface loading rates and performing the economic analysis for each operating line. The theoretical minimum exhaustion rate is based on the equilibrium isotherm for the carbon and the wastewater. The rate at which carbon is being exhausted is equal to the rate at which adsorbate is being removed. Re;min q1 ˆ QC 0 ) Re;min ˆ

QC 0 q1

(15.112a)

If there is a concentration, Cn, of nonadsorbable matter then Re;min ˆ

Q…C 0 C n † q1

(15.112b)

Figure 15.31 Operating line and theoretical minimum EBCT and exhaustion rate when only adsorption is significant.

15 Physical–Chemical Treatment for Dissolved Constituents

The operating line asymptotically approaches the minimum EBCT, and exhaustion rate lines. The theoretical minimum exhaustion rate assumes that only adsorption is significant and that the adsorption zone is a square wave moving through the column. The actual exhaustion rate will be less than the theoretical minimum rate. If biological activity or other phenomena contribute to the removal, the actual rate of exhaustion may be lower than the theoretical minimum rate depending on the extent of phenomena other than adsorption (see the following example). Example 15.5 BDST Carbon Adsorber Design The breakthrough curves for COD shown in Figure 15.29 were obtained for three columns with depths of 1.5, 3.0, and 4.5 m. The loading rate applied to each column was 300 m d 1. The influent flow rate is 960 m3 d 1. The influent contains 150 mg/L COD, of which 15 mg L 1 is nonadsorb­ able. The breakthrough concentration was set at 0.28C0, which is indicated in Figure 15.29. The carbon has a porosity of 0.50 and packed density of 0.50 g cm 3. From isotherms obtained with the wastewater and carbon, q1 is 0.31 kg COD/kg carbon. Find the operating line. Depth (m)

tb (d)

ts (d)

Re (kg d − 1)

EBCT (min)

1.5

6.5

6.5

369

7.2

3.0

15.0

15.0

320

14.4

4.5

24.0

24.0

300

21.6

The required area of the column is Aˆ

Q 960 m3 d 1 ˆ ˆ 3:20 m2 Q=A 300 m d 1

The time to breakthrough in each column was found from Figure 15.29 and entered in the table above. Equation (15.108) was applied to find the service time. For the 1.5 m bed, ts ˆ tb

eAD ˆ 6:5 d Q

…0:50† 3:20 m2 …1:5 m† 960 m3 d

1

ˆ 6:5

0:0025 d ˆ 6:5 d

The other ts values were calculated similarly and are given in the table. These data are plotted in Figure 15.30. The curve in Figure 15.30 was regressed using the model D = Ke mt to find the intercept value, Dmin. The regression equation is D ˆ 1:054e

0:063t s

R2 ˆ 0:973

The regression indicates that Dmin = 1.05 m. The overall carbon exhaustion rates were calculated from Eq. (15.109). For the 1.5 m bed, Re ˆ

ρp AD 0:50 g cm 3 3:20 m2 …1:5 m† 1 kg ˆ 6:5 d 1000 g tb

106 cm3 m

3

ˆ 369 kg d

The other values were calculated similarly and entered into the table. EBCTs for the bed were calculated from Eq. (15.110). For the 1.5-m depth column, td ˆ

D 1:5 m VB ˆ ˆ Q Q=A 300 m d

1

ˆ 0:005 d ˆ …0:005 d† 1440 min d

1

ˆ 7:2 min

1

493

494

Theory and Practice of Water and Wastewater Treatment

The other values were calculated similarly and entered into the table. The Re and EBCT data were plotted as in Figure 15.27 to obtain the operating line. From the limited amount of data it appears that Re,min = 250 kg d 1. A regression using a power law indicates this is reasonable. The equation is Re ˆ 537t d 0:191

R2 ˆ 0:996

The theoretical minimum carbon exhaustion rate [Eq. (15.112b)] is Re;min ˆ

960 m3 d 1 150 15 mg L Q …C 0 C n † ˆ q1 0:31 kg=kg

1

1 kg 106 mg

1000 L m

3

ˆ 418 kg d

1

The minimum contact time is t d;min ˆ

Dmin 1:05 m

ˆ 1440 min d Q=A 300 m d 1

1

ˆ 5:0 min

The minimum contact time, theoretical minimum exhaustion rate and actual exhaustion rates were drawn with dashed lines in Figure 15.32. The operating line asymptotically approaches the minimum contact time and actual exhaustion rate lines. Any point on the operating line is a valid design. Costs of greater depth of the bed must be compared against carbon regeneration rates to establish the optimum point.

Figure 15.32 Operating line.

Comparing the actual and theoretical exhaustion rate lines it is apparent that biological activity is very significant in these columns. A power law does not yield a definite minimum exhaustion rate. More data would allow better definition of the line but the data clearly indicate that the actual minimum carbon exhaustion rate is well below the theoretical minimum.

15.12.4 Rapid Small-Scale Column Tests Pilot studies for designing a fixed-bed carbon adsorber are tedious, time-consuming exercises. Theory presented above for fixed-bed adsorber systems incorporates many phenomena into the first-order expression that was used but it yields very good results. At a fundamental level, the overall process is governed by a number of phenomena including movement of the liquid through the bed, diffusion within the bed, diffusion within fissured particles, and the adsorption relation, among other phenomena. The equations describing the process are complex, beyond the scope of this text, requiring numerical solutions. However, for any set of differential equations, reducing them to nondimensional forms yields nondimensional terms (e.g., the

15 Physical–Chemical Treatment for Dissolved Constituents

Reynolds number for the nondimensional form of the momentum equation in fluid flow) that govern the solution to the equations. If similarity, i.e., the same values exist for the dimensionless parameters, for two systems, they have the same solution. This principle of similarity is used to scale large-scale systems into smaller, less costly, less time-consuming pilot or lab-scale systems that are used to gauge the performance of the larger system. The detailed theory describing adsorber performance is given in Crittenden et al. (1986, 2005, 2012). Crittenden et al. (2005) provide a list of dimensionless groups involved in the development. Only a select group of the dimensionless parameters is required to design and scale-up results of the small-scale system. The protocol is termed the rapid small-scale column test (RSSCT). Operating times for a RSSCT are a few days compared to months in full-scale units; it is indeed, rapid. The important similitude relations for a RSSCT are taken from Crittenden et al. (1991). In the equations, SC refers to the small-scale column and LC refers to the large-scale column. The EBCT relation is EBCTSC ˆ EBCTLC

dp;SC d p;LC

2 x

ˆ

t SC t LC

(15.113)

where dp is the adsorbent particle size, t is the time from the beginning of a run, and x depends on intraparticle diffusivity. Particle size distributions are typically log-normal and the mean log-normal size should be used for dp. If intraparticle (pore) diffusivities do not change with particle size, x = 0 in Eq. (15.113). If intraparticle diffusion causes most of the spreading in the mass transfer zone and intraparticle diffusivity is proportional to particle size, x = 1. The constant diffusivity model, x = 0 applies to SOCs; the variable diffusivity model, x = 1 should be used when NOM removal is the treatment objective or NOM concentrations are high, such as in surface waters. When the Reynolds numbers for the small and large columns are equal, spreading in the mass transfer zones in relation to column lengths are equal. The Reynolds number (Re) relation is Re ˆ

ρL vi d p;SC μL

(15.114)

where ρL is liquid density, g L 1; vi is interstitial velocity in the column, m h 1; and μL is dynamic viscosity of liquid, g (m s) 1. The interstitial velocity is related to the superficial velocity by vi ˆ v=ε ˆ …Q=A†=ε

(15.115)

where v is the superficial velocity (Q/A) of the column, ε is the adsorber bed void fraction, Q is the volumetric flowrate, and A is the surface area of the column. Minimizing the SC column superficial velocity, vSC, shortens the column length, reduces headloss, and the amount of water used in the RSSCT. For situations in which intraparticle diffusivity varies with particle size (x = 1), the practical minimum superficial velocity should be based on a minimum Reynolds number. Crittenden et al. (1991, 1987) recommend a minimum Re value of 1 but also note that lower values can be used if headloss and column length are unacceptable. Chowdhury et al. (2013) and Crittenden et al. (2012) recommend minimum Re values of 0.5 and 0.1, respectively. Larger velocities can be used in either case if the short column length is too short. The equation to be used for scaling superficial velocity in the constant diffusivity case is vSC dp;LC ˆ vLC d p;SC

(15.116a)

495

496

Theory and Practice of Water and Wastewater Treatment

For a nonconstant diffusivity, the equation to be used for superficial velocity is based a minimum Reynold’s number of 0.1 for the SC. vSC ˆ vLC

dp;LC ReSC;min d p;LC 0:1μL ε 0:1 ˆ vLC ˆ d p;SC ReLC d p;SC ρL …vLC =ε†dp;LC ρL dp;SC μL

(15.116b)

If the vSC determined by Eq. (15.116b) results in a column length that is too short, a larger vSC can be chosen. The essential criterion is that ReSC,min is above 0.1. The impact of bulk density and void fraction differences between the pilot and RSSCT will be insignificant as the mass of carbon in the RSSCT is calculated from M SC ˆ EBCTLC

d p;SC dp;LC

2 x

QSC ρb;LC

(15.117)

where MSC is mass of carbon, QSC is volumetric flow rate to SC, and ρb,LC is bulk density of the carbon in the full-scale column. The RSSCT method has been shown to be an appropriate design tool for many situations (Crittenden et al. 1991; Chowdhury et al. 2013). The short time required to complete the test makes it a good choice for bidding and designing a pilot scale adsorber that provides the best design results. There are, however, limitations or disadvantages of the RSSCT method of which the designer must be aware. (i) The RSSCT uses water containing constant concentrations of the target and other possibly involved species which will not be the case for the actual situation; (ii) The appropriate constant or variable diffusivity model is not necessarily known. One should compare RSSCT results to a pilot or full-scale operation to determine definitively the correct model; (iii) There will be no biological growth in a RSSCT. Choosing an average concentration of the target species will not necessarily reflect performance of an adsorber over a long time period with variable influent concentrations. Biological removal of targeted species will be a benefit, but diffusivity may be affected with biological growth. Diameters of columns used in RSSCTs are 1–1.1 cm, and GAC is crushed to nominal diameters of 0.1–0.2 mm (Crittenden et al. 1986, 1987, 1991). Procedure for Applying the RSSCT Method The procedure to determine SC parameters from a projected LC adsorber is as follows: 1) Set the treatment objective. If NOM removal is the objective, x = 1 in Eq. (15.113). If SOC is the target compound, x = 0. Consult the literature to further confirm the choice of x. 2) Select typical values for design features of the large unit, including EBCT, carbon density (ρb,LC), particle diameter (dp,LC), loading rate (vLC), flowrate (QLC), length of adsorber (LLC), and operating time (tLC). Ensure that relations are consistent (e.g., column diameter must provide an area consistent with the loading rate, QLC/ALC). 3) Choose one of EBCTSC, dpSC, or tSC at a convenient value. Using Eq. (15.113), determine the remaining two parameters in the small column. Now find vSC from Eq. (15.116a) or (15.116b). Crush and sieve GAC to the appropriate size. Measure the porosity of the crushed GAC. 4) Choose the column area (ASC). The column length can now be calculated from LSC ˆ vSC EBCTSC If LSC is too short, choose a larger value of vSC.

Then find the flowrate: QSC = vSC ASC

The volume of water required for the RSSCT is determined from

volume of water = QSCtSC.

15 Physical–Chemical Treatment for Dissolved Constituents

5) Use Eq. (15.117) to find the mass of carbon required.

6) Check the Reynolds number [Eq. (15.114)] to determine if it is satisfactory.

There is a fair amount of flexibility in determining the SC setup.

The procedure for scaling RSSCT data to the LS unit is a straightforward application of Eq.

(15.113) that determines the scale-up factor (EBCTLC/EBCTSC) for operating time and volu­ metric throughput of the large column. t LC ˆ t SC

EBCTLC EBCTSC

Example 15.6 RSSCT Design and Scale-up A RSSCT study is to be carried out to gauge the time for breakthrough of a SOC. Literature review has confirmed that a constant diffusivity model is appropriate for the agent. The large column design EBCT is 10 minutes with a loading rate of 12 m h 1. The carbon to be used has a density of 0.65 g cm 3 and particle diameter of 1.3 mm. The GAC porosity is 0.50. The RSSCT column will have a diameter of 1.0 cm. The GAC will be crushed to a nominal diameter of 0.20 mm and its porosity is unchanged from 0.50. The RSSCT will be carried out at 20 °C. Determine EBCTSC, QSC, vSÇ length of the column, and mass of carbon for the column in the RSSCT test. Also determine the run time of the small-scale column that is equivalent to a run time of 100 days for the large column. In Eq. (15.113), x = 0 for a constant diffusivity case EBCTSC ˆ EBCTLC vSC ˆ vLC

2

d p;SC d p;LC

dp;LC ˆ 12 m h dp;SC

ˆ …10 min†

0:20 mm 1:3 mm

1:3 mm 0:20 mm

ˆ 78 m h

1

LSC ˆ vSC EBCTSC ˆ 78 m h

…0:237 min†

1

2

ˆ 0:237 min ˆ 14:2 s 1

1h 60 min

ˆ 0:308 m

The area of the column is ASC = π(dSC/2)2 = π(0.50 cm)2 = 0.785 cm2 QSC ˆ vSC ASC ˆ 78 m h

1

0:785 cm2 100 cm m

ˆ 102 cm3 min M SC ˆ EBCTLC

ˆ 102 mL min

2

dp;SC dp;LC

102 cm3 min

1

QSC ρb;LC ˆ …10 min†

1

0:65 g cm

1

1h 60 min

1

0:20 mm 1:2 mm

2

ˆ 18:4 g

3

Check Re. The interstitial velocity in the column is v;i ˆ vSC =e ˆ 78 m h

1

=0:50 ˆ 156 m h

At 20°C; μL ˆ 1:005 x 10 Re ˆ

3

…N

s† m

2

1

ˆ 1:005 g …s m† 1 ; ρL ˆ 998 kg m

998 g L 1 156 m h 1 …0:20 mm† ρL vi d p;SC 1h ˆ μl 3600 s 1:005 g …s m† 1

1m 1000 mm

The Re number is well above the minimum recommended values.

3

ˆ 998 g L

1000 L m

3

1

ˆ 8:61

497

498

Theory and Practice of Water and Wastewater Treatment

The time-scale relation is t LC ˆ t SC

EBCTLC 10 min ˆ t SC ˆ 42:2 EBCTSC 0:237 min

For tLC = 100 days, the run time of the small column is t SC ˆ t LC =42:2 ˆ …100 d†=42:2 ˆ 2:37 d

15.12.5 Granular Activated Carbon Reactors in Series To fully utilize the capacity of GAC, fixed-bed adsorbers can be designed in a series of two or more as shown in Figure 15.33. The operation occurs in phases as indicated in the figure. The lead column is replaced when the lag column reaches the treatment objective. Exposing the lag column to residual NOM may reduce the capacity and somewhat alter the capacity of the lag column (Jarvie et al. 2005). The design has increased capital costs for piping and controls as well as a minor increase in operational costs but less carbon will be required. Another alternative is to operate a system of multiple columns in an offset fashion, blending their effluents. This can result in significantly less activated carbon usage. 15.12.6 Design of a Suspended Media PAC or GAC Continuous Flow Reactor PAC or GAC is added to a mixed reactor to increase contact between the liquid and the carbon. In a steady-state condition, the carbon will be in equilibrium with the liquid concentration of adsorbate. The mass balance (on a mass basis as opposed to a mass flowrate basis) for the adsorbate in the reactor which is assumed to be completely mixed is QC 0

QC e ‡ rV ˆ 0

(15.118)

where C is the concentration of adsorbate and r is the rate of adsorption. Subscripts 0 and e refer to influent and effluent, respectively The mass flowrate of carbon into and out of the reactor is mass flowrate of carbon ˆ QC C

(15.119a)

where CC is the concentration of carbon. The rate of adsorbate removed by carbon is mass flowrate of adsorbate removed ˆ qe QC C ˆ r ˆ Solving for C C : C C ˆ

C0 Ce qe V

Figure 15.33 Operation of two adsorbers in series.

Q…C 0 C e † V

(15.119b) (15.120)

15 Physical–Chemical Treatment for Dissolved Constituents

The appropriate isotherm can be used to determine qe at the effluent concentration Ce. The retention time in the reactor must be sufficient to provide the contact time required to achieve equilibrium at the desired effluent adsorbate concentration. Pilot studies are needed to ascertain the required retention time in the reactor.

Questions and Problems 15.1 What are the total and calcium hardnesses as CaCO3 of water that contains ions at the concentrations given in the following table? Ion

Ca2+

Mg2+

Fe3+

Na+

K+

CO2

48

20

0.3

1.2

0.7

2

1

Concentration (mg L )

15.2 If a significant amount of alkalinity in a water came from species other than inorganic carbon species or OH ions, would this portion of alkalinity contribute to carbonate hardness and assist softening? 15.3 Calculate the Ksp for CaCO3 based on the practical limits of CaCO3 removal in a treatment process (0.6 meq/L of CaCO3). Using the Ksp based on the practical removal limit, calculate how much CaCO3 would precipitate from initial concentrations of Ca2+ and CO23 of 1.90 and 2.20 meq L 1, respectively. 15.4 The following compositions of major constituents of four natural waters are given in Faust and Aly (1981). Draw meq/L bar graphs for waters 2 and 4. List the concentrations of hypothetical compounds in the waters. Constituenta)

1

Ca Mg

2

3

4

40

94

140

126

22

40

43

43

13

Na

0.4

17

b)

K

1.2

2.2

b)

2.1

HCO3

213

471

241

440

SO24

4.9

49

303

139

Cl

2.0

9.0

38

8.0

a) All concentrations are in mg L 1 of the constituent.

b) Na and K concentrations were not specified.

15.5 What pH is attained by adding 0.7 meq L 1 of lime to pure water? What pH is attained by adding the same amount of lime to water with a pH of 7.00 that has an alkalinity of 100 mg L 1 as CaCO3? 15.6 a For hardness removal in Example 15.1(a), prepare bar graphs describing the state of the water at stages given below. It will be assumed that reactions occur in a definite sequence.

499

500

Theory and Practice of Water and Wastewater Treatment

i) The initial state of the water. ii) After addition of lime.

iii) After reaction of lime with CO2 and bicarbonate.

iv) After precipitation of CaCO3.

b Prepare bar graphs as indicated in part (a) for Example 15.1(b). In addition, prepare the following bar graphs. v) Showing the state of the water after Mg(OH)2 has precipitated. vi) After Na2CO3 has been added. vii) The final state of the water. 15.7 For each water in Problem 4, what are the stoichiometric quantities of lime and soda required to remove hardness to practical limits? Ignore excess lime requirements for Mg removal. 15.8 Calculate the lime dose, carbon dioxide dose after the first stage, and Na2CO3 dose for hardness removal of water no. 3 in Problem 4 using a split two-stage recarbonation process. The effluent Mg concentration is to be 40 mg L 1 as CaCO3 and the total hardness should not exceed 120 mg L 1 as CaCO3. The excess lime for Mg removal is 1.00 meq L 1. The carbon dioxide concentration of the water was 0.45 meq L 1. 15.9 Calculate the lime and soda ash requirements for a split-treatment process. What is the fraction of bypassed flow? The influent water contains Ca2‡

4:80 meq L

1

Mg2‡

2:70 meq L

1

CO2

0:60 meq L

1

Alkalinity

6:30 meq L

1

The product water should have a magnesium concentration of 0.80 meq L 1 and a total hardness concentration of 2 meq L 1 or lower. Perform the analysis when the concentration of magnesium from the first-stage process is0. Assume the excess lime required to precipitate Mg(OH)2 is 1.0 meq L 1 and an excess of 0.6 meq L 1 of CaCO3 is required for Ca2+ removal. What are the concentrations of all softening-related species after each stage? 15.10 Determine the total hardness, alkalinity, Cl , and HCO3 concentrations for the following water analyzed at 25 °C. The pH of the water is 7.9. Assume that other species are present in insignificant amounts. Constituent

Concentration (mg L − 1)

CO2

3.0

Ca

2+

K+ Mg

45.5 37.5

2+

21.0

Cl

?

HCO3

?

SO24

46.5

15.11 Algae in stabilization ponds or reservoirs may cause hardness removal during daylight hours on hot summer days when their metabolic activities are high. Explain how this may happen.

15 Physical–Chemical Treatment for Dissolved Constituents

15.12 Calculate the saturation concentration of Ca2+ in equilibrium with an atmospheric CO2 concentration of 10 3.5 atm at pH values of 5, 6, 7, 8, and 9. Ignore ionic strength effects and use KH = 1470 mg (L atm) 1 [Eq. (3.24)], K0 = 10 2.8 for Eq. (3.25), and K ´1 = 10 3.5 for Eq. (3.26a). Other equilibrium constants are given in Tables 1.5 and 3.2. 15.13 What is the error in estimation of the saturation pH of water that has an alkalinity concentration of 190 mg L 1 and Ca2+ concentration of 120 mg L 1, both measured as CaCO3? Perform the calculations for water samples with pH of 7.00 and 8.50 at a temperature of 25 °C. Assume S = 0.145. (Hint: Equation (15.15) assumes that all alkalinity is HCO3 ions.) 15.14 (a) What is the saturation pH for a water at 10 °C that contains Ca2+ at 100 mg L 1 as CaCO3, an alkalinity of 150 mg L 1 as CaCO3, and TDS of 400 mg L 1? (b) Perform the same exercise for this water at a temperature of 25 °C. What is the SI of these waters if the pH of the water is 7.80? 15.15 What is the error in estimating the saturation pH of the water in Problem 14 if the salinity correction is ignored? 15.16 How much strong acid or Ca(OH)2 must be added to the water in Problem 14 to achieve a saturation index of 0.10? Only consider activity corrections in calculating pHs. 15.17 (a) Although the oxidation of iron with permanganate is favored by a low pH, why is the overall reaction for the removal of iron favored by a high pH? (b) Considering that each oxygen atom can acquire two electrons, what are the overall equations for the removal of iron and manganese using ozone as the oxidant? 15.18 Describe the methods for iron and manganese removal, including chemical equations. 15.19 What is the stoichiometric requirement for HOCl to remove iron and manganese in mg HOCl/mg X where X is either Fe2+ or Mn2+? 15.20 (a) In the case of iron and aluminum, why is significantly more metal coagulant required beyond the stoichiometric amount to reduce soluble orthophosphate concentrations to 0.5–1 mg P L 1? (b) What is the largest factor regulating the dose of calcium to remove phosphorus? 15.21 Is the MFI sensitive to temperature? 15.22 The following data were taken for a MFI test. Determine the MFI. Time (s)

0

30

60

90

120

150

180

210

240

270

300

Volume (L)

0

6.1

7.5

9.0

10.6

11.9

13.0

14.1

15.0

16.0

16.9

Time (s)

330

360

390

420

450

Volume (L)

17.8

18.6

18.9

19.2

19.3

501

502

Theory and Practice of Water and Wastewater Treatment

15.23 A metal plating industry produces effluent containing Cu2+, Fe2+, and Ni2+ at concen­ trations of 25, 25, and 35 mg L 1, respectively. The wastewater flow rate averages 290 m3 d 1. An ion exchanger with a capacity of 3.00 meq g 1 is to be used to remove these metals before discharge of the waste to the sewer. The exchange capacity of the resin on a volumetric basis is 2.0 meq L 1. The regenerant solution will be H2SO4 at a concentration of 50 g L 1. What is the volume of the ion exchanger and mass of resin in it to remove these ions if the resin is to be regenerated once per week? What is the volume of regenerant used on a weekly basis? 15.24 On which side of a solute-removing membrane is precipitation a problem? Why? 15.25 Explain why and how an increase in flow-through velocity affects the breakthrough curve. Draw the breakthrough curve for a nominal flow-through velocity near zero. 15.26 How can steady state be assumed in deriving Eq. (15.87) when the adsorption zone is moving through the bed, causing the accumulation on any layer of carbon within the adsorption zone to change with time? 15.27 Would recycle of the effluent improve the efficiency of a continuous flow fixed-bed carbon adsorption unit? Why or why not? 15.28 Design a fixed-bed carbon adsorber to remove a contaminant down to 0.005 mg L 1 with a breakthrough criterion of 0.004 mg L 1. Plot the breakthrough curve and find the run time. The influent concentration is 0.5 mg L 1 in a flow of 850 m3 d 1. Virgin carbon with a density of 0.52 g cm 3 is to be used in the unit. The column will have a depth of 4.0 m and the surface loading rate is 0.25 m3 (m2 min) 1. kdα = 6.0 g (cm3 min) 1. Use Ce of 0.46 mg L 1. The following table provides the adsorption data for the compound. C (mg L 1) 1

0.000 20

0.000 42

0.000 59

0.001 0

0.002 5

0.004 2

qe (mg g )

0.016 5

0.028 8

0.041 5

0.062 6

0.121

0.170

C (mg L 1)

0.019

0.068

0.11

0.36

0.95

1.33

qe (mg g 1)

0.667

1.61

2.58

5.55

12.4

17.1

0.008 8 0.322

15.29 A painter is cleaning her paintbrush with a solvent. The mass of the brush fibers is 250 g, and the fibers have 50 g of paint on them. The brush will be immersed in the solvent and agitated until equilibrium conditions are established. A Freundlich isotherm applies with KF = 0.90 and n = 3.14 for q in mg g 1 and C in mg L 1. a Find the mass of paint remaining on the brush if it is immersed in a solvent volume of 1.0 L. b Find the total volume of solvent required to achieve the same degree of cleansing as in (a) if two equal volumes of solvent will be used in succession. The brush is immersed in the first volume and agitated until equilibrium is established, removed, and the exercise is repeated with the second volume. Assume that the solvent remaining in the brush after it is removed from the first batch is insignificant. 15.30 Solve Example 15.5 in US units for the operating line with the following changes. Assume the curves in Figure 15.31 apply, but they were obtained for column depths of 4.0, 8.0,

15 Physical–Chemical Treatment for Dissolved Constituents

and 12.0 ft at a nominal velocity (Q/A) of 6000 gal/ft2-d. The influent flow rate is 0.30 Mgal d 1. Influent COD is 150 mg L 1 of which 15 mg L 1 is nonabsorbable. The breakthrough concentration is 0.28C0. The carbon porosity is 0.48, and its packed density is 35 lb/ft3. The value of q1 is 0.30 lb COD/lb carbon. 15.31 Use a nonconstant diffusivity model (intraparticle diffusivity is proportional to particle size and x = 1) to determine small column parameters for the conditions given in Example 15.6.

References Al-Shemmeri, T. (2012). Engineering Fluid Mechanics. Bookboon. ASTM (2001a). Standard Test Method for Silt Density Index (SDI) of Water. Philadelphia, PA: American Society for Testing and Materials. ASTM (2001b). Standard practice for standardizing reverse osmosis performance data. In: Annual Book of Standards, vol. 11.2. Philadelphia, PA: American Society for Testing and Materials. ASTM (2014). G4-01 Standard Guide for Conducting Corrosion Tests in Field Applications. West Conshohocken, PA: ASTM International. doi: 10.1520/G0004-01R14. AWWA (2007). Reverse Osmosis and Nanofiltration, AWWA Manual M46. American Water Works Association: Denver, CO. AWWA Joint Task Group (1990). Suggested methods for calculating and interpreting calcium carbonate saturation indexes. J. Am. Water Resour. Assoc. 82 (7): 71–77. http://www.jstor.org/ stable/41292976. AWWA Membrane Committee (1992). Committee report: membrane processes in potable water treatment. J. Am. Water Resour. Assoc. 84 (1): 59–67. http://www.jstor.org/stable/41294239. Benedek, A. (1974). Carbon evaluation and process design. Proceedings of the Physical–Chemical Treatment Activated Carbon Adsorption in Pollution Control, Environment Canada, Ottawa, ON. Benjamin, M.M. (2015). Water Chemistry, 2e. Long Grove, IL: Waveland Press. Bergman, R.A., Garcia-Aleman, J., and Morgan, R. (2013). Membrane Processes. In: Water Treatment Plant Design, 5e. New York: American Water Works Association and American Society of Civil Engineers, McGraw-Hill. Brinck, W.L., Schliekelman, R.J., Bennet, D.L. et al. (1978). Radium removal efficiencies in water treatment processes. J. Am. Water Resour. Assoc. 70 (1): 31–35. Brunauer, S., Emmett, P.H., and Teller, E. (1938). Adsorption of gases in multimolecular layers. J. Am. Chem. Soc. 60: 309–319. doi: 10.1021/ja01269a023. Bulusu, K.R., Sundaresan, B.B., Pathak, B.N. et al. (1979). Fluorides in water, defluoridation methods and their limitations. J. Inst. Eng. (India) 60 (1): 1–25. Caldwell, D.H. and Lawrence, W.B. (1953). Water softening and conditioning problems. Ind. Eng. Chem. 45 (3): 535–547. doi: 10.1021/ie50519a027. Carlson, K.H., Knocke, W.R., and Gertig, K.R. (1997). Optimizing treatment through fractionation. J. Am. Water Resour. Assoc. 89 (4): 162–171. doi: JAW_0046155. Chan, S.H.S., Wu, T.Y., Juan, J.C., and The, C.Y. (2011). Recent developments of metal oxide semiconductors as photocatalysts in advanced oxidation processes (AOPs) for treatment of dye waste-water. J. Chem. Technol. Biotechnol. 86: 1130–1158. doi: 10.1002/jctb.2636. Chowdhury, Z.K., Summers, R.S., Westerhoff, B.P. et al. (2013). Activated Carbon: Solutions for Improving Water Quality. American Water Works Association: Denver, CO.

503

504

Theory and Practice of Water and Wastewater Treatment

Cleasby, J.L. and Dillingham, J.H. (1966). Rational aspects of split treatment. J. Sanitary Eng. Div. ASCE 92 (2): 1–7. Clifford, D.A. (1990). Ion exchange and inorganic adsorption. In: Water Quality and Treatment (ed. F.W. Pontius), 561–639. Denver, CO: American Water Works Association. Cole, L.D. (1976). Surface-water treatment in cold weather. J. Am. Water Resour. Assoc. 68 (1): 22–25. http://www.jstor.org/stable/41268152. Comstock, G.W. (1979). Water hardness and cardiovascular disease. Am. J. Epidemiol. 110: 375–400. doi: 10.1093/oxfordjournals.aje.a112823. Crittenden, J.C., Berrigan, J.K., and Hand, D.W. (1986). Design of rapid small-scale adsorption tests for a constant diffusivity. J. Water Pollut. Control Fed. 58 (4): 312–319. http://www.jstor.org/ stable/25042907. Crittenden, J.C., Berrigan, J.K., Hand, D.W., and Lykins, B. (1987). Design of rapid fixed-bed adsorption tests for nonconstant diffusivities. J. Environ. Eng. ASCE 113 (2): 243–259. doi: 10.1061/(ASCE)0733-9372(1987)113:2(243). Crittenden, J.C., Reddy, P.S., Arora, H. et al. (1991). Predicting GAC performance with rapid smallscale column tests. J. Am. Water Resour. Assoc. 83 (1): 77–87. jstor.org/stable/41293124. Crittenden, J.C. et al. (2005). Water Treatment: Principles and Design, 2e. Hoboken, NJ: Wiley. Crittenden, J.C., Trussell, R.R., Hand, D.W. et al. (2012). Water Treatment: Principles and Design, 3e. Hoboken, NJ: John Wiley & Sons. Culp, R.L. and Stoltenberg, H.A. (1958). Fluoride reduction at LaCrosse, Kan. J. Am. Water Resour. Assoc. 50 (3): 423–431. http://www.jstor.org/stable/41255125. Dalla Costa, R.F., Klein, C.W., Bernardes, A.M., and Zoppas Ferreira, J. (2002). Evaluation of the electrodialysis process for the treatment of metal finishing wastewater. J. Braz. Chem. Soc. 13 (4): 540–547. doi: 10.1590/S0103-50532002000400021. Deng, Y. and Zhao, R. (2015). Advanced oxidation processes (AOPs) in wastewater treatment. Curr. Pollut. Rep. 1: 167–176. doi: 10.1007/s40726-015-0015-z. Department of National Health and Welfare (1980). Guidelines for Canadian Drinking Water Quality 1978, Supporting Documentation. Ottawa, ON: Supply and Services Canada. Dietrich, M., Andaluri, G., Smith, R.C., and Suri, R. (2017). Combined ozone and ultrasound for the removal of 1,4-dioxane from drinking water. Ozone Sci. Eng. 39 (4): 244–254. doi: 10.1080/ 01919512.2017.1321981. Duranceau, S.J. and Taylor, J.S. (2011). Membranes. In: Water Quality & Treatment, 6e (ed. J.K. Edzwald), 11-1–11-106. New York: American Water Works Association, McGraw-Hill. Eckenfelder, W.W. Jr. and Argaman, Y. (1991). Principles of biological and physical/chemical nitrogen removal. In: Phosphorus and Nitrogen Removal from Municipal Wastewater, 2e (ed. R. Sedlak), 3–42. Chelsea, MI: Lewis Publishers. Edwards, M., Schock, M.R., and Meyer, T.E. (1996). Alkalinity, pH, and copper corrosion byproduct release. J. Am. Water Resour. Assoc. 88 (3): 81–94. http://www.jstor.org/stable/ 41295473. Edwards, M., Hidmi, L., and Gladwell, D. (2002). Phosphate inhibition of soluble copper corrosion by-product release. Corros. Sci. 44: 1057–1071. doi: 10.1016/S0010-938X(01)00112-3. Egan, D. (2017). The Death and Life of the Great Lakes. New York: W. W. Norton & Company, Inc. Erskine, D.B. and Schuliger, W.G. (1971). Graphical method to determine the performance of activated carbon processes for liquids. Am. Inst. Chem. Eng. Symp. Ser. 68 (124): 185–190. Faust, S.D. and Aly, O.M. (1981). Chemistry of Natural Waters. Ann Arbor, MI: Ann Arbor Science. Faust, S.D. and Aly, O.M. (1987). Adsorption Processes for Water Treatment. Toronto: Butterworth Publishers.

15 Physical–Chemical Treatment for Dissolved Constituents

Flick, E.W. (1991). Water Treatment Chemicals: An Industrial Guide. Park Ridge, NJ: Noyes Publications. Foo, K.Y. and Hameed, B.H. (2010). Insights into the modeling of adsorption isotherm systems. Chem. Eng. J. 156 (1): 2–10. doi: 10.1016/j.cej.2009.09.013. Freundlich, H. (1926). Colloid and Capillary Chemistry. London: Methuen & Co. Garcia-Segura, S. and Brillas, E. (2017). Applied photoelectrocatalysis on the degradation of organic pollutants in wastewaters. J. Photochem. Photobiol., C 31: 1–35. doi: 10.1016/j. jphotochemrev.2017.01.005. Hand, D.W., Herlevich, J.A., Jr, D.L.P., and Crittenden, J.C. (1994). Synthetic adsorbent versus GAC for TCE removal. J. Am. Water Resour. Assoc. 86 (8): 64–72. http://www.jstor.org/stable/ 41294767. Hassan, M., Zhao, Y., and Xie, B. (2016). Employing TiO2 photocatalysis to deal with landfill leachate: current status and development. Chem. Eng. J. 285: 264–275. doi: 10.1016/j. cej.2015.09.093. Heinke, K.R. (2009a). Preface. In: Arsenic: Environmental Chemistry, Health Threats and Waste (ed. K.R. Heinke). Hoboken, NJ: Wiley. Heinke, K.R. (2009b). Waste treatment and remediation technologies for arsenic. In: Arsenic: Environmental Chemistry, Health Threats and Waste (ed. K.R. Heinke). Hoboken, NJ: Wiley. Heinke, K.R. and Hutchinson, A. (2009). Arsenic chemistry. In: Arsenic: Environmental Chemistry, Health Threats and Waste (ed. K.R. Heinke). Hoboken, NJ: Wiley. Hutchins, R.A. (1973). Optimum sizing of activated carbon systems. Ind. Water Eng. 10 (3): 40–43. Jarvie, M.E. et al. (2005). Simulating the performance of fixed bed granular activated carbon adsorbers: removal of synthetic organic chemicals in the presence of background organic matter. Water Res. 39 (11): 2407–2421. doi: 10.1016/j.watres.2005.04.023. Jenkins, D. and Hermanowicz, S.W. (1991). Principles of chemical phosphate removal. In: Phosphorus and Nitrogen Removal from Municipal Wastewater, 2e (ed. R.I. Sedlak), 91–110. Chelsea, MI: Lewis Publishers. Kuchenrither, R.D., Elmund, G.K., and Houck, C.P. (1992). Sludge quality benefits realized from drinking water stabilization. Water Environ. Res. 64 (2): 150–153. doi: 10.2175/WER.64.2.8. Langelier, W.F. (1936). The analytical control of anti-corrosion water treatment. J. Am. Water Resour. Assoc. 28: 1500–1521. http://www.jstor.org/stable/41226418. Kusakabe, K., Aso, S., Hayashi, J.-I. et al. (1990). Decomposition of humic acid and reduction of trihalomethane formation potential in water by ozone with u.v. irradiation. Water Res. 24 (6): 781–885. doi: 10.1016/0043-1354(90)90036-6. Langmuir, I. (1918). The adsorption of gases on plane surfaces of glass, mica and platinum. J. Am. Chem. Soc. 40 (9): 1361–1402. doi: 10.1021/ja02242a004. Larson, T.E., Lane, R.W., and Neff, C.H. (1959). Stabilization of magnesium hydroxide in the solids-contact process. J. Am. Water Resour. Assoc. 51 (12): 1551–1558. http://www.jstor.org/ stable/41256160. Liu, X. and Clifford, D.A. (1996). Ion exchange with denitrified brine reuse. J. Am. Water Resour. Assoc. 88 (11): 88–99. http://www.jstor.org/stable/41295914. Lytle, D.A., Williams, D., and White, C. (2012). A simple approach to assessing copper pitting corrosion tendencies and developing control strategies. J. Water Supply Res. Technol. AQUA 61 (3): 164–175. doi: 10.2166/aqua.2012.079. McNeill, L.S. and Edwards, M. (2001). Iron pipe corrosion in distribution systems. J. Am. Water Resour. Assoc. 93 (7): 88–100. http://www.jstor.org/stable/41297605. Merrill, D.T. (1978). Chemical conditioning for water softening and corrosion control. In: Water Treatment Plant Design (ed. R.L. Sanks), 497–565. Ann Arbor, MI: Ann Arbor Science.

505

506

Theory and Practice of Water and Wastewater Treatment

Metcalf and Eddy:AECOM (2014). Wastewater Engineering: Treatment and Resource Recovery (ed. G. Tchobanoglous, H.D. Stensel, R. Tsurchihashi and F.L. Burton). New York: McGraw-Hill. Mouchet, P. (1992). From conventional to biological removal of iron and manganese in France. J. Am. Water Resour. Assoc. 84 (4): 158–167. http://www.jstor.org/stable/41294280. Narbaitz, R.M. (1985). Modelling the competitive adsorption of 1,1,2-Trichloroethane with naturally occurring background organics onto activated carbon. PhD Thesis. McMaster University, Hamilton, ON. Nawlakhe, W.G. and Paramasivam, R. (1993). Defluoridation of potable water by Nalgonda technique. Curr. Sci. 65 (10): 743–748. jstor.org/stable/24095996. Neri, L.C. and Johansen, H.L. (1978). Water hardness and cardiovascular disease. Ann. N.Y. Acad. Sci. 304: 203–219. Pearce, G.K. (2011). UF/MF Membrane Water Treatment: Principles and Design. Bangkok: Water Treatment Academy, TechnoBiz Communications Co., Ltd. Peng, C.G., Qi, J., and Rubin, A.J. (1996). Evaluating fluorspar for fluoridation. J. Am. Water Resour. Assoc. 88 (1): 97–106. http://www.jstor.org/stable/41295416. Pignatello, J.J., Oliveros, E., and MacKay, A. (2006). Advanced oxidation processes for organic contaminant destruction based on the Fenton reaction and related chemistry. Crit. Rev. Environ. Sci. Technol. 36 (1): 1–84. doi: 10.1080/10643380500326564. Plummer, L.N. and Busenberg, E. (1982). The solubilities of calcite, aragonite, and vaterite in CO2– H2O solutions between 0 and 90°C and an evaluation of the aqueous model for the system CaCO3–CO2–H2O. Geochim. Cosmochim. Acta 46 (6): 1011–1040. doi: 10.1016/0016-7037(82) 90056-4. Pratt, C., Parsons, S.A., Soares, A., and Martin, B.D. (2012). Biologically and chemically mediated adsorption and precipitation of phosphorus from wastewater. Curr. Opin. Biotechnol. 23 (6): 890–896. doi: 10.1016/j.copbio.2012.07.003. Randtke, S.J. and Snoeyink, V.L. (1983). Evaluating GAC adsorptive capacity. J. Am. Water Resour. Assoc. 75 (8): 406–413. http://www.jstor.org/stable/41273004. Robinson, R.A. and Stokes, R.H. (1959). Electrolyte Solutions; the Measurement and Interpretation of Conductance, Chemical Potential, and Diffusion in Solutions of Simple Electrolytes. London: Butterworths. Robinson, R.B., Reed, G.D., and Frazier, B. (1992). Iron and manganese sequestration facilities using sodium silicate. J. Am. Water Resour. Assoc. 84 (2): 77–82. http://www.jstor.org/stable/41293636. Roig, M., Manzano, T., and Díaz, M. (1997). Biochemical process for the removal of uranium from acid mine drainages. Water Res. 31 (8): 2073–2083. doi: 10.1016/S0043-1354(97)00036-5. Rossum, J.R. and Merrill, D.T. (1983). An evaluation of the calcium carbonate saturation indexes. J. Am. Water Resour. Assoc. 75 (2): 95–100. http://www.jstor.org/stable/41271574. Rubel, F. Jr. and Woosely, R.D. (1979). The removal of fluoride from drinking water by activated alumina. J. Am. Water Resour. Assoc. 71 (1): 45–48. http://www.jstor.org/stable/41269576. Sanks, R.L. (1978). Ion exchange. In: Water Treatment Plant Design (ed. R.L. Sanks), 597–622. Ann Arbor Science Publishers, Ann Arbor, MI. Sayell, K.M. and Davis, R.R. (1975). Removal of iron and manganese from raw water supplies using manganese greensand zeolite. Ind. Water Eng. 12 (5): 20–23. Schippers, J.C. and Verdouw, J. (1980). The modified fouling index, a method for determining fouling characteristics of water. Desalination 32: 137–148. doi: 10.1016/S0011-9164(00)86014-2. Schock, M.R. (1990). Internal corrosion and deposition control. In: Water Quality and Treatment, 4e (ed. F.W. Pontius). New York: McGraw-Hill. Schrotter, J.-C. and Schrotter, B.B. (2010). Current and emerging membrane processes for water treatment. In: Membrane Technology: Membranes for Water Treatment, vol. 4 (ed. K.V. Peinemann and S.P. Nunes), 53–91. Wiley. doi: 10.1002/9783527631407.

15 Physical–Chemical Treatment for Dissolved Constituents

Sims, R.C. and Hindin, E. (1978). Use of clinoptilolite for removal of trace levels of ammonia in reuse water. In: Chemistry of Wastewater Technology (ed. A.J. Rubin), 305–323. Ann Arbor, MI: Ann Arbor Science Publishers. Singh, R. (2011). Analysis of energy usage at membrane water treatment plants. Desalin. Water Treat. 29 (1–3): 63–72. doi: 10.5004/dwt.2011.181029. Snoeyink, V.L. and Jenkins, D. (1980). Water Chemistry. New York: John Wiley & Sons. Sorial, G.A., Suidan, M.T., Vidic, R.D., and Brenner, R.C. (1993). Effect of GAC characteristics on adsorption of organic pollutants. Water Environ. Res. 65 (1): 53–57. doi: 10.2175/WER.65.1.7. Stephenson, T., Judd, S., Jefferson, B., and Brindle, K. (2000). Membrane Bioreactors for Wastewater Treatment. London: IWA Publishing. Stumm, W. and Morgan, J.J. (1996). Aquatic Chemistry, 3e. New York: Wiley. Suffet, I.H., Corado, A., Chou, D. et al. (1996). AWWA taste and odor survey. J. Am. Water Resour. Assoc. 88 (4): 168–180. http://www.jstor.org/stable/41295508. Suzuki, H., Araki, S., and Yamamoto, H. (2015). Evaluation of advanced oxidation processes (AOP) using O3, UV, and TiO2 for the degradation of phenol in water. J. Water Process Eng. 7: 54–60. doi: 10.1016/j.jwpe.2015.04.011. Szabo, A., Takacs, I., Murthy, S. et al. (2008). Significance of design and operational variables in chemical phosphorus removal. Water Environ. Res. 80 (5): 407–416. doi: 10.2175/ 106143008X268498. Trussell, R.R., Trussell, A., and Kreft, P. (1980). Selenium Removal from Groundwater Using Activated Alumina. USEPA Rep. No. 600/2-80-153, Cincinnati, OH. Tsitonaki, A., Petri, B., Crimi, M. et al. (2010). In situ chemical oxidation of contaminated soil and groundwater using persulfate: a review. Crit. Rev. Environ. Sci. Technol. 40 (1): 55–91. doi: 10.1080/10643380802039303. USAID (1980). The USA.I.D. Desalination Manual. United States International Development Cooperation Agency, Office of Engineering, August, pp. 467. USEPA (1973). Process Design Manual for Carbon Adsorption. Cincinnati, OH: Center for Environmental Research Information. USEPA (1987a). Dewatering Municipal Wastewater Sludges. Design Manual No. EPA/625/1-87/ 014, Center for Environmental Research Information, Cincinnati, OH. USEPA (1987b). Phosphorus Removal. Design Manual No. EPA/625/1–87/001, Center for Environmental Research Information, Cincinnati, OH. USEPA (1987c). Retrofitting POTWS for Phosphorus Removal in the Chesapeake Bay Drainage Basin. Handbook No. EPA/625/6–87/017, Center for Environmental Research Information, Cincinnati, OH. USEPA (1991). Lead and Copper: Final Rule, Federal Registry 5611026460. USEPA (2000). Arsenic Removal from Drinking Water by Ion Exchange and Activated Alumina Plants, No. EPA 600/R-00/088. Cincinnati, OH: US Environmental Protection Agency. Vandenabeele, J., de Beer, D., Germonpré, R. et al. (1995). Influence of nitrate on manganese removing microbial consortia from sand filters. Water Res. 29 (2): 579–587. doi: 10.1016/0043­ 1354(94)00173-5. WHO (2011). Guidelines for Drinking-Water Quality, Health Criteria, 4e. Geneva: World Health Organization. Yoo, R.S., Brown, D.R., Pardini, R.J., and Bentson, G.D. (1995). Microfiltration: a case study. J. Am. Water Resour. Assoc. 87 (3): 38–39. http://www.jstor.org/stable/41294989.

507

509

16 Disinfection Disinfection is the inactivation or removal of pathogenic microorganisms in a water. The water is not necessarily sterilized. The protection of public health from waterborne disease transmission by disinfection of water has been recognized since the early 1900s. The eradication of waterborne pathogens is the most important treatment component of drinking water. Disinfection is also applied to wastewater effluents to reduce the risk of disease for recreational users of surface waters and to minimize the contamination to downstream drawers of water. Pathogen inactivation efficiency is not the only consideration in selecting a disinfectant. The characteristics of an ideal disinfectant are as follows: 1) 2) 3) 4) 5) 6) 7)

Effective inactivation or removal of pathogenic microorganisms Nontoxic to humans or domestic animals Nontoxic to fish and other aquatic species Easy and safe to store, transport, and dispense Low cost Easy and reliable analysis of dose and residuals in water (and wastewater, if necessary) Provides residual protection in drinking water.

Boiling water for 15–20 minutes eradicates pathogenic microorganisms, but it is too energy intensive to be used for water or wastewater disinfection. It was used for a short time in only one municipality, Parthenay, France, in the early 1900s (Baker 1981). However, when municipal water treatment works fail to produce water of suitable microbiological quality, boil-water orders are immediately issued to the public. Disease outbreaks associated with drinking water are with very few exceptions due to viruses, bacteria, or protists. The majority (66%) of outbreaks in 2011–2012 in the United States were due to Legionella spp., which grew in building plumbing systems and untreated groundwater (Beer et al. 2015). From 1970 to 2014, approximately 50% of all waterborne illnesses in Canada and the United States were associated with small noncommunity drinking water systems (Pons et al. 2015). Of the 293 outbreaks, there were 41 862 illnesses but only 3 deaths; almost 30% of the illnesses were due to Giardia intestinalis (also known as Giardia lamblia) and 24.2% to Norovirus. Such outbreaks are rare with large-scale systems, because they generally include multiple barriers and employ professional operators. Nevertheless, failures in these systems can be more catastrophic. An example is the case in Walkerton, Canada, which occurred in May 2000. The technical failure was the chlorination system; however, several other factors, including poor operator training and fraudulent record-keeping, led to the outbreak which resulted in 2300 illnesses, 7 deaths, and many more people with permanent kidney and other organ damage. The etiological agent was Escherichia coli O157:H7 (Salvadori et al. 2009).

Theory and Practice of Water and Wastewater Treatment, Second Edition. Ronald L. Droste and Ronald L. Gehr.

 2019 John Wiley & Sons Inc. Published 2019 by John Wiley & Sons Inc.

Companion website: www.wiley.com/go/droste/water

510

Theory and Practice of Water and Wastewater Treatment

There are many agents that effect disinfection, including chemical oxidants, irradiation, thermal treatment, and electrochemical treatment, but the history of water and wastewater disinfection is the history of chlorination (Black & Veatch Corp 2010). They report that the first continuous application of chlorine for disinfection of a municipal water supply occurred in Belgium in 1902 (Middelkerke) and 1903 (Ostende); Chicago was the first city in the United States (1908). Chlorine satisfies many of the characteristics of an ideal disinfectant. Developments in disinfection technology have produced many alternatives to chlorination, the most widely used alternatives for drinking water being ozonation, ultraviolet (UV) irradiation, and chlorine dioxide as the primary disinfectant, and chloramines as both primary disinfectant and to provide a residual in the distribution system. Ultraviolet irradiation has become the greatest challenger to chlorination for wastewater disinfection; others include peracetic acid (PAA), performic acid (PFA), and ozone. Silver is used as a bactericide for point-of-use (POU) applications, and other halogens, such as bromine and iodine, are available for specific applications, but are not employed at large-scale facilities. The focus of this chapter is on the disinfectants and practices for pathogen removal from drinking water and wastewater. Beyond the disinfectants discussed in this chapter, there are many synthetic biocides used for general control of microorganism growth in pipelines, cooling towers, aquaculture, and other industrial applications. These agents may be present in waste­ waters with potential to affect downstream biological treatment processes and to contribute toxicity in the effluent.

16.1 Kinetics of Disinfection The kinetics of disinfection is described by a first-order law from studies by Chick (1908). r N ˆ kN

(16.1)

where k is the dieoff coefficient, N is the number of microorganisms, and rN is rate of inactivation of microorganisms. The dieoff coefficient is a function of disinfectant dose, type of microorganism, and conditions in the water. Watson (1908) refined the rate coefficient to explicitly include the concentration of disinfectant and another term related to disinfection power of the disinfectant. k ˆ αC n

(16.2)

where C is the concentration of disinfectant, n is termed a constant of dilution, and α is an inactivation constant. In Eq. (16.2), the exponent n is commonly assumed to be 1 although this should be experimentally verified; indeed, the Ct values in Table 16.1 imply that n is less than 1.0. Combining the above two equations and integrating (for a batch reactor), the Chick–Watson disinfection model [Eq. (16.3)] is obtained. ln

N ˆ αC n t N0

(16.3)

where N0 is the initial number of microorganisms and t is time (in conformity with current practice, “t” will be used for time here as opposed to θ used in some other chapters.) The inactivation constant is specific to the microorganism and disinfectant being used and is also sensitive to environmental conditions in the water. Furthermore, disinfectants

16 Disinfection

Table 16.1 Selected Ct values for Giardia and virus inactivation by free chlorine, chlorine dioxide, chloramine, and ozone. Free chlorine, 20 °C, pH 7

Inactivation (log)

Chlorine dioxide, 20 °C, pH 6–9

Chloramine, 20 °C, pH 6–9

Ozone, 20 °C

Chlorine concentration (mg L − 1)

Giardia cysts 1

2

3

1

12

14

15

5.0

370

0.24

2

25

27

31

10.0

735

0.48

3

37

41

46

15.0

1100

0.72

Enteric viruses, pH 6–9a) Free chlorine, 20 °C, pH 6–9 2

1.0

2.1

321

0.25

3

2.0

6.4

534

0.40

4

3.0

12.5

746

0.50

a)

Enteric viruses are of fecal origin, which are infectious to humans by waterborne transmission. Maximum Ct values from hepatitis A and rotavirus were used above. Source: Adapted from USEPA (1999a) and USEPA (1991).

undergo chemical reactions that diminish their disinfection power but do not necessarily destroy it. Therefore, deviations from the Chick–Watson law are common. The general principle embodied in the law is that as concentration or contact time (C or t, respectively) is increased, inactivation of microorganisms is increased. Practically, C × t (or simply Ct) tables are prepared for in situ conditions to arrive at the dose and contact times required to attain the desired removal. Suggested values are also available from various sources, including government regulators, and, in some cases, minimum values take on the status of regula­ tions. A compilation of Ct values taken from the USEPA Guidance Manual (1999a) is given in Table 16.1. The different capabilities of the four disinfectants shown are evident; furthermore, neither α nor n in Eq. (16.3) can be assumed equal to 1.0, because the values do not always increase in the same proportion as the log-inactivation values. In addition, for chlorine, the Ct values increase slightly as chlorine concentrations and pH levels increase (other pH values are not shown in the table). Finally, the table gives Ct values for one temperature only; the Guidance Manual shows that Ct decreases as temperature increases, for all disinfectants. The bactericidal action of a disinfectant is, of course, not limited to pathogens. Fortunately, waterborne-disease-causing microorganisms are removed to levels generally considered safe at lower doses of disinfectant than required for complete sterilization of water. Protista (formerly protozoa) and some viruses require higher doses of disinfectant than do most bacteria. Another common use for disinfecting agents is to control microbial growth in treatment units, pipes, and other appurtenances in the operation. Slime growth, which is primarily due to bacteria and archaea, can result in the addition of taste- and odor-causing compounds to a water. Microbial buildup increases headloss in conveyance systems, where the additional energy costs must be weighed against the costs of disinfectant addition. Many disinfectants increase the corrosiveness of a water, which is another cost consideration.

511

512

Theory and Practice of Water and Wastewater Treatment

Figure 16.1 Distribution of HOCl and OCl .

16.2 Chlorination Chlorine participates in a number of reactions that affect its disinfecting capability. 16.2.1 Chemistry of Chlorine Chlorine is a gas, and dissolved chlorine has a tendency to escape to the atmosphere. Henry’s law describes the equilibrium relation. Cl2 …aq† ⇆ Cl2 …g†

(16.4)

The Henry’s constant as shown in Table 16.4 is 1.88 × 10 3 (mole fraction solubility); however, the loss of chlorine by volatilization is minimal because chlorine rapidly hydrolyzes in water according to the following reactions. The reaction in Eq. (16.5) occurs in a fraction of a second at 20 °C and takes only a few seconds at 0 °C (Black & Veatch Corp 2010). Cl2 ‡ H2 O ⇆ HOCl ‡ H‡ ‡ Cl

(16.5)

HOCl ⇆ H‡ ‡ OCl

(16.6)

Hypochlorous acid (HOCl) is also volatile, but it is on the order of 1.28 × 105 less volatile than Cl2 (i.e., its Henry’s law constant is 1.28 × 105 larger than that for Cl2; Blatchley et al. 1992). The equilibrium expressions for reactions (16.5) and (16.6) are Kˆ

‰H‡ Š‰Cl Š‰HOClŠ ˆ 4  10 ‰Cl2 Š

Ka ˆ

4

‰H‡ Š‰OCl Š ‰HOClŠ

at 25 °C

(16.7) (16.8)

From Eq. (16.7), it can be seen that Cl2 is much less than 1% of the total moles of chlorine species (Cl2, HOCl, and OCl ) in the pH range 6–9 of treated waters. Free available chlorine is chlorine in the form of Cl2, HOCl, and/or OCl (hypochlorite ion). The dissociation of HOCl [Eq. (16.6)] is also temperature and pH dependent, as shown in Figure 16.1. The variation of pKa in Eq. (16.8) as a function of temperature was studied by Morris (1966), who found the following equation: pK a ˆ

3000:00 T

10:0686 ‡ 0:0253T

(16.9)

16 Disinfection

where T is in °K. The following two observations arise from Figure 16.1: At pH 5.0 and below, almost all chlorine is in the form of HOCl. At pH 10.0 and above, almost all chlorine is in the form of OCl . Given that HOCl is a very strong disinfectant, about 80–200 times as strong as OCl , pH exerts a strong influence on the effectiveness of chlorine. HOCl reacts with the enzymes essential to the metabolic processes of living cells. If hypochlorite salts (e.g., Ca(OCl)2) are used, the following reactions occur: Ca…OCl†2 → Ca2‡ ‡ 2OCl

(16.10)

H‡ ‡ OCl ⇆ HOCl

Chlorine reacts chemically with the following substances:

1) Reducing agents such as S2 , Fe2+, Mn2+, and NO2 Using Table 16.3, the overall reaction between chlorine and a reducing agent may be formed. An example reaction is H2 S ‡ 4Cl2 ‡ 4H2 O → H2 SO4 ‡ 8HCl

(16.11)

2) Organic matter Chlorine is able to react in a variety of ways with functional groups and other reaction sites on organic molecules. As an example, the electrophilic addition of Cl2 to a double bond of an alkene is shown in Figure 16.2. 3) Ammonia Hypochlorous acid reacts with ammonia to produce chloramines as shown in the following three reactions: NH3 ‡ HOCl → NH2 Cl ‡ H2 O monochloramine

(16.12)

NH3 ‡ 2HOCl → NHCl2 ‡ 2H2 O dichloramine

(16.13)

NH3 ‡ 3HOCl → NCl3 ‡ 3H2 O trichloramine

(16.14)

Chloramines are called combined chlorine residuals. The distribution of the three types of chloramines is a function of pH. The electron uptake capability of a chloramine is directly dependent on the number of hypochlorous acid molecules used to form it. Ammonia and chloride are released by the reaction of a chloramine with a reducing agent. For instance, the balanced half-reaction for dichloramine is NHCl2 ‡ 2H‡ ‡ 4e ˆ NH3 ‡ 2Cl

(16.15)

There were two HOCl molecules required to form NHCl2 and each HOCl molecule can take up two electrons.

Figure 16.2 Electrophilic addition of chlorine to a double bond.

513

514

Theory and Practice of Water and Wastewater Treatment

Figure 16.3 The classic breakpoint curve. Source: Black & Veatch Corp (2010). Reproduced with permission from John Wiley & Sons.

Dichloramine produces a disagreeable odor and taste, whereas monochloramine does not. Trichloramine is not stable and it breaks down to yield N2 with a loss of oxidation power. Ammonia and chlorine are both biocides; chloramines are disinfectants, but their strength is significantly less than HOCl or OCl . However, chloramines are much longer lasting in water and provide a degree of residual protection. It was found in one study that the combined disinfection effect of free chlorine and monochloramine was greater than the additive effects of each agent; i.e., there was synergism for inactivation of E. coli (Kouame and Haas 1991). As a result of ammonia and other impurities in a water, there is a chlorine demand. The chlorine demand is assessed by adding known amounts of chlorine to a water and measuring the residual chlorine concentrations after a specified contact time. A typical chlorine demand curve is shown in Figure 16.3. Note that this is an equilibrium curve. Inorganic reducing substances that convert chlorine into chloride, which has no residual oxidizing power, may consume the initial small dose of chlorine (not shown in the figure). Chlorine dosed beyond this point is converted into chloramines and chloro-organics, but the former usually outweigh the latter. The dominant chloramine species formed depends on the molar ratio of chlorine added to ammonia nitrogen present. When the molar ratio exceeds 1, reaction (16.13) begins to dominate; when the ratio exceeds 2, reaction (16.14) begins to dominate. There is some formation of NCl3 when the ratio of chlorine to nitrogen is less than 1, even though monochloramine is by far predominant. Unstable nitrogen trichloride is formed when the molar ratio of chlorine to nitrogen exceeds 2 : 1 to 3 : 1 and a large amount of it begins to breakdown to N2. There are also many other inorganic forms of nitrogen with various oxidation states such as nitrate that may be formed from oxidation of ammonia nitrogen as more chlorine is added. Formation of these inorganic species reduces the combined residual. The reactions that occur after the hump in the curve (zone 2) are not well understood. Monochloramines dominate the combined residual in zone 1. After the dip in the curve, in zone 3, the ammonia reactions are completed and further addition of chlorine will

16 Disinfection

result in a free chlorine residual (HOCl and OCl ). The dip in the curve is known as the breakpoint and addition of chlorine beyond this point is called breakpoint chlorination. There will be a relatively small irreducible residual concentration of chloramines (all three species) remaining after the breakpoint that contribute to the total chlorine residual. Reactions of chlorine with organic matter will occur coincidentally with the chloramine and other reactions. Some organic compounds will completely oxidize chlorine but other chloro­ organics formed will have some oxidizing power. Reactions of chlorine with organic matter for the formation of combined chlorine residuals are usually not significant compared to the ammonia reactions. The chlorine demand curve may change as time progresses and the slower reactions consume or convert chlorine into other forms. It is not desirable to be in the region immediately after the hump in the curve (beginning of zone 3); the steep change in the curve is an unstable region. Dichloramines exist in this region, which are malodorous as noted above, and although they are more powerful disinfectants than monochloramines, they are more unstable. Also, NCl3 is being formed, which is the most unstable chloramine. Example 16.1 Chlorine Demand

Chlorine residual (mgL–1)

The chlorine demand curve on the graph was obtained for a drinking water for a 1 hour contact time. Determine the daily amount of NaOCl to be applied to this water to produce (a) a combined residual of 0.4 mg L 1 and (b) a free residual of 0.5 mg L 1 after a contact time of 1 hour in a flow of 24 000 m3 d 1. From the graph, the chlorine dose to achieve a combined residual of 0.4 mg L 1 is 0.60 mg L 1. 1 0.8 0.6 0.4 0.2 0 0

0.2

0.4

0.6

0.8

1

Chlorine dose

1.2

1.4

1.6

1.8

2

(mgL–1)

The combined residual remaining after the breakpoint is approximately 0.08 mg L 1. The chlorine dose to reach the breakpoint is 1.1 mg L 1. After the breakpoint, the curve shows a linear increase in chlorine residual with chlorine dose. The dose of chlorine required to obtain a free residual of 0.5 mg L 1 is 1.1 mg L 1 + 0.50 mg L 1 = 1.60 mg L 1. The total residual is 0.58 mg L 1 at this dose. NaOCl is a salt that dissociates to NaOCl → Na‡ ‡ OCl The MW of NaOCl is 23 + 16 + 35.5 = 74.5 g. One mole of OCl is equivalent to 1 mol of Cl2 from reactions (16.5) and (16.6). a) The amount of NaOCl that must be added each day for a combined residual of 0.4 mg L a 1 hour contact time is

1

after

515

516

Theory and Practice of Water and Wastewater Treatment

QC ˆ 24 000 m3 d

1

0:60 mg Cl2 L

74:5 mg NaOCl 71 mg Cl2

1

1000 L m

3

1 kg 106 mg

ˆ 15:1 kg d

b) Similarly, the daily amount of NaOCl to be added to achieve a free residual of 0.5 mg L QC ˆ

1:60 mg L 0:60 mg L

1 1

15:1 kg d

1

1

1

is

ˆ 40:3 kg d 1 :

The chlorine demand for wastewaters will be much higher than the demand for lake, river, or ground water treated for consumption. The major factor controlling chlorine demand in wastewaters is the concentration of ammonia. Secondary biological treatment processes are often not designed to remove ammonia from the effluent. The amount of chlorine required to achieve a free residual in this case will be excessively high. 16.2.2 Measurement of Free and Residual Chlorine As described previously, chlorine in water and wastewater can exist in two possible forms, free and combined; combined chlorine can be further separated as mono-, di-, or trichloramine, although the last mentioned species is rare. Measurement methods should ideally be able to distinguish between these four forms. The most recent editions of Standard Methods (APHA et al. 2012; unchanged in 2017) give details of four suitable methods. In addition to practicality and sensitivity, the analyst should be aware of possible interferences, such as the presence of other oxidizing agents, when choosing the method to use. Iodometric methods, which use titration, are suitable for determining total chlorine concen­ trations (free plus combined chlorine) greater than 1 mg L 1. For water samples, excess iodine in the form of potassium iodide (KI) is added to a sample buffered at pH ∼4.0. Chlorine liberates the iodine, which is titrated with a standard sodium thiosulfate solution and starch as an indicator. For wastewater samples, excess reducing agent (phenylarsine oxide or thiosulfate) is first added, and the remainder after oxidation by the chlorine is titrated with a standard iodine or iodate solution. Amperometric titration is a more sensitive method, and it is also able to distinguish between the four chlorine forms; however, its application is more tricky and some practice is needed to obtain accurate results. The apparatus consists of a microammeter connected to two electrodes, usually a platinum indicating electrode and a silver–silver chloride electrode as reference. Specialized commercial equipment is readily available. A standardized solution of phenylarsine oxide is the titrant. When buffered at pH 6.5–7.5, free chlorine is determined; adding KI and acetate buffer to pH 3.5–4.5 allows combined chlorine to be measured. In all cases, the endpoint occurs when the microammeter signal stops changing, hence titrating slightly beyond the endpoint is required. Using N,N-diethyl-p-phenylenediamine (DPD) as the indicator and standard ferrous ammo­ nium sulfate (FAS) as the titrant is a popular method because of its relative simplicity and ease of application. There are two approaches to the procedure. One is to titrate with FAS until the red color disappears. Free chlorine is titrated first, then by adding ever larger amounts of KI to the same sample, followed by the same titration procedure as earlier, mono-, di-, and trichloramine can be determined. The second requires a calibration curve prepared with a range of standard­ ized free chlorine solutions, and quantified by spectroscopy at 515 nm. Then, appropriate volumes of sample, phosphate buffer, and DPD indicator are placed in the spectrophotometer cuvette and the free chlorine concentration obtained. Mono- and di-chloramine concentrations are obtained as previously by adding KI crystals. A slightly revised procedure is necessary for trichloramine.

16 Disinfection

The free (available) chlorine test syringaldazine (FACTS) is another spectrophotometric method requiring a calibration curve. The syringaldazine indicator is oxidized by free chlorine at a 1 : 1 molar basis to produce a colored product at an optimum pH of 6.5–6.8. Finally, commercially available combination platinum plus iodide ion-selective electrodes are available, which respond to the iodine released when an acetate-buffered solution containing KI is exposed to chlorine. The electrode pair is used with a typical pH/mV meter, and a calibration curve must be developed. 16.2.3 Chlorine Decay In drinking water treatment plants and distribution systems, the full chlorine demand is not exerted immediately upon addition of chlorine to the water. Excluding volatilization and conversion to chloramines, Brown et al. (2011) noted that there are two types of chemical chlorine decay reactions, as well as two physical environments for the interactions. Chlorine reacts within seconds with reduced inorganic substances, but much slower with organics. Chlorine decay can occur in the bulk solution, as well as in interactions with tank walls, pipes, and fittings. They reviewed six decay models and concluded that a first-order model was the most practical and gave an adequate fit to the data from a large number of installations. Combining the bulk and wall decays into a single constant, kT, they reported a range of values from 0.03 to 1.4 h 1, with most values falling between 0.1 and 0.3 h 1. This translated to half-lives of several hours to several days. Abdel-Gawad and Bewtra (1988) studied the decay of total residual chlorine from chlorination of physical–chemical treatment effluent in natural river water. They found a first-order decay model to be suitable with the rate constant influenced by turbulence, evaporation, photolysis, and temperature. The rate constant was formulated as k T ˆ F TB …k ev ‡ k S ‡ k ox †θ…T

20†

(16.16)

where FTB is a turbulence factor, kT is overall decay coefficient at temperature T, kev is rate constant for evaporation, kS is rate constant for photo-oxidation, kox is rate of free radical oxidation by chlorine, T is temperature (°C), and θ is the Arrhenius constant. The values of constants in the above equation were as follows: FTB = 2.05 for turbulent conditions, FTB = 1.0 for quiescent conditions; kev = 0.013 H 1 d 1 (H is the depth of the flow in meters); kS = 0.03 d 1; kox = 0.065 d 1; and θ = 1.08. Nowell and Hoigné (1992) also found that photodegradation of chlorine followed a first-order reaction. Sunlight in the wavelength range 320–340 nm controls photolysis. The hypochlorite ion was more sensitive to sunlight than hypochlorous acid. Covering basins minimizes the loss of chlorine. Bleach [NaOCl or Ca(OCl)2] is simply a solution of HOCl that can be used as a source of chlorine. Chlorate ion (ClO3 ) is a decomposition product of OCl . Various factors including pH, temperature, and metals affect the rate of decomposition of OCl (Gordon et al. 1995, 1997). The rate and amount of formation of chlorate ion in product waters due to decomposition of hypochlorite ion is insignificant because of the low concentration of OCl . However, when bleach is used as the source of chlorine, decomposition of the OCl in the concentrated bleach feed tank can result in significant ClO3 concentrations in the feed solution (Gordon et al. 1995). Dilution of the feed bleach solution with the product water will reduce chlorate concentrations in the product water. Concentrations of chlorate ion in the feedstock can be minimized by maintaining a high pH between 12 and 13, maintaining a low temperature, shielding bleach storage tanks from sunlight, and minimizing the bleach storage time. For single tank

517

518

Theory and Practice of Water and Wastewater Treatment

installations, the bleach storage tank should be periodically flushed and cleaned to remove accumulated chlorate buildup from hypochlorite decomposition. 16.2.4 Drinking Water Disinfection by Chlorine Current US practice for surface waters is based on two rules. The first, for systems serving fewer than 10 000 persons, is the Long Term 1 Enhanced Surface Water Treatment Rule (LT1ESWTR) (Federal Register 2002). The second, for all public water systems (PWS), is the Long Term 2 Enhanced Surface Water Treatment Rule (LT2ESWTR) (Federal Register 2006a). In both cases, in addition to requiring 99.9% (3-log) inactivation and/or removal of G. lamblia and 99.99% (4­ log) inactivation and/or removal of viruses, there is a requirement regarding Cryptosporidium, for which the maximum contaminant level goal (MCLG) is zero. In addition, consideration is given to filtration and watershed protection to reduce Cryptosporidium concentrations before disinfection. The interesting feature of these rules is that there are no guidelines or requirements for chlorine, as chlorine is essentially ineffective against Giardia or Cryptosporidium. Therefore, these need to be removed by filtration or inactivated by ozone, chlorine dioxide, or UV. Chlorine is nevertheless required as a secondary disinfectant to act as a residual and to prevent microbial growth in the plant as well as in the distribution pipes. Doses are therefore determined by experiments to obtain the complete breakpoint curve in order to provide the required residual concentrations, as free chlorine, chloramines, or both. For example, Canadian drinking water quality guidelines comment that free chlorine concentrations in most Canadian drinking water distribution systems range from 0.04 to 2.0 mg L 1; however, the maximum acceptable concen­ tration for chloramines is 3.0 mg L 1 (Health Canada 2017). Total chlorine dosages at treatment plants vary over a wide range, and one would be hard-pressed to find typical or recommended values in the modern literature. Significant concentrations of chloramines in water are toxic to pet fish, and the chloramines must be removed before using the water in aquariums. Example 16.2 Disinfection Design According to the Ct Concept Calculate the quantity of chlorine consumed on a daily basis and the detention time required for 99.9% reduction of Giardia cysts according to the USEPA Guidance Manual (1999a). The free chlorine residual in the effluent (temperature 20 °C) from the basins is to be 2 mg L 1, the decay rate constant for chlorine is assumed to be 0.2 h 1, and the flow rate is 1.50 × 104 m3 d 1. The pH of the water varies between 6.9 and 7.1. Make the calculations for both a complete mixed (CM) and a plug flow (PF) basin. The appropriate Ct value is taken from Table 16.1. A reduction of 99.9% is equivalent to 3-logs. At a pH of 7.00 and temperature of 20 °C, the Ct value is 41. The effective contact time, t = 41/2 = 20.5 minutes at these conditions. If the basin is PF, the detention time in the basin is 20.5 minutes because td = t10. Using a first-order decay model with a rate constant of 0.2 h 1 [see Eq. (10.4f)], dC ˆ kC ) C ˆ C 0 e dt

ˆ 2:29 mg L 1 :

kt d

or C 0 ˆ C ektd ˆ 2:00 mg L

1

1 e…0:2 h †…41 min†…1 h=60 min†

The quantity of chlorine consumed is QC ˆ 1:50  104 m3 d

1

1000 L m

3

2:29 mg L

1

1 kg 106 mg

ˆ 34:4 kg d 1 :

16 Disinfection

If the basin is CM, the detention time in the basin must be t d ˆ t 10 =0:105 ˆ …20:5 min†=0:105 ˆ 195 min ˆ 3:25 h: Equation (10.21) was derived for first-order decay in a CM basin. Cˆ

C0 ) C 0 ˆ C…1 ‡ kt d † ˆ 2:00 mg L 1 ‡ kt d

1

1 ‡ 0:2 h

1

…3:25 h† ˆ 3:3 mg L 1 :

The quantity of chlorine consumed in this case is QC ˆ 1:50  104 m3 d

1

1000 L m

3

3:3 mg L

1

1 kg 106 mg

ˆ 49:5 kg d 1 :

The benefits of a PF regime are apparent from both the size of the basins and the quantity of chlorine required. Note that in both cases, the Giardia will be reduced by more than 3-logs. For the PF reactor, the chlorine concentration will be higher than 2.0 except in the effluent, and for the CM reactor not all the chlorine will have decayed to 2.0 mg L 1. One needs to keep in mind, though, that each miniscule element of chlorine within the reactor must be at least 2.0 mg L 1 to achieve 99.9% inactivation of the Giardia; it is thus recognized by the Guidance Manual that the design will indeed be conservative and that it will take into account possible short-circuiting in real PF or CM reactors.

16.2.5 Wastewater Disinfection by Chlorine Effluents from wastewater treatment plants (WWTPs) must often be disinfected before discharge to receiving waters, particularly during warm weather when water recreation activity is high. Those plants connected to combined sewers will generally experience high fluctuations in flows and in microbial concentrations, leading to challenging design situations. In some cases, overflows of wastewater from combined sewers – called combined sewer overflows, CSOs – will occur, and these may be totally untreated, or at best, have minimal particle removal followed by disinfection, usually chlorine. Chlorination practice in North America for wastewater treatment has changed rather drastically from 1980 onward. According to surveys reported by Leong et al. (2008), whereas in 1979, 95% of WWTPs used chlorination for disinfection, and of those, 95% used chlorine gas, by 2003, those using chlorine dropped to 72%, of which only 56% were using chlorine gas. The remainder used hypochlorite. Of those using chlorine gas in 2003, more than half were planning to change before 2013 (over 60% to UV, the remainder to hypochlorite). Chlorine dosages vary depending on treatment objectives and the degree of purification of the wastewater. The most common objective is to protect swimmers and recreational users who use the receiving water body, which may be fresh or marine waters. In this case, the demands on the disinfection system are reduced, as water intake by users is much lower. Hence, indicator bacteria, such as E. coli, enterococci, or fecal coliforms are commonly used, and levels need not be reduced to zero. For example, the Canadian Guidelines for Recreational Water Quality (Health Canada 2012) recommend a geometric mean concentration (minimum of five samples) 200 E. coli/100 mL, and single-sample maximum concentration 400 E. coli/100 mL, for fresh recreational waters used for primary contact activities. For marine waters, the same guidelines recommend a geometric mean concentration (minimum of five samples)  35 enterococci/ 100 mL, and single-sample maximum concentration  70 enterococci/100 mL. In all cases, these values would be after dilution; hence, WWTP operators need to back-calculate their target

519

520

Theory and Practice of Water and Wastewater Treatment

Table 16.2 Combined chlorine doses in wastewater treatment. Design chlorine dose (mg L − 1) Metcalf and Eddy:AECOM (2014)b) Effluent Standard (MPN/100 mL)

Ten States Standard (2014)a)

Primary effluent

1000

200

8–12

18–24

23

2.7

Nitrate (mg L 1)

4

12

36

108

>108

Chloride (mg L )

50

150

300

620

>620

Iron (mg L 1)

0.1

0.3

0.9

2.7

>2.7

0.05

0.17

0.5

1.0

>1.0

1

1

1

Manganese (mg L )

Source: Reprinted from Water Res, 5, 9, L Prati, R Pavanello and JT Haney, “Assessment of Surface Water Quality by a Single Index of Pollution,”, 741–751, Copyright 1971, with permission from Elsevier.

Table 22.5 Maximum average weekly temperatures for growth of fish. Fish species

Temperature (°C) Growth

Spawning

Bluegill

32

25

Channel catfish

32

27

Largemouth bass

32

21

Emerald shiner

30

24

Smallmouth bass

29

17

Yellow perch

29

12

White crappie

28

18

Northern pike

28

11

White sucker

28

10

Black crappie

27

—

Sauger

25

10

Atlantic salmon

20

5

Brook trout, Rainbow trout

19

9

Coho salmon, sockeye salmon

18

10

Lake herring (Cisco)

17

3

Source: Compiled from USEPA (1986).

863

864

Theory and Practice of Water and Wastewater Treatment

fish survival. The USEPA (1986) (which was current in 2017) used the formula: Maximum weekly 1 optimum average ˆ ‡ temperature 3 temperature

ultimate upper incipient lethal temperature

optimum temperature

as well as other data to calculate the maximum temperatures in Table 22.5 for spawning and growth of fish. Note that the more desirable fish species generally need lower temperatures for spawning and growth. Nutrient Loadings to Lakes

Nitrogen and phosphorus are the key nutrient elements. The concentration and loading of nutrients in a lake directly influence the trophic state (abundance and activity of organisms in the lake). As the overall nutrient level rises, a lake passes from oligotrophic (paucity of nutrients) to mesotrophic to eutrophic (abundance of nutrients) conditions. The total biomass of virtually all groups of organisms increases along this gradient. However, the types and diversity of species change with the trophic condition. The diversity of species is highest in moderately oligotrophic systems and lower in very low or in higher primary production systems. As a lake becomes nutrient rich, less desirable types of life are present. Therefore, one indicator of the trophic level of a lake is the species types and diversity index. The number of objectionable algae blooms increases as a lake becomes more eutrophic. Total algal biomass increases in proportion to nutrient levels, which affect water transparency as well as DO concentrations. Cyanobacteria (commonly referred to as blue-green algae, although they are not algae) comprise an increasing proportion of plankton biomass at higher nutrient levels. Cyanobacteria form surface blooms, produce undesirable tastes and odors in the water and can produce neurotoxins as noted in Section 6.3.1. There are many approaches to determining acceptable nutrient concentrations in streams and lakes (CCME 2016). Values are ecoregion and waterbody specific in both the US and Canada, and the states or provinces set the regulations. There is considerable variation in natural features of an ecosystem, from flow volumes and climate to soil and land features, among other factors, which influence nutrient levels. In Canada total phosphorus (TP) concentration guidelines range from 0.002 to 0.101 mg L 1 and total nitrogen (TN) guide­ lines range from 0.100 to 1.28 mg L 1; for the US, the ranges are 0.01–0.105 mg L 1 for TP and 0.3–1.4 mg L 1 for TN (CCME 2016). Long-term studies of lakes have shown that reducing only phosphorus input is the solution to eutrophication (Schindler and Hecky 2009; Schindler 2012). Reducing nitrogen concentrations can have adverse consequences such as increasing blooms of Cyanobacteria (Egan 2017) or possibly enhancing phosphorus release from lake sediments, or may have no effect on oligotrophication of lakes (Schindler 2012). Discharge of nutrient-laden waters to the ocean can also result in ecosystem upset. Bell (1992) reported that runoff and sewage effluents are implicated in eutrophication and damage to coral reefs off the coast of Australia. Similar problems exist off the Caribbean islands and in many other regions with coral reefs near human population centers. The concentrations of N and P associated with the onset of eutrophication in coral reef communities are less well defined than for freshwater systems, but inorganic nitrogen and phosphorus levels of approximately 1 μM and 0.1–0.2 μM, respectively, were suggested. These levels are in accord with threshold levels for eutrophication in sensitive freshwater systems.

22 Effluent Disposal in Natural Waters

22.2 Loading Equations for Streams Discharge of pollutants into a stream may originate from a point source or a distributed source. Point sources include sewer outfalls from municipal or industrial sewage treatment plants or other distinct locations where liquid wastes are discharged such as meltoff from a snow dump site. Distributed sources, such as runoff or irrigation, discharge over a wide area. The distinction between point and nonpoint (distributed) sources depends on the time and space scales of the pollutant. Whether the source is point or nonpoint becomes irrelevant for substances with long space–time scales. Definition sketches for the two types of sources are given in Figures 22.3a and b. In Figure 22.3, the subscript “r” refers to the stream condition immediately above the point where the pollution source enters the stream; subscript “w” refers to the waste source. Subscript “0” will be used in the equations below to refer to the location immediately after the waste discharge mixes with the stream. It is assumed that a point source discharge mixes immediately in the cross-section with the receiving stream water. For a distributed source, the discharge occurs over a distance L, and the concentration and stream flow rate vary in this reach primarily as a result of the discharge of the waste. A uniform discharge per unit length of stream, qw, is assumed and the total volumetric flow from the distributed source is Qw. The relations among discharges are Qw L qw x ˆ Qwx qw ˆ

(22.8a) (22.8b)

where Qwx is cumulative flow discharged from the distributed source over the reach from 0 to x and x is distance downstream of the first point of entry of pollution source. 22.2.1 Pollutant Decay in Streams The concentration of a pollutant in the stream is influenced by dilution and various other physical, chemical, and biological processes. The setup of mass balances for this type of problem has been discussed in Chapter 10. The equations developed in the following sections are used to quantify allowable loadings that can be delivered to a stream. Steady-state conditions will be assumed to exist in the developments that follow. The flow regime in the stream is assumed to be plug flow (PF). Actual flow regimes in streams will exhibit some dispersion but over typical space–time scales for most substances, PF will be more realistic than complete mixed conditions. Tracer tests will provide the definitive definition of the flow regime. There can be instances where mixing in the longitudinal direction may be significant. Throughout these developments, it is assumed that the substance is carried along at the average velocity of the stream.

Figure 22.3 (a) Point and (b) distributed pollutant discharges.

865

866

Theory and Practice of Water and Wastewater Treatment

22.2.2 Conservative Substance Conservative substances are not affected by any process other than dilution. Chlorides are generally a good example of a conservative substance. They are soluble and participate in insignificant amounts only in complexation reactions and are taken up in minimal amounts by microorganisms and higher life forms. However, if the concentration of chlorides is high in the waste stream, a density difference exists between the waste stream and the receiving stream that impairs the mixing and dispersion of the waste stream in the receiving body of water. The equations for a conservative substance are simple. Point Source

The flow of water immediately increases at the point of discharge. Q ˆ Qr ‡ Qw

(22.9a)

The concentration in the receiving water at the discharge point is determined from a mixing equation that results from a mass balance at the point of confluence. C0 ˆ C ˆ

C r Qr ‡ C w Qw Q

(22.9b)

Distributed Source

The corresponding equations for flow and concentration in the reach receiving the discharge and downstream of the discharge are as follows: 0→L : Cx ˆ

Qx ˆ Qr ‡ qw x ˆ Qr ‡ Qwx

Qr C r ‡ C w qw x Qr C r ‡ C w Qwx ˆ Qx Qx

(22.10a) (22.10b)

after L :

Q ˆ Qr ‡ qw L ˆ Qr ‡ Qw

(22.10c)

C0 ˆ C ˆ

Qr C r ‡ C w qw L Qr C r ‡ C w Qw ˆ Q Q

(22.10d)

22.2.3 Substances That Are Transformed by One Reaction Many substances are transformed by one or more natural phenomena. In some instances, only one mechanism is dominant. In other cases, many mechanisms are significant, but the overall process is too complex to separate it into its individual components; therefore, all phenomena are lumped into a single reaction model. The reaction may incorporate physical, chemical, and biological phenomena. Indicator bacteria are a good example. Temperature, sedimentation, sunlight, predation, and nutrient availability are some phenomena that influence the survival of bacteria, but generally indicator bacteria disappearance is modeled according to a first-order model, in which the rate constant is a function of temperature (Section 7.6). All rate constants are normally assumed to follow the Arrhenius equation. k T ˆ k 20 θ…T

20†

(22.11)

where k20 is the value at 20 °C, kT is the value at temperature T in °C, and θ is an Arrhenius constant. Point Source

For a substance that decays according to first-order kinetics, the equations have been developed in Chapter 10 for a point source and PF situation. The reaction model is r ˆ kC where k is the rate constant and r is the rate of reaction.

(22.12a)

22 Effluent Disposal in Natural Waters

As developed in Section 10.1.2, the mass balance results in dC ˆ dx

k C v

or

dC ˆ kC dt

(22.12b)

where v is the average flow velocity in the stream. Time and distance are synonymous, although irregular stream geometries cause variable stream velocities. In reaches where the channel geometry is irregular, the distance can be divided into reaches where the velocity is approximately constant and the first form of Eq. (22.12b) can be applied over each reach. Otherwise, a distance–velocity relation must be determined before this equation can be integrated (see Section 10.1.2). Regardless of velocity variation, if the travel time between two points is known, the second form of Eq. (22.12b) can be used. The solution of the second form for Eq. (22.12b) is C ˆ C0e

kt

ˆ

C r Qr ‡ C w Qw e Q

kt

(22.13)

The starting concentration is given by the mixing equation, Eq. (22.9b). Distributed Source

In the reach receiving waste, concentration varies from dilution as well as according to the reaction model which will again be assumed to be a first-order decay. Setting up a mass balance for an elemental volume (Figure 22.4): Qx C x ‡ qw C w Δx

…Qx ‡ ΔQx †…C x ‡ ΔC x †

kC x ΔV ˆ

@C x ΔV ˆ 0 @t

(22.14)

where ΔV is volume (AΔx) of the element. It is further assumed that the velocity of flow, v (Qx/A), and that area, A, are constant in the reach (i.e., the flow contribution of the distributed source, Qw or qwL, is relatively small compared to the stream flow, Qr). Expanding and simplifying the mass balance and dividing by A and Δx results in qw Cw A

Qx dC x A dx

C x dQx ˆ kC x A dx

In addition to v = Qx/A, it is noted that dQx/dx = qw.

Using these relations, the equation may be further simplified to

qw Cw A

v

dC x dx

Cx q ˆ kC x A w

Finally, the equation may be divided by v and rearranged.

dC x ˆ dx

k Cx v

qw q Cx ‡ w Cw Av Av

(22.15)

Figure 22.4 Elemental volume in the reach receiving distributed flow.

867

868

Theory and Practice of Water and Wastewater Treatment

The differential equation, Eq. (22.15), is readily integrated. Defining two new lumped constants, K1 and K2, as follows: K1 ˆ

k v

qw 1 ˆ Av v

qw A

k

K2 ˆ

and

qw Cw Av

The differential equation is now dC x ˆ K 1Cx ‡ K 2 dx Cx

∫ Cr

x dC x ˆ dx K 1Cx ‡ K 2 ∫ 0

The solution of the integral is Cx ˆ

Cr ‡

K 2 K 1x e K1

qw x K2 ˆ C r e …k‡ A †v K1

qw C w q A k‡ w A

qw x e …k‡ A †v

1

Substituting t for x/v, qw C x ˆ C r e …k‡ A †t

qw C w q A k‡ w A

qw e …k‡ A †t

1

(22.16)

By solving Eq. (22.16) with x = L or t = L/v, the concentration, C0 (in the stream immediately after the reach receiving the distributed flow), will be found. Equation (22.13) may then be used to find the concentration, C, at any distance downstream of the reach receiving the discharge.

Example 22.1 Indicator Bacteria Decay from Point and Distributed Sources

A stream is receiving waste containing indicator microorganisms from a point source and from a distributed source located 1.0 km downstream of the point source (refer to the accompanying sketch). The subscripts p and d refer to the point source and the distributed source, respectively. The distributed source discharges over a distance of 0.8 km. Other quantities indicated in the sketch have the following values: Qr ˆ 9500 m3 d 1 Qp ˆ 1050 m3 d 1 qd ˆ 500 m3 …km d†

1

C r ˆ 22=100 mL C p ˆ 1400=100 mL C d ˆ 2700=100 mL

The velocity of the stream is constant at 3 km d 1. A first-order model describes microorganism decay and the rate constant (base 10) is 0.072 h 1. Find the concentration of indicator microorganisms 2.0 km below the distributed source. Assume that the point and distributed source discharges mix well with the stream.

22 Effluent Disposal in Natural Waters

The stream concentration immediately after mixing with the point source discharge is calculated from Eq. (22.9b): C0 ˆ

C r Qr ‡ C p Qp 9500 m3 d ˆ Qr ‡ Qp

…22=100 mL† ‡ 1050 m3 d

1

9500 m3

1

d

‡

1050 m3

1

d

…1400=100 mL†

1

ˆ 159=100 mL

The time to travel from the point source down to the beginning of the distributed source (separated by a distance, d1) is t ˆ d 1 =v ˆ …1:0 km†= 3 km d

ˆ 0:333 d ˆ 8:00 h

1

Using Eq. (22.13), the concentration of indicator microorganisms in the stream at the beginning of the distributed source is C ˆ C 0 10

kt

ˆ

1 159 10 …0:072 h †…8:00 h† ˆ 42:2=100 mL 100 mL

Equation (22.16) applies to the reach to which the distributed source discharges. The area of flow is required to solve the equation. The area is calculated based on the average flow in the reach. At the beginning and end of the reach, the flows are Q1 and Q2, respectively. Q1 ˆ 9500 m3 d

‡ 1050 m3 d

1

ˆ 10 550 m3 d

1

Q2 ˆ Q1 ‡ qd d2 ˆ 10 550 m3 d

1

‡ 500 m3 km

1

1

1

d

…0:800 km† ˆ 10 950 m3 d

1

The average flow in the reach is Q1 ‡ Q2 10 550 ‡ 10 950 ˆ ˆ 10 750 m3 d 2 2

1

Now, v ˆ Q=A and A ˆ Q=v ˆ 10 750 m3 d

1

= 3000 m d

1

ˆ 3:58 m2

The term 500 m3 d 1 qd ˆ ˆ 0:140 d A …1000 m†…3:58 m2 †

1

Converting the die-off rate coefficient to base e: k e ˆ 2:3k 10 ˆ 2:3 0:072 h

1

ˆ 0:166 h

1

ˆ 3:97 d

1

The time to travel 0.8 km is t ˆ x=v ˆ …0:8 km†= 3 km d

1

ˆ 0:267 d

Substituting these quantities into Eq. (22.16), the concentration of indicator microorganisms at the end of the reach is C2 ˆ

42:2 e 100 mL

…0:140‡3:97†…0:267†

‡

0:140 d

1

2:7  103 =100 mL

…0:140 ‡ 3:97†d

1

1

e

…0:140‡3:97†…0:267†

ˆ 75:2=100 mL

Finally, applying Eq. (22.13) again for a distance of 2.0 km downstream of the distributed source: C ˆ C 0 10

kt

ˆ

1 75:2 10 …0:072 h † 100 mL

2:00 km 3 km d 1

…24 h d 1 † ˆ 5:3=100 mL

869

870

Theory and Practice of Water and Wastewater Treatment

Figure 22.5 Natural reaeration in a stream.

22.3 Dissolved Oxygen Variation in a Stream The DO concentration variation in a stream depends on many factors. The classical development of Streeter and Phelps (1925) considers the two main influences: oxygen decrease resulting from the exertion of BOD and oxygen replenishment by natural reaeration from the atmosphere. The approach is similar to the developments presented in Section 22.2 except that the reaction is a coupled phenomenon. Flow in the stream is assumed to be PF, and steady-state conditions are also assumed to exist. The consumption of BOD (which is conveniently expressed in terms of oxygen) is expressed by rL ˆ k 1L

(22.17)

where k1 is the rate constant for BOD exertion, L is BOD concentration, and rL is the rate of BOD exertion. The saturation concentration of DO is governed by Henry’s law. Natural turbulence in the stream enhances the mass transfer of oxygen between the stream and atmosphere. The change in DO caused by reaeration (Figure 22.5) is expressed by r aer ˆ k 2 …C s −C†

(22.18)

where C is concentration of DO, Cs is saturation concentration of oxygen, k2 is the rate constant for reaeration, and raer is the rate of reaeration. Similar to these developments, only one-dimensional flow and concentration variation are considered; i.e., the stream is well mixed in any cross-sectional plane. An elemental volume of the stream is shown in Figure 22.6.

Figure 22.6 Stream elemental volume.

22 Effluent Disposal in Natural Waters

The equation for the mass balance is QC

Q C‡

@C @C Δx ‡ rΔV ˆ ΔV @x @t

where r is the volumetric reaction rate and ΔV is the volume (AΔx) of the element. Assuming steady state (@C/@t = 0), simplifying the preceding equation, and substituting the relations for velocity and volume results in the usual expression with the application of the chain rule. v

@C dC dx dC ˆ ˆ ˆr @x dx dt dt

(22.19)

The reaction rate is a coupled phenomenon depending on BOD consumption and reaeration. r ˆ k 1 L ‡ k 2 …C s

(22.20)

C†

Substituting this equation into Eq. (22.19), the resulting differential equation is dC ˆ k 1 L ‡ k 2 …C s dt

(22.21)

C†

Differential equations that include a constant coupled with the variable are usually simplified by substitution of a new variable as follows. Define the oxygen deficit, D, as D ˆ C s −C

(22.22)

and substitute it into Eq. (22.21) to yield, dD ˆ k 1L dt

k2D

Before attempting to solve this equation, it must be recognized that BOD (L) is a function of time. The equation describing the relation of L and t is readily found by making a mass balance for BOD from which it will be determined that dL ˆ k1L dt

)

L ˆ L0 e

k1t

where L0 is the initial ultimate BOD. Substituting this into the differential equation: dD ˆ k 1 L0 e dt

k1t

k 2D

(22.23)

Equation (22.23) must be integrated between the following limits:

t ˆ 0; D ˆ D0 t ˆ t;

DˆD

also, from Section 22.2.3, t ˆ 0; L ˆ L0 D0 and L0 are the values after mixing of the waste flow with the stream and are found by applying the mixing equation [Eq. (22.9b)].

871

872

Theory and Practice of Water and Wastewater Treatment

Rearranging Eq. (22.23) and applying an integrating factor:

dD ‡ k 2D dt

ek 2 t ∫

ˆ ek 2 t k 1 L 0 e

d Dek 2 t ˆ k 1 L0 e…k 2 ∫

k 1 †t

k1t

dt

The expression resulting from the integration and substitution of the initial condition is Dˆ

k 1 L0 e k2 k1

k1t

e

k2t

‡ D0 e

k2t

(22.24)

or, expressing the equation in base 10 (where the primed constants are referred to base 10): Dˆ

k ´1 L0 10 k ´2 k 1´

k ´1 t

10

k ´2 t

‡ D0 10

k ´2 t

A plot of the deficit curve, known as a DO sag curve, is shown in Figure 22.7. The resulting deficit curve is the summation of the two contributing phenomena. The maximum deficit defines the critical point, which is defined by d2 D 99

H2O2

90–129

30–70

KMnO4

138–160

97–99

NaHSO3 (liq)

133

40

NaOCl (liq)

122

13.1–14.9

CaO (dry)

96

80–95

Ca(OH)2 (dry)

48

82–95

Other

(continued )

917

APPENDIXA

07/16/2018

918

Appendix A

12:40:22

Page 918

Table A.3 (Continued ) Bulk density g/100 mL

Agent

Active ingredients %

CuSO4 (dry)

120–144

99

H2SO4

170–182

62–93

H3PO4

157–169

75–85

H2SiF6 (fluosilicic acid)

123

24

Greensand

136

—

KCl (dry)

103

>99

NaCl (dry)

105–121

98.6–99.9

Na2CO3 (dry)

253

>99

NaF (dry)

144–168

95–98

NaOH (liq)

153–170

50–73

NH4OH (liq)

90

%NH3, 28–30

a) MW: 2000–60 000.

Source: Adapted from Flick (1991) and Kawamura (1991).

Table A.4 Conversion factors and constants. Acceleration of gravity, g = 9.807 m s

2

(32.174 ft s 2)

R ˆ 1:9872 cal deg

Universal gas constant;

1

ˆ 0:082 054 l atm deg ˆ 8:3144 J deg

1

1

mol

mol

1

mol

Standard atmosphere ˆ 101:325 kN m 14:696 lbf in: ˆ 101:325 kPa…1:013 bar† ˆ 10:333 m …33:899 ft† of water 2

1 bar = 105 N m

2

1

1 2

(14.504 lbf in. 2)

Density of dry air at 0 °C and 1 atm = 0.001 293 g cm

3

1 g molecular volume of a gas at 0 °C and 1 atm = 22.4 L 1 m head of water 20° C ˆ 0:009 79 N mm ˆ 9:790 kN m 1 acre = 43 560 ft

2

1:420 lbf in:

2

2

1 ft2 = 2.296 × 10

2

1 A (amp) = 1 C (coulomb)s

5

acre

1

1 atm = 760 mm of Hg = 29.9213 in. of Hg 1 Btu = 1.0551 kJ

1 kJ = 0.9478 Btu

1 cal (20 °C) = 0.003 966 Btu

1 Btu = 252.1 cal (20 °C)

1 cal = 4.184 J 1 cP = 0.01 g cm

1 J = 0.2390 cal 1

s

1

1 °C = 1.800 °F

APPENDIXA

07/16/2018

12:40:22

Page 919

Appendix A

Table A.4 (Continued ) 1 °F = 0.5556 °C o

C ˆ 59 …o F

o

32†

F ˆ 95 o C ‡ 32

°K = °C + 273.16 1 Faraday ˆ 96 493 C=eq ˆ 23 062:4 cal=V eq 1 fathom = 1.829 m

1 m = 0.5468 fathom

1 ft3 = 7.480 51 US gal

1 US gal = 0.133 680 ft3

1 ft3 = 28.316 L 1 ft lb s

1

1 L = 0.035 31 ft3 1

= 1.3558 J s

1Js

1

1

= 0.7376 ft lb s

1 grain = 0.064 80 g

1 g = 15.43 grain

1 Imp. gal = 4.546 L

1 L = 0.2200 Imp. gal

1 US gal = 3.785 L

1 L = 0.2642 US gal

1 US gal = 0.133 68 ft3

1 ft3 = 7.4805 US gal

1 US gal = 0.003 785 m3

1 m3 = 264.17 US gal

1 ha ˆ 2:471 acres

1 acre = 0.4047 ha

ˆ 10 000 m

2

1

1 hp = 745.7 J s

1Js

1

= 0.001 341 hp

1 in. = 25.4 mm 1J = 1Nm

1 mm = 0.0394 in. 1

1 knot = 1.852 km h

1 km h

1

= 0.5400 knot

1 kg = 2.204 62 lb

1 lb = 0.453 59 kg

1 km = 0.6214 mi

1 mi = 1.609 km

3

1000 L = 1 m

1 L = 0.001 00 m3

1 lm (at 5550 A) = 0.001 471 W

1 W = 679.8 lm (at 5550 A)

2

1 lm cm 1 lm m

2

= 1 lambert

= 0.0929 ft-c

1 ft-c = 10.76 lm cm

1 m = 3.281 ft

1 ft = 0.3048 m

1 m3 = 35.31 ft3

1 ft3 = 0.028 32 m3

1 mi = 5280 ft

1 ft = 1.894 × 10

1 N = 1 kg m s

2

4

mi

4

tonne

2

pi = 3.141 592 65 1 lb ˆ 453:6 g ˆ 16 oz 1 stoke = 1 cm2 s

1 g = 0.002 204 lb 1

1 ton (US) = 2000 lb 1 tonne (metric) = 1000 kg 1 lb = 4.536 × 10

1 tonne = 2204.6 lb 1W = 1Js

1

1 W = 0.7376 ft lb s

1

1 ft-lb s

1

= 1.3558 W

919

APPENDIXA

07/16/2018

920

Appendix A

12:40:22

Page 920

Table A.4 (Continued ) Loading rate and other conversion factors commonly used in environmental engineering 1 kg ha

1

1 kg m

2

1 kg m

3

1 kg m

3

3

d

1

d

1

d

1

3

1m m 1 m3 kg

1

1Lm

1

d

s

d 3

= 24.54 gal ft

2

1

d 1

d

1

= 80.5196 gal ft

d

1

1 1

= 1.473 gal min -ft

1 kJ kg

1

= 0.4303 Btu lb

1 kg ha

1

= 0.8922 lb/acre

2

1 m3 ha 1 L mg

1

1 lb ft

3

1 lb ft

3

d

1

d

1

1

d

= 4.882 kg m

2

d

1

= 16.01 kg m

3

d

1

= 16.019 kg m

3

1 gal ft

2

1 gal ft

2

d

1

= 0.0407 m3 m

d

1

= 0.1337 ft d

1

1 gal ft

2

d

1

1 gal ft

1

= 1.547 × 10

d

1

1 ft3 lb

1

= 0.06243 m3 kg

6

2

d

1

ft s

3

1

1

= 0.012 42 m m d

1

1

1 gal min -ft = 0.6791 L m 2-s

1

1

2

= 2.324 kJ kg

1

1 lb/acre = 1.1209 kg ha

1

1 Btu lb

1 lb/ton = 0.500 kg/tonne

3

= 0.0380 hp/1000 ft3

1

d

1

1 lb ft

2

1

1 kg/tonne = 2.000 lb/ton 1 kW m

1

1 lb/acre-d = 1.1209 kg ha

1

3

= 16.019 ft3 lb 1

2

= 0.062 43 lb ft

1

d

1

2

= 0.2048 lb ft

= 0.062 43 lb ft

2

1m m

= 0.8922 lb/acre-d

1 hp/1000 ft3 = 26.33 kW m

1 gal/acre-d = 0.009 354 m3 ha

= 106.91 gal/acre-d

= 0.1198 Mgal lb

1

1 kPa (gauge) = 0.1450 lb in.

1 Mgal lb 2

3

(gauge)

1 lb in.

2

1

= 8.344 L mg

1

d

1

1

(gauge) = 6.8948 kPa (gauge)

A.1 Normal Distribution Density function for normally distributed variable with mean, μ, and standard deviation, σ: 1 p…x† ˆ p e σ 2π

…x μ†2 =2σ 2

or

1 2 f …z† ˆ p e z =2 2π

where zˆ

μ

x σ

Cumulative distribution function for normally distributed variable: 1 z e F …z † ˆ p 2π ∫ 0

z2 =2

dz

A.2 Integrating Factor for Linear Differential Equations of the First Order A first-order linear differential equation of the form y´ ‡ P …x†y ˆ Q…x†

APPENDIXA

07/16/2018

12:40:23

Page 921

Appendix A

where P and Q are continuous functions of x can be solved by setting Q(x) = 0. This leads to the separable equation y´ ˆ P …x† y if y ≠ 0. Integrating, ln jyj ˆ

∫

P …x† dx ‡ ln jC j

where C is a constant. Raising each side to the base e, y ˆ Ce

∫

P …x† dx

or ye∫

P …x† dx

ˆC

Now,

P …x† dx d ye∫ dx

ˆ y´ e ∫

P …x† dx

‡ P…x†ye∫

P …x† dx

ˆ e∫

P …x† dx

y´ ‡ P …x†y

If each side of the original differential equation is multiplied by e∫P(x) dx, the equation may be written as P …x† dx d ye∫ dx

ˆ Q…x†e∫

P …x† dx

The implicit solution of the above equation is ye∫

P …x† dx

ˆ Q…x†e∫ ∫

P …x† dx

dx ‡ C

which can be solved for an explicit solution. The expression e∫P(x) dx is known as an integrating factor.

921

APPENDIXA

07/16/2018

922

Appendix A

12:40:23

Page 922

Table A.5 Physical properties of water (SI units). Specific weight γ kN m − 3

ρ kg m − 3

Dynamic viscositya) μ × 103 N s m−2

Kinematic viscositya) υ × 106 m2 s − 1

Surface tension σ N m−1

Vapor pressure pv kN m − 2

0

9.805

999.8

1.781

1.785

0.0765

0.61

5

9.807

1000.0

1.518

1.519

0.0749

0.87

10

9.804

999.7

1.307

1.306

0.0742

1.23

15

9.798

999.1

1.139

1.139

0.0735

1.70

20

9.789

998.2

1.002

1.003

0.0728

2.34

25

9.777

997.0

0.890

0.893

0.0720

3.17

30

9.764

995.7

0.798

0.800

0.0712

4.24

40

9.730

992.2

0.653

0.658

0.0696

7.38

50

9.689

988.0

0.547

0.553

0.0679

12.33

60

9.642

983.2

0.466

0.474

0.0662

19.92

70

9.589

977.8

0.404

0.413

0.0644

31.16

80

9.530

971.8

0.354

0.364

0.0626

47.34

90

9.466

965.3

0.315

0.326

0.0608

70.10

100

9.399

958.4

0.282

0.294

0.0589

101.33

Temperature °C

Density

a) In this text, when a table heading indicates that the number is multiplied by some factor, the recorded number must be divided by this factor to obtain the correct result. For instance, the dynamic viscosity at 0 °C is z = 1.781 × 10 3 N s m 2. The number recorded is z × (103) = 1.781 × 10 3 × 103 = 1.781. Source: Vennard and Street (1975). Reprinted by permission of Wiley.

APPENDIXA

07/16/2018

12:40:23

Page 923

Appendix A

Table A.6 Physical properties of water (US units). Specific weight γ lb ft − 3

Density ρ slug ft − 3

Dynamic viscositya) μ × 105 lb s ft − 2

Kinematic viscositya) υ × 105 ft2 s − 1

Surface tension σ lb ft − 1

Vapor pressure pv lb ft − 2

32

62.42

1.940

3.746

1.931

0.005 18

0.09

40

62.43

1.940

3.229

1.664

0.006 14

0.12

50

62.41

1.940

2.735

1.410

0.005 09

0.18

60

62.37

1.938

2.359

1.217

0.005 04

0.26

70

62.30

1.936

2.050

1.059

0.004 98

0.36

80

62.22

1.934

1.799

0.930

0.004 92

0.51

90

62.11

1.931

1.595

0.826

0.004 86

0.70

100

62.00

1.927

1.424

0.739

0.004 80

0.95

110

61.86

1.923

1.284

0.667

0.004 73

1.27

120

61.71

1.918

1.168

0.609

0.004 67

1.69

130

61.55

1.913

1.069

0.558

0.004 60

2.22

140

61.38

1.908

0.981

0.514

0.004 54

2.89

150

61.20

1.902

0.905

0.476

0.004 47

3.72

160

61.00

1.896

0.838

0.442

0.004 41

4.74

170

60.80

1.890

0.780

0.413

0.004 34

5.99

180

60.58

1.883

0.726

0.385

0.004 27

7.51

190

60.36

1.876

0.678

0.362

0.004 20

9.34

200

60.12

1.868

0.637

0.341

0.004 13

11.52

212

59.83

1.860

0.593

0.319

0.004 04

14.70

Temperature °F

a) In this text, when a table heading indicates that the number is multiplied by some factor, the recorded number must be divided by this factor to obtain the correct result. For instance, the dynamic viscosity at 32 °C is z = 3.746 × 10 3 lb s ft 2. The number recorded is z × (103) = 3.746 × 10 3 × 103 = 3.746. Source: Vennard and Street (1975). Reprinted by permission of Wiley.

Table A.7 Derived units. Quantity

SI unit

Symbol

Expression in other units

Electric charge

coulomb

C

Electric potential

volt

V

WA

Electric resistance

ohm

Ω

VA

Expression in SI basic units

As 1 1

m2 kg s

3

A

1

m2 kg s

3

A

1

2

Energy, work, heat

joule

J

Frequency

hertz

Hz

s

Force

newton

N

m kg s

Pressure, stress

pascal

Pa

Nm

W

1

Power

watt

Nm

Js

1

m kg s

2

1

kg m 2

1

m kg s

2 2

s 2

923

APPENDIXA

07/16/2018

924

Appendix A

12:40:24

Page 924

Table A.8 Commercial pipe sizes.

Metric sizes mm

Existing US pipe sizes in.

mm

4

101.6

5

127.0

150

6

152.4

200

8

203.2

250

10

254.0

300

12

304.8

14

355.6

15

381.0

16

406.4

18

457.2

20

508.0

400 500

21

533.4

600

24

609.6

700

27

685.8

800

30

762.0

900

36

914.4

1000

42

1066

1200

48

1219

1400

54

1371

1600

60

1524

66

1676

1800

72

1829

2000

78

1981

84

2134

2200

90

2286

2400

96

2438

102

2591

108

2743

114

2896

120

3048

144

3658

180

4572

204

5182

240

6096

2700 3000 3300 3600

APPENDIXA

07/16/2018

12:40:24

Page 925

Appendix A

Table A.9 Typical wall roughness factors for commercial conduits. Roughness height Material (new)

mm

ft

Asphalted cast iron

0.12

0.000 4

Cast iron

0.26

0.000 85

Commercial steel or wrought iron

0.046

0.000 15

Concrete

0.3–3.0

0.001–0.01

Drawn brass or copper tubing

0.000 15

0.000 005

Galvanized iron

0.15

0.000 5

Glass and plastic

Smooth

Smooth

Riveted steel

0.9–0.9.0

0.003–0.03

Wood stave

0.18–0.9

0.000 6–0.003

Figure A.1 Moody Diagram. Source: Moody (1944). Republished with permission of American Society of Mechanical Engineers. Permission conveyed through Copyright Clearance Center, Inc.

References APHA, AWWA, and WEF (2012). Introduction to conductivity. In: Standard Methods for the Examination of Water and Wastewater, 22e (ed. E.W. Rice, R.B. Baird, A.D. Eaton and L.S. Clesceri), 4–141. Washington, DC: American Public Health Association. Flick, E.W. (1991). Water Treatment Chemicals: An Industrial Guide. Park Ridge, NJ: Noyes Publications. Kawamura, S. (1991). Integrated Design of Water Treatment Facilities. Wiley. Moody, L.F. (1944). Friction Factors for Pipe Flow. Trans. Am. Soc. Mech. Eng. 66: 671–684. Vennard, J.K. and Street, R.L. (1975). Elementary Fluid Mechanics, 5e. Wiley.

925

APPENDIXA

07/16/2018

12:40:24

Page 926

927

Author Index Abbas, H, 796 Abdel-Gawad, ST, 517 Abel-Magid, IM, 808 Abis, KL, 793 Abraham, K, 761, 762 Absi, F, 110, 535 Acra, A, 547 Adams, CE, 666 Adrian, DD, 873 Ahmad, M-L, 380 Ahmad, R, 412 Ahmed, AU, 850 Ahmed, MT, 889, 892, 893 Akgiray, Ö, 402 Aksela, R, 531 Aksolekar, SR, 803 Al-Adham, SS, 422 Alaerts, GJ, 799 Alam, MZB, 525 Albertson, OE, 315, 708, 709 Alexiou, GE, 793 Ali, M, 627 Al-Shemmeri, T, 465 Alvarez, M, 387 Aly, OM, 473, 499 Ames, BN, 178, 181 Amirtharajah, A, 230, 358, 372, 387, 389, 403, 411, 412, 418 Anderson, B.C., 669 Anderson, DA, 339 Anderson, DL, 193 Anderson, EJ, 147 Anderson, GK, 105, 250, 757 Andrews, RC, 551 Anouk Simard, M, 552 Antonie, RL, 648, 649 APHA, 95, 96, 105, 106, 110, 137, 139, 156, 529, 581, 663, 704 Appeldoorn, KJ, 628

Arceivala, SJ, 803 Ardern, E, 568 Argaman, Y, 458 Arvai, A, 171 Asami, M, 423 ASCE, 220, 275, 277, 286, 291, 292, 294, 314, 315, 341, 344, 346, 588, 622, 628, 639, 643, 647, 648, 649, 651, 652, 654, 690, 692, 708, 753, 772, 832, 837, 840, 846 ASME, 231 ASTM, 443, 466, 469 ATV, 651 Au, K-K, 530 AWWA, 314, 341, 344, 346, 359, 418, 444, 445, 446, 469, 470 Ayoub, GA, 379 Babbitt, HE, 726 Bader, H, 525, 526 Baeyens, J, 838 Baier, U, 761 Baker, MN, 351, 387, 509 Baldry, MGC, 527 Balls, M, 183 Bandara, WM, 731 Bányai, É, 72 Barbosa, R.A., 756 Bare, JC, 900 Bark, K, 628 Barnes, RSK, 131 Barnett, E, 417 Barnhart, EL, 799, 800 Barr, K, 765, 766 Barrett, MJ, 699 Bastian, R, 849 Basu, OD, 523 Baumann, ER, 650, 726 Beck, SE, 541 Beer, KD, 149, 185, 509

Theory and Practice of Water and Wastewater Treatment, Second Edition. Ronald L. Droste and Ronald L. Gehr.  2019 John Wiley & Sons Inc. Published 2019 by John Wiley & Sons Inc. Companion website: www.wiley.com/go/droste/water

928

Author Index

Belanger, SE, 552 Bell, PRF, 864 Bellar, TA, 548 Benedek, A, 490 Benefield, LD, 226, 310, 802 Benjamin, MM, 25, 353, 355, 356, 358, 359, 453 Beran, B, 790 Bergman, EA, 466 Bergman, RA, 466 Berjenbruch, M, 645, 646 Berquist, SA, 619 Bewtra, JK, 345, 362, 517 Bhargava, DS, 365, 366, 367, 376, 669 Binxin, W, 750 Bishop, E, 98 Bishop, JW, 174 Biswal, BK, 143 Bitton, G, 713, 714, 715, 725, 749 Black & Veatch Corp, 510, 512, 514, 520, 523, 524, 526, 537 Black, JG, 122 Blatchley, ER, 512, 538 Blaustein, RA, 159 Block, P, 531 Blum, DJW, 183, 749 Bock, E, 626 Boehm, AB, 157 Boehm, WA, 148 Bohrerova, Z, 539 Bolton, JR, 542, 545 Boltz, JP, 640, 641, 651, 652, 658, 660 Bordeleau, EI, 761, 762 Bosich, JF, 53 Bouchard, DC, 173 Bougrier, C, 762 Bourgin, M, 538 Boussaid, AL, 762 Bouzas, A, 631 Boyko, B, 607 Boyle, MA, 548 Boyle, WC, 626, 695 Brady, TJ, 552 Breese, S, 314 Brillas, E, 456 Brinck, WL, 449 Brock, TD, 713 Brown, D, 517, 524 Brown, GD, 714 Brown, JP, 761 Brown, MJ, 664 Brunauer, S, 476 Brundtland, G, 887 Budd, GC, 215

Buisman, C, 132 Bulusu, KR, 459, 460 Bumbac, C, 635 Burger, G, 672 Burkehead, CE, 573 Busenberg, E, 73, 445 Cabral, JPS, 147 Cadavid-Rodriguez, LS, 277 Cakir, FY, 721 Caldwell, DH, 446 California EPA, 550 Cameron, GN, 552 Camp, TR, 173, 281, 282, 294, 303, 356, 396, 874 Canter, LW, 795, 796 Cantwell, RE, 540, 541 Cao, Y, 901 Carlson, KH, 447, 448 Carman, PC, 394 Carny, P, 175, 176 Cartmell, E, 761 Casanova, LM, 196, 197 Cashman, S, 901 Cavallini, G, 414 CCME, 184, 192, 849, 860, 862, 864, 876 CDC, 174 Cecen, F, 423 Cha, DK, 713 Chakrabarti, T, 756 Chamberlin, C, 159, 161, 162 Chan, SHS, 456 Chaney, RL, 848, 849 Chang, JC, 183 Chang, M, 396 Chang, SD, 549 Chemguide, 72 Chen, X, 635 Chen, Z, 901 Cheng, J, 671 Chhetri, RK, 530, 531 Chick, H, 510 Chiesa, SC, 663, 712, 714 Chisolm, KA, 614 Chiu, YC, 761 Chorus, I, 176 Chow, VT, 312, 364, 366, 367 Chowdhury, ZK, 474, 476, 478, 495, 496 Christensen, DR, 601, 873 Chu, CP, 761, 762, 607, 712 Chuboda, J, 602 Chung, YC, 87, 738 Churchill, M, 877 City of Ottawa, 443, 878

Author Index

Claros, J, 626 Clasen, T, 547 Cleasby, JL, 58, 353, 356, 368, 403, 404, 405, 412, 413, 417, 435, Clifford, DA, 457, 458 CML, 901 Coackley, P, 843 Cockburn, A, 130 Coelho, NMH, 761 Cohen, A, 174 Cole, LD, 434 Colomer, FL, 790 Colwell, RR, 149 Comeau, Y, 196 Comstock, GW, 431 Corominas, L, 888, 889, 892, 893, 899, 901 Cortruvo, JA, 182 Corvi, R, 181 Council of the European Union, 549 Covar, AP, 877 Cox, HHJ, 760 Crapper, DR, 173 Craun, GF, 185, 387 Crites, RW, 806, 807, 808, 809, 814, 816 Crittenden, JC, 341. 344, 371, 372, 375, 390, 402, 412, 434, 456, 469, 470, 472, 495, 496, 532, 534, 537, 544, 550, 831 Crone, BC, 731 Cullis, IF, 762 Culp, R., 460 Culp, RL, 412 Curds, CR, 130 Currie, J, 750 Curtis, TO, 789 Dague, RR, 747 Daigger, GT, 651, 652, 658, 660, 704, 714 Daims, H, 143 Dalla Costa, RF, 472 Damewood, M, 419 Dammel, EE, 826 Danckwerts, PV, 336 Daneshmand, TN, 834 Daniel, FB, 181, 551 Darby, JL, 539, 541 Das, D, 607 Dasgupta, A, 196 Datar, MT, 669 Davis, RR, 449 De Baere, L, 762 De Clercq, J, 599 De Leon, R, 414 de Rijk, SE, 838

Degrémont, 419 Dell’Erba, A, 550 DeLong, EF, 141 Deng, Y, 456 Dept. of National Health and Welfare, 459 Desjardins, C, 381 Dharmarajah, AH, 403, 404 Di Bernardo, L, 307, 308, 413 Diak, J, 834 Dick, RI, 703 Dietrich, M, 456 Dillingham, JH, 435 Disinfection Committee, 549 Dixon, KL, 526, 527 Dobbins, WE, 877 Dohanyos, M, 761 Dohmann, M, 414 Dowbiggin, WR, 314 Doyle, JD, 764 Driscoll, CT, 353 Droste, RL,666, 667, 668, 669, 730, 745, 749, 758, 776, 799 Ducoste, JJ, 796 Dunkin, N, 530 Dunn, G, 187 Duranceau, SJ, 468, 469, 470, 472 Dzombak, DA, 649 Eastman, JA, 741, 746 Eckenfelder, WW, Jr, 113, 254, 458, 572, 585, 653 Edwards, FG, 380, 381 Edwards, M, 212, 353, 355, 427, 442, 443, 446 Edzwald, JK, 837, 839 Egan, D, 473, 864 Eismann, F, 727 Ekama, GA, 318, 581, 602, 708 Elder, D, 215 Eliassen, R, 387, 397 Ellis, KV, 422 El-Rehaili, AM, 110 Emmerson, RHC, 888 Englande, AJ, Jr, 570, 571, 596, 668, 795, 796 Environment Canada, 195, 847, 849 Environmental Health Directorate, 187 Ergun, S, 396 Ericsson, B, 714 Eriksson, L, 714 Erskine, DB, 492 Eskicioglu, C, 761 Espino de la O, E, 794 Espino, E, 793 Evett, JB, 228, 231

929

930

Author Index

Fair, GM, 339, 342, 344, 390, 392, 410, 411 Falconer, IR, 171 Falkinham, JO, 149 Falsanisi, D, 528 Fan, K, 403 Farooq, S, 421, 422 Fattah, KP, 765, 766 Faust, SD, 473, 499 Fdz-Polanco, F, 766 Feachem, RG, 147, 195 Federal Register, 518, 524, 525, 533, 544, 549, 550 Ferguson, JE, 168, 741, 746 Finch, GW, 797 Finkbeiner, M, 890, 892, 898 Finney, DJ, 139 Flick, EW, 457, 474 Flores, MJ, 528, 529 Fongastitkul, P, 759 Foo, KY, 476 Ford, DL, 254, 572, 585, 794 Forster, CF, 414, 670, 671, 762 Fox, JC, 154 Fox, P, 759 Fraleigh, PC, 552 France, PW, 365 Franck, EU, 58 Fraser, CG, 799 French, MS, 527 Freundlich, H, 476 Friedrich, E, 889, 892, 894, 895, 897, 901 Fritz, JJ, 790 Gagnon, G, 525 Galvagno, G, 765 Ganczarczyk, J, 668 Gao, DXY, 635 Garber WF, 778 Garcia-Segura, S, 456 Gehr, R, 527, 531, 539, 541 Geldreich, EE, 156, 157, 217 Gerba, CP, 130 Germain, JE, 653 Gerval, R, 534 Ghosh, S, 760 Gibbs, JW, 35 Gillot, S, 671, 800 Gilmore, KR, 898 Gloyna, EF, 791, 792, 793, 817 GLUMRB, 520, 691, 796 Goedkoop, M, 896 Golgrabe, JC, 423 Gomes, J, 538 Gonce, N, 527

Gonze, E, 761 Goodman, BL, 570, 571, 596, 668 Gordon, G, 517 Grady, CPL, Jr, 649, 658 Grady, LE, 587 Grasso, D, 536 Gray, DMD, 714 Gray, JB, 359 Greenfield, PF, 570 Greenwood, M, 138 Gregory, J, 351 Griese, MH, 527 Groves, KP, 692, 694 Guibai, L, 351 Guillot, E, 148 Gujer, W, 714, 723, 740, 746 Gupta, A, 727 Gutiérrez-Alfaro, S, 548 Haarhoff, J, 353, 839 Haas, CN, 179, 180, 514, 528 Haberman, WL, 346 Haji Malayeri, A, 542, 543 Hallmich, C, 539 Hamann, CL, 215 Hameed, BH, 476 Hammer, MJ, 840 Hamoda, MF, 414 Hand, DW, 473 Haney, JT, 863 Hansen, J, 187 Hanson, AT, 58, 353 Hao, OJ, 622 Hao, X, 628 Harada, H, 767 Harper, SR, 724 Harremoës, P, 727, 745, 747 Harris, WW, 747 Hartwig, TL, 763 Hassan, M, 456 Hassard, F, 648, 649 Hauschild, MZ, 888 Hazen, A, 294, 392 HDR Engineering, 831 Health and Welfare Canada, 195 Health Canada, 158, 518, 519, 549, 550 Hecky, RE, 864 Heinke, GW, 796, 797 Heinke, KR, 454, 455 Hellinga, C, 626 Hellman, T.M., 176 Hendricks, D, 387, 420, 421 Henrie, T, 171, 172

Author Index

Henze, M, 196, 715, 727, 745, 747 Hermanowicz, SW, 451, 452 Hermant, BM, 523 Herschel, F, 192, 194 Higbie, R, 336 Higgins, MJ, 704, 834 Hijnen, WAM, 540 Himberg, K, 212, 215 Hindin, E, 458 Hlavsa, 185 Hoffman, FA, 174 Hofkes, EH, 420 Hofmann, R, 540, 541, 551 Hoigné, J, 517, 525, 526 Højri, B, 140 Holden, GW, 550 Holeton, C, 174 Hom, LW, 528 Hongve, D, 551 Hoover, SR, 573 Horan, NJ, 276, 277 Howe, KJ, 336 Huang, J, 622 Hubbell Inc., 53 Hudson, HE, Jr, 319, 320, 321 Huijbregts, MAJ, 888, 900 Huisman, L, 363, 364, 376, 377, 378, 379, 420, 421 Hulshoff Pol, LW, 754, 755 Humbert, S, 900 Humphries, H, 409 Hurst, GH, 527 Hutchins, RA, 490 Hutchinson, A, 454 Iatrou, A, 527 IAWPRC, 149, 150 Ikumi, DS, 744 Imasuen, E, 380, 381 IMO, 553 InfilcoDegremont, 840 Inland Waters Directorate, 170 Ioannou-Ttofa, L, 892 IPNI, 851 Iranpou, R, 760 Irvine, RL, 712 Isom, BG, 183 IUVA, 540 Ives, KJ, 387 IWA Task Group, 598, 601, 603, 619, 722, 745 Iza, J, 759 Jagger, J, 539 Jardin, N, 629

Jarvie, ME, 498 Jenkins, D, 87, 445, 451, 452, 570, 571, 713 Jenkins, SR, 763, 764 Jeris, JS, 723 Jetten, MSM, 627 Jiminez, B, 250 Joffe, J, 528 Johansen, HL, 431 Johnson, R.W., 150 Joilliet, O, 888 Jolis, D, 380, 761 Jolliet, O, 888, 898, 900 Jones, BRS, 843 Jones, GN, 702 Jørgensen, PE, 570 Jouanneau, S, 110 Judd, C, 637, 638, 712 Judd, S, 637, 638, 665, 712 Julien, L, 879, 881 Jungfer, C, 540 Junli, H, 524 Kadir, K, 159 Kappeler, J, 714 Kargi, F, 790 Karpova, T, 531 Kartal, B, 627 Kawamura, S, 226, 230, 354, 358, 369, 372, 400, 412, 413, 421 Keinath, TM, 703, 704, 709 Keiser, DA, 174 Kelly, HG, 669, 670 Kennedy, JB, 112, 138 Kennedy, KJ, 745, 749, 758 Kepp, U, 761, 762 Kim, BR, 105 Kim, K, 168 Kingsley, C, 834 Kinner, NE, 130 Kinsils, MG, 423 Klenk, K, 527 Klerks, PL, 552 Knocke, WR, 527, 834, 840 Knoll, H, 194, 791, 797, 798 Kobayashi, T, 760 Kobylinski, EA, 530 Koch, G, 625 Köck-Schulmeyer, M, 171 Koers, DA, 668 Koesser, H, 646 Koorse, SJ, 186 Kopeloff, LM, 173 Koschelnik, J, 140

931

932

Author Index

Koster, IW, 724 Kouame, Y, 514 Kovács, R, 670 Krasner, SW, 524 Krishnamoorthy, R, 666, 668, 669 Krzeminski, P, 637 Kuchenrither, RD, 446 Kuczynski, L, 379 Kumara, S, 380, 381 Kusakabe, K, 456 Kynch, GJ, 702 Lackner, S, 626, 627 Lafitte-Trouque, S, 762 Laidler, KJ, 12 Langelier, W.F., 14, 443 Langenbach, K, 422 Langmuir, I, 475 Lanham, AB, 628 Larson, TE, 435 Latkar, M, 756 Lawler, DF, 249, 336, 356, 358, 359, 521 Lawrence, AL, 584, 588, 608, 729 Lawrence, WB, 446 Layden, NM, 669, 670 Lazarova, V, 537, 538 Lee, KM, 901 Lee, RG, 51, 526, 527 Lei, L, 630, 631 Leland, DE, 419 Leong, LYC, 519, 533, 536, 537, 540, 550, 551 Lester, JN, 664 Letterman, RD, 353, 356 Lettinga, G, 722, 727, 754, 755, 756 Levenspiel, O, 249, 250 Levin, GV, 87 Lewis, WK, 332 Li, A, 759, 760 Li, H, 639 Li, JW, 524 Li, W-W, 887 Li, Y-Y, 760 Lichtenberg, JJ, 548 Lim, HC, 649 Linden, KG, 541, 542 Lindenauer, KG, 539 Linne, SR, 714 Liu, C, 228, 231 Liu, HW, 762 Liu, X, 458 Lockett, WT, 568 Lodgson, GS, 418 Loehr, RC, 666, 668, 669

Lopman, BA, 150 Loret, JF, 148 Lotti, T, 627 Lowry, JD, 175, 176 Lowry, SB, 175, 176 Lund, V, 551 Lundin, M, 892 Luukkonen, T, 527, 529, 530, 531, 550 Lytle, DA, 337, 442 Mackie AL, 381 Madia, F, 181 Madigan, M, 725 Madigan, MT, 124, 125, 126, 128, 129, 142 Madoux-Humery, AS, 878 MAF, 850, 852, 853 Maillacheruvu, KY, 764 Malina, JF, 802 Malley, JP, 423 Mamais, D, 764 Mamane, H, 540 Mancini, JL, 799, 800 Mara, DD, 793, 794 Marsalek, J, 878 Marshall, WL, 58 Martel, AE, 25 Martel, CJ, 834 Marten, WL, 714 Martin, JH, 672, 673 Martin, N, 539 Martinez, JA, 794 Martins, AP, 714 Masschelein, WJ, 538, 540 Massé, D, 749 Masten, SJ, 171 Mattila, T, 531 Mavinic, DS, 668, 669, 765 Mayhugh, JR, 397 Mayo, A, 802 McCarty, PL, 584, 588, 601, 608, 625, 723, 729, 752, 767, 873 McClelland, NI, 861 McCrady, MH, 139 McGarry, MG, 792 McGhee, TJ, 323 McGuigan, KG, 547, 548 McKinney, RE, 790 McMahon, RF, 552 McNeill, LS, 212, 443, 446 McQuarrie, JP, 640, 641 Meadows, D, 888 Meier, JR, 181 Merkel, A, 887

Author Index

Merlo, R, 414 Merrill, DT, 445, 446 Merrill, MS, 615 Meserve, RL, 173, 874 Metcalf and Eddy, 279, 397, 616, 663, 832, 848 Metcalf and Eddy:AECOM, 193, 194, 195, 196, 277, 291, 292, 307, 314, 315, 361, 412, 414, 474, 520, 571, 588, 614, 616, 646, 647, 648, 651, 652, 657, 658, 672, 694, 708, 710, 745, 753, 767, 825, 840 Middlebrooks, EJ, 801 Miles, SL, 140 Miller, RG, 353 Miltner, RJ, 423 Ministry of Housing, Spatial Planning and the Environment, 901 Mitch, W, 550 Mitchell, R, 159, 161, 162 MOE, 848, 850, 852, 853 Moelling, K, 130, 131 Mohanrao, GJ, 794 Monk, RDG, 363, 372, 417 Monod, J, 134, 569 Monteith, HD, 748 Moore, EW, 111 Moregenroth, E, 634 Moreira, RA, 131 Morgan, JJ, 451, 461 Morowitz, HJ, 544 Morris, JC, 512 Morton, RK, 346 Mouchet, P, 449, 450 Mueller, JA, 689, 694, 696, 873 Muller, CD, 761, 762 Muller, J, 761, 762 Münch, E, 765, 766 Munro, PM, 149 Munz, C, 336 Murphy, KL, 607 Murray, WD, 727 Mutoti, G, 524 MWH, 369 Nakasone, H, 699, 700 Nakhla, G, 421, 422 Narbaitz, RM, 466, 469, 474, 486 Nasserddine, M, 53 Naunovic, Z, 541 Nawlakhe, WG, 459, 460 Nazzal, FF, 379 Neethling, JB, 87, 738 Neff, CH, 170 Nelson, DL, 167

Nelson, KL, 159, 793 Nemerow, NL, 196 Neri, LC, 431 Nessim, Y, 541 Neville, AM, 112, 138 Nicholas, WR, 345, 362 Nicholson, GA, 714 Noot, DK, 551 Novak, JT, 704, 834 Novák, L, 714 Nowak, G, 714 Nowell, LH, 517 Nozhevnikova, AN, 626, 627 NWRI, 545 O’Connor, DJ, 877 O’Day, DK, 51, 53 Ødegaard, H, 641, 642 Ohanian, EV, 185 Ojha, CSP, 365, 366, 367, 376 Okun, DA, 194, 226, 354, 363, 365, 370, 410 Oldshue, JY, 361 Oleskiewicz, JA, 619 Olukanni, DO, 796 OMAFR, 851 OMOECC, 193, 194, 197, 198, 878 ÖNORM, 545 Oomen, JHCM, 420 Oragui, JI, 802 Ordon, CJ, 660 Orgeron, DJ, 794 Örmeci, B, 834 OSHA, 338 Ostrem, K, 762 Oswald, WJ, 791, 793 Ott, WR, 859 Ottaway, JM, 98 Otter, JA, 163 Owen, WF, 568, 569, 648, 729, 760, 764, 778 Owens, M, 877 Paramasivam, R, 459, 460 Park, C, 761 Parker, CD, 794 Parker, DG, 834 Parker, DS, 615 Parker, W, 672 Parker, WJ, 749 Parkin, GF, 760, 764, 778 Parsons, SA, 764 Pasqualino, JC, 901 Patel, GB, 724 Pavanello, R, 863

933

934

Author Index

Pawlowski, A, 901 Pearce, GK, 461, 470 Pehkonen, SO, 527, 529, 530, 550 Peng, CG, 458 Perry, RH, 672 Persich, B, 192 Pescod, MB, 792 Peters, GM, 901 Phelps, EB, 870 Philippe, N, 130 Pignatello, JJ, 456 Pilli, S, 761, 762 Pincince, AB, 698 Pinkernell, U, 531 pinpoint floc, 712 Pitt, P, 713 Plum, V, 380 Plummer, L.N., 73, 445 Pohland, FG, 724 Pollard, PC, 570 Pons, W, 185, 509 Pontius, F.W., 506 Pöpel, HJ, 629 Porges, N, 573 Porter, R, 153 Post Mixing, 361 Post, GB, 174 Prati, L, 862, 863 Prest, EI, 524 Preuss, A, 531 Price, DS, 798 Pronk, M, 634, 635 Pryshlakivsky, J, 888, 898 Pujol, R, 646 Qakim, SR, 226 Rabb-Waytowich, D, 174 Racault, Y, 793, 795 Ragazzo, P, 527, 531, 550 Ragush, Y, 798 Railsback, SF, 878 Rakness, KL, 531, 533, 534 Ramalho, RS, 299 Ramani, R, 791 Randall, CW, 802 Randtke, SJ, 475 Rasmussen, H, 632, 834 Ratsak, CH, 567 Raunkjær, K, 196 Rebhun, M, 520 Redmon, DT, 692 Reed, SC, 409, 795, 796, 801, 806, 807, 852

Regunathan, P, 217 Reiber, S, 173 Reynolds, TD, 361, 668 Rich, LG, 803 Richards, PA, 361 Richardson, JF, 403 Richardson, SD, 550 Richter, CA, 306, 379 Rico, DP, 790 Riggs, JL, 150 Riley, DW, 670, 671 Rinzema, A, 727 Rittman, BE, 422, 423, 767 Roberts, PV, 336 Robinson, J, 840 Robinson, RA, 471 Robinson, RB, 449 Roig, MG, 454 Rojas, J, 670 Rook, JJ, 548 Rose, JB, 196, 197 Ross, B, 648 Rossum, JR, 445, 446 Rouse, H, 231 Rowe, DR, 808 Rowley, HV, 901 Rubel, F, Jr, 457, 458 Rushton, JH, 361 Russell, LL, 14 Rusten, B, 643 Ryan, JA, 849 Ryckebosch, E, 732 Saatçi, AM, 401, 402 Salvadori, MI, 509 Sancar, A, 539 Sanchez, WA, 666, 668, 669 Sanders, TG, 873 Sandford, DS, 614 Sanin, FD, 833, 834 Sanks, RL, 226, 457, 831 Sant’Anna, GL, Jr, 756 Santoro, D, 528 Santos, A, 665 Sanz, I, 766 Sarathy, SR, 528, 529, 531, 556 Sassoubre, LM, 157 Sastry, CA, 794 Sawicki, JM, 291, 292 Sawyer, CN, 81 Sayell, KM, 449 Sayre, IM, 184 Schindler, DW, 864

Author Index

Schippers, JC, 467 Schlegel, S, 646 Schmidheiny, P, 761 Schmidtke, NW, 847 Schock, MR, 170, 442 Schön, G, 632 Schroeder, ED, 588, 656, 826 Schrotter, BB, 460, 468 Schrotter, JC, 460, 468 Schuler, PF, 420 Schuliger, WG, 492 Schultz, KL, 653 Schulz, CR, 226, 354, 363, 365, 376 Schulz, JR, 589 Searcy, C, 888, 898 Sedlak, D, 550 Selna, MW, 588 Seviour, EM, 713 Seviour, RJ, 713 Shahriari, H, 598 Shammas, NK, 193, 197, 198 Shao, YJ, 713 Shindala, A, 415 Sibony, J, 381 Siegrist, H, 625, 760 Siegrist, RL, 193 Sillén, LG, 25 Simpkin, TJ, 626 Sims, RC, 458 Singer, PC, 249, 354, 521, 537, 549, 550 Singh, R, 460 Sinton, LW, 159, 160, 161 Skiadas, VI, 761 Slawson, RM, 539 Small, FH, 176 Smith, DW, 194, 791, 797, 798 Smoluchowski, Mv, 368 Snoeyink, VL, 422, 423, 445, 475 Soddell, JA, 713 Sommers, LE, 848 Sorenson, DL, 850 Sorial, GA, 474 Speece, RE, 183, 725, 747, 749, 764, 767 Speight, J, 38 Spengel, DB, 649 Sprott, GD, 724 Stacey, G, 178 Stante, L, 629 Statistics Canada, 851 Stehouwer, PP, 553 Stein, PC, 356 Steinbach, S, 839 Stenstrom, MK, 721

Stephenson, JR, 748 Stephenson, R, 761 Stephenson, T, 463, 465, 616, 636, 638, 760 Stokes, RH, 471 Stoltenberg, HA, 460 Street, RL, 310 Streeter, HW, 870 Strious, M, 627 Stukenberg, JR, 778 Stumm, W, 354, 451, 461 Su, MC, 714 Suffet, IH, 473, 522 Surampalli, RY, 650 Sutton, PM, 759, 760 Suzuki, H, 456 Sykes, RM, 607 Szabo, A, 452 Takács, I, 599 Talinli, I, 105 Tambo, N, 358 Tate, C, 532, 536 Taylor, JS, 469, 470, 472 Taylor, SJ, 468 Tchio, M, 418 Tchobanoglous, G, 387, 397, 656 Terry, DW, 690 Tetra Tech, 873, 878 Thiem, LT, 690 Thirumurthi, D, 797 Thomann, RV, 873 Thomas, HA, Jr, 111 Thompson, C, 535 Tiedje, J, 725 Tijani, JO, 172 Tischler, LF, 791, 792, 794, 817 Tobiason, JE, 413 Toerien, DF, 722 Toprak, H, 794 Tratnyek, PG, 526 Troester, M, 172, 173 Truax, DD, 412 Trussell, RR, 363, 372, 396, 457, 458 Tsitonaki, A, 456 Uemura, S, 767 Uhl, VW, 359 Unz, RF, 713 US Government, 849, 852 USEPA, 157, 171, 181, 187, 189, 191, 212, 213, 254, 278, 279, 281, 254, 278, 279, 291, 405, 406, 412, 443, 451, 452, 454, 455, 477, 511, 521, 523, 526, 535, 545, 549, 619, 620, 621, 622, 626,b 629,

935

936

Author Index

633, 670, 672, 692, 694, 752, 795, 796, 798, 801, 802, 804, 805, 806, 807, 808, 809, 810, 811, 814, 815, 816, 825, 832, 833, 834, 839, 843, 845, 846, 847, 848, 851, 852, 860, 861, 862, 863, 873, 876 USGS, 552 Uslu, G, 541 van den Berg, L, 727 van Dijk, JC, 420 van Haandel, AC, 755, 756 van Kessel, MAHJ, 143 Vandenabeele, J, 450 Vanparys, R, 181 Vanrolleghem, PA, 671, 800 Velz, CJ, 652, 799 Vennard, JK, 310 Verdouw, J, 467 Verschueren, K, 168, 176, 177 Verstraete, W, 887 Vesilind, PA, 702, 703, 827, 828, 834, 840, 842 Viacoumi, S, 353, 356 Viessman, W, Jr, 840 Vlaeminck, SE, 887 Vogel, J, 176 Vogt, GM, 762 Volk, CJ, 525 von Gunten, U, 536, 550, 551 von Sonntag, C, 536, 550, 551 Voudrais, EA, 527 Waddell, SL, 573 Wagner, M, 110, 529 Wahlberg, EJ, 703, 704 Wallis, GB, 346 Walsh, P, 859 Wang, D, 538 Wang, LK, 193, 197, 198 Wanner, J, 620, 714 Watson, HE, 510 Webber, MA, 522 Weber, SG, 147 Weber, WJ, Jr, 536 Weddle, CL, 87, 570, 571 WEF, 220, 275, 277, 286, 291, 292, 294, 314, 315, 540, 588, 620, 622, 628, 629, 630, 631, 637, 639, 640, 641, 642, 643, 646, 647, 648, 649, 651, 652, 653, 654, 657, 692, 708, 712, 753, 772, 753, 772, 832, 837, 840, 846, 852

Wen, CY, 402 Wenzel, H, 901 Wernet, G, 901 Wert, EC, 534, 535, 551 Wheeler, W, 859 White, FM, 657 White, MJD, 581 Whitley, G, 797 Whitman, WE, 332 WHO, 147, 174, 185, 187, 191, 207, 212, 213, 447, 549, 550, 551 Wid, N, 276 Wild, D, 766 Wilkinson, JL, 173 Willey, JM, 124 Williams, TM, 713 Wittmann, JW, 791, 792, 817 Woese, C, 121 Wong, JM, 381 Wood, WE, 420, 421 Woosely, RD, 457, 458 WPCF, 338, 652 WRC, 380, 381, 588, 603, 619, 745 Wrigley, TJ, 795 Wu, W, 750 Yang, BS, 757, 758, 766 Yao, KM, 304, 305, 387 Yeh, AC, 757 Yezli, S, 163 Yoo, RS, 460 Young, JC, 380, 381, 757, 758, 766 Yu, YH, 402 Yule, GM, 138 Zaki, WN, 403 Zehnder, AJ, 723, 725 Zehnder, JB, 723, 740, 746 Zeikus, JG, 724 Zhang, J, 536 Zhang, Z, 525 Zhao, R, 456 Zhelev, T, 670 Zimmer, JL, 539 Zinder, SH, 724 Zitomer, DH, 725 Zoetemeyer, RJ, 722

937

Subject Index χ2 criterion, 603 7Q10, 875 7Q20, 875 acetoclastic bacteria, 724 acetogenesis, 723 acetotrophic. See acetoclastic acid definitiion, 58 strong, 66 weak, 65, 68, 73 acids:alkalinity ratio anaerobic digestion, 769 acid attack, 50 acid-base effects on gas transfer, 336 indicators, table, 72 titration, 96 acid formers, 724 acidity, 75, 769 acidogenesis, 723 acidophile, 127 acid rain, 171 Acinetobacter, 629 Actinomycetes, 176 activation energy for reaction, 19 activated alumina arsenic removal, 454 activated carbon, 207, 217, 459, 549 dechlorination, 523 granular, 207, 473, 537, 759 powdered, 207, 473 regeneration, 473, 477 synthetic, 473 treatment, 217 activated primary tank, 630 activated silica, 354 activated sludge, 218, 665, 668, 701, 729 advanced model, 596, 618

design parameters, 615 dissolved oxygen, 615, 663, 691 extended aeration, 612 high rate, 614 modified aeration, 614 nitrification, 616 nitrogen requirement, 573 oxygen uptake, 663 phosphorus requirement, 573 process variations, 609 pure oxygen, 615 recycle, 593 sludge production, 589, 604 specific gravity, 826 activity, 36, 46 coefficient, 15, 36, 444 electrochemical, 50 acquired immunodeficiency syndrome, 150 adenosine diphosphate, 87, 123, 126 adenosine monophosphate, 87 adenosine triphosphate, 87, 570, 738 adenovirus, 150 adsorbate, definition, 472 adsorber continuous flow, 477 fixed bed, 477 fixed bed design, 478 adsorption, 159, 353 isotherms, 474 operating line, 484 zone, 478 zone capacity, 482 advanced oxidation, 208, 217 processes, 455, 537 advection, 235 aerated lagoon, 790, 802 mixing, 802 power input, 802 aerated pond, 795

Theory and Practice of Water and Wastewater Treatment, Second Edition. Ronald L. Droste and Ronald L. Gehr.  2019 John Wiley & Sons Inc. Published 2019 by John Wiley & Sons Inc. Companion website: www.wiley.com/go/droste/water

938

Subject Index

aeration, 208,390 biological treatment, 689 equalization basins, 254 iron and manganese removal, 449 membrane fouling, 637 theory, 337 water treatment, 336 aeration devices in water treatment, 339 aerator cascade, 339 diffused, 344 spray, 341 wind effects, 344 aerobe obligate, 126 aerobic digestion, 748, 830 Adams model, 667 autothermal, 669 dissolved oxygen, 691 oxygen uptake, 668 pH, 669 thermophilic, 669 aerobic pond. See maturation pond aesthetic parameters, 191 AIDS. See acquired immunodeficiency syndrome air pollution, 171 air saturation concentrations, 838 air scour system, 412 air stripper, 218 air stripping, 208, 218, 337 alcohol, 82, 83 aldehyde, 83, 84, 177 algae, 128, 132, 161, 278, 337, 789, 793, 874 composition, 791 growth in filters, 420 red, 128 stabilization pond, 791 alginate, 354 aliphatic compound, 82, 83 alkali metal, 7 alkalinity, 74, 351, 431, 433, 442, 451, 536, 619, 726, 769 denitrification, 626 usable in anaerobic digestion, 764 alkaliphiles, 127 alkane, 83 alkene, 83, 177, 513 alkylbenzenesulfonates, 174 alkyne, 83 alloys, 442 alpha decay, 27 alpha radiation, 175

alum, 351 addition to ponds, 799 as dewatering agent, 833 fluoride removal, 459 recovery, 355, 830 alumina, activated, 457 aluminate, 450 aluminum, 173 Alveolates, 129 Alzheimer’s disease, 173, 353 Ames test, 178, 181, 550, 551 amide, 83 amine, 83, 177, 353 amino acid, 83, 86, 125 ammonia, 169, 208, 442 anaerobic digestion, 726 content in urine, 195 interference in COD test, 105 oxidizing bacteria, 109, 616 reaction with chlorine, 513 removal by ion exchange, 458 stripping, 218 wastewater treatment plant effluent, 873 Amoebozoa, 130 amperometric titration, 516, 529 Amstichthys nobilis, 795 Anabaena, 171 anabolism, 87, 124 anaerobe aerotolerant, 126 obligate, 126 anaerobic, 447, 714 biofilters, 757 decay, 789, 874 contact process, 218, 722, 753 expanded bed reactor, 759 expanded bed UASB reactor, 756 fixed film reactor, 756 fluidized bed reactor, 759 hydrolysis, 722 treatment, 218 UASB reactor, 754 anaerobic digestion, 59, 750, 830 alkalinity, 726 ammonia, 727 mixing, 726 nutrient requirements, 727 pH, 726 phase separation, 740, 759 temperature, 725, 747 trace elements, 727 anaerobic stabilization pond, 792

Subject Index

anaerobic reactor, 250 oxidation-reduction potential, 49 anaerobic state, 49 anaerobic toxicity assay, 730 anaerobic treatment conventional process, 750 energy balance, 774 energy consumption, 767 kinetic coefficients, 745 mixing, 748, 752 Anammox, 219, 627 anode, 47, 472 anodic inhibitors, 443 Anopheles, 155 anoxic conditions, 48, 714 anoxic environment, 621 antimony, 168 A/O (anaerobic/oxic) process, 629 apatite, 173, 452 appurtenance headloss, 229 aquaculture, 221, 795 aragonite, 444 Archaea, 125 phylogeny, 121 argentometric method, 95 argyria, 169 aromatic compound, 84 Arrhenius, 19 theory of ionization, 58 Arrhenius equation, 161, 571, 640, 654, 668, 673, 866, 877 arsenic, 168 acid, 454 removal, 457 arsenous acid, 454 arsine, 454 ascarid, 132 Ascaris lumbricoides, 151, 154 Ascaris suum, 834 ascorbic acid dechlorination, 523 Asian clam, 552 assimilation, 87 atomic number, 3 atomic weight, 3 ATP. See adenosine triphosphate autocatalytic reaction, 17 for ozone, 536 autotroph, 110, 125, 617 yield factor, 617 auxiliary wash systems, 411 Avogadro number, 3 Avogadro’s law, 19

Bacillus, 124 backwash, 829 bed expansion, 403 biologically active filters, 646 backwashing filters, 398 porosity, 400 velocity, 403 bacteria filamentous, 124 iron oxidizing, 51 shapes, 124 phylogeny, 121 size, 124 sulfate reducing, 51 survival, 866 bacteriophage, 121 baffle headloss, 231 baffle wall, 371, 613 baffling, 228, 361, 376 Balantidium coli, 151 ballasted flocculation, 208, 219, 380 performance in wastewater treatment, 380 ballast waters, 553 Bardenpho process, 619, 632 barium, 168 bar graph, 439 bar racks, 211, 220, 275 base definition, 58 bass, 863 batch digestion, 665 batch process, 244, 568 BDST. See bed-depth-service time method Becquerel, 28 bed-depth-service time method, 490 Beer-Lambert law, 99, 540, 789, 801 Beer’s law, 99 Beggiatoa, 713 benthos, 789, 798 benzene, 84 bentonite, 354 beryllium, 168 berl saddles, 218 Bernoulli’s equation, 226, 275, 283, 364, 365, 416, 838 BET isotherm, 476 beta decay, 27 beta-Poisson model, 179 beta radiation, 175 bilharzia. See schistosomiasis biochemical methane potential, 729

939

940

Subject Index

biochemical oxygen demand, 105, 195, 571, 729, 791 chlorine effects on, 520 exertion in streams, 870 progression, 111 regression method, 112 relation to COD and TOC, 571 temperature effect, 108 ultimate, 106 binomial distribution, 138 biodisks, 217 biofilm, 51 biogas, 721, See also methane composition, 730 biological filtration, 421 biologically activated filter, 645 operating conditions, 646 biologically aerated filter. See biologically activated filter biological nutrient removal, 219, 222 biological treatment, 218, 351 biomass active, 570 composition, 573 synthesis, 591 bio-P process, 219, 628 operating conditions, 633 bisphenol A, 171 bluegreen algae, 212, 214, 790, 864, See also cyanobacteria BOD. See biochemical oxygen demand boiling point, 82 Boltzman equation, 19 bond covalent, 82 ionic, 82 bone char, 459 boron, 169 bottled water, 184 Boyle’s law, 19, 362 breakpoint chlorination, 515 breakthrough curve, 478, 491 brine use factor, 458 bromate, 169, 551 formation with ozone, 423 bromine, 169 bromoform, 549 Brönsted and Lowry acid-base definition, 58 Brownian motion, 332, 356 bubble rise velocity, 346 budding, 129 buffers, 59, 63

natural water, 73 bulk flow, 235 bulking, 639 filamentous, 712 nonfilamentous, 714 bypasses, 207 byproduct formation, 607 cadmium, 169 calcite, 54, 444 Caldwell-Lawrence, 446 Cambi process, 761 Camp–Shields equation, 282 Campylobacter, 148 CANON process, 628 capsid, 130 capsule, 123, 714 carbohydrate, 85, 86, 770 disaccharide, 85 monosaccharide, 85 polysaccharide, 86 carbon, 102 adsorption, 472 carbon dioxide, 81, 337 atmospheric content, 21 effects on softening, 433 carbonaceous biochemical oxygen demand, 109 carbonic acid, 73 carbonization, 473 carbon monoxide, 338 carbonyl group, 82 carboxyl group, 82 carcinogenicity determination, 180 carcinogens, 174 cardiovascular disease, 431 Carman–Kozeny equation, 379, 394, 396, 398, 404, 415, 464, 465, 843 carp, 795 cascade aerator, 340, 699 catalase, 529 catabolism, 87, 88, 124 catalyst, 18, 86 cathode, 47, 472 cathodic protection, 52 cattails, 791 caustic soda, 433 cavitation, 838 cell potential, 46 cell transformation assays, 181 cell wall, 122 cellulose, 86, 128 centrifugation, 211, 221, 830

Subject Index

centrifuge, 766, 840 Ceriodaphnia, 183 cesspool, 722 cestode, 151, 154 channel best hydraulic section, 367 velocity in treatment plant, 210, 220 characterization models, 896 charge balance, 48, 104 charge neutralization, 359 Charles’s law, 19 chelating agent, 96 chemically enhanced primary treatment, 219 chemical oxygen demand, 103, 571, 587, 666, 729 interferences, 105 chemical potential, 36, See also free energy chemolithotroph, 125 chemoorganoheterotroph, 129 chemoorganotrophs, 125 chemotroph, 125 Chezy equation, 312 Chick-Watson, 530 disinfection model, 510, 520 Chick-Watson law, 520 chitin, 128 chitosan, 354 chloramines, 189, 209, 447, 473, 513, 524, 525, 526, 549, 551 deoxygenation, 690 Chlorella, 790 chlorination, 874 chlorine, 169, 210, 660 biofilm sloughing, 649 bulking control, 715 iron and manganese removal, 448 photodegradation, 517 reaction with organic matter, 513 removal of iron and manganese, 447 chlorine demand curve, 514 chlorine dioxide, 524 iron and manganese removal, 448 chlorine residual, 195 corrosion, 442 chloroform, 549 chloro-organics, 210, 515 chlorophyll, 128 chloroplast, 128 cholera, 149 Citrobacter, 453 clarifiers depth, 318 rising sludge, 715

solids contact, 313, See also upflow solids contact clarifier wastewater treatment, 315 water treatment, 314, 708 chromium, 169 hexavalent, 189 chromosome, 123 Chrysophyte, 130 cilia, 131 ciliate, 129 circuit external, 44 internal, 44 citric acid cycle. See tricarboxylic acid cycle Cladocera, 154 clinophlolite, See clinoptilolite clinoptilolite, 458 Clostridium, 124 coagulant, 211, 298, 364, 829 aid, 356 recovery, 208, 219, 355 coagulation, 159, 209, 219, 337, 351, 387, 418, 434, 448 See also upflow solids contact clarifier cobalt, 169 deoxygenation, 690 Coccidia, 154 coccus, 124 COD. See chemical oxygen demand co-digestion, 762, 766 cogeneration, 219 coke, 436, 449 coliform freshwater decay rate, 162 seawater decay rate, 162 thermotolerant, 156 coliphage, 154, 530 colloid, 99, 351, 567, 586 colmatage, 466 color, 102, 431, 449 colorimetric analysis, 99 column settling test, 701 combined chlorine doses wastewater teatment, 520 combined chlorine residual, 513 combined sewer overflow, 519, 527, 530, 878 comminutors, 219 common ion effect, 63 comparator, 102 competitive adsorption, 490 competitive parallel reaction, 17 complete mixed, 235, 242, 865

941

942

Subject Index

complete mixed reactor, 574, 577 in series, 242 completely stirred tank reactor. See complete mixed reactor complexes, 25, 63, 95, 450 concentration, 4 gas, 39 ppm, 4 ppt, 4 concentration polarization, 469 condensed phosphate. See polyphosphate conductance, 14, 442 conjugate acid and base, 61 conservative substance, 866 constant rate filtration, 418 contact stabilization, 614 Contaminant Candidate List, 550 convection, 235 coordination compound, 25 copepod, 154 copper, 169, 170 catalysis of Mn oxidation, 447 standards for drinking water, 191 copper sulfate biocide, 552 coral reef, 864 Corbicula fluminea. See Asiastic clam corona discharge, 533 corrosion, 43, 49, 170, 446, 511, 732 chemical factors, 442 environmental conditions, 52 microbial, 51 prevention, 442 corrosiveness, 337 crappie, 863 critical deficit, 872 critical flow, 227 crop nutrient uptake, 807, 851 yield, 851 crossflow media, 757 couple potential. See cell potential coupon, 443 crustacean, 568 cryogenic process, 615 Cryptosporidium, 151, 158, 185, 187, 188, 420, 460, 518, 525, 834 inactivation by chlorine dioxide, 525 inactivation by ozone, 533 UV dose, 543 Ct concept, 511 Ctenopharyngodon idella, 795 Culex pipens, 155

curie, 28 cyanide, 169, 214 Cyanobacteria, 128, 171, 176, 864 cyanotoxin, 128 cyclone degritter, 294 Cyprinus carpio, 795 cyst, 150, 154 cysteine, 84 cytoplasm, 123, 127 cytoplasmic membrane, 123 Dalton, 447, 461 Dalton’s law, 20 DALY. See Disability-Adjusted Life Year dams reaeration, 878 Darcy’s law, 843 Darcy–Weisbach equation, 227, 320, 376, 394, 400, 411, 661, 828 Darcy–Weisbach friction factor, 282 DDT, 171 dead volume, 249, 695, 748 dechlorination, 59, 208, 209, 211, 473, 522 declining rate filtration, 418 defluoridation, 458, See fluoride removal degree-day product, 668 dehydrogenase, 87, 570 denaturing gradient gel electrophoresis, 142 dengue fever, 155 denitrification, 208, 422, 458, 621, 706, 714, 715, 798, 810, 811, 814 growth processes, 621 deoxyribonucleic acid, 123, 539, 570 dermotoxin, 128 desalination, 460 design basis unit process, 198 detergents, 174, 431 detritus tank, 292, 294 dewatering unit performance, 832 dextran blue, 250 diarrhea, 149 diatom, 130, 790 diatomaceous earth, 843 dichloramine, 513 Diptera, 154 differential media, 137 diffused aeration, 218, 648 factors affecting oxygen transfer, 694 diffused aerator, 362 diffuser, 344, 692 diffuser wall. See baffle wall diffusion, 236, 331, 356, 387, 470

Subject Index

dilution rate, 242 dinoflagellate, 130 diode array lamps, 541 dioxin, 191 direct current, 50 direct filtration, 418 direct plate count, 136 Disability-Adjusted Life Year, 147 disinfection, 59, 209, 213, 219, 509, 701 byproducts, 469, 520, 525, 527, 549, 551 kinetics, 510 dispersion, 235, 250, 866 coefficient, 250 index, 250 dissimilation, 87 dissociation constants, acid, 59 dissolved air flotation, 219, 314, 832 water treatment, 839 dissolved oxygen, 102, 870 corrosion, 442 factors affecting sag curve, 874 guidelines for Ontario surface waters, 876 impact of algae, 791 sag curve, 872 sag curve critical deficit, 876 temperature effects, 874 distillation, 217 distributed source, 865, 866 distribution box, 319 distributors trickling filter, 660 DNA. See deoxyribonucleic acid dosing siphon, 801 downflow fixed film reactor, 745, 758 dracunculiasis, 151 draft trickling filter, 656 Dreissena polymorpha. See zebra mussel drying beds, 221 dual distribution, 194 duckweed, 799 dysentery, 147 Eadie–Hoftsee plot, 91 EC50, 183 E. coli, 140 eddy size, 356 EDTA. See ethlyenediaminetetraacetic acid effective size, 391, 801 effective SRT in SBR, 612 egg-shaped digester, 750 Eichhornia crassipes. See water hyacinth

electrochemical cell, 43, 48 electrode, 43 electrodialysis, 211, 472 electrolytic cathodic protection, 53 electrolytic corrosion, 50 electromotive force, 443 series for metals table, 40 electron balance, 104 free energy, 37 electronegativity, 81, 83 electrophilic addition, 513 elements, 3 toxicity, 167 elementary reaction, 12, 17 elephantiasis, 155 Embden-Meyerhof-Parnas pathway. see glycolysis emerging contaminants, 171, 208 EMF. See electromotive force empty bed contact time, 423, 457, 479, 491 endocrine disruptors, 171 endogenous decay, 572, 592, 594, 607, 612, 633, 665, 740 coefficients for activated sludge, 588 endogenous VSS, 570 endoplasmic reticulum, 127 endospore. See spore energy potential, 777 enhanced coagulation, 351 enhanced phosphorus uptake, 628 Entamoeba coli, 154 Entamoeba histolytica, 151 Entamoebas, 130 Enterobacter aerogenes, 157 enterovirus, 149 enthalpy, 42 entropy, 35 enumerating microorganisms, 156 enumeration techniques microorganisms, 136 environmental impact assessment, 889 enzymatic defined substrate technology, 139 enzyme, 18, 86, 87, 89, 626 extracellular, 723 substrate complex, 89 epidemiology, 178 equalization, 213, 219, 254, 769, 811, 814 equilibrium, 12 equilibrium constant, 13, 37 temperature effect, 42 equilibrium constant water, 57 temperature effects, 57 equivalence pH, 67

943

944

Subject Index

equivalence point, 67 redox titration, 96 equivalent conductance, 14 equivalent weight, 6, 7, 14, 52, 61 Ergun equation, 396, 400, 401, 409 Escherichia coli, 148, 157, 420,514. See also E. coli ester, 83 estrogens, 171 ether, 83, 177 ethlyenediaminetetraacetic acid, 96 Euglenozoans, 129 Eukarya phylogeny, 121 eukaryote, 127, 567 eutrophication, 128, 174, 450, 791, 864 exchange capacity, 457 excimer lamp, 541 excreta, 195 extended aeration, 612 facultative organism, 126, 567 facultative pond, 792, 793 loading rate, 796 Faraday constant, 46, 52 fat, 85, 86, 770 fathead minnow, 184 fatty acid, 86 fecal coliform, 156, 673, 777 fecal streptococci, 673 fermentation, 87, 125, 722 primary sludge, 630 fermenter, 220 ferric chloride, 352 ferric sulfate, 352 ferrous sulfate, 352 Fick’s Law, 331, 479, 657, 770 filamentous microorganisms, 129, 712, 713, 714 filamentous sulfur bacteria, 713 filariasis, 155 filters air binding, 415 algae, 415 backwash, 307, 389, 398 collection launders, 400 conduit design, 411 design for wastewater treatment, 412 design for water treatment, 412 flies. See Psychoda flies granular activated carbon, 417 gravel support layers, 400, 409 headloss, 394 hydraulic radius, 395 iron and manganese removal, 450

media characteristics, 391 minimum fluidization, 402 pressure, 449 removal mechanisms, 387 roughing, 351 for stabilization ponds, 412 terminal headloss, 397 underdrains, 409 wastewater, 397 for wastewater treatment, 414 for water treatment, 413 weir location, 416 filtration, 208, 209, 213, 220, 337, 387, 448, 460, 599 hydraulic shear, 389 in-line, 353 iron and manganese removal, 447 fimbriae, 123 first-order reaction, 15 fish chloramine toxicity, 518 temperature tolerance, 863 fixed film process, 218, 637 activated sludge, 639 anaerobic, 757 fixed-film-suspended growth systems, 639 flagella, 128 flagellum, 123 flat sheet membrane, 637 flatworm, 131, 150 flavin adenine dinucleotide, 87 flexirings, 757 flies, 155 black, 155 Mangrove, 155 Tsetse, 155 flocculation, 159, 210, 211, 220, 298, 313, 336, 337, 390, 434, 448, 525, 607 of activated sludge, 704 tapered, 369 flocculator, 210, 368 Alabama, 377 baffled channel, 376 paddle, 369 pebble bed, 368, 379 pipe, 376 propeller, 372 spiral flow, 378 vertical flow, 376 vertical-shaft turbine, 375 flocculators comparison, 369

Subject Index

flooding basin technique, 805 flotation, 754, 769. See also dissolved air flotation thickening, 766 fluidization velocities, 405 temperature effect on, 406 fluidized bed, 422 fluke, 132, 150 fluorescein, 250 fluorescence in-situ hybridization, 141 fluoridation, 59, 173, 210, 458 fluoride, 169, 173 removal, 210, 457, 459 fluorosilicic acid, 458 fluorosis, 173 fluorspar, 458 fly, Psychoda, 660 F:M, 694, 715. See also food:microorgamisnm ratio foam, 318, 712 food:microorganism ratio, 594 selectors, 714 food poisoning, 149 forest crops, 807 free energy, 35, 88 formation, table, 38 standard state, 37 Freundlich isotherm, 476 Froude number, 227, 312 grit chamber, 285 hydraulic jump, 365 fruiting bodies, 129 fulvic acids, 102. 422 functional groups, 82, 456 fungi, 129, 568 nitrogen and phosphorus requirement, 129 GAC. See activated carbon, granular Galileo number, 401 galvanic cathodic protection, 52 galvanic corrosion, 50 gamma radiation, 27, 175 gas equation of state, 333 gases specific heats, 672 gas transfer, 332 gastroenteritis. See dysentery gas vesicles, 123 generation time, 132, 133 genome, 130 genomics microbial, 141 geometric mean, 140 geosmin, 128, 176, 522

Giardia lamblia, 151, 158, 188, 213, 213, 420, 460, 509, 518, 525 inactivation by disinfectants, 511 UV dose, 543 glauconite. See greensand GLUMRB, 691 glutamic acid, 84 glycogen, 123, 628 accumulating organisms, 628 glycolysis, 87, 126 Golgi body, 128 Gordona, 713 gram atomic weight, 3 gram molecular weight, 3 granular activated carbon, 417, 526, 757 biological filters, 422 granular activated sludge, 634 design and performance, 635 graphite, 53 gravel porosity, 379, 409 support layers in filters, 409 gravity aerators, 339 gravity flux, 703 greenhouse gas production, 721 greensand, 448 greywater, 196 grit composition, 281 definition, 281 quantities, 281 removal, 769 grit chamber, 220, 281 aerated, 290 horizontal flow, 282 inlet, 319 parabolic, 284 square tank, 292 grit washing, 294 growth factor, 125 groundwater, 213, 336, 447, 459, 810, 815, 851 guanosine triphosphate, 89 Guinea worm. See dracunculiasis Gymnamoebas, 130 Haemagogus, 155 Hagen-Poisuelle law, 305, 463, 844 half-life, 27 isotopes, 28 half-reaction, 10 table, 8 haloacetic acid, 549, 551 haloacetonitrile, 551

945

946

Subject Index

halo-obligatory, 127 halophile, 127 halotolerant, 127 hardness, 431. See also softening carbonate, 431 noncarbonate, 432 ranges, 433 relations, 432 heat exchangers, 770 heat transfer, 799 coefficients, 672, 800 heavy metals, 223, 569 removal, 714 helminth, 131, 147, 150, 420 standards, 187 Henderson-Hasselbach equation, 64, 65, 69, 72 Henry’s constant, 20 Henry’s law, 20, 39, 62, 74, 333, 336, 512, 689, 722, 838, 870 temperature conversion factors, 22 Henry’s law constants for air, 838 table, 22 hepatitis, 149, 150 hepatotoxin, 128, 171, 214 heterotrophic organisms, 87, 125, 583, 590, 616, 619, 790 heterotrophic plate count, 136, 188 hexametaphosphate, 397 hindered settling, 703 histidine, 181 hollow fiber membrane, 637 hookworm, 132, 150 horizontal belt press, 833 humic acid, 102, 422, 456, 476, 540 hydraulic conductivity, 805 measurement, 805 hydraulic design of treatment works, 226 hydraulic jump, 363 hydraulic radius, 227 definition for filters, 395 hydride, 5 hydrocarbon, 82 hydrogen peroxide, 455 hydrogen sulfide, 85, 170, 176, 191, 337, 338, 442, 449, 725, 727, 764, 793 hydrogenophilic methanogens, 724 hydrolysis, 596, 665, 666, 722, 723, 735, 740, 741, 761 in anaerobic fermentation, 740 hydroxyl radical, 455, 537, 548, 551

hyphae, 129 hypochorous acid decay, 517 Hypophthalmichthys molitrix, 795 ideal gas law, 20 ilmenite, 417 Imhoff tank, 726 imine, 353 impellers, 360 mixer, 359 types, 360 impoundment, 217 incineration, 220, 829 incinerator, 221 indicator acid-base, 71 acid–base, table, 72 redox, 98 indicator microorganism, 209, 672, 673, 801, 866, 868 traits, 155 industrial waste, 196 inert organics, 577 inert solids, 578, 667, 744 infective dose, 162 infiltration capacity, 805 infiltrometer, 805 inlet arrangements sedimentation tanks, 322 integrated fixed film activated sludge, 640 integrating factor, 872 interstitial velocity, 404, 840 intrinsic resistance, 844 invasive species, 522, 551 inventory analysis, 889 iodometric method, 516 ion exchange, 210, 217, 456 softening, 456 ionic strength, 14, 73, 445 inert solids, 101 irrigation, 223 system. See slow rate system iron, 170, 337, 447 oxidation, 339 salts, 351 iron and manganese biological removal, 449 removal, 422, 447, 525 isomer, 81, 82 isotherm, definition, 474 isotope, 3, 27 IUPAC, 3

Subject Index

jar test, 355, 451, 452 jar testing apparatus, 355 ketone, 83, 84 Kjeldahl nitrogen, 616 Klebsiella pneumoniae, 156 Kolmogorov length microscale, 356 Krebs cycle. See tricarboxylic acid cycle lagoon, 220, 789, See also aerated lagoon storage, 831 Lambert’s law, 99, 544 lamella clarifier, 211, 303, 712, 754 land application of sludge, 847 landfill, 211, 220, 221, 223, 829 Langelier Index, 445 Langmuir adsorption isotherm, 475 land treatment overland flow systems, 805 rapid infiltration systems, 804 slow rate systems, 804 of wastewater, 804 latent heat of water, 671 launder, 309 law of mass action, 9, 39 LC50, 183, 749 leachate, 758 lead, 51, 169, 179 and copper rule, 443 Legionella, 148, 149, 185, 188, 509 Leptospira, 148 Leucothrix, 713 lichen, 128 ligand, 25, 96, 450, 664 lime for coagulation, 351 recovery, 441 soda softening, 397 limiting flux, 706 Lineweaver-Burk plot, 91, 589, 602 lipid, 86 lithium, 169 as a tracer, 250 lithotroph, 125 loiasis, 155 long-term acceptance rate, 810 Louse, 155 Ludzack-Ettinger process, 619 macrophyte, 791 magnesium in biological phosphorus uptake, 629

magnetite, 123 magnetosomes, 123 malaria, 153, 155 manganese, 337, 447 oxidation, 339 manifold design, 319 Manning’s equation, 227, 312, 366 Mansonia, 155 marsh gas, 337, 721 mass balance, 63, 235 system, 256 mass transfer, 331 mass transfer coefficient, 536 calculation, 335 maturation pond, 162, 795 maximum contaminant level, 185, 518 MCL. See maximum contaminant level mean cell residence time. See solids retention time media anaerobic downflow reactors, 758 anaerobic fluidized bed reactors, 759 anaerobic upflow filter, 757 specific surface, 653 surface area, 651 void volume, 651 melting point, 82 membrane, 44, 48 anaerobic treatment, 760 backwashing, 466, 637 bioreactors, 635 filter analyses, 137 flat sheet, 637 treatment, 210 membrane processes, 212, 460, See also reverse osmosis operating pressures, 466 membrane solids separation unit, 711 design, 712 membrane technologies for water treatment, 468 mercaptan, 85, 177 mercury, 168, 169, 214 based UV lamps, 540 mesh size. See sieve size openings mesotrophic, 864 metabolism, 86, 568 metals irrigation, 807 removal, 353, 453, 461 removal in activated sludge, 664 methane, 337, 721 dissolved, 731 production from acetate, 724

947

948

Subject Index

methanogenesis, 724 methanogens, 792 methanol, 625 Methanosaeta, 724 Methanosarcina, 724 Methanothrix, 724 Methanothrix sohengenii, 724 methemoglobinemia, 173 methylisoborneol, 128, 522 methyl mercury, 168, 169 MIB. See methylisoborneol micelle, 100 Michaelis–Menten equation, 90, 134, 236, 475 microaerophilic organism, 126 microbiological quality drinking water, 187 microconstituent. See emerging contaminant Microthrix parvicella, 713 microcystin, 171 Microcystis, 171, 473 microfiltration, 212, 221, 460 microorganism ATP content, 87 die-off, 159 growth cycle, 133 growth rate, 573 indicator, 153 model, 154 test, 154 micropollutants. See emerging contaminants microscopic reversibility, 12, 89, 474 microstrainer, 277, 801 design parameters, 279 Microtox® , 183 minimum fluidization velocity, 402 mite, 154 mitochondrion, 127 mitosis, 128 mixed culture growth, 135 mixed liquor, 568 mixers, 208, 219, 359 hydraulic, 363 pneumatic, 362 mixing, 689, 866 aeration basin, 568 indexes, 249 power input correlations, 359 power requirement, 695 modified fouling index, 466, 468 molal concentration, 4 molar concentration, 4 mold, 129 mole fraction, 4, 39

molecular microbiology tools, 141 molecular weight cutoff, 447, 461 molybdenum, 169 mosquitoes, 155 monochloramine, 189, 513 Monod equation, 236, 569, 583, 590, 735 monosaccharide, 86, 88 Moody diagram, 227, 360, 376, 657 most probable number, 137, 156 moving bed bioreactor media, 640 MPN. See most probable number multi-media filter, 417 municipal solid waste digestion, 762 mushroom, 129 mutagen, 181 MWCO. See moelcular weight cutoff Mycobacterium, 148, 713 NAD. See nicotinamide adenine dinucleotide NADH, 126 Nalgonda technique, 460 nanofiltration, 212, 221, 460, 469 nanoparticles, 172 NDMA. See N-nitrosodimethylamine negative head, 415 nematode, 132, 150, 154 nephelometry, 101 Nernst equation, 45, 96, 98 net driving pressure, 471 neurotoxin, 128, 171, 214, 864 neutron capture, 27 neutralization, 220 Newtonian fluid, 827 nickel, 169, 214 nicotinamide adenine dinucleotide, 87, 125 phosphate, 87 nitrate, 169, 173, 189, 214 removal in drinking water, 458 nitrification, 59, 109, 450, 520, 615, 629, 650, 658, 664, 669, 714, 770, 814, 873, 874 alkalinity, 619 autothermal aerobic digestion, 670 nitrifier, 576, 617, 619, 620, 621, 642, 668, 670 nitrilotriacetic acid, 190 nitritation, 617 nitrite, 169 oxidizing bacteria, 616 Nitrobacter, 109, 626 nitrogen atmospheric content, 21 nitrogenous biochemical oxygen demand, 110

Subject Index

nitrogenous oxygen demand, 873, 882 nitrogen trichloride, 514 Nitrosomonas, 109 N-nitrosodimethylamine, 525 noble metal, 443 Nocardia amarae, 713 NOEC. See no-observed-effect concentration nondegradable organic matter, 588 no-observed-effect concentration, 183 normality, 61, 62 Northwest Territories, 796, 797 nozzle for aerators, 344 discharge coefficients, 342 nuclear chemistry, 27 nuclide, 27 nutrients, 171 activated sludge process, 573 anaerobic treatment, 766 crops, 809 loadings for lakes, 864 in sludge, 851 observed yield, 573, 639 odor, 338, 793 index, 176 Ohm’s law, 14 oils, 86 OLAND process, 628 oleic acid, 86 oligotrophic, 864 oocyst, 150 Oomycetes, 130 organic acids, 724 organotroph, 125 ORP. See oxidation-reduction potential orthokinetic motion, 356 orthophosphate, 442 equilibrium reactions, 450 osmophilic, 127 osmotic pressure, 127, 461 overland flow land treatment, 815 oxidation ditch, 220, 695 oxidation number, 5, 104, 113 oxidation pond, 789 oxidation-reduction, 6, 10, 87 oxidation-reduction potential, 48, 125, 126, 450 in anaerobic reactors, 725 effects on gas transfer, 336 oxidizing agent, 10 oxygen atmospheric content, 21 hazards, 338

oxygen deficit, 871 oxygen demand, theoretical, 102 oxygen transfer efficiency, 695 oxygen uptake rate, 601, 607 ozone, 210, 423, 531, 532, 551 coagulant aid, 354 generator, 533 p notation, 57 PAC. See activated carbon, powdered parabolic grit chamber design notes, 284 parasites, 154 partial pressure, 20, 36 paratyphoid fever, 148 Pareto analysis, 898 Parshall flume, 283, 364 path function, 35 pathogens, 147 bacteriological, 148 helminth, 152 protozoa, 150 virus, 149 pathogen reduction anaerobic process, 777 peaking factors, 194 water distribution, 194 pentachlorophenol, 85, 190 peracetic acid, 110, 510, 527 properties, 527 peracid, 527 perfluorinated compounds, 172 performic acid, 510, 531 perikinetic motion, 356 periodic table, 3, 7 permanent hardness, 432 permanganate greensand regeneration, 448 removal of iron and manganese, 447 permeability, 810, See also hydraulic conductivity peroxyacid. See peracid personal care products, 172 pesticides, 171 pH anaerobic treatment, 724 definition, 11, 57 effects on coagulation-flocculation, 353 effects on gas transfer, 336 for Mg removal in softening, 434 stabilization ponds, 798 standards for drinking water, 191 variation in titration, 67 phage, 131

949

950

Subject Index

phagocytosis, 129 pharmaceuticals, 171 phenol, 84, 169, 176 phosphate accumulating organisms, 628 precipitates, 451 removal, 457 phosphine, 727 phosphorus, 766 anaerobic digestion effluent, 766 precipitation, 450 phosphorylation, 88 Phostrip process, 632 Photobacterium phosphoreum, 183 photolyase, 539 photolysis, 455 photoreactivation, 539 photosynthesis, 128, 789, 791 photosynthetically active radiation, 789 phototroph, 125 phycobiliprotein, 128 physical-chemical treatment, 220, 222, 223, 351, 828 phylogeny, 121 phytotoxicity, 169 pickle liquor, 450 pili, 123 pinpoint floc, 594 pipe velocity for wastewater, 220 velocity for water, 210 pitch, 360 plain sedimentation, 211, 420 plastic fluid, 827 platinum, 45 plug flow, 235, 237, 865, 870 analysis, 247 plug flow reactor, 574 in series, 247 plumbing, 170 polarity, 81 point-of-entry device, 217 point-of-use device, 217 point source, 865, 866 Poisson distribution, 138, 179 polarization, 51, 469 pollution indices, 859 polyacrylamide, 353 polyelectrolytes, 353, 839, 840 natural, 354 poly-β-hydroxyalkanoate, 123 poly-ß-hydroxybutyric acid, 123 polyhydroxyalkanoates, 628

polymerase chain reaction, 136, 141 polychlorinated biphenyls, 171 polymers, 353 polyphosphate, 123, 174, 436, 442, 449, 450, 628 pond. See also stabilization pond heat balance, 799 indicator microorganism removal, 801 nitrogen removal, 798 porosity, 389, 480 positron, 27 post denitrification, 622 potassium in biological phosphorus uptake, 629 potassium chloroplatinate, 102 potentiometer, 44 potentiometric analysis, 98 powdered activated carbon, 615 power number, 359, 360 power plant, 862 prechlorination, 210, 549 precoat filter, 209, 843 preliminary treatment, 387 pre-oxidation, 210 presence/absence rule, 187 pressure filter, 209, 419, 766, 830, 846 pressure swing adsorption, 615 primary clarifier. See sedimentation, type I and type II primary sedimentation, 768 primary sludgea composition, 848 degradability, 668 primary treatment, 222 primers, 141 process loading factor. See food:microrganism ratio prokaryote, 121, 132 propeller, 360 protective coating, 442, 443 protein, 85, 86, 727, 770 transport, 86 protist, 129 protozoa, 567 pathogenic, 187 protozoan, 108, 209 Pseudomonas denitrificans, 621 pseudo-plastic fluid, 828 Psychoda flies, 660 psychrophilic, 725 pure culture growth, 132 pyrolusite, 448 pyrolysis, 829 pyruvate, 89

Subject Index

QUAL2E, 873 quaternary ammonium salt, 83, 353 quicklime, 433 rad, 175 radiation, 175 cosmic, 27 sources, 175 radical, 5, 455 radioactivity, 27 characteristics of isotopes, 28 radium-226, 189, 214 removal, 449 radon, 175, 176, 215 rainbow trout, 184 Raleigh’s law, 101 rapid infiltration loading cycles, 814 rapid sand filters, 209, 388 rapid small-scale column test, 495 raschig rings, 757 rats, 155 reaction order, 15 temperature, 19 reaeration, 870 coefficient, 877 recalcination, 208 recarbonation, 210, 436 split treatment, 435 recycle flow effect on liquid retention time, 251 redox. 37, 62 See also oxidation–reduction indicator for titration, 98 potentials, 43 reaction, 7, 10, 41 titration, 96 reducing agent, 10 reference cell, 45 reference electrode, 48 relapsing fever, 155 relative error criterion, 603 rem, 175 reovirus, 150 resazurin, 114, 730 reservoirs, 213 resin, 216, 456 exchange capacity, 457 fouling, 457 residence time, 252 resistance, 14 in vacuum filtration, 844 resonance, 84

respiration, 87, 125 respirometric methods, 601 response-fluence curve, 542 resuspension, 307 retardant reaction, 17 reverse osmosis, 210, 217, 331, 460, 469 Reynolds number, 394 backwash velocity, 403 filters, 396 impeller, 359 mixing devices, 359 particle, 280 rapid small-scale column test, 495 tube, lamella clarifiers, 306 retardant, 666 retardant reaction, 570 Rhodamine B, 250 Rhodococcus, 713 ribosomal ribonucleic acid, 121 ribosome, 123 Richardson-Zaki equation, 403 riffle plate, 339 risk assessment, 179, 889 risk factor, 174 river blindness, 155 rms velocity gradient, 358 roentgen, 175 rotating biological contact sludge production, 649 rotifer, 568 rotavirus, 150 rotifer, 131 roundworms, 150 roughing filter, 209, 387, 420 RSSCT. See rapid small-scale column test runoff agricultural, 171 saccharophylic, 127 sacrificial anode, 52 Safe Drinking Water Act, 185, 186 sag curve. See dissolved oxygen sag curve salmon, 863 Salmonella, 149, 850 Salmonella paratyphi, 148 Salmonella typhi, 148 Salmonella typhimurium, 181 salt definition, 58 density index, 466 passage, 472 sand filtration, 537 Sarotheroden mossambicus, 795

951

952

Subject Index

saturation index. See Langelier index scale formation, 431 Scenedesmus, 790 schistosomiasis, 151, 152 schmutzedecke, 420 Schultz-Germain formula, 649 Schultze–Hardy rule, 351 scour velocity grit chamber, 282 screenings, 276 screens, 211, 220, 275 water treatment, 276 scum, 318 sea urchin, 184 seaweed, 127 Secchi disk, 789 second-order reaction, 15 secondary clarifier. See sedimentation, type II secondary treatment, 221, 222 sedimentation, 159, 211, 221, 237, 275, 278, 337, 387, 390, 434, 448, 525, 726, 829 type I, 294, 297 type II, 297 type III, 568, 700 type IV, 836 sedimentation basin. See also clarifiers inlet hydraulics, 319 inlets, 307, 319 selective media, 137 selectivity quotient, 456 selector reactor, 713 selenium, 169, 214 removal, 457 Selnastrum, 184 semi-permeable membrane, 44, 460 sensitivity analysis, 898 septage, 834, 848, 853 septic tank, 722, 726 sequencing batch reactor, 574, 609, 754 sequestering agent, 96, 449, 470 settleable solids, 874 settling column, 298 settling velocity, 279 sewage land application, 223 SHARON process, 626 Shigella, 148 short circuiting, 237, 248, 251, 748 shroud, 360 sidestream membrane unit, 711 sievert, 175 sieve size openings, 391

silica removal, 457 silicate, 449 siloxanes, 732 silver, 103, 169, 217, 510 slaked lime, 209, 433 sleeping sickness, 155 slime layer, 124 slowly degraded substrate, 596 slow rate land treatment, 807 slow sand filter, 388, 419 slow sand filtration, 157, 209 sludge anaerobic digestion, 750 bulking, 712 chemical conditioning, 829 compactibility, 580, 593 compaction in settling, 701 concentration processes, 211, 221 concentrations in water treatment processes, 831 concentration in wastewater operations, 832 conditioning, 833 dewaterability, 452 dewatering, 776 disposal, 721 land application, 223 land application rates, 849 nitrogen content of digested sludge, 847 ocean disposal, 829 softening, 829 solubilization in aerobic digestion, 666 specific gravity, 825 specific resistance, 846 thickening wastewater sludge, 832 US quality classification, 849 viscosity, 827 sludge age. 727, See alsosolids retention time anaerobic sludge digestion, 752 sludge composition, 848 aerobic treatment, 573 anaerobic treatment, 727 sludge conditioning freeze-thawing, 834, 840 thermal, 834 sludge digestion, 221, 633 aerobic, 664 anaerobic, 722 ponds, 797 sludge production, 847 anaerobic treatment, 727 due to metal coagulants, 452 sludge thickeners, 830

Subject Index

sludge volume index, 580, 704 sludge worms, 568 smallpox virus, 130 S. niloticus, 795 soap, 100, 431 soda ash, 433 sodium, 170 adsorption ratio, 807 sodium aluminate, 352 sodium arsenite, 454 sodium fluoride, 458 sodium silicofluoride, 458 softening, 211, 431 lime-soda, 433 nanofiltration, 461 natural, 213 soil aquifer treatment. See rapid infiltration soil corrosivity, 53 sol, 100 silica, 354 solar disinfection, 547, 801 solar radiation, 159 solids categories, 99 colloidal, 100 crystalline state, 39 dissolved, 99 flux, 703 generation, 828 settleable, 100 volatile, 100 solids retention time, 580, 727 food:microorganism relation, 594 minimum, 584 in rotating biological contactors, 649 trickling filter, 656 solubility product, 23, 39, 93, 213, 432, 444 constants table, 23, 24 solute, 4 solvent, 4 sources of water pollution, 170 specific aeration demand, 637 specific conductance, 14 specific deposit, 844 specific growth rate, 572 specific ion electrode, 460 specific resistance sludge cake, 844 sphericity, 389 spirillum, 124 split recarbonation, 435 split treatment, 256, 434

spore, 124 spray aerator, 342 sprinkling infiltrometer, 806 stability constant, 26 stabilization pond, 128, 221, 278, 387, 789, See also pond Northwest Territories, 797 operation, 790 sludge accumulation, 791 stabilization of water, 441 standard enthalpies table, 38 standard free energy, 37 standard hydrogen electrode, 45 standard plate count. See heterotrophic plate count standard potential, 46 standards for Canadian drinking water, 186 drinking water, 178 environmental water quality, 184 hazardous organics, 190 inorganic substances, 189 radiological, 192 US for drinking water, 187 various water uses, 192 starch, 86 startup of anaerobic systems, 749 state-point analysis, 709 stator, 369 steady state condition, 235, 236 stearic acid, 86 stepwise formation constant, 26 table of, 25 stirred SVI, 581 stoichiometric number, 10, 37 Stoke’s range, 280 stormwater runoff, 275, 278 stramenopile, 130 Streeter-Phelps equation, 870 Stronglyoides, 151 strontium, 170 struvite, 221, 764 harvesting, 764 subcritical flow, 227 submerged membrane reactors operating data, 638 sulfate as laxative, 188 reducing bacteria, 725 sulfide, 132, 170, 177, 749 removal, 449, 522 toxicity in anaerobic treatment, 727

953

954

Subject Index

sulfite deoxygenation, 690 sulfur droplets, 123 super-chlorination, 211 supercritical flow, 227 superficial velocity, 396 surface aerator, 336, 693 Surface Water Treatment Rule, 521 suspended growth process, 218 suspended solids, 100 surface loading rate. 396, See also surface overflow rate surface overflow rate, 295, 303 surface wash system, 412 surface water quality, 863 Svedberg unit, 121 SVI. See sludge volume index sweep coagulation, 359 SWTR. See Surface Water Treatment Rule symbols definition activated sludge, 575 synthesis, 41, 87 synthetic organic chemical, 214 Syrian Hamster Embryo, 181 system potential, 48, 96, 97 t90, 159 Talipia, 795 tangential flow, 465 tapeworms, 151 tapered aeration, 607 Taq DNA polymerase, 142 Tardigrada, 154 taste and odor, 176, 191, 208, 211, 217, 337, 338, 423, 447, 511, 522, 524, 532, 550, 551, 552 control, 473, 522, 532 removal, 455 TCA cycle. see tricarboxylic acid cycle TDS. See total dissolved solids temperature fish survival, 862 reaction rate, 19 temperature effect aerobic digestion, 668 anaerobic sludge digestion, 751 anaerobic treatment, 725, 747 BOD, 108 CAS systems, 571 coagulation–flocculation, 353 equilibrium constant, 42 filter fluidization velocity, 406 filtration, 405, 415

on fish growth, 863 flocculation, 368 indicator dieoff in aerobic digestion, 673 indicator mortality, 161 trickling filters, 654 tensiometers, 806 teratogenicity, 181 thallium, 170 thermophilic, 725 anaerobic digestion, 747, 760 thermotolerant coliforms, 187 thickening, 211, 221, 701 dissolved air flotation, 837 gravity, 836, 837 options for wastewater treatment, 830 options for water treatment, 830 Thiotrix, 713 thiourea, 601 thymidine, 570 titration, 66, 68 complex formation, 96 resin capacity, 457 TOC. See total organic carbon total barrier oxidation ditch, 695 total coliform, 139, 187, 188, 673 group, 157 monitoring requirements, 186 total dissolved solids, 14, 99 relation to corrosion, 442 standard for drinking water, 191 total organic carbon, 102, 113, 571 total suspended solids, 100 toxicity, 182 assay, 183 determination, 182 of elements and compounds, 168 removal, 749 toxicity bioassays test organisms, 184 toxic wastes, 763 Toxocara, 154 tracer, 245, 250, 602, 607 dyes, 250 transmembrane pressure, 464, 637 transmittance, 99, 540 tray aerator, 341, 449 trematode, 132, 150 tricarboxylic acid cycle, 88, 126 Trichuris, 154 trickling filter, 217, 650, 665, 729, 769 air supply, 656 design information, 652 distribution system, 660

Subject Index

Eckenfelder model, 653 media, 651 odor problems, 660 Schultz-Germain formula, 653 sludge production, 656 Velz model, 652 trihalomethane, 191, 422, 473, 524, 549, 551 trout, 863 Tsukamurella, 713 tube clarifier, 211, 303, 754 tubercle, 51 tuberculosis, 148 tungsten, 170 turbidity, 101, 158, 449 standards for drinking water, 191 turbine, 360 turbulence, 874 two-film theory, 332 typhoid fever, 148, 149 typhus, 155 ultrafiltration, 212, 221, 460 iron and manganese removal, 447 ultraviolet radiation, 209, 510, 536 uncertainty analysis, 898 underdrain, 398, 410 underdrain blocks trickling filter, 658 underflow flux, 705 uniformity coefficient, 391 University of Cape Town process, 619 unsaturated compound, 83 upflow anaerobic sludge blanket process, 218 upflow sedimentation tanks, 297 upflow solids contact clarifier, 314, 377 uranium, 170 urea, 195 urine, 151 production, 195 UV radiation. See ultraviolet radiation vacuole, 128 vacuum dewatering, 211, 221 vacuum degasification, 754 vacuum filter, 830, 843 vacuum filtration, 843 valence, 61, 82 vanadium, 170 van’t Hoff equation, 21, 42 vectors of disease, 153 velocity, 210, 220, 660 minimum velocity backwashing, 403 velocity gradient, 358

anaerobic reactors, 752 diffused aeration, 696 Venturi section, 363 veterite, 444 vibrio, 124 Vibrio cholerae, 149, 553, 802 vicinal water, 840 virophage, 130 virus, 121, 130, 149, 567 common cold, 150 inactivation by disinfectants, 511 UV dose, 543 viscosity, Newton’s law, 356 vitamin, 125 volatile acids, 763 volatile fatty acids, 220, 628, 723 volatile organics, 214, 338 chemicals, 208 volatile solids definition, 100 volatile suspended solids anaerobic treatment, 738 volutin, 123 vortex grit removal, 293 waste activated sludge, 837 wastewater composition, 196 effluent guidelines, 195 flows, 197 organics variation, 572 protein, carbohydrate, lipid content, 196 reclamation, 194 wastewater treatment effectiveness, 222 wastewater treatment processes, 217 contaminant removal, 224 water dissociation, 62 properties, 5 water activity, 129 water consumption, 192 northern Canadian communities, 194 water demand commercial and institutional, 193 water hyacinth, 798 water quality Canadian guidelines, 186 index, 859 water stabilization, 212 waterborne disease, 185 water treatment, home, 217

955

956

Subject Index

water treatment plant flow charts, 212 water treatment processes, 207 effectiveness, 214 weir, 228, 309, 416 loading rates, 309 for mixing, 363 suppressed, 228 V-notch, 309 wet-air oxidation, 829 wetlands, 223, 804 Winkler method, 690 worms, 131 X-rays, 175 Years of Life Lost, 147 yeast, 129, 132, 568 yellow fever, 153 Yersinia, 148

yersiniosis, 148 yield observed (net), 573 yield factor, 133, 134, 572 activated sludge, 588 anaerobic treatment, 746 trickling filter, 656 zebra mussel, 176, 552 zeolite, 210, 453, 458 zero-order reaction, 15 zinc, 17 zone of mixing, 803 zone of oxygen dispersion, 803 zone sedimentation, 701, See sedimentation type III zone settling. See sedimentation type III zoogleal floc, 613 zoogleal growth, 388

ATOMATIC

05/11/2018

20:13:8

Page 1

Atomic Weights of Elements Element

Symbol Atomic Number

Atomic Weight

Element

Symbol Atomic Number

Atomic Weight

Actinum Aluminum Americium Antimony Argon Arsenic Astatine Barium Berkelium Beryllium Bismuth Boron Bromine Cadmium Calcium Californium Carbon Cerium Cesium Chlorine Chromium Cobalt Copper Curium Dysprosium Einsteinium Erbium Europium Fermium Fluorine Francium Gadolinium Gallium Germanium Gold Hafnium Helium Holmium Hydrogen Indium Iodine Iridium Iron Krypton Lanthanum Lawrencium Lead Lithium Lutetium Magnesium Manganese Mendelevium

Ac Al Am Sb Ar As At Ba Bk Be Bi B Br Cd Ca Cf C Ce Cs Cl Cr Co Cu Cm Dy Es Er Eu Fm F Fr Gd Ga Ge Au Hf He Ho H In I Ir Fe Kr La Lr Pb Li Lu Mg Mn Md

[227] 26.98 [243] 121.8 39.95 74.92 [210] 137.3 [247] 9.012 209.0 10.80 79.90 112.4 40.08 [251] 12.00 140.1 132.9 35.45 52.00 58.93 63.55 [247] 162.5 [252] 167.3 152.0 [257] 19.00 [223] 157.3 69.72 72.63 197.0 178.5 4.003 164.9 1.007 114.8 126.9 192.2 55.85 83.80 138.9 [262] 207.2 6.938 175.0 24.30 54.94 [258]

Mercury Molybdenum Neodymium Neon Neptunium Nickel Niobium Nitrogen Nobelium Osmium Oxygen Palladium Phosphorus Platinum Plutonium Polonium Potassium Praseodynium Promethium Protactinium Radium Radon Rhenium Rhodium Rubidium Ruthenium Samanium Scandium Selenium Silicon Silver Sodium Strontium Sulfur Tantalum Technetium Tellurium Terbium Thallium Thorium Thulium Tin Titanium Tungsten Uranium Vanadium Xenon Ytterbium Yttrium Zinc Zirconium

Hg Mo Nd Ne Np Ni Nb N No Os O Pd P Pt Pu Po K Pr Pm Pa Ra Rn Re Rh Rb Ru Sm Sc Se Si Ag Na Sr S Ta Tc Te Tb Tl Th Tm Sn Ti W U V Xe Yb Y Zn Zr

200.6 95.95 144.2 20.18 [237] 58.69 92.91 14.00 [259] 190.2 15.99 106.4 30.97 195.1 [244] [209] 39.10 140.9 [145] 231.0 [226] [222] 186.2 102.9 85.47 101.1 150.4 44.96 78.97 28.08 107.9 22.99 87.62 32.05 180.9 [97] 127.6 158.9 204.3 232.0 168.9 118.7 47.87 183.8 238.0 50.94 131.29 173.0 88.91 65.38 91.22

89 13 95 51 18 33 85 56 97 4 83 5 35 48 20 98 6 58 55 17 24 27 29 96 66 99 68 63 100 9 87 64 31 32 79 72 2 67 1 49 53 77 26 36 57 103 82 3 71 12 25 101

80 42 60 10 93 28 41 7 102 76 8 46 15 78 94 84 19 59 61 91 88 86 75 45 37 44 62 21 34 14 47 11 38 16 73 43 52 65 81 90 69 50 22 74 92 23 54 70 39 30 40

A value given in brackets denotes the mass number of a selected isotope. Adapted from International Union of Pure and Applied Chemistry (2016) and Pure and Applied Chemistry (1988), vol. 60, pp. 842–854.