Product, Process and Plant Design Using Subcritical and Supercritical Fluids for Industrial Application 3031346351, 9783031346354

Table of contents :
Preface
Contents
1 Introduction
1.1 Introduction
1.1.1 High Pressure and Nature
1.2 High Pressure Through Time
1.3 Why High Pressure for Products Design
1.4 Summary
References
2 What Are Supercritical Fluids?
2.1 Definition of Supercritical Fluids
2.2 Thermodynamic Properties of Supercritical Fluids
2.2.1 Phase Equilibria
2.3 Mass and Heat Transport Properties of Supercritical Fluids
2.3.1 Mass Transfer
2.3.2 Heat Transfer
2.4 Examples: Experimental Techniques for Determination of Thermodynamic Properties
2.4.1 Phase Equilibria—Static Method
2.4.2 Modified Capillary Method for Determination of Melting Points
2.4.3 External Balance Method for Determination of Diffusion Coefficients
2.4.4 Gravimetric Methods for Diffusion Coefficients, Density and Solubility of Gas by Magnetic Suspension Balance (MSB)
2.4.5 Viscosity Measurements in High-Pressure View Cell
2.4.6 Density of Measurements by High-Pressure Vibration Tube Densitometer
2.4.7 Capillary Rise Method for Determination of Interfacial Tension
2.4.8 Transport Properties from Drop Geometry in Dense Fluid
References
3 Industrial Scale Applications: Physical-Based Processes
3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid Materials
3.1.1 Thermodynamic Fundamentals of Extraction Processes
3.1.2 Solid-Supercritical Fluid Equilibrium
3.1.3 Liquid-Supercritical Fluid Equilibrium
3.1.4 Cycle Processes for Extraction Using Supercritical Fluids
3.1.5 Separation of Solute in Extraction Processes Using SCF
3.1.6 Basic Design Considerations of Extraction Plants for Solids
3.1.7 Supercritical Fluids Extraction at Ultra-High Pressure
3.1.8 Extraction of Solids Using SCF—Industrial Scale Units
3.1.9 Design of Extraction Plant for Liquids
3.1.10 Extraction of Liquids Using SCF-Industrial Scale Units
3.1.11 Conclusion
3.2 High Pressure Polymer Processing
3.2.1 Polymer Particles
3.2.2 Polymer Foaming with Subcritical or Supercritical Fluids
References
4 Industrial Scale Applications: Reaction-Based Processes
4.1 Chemical and Biochemical Reactions in SCFs
4.2 Chemical Reactions in SCFs
4.2.1 Polymerization and Depolymerization
4.2.2 Carbonylation
4.2.3 Oxidation
4.2.4 Hydrogenation
4.2.5 Hydroformylation
4.2.6 Hydrothermal Synthesis
4.2.7 Advantages on Using SCFs as Media for Chemical Reactions
4.3 Biochemical Reactions in SCFs
4.3.1 Influence of SCFs on Enzyme Activity and Stability
4.3.2 Enzyme-Catalysed Polymerization in SCFs
4.3.3 Reactors for Enzyme-Catalysed Processes Under High Pressure
4.3.4 Investigations to Perform Biochemical Reaction in High Pressure Batch Stirred Tank Rector—HP BSTR
4.4 Conclusions
4.4.1 Enzyme-Catalysed Synthesis of Biodiesel and Lignocellulosic Biomass Bioconversion in SCFs
4.4.2 Enzymatic Reactions in IL/SCFs Media
4.4.3 Future Trends
References
5 Design of High Pressure Plants for Research, Pilot and Production Scale
5.1 Basic Considerations for Effective Process Synthesis
5.2 Equipment for Bench-Scale Tests
5.3 Equipment for Pilot-Scale Tests
5.4 Equipment for Commercial Scale Plants
5.4.1 Supercritical Fluid Extraction Plants
5.4.2 Additional Features of High Pressure Plant Equipment
5.5 Turning Process Development into a Commercial Plant
5.5.1 Considerations for a Commercial Plant Project
5.5.2 Execution of a Commercial Plant Project
References
6 Safety and Control in High Pressure Plant Design and Operation
6.1 Safety Considerations for High Pressure Equipment and Plants
6.1.1 Diligence in Design and Fabrication
6.1.2 Diligence in Operation and Care for Longevity
6.1.3 Externally Induced Error or Defects
6.2 Control of High Pressure Plants
6.2.1 Functional Safety
6.2.2 Considerations for Engineering of Instrumentation and Control System
6.3 Engineering Ethics
References
7 Conclusion and Future Perspectives
7.1 Future Directions/Perspectives
7.1.1 Extraction
7.1.2 Micronisation
7.1.3 Impregnation with SC Fluids
7.1.4 Chemical and Biochemical Reactions
References

Citation preview

Željko Knez Christoph Lütge

Product, Process and Plant Design Using Subcritical and Supercritical Fluids for Industrial Application

Product, Process and Plant Design Using Subcritical and Supercritical Fluids for Industrial Application

Željko Knez · Christoph Lütge

Product, Process and Plant Design Using Subcritical and Supercritical Fluids for Industrial Application

Željko Knez Faculty of Chemistry and Chemical Engineering University of Maribor Maribor, Slovenia

Christoph Lütge Unna, Germany

ISBN 978-3-031-34635-4 ISBN 978-3-031-34636-1 (eBook) https://doi.org/10.1007/978-3-031-34636-1 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Preface

One of the most recent books in the field of subcritical fluid applications at high pressures and supercritical fluid applications was published in 2012 by Willey-VCH under the title Industrial High Pressure Applications. The editor was Prof. Rudolf Eggers. In the time to date, knowledge in the field of high-pressure applications has advanced considerably due to the need for new processes and products. This has led to the development of new high-pressure processes and, of course, to the development of techniques to determine the fundamental data needed for industrial development. Over the last decades, numerous publications have been published on the isolation and fractionation of various compounds such as plant extracts, essential oils and purified compounds. In addition, numerous publications have appeared on the processing of materials using sub- and supercritical fluids. In addition, many manuscripts on phase equilibrium research and modelling are available. However, publications on fundamental thermodynamics and mass transfer data necessary for the design of industrial plants are scarce. Further, there are virtually no manuscripts on the design of real industrial plants, the commissioning of industrial plants and their maintenance. Based on our many years of experience in both laboratory and pilot scale research, and the subsequent transfer of these technologies into industrial practice, i.e. the manufacture of equipment, its installation and subsequent commissioning and maintenance, we have decided to bring our knowledge and experience together in this book. This book provides missing information on the basic data needed for the design of an industrial plant and should give readers a detailed insight into the design, construction, commissioning, operation and maintenance of industrial plants for different processes. The book should be of interest to industry that operates, or wishes to operate, high-pressure plants. On the other hand, it might also be of interest to industry trying to find the advantages of using supercritical fluid technologies to solve problems of product fractionation, product formulation, synthesis in subcritical and supercritical fluids, etc. v

vi

Preface

The book finally aims at being a worthwhile resource for academics and researchers focused on product research using supercritical fluids, as well as for postgraduate and Ph.D. students who could use the proposed book as a textbook. Maribor, Slovenia Unna, Germany March 2023

Željko Knez Christoph Lütge

Contents

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1.1 High Pressure and Nature . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 High Pressure Through Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3 Why High Pressure for Products Design . . . . . . . . . . . . . . . . . . . . . . . 1.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 1 1 4 4 8 8

2 What Are Supercritical Fluids? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Definition of Supercritical Fluids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Thermodynamic Properties of Supercritical Fluids . . . . . . . . . . . . . . 2.2.1 Phase Equilibria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Mass and Heat Transport Properties of Supercritical Fluids . . . . . . . 2.3.1 Mass Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.2 Heat Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Examples: Experimental Techniques for Determination of Thermodynamic Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.1 Phase Equilibria—Static Method . . . . . . . . . . . . . . . . . . . . . . 2.4.2 Modified Capillary Method for Determination of Melting Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.3 External Balance Method for Determination of Diffusion Coefficients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.4 Gravimetric Methods for Diffusion Coefficients, Density and Solubility of Gas by Magnetic Suspension Balance (MSB) . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.5 Viscosity Measurements in High-Pressure View Cell . . . . . 2.4.6 Density of Measurements by High-Pressure Vibration Tube Densitometer . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.7 Capillary Rise Method for Determination of Interfacial Tension . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

11 11 14 14 20 20 25 26 26 27 30

31 35 38 39

vii

viii

Contents

2.4.8

Transport Properties from Drop Geometry in Dense Fluid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

42 46

3 Industrial Scale Applications: Physical-Based Processes . . . . . . . . . . . . 49 3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 3.1.1 Thermodynamic Fundamentals of Extraction Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 3.1.2 Solid-Supercritical Fluid Equilibrium . . . . . . . . . . . . . . . . . . 56 3.1.3 Liquid-Supercritical Fluid Equilibrium . . . . . . . . . . . . . . . . . 67 3.1.4 Cycle Processes for Extraction Using Supercritical Fluids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 3.1.5 Separation of Solute in Extraction Processes Using SCF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 3.1.6 Basic Design Considerations of Extraction Plants for Solids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 3.1.7 Supercritical Fluids Extraction at Ultra-High Pressure . . . . 77 3.1.8 Extraction of Solids Using SCF—Industrial Scale Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 3.1.9 Design of Extraction Plant for Liquids . . . . . . . . . . . . . . . . . 93 3.1.10 Extraction of Liquids Using SCF-Industrial Scale Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 3.1.11 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 3.2 High Pressure Polymer Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 3.2.1 Polymer Particles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 3.2.2 Polymer Foaming with Subcritical or Supercritical Fluids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141 4 Industrial Scale Applications: Reaction-Based Processes . . . . . . . . . . . 4.1 Chemical and Biochemical Reactions in SCFs . . . . . . . . . . . . . . . . . . 4.2 Chemical Reactions in SCFs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.1 Polymerization and Depolymerization . . . . . . . . . . . . . . . . . 4.2.2 Carbonylation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.3 Oxidation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.4 Hydrogenation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.5 Hydroformylation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.6 Hydrothermal Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.7 Advantages on Using SCFs as Media for Chemical Reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3 Biochemical Reactions in SCFs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.1 Influence of SCFs on Enzyme Activity and Stability . . . . . 4.3.2 Enzyme-Catalysed Polymerization in SCFs . . . . . . . . . . . . . 4.3.3 Reactors for Enzyme-Catalysed Processes Under High Pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

151 151 153 153 157 158 161 162 162 166 166 167 171 173

Contents

ix

4.3.4

Investigations to Perform Biochemical Reaction in High Pressure Batch Stirred Tank Rector—HP BSTR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4.1 Enzyme-Catalysed Synthesis of Biodiesel and Lignocellulosic Biomass Bioconversion in SCFs . . . . . 4.4.2 Enzymatic Reactions in IL/SCFs Media . . . . . . . . . . . . . . . . 4.4.3 Future Trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Design of High Pressure Plants for Research, Pilot and Production Scale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 Basic Considerations for Effective Process Synthesis . . . . . . . . . . . . 5.2 Equipment for Bench-Scale Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3 Equipment for Pilot-Scale Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4 Equipment for Commercial Scale Plants . . . . . . . . . . . . . . . . . . . . . . . 5.4.1 Supercritical Fluid Extraction Plants . . . . . . . . . . . . . . . . . . . 5.4.2 Additional Features of High Pressure Plant Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.5 Turning Process Development into a Commercial Plant . . . . . . . . . . 5.5.1 Considerations for a Commercial Plant Project . . . . . . . . . . 5.5.2 Execution of a Commercial Plant Project . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

175 177 178 179 180 181 193 193 196 200 207 208 214 217 217 218 221

6 Safety and Control in High Pressure Plant Design and Operation . . . 6.1 Safety Considerations for High Pressure Equipment and Plants . . . . 6.1.1 Diligence in Design and Fabrication . . . . . . . . . . . . . . . . . . . 6.1.2 Diligence in Operation and Care for Longevity . . . . . . . . . . 6.1.3 Externally Induced Error or Defects . . . . . . . . . . . . . . . . . . . 6.2 Control of High Pressure Plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2.1 Functional Safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2.2 Considerations for Engineering of Instrumentation and Control System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3 Engineering Ethics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

223 225 225 241 242 243 243

7 Conclusion and Future Perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1 Future Directions/Perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1.1 Extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1.2 Micronisation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1.3 Impregnation with SC Fluids . . . . . . . . . . . . . . . . . . . . . . . . . 7.1.4 Chemical and Biochemical Reactions . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

249 251 251 252 253 254 256

244 246 247

Chapter 1

Introduction

Abstract The definition of high pressure in nature as well as in industry is presented. An overview on the occurrence of high pressure in nature is provided. Discovery of critical point, a brief history of supercritical fluids, along with early developments on the use of high pressure and supercritical fluids are presented. The motivation for use of high pressure in industrial scale operations is described. Keywords High pressure · Critical point · Supercritical · Subcritical · Fluid · Application · Industrial

1.1 Introduction 1.1.1 High Pressure and Nature The term high pressure has different meanings in different contexts for human beings. High air pressure in weather prognosis describes pressures above 1 bar at sea level. High blood pressure is very dangerous when exceeds 0.019 bar. But, these pressures are far away from the values of pressure used and described for technical processes in this book. Terminology of high pressure of industrial processes is presented in Table 1.1. In technical applications 10,000 bar should be taken as the upper practical end. Any other pressure mentioned somewhere in the world is most likely a calculated value. Calibration of pressure transducers is only possible up to ca. 10,000 bar (except in military applications). Beyond, pressure ratings cannot be measured reliably as the transducers are not calibrated (HBM Company 2020). It should be noted that, the pressures employed in industrial applications, even though high, are rather low in comparison with the high pressures encountered in geosciences. As we could see from Fig. 1.1 in the center of the earth, we have pressures approaching 4000 kbar, at temperatures around 10,000 °C with the density of material being around 13 g/cm3 .

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Ž. Knez and C. Lütge, Product, Process and Plant Design Using Subcritical and Supercritical Fluids for Industrial Application, https://doi.org/10.1007/978-3-031-34636-1_1

1

2 Table 1.1 Terminology of high pressure of industrial processes

1 Introduction

Term

Pressure (bar) From

To

1

20

Low pressure

20

100

Medium Pressure

100

1000

High pressure

1000

10,000

Ultra high pressure

Fig. 1.1 Earth structure and physical properties (UHPT 2017)

At geological conditions, different types of minerals could be formed, such as silicate perovskite, which is thought to make up half of the Earth’s bulk, and post perovskite, which occurs at the core-mantle boundary. Silicate perovskite is either (Mg, Fe)SiO3 (magnesium, iron silicate) or CaSiO3 (calcium silicate) (Tomioka and Fujino 1997). In man living environments, pressures are much lower. In the Pacific Ocean at depth of 10,994 m (Mariana Trench) the pressure is over 1000 bar, where beside the microorganisms, amphipods and other animals were found. Toothed whales are able to dive 3000 m deep, where the pressure is round 300 bar. These whales have very sophisticated system to compensate for the pressure difference when they rise to the sea surface to get air (Quick et al. 2020). In the deep sea (below 2000 m depth), beside high pressure, there are also places with high temperatures up to several hundred degrees Celsius. These places are hydrothermal vents known as “black smokers”. These are large (several meters high) “chimneys” of sulfide and sulfate minerals which vent fluids up to 400 °C (ColínGarcía et al. 2016). The fluids appear like great black billowing clouds of smoke due to the precipitation of dissolved metals in the fluid. It is likely that at lower depths

1.1 Introduction

3

Fig. 1.2 Phase diagram of CO2

many of these vent sites reach supercritical conditions, but most cool sufficiently by the time they reach the sea floor to be subcritical. The definition of supercritical fluids (SCFs) is provided by the International Union of Pure and Applied Chemistry (IUPAC) as “The temperature and pressure at which the liquid and vapor intensive properties (density, heat capacity, etc.) become equal. It is the highest temperature (critical temperature) and pressure (critical pressure) at which both a gaseous and a liquid phase of a given compound can coexist.” (Gold 2019). In literature (Jessop and Leitner 1999) more precise definition is given: “SCF is defined as a state of a compound, mixture or element above its critical pressure (Pc ) and critical temperature (Tc ) but below the pressure required to condensate into solid” (Fig. 1.2). SCFs have gas like viscosity and diffusivity, liquid like density and dielectric constant and solvating properties which could be adjusted by operating pressure and/or temperature. All these properties make them suitable solvents for various applications. Chemical composition of atmosphere of planet Venus is 96.5% carbon dioxide and 3.5% nitrogen (Taylor 2014). At the surface of Venus, the pressure is 93 bar and the surface temperature is 462 °C. This is well above the critical points of both constituents of its atmosphere and therefore the atmosphere at the surface of Venus is a supercritical fluid. The interior atmospheres of the solar system’s gas giant planets are composed mainly of hydrogen and helium at temperatures well above their critical points. The gaseous outer atmospheres of Jupiter and Saturn transition smoothly into the dense liquid interior, while the nature of the transition zones of Neptune and Uranus are unknown.

4

1 Introduction

1.2 High Pressure Through Time Early applications of high pressure are dated back to the Roman times, when wooden pumps were used to supply water from 100 m deep fountains (Oleson 2015). Application of high pressure in “food sector” was a “digester” from Denis Papin in 1679 (Papin 1681). He made experiments in high pressure vessels to prove that the boiling point of water could be increased by elevation of pressure. The practical application was cooking meals for King Charles II at elevated pressures. Several hundred years later high pressure steam engines were invented by James Watt around 1785 (Dickinson 1939). In 1822 Baron Cagniard de la Tour published the first manuscript on the observation of the critical point (de La Tour 1822). The thermodynamic fundamentals on P–V–T behavior of substances were reported in 1869 by Thomas Andrews (Jessop and Leitner 1999). Hannay and Hogarth did several studies in 1879 and 1880 on the solubility of substances in supercritical fluids (Hannay and Hogarth 1879; Jessop and Leitner 1999). One of the most important inventions for the application of high pressure in industry was the synthesis of ammonia by Haber and Bosch. For this invention they were awarded the Nobel prize in 1918. This application of high pressure is an important milestone for several other processes which are employed in industry today. One of the most sophisticated large scale applications is the process for synthesis of low density polyethylene (LDPE) where reactor pressures are between 1200 bar and 3500 bar and temperatures around 300 °C. In high pressure pasteurization of food stuff, static pressures up to 8000 bars are used. Today high pressure is applied in several other industrial sectors. An overview of applications and motivation for the use of high pressure nowadays is presented in Table 1.2. The highest pressures are used for production of diamonds from graphite at pressures of 120,000 bar at 3000 °C.

1.3 Why High Pressure for Products Design Using supercritical fluids has numerous advantages, but the main drivers are environmental, health and safety, process and chemical benefits. Nowadays, there is a trend to develop alternative technologies with minimal environmental impact for products with special custom-designed properties. Reduced energy consumption, less toxic residues, efficient conversion of reactants to products, less by-products and higher quality and safety of final products are crucial requirements for the future processes. High pressure technologies are a relatively new tool to satisfy the mentioned demands. Technologies using high pressure processes resulted in completely new products with special characteristics (Aymonier et al. 2011; Reverchon 2002). The advantages applying high pressure processes are summarized in Table 1.3.

1.3 Why High Pressure for Products Design

5

Table 1.2 Examples of industrial high pressure processes Process

Pressure range (bar)

Motivation for high pressure

Hydrogenation

70–300

Gas solubility, reaction rate

Urea synthesis

140–250

Higher rate in liquid phase, reaction rate

Hydroformylation

150–300

Gas solubility, catalyst stability

Ammonia synthesis

150–350

Reaction equilibrium, volume reduction

Tobacco expansion (with N2 )

~700

Product properties

CO2 extraction

80–1000

Product properties (avoiding problematic solvent residues), selectivity

LDPE polymerization

1200–3500

Polymerization in homogenous phase, easy polymer separation by pressure decrease, achieving of special polymer properties

Isostatic pressing

1000–4000

Product properties

HP Pasteurization

3000–7000

Gentle food processing; improved product properties

Diamond synthesis

50,000–120,000

Promotion of the desired polymorph of carbon

Table 1.3 Advantages of using of high pressure in process industries – Influencing the reaction equilibrium

– In reactions with a mole number reduction, the equilibrium at high pressure is on the side of the reaction products (e.g. ammonia and methanol synthesis)

– Increasing the reaction – The reaction rate of gas reactions can be increased by pressure rate increase (via increase of the concentration) by a factor of 102 to 103 . Furthermore, the reaction rate constant will be influenced by the higher pressure – Influencing the selectivity

– Due to different influence of the pressure on the reaction rate of the reaction steps the selectivity can be altered

– Changing the solubility

– The solubility e.g. of gases in liquids and of solutes in supercritical fluids is increased at higher pressure. This effect is used for separation processes. Polymerization reactions can be performed at high pressure in homogeneous phase without additional solvent, the monomer acting as both reactant and as solvent

– Influencing the mass transfer rate

– In gas/liquid reactions the mass transfer rate is increased at higher pressure

– Volume reduction of gases

– The volume reduction of gases with higher pressure reduces the equipment dimensions

– Increasing the energy density

– By applying high pressure, the increased stored energy can be used for water jet cutting, for formulation of fine particles (micronization) and for destruction of cell membranes (tobacco and algae)

6

1 Introduction

Among different high pressure processes, sub-critical and supercritical fluids offer development of several excellent process technologies due to their tunable physicochemical properties. The processes with SCFs are sustainable, environmentally friendly, and in some applications also cost efficient, and what is the most important offer the possibility to obtain new products with special costumer-specific designed properties. One of their major advantages is the possibility of adjusting the thermophysical properties such as diffusivity, viscosity, density, dielectric constant by simply varying the operating pressure and/or temperature. Moreover, they have excellent heat transfer properties and could be used as heat transfer fluids with practically no impact to the environment compared to conventional ones like chlorofluorocarbon (CFC), hydrochlorofluorocarbon (HCFC) freons and for higher temperatures hydrocarbon fluids, synthetic hydrocarbon, silicone based fluids, molten salts and molten metals which are toxic or potent greenhouse gases (Kravanja 2018). In some parts of the world scCO2 is also often considered to be a greenhouse gas and the emissions through batch change is viewed as a concern. However, it should be noted that CO2 used in scCO2 processes is typically of natural origin (underground sources) and not separately produced. Supercritical fluids such as carbon dioxide are generally considered as “green solvents for the future” due to the ecological benefits they offer They are already applied in several processes which are operating on industrial scale in pharmaceutical, food, textile and chemical industries (Brunner 2010; Cansell et al. 2003; Pilz et al. 2006; Yoo and Fukuzato 2006). They can be used as processing solvents, reaction media or as reactants. A review of literature on the application of supercritical fluids shows that they are finding increasing use in the field of chemical and biochemical reactions (Knez et al. 2010), for synthesis of new materials and new catalyst supports such as aerogels (Zheng et al. 2010), for special separation techniques such as chromatography (Saito 2013), extraction processes (Knez et al. 2010), and for particle formation and product formulation (Jung et al. 2004). Supercritical fluids exhibit a pressuretunable dissolving power, they possess a liquid like density (and thus a high solvent strength), and their gas like transport properties allow facile extraction from solid natural materials to be accomplished. This distinctive combination of properties is preferably matched for developing processes for extracting, purifying, and recrystallizing fine chemicals and pharmaceuticals and producing new product forms that cannot be achieved by industry’s more conventional processing ways. Other industrial applications include dry cleaning, dyeing and impregnation (Weidner 2018), high pressure sterilization (Perrut 2012), jet cutting (Shen et al. 2011), thin-film deposition for microelectronics (Jianzhong et al. 2009), separations of value added products from fermentation broths in the field of biotechnology (Fabre et al. 1999), syntheses involving transition metals, where SCFs are used as solvents that can easily be removed from the solutes by mere expansion to ambient pressure (Sokolov et al. 2010). There is a great potential for applications of SCFs in the industrial processing of fats, oils and their derivatives (Stahl et al. 2013). Owning to their heat transfer properties, heat transfer studies for fluids in supercritical conditions are

1.3 Why High Pressure for Products Design

7

necessary for designing and optimizing trans-critical CO2 power cycles in refrigeration systems, nuclear reactors, air-conditioning, and advance heat pumps (Knez et al. 2014). Supercritical fluid based technologies offer important advantages over organic solvent technology, such as ecological friendliness and ease of product fractionation. Sub and scCO2 and SC water are the solvents of choice for the extraction processes (Knez et al. 2010, 2014; Knez 2018). Carbon dioxide is the most common supercritical solvent. It is used on large scale for the decaffeination of green coffee beans and tea, the extraction of hops for beer production and the production of essential oils and pharmaceutical products from plants. Water is at the moment the cheapest solvent and several substances are highly soluble in water. Therefore, more and more research on the use of subcritical or SC water for isolation and fractionation of substances is under investigation. Nowadays for some special applications, the trends shift towards the use of unconventional sub- or supercritical fluids such as SF6 , propane, and argon (Hrnˇciˇc et al. 2014, 2018; Ili´c et al. 2009). Propane is a very selective solvent for high pressure extractions (Knez et al. 1999). However, it is highly flammable. Argon is quite expensive, and SF6 is environmentally problematic. These are the main drawbacks in comparison with CO2 , but the benefits of these fluids are their complete non reactivity with natural substances and immiscibility with water. Propane is excellent solvent for fat and oils and the extraction processes could be performed already at relatively low pressure and what is important from economic point of view at very low solvent/feed ratios (Hrnˇciˇc et al. 2014, 2018; Ili´c et al. 2009; Knez et al. 1999). The main advantages of using supercritical fluids for isolation of natural products are solvent free products, no co-products and low temperature in separation process. In addition, the processes can be easily linked with direct micronisation and crystallization from scCO2 by fluid expansion. The most important advantage of the use of supercritical fluids is selective extraction of components or fractionation of total extracts. The limitation of further application of extracts obtained by high pressure technology is in the price of the product, which is in comparison with conventionally obtained products relatively high. The legal limitations of solvent residues and solvents (for products used in human applications) and isolation/fractionation of special components from total extracts in combination with different formulation (for example controlled release), chromatography and sterilization processes will increase the use of dense gases for extraction applications. In this book, we will present some technically relevant processes that have already been realized on an industrial scale. Some of them will be discussed in more detail in the continuation of the book, as they are more common in terms of the scope of working devices or according to the authors opinion, have great potential. In principle, all high-pressure processes can be divided into a physical based processes where equilibrium between solute and dense fluid are important, or processes where the dense media is used as reactant.

8

1 Introduction

1.4 Summary If we summarize, the supercritical fluids are solvents for the future, because—from environmental point of view—supercritical fluid-based technologies offer important advantages over organic solvent technologies. Due to high environmental concerns and due to increase of organic solvent prices, the demands on the use of new solvents increase. On the other hand, the demand on new products with special characteristics, high purity, and lower energy consumption due to low process temperatures for applications in all fields of human activity are increasing. By use of SCFs such products could be obtained. Therefore, we conclude that the advances in the field of high pressure technologies have opened up new pathways for substances and products obtained with cheap and environmentally friendly processes and will continue to do so in future.

References Aymonier C, Le Meur AC, Heroguez V (2011) Synthesis of nanocomposite particles using supercritical fluids: a bridge with bio-applications. Nanocomposite particles for bio-applications materials bio-interfaces. Pan Stanford Publishing Ltd., Singapore, pp 145–164 Bach E, Schollmeyer E (2007) Supercritical fluid textile dyeing technology. In: Environmental aspects of textile dyeing. Elsevier, pp 93–115 Brunner G (2010) Applications of supercritical fluids. Annu Rev Chem Biomol Eng 1(1):321–342 Cansell F, Aymonier C, Loppinet-Serani A (2003) Review on materials science and supercritical fluids. Curr Opin Solid State Mater Sci 7(4–5):331–340 Colín-García M, Heredia A, Cordero G, Camprubí A, Negrón-Mendoza A, Ortega-Gutiérrez F, Beraldi H, Ramos-Bernal S (2016) Hydrothermal vents and prebiotic chemistry: a review. Bol Soc Geológica Mex 68(3):599–620 de La Tour CC (1822) Exposé de quelques résultats obtenu par l’action combinée de la chaleur et de la compression sur certains liquides, tels que léau, l’alcool, l’ether sulfurique et l’essence de pétrole rectifiée. Ann Chim Phys 21(2):127 Dickinson HW (1939) A short history of the steam engine. Cambridge University Press Fabre CE, Condoret J-S, Marty A (1999) Extractive fermentation of aroma with supercritical CO2 . Biotechnol Bioeng 64(4):392–400 Gold V (ed) (2019) The IUPAC compendium of chemical terminology: the gold book, 4th edn. International Union of Pure and Applied Chemistry (IUPAC), Research Triangle Park, NC Hannay JB, Hogarth J (1879) On the solubility of solids in gases. Proc R Soc Lond 29:324 HBM Company (2020) Druck-Kalibrierung. In: Druck-Klibrierung. https://www.hbm.com/de/ 0151/druck-kalibrierung/ Hrnˇciˇc MK, Škerget M, Knez Ž (2014) Argon as a potential processing media for natural and synthetic substances. J Supercrit Fluids 95:252–257 Hrnˇciˇc MK, Cör D, Knez Ž (2018) Subcritical extraction of oil from black and white chia seeds with n-propane and comparison with conventional techniques. J Supercrit Fluids 140:182–187 Ili´c L, Škerget M, Hrnˇciˇc MK, Knez Ž (2009) Phase behavior of sunflower oil and soybean oil in propane and sulphur hexafluoride. J Supercrit Fluids 51(2):109–114 Jessop PG, Leitner W (1999) Supercritical fluids as media for chemical reactions. Chem Synth Using Supercrit Fluids. Wiley–VCH, Weinheim, pp 1–36 Jianzhong Y, Xianzhen Z, Qinqin X, Chuanjie Z, Aiqin W (2009) Supercritical fluids deposition techniques for the formation of nanocomposites. Prog Chem 21(4):606–614

References

9

Jung J-I, Bae JY, Bae B-S (2004) Preparation and characterization of structurally stable hexagonal and cubic mesoporous silica thin films. J Sol-Gel Sci Technol 31(1):179–183 Knez Ž (2018) Enzymatic reactions in subcritical and supercritical fluids. J Supercrit Fluids 134:133–140 Knez Ž, Hadolin M, Debrunner B (1999) Separation of pyrrolizidine alkaloids from Petasites hybridus with high pressure n-propane extraction. Chem Tech (Berl DDR 1949) 51(3):141–143 Knez Ž, Markoˇciˇc E, Leitgeb M, Primožiˇc M, Knez Hrnˇciˇc M, Škerget M (2014) Industrial applications of supercritical fluids: a review. Energy 77:235–243 Knez Ž, Škerget M, Knez Hrnˇciˇc M (2010) Principles of supercritical fluid extraction and applications in the food, beverage and nutraceutical industries. In: Separation, extraction and concentration processes in the food, beverage and nutraceutical industries, pp 3–38 Kravanja G (2018) High-perssure process design for polymer treatment and heat transfer enhancement. Ph.D. thesis, University of Maribor Oleson JP (2015) The final word on Roman wooden pumping machinery. J Roman Archaeol 28:707– 708 Papin D (1681) A new digester or engine for softening bones. Royal society Perrut M (2012) Sterilization and virus inactivation by supercritical fluids (a review). J Supercrit Fluids 66:359–371 Pilz S, Lack E, Seidlitz H, Steinhagen V, Stork K (2006) Überkritische extraktion aus sicht der industrie. Fluid-Verfahrenstechnik Grundlagen Method Tech Prax 2(11.2):1074–1130 Quick NJ, Cioffi WR, Shearer JM, Fahlman A, Read AJ (2020) Extreme diving in mammals: first estimates of behavioural aerobic dive limits in Cuvier’s beaked whales. J Exp Biol 223(18):jeb222109 Reverchon E (2002) Micro and nano-particles produced by supercritical fluid assisted techniques: present status and perspectives. Chem Eng Trans 2:1–10 Saito M (2013) History of supercritical fluid chromatography: Instrumental development. J Biosci Bioeng 115(6):590–599 Shen Z, Wang H, Li G (2011) Numerical simulation of the cutting-carrying ability of supercritical carbon dioxide drilling at horizontal section. Pet Explor Dev 38(2):233–236 Sokolov VI, Nikitin LN, Bulygina LA, Khrustalev VN, Starikova ZA, Khokhlov AR (2010) Supercritical carbon dioxide in organometallic synthesis: combination of sc-CO2 with Nafion film as a novel reagent in the synthesis of ethers from hydroxymethylmetallocenes. J Organomet Chem 695(6):799–803 Stahl E, Quirin K-W, Gerard D (2013) Verdichtete gase zur extraktion und raffination. Springer Taylor FW (2014) Venus: atmosphere. In: Spohn T, Breuer D, Johnson T (eds) Encyclopedia of the solar system. Elsevier Science & Technology, Oxford Tomioka N, Fujino K (1997) Natural (Mg, Fe) SiO3 -ilmenite and-perovskite in the Tenham meteorite. Science 277(5329):1084–1086 UHPT (2017) Uhde high pressure technologies Weidner E (2018) Impregnation via supercritical CO2 –what we know and what we need to know. J Supercrit Fluids 134:220–227 Yoo KP, Fukuzato R (2006) Current status of commercial development and operation of SCF technology in China, Japan, Korea and Taiwan. In: Proceedings of the 8th international symposium of supercritical fluids, Kyoto, Japan Zheng S, Hu X, Ibrahim A-R, Tang D, Tan Y, Li J (2010) Supercritical fluid drying: classification and applications. Recent Pat Chem Eng 3(3):230–244

Chapter 2

What Are Supercritical Fluids?

Abstract In this chapter brief definitions of subcritical and supercritical fluids are presented. The thermodynamic properties of systems with supercritical fluids like density, enthalpy, heat capacity, Joule–Thomson coefficient, dielectric constant are described. Special attention is given to phase equilibria of systems with supercritical fluids. Mass and heat transport properties of systems with supercritical fluid like viscosity, diffusivity, thermal conductivity, surface tension are described. Experimental techniques for determination of thermodynamic properties like density, solubility of substances in supercritical and vice versa, and solid–liquid–gas (S–L–G) equilibria are described. Different experimental techniques for determination of mass transport properties like diffusion coefficients, viscosity, interfacial tension are described in details. Keywords Supercritical fluid · Experimental · Phase equilibrium · Solubility · Density · Diffusion coefficients · Viscosity · Interfacial tension

2.1 Definition of Supercritical Fluids Today, there is a strong trend in the development of alternative technological processes with minimal environmental impact, such as reduced energy consumption, less toxic residues, better use of by-products, no by-products or artefacts, and at the end, also better quality and safety of the final products. High pressure technologies are relatively new processing tools, which led to the development of several processes resulting in completely new products with special customer-designed properties. A particular area of high pressure technologies is occupied by processing media named as supercritical fluids. By IUPAC definition (McNaught and Wilkinson 1997) “Supercritical Fluid” is the “defined state of a compound, mixture or element above its critical pressure (pc ) and critical temperature (Tc )” (Fig. 2.1). “A mixture of components is considered to be supercritical with respect to pressure, temperature or concentration, if conditions

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Ž. Knez and C. Lütge, Product, Process and Plant Design Using Subcritical and Supercritical Fluids for Industrial Application, https://doi.org/10.1007/978-3-031-34636-1_2

11

12

2 What Are Supercritical Fluids?

Fig. 2.1 P–V–T behaviour of pure fluid (Brunner 1994)

of state of a mixture (pressure, temperature, concentration) are beyond the critical point of certain mixture” (Brunner 1994). Baron Charles Cagniard de la Tour in year 1822 (Cagniard De La Tour 1822) proved the existence of a critical point. By conducting acoustics experiments in a sealed cannon, he noticed that a splashing sound generated by a solid ball in a liquid phase inside the cannon ceased above a certain temperature and pressure. That, he argued, indicates no liquid–gas phase boundary and surface tension in a supercritical fluid phase. Cagniard further showed that CO2 could be liquefied at 31 °C at a pressure of 73 bar, but not at a slightly higher temperature, even under pressures as high as 3000 bar. For pure substances, there is an inflection point in the critical isotherm (or temperature contour line) on a pV diagram. This means that at the critical point (the term was introduced by Thomas Andrews (Andrews 1870), the first and second derivatives of pressure with respect to volume are zero: (

∂P ∂V

)

( = T

∂2 P ∂V 2

) =0

(2.1)

T

Above the critical point there exists a state of matter that is continuously connected with (can be transformed without phase transition into) both the liquid and the gaseous state.

2.1 Definition of Supercritical Fluids Table 2.1 Characteristic values of liquid, supercritical fluid and gas

13

Property

Liquid

Supercritical fluid Gas

ρ (kg/m3 )

1 × 103

0.2 × 103 –0.9 × 103

0.6–1.0

η (kg m−1 s−1 )

1 × 10–3

1 × 10–5 –1 × 10–4

1 × 10–5

Dab (m2 s−1 )

0 x = l

(2.6)

where c1 is the concentration of the diffusing CO2 on the sample surface (mol/L), c0 is the initial concentration of the diffusing CO2 (mol/L) and l resprisents the thickness of sample plat (mm). After considering the boundary conditions, the solution of Fick’s second law can be expressed in the simplified form as: √ D√ mt 2 t =√ m eq π l

(2.7)

where m t is the mass of gas CO2 absorbed in the polymer at time t (g) and m eq is the mass of gas CO2 absorbed (√ in ) the polymer at equilibrium (g). From the initial gradient of the plot mmeqt = f t the diffusion coefficient (D) can be determined as follows: D=

π 22 ┌ l 4

(2.8)

where ┌ is the initial gradient of the plot (s–1/2 ). Results The increase of the diffusion coefficient with increasing pressure up to 150 bar in excipients PEG 600/CO2 and Brij52/CO2 system and reduction of the diffusion coefficient with further increasing of the pressure can be explained by the fact that CO2 acts as a plasticizer and additionally by the effect of hydrostatic pressure. When a polymer absorbs CO2 , its molecules are distributed in a new equilibrium state. At lower concentrations of absorbed CO2 , which coincides with the lower pressures, the plastic effect of CO2 causes higher mobility of the polymer chains and leads to higher diffusion coefficient. At higher concentrations of CO2 , the hydrostatic pressure plays a more important role. This pressure reduces free volume between chains of the polymer, thereby reducing the diffusion coefficient (Kegl et al. 2017). The obtained diffusion coefficients at pressures up to 150 bar are higher at lower temperatures, meanwhile, at pressures above 150 bar, higher diffusion coefficients are obtained at higher temperatures. This temperature effect is more explicit for the system of the Brij 52/CO2 system (Fig. 2.19).

2.4 Examples: Experimental Techniques for Determination …

35

Fig. 2.19 Diffusion coefficient as a function of pressure at various temperatures for the systems of Brij52/CO2 , PEG 600/ CO2 , PEG 1500/CO2 and PEG 6000/CO2

2.4.5 Viscosity Measurements in High-Pressure View Cell Apparatus The viscosity of CO2 saturated solutions of PEG 6000 was measured using a highpressure variable-volume view cell (NWA Gmbh, Lorrah, Germany) with a capacity of 60 ml. The method is based on the power requirement for mixing. The cell consists of two safire windows, an opening for inserting and emptying the CO2 gas, and for placing a thermocouple. The cell is designed for a maximum operating pressure of 750 bars and a maximum operating temperature of 200 °C. Viscosity is measured using a step-by-step procedure at temperatures higher than the melting point of an investigated polymer. Approximately 10 g of the sample was placed inside the high-pressure view cell. The pressure was increased with an inlet CO2 gas powered by a high-pressure liquid pump PM-101 (NWA Gmbh, Lorrah, Germany). The pressure was measured by an electronic pressure gauge (WIKA Alexander Wiegand GmbH & Co. KG Klingenberg, Germany) to within ± 0.1%, and the cell was electrically heated by a heating jacket accurate to within ± 0.5 °C. After 1 h, when the system reached thermodynamic equilibrium, mixing was turned on. Temperature, pressure, voltage, mixing rate and electric current were observed for every experiment. From the data obtained, Re and Ne nondimensional numbers were calculated. Re and Ne curves were also obtained for pure CO2 , under the same conditions as the polymer samples, from the variation of voltage, mixing rate and electronic current. Detailed description of the apparatus and the methods are described in (Borjan et al. 2022a, 2022b; Knez Hrnˇciˇc et al. 2014b) (Fig. 2.20). Viscosity was derived from the Reynolds number calculated from the Newton number provided in the equation below:

36

2 What Are Supercritical Fluids?

Fig. 2.20 A scheme of a high-pressure apparatus for viscosity determination. Kravanja et al. (2018)

Ne =

P ρ · ω3 · d 5

(2.9)

where P (W) presents power (a function of experimentally determined electric current and voltage), ρ (kg/m3 ) is the density of the fluid at certain conditions, ω (s-1) is rotation speed, d (m) is the diameter of the propeller mixer. Viscosity is calculated from the equation for the Reynolds number according to: Re =

ρ · ω · d2 , η

(2.10)

η=

ρ · ω · d2 , Re

(2.11)

where Re is the Reynolds number determined (from Ne versus Re curves for pure CO2 ) at each operating condition. Each data point represents the average of at least three measurements at a certain voltage. Results For system PEG 6000/CO2 at 343 K the viscosity decreases with increasing pressure due to higher solubility of gas in the sample. The same observation was found for all investigated systems of polyethylene glycols of different molar masses. For all it was found that, viscosity decreases with increasing pressure. Due to decreased solubility of gas at higher temperatures the viscosity of polyethylene glycols/CO2 systems increases at constant pressure (Fig. 2.21). To illustrate the dependence of pressure and temperature on viscosity of CO2 saturated solution of PEG 6000, a three-dimensional plot in temperature range of

2.4 Examples: Experimental Techniques for Determination …

37

Fig. 2.21 Viscosity (⟁) and solubility (◦) versus pressure for system PEG 6000/CO2 at 343 K. Kravanja et al. (2018)

323 K < T < 373 K and pressure range of 5 bar < p < 350 bar is presented in Fig. 2.22.

Fig. 2.22 Three-dimensional plot of viscosity CO2 saturated solution of PEG 6000 as a function of temperature and pressure (Kravanja 2018)

38

2 What Are Supercritical Fluids?

2.4.6 Density of Measurements by High-Pressure Vibration Tube Densitometer The density of CO2 saturated polymer solution was measured by a vibrating Anton Paar DMA 602 U-tube densitometer (Anton Paar, Graz, Austria) with an Anton Paar DMA 60 (Anton Paar, Graz, Austria) electronic control unit. Nitrogen and Milli Q water were used as calibration fluids (Fig. 2.23). Based on the oscillating time of nitrogen τN2 and Milli Q water τMilli Q water , determined experimentally, and known densities ρN2 and ρMilli Q water , the characteristic constant K of the device has been calculated: K =

ρN2 − ρMilli Q water 2 τN2 2 − τMilli Q water

(2.12)

When the U-tube was filled with a sample under the same experimental conditions, the oscillating times τpolymer/CO2 were measured and the density of the sample ρpolymer/CO2 determined by: ) ( 2 2 + ρN2 ρpolymer/CO2 = K τpolymer/CO − τ N 2 2

(2.13)

In order to reach equilibrium, approximately 10 min was needed at each pressure to stabilize the system. Detail operating procedure can be found in the literature (Kravanja et al. 2018). The U-tube was thermostated by means of an external temperature controlled circulating bath, which controls the temperature within ± 5 × 10−3 K. The temperature and pressure inside the U-tube were measured with an Anton Paar CKT 100 platinum resistance thermometer with an uncertainty of ± 0.01 K and a manometer (Nuova Fima EN837-1) with an accuracy of 0.25% for pressures lower than 600 bar. The reported uncertainty in the density of reference fluids is generally

Fig. 2.23 A scheme of a high-pressure cell with a vibrating U-tube for density determinations. Kravanja (2018)

2.4 Examples: Experimental Techniques for Determination …

39

Fig. 2.24 Density CO2 saturated PEG 6000 as a function of temperature and pressure. Kravanja (2018)

less than 0.1% with an estimation of ± 0.05 kg/m3 . Detailed operating procedure can be found in the literature (Kegl et al. 2017; Kravanja 2018). Results Density for all observed systems increases linearly with increasing pressure as a consequence of liquid compression and low gas solubility in the polymer matrix. Dissolving gas in the polymer causes an increase of free volume in the polymer. Therefore increasing pressure of gas result increasing gas concentration and therefore polymer swelling Conversely, density is significantly reduced with increasing temperature at isobaric conditions. The influence of pressure and temperature on the density of PEG 6000 saturated with CO2 , in temperature range of 323 K < T < 373 K and pressure range of 5 bar < p < 350 bar, is shown in a three-dimensional plot presented in Fig. 2.24.

2.4.7 Capillary Rise Method for Determination of Interfacial Tension Apparatus By means of the capillary rise method, accurate data of the equilibrium height of molten biodegradable polymers were measured in a high-pressure view equilibrium cell made of stainless steel (SITEC, Zurich, Switzerland). The cell volume is 500 mL

40

2 What Are Supercritical Fluids?

Fig. 2.25 A scheme for measuring the equilibrium heights by capillary rise (CR) method in a high-pressure equilibrium view cell. https://doi.org/10.1016/j.supflu.2015.10.013

and is designed for a pressure of 500 bar and temperature of 150 °C. The pressure inside the cell was measured by an electronic pressure gauge (WIKA Alexander Wiegand GmbH & Co. KG Klingenberg, Germany). The temperature of the cell was kept constant using a heating jacket and was observed using calibrated thermocouple immersed in the cell. The uncertainty of the pressure was 0.1 bar and the total uncertainty of the temperature was 0.1 °C (Fig. 2.25). Capillary Rise Method The capillary rise technique is a well-established technique used to measure surface (interface) tension. When a glass capillary tube with known inner diameter is immersed in a wetting liquid, the liquid rises due to the action of surface tension forces (Fig. 2.26). ( Equilibrium ( ) )height (h) occurs when the force of gravity on the volume of liquid ρ · h · πr 2 g balances the force due to surface tension (r (2πr )). Since some of the liquid remains above the meniscus (the surface is not flat), equilibrium height h should be replaced with correlation (h + r/3) that results in well-known equation for surface tension: γ =

( ) r) ( 1 ·r ·g· h + · ρ p/CO2 − ρCO2 2 3

(2.14)

where r is the radius of the applied capillary, Δρ is the is the difference in density of the interfacing components (gas-saturated liquid in equilibrium with the CO2 phase), g = 9.8 m/s2 is the acceleration due to the gravity and since it is experimentally evident that liquid fully wets the glass of capillary tube, the contact angle is assumed to be zero if θ < 8°. Two different sizes of capillaries were placed vertically inside the measuring cell on a stillness holder. One was with a radius of 0.1500 mm and other with a radius 0.4780 mm. First, equilibrium height was determined of a water

2.4 Examples: Experimental Techniques for Determination …

41

Fig. 2.26 Equilibrium height (h) occurs when the force of gravity on the volume of liquid balances the force due to surface tension. Kravanja (2018)

rise inside a capillary at ambient and also at elevated pressure, followed by the determination of equilibrium height of the melted polymer rise under the pressure of CO2 and argon. While equilibrium height of a water rise inside a capillary, was reached approximately in 5 min, equilibrium height of the melted polymers needed more time: approx. 30 min due to the higher viscosity of the system. However, with increasing pressure due to viscosity reduction of polymers arising from higher solubility of gas in the polymer, lower time was needed to establish equilibrium inside a capillary. Second, the capillary rise method was validated by measuring surface tension at the CO2 + water interface within a pressure range from 1 to 200 bar at a constant temperature 318.15 K. The obtained experimental results were compared to the data available in the literature. At least three measurements were performed at a certain pressure and temperature, and the relative standard deviation between the experimental data and the literature data ranged below 5%. The main difficulties of the method lie in determining the uniform capillary diameter and ensuring system cleanliness. However, recent progress in the manufacturing of glass tube and laser techniques allow knowledge of the capillary diameter up to a high precision (∓ 0.001 mm). In our research, the radii of capillaries, at both ends, were determined by a laser coordinate measuring machine (Zeiss, UMC Zeiss UMC 850) (Fig. 2.27). Results It is evident that pressure has a significant effect on interfacial tension reduction; where the temperature effect (measured for system PEG 6000/CO2 (Kravanja et al.

42

2 What Are Supercritical Fluids?

Fig. 2.27 Correlation between interfacial tension and solubility of system CO2 /PEG 1500 at 343 K

2016) is much lower. Interfacial tension is reduced by elevated pressure due to higher solubility of gas in heavy component. Influence of pressure and temperature on interfacial tension presented in a threedimensional plot is presented in Fig. 2.28 for temperature range of 323 K < T < 373 K and pressure range of 5 bar < P < 350 bar.

2.4.8 Transport Properties from Drop Geometry in Dense Fluid Apparatus The central part of the experimental setup comprises an optical high-pressure cell (NWA Gmbh, Lorrah, Germany). Liquid phase was injected by a high-pressure manual pump (mod. 750.1100, SITEC, Zurich, Switzerland). Pendant drops of adequate size were formed on a stainless steel tip that was placed vertically in the cell between two sapphire windows. Measurements of dynamic drop volume were filmed with a Basler Aca1300200um digital camera (Basler, Ahrensburg, Germany) equipped with a CCTV lens (Tamron, Saitama, Japan), connected to a computer by using the Open Drop algorithm. To avoid optical aberrations and the fake reflections from other sources that can occur at the drop edge, the drop was lit from the other side with a diffusion light, which was achieved by placing a glass diffuser between the light source and the hanging drop. The undesirable effect of droplet oscillation was minimized with an antivibration table. Pressure inside the cell was increased with inlet gas by a high-pressure

2.4 Examples: Experimental Techniques for Determination …

43

Fig. 2.28 Interfacial tension CO2 saturated solution of PEG 6000 as a function of temperature and pressure

pump PM-101 (NWA Gmbh, Lorrah, Germany) or gas booster (Maximator DLE 751-GG-H2, MAXIMATOR GmbH, Nordhausen, Germany), depending on the type of gas. Pressure was monitored during the entire experiment by an electronic pressure gauge (WIKA Alexander Wiegand GmbH & Co. KG Klingenberg, Germany) with an uncertainty of 0.1 bar. Total uncertainty of the temperature measurement was 0.1 °C. The entire experimental setup is presented in Fig. 2.29. Pendant Drop Method Details on drop tensiometry measurements are presented in (Kravanja 2018). The basis of the method is to form a drop of investigated substance in dense gas. From drop shape the surface tension as well the diffusivity in the system substance/CO2 could be determined. Results The high-pressure tensiometry method was used to measure diffusion coefficients and surface tension for propylene glycol in supercritical CO2 . Diffusion Coefficient of Propylene Glycol/CO2 System Diffusion coefficients of propylene glycol in supercritical CO2 were measured at temperatures of 125 °C and 150 °C and at pressures ranging from 50 bar, up to 175 bar. As could be seen from Fig. 2.30, diffusivity decreases as the pressure increases under isothermal conditions, where the decrease is more noticeable at lower pressures.

44

2 What Are Supercritical Fluids?

Fig. 2.29 Scheme of the experimental setup for measuring interfacial tension and diffusion coefficients by a high-pressure view cell. https://doi.org/10.1016/j.supflu.2017.09.022 Fig. 2.30 Diffusion coefficients in propylene glycol saturated the solution with CO2 at 125 °C and 150 °C in the pressure range from 50 bar up to 175 bar

2.4 Examples: Experimental Techniques for Determination …

45

Fig. 2.31 Density of the propylene glycol saturated the solution with CO2 at 125 °C and 150 °C in the pressure range from 50 bar up to 175 bar

This can be explained as the result of a high number of molecular collisions and the smaller mean free path between them. On the other hand, diffusivity increases as the temperature increases at constant pressure, where greater dependence on the temperature is observed at a lower pressure. Interfacial Tension of a CO2 -saturated Propylene Glycol Solution Interfacial tension and density of a CO2 -saturated propylene glycol solution were obtained using the pendant drop method, simultaneously with drop diameter measurements. Density and interfacial tension as functions of pressure and temperature are illustrated in Figs. 2.31 and 2.32. Density for all observed systems increased linearly with pressure as a consequence of liquid compression and low solubility. The interfacial tension for the propylene glycol/CO2 binary system decreased with increasing pressure.

46

2 What Are Supercritical Fluids?

Fig. 2.32 Interfacial tension of the propylene glycol saturated the solution with CO2 at 125 °C and 150 °C in the pressure range from 50 bar up to 175 bar

References Andrews T (1870) XXI. Bakerian lecture—on the continuity of the gaseous and liquid states of matter. Proc R Soc Lond 18(114–122):42–45 Bertucco A, Vetter G (2001) High pressure process technology: fundamentals and applications. Elsevier Borjan D, Cör D, Knez Marevci M, Grˇcar I, Knez Ž (2022a) Phase equilibrium data of tetrabutylurea, tetramethylurea, and tetramethylthiourea/carbon dioxide at pressures up to 200 bar at 313.15 and 333.15 K. J Chem Eng Data 67(9):2378–2383 Borjan D, Knez Marevci M, Knez Ž (2022b) P-x,y equilibrium data of the binary systems of 2propanol, 1-butanol and 2-butanol with carbon dioxide at 313.15 K and 333.15 K. Molecules 27(23):8352 Brignole EA, Pereda S (2013) Phase equilibrium engineering. Elsevier, Amsterdam Brunner G (1994) Gas extraction-an introduction to fundamentals of supercritical fluids and the application to separation processes. In: Baumgärtel H, Franck EU, Grünbein W (eds) Topics in physical chemistry, vol 4. Springer, Steinkopff, Darmstadt, New York, 387 S., DM 64 Budisa N, Schulze-Makuch D (2014) Supercritical carbon dioxide and its potential as a lifesustaining solvent in a planetary environment. Life 4(3):331–340 Cabeza LF, de Gracia A, Fernández AI, Farid MM (2017) Supercritical CO2 as heat transfer fluid: a review. Appl Therm Eng 125:799–810 Cagniard De La Tour C (1822) Exposé de Quelques Résultats Obtenupar L’action Combinée de la Chaleur et de la Compression Sur Certains Liquides, Tels Que L’eau, L’alcool, L’ether Sulfurique et L’essence de Pétrole Rectifiée. Ann Chim Phys 21:127–132 Deiters UK, Kraska T (2012) High-pressure fluid phase equilibria: phenomenology and computation. Elsevier, Amsterdam Dohrn R, Brunner G (1995) High-pressure fluid-phase equilibria: experimental methods and systems investigated (1988–1993). Fluid Phase Equilib 106(1–2):213–282 Fornari RE, Alessi P, Kikic I (1990) High pressure fluid phase equilibria: experimental methods and systems investigated (1978–1987). Fluid Phase Equilib 57(1):1–33

References

47

Fukné-Kokot K, König A, Knez Ž, Škerget M (2000) Comparison of different methods for determination of the S-L–G equilibrium curve of a solid component in the presence of a compressed gas. Fluid Phase Equilib 173(2):297–310 Hrnˇciˇc MK, Škerget M, Knez Ž (2014) Argon as a potential processing media for natural and synthetic substances. J Supercrit Fluids 95(Supplement C):252–257 Jarrahian A, Heidaryan E (2012) A novel correlation approach to estimate thermal conductivity of pure carbon dioxide in the supercritical region. J Supercrit Fluids 64:39–45 Kegl T, Kravanja G, Knez Ž, Knez Hrnˇciˇc M (2017) Effect of addition of supercritical CO2 on transfer and thermodynamic properties of biodegradable polymers PEG 600 and Brij52. J Supercrit Fluids 122:10–17 Knez Hrnˇciˇc M (2014) Thermodynamic and physical properties for high pressure process design. Ph.D. Univerza v Mariboru (Slovenia) Knez Hrnˇciˇc M, Markoˇciˇc E, Trupej N, Škerget M, Knez Ž (2014a) Investigation of thermodynamic properties of the binary system polyethylene glycol/CO2 using new methods. J Supercrit Fluids 87:50–58 Knez Hrnˇciˇc M, Škerget M, Knez Ž (2014b) Density and viscosity of the binary polyethylene glycol/ CO2 systems. J Supercrit Fluids 95:641–668 Knez Ž, Škerget M, Mandžuka Z (2010) Determination of S–L phase transitions under gas pressure. J Supercrit Fluids 55(2):648–652 Kravanja G (2018) High-perssure process design for polymer treatment and heat transfer enhancement. Ph.D., Univerza v Mariboru (Slovenia) Kravanja G, Hrnˇciˇc MK, Škerget M, Knez Ž (2016) Interfacial tension and gas solubility of molten polymer polyethylene glycol in contact with supercritical carbon dioxide and argon. J Supercrit Fluids 108:45–55 Kravanja G, Knez Ž, Hrnˇciˇc MK (2018) Density, interfacial tension, and viscosity of polyethylene glycol 6000 and supercritical CO2 . J Supercrit Fluids 139:72–79 Magalhães AL, Lito PF, Da Silva FA, Silva CM (2013a) Simple and accurate correlations for diffusion coefficients of solutes in liquids and supercritical fluids over wide ranges of temperature and density. J Supercrit Fluids 76:94–114 Magalhães AL, Vaz RV, Gonçalves RMG, Da Silva FA, Silva CM (2013b) Accurate hydrodynamic models for the prediction of tracer diffusivities in supercritical carbon dioxide. J Supercrit Fluids 83:15–27 Mamaliga I, Schabel W, Kind M (2004) Measurements of sorption isotherms and diffusion coefficients by means of a magnetic suspension balance. Chem Eng Process Process Intensif 43(6):753–763 Martín A, Cocero MJ (2007) Mathematical modeling of the fractionation of liquids with supercritical CO2 in a countercurrent packed column. J Supercrit Fluids 39(3):304–314 McNaught AD, Wilkinson A (1997) Compendium of chemical terminology. Blackwell Science Oxford Michels A, Michels C (1933) The influence of pressure on the dielectric constant of carbon dioxide up to 1000 atmospheres between 25 and 150 C. Philos Trans R Soc Lond Ser Contain Pap Math Phys Character 231(694–706):409–434 NIST Office of Data (2019) NIST WebBook. https://webbook.nist.gov/ Peper S, Fonseca JMS, Dohrn R (2019) High-pressure fluid-phase equilibria: trends, recent developments, and systems investigated (2009–2012). Fluid Phase Equilib 484:126–224 Poletto M, Reverchon E (1996) Comparison of models for supercritical fluid extraction of seed and essential oils in relation to the mass-transfer rate. Ind Eng Chem Res 35(10):3680–3686 Ratnakar RR, Dindoruk B (2016) On the exact representation of pressure decay tests: new modeling and experimental data for diffusivity measurement in gas-oil/bitumen systems. In: SPE annual technical conference and exhibition. Society of petroleum engineers Reid RC, Prausnitz JM, Poling BE (1989) The properties of gases and liquids, 4th edn. McGrawHill, United States

48

2 What Are Supercritical Fluids?

Simoes PC, Matos HA, Carmelo PJ, Gomes de Azevedo E, Nunes da Ponte M (1995) Mass transfer in countercurrent packed columns: application to supercritical CO2 extraction of terpenes. Ind Eng Chem Res 34(2):613–618 Suárez JJ, Medina I, Bueno JL (1998) Diffusion coefficients in supercritical fluids: available data and graphical correlations. Fluid Phase Equilib 153(1):167–212 Sun Y, Shekunov BY (2003) Surface tension of ethanol in supercritical CO2 . J Supercrit Fluids 27(1):73–83 SWEP (2019) Appendix B—log P/h diagrams for refrigerants. https://www.swep.de/refrigeranthandbook/appendix/appendix-b/ Trupej N, Hrnˇciˇc MK, Škerget M, Knez Ž (2015) Solubility and binary diffusion coefficient of argon in polyethylene glycols of different molecular weights. J Supercrit Fluids 103:10–17 Valle JMD, De La Fuente JC (2006) Supercritical CO2 extraction of oilseeds: review of kinetic and equilibrium models. Crit Rev Food Sci Nutr 46(2):131–160 Xiong T, Yan X, Huang S, Yu J, Huang Y (2013) Modeling and analysis of supercritical flow instability in parallel channels. Int J Heat Mass Transf 57(2):549–557 Zang D, Yu Y, Chen Z, Li X, Wu H, Geng X (2017) Acoustic levitation of liquid drops: dynamics, manipulation and phase transitions. Adv Colloid Interface Sci 243:77–85 Zhao Z, Su S, Si N, Hu S, Wang Y, Xu J, Jiang L, Chen G, Xiang J (2017) Exergy analysis of the turbine system in a 1000 MW double reheat ultra-supercritical power plant. Energy 119:540–548

Chapter 3

Industrial Scale Applications: Physical-Based Processes

Abstract In this chapter industrial scale applications based on solvent properties of supercritical fluids are described in first sub-chapter. Thermodynamic fundamentals of extraction processes of solids and liquids with supercritical fluids are presented. Different operating modes of extraction units as well as different separation strategies are presented. Energy consumption and design considerations for design of industrial scale extraction units for solids extraction are given. A special part of sub-chapter is devoted to ultrahigh pressure extraction where operating pressure in extraction process is more than 1000 bar. Benefits of ultrahigh pressure extraction are presented in details. Scale-up of different extraction processes which are realized on industrial scale are presented. In the second sub-chapter processes for polymer particles production as well for polymer foaming are presented. Fundamental data for design and scale-up of Crystallization, Rapid expansion of supercritical solution— RESS, Gas anti-solvent—GAS and Particles from gas saturated solution—PGSS™ processes are reviewed. Keywords Supercritical fluid · Extraction · Solid · Liquid · Polymer · Particle formation · Polymer foaming · Crystallization · RESS · GAS · PGSS™

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid Materials The application of subcritical or supercritical fluids as relatively new processing media can provide new products with special customer-designed and completely new properties. These processes are typically environmentally friendly, of low costs and sustainable. Benefits of most subcritical or supercritical fluids in industrial scale processes are in replacement of environmentally far more problematic conventional organic solvents. Another environmental impact is in most processes the lower energy consumption compared to conventional process operation.

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Ž. Knez and C. Lütge, Product, Process and Plant Design Using Subcritical and Supercritical Fluids for Industrial Application, https://doi.org/10.1007/978-3-031-34636-1_3

49

50

3 Industrial Scale Applications: Physical-Based Processes

SCCO2 and SCH2 O, which are mostly used as SCFs, are non-toxic, noncarcinogenic, non-mutagenic, non-flammable and are both thermodynamically stable. Typical thermo-physical properties of SCFs are low viscosity, high diffusivity, density and the dielectric constant of SCF, which can easily be changed by varying the operating pressure and/or temperature (Knez 2016). The benefits using subcritical or supercritical fluids for extraction and fractionation of natural products from solids is well described in the literature (Marr and Gamse 2000) (solvent free products, no side-products, no artefacts, low processing temperature, etc.). Selective extraction as one of the most important advantages applying subcritical or supercritical fluid extraction is possible by using different gases for isolation/fractionation of components and/or varying process parameters. The price of the product presents in some cases the limitation of further application of extracts obtained by high pressure technology. But the legal limitations (e.g. 2009/32/EC) of solvent residues and solvents and isolation/fractionation of special components from total extracts in combination with different formulation and sterilization processes (controlled release for example) will increase the use of dense gases in extraction processes applications. In general, several feedstocks could be extracted by different extraction methods. An overview of extraction methods versus feed materials are presented in Table 3.1. Supercritical fluid extraction from solid materials can be found in both on pilot and on an industrial scale (McHugh and Krukonis 2013). Different gases for isolation or fractionation of components, are used for extraction of valuable compounds from hop, for decaffeination of tea and coffee which are the largest scale processes and are mostly performed in industrial scale. Several industrial plants are in operation also for extraction of spices for food industry and natural substances for use in cosmetics. (Knez et al. 2013). These plants are mainly applying batch extraction (refer to Table 3.1). An extensively investigated area represents the application of subcritical or supercritical fluid extraction in vegetable oil industry offering an alternative to conventional refining, separation and fractionation processes (Temelli 2009). Solvent-free products, low temperature during the process, no by-products are some of the main advantages using SCFs for extraction, isolation of several natural compounds (Marr and Gamse 2000). The legal limitations on solvent residues and solvents especially in products for human use as well as isolation/fractionation of specific components from extracts and further formulation encourage the use of dense gases in extraction applications. There are relatively few industrial units using supercritical fluids for separation of compounds from liquid phases. Lab scale or pilot scale extraction units, however, often involve liquid/supercritical fluid features offering a wider range of process development or applications to be investigated in depth. There are also some available data on binary systems liquid/supercritical fluid but there is a lack of data on liquid/ liquid/supercritical fluid systems. The basic data needed for separation processes is the solubility of a single compound or a mixture of compounds in supercritical fluid. For successful separation it is necessary that compounds or a mixture of compounds are soluble in supercritical fluid. As in all extraction processes also in supercritical

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

51

Table 3.1 Extraction methods for different materials Method

Feedstock

Advantages

Batch extraction

Solids (liquids)

Efficient, no Discontinuous pressure filling and limitation, easy emptying operation

Disadvantages

Continuous operation

Suitable for viscous feedstocks

Discontinuous − feedstock handling

Liquid–liquid Liquids extraction

Highly efficient, low cost

Limited pressure Yes (hydrodynamic), fouling

−

Agitated extraction vessel

Liquids, solids

Mixing generates required mass transfer surface

Difficult if from a liquid feedstock a solid raffinate is targeted

Yes, if feedstock and raffinate are free-flowing

+

Extruder extraction

Liquids

Mixing generates required mass transfer surface, low cost

Sealing extraction homogeneity

Yes

++

Spray extraction

Liquids Particle Not suitable for (suspensions) formation dry, powdery provides large feedstocks mass transfer surface and small diffusion length

Yes, if raffinate is free-flowing

++

extraction of solid and liquid mixtures the solubility of single component or mixture of components in subcritical or supercritical fluid are the basic data for design of separation processes. The components or mixture of compounds which have to be extracted have to be soluble in SCF/dense gas. The solubility of components in supercritical fluids primarily depends on the density of the supercritical fluid. The density of a supercritical fluid can be changed by two process parameters pressure and temperature. Another significant parameter manipulating the solubility of compounds in supercritical fluid is its dielectric constant which can be controlled by temperature and/or pressure of supercritical fluid. The phase equilibrium data are significant to establish proper operating parameters (pressure and temperature) of supercritical fluid in an extraction plant. In extraction stage, the solubility of a compound or mixture of compounds has to be the highest while in separation step the solubility of compound in SCF has to be the lowest (Knez et al. 2013). Therefore, the phase equilibrium data are the most important data for the design of operating pressure and operating temperature of SCF in extraction plant. Based

52

3 Industrial Scale Applications: Physical-Based Processes

on phase equilibrium data the theoretical amount of supercritical fluid necessary for separation of compound from solid or liquid mixture could be calculated. Properly selected process parameters are crucial for economics of the process, since design parameters have a very important influence on the investment and operating cost for the high pressure plant and subsequently on the economy of the process. Mass transfer also effects on the economics of extraction process due to the capacity of extraction unit at certain capacity and subsequently the investment costs. Mass transfer models are most commonly presented as yield of extraction process (mass of supercritical solvent versus mass of solid material—solvent to feed ratio S/ F) (Knez et al. 2013). Realizing the process of supercritical extraction and translating the previously discussed theory into the world of “heavy metal” can be achieved in different scale and complexity. The simplest approach is shown by a basic flow-sheet of the extraction process in Fig. 3.1. The solid feedstock or raw material is placed within the extractor. Supercritical fluid is taken from the solvent tank and brought up to the desired process conditions of pressure and temperature. The solvent flows through the bed placed in the extractor whereby removing soluble ingredients and carrying those into the separator. By releasing the pressure, the extract can be collected in the separator and the cleaned fluid is recycled into the solvent tank. This simple flowsheet is typically realized in pilot plants allowing to extract larger amounts of material, optimize process parameters and investigate the influence of the solvent recycle, resp. the accumulation of components lowering the efficiency of the process.

Fig. 3.1 Basic flow sheet of supercritical extraction plant. https://doi.org/10.1533/978085709075 1.1.3

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

53

Figure of a “real” pilot plant could be seen on Fig. 5.13 in Chap. 5. Cascade operation is used in industrial scale plants to increase the economy of the solid—SC solvent extraction process. Till now, there was some research on continuous operation of plants for extraction of solids with SC solvent (Eggers 1996). So far no commercial application for continuous feeding of solids was yet applied on industrial scale. Feeding a solid material into a pressure vessel is technically a difficult task with regard to the equipment—it can be realized, however, the cost of the equipment would be high and only bring benefits in very large scale. Therefore, in industrial scale usually a number of extractors is combined in series. By cycle operation of such battery of extractors, a quasi–continuous solid flow in counter-current to the extraction fluid can be achieved. Such mode of operation gives extremely high extraction yields, because the (almost) pure solvent is contacted with pre-extracted (or soaked) material and thus the solvent is loaded to maximum solvent capacity. In case the process aims at removing a substance from the solid matrix (e.g. cleaning) and the solid matrix or raffinate is the final product, such operation also achieves very low amounts of residue traces and high value of the product. A set-up of such commercial plant is shown in Fig. 3.2. A major advantage of supercritical fluid extraction is the possibility to easily produce fractions of the extract with strongly differing properties. Such multi-step separation can be achieved by use of several separators and stepwise reducing the solvent power. A basic flow sheet of such extraction unit is presented in Fig. 3.2. Reduced solvent power can be accomplished by several different methods which will be described in a following subchapter. There are numerous applications for extraction of solids using supercritical fluids. Several overviews are available (Brunner 2013; Catchpole et al. 2009; Diaz and Brignole 2009; Eltringham and Catchpole 2007; Fang et al. 2007; Gardner 1993;

Fig. 3.2 Cascade operation extraction of solids—3 extractors with 2 step separation (fractionation of extract)

54

3 Industrial Scale Applications: Physical-Based Processes

Lack and Seidlitz 1993; Li 2007; Mendes 2007; Moyler 1993; Mukhopadhyay 2007a; Temelli 2009; Temelli et al. 2007; Stahl et al. 2013). The highest capacity equipment is installed for coffee and tea decaffeination followed by second largest application for the extraction of hop compounds. Extraction of spices for the production of oleoresins are also widely used. As well as applications of extraction of bioactive compounds from plants. One of the latest applications is the extraction of oil from the degumming residue to obtain highly concentrated and very pure lecithin powder (plant designed, manufactured and erected by Uhde HPT). There are fewer industrial units for separation of components from liquid mixtures using supercritical fluids. Literature review shows some laboratory scale studies on extractions in the systems liquid/SCF. Numerous data on binary systems liquid/SCF could be found in the literature, but there is lack of data on systems liquid/liquid/ SCF (Gupta et al. 2006). Extraction of liquid mixtures with supercritical fluids is similar to liquid–liquid extraction, where supercritical fluid is used instead of an organic solvent. In liquid–supercritical fluid extraction processes the pressure represents an important factor. By changing pressure and/or temperature, the physicochemical properties (density, viscosity, surface tension, and dielectric constant) of the supercritical fluid are changed. Selective extraction of components as well as fractionation of total extracts are possible by use of different fluids. Another advantage is that—depending on the feed material—the density difference between the two counter-current flowing phases can be adjusted. Easy solvent regeneration is one of the major advantages of use of supercritical fluids. In liquid–liquid extraction solvent regeneration in many cases involves a re-extraction or distillation step, which is energy consuming and increases the cost of the extraction process. Heat treatment of extract or raffinate may degrade thermolabile substances. For extraction plant where supercritical fluids are used as solvent media, regeneration is reached by changing process parameters pressure and/or temperature after the extraction step. Consequently, density is changed and the fluid can be easily recycled after separation of solute. Compared to extraction of solids with supercritical fluid, liquids could be continuously introduced. This allows higher flowrates in continuously operating counter-current processes. Literature review indicates some laboratory scale studies on extractions in the systems liquid/supercritical fluid (Knez 2016). Liquid/sub- or supercritical fluid extraction were used for separation of ethanol from water (Hsu and Tan 1994; Knez et al. 1994), separation of aromas from different alcoholic beverages (Gamse et al. 1999), separation of components from citrus oils (Knez 1989), for purification of tocopherols (Fleck et al. 2000). Separation of caffeine from CO2 with water is used widely in decaffeination process. In the future, further restrictions on the use of organic solvents and the demands of new applications will be the deciding factors for sustainable processing (Knez et al. 2013). Several operating methods of extraction liquid/sub- or supercritical solvent are available. The simplest method is single-stage extraction. It is used in systems where the separation factors for a solute are high. When the separation factor is in the order of 1–10 multistage separation is used. Several modes of operation in multistage processes are used, like multistage crossflow with relatively low loading of

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

55

solvent with the extract obtained in each stage. In multistage counter-current extraction, high loading of solvent with extracts is possible and a different configuration of the apparatus is possible. Counter-current liquid/sub- or supercritical fluid extraction (SFE) may be designed by the use of basic equations: mass balance, energy balance, equilibrium distribution coefficients, and mass transfer rate equations. Determination of the number of theoretical stages/transfer units, size and type of a separation device, design of the solvent cycle are essential for evaluation of extraction costs. For instance, batch processes have higher operating costs, whereas in continuous processes the costs are lower. For continuous process 60,000 tonne per year approximately 0.06 e/gf feed while for batch around 60 e/kg for c. 200 tonne per year (Brunner 2009).

3.1.1 Thermodynamic Fundamentals of Extraction Processes The difference in concentration of compounds from equilibrium state at certain process conditions is the driving force for mass transfer in all extraction/leaching processes. All these processes are based on the solubility of compounds in phases present in the extraction system. Phase behaviour near the critical point can be very complex even for relatively simple binary mixtures, especially when mixture components differ in molecular size and shape, structure and polarity. Phase behaviour can be presented by a P–T and P–x projections of P–T–x diagrams. Using Gibbs phase rule (Eq. 3.1) determines the geometrical constrains for presentation of multiphase regions. F=C+2−P

(3.1)

In (Eq. 3.1) F is the number of independent variables, C is the number of components and P is the number of phases. The phase equilibrium of substances (1, 2, 3,…, N) distributed between phases (α, β, γ, …, π) present in the extraction system can be defined in terms of the fugacity (f) by Eq. (3.2): β

γ

fi α = fi = fi = . . . . . . = fi π i = 1, 2, 3, . . . , N

(3.2)

Only binary systems will be presented in more details in this present chapter. Phase equilibria of multi-component mixtures are discussed in details in the literature (Brunner 1994; Sadus 2012; McHugh and Krukonis 2013).

56

3 Industrial Scale Applications: Physical-Based Processes

3.1.2 Solid-Supercritical Fluid Equilibrium 3.1.2.1

Phase Diagrams

In the system solid–supercritical fluid mostly the normal melting temperature of the solid matter is higher than the critical temperature (Tc ) of the supercritical fluid. In that case two possible phase behaviours can be present. For chemically similar components the simplest behaviour is presented in Fig. 3.3a, where the critical mixture curve runs continuously among the critical points of both components. The solid– liquid–gas (SLG) line is continuous. It begins at the normal melting point of the heavier component and goes toward lower temperatures as the pressure is increased. Usually the line ends at a temperature below the critical temperature of the lighter component. The melting point of the pure solid generally increase with an increase in the hydrostatic pressure. In the presence of dense gas, the melting point of the solid decreases as the pressure increases due to the increasing solubility of gas in the solid (Knez et al. 2013). Another type of solid–supercritical fluid phase behaviour is shown in Fig. 3.3b and is representative for systems in which the solid and the supercritical fluid vary significantly in molecular size, form and/or polarity. This kind of behaviour can be understood as type III fluid-phase behaviour (de Loos 2006) according to the classification of van Konynenburg and Scott (Konynenburg et al. 1980). In that case the gas which is lighter is poorly soluble in the heavy liquid, even at higher pressures. It results in the melting-point depression of the solid which is quite small. The solid–liquid-gas curve is no longer continuous. In that case three phase solid–liquid-gas equilibria are presented by two branches of the solid–liquid– gas line in P–T diagram (Knez et al. 2013). Means that high-temperature branch of the solid–liquid–gas line begins at the normal melting point of the solid and crosses with the critical-mixture curve at the upper critical end point (UCEP).

Fig. 3.3 Solid-supercritical fluid equilibrium. P–T phase diagram for: a similar binary systems, b asymmetrical binary systems. C, critical point; TP, triple point; L, liquid; G, gas; S, solid; UCEP, upper critical end point; LCEP, lower critical end point. (----) critical lines, line 1 and 2 vapour pressure curves for two components. https://doi.org/10.1533/9780857090751.1.3

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

57

The low-temperature branch of the solid–liquid–gas line crosses with the criticalmixture curve at the lower critical end point (LCEP). At these two points both the liquid and gas phases combine into a single fluid phase in the presence of excess solid. Among these two branches of the SLG line exists only a solid–gas equilibrium. Potential phase behaviour for type III systems is in detail presented in literature by de Loos (de Loos 2006). The course of the three-phase solid–liquid–gas line of a binary system depends on the solubility of the gas in the liquid phase. If the gas dissolves in the melt of a heavy component temperature two effects appear and are given by Eq. (3.1): ( dT =

∂T ∂P

)

( dP +

xA

∂T ∂XA

) dxA = P

T ΔV fus RT 2 dP + dxA ΔH fus xA ΔH fus

(3.3)

By means when hydrostatic pressure increases the melting temperature of heavy component, and the dissolved gas in heavy component decreases the melting temperature. Higher solubility of gas in the melting results in larger melting-point depression (Knez et al. 2013). Four specific shapes of solid–liquid–gas equilibrium lines in P–T diagram were experimentally observed and are presented in Fig. 3.4. These are typical for asymmetric binary systems of compressed gases and non-volatile compounds (Arons and Diepen 1963; de Loos 2006; Knez et al. 2013; Weidner et al. 1997). – negative dP/dT slope—the effect of gas solubility prevails; – positive dP/dT slope—the effect of pressure prevails; Fig. 3.4 Specific shapes of solid–liquid–gas lines. https:/ /doi.org/10.1533/978085709 0751.1.3

58

3 Industrial Scale Applications: Physical-Based Processes

– temperature minimum—both effects are competing; – temperature maximum and a temperature minimum. The last shape of solid–liquid–gas is a very rare phenomenon and was reported for systems CO2 + polyethylene glycol (Weidner et al. 1997), CO2 + tripalmitin (O’Connell et al. 2003) and can be explained by the higher solubility of the supercritical gas in the liquid phase than in the solid phase (de Loos 2006). From the literature can be found that in these equilibria the supercritical fluid is insoluble in the phase of the non-volatile component (de Loos 2006). The form of the solid–liquid–gas line is reliant on the gas and the chemical structure of the compound, i.e. the type and position of the functional groups. An example which demonstrates that isomers may have a different type of solid–liquid–gas line in the presence of a dense gas, is the vanillin–gas system. Binary systems of vanillin (V) and o-vanillin (o-V) with fluorinated hydrocarbons (R23, R134a, R236fa) and CO2 were investigated. Solid–liquid-gas phase curves are presented in Fig. 3.5. It can be observed that for vanillin with –OH group in the para position, the melting point depression in CO2 and fluorinated hydrocarbons is generally lower than for vanillin with –OH group in the ortho position. Experimentally determined melting point of vitamin K3 under pressure of different gases is presented in Fig. 3.6 and indicates various paths of the solid–liquid–gas line (Knez and Škerget 2001). In the presence of CO2 and dimethyl-ether the negative

Fig. 3.5 Solid–liquid–gas lines for vanillin (V) and o-vanillin (o-V) in presence of dense gases. https://doi.org/10.1533/9780857090751.1.3

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

59

Fig. 3.6 Solid–liquid–gas equilibria for binary system K3 —gas (CO2 , propane, dimethyl ether, argon or nitrogen). https://doi.org/10.1533/9780857090751.1.3

slope dP/dT can be observed. It can be found that the melting point depression of vitamin K3 is the highest under the pressure of dimethyl-ether. While the meltingpoint depression of vitamin K3 is less definite in the presence of propane. On the other hand, under the pressure of inert gas (nitrogen and argon), the solid–liquid–gas curve has a positive dP/dT slope due to the low solubility of gas in vitamin K3. It is generally known that pharmaceutical substances are poorly soluble in supercritical CO2 . For example phase equilibria, namely equilibrium solubility and melting point depression of fenofibrate in several dense gases (carbon dioxide, propane, trifluoromethane) was investigated and solid–liquid–gas curves are presented in Fig. 3.7 (Ljubec et al. 2018). A melting point depression with a temperature minimum in the p, T –projection of the SLG line has been noticed for all dense gases used.

3.1.2.2

Thermodynamic Modelling

Solid-Supercritical Fluid At equilibrium in a binary solid-supercritical two phase system, the fugacity of the solute in the solid phase is equal to that in the supercritical phase: fi S (P, T , x) = fiG (P, T , y)

(3.4)

Gas is marked as component 1 and non-volatile compound as component 2. Solubility of the gas in the solid phase is assumed to be negligible (x 2 = 1), therefore the fugacity of the solid in the solid phase is equal to the fugacity of pure solid (Prausnitz

60

3 Industrial Scale Applications: Physical-Based Processes

Fig. 3.7 Melting point depression curves for binary systems of fenofibrate and dense gases (Ljubec et al. 2018)

et al. 1998): S,pure

f2S (P, T , x) = f2

(P, T )

(3.5)

Fugacity of a pure compound is defined as: S,pure

RT ln

f2

(P, T ) = P

∫P ( v2 −

) RT dP P

(3.6)

0 S,pure

RT ln

f2

(P, T ) = P

) ) ∫ ( ∫P ( RT RT S v2 − v2 − dP+ dP P P P2S

0

(3.7)

P2S

In this equation P2S is the sublimation pressure at the process temperature, v2S is the molar volume of the pure solid component. In Eq. (3.8) the left side is the fugacity of the saturated vapour, which is equal to fugacity of saturated solid phase. Second term in Eq. (3.8) is the correction due to the compression of solid phase to pressure P.

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

∫P

S,pure

RT ln

fS (P, T ) = RT ln 2S + P P2

f2

v2S dP − RT ln P2S

61

P P2S

(3.8)

By rearranging Eq. (3.8) and inserting the expression for fugacity coefficient fS ϕ2S = P2S : 2

⎛ S,pure

f2

∫P

⎜ 1 (P, T ) = P2S ϕ2S exp⎝ RT

⎞ ⎟ v2S dP ⎠

(3.9)

P2S

where ϕ2S is the fugacity coefficient at T and P2S . Fugacity of the gas phase is: f2G (P, T , y) = ϕ2G y2 P

(3.10)

ϕ2G is the fugacity coefficient of solid component 2 in the system. By rearranging/ insert of Eqs. (3.9) and (3.10) into (3.5) and (3.4) the equation for solubility of solids in gas phase is obtained: ( P2S ϕ2S exp y2 =

1 RT

∫P

) v2S dP

P2S

(3.11)

Pϕ2G

Equation (3.11) for the solubility of solids in supercritical fluid can be expressed also in the form: y2 =

P2S E P

( ϕ2S exp E≡

1 RT

∫P P2S

ϕ2G

(3.12) ) v2S dP (3.13)

Factor E is the enhancement and is the correction of expression of the ideal-gas that is valid only at low pressures and contains three factors: • ϕ2S : takes into account non-ideality of the pure saturated vapour; in case the sublimation pressure of solid is low, P2S , the ϕ2S is practically equals unity. • Poynting correction (exponential term), which presents the effect of pressure on the fugacity of the pure solid; it is a small value at low pressures but could become higher at high pressures or at low temperatures.

62

3 Industrial Scale Applications: Physical-Based Processes

• ϕ2g : gas phase fugacity coefficient in the high pressure system of gas mixture with solid. Gas phase fugacity coefficient is the most important, because it is much lower than 1 and therefore enhancement factors can be very large (103 or higher). ( ) Supposing that the molar volume of the pure solid v2S at the system temperature is pressure independent, the Poyning correction takes the simple form:

E=

ϕ2S exp

(

v2S (P−P2S ) RT

) (3.14)

ϕ2G

ϕ2G can be calculated from an equation of state using following equation: 1 ln ϕi = RT

∫∞ [( V

∂P ∂ni

) T ,V,nj

] RT PV − d V − ln V nT RT

(3.15)

where ni is the molar number of species i in the mixture and V is the total volume of the system, v is the molar volume, R is the gas constant. For calculation of fugacity coefficients different equations of state (EOS) are applied. Cubic equations of state (EOS) based on attractive and repulsive forces in combination with mixing rules are currently the most widely used. Several types and/ or modifications of cubic EOS exist for van-der-Waals (Anderko 2000), RedlichKwong (RK) (1949), Soave–Redlich–Kwong (SRK) (Soave 1972), Peng-Robinson (PR) (1976). General form of cubic EOS is p=

a RT − V − b V 2 + ubV + wb2

(3.16)

using different values for u and w, which are presented in Table 3.2. Table 3.2 Cubic equations of state. https://doi.org/10. 1533/9780857090751.1.3

General form

RT − P = v−b (3.17)

Van-der-Waals (VDW: b = 0, c = 0)

P=

RT v−b

−

ac v2

Redlich-Kwong (RK: c = 0)

P=

RT v−b

−

ac Tc v(v+b)T 1/2

Redlich-Kwong-Soave (RKS: c = 0)

P=

RT v−b

−

a v(v+b)

Peng-Robinson (PR: c = b)

RT P = v−b − (3.21)

a v(v+b)+c(v−b)

(3.18) 1/2

(3.19)

(3.20)

a v(v+b)+b(v−b)

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

63

where: a = ac α ac = Ωa

(3.22)

R2 Tc2 Pc

(3.23)

)] 2 [ ( α = 1 + χ 1 − Tr1/2

(3.24)

χ = A0 + B0 ω + C0 ω2

(3.25)

b = Ωb

RTc Pc

(3.26)

The advantage of cubic EOS is that in engineering practice they are used to predict the thermodynamic properties like enthalpy, internal energy, entropy, etc.…, of fluids and describe phase behaviour of mixtures over large ranges of temperature and pressure. Thermodynamic models applied for prediction of phase behaviour and modeling of systems of supercritical fluid mixtures are reviewed by several authors (Anderko 1990; Anisimov and Sengers 2000; Dohrn 1994; Tuminello et al. 1995). Cubic EOS in combination with mixing rules are currently the most widely used models for the calculation of solubility of components in systems with supercritical fluids (Tables 3.1 and 3.2). In the equations (Tables 3.2 and 3.3) ω is the acentric factor and is a measure of acentric nature of intermolecular forces, parameters a and b reflect the contribution of attractive forces and molecular volume. If it is proposed to extend the use of a pure-fluid equation of state to mixtures, the EOS for the mixture is the same as for a hypothetical pure fluid (Cotterman et al. 1986) and characteristic constants a and b are dependent on composition. In this case, van-der-Waals one-fluid mixing rules with one or two adjustable parameters are used: ∑∑ a= yi yj aij (3.27) i

j

Table 3.3 Constants in cubic equations of state. https://doi.org/10.1533/9780857090751.1.3 Constant

VDW

RK

RKS

PR

Ωa

0.42188

0.42478

0.42747

0.45724

Ωb

0.125

0.08664

0.08664

0.0778

A0

–

–

0.48

0.37464

B0

–

–

1.574

1.54226

C0

–

–

−0.176

−0.26992

64

3 Industrial Scale Applications: Physical-Based Processes

b=

∑∑ i

yi yj bij

(3.28)

j

a reflects interactions between the species in the mixture and b accounts for the excluded volume of the species of the mixture. The coefficients aij and bij are related to the corresponding pure-component parameters by the rules: aij = bij =

√

( ) aii ajj 1 − kij

)( ) 1( bii + bjj 1 − cij 2

(3.29) (3.30)

where k ij and cij are the binary interaction and size parameters. If cij is set to zero, the co-volume b is expressed by a linear mixing rule. Equation (3.30) reduces to: b=

∑

yi bi

(3.31)

i

A linear mixing rule is used for the co-volume b (Anderko 1990) in most of the practical applications,. Several authors use a quadratic mixing rule for b in analogy with the mixing rule for a. Therefore, a second binary parameter is introduced, which is found to be practical for correlating gas–liquid equilibria in system with components of very different size of compounds (Anderko 1990). k ij and cij are in general in the same order of magnitude (Dohrn 1994) and both are expected to have an absolute value much less than 1 (McHugh and Krukonis 2013). Both of them could be positive or negative. If k ij has a negative value it mostly indicates that specific chemical interactions like hydrogen bonds between the components are present in the mixture (McHugh and Krukonis 2013). But it is less outward to interpret a negative value for cij . By rearranging Eq. (3.15) the expression for the fugacity coefficient of component i is gained, where for fugacity coefficient of component i PR EOS and van-der-Waals mixing rules is used: ( ) P(v − b) bN Pv − 1 − ln ln ϕi = b RT RT ⎛ N ⎞ ∑ ( √ ) 2 y a j ij ⎜ j ⎟ v+ 1+ 2 b b a N⎟ ⎜ ( − − √ ⎜ ⎟ · ln √ ) a b⎠ 2 2RT b ⎝ v+ 1− 2 b (3.32)

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

65

where: bN = 2

N ∑ k

yk bik −

N ∑ j

yj2 bjj − 2

N −1 ∑

N ∑

yi yi−j bij

(3.33)

j=1 i=j+1

and for binary system: bN = 2 · yi · bij + 2 · yj · bjj − yi2 · bii − yj2 · bjj − 2 · yi · yj · bij

(3.34)

The critical properties and acentric factors in the above equations can be estimated using group contribution methods. Molar volume is usually determined experimentally e.g. with pycnometer. The binary interaction and size parameters, k ij and cij are obtained by fitting the equation of state to experimentally determined phase-equilibrium data. Mixing rules that are in use in equations of state developed for pure fluids to mixtures are only applicable to mixtures which exhibit relatively moderate solution non-idealities. Problems, which endure in modelling phase equilibria in supercritical system mixtures with the use of cubic equations of state, where systems are highly nonideal due to high pressures involved, are (Prausnitz et al. 1998; Škerget et al. 2002): (a) the interaction parameters are usually temperature dependent; (b) reliable values of the physical parameters of substances in the system are not always available; (c) the equation does not fit the data equally well at a wide temperature and pressure range and especially near the critical point the deviation of the model is high. Most attempts to extend the range of equations of state (EOS) retain the vander-Waals separation of repulsive and attractive term. They have introduced some modifications to the attractive or repulsive term or both. Therefore, a number of empirical and theoretical models have been developed (Sadus 2012; Sandler and Orbey 2000) which permit extrapolation and prediction over wide ranges of temperature and pressure and can describe higher degrees of non-ideality. To eliminate the shortcomings of the van-der-Waals one-fluid model for a cubic EOS to provide additional composition dependence or density dependence by adding parameters to the combining rule for the parameter a, leaving the mixing rule for the parameter b unchanged, an empirical approach has been made (Sadus 2012; Sandler and Orbey 2000; Škerget et al. 2002). In addition, an approach on the density dependent mixing rules has been reviewed by Anderko (1990) and Danner and Gupte (1986). In general, there are several problems associated with this multiparameter combining rules that limit their use and are reviewed in literature. The main common difficulty is that they do not result in correct treatment of ternary and multi-component mixtures (de Loos 2006). An alternative method is based on the combination of the equations of state with activity-coefficient models (Huron and Vidal 1979) and Wong

66

3 Industrial Scale Applications: Physical-Based Processes

and Sandler mixing rules (Wong and Sandler 1992). They are mostly fairly reliable for the prediction of multicomponent phase equilibria from binary data, except for the most of strongly non-ideal systems. Three phase Solid/Liquid/Gas (SLG) for binary systems at equilibrium, the parameters needed in the Peng-Robinson equation of state to model SLG line are determined by solving following equations: f2G (T , P, y2 ) = f2L (T , P, x2 )

(3.35)

f2S (T , P) = f2G (T , P, y2 )

(3.36)

f1G (T , P, y1 ) = f1L (T , P, x1 )

(3.37)

Equation (3.35) transforms to Eq. (3.11) and Eqs. (3.37) and (3.35) become: y1 ϕ1G = x1 ϕ1L

(3.38)

y2 ϕ2G = x2 ϕ2L

(3.39)

Calculation of the Pure Solute Fugacity Fugacity of a solute in a solid phase cannot be directly calculated by a conventional equation of state, instead it is calculated by means of a subcooled liquid reference state. Fugacity of the subcooled liquid at temperature T in terms of measurable thermodynamic properties can be therefore expressed by (Prausnitz et al. 1998): fus (

L,pure

ln

f2

S,pure

f2

=

ΔH2 RTt,2

) ( ) Δcp,2 Tt,2 Δcp,2 Tt,2 Tt,2 −1 − −1 + ln T R T R T

(3.40)

fus

where T t,2 is the triple-point temperature, ΔH2 the enthalpy of fusion for component 2 at temperature T. Usually only the first term is considered in the above equation, especially, if T and T t are not far apart and to neglect the contribution in cp . Because there is a small difference between the triple-point temperature and the normal melting temperature, it is common to substitute T t for its melting point temperature at normal conditions Tfus,i . Simplified equation, valid at the triple point pressure Pt of the solute is therefore: L,pure

ln

f2

S,pure

f2

=

) fus ( Tfus,2 ΔH2 −1 RTfus,2 T

The effect of pressure, to the fugacities can be written as:

(3.41)

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

⎛ 1 S,pure S,pure f2 (P, T ) = f2 (Pt , T ) exp⎝ RT

∫P

67

⎞ v2S dP ⎠

(3.42)

Pt

⎛ L,pure f2 (P, T )

=

L,pure f2 (Pt , T ) exp⎝

1 RT

∫P

⎞ v2L dP ⎠

(3.43)

⎞ ( S ) v2 − v2L dP ⎠

(3.44)

Pt

Therefore the fugacity ratio is: S,pure

(P, T )

S,pure

⎛ (Pt , T )

1 exp⎝ = L,pure L,pure RT f2 f2 (P, T ) (Pt , T )

f2

f2

∫P Pt

Combination of Eqs. (3.40) and (3.43) results in: ⎛ S,pure

f2

L,pure

(P, T ) = f2

(P, T ) exp⎝

1 RT

∫P

(

⎞ ) fus ( ) T ΔH2 fus,2 ⎠ 1− v2S − v2L dP + RTfus,2 T

Pt

(3.45) and by integration of Eq. (2.44) the final equation describing the fugacity is: S,pure f2 (P, T )

=

L,pure f2 (P, T ) exp

(( ) )) fus ( v2S − v2L (P − Pt ) Tfus,2 ΔH2 1− + RT RTfus,2 T (3.46)

3.1.3 Liquid-Supercritical Fluid Equilibrium 3.1.3.1

Phase Diagrams

Six main types of fluid-phase behaviour in binary systems are esteemed by their critical properties; their P–T projections are shown in Fig. 3.8 (Konynenburg et al. 1980). Type I and type II—similar behaviour is observed. LG critical line is continuous and connects the C1 and C2 critical points of pure components. Difference between both types can be observed because liquid mixtures in type II are not miscible in all proportions and display a LLG three phase line at low temperature. At UCEP two liquid phases merge into one liquid phase. Then LL critical line growths rapidly to higher pressures.

68

3 Industrial Scale Applications: Physical-Based Processes

Fig. 3.8 Classification of the phase behaviour of binary fluid systems. Where: C, critical point; TP, triple point; L, liquid; G, gas; S, solid; UCEP, upper critical end point; LCEP, lower critical end point. (----) critical lines, line 1 and 2 vapour pressure curves for two components (Brunner 1994)

Type III—LG line is not continuous it has two branches. The critical line starting from C2 with a positive slope shows the presence of the so-called gas–gas equilibria. The two phases are in equilibrium at a temperature higher than the critical temperature of either pure component. The three-phase LLG line can be seen near the vapour pressure line of the most volatile compound (Sadus 2012). Type IV—the critical line among the critical points of pure compounds is broken by a three-phase LLG line. One branch of the critical line goes from the critical point of the component with the higher critical temperature to the LCEP and other branch goes from C1 to the UCEP. Type V—similar to type IV excluding an extra LL critical line at low temperatures which ends at another UCEP. For this type of mixture, two regions of limited liquid miscibility at lower and higher temperatures and pressures are possible. Type VI—system including two critical curves. One is a continuous LG critical line, which connects the critical points of the pure compounds, and another an LL critical curve, which connects the UCEP and LCEP.

3.1.3.2

Thermodynamic Modelling of High-Pressure V-L Equilibria

Consider binary liquid in equilibrium with a gas phase, the equilibrium equation for a binary liquid–gas system at temperature T and pressure P is expressed as:

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

69

f1G (T , P, y1 ) = f1L (T , P, x1 )

(3.46)

f2G (T , P, y2 ) = f2L (T , P, x2 )

(3.47)

where x i is mole fraction of component i in the liquid phase and yi is mole fraction of component i in gas phase. The fugacity in each phase can be written as: fi L = xi ϕiL P

(3.48)

fi G = yi ϕiG P

(3.49)

where ϕiG and ϕiL are the fugacity coefficients of i in gas and liquid phase, respectively. y1 ϕ1V = x1 ϕ1L

(3.50)

y2 ϕ2V = x2 ϕ2L

(3.51)

Fugacity coefficients for each component in the liquid and gas phase can be calculated by using Eq. (3.15).

3.1.4 Cycle Processes for Extraction Using Supercritical Fluids A typical high-pressure extraction process principally includes a separation stage for the feedstock (extractor and separator) and a regeneration stage for the solvent. Firstly, in the separation stage, the components to be extracted become concentrated in the gas and are then precipitated in the separator by applying appropriate methods. Afterwards the gas must be eliminated from the extract and the raffinate and cleaned for reuse in the extraction process. If a solvent mixture is applied in the extraction process, the composition of the mixture must be adjusted before reuse. The solvent recovery can be achieved in different ways. The appropriate procedure of solvent recovery depends on the nature of substances, the scale of the process unit and the operating conditions (Brunner 2013). The separation of extract from the solvent can be achieved by several ways like (Brunner 2013): – Isenthalpic throttling (expansion) to subcritical conditions by changing pressure and temperature, – Varying the temperature and keeping supercritical conditions, or cooling down to subcritical conditions, – Using an additional mass separating agent e.g. absorbing medium, membrane adsorbents), while maintaining supercritical conditions for the solvent. The proper method and operating conditions in the separator are selected according to the phase equilibrium data. Overall, a variation in temperature will not

70

3 Industrial Scale Applications: Physical-Based Processes

be effective in cleaning the solvent appropriately for reuse in the extraction process. Method in temperature changing can be applied in addition, if a total regeneration of the solvent is not necessary (Catchpole et al. 2009). The benefit of using an absorbent or adsorbent for separation of the extract and the solvent is that the separation can be achieved without drastically reducing the pressure. Due to operating at constant pressure the energy consumption is lower as only the pressure drop in the extraction plant has to be compensated. The disadvantage of that kind of separation is that the extract in the absorbent/adsorbent have to be processed subsequently. This may be difficult and should be considered when designing the process and calculating its operating costs. The solvent in the subcritical (liquid) or supercritical state can be delivered by a compressor or by a pump. T–S diagrams are a widely applied means to present solvent cycles with corresponding changes of conditions of state. These diagrams are used for calculating the heat balance of each step of the extraction process (Knez et al. 2013).

3.1.4.1

Pump Process

The energy for the extraction process using a pump could be evaluated from temperature–entropy (T–S) diagram shown in Fig. 3.9. It is presented with areas of homogeneous liquid, gas, supercritical fluid and two-phase liquid–gas region (L–G). The dotted line presents no border line, however, it visibly separates the area of supercritical fluid from the gas and liquid state areas. The extraction process is performed under constant conditions in steady state (3 in Fig. 3.9). The supercritical fluid phase leaves the extraction unit and the dissolved substance is afterwards separated from the solvent by changing the pressure and temperature (isenthalpic throttling 3–4). The two-phase region of the solvent is obtained, and the Fig. 3.9 T –S diagram of gas extraction process: solvent circuit in the pump mode (Brunner 1994)

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

71

substance dissolved in the solvent precipitates to a state which can be removed (solid or liquid). The gaseous part of the solvent is cooled and condensed to a liquid state (6) and then sub-cooled (1). To eliminate the substances that remain dissolved, the liquid solvent can be evaporated (6–5), afterwards liquefied (5–6) and sub-cooled (1). After that the sub-cooled liquid is pumped to the working pressure (1–2) and heated to the operating temperature (2–3). The area enclosed by the solvent circuit in the T –S diagram characterizes the thermodynamic work needed for the process of cycling the supercritical fluid. The lower is the pressure drop for separating the extract and the solvent, the lower work will be required (Knez et al. 2013).

3.1.4.2

Compressor Process

The extraction process using a compressor is schematically shown in the T–S diagram in Fig. 3.10. The extraction process is performed under constant conditions (3), then the solute is removed from the supercritical fluid (isenthalpic throttling) into the subcritical region (3–4) then the evaporation is followed (4–5–1). The dissolved substances precipitate. The solvent, which is in a gaseous state, is then compressed to the pressure of the extraction, shown in Fig. 3.10 as an idealized, isentropic compression (1–2). During compression, the temperature of the supercritical fluid increases, thus the solvent is cooled to the temperature of extraction (2–3). This finishes the cycle and may be repeated. The balance for real processes should be calculated and all mass and heat losses need to be considered (Knez et al. 2013). Fig. 3.10 T –S diagram of gas extraction process: solvent circuit in the compressor mode (Brunner 1994)

72

3 Industrial Scale Applications: Physical-Based Processes

3.1.5 Separation of Solute in Extraction Processes Using SCF The dissolved substances in the supercritical fluid must be separated from the solute– supercritical fluid mixture. Numerous separation processes may be used, but the ideal option is to use processes in which no further substances are introduced into the system. If additional substances are needed to separate the solute from the supercritical fluid that increases the costs of separation. While re-compression and cooling conditioning of the supercritical fluid to operating extraction conditions increases the energy costs of the process. As it was already mentioned one of main advantages of using supercritical fluid for extraction is the relatively easy fractionation of substances. Fractionation of substances from extracts can be achieved by a stepwise reduction of solvent power in mostly static equipment.

3.1.5.1

Separation of Solute by Reduced Solvent Power

Reduction of the solvent power is the most commonly used process for the separation. A condensed phase is formed, and the gaseous phase is removed. The solvent power of supercritical solvents depends upon pressure and temperature which effect the density of the supercritical fluid. Commonly, the solubility of a solute increases with increasing density and decreases with decreasing density. Density can be decreased by decreasing pressure and/or increasing temperature. In various industrial processes, the concentration of solute in a solvent is reduced by decreasing the pressure. Consequently, the density decreases, while an increase in temperature increases the vapour pressure of the solute. In a region of low pressure, the density decreases intensely with a temperature increase and the concentration of solute in supercritical fluid decreases with increasing temperature when the vapour pressures of the substances are relatively low. When the vapour pressure of a solute is high and increases with a rise in temperature, the separation of a solute with temperature increase is not an appropriate separation method. One example is the separation of essential oils from supercritical fluids. At high pressure the density decreases slightly with an increase in temperature, which will lead to a higher concentration of solute with increasing temperature (at constant pressure) in the supercritical fluid. Based on these observations, it is apparent that the increase in temperature, which decreases the density and solvent power, will lead to efficient separation of a solute with low or moderate vapour pressure within the low-pressure range (Knez et al. 2013). For some substances, the solubility decreases with increasing pressure. In these cases, separation procedures of solutes by pressure reduction or reduction of the solvent power are impossible. It is obvious that an efficient separation process can only be designed if for such a system phase equilibrium data exist. After implementing thermodynamic conditions for reducing the solubility in a separator, a condensed phase is formed. Which is then separated from the solvent with lower

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

73

density in one, or several separators in series. In the last separator separation takes place from the gaseous phase. Separation in series will operate at different pressures and/or temperatures in purpose to fractionate certain components of extracts. The solute–solvent system should stay in the separator long enough time to reach the phase equilibrium.

3.1.5.2

Separation by Expansion

The solute-loaded solvent phase is expanded into the separator where the solvent is below supercritical conditions. At this point, both the gaseous and the liquid solvent containing the solute are present in the separator. If the solute is unsolvable or less soluble in the liquid and gaseous phases a three-phase mixture is present. Generally, the liquid solvent evaporates and afterwards the solute can be removed from the separator. Then the gaseous phase can be recycled. This kind of separation process is appropriate for the separation of a solute from solute-loaded supercritical fluid, but operation costs are very high due to high energy input. Auxiliary devices are used to escalate the separation of the solid or liquid phase from the low-density solvent. These kinds of systems include demisters consisting of wire mesh packing, deflectors and filters, and cyclones (Knez et al. 2013).

3.1.5.3

Separation of Solute and Solvent by a Mass Separating Agent

The separation of solute by a mass separating agent can be done by: • • • •

absorption, adsorption, use of membranes, or adding a substance of low solvent power.

For separation by absorption, the solvent circuit may run almost at constant pressure. It is important that the absorbing liquid dissolves the solute and does not absorb the solvent. Separation by adsorption can be a very effective process and can operate with almost no pressure and/or temperature drop. Consequently, both separation processes have a major impact on the economy of entire extraction process. For the separation of solute from a gas phase—due to the difference in molecular mass—a membrane separation processes can be used. In that case the pressure drop through the membrane is relatively low, so the solvent regeneration costs are minor. Reducing the solvent power of a supercritical fluid may further be achieved by adding a substance of low solvent power (e.g. adding nitrogen to supercritical CO2 for several substances). A similar effect may be found, if an entrainer was applied in the extraction process and separated from the supercritical solution by adsorption or absorption afterwards. Such a separation process operates at an almost constant pressure, so operating costs are minimal (Knez et al. 2013).

74

3 Industrial Scale Applications: Physical-Based Processes

3.1.6 Basic Design Considerations of Extraction Plants for Solids For successful design of a supercritical extraction process and the subsequent engineering of a commercial plant the mass and energy balance of the process has to be elaborated. Basically the following parameters should be determined: – – – –

basic data specific for the product thermodynamic conditions for operation of extraction and separation process mass transfer data for the system energy consumption for single process steps (incl. T –S diagram).

For exact sizing of the extractors specific basic data and thermodynamic conditions should be defined. Mass transfer data regulate the time solids remain in contact with the supercritical solvent. Since most processes apply the extraction from solid materials and require a batch-wise operation of the extractor, also the time for changing the batch needs to be considered carefully and has a great influence on the overall economics. Quick batch-change and minimal down-time without compromising plant and operator safety are the main targets here. For the sizing of the separator(s) favourable velocities and residence times need to be taken into account to make sure the desired separation result is obtained. The heating and cooling consumption and the capacity of the pump/compressor and heat exchangers area and piping system need to be determined for the whole solvent circulating system. Specific data and pre-treatment of solids: For the design of an industrial plant following data are crucial: • • • •

Nature of raw material Final product requirement Desired plant capacity Plant location (usually local conditions, codes & standards prevail).

The nature of the raw materials effects the quality of the extract or raffinate obtained (depending on which part will be the desired product) and also the complete economy of the process. In case the raw material is a fluffy matter, compacting or pelletizing may be an important step prior to extraction. On the other hand, it may be necessary to break larger bulky matter into smaller particles and obtain a more uniform size distribution to achieve similar mass transfer conditions within the bed to be extracted. It should also be investigated whether the raw material is contaminated. The desired purity of the extract may not be achievable at all or some waste of product must be accepted. Further, in case the raw material contains a low concentration of desirable constituents only, the question is how economically feasible the process will be. The final product specification is generally based on customer and market requirements. Prospects and specifications have a significant influence on the costs of the extraction process. The maximum yield and the highest selectivity at accurate process

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

75

conditions to separate the required substance at lowest separation costs have to be determined. In various cases a multistep separation is appropriate for the fractionation of extracts. It is known that the extractor volume effects the investment costs for an extraction unit. Materials which might be extracted with supercritical fluids can be separated into two categories: 1. Raw materials where only undesirable substances will be removed, and the form of the raw material is maintained during the process. Examples for this group of materials are decaffeination of coffee or tea (although the natural caffeine is a valuable by-product), some seed plant materials and the separation of pesticides, from rice or ginseng. In this case of separation, a high selectivity of undesirable compounds is necessary. Thus the process parameters have to be chosen carefully. Such processes are used for high-volume or relatively high-value market products. 2. Raw materials where pre-treatment is permitted and extracted substances are the main product. Any pre-treatment of material can be used to reach a high yield at a low solvent to feed ratio with a low energy consumption. Usually the raw material is ground to increase the bulk density. Bulk density of non-ground material is usually between 150 and 250 kg/m3 , while ground materials reach a density around 350–500 kg/m3 . It is known that mass flux through smaller particles is higher than through the bigger ones. Nevertheless, the linear solvent velocity through the ground raw material is reduced and therefore a greater pressure drop is obtained. Normally the linear velocity through material is chosen from 0.7 to 0.8 cm/s. Too finely ground material may cause clogging, so if the bulk density is less than 250 kg/m3 the material should be formed into pellets. Generally, the bulk density has a major impact on the economics of the process: if the bulk density is low, less material will be introduced into the extraction vessel and the yield per batch is reduced. The quantity of water in the plant material effects on the economy of the process as well. Usually the moisture content of raw plant materials is between 8 to 15 wt%. If the water content is higher polar substances are extracted at relatively low pressure. If the water amount in the raw plant material is too low, the cells may shrink and delay the mass transfer of extracted substances. However, the ideal moisture content of plant material for extraction process should be optimized during laboratory-scale tests. The decaffeination of coffee and tea, the water content of green coffee beans should range from 35 to 45 wt% (Zosel 1981). The water amount in the extraction of astaxanthin from algae should be as low as possible while water–oil emulsions are formed so the yield of astaxanthin would therefore be also very low. The content of oils, essential oils and waxes has also a large influence on the extraction yield. These ingredients contained in raw plant material may act as entrainers for the extraction of valuable compounds. This could be observed for extraction of carotenoids from ground paprika. Carotenoids have low solubility in pure CO2 even at ultra-high pressure (1,500 bar), and could not be separated quantitatively from plant material.

76

3 Industrial Scale Applications: Physical-Based Processes

Thus, in some processes, oils, essential oils and waxes are used as entrainers for extraction of valuable compounds (as colorants, antioxidants etc.) (Knez et al. 2013). Thermodynamic Data The solubility of desired extracted components in a supercritical solvent is also crucial to the economics of the extraction process. It is desirable that the highest possible loading of supercritical solvent should be achieved. In the separation stage, the solubility of solute in solvent should be as low as possible. In some cases, the solvent power for solutes should be slowly reduced and the fractionated separation of solute from the solvent may be achieved. Mass Transfer Mass transfer data are very important for determining the extraction time and the capacities of pumps, compressors, and heat exchangers. Mass transfer data should be optimised experimentally. To the extracted raw plant material humidity, particle size, and particle size distribution have to be defined. The extraction pressure and optimal extraction temperature obtained from thermodynamic investigations for a defined SC solvent need to be determined. Based on mass transfer experiments, the mass of solvent can be determined for the highest possible extraction yield. The most common extraction curve for the isolation of a substance from solids is presented in Fig. 3.11. The curve for the isolation of substances from solids is normally divided into a period of constant extraction rate and a period of falling extraction rate (Sovová 2005). In the initial period of the process solubility controlled mass transfer prevails. At higher S/F or at longer extraction time, the mass transfer is controlled by diffusion. The diffusion and hydrodynamics influence the mass transfer rates. In order to optimize the overall process economically it may be advised not to operate to the maximum yield and stop the extraction at a viable S/F ratio instead.

Fig. 3.11 Typical extraction curve (Brunner 1994)

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

77

Diffusion The desirable compound to be extracted may be placed in the cell depth or adsorbed on the surface of a solid matrix. Consequently, the mass transfer depends on the location of the compound. If it is adsorbed on the surface, the mass transfer rates are high. In case the solid particles of the substances are in the cell depth, they have to diffuse through cell walls, therefore the mass transfer rates tend to be low. In some examples, the substances to be extracted also form complexes which necessitate release by a chemical reaction (usually hydrolysis with water). For substances that do not form complexes, the diffusion may be influenced by particle size reduction (reduction of diffusion path) or cells destruction (by swelling or cracking cells, by ultrasound, by milling procedure) (Knez et al. 2013). Hydrodynamics Inappropriate particle size, particle form and particle size distribution of the raw material may cause channelling. In that case the even flow of the supercritical fluid solvent through the extracted material is reduced. Also, the swelling of the material during the extraction may reduce the flow of supercritical solvent. Likewise, the direction of the flow of the supercritical fluid is in some cases important. For example, in industrial scale processes, the flow is normally directed from bottom to top, while under some process conditions in the opposite way. Energy Consumption Energy consumption for extraction process using supercritical solvents can be determined from the energy balance and visualized in a T–S diagram as described in Sect. 1.4.

3.1.7 Supercritical Fluids Extraction at Ultra-High Pressure Industrial production is continuously driven by ongoing improvements looking for the most economical solution of producing the goods. This driving force is founded on steady development of new processes or production techniques, economies of scale as well as optimization of the energy demand during production. Some examples can be given from the automotive industry, chemical and petrochemical production as well as from mechanical engineering and food processing. Comparing these trends and developments in various industries with R&D (research&development) efforts as well as industrial applications of SFE we see some differences. The early industrial application of SFE (decaffeination of coffee, hops extraction) was a big step forward using approx. 300 bar processes in relatively large scale up to 20 m3 . Thereafter, we do not find too many new processes using

78

3 Industrial Scale Applications: Physical-Based Processes

higher pressures or larger scale. R&D often focusses on process development at socalled high pressure meaning 80 bar to 150 bar instead of really challenging technical possibilities. From the literature (Giddings et al. 1969, 1970; Stahl et al. 1988) it is well known that the solubility of substances increases with increasing pressure at constant temperature until solubility maximum and afterwards with increasing pressure the solubility decreases. Therefore, with increasing pressure and in most cases with increasing temperature, the total yield increases, but the selectivity for a certain substance decreases. Such substances are for example soya oil and jojoba triglycerides, where the solubility decreases at certain temperature already after pressure of 600 bar (Stahl et al. 1987). Several patents could be found where higher pressure is used for separation of substances from plant material. In patent (Forchhammer et al. 2006), WO 2006/05537 A1)) a pressure for separation of caffeine from tea is described. Applied pressures were up to 1,000 bar and temperature up to 80 °C. Isolation of Xanthohumol rich hop extracts at 60 °C is described in patents (Forchhammer et al. 2006; Forster et al. 2004) (WO 2006/05537 A1., EP 1 424 385 B1). For industrial application pressures above 500 bar are unusual and most of the industrial plants operate at pressures of 300 to 500 bar (Bork 2000; Gehrig and Schulmeyr 1997; Sebald et al. 1995) (DE 195 24 481 C2., DE 4400 0 96 C2., DE 198 54 807 A1.). In published literature some articles could be found where ultra-high pressure technique was employed to extract ginsenosides from roots of ginseng (Panax ginseng) (Chen et al. 2009), for pectin extraction from navel orange peel (Citrus sinensis) (Guo et al. 2012) and extraction of catechins from green tea leaves (Jun et al. 2010). In these experiments water or its solutions with some organic solvent were used at hydrostatic pressures up to 6000 bar at temperatures up to 60 °C. These processes were performed in lab scale, were not operated in continuous mode and not enable separation of pure solute from the raw material. Ultra-high pressure extraction using SC CO2 can overcome the drawbacks of hydrostatic processes. In case of a relatively low solubility of substances in extraction process operating at a pressure range up to 500 bar, entrainers are added to CO2 , to enhance the solubility of such substances. The disadvantage using co-solvents is a potential contamination of extracts with organic solvents. Therefore, expensive separation processes have to be applied for separation of residual solvents, which could not be separated below certain threshold limits. Based on the review of literature data on solubility trends in supercritical fluids and own experiments for practical determination of solubility in different dense fluids up to pressures of 1000 bar (please see Fig. 3.12), a multi-purpose extraction plant being suitable for extraction up to 2500 bar was built to evaluate the benefits of “real” high pressure (Luetge et al. 2009, 2007). Ultra-high pressure extraction experiments with CO2 as solvent for many different raw materials were performed and the advantages are presented in the following subchapters.

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

79

Fig. 3.12 Solubility of concentrated rosemary extract versus pressure and temperature

3.1.7.1

Ultra High Pressure Extraction Equipment

The ultra-high-pressure CO2 extraction apparatus shown in Fig. 3.13 was used for the extraction experiments. Unit

Parameters

Extractor 1

Volume: 0.64L Max. op. press.: 2800 bar Max. op. temp.: 90°C

Extractor 2

Volume: 2.0L Max. op. press.: 2800 bar Max. op. temp.: 90°C

Separator 1

Volume: 0.64L Max. op. press.: 2800 bar Max. op. temp.: 90°C

Separator 2

Volume: 1.0L Max. op. press.: 1000 bar Max. op. temp.: 120°C

Flow rate of CO2

Max 16 kg CO2/h

Fig. 3.13 Ultra high-pressure CO2 extraction apparatus and technical characteristics of the unit

80

3 Industrial Scale Applications: Physical-Based Processes

Fig. 3.14 Total yield of rosemary extraction and composition versus pressure at 60 °C and S/F = 40 kg/kg

The laboratory plant can be operated with two step cascade separation or single stage separation when only one separator is in operation. The direction of the flow of the supercritical solvent through the fixed bed in the extractor can be upwards or downwards. At the bottom of the extractor of ultra-high-pressure CO2 extraction apparatus an inlet of capillary tube which was connected to a HPLC pump (Spectra- Physics Spectra SYSTEM P1500) to introduce co-solvents to the flow of CO2 . In some cases, co-solvents (in most cases water) are added, even ultra-high pressure is applied, due to cleavage of bonds in substances containing alkaloids.

3.1.7.2

Examples of Ultra-High Pressure Extraction of Different Materials

Rosemary (Rosmarinus Officinalis) In Fig. 3.14 the effect of extraction pressure to the yield and composition for rosemary extraction with CO2 is shown. The carnosolic acid content of the extract can be varied to a high degree when changing the pressure between 1000 and 1500 bar. These phenomena can be used to adjust the product composition in situ during the extraction processing step. Generally, we could conclude that the extraction of rosemary with CO2 at temperature 80 °C gave the highest yields and total yield on carnosolic acid and carnosol at highest applied pressure (1500 bar) and high S/F (mCO2 /m material), but the total

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

81

Fig. 3.15 Yield versus solvent/feed ratio for extraction of nut oil

yield (at S/F = 40) of carnosolic acid is lower than at temperature 60 °C and higher than at temperature 100 °C. With increasing S/F the total yield on carnosolic acid was the highest at S/F = 60 and a pressure of 1500 bar at temperature 80 °C. At a temperature of 100 °C extraction yields increase with increasing pressure at constant ratio S/F of 40 kg/kg from 500 bar up to 1000 bar, the content of carnosolic acid decreases. With further increase of pressure from 1000 bar up to 1500 bar the total yield decreases but the content of carnosolic acid in extract increases. The ratio between carnosolic acid/carnosol does not change with pressure, which confirms that no oxidation of products of carnosolic acid during ultra-high pressure extraction process occurs.

Nut Kernels (Juglans Regia) Sliced nut kernels (Juglans regia) with oil content of about 58% (w/w) was filled into the extractor and extracted at temperature 40 °C and pressures 300 bar, 500 bar, 1000 bar and 1500 bar. From results on Fig. 3.15 it is evident that for the same yield the solvent/feed ratio is decreased with increasing pressure. For an extraction yield of e.g. 50% the process can be optimized by reducing the extraction time sharply which is indicated by the S/F ratio. At 300 bar S/F should be ca. 55, whereas at a pressure of 1500 bar the S/F value reduces to ca. 10.

Chamomile (Matricaria Chamomilla) Extracts of chamomile (Matricaria chamomilla) are used in medicine and cosmetics due to their anti-inflammatory and anti-spasmolitic properties.

82

3 Industrial Scale Applications: Physical-Based Processes

Main active ingredients are sesquiterpenelacton Matricin, sesquipenterpene alcohols like (-)-α- Bisabolol, Bisabolol oxide A and Bisabolol oxide B and En-InDicycloether. All these substances are very heat sensitive and are decomposed already at a temperature of 80 °C. During the experiments it was observed that with increasing pressure at constant temperature the solubility of substances, which could be extracted from chamomile flowers increases dramatically with pressure. Therefore, the total extraction yield and composition of extracts is mostly dependent on pressure. Experiments of extraction were performed at 1500 bar and temperature 40 °C. Fractionated separation of extracts was done in 4 consecutive separators operated at temperature 40 °C and at pressures from 1300 bar to 50 bar. The process parameters were determined in such manner, that extremely efficient separation of waxes, water, fragrant components and sesquiterpenelacton Matricin, sesquipenterpene alcohols like (-)-α- Bisabolol, Bisabolol oxide A and Bisabolol oxide B and En-In-Dicycloether were successfully obtained with huge economic benefits.

Cocoa (Theobroma Cacao) Cocoa (Theobroma cacao) powder with fat content of 10–12% (w/w) was extracted at 1500 bar and 40 °C solvent/feed ratio of 40 kg/kg. From the cocoa powder fat was removed and the theobromine content was reduced for 50%. Compared to extraction process performed at the same temperature but much lower extraction pressure of 480 bar the S/F increased to 120.

Extraction of Pigments A very intensive research was done in the field of ingredients from algae biomass. Fundamental research was transferred to pilot plant tests and later realized in an industrial scale production process as presented in Sect. 3.1.8.1. Pigments were also extracted from agricultural waste products form tomato industry. Lycopene was extracted in extreme yield at pressure of 1500 bar and temperature 80 °C at relatively low S/F. Using two step separation process very concentrated lycopene was collected in the first separator. Carotenoids were extracted from Krill (Euphausia superba) meal with SC CO2 at pressures between 200 bar and 1000 bar at temperatures 60 °C and 80 °C. In these experiments also lipids and phospholipids were co-extracted. It was found that with increasing pressure at constant temperature and S/F the total extraction yields as well the yield of carotenoids highly increase with increasing pressure.

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

3.1.7.3

83

Conclusions

Many active ingredients of plant material and other natural resources cannot be extracted using SFE at a pressure below 500 bar or could be extracted at very low yield at reasonable S/F ratios. In these cases, the process is uneconomic and cannot be realised on industrial scale. Either co-solvent is needed with the additional task to remove the co-solvent from the product. Therefore, more process development is needed and a first pilot plant capable of using pressures up to 2500 bar is installed and operational. Economy of the process was studied in details also on other products like coffee decaffeination process, extraction of valuable substances from saw palmetto berries, isolation of piperine from pepper, isolation of xantohumol from hop and many other plant or animal raw materials. The benefit of using ultra-high pressure in extraction processes is fractionation of extracted substances by pressure and/or temperature dependent precipitation in a multi-step separation and/or by chromatographic methods incorporated in the extraction plant. Based on the performed research in domain of ultra-high pressure extraction patent application was made (Bork and Luetge 2008). The economy of the process shows decrease of specific processing costs with increased operating pressure of apparatus.

3.1.8 Extraction of Solids Using SCF—Industrial Scale Units The application of extraction of solids using supercritical fluids is numerous. In the literature several overviews can be found (Brunner 1994; Eltringham and Catchpole 2007; Fang et al. 2007; Gardner 1993; Lack and Seidlitz 1993; Lack and Simandi 2001; Mendes 2007; Moyler 1993; Mukhopadhyay 2007b; Stahl et al. 1987; Temelli 2009; Temelli et al. 2007). On internet pages of process design companies and equipment producers (Uhde HPT, Natex, Sitec, Nova Swiss, etc.) references to their plants are given. From these data it can be seen that the highest capacities are installed for coffee and tea decaffeination. Actually the large scale decaffeination of coffee was the very first commercial application realized in Germany in the 1980s. From this initiating point many other technical applications were developed such as the second largest application which is the extraction of hops for brewing bear with standardized ingredients obtaining constant quality. Extraction of spices for production of oleoresins and extraction of bioactive compounds from plants are very widely used applications of SC fluids for extraction of solids. Latest applications are a plant for extraction of oil from degumming residue to obtain highly concentrated and very pure lecithin and applications for cleaning of natural cork or sensitive industrial parts.

84

3 Industrial Scale Applications: Physical-Based Processes

Fig. 3.16 Structural formula of astaxanthin

3.1.8.1

Extraction of Astaxanthin for Algae Extraction Using Supercritical CO2

Lab Scale Experiments As known from the literature the microalgae Haematococcus pluvialis can be forced to produce astaxanthin, a valuable ingredient for the food supplement market. It is a dietary antioxidant shown to support and maintain the body’s natural inflammatory response, to enhance skin, and to support eye and joint health (Fig. 3.16). Microalgae can be cultivated and grown in open basins with forced flow to get access of the sunlight to all algae biomass. Harvested algae biomass is yielded stress to produce the targeted astaxanthin. Cells have to be cracked to get access to the carotenoids by the solvent CO2 . Finally, the product is dried to produce flakes or a powder. There are some data in the literature on high pressure extraction of algae, but they present too few data for design of an industrial plant (Moyler 1993). Based on observation of phase behaviour of CO2 —algae extracts, fractionated two step separation should be very efficient to obtain products with higher astaxanthin content. For design of fractionated separation, the solubility of algae oil in CO2 was measured. Results are shown in Fig. 3.17. It could be concluded that the solubility of substances increases with increasing pressure at constant temperature. At temperature 40 °C and pressure 300 bar the solubility is 6x higher than at 100 bar. At pressures over 300 bar the selectivity for the system oil/carotenoids/CO2 decreases and therefore the results for the pressures over 300 bar does not give the values for the solubility of pure oil in CO2 . The extraction experiments were performed at a pressure higher than 700 bar. The algae oil is separated in two separators with different operating pressures to allow a fractionation. The gained extracts in the two separators differ mainly in the astaxanthin concentration, the higher the pressure in the first separator the higher is the astaxanthin concentration in the oil of this separator.

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

85

Fig. 3.17 Solubility of algae oil in CO2 at 40 °C

Design of Industrial Scale Plant Based on the results of lab and pilot scale experiments the production plant was designed with two alternating operated extraction vessels as could be seen from Fig. 3.18.

3.1.8.2

Extraction of Pine Nut Press Cake

Lab Scale Experiments Pine nut press cake and pine nut oil is used in food and cosmetic industry. The oil is rich in pinolenic acid and antioxidants which may have some health benefits. The extractions were performed at a pressure between 350 and 450 bar, different solvent to feed ratios, different direction of CO2 flow, as well as the influence of extraction time were investigated. The oil was collected stepwise in two sedimentation separators. By tuning the operating conditions in the separators the composition of the extract was varied.

Design of Industrial Scale Plant Based on lab and pilot scale experiments industrial scale plant was designed. The pictures of the plant are in Fig. 3.19.

86

3 Industrial Scale Applications: Physical-Based Processes

Fig. 3.18 Production plant with two alternating operated extraction vessels for algae extraction (photo: courtesy Uhde High Pressure Technologies, Hagen, Germany)

Fig. 3.19 Industrial plant for CO2 extraction of pine press cake (photo: courtesy Uhde High Pressure Technologies, Hagen, Germany)

3.1.8.3

Spray Extraction of Highly Viscous Liquid/SCF—Deoiling of Degumming Residue

Lecithin is a natural emulsifier which is found in high concentrations in egg yolk and soy beans. It is used as a neutraceutical, as emulsifying agent in the food industry and as a source for phosphatidylcholine (PC) in the pharmaceutical industry. Mainly

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

87

used in the food industry is soy lecithin, while egg yolk lecithin is used to recover PC, because the PC-content in egg lecithin is higher than in soy lecithin. Lecithin is not a single substance but a mixture of different phospholipids: phosphatidylcholine (PC), phosphatidylethanolamin (PE), phosphatidylinositol (PI), phosphatoic acid (PA) and others. Soy lecithin is a by-product of the soy bean oil production. Raw soy bean oil is usually extracted by hexane. In the first refining step gummy substances are removed by hot water and collected as de-gumming residue. Further refining steps such as hydrolysis, bleaching and others lead to the fine oil. De-gumming residue is dried giving raw lecithin as the product. Different lecithin products are produced. The first is the liquid lecithin, which is a highly viscous fluid with a typical content of 62% of acetone insoluble matter (AIM), the parameter for the purity of lecithin. This liquid lecithin is conventionally de-oiled by acetone to form a pure lecithin in powder or granular shape with AIM-content of approx. 97%. To overcome the regulatory problems from the use of acetone, residues in the de-oiled lecithin and the extracted oil need thorough solvent recovery; furthermore, plants have to be explosion-proof. Supercritical CO2 processing of de-gumming residue could be an alternative. The demand of green products requires the use of a green solvent-like CO2 . All these processes developments led, up to now, to processes which were not applied on industrial scale, besides the process using propane. A development was started to establish a feasible process leading to a production scale plant which produces lecithin powder with a minimum content of 95% AIM.

Phase Equilibrium Experiments The studies started with solubility measurements a phase equilibrium apparatus on a real de-gumming residue (DR) containing 50% AIM. Determined was the solubility of the soy bean oil in CO2 as well as the solubility of CO2 in the soy bean oil. Measurements were performed in pressure range up to 600 bar and at temperatures 40 and 60 °C. The apparatus used was high pressure view cell (60 mL, 700 bar, 250 °C). 25–30 g of DR was introduced in to the view cell and pressurized with CO2 to desired pressure. Stirring and settling was applied for 1 h each, and samples of lower and upper phase were taken. Results are given as mg of oil per g of CO2 for the upper (gaseous) phase and mg of CO2 per g DR for the lower (liquid) phase (Fig. 3.21). On the mass basis, results are also given as weight % of CO2 in liquid (left side of diagram) and in gaseous phase (right side of diagram) and are shown in Fig. 3.21. The solubility of CO2 in DR increased up to approx 30 wt. % at 325 bar and remained constant up to 600 bar. The maximum wt. % of oil in CO2 was obtained at 1.5 wt. % of oil (Table 3.4). As shown in Fig. 3.20 the solubility of oil in CO2 as well as the solubility of CO2 in oil increases with increasing pressure and temperature. A solubility maximum in the applied pressure range was not observed for the solubility of oil in CO2 .

88

3 Industrial Scale Applications: Physical-Based Processes

Table 3.4 Phase equilibrium data for system degumming residue (DR)-CO2 Pressure (bar)

Temperature (°C)

Solubility of oil in CO2 (mg oil/g CO2 )

Solubility of CO2 in oil (mg CO2 /g DR)

202

43

2.791

119.930

389

42

4.603

405.670

181

61

6.905

136.413

378

61

11.717

362.207

601

60

15.313

363.689

Fig. 3.20 Solubility data for the system soy bean oil/CO2 (CO2 in oil and oil in CO2 )

Fig. 3.21 Phase observations in system de-gumming residue/CO2

Phase Behaviour Observations of System Degumming Residue—CO2 in View Cell High pressure view cell was filled with de-gumming residue and was pressurized with CO2 to certain pressure. Temperature was set to desired value and the system was

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

89

pressurized to desired value. Mixture was stirred and left for the phases to separate, during which phase behaviour was observed: Influence of process parameters on: • Distribution of liquid and gaseous phases—possible phase inversions, • Qualitative evaluation of viscosity of mixtures, • Separation of phases after intensive stirring, was observed. In the range of applied pressure and temperature (up to 520 bar at 62 °C and 700 bar at 40 °C) the distribution of phases was such that degumming residue rich phase (liquid phase) was the bottom phase and CO2 rich phase containing dissolved oil (gaseous phase) was the upper phase. No inversion of phases was observed as could be seen from Fig. 3.21. It was found that the viscosity of system was reduced with increasing pressure at constant temperature and at constant pressure the viscosity was reduced with increasing temperature. The separation of phases was fast after intensive mixing.

Laboratory Scale Extraction Experiments From the results of the solubility measurement the conditions for the experiments in the next scale were fixed. The experiments were carried out in a 10 L laboratory plant with a maximum operating pressure of 750 bar. Extraction was carried out in the way that the de-gumming residue is fed at the top into the extractor and is counter-currently extracted by CO2 , which is fed at the bottom of the vessel. The extracted oil is recovered in the separator while the de-oiled powder is collected in the extractor. The flow diagram of the experimental extraction set-up is presented in Fig. 3.22. Tests were carried out at 400 bar and 500 bar at a constant temperature of 60 °C. The specific CO2 -flow rate, which is defined as kilogram of CO2 per kilogram of raw material, was varied between 75 and 225 kg/kg. As raw material the same de-gumming residue as for the solubility measurements were used. The results of lab-scale experiments are shown in Fig. 3.23. As the target the content of purity of lecithin were expressed in AIM (acetone insoluble material). As it can be seen from Fig. 3.23 the AIM-content in the de-oiled product is widely independent from the specific CO2 -demand at the tested condition.

Pilot—Small Scale Production Tests The results from the tests in the laboratory plant, pressure, temperature, CO2 -flow rate were used for scale-up tests in a small scale production plant. This plant can be operated up to an extraction pressure of 500 bar at a CO2 -flow rate of 1,000 kg/ h. The tests were carried out as described above at 400 bar and 60 °C with specific CO2 -flow rates ranging from 100 kg/kg to 200 kg/kg at a scale-up factor of 1:30.

90

3 Industrial Scale Applications: Physical-Based Processes

Fig. 3.22 Principle flow diagram (photo: courtesy Uhde High Pressure Technologies, Hagen, Germany) Fig. 3.23 Results of lab-scale tests: purity of lecithin versus specific CO2 consumption

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

91

Fig. 3.24 Basic flow sheet of industrial plant (operating pressure 500 bar, operating temperature 100 °C) (photo: courtesy Uhde High Pressure Technologies, Hagen, Germany)

During all test runs totally 200 kg of de-gumming residue were extracted resulting in approx. 100 kg of lecithin powder. As it was found out during the lab-scale tests also the scale-up tests show, that the extraction result is independent from the specific CO2 -flow rate in the range between 100 kg/kg and 200 kg/kg. Besides the degree of de-oiling also the target of producing a dry powder was reached. During the experimental study a safe design basis for production plants was established. The experiments range from a scale of some milli-litres to 200 L extractor volume. In every scale it was demonstrated that both of the targets, high degree of de-oiling and a dry powder product, can be reached.

Production Plant Design, Construction and Start up As the result of the experimental studies a production plant was designed based on the parameters, which were obtained and verified in different scales. This plant has to de-oil 200 kg of liquid lecithin per hour (Fig. 3.24). Assuming 62% AIM in the raw material approx. 120 kg/h of pure lecithin powder will be produced. Besides the degree of de-oiling and the powdery shape it is essential to prevent oxidation in the final product. For the production plant a continuous feeding of the raw material and uninterrupted extraction is applied. Because the product is a solid material, de-oiled lecithin powder, the extractor has to be emptied batch-wise during the extraction

92

3 Industrial Scale Applications: Physical-Based Processes

Fig. 3.25 Particle size distribution of lecithin powder produced at different operating conditions

Fig. 3.26 View of lecithin powder particles produced at different operating conditions

without interruption of feeding. This has the advantage of a more effective and economic operation of the de-oiling process and ensures that the sensible product will not come into contact with oxygen from air. The plant will be operated at extraction pressures up to 500 bar and should produce a “green” lecithin from non-GMO soy beans. The tests were performed on the properties of deoiled lecithin particles shape and particle size distribution. The view of particles and their particle size distribution is presented in Figs. 3.25 and 3.26.

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

93

Fig. 3.27 Samples of the deoiled lecithin from industrial plant (photo: courtesy Uhde High Pressure Technologies, Hagen, Germany)

After accomplishing the plant installation, the plant was started up and production tests were carried out. The parameters applied for the de-oiling were the minimum figures found out in the experimental studies. In all tests the minimum requirement on de-oiling was fulfilled. The content of acetone insoluble matter reached 95% and even more. Figure 3.27 displays samples of the product confirming the viability of the process, the engineering and design as well as the plant set-up. Figure 3.28 shows the constant improvement of the process with the number of production batches. As the plant is fed with degumming residue continuously, a batch is defined as a filling of dried lecithin powder removed from the extractor resp. the hopper vessel after 2 h of spray drying. The lower AIM-limit for a batch according to specification is indicated by the red line (i.e. 95% AIM). All batches are above the commercial limit and the improvement over the batches produced is indicated by the green dashed line indicating that after ca. 25 batches (i.e. less than 2 days of operation) a consistent drying beyond 96.5% AIM was achieved.

3.1.9 Design of Extraction Plant for Liquids There are less industrial units for separation of components from liquid mixtures using supercritical fluids. Extraction of liquid mixtures with supercritical fluids is comparable to liquid–liquid extraction, where compressed gas is used instead of an organic solvent. In liquid-SCF extraction processes the pressure plays an important role. Changing pressure and/or temperature, the physico-chemical properties of the SCF, like density, viscosity, surface tension, dielectric constant, etc. are changed. Selective extraction of components or fractionation of total extracts is possible by use of different gases for isolation/fractionation of components and/or changing

94

3 Industrial Scale Applications: Physical-Based Processes

Fig. 3.28 AIM content of production batches from industrial plant. (photo: courtesy Uhde High Pressure Technologies, Hagen, Germanyp)

process parameters. Another advantage is that depending on the feed material the density difference between the two counter-current flowing phases can be adjusted. One of the most important advantages of use of supercritical fluids is the simple solvent regeneration in comparison with liquid–liquid extraction, where in the most cases solvent regeneration includes a re-extraction or distillation step, which is energy consuming and therefore cost intensive. Heat treatment of extract or raffinate phase may cause degradation of heat sensitive substances. In SCF extraction plant the solvent regeneration is achieved by changing pressure and/or temperature after the extraction step, thus changing the density and the solvent power of gas, which can be later easily recycled after separation of solute. Compared to extraction of solids with SCF, liquids could be continuously introduced in and withdrawn from the high pressure extraction unit. This gives the benefit of higher throughputs in continuously operating counter-current processes. Literature search shows some laboratory scale studies on extractions in the systems liquid/supercritical fluid. Several data on binary systems liquid/SCF could be found, but there is less data on systems liquid/liquid/supercritical fluid which are necessary for design of extraction processes of liquid mixtures with supercritical fluids. Like in conventional continuous liquid–liquid extraction in liquid/sub- or supercritical solvents extraction several modes of operation are available. Single stage extraction is the simplest one and is used for the systems where separation factors for solute are high. Multi-stage separation is necessary when the separation factor between components is in the order of 1–10. Different modes of operation of multistage processes are used like multi-stage cross-flow where relatively low loading of solvent with extract are obtained in each stage. In multi-stage counter-current extraction high loading of solvent with extracts is possible as well as different geometry of apparatus.

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

95

Counter-current liquid/sub- or supercritical fluid extraction can be modeled by the use of typical common used basic equations: mass balance, energy balance, equilibrium distribution coefficients, mass transfer rate equations. For extraction, following data are necessary: – determination of number of theoretical stages/transfer units, – size and type of a separation device with respect to separation performance, – design of solvent cycle. Based on the above facts and experimental data the costs of separation using liquid/sub- or supercritical process can be determined. The costs per ton of the feed are influenced by throughput and mode of operation (batch processes have higher operating costs while in continuous the process costs are lower) and are in the range from ca60. In the literature (Johnston and Penninger 1989) design, scale-up and economy of supercritical extraction of solids are presented. In (Brunner 2009) it is reported that common estimation methods yield results with an error of ±30%, while even after the project has been completed, costs are difficult to be determined better than with 5% error. Main reasons can be—since typically natural products are processedvarying characteristics depending on origin or provenience, characteristics of season or environmental conditions (weather, soil, etc.), unevenly distributed moisture, etc. Furthermore, the process design takes into consideration the main components only, whereas other constituents or trace components within the raw material as well as effects from recycling the solvent may influence the overall economy.

3.1.10 Extraction of Liquids Using SCF-Industrial Scale Units Applications of liquid/sub- or supercritical fluid extraction are numerous and were used for separation of ethanol from water (Hsu and Tan 1994; Knez et al. 1994), separation of aromas from different alcoholic beverages (Gamse et al. 1999), separation of components from citrus oils (Knez 1989), for purification of tocopherols (Fleck et al. 2000). There are several research projects for use of liquid/SCF fluid extraction processes in chemical industry. Separation of caffeine from CO2 is used widely in decaffeination process. The high pressure column of an industrial scale separation processes is presented (Fig. 3.29). In the future further limitations on use of organic solvents and new application demands will be driving force for processing in sustainable manner. Example on Liquid/SCF Extraction: The task of a project proposed by chemical industry was purification of water solution of an inorganic chemical from non-polar organic impurities by supercritical extraction.

96

3 Industrial Scale Applications: Physical-Based Processes

Fig. 3.29 High pressure column of industrial scale separation of liquids with sc CO2 350 bar, 120 °C Di = 0.07 m, L = 5 m. (photo: courtesy Uhde High Pressure Technologies, Hagen, Germanyp)

Fig. 3.30 Flow sheet of extraction unit for purification of organic impurities from water

3.1 Supercritical Fluid Extraction and Fractionation from Solid and Liquid …

97

Based on experiments in view cell basic thermodynamic data necessary for the design of a first pilot scale process were determined. Later on, the mass transfer was studied in lab-scale with a continuously operated counter-current extraction column (filled with packing) and further optimisation of process was done. The basic design of the process is presented in Fig. 3.30. The process parameters are: Counter Current Extraction Column: • extraction pressure of 350 bar, • extraction temperature 40 °C to 60 °C, • S/F 10 to 30 kg of CO2 /kg feed. The ratio S/F depends on the desired purity of the final product. Separator: • separator pressure of 50 bar to 55 bar, • separator temperature 40 °C.

3.1.11 Conclusion Technologies using supercritical fluids offer several benefits compared to organic solvent technology. Those technologies are ecologically friendly and enable easier product fractionation. The largest scale extraction processes using sub- or supercritical solvents realized on industrial scale are decaffeination of tea and coffee and extraction of hop ingredients. Some industrial plants like extraction of spices for food industry and natural substances for use in cosmetics also use sub- and supercritical fluids in their processes. There are fewer industrial units for separation of components from liquid mixtures using sub- or supercritical fluids. Supercritical fluids have many favourable properties among them the main advantages present solvent free products, low temperature during the separation process and no co-product. These processes can also be easily connected directly to micronisation or crystallization from supercritical CO2 by fluid expansion. The major advantage of use of supercritical fluids is selective extraction of components or fractionation of total extracts. This can be achieved by the use of different gases for isolation/fractionation of components and/or by adjusting the process parameters. Besides the commonly used gas, carbon dioxide, for sub- or supercritical extraction, other sub- or supercritical solvents are being used. Between sub- and supercritical fluids CO2 and H2 O indicates the most favourable properties. They are both non-toxic, non-carcinogenic, non-mutagenic, non-flammable and thermodynamically stable. Furthermore, CO2 does not normally oxidize substrates or products and allows the process operations at low temperatures. Currently the water represents the cheapest solvent with high solubility of numerous substances.

98

3 Industrial Scale Applications: Physical-Based Processes

Consequently, a lot of research is now focused on sub- or supercritical water for isolation and fractionation of substances. Thermo-physical properties of SCFs such as: high diffusivity, low viscosity, the density, dielectric constant present the major process benefits of SCF and can be fine-tuned by changes of operating pressure and/or temperature. Price of the final product is the limitation of further application of extracts obtained by high pressure technology. In comparison to conventionally obtained products the price of products obtained by high pressure technology are usually higher. The legal limitations of solvent residues and solvents (for products used in human applications such as food, supplements/nutraceuticals, pharmaceuticals, etc.) and isolation/fractionation of special compounds from total extracts in combination with different formulation (like controlled release) (Reverchon et al. 2009; Weidner 2009), chromatography (Taylor 2009) and sterilization processes will increase the use of supercritical fluids for extraction applications. There is no trend evidently related to higher or lower pressure. However, a part of research and development and patent filings of SC extraction processes are focused on lower pressure and the other is orientated towards higher pressure. Various substances have limited solubility at normal pressure and the solubility of several substances increases for some orders of magnitude using pressures over 500 bar. The solubility of several dense gas low soluble substances increases with increasing pressure and temperature. Due to this, ultra-high-pressure range (up to 2500 bar) allows further fractionation of substances in the extract by pressure and/ or temperature dependant precipitation. Estimation of processing costs for few plant materials processed in so called ultrahigh pressure extraction process using scCO2 is presented in Fig. 3.30. The trend shows that with increasing operating pressure of the SC extraction plant the total processing costs decrease at a given plant capacity. The main reason is the sheer speed of the process, i.e. the extraction time is so much shorter that the equipment can be smaller due to far enhanced process efficiency. In some cases it might be argued that pressures as high as 1500 bar are too difficult to manage and operate safely in large volumes for the extraction with scCO2 . However, it should be considered that another large scale process—the synthesis of LDPE in an autoclave reactor—is accomplished in reactor volumes up to 2500 l under similar pressure, much higher temperature (ca. 300 °C) and within reactive ethylene and peroxide under very stringent process safety measures. Therefore, the potential advantages of the inert scCO2 for process intensification should be considered and elaborated diligently in future work opening new fields of application for sc fluids. From this time forth, additional limitation on the use of health-hazardous organic solvents, new applications of numerous substances, new customer desired properties of products, sustainable production and processing of substances will open new applications of high-pressure processing. Furthermore, all the advantages in highpressure research area such as cheap and environmental friendly solvents, like CO2 and some other gases and sub- or supercritical water will open up new pathways for substances and products produced at high pressure.

3.2 High Pressure Polymer Processing

99

3.2 High Pressure Polymer Processing Supercritical fluids are well established for use as a green processing solvent in polymer applications such as polymer modification, the formation of polymer composites, polymer blending, microcellular foaming, polymerization and particle production (Nalawade et al. 2006). Powders and composites with special characteristics can be produced by applying supercritical fluids for particle formation and present a possibility to overcome the drawbacks of conventional particle size reduction processes like crushing, milling, cryo-milling, grinding, sublimation and recrystallization. Impregnation of solid particles, for formation of solid powderous emulsions, particle coating, e.g. for formation of solids with unique properties for the use in different applications are some of the proceses benefiting from the unique thermodynamic and fluid-dynamic properties of supercritical fluids (Knez and Weidner 2003). Using high pressure processes in presence of supercritical fluids it is possible to generate powders with properties that are difficult or even impossible to achieve in conventional ways and can be applied for polymers, waxes and resins, natural products, fats and fat derivatives, pharmaceuticals, synthetic and natural antioxidants, surface-active compounds, UV-stabilizers. In particular, special attention is dedicated to using biodegradable polymers in particle size reduction processes that are related to pharmaceutical applications for controlled drug release (Marizza et al. 2014). Advantages of polymers in forms such as oils, tars, resins, and gums have been exploited since centuries ago. After industrial revolution, modern polymer industry began to develop when Charles Goodyear succeeded in producing a useful form of natural rubber through a process known as “vulcanization.” But anyway, progress in polymer science was slow until the 1930s when materials like vinyl, neoprene, polystyrene, and nylon were discovered. The diversity of polymer properties makes polymers such as cotton, wool, rubber, teflon, and all plastics widely applied in every day life and also in industrial applications where natural and synthetic polymers with a wide range of stiffness, strength, heat resistance, density, and even varying price are produced. As mentioned above, low cost, low density, ease of processing are just some of the advantages of polymers which make them highly applicable. But there is also a disadvantage, poor mechanical properties, which also explaines the fact that polymers are higly investigated also in the fields of scientific literature in order to moderate this disadvantage. One possibility is to modify the structure by producing polymer-matrix composite materials. Lately, extensive research has been focused on the use of supercritical fluids for obtaining polymers and composite microparticles. Polymer Matrix Composites (PMC) are very common due to their low cost and simple fabrication methods. Independently on the type of polymer (thermoplastic or thermoset) low temperature is required in polymer processing due to possible polymer degradation. Conventional methods generally demand high temperatures and/or use of organic solvents which may be toxic, flammable, environmentally unfriendly and expensive and therefore

100

3 Industrial Scale Applications: Physical-Based Processes

they are more and more often substituted by novel methods applying high pressure processes in the presence of supercritical fluids (Kiran 2009; Nalawade et al. 2006). As absorption of compressed gas in polymer matrices results in a wide spectrum of possible applications, for instance production of fibres, micro-particles and foams, processes as polymer impregnation, separation of gas mixtures through polymer membranes are of a high interest and therefore investigation of thermodynamic properties of binary systems polymer/CO2 are a topic under an intense research. Several modifications of polymers are available since either polymer swelling or CO2 dissolution can be expected (Kiran 2009; Nalawade et al. 2006). Addition of supercritical fluid reflects in modification of several physical properties of polymer such as glass transition temperature, melting temperature, surface tension, and viscosity, which are changed depending on solubility rate of supercritical fluid in the polymer (Kegl et al. 2017). One of the most frequently used biodegradable polymer is polyethylene glycol (PEG), which is a water-soluble polymer, phsiological acceptable and biocompatible (Trupej et al. 2015). Mathematical models are commonly used to predict phase behavior of the investigated system, but anyhow, several parameters controlling feasibility of the process should be determined experimentally prior to design of different processes involving substances in supercritical fluids (Knez Hrnˇciˇc et al. 2014a). Processes like polymer synthesis and separation, formation of particles, fibers, foams and blends, polymer impregnation, morphological modifications and transformations are nowadays successfully performed by indruduction of supercritical fluids replacing thermally induced processes or processes involving organic solvents. Major drawbacks of these approaches are well known; the difficulty to process heat-sensitive materials and the possible presence of residual solvent in the final product. Macromolecular chains of polymers (PEGs) allow absorbtion of significant amount of gas in sub- or supercritical state. The gas, in most cases CO2 , weakly interacts with basic sites of the polymer chain, which results in polymer plasticization, resulting in an increase of the inter-chain distances and higher chain mobility. Glass transition temperature in the case of amorphous polymers and temperature of melting point in the case of crystalline polymers are significantly lowered by sorption of scCO2 . Theses concepts are widely applied in several industrial branches; mainly food, pharmaceutical and cosmetic industries and also medicine, where products of high quality are required (Knez et al. 2014).

3.2.1 Polymer Particles For the production of powders (solid components, composite with defined particles size and size distribution as well as particle shape in industrial scale) conventional particle size reduction processes like crushing, milling, cryo-milling, grinding, sublimation and recrystallization are ordinarily applied.

3.2 High Pressure Polymer Processing

101

Supercritical/dense gas processes have been used for particle design and drug formulation. Thedriving force of the research is derived from the fact that conventional processes for manipulating the characteristics of particles used in drug formulation are limited in their flexibility. The physical properties of the particles such as particle shape, the particle size distribution can not be precisely fine-tuned. Formulation of composites using conventional techniques for particle size reduction is practically not possible. Another problem represent substances that are heat sensitive and could therefore not be treated by mechanical particle size reduction procedures. In recrystallization processes substances are contaminated with solvents in sublimation processes it is not easy to control particle size and particle shape, while the particle shape and particle size distribution could not be influenced by variation of process parameters. The unique thermo-dynamic and fluid-dynamic properties of supercritical fluids can be used also for impregnation of solids, for formation of solid emulsions, particle coating and for formation of composites, which are traditionally produced by using solvents. Though, particularly in food and pharmaceutical applications the use of certain solvents is limited or even restricted (EU Council Directive 88/344/EEC). Due to disadvantages of conventional processes new techniques have been developed which use sub- and supercritical fluids and overcome the drawbacks of conventional processes (Knez and Weidner 2001). By applying supercritical fluids powders and composites with special characteristics can be produced. Several processes for formation and design of solid particles using dense gases are studied intensively. There are several examples of particle formation and drug formulation by dense gas techniques which demonstrate the potential of dense gas technology for drug formulation purposes (Hannay et al. 1880). Recent developments of dense gas techniques, fundamentals and literature review on particle production using sub-critical and supercritical fluids are discussed in the following subchapters. Particle formation techniques and formulation of materials with sub- and supercritical fluids is today an important topic of research. Approximately 70% of new drug candidates possess poor aqueous solubility and about 40% of the immediate release oral drugs on the market are considered to be practically insoluble in water (Pestieau et al. 2015). The pharmaceutical industry is interested in obtaining the successful formulation of poorly soluble active compounds in order to increase their bioavailability and dissolution rate. The current delivery options for improving the dissolution properties of drugs are particle size reduction, crystal modification, pH modification, self-emulsification, amorphization and the formulation of drugs with surfactant carriers and amorphous polymers (Jug et al. 2018). Particle size reduction and drug formulation with polymeric carriers incorporate the most promising options in this regard. A number of conventional methods have been developed to improve the dissolution properties of drugs (Kawabata et al. 2011). Many of these methods possess drawbacks, such as thermal and chemical degradation of drugs, large quantity organic solvent use, broad particle size distribution and low drug load (Huang et al. 2005).

102

3 Industrial Scale Applications: Physical-Based Processes

To overcome these limitations, supercritical fluid technology promises to be an excellent option. The reasons for the widespread use of supercritical fluids mainly lie in the simplicity of the processes, high purity of products, no organic solvents in the process, no communication steps in drug preparation, mildness of operating conditions and the possibility of obtaining non-contaminated fine particles with narrow size distributions (Knez et al. 2011; Weidner et al. 2003). Production of fine particles with improved characteristics using supercritical fluids has been obtained with rapid expansion of supercritical solutions (RESS) (Paisana et al. 2016), the gas antisolvent process (GAS) (L˝orincz et al. 2016), supercritical antisolvents (SAS) (Prosapio et al. 2016) solution enhanced supercritical dispersion processes (SEDS) (Dal Magro et al. 2017), aerosol solvent extraction systems (ASES) (Bleich et al. 1993), supercritical fluid extraction of emulsions (SFEE) (Lévai et al. 2017) and particles from gassaturated solution (PGSS™ ) (Martín and Weidner 2010). In particular, the PGSS™ is an organic solvent-free process which has found a wider spectrum of industrial applications. In the process, polymeric carriers with the target pharmaceutical drug to be micronized and encapsulated are loaded into a high-pressure autoclave together with supercritical CO2 . By heating the autoclave, the content is melted and then, after saturation with the supercritical CO2 solution, rapidly expanded through a nozzle. Fine particles with irregularly shaped morphology, which normally release the active compound in a very short period of time, are produced (Couto et al. 2017). Several bioactive substances, including active pharmaceutical drugs, flavors, and vitamins, have been successfully micronized and formulated using the PGSS™ process on a polymeric carrier. Weidner et al. (1996) studied particle precipitation of polyethylene glycol (PEG) of different molar weights that has been frequently employed in the preparation of solid dispersions. Chen et al. (2013) successfully micronized the non-steroidal anti-inflammatory drug ibuprofen with PEG 6000 as a carrier material. Božiˇc et al. and Kerˇc et al. (1999), (Senˇcar-Božiˇc et al. 1997) used PEG 4000 for the powder generation of the practical water-insoluble calciumchannel blockers nifedipine and felodipine and the hypolipidemic agent fenofibrate. Meanwhile, Marizza et al. (2014) report on the drug release of ketoprofen from polyvinylpyrrolidone (PVP) using supercritical CO2 . Pestieau et al. (2015) developed a formulation containing fenofibrate and Gelucire 50/13 as a career in order to improve the bioavailability of the insoluble drug. García-González et al. (2010) investigated the encapsulation efficiency of solid lipid hybrid particles prepared using the PGSS technique with ketoprofen, glutathione, and caffeine. Knez et al. (2004) measured phase equilibria and performed micronization of the flavors vanillin and ethyl-o-vanillin in the presence of various compressed gases, where vanilla was mainly used to mask the unpleasant flavors of drugs. Couto et al. (2017) research encapsulated vitamin B2 in solid lipid nanoparticles using supercritical CO2 with the aim of protecting active substrate and longer shelf-life. Production of ultrafine (micro- or nano-sized) particles with desired properties and precise control of particle size and morphology is one of the objectives of many industries.

3.2 High Pressure Polymer Processing

3.2.1.1

103

State of the Art of Material Processing Using Supercritical Fluids

Several papers about micronisation using sub-critical or super-critical fluids have been published recently, which means that the topic is still is still a subject of intensive research. Polymer processing considering compounding and mixing of polymers requires previous knowledge of S-L-G phase equilibria, density, viscosity, interfacial tension, gas solubility, and diffusivity in multipolymer and supercritical fluid systems. In recent years, many experimental and numerical investigations of various molecular weight polyethylene glycols (PEGs) using carbon dioxide (Gourgouillon and Ponte 1999; Knez Hrnˇciˇc et al. 2014a), argon (Trupej et al. 2015), nitrogen and propane (Trupej et al. 2015) as an sub or supercritical fluid have been done. Therefore, due to the several considerations, related to the complicated thermodynamic and mass transfer inquiries, compared to several hundred commercial high pressure extraction units, the number of micronisation units developed to industrial scale is still relatively low, but is increasing rapidly due to the growing interest. Today, supercritical fluid technology allows the production, with a reduction of the environmental impact, of well designed nanomaterials with controlled properties for applications in many interesting fields, such as catalysis, electronics, energy, optics, pharmacology, etc. Generally, the properties of the obtained powder product like particle size, size distribution and morphology depend on phase equilibria and thermodynamic behavior of the system, fluid dynamics, mass transfer and nucleationgrowth kinetic. Beneficial thermodynamic and mass transfer properties of supercritical fluids allow an easy tuneability of solvent properties, which is a major advantage of using using sub-critical or super-critical fluids for micronisation processes, which can furthermore be easily integrated with subcritical or supercritical extraction processes, or to a downstream processing for products of chemical or biochemical synthesis in subcritical or supercritical fluids. Supercritical fluids are already commercially applied in a variety of fields—from pharmacy, food sciences to the textile industry and as research continues to investigate the capabilities of CO2 , new applications of the technology are developing daily. Water and carbon dioxide, especially in their supercritical states, are possible and very suitable replacement of conventional solvents for greener, environmental friendly processes, which require lower energy consumption, but as an disadvantage, higher investment costs. Use of CO2 also prevents oxidation of products during processing steps. Carbon dioxide is the most commonly used supercritical fluid because of its low critical temperature, low toxicity and high purity at a low cost. It is non-flammable and its use does not contribute to the net global warming effect. Being a gas under ambient conditions favours its easy removal from polymeric products by simply reducing the pressure. It can be easily recovered for recycling and can be used to replace harmful or toxic materials such as freons.

104

3 Industrial Scale Applications: Physical-Based Processes

One of the major reasons for the fast development in the field of application of subcritical and supercritical fluids are restrictions regarding the use of conventional organic solvents in food, feed and pharmaceutical industry. All in all, current research of S-L-G phase equilibria, density, viscosity, interfacial tension, gas solubility, and diffusivity in polymer and supercritical fluid systems demonstrates that the field of micronisation and formulation processes using subcritical and supercritical fluids will open up new ways for substances produced at an industrial scale in the near future. Even if water is the cheapest solvent, the use of water as recrystallization solvent is limited due to the high polarity; therefore CO2 with its several advantages (Careno et al. 2012) is still most recently used. Supercritical carbon dioxide acting as a plasticizer was used in investigations of polycaprolactone (Fanovich and Jaeger 2012).,polypropylene, polystyrene (Sato et al. 1999), poly(βhydroxybutyrate) (Khosravi-Darani et al. 2003), poly(3-hydroxybutyrate-co-3hydroxy valerate) (Cravo et al. 2007), and polyethylene glycol nonyl phenyl ether (Dimitrov et al. 1998). Weidner et al. (1997) studied phase equilibrium (solid–liquid–gas) in polyethylene glycol (PEG)-carbon dioxide systems that has been lately frequently employed in the preparation of solid dispersions. Phase behavior with a negative dp/dT slope is desirable, where the SCF is highly soluble in the molten heavy component (Knez et al. 2010). In particular, the solubility of the SCF biodegradable carrier PEG of different molar weights was found to increase significantly with pressure (Kegl et al. 2017; Trupej et al. 2016). Knez Hrnˇciˇc et al. (2014a) measured the solubility and diffusivity of CO2 in PEGs of different molecular weight. Results indicated that solubility of CO2 in PEG increases with increasing pressure. In contrast, diffusion coefficients in the system are mostly influenced by the amount of CO2 already present in PEG. The same research group has also reported on density and viscosity of the binary polyethylene glycol/CO2 systems (Knez Hrnˇciˇc et al. 2014b). They found out that increase of dissolved CO2 in the polymer matrix is related to viscosity reduction and density increase. Fanovich and Jaeger (2012) published a research on determining solubility and diffusivity of CO2 within polycaprolactone and thus on the swelling of the polymer using magnetic suspension balance. Sorption measurements indicate that a higher amount of gas is absorbed when pressure is increased, meanwhile, this amount usually decreases when the temperature is raised. Sato et al. (1999) addresses the behavior of the CO2 and N2 in the polypropylene and polystyrene. The solubility of CO2 and N2 in polypropylene were measured at temperatures from 160 to 200 °C and pressure up to 170 bar. The solubility of CO2 decreased with increasing temperature, while the solubility of N2 increased with increasing temperature. The solubility of both gases in polypropylene is much higher than in polystyrene. Khosravi et al. (2003) investigated the effect of pressure and temperature on the solubility of CO2 in polyhydroxybutyrate. Results clearly demonstrate that solubility increases with increasing pressure and temperature at a pressure above 182 bar, meanwhile, below this pressure, the solubility increases with increasing pressure but decreases with increasing temperature. Cravo et al. (2007) report on the determination of diffusion

3.2 High Pressure Polymer Processing

105

coefficient of CO2 in natural biodegradable poly (3-hydroxybutyrate-co-3-hydroxy valerate). A maximum diffusion coefficient has been achieved at lower temperatures and higher pressures. Dimitrov et al. (1998) report on the solubility of CO2 within polyethylene glycol nonyl phenyl ether of molecular weight of 1980 g/mol, 2420 g/ mol, and 2860 g/mol. The results show, that the solubility increases with increasing the pressure for all molecular mass of polyethylene glycol nonyl phenyl ether at a constant temperature. Review by Nalawade et al. (2006) provides extensive information on useful processing aspects and applications of polymers melts at high pressures, where the high diffusivity of CO2 and the low interfacial tension are key factors in determining a wide range of applications. It is well known that polymer-gas mixture exposed to high pressure and temperature has a lower interfacial tension than pure polymers. Mahmood et al. (2014) investigated the interface of molten polylactide (PLA) by analyzing sessile drops in high-pressure and high-temperature visualization chamber from 68.9 to 206.8 bar and from 463 to 473 K, respectively. Interesting work was carried out by Carbone et al. (2012) were interfacial tension, solubility, diffusivity and specific volume of molten poly(caprolactone)/CO2 solutions were simultaneously measured to reduce errors by using costume-designed measurement device consisting of a rod to which polymer-gas solution is stuck (by patent drop method) and placed in a magnetic suspension cell. Gutierrez et al. (2014) studied interfacial tension and glass transition of polystyrene in scCO2 within the pressure range from vacuum to 90 bar, and temperatures from 30.15 to 40.15 °C by the pendant drop method. Although it is difficult to achieve accurate and comparable data in modelling of thermodynamic properties of molten polymers at high pressures with experimental ones Enders et al. (2005) presented the Chan- Hilliard theory with equations of state (the original statistical associating fluid theory, the perturbed-chain statistical associating fluid theory or the Sanchez–Lacombe lattice theory) in order to describe both, the solubility of carbon dioxide in polystyrene, and the interfacial properties between the liquid mixture and the pure gas phase. Operating pressure during processing with supercritical fluids is of fundamental importance considering morphology of the prepared materials. In some cases, operating pressure showed a linear relation to the porosity of the matrices (Silva et al. 2011) A theoretical approach to describing the experimental data on the sorption and diffusion of gaseous mixtures in polymers under high pressures has been developed by Kiran (Krykin 1990). Markoˇciˇc et al. (2013) investigated the effect of pressure on a system of polycaprolactone and CO2 . Temperature and pressure influence foam morphology through their effect on gas solubility in the polymer and on the viscosity of the substrate. The number of pores and their size depends on nucleation and growth rates. At higher gas concentration, the influence of nucleation predominates, therefore higher solubility enables more pores with smaller diameters. The results show that supercritical CO2 is suitable to obtain the desired porosity of polycaprolactone. Gutierez et al. (2014) shows the behavior of the system polystyrene /CO2 in the presence of terpene oil limonene in the pressure range from 50 to 150 bar and concentration from 0.05 g to 0.2 g polystyrene/mL limonene at a temperature of 30 and 40 °C. At the pressure of 80 bar, the limonene is fully miscible with CO2 , which results in lower absorption of CO2

106

3 Industrial Scale Applications: Physical-Based Processes

into polystyrene at a further increase of pressure. The most suitable condition to foam polystyrene from limonene solution of 0.1 g polystyrene/mL limonene is a pressure of 90 bar, the temperature of 30 °C, contact time of 240 min, and depressurization time of 30 min. Nalawade et al. (Nalawade et al. 2007) investigated phase behavior of PEGs with molecular weight 6000 g/mol and 10,000 g/mol with a goal to produce micron size particles from gas-saturated solution (PGSS). They report that particle size, shape, morphology and particle size distribution depending on the molecular weight of polymer, pressure, and temperature and nozzle diameter.

3.2.1.2

Crystallization from Supercritical Solutions (CSS)

Fundamentals A large number of particle engineering methodologies have been developed, tested, and applied in the synthesis and control of particle size/particle-size distributions, crystallinities, and polymorphic purities of drug micro- and nano-particles/crystals. In recent years pharmaceutical processing using supercritical fluids, in general, and supercritical carbon dioxide, in particular, have attracted a great attention from the pharmaceutical industry. CSS was not widely used and is limited to some products experiments for various products. Crystallization from supercritical fluids has been proposed as a noncontaminating, non-toxic technique for purification and particle size manipulation of thermally labile pharmaceutical compounds. Simple control of process parameters and ommision of solvents are the main features of the crystallization from supercritical solutions, while the disadvantages of high pressure crystallization are high volumes of solvents (gasses), owing to the low solubility in the gas of the substances to be crystallized, quite high pressures, batch wise operation, and lengthy cooling times. When the solute-laden solution is a supercritical fluid, supersaturation may be induced not only by varying the temperature but also by pressure variation. Thus pressure and pressure-gradients would be additional means for generation of particles with the desired size, form and morphology. Applying supercritical fluids in CSS allows to obtain supersaturation and to control nucleation- and growth rates by temperature-induced variation of the concentration of the solute in systems where no volatile organic solvents are present. The formation of small particles is favoured when solids formation is maintained via primary nucleation throughout the batch crystallization. First report on CSS was in 1989, when benzoic acid was recrystallized from supercritical CO2 (Tavana and Randolph 1989). The solvent power of CO2 was reduced by simultaneous reduction of pressure and temperature. Till today no large scale process has been installed. Very probably the reason is that the process has to be performed in large vessels where long time for cooling process is required, while the pressure reduction should not be problematic.

3.2 High Pressure Polymer Processing

107

Fig. 3.31 Batch crystallization apparatus

Advantage of HP batch crystallization is that it is relatively simple process, while the major disadvantages are high volume of apparatus (due to low solubility of substances in supercritical fluids), high pressure (to solubilize as much as possible of the substances), batch wise operation, long cooling times and high volumes of gas (due to relatively low solubility of substances in SC fluids).

Design Criteria Crystal size as well crystal size distribution is a function of supersaturation and could be represented by following equation: ( ) B0 α exp −K1 /log2 S

(3.52)

where B0 is the nucleation rate, S is super saturation (C/C*) where C is concentration of solute in solution and C* is equilibrium solubility of component in solvent. Batch crystallization apparatus (Fig. 3.31) consists of an HP autoclave, gas supply system as well the system for pressure reduction. Although crystalization from supercritical solution this is a relatively simple process, no large scale process have been installed till today, probably due to several disadvantages of the process, such as that due to low solubility of substances in supercritical fluids high volume of apparatus and high pressures to solubilize as much as possible of the substances are required, batch wise operation, long cooling times due to huge mass of high pressure apparatuses what also cause low cooling rates and thus the kinetics of crystall growth is limited and high volumes of gas (due to relatively low solubility of substances in SC fluids). Beside this the yield of the process is low.

108

3 Industrial Scale Applications: Physical-Based Processes

Fig. 3.32 High-pressure crystallization of dicarboxylic acid from CO2 . a View of product in highpressure cell. b Recrystallized product

Very problematic could be also separation of formed crystals from high volumes of gas (due to relatively low solubility of substances in supercritical fluids or dense gas) is rather complicated and could be very complex especially for highly dangerous substances. Due to all these facts application of CSS is very limited and is mostly to research activities or on larger scale lab units to highly sophisticated (toxic, explosive,…) product.

3.2.1.3

Rapid Expansion of Supercritical Solutions (RESS)

Fundamentals Another process where fine particles can be obtained from supercritical solution is RESS. RESS process was proposed in 1986 (Matson et al. 1987). In this process the substance to be micronized is solubilized in SC fluid. Due to relatively low solubility at lower pressure and temperature, higher pressures as well as higher temperatures are usually applied. Later homogeneous solution of substance in supercritical fluid is expanded to lower pressure or to ambient conditions. Super saturation which occurs due to sudden depressurization gives extremely fine particles. The process is comprised of two-steps: solubilisation and particle formation. After the substance is solubilized in supercritical fluid, the mixture is suddenly depressurized in a nozzle causing fast nucleation and fine particle generation (Fig. 3.32). The main features of RESS process are the possibility to obtain very fine particles (even of nano size), particle size distribution could be controlled, the process

3.2 High Pressure Polymer Processing

109

is a typical solvent free process, and above all, scale up is due to knowledge of fundamentals relatively easy. On the other hand, high ratios of gas and substance, high pre expansion pressure of the process and high temperature during the operation are the main disadvantages. Additional problem is that separation of fine particles from large volumes of gas is often difficult, process are mostly operated in batch wise, therefore vessels are relatively large, processing costs are due to all disadvantages high.

Design Criteria In designing a RESS process, it is essential to obtain the knowledge on what occurs during the extraction step. Consequently, it is required either to collect data from the literature if possible, or to perform experiments or modelling, about the substance solubility in the supercritical fluid (Krukonis and McHugh 1994). The key parameters of the extraction step that influence solubility are operating pressure and temperature. Another important is the flow rate, namely equilibration may not be achieved in extraction autoclave. Kinetics of the dissolution must be considered due to the possible and diffusion limitations. The process parameters can be controlled relatively easily when the solute subjected to extraction is a pure component. If not, additional problems may be encountered. Certainly, after pressure decrease, fractionation of the load can result in a variation of the composition of the obtained particles. Varying the depressurization rate influences also nucleation rate and the supersaturation. This impacts the morphology, as well as particle size and particle size distribution of the final product. The nucleation rate could be expressed by following equation, which describes the number of critical nuclei formed per unit time in a unit volume /

ν2 I = 2 · Ntot · β · σ · 1 · T k | | ( )3 ( )2 2/3 e · exp −16 · π/3 · σ · ν1 · 1/ ln S − K · y1 · (S − 1)

(3.53)

With: N tot total solute concentration in the bulk fluid phase; β, thermal flux of solute molecules; σ, interfacial tension; ν 1 , molecular volume of solid solute; k, Boltzmann’s constant; S, super saturation ratio; K, a function of temperature and pressure; and y1e , mole fraction of solute in equilibrium (Debenedetti 1990). Experimental setup for RESS process is shown on Fig. 3.33. The RESS micronisation apparatus consists of high pressure autoclave, with pressure and temperature regulation and expansion vessel with filter for collection of produced fine particles. The most important part of the unit is decompression nozzle—which are usually of two dimensions (~100 μm) or laser drilled nozzles of 20–60 μm diameter. The nozzles are during operation exposed to large forces

110

3 Industrial Scale Applications: Physical-Based Processes

Fig. 3.33 Experimental apparatus for RESS process

can easily be destroyed during micronization. Therefore they are produced from special materials and are relatively expensive (also due to processing costs for nozzle production). The systems which are micronized by RESS process are highly asymmetrical binary systems with substances of large differences in molecular size, structure and intermolecular interactions. The features of the phase behavior for such systems are: – the triple-point temperature of the heavy component is much higher than the critical temperature of the light one; – solubility of the light component in the liquid phase is quite limited. For design of RESS process, as well for PGSS™ process, the course of S-L-V line is very important which intersects the gas–liquid critical curve in the lower critical end point (LCEP) and the upper critical end point (UCEP). At these points the liquid and the gaseous phase merge to a single phase, where excess of solid phase is present. Solid vapor equilibrium is observed at temperatures between LCEP and UCEP. The solubility of heavy component in the supercritical fluid is near the LCEP relatively low due to low temperature while near UCEP the solubility of heavy component in the gaseous phase is relatively high due to higher temperature. But in all cases the solubilities of solutes are relatively low and therefore the gas demand in RESS process is extremely high. This is one of the main disadvantages of the process. On the other hand also separation of fine particles from large volumes of gas is difficult. The major advantage of the RESS process is that very fine particles could be produced. Due to above mentioned disadvantages and relatively high preexpansion pressure and sometimes also high processing temperature the process is rather expensive and should be applied for production of powders where price of the product is not of prime importance. There are number of papers dealing with hydrodynamic Modelling of the RESS process, phase behavior of expanding mixture and particle formation mechanism

3.2 High Pressure Polymer Processing

111

and some are reviewed by Martin et al. (Martín and Cocero 2008; Reverchon and De Marco 2011). To understand the behavior during fluid expansion, phase diagrams are normally used. Normally, the fluid phases are modelled using the Peng-Robinson’s equation of state in combination with various mixing rules (Fages et al. 2004). These phase equilibria comprise pure solid components, therefore the sublimation pressure is not taken into account. For designing the RESS process it is important to know the solubility data of the heavy compound in the supercritical fluid, and similar for the SAS process the solubility data of both solvent and heavy compound in supercritical fluid in order to determine the conditions of pressure and temperature where solvent is highly miscible while solute is immiscible with supercritical fluid. For designing of PGSS process it is essential to know the solubility data of the dense gas in the molten heavy compound and the course of three phase S-L-V line in P–T projection of phase diagram, i.e. how the gas is influencing the melting point of the heavy compound. In binary systems three phases can be present and equilibrium is described by following equations: fi V (T , P, yi ) = fi L (T , P, xi ) i = 1, 2

(3.54)

f2S (T , P) = f2V (T , P, y2 )

(3.55)

where 1 and 2 denote light (supercritical fluid) and heavy component, respectively, fi is the fugacity of component i, V is the vapor phase, L the liquid phase and S pure solid phase. Equation (X.1) becomes: yi ϕiV P = xi ϕiL P

(3.56)

where ϕi is the fugacity coefficient of component i and xi and yi are mole fractions of compound i in liquid and vapor phase, respectively. Because the fugacity of the pure solute in a solid phase cannot be directly calculated by a conventional equation of state, the fugacity in the solid phase can be estimated by using the fugacity of a pure solute in the subcooled liquid phase, f2subcL (Prausnitz et al. 1986): In

) ( ) fus ( Δcp,2 Ttp,2 Ttp,2 f2subcL ΔH2 − 1 − − 1 = RTtp,2 T R T f2S Δcp,2 Ttp,2 + In R T

(3.57)

By neglecting the difference in the heat capacities in the solid and liquid states and substituting the triple point temperature Ttp,2 with the normal melting temperature, Tm, of component 2 at atmospheric pressure P0, following expression is obtained:

112

3 Industrial Scale Applications: Physical-Based Processes fus (

ΔH2 f subcL In 2 S = RTm,2 f2 fus

) Tm,2 −1 T

(3.58)

where ΔH2 is the heat of fusion at the normal melting temperature. Concentration of the solute and the solubility (the concentration at saturation) define supersaturation ratio. The value of the enhancement factor (ratio of the solubility of a given solid in supercritical fluid over the theoretical solubility in an ideal gas) can be as high as 105 or 106 . When the expansion pressure is close to atmospheric pressure, the fluid after expansion can be considered to be an ideal gas. The rate of nucleation as well as the inverse of the nuclei size increase with supersaturation ratio. Very high values of supersaturation ratio will therefore give a very large number of small nuclei. One possible way to control particle size is therefore to fine-tune the supersaturation by varying the drop in pressure through the nozzle. During the expansion step a drop in pressure denotes an important reduction in temperature. When considering the phase diagram, the starting point, as well as the final point must be in the single-phase zone of the diagram. The temperature before expansion must be high enough to guarantee such conditions, which may be incompatible with the stability of the extracted solute. A lower pre-expansion temperature may lead to a condensation or freezing of the supercritical fluid. To compensate this temperature decrease and to eliminate potential problems due to congestion, a heated nozzle or a heating device just upstream from the nozzle can be used. The post-expansion pressure is also significant. Manipulating processing temperature was found to influence the morphology of particles. As supercritical solutions expand across a nozzle, the pressure drop inside the nozzle can stimulate precipitation of the solute from the solution driven by nucleation, condensation and particle coagulation. The location of the initial solute condensation along the expansion path decides the particle size and morphology. Sub-micron particles are produced in case when particles begin to nucleate from solution as a homogenous solution is expanded across the nozzle. Micron-size particles formation after spraying is induced as particles begin to nucleate at the nozzle entrance and grow very quickly to form micron size particles. Formation of fibers takes place as the precipitation starts upstream of the nozzle entrance, since the solution already has crossed the cloud point. When the solution is expanded through the nozzle, particles grow and coalesce. Fibers are drawn from the nozzle exit or the nozzle is plugged completely by reason of agglomeration. Larger particles are attained by a prolonged path length after nucleation. The effect of nozzle diameter on particle size is such that the particle size increases with an increase of nozzle diameter. Development of a systematic procedure for the design and scale up of the supercritical micronization processes is a demanding task. A number of investigations were performed in order to establish mechanism of particle formation and the influence of process parameters on the properties of obtained powderous product. Some of the key parameters in these processes are pre-expansion conditions, phase behavior of expanding mixture and nozzle design, which determines the fluid mechanics of the process. A general model to predict the particle characteristics has not been developed

3.2 High Pressure Polymer Processing

113

yet. Generally the models that have been developed enable qualitative predictions of the product properties depending on process parameters. Several reviews of the investigations dealing with fundamentals and modelling were published (Martín and Cocero 2008; Reverchon and De Marco 2011). Due to all disadvantages of RESS process currently no large scale unit is in operation in industry. Some smaller apparatus could be applied mostly in pharmaceutical industry for specially substances with very specific particle size, particle shape and particle size distribution.

3.2.1.4

Gas Antisolvent Processes (GAS, GASP, SAS, PCA, SEDS)

Fundamentals Supercritical fluids can be used also as anti-solvents. After a first application in the 1920s (Zsigmondy and Bachmann 1918) the antisolvent technique was further technologically advanced in the late 1980s (Gallagher et al. 1989). The low solubility of therapeutics in dense gases is a serious omission for the large-scale production of micronised particles by the rapid expansion of supercritical solution (RESS) technique. The low solubility can also be exploited as an advantage when dense gases are used as anti-solvents. In anti-solvent processes solute is solubilized in a conventional solvent. This solution is contacted with a supercritical fluid acting as an antisolvent. Depending on pressure, temperature and mass transfer the supercritical fluid is able to remove the classical solvent. The concentration of the solute increases, reaches saturation, then supersaturation and finally nucleates and forms nano- or microsized particles. Application of sub- or supercritical fluids as anti-solvent is an alternative recrystallization technique for processing of solids that are insoluble in sub- or supercritical fluids. The most commonly used supercritical antisolvent is CO2 , but other compounds have also been used. Dense gas anti-solvent processes have been operated primarily in two approaches. A method that comprises the gradual addition of a dense gas to expand a static solution until supersaturation occurs and the solute precipitates is known as method known as gas anti-solvent system (GAS). The GAS process has been used for micronisation, fractionation, synthesis and coating of several therapeutic compounds whereas micro-particles of low and high molecular weight drugs have been produced (Foster et al. 2003a; Jung and Perrut 2001). Particle size and morphology is controlled by pressurization rate, concentration, solvent and anti-solvent system, temperature and stirring rate can influence the particle characteristics. After precipitation of the solute is achieved, addition of dense gas follows to remove the residual solvent. After depressurization a dry precipitate is attained (Dehghani and Foster 2003). Another technique involves spraying or dispersing an organic solution via a nozzle into a flowing or static dense gas and is known as aerosol solvent extraction system (ASES) (Foster et al. 2003b).In both cases, the

114

3 Industrial Scale Applications: Physical-Based Processes

organic solvent must be miscible with the dense gas at operating conditions, while the solute must be insoluble (or minimally soluble) in the dense gas. In the ASES technique, high degrees of supersaturation result in the precipitation of fine particles as the solution is sprayed into the dense gas. In this process, the higher mass transport and hydrodynamic mixing provide additional variables that may be used to control particle size, size distribution, and morphology. Precipitation using dense gas antisolvent processes requires mild operating pressures and temperatures. The technique is rapid and does not involve an additional solvent removal step. The ASES has been successfully applied to the micronisation of drug from both organic and aqueous solutions (Foster et al. 2003a) poorly water-soluble drugs in order to improve their dissolution rates (Dehghani and Foster 2003), and for the re-engineering of proteins, steroids and antibiotics. It has been found that material produced by gas anti-solvent processes have residual solvent levels below the limits specified by the United States Pharmacopoeia (USP). There have also been numerous investigations of the feasibility of utilizing the ASES process for the micronisation of fragile and heat labile molecules, such as proteins which are difficult to process by conventional techniques. Gas anti-solvent processes (GASR, gas anti-solvent recrystallization; GASP, gas anti-solvent precipitation; SAS, supercritical anti-solvent fractionation; PCA, precipitation with a compressed fluid anti-solvent; SEDS, solution-enhanced dispersion of solids) differ in the way the contact between solution and anti-solvent is achieved. This may be by spraying the solution into a supercritical gas, or by spraying the gas into the liquid solution (Knez and Weidner 2003). These processes are characterised by very mild conditions of temperature and smaller particles are obtained when compared to the common industrial comminution techniques like jet milling, The process is based on the phenomenom of liquid antisolvent precipitation and crystallisation and on the ability of sub- or supercritical fluids to solubilize in conventional solvents. By that, the solvent power of the solvents for compounds in solution is decreased. Due to lower solvent power of solvent extremely high super saturation could be achieved and therefore solids particles of solute precipitate.

Design Criteria Depending on process parameters that influence the supersaturation and the nucleation rate, the particle size, particle shape and particle size distribution could be varied over a wide range. Technically in these processes conventional solvent is saturated with the substance which should be micronized in an autoclave. To saturated solution, sub- or supercritical fluids is introduced in an autoclave where precipitation of solute is performed. Later gas saturated solution is removed, while precipitated particles remain on a filter in the autoclave. Solution could be recycled, while the solvent adsorbed on the particles could be removed either by drying process or extracted by sub- or supercritical fluid (Fig. 3.34).

3.2 High Pressure Polymer Processing

115

Fig. 3.34 Scheme of experimental apparatus for anti-solvent process

Mass transfer limits the whole recrystallization process using sub- or supercritical fluids. In anti-solvent recrystallization process precipitation consists homogeneous phase, similar as in conventional recrystallization process, consists from initial nucleation stage, where nucleus is formed and subsequent growth of the formed nuclei. The rate of formation of nuclei can be defined by equation: J = Zexp(ΔGmax /R · T )

(3.59)

where J is production rate of nuclei (number of nuclei per unit time and per volume), Z is collision frequency—can be calculated from conventional kinetic theory, Gmax is the Gibbs free energy, therefore it is clear that the particle size and particle size distribution should be varied and regulated via the rate and mode of addition of anti-solvent fluid (sub- or supercritical) to the solution. A numerous anti-solvent processes that are different in way of contacting solution and sub- or supercritical fluids have been developed and patented (Hickey and Giovagnoli 2018). Thermodynamics of anti-solvent process is relatively more sophisticated compared to RESS process, while 3 phases are present (solid, liquid and gas). Modelling of supercritical antisolvent processes has been reviewed recently by (Martín and Cocero 2008; Tabernero et al. 2012). In the following section jus main characteristics of the processes will be emphasized. In GAS process the most important parameter is the volumetric expansion based on the variation of the partial molar volume of the solvent (de la Fuente Badilla et al. 2000). Based on relative molar volume expansion the optimum combination of solvent, pressure and temperature can be determined. The optimum conditions for

116

3 Industrial Scale Applications: Physical-Based Processes

the GAS process are located at the minimum of the volumetric expansion curve of the solvent (de la Fuente Badilla et al. 2000; de la Fuente et al. 2004). For the design of SAS, SEDS and ASES processes the knowledge about the position of mixture critical point (CP) is crucial. Reverchon et al. (2003a) found that when operating at pressures above the mixture CP, the mixing parameters (type of nozzle and precipitator) are not influencing the precipitation, while at pressures below mixture CP these parameters are influencing the precipitation significantly. When operating above the mixture CP the initial concentration of the solution is the most influential parameter and by scale-up of the process the fluid mechanics parameters including nozzle and precipitator design have limited influence on the process (Reverchon et al. 2003b). It was observed that the precipitation mechanism and particle characteristics are different when operating above or below mixture CP (De Gioannis et al. 2004; Reverchon et al. 2007, 2010; Reverchon and De Marco 2004; Tavares Cardoso et al. 2008; Tenorio et al. 2007; Weber et al. 2005). Numerous studies on Modelling of SAS process were performed including investigations of different jet disintegration regimes at high pressures, mass-transfer between the droplet and the surrounded atmosphere after atomization, influence of buoyancy effect and thermodynamics. One of the most complete studies of SAS was performed by Martin et al. (Mart´ın and Cocero 2004) and a mathematical model was developed, which represents phase equilibrium, mass transfer, jet hydrodynamics and crystallization kinetics simultaneously. The model has been used to analyze the mechanism of particle formation in the SAS process, and to study the effects of the operating parameters on the particle size and solid recovery. The work contributes to understanding of the SAS process; the model developed is able to describe the main trends, e.g. the variation of particle size with the operating parameters, however it fails to predict the mean particle size. In recent study published by Reverchon et al. (2010) it was investigated how phase equilibria, jet mixing and mass transfer interactions are influencing the morphology of the particles produced by SAS. In the work of Marra et al. (2012) jet break –up time and of dynamic surface tension vanishing have been considered as mechanisms in competition during supercritical antisolvent precipitation and a mathematical model, based on these two characteristic phenomena, has been presented.

3.2.1.5

Particles from Gas Saturated Solutions (PGSS™ )

Process Description This process allows to form particles from substances that are non soluble in supercritical fluid, but absorb a large amount of gas that either swells the substance or decreases the melting point (for polymers glass transition temperature). This process can also be used for micronization of liquids, suspensions and emulsions. In PGSS™ process compressible media is solubilized in the substance or mixture of substances (pure substance, solution, dispersion, emulsion, …) which has to be micronized (Knez et al. 1995). Several substances which are practically insoluble in

3.2 High Pressure Polymer Processing

117

sub-critical or supercritical fluids dissolve considerable mass of gasses in the liquid phase. For solid substances it was observed that in the presence of gas, melting point of several substances decreases with increasing pressure of gas due to solubilization of gas in the solid. By expansion of gas saturated solution in an expansion with the compressible media is evaporated and due to Joule–Thomson effect, the solution is cooled below melting point. Technically for batch wise operated plant the substance to be micronized is filled in an autoclave, later the gas is loaded and the system is equilibrated (gas is solubilized in the substances or mixture of substances). The gas saturated system is expanded via a nozzle. In a continuous operated plant the substance to be powdered (molten or liquid, emulsion or suspension) is feed to static mixer, where it is mixed with subcritical or super-critical fluid. After mixing the multicomponent system is expanded via nozzle. The basic flow sheet of batch and continuous operated equipments are presented in Figs. 3.5 and 3.6. Produced particles, which are of micron size, are easily separated from gas stream in a spray tower cyclone. The PGSS™ process could be used for production of particles of pure substances, but the main advantage of process is production of composites of miscible or even immiscible substances. In this way frozen emulsions or liquid filled particles could be produced. As mentioned before, PGSS™ process could operate batch wise or continuously and today several plants from the lab up to industrial scale are in use in different industry. As mentioned before, in PGSS™ process the substance or the mixture of substances to be powdered must be converted into a sprayable form by liquefaction/dissolution. This can be achieved by melting or/and dissolving of the substance or mixture of substances in a liquid solvent, or by dispersing solids or liquids in a melt or solution, and saturation of the melt/solution/dispersion with the gas. Then the gas-containing solution is rapidly expanded in an expansion unit and the gas is evaporated. Owing to the Joule–Thomson effect and/or the evaporation and the volumeexpansion of the gas, the solution cools down below the solidification temperature of the solute, and fine particles are formed. The solute is separated and fractionated from the gas stream by a cyclone and electro-filter.

Design Criteria The process allows a careful processing in an inert gas atmosphere at low temperatures, moderate pressures and is suitable for producing powders and composites of solids, very viscous melts and even liquid substances. In PGSS™ process the content of the gas in the heavy phase is of great importance for process design. A clear influence of the pre-expansion pressure and the temperature in the spray tower on the particle size distribution of the product powders was found. With increasing temperatures in the spray tower, the solidification process of the particles takes longer than at lower temperatures. Agglomerated particles were observed at high temperatures in the spray tower. The effect of the pre-expansion pressure is more complex. A high

118

3 Industrial Scale Applications: Physical-Based Processes

pre-expansion pressure leads to a strong Joule–Thomson phenomenon. Droplets of the melted shell material and the liquid dispersed inside are cooled down faster at higher than at lower pressures. So at higher pressure smaller particles are formed and less or no agglomeration occurs. When the liquefaction is achieved by melting, the knowledge of the P–T trace of the S-L-V equilibrium curve gives information on the pressure needed to melt the substance to be micronized and form a liquid phase at a given temperature, and to calculate its composition (Fukné-Kokot et al. 2003). When the supercritical fluid has a relatively high solubility in the molten heavy component, the S-L-V curve can have a negative dP/dT slope. The second type of three-phase S-L-V curve shows a temperature minimum (Tuminello et al. 1995). In the third type, where the S-L-V curve has a positive dP/dT slope, the supercritical fluid is only slightly soluble in the molten heavy component, and therefore the increase of hydrostatic pressure will raise the melting temperature and a new type of three-phase curve with a temperature minimum and maximum may occur (Weidner et al. 1997). In general, for substances for which the liquefaction is achieved by melting the systems with a negative dP/dT slope or with a temperature-minimum in the S-LV curve could be processed by PGSS™ (Jung and Perrut 2001; Knez and Weidner 2001). Basic flow sheet of batch operated apparatus is presented on Fig. 3.35. In Fig. 3.36 the basic scheme of continuous operated PGSS™ apparatus is presented. The PGSS™ process was tested in the pilot- and technical size on various classes of substances. Up to the present time the application of the PGSS™ process has been investigated for polymers, waxes and resins, natural products, fats and fat derivatives, pharmaceuticals, synthetic and natural antioxidants, surface-active compounds, UVstabilizers, etc. (Couto et al. 2017; Knez et al. 2019; Temelli 2018). The highly compressible fluids which have been used were carbon dioxide, propane, butane, dimethyl ether, freons, nitrogen, alcohols, esters, ethers, ketones and mixtures of above-mentioned gasses and solvents. The powders produced show narrow particle-size distributions, and have improved properties compared to the conventional produced powders. The characteristics (properties) of the substance to be micronised (crystallineamorphous, pure or composite), the process parameters (pre-expansion pressure, temperature, gas to substance ratio (GSR), viscosity of melt/solution/dispersion) of the PGSS™ process and geometry of the process equipment influence particle size, particle size distribution, bulk density, the morphology (particle shape) and ratios crystalline/amorphous of the processed substances. There are several advantages of PGSS™ process compared to HP crystallization, RESS and anti-solvent processes like moderate operating pressure and temperature, very low gas consumption, separation of formed solid particles from gas stream is easy, production of composites is possible, process could operate batch wise or continuously, scale up is relatively simple and low investment costs (Knez et al. 2014). Disadvantage is that mostly particles of micron size and bigger could be produced, while production of submicron sized particles is nearly impossible.

3.2 High Pressure Polymer Processing

Fig. 3.35 Flow sheet of batch operated PGSS™ apparatus Fig. 3.36 Basic scheme of continuous operated apparatus for PGSS™ micronisation

119

120

3 Industrial Scale Applications: Physical-Based Processes

Due to the low operating costs and low investment costs, the process is applied in different industries. There are some modifications of PGSS™ process like CPF process (de Paz et al. 2012; Li et al. 2012).

Modelling In the PGSS™ processes supercritical CO2 saturated solution is expanded through a nozzle to ambient pressure and due to evaporation of CO2 and Joule–Thomson effect a large cooling effect occurs what leads to formation of a product, which can be in a form of solid particles or liquid droplets. The obtained particles can have different size distribution (narrow or wide) and can have different morphology, from porous structures, spheres, sponges, fibers, what depends on the pre-expansion and afterexpansion conditions, ratio of solvent/saturated solution flow rates, nozzle diameter and particle formation mechanism. Empirical correlations to determine the particle morphology and bulk density as a function of process conditions were published by Kappler et al. (2003). To design particles with specific characteristics it is crucial to understand many different phenomena, including thermodynamics, nozzle hydrodynamics, droplet fluid dynamics, crystallization kinetics, bubble formation and droplet coalescence (Strumendo et al. 2007). Thermodynamics A first thermodynamic model for calculating the enthalpy variation of PGSS process was developed by Elvassore et al. (2003) aiming to be used to interpret the influence of temperature and pressure on the properties of product. Assuming that changes of potential and kinetic energy can be neglected the energy balance for adiabatic expansion of CO2 saturated solution through the nozzle according to Elvasore et al. is: Hsol = x1 H1 + x2 H2

(3.60)

where H sol is the enthalpy of saturated solution before expansion and the sum on the right side of equation presents the enthalpy of the final state after expansion. H 1 and H 2 are the enthalpies of pure CO2 (1) and pure heavy compound (2), respectively and x 1 and x 2 are the mole fractions of both components in the starting mixture. For solving Eq. (3.1) proper equation of state has to be used. Elvassore et al. (2003) coupled this energy balance with a PHSC EoS and constructed P–T operating charts for CO2 -lipid systems showing three phase S-L-V line for a compound to be micronized in presence of CO2 and regions of pre-expansion conditions at which solid, partially solid or liquid product is obtained after expansion through a nozzle. Enthalpy of solution is calculated from the ideal gas contribution and the residual enthalpy:

3.2 High Pressure Polymer Processing

121

Hsol =

∑

R xi HiIG + Hsol

(3.61)

i R where HiIG is the ideal gas enthalpy of component i and H sol and Hsol are the enthalpy and residual enthalpy of the solution, respectively. Ideal gas enthalpy of component i is calculated as:

∫T HiIG =

IG cpi (T )dT

(3.62)

Tm

where cpIGi is the ideal gas heat capacity of component i. The normal melting point T m of solid component is used as reference for zero enthalpy. The residual enthalpy is calculated as: ( R ) ∂Asol R R (3.63) Hsol = PVsol + ARsol − T ∂T V,xi R where Vsol is the solution residual volume obtained from EOS and ARsol is the solution residual Helmholtz free-energy:

ARsol = RT

∑

R xi ln ϕ i − PVsol

(3.64)

i

where ϕ i is the partial molar fugacity coefficient obtained from EOS. Assuming that after the expansion of saturated solution to atmospheric pressure, compounds 1 and 2 are completely separated, enthalpy of final state can be calculated. Enthalpy of CO2 (compound 1) at final conditions is derived assuming that CO2 follows an ideal behavior. The enthalpy of compound 2 at normal melting temperature (at 1 bar) depends on the fraction of solid phase in the final product ϕS : formation

H2 = ϕS ΔH2

+ H2L,sat

(3.65)

where H2L,sat is the enthalpy of saturated liquid 2 at normal melting temperature and formation ΔH2 is the formation enthalpy of 2. The enthalpy of the saturated liquid 2 is calculated by eq. (X.17) at normal melting temperature and 1 bar. The enthalpy of sub-cooled solid product is evaluated by subtracting: sub-cooled ΔHsystem

∫T ( ) IG S x1 cp,1 = (T ) + x2 cp,2 (T ) dT

(3.66)

Tm

from the enthalpy of the final product. In eq. (X.22) cp,i (T ) is the heat capacity of ideal CO2 and solid compound 2.

122

3 Industrial Scale Applications: Physical-Based Processes

In the work of Calderone et al. (2007) the thermodynamic balance for the PGSS micronization of CO2 saturated fat melt (Precirol—mixture of mono-, di- and triglyceride) in excess of supercritical CO2 was performed. The expansion was found to be isenthalpic m ˙ hin = m ˙ hout

(3.67)

where m ˙ is mass flow rate and h enthalpy per unit of mass and the final relation derived was: ] [ w1 h1V (TF , Patm ) − hSC 1 (TSC , PSC ) − SΔhdiss (TSC , PSC ) ) )( ) ( [( + (1 − w1 ) 1 − f2L Δhcrist − cp,L (TSC − Tm ) − 1 − f2L Δcp,m ] (3.68) (TF − Tm ) − v2L (PSC − Patm ) = 0 ( )) ( where: f2L is the mass fraction of liquid fat after expansion f2L = mL2 / mL2 + mS2 , Δhcrist and Tm are crystallization enthalpy and melting temperature of fat, Δhdiss is the dissolution enthalpy per unit mass of CO2 , w1 is the mass ratio of CO2 , v2L constant specific volume of melted fat, subscripts F and SC denote final state an supercritical conditions, S is the mass ratio of dissolved CO2 compared to the total CO2 . Based on the obtained results “operating charts” were constructed which show a thermodynamic link between initial and final states and give information in order to anticipate the final state of the lipid after its expansion, versus the quantity of carbon dioxide expanded and when initial conditions in the vessel are known. The second point of the investigation by Calderone et al. (2007) was based on the assumption that crystallization takes place after expansion at atmospheric conditions and so classical melt media theory was applied and the kinetics of the crystallization of the fat was investigated. Driving force, critical radius of the formed nuclei above which the nucleus is stable and will grow and nucleation rates were estimated. Nozzle Hydrodynamics and Atomization Li et al. (2004) introduced a mathematical model of a PGSS process based on assumption of one-dimensional homogeneous two-phase flow of the CO2 -rich phase and the mixed CO2 /hydrogenated palm oil (HPO) phase through a capillary nozzle. Model was derived by the coupled resolution of: – mass, energy and momentum equations describing pressure, temperature, velocity and density change along the nozzle, – the aerosol GDE to model particle formation by nucleation, and growth by condensation and coagulation and – Peng-Robinson EOS for phase equilibrium. The model was applied to study the precipitation of HPO. The simulation results showed that relatively small particles with narrow size distribution may be obtained,

3.2 High Pressure Polymer Processing

123

and that pre-expansion temperature, pre-expansion pressure and flow rate ratio CO2 / HPO influence the mean particle size. Limits of pre-expansion temperature and pressure were observed where no particles are produced by crystallization. Higher flow rate ratios CO2 /HPO produced particles with smaller diameter. Furthermore it was observed, that larger nozzle diameters produce slightly larger particles; however this effect is not very strong, what suggests that increasing the nozzle diameter is a rather convenient choice while larger capacities can be obtained (g powder product/min). However particle sizes predicted by the simulation were several orders of magnitude smaller than those obtained by experiments, what could be explained by the fact that the simulation is carried out at the nozzle only and it does not take into account the particle coagulation after the nozzle. A year later Li et al. (2005) developed a general model based on both the atomization and the crystallization mechanisms that considers both the melt crystallization and the gas-solution crystallization. The model was applied to study the precipitation of HPO. Based on results it could be concluded that particles produced by PGSS can have unimodal or bimodal distribution, depending on the prevailing particle formation mechanisms at different operating conditions. A narrow size distribution of small particles at nozzle exit is obtained by melt crystallization, while a wide size distribution of large particles is obtained by atomization. The operating conditions with small nozzle diameters or high pre-expansion temperatures or mild pre-expansion pressures tend to produce bimodal particle distribution curves. However the calculated particle sizes were still about 30 times smaller than the experimental sizes. Temperature distribution in the CO2 jet during depressurization was analyzed recently by Pollak et al. (2011) by heat imaging method. Based on investigations a simple model was introduced for estimating the temperature field in a polymer droplet of polybutylene terephthalate (PBT) during spray process which provides a plausible explanation for the formation of differently shaped particles by PGSS-process. Isolated Droplet Fluid Dynamics, Evaporation and Solidification Strumendo et al. (2007) investigated the fluid dynamics of an isolated droplet of a tristearin and CO2 solution in stagnant CO2 atmosphere, considering mass and heat transfer phenomena and the CO2 evaporation until the solidification of the droplet. Equilibrium is computed with PHSC EoS and Modelling approach follows the concepts of the Stefan problem formulation for the solution of moving boundary problem. However, convective motion, gravity and droplet interactions are not taking into account. The model was simultaneously solved until S-L equilibria at the droplet boundary is reached, which is given by the melting point temperature curve as a function of dissolved gas content. The results showed that mass transport in the gas saturated solution is much slower than heat conduction and at the final time internal parts of the droplet still contain initial amount of the CO2 , while temperature has changed in the entire droplet and the temperature profile within the droplet is relatively constant. Time for reaching S-L equilibria increases with increasing initial droplet temperature and decreases with increasing initial mass fraction of CO2 , what is in agreement with the experimental data which showed that particle size decreases

124

3 Industrial Scale Applications: Physical-Based Processes

at increasing pre-expansion pressure and decreasing pre-expansion temperatures. The simulation is stopped, when S-L equilibria is achieved so this model does not takes into account solid phase growth and crystallization kinetics and CO2 bubble nucleation inside the droplet, which would provide valuable information about the morphology of particles and particle design.

3.2.1.6

PGSS™ -Drying

Fundamentals of PGSS-drying process were studied by Martin and Weidner (2010) based on experimental results of micronization of polyethylene glycol from aqueous solutions. Analysis includes mass and energy balances, phase equilibrium conditions, mass transfer rates and atomization mechanisms. It has been shown that a global mass balance can be used to calculate the minimum gas/liquid flow ratio required for producing dry powder. Some differences have been observed in concentration of moisture in powder obtained experimentally and calculated based on the mass balance and the phase equilibrium, which have been attributed to the kinetic factors not considered in mass balance calculations, i.e. the evolution of pressure and temperature along the expansion path. Furthermore a global energy balance was used for estimation of spray tower temperature as a function of pre-expansion conditions, and the results were in agreement with experimentally measured temperatures. It has been proposed that atomization of the biphasic mixture leaving the static mixture occurs by two simultaneous processes: flash-boiling atomization and effervescent atomization. Based on this mechanism, variation of particle size with process conditions observed in experiments can be explained as a function of CO2 concentration in the liquid phase: process conditions which cause an increase of CO2 concentration in the liquid phase, promote a more efficient atomization and therefore a reduction of particle size.

3.2.1.7

Example—Design of PGSS™ Micronisation Process of Polyethylene Glycol

For the design of micronisation of polyethylene glycol several process parameters have to be studied to obtain particles with defined particle size, particle shape, particle size distribution and crystallinity. In the first step as shown in previous chapters’ solubility of gas in polymer melt and melting point depression of polyethylene glycol has to be determined. Based on experimental techniques as presented in Chap. 2 melting point depression and solubility of CO2 in polyethylene of different molar masses were determined. All obtained data are summarized in Fig. 3.37. Reduced temperature is defined as: Reduced temperature = melting point Temp.(Gas, temp.)/Tfus

3.2 High Pressure Polymer Processing

125

Fig. 3.37 General P-Tred diagram of the S-L-G line with isopleths for PEG 1500-35000/CO2 (Bertucco and Vetter 2001)

Based on the data in Fig. 3.37 process parameters could be determined pressure and operating temperature. This process parameters are usually at the highest solubility of gas in the melt/suspension/solution and at pressure where the melting point line has its minimum. In early stage of research, it was evident that the energy balances play an important role to obtain solid particles or liquid droplets in the spray tower. Therefore, a process simulation program for establishing the energy balance was developed. The details of the simulation program as well the thermodynamic model for evaluation of heat balances are presented in Figs. 3.38 and 3.39. In this early stage of development of micronisation process of polyethylene glycol b PGSS™ it was found that the process parameters highly influence the particle shape, particle size and particle size distribution as could be seen on Fig. 3.40. Therefore, an extended study on the influence of gas solubility in polymer melt, the influence of solubility of gas on the viscosity and surface tension of gas saturated melt on particle size, particle shape and bulk density of obtained powder were performed. Investigation of thermodynamic and transport properties of binary systems polyethylene polymer/CO2 and correlation to particle size, particle shape and bulk density was a topic under an intense research by Kappler et al. (2003) and Kravanja (2018). Kappler developed a regression function (F) that can provide a certain type of PEG 6000 particle morphology as follows:

126

3 Industrial Scale Applications: Physical-Based Processes

Fig. 3.38 Flow scheme for establishing the energy balance by a process simulation program

F = 6.9 · 10−11 ·

T [K]4.247 p[bar]0.403 · GTP 0.105

(3.69)

where T is processing temperature, p is processing pressure and GTP is gas to product ratio (CO2 mass flow/polymer mass flow). Relative deviation of results obtained was 14%, at a temperature range of 323 < T < 383 and pressure range of 5 bar < p < 350 bar. Different particle shapes are formed, shapes are formed, depending on the processing conditions applied. In study (Kravanja 2018) a PEG 6000/CO2 system was used as a model in order to investigate the basic thermodynamic and transport data including density, viscosity, and interfacial tension of a CO2 saturated PEG 6000 solution; to provide better insights for the design of a particle formation process. Density, viscosity, interfacial tension of CO2 saturated PEG 6000 solution were determined at four temperatures: 60, 70, 80 and 90 °C in the pressure range from ambient pressure up to 350 bar. Additionally, particle size and particle morphology (obtained by PGSS™ ) were correlated for the first time with the interfacial tension and viscosity of a PEG 6000 CO2 saturated solution.

Density of CO2 Saturated PEG 6000 Solution To ensure the reliability of the measurements, the density of pure carbon dioxide was determined beforehand at different temperatures and compared with NIST values (NIST Office of Data 2019). The deviation was about 2% over the entire pressure range from 50 bar up to 200 bar. Density for all observed systems increases linearly with increasing pressure as a consequence of liquid compression and low

3.2 High Pressure Polymer Processing

Fig. 3.39 The thermodynamic model of PGSS™ process

Fig. 3.40 Morphology of PGSS™ micronised polyethylenglycols (Bertucco and Vetter 2001)

127

128

3 Industrial Scale Applications: Physical-Based Processes

Fig. 3.41 Density of CO2 saturated PEG 6000 as a function of temperature and pressure. https:// doi.org/10.1016/j.supflu.2018.05.012

Table 3.5 Constants a polynomic function for density calculations, R2 of a function is 0.9838 a

b

c

1.09

−6.78 ·

10–3

4.96 ·

d 10–3

4.66 ·

e 10–4

7.49 · 10–4

gas solubility in the polymer matrix (Ilieva et al. 2016). Oppositely, density is significantly reduced with the increasing temperature at isobaric conditions. To illustrate the dependence of pressure and temperature on the density, a three-dimensional plot is presented in Fig. 3.41. A polynomic function was used to describe the trend with more than 95% confidence bounds at the temperature range of 50 °C < T < 90 °C and pressure range of 50 bar < p < 350 bar (Eq. 3.19). Constants of a simple polynomic function for calculating of density CO2 saturated PEG 6000 solution are shown in Table 3.5. [ ] ρ g/ml = a + b · T + c · p + d · T 2 + e · T · p

(3.70)

The results obtained by the vibrating U-tube densitometer show patterns that are consistent measurements of densities by the external balance method (Knez Hrnˇciˇc et al. 2014b) and by other gravimetrical methods involving a magnetic suspension balance. Additionally, densities obtained in our study fit closely to those of recently published by Avelino et al. (2017) and by Gourgouillion et al. (1998). The difference in results could be due to the different method used for density measurements.

3.2 High Pressure Polymer Processing

129

Table 3.6 Constants for interfacial tension of calculations, R2 of a function is 0.997 a

b

c

d

E

f

64.77

−9.627 · 10–2

−0.3067

8.56 · 10–5

9.57 · 10–4

4.941 · 10–8

Interfacial Tension of CO2 Saturated PEG 6000 Solution Calculation of interfacial tension using experimental data obtained by the capillary rise method requires previous data on the density of CO2 saturated PEG 6000 and the density of pure CO2 . It can be seen that pressure has a significant effect on interfacial tension reduction, on the other hand, the temperature effect is minimal. Interfacial tension is reduced by elevated pressure. For instance, at a constant temperature of 80 °C, interfacial tension decreases from 21.28 mN/m at 42 bar to 3.46 mN/m at 300 bar. This could be a consequence of CO2 solubility in the polymer matrix as pressure increases. At lower pressures, the decrease was sharper, and the opposite was observed at higher pressures (p > 100 bar); the interfacial tension decrease became slower and finally vanished and inclined asymptotically to a constant value (p > 150 bar). The interfacial tension values presented are in good agreement with those of Harrison et al. (1996), measured by the pendant drop method. To illustrate the dependence of pressure and temperature on interfacial tension, a three-dimensional plot is presented in Fig. 3.12. A polynomic function (Eq. 3.20) was used to describe the trend with more than 95% confidence bounds at the temperature range of 50 °C < T < 90 °C and pressure range of 5 bar < p < 350 bar. The aforementioned constants for calculating the interfacial tension CO2 saturated PEG 6000 solution by a polynomic function are shown in Table 3.6 and calculated results are presented in Fig. 3.42. ITF[mN/m] = a − b · T − c · p + d · T · p + e · p2 + f · T · p2

(3.71)

Viscosity of CO2 Saturated PEG 6000 Solution Viscosity measurements were conducted at four temperatures: 60, 70, 80 and 90 °C at a pressure range from 100 bar up to 350 bar for PEG 6000 in carbon dioxide. After PEG 6000 is saturated with CO2 at 90 °C, its viscosity decreases from 652 mPas at atmospheric pressure to 2.24 mPas at 200 bar. Further addition of CO2 and increasing pressure result in even lower viscosity. As expected, the highest viscosity reduction was reached at the highest investigated pressure and temperature; at 350 bar and 90 °C the viscosity of the system, PEG 1500/CO2 is only 0.995 mPas. Temperature variation has a significant effect on viscosity. With increasing temperature, a rapid reduction in viscosity can be achieved. However, that is not an optimal choice for viscosity reduction since it leads to polymer degradation. The highest viscosity reduction was achieved at the lowest temperatures.

130

3 Industrial Scale Applications: Physical-Based Processes

Fig. 3.42 Interfacial tension of CO2 saturated solution of PEG 6000 as a function of temperature and pressure. https://doi.org/10.1016/j.supflu.2018.05.012

Table 3.7 Constants of a polynomic function for viscosity calculations, R2 of a function is 0.9877 a 12.62

b 2.78 ·

c 10–2

7.5 ·

d 10–3

5.08 ·

E 10–6

3.48 · 10–6

To illustrate the dependence of pressure and temperature on viscosity, a threedimensional plot is presented in Fig. 3.13. A polynomic function (Eq. 3.21) was used to describe the trend with more than 95% confidence bounds at the temperature range of 50 °C < T < 90 °C and pressure range of 5 bar < p < 350 bar. The abovementioned constants for calculating the viscosity CO2 saturated PEG 6000 solution with a polynomic function are shown in Table 3.7 and calculated data are presented in Fig. 3.43. viscosity[mPas] = a + b · T + c · p + d · T · p + e · p2

(3.72)

These data provide a good comparison with our previously published results after testing a PEG/CO2 system of different molecular weight at 70 °C using the same method (Knez Hrnˇciˇc et al. 2014b). The high-pressure method is considered as rigorous, but there are still problems relating to accuracy. The mixing rate deviated to a small extent and as in several other methods, accurate viscosity measurements and the prevention of gas leakage are still major challenges.

3.2 High Pressure Polymer Processing

131

Fig. 3.43 Viscosity of a CO2 saturated solution of PEG 6000 as a function of temperature and pressure. https://doi.org/10.1016/j.supflu.2018.05.012

Interfacial Tension and Viscosity at Different Spraying Pressure and Temperature Figure 3.44 presents the interfacial tension of CO2 saturated PEG 6000 solution at different spraying pressure and temperature conditions. It can be seen that interfacial tension is significantly low, varying from 13.43 mN/m at 100 bar down to 5.02 mN/m at 250 bar. The interfacial tension of a CO2 saturated PEG 6000 solution is reduced to 37% with a pressure change from 100 bar up to 350 bar at a constant temperature of 60 °C. Effect of a temperature on interfacial tension is minor. On the other hand, the pre-expansion temperature has a strong influence on morphology (Martín and Weidner 2010). Spheres are formed at high pre-expansion temperatures and at lower mixing pre-expansion pressures. When obtaining micro-foams, at lower temperatures around 70 °C the pressure range is much larger. Particle size is reduced when the temperature is decreased and pressure is increased (Yeo and Kiran 2005). That indicates that higher interfacial tension boosts the formation of smaller particle droplets after spraying through the nozzle. Compared to interfacial tension, the pre-expansion temperature has a great effect on viscosity, as it does for particle morphology and size. Figure 3.15 presents the viscosity of CO2 saturated PEG 6000 solution at different spraying pressure and temperature conditions. Viscosity is significantly low, varying from 2.8 mPas at 100 bar down to 2.0 mPas. Additionally, when the pre-expansion temperature is increased, viscosity decreases even more. Spheres are formed under pressure and temperature conditions that are related to the highest viscosity of the system. As the viscosity is reduced particles of smaller mean size are expected. Like interfacial

132

3 Industrial Scale Applications: Physical-Based Processes

Fig. 3.44 Influence of interfacial tension of CO2 saturated PEG 6000 melt at different spraying pressure and temperature on particle shape and size, GTP (Gas to Product) = 1. https://doi.org/ 10.1016/j.supflu.2018. 05.012

Fig. 3.45 Influence of viscosity of CO2 saturated PEG 6000 melt at different spraying pressure and temperature on particle shape and size, GTP (Gas to Product) = 1. https://doi.org/ 10.1016/j.supflu.2018. 05.012

tension and viscosity, the density of CO2 saturated PEG 6000 solution is also reduced when the pre-expansion temperature is increased (Fig. 3.45).

Economy of the PGSS™ Process Basic economy for PGSS™ process was calculated for micronisation of pure substance with no gas added to the system. In this calculation it was foreseen that the investment are 16%, labour 40%, energy 4% and gas 40% of total costs. As could be seen from Fig. 3.46 the total processing costs are under 1.0 Euro/kg of processed powder. With increasing capacity of PGSS™ plant from 200 kg/h to 500 kg/h and

3.2 High Pressure Polymer Processing

133

Fig. 3.46 Production costs of PGSS™ process for pure substances (Weidner 2009)

with increasing annual operating time the processing costs should be significantly decreased.

3.2.1.8

Micronisation Using Sub- and Supercritical Fluids—Conclusions/Future Trends

Micronisation using sub-critical or super-critical fluid is still subject of intensive research. There are already several plants on industrial scale in operation, compared to several hundred commercial high pressure extraction units, the number of micronisation units is relatively low. There is certain time gap before research is converted to industrial application, and based on developments in area of high pressure extraction we could be sure that the number of high pressure micronisation units will increase in the near future. Main advantage of the use of sub-critical or super-critical fluids for production of fine particles is the tunability of solvent properties. The unique thermodynamics and fluid dynamic properties of subcritical or supercritical fluids can be used for formation of products with unique customer designed properties for the use in different applications. Micronisation processes could be easily connected to subcritical or supercritical extraction processes, or to a downstream processing for products of chemical or biochemical synthesis in subcritical or supercritical fluids. Use of CO2 also prevents oxidation of products during processing steps, processing of substances using PGSS™ process could be performed even below their melting point.

134

3 Industrial Scale Applications: Physical-Based Processes

Table 3.8 Technological features of CSS, RESS, GAS/SAS/PCA and PGSS™ process Establishing gas-containing solution

CSS

RESS

GAS/SAS/PCA

PGSS™

Discontinuous

Discontinuous

Semicontinuous

Continuous

Gas demand

Low

High

Medium

Low

Pressure

High

High

Low to medium

Low to medium

Solvent

None

None

Yes

None

Volume of pressurized equipment

Small to large

Large

Medium to large

Small

Separation gas/solid

Easy

Difficult

Easy

Easy

Separation gas/solvent

Not required

Not required

Difficult

Not required

The other reasons for the fast development in the field of application of subcritical and supercritical fluids are restrictions regarding the use of conventional organic solvents in food, feed and pharmaceutical industry. Technological features of different micronisation processes using subcritical or supercritical fluids (CSS, RESS, GAS/SAS/PCA and PGSS™ ) are presented in Table 3.8. Water is the cheapest solvent, but due to the high polarity, the use of water as recrystallization solvent is limited and therefore the second cheapest solvent with pressure and temperature unable properties is carbon dioxide. Beside the tunability of thermodynamic and fluid dynamic properties and low price of CO2 , CO2 has also several other advantages (Badens et al. 2018). Therefore we can be sure that advances in the field of micronisation and formulation processes using subcritical and supercritical fluids will in the near future open up new ways for substances produced at an industrial scale.

3.2.2 Polymer Foaming with Subcritical or Supercritical Fluids By the definition, foams are multiphase materials, where one of the phases is in gaseous state. Due to their cellular structure, the foams are materials with wide range of applications. They can be used for thermal and sound insulation, as structural materials, as packing materials, airplane and automotive parts, as microelectronic and optical devices, as scaffolds for tissue engineering, as drug formulations with controlled release (Eaves 2004; Forest et al. 2015; Horvat et al. 2022; Ibeh and Bubacz 2008; Kravanja et al. 2022; Mi et al. 2013; Shinde et al. 2013). How the structure of the foams and pores inside differs, also the field of the application differs (Di Maio and Kiran 2018). Nowadays, polymeric foams represent the major part of the overall market. For example, polymer foams comprised 19.1 million tonnes in 2013, which represents

3.2 High Pressure Polymer Processing

135

nearly $87 billion. By 2019, the market is expected to consume 25.1 million tonnes (Global Information 2019). Polyurethanes, polystyrene, polypropylene and polyethylene foams are the most spread. However, biopolymer foams, such as poly (lactic acid), cellulose, starch are finding their way to the market as well.

3.2.2.1

Basics of the Foaming Process

For the production of porous polymers, supercritical fluids (mainly CO2 and N2 ) can be used as physical blowing agents, which could substantially improve the conventional manufacturing techniques (Tomasko et al. 2009; Wong et al. 2014). By this way, porous polymers can be produced through the gas foaming method. The foaming process consists out of two basic steps. (1) The process begins with the saturation of the polymer with supercritical fluid at constant temperature and pressure conditions. It usually involves sorption or dissolution of carbon dioxide, where a polymer/gas solution is formed. (2) The process continues either by rapidly increasing temperature (temperature induced phase separation) or reducing pressure (pressure induced phase separation), where the system is induced to a supersaturated state. In this moment, the nucleation and growth of gas bubbles occurs, forming the cells inside the polymer matrix (Goel and Beckman 1994; Kumar and Suh 1990). The foaming continues until the viscosity of the polymer is so high that the force opposing the expansion of the foam is sufficient enough. In the other case, the foaming continues until the system passes from the rubbery to the glassy state (Tsivintzelis et al. 2016). Figure 3.47 shows a scheme of the two-phase foaming process. As mentioned, the most commonly used supercritical fluid for the polymer foaming is CO2 due to its unique properties. First of all, the diffusivity of CO2 in polymer melt is large, which ensures a quick mixing process. Introducing CO2 as a foaming agent targets two large-scale applications: low-density insulation foams and high-density microcellular foams. Lastly, CO2 is environmentally benign and costeffective (Tomasko et al. 2003). Furthermore, foaming of polymers with CO2 does not require the use of harmful solvents. They are used to replace traditional chlorofluorocarbons or volatile organic solvents which are hazardous to health and environment.

Fig. 3.47 Schematic presentation of the basic principles of the foaming process

136

3 Industrial Scale Applications: Physical-Based Processes

This is a huge advantage for processing of polymers when it comes to biomedical applications. Biodegradable and biocompatible polymers can be produced, which are proper candidate as scaffolds for tissue engineering (Ivanovic et al. 2016; PintadoSierra et al. 2014; Salerno et al. 2014). Scaffolds, which provide a temporary artificial matrix for cell seeding, should meet certain fundamental characteristics such as high porosity, appropriate pore size, biocompatibility, biodegradability and proper degradation rate. Therefore, scaffold fabrication methods should allow control of pore size and enhance the maintenance of mechanical properties and materials’ biocompatibility (Ma 2004). However, there are also few challenges of CO2 as a foaming agent. They are associated with the higher pressure operation, dimensional instability during the foam-shaping process, and the high diffusivity of CO2 out of the foam resulting in a poor cell growth control, a lower nucleation density, and a low R-value (Tomasko et al. 2003).

3.2.2.2

Foaming Procedures

Foaming process is most commonly operated by batch system, continuous extrusion process and foam injection molding.

Batch Foaming Batch foaming or solid state foaming is the simplest foaming method where the preshaped polymer samples are placed in a high-pressure autoclave, which is saturated with CO2 for a predetermined period of time. Afterwards, decompression follows, upon which the gas bubbles are nucleated. Nucleation and cell growth are controlled by the pressure-release rate and foaming temperature. The other possibility is to remove the polymer from the vessel with trapped CO2 inside. The polymer is then heated. When the polymer is soften, the gas bubbles are nucleated. The batch foaming is mostly applied in research, due to its simplicity and ability to easily control processing parameters (time, temperature and pressure) (Di Maio and Kiran 2018).

Continuous Extrusion Process Since the batch process has limited productivity, it has low applicability in industry. Industry requires high production rates, which cannot be achieved with a simple batch process. On the other hand, extrusion foaming is preferred process for the industry, due to adaptable continuous production and relatively easy scale up process (Di Maio and Kiran 2018). Extrusion process begins with introducing CO2 into the extrusion barrel, where it is mixed with molten polymer at high pressure. CO2 and polymer are mixed together by single or double screw systems or systems involving two extruders that are coupled together, where a single phase is formed. Nucleation is initiated by

3.2 High Pressure Polymer Processing

137

a rapid pressure drop at nozzle exit, which reduces the solubility and creates a supersaturated solution. Nucleation can be also initiated via temperature increase near the nozzle exit (Tomasko et al. 2003). Foaming temperature, pressure drop or pressure drop rate, CO2 concentration determine the changes in the viscosity, solubility, surface tension, diffusivity and other physical properties, as well as play important roles in mechanisms of cell nucleation and cell growth. Challenges concerning extrusion process are limited pressure due to extruder barrel at the CO2 charge point, time for CO2 sorption and dissolution in the polymer and melt viscosity requirements to prevent CO2 escape and foam collapse (Di Maio and Kiran 2018).

Foam Injection Molding Foam injection molding is used for the production of structurally more complex (three-dimensional) foams. This method requires special equipment that consists of a gas dosage unit and controller, a gas injector, a specially designed screw, and a high-pressure gas source. In addition, the machine barrel needs to be modified to accommodate the gas injector (Lee et al. 2011). Contrary to extrusion process where the expansion starts immediately after die exit, in injection molding, expansion is induced with enlarging the mold cavity after it is firstly filled with polymer/CO2 solution. When the cavity is expanded, a gas that is dissolved in the polymer suffers pressures drop resulting in nucleation and foaming process (Di Maio and Kiran 2018). The batch foaming technique allows the processing of polymers at significantly lower temperatures than those needed in extrusion or injection molding. At lower temperatures, the solubility of CO2 in amorphous polymers increases, resulting in higher cell densities and smaller cell sizes. Foam characteristics, such as cell size and cell density can be improved by the addition of nanofillers inside the polymer matrix. Consequently, this results in porous structures with improved mechanical properties, increased cell density and reduced cell size (Tsivintzelis et al. 2016).

3.2.2.3

Polymer Types

Not all the polymers are appropriate, and they have to possess specific properties to be considered for the foaming process. To name just a few, solubility of CO2 (blowing agent), high melt strength and possibility of either crystallization or vitrification are important parameters for the process. One of the most significant effect of CO2 as blowing agent is to cause plasticization of the polymer (Li et al. 2016; Takahashi et al. 2012). Foaming of polymers with CO2 is usually carried out with glassy, amorphous polymers. Foaming of semi-crystalline polymers is, however, more challenging due to insufficient melt strength at high temperatures and crystallization at low temperatures (Di Maio and Kiran 2018).

138

3 Industrial Scale Applications: Physical-Based Processes

In the most cases, the production of foams comes from usual thermoplastics or synthetic biodegradable polymers. Polystyrene is the most frequently foamed polymer. Native biopolymers, such as cellulose, chitin, and starch are less investigated since these materials are difficult to process. They have a crystalline and intramolecular structure, which is difficult to break (Tsioptsias and Panayiotou 2008; Tsioptsias et al. 2011). The foaming of polymers with supercritical fluids has attracted particular interest mainly for producing microcellular materials. Microporous materials have average pore diameter smaller than 10 μm and pore population density larger than 109 pores per cm3 (Kumar and Suh 1990). The advantage of these foams compared to corresponding compact materials is that they have reduced bulk density. Reduced bulk density induces materials savings and costs. On the other hand, such materials often exhibit high toughness, high impact strength, high fatigue life, high stiffness to weight ratio, as well as low thermal conductivity (Kumar and Suh 1990). The most important properties of the polymer gas (CO2 ) system that define the final porous structure are: • the degree of crystallinity of the polymer matrix (in case of semicrystalline polymers and solid state foaming), • the initial (equilibrium) amount of the dissolved gas, • the degree of supersaturation that is induced from the depressurization or the heating of the system, • the surface energy between the gas nuclei and the surrounding polymer matrix, and • the plasticization profile of the polymer–gas system (i.e. the change in the glass transition temperature, which is induced by the sorption of the gas) (Curia et al. 2015; Goel and Beckman 1994; Kiran 2016; Tsivintzelis and Panayiotou 2013). 3.2.2.4

Example—Design of Process for Batch PLLA and PLGA Polymers Foaming

A magnetic suspension balance was used for measuring the solubility of CO2 in both PLLA (poly(L-lactide)) and PLGA (poly(D,L-lactide-co-glycolide)) polymers. The general scheme of the balance is presented in Fig. 3.48. The results represent an insight into the phenomena that take place during polymer foaming. These two polymers bahave differently under scCO2 . PLLA is a semicrystalline polymer with a high melting point, while PLGA is amorphous. The solubility of CO2 in PLLA and PLGA was measured for three different temperatures (308, 313 and 323 K) for a pressure range of 100–300 bar, and the results are presented in Figs. 3.49 and 3.50 (Aionicesei et al. 2008). From the data presented above it is clear that for all studied temperatures the solubility of CO2 increases with increasing pressure. By raising the pressure, the gas molecules are forced between polymer chains, expanding the space between molecules and thus increasing their mobility. Increased mobility of the chains allows more gas molecules to be adsorbed as the pressure increases. Higher pressures drive

3.2 High Pressure Polymer Processing

139

Fig. 3.48 A scheme of magnetic suspension balance Aionicesei (2009)

Fig. 3.49 Solubility of CO2 in PLLA at 308, 313 and 323 K. https://doi.org/10.1016/j.supflu.2008. 07.011

140

3 Industrial Scale Applications: Physical-Based Processes

Fig. 3.50 Solubility of CO2 in PLGA at 308, 313 and 323 K. https://doi.org/10.1016/j.supflu.2008. 07.011

more carbon dioxide into the polymer, while at the same time, they produce a hydrostatic effect that hinders the swelling and tends to squeeze out the free volume. Solubility of CO2 in PLGA is lower then in PLLA, despise the semicrystalline nature of PLLA. However, PLLA possess a larger free volume than PLGA due to its structure and extra methyl group. The solubility of gasses in glassy polymers is higher then in rubbers due to the existence of additional free volume into which sorption can occur (Aionicesei et al. 2008). Figure 3.51 presents some of the PLGA samples obtained at different process parameters studied by SEM. It is clear that the mean pore diameter decreases, while the cell density decreases with increase of operating pressure. Furthermore, by increasing the depressurization rate, the mean pore diameter decreases, while the cell population density increases. The effect of pressure and depressurization rate on the internal structure of the foams is the same for composite materials PLGA-Hydroxyapatite. Lastly, the effect of temperature showed that pore diameter increases, while the cell population density decreases with the increase of temperature. Similar results can be found in literature.

3.2.2.5

Future Trends

Most certainly, the main focus will be on the development of new methodologies that will allow industrial implementation. Furthermore, it is expected that the research area is even more spread to semi-crystalline, rubbery polymers which are currently neglected because of amorphous-glassy polymers, which are easier to handle. Lastly, it is expected that the research grows in the direction of generating nano-porous materials and systems that incorporate nanoparticles.

References

141

Fig. 3.51 The effect of pressure and depressurization rate on the structure of PLGA and PLGA-HA samples (Aionicesei 2009)

References Aionicesei E, Škerget M, Knez Ž (2008) Measurement of CO2 solubility and diffusivity in poly(llactide) and poly(d, l-lactide-co-glycolide) by magnetic suspension balance. J Supercrit Fluids 47(2):296–301 Aionicesei E (2009) Polymer processing using supercritical fluids. Ph.D., Univerza v Mariboru, Slovenia Anderko A (1990) Equation-of-state methods for the modelling of phase equilibria. Fluid Phase Equilib 61(1–2):145–225 Anderko A (2000) 4 Cubic and generalized van der waals equations. In: Sengers JV, Kayser RF, Peters CJ, White HJ (eds) Experimental thermodynamics. Elsevier, pp 75–126 Anisimov MA, Sengers JV (2000) Equations of state for fluids and fluid mixtures. IUPAC Elsevier, pp 381–434 Avelino HMNT, Fareleira JMNA, Gourgouillon D, Igreja JM, Nunes da Ponte M (2017) Viscosity of poly(ethyleneglycol) 200 [PEG 200] saturated with supercritical carbon dioxide. J Supercrit Fluids 128:300–307 Badens E, Masmoudi Y, Mouahid A, Crampon C (2018) Current situation and perspectives in drug formulation by using supercritical fluid technology. J Supercrit Fluids 134:274–283

142

3 Industrial Scale Applications: Physical-Based Processes

Bertucco A, Vetter G (2001) High pressure process technology: fundamentals and applications. Elsevier, Amsterdam Bleich J, Müller BW, Waßmus W (1993) Aerosol solvent extraction system—a new microparticle production technique. Int J Pharm 97(1):111–117 Bork M (2000) Production process, used in particular for obtaining lecithin from dehydrated egg, DE patent 198 54 807 A1 Bork M, Luetge C (2008) Method for the selective extraction and separation of organic substances by means of high pressure Brunner G (1994) Gas extraction: an introduction to fundamentals of supercritical fluids and the application to separation processes. In: Baumgärtel H, Franck EU, Grünbein W (eds) Topics in physical chemistry, vol 4. Steinkopff, Darmstadt/Springer, New York, 1994, 387 S., DM 64 Brunner G (2009) Counter-current separations. J Supercrit Fluids 47(3):574–582 Brunner G (2013) Gas extraction: an introduction to fundamentals of supercritical fluids and the application to separation processes. Springer Science & Business Media Calderone M, Rodier E, Letourneau J-J, Fages J (2007) Solidification of Precirol® by the expansion of a supercritical fluid saturated melt: from the thermodynamic balance towards the crystallization aspect. J Supercrit Fluids 42(2):189–199 Careno S, Boutin O, Badens E (2012) Drug recrystallization using supercritical anti-solvent (SAS) process with impinging jets: effect of process parameters. J Cryst Growth 342(1):34–41 Catchpole OJ, Tallon SJ, Eltringham WE, Grey JB, Fenton KA, Vagi EM, Vyssotski MV, MacKenzie AN, Ryan J, Zhu Y (2009) The extraction and fractionation of specialty lipids using near critical fluids. J Supercrit Fluids 47(3):591–597 Chen R, Meng F, Zhang S, Liu Z (2009) Effects of ultrahigh pressure extraction conditions on yields and antioxidant activity of ginsenoside from ginseng. Sep Purif Technol 66(2):340–346 Chen W, Hu X, Hong Y, Su Y, Wang H, Li J (2013) Ibuprofen nanoparticles prepared by a PGSSTMbased method. Powder Technol 245:241–250 Cotterman RL, Schwarz BJ, Prausnitz JM (1986) Molecular thermodynamics for fluids at low and high densities. Part I: pure fluids containing small or large molecules. AIChE J 32(11):1787–1798 Couto R, Alvarez V, Temelli F (2017) Encapsulation of vitamin B2 in solid lipid nanoparticles using supercritical CO2 . J Supercrit Fluids 120:432–442 Cravo C, Duarte ARC, Duarte CM (2007) Solubility of carbon dioxide in a natural biodegradable polymer: determination of diffusion coefficients. J Supercrit Fluids 40(2):194–199 Curia S, De Focatiis DSA, Howdle SM (2015) High-pressure rheological analysis of CO2 -induced melting point depression and viscosity reduction of poly(ε-caprolactone). Polymer 69:17–24 Dal Magro C, Aguiar GPS, Veneral JG, dos Santos AE, de Chaves LMPC, Oliveira JV, Lanza M (2017) Co-precipitation of trans-resveratrol in PHBV using solution enhanced dispersion by supercritical fluids technique. J Supercrit Fluids 127:182–190 Danner RP, Gupte PA (1986) Density dependent local composition models: an interpretive review. Fluid Phase Equilib 29:415–430 de Arons JS, Diepen GAM (1963) Thermodynamic study of melting equilibria under pressure of a supercritical gas. Recl Trav Chim Pays-Bas 82(3):249–256 De Loos ThW (2006) On the phase behaviour of asymmetric systems: the three-phase curve solid– liquid–gas. J Supercrit Fluids 39(2):154–159. https://doi.org/10.1016/j.supflu.2006.02.002 De Gioannis B, Gonzalez AV, Subra P (2004) Anti-solvent and co-solvent effect of CO2 on the solubility of griseofulvin in acetone and ethanol solutions. J Supercrit Fluids 29(1):49–57 De la Fuente Badilla JC, Peters CJ, de Swaan AJ (2000) Volume expansion in relation to the gas–antisolvent process. J Supercrit Fluids 17(1):13–23 De la Fuente JC, Shariati A, Peters CJ (2004) On the selection of optimum thermodynamic conditions for the GAS process. J Supercrit Fluids 32(1):55–61 De Paz E, Martín Á, Duarte CMM, Cocero MJ (2012) Formulation of β-carotene with poly-(εcaprolactones) by PGSS process. Powder Technol 217:77–83 Debenedetti PG (1990) Homogeneous nucleation in supercritical fluids. 9(36):1289–1298

References

143

Dehghani F, Foster NR (2003) Dense gas anti-solvent processes for pharmaceutical formulation. Curr Opin Solid State Mater Sci 7(4):363–369 Di Maio E, Kiran E (2018) Foaming of polymers with supercritical fluids and perspectives on the current knowledge gaps and challenges. J Supercrit Fluids 134:157–166 Diaz MS, Brignole EA (2009) Modeling and optimization of supercritical fluid processes. J Supercrit Fluids 47(3):611–618. https://doi.org/10.1016/j.supflu.2008.09.006 Dimitrov K, Boyadzhiev L, Tufeu R, Cansell F, Barth D (1998) Solubility of poly (ethylene glycol) nonylphenyl ether in supercritical carbon dioxide. J Supercrit Fluids 14(1):41–47 Dohrn R (1994) Berechnung von phasengleichgewichten, Friedr. Vieweg Sohn Verlagsgesellschaft MbH Braunschw Ger Eaves D (2004) Handbook of polymer foams. Rapra Technology Ltd., Shawbury Eggers R (1996) Supercritical fluid extraction (SFE) of oilseeds/lipids. In: Supercritical fluid technology in oil and lipid chemistry. AOCS Press, Illinois, p 34 Eltringham W, Catchpole OJ (2007) Processing of fish oils by supercritical fluids. In: Supercritical fluid extraction of nutraceuticals bioactive compounds CRC Press, Boca Raton, pp 141–188 Elvassore N, Flaibani M, Bertucco A, Caliceti P (2003) Thermodynamic analysis of micronization processes from gas-saturated solution. Ind Eng Chem Res 42(23):5924–5930 Enders S, Kahl H, Winkelmann J (2005) Interfacial properties of polystyrene in contact with carbon dioxide. Fluid Phase Equilib 228–229:511–522 Fages J, Lochard H, Letourneau J-J, Sauceau M, Rodier E (2004) Particle generation for pharmaceutical applications using supercritical fluid technology. Powder Technol 141(3):219–226 Fang T, Goto M, Sasaki M, Yang D (2007) Extraction and purification of natural tocopherols by supercritical CO2 . In: Supercritical fluid extraction nutraceuticals and bioactive compounds. CRC Press, Boca Raton, pp 103–140 Fanovich MA, Jaeger P (2012) Sorption and diffusion of compressed carbon dioxide in polycaprolactone for the development of porous scaffolds. Mater Sci Eng C 32(4):961–968 Fleck U, Brunner G, Karge R (2000) Purification of synthetic crude tocopherol acetate by means of supercritical fluid extraction. In: Proceedings of the fifth international symposium on supercritical fluids Forchhammer B, Gehrig M, Geyer S, Schulmeyr J, Simon K (2006) Method for decaffeinating tea Forest C, Chaumont P, Cassagnau P, Swoboda B, Sonntag P (2015) Polymer nano-foams for insulating applications prepared from CO2 foaming. Prog Polym Sci 41:122–145 Forster AD, Schulmeyr J, Schmidt R, Simon K, Ketterer MD, Forchhammer B, Geyer S, Gehrig MD (2004) Method for preparing a xanthohumol enriched hop extract and use thereof Foster N, Mammucari R, Dehghani F, Barrett A, Bezanehtak K, Coen E, Combes G, Meure L, Ng A, Regtop HL, Tandya A (2003a) Processing pharmaceutical compounds using dense gas technology. Ind Eng Chem Res 42(25):6476–6493 Foster NR, Dehghani F, Charoenchaitrakool KM, Warwick B (2003b) Application of dense gas techniques for the production of fine particles. AAPS PharmSci 5(2):32–38 Fukné-Kokot K, Škerget M, König A, Knez Ž (2003) Modified freezing method for measuring the gas solubility along the solid–liquid–gas equilibrium line. Fluid Phase Equilib 205(2):233–247 Gallagher PM, Coffey MP, Krukonis VJ, Klasutis N (1989) Gas antisolvent recrystallization: new process to recrystallize compounds insoluble in supercritical fluids. In: Supercritical fluid science and technology. American Chemical Society, pp 334–354 Gamse T, Rogler I, Marr R (1999) Supercritical CO2 extraction for utilisation of excess wine of poor quality. J Supercrit Fluids 14(2):123–128 García-González CA, Argemí A, de Sousa ARS, Duarte CMM, Saurina J, Domingo C (2010) Encapsulation efficiency of solid lipid hybrid particles prepared using the PGSS® technique and loaded with different polarity active agents. J Supercrit Fluids 54(3):342–347 Gardner DS (1993) Commercial scale extraction of alpha-acids and hop oils with compressed CO2 . In: King MB, Bott TR (eds) Extraction of natural products using near-critical solvents. Springer, Netherlands, Dordrecht, pp 84–100

144

3 Industrial Scale Applications: Physical-Based Processes

Gehrig MDCD, Schulmeyr JDI (1997) Decaffeination of tea leaves—uses pressure and temp to separate caffeine from carbon dioxide solvent, DE patent 4400 0 96 C2 Giddings JC, Myers MN, King JW (1969) Dense gas chromatography at pressures to 2000 atmospheres. J Chromatogr Sci 7(5):276–283 Giddings JC, Czubryt JJ, Myers MN (1970) Solubility phenomena in dense carbon dioxide gas in the range 270–1900 atmospheres. J Phys Chem 74(24):4260–4266 Global Information (2019) Global information, Inc. (GII)—The future of polymer foams. https:// www.giiresearch.com/report/rap304492-future-polymer-foams.html Goel SK, Beckman EJ (1994) Generation of microcellular polymeric foams using supercritical carbon dioxide. I: effect of pressure and temperature on nucleation. Polym Eng Sci 34(14):1137– 1147 Gourgouillon D, Avelino HMNT, Fareleira JMNA, Nunes da Ponte M (1998) Simultaneous viscosity and density measurement of supercritical CO2 -saturated PEG 400. J Supercrit Fluids 13(1):177– 185 Gourgouillon D, da Ponte MN (1999) High pressure phase equilibria for poly(ethylene glycol)s + CO2 : experimental results and modelling. Phys Chem Chem Phys 1(23):5369–5375. https://doi. org/10.1039/A906927E Guo X, Han D, Xi H, Rao L, Liao X, Hu X, Wu J (2012) Extraction of pectin from navel orange peel assisted by ultra-high pressure, microwave or traditional heating: a comparison. Carbohydr Polym 88(2):441–448 Gupta RB, Shim J-J, Shim J-J (2006) Solubility in supercritical carbon dioxide. CRC Press Gutiérrez C, Rodríguez JF, Gracia I, de Lucas A, García MT (2014) Preparation and characterization of polystyrene foams from limonene solutions. J Supercrit Fluids 88:92–104 Hannay JB, Hogarth J, Stokes GG (1880) On the solubility of solids in gases. Proceedings of royal society of London series A, London Harrison KL, Johnston KP, Sanchez IC (1996) Effect of surfactants on the interfacial tension between supercritical carbon dioxide and polyethylene glycol. Langmuir 12(11):2637–2644 Hickey AJ, Giovagnoli S (2018) Recent and evolving manufacturing processes. In: Recent and evolving manufacturing processes|SpringerLink. Springer, Cham Horvat et al. (2022) https://doi.org/10.1016/j.cej.2022.138251 Hsu J, Tan C (1994) Separation of ethanol/water solution with supercritical CO2 in the presence of a membrane. In: Rizvi S (ed) Supercritical fluid processing of food and biomaterials. Chapman & Hall, New York. Chapman & Hall, pp 114–122 Huang Z, Sun G-B, Chiew YC, Kawi S (2005) Formation of ultrafine aspirin particles through rapid expansion of supercritical solutions (RESS). Powder Technol 160(2):127–134 Huron M-J, Vidal J (1979) New mixing rules in simple equations of state for representing vapourliquid equilibria of strongly non-ideal mixtures. Fluid Phase Equilib 3(4):255–271 Ibeh CC, Bubacz M (2008) Current trends in nanocomposite foams. J Cell Plast 44(6):493–515 Ilieva P, Kilzer A, Weidner E (2016) Measurement of solubility, viscosity, density and interfacial tension of the systems tristearin and CO2 and rapeseed oil and CO2 . J Supercrit Fluids 117:40–49 Ivanovic J, Knauer S, Fanovich A, Milovanovic S, Stamenic M, Jaeger P, Zizovic I, Eggers R (2016) Supercritical CO2 sorption kinetics and thymol impregnation of PCL and PCL-HA. J Supercrit Fluids 107:486–498 Johnston KP, Penninger JML (1989) American chemical society symposium series 406. American Chemical Society, Washington DC ˇ Jug M, Hafner A, Lovri´c J, Kregar ML, Pepi´c I, Vani´c Ž, Cetina-Cižmek B, Filipovi´c-Grˇci´c J (2018) An overview of in vitro dissolution/release methods for novel mucosal drug delivery systems. J Pharm Biomed Anal 147:350–366 Jun X, Shuo Z, Bingbing L, Rui Z, Ye L, Deji S, Guofeng Z (2010) Separation of major catechins from green tea by ultrahigh pressure extraction. Int J Pharm 386(1–2):229–231 Jung J, Perrut M (2001) Particle design using supercritical fluids: literature and patent survey. J Supercrit Fluids 20(3):179–219

References

145

Kappler P, Petermann M, Weidner E (2003) Size and morphology of particles generated by spraying polymer melts with carbon dioxide. Tome 3. Vandoeuvre Cedex: Institut National Polytechnique de Lorraine, Versailles, France, pp 1891–1896 Kawabata Y, Wada K, Nakatani M, Yamada S, Onoue S (2011) Formulation design for poorly watersoluble drugs based on biopharmaceutics classification system: basic approaches and practical applications. Int J Pharm 420(1):1–10 Kegl T, Kravanja G, Knez Ž, Knez Hrnˇciˇc M (2017) Effect of addition of supercritical CO2 on transfer and thermodynamic properties of biodegradable polymers PEG 600 and Brij52. J Supercrit Fluids 122:10–17 Kerˇc J, Srˇciˇc S, Knez Ž, Senˇcar-Božiˇc P (1999) Micronization of drugs using supercritical carbon dioxide. Int J Pharm 182(1):33–39 Khosravi-Darani K, Vasheghani-Farahani E, Yamini Y, Bahramifar N (2003) Solubility of poly (β-hydroxybutyrate) in supercritical carbon dioxide. J Chem Eng Data 48(4):860–863 Kiran E (2009) Polymer miscibility, phase separation, morphological modifications and polymorphic transformations in dense fluids. J Supercrit Fluids 47(3):466–483 Knez Z (1989) Separation of components of citrus oils—industrial project Kiran E (2016) Supercritical fluids and polymers—the year in review—2014. J Supercrit Fluids 110:126–153 Knez Ž (2016) Food processing using supercritical fluids. In: Nedovi´c V, Raspor P, Levi´c J, Tumbas Šaponjac V, Barbosa-Cánovas GV (eds) Emerging and traditional technologies for safe, healthy and quality food. Springer International Publishing, Cham, pp 413–442 Knez Ž, Škerget M (2001) Phase equilibria of the vitamins D2, D3 and K3 in binary systems with CO2 and propane. J Supercrit Fluids 20(2):131–144 Knez Ž, Weidner E (2001) Precipitation of solids with dense gasses. In: Bertucco A, Vetter G (eds) High pressure process technology: fundamentals and applications. Elsevier, London, p 571 Knez Z, Weidner E (2003) Particles formation and particle design using supercritical fluids. Curr Opin Solid State Mater Sci 7(4):353–361 Knez Ž, Posel F, Krmelj I (1994) Supercritical fluid processing of food and biomaterials. In: Rizvi S (ed) High pressure extraction of organics from water. Chapman & Hall, New York, pp 181–186 Knez Ž, Škerget M, Mandžuka Z (2010) Determination of S–L phase transitions under gas pressure. J Supercrit Fluids 55(2):648–652 Knez Ž, Markoˇciˇc E, Novak Z, Hrnˇciˇc MK (2011) Processing polymeric biomaterials using supercritical CO2 . Chem Ing Tech 83(9):1371–1380 Knez Ž, Markoˇciˇc E, Leitgeb M, Primožiˇc M, Knez Hrnˇciˇc M, Škerget M (2014) Industrial applications of supercritical fluids: a review. Energy 77:235–243 Knez Ž, Panti´c M, Cör D, Novak Z, Knez Hrnˇciˇc M (2019) Are supercritical fluids solvents for the future? Chem Eng Process—Process Intensif 141:107532 Knez Ž, Novak Z, Weidner E (1995) Process for preparing particles or powders, WO patent WO1995021688A1 ˇ Knez Ž, Cretnik L, Škerget M (2004) Micronisation of vanillins by PGSSTM using various compressed gases. In: Kikic I, Perrut M (eds) Proceedings of the 9th meeting on supercritical fluids. Trieste, Italy Knez Ž, Škerget M, KnezHrnˇciˇc M (2013) Principles of supercritical fluid extraction and applications in the food, beverage and nutraceutical industries. In: Rizvi SSH (ed) Separation, extraction and concentration processes in the food, beverage and nutraceutical industries. Woodhead Publishing, pp 3–38 Knez Hrnˇciˇc M, Markoˇciˇc E, Trupej N, Škerget M, Knez Ž (2014a) Investigation of thermodynamic properties of the binary system polyethylene glycol/CO2 using new methods. J Supercrit Fluids 87:50–58 Knez Hrnˇciˇc M, Škerget M, Knez Ž (2014b) Density and viscosity of the binary polyethylene glycol/ CO2 systems. J Supercrit Fluids 95:641–668 Kravanja G (2018) High-perssure process design for polymer treatment and heat transfer enhancement. Ph.D., Univerza v Mariboru (Slovenia)

146

3 Industrial Scale Applications: Physical-Based Processes

Kravanja et al. (2022) https://doi.org/10.1016/j.matdes.2022.110653 Krukonis V, McHugh M (1994) Supercritical fluid extraction, 2nd edn. Butterworth-Heinemann, Boston, MA Krykin MA (1990) Sorption and diffusion of gas mixtures in polymers at high pressures. Mutual effects of components. Polym Sci USSR 32(2):301–307 Kumar V, Suh NP (1990) A process for making microcellular thermoplastic parts. Polym Eng Sci 30(20):1323–1329 Lack E, Seidlitz H (1993) Commercial scale decaffeination of coffee and tea using supercritical CO2 . In: King MB, Bott TR (eds) Extraction of natural products using near-critical solvents. Springer Netherlands, Dordrecht, pp 101–139 Lack E, Simandi B (2001) Supercritical fluid extraction and fractionation from solid materials. Ind Chem Libr 9:537–575 Lee J, Turng L-S, Dougherty E, Gorton P (2011) Novel foam injection molding technology using carbon dioxide-laden pellets. Polym Eng Sci 51(11):2295–2303 Lévai G, Albarelli JQ, Santos DT, Meireles MAA, Martín Á, Rodríguez-Rojo S, Cocero MJ (2017) Quercetin loaded particles production by means of supercritical fluid extraction of emulsions: process scale-upstudy and thermo-economic evaluation. Food Bioprod Process 103:27–38 Li J, Matos HA, de Azevedo EG (2004) Two-phase homogeneous model for particle formation from gas-saturated solution processes. J Supercrit Fluids 32(1):275–286 Li J, Rodrigues M, Paiva A, Matos HA, de Azevedo EG (2005) Modeling of the PGSS process by crystallization and atomization. AIChE J 51(8):2343–2357 Li Y, Yang DJ, Zhou W, Chen SB, Chen SL (2012) Recrystallization of puerarin using the supercritical fluid antisolvent process. J Cryst Growth 340(1):142–148 Li R, Zhang Z, Fang T (2016) Experimental research on swelling and glass transition behavior of poly(methyl methacrylate) in supercritical carbon dioxide. J Supercrit Fluids 110:110–116 Li S (2007) Application of supercritical fluids in traditional Chinese medicines and natural products. In: Supercritical fluid extraction of nutraceuticals and bioactive compounds. CRC Press, Boca Raton, pp 215–242 Ljubec B, Knez Hrnˇciˇc M, Cör D, Kravanja G, Knez Ž (2018) Phase equilibria of the binary systems of fenofibrate and dense gases (carbon dioxide, propane, trifluoromethane). Fluid Phase Equilib 474:110–115 L˝orincz L, Bánsághi G, Zsemberi M, de Simón BS, Szilágyi IM, Madarász J, Sohajda T, Székely E (2016) Diastereomeric salt precipitation based resolution of ibuprofen by gas antisolvent method. J Supercrit Fluids 118:48–53 Luetge C, Bork M, Knez Z, Kreiner M (2007) Ultra high pressure dense gas extraction and fractionation. In: Proceedings international symposia high press. Process technology chemistry engineering, 5th, Segovia, Spain. European federal chemistry engineering. Segovia, Spain, pp 1–4 Luetge C, Steinhagen V, Bork M, Knez Z (2009) Supercritical carbon dioxide extraction of plant materials at ultrahigh pressure. In: Proceedings of the IX international symposium on supercritical fluids Ma PX (2004) Scaffolds for tissue fabrication. Mater Today 7(5):30–40 Mahmood SH, Ameli A, Hossieny N, Park CB (2014) The interfacial tension of molten polylactide in supercritical carbon dioxide. J Chem Thermodyn 75:69–76 Marizza P, Keller SS, Müllertz A, Boisen A (2014) Polymer-filled microcontainers for oral delivery loaded using supercritical impregnation. J Controlled Release 173:1–9 Markoˇciˇc E, Knez Z, Škerget M (2013) Effect of temperature and pressure on the behavior of poly(ε-caprolactone) in the presence of supercritical carbon dioxide. 44(52):15594–15601 Marr R, Gamse T (2000) Use of supercritical fluids for different processes including new developments—a review. Chem Eng Process Process Intensif 39(1):19–28 Marra F, De Marco I, Reverchon E (2012) Numerical analysis of the characteristic times controlling supercritical antisolvent micronization. Chem Eng Sci 71:39–45

References

147

Martìn A, Cocero MJ (2004) Numerical modeling of jet hydrodynamics, mass transfer, and crystallization kinetics in the supercritical antisolvent (SAS) process. J Supercrit Fluids 32(1):203–219 Martín A, Cocero MJ (2008) Micronization processes with supercritical fluids: fundamentals and mechanisms. Adv Drug Deliv Rev 60(3):339–350 Martín Á, Weidner E (2010) PGSS-drying: mechanisms and modeling. J Supercrit Fluids 55(1):271– 281 Matson DW, Petersen RC, Smith RD (1987) Production of powders and films by the rapid expansion of supercritical solutions. J Mater Sci 22(6):1919–1928 McHugh M, Krukonis V (2013) Supercritical fluid extraction: principles and practice. Elsevier Mendes RL (2007) Supercritical fluid extraction of active compounds from algae. Supercritical fluid extraction nutraceuticals and bioactive compounds. CRC Press, Boca Raton, pp 189–213 Mi H-Y, Salick MR, Jing X, Jacques BR, Crone WC, Peng X-F, Turng L-S (2013) Characterization of thermoplastic polyurethane/polylactic acid (TPU/PLA) tissue engineering scaffolds fabricated by microcellular injection molding. Mater Sci Eng C 33(8):4767–4776 Moyler DA (1993) Extraction of flavours and fragrances with compressed CO2 . In: Extraction of natural products using near-critical Solvents. Springer, pp 140–183 Mukhopadhyay M (2007a) 11 Processing of spices using supercritical fluids. Supercritical Fluid Extr Nutraceuticals Bioact Compd :337 Mukhopadhyay M (2007b) Processing of spices using supercritical fluids. In: Martinez JL (ed) Supercritical fluid extraction of nutraceuticals and bioactive compounds. CRC Press, pp 337–366 Nalawade SP, Picchioni F, Janssen LPBM (2006) Supercritical carbon dioxide as a green solvent for processing polymer melts: processing aspects and applications. Prog Polym Sci 31(1):19–43 Nalawade SP, Picchioni F, Janssen LPBM (2007) Batch production of micron size particles from poly(ethylene glycol) using supercritical CO2 as a processing solvent. Chem Eng Sci 62(6):1712–1720 NIST Office of Data (2019) NIST WebBook. https://webbook.nist.gov/ O’Connell JP, Weber W, Brunner G (2003) Measurement and thermodynamics of triglyceride melting in near-critical fluids. In: AIChE annual meeting, Indianapolis, USA Paisana MC, Müllers KC, Wahl MA, Pinto JF (2016) Production and stabilization of olanzapine nanoparticles by rapid expansion of supercritical solutions (RESS). J Supercrit Fluids 109:124– 133 Pastore Carbone MG, Di Maio E, Scherillo G, Mensitieri G, Iannace S (2012) Solubility, mutual diffusivity, specific volume and interfacial tension of molten PCL/CO2 solutions by a fully experimental procedure: effect of pressure and temperature. J Supercrit Fluids 67:131–138 Peng D-Y, Robinson DB (1976) A new two-constant equation of state. Ind Eng Chem Fundam 15(1):59–64 Pestieau A, Krier F, Lebrun P, Brouwers A, Streel B, Evrard B (2015) Optimization of a PGSS (particles from gas saturated solutions) process for a fenofibrate lipid-based solid dispersion formulation. Int J Pharm 485(1):295–305 Pintado-Sierra M, Delgado L, Aranaz I, Marcos-Fernández Á, Reinecke H, Gallardo A, Zeugolis D, Elvira C (2014) Surface hierarchical porosity in poly (E-caprolactone) membranes with potential applications in tissue engineering prepared by foaming in supercritical carbon dioxide. J Supercrit Fluids 95:273–284 Pollak S, Kareth S, Kilzer A, Petermann M (2011) Thermal analysis of the droplet solidification in the PGSS-process. J Supercrit Fluids 56(3):299–303 Prausnitz JM, Lichtenthaler RN, de Azevedo EG (1986) Molecular thermodynamics of fluid-phase equilibria, 2nd edn. Prentice-Hall Inc, Englewood Cliffs, New Jersey Prausnitz JM, Lichtenthaler RN, Azevedo EG de (1998) Molecular thermodynamics of fluid-phase equilibria. Pearson education Prosapio V, Reverchon E, De Marco I (2016) Formation of PVP/nimesulide microspheres by supercritical antisolvent coprecipitation. J Supercrit Fluids 118:19–26

148

3 Industrial Scale Applications: Physical-Based Processes

Redlich O, Kwong JN (1949) On the thermodynamics of solutions. V. An equation of state. Fugacities of gaseous solutions. Chem Rev 44(1):233–244 Reverchon E, De Marco I (2004) Supercritical antisolvent micronization of Cefonicid: thermodynamic interpretation of results. J Supercrit Fluids 31(2):207–215 Reverchon E, De Marco I (2011) Mechanisms controlling supercritical antisolvent precipitate morphology. Chem Eng J 169(1):358–370 Reverchon E, Caputo G, De Marco I (2003a) Role of phase behavior and atomization in the supercritical antisolvent precipitation. Ind Eng Chem Res 42(25):6406–6414 Reverchon E, De Marco I, Caputo G, Della Porta G (2003b) Pilot scale micronization of amoxicillin by supercritical antisolvent precipitation. J Supercrit Fluids 26(1):1–7 Reverchon E, De Marco I, Torino E (2007) Nanoparticles production by supercritical antisolvent precipitation: a general interpretation. J Supercrit Fluids 43(1):126–138 Reverchon E, Adami R, Cardea S, Della Porta G (2009) Supercritical fluids processing of polymers for pharmaceutical and medical applications. J Supercrit Fluids 47(3):484–492 Reverchon E, Torino E, Dowy S, Braeuer A, Leipertz A (2010) Interactions of phase equilibria, jet fluid dynamics and mass transfer during supercritical antisolvent micronization. Chem Eng J 156(2):446–458 Sadus RJ (2012) High pressure phase behaviour of multicomponent fluid mixtures. Elsevier Salerno A, Fanovich MA, Pascual CD (2014) The effect of ethyl-lactate and ethyl-acetate plasticizers on PCL and PCL–HA composites foamed with supercritical CO2 . J Supercrit Fluids 95:394–406 Sandler SI, Orbey H (2000) Mixing and combining rules in equations of state for fluids and fluid mixtures. Elsevier, New York Sato Y, Fujiwara K, Takikawa T, Sumarno TS, Masuoka H (1999) Solubilities and diffusion coefficients of carbon dioxide and nitrogen in polypropylene, high-density polyethylene, and polystyrene under high pressures and temperatures. Fluid Phase Equilib 162(1):261–276 Sebald JDI, Schulmeyr JDI, Forster ADID (1995) Decaffeination of black tea, DE patent 195 24 481 C2 Senˇcar-Božiˇc P, Srˇciˇc S, Knez Z, Kerˇc J (1997) Improvement of nifedipine dissolution characteristics using supercritical CO2 . Int J Pharm 148(2):123–130 Shinde NG, Aloorkar NH, Bangar BN, Deshmukh SM, Shirke MV, Kale BB (2013) Pharmaceutical foam drug delivery system: general considerations. Indo Am J Pharm Res 3:1322–1327 Silva SS, Duarte ARC, Carvalho AP, Mano JF, Reis RL (2011) Green processing of porous chitin structures for biomedical applications combining ionic liquids and supercritical fluid technology. Acta Biomater 7(3):1166–1172 Škerget M, Novak-Pintariˇc Z, Knez Ž, Kravanja Z (2002) Estimation of solid solubilities in supercritical carbon dioxide: Peng-Robinson adjustable binary parameters in the near critical region. Fluid Phase Equilib 203(1–2):111–132 Soave G (1972) Equilibrium constants from a modified Redlich-Kwong equation of state. Chem Eng Sci 27(6):1197–1203 Sovová H (2005) Mathematical model for supercritical fluid extraction of natural products and extraction curve evaluation. J Supercrit Fluids 33(1):35–52 Stahl E, Quirin K-W, Gerard D (1987) Verdichtete gase zur extraktion und raffination. Springer Stahl E, Quirin K-W, Gerard D (1988) Applications of dense gases to extraction and refining. In: Dense gases for extraction and refining. Springer, pp 72–217 Stahl E, Quirin K-W, Gerard D (2013) Verdichtete Gase zur Extraktion und Raffination. Springer Strumendo M, Bertucco A, Elvassore N (2007) Modeling of particle formation processes using gas saturated solution atomization. J Supercrit Fluids 41(1):115–125 Tabernero A, Martín del Valle EM, Galán MA (2012) Supercritical fluids for pharmaceutical particle engineering: methods, basic fundamentals and modelling. Chem Eng Process Process Intensif 60:9–25 Takahashi S, Hassler JC, Kiran E (2012) Melting behavior of biodegradable polyesters in carbon dioxide at high pressures. J Supercrit Fluids 72:278–287

References

149

Tavana A, Randolph AD (1989) Manipulating solids CSD in a supercritical fluid crystallizer: CO2 benzoic acid. 10(35):1625–1630 Tavares Cardoso MA, Antunes S, van Keulen F, Ferreira BS, Geraldes A, Cabral JM, Palavra AM (2008) Supercritical antisolvent micronisation of synthetic all-trans-β-carotene with tetrahydrofuran as solvent and carbon dioxide as antisolvent—Tavares Cardoso—2009. J Chem Technol Biot Wiley Online Library. J Chem Technol Biot (84):115–122 Taylor LT (2009) Supercritical fluid chromatography for the 21st century. J Supercrit Fluids 47(3):566–573 Temelli F (2009) Perspectives on supercritical fluid processing of fats and oils. J Supercrit Fluids 47(3):583–590 Temelli F (2018) Perspectives on the use of supercritical particle formation technologies for food ingredients. J Supercrit Fluids 134:244–251 Temelli F, Saldaña MD, Moquin PH, Sun M (2007) Supercritical fluid extraction of specialty oils. In: Supercritical fluid extraction of nutraceuticals and bioactive compounds. CRC Press, pp 60–110 Tenorio A, Gordillo MD, Pereyra C, de la Ossa EJM (2007) Controlled submicro particle formation of ampicillin by supercritical antisolvent precipitation. J Supercrit Fluids 40(2):308–316 Tomasko DL, Li H, Liu D, Han X, Wingert MJ, Lee LJ, Koelling KW (2003) A review of CO2 applications in the processing of polymers. Ind Eng Chem Res 42(25):6431–6456 Tomasko DL, Burley A, Feng L, Yeh S-K, Miyazono K, Nirmal-Kumar S, Kusaka I, Koelling K (2009) Development of CO2 for polymer foam applications. J Supercrit Fluids 47(3):493–499 Trupej N, Hrnˇciˇc MK, Škerget M, Knez Ž (2015) Solubility and binary diffusion coefficient of argon in polyethylene glycols of different molecular weights. J Supercrit Fluids 103:10–17 Trupej N, Škerget M, Knez Ž (2016) Thermodynamic data for processing polyethylene glycol with non-conventional fluids. J Supercrit Fluids 118:39–47 Tsioptsias C, Panayiotou C (2008) Foaming of chitin hydrogels processed by supercritical carbon dioxide. J Supercrit Fluids 47(2):302–308 Tsioptsias C, Paraskevopoulos MK, Christofilos D, Andrieux P, Panayiotou C (2011) Polymeric hydrogels and supercritical fluids: the mechanism of hydrogel foaming. Polymer 52(13):2819– 2826 Tsivintzelis I, Panayiotou C (2013) Designing issues in polymer foaming with supercritical fluids. Macromol Symp 331–332(1):109–114 Tsivintzelis I, Sanxaridou G, Pavlidou E, Panayiotou C (2016) Foaming of polymers with supercritical fluids: a thermodynamic investigation. J Supercrit Fluids 110:240–250 Tuminello WH, Dee GT, McHugh MA (1995) Dissolving perfluoropolymers in supercritical carbon dioxide. Macromolecules 28(5):1506–1510 van Konynenburg PH, Scott RL, Shipley RJ (1980) Critical lines and phase equilibria in binary van der Waals mixtures. Philos Trans R Soc Lond Ser Math Phys Sci 298(1442):495–540 Weber A, Yelash LV, Kraska T (2005) Effect of the phase behaviour of the solvent–antisolvent systems on the gas–antisolvent-crystallisation of paracetamol. J Supercrit Fluids 33(2):107–113 Weidner E (2009) High pressure micronization for food applications. J Supercrit Fluids 47(3):556– 565 Weidner E, Wiesmet V, Knez Ž, Škerget M (1997) Phase equilibrium (solid-liquid-gas) in polyethyleneglycol-carbon dioxide systems. J Supercrit Fluids 10(3):139–147 Weidner E, Petermann M, Knez Z (2003) Multifunctional composites by high-pressure spray processes. Curr Opin Solid State Mater Sci 7(4):385–390 Weidner E, Steiner R, Knez Z (1996) Powder generation from polyethylene glycols with compressible fluids. In: von Rohr PR, Trepp C (eds) Process technology proceedings. Elsevier, pp 223–228 Wong DSH, Sandler SI (1992) A theoretically correct mixing rule for cubic equations of state. AIChE J 38(5):671–680 Wong A, Mark LH, Hasan MM, Park CB (2014) The synergy of supercritical CO2 and supercritical N2 in foaming of polystyrene for cell nucleation. J Supercrit Fluids 90:35–43

150

3 Industrial Scale Applications: Physical-Based Processes

Yeo S-D, Kiran E (2005) Formation of polymer particles with supercritical fluids: a review. J Supercrit Fluids 34(3):287–308 Zosel K (1981) Process for the decaffeination of coffee, US patent 4260639A Zsigmondy R, Bachmann W (1918) Über neue filter. Z Für Anorg Allg Chem 103(1):119–128

Chapter 4

Industrial Scale Applications: Reaction-Based Processes

Abstract In this chapter applications of sub- and supercritical fluids as reaction media for chemical and biochemical reactions as well as the reactant for different products are presented. The solvent properties of supercritical fluids enable them to apply them as solvent and as reactant in polymerization reactions and in hydrothermal synthesis. As solvent subcritical and supercritical fluids are applied for carbonylation, oxidation, hydrogenation, hydroformylation and as reaction media for biochemical reactions. Chemical reactions in supercritical media are already realized on industrial scale. The highest volume of use of supercritical fluids as reactant and as solvent media are processes on production of various polymer grades of polyethylene. Biochemical reactions in supercritical media were not yet applied in industrial scale. But most probably—due to excellent solvent properties of dense gases, low costs of some subcritical or supercritical fluids, possibilities of products fractionation and product formulation—these processes will sooner or later be applied in industrial scale. Keywords Supercritical fluid · Subcritical fluid · Reaction media · Polymerization · Hydrothermal synthesis · Carbonylation · Oxidation · Hydrogenation · Hydroformylation · Biochemical reaction

4.1 Chemical and Biochemical Reactions in SCFs Sub (Sub SCFs) and supercritical fluids (SCFs) are increasingly attractive as a medium for chemical and biochemical reactions due to their properties such as a low viscosity like the viscosity of a gas and the high density like the density of a liquid. They may be alternative solvents to conventional ones due to their tunable properties, and enable to perform reactions with environmental, health and safety process advantages as well as chemical benefits. The main reasons for reactions performed in SCFs (Clifford 1994) are presented in Fig. 4.1.

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Ž. Knez and C. Lütge, Product, Process and Plant Design Using Subcritical and Supercritical Fluids for Industrial Application, https://doi.org/10.1007/978-3-031-34636-1_4

151

152

4 Industrial Scale Applications: Reaction-Based Processes

Fig. 4.1 The main reasons for performance of chemical and biochemical reaction in SCFs

Running the reaction under supercritical conditions gives the ability to manipulate solvent properties by manipulating pressure and/or temperature. Therefore, the solubility of the reactants and products is increased and the interfacial transport with respect to the increase in the reaction rate is constrainted. Consequently, the selectivity can be improved, and reaction and separation can be integrated in one step. The application of sub and SCFs as solvents for chemical and biochemical reactions are advantaged due to simple downstream processing of reaction systems or even the separation of interfering components, which inhibit the reaction, can be achieved. Solubility of gases in SCFs is in general very high in comparison to the conventional solvents where their solubility is limited. SCFs may be advantageous as a solvent for reactions involved in fuels processing, biocatalysis, biomass conversion, homogeneous and heterogeneous catalysis, environmental control, polymerization, materials synthesis, chemical synthesis etc. (Brunner 2005, 2010; Knez et al. 2014). Another very useful application of SCFs is their possible usage as reactants in chemical and biochemical reactions. The advantage of using SCFs as a reactant in a reaction is either to avoid using an additional solvent or to maximize the concentration of reactant. Sub and SCFs could be also used as catalysts for different reactions. Subcritical water (SubH2 O) or supercritical water (SCW) have relatively high dissociation constants, and at the same time with increase in pressure and temperature of SCW, the dielectric constant is significantly decreased leading to an increased solubility of non-polar substances. During the production of fine chemicals considerable volumes of waste are generated, since the syntheses typically consist of several steps. Each of these steps usually achieves a yield between 60 and 90%. Thus it can be concluded, that approximately 1 kg of end-product leads to the generation of 15 kg of wastes or more. Most of wastes generated are solvents and by-products from solvents and intermediates. A

4.2 Chemical Reactions in SCFs

153

potential solution for reducing those wastes during synthesis of fine chemicals is using reactions performed in water or in SCFs (Knez et al. 2015a). SCFs are used in a wide range of chemical reactions and processes on large industrial scale for production of several bulk chemicals. Nowadays, studies on the use of SCFs as a solvent for chemical and biochemical processes are focused on the synthesis of complex organic molecules.

4.2 Chemical Reactions in SCFs SCFs in chemical reactions could be also used as reactants. One of the first chemical process, where SCFs were used as reactants, was the polymerisation of ethylene to form low-density-polyethylene (LDPE). Additional, due to lower dielectric constant and polarity of SCW at high pressure and temperature, the SCW could be efficiently used for acid catalysed reactions like Friedl-Crafts alkylation and hydrolytic reactions of esters. SCFs, especially SC CO2 , have been used in different important chemical reactions, such as hydroformylation, hydrogenation, alkylation etc. as an alternative reaction medium.

4.2.1 Polymerization and Depolymerization The use of CO2 as an inert solvent has emerged as an alternative to classical synthesis in polymer chemistry. The ability to tailor the properties with SC CO2 has enabled the synthesis of polymers with controlled molecular weight, morphology, polydispersity and with minimal contamination. Polymer production and processing using SC CO2 as reaction medium include polymerization, production of polymer composite, polymer blending, particle production, and microcellular foaming. Several applications, particularly those involving low pressures, have been successfully transferred to an industrial scale. Most suitable polymerization processes for the SCF technology application are solution, suspension and emulsion type of reactions, since the reduction of wastewater and/or solvent during polymerisation and improvement of properties occurred (Goodship and Ogur 2004). An overview of different types of polymerization (free radical, cationic, ring-opening, melt-phase condensation, catalytic chain transfer, oxidative coupling, sol–gel polymerization as well as polymer blend synthesis and simultaneous one-pot combination of enzymatic and chemical polymerization) in SC CO2 was done by Lutz et al. (2016). An overview of fluoropolymer syntheses (acrylate- and styrene-based systems, poly(vinyl-ether)s, tetrafluoroethylene- and vinylidenefluoride-based, as well as novel fluorinated elastomers and thermoplastics) in SC CO2 with discussions on the synthesis of mainchain and side-chain fluoropolymers conducted via a chain-growth or continuous process is given by Du et al. (2009). Fluoropolymers (e.g. fluorinated ethylenepropylene) are already produced commercially using emulsion process in SC CO2

154

4 Industrial Scale Applications: Reaction-Based Processes

by the DuPont company, Wilmington, Delaware, USA with the production capacity of 1100 metric tons per year (m.t./yr) (Goodship and Ogur 2004). In recent years, there has been a significant increase in the use of SC CO2 as a reaction medium for free-radical dispersion polymerizations (Wang et al. 2005; McAllister et al. 2016; Oliveira et al. 2016; Alaimo et al. 2017; Parilti et al. 2017, 2018; Costa and Storti 2018). Sulindac loaded poly(HEMA) cross-linked microparticles with well-defined spherical morphology and sizes between 250 and 350 nm were synthesized via one-pot free-radical dispersion polymerisation in SC CO2 in presence of photocleavable diblock stabilisers (Parilti et al. 2018). Controlled/ living radical polymerization has attracted much attention and was frequently studied (Alzahrani et al. 2019; Kazaryan et al. 2019), since it allows the synthesis of complex macromolecular structures and is generally more tolerant of polar functionalities and impurities than ionic and coordination polymerizations (Wang et al. 2005). An overview of the synthesis of polymer-inorganic filler nanocomposites in SC CO2 was given by Haldorai et al. (2012) with conclusion that for all three methods for the preparation of nanocomposites (blending, sol–gel and in situ polymerizations) SC CO2 was demonstrated to be a viable alternative to the conventional solvents. A novel and green method using SC CO2 for the preparation of pH sensitive biopolymer-based membranes with grafting copolymerization of acrylic acid (AA) on a cross-linked porous chitosan membrane was developed where benzyl-peroxide (BPO) as the reduction–oxidation free radical initiator was used (Cao et al. 2015). The water flux of the grafted chitosan membranes decreases with pH from 2 to 7, even at considerably low grafting yield. The ring-opening copolymerization (ROCOP) of CO2 and epoxides was discovered more than 40 years ago by Inoue and co-workers (Inoue et al. 1969). Due to environmental oriented method and their simplicity, ROCOP predominated over conventional polymerization processes using high-cost and/or hazardous reagents to synthesize a range of aliphatic polycarbonates (Dai et al. 2009). It is an interesting method to add value to CO2 including from waste sources, and to reduce pollution associated with commodity polymer manufacture, since it is also 100% atom economical. With the selection of the proper catalyst, the properties of the synthesized polymers can be controlled. Polymerisation in heterogeneous (Bhanage et al. 2001; Sun et al. 2009; Lang and He 2015) and homogeneous (Darensbourg and Yarbrough 2002; He et al. 2003; Song et al. 2012) catalyst systems have been often studied, but the activity and selectivity of those catalysts still remain low (Klaus et al. 2011). Most homogeneous catalysts operate under high pressures of CO2 (higher than 10 bar), but they do yield highly in comparison to heterogeneous catalysts. However, the interest in catalysts, which show high activities under low pressures (under 1 bar) is enormous. A promising path to reduce atmospheric CO2 levels is the efficient production of value-added chemicals from CO2 as a raw material. Among all possible CO2 capture utilization (CCU) transformations of CO2 into C1 or higher species, the most promising approach is currently the formation of cyclic carbonates via epoxide coupling (Zhang et al. 2014; Monfared et al. 2019). Usage of organocatalysts for

4.2 Chemical Reactions in SCFs

155

cycloaddition reaction can improved the sustainability and carbon those type of reactions (Cokoja et al. 2015). Lignin provides a valuable source of value-added chemicals, particularly phenolics. Under the condition of an adapted choice of process parameters, high-pressure biorefining allows the connection and combination of different highly efficient and selective unit operations into a single, one-step process of producing chemicals from lignocellulosic biomass that can be realized on different scales (Zetzl et al. 2011). Worldwide Kraft lignin (KL) production by pulp and paper manufacturers can exceed 50 million metric tons annually (Mahmood et al. 2013). Because of the diversity of its monomers and its markedly lower oxygen content, lignin is a promising feedstock for the production of renewable replacements for petroleum-derived products. The advantage of using SCF technology processes of lignin opens the possibility to produce smaller fragments through breakage of the ether linkages and produce larger fragments through cross linking between the reactive fragments (Wang et al. 2013). In Table 4.1, a literature review of novel studies of lignin depolymerisation is presented. For a reaction behaviour study of the lignin model compound in subH2 O and SCW, a batch type reactor was used (Wahyudiono et al. 2007, 2009). Goto et al. (2004) performed hydrothermal conversion of municipal organic waste into resources at sub-critical conditions in a batch reactor within a temperature range from 473 to 623 K and in a semi-continuous reactor with a temperature profile from 473 to 573 K. SCW was also used for the ultrafast re-polymerization of sulfonated Kraft lignin (SKL). Chemical structure of KL and its polymeric product after the SCW process was remarkably similar (Adamovic et al. 2022). Recycling of waste plastics by chemical decomposition in SCFs proceeds rapidly and selectively compared to conventional processes. Condensation polymerization plastics (as PET, polyurethane, and nylon) are relatively easily depolymerized to their monomers in SCW, SC MeOH (Kamimura et al. 2014) or alcohols. Selective decross-linking reactions in SCFs shows promising path for plastic recycling without severe decomposition of the backbone chains. SubH2 O, SCW as well as subcritical and supercritical benzyl alcohol are promising solvents for chemical recycling of thermoset composite materials, resulting in very clean carbon fibers after treatment, and the process is highly acceptable from both economic and environmental point of view (Morales Ibarra et al. 2015). Elmanovich et al. (2022) have concluded, that combination of the processes of polysiloxane depolymerization and plastic recycling in sc media can potentially lead to the emergence of a synergistic effect necessary for a breakthrough in the real industrial applications of both the chemical recycling of siloxanes and supercritical fluid technologies. Novel studies of plastic depolymerisation are presented in Table 4.2.

156

4 Industrial Scale Applications: Reaction-Based Processes

Table 4.1 Lignin depolymerisation using different SCFs and catalysts Material

Medium

Catalyst

Products

References

KL

Supercritical ethanol

Molybdenum-based catalysts

Highly-valued chemicals

Ma et al. (2015)

Concentrated sulphuric acid hydrolysed lignin (CSAHL) and KL

SubH2 O and supercritical methanol (SC MeOH)

NaOH, KOH, and Na2 CO3

High-yield aromatic monomers

Hidajat et al. (2017)

Alkali lignin

SC CO2

NiO

Phenolic monomers

Rajappagowda et al. (2017)

KL

SCW

/

Aromatic-based chemicals

Zurbel et al. (2019)

Lignin

Supercritical ethanol

CuMgAlOx catalyst

High monomer yield (23 wt%) without char formation

Huang et al. (2014)

Lignin

Supercritical methanol

Acidified modified SEP (m-SEP) and as-prepared CoO/ m-SEP catalysts

Monomers

Chen et al. (2019)

KL

SubH2 O

ZrO2 , K2 CO3 , and KOH

Phenolic monomers

Belkheiri et al. (2018)

Lignin

Supercritical ethanol

Ru/C and MgO/ZrO2 Phenolic monomers

Limarta et al. (2018)

Hydrolyzed lignin

Supercritical ethanol

Ni/Al-SBA-15(20)

Chen et al. (2017)

Monomers without char formation

Table 4.2 Plastic depolymerisation using different SCFs Material

System

Products

Nylon

NO2 /SC CO2

Aliphatic alpha, Yanagihara omega-diacids (succinic, and Ohgane glutaric and adipic acids) (2013)

References

Polybutylene terephthalate (PBT)

SC MeOH

100% degradation in 20 min at temperature > 513 K

Huang et al. (2005)

Nylon 12

Glycolic acid/supercritical MeOH

Hydroxyl esters

Kamimura et al. (2017)

Carbon fiber reinforced plastic

Supercritical solvent (methanol, 1-propanol, 2-propanol, 1-butanol, 2-butanol, tert-butanol, acetone, or methyl ethyl ketone)

Recovered carbon fibers and monomers

Okajima et al. (2017)

Poly(ethylene naphthalate) (PEN)

SubH2 O

2,6-naphthalene dicarboxylic acid (2,6-NDA) and ethylene glycol (EG)

Bei et al. (2017)

4.2 Chemical Reactions in SCFs

157

4.2.2 Carbonylation Carbonylation, acylation, hydrogenation and oxidation are multiphase catalytic processes, which are widely used in petroleum refining, petrochemicals, environmental pollution control and for the manufacture of a large variety of intermediate and products (Subramaniam et al. 2014). Among all carbonylation reactions performed in SCFs, hydroformylation of 1- alkenes and vinyl-arenes has been deeply studied in SC CO2 (Bektesevic et al. 2006; Jessop 2006). The first cobalt carbonyl catalyzed hydroformylation of propene in SC CO2 was performed in year 1991 by Rahtke et al. (1991). Conversion and regioselectivity is always very important for metalcatalyzed carbonylation. Among many parameters for increase in reactant conversion and product regioselectivity, the choice of a suitable solvent is one of the most important. Zhao et al. (2014) reported about improvement in alcoholysis of ethyl carbamate under catalyst-free conditions in supercritical ethanol. SC CO2 can be also used as a suitable reaction medium for free-radical carbonylation of alkyl halides. The carbonylation of aryl halides catalyzed by SC CO2 soluble Pd complexes with trialkyl or aryl phosphite ligands proceeds rapidly in SC CO2 and consequently the rate of the reaction is higher in comparison to solution phase reaction rates (Ikariya et al. 2000). Carbonylation annulation reaction between benzaldehyde and 2-hydroxy acetophenone in one-pot by using subH2 O as a solvent produced 64% of flavone product without by-products formation (Sirin et al. 2013). The alkylation of phenol in SC CO2 was carried out using gamma-Al2 O3 and cyclohexene and cyclohexanol as alkylating agents. The inhibition of catalyst was observed due to water formation in the reaction therefore, the reaction was proposed to be carried out as a two-step process (Amandi et al. 2007). The Fischer–Tropsch synthesis (FTS), which is classical heterogeneous reaction, is an attractive route for production of higher hydrocarbons and oxygenates, such as clean fuels, from syngas from different carbon sources like coal, natural gas, biomass etc. Performance of FTS reactions in SCF media (e.g. hexane, propane) have certain advantages over the traditional routes (Fan and Fujimoto 1999; Elbashir et al. 2010; Hao et al. 2012; Sari 2014; Durham et al. 2014). A large amount of solvent with high partial pressure is required during the supercritical Fischer–Tropsch synthesis (SC-FTS) therefore, recycling of the solvent is recommended in the SC-FTS process (Malek Abbaslou et al. 2009). Under identical temperature, pressure and contact time, CO conversion in supercritical media is more than or very close to that in gas phase FTS (GP-FTS). The syngas concentration and contact time decreased with addition of solvent and application of higher pressure at supercritical conditions. Higher thermal conductivity in supercritical phase leads to enhanced heat transfer. The better stability of the catalyst in the supercritical media is related to more uniform temperature distribution inside the supercritical fixed-bed reactor and to desorption of heavy hydrocarbons (Malek Abbaslou et al. 2009). Performing FTS in supercritical

158

4 Industrial Scale Applications: Reaction-Based Processes

media and using nitrogen-rich synthesis gas may also be beneficial for heavy product formation (Eliseev et al. 2018). Durham et al. (2013) constructed a novel supercritical adiabatic reactor design for FTS with lower capital cost requirements than traditional fixed-bed FTS reactors under both gas-phase and supercritical phase conditions and allows much easier catalyst replacement. The iron-based catalyzed production of lower olefins from alternative feedstocks (e.g. coal, natural gas, biomass) via FTS to olefin (FTO) process was recommended by Galvis and Jong (2013). This process can be performed using CO-reach syngas directly, without any H2 /CO ration adjustment, in view of their high water gas shift activity.

4.2.3 Oxidation Wastewater usually contains different organic compounds, such as pesticides, PCBs, herbicides, phenols, polycylic aromatic hydrocarbons, aliphatic and hetercyclic compounds. A potential treatment technology for organic wastewater is chemical oxidation technology. SCW is emerging as a medium that can provide the optimum conditions for a variety of chemical reactions, among them the destruction of hazardous waste (Suzuki et al. 1999; Tagaya et al. 1999; Dai et al. 2002; Bermejo and Cocero 2006; Sánchez-Oneto et al. 2007; Brunner 2009; Savage 2009; Subramaniam et al. 2014). The process for sub-critical process conditions is also referred to as wet air oxidation (WAO). WAO is based on the oxidizing properties of air’s oxygen. Typical conditions for wet oxidation range are: temperature from 180 °C to 315 °C, pressure from 20 to 150 bar, residence times may range from 15 to 120 min, and the chemical oxygen demand (COD) and total organic carbon (TOC) removal may typically be about 75–90% (Zheng et al. 2013). This technique is often used as a pre-treatment technique. The conversion of reaction intermediates (for example, acetic acid and ammonia) is usually very difficult in the absence of catalysts therefore the catalytic wet air oxidation (WACO) is performed (Fang 2016). When the process is performed under supercritical conditions, this is referred to as supercritical water oxidation (SCWO). Applications for supercritical water technology for can be divided in three main areas: (i) supercritical water oxidation (SCWO); (ii) supercritical water (SCW) biomass gasification; and (iii) hydrolysis of polymers in SCW for composites/plastics recycling. SCWO is a promising wastewater treatment technology. It has high destruction efficiency for a broad range of organic wastes and it is well used for commercial purposes in full-scale SCWO plants (Marrone 2013; Yang et al. 2019b), with the aim of performing total oxidation without a catalyst, in a few minutes with the possibility of precipitating inorganic pollutants. However, corrosion, salt deposition

4.2 Chemical Reactions in SCFs

159

and plugging, and high energy consumption and operating costs have to be taken in the consideration. Drilling fluid wastewater is waste originating from industries primarily engaged in refining crude oil and is a very complex compound of various oily wastes, water and heavy metals. SCWO is an effective process for removing organic compounds from drilling fluid wastewater, where chemical COD removal can reach 99% (Chao 2014; Liu et al. 2016). Some application of SCWO for wastewater-treatment are presented in Table 4.3. Table 4.3 SCWO for wastewater-treatment Components

Findings

References

Pyridine and 3-cyanopyridine

TOC removal efficiencies of pyridine and 3-CP were significantly improved as the oxidant dose ratio rose from 0 to 5, temperatures increased from 350 °C to 550 °C, and reaction time extended from 0.5 to 6 min

Bowen and Zhemin (2019)

Organophosphate flame retardants

TOC removal efficiencies of the OPFRs were significantly enhanced, when SCWO temperatures raised from 623 to 823 K and reaction time prolonged from 1 to 6 min

Yang et al. (2019a)

Quinoline

Moderate preheat temperature (420–510 °C) and Ren et al. (2019) initial concentration (1–10 wt%) are selected to address the possibility of utilizing the heat released during the reaction, in order to realize high conversion rate at relatively low preheat temperature

Nitrogen-containing compounds

Decomposition of nitrogen-containing compounds in the temperature range of 350–550 °C resulted in total nitrogen (TN) removal from 55 to 94%

Landfill leachate

Increasing temperature, OC and reaction time Gong et al. enhances the degradation of TOC and NH3 -N in (2018) landfill leachate undergoing SCWO without catalyst. The co-oxidation of landfill leachate with methanol increased the capability of SCWO remarkbly. Adding 5 vol% methanol to landfill leachate SCWO at 600 °C, 600 s and 1.7 OC, TRE and NRE increased from 89.2% to 98.9% and from 42.1% to 95.5%, respectively

Pesticide

X-COD and X-TN (removal efficiency of total Xu et al. (2015) nitrogen) were 99.42% and 86.70% at 600 °C, 250 bar, OC = 3.0 and t = 2.0 min, respectively. More than 92 wt% total organic carbon (TOC) and 86.70 wt% total nitrogen in the wastewater were converted into CO2 and N-2 under the above conditions

Yang et al. (2018)

160

4 Industrial Scale Applications: Reaction-Based Processes

Scandelai et al. (2018) performed landfill leachate degradation, which contained high concentrations of many organic and inorganic compounds, using combined processes of ozonation and SCWO. A combination of ozonation (30 min) and SCWO process (O3 -30’/SCWO) was the most efficient technique for the degradation of the leachate assessed. SCW technology also allows fast conversion of cellulose into sugars, and constitutes a suitable reaction medium for the synthesis of selected chemicals from biomass (Wu et al. 2014). A fast reaction rate, high selectivity and conversion yield of many biomass feedstocks can be achieved, when as a reaction medium SCW is used (Martínez et al. 2015). Besides, biomass can be selectively hydrolyzed in SCW to sugars within short reaction times (Cantero et al. 2014, 2015) and at low concentrations (Martínez et al. 2015). SubH2 O and SCW treatment of agricultural and food-processing waste is an alternative to conventional waste treatment technologies due to improvement of process performance, energy and economic advantages (Pavloviˇc et al. 2013). The main obstacles to scaling up this process are related to the harsh conditions and potential corrosion caused by usage of SCW; therefore, high-temperature- and high-pressureresistant materials such as Ni alloys with the addition of Cr and Mo are proposed to be used as material for equipment (Kim et al. 2010). The transformation of hydrocarbons in SCW is a synergistic process of extraction, fractionation and reaction. SCW oxidation of polycyclic aromatic hydrocarbons (PAH) was successfully performed in a batch reactor with 99.9 wt.% destruction of the PAH (Onwudili and Williams 2006). With the simultaneous increase in water density and temperature, the miscibility of PAHs with subH2 O/SCW is improved by the enhanced attractive electrostatic interaction between PAHs and hydrothermal environment and the weakened interaction between PAHs (Qu et al. 2019). SubH2 O and SCW were also used for oxidation of polycyclic aromatic compounds (phenanthrene and naphthalene). At lower temperatures, thermal degradation was achieved, but with increase in temperature hydroxylation of the aromatic moiety occurred, consequently the ring-opening and rearrangement reactions occurred (Onwudili and Williams 2007). Heterocyclic aromatic organic compound quinazoline was treated using SCWO. Most of the organic compounds decomposed remarkably in initial 60 s, but an appreciable improvement was not received after prolonging reaction time. Both temperature and time significantly influenced the CO yields. At temperature above 550 °C, CO formed from incomplete oxidation can be fully oxidized to CO2 as the temperature and time increased. Long reaction time has remarkable effect to promote TN removal efficiency (NRE) at temperature higher than 500 °C (Gong et al. 2016). Additionally, quantitative kinetic model for SCWO of quinazoline that describes the formation and interconversion of intermediates and final products at 673–873 K was also developed by Gong et al. (2017). SCW is uniquely green medium for diverse applications because of its changing nature from polar to non-polar. Owing to this property, it is being considered for heavy oil upgradation since it dissolves both organics (oil) and hydrogen while inorganics behave conversely. Upgrading heavy oil in SCW may be a promising alternative technique for acquiring clean light oil, given its relatively superior performance in

4.2 Chemical Reactions in SCFs

161

the transformation of hydrocarbons, the suppression of coke and the removal of heteroatoms (Canıaz and Erkey 2014; Li et al. 2015). The crude oil upgrading was conducted also in supercritical methanol (scMeOH) using batch reactors. The crude oil was upgraded into the light oil with more saturate and less aromatic resin and the decomposition of asphaltene is the most dominant reaction in the upgrading of crude oil while the generation of coke is suppressed (Kang et al. 2018). Pd(II)-catalyzed acetalization of terminal alkenes with alcohols, which undergo a Wacker-type process, ranks among the most important reactions for the functionalization of alkene feedstocks (Jiang et al. 2008). The acetalizations using Pd(II) as catalyst can be performed under supercritical conditions in SC CO2 and are defined as an environmentally friendly process (Jia et al. 1999; Wang et al. 2006). Wacker reaction can be carried out smoothly in SC CO2 or ROH/SC CO2 , where CO2 and the co-solvent can remarkably affect the selectivity towards methyl ketone, and the presence of ROH accelerates the reaction (Jiang et al. 2000). SCWO is also a promising method for the removal of emerging organic pollutants from hospital wastewater in one step and a very short reaction time without the need of additional treatment steps (Top et al. 2020).

4.2.4 Hydrogenation Hydrogenation reaction rates are often limited by the low hydrogen solubility in conventional solvents. Since SCF can completely solubilize hydrogen, the gas– liquid transport resistance in solid-catalyzed hydrogenations is therefore minimized or even eliminated. Both homogeneous and heterogeneous catalysts have been used, but heterogeneous catalysts are much more suitable for large-scale chemical productions because they can be applied in fixed-bed and continuous-flow reactor systems (Seki et al. 2008) and are among the most extensively studied conversions in SCF. The combination of an appropriate enantioselective catalyst with SC CO2 enables conducting continuous asymmetric hydrogenation (Cole-Hamilton 2006; Stephenson et al. 2006; Theuerkauf et al. 2013). Hydrogenation of vinyl benzene in supercritical media was researched by Altinel and co-workers (Altinel et al. 2009a, b). The heterogeneous hydrogenation of polystyrene (PS) was studied in the presence of SC CO2 , decahydronaphthalene (DHN) and palladium-based porous catalysts at 150 °C. Relatively high degrees of hydrogenation were obtained with monometallic palladium catalysts for the reaction conducted in neat DHN. However, when either palladium catalyst was used in SC CO2 -DHN, hydrogenation ceased within 15 min of CO2 addition to the reactor (Dong et al. 2010). A range of hydrogenation reactions in a continuous flow carbon nanoreactor in SC CO2 was successfully performed by Chamberlain et al. (2014). Hydrogenation of limonene into valuable chemicals was performed in SC CO2 in the presence of Pt or Pd as catalyst. The composition of products was affected by the flow rate of the reaction mixture through the stationary catalyst bed; the rate of limonene hydrogenation was increased wit increase in flow rates (Bogel-Łukasik et al. 2009). The SC CO2

162

4 Industrial Scale Applications: Reaction-Based Processes

hydrogenation (using Pd/CaCO3 catalyst) of complex pharmaceutical intermediate, rac-sertraline imine was performed in a continuous flow process, where superior levels of selectivity were obtained in the flow system, owing to the heat transfer properties of SC CO2 (Clark et al. 2007). Additional pharmaceutical drug phenylethanol (PhE), which is widely used as an anti-inflammatory and analgesic drug, was selective hydrogenated from acetophenone in SC CO2 over polyurea-microencapsulated mono- and bimetallic catalysts (More and Yadav 2018). The dominant effect of solvent polarity and density on the hydrogenation reaction was reported by different authors (Jessop and Leitner 1999; Combes et al. 2000).

4.2.5 Hydroformylation The oxo synthesis is an increasingly important chemical synthesis route, as it is one of the few reactions through which a single carbon atom can be added to a long chain species. The industrial processes vary depending on the chain length of the olefin to be hydroformylated, the catalyst metal and ligands, and the recovery of the catalyst. SC CO2 is an environmentally benign reaction medium for highly efficient noble metal-catalyzed hydroformylation reactions (Table 4.4). Usage of catalysts for hydroformylation of alpha-olefins to form linear aldehydes and for asymmetric hydroformylation of styrene in SC CO2 , along with the effect of the supercritical fluid solvent on these reactions was well documented Bektesevic et al. (2006). An overview of continuous flow homogeneous hydroformylation of alkenes using SCFs and design of a plant suitable for operating these reactions in a totally emission-free fashion as well as comparisons of these systems with current commercial systems was done by Webb et al. (2005). Highly efficient continuousflow asymmetric catalysis was achieved by combination of supported ionic liquid phase (SILP) catalysts with SC CO2 as the mobile phase, as demonstrated for enantioselective hydrogenation in the presence of a molecular rhodium–QUINAPHOS complex. The integrated reaction and separation process yielded chemically and enantiomerically pure products without the need for organic solvents (Hintermair et al. 2010).

4.2.6 Hydrothermal Synthesis The idea of recreating on land the state of high temperature and pressure that is characteristic of hydrothermal deposits in the depths of the oceans was already expressed 25 years ago. Excellent examples of hydrothermal reactions are supplied by nature, since numerous minerals have been formed under these circumstances. Water in the supercritical state, it mixes together with oil, and the water molecules themselves have the function of an acid and base catalyst. Therefore, no need to use

Catalyst

Fluoroacrylate polymer supported Rh-catalysts

Tris(3,5-bis[trifluoromethyl]phenyl)phosphine-modified Rh-catalyst

Rh-catalyst

Molecular rhodium–QUINAPHOS complex

Substrates

Styrene

1-Octene

1-Octene

Long-chain alkenes

Table 4.4 Hydroformylation in SC CO2 Findings

References

Koeken et al. (2011)

Kani et al. (2004)

Successful performance of highly selective hydroformylation in a supercritical fluid ionic liquid biphasic system was done

(continued)

Kunene et al. (2011)

The presence of peracetylated Tortosa Estorach et al. beta-cyclodextrin in the reaction allowed (2008) for an increase in the solubility of rhodium species modified by alkyl P-donor ligands

The reaction rate had a positive order in 1-octene at a concentration lower than 0.5 mol/L while saturation kinetics were observed at a higher concentration

Conversions up to almost 100% and branched aldehyde selectivities of 95–100% were obtained under most reaction conditions (T = 323 and 348 K and p = 17.2, 20.7, and 241 bar)

4.2 Chemical Reactions in SCFs 163

Enantioselectivity of >99% ee and Hintermair et al. (2013) quantitative conversion were achieved after 30 h of hydoformylation, reaching turnover numbers beyond 100 000 for the catalyst. The optimised system reached stable selectivities and productivities that correspond to 0.7 kgL−1 h−1 space–time yield and at least 100 kg product per gram of rhodium, thus making such processes attractive for larger-scale application

Chiral transition-metal complex (QUINAPHOS–rhodium complex)

References

Findings

Catalyst

Substrates

Dimethylitaconate

Table 4.4 (continued)

164 4 Industrial Scale Applications: Reaction-Based Processes

4.2 Chemical Reactions in SCFs

165

a toxic organic solvent and a catalyst arises when the SCW is used as the reaction medium. Another benefit of using SCW is that the equilibrium and speed of reaction can be varied by small changes in pressure and/or temperature therefore, various reactions, such as biomass conversion, waste recycling, organic synthesis and nanoparticle synthesis can be performed in SCW. Hydrothermal techniques enables the synthesis of some new materials including single crystals materials, zeolites and related microporous materials, ionic conductors, complex oxides and fluorides, inorganic–organic hybrid materials, and nanomaterials (Feng and Li 2017). The characteristic phase behavior of SCW enables organic ligand modification, and fine tuning of particle size and morphology can be realized by the method. Synthesis of metal nanoparticles is possible in homogeneous reductive atmosphere provided in SCW (Darr et al. 2017). Nanomaterials with high vapor pressures can be produced by the hydrothermal method with minimum loss of materials. The compositions of nanomaterials to be synthesized can be well controlled in hydrothermal synthesis through liquid phase or multiphase chemical reactions. During production of nanoparticles in SCW (Adschiri et al. 2012; Taguchi et al. 2014), the oil, as a surfactant, could be added to bind to the surface of metal oxides to produce organically-modified nanoparticles of ≤ 10 nm. Hydrothermal technology in the twenty-first century is not just confined to the crystal growth or leaching of metals, but assumes a very broad scope, covering several interdisciplinary branches of science (Byrappa and Adschiri 2007). The principles of hydrothermal synthesis as well as the chemistry of green materials synthesized with SCFs (Adschiri et al. 2011) are well documented by Adschiri et al. (Adschiri and Byrappa 2009; Adschiri et al. 2013). Octanoic acid concentration influence the formation of an intermediate with the octanoic acid which controls particle growth of boehmite (AlOOH) in subH2 O and SCW (Fujii et al. 2016a). Additionally, the morphology of AlOOH nanoparticles can be controlled over a short timespan by supercritical hydrothermal treatment in the presence of alkyl carboxylic acids (e.g. hexanoic, octanoic, decanoic, tetradecanoic and octadecanoic acids) (Fujii et al. 2016b). Flow-type reactors are effective for the precise control of reaction conditions and high throughput production. To enhance the effectiveness of this process, Aoki et al. (2016) studied the kinetics to identify reaction-controlled conditions for supercritical hydrothermal nanoparticle synthesis with flow-type reactors. Ferroelectrics nanoparticles were obtained using SCFs technology which exhibits very interesting characteristics such as fast continuous synthesis (few seconds) of high quality nanoparticles (well crystallized nanoparticles with narrow size distribution) at intermediate synthesis temperatures (4000 bar) can lead to irreversible structural changes in the enzyme, but within the pressure range of 100–400 bar, only reversible conformational changes can occur. Many of these reversible changes do not affect the overall catalytic performance of the enzyme. A high temperature is always destructive, especially when applied over a long period (Rezaei et al. 2007). An overview of process parameters influencing enzyme activity and stability, reaction rates and productivity, applications of various types of reactors for enzymatic reaction in SCFs and the limitations of using enzymes as biocatalysts in SCF are well presented by Knez and co-authors (Knez 2009, 2018; Knez et al. 2015b). Several publications on oxidation (Randolph et al. 1988a, b; Findrik et al. 2005), hydrolysis (Muratov et al. 2005; Habulin et al. 2005; Primozic et al. 2006; Guthalugu et al. 2006; Guthalugu et al. 2006; Bártlová et al. 2006; Salgın et al. 2007; Sovová et al. 2008), transesterification (Madras et al. 2004; Weber et al. 2008), esterification (Šabeder et al. 2005; Romero et al. 2005; Lin et al. 2006; Ghaziaskar et al. 2006; Varma and Madras 2007a; Varma and Madras 2007a; Laudani et al. 2007; Knez et al. 2007), and enantioselective synthesis (Nakamura et al. 2003; Matsuda et al. 2004; Salgın et al. 2007; Palocci et al. 2008) have proved the feasibility of enzymatic reactions in SCFs. Lately, some new applications of enzymes in SCFs as reaction media have been reported (Table 4.5).

4.3.1 Influence of SCFs on Enzyme Activity and Stability Enzymes after recovering from reaction media, can often lost their catalytic activity and denaturation may occur. Therefore, enzyme activity enhancement represents a great opportunity for biotechnological production. Many authors reported that activity and stability of enzyme can be improved by using SCFs as potential media for enzyme-catalysed reactions (Matsuda 2013; Senyay-Oncel and Yesil-Celiktas 2013, 2015). The crucial point in and basis for developing and applying green, environmentally friendly enzyme-catalysed processes and/or reactions in SCFs for different industrial applications is usage of active and stable enzyme. The activity and stability of different enzymes in SCFs (Table 4.6) could be improved due to conformational changes (Peng et al. 2016), but enzyme activity changes are significantly dependent on the enzyme species and on the experimental conditions (e.g. temperature and pressure) in the reaction system (Carvalho et al. 2014). Nowadays, enzyme-catalysed reaction in SCFs are oriented toward polymerisation process, lignocellulosic biomass conversion and synthesis of biodiesel. Enzyme-catalysed reaction are often also carried out in IL/SCFs media.

168

4 Industrial Scale Applications: Reaction-Based Processes

Table 4.5 Enzyme-catalysed reactions in SCFs Product

Enzyme

SCFs

Findings

References

Galacto-oligosaccharides Beta-galactosidase from a (GOS) permeabilized cell of Kluyveromyces marxianus

Various GOS production in Manera compressed the SC CO2 led to et al. fluids (2012) maximum production of 83 g/ L while, for propane and n-butane, values of 63 and 75 g/L were verified, respectively

Geranyl butyrate

Novozym 435, lipase B from C. antarctica

Various SCFs

Initial rate of transesterification of butyl butyrate in different SCFs followed in this order: SC CO2 < SC C2 H6 < SC C2 H4 < SC CH4

Varma and Madras (2010)

Eugenyl acetate

Lipozyme 435, lipase from C. antarctica and Novozym 435, lipase B from C. antarctica

SC CO2

Optimal condition for the synthesis of eugenyl acetate in SC-CO2 was determined at 60 °C and 100 bar

dos Santos et al. (2016)

Isoamyl acetate

Novozym 435, lipase B from C. antarctica

SC CO2

The highest dos Santos productivity of et al. isoamyl acetate (2017) was obtained in packed bed reactor, although its conversion was lower than that obtained by the batch reactor

Ethanolysis of fish oil

Lipozyme RM IM, SC CO2 lipase from Rhizomucor miehei

Lipase-catalyzed ethanolysis in SC CO2 has been shown as suitable method to obtain less oxidized n-3 PUFA FAEE compared to other reaction media

Melgosa et al. (2017)

(continued)

4.3 Biochemical Reactions in SCFs

169

Table 4.5 (continued) Product

Enzyme

Geranyl acetate

Lipozyme RM IM, SC CO2 lipase from Rhizomucor miehei and Novozym (R) 435, lipase B from C. antarctica

SCFs

Findings

References

Novozym 435 Tavares took nearly half the et al. (2022) time to achieve similar reaction Conversions compared to Lipozyme RM IM

Table 4.6 Activity and stability of enzymes in SCFs Enzyme

SCFs

Findings

Burkholderia cepacia lipase (BCL)

Compressed propane

In compressed propane, the lid Housaindokht and of the lipase was opened and so Monhemi (2013) the active conformation of the enzyme was resulted and the enzyme has native conformation, similar to the aqueous solution

References

Inulinases from Subcritical Kluyveromyces propane marxianus NRRL Y-7571 immobilized onto natural montmorillonite

The enzyme activity changed Kuhn et al. (2011) significantly depending on the investigated experimental conditions. Stability of the enzyme after high-pressure pre-treatment was improved; the activity of the treated biocatalyst was always higher than the activity of the non-treated one

Compressed CO2, Beta-galactosidase from Kluyveromyces propane and marxianus CCT n-butane 7082

The use of the n-butane-pretreated enzyme led to very high reaction conversions and selectivity

Manera et al. (2011)

α-amylase

SC CO2

SC CO2 pretreatment helped to enhance the activity yielding in 67.7% higher activity than the untreated enzyme at optimum conditions: 240 bar, 41 °C, 4 g/ min CO2 flow and 150 min of process duration

Senyay-Oncel and Yesil-Celiktas (2011)

Compressed propane

An enhancement in residual lipase activity in most of the investigated conditions for free and immobilized lipases was observed

Franken et al. (2010)

(continued)

170

4 Industrial Scale Applications: Reaction-Based Processes

Table 4.6 (continued) Enzyme

SCFs

Findings

Horseradish peroxidase (HRP)

Compressed CO2 and propane

HRP treatment in CO2 promotes Fricks et al. a significant decrease in the (2009) enzyme activity, whereas HRP showed good stability in propane; HRP is more stable in propane than in CO2 . Treatment of the HRP with propane did not induce changes in the secondary structure, while incubation in CO2 led to significant loss of HRP secondary structure

Novozym 435, lipase B from C. antarctica

SC CO2

Significant increase of activity of Yu et al. (2007) Novozym 435, and a maximum residual activity of 300% was measured at 40 bar, 30 °C after 7 h incubation was observed

Immobilized lipases Compressed CO2 , (Lipozyme RM IM, propane and lipase from n-butane Rhizomucor miehei and Novozym 435, lipase B from C. antarctica) Yarrowia lipolytica lipase

References

Activity losses were verified for Lipozyme IM in all compressed solvents, markedly in SC CO2 . For Novozym 435, treatment in SC CO2 also led to activity losses, while the use of propane and n-butane improved enzyme activity

Oliveira et al. (2006)

Compressed CO2 , Significant activity losses were propane and obtained when the lipase was n-butane treated with CO2 , while negligible losses were observed in both propane and n-butane

Oliveira et al. (2006)

Tyrosinase from Compressed CO2 mushroom Agaricus bisporus

Complete inactivations were achieved when the enzyme solution was subjected to pressures as high as 200 bar for less than 9 min within dependence of the temperature used (25–45 °C)

Horseradish peroxidase from Amoracia rusticana root and mushroom polyphenol oxidase from Agaricus bisporus

SC CO2 had a significant Marszałek et al. influence on polyphenol oxidase (2019) (PPO) and peroxidase (POD) activities, while observing an increased reduction in the residual activities of both enzymes when the pressure was increased

SC CO2

Benito-Román et al. (2019)

(continued)

4.3 Biochemical Reactions in SCFs

171

Table 4.6 (continued) Findings

References

Immobilized C. SC CO2 antarctica Lipase B (Novozyme NZL-102-LYO-HQ)

The residual activity of CALB immobilized in polyurethane was 315%, reaching 2486 U/g and maintained 100% of its initial activity when reused for 10 cycles

Nyari et al. (2018)

Hen egg-white lysozyme

Compressed CO2 and R134a

The higher residual activity (141 Souza et al. (2017) ± 1.8%) was obtained for SC CO2 at 150 bar, for 2 h and with depressurizing rate of 30 bar/ min. Increase in residual activity was detected also after enzyme exposure in compressed R134a

Pectinase from Aspergillus niger

SC CO2

Significant increases in activity Peng et al. (2016) and stability of treated pectinase was detected at pressure lower than 150 bar, whereas, temperature tends to reduce enzymatic activity and stability

Lipase B from Candida antarctica

SC CO2

Residual specific hydrolytic activities of 132% was observed when CALB was exposed to SC CO2 at 35 °C, 75 bar and 1 h

Enzyme

SCFs

Feiten et al. (2023)

4.3.2 Enzyme-Catalysed Polymerization in SCFs Polymerization processes catalyzed by enzymes in SCFs are becoming more attractive, owing to the importance of clean processes, which produce substances free of residues and ideal for pharmaceutical and food applications. In the last decade, an enormous increase in application of enzymes as catalysts for polymerization has been observed. Mild polymerization conditions, high enantio- and regio-selectivity and recyclability of enzymes predominated over the use of organo-metallic catalysts for polymerization processes (Albertsson and Srivastava 2008). To replace the petro-chemical-based synthetic polymers, aliphatic polyesters, which are thermoplastic, biodegradable and biocompatible, could be synthetized by enzyme-catalyzed ring-opening polymerization (ROP) in SCF solvents (Zhao 2018). ROP is the most common method of synthesizing polyesters for biomedical applications due to ability to control polymer properties and architecture during this process. In Table 4.7, latest researches on ROP are presented. Other SCFs e.g. freon (compressed R134a) was also used as a reaction medium for enzyme-mediated synthesis of linear and hyperbranched polyesters (López-Luna et al. 2010). Among ROP, which is most often enzyme-catalysed polymerization process performed in SCFs, the enzymatic transformation of aliphatic polyesters

SC CO2

SC CO2

Immobilized Novozym 435, lipase B from C. antarctica

lipase from Burkholderia cepacia

Immobilized Novozym 435, lipase B from C. antarctica

Different comercial available lipases

Poly(epsilon-caprolactone)

L-lactide

Omega-pentadecalactone

Polylactides and polylactones

subcritical and supercritical 1,1,1,2-tetrafluoroethane (scR134a)

SC CO2

SCFs SC CO2

Enzyme

Immobilized Novozym 435, lipase B from C. antarctica

Reactant

Omega-pentadecalactone

Table 4.7 Enzyme-catalyzed ring-opening polymerization in SCFs Findings

References

Key factors controlling the ROP reactions include the types of lipases and solvents, the solvent concentration, water contents in enzymes, substrates and solvents, and the reaction temperature

Reaction yield using only SC CO2 as solvent was 60 wt% and molecular weights was up to 33,000 g/mol. A strong influence of water content on enzyme and monomer concentration in reaction medium over the final polymer molecular weight and reaction yield was detected

Best results were obtained at 105 °C and 50 bar using the thermostable, with a maximum poly-L-lactic acid yield above 50%

Polymerization results for the kinetic experiments indicate reaction yields up to 90 wt%, M–n up to 13,700 Da and M–w up to 22,200 Da, with P.I. ranging from 1.2 to 1.7

Zhao (2018)

Polloni et al. (2017)

Guzmán-Lagunes et al. (2012)

Comim Rosso et al. (2013)

Higher molecular weights were obtained when a Polloni et al. (2017) co-solvent was used, and a strong influence of water content on enzyme and monomer concentration in the reaction medium over the final polymer molecular weight and reaction yield was observed

172 4 Industrial Scale Applications: Reaction-Based Processes

4.3 Biochemical Reactions in SCFs

173

into cyclic oligomers was performed in enzyme-packed column reactor with immobilized lipase from C. antarctica (Novozym 435) under a continuous flow of SC CO2 , with toluene as a co-solvent (Osanai et al. 2006). All tested polymers were quantitatively transformed into the corresponding cyclic oligomers at 40 °C under a continuous flow of SC CO2 with toluene. Degradation of the polymer was significantly promoted with respect to the reaction time, temperature and polymer concentration for complete transformation of the polyesters into oligomers through the enzyme column when SC CO2 was added as a mobile phase. Baheti et al. (2018) synthesised star-shaped polymers composed of a d-sorbitol core and polycaprolactone arms by ROP in SC CO2 using two different catalysts: conventional metal catalyst tin(ii) 2-ethylhexanoate (Sn(Oct)2 ) and an immobilized enzyme, Novozym 435 (Lipase B from C. Antarctica). Regular star polymers were obtained in the presence of Sn(Oct)2 whereas Novozym 435 gave access to miktoarm-type star PCL. Continuous mode enzyme-catalysed polymerizations using CO2 as solvent and was Novozym435 as a biocatalist, were also carried out for reactions at 120 bar and 200 bar, 65 °C and solvent/monomer mass ratios of 2:1 and 1:2. Results show, that the pressure has no significant influence over the parameters evaluated, while the solvent/monomer mass ratio and enzyme content presented significant effect on reaction yield (Rosso Comim et al. 2015).

4.3.3 Reactors for Enzyme-Catalysed Processes Under High Pressure In Fig. 4.2 most common used reactors for enzymatic reaction in SCFs such as high-pressure enzymatic batch stirred tank reactor and high-pressure packed-bed enzymatic reactor are presented. Hence, it is necessary to understand enzyme functions in SCFs before it becomes possible to move supercritical fluid enzymology from laboratory to industry scale (Figs. 4.3 and 4.4).

Fig. 4.2 Continuous tubular reactor

174

4 Industrial Scale Applications: Reaction-Based Processes

Fig. 4.3 Continuous packed-bed reactor

Fig. 4.4 HP continuous flat-shape membrane reactors

4.3 Biochemical Reactions in SCFs

175

4.3.4 Investigations to Perform Biochemical Reaction in High Pressure Batch Stirred Tank Rector—HP BSTR Biocatalysis in CO2 at supercritical conditions were in the past mostly performed in single phase reaction mixture (Comim Rosso et al. 2013). The major drawback of enzymatic reactions in SC-CO2 carried out in a single supercritical phase is the high pressures (on the order of hundreds of bars) required to ensure entire solubility of mostly used organic compounds in CO2 . Today are only few reports on performing catalytic reactions carried out in subcritical conditions (Darr and Poliakoff 1999; Chouchi et al. 2001). The challenge in developing catalysis in “CO2 expanded” reaction media have been recently explained on the basis of the high solubility of carbon dioxide in many organic liquids (Wei et al. 2002). Due to the high solubility of CO2 in the liquid reactants makes it an “expanded fluid” avoiding the need to generate a single supercritical phase and providing unexpected advantages (Musie et al. 2001). Indeed, simply ensuring that a significant amount of CO2 is present in the liquid phase may be sufficient to gain kinetic control over the reaction. Performing biocatalysis in CO2 expanded reaction media, offers several advantages similar to those where reactions are performed in single phase of reaction media. The investigated model system contain is focused on the biosynthesis of longchain fatty acid esters in high-pressure carbon dioxide catalysed by lipase from Rhizomucor miehei (Comim Rosso et al. 2013). For design of HP BSTR the influence of several process parameters on the reaction rate and the product yield has to be studied.

4.3.4.1

Phase Equilibrium of Reaction System

Visual observations and quantification of phase equilibrium has to be determined to define equilibrium pressure–temperature at which phase separation occurred at known composition of the mixture, to obtain the location of liquid–vapour phase boundaries and evaluate solubility of mixture by using the synthetic method. In Fig. 4.5 the phase equilibrium phenomena of reaction system oleic acid/1-octanol/ CO2 at different temperatures and pressures is presented.

4.3.4.2

Pressure Effect

Pressure affect the reaction performance by changing either the rate constant or the solubility of reactants in reaction media. The effect of the pressure on enzymes activity and stability has to be investigated in details!

176

4 Industrial Scale Applications: Reaction-Based Processes

Fig. 4.5 The phase equilibrium phenomena of reaction system oleic acid/1-octanol/CO2 at temperature of 323.15 K; at a 100 bar; b 200 bar; c 300 bar

Table 4.8 Thermodynamic properties for synthesis of n-octyl oleate at 1 and 300 bar Pressure (bar) 1

80

150

200

300

Deactivation enthalpy (kJ/mol)

159.54

165.22

163.47

222.01

305.58

Deactivation constant

4.46

0.62

0.99

1.23

1.84

Free energy of deactivation (kJ/mol)

−4.015

1.284

0.027

−0.556

−1.637

Entropy of deactivation (kJ/molK)

0.506

0.507

0.506

0.689

0.951

Reaction conditions: equimolar ratio of substrates (2.1 M), biocatalyst concentration of 5.45% (w/ w of substrates), 500 rpm

4.3.4.3

Temperature Effect

Temperature significantly affects enzyme catalysis in CO2 expanded reaction media, as the temperature effect is related to the enzyme activity and stability and to the CO2 solvating power. Based on the obtained data Arrhenius plot could be constructed where both pressures the activation and deactivation energies, the equilibrium constant, the deactivation entropy and deactivation Gibb’s free energy were estimated. These values are presented in Table 4.8 (Fig. 4.6).

4.3.4.4

Stirrer Speed Effect

The influence of the stirrer speed on esterification reaction has to be investigated to determine the stability of biocatalyst but more important to determine the initial reaction rate vs. stirring rate which influence the transfer of reactants between the organic phase and the biocatalyst surface.

4.4 Conclusions

177

Fig. 4.6 Arrhenius plot for the synthesis of n-octyl oleate in SC-CO2 at different pressures

4.3.4.5

Water Content in Reaction Media

The influence of the water content on enzyme activity and equilibrium conversion has to be investigated to define the amount of water required to reach the highest enzymatic activity without shifting the equilibrium towards hydrolysis.

4.3.4.6

Effect of Substrates Molar Ratio

The influence of substrates molar ratio on the reaction rate and on the conversion of fatty acid in ester has to investigated for the reaction system. Since the esterification reactions are reversible, an increase of alcohol concentration may result in higher FAA conversion, shifting the chemical equilibrium towards the product synthesis.

4.4 Conclusions Till now practically no process on enzymatic synthesis in sub or supercritical fluids were performed on industrial scale. But based on the several advantages of these processes we could be sure that in the future they will play an important role in industrial production of several substances.

178

4 Industrial Scale Applications: Reaction-Based Processes

Main advantages of using sub and supercritical fluids as reaction media for biocatalyzed reactions are: • the tunability of solvent properties which influences the activity, stability and selectivity of several enzyme species, • CO2 is an inert solvent and therefore does not usually oxidize substrate, CO2 has also extremely favourable attributes for it application as reaction media in biocatalysed reactions such as inflammability, nontoxicity and low costs, • advantages of enzymatic synthesis in sub or supercritical fluids is the ease of substrate and product fractionation and what is more and more important in the last time, integration of reaction and separation units of production process. More benefits and drawbacks on enzymatic synthesis in sub or supercritical fluids that are well presented in Knez (2018). Research on new types of reactors (microreactors, membrane bioreactors) for continuous biocatalysis in dense fluids will reduce the high pressure part of processing units (what will also to reduce investment costs) and will employ continuous processing what will reduce operating costs. Most of the research on biocatalysis in subcritical and supercritical fluids was made in carbon dioxide as solvent which have several limitations. Drawbacks on enzymatic synthesis in sub or supercritical CO2 as reaction media are, that CO2 shows low solubility of several, the enzyme activity could be changed or enzymes could be completely deactivated. Therefore, new solvent systems are investigated. Very promising processes are in development for synthesis of biodisel and bioconversion of lignocelulosic materials.

4.4.1 Enzyme-Catalysed Synthesis of Biodiesel and Lignocellulosic Biomass Bioconversion in SCFs Depleting fossil fuel reserves and an increasing awareness of the impact of energy production on society and the environment has encouraged the research on the field of cleaner energy resources. Biodiesel and bioethanol are set to remain the primary replacement fuels for fossil-based diesel and petroleum respectively. Biodiesel produced from oil-rich feedstocks is known as a green replacement for conventional petroleum diesel. Using alternative solvents, (SCFs, ILs, SCFs/ILs system, etc.) for enzyme-catalysed biodiesel production (Varma and Madras 2007b; Lee et al. 2009; Taher et al. 2011; Ciftci and Temelli 2013; Gutiérrez-Arnillas et al. 2016; Taher and Al-Zuhair 2017) provided to be a prospective synthesis technique. Dimethyl carbonate and methyl acetate have been shown to produce high FAME yields and valuable by-products when used as acyl acceptor for biodiesel production. Dimethyl carbonate is favoured in literature for use with lipase-based biocatalysts while methyl acetate is favoured to be used under supercritical reaction conditions (Marx 2016). The continuous processes for enzyme-cazalysed synthesis of biodiesel

4.4 Conclusions

179

in SC CO2 were demonstrated as a promising technology for large-scale purposes by various authors (Ciftci and Temelli 2011; Colombo et al. 2015; Gameiro et al. 2015). Biodiesel was synthesised not only in SC CO2 but also in supercritical alcohols (methanol and ethanol) by non-catalytic path. Total conversion was observed in supercritical alcohols, whereas only a 70% conversion was observed for the enzymatic synthesis in SC CO2 (Varma and Madras 2010). Besides, the production of fatty acid ethyl esters from soybean oil in compressed CO2 , propane and n-butane using immobilized lipase B from C. antartica (Novozym 435) as a catalyst was performed in continuous tubular reactor. Lipase-catalyzed alcoholysis in a continuous tubular reactor using compressed propane might be a potential biodiesel production route with high reaction conversion (Dalla Rosa et al. 2009). Pre-treatments with near critical fluids and SCFs have the potential to reduce structural obstacles of lignocellulosic materials and to enhance their anaerobic biodegradability (Rosero-Henao et al. 2019). Zetzl et al. (2011) reported about the different high-pressure processes for enzymecatalysed production of chemicals from lignocellulosic biomass and the possible combination of different process steps as well as the scale-up possibilities. Another promising solvent are ionic liquids, even due to their non-green solvents due to environmental negative effect.

4.4.2 Enzymatic Reactions in IL/SCFs Media Enzymatic reactions based on ionic liquids (ILs) and SCFs, as non-aqueous reaction media, presents alternatives to organic solvents for designing clean synthetic chemical processes that provide pure products directly. Biphasic systems based on ILs and SC CO2 have been proposed as the first approach to achieving integral green bioprocesses in non-aqueous media. The usabillity of IL/SC CO2 biphasic systems, as reaction media for enzymatic reaction, is based on the fact that ILs provide an adequate microenvironment for the high catalytic efficiency of enzymes, while SC CO2 can act as extracting, disolving or transporting phase, making possible the easy recovery of the products. An enormous advantage of SC CO2 is expressed in the miscibility switch phenomenon of CO2 (Bermejo et al. 2008). Due this ability of SC CO2 , two immiscible phases can form one homogeneous fluid phase for the reaction performance. After the completion of the reaction, the homogeneous fluid phase could be split into two or three phases upon pressure decrease in order to facilitate product recovery. Enzyme behavior in SC CO2 and ILs, as well as the phase behavior of ILs/SC CO2 , are key parameters for carrying out integral green bioprocesses in continuous operation. Using continuous flow techniques for multi-step synthesis (i.e. dynamic kinetic resolution (DKR) of sec-alcohols, synthesis of biodiesel, etc.) in ILs/SC CO2 media enables multiple reaction steps to be combined into a single continuous operation (Lozano et al. 2011, 2017).

180

4 Industrial Scale Applications: Reaction-Based Processes

Operational stability of the Novozym 435, immobilized Candida antarctica lipase B was improved by coating it with IL (1-methyl-3-octadecyl imidazolium hexafluorophosphate) and it showed excellent catalytic behaviour in continuous operation under SC CO2 conditions (up to 82% biodiesel yield after 12 cycles). Enzymatic DKR of rac-1-phenylethanol in IL/SC CO2 in the was successfully performed with high yield (up 98.0%) for R-phenylethyl propionate ester and excellent enantioselectivity of Novozym 435 (up to 97.3%), resulting in no activity loss after 14 days of operation (Lozano et al. 2009). Dynamic membranes with immobilized C. antarctica lipase B for butyl propionate synthesis in a recirculating bioreactor in SC CO2 and in IL/SC CO2 biphasic systems was used. When IL/SC CO2 biphasic systems were used, the selectivity of the process increased up to > 99.5% compared with when SC CO2 assayed alone, although the synthetic activity (Ucm−2 ) was lower (Hernández et al. 2006). Lipase-catalyzed synthesis of lauroyl glucose ester was continuously performed in a IL/SC CO2 biphasic system. The enzyme activity in the biphasic system was higher than that in the pure IL. The combination of supercritical CO2 and ILs could not only improve the reaction rate and yield, but also make the product separation and enzyme/ILs recycling easier. Moreover, the successful enzyme-mediated synthesis of linear and hyperbranched poly-L-lactides was performed in homogenous media, composed of ionic liquid [C4 MIM] [PF6 ] in combination with miscible compressed R134a, above the critical point of the mixture, at 338.15 and 363.15 K and at a pressure of 300 bar. Nonenzymatic propagation side mechanisms were minimized or eliminated with the addition of the green compressed R134a, as well (Mena et al. 2015).

4.4.3 Future Trends From ecology point of view supercritical fluid based technologies however offer important advantages over organic solvent technologies. Water is till now the cheapest solvent, high solubility of several substrates in water and high activity of enzymes in water are probably one of the major reasons that there are no enzyme biotransformations performed in subcritical or supercritical fluids on industrial scale till now. Due to high environmental concerns and due to increase of organic solvent prices the demand on use of new solvents increases. Natural biocatalysts (enzymes, whole cells, cell debris) in environmental friendly second cheapest solvent, CO2 , and with the new challenges and opportunities will overcome the current gaps and will in the future open new pathways for production of several organic substances substances by biocatalysis in dense fluids on an industrial scale.

References

181

References Adamovic T, Zhu X, Perez E, Balakshin M, Cocero MJ (2022) Understanding sulfonated kraft lignin re-polymerization by ultrafast reactions in supercritical water. J Supercrit Fluids 191:105768 Adschiri T, Byrappa K (2009) Supercritical hydrothermal synthesis of organic-inorganic hybrid nanoparticles. In: Muramatsu A, Miyashita T (eds) Nanohybridization of organic-inorganic materials. Springer, Berlin, Heidelberg, pp 247–280 Adschiri T, Lee Y-W, Goto M, Takami S (2011) Green materials synthesis with supercritical water. Green Chem 13(6):1380–1390 Adschiri T, Takami S, Arita T, Hojo D, Minami K, Aoki N, Togashi T (2013) Handbook of advanced ceramics: Chapter 11.1.5. In: Supercritical hydrothermal synthesis. Elsevier Inc. Chapters Adschiri T, Takami S, Minami K, Yamagata T, Miyata K, Morishita T, Ueda M, Fukushima K, Ueno M, Okada T, Oshima H, Mitani Y, Asahina S, Unno S (2012) Super hybrid materials. Mater Sci Forum 700:145–149 Alaimo D, Grignard B, Kuppan C, Adriaensen Y, Genet MJ, Dupont-Gillain C, Gohy J-F, Fustin CA, Detrembleur C, Jérôme C (2017) A photocleavable stabilizer for the preparation of PHEMA nanogels by dispersion polymerization in supercritical carbon dioxide. Polym Chem 8(3):581– 591 Albertsson A-C, Srivastava RK (2008) Recent developments in enzyme-catalyzed ring-opening polymerization. Adv Drug Deliv Rev 60(9):1077–1093 Altinel H, Avsar G, Guzel B (2009a) Fluorinated rhodium-phosphine complexes as efficient homogeneous catalysts for the hydrogenation of styrene in supercritical carbon dioxide. Transit Met Chem 34(3):331–335 Altinel H, Avsar G, Yilmaz MK, Guzel B (2009b) New perfluorinated rhodium–BINAP catalysts and hydrogenation of styrene in supercritical CO2 . J Supercrit Fluids 51(2):202–208 Alzahrani A, Zhou D, Kuchel RP, Zetterlund PB, Aldabbagh F (2019) Polymerization-induced self-assembly based on ATRP in supercritical carbon dioxide. Polym Chem 10(21):2658–2665 Amandi R, Scovell K, Licence P, Lotz TJ, Poliakoff M (2007) The synthesis of o-cyclohexylphenol in supercritical carbon dioxide: towards a continuous two-step reaction. Green Chem 9(7):797–801 Aoki N, Sato A, Sasaki H, Litwinowicz A-A, Seong G, Aida T, Hojo D, Takami S, Adschiri T (2016) Kinetics study to identify reaction-controlled conditions for supercritical hydrothermal nanoparticle synthesis with flow-type reactors. J Supercrit Fluids 110:161–166 Baheti P, Gimello O, Bouilhac C, Lacroix-Desmazes P, Howdle SM (2018) Sustainable synthesis and precise characterisation of bio-based star polycaprolactone synthesised with a metal catalyst and with lipase. Polym Chem 9(47):5594–5607 Bártlová M, Bernášek P, Sýkora J, Sovová H (2006) HPLC in reversed phase mode: tool for investigation of kinetics of blackcurrant seed oil lipolysis in supercritical carbon dioxide. J Chromatogr B 839(1):80–84 Bei K, Ma P, Wang J, Li K, Lyu J, Hu Z, Chou I-M, Pan Z (2017) Depolymerization of poly(ethylene naphthalate) in fused silica capillary reactor and autoclave reactor from 240 to 280 °C in subcritical water. Polym Eng Sci 57(12):1382–1388 Bektesevic S, Kleman AM, Marteel-Parrish AE, Abraham MA (2006) Hydroformylation in supercritical carbon dioxide: Catalysis and benign solvents. J Supercrit Fluids 38(2):232–241 Belkheiri T, Andersson S-I, Mattsson C, Olausson L, Theliander H, Vamling L (2018) Hydrothermal liquefaction of Kraft Lignin in subcritical water: influence of phenol as capping agent. Energy Fuels 32(5):5923–5932 Benito-Román Ó, Teresa Sanz M, Melgosa R, de Paz E, Escudero I, Beltrán S (2019) Studies of polyphenol oxidase inactivation by means of high pressure carbon dioxide (HPCD). J Supercrit Fluids 147:310–321 Bermejo MD, Cocero MJ (2006) Supercritical water oxidation: a technical review. AIChE J 52(11):3933–3951

182

4 Industrial Scale Applications: Reaction-Based Processes

Bermejo MD, Kotlewska AJ, Florusse LJ, Cocero MJ, van Rantwijk F, Peters CJ (2008) Influence of the enzyme concentration on the phase behaviour for developing a homogeneous enzymatic reaction in ionic liquid–CO2 media. Green Chem 10(10):1049–1054 Bhanage BM, Fujita S, Ikushima Y, Arai M (2001) Synthesis of dimethyl carbonate and glycols from carbon dioxide, epoxides, and methanol using heterogeneous basic metal oxide catalysts with high activity and selectivity. Appl Catal Gen 219(1):259–266 Bogel-Łukasik E, Bogel-Łukasik R, da Ponte MN (2009) Pt- and Pd-catalysed limonene hydrogenation in high-density carbon dioxide. Monatshefte Für Chem Chem Mon 140(11):1361 Bowen Y, Zhemin S (2019) Supercritical water oxidation of pyridine and 3-cyanopyridine: TOC removal, kinetics, and degradation pathway. J Environ Eng 145(4):1–5 Brunner G (2009) Near and supercritical water. Part II: Oxidative processes. J Supercrit Fluids 47(3):382–390 Brunner G (2010) Applications of supercritical fluids. Annu Rev Chem Biomol Eng 1:321–342 Brunner G (2005) Supercritical fluids: technology and application to food processing. J Food Eng 67(1):21–33 Byrappa K, Adschiri T (2007) Hydrothermal technology for nanotechnology. Prog Cryst Growth Charact Mater 53(2):117–166 Canıaz RO, Erkey C (2014) Process intensification for heavy oil upgrading using supercritical water. Chem Eng Res Des 92(10):1845–1863 Cantero DA, Bermejo MD, Cocero MJ (2015) Governing chemistry of cellulose hydrolysis in supercritical water. Chemsuschem 8(6):1026–1033 Cantero DA, Martínez C, Bermejo MD, Cocero MJ (2014) Simultaneous and selective recovery of cellulose and hemicellulose fractions from wheat bran by supercritical water hydrolysis. Green Chem 17(1):610–618 Cao L, Wang X, Wang G, Wang J (2015) A pH-sensitive porous chitosan membrane prepared via surface grafting copolymerization in supercritical carbon dioxide. Polym Int 64(3):383–388 Carvalho NB, Silva MA de O, Fricks AT, Franceschi E, Dariva C, Zanin GM, Lima ÁS, Soares CMF (2014) Evaluation of activity of Bacillus lipase (free and immobilized) treated with compressed propane. J Mol Catal B Enzym 99:130–135 Chamberlain TW, Earley JH, Anderson DP, Khlobystov AN, Bourne RA (2014) Catalytic nanoreactors in continuous flow: hydrogenation inside single-walled carbon nanotubes using supercritical CO2 . Chem Commun 50(40):5200–5202 Chao M (2014) Supercritical water oxidation of wastewater-based drilling fluid with glycol addition. J Adv Oxid Technol 17(2):385–388 Chen M, Cao Y, Wang Y, Yang Z, Wang Q, Sun Q, Wang J (2019) Depolymerization of lignin over CoO/m-SEP catalyst under supercritical methanol. J Renew Sustain Energy 11(1):1–9 Chen P, Zhang Q, Shu R, Xu Y, Ma L, Wang T (2017) Catalytic depolymerization of the hydrolyzed lignin over mesoporous catalysts. Bioresour Technol 226:125–131 Chouchi D, Gourgouillon D, Courel M, Vital J, Nunes da Ponte M (2001) The influence of phase behavior on reactions at supercritical conditions: the hydrogenation of α-Pinene. Ind Eng Chem Res 40(12):2551–2554 Ciftci ON, Temelli F (2013) Enzymatic conversion of corn oil into biodiesel in a batch supercritical carbon dioxide reactor and kinetic modeling. J Supercrit Fluids 75:172–180 Ciftci ON, Temelli F (2011) Continuous production of fatty acid methyl esters from corn oil in a supercritical carbon dioxide bioreactor. J Supercrit Fluids 58(1):79–87 Clark P, Poliakoff M, Wells A (2007) Continuous Flow Hydrogenation of a Pharmaceutical Intermediate, [4-(3,4-Dichlorophenyl)-3,4-dihydro-2H-naphthalenyidene]methylamine, in supercritical carbon dioxide. Adv Synth Catal 349(17–18):2655–2659 Clifford AA (1994) Reactions in supercritical fluids. In: Kiran E, Sengers JMHL (eds) Supercritical fluids: fundamentals for application. Springer, Netherlands, Dordrecht, pp 449–479 Cokoja M, Wilhelm ME, Anthofer MH, Herrmann WA, Kühn FE (2015) Synthesis of cyclic carbonates from epoxides and carbon dioxide by using organocatalysts. Chemsuschem 8(15):2436– 2454

References

183

Cole-Hamilton DJ (2006) Asymmetric catalytic synthesis of organic compounds using metal complexes in supercritical fluids. Adv Synth Catal 348(12–13):1341–1351 Colombo TS, Mazutti MA, Di Luccio M, de Oliveira D, Oliveira JV (2015) Enzymatic synthesis of soybean biodiesel using supercritical carbon dioxide as solvent in a continuous expanded-bed reactor. J Supercrit Fluids 97:16–21 Combes GB, Dehghani F, Lucien FP, Dillow AK, Foster NR (2000) Chapter 14—Asymmetric catalytic hydrogenation in CO2 expanded methanol—an application of gas anti-solvent reactions (GASR). In: Abraham MA, Hesketh RP (eds) Reaction engineering for pollution prevention. Elsevier Science, Amsterdam, pp 173–181 Comim Rosso SR, Bianchin E, de Oliveira D, Oliveira JV, Ferreira SRS (2013) Enzymatic synthesis of poly(E-caprolactone) in supercritical carbon dioxide medium by means of a variable-volume view reactor. J Supercrit Fluids 79:133–141 Costa LI, Storti G (2018) Kinetic modeling of precipitation and dispersion polymerizations. In: Pauer W (ed) Polymer reaction engineering of dispersed systems, vol II. Springer International Publishing, Cham, pp 45–77 Dai W-L, Luo S-L, Yin S-F, Au C-T (2009) The direct transformation of carbon dioxide to organic carbonates over heterogeneous catalysts. Appl Catal Gen 366(1):2–12 Dai Z, Hatano B, Kadokawa J, Tagaya H (2002) Effect of diaminotoluene on the decomposition of polyurethane foam waste in superheated water. Polym Degrad Stab 76(2):179–184 Dalla Rosa C, Morandim MB, Ninow JL, Oliveira D, Treichel H, Oliveira JV (2009) Continuous lipase-catalyzed production of fatty acid ethyl esters from soybean oil in compressed fluids. Bioresour Technol 100(23):5818–5826 Darensbourg DJ, Yarbrough JC (2002) Mechanistic aspects of the copolymerization reaction of carbon dioxide and epoxides, using a Chiral Salen Chromium Chloride catalyst. J Am Chem Soc 124(22):6335–6342 Darr JA, Poliakoff M (1999) New directions in inorganic and metal-organic coordination chemistry in supercritical fluids. Chem Rev 99(2):495–542 Darr JA, Zhang J, Makwana NM, Weng X (2017) Continuous hydrothermal synthesis of inorganic nanoparticles: applications and future directions. Chem Rev 117(17):11125–11238 de Oliveira D, Feihrmann AC, Dariva C, Cunha AG, Bevilaqua JV, Destain J, Oliveira JV, Freire DMG (2006a) Influence of compressed fluids treatment on the activity of Yarrowia lipolytica lipase. J Mol Catal B Enzym 39(1):117–123 de Souza MM, Veneral JG, Furigo Junior A, de Oliveira JV, Di Luccio M, Prando LT, Terenzi H, de Oliveira D (2017) Effect of compressed fluids on the enzymatic activity and structure of lysozyme. J Supercrit Fluids 130:125–132 Dong LB, McVicker GB, Kiserow DJ, Roberts GW (2010) Hydrogenation of polystyrene in CO2 expanded liquids: the effect of catalyst composition on deactivation. Appl Catal Gen 384(1):45– 50 dos Santos P, Meireles MAA, Martínez J (2017) Production of isoamyl acetate by enzymatic reactions in batch and packed bed reactors with supercritical CO2 . J Supercrit Fluids 127:71–80. https://doi.org/10.1016/j.supflu.2017.03.019 dos Santos P, Zabot GL, Meireles MAA, Mazutti MA, Martínez J (2016) Synthesis of eugenyl acetate by enzymatic reactions in supercritical carbon dioxide. Biochem Eng J 114:1–9 Du L, Kelly JY, Roberts GW, DeSimone JM (2009) Fluoropolymer synthesis in supercritical carbon dioxide. J Supercrit Fluids 47(3):447–457 Durham E, Stewart C, Roe D, Xu R, Zhang S, Roberts CB (2014) Supercritical Fischer-Tropsch synthesis: heavy aldehyde production and the role of process conditions. Ind Eng Chem Res 53(23):9695–9702 Durham E, Xu R, Zhang S, Eden MR, Roberts CB (2013) Supercritical adiabatic reactor for FischerTropsch synthesis. Ind Eng Chem Res 52(9):3133–3136 Elbashir NO, Bukur DB, Durham E, Roberts CB (2010) Advancement of Fischer-Tropsch synthesis via utilization of supercritical fluid reaction media. AIChE J 56(4):997–1015

184

4 Industrial Scale Applications: Reaction-Based Processes

Eliseev OL, Savost’yanov AP, Sulima SI, Lapidus AL (2018) Recent development in heavy paraffin synthesis from CO and H2 . Mendeleev Commun 28(4):345–351 Elmanovich IVV, Sizov VEE, Zefirov VVV, Kalinina AAA, Gallyamov MOO, Papkov VSS, Muzafarov AMM (2022) Chemical recycling of high-molecular-weight organosilicon compounds in supercritical fluids. Polymers 14(23):5170 Fan L, Fujimoto K (1999) Fischer-Tropsch synthesis in supercritical fluid: characteristics and application. Appl Catal Gen 186(1):343–354 Fang L (2016) Catalytic wet oxidation of waste drilling fluid. Oxid Commun 39(3 A):2728–2732 Feiten MC, Morigi I, Di Luccio M, Oliveira JV (2023) Activity and stability of lipase from Candida Antarctica after treatment in pressurized fluids. Biotechnol Lett 45(2):287–298 Feng S-H, Li G-H (2017) Chapter 4—hydrothermal and solvothermal syntheses. In: Xu R, Xu Y (eds) Modern inorganic synthetic chemistry, 2nd edn. Elsevier, Amsterdam, pp 73–104 Findrik Z, Vasi´c-Raˇcki Ð, Primožiˇc M, Habulin M, Knez Ž (2005) Enzymatic activity of Lamino acid oxidase from snake venom Crotalus adamanteus in supercritical CO2 . Biocatal Biotransformation 23(5):315–321 Franken LPG, Marcon NS, Treichel H, Oliveira D, Freire DMG, Dariva C, Destain J, Oliveira JV (2010) Effect of treatment with compressed propane on lipases hydrolytic activity. Food Bioprocess Technol 3(4):511–520 Fricks AT, Oestreicher EG, Filho LC, Feihrmann AC, Cordeiro Y, Dariva C, Antunes OAC (2009) Effects of compressed fluids on the activity and structure of horseradish peroxidase. J Supercrit Fluids 50(2):162–168 Fujii T, Kawasaki S, Adschiri T (2016a) Kinetic study of octanoic acid enhanced crystal growth of boehmite under sub- and supercritical hydrothermal conditions. J Supercrit Fluids 118:148–152 Fujii T, Kawasaki S, Suzuki A, Adschiri T (2016b) High-speed morphology control of Boehmite nanoparticles by supercritical hydrothermal treatment with carboxylic acids. Cryst Growth Des 16(4):1996–2001 Gameiro M, Lisboa P, Paiva A, Barreiros S, Simões P (2015) Supercritical carbon dioxide-based integrated continuous extraction of oil from chicken feather meal, and its conversion to biodiesel in a packed-bed enzymatic reactor, at pilot scale. Fuel 153:135–142 Ghaffari-Moghaddam M, Eslahi H, Aydin YA, Saloglu D (2015) Enzymatic processes in alternative reaction media: a mini review. J Biol Methods 2(3):e25 Ghaziaskar HS, Daneshfar A, Calvo L (2006) Continuous esterification or dehydration in supercritical carbon dioxide. Green Chem 8(6):576–581 Gong Y, Guo Y, Sheehan JD, Chen Z, Wang S (2018) Oxidative degradation of landfill leachate by catalysis of CeMnOx/TiO2 in supercritical water: mechanism and kinetic study. Chem Eng J 331:578–586 Gong Y, Guo Y, Wang S, Song W (2016) Supercritical water oxidation of Quinazoline: effects of conversion parameters and reaction mechanism. Water Res 100:116–125 Gong Y, Guo Y, Wang S, Song W, Xu D (2017) Supercritical water oxidation of quinazoline: reaction kinetics and modeling. Water Res 110:56–65 Goodship V, Ogur EO (2004) Polymer processing with supercritical fluids. iSmithers Rapra Publishing Goto M, Obuchi R, Hirose T, Sakaki T, Shibata M (2004) Hydrothermal conversion of municipal organic waste into resources. Bioresour Technol 93(3):279–284 Guthalugu NK, Balaraman M, Kadimi US (2006) Optimization of enzymatic hydrolysis of triglycerides in soy deodorized distillate with supercritical carbon dioxide. Biochem Eng J 29(3):220–226 Gutiérrez-Arnillas E, Álvarez MS, Deive FJ, Rodríguez A, Sanromán MÁ (2016) New horizons in the enzymatic production of biodiesel using neoteric solvents. Renew Energy 98:92–100 Guzmán-Lagunes F, López-Luna A, Gimeno M, Bárzana E (2012) Enzymatic synthesis of poly-llactide in supercritical R134a. J Supercrit Fluids 72:186–190 Habulin M, Primožiˇc M, Knez Ž (2005) Enzymatic reactions in high-pressure membrane reactors. Ind Eng Chem Res 44(25):9619–9625

References

185

Haldorai Y, Shim J-J, Lim KT (2012) Synthesis of polymer–inorganic filler nanocomposites in supercritical CO2 . J Supercrit Fluids 71:45–63 Hao Q-Q, Zhao Y-H, Yang H-H, Liu Z-T, Liu Z-W (2012) Alumina grafted to SBA-15 in supercritical CO2 as a support of cobalt for Fischer-Tropsch synthesis. Energy Fuels 26(11):6567–6575 He L-N, Yasuda H, Sakakura T (2003) New procedure for recycling homogeneous catalyst: propylene carbonate synthesis under supercritical CO2 conditions. Green Chem 5(1):92–94 Hernández FJ, de los Ríos AP, Gómez D, Rubio M, Víllora G (2006) A new recirculating enzymatic membrane reactor for ester synthesis in ionic liquid/supercritical carbon dioxide biphasic systems. Appl Catal B Environ 67(1):121–126 Hidajat MJ, Riaz A, Park J, Insyani R, Verma D, Kim J (2017) Depolymerization of concentrated sulfuric acid hydrolysis lignin to high-yield aromatic monomers in basic sub- and supercritical fluids. Chem Eng J 317:9–19 Hintermair U, Franciò G, Leitner W (2013) A fully integrated continuous-flow system for asymmetric catalysis: enantioselective hydrogenation with supported ionic liquid phase catalysts using supercritical CO2 as the mobile phase. Chem Eur J 19(14):4538–4547 Hintermair U, Höfener T, Pullmann T, Franciò G, Leitner W (2010) Continuous enantioselective hydrogenation with a molecular catalyst in supported ionic liquid phase under supercritical CO2 flow. Chem Cat Chem 2(2):150–154 Hobbs HR, Thomas NR (2007) Biocatalysis in supercritical fluids, in fluorous solvents, and under solvent-free conditions. Chem Rev 107(6):2786–2820 Housaindokht MR, Monhemi H (2013) The open lid conformation of the lipase is explored in the compressed gas: New insights from molecular dynamic simulation. J Mol Catal B Enzym 87:135–138 Huang J, Qi W, Wu Y, Zhu Z (2005) Depolymerization of polybutylene terephthalate in supercritical methanol. Acta Polym Sin 2(2):309–312 Huang X, Korányi TI, Boot MD, Hensen EJM (2014) Catalytic depolymerization of lignin in supercritical ethanol. Chemsuschem 7(8):2276–2288 Ikariya T, Kayari Y, Kishimoto Y, Noguchi Y (2000) Highly efficient carbonylation reactions of organic halides in supercritical carbon dioxide. Prog Nucl Energy 37(1):429–434 Inoue S, Koinuma H, Tsuruta T (1969) Copolymerization of carbon dioxide and epoxide. J Polym Sci [b] 7(4):287–292 Jessop PG (2006) Homogeneous catalysis using supercritical fluids: recent trends and systems studied. J Supercrit Fluids 38(2):211–231 Jessop PG, Leitner W (1999) Chemical synthesis using supercritical fluids. WILEY-VCH Verlag GmbH Jia L, Jiang H, Li J (1999) Palladium(II)-catalyzed oxidation of acrylate esters to acetals in supercritical carbon dioxide. Chem Commun 11:985–986 Jiang H, Jia L, Li J (2000) Wacker reaction in supercritical carbon dioxide. Green Chem 2(4):161– 164 Jiang H-F, Shen Y-X, Wang Z-Y (2008) Palladium-catalyzed aerobic oxidation of terminal olefins with electron-withdrawing groups in scCO2 . Tetrahedron 64(3):508–514 Kamimura A, Ikeda K, Suzuki S, Kato K, Akinari Y, Sugimoto T, Kashiwagi K, Kaiso K, Matsumoto H, Yoshimoto M (2014) Efficient conversion of polyamides to ω-hydroxyalkanoic acids: a new method for chemical recycling of waste plastics. Chemsuschem 7(9):2473–2477 Kamimura A, Ikeda K, Suzuki S, Kato K, Matsumoto H, Kaiso K, Yoshimoto M (2017) A kinetic study on the conversion of nylon 12 to methyl 12-hydroxydodecanoate in supercritical MeOH in the presence of carboxylic acid. Polym Degrad Stab 146:95–104 Kang J, Myint AA, Sim S, Kim J, Kong WB, Lee Y-W (2018) Kinetics of the upgrading of heavy oil in supercritical methanol. J Supercrit Fluids 133:133–138 Kani I, Flores R, Fackler JP, Akgerman A (2004) Hydroformylation of styrene in supercritical carbon dioxide with fluoroacrylate polymer supported rhodium catalysts. J Supercrit Fluids 31(3):287–294

186

4 Industrial Scale Applications: Reaction-Based Processes

Kavˇciˇc S, Knez Ž, Leitgeb M (2014) Antimicrobial activity of n-butyl lactate obtained via enzymatic esterification of lactic acid with n-butanol in supercritical trifluoromethane. J Supercrit Fluids 85:143–150 Kazaryan PS, Tyutyunov AA, Kondratenko MS, Elmanovich IV, Stakhanov AI, Zefirov VV, Gallyamov MO, Blagodatskikh IV, Khokhlov AR (2019) Superhydrophobic coatings on textiles based on novel poly(perfluoro-tert-hexylbutyl methacrylate-co-hydroxyethyl methacrylate) copolymer deposited from solutions in supercritical carbon dioxide. J Supercrit Fluids 149:34–41 Kim H, Mitton DB, Latanision RM (2010) Corrosion behavior of Ni-base alloys in aqueous HCl solution of pH 2 at high temperature and pressure. Corros Sci 52(3):801–809 Klaus S, Lehenmeier MW, Anderson CE, Rieger B (2011) Recent advances in CO2 /epoxide copolymerization—New strategies and cooperative mechanisms. Coord Chem Rev 255(13):1460–1479 Knez Ž (2009) Enzymatic reactions in dense gases. J Supercrit Fluids 47(3):357–372 Knez Ž (2018) Enzymatic reactions in subcritical and supercritical fluids. J Supercrit Fluids 134:133–140 Knez Z, Laudani CG, Habulin M, Reverchon E (2007) Exploiting the pressure effect on lipasecatalyzed wax ester synthesis in dense carbon dioxide. Biotechnol Bioeng 97(6):1366–1375 Knez Ž, Leitgeb M, Primožiˇc M (2015a) Biochemical Reactions in Supercritical Fluids. Funct Food Ingred Nutraceuticals Process Technol 13:127 Knez Ž, Leitgeb M, Primožiˇc M (2015b) Enzymatic reactions in supercritical fluids. In: High pressure fluid technology for green food processing. Springer, pp 185–215 Knez Ž, Markoˇciˇc E, Leitgeb M, Primožiˇc M, Knez Hrnˇciˇc M, Škerget M (2014) Industrial applications of supercritical fluids: a review. Energy 77:235–243 Koeken ACJ, van den Broeke LJP, Deelman B-J, Keurentjes JTF (2011) Full kinetic description of 1-octene hydroformylation in a supercritical medium. J Mol Catal Chem 346(1):1–11 Kuhn G de O, Coghetto C, Treichel H, de Oliveira D, Oliveira JV (2011) Effect of compressed fluids treatment on the activity of inulinase from Kluyveromyces marxianus NRRL Y-7571 immobilized in montmorillonite. Process Biochem 46(12):2286–2290 Kunene TE, Webb PB, Cole-Hamilton DJ (2011) Highly selective hydroformylation of long-chain alkenes in a supercritical fluid ionic liquid biphasic system. Green Chem 13(6):1476–1481 Lang X-D, He X-FL (2015) Sustainable solid catalysts for cyclic carbonate synthesis from CO2 and epoxide. Curr Org Chem 19(8):681–694 Laudani CG, Habulin M, Knez Ž, Porta GD, Reverchon E (2007) Lipase-catalyzed long chain fatty ester synthesis in dense carbon dioxide: kinetics and thermodynamics. J Supercrit Fluids 41(1):92–101 Lee JH, Kwon CH, Kang JW, Park C, Tae B, Kim SW (2009) Biodiesel production from various oils under supercritical fluid conditions by Candida antartica lipase B using a stepwise reaction method. Appl Biochem Biotechnol 156(1–3):24–34 Leitgeb M, Primožiˇc M, Knez Ž (2007) Supercritical fluids as solvents for enzymatic reactions. Acta Chim Slov 54(4):667–677 Li N, Yan B, Xiao X-M (2015) A review of laboratory-scale research on upgrading heavy oil in supercritical water. Energies 8(8):8962–8989 Limarta SO, Ha J-M, Park Y-K, Lee H, Suh DJ, Jae J (2018) Efficient depolymerization of lignin in supercritical ethanol by a combination of metal and base catalysts. J Ind Eng Chem 57:45–54 Lin T-J, Chen S-W, Chang A-C (2006) Enrichment of n-3 PUFA contents on triglycerides of fish oil by lipase-catalyzed trans-esterification under supercritical conditions. Biochem Eng J 29(1):27–34 Liu H, Meng Y, Zhang G, Li G (2016) Supercritical water oxidation of drilling fluid wastewater. Oxid Commun 39(2):1687–1693 López-Luna A, Gallegos JL, Gimeno M, Vivaldo-Lima E, Bárzana E (2010) Lipase-catalyzed syntheses of linear and hyperbranched polyesters using compressed fluids as solvent media. J Mol Catal B Enzym 67(1):143–149

References

187

Lozano P, Diego TD, Vaultier M, Iborra JL (2009) Dynamic kinetic resolution of Sec-Alcohols in ionic liquids/supercritical carbon dioxide biphasic systems. Int J Chem React Eng 7(1) Lozano P, Garcia-Verdugo E, V Luis S, Pucheault M, Vaultier M (2011) (Bio) catalytic continuous flow processes in scCO2 and/or ILs: towards sustainable (Bio) catalytic synthetic platforms. Curr Org Synth 8(6):810–823 Lozano P, Nieto S, L Serrano J, Perez J, Sanchez-Gomez G, Garcia-Verdugo E, V Luis S (2017) Flow biocatalytic processes in ionic liquids and supercritical fluids. Mini-Rev Org Chem 14(1):65–74 Lutz J-F, Lehn J-M, Meijer E, Matyjaszewski K (2016) From precision polymers to complex materials and systems. Nat Rev Mater 1(5):16024 Ma X, Ma R, Hao W, Chen M, Yan F, Cui K, Tian Y, Li Y (2015) Common pathways in ethanolysis of Kraft Lignin to platform chemicals over molybdenum-based catalysts. ACS Catal 5(8):4803– 4813 Madras G, Kolluru C, Kumar R (2004) Synthesis of biodiesel in supercritical fluids. Fuel 83(14):2029–2033 Mahmood N, Yuan Z, Schmidt J, (Charles) Xu C (2013) Production of polyols via direct hydrolysis of kraft lignin: Effect of process parameters. Bioresour Technol 139:13–20 Malek Abbaslou RM, Soltan Mohammadzadeh JS, Dalai AK (2009) Review on Fischer-Tropsch synthesis in supercritical media. Fuel Process Technol 90(7):849–856 Manera AP, Kuhn G, Polloni A, Marangoni M, Zabot G, Kalil SJ, de Oliveira D, Treichel H, Vladimir Oliveira J, Mazutti MA, Maugeri F (2011) Effect of compressed fluids treatment on the activity, stability and enzymatic reaction performance of β-galactosidase. Food Chem 125(4):1235–1240 Manera AP, Zabot GL, Vladimir Oliveira J, de Oliveira D, Mazutti MA, Kalil SJ, Treichel H, Filho FM (2012) Enzymatic synthesis of galactooligosaccharides using pressurised fluids as reaction medium. Food Chem 133(4):1408–1413 Marrone PA (2013) Supercritical water oxidation—Current status of full-scale commercial activity for waste destruction. J Supercrit Fluids 79:283–288 Marszałek K, Doesburg P, Starzonek S, Szczepa´nska J, Wo´zniak Ł, Lorenzo JM, Sk˛apska S, Rzoska S, Barba FJ (2019) Comparative effect of supercritical carbon dioxide and high pressure processing on structural changes and activity loss of oxidoreductive enzymes. J CO2 Util 29:46–56 Martínez CM, Cantero DA, Bermejo MD, Cocero MJ (2015) Hydrolysis of cellulose in supercritical water: reagent concentration as a selectivity factor. Cellulose 22(4):2231–2243 Marx S (2016) Glycerol-free biodiesel production through transesterification: a review. Fuel Process Technol 151:139–147 Matsuda T (2013) Recent progress in biocatalysis using supercritical carbon dioxide. J Biosci Bioeng 115(3):233–241 Matsuda T, Watanabe K, Harada T, Nakamura K (2004) Enzymatic reactions in supercritical CO2 : carboxylation, asymmetric reduction and esterification. Catal Today 96(3):103–111 McAllister TD, Farrand LD, Howdle SM (2016) Improved particle size control for the dispersion polymerization of methyl methacrylate in supercritical carbon dioxide. Macromol Chem Phys 217(20):2294–2301 Melgosa R, Sanz MT, Solaesa ÁG, de Paz E, Beltrán S, Lamas DL (2017) Supercritical carbon dioxide as solvent in the lipase-catalyzed ethanolysis of fish oil: Kinetic study. J CO2 Util 17:170–179 Mena M, Shirai K, Tecante A, Bárzana E, Gimeno M (2015) Enzymatic syntheses of linear and hyperbranched poly-l-lactide using compressed R134a–ionic liquid media. J Supercrit Fluids 103:77–82 Monfared A, Mohammadi R, Hosseinian A, Sarhandi S, Nezhad PDK (2019) Cycloaddition of atmospheric CO2 to epoxides under solvent-free conditions: a straightforward route to carbonates by green chemistry metrics. RSC Adv 9(7):3884–3899

188

4 Industrial Scale Applications: Reaction-Based Processes

Morales Ibarra R, Sasaki M, Goto M, Quitain AT, García Montes SM, Aguilar-Garib JA (2015) Carbon fiber recovery using water and benzyl alcohol in subcritical and supercritical conditions for chemical recycling of thermoset composite materials. J Mater Cycles Waste Manag 17(2):369–379 More SR, Yadav GD (2018) Effect of supercritical CO2 as reaction medium for selective hydrogenation of acetophenone to 1-phenylethanol. ACS Omega 3(6):7124–7132 Muratov G, Seo KW, Kim C (2005) Application of supercritical carbon dioxide to the bioconversion of cotton fibers. J Ind Eng Chem 11(1):42–46 Musie G, Wei M, Subramaniam B, Busch DH (2001) Catalytic oxidations in carbon dioxide-based reaction media, including novel CO2 -expanded phases. Coord Chem Rev 219–221:789–820 Nakamura K, Yamanaka R, Matsuda T, Harada T (2003) Recent developments in asymmetric reduction of ketones with biocatalysts. Tetrahedron Asymmetry 14(18):2659–2681 Nyari NLD, Zabot GL, Zamadei R, Paluzzi AR, Tres MV, Zeni J, Venquiaruto LD, Dallago RM (2018) Activation of Candida antarctica lipase B in pressurized fluids for the synthesis of esters. J Chem Technol Biotechnol 93(3):897–908 Okajima I, Watanabe K, Haramiishi S, Nakamura M, Shimamura Y, Sako T (2017) Recycling of carbon fiber reinforced plastic containing amine-cured epoxy resin using supercritical and subcritical fluids. J Supercrit Fluids 119:44–51. https://doi.org/10.1016/j.supflu.2016.08.015 Oliveira D, Feihrmann AC, Rubira AF, Kunita MH, Dariva C, Oliveira JV (2006b) Assessment of two immobilized lipases activity treated in compressed fluids. J Supercrit Fluids 38(3):373–382. https://doi.org/10.1016/j.supflu.2005.12.007 Oliveira PF, Machado RAF, Barth D, Acosta ED (2016) Dispersion polymerization of methyl methacrylate in supercritical carbon dioxide using vinyl terminated poly(dimethylsiloxane). Chem Eng Process Process Intensif 103:46–52 Onwudili JA, Williams PT (2006) Flameless supercritical water incineration of polycyclic aromatic hydrocarbons. Int J Energy Res 30(7):523–533 Onwudili JA, Williams PT (2007) Reaction mechanisms for the decomposition of phenanthrene and naphthalene under hydrothermal conditions. J Supercrit Fluids 39(3):399–408 Osanai Y, Toshima K, Matsumura S (2006) Enzymatic transformation of aliphatic polyesters into cyclic oligomers using enzyme packed column under continuous flow of supercritical carbon dioxide with toluene. Sci Technol Adv Mater 7(2):202–208 Palocci C, Falconi M, Chronopoulou L, Cernia E (2008) Lipase-catalyzed regioselective acylation of tritylglycosides in supercritical carbon dioxide. J Supercrit Fluids 45(1):88–93 Parilti R, Alaimo D, Grignard B, Boury F, Howdle SM, Jérôme C (2017) Mild synthesis of poly(HEMA)-networks as well-defined nanoparticles in supercritical carbon dioxide. J Mater Chem B 5(29):5806–5815 Parilti R, Riva R, Howdle SM, Dupont-Gillain C, Jerome C (2018) Sulindac encapsulation and release from functional poly(HEMA) microparticles prepared in supercritical carbon dioxide. Int J Pharm 549(1):161–168 Pavloviˇc I, Knez Ž, Škerget M (2013) Hydrothermal reactions of agricultural and food processing wastes in sub- and supercritical water: a review of fundamentals, mechanisms, and state of research. J Agric Food Chem 61(34):8003–8025 Peng Y-K, Sun L-L, Shi W, Long J-J (2016) Investigation of enzymatic activity, stability and structure changes of pectinase treated in supercritical carbon dioxide. J Clean Prod 125:331–340 Philippot G, Elissalde C, Maglione M, Aymonier C (2014) Supercritical fluid technology: a reliable process for high quality BaTiO3 based nanomaterials. Adv Powder Technol 25(5):1415–1429 Polloni AE, Veneral JG, Rebelatto EA, de Oliveira D, Oliveira JV, Araújo PHH, Sayer C (2017) Enzymatic ring opening polymerization of ω-pentadecalactone using supercritical carbon dioxide. J Supercrit Fluids 119:221–228 Primozic M, Habulin M, Knez Z (2006) Proteinase-catalyzed hydrolysis of casein at atmospheric pressure and in supercritical media. Chem Biochem Eng Q 20(3):255–261

References

189

Qu H, Gong J-H, Tan X-C, Yuan P-Q, Cheng Z-M, Yuan W-K (2019) Dissolution of polycyclic aromatic hydrocarbons in subcritical and supercritical Water: a molecular dynamics simulation study. Chem Eng Sci 195:958–967 Rajappagowda R, Numan-Al-Mobin AM, Yao B, Cook RD, Smirnova A (2017) Toward selective lignin liquefaction: synergistic effect of hetero- and homogeneous catalysis in sub- and supercritical fluids. Energy Fuels 31(1):578–586 Randolph TW, Blanch HW, Prausnitz JM (1988a) Enzyme-caytalyzed oxidation of cholesterol in supercritical carbon dioxide. AIChE J 34(8):1354–1360 Randolph TW, Clark DS, Blanch HW, Prausnitz JM (1988b) Enzymatic oxidation of cholesterol aggregates in supercritical carbon dioxide. Science 239(4838):387–390 Rathke JW, Klingler RJ, Krause TR (1991) Propylene hydroformylation in supercritical carbon dioxide. Organometallics 10(5):1350–1355 Ren M, Wang S, Yang C, Xu H, Guo Y, Roekaerts D (2019) Supercritical water oxidation of quinoline with moderate preheat temperature and initial concentration. Fuel 236:1408–1414 Rezaei K, Temelli F, Jenab E (2007) Effects of pressure and temperature on enzymatic reactions in supercritical fluids. Biotechnol Adv 25(3):272–280 Romero MD, Calvo L, Alba C, Habulin M, Primožiˇc M, Knez Ž (2005) Enzymatic synthesis of isoamyl acetate with immobilized Candida antarctica lipase in supercritical carbon dioxide. J Supercrit Fluids 33(1):77–84 Rosero-Henao JC, Bueno BE, de Souza R, Ribeiro R, Lopes de Oliveira A, Gomide CA, Gomes TM, Tommaso G (2019) Potential benefits of near critical and supercritical pre-treatment of lignocellulosic biomass towards anaerobic digestion. Waste Manag Res J Int Solid Wastes Public Clean Assoc ISWA 37(1):74–82 Rosso Comim SR, Veneral JG, de Oliveira D, Ferreira SRS, Oliveira JV (2015) Enzymatic synthesis of poly(E-caprolactone) in liquified petroleum gas and carbon dioxide. J Supercrit Fluids 96:334– 348 Šabeder S, Habulin M, Knez Ž (2005) Comparison of the esterification of fructose and palmitic acid in organic solvent and in supercritical carbon dioxide. Ind Eng Chem Res 44(25):9631–9635 Salgın U, Salgın S, Takaç S (2007) The enantioselective hydrolysis of racemic naproxen methyl ester in supercritical CO2 using Candida rugosa lipase. J Supercrit Fluids 43(2):310–316 Sánchez-Oneto J, Portela JR, Nebot E, Martínez de la Ossa E (2007) Hydrothermal oxidation: application to the treatment of different cutting fluid wastes. J Hazard Mater 144(3):639–644 Sari A (2014) Investigation of the supercritical conditions for Fischer-Tropsch reaction over an industrial Co–Ru/γ-Al2O3 catalyst. Chem Eng J 244:317–326 Savage PE (2009) A perspective on catalysis in sub- and supercritical water. J Supercrit Fluids 47(3):407–414 Scandelai APJ, Cardozo Filho L, Martins DCC, Freitas TKF de S, Garcia JC, Tavares CRG (2018) Combined processes of ozonation and supercritical water oxidation for landfill leachate degradation. Waste Manag 77:466–476 Seki T, Grunwaldt J-D, Baiker A (2008) Heterogeneous catalytic hydrogenation in supercritical fluids: potential and limitations. Ind Eng Chem Res 47(14):4561–4585 Senyay-Oncel D, Yesil-Celiktas O (2015) Characterization, immobilization, and activity enhancement of cellulase treated with supercritical CO2 . Cellulose 22(6):3619–3631 Senyay-Oncel D, Yesil-Celiktas O (2013) Treatment of immobilized α-amylase under supercritical CO2 conditions: Can activity be enhanced after consecutive enzymatic reactions? J Mol Catal B Enzym 91:72–76 Senyay-Oncel D, Yesil-Celiktas O (2011) Activity and stability enhancement of α-amylase treated with sub- and supercritical carbon dioxide. J Biosci Bioeng 112(5):435–440 sirin ÿzlem Z, Demirkol O, Akbaslar D, Giray ES (2013) Clean and efficient synthesis of flavanone in sub-critical water. J Supercrit Fluids 81:217–220 Song Q-W, He L-N, Wang J-Q, Yasuda H, Sakakura T (2012) Catalytic fixation of CO2 to cyclic carbonates by phosphonium chlorides immobilized on fluorous polymer. Green Chem 15(1):110–115

190

4 Industrial Scale Applications: Reaction-Based Processes

Sovová H, Zarevúcka M, Bernášek P, Stameni´c M (2008) Kinetics and specificity of Lipozymecatalysed oil hydrolysis in supercritical CO2 . Chem Eng Res Des 86(7):673–681 Stephenson P, Kondor B, Licence P, Scovell K, Ross SK, Poliakoff M (2006) Continuous asymmetric hydrogenation in supercritical carbon dioxide using an immobilised homogeneous catalyst. Adv Synth Catal 348(12–13):1605–1610 Subramaniam B, Chaudhari RV, Chaudhari AS, Akien GR, Xie Z (2014) Supercritical fluids and gas-expanded liquids as tunable media for multiphase catalytic reactions. Chem Eng Sci 115:3– 18 Sun J, Cheng W, Fan W, Wang Y, Meng Z, Zhang S (2009) Reusable and efficient polymersupported task-specific ionic liquid catalyst for cycloaddition of epoxide with CO2 . Catal Today 148(3):361–367 Suzuki Y, Tagaya H, Asou T, Kadokawa J, Chiba K (1999) Decomposition of prepolymers and molding materials of phenol resin in subcritical and supercritical water under an Ar atmosphere. Ind Eng Chem Res 38(4):1391–1395 Tagaya H, Katoh K, Kadokawa J, Chiba K (1999) Decomposition of polycarbonate in subcritical and supercritical water. Polym Degrad Stab 64(2):289–292 Taguchi M, Yamamoto N, Hojo D, Takami S, Adschiri T, Funazukuri T, Naka T (2014) Synthesis of monocarboxylic acid-modified CeO2 nanoparticles using supercritical water. RSC Adv 4(91):49605–49613 Taher H, Al-Zuhair S (2017) The use of alternative solvents in enzymatic biodiesel production: a review. Biofuels Bioprod Biorefining 11(1):168–194 Taher H, Al-Zuhair S, AlMarzouqui A, Hashim I (2011) Extracted fat from lamb meat by supercritical CO2 as feedstock for biodiesel production. Biochem Eng J 55(1):23–31 Tavares MVL, Kanda LRS, Giacomin Junior WR, Ramos LP, Vandenberghe LPS, Corazza ML (2022) Supercritical carbon dioxide effect on lipase-catalyzed geranyl acetate synthesis. J Braz Chem Soc 33(7):715–724 Theuerkauf J, Franciò G, Leitner W (2013) Continuous-flow asymmetric hydrogenation of the βKeto ester methyl propionylacetate in ionic liquid-supercritical carbon dioxide biphasic systems. Adv Synth Catal 355(1):209–219 Top S, Akgun M, Kipcak E, Bilgili MS (2020) Treatment of hospital wastewater by supercritical water oxidation process. Water Res 185:116279 Torres Galvis HM, de Jong KP (2013) Catalysts for production of lower olefins from synthesis gas: a review. ACS Catal 3(9):2130–2149 Tortosa Estorach C, Giménez-Pedrós M, Masdeu-Bultó AM, Sayede AD (2008) Monflier E (2008) Hydroformylation of 1-octene in supercritical carbon dioxide with Alkyl P-Donor ligands on rhodium using a peracetylated β-cyclodextrin as a solubiliser. Eur J Inorg Chem 17:2659–2663 Varma MN, Madras G (2007a) Synthesis of isoamyl laurate and isoamyl stearate in supercritical carbon dioxide. Appl Biochem Biotechnol 141(1):139–147 Varma MN, Madras G (2010) Kinetics of enzymatic synthesis of geranyl butyrate by transesterification in various supercritical fluids. Biochem Eng J 49(2):250–255 Varma MN, Madras G (2007b) Synthesis of biodiesel from castor oil and linseed oil in supercritical fluids. Ind Eng Chem Res 46(1):1–6 Wahyudiono KT, Sasaki M, Goto M (2007) Decomposition of a lignin model compound under hydrothermal conditions. Chem Eng Technol 30(8):1113–1122 Wahyudiono SM, Goto M (2009) Conversion of biomass model compound under hydrothermal conditions using batch reactor. Fuel 88(9):1656–1664 Wang W, Irvine DJ, Howdle SM (2005) Dispersion catalytic chain transfer polymerizations of methyl methacrylate in supercritical carbon dioxide. Ind Eng Chem Res 44(23):8654–8658 Wang X, Zhou JH, Li HM, Sun GW (2013) Depolymerization of lignin with supercritical fluids: a review. Adv Mater Res 821–822:1126–1134 Wang Z-Y, Jiang H-F, Ouyang X-Y, Qi C-R, Yang S-R (2006) Pd(II)-catalyzed acetalization of terminal olefins with electron-withdrawing groups in supercritical carbon dioxide: selective control and mechanism. Tetrahedron 62(42):9846–9854

References

191

Webb PB, Kunene TE, Cole-Hamilton DJ (2005) Continuous flow homogeneous hydroformylation of alkenes using supercritical fluids. Green Chem 7(5):373–379 Weber A, Catchpole O, Eltringham W (2008) Supercritical fluid assisted, integrated process for the synthesis and separation of different lipid derivatives. J Sep Sci 31(8):1346–1351 Wei M, Musie GT, Busch DH, Subramaniam B (2002) CO2 -expanded solvents: unique and versatile media for performing homogeneous catalytic oxidations. J Am Chem Soc 124(11):2513–2517 Wu Y, Chen Y, Wu K (2014) Role of co-solvents in biomass conversion reactions using sub/ supercritical water. In: Fang Z, Xu C (Charles) (eds) Near-critical and supercritical water and their applications for biorefineries. Springer Netherlands, Dordrecht, pp 69–98 Xu D, Wang S, Zhang J, Tang X, Guo Y, Huang C (2015) Supercritical water oxidation of a pesticide wastewater. Chem Eng Res Des 94:396–406 Yanagihara N, Ohgane K (2013) Studies on the oxidative degradation of nylons by nitrogen dioxide in supercritical carbon dioxide. Polym Degrad Stab 98(12):2735–2741 Yang B, Cheng Z, Shen Z (2019a) Decomposition of 14 organophosphate flame retardants during supercritical water oxidation. J Taiwan Inst Chem Eng 95:40–47 Yang B, Cheng Z, Yuan T, Gao X, Tan Y, Ma Y, Shen Z (2018) Temperature sensitivity of nitrogencontaining compounds decomposition during supercritical water oxidation (SCWO). J Taiwan Inst Chem Eng 93:31–41 Yang J, Wang S, Li Y, Zhang Y, Xu D (2019b) Novel design concept for a commercial-scale plant for supercritical water oxidation of industrial and sewage sludge. J Environ Manage 233:131–140 Yu G, Xue Y, Xu W, Zhang J, Xue CH (2007) Stability and activity of lipase in subcritical 1,1,1,2tetrafluoroethane (R134a). J Ind Microbiol Biotechnol 34(12):793–798 Zetzl C, Gairola K, Kirsch C, Perez-Cantu L, Smirnova I (2011) High pressure processes in biorefineries. Chem Ing Tech 83(7):1016–1025 Zhang F, Xie Y, Liu P, Hao F, Yao Z, Luo H (2014) Cycloaddition reaction of propylene oxide and carbon dioxide over NaX Zeolite supported metalloporphyrin catalysts. Catal Lett 144(11):1894–1899 Zhao H (2018) Enzymatic ring-opening polymerization (ROP) of polylactones: roles of non-aqueous solvents. J Chem Technol Biotechnol Oxf Oxfs 93(1):9–19 Zhao L-C, Hou Z-Q, Liu C-Z, Wang Y-Y, Dai L-Y (2014) A catalyst-free novel synthesis of diethyl carbonate from ethyl carbamate in supercritical ethanol. Chin Chem Lett 25(10):1395–1398 Zheng C, Zhao L, Zhou X, Fu Z, Li A (2013) Treatment technologies for organic wastewater. Water Treat 11:250–286 Zurbel A, Kaiser D, Hippmann S, Bertau M (2019) Thermochemische depolymerisation von lignin zur Gewinnung von aromaten. Chem Ing Tech 91(4):484–493

Chapter 5

Design of High Pressure Plants for Research, Pilot and Production Scale

Abstract Turning a process idea into an operational plant always has been an exciting task. A good balance between experiments in different scales and theoretical considerations has to be found. Thus, collaboration of different technical disciplines is essential to go from lab-scale to commercial scale solving questions that arise along the way. Besides working on technical details it is essential to manage risks, overcome drawbacks and fulfil requirements of budget and schedule. The chapter summarizes considerations for the different stages of process development complemented by practical examples from successfully realized high pressure applications. Keywords Equipment · Bench-scale · Pilot-scale · Commercial scale · Plant design

5.1 Basic Considerations for Effective Process Synthesis Many considerations for the applications of SC fluid processes are triggered by the shortcomings of conventional process technologies: be it too much by-product, too complicated down-stream processing, poor product quality or residues of solvents in the product, too long processing time or several other reasons. In many cases, unfortunately, high pressure processes are considered a kind of last resort for a process challenge, although they can often solve several issues with conventional processing technologies. Developments therefore often start with some reservation about the principal viability of high pressure processing since mankind tries to avoid unusual things or seemingly risky steps in general. It is a big part of human nature as this attitude secured the survival over many millenniums and brought us where we stand today. Nevertheless, in order to build a better world for the future, we also need to try the seemingly unusual and be brave enough walking more often unconventional pathways—like applying high pressures. A perfect example from industrial history is the development of LDPE which comprised all the aspects mentioned before. A closely collaborating team of engineers and chemists developed the process and the relevant equipment, worked through © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Ž. Knez and C. Lütge, Product, Process and Plant Design Using Subcritical and Supercritical Fluids for Industrial Application, https://doi.org/10.1007/978-3-031-34636-1_5

193

194

5 Design of High Pressure Plants for Research, Pilot and Production Scale

scale-up and ended in commercial production in relatively short time. Just by chance the first few “unusual” molecules of a “waxy solid” (i.e. LDPE) were discovered 1932/1933 in high pressure equipment after trying a new chemical synthesis aiming at a different molecule. Upon repetitions of the experiment many decompositions (i.e. ethylene suddenly reacting into hydrogen and carbon under extreme pressure) happened giving various reasons for stopping the endeavour and giving-up. However, the engineers and chemists were courageous enough to continue their work by analyzing the problem and finding ways to master the development—they had a clear purpose in mind to accomplish the needful. In 1938 a pilot plant was established and they achieved a conversion rate of 33% in a 9 L pressure vessel (which is still a good KPI for the output of the reaction today). As early as 1939 another plant with a 250 L vessel was designed aiming at a 500 t/a production as soon as possible. Production commenced in March 1942 due to many drawbacks also from the ongoing war. A more detailed description of the developments may be found in Ellis (2005) and may encourage young generations of scientists to look ahead, try the unusual and overcome difficulties. Which lessons can be learnt from the aforementioned historical development of LDPE for novel high pressure process developments? The essence was a trustful and close collaboration within a multi-disciplined team, curiosity for all the new discoveries, the smart and creative combination of theoretical considerations and practical work as well as steadfastness and will to learn quick and overcome difficulties. Finally, they had a clear purpose driving the whole team forward: to invent a novel material being of high value at that time and still today. An overview of said concept in a more engineering-like way can be found in Fig. 5.1. Upfront any experimental work we usually go through literature and intellectual property and check for published prior work that was already elaborated elsewhere. From this theoretical literature and patent research comprehensive ideas can be collected and, often very surprising, also interesting gaps may be found to define unknown areas and room for own future research. A collection of relevant property data as well as phase equilibria is also to be carried out at this stage. Thereafter an experimental program should be worked through in laboratory scale (i.e. vessel volume in ml-scale) including proper analyses and optimization in the best case by a factorial design of experiments and milestones. This enables the researcher to verify the essential parameters of the process and give enough information for a second step of theoretical consideration: the applicability of the new process in competition to other options or alternative process routes. A first approach through a simple empirical model might form the basis for a simulation of the process and give an initial clue of the process concept, the mass and energy balance as well as ideas for the product quality and probable cost. The result should be a sound decision to abort or continue the effort based on the clear understanding of the reasons and the room for improvements. As soon as the bench-scale work proofs promising, further experimental work is needed in larger scale (vessel volume in litre-scale). Initial work was typically done in very small high pressure apparatus keeping the equipment and the sample material small and the cost low. Figure 5.2 summarizes the working steps in the next larger

5.1 Basic Considerations for Effective Process Synthesis

195

Fig. 5.1 Stepwise approach during exploratory work (bench-scale) and early study of general feasibility

scale which may scale-up by a factor of 1:200 or even greater. It is very essential for these experiments to realize the work ideally in the same plant set-up that is likely to be used in commercial scale or—at least—to keep the adaptability of the pilot system for necessary adjustments. It must be verified at this stage e.g. which influence recycling of certain media streams will have on the yield, the formation of by-products or the efficiency of separation steps, whether or not an accumulation of unwanted components occurs and hinders the process, etc. Further, a clearer picture for the mass and energy balance, yields, processing time and consumption figures must be elaborated. Also the general handling of the plant, operation of the process and handling of the materials must be rehearsed. Only a practical doing of all the steps repeatedly will give the final judgement of process feasibility in commercial scale. In case market introduction should already be tested, larger amounts of product can either be collected from various batches of “production” or obtained from a rental phase of an existing commercial plant. Suppliers of high pressure equipment and plants typically entertain a network with their customers and support the plant design as well as sound cost estimates. Before elaborating on the details of equipment and plants in the different scales, a more generic differentiation of process principles (physical or chemical process), the phases involved for feedstock (F), solvent (S) and product (P) and applied scale might be discussed using practical examples. Table 5.1 summarizes an overview of existing high pressure applications in commercial scale and also tries to give some guidance for transfer opportunities of practical know-how existing elsewhere and expertise that might be converted into new areas of current and future developments. The most widely used principles of high pressure processes in industrial operation are (so far) based on 2 general patterns: firstly, the treatment of solid matter in batch processes which can be considered a physical process of separating or adding particular ingredients by means of a SC fluid. Secondly, the other end of Table 5.1, the production of LDPE in a continuous process using the feedstock ethylene also as a SC solvent to form a polymer both in an autoclave system or a tubular

196

5 Design of High Pressure Plants for Research, Pilot and Production Scale

Fig. 5.2 Stepwise approach during later stage feasibility study and pilot-scale (scale-up)

reactor system. This general differentiation also defines special requirements for the equipment involved in the particular scale which will be discussed in the following sub-chapters.

5.2 Equipment for Bench-Scale Tests Following the explanations along Fig. 5.1 and being able to focus on the process development itself puts a clear emphasis on the tools to be used: reliable bench-scale equipment, reliable analytical tools together with experienced analytical skills, while access to property data bases, literature and IP should be considered a given in the twenty-first century. The equipment can be self-made (which is often the case in university research) or purchased from professional suppliers. However, in both cases a number of prerequisites need to be strictly obeyed which are definitely not a subject for compromise: safety of personnel and equipment, robustness, reliability as well as flexibility/ variability. The focus must be put on the experimental program itself and on obtaining reliable results within short time. On the other hand, constant maintenance and repair exercises on the high pressure equipment and lack of repeatability of experiments must be avoided. Too many process options have been considered poor or useless due to unreliable test equipment, shaky installation and improvised experimental setup. Those unfortunate findings can be circumvented by robust, reliable and versatile hardware, safe and simple to operate. In bench-scale experiments the SC fluid is usually withdrawn after the process and neither collected or recycled. Table 5.2 summarizes the essential equipment and

Chemical

Physical

Process principle

SC

SC

SC

Liquid

SC

SC

Liquid

Solid

SC

Solvent

Solid

Feedstock

Phases involved

Liquid

Solid

Solid

Liquid

Solid

Product

LDPE

Enzymatic reactions in SCF

De-oiling of raw lecithin Particle formation

Washing of caffeine from SC CO2 , fish-oil, SC chromatography,

Most SCF extraction applications, dyeing (e.g. textile), impregnation (e.g. wood, tobacco), cleaning (e.g. parts, cork), drying of aerogels, etc.

Practical examples

Table 5.1 General overview of practical high pressure process principles in various scales

Mostly autoclave, batch process

Autoclave, batch process

Autoclave, batch process

Column, continuous

Autoclave, batch process

Bench-scale

Commercial scale

Mostly autoclave, batch process

Autoclave, batch process

Autoclave, batch process

Column, continuous

Autoclave and tubular reactor system, continuous processes

/

Autoclaves, semi-continuous Autoclave, batch process

Column, continuous, SMB

Autoclave (s), Autoclaves, batch batch process process

Pilot-scale

5.2 Equipment for Bench-Scale Tests 197

198

5 Design of High Pressure Plants for Research, Pilot and Production Scale

gives a few remarks for careful consideration. It must be understood that working under high pressure conditions requires care and diligence by the researchers since the equipment carries a high amount of potential energy and thus may be dangerous to operate. Official regulations, codes and standards (e.g. PED, ASME, etc.) do not cover non-commercial experimental equipment, however, it is clearly advised to comply with the proven principles of safety measures and use equipment from knowledgeable suppliers carrying relevant approvals. A proper documentation of materials or equipment used, the source of supply, etc. is recommended. Figure 5.3 shows a typical process and instrumentation diagram (P&ID) for a bench-scale set-up for process development giving detailed information of the sequence of equipment and installation. The core is the autoclave A situated in a heated (water) bath volume for constant temperature conditions. The SCF is supplied from a steel bottle S with riser pipe (CO2 ), sub-cooled and then brought to operating pressure by a membrane pump M. Supply of entrainer could be realized through a supply pump SP. To the right we find devices for cleaning and washing the SCF from the extract, flow control valve and gasmeter G to register the amount of fluid used during the process. The whole set-up is installed under a fume hood making sure released streams are controlled and cleaned prior to their exit into the environment. A simple set-up for chemical reactions under high pressure in this scale is very much identical. Table 5.2 List of essential equipment for high pressure process development in bench-scale Equipment

Remark

Classification

Media supply

Bottles (incl. riser pipe for liquid phase)

Rented

Pump

Membrane pump or Air driven booster

Safety relevant

Autoclave

Proper sealing Safety relevant Easy opening only when de-pressurized/easy closing Avoid fretting between moving parts Preferably no disassembly/assembly of tubing

Control valve

Needle valve with convertible seat

Safety relevant

Safety device

Burst discs (or safety valves)

Safety relevant

Instruments: flow meter, pressure gauges, temperature sensors

Indicating constant flow and registrating total volume used Indicating (plus registrating) pressure and temperature

Safety relevant

Tubing and fittings

Commercial classified materials only

Safety relevant

Utilities

Dry air, heating, cooling

Reliable supply

Laboratory

Spacious; incl. HSE devices

Safety relevant

5.2 Equipment for Bench-Scale Tests

199

Fig. 5.3 P&ID example for a SFE plant in bench-scale (Michel 1992)

Figure 5.4 shows a few examples of laboratory or bench-scale reactors and autoclaves which can be purchased as fully tested and certified equipment for maximum safety. It is advised to carefully consider the necessary instrumentation and rather opt for an autoclave with more ports in the body than initially thought. These can easily be blocked by plugs. Often times more measurements or injection points are wished at a later stage of process development. However, adding holes or ports to high pressure equipment later without having clear mechanical expertise can compromise the integrity of the equipment and must be avoided. Further it is advised to consider a sealing system that withstands a number of experiments and is not damaged too early when tightening the closure to the vessel. Ideally, the lid is covered and held by a large nut and does not turn (and damage the seal) when being closed.

Fig. 5.4 Examples for experimental reactors or autoclaves in bench-scale manufacturer NWA (left), Natex (right)

200

5 Design of High Pressure Plants for Research, Pilot and Production Scale

5.3 Equipment for Pilot-Scale Tests The next step in process verification and scale-up should be performed in pilotscale, meaning to operate a plant in small scale that allows to emulate a commercial set-up in smaller size. In many cases those pilot systems are very versatile settings allowing multi-purpose operation. In case of SCF extraction such systems often contain extractors for solid material processing as well as columns for liquid-sc phase separations. Further it is advised to allow for testing the effect of entrainers which results in a layout with 2 separators: one of which for the product, the other for the entrainer to be removed from the SCF. Such plants typically have to comply with explosion-proof design to be as flexible as possible with the choice of the entrainer. Several examples are given in Fig. 5.5 showing extraction units (solid and liquid) as well as SC reaction set-ups from some commercial manufacturers. These modern pilot scale plants are typically assembled in rigs or frames containing all equipment of a given process set-up. A big advantage of such skid-mounted plants is the possibility to assemble them under proper conditions of a professional workshop and perform all functional and tightness tests including a factory acceptance test (FAT) in that environment. Thus, it is made sure that any mechanical or electrical de-bugging can be done professionally in a workshop and reduces the risk to improvise. Also, the start-up at the final destination is kept very short in time and the user can focus on the operation rather than on mechanical work. Pilot-scale plants are used for two main purposes: firstly, to confirm and further optimize the process designed as well as—one of the highest quality tasks in process design—to compare experimental work with its theoretical simulation in order to minimize further practical work and its associated high cost. Secondly, to complement a commercial operation for training of operators, further continuous improvement measures or process developments, testing various (natural) raw materials from

Fig. 5.5 Example for experimental set-up in pilot-scale from manufacturer Sitec SCF extraction unit for solid and liquid processes (left); SC reaction unit (right) (photo: courtesy Sitec, Switzerland)

5.3 Equipment for Pilot-Scale Tests

201

other sources or provenience and many more reasons to keep a commercial producer competitive. Therefore, these installations should be equipped very much like a commercial plant in smaller scale and, in parallel, be kept very versatile for process development and open for further additions of single equipments or process units. During the early days of chemical engineering such plants might have grown over a longer period of time and may have ended like the installation shown in Fig. 5.6. The picture shows a versatile high pressure plant for sophisticated chemical synthesis at a chemical company. Operating such plants required a lot of experience and a good memory of the operator to remember the correct valves to be operated manually without too detailed indications in the plant. Since high pressure inside the equipment means a lot of stored energy and may cause damage, safety of operators or equipment must never be a matter for compromise. Therefore, the plant shall contain all relevant safety devices that may be found during a typical exercise during engineering: the HAZOP-study (HAZOP stands for hazard and operability). Along the engineering of the plant the process flow-sheets are subject to a thorough theoretical study in the presence of an independent auditor to list all consequences that may be seen in case of wrongful operation or failure of equipment and also the counter-measures taken to avoid damage. Further details about safety in layout, engineering and operation of high pressure plants will be discussed in Chap. 6.

Fig. 5.6 Example for experimental set-up in pilot-scale from Ruhrchemie ca. 1968, © (photo: courtesy of LVR-Industriemuseum, Oberhausen, Germany)

202

5 Design of High Pressure Plants for Research, Pilot and Production Scale

For efficient and safe operation of high pressure vessels it is a proven concept since long that the vessels may be opened and closed without disassembling or assembling pipelines or tubing. Since most SCF plants are used for processing solid materials those solids are typically put into the high pressure extractor in a basket. The basket is like a cartridge carrying the material; in most cases it is a tube with bottom and top plates made of porous material (sinter plates, mesh, filter material, etc.) depending on the size of the raw material to be extracted. The basket can be put into or out of the pressure vessel quite fast to minimize down-time, keep the vessel clean and avoid to blow the material through the solvent circuit. Of importance for plant layout and operation are two more things: the seal ring at the basket needs to be taken care of and be intact at all times to make sure the SCF flows through the bed of solids within the basket and cannot flow through the circumference between basket and pressure vessel inner wall thus being spent unused. Further, depending on the characteristics of the raw material (coarse, fine, fluffy, compacting, etc.) it is advised to install a by-pass between inlet and outlet of the extractor to avoid heavy compacting of the material placed within the basket during pressure equalization or start-up. This by-pass is shut after pressure is balanced in the extractor and before the pressure increase is started. Finally, it should be checked after an extraction whether or not the raw material was compacted during extraction and if channelling occurred. This channelling effect may also cause SCF to flow through the bed without effect and lower the efficiency of the process. Counter-measures could be a different mechanical preparation of the solid or a change of flow direction in the extractor (down, up). Ideally one extractor is equipped with at least two baskets to make sure the batch change is kept short and preparation of raw material can be done while one extraction is underway. In pilot-scale the basket is often connected to the lid of the pressure vessel, allowing to take out the basket when lifting the top cover. The design parameters of the extractor should not be chosen and fixed at too low “standard” values (i.e. 300–500 bar): a little higher design pressure may certainly cause a little higher investment cost, however, many processes in SCF are not running at the optimum parameters regarding yield, time, product characteristics, etc. When looking into literature many “process optimizations” are limited by the design or operation limits of the equipment in the test plants. The yield curves do not clearly contain an optimum, rather they often end at the experimental limits dictated by the hardware used. Therefore, it is highly recommended to chose the design pressure beyond the usual standards and keep the option open for a wider operating field during process optimization. This allows to really find an experimental process optimum. Whether or not this is also an economical optimum has to be worked out in a separate commercial exercise. A number of novel applications are leveraging the benefit of SCF under higher pressure (i.e. up to 1000 bar and beyond) and open a completely new arena for SCFs. The same applies for the design temperature, however, most of the extracts are sensitive to high temperature and therefore a design temperature of max. 100 °C seems sufficient for flexible process development. Nevertheless, in case adsorption/desorption are playing a major role in mass transfer during SFE, a higher operating temperature should be considered. This is of particular relevance for extracting harmful compounds in environmental processes.

5.3 Equipment for Pilot-Scale Tests

203

In order to obtain practical data for scale-up the size of the extractor in pilot-scale should not be chosen too small. Although large scale-up factors of 1:100 and beyond can be accepted in SCF processes, the pilot scale should be done at least in 4 L (10 L) scale for a process targeting on a commercial scale unit of 500 l extraction volume or larger. A general design of an extractor can be found in Fig. 5.7. The inlet and outlet of the SCF at the extractor are located at the bottom centre and at the upper end of the side wall. Through this design the top cover of the pressure vessel can be lifted and taken off without disassembling pipe-work. During this procedure also the basket—which is connected to the lid—is lifted as well and can be removed. The separator is collecting the extract after having adjusted the pressure and temperature of the media. Typically, the solubility of the extract in the solvent is several orders of magnitude lower than under extraction conditions. Thus, the separation of extract and fluid can in most cases be carried out be changing flow velocity and direction of the two constituents. In a special case the extract is a solid and it might be helpful to attach the control valve for pressure release from extraction pressure to separation pressure directly on top of the lid of the separator. The advantages are the extract is directly routed into the collecting vessel, the solid cannot precipitate in tubing and force clogging, cleaning of the plant is simpler and faster, a mass balance may be more correct, etc. Some internals can be further added to simulate a cyclone-like flow of the extraction medium (tangential flow) and filter fine solids from the fluid by a centrally positioned

Fig. 5.7 Extractor with basket attached to the top cover (Lütge 1993)

204

5 Design of High Pressure Plants for Research, Pilot and Production Scale

sinter candle. In case the extraction focuses on harmful substances the candle can be filled with adsorbent and act as a police filter making sure the fluid is not accumulating harmful ingredients. A design example as described before can be found in Fig. 5.8. In such case opening the separator requires the disassembly of the pipe towards the control valve situated on top of the lid. However, comparing the time saved for less cleaning of the plant makes it a simple manoeuvre. Upon closing the vessel, the connecting pipe needs to be safely assembled and properly tightened to close the solvent circuit again. Pumps for lifting the fluid pressure from buffer vessel level (in case of CO2 ca. 60 bar) to extraction pressure are typically membrane pumps with a hydraulic medium being the transmitter of force from the oscillating piston onto the membrane. A couple of parameters should be considered carefully when choosing the pump: the maximum operating pressure of the pump is defined by the target extraction pressure. The volume flow needs to be evaluated from prior tests and some references in literature, however, also here the throughput of the pump must not become the limiting factor of the experiments and their optimization. As a rule of thumb—in case of CO2 being the SCF—the pump capacity in kg/h should be rather 10 times the volume of the extractor in litre. Otherwise, the process optimization may lead to far too long

Fig. 5.8 Separator with direct release device incorporated in the top cover (Lütge 1993)

5.3 Equipment for Pilot-Scale Tests

205

extraction times making the overall process apparently uneconomical. Under normal operating conditions the pressure drop in the process loop is typically small (as is the viscosity of the SCF). Thus the volume flow can be high and is usually limited by solubility and mass transfer characteristics. As the pumps are reciprocating machines a pulsation dampener is advised to stabilize and equalize the process characteristics of the fluid flow. For larger volume flows a multi-head design of the pump should be considered to achieve even better steady state flow conditions. A lot of research and development work has been accomplished on this topic both at universities and at pump manufacturers. Therefore, for more details reference should be made within the relevant literature elsewhere. For isobaric processes (i.e. processes with little pressure drop) also a compressor can be considered to move the fluid within the circuit. Such process is of particular economic advantage in case of long extraction times, large fluid volumes and separation conditions close to extraction conditions. A practical example is the decaffeination of coffee where the SCF carries the caffeine away from the coffee beans and can easily be cleaned by washing the SCF with water in an absorption column. Heat Exchangers in pilot plants are typically tube-in-tube designs. This means a rated high pressure tube is surrounded by an outer shell which is assembled by simply mounting the shell with standard fittings and sealing packings on the outside. This leaves a lot of flexibility in case of process adjustments and also is a mechanically simple yet robust design. A Buffer Vessel for the SCF is needed to complement the solvent circuit and has two main tasks: firstly, it should carry enough solvent to fill the circuit under process conditions and provide sufficient liquid level and pressure of the solvent on the suction side of the pump. Secondly, when shutting the plant down the solvent contained in the circuit should be fully re-collected in the buffer and save solvent cost. As the pump typically needs liquid solvent on the suction side and the pump characteristics require avoiding cavitation, a certain sub-cooling of the liquid solvent is essential. It is a further advantage, if the sub-cooling is achieved with minimum temperature fluctuations. As the solvent is at its boiling point in the buffer vessel (vapour-liquidequilibrium), temperature fluctuations can induce remarkable pressure fluctuations and finally result in volume flow fluctuations of the solvent in the cycle. To obtain reliable and repeatable experimental results, a steady flow of SCF in the process is advised. Piping, Fittings, Instruments, etc. can be found from several specialized manufacturers of those plant components and will not be described in every detail here. However, it should be mentioned another important topic for consideration: it might be very helpful for robust plant operation to install some inline filter devices to make sure fine solid particles (raw material dust, ice, etc.) are not being transported in the SCF circuit and may attach to sensitive moving parts in non-static equipment. E.g. the ball valves inside the pumps may not close properly and cause unnecessary experimental failure or maintenance and repair. Also the valve seats or stems may be

206

5 Design of High Pressure Plants for Research, Pilot and Production Scale

influenced or damaged by fine particles resulting in incorrect process flow, pressure control and insufficient operation. A typical example for a P&ID of a pilot scale extraction plant for solid materials is given in Fig. 5.9. The pressure vessels are extractor E, separator A and buffer vessel V. All heating and cooling supply is done by independent devices allowing for very flexible choice of temperatures at the various process stages. Tube-in-tube heat exchangers W are installed at relevant places as well as a number of filters F making sure no solids are transported within the loop and sensitive equipment (pump, valves) is protected. Safety against over pressure is realized through burst discs, header lines collecting vent streams are connected to proper external cleaning and routed to the outside. The Process Control System can be kept simple and only needs to contain control loops for the most important parameters such as extractor pressure, separator pressure and volume flow of SCF, ideally equipped with an integration device to control the total amount of SCF spent. Temperatures may be adjusted manually and monitored from time to time to keep the plant investment cost low. Most extractions focus on natural matter which properties may deviate depending on the particular harvest, provenience, etc. In such applications an online monitoring of the SCF leaving the extractor could be helpful to learn about the propagation of the extraction and optimize the process. This additional tool also enables the operator to quickly decide interrupting or stopping the process in case of unusual observations. A very schematic idea is presented in Fig. 5.10 applying an online monitoring system attached to the extractor.

Fig. 5.9 Example for a typical P&ID of a pilot plant for SFE (Lütge 1993)

5.4 Equipment for Commercial Scale Plants

207

Fig. 5.10 Example for on-line monitoring of extraction progress (Lütge 1993)

5.4 Equipment for Commercial Scale Plants Often times the question arises which size of a plant should be considered “commercial scale”. Due to very different valuations of the products being synthesised or processed one single precise figure differentiating pilot scale and commercial scale cannot be given. In pharmaceutical applications a commercial scale plant can be as small as a few litres up to 100 l pressure vessel volume. In contrast to such relatively small commercial size a big system—as is used for decaffeination of coffee or for treatment of other natural substances such as rice or wood—a commercial plant will range up to 20 m3 of extractor volume. Most commercial scale batch-process plants, however, are in operation in the range of ca. 200 l up to 3500 l extractor volume. Typical modern LDPE plants of world-scale size are built in the range of ca. 400 kt/ a product output and consist of a tubular reactor system (consisting of intercooler, compressors, reactor, high pressure control valve, high/low pressure separator); the core equipment of such reaction system is operating under different pressure stages of ca. 300 bar, ca. 1600 bar up to 3600 bar and offer a total processing volume of more than 50 m3 . Nevertheless, these systems are filled with the reactive component ethylene and carry a much higher process pressure compared to typical SFE applications. Thus, the highest diligence is essential and strict quality requirements must be met during all stages of manufacturing the equipment, its testing procedures and the plant assembly. The target must be to set-up a robust production plant which can be

208

5 Design of High Pressure Plants for Research, Pilot and Production Scale

operated safely and long-lasting during its daily operation over the long depreciation period of the asset.

5.4.1 Supercritical Fluid Extraction Plants In case solid material is processed, the vessels are typically arranged for vertical design; the solid material is fed through the top opening and in some cases a basket is no longer utilized to position the solid bed within the extractor. Rather the exhausted solids are removed from the bottom opening using gravity to initiate the solid flowing out through a lower port automatically opened and closed by another lid. For some applications such as wood treatment, aerogel drying or parts cleaning a horizontal design has proven beneficial. In these plants the material is fed and removed in and out the vessel through a front door which might be facilitated by suitable carrying or handling devices similar to a dish washer in a kitchen. A principle flow-sheet of a typical commercial plant for SFE is shown in Fig. 5.11. Three extractors are operated in cascade mode to utilize the solvent at best to the solubility limit and extract the raw material as far as economical or required. The extraction can be supported by adding a co-solvent from the tank via a pump displayed to the left. Such set-up can be an advantage in case polar substances are to be extracted with a non-polar solvent. This can either be accomplished by adding the co-solvent over the complete cycle and utilize the staged 2-step separation to separate the polar fraction and the co-solvent from the mixture in one separator and the less polar extract in the other. Another option can be to operate in 2 different extraction modes: Phase 1 comprises a pure SCF extraction taking out a particular portion of non-polar product. Phase 2 would then be a SFE process supported by entrainer producing a more polar product fraction. In some natural extraction applications using SC CO2 only, the 2 separators can further be used to collect different fractions or phases of product as is done for processing of spices to separate a hot solid product and a more tasty viscous fraction. During process optimization it also has to be worked out, which number of extraction vessels at a given volume is the optimum number for the plant capacity in focus. This evaluation was executed many times and published in various papers. The solution mostly arrives at 2–4 extraction vessels as can be seen from Fig. 5.12. A higher number of vessels increases the investment cost while a small number lowers the efficient use of the solvent due to weaker load with extract. The economical optimum has to balance both cost drivers to define a proper compromise between OPEX and CAPEX. A real plant photograph is shown in Fig. 5.13 for an industrial scale plant for SFE. The technical parameters are given in the description of the figure. The picture shows the top covers of the extractors in the front row; the extractors are equipped with quick-acting clamp closures and carry baskets inside (not shown). The steel structure in the background carries heat exchangers, control valves, pipe work, etc. and allows sufficient clearance for the extraction baskets to be manipulated.

5.4 Equipment for Commercial Scale Plants

209

Fig. 5.11 Cascade operation extraction of solids—3 extractors with 2 step separation

Fig. 5.12 Specific cost of SFE relative to plant capacity and number of extraction vessels

The pumps and sub-coolers of liquid CO2 are located on the ground floor one level lower. A particular feature of this plant is the relatively high design pressure of 700 bar while using an extractor volume at the lower end of the scale of 150 l only. Most SFE plants are designed for operating pressures around 300 or 500 bar. In particular applications, however, a higher extraction pressure is making the process more efficient. This plant is the first commercial example using an design pressure of 700 bar allowing higher extraction pressures of ca. 650 bar while reducing the volume of the extractors to 150 L due to process intensification and acceleration under such elevated process conditions for the extraction of algae. Another practical

210

5 Design of High Pressure Plants for Research, Pilot and Production Scale

Fig. 5.13 Industrial extraction plant with 3 Extractors 150 l, 700 bar, 2 Separators 150 l, 350 bar and 100 bar, 1 CO2 Pump, 2200 kg/h (photo: courtesy Uhde High Pressure Technologies, Hagen, Germany)

example is the use of even 1000 bar for extracting Xanthohumol from hops. These polyphenols are health supporting, however, under typical conditions of supercritical hops extraction still insoluble. In case of larger size of the extractors a different closure system for the top cover is applied which is known as segment ring closure (see Fig. 5.14). The purpose of the design is to have the maximum diameter available as opening port and use the whole diameter of the forged vessel without a bottle-neck to lock the lid safely into its position. Figure 5.14 shows the cover seal as yellow coloured ring at the bottom of the lid piece. As soon as the lid is lowered into its position to close the vessel, segment rings are radially moved out into a shoulder in the extractor wall. Those segments are then mechanically locked by the top ring and hindered to move out of their position safely holding the lid. The whole system is operated by hydraulic actuation and interlocked to make sure no wrongful operation is allowed. Figure 5.15 shows to the left the bottom part of another extraction vessel which is used for a spray drying process forming dry lecithin powder particles. The special feature of this commercial scale plant is the continuous feeding of liquid degumming residue into a SFE plant. While the liquid raw material can be pumped into the extractor continuously, a special removal system was designed and built to take the dried solid powder out of the extractor without de-pressurizing the whole unit every now and then. The lecithin powder is collected at the bottom of the extractor for a period of time. Thereafter, the ball valve at the bottom is opened (centre piece with blue vertical actuator), connecting upper extraction vessel and the lower hopper

5.4 Equipment for Commercial Scale Plants

211

Fig. 5.14 Large scale industrial extractor with segment ring closure 500 bar, 3,500 l (photo: courtesy Uhde High Pressure Technologies, Hagen, Germany)

vessel at the same pressure and the dry powder is filled into the hopper vessel by gravity. Upon closing the ball valve the hopper vessel is de-pressurized and the lecithin powder can be removed to the environment (see right picture in Fig. 5.15). In the next step the hopper vessel is closed and brought back to the pressure level of the spray drying and extraction process and the ball valve can be opened again to remove the next batch of dry lecithin powder. Through this process pattern the extraction is running in continuous mode and only the hopper pressure is fluctuating. The plant is equipped with 2 similar extraction lines and runs continuously to produce high value dry lecithin powder from a residue stream of a conventional soy bean oil mill. SCF is used to remove traces of conventional solvent, water from the steaming/ degumming step as well as soy bean oil left in the stream. More process details can be found in Chap. 3. The layout of the aforementioned plant is shown in Fig. 5.16 giving a more general overview of the lay-out of the plant. The 2 extraction vessels and 2 hoppers are shown in the foreground, pumps are located in the basement, 2 separators are situated at the backside while heat exchangers are implemented within the pipe work at suitable positions. Control valves, filters and heat exchanger as well as piping are positioned in the space between the extractor and the separators to allow good access to the core of the plant (i.e. the extractors and separators). The heat exchangers are mostly tube bundle heat exchangers that are connected to the supply of cooling or heating media from a utility section elsewhere. The pressurization and flow of SCF within the plant circuit is achieved and maintained by piston pumps as shown in Fig. 5.17. The horizontal line in the foreground

212

5 Design of High Pressure Plants for Research, Pilot and Production Scale

Fig. 5.15 Bottom section of spray drying extractor vessel with ball valve connecting a hopper vessel (left); view to bottom of hopper vessel with dry lecithin powder product (right) (photo: courtesy Uhde High Pressure Technologies, Hagen, Germany)

Fig. 5.16 3 D model of a spray drying plant using supercritical CO2 for de-oiling raw lecithin (photo: courtesy Uhde High Pressure Technologies, Hagen, Germany)

shows the CO2 supply fed into the pump head. The vertical line to the left is complemented by a ball-type pulsation dampener on top of the pump head to equalize the CO2 flow supplied into the process. The whole machine is positioned on a strong, slightly elevated foundation with sufficient space around to facilitate maintenance at

5.4 Equipment for Commercial Scale Plants

213

Fig. 5.17 Piston pump for supply of supercritical CO2 in de-oiling plant of raw lecithin (photo: courtesy Uhde High Pressure Technologies, Hagen, Germany)

the pump. Further, a soft material between the foundation and the rotating machinery makes sure vibration is also de-coupled and reduced from the rest of the plant. Most valves, i.e. control valves, shut-off valves, safety valves etc. are often arranged in a plant section between the extraction vessels and the separators as can be seen in Fig. 5.16. The control and shut-off valves typically consist of a valve body made from a forged block of suitable material. The valve body and the size and shapes of its internals are calculated and designed acc. to the flow conditions prior to and behind the valve. It is important to avoid too high velocity of the fluid which may cause high noise levels or even more severe danger of damage such as cavitation. A few examples from a practical plant can be found in Fig. 5.18. The valves are designed as angle valves with a 90° direction change of the fluid and are equipped with pneumatic actuators on top moving the stem within the assembly. The pipe-work within the plant can be established from seamless tube parts and fittings of suitable sizing. As many SCF plants are producing materials used in the food and beverage industries, for nutraceuticals or even pharmaceuticals the materials chosen are often stainless steels. Therefore the pipe-work is typically welded and all welding seams are checked by x-ray prior to pressure test of appropriate sections within the whole plant assembly. Most plants using supercritical fluids in physical separation processes are being built so compact and narrow and are often extended over time making it difficult to

214

5 Design of High Pressure Plants for Research, Pilot and Production Scale

Fig. 5.18 Section of a SCF plant with special designed control and shut-off valves (photo: courtesy Uhde High Pressure Technologies, Hagen, Germany)

present nice photographs giving a good overview of such plants. This is particularly true for absorption towers or columns in commercial size. While a column in pilotscale is easily integrated into the skid-mounted plant (refer to Fig. 5.5) this is mostly not possible in large scale as the columns are typically the tallest equipment and positioned apart. One example for an absorption tower used for washing caffeine from SC CO2 is shown in Fig. 5.19. The tower consists of 3 seamless forged segments of high strength steel connected by metal seals and flanges. The photograph shows the column under assembly with bolt tensioning devices to prepare the equipment for the pressure test. For final operation the column will be equipped with structured metal packing inside the hollow body to intensify the mass transfer from the SC CO2 into the washing water. A bottom section and a top head carrying pipe connections are also added and the total assembly is finally positioned upright.

5.4.2 Additional Features of High Pressure Plant Equipment The pressures applied in separation processes using supercritical fluids mostly allow for weldable alloys as the main materials for equipment and plants. Only in very rare and exceptional cases the body material of pressure vessels is made from high strength, non-weldable steel. However, in high pressure processes such as the synthesis of LDPE from ethylene the pressures in large sections of the plant are much higher. Thus, these equipments require a particular material choice using alloys with yield strengths of 800 MPa and above (SA 723 and similar). These steels do not allow for welding as their carbon content is above 0.22% and makes them susceptible to embrittlement which would jeopardize equipment and plant safety. An overview of steels typically used in high pressure equipment is given in Table 5.3. Due to these particular material properties some special requirements for the design of equipment used under such harsh process conditions have to be met. In

5.4 Equipment for Commercial Scale Plants

215

Fig. 5.19 Absorption tower (column) under preparation for pressure test (photo: courtesy Uhde High Pressure Technologies, Hagen, Germany) Table 5.3 List of weldable and non-weldable steels for high pressure equipment (Stahlschlüssel 2007) Steel type (typical grades)

Yield strength at 20 ºC (MPa)

Steel type (typical grades)

Yield strength at 20 ºC (MPa)

High-alloy (>12% Cr) steel or Austenitic stainless (1.4429, 316LN)

~300

Low-alloy (≤12% Cr) steel, Fine-grain and high-temperature structural steel (1.6368)

~450

Weldable

Austenitic-ferritic duplex (1.4462)

~450

Cr–Mo H2* resistant steel (1.7779)

~500

Weldable

Soft martensitic (1.4418)

~800

HSLA (1.6580, A723Gr1)

~850

Nonweldable

Precipitation hardening ~850 (1.4545, 15-5PH)

HSLA (1.6957, A723Gr3)

~900

Nonweldable

216

5 Design of High Pressure Plants for Research, Pilot and Production Scale

case welding is impossible (and therefore not allowed), piping has to be designed with threaded flanges connecting the tube parts with bolts and nuts. The sealing between the pieces is accomplished by a spherical metal lens ring in contact with conical surfaces of the tube pieces. A principle example is shown in Fig. 5.20. The disadvantage of the high tightening forces is compensated by two different benefits: the tightness of the connection is supported by the internal system pressure and the connection can be considered safe as a leakage would indicate a coming failure in case one of the bolts would be stressed beyond its limits. Also tubular reactor systems for LDPE production are made up of tube elements applying this design principle. The main difference is the wall thickness applied depending on the process pressure within the particular plant section. The same principle is valid for the valves and safety devices used in such ultrahigh pressure processes. Figure 5.21 shows a reactor pressure control valve from a commercial LDPE plant made up from a forged body and bolted connections towards the pipe-work. Channels in the body are used to heat or cool particular areas of the body making sure the viscous product is not solidifying or the stem packing is not worn too early due to high temperature.

Fig. 5.20 Threaded flange connection with lens ring seal (metal to metal)

5.5 Turning Process Development into a Commercial Plant

217

Fig. 5.21 Reactor control valve from a LDPE plant (photo: courtesy Uhde High Pressure Technologies, Hagen, Germany)

5.5 Turning Process Development into a Commercial Plant 5.5.1 Considerations for a Commercial Plant Project During the final phase of a process development towards its commercial application further perspectives come into play and dominate the investment decision at this stage. The final considerations—among others—will be whether, how and to who the product can be sold on the market, at which price and to what market demand. From the main answers a plant capacity and some plant features can be derived that are determining the associated conversion cost. Additional topics to be solved are the availability of raw material or feedstock as well as an appropriate location to install the plant. Important site factors that strongly influence the overall conversion cost are e.g. energy cost or availability of utilities and skilled labour, logistics cost to/from the site, construction cost and more factors being constraints to plant design and construction. In commercial SCF applications the number of plants for one single process application is often limited due to the quite specific use of the individual SC processes. E.g. the commercial plants for hops extraction are concentrated at a very little number of regions where hops is grown. The overall number of commercial plants is around 10

218

5 Design of High Pressure Plants for Research, Pilot and Production Scale

and represents the largest single SFE process application worldwide. Other processes satisfy the world demand at even smaller scale or a smaller number of plants going down to one or two plants worldwide only. Therefore, flexible plant designs for more than one process application are often required to make the production commercially feasible. Thus, we find several multi-purpose plants in the world that are operated for different separation processes and produce different products in campaigns on the same commercial plant. In such case a proper, safe and quick cleaning of the plant equipment is an important factor and avoiding cross-contamination of different materials is essential. To facilitate such plant operation CIP features (i.e. clean in place) may complement the high pressure equipments to allow for qualified cleaning manoeuvres. In summary we are seeing a highly sophisticated technology requiring huge expertise being applied in a technological niche. However, this must not be understood as a hurdle too high to be overcome for novel process solutions. Future options to leverage more high value commercial applications into the market could be the following three ideas: firstly, the “new” product for market introduction could be processed in another, already existing plant buying toll manufacturing capacity elsewhere. This gives a much better impression of market demand and customer behaviour and postpones the associated investment cost to a later stage with less commercial risk. The conversion is based on variable cost instead of fix cost and the burdens of interest rate or depreciation in case of lower demands and an idle asset can be avoided. We find such models also applied in beer brewing or production of spirits and the idea can be transferred. A second option could be to share the investment with other owners dedicated to similar processes and co-own a joint facility. As many process developments envisage the same capacity difficulty to commercialize their product, a community of fellow sufferers exists and bundling the interests might be a way forward for all together. Thirdly, the brave approach could be to set-up a plant suitable for ones’ own purpose and vision and actively offer the spare capacity on the market. Taking the initiative might bring some advantage in the long-term, however, the early stage risk must be taken consciously.

5.5.2 Execution of a Commercial Plant Project Different to the earlier stages of setting up smaller scale systems for process development a large commercial facility should not be understood as buying equipment that has “only” to be assembled. Too many new factors—some of them have been mentioned before—are strongly influencing the investment and will finally also determine the conversion cost. Therefore, it is essential to handle a commercial plant installation as a serious project, implement professional project management and do not underestimate project logistics, the overall complexity and the importance of a well defined and managed project schedule. The general definitions of the process, the project and its execution are laid down in the conceptual phase. The project typically goes on with a documented summary of any and all details for the site, the process, codes and standards, user requirements, etc. that are compiled

5.5 Turning Process Development into a Commercial Plant

219

in one very important document: the design basis. This document is to be understood as the solid foundation assembling all necessary information and requirements definitions to be supplied to the various engineering disciplines contributing to the project. Working out the design basis may take its time, however, clear information for engineering at the beginning avoids misunderstandings or confusion later on. Also the project schedule needs to be fixed early on. Too many projects derailed (in terms of schedule and/or budget) due to lack of transparency and clear definitions or along changes that became necessary to comply with basic requirements that were initially overlooked. Thereafter, the process design needs to be finalized and double-checked in all details. Since process engineers are very creative humans and tend to change many things still during project execution, freezing the process design is an essential milestone of the project. Important deliverables are the Process Flow Diagram (PFD), equipment specifications (incl. sizing, material choice, etc.), utility requirements, equipment lists as well as the P&ID (piping and instrumentation diagram). The final stage of process engineering forms the HAZOP study (hazard and operability) which might have a last influence on the process design and some instrumentation or piping and equipment features. After this point the process engineering should not be revised any more to keep the rest of the engineering disciplines on track and avoid costly iterations. Next phase is the detailed mechanical design of all equipment, the plant lay-out incl. steel structure design, further the pipe routing as well as electrical and instrument engineering are carried out. From all engineering disciplines the respective material take-off (MTO) is forming the basis for procurement. At this stage it must be considered which items are long-lead items and need to be ordered early on in contrast to items being purchased off the shelve. Also a well defined strategy for spare parts that will be needed in the different project phases (construction, commissioning/start-up, 2 years operation) is helpful. A summary of critical project phases is given in Table 5.4. As some of the engineering tasks can be performed in an overlapping collaboration, the overall execution schedule of such commercial scale project can be expected in the range of 18–24 months. Decisive factors for the project duration are the procurement of the main materials for the heavy equipments on the one hand and the execution of the construction on-site on the other. The main materials are the forgings coming from a limited number of forging companies delivering high quality and reliable materials. In case those companies are loaded with work for other large scale parts or projects, the supply of the pre-material to fabricate the pressure vessels, heat exchangers and other heavy equipment may take as long as 6–12 months. Only thereafter the fabrication can start and must not contain any failure during welding procedures or machining. Otherwise the schedule is corrupted and the project delayed. Therefore, early definition of these heavy pieces and their procurement from reliable sources is essential for the successful project. Since most projects are financed the schedule must be controlled very accurately. A good means to get better control over the project schedule, shorten the time required for installation and also realize a much higher quality of the assembly is to build the

220

5 Design of High Pressure Plants for Research, Pilot and Production Scale

Table 5.4 Summary of critical project phases and their duration (Shallice 2006)

Phase

Duration weeks

Conceptual phase

8–

Design basis document

4–8

Process engineering

8–12

HAZOP study

2–4

Mechanical design, electrical and instrument engineering

12–20

Procurement and fabrication

12–26

Construction (modules vs. stick-built)

12–32

Commissioning and start-up

8–12

plant in modules. Even in case the assembly manpower cost are slightly higher than on-site the project location, the shorter assembly schedule and higher quality often pay-off. The modules are shipped to the site and only few connections have to be made between them in order to keep the on-site work as short as possible. Figure 5.22 shows the approach with pre-assembled modules for a SFE plant and gives an idea of the delivery to the site. Logistics may be slightly more complex and need to be considered already during design phase in order to plan for modules which fit in size and weight to the logistical conditions between fabrication and installation site.

Fig. 5.22 Pre-fabrication of plant module (left); delivery of module at site (right) (photo: courtesy Shallice 2006)

Figure 5.23 gives another example of a pre-assembled intercooler module for a world-scale LDPE plant being ready for delivery ex works after full assembly and pressure test. The module size and weight are even heavier than the aforementioned

References

221

Fig. 5.23 Intercooler module ready for delivery ex works (photo: courtesy Uhde High Pressure Technologies, Hagen, Germany)

examples and need to be considered prior to design in order to make sure the logistics can be handled properly. Project schedules tend to be of ever more importance and shortening the schedules of different comparable projects over time is another trend in engineering and construction. Therefore, the modularization of process plants is enjoying more and more application during design. This trend also helps improving the plant quality and safety in high pressure.

References Ellis ER (2005) Polythene came from Cheshire: the discovery, early development and production by the high-pressure process, 1933–1971. E.R. Ellis, Chester Lütge C (1993) PhD Thesis. PhD, University of Dortmund, Institute of Thermodynamics Michel S (1992) PhD Thesis. PhD, University of Dortmund, Institute of Thermodynamics Shallice C (2006) Execution of major supercritical projects. In: 7th International Symposium on Supercritical Fluids, Orlando, Florida May 1–4, 2005. J Supercritical Fluids. Elsevier Science BV, Amsterdam, Netherlands Stahlschlüssel (2007) Verlag Stahlschlüssel Wegst GmbH

Chapter 6

Safety and Control in High Pressure Plant Design and Operation

Abstract High pressure is applied in industry since long e.g. for fired boilers or in chemical industry. Through learning from many incidents a general framework evolved handling the challenges of using high pressure safely. National and international codes and standards are embedded in laws or directives to ensure proper design, equipment manufacturing, process control, plant operation, etc. Through clearly defined sequences for engineering and design a safe and reliable practise of high pressure can be considered "state of the art" today. Keywords Plant safety · Equipment design · Plant layout · Fabrication · Pre-stressing · Testing · Operation · HAZOP · Engineering ethics

The practical application of high pressure in industry started as early as in the eighteenth century and was initially driven by the development of the steam engine. James Watt improved the early models of steam engines combining in his own work what today translates into a smart connection of mechanical engineering, thermodynamics, process engineering and material science. He firstly improved the mechanical features of the steam engine. In parallel, he made a reduction of primary energy consumption of more than 50% possible by adding a steam jacket to the cylinder and moving the condensation into a separate process step. Thirdly, he found a better material to build his installations. The overall development and industrialization took its time, consumed a lot of money and, unfortunately, also caused numerous casualties and destroyed many assets. Only after a large number of accidents and casualties in industrializing countries, rules were developed and enforced aiming to reduce and avoid further incidents with fired boilers. Within the second half of the nineteenth century various institutions were established who are still ruling and regulating the application of steam engines and fired boilers. These institutions were established to primarily supervise the testing of fired pressure vessels and, secondly, make sure regular inspections are carried out and, finally, common design principles are accepted to reduce the number of incidents and its severity. Such institutions are e.g. the TÜV in Germany (Technischer Überwachungs Verein, est. 1866), the ASME (Peters and Ritter 2019) in the USA (American Society of Mechanical Engineers, © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Ž. Knez and C. Lütge, Product, Process and Plant Design Using Subcritical and Supercritical Fluids for Industrial Application, https://doi.org/10.1007/978-3-031-34636-1_6

223

224

6 Safety and Control in High Pressure Plant Design and Operation

Fig. 6.1 Left: Grover shoe factory disaster (Times 1905) right: Steam locomotive boiler accident at Darlington on 2nd February 1850 (Wikipedia 2022)

est. 1884) and similar bodies elsewhere. Unfortunately, founding such institutions mostly was the result of a number of disasters and the response to bitter lessons learnt. Various summaries of accidents on railroad, with shipping vessels or fixed boiler installations can be found in literature. Figure 6.1 shall only be understood as harsh visual examples to emphasize the importance and severity of plant safety in pictures rather than words. A second driving force pushed for the broader application of high pressure mainly in chemical industry which is known as the principle of Le Chatelier which was formulated at the end of the nineteenth century. Based on this principle an increasing pressure forces a reaction equilibrium to shift to that side of the reaction that requires less volume. A very popular example often taught in chemical engineering schools is the formation of ammonia from hydrogen and nitrogen under high pressure which was developed and industrialized at the beginning of the twentieth century. Therefore, the early rules and regulations for safe high pressure boiler installations needed to be extended to pressure vessels in general and had to include unfired containments of high pressure as well. State of the art for design and manufacturing of high pressure equipment, piping and plants are Codes and Standards that are enforced in their respective regional area of application. Examples are the ASME framework (Peters and Ritter 2019) which is valid in the USA and widely applied in international markets. Within the European Union the law is known as Pressure Equipment Directive (EU-OSHA 2014) and conform design and manufacturing have to follow harmonized norms EN 13,445 (European Standard 2021) for unfired pressure vessels resp. EN 13,480 (European Standard 2012) for piping. In other regions of the world comparable laws have to be followed (i.e. China/SELO, Korea/KGS, Japan/JIS, etc.) and regulate the technical application of high pressure. Any such rules are typically incorporated into laws, must strictly be obeyed for design, manufacturing, inspection and testing as well as in operation, and typically need to be supervised by independent thirdparty inspectors (notified bodies). Through these frameworks a safe design, proper

6.1 Safety Considerations for High Pressure Equipment and Plants

225

manufacturing and testing as well as a reliable operation of high pressure equipment and plants are ensured. The safety of people working or living near a technical high pressure installation is realized on a very high level and insurance companies offer coverage for the assets under operation. Finally, unfortunate incidents of failure are meanwhile mostly reduced to human error. In essence this means any error or failure working under high pressure must be analysed in depth and utilized as a means for learning and progress. Current statistics show that the framework of safety for high pressure equipment and plants is covering the field properly; the number of incidents is kept on a very low level as is the number of casualties. Making sure the high pressure process is properly operated as designed and the equipment and plant safety are not compromised, the instrumentation of the asset and its control system are forming integral parts of the modern overall safety philosophy. Therefore, both elements will be discussed in more detail in the following chapter.

6.1 Safety Considerations for High Pressure Equipment and Plants The safe operation of a high pressure process covers many aspects that shall be discussed along the sequence of turning an initial process idea into the reality of hardware and into operation. Therefore, the consideration commences with explanations and recommendations for engineering and manufacturing of high pressure equipment as well as the assembly of a plant. Thereafter, the necessary diligence in operation and care for the equipment or plant is discussed. Finally, some considerations are put forward how to safely handle unforeseen external forces possibly compromising the integrity of equipments.

6.1.1 Diligence in Design and Fabrication Before beginning with engineering and design of equipment a general understanding of the applicable requirements of the installation or “the project” is necessary. As laid out in Sect. 5.5.2 the best way to compile this information is one comprehensive document called Design Basis. This document defines—among many other details— the requirements of the site and thus the applicable pressure vessel and piping code to be complied with. Within the relevant code also materials for design and fabrication are defined and it is indispensable to dive into the current high pressure material knowledge and its practical know-how as a first step.

226

6.1.1.1

6 Safety and Control in High Pressure Plant Design and Operation

Considerations for Design Materials

In general it can obviously be stated that for equipment used under high pressure conditions a high strength material is required. Such materials are still today mostly steels, often tuned to high-strength properties by particular emphasis on reducing trace components and improved heat treatment procedures. Through this also a sustainable design becomes possible as the material and energy consumed are reduced compared to older design materials with weaker mechanical properties and shorter life-time. Nevertheless, the number of applicable alloys remains small as the demanding conditions for operation under high pressure are very particular. The main criterion for the material properties is the ability to withstand the static forces induced by the inside high pressure. Further properties define the ductility of the material and particularly keep the resilience of the material intact also at low operating and/or environment temperatures. Next—in case of a fluctuating pressure (e.g. in pump heads)—the fatigue properties have to be considered. Furthermore, in case of a chemically challenging process regime the resistance to corrosive attack or hydrogen corrosion have to be mastered. Finally, long-term characteristics against high temperature (e.g. creeping) shall be considered. A summary of widely applied alloys can be found in Table 5.3. Typical defects induced to the material already during fabrication in the steel mill may be—among others—inclusions due to impurities in the raw material or from the melting process in the furnace, porosity or cavities due to a lower degree of deformation during hot forming, or cracks from wrongful heat treatment. All these deficiencies have to be avoided by closely monitoring the mill process and by a defined pattern of destructive and non-destructive testing prior to the start of mechanical manufacturing of the high pressure equipment. The tests giving evidence for the proper material characteristics and in some cases also reflecting a proper part design are summarized in Table 6.1. The destructive test methods of a tensile test and an impact test are usually applied along the steel production process in the mill and witnessed and documented by an Table 6.1 Examples for destructive and non-destructive tests during fabrication Destructive testing methods

Non-destructive testing methods

Specimen test

Actual part test

Tensile test (yield strength, elongation)

VT—visual testing

Notch impact test (impact strength)

PT—penetration testing

Actual part test

MT—magnetic particle testing

Fatigue test (longevity under fluctuating pressure) UT—ultrasonic testing LBB test (leak-before-burst evidence)

RT—radiographic testing eddy current (plus specimen for calibration) Pressure test Leakage test

6.1 Safety Considerations for High Pressure Equipment and Plants

227

independent, third party inspector. In case of a large number of similar size working pieces a certain specimen pattern may be arranged as long as the mill statistics allow for such simplification. Passing the stringent test regime also means a hard stamping of each and every metal piece. Thus a firm and unambiguous assignment of the raw steel part to the documented test certificate is done. Through this principle a clear tracing and tracking of any and all parts the whole way back towards the steel production is strictly documented and practically ensured. Only in special cases of a particular new material application or a changed resp. a new design the two other mentioned destructive tests on actual parts are used as they are typically time consuming and more costly. A fatigue test helps to understand the material behaviour under a fluctuating pressure regime and also gives a good indication for the longevity of a given design of a high pressure containing part in combination with a chosen alloy. This is often applied to prove new materials on a given part design or to check the new material against previously used material properties. Alternatively, the improved longevity of different part designs for the same application of a particular high pressure part may be practically demonstrated using the same material. Thus, the comparison of the actual test result with previous experience from reference parts is forming the basis of this test, its interpretation and should be documented in detail. A lot of scientific literature also exists around the final principle of destructive tests for thick-walled high-pressure parts mentioned in Table 6.1 which is called leak-before-burst (LBB) and can be studied in more detail elsewhere. The general idea, however, can be explained in brief as follows: assuming high pressure parts are susceptible to failure through a variety of reasons anyway, it is essential to understand the failure mode beforehand and control the failure to a minimum damage. This can be achieved by leaving a reasonable time between the first indication of a near disintegration of a part or an assembly of parts (typically “a leak”) before the part(s) break(s) (the “burst”). The time in between those two incidents must be long enough to depressurize the equipment and evacuate people in the vicinity safely while the residual material ligament around the leak port needs to maintain the overall integrity of the construction until complete shut-down is accomplished. This general idea can also be used as a destructive testing method and a proof of concept for a given design. Figure 6.2 (Lütge 2011) shows an LBB proof part with the outside wall picture to the left and the almost ideal pattern of material destruction within the wall to the right. The indication of the end of its working life was achieved by a leak and the surrounding material held the overall part in one piece—thanks to the ductility of the material chosen. When breaking the wall open it becomes evident in the right part of the figure that the starter was a crack on the inside of the wall. For test purpose such starters can be induced in a defined mode as inside scratches and the failure mode of the part may be studied through a pulsation test until destruction of the part. The non-destructive test methods mentioned in Table 6.1 are often applied in addition during the fabrication sequence of high pressure parts to check the outer surface quality (VT, PT), the inner surface quality (eddy current) or the uniformity of properties and the absence of pores, inclusions or crack starters within the material volume (MT, UT, RT). Since the specimens taken for destructive testing may be

228

6 Safety and Control in High Pressure Plant Design and Operation

Fig. 6.2 Left: leaking fissure at a part surrounded by an area of plastic deformation, right: typical pattern of destruction within the wall commencing from a crack

small in size and taken from selected areas compared to the real long and heavy working piece, such non-destructive tests provide additional security in terms of uniformity of the material used. The applicable codes give additional guidance and a more comprehensive overview can be found elsewhere. Understanding the failure mode was historically essential to improve the materials used and the design principles applied in high pressure in order to drive the number of incidents down by additional safety considerations and practical measures. A sudden rupture and catastrophic destruction must be avoided at all means and the material properties can support achieving this target in a very important manner. We have to bear in mind that the stored energy within the pressure containing parts may be critically high, especially when a compressed gas is contained in the large volume of a high pressure system. However, even a liquid compressed to a very high pressure does not have a negligible compressibility and should not be underestimated in this context. E.g. the compressibility of water is as high as ca. 15% when water is used at 6000 bar (90,000 psi). A good historical example is again the development of the LDPE process in Great Britain and is illustrated by Fig. 6.3 (Ellis 2005). The reaction to form LDPE may get out of control with a fast runaway reaction leading to a decomposition of ethylene into carbon, hydrogen, and methane under exothermic conditions. Thus, the reactor temperature is exceeding the design limits and the pressure is also elevated beyond the yield strength of the material. Fortunately, the material used to build the autoclave reactor shown in Fig. 6.3 was ductile enough to remain in one piece. Further, the rupture followed the mechanical theory which is indicated by the longitudinal openings in the shell as a result of the higher tangential stresses prevalent in the wall. The essential step after such incidents is to understand the real root cause and drawing appropriate conclusions within an open-minded, multi-discipline expert team to avoid a repetition and improve the safe application of a high pressure process. It should further be considered that a sudden rupture of a high pressure system— as happened with the autoclave from Fig. 6.3—may cause a shock wave in the environment. In case of chemicals under high pressure, their properties may cause

6.1 Safety Considerations for High Pressure Equipment and Plants

229

Fig. 6.3 Left: autoclave reactor assembly after a decomposition incident, right: same body with longitudinal fractures within the ductile material

harm to the people around, damage the environment, or cause sudden subsequent reactions under ambient conditions (e.g. with oxygen). Finally, debris from breaking high pressure parts may also cause injuries or damage. All these potential risks can be managed and be avoided by appropriate material choice, by proving material properties along manufacturing, by well thought equipment design and by smart plant layout. Therefore, the aforementioned details about design materials shall be understood as lessons already learnt which brought high pressure process applications to a very safe level and form a solid, experienced basis for further robust and safe utilization.

6.1.1.2

Considerations for the Design of High Pressure Processes

During the effort to compile the Design Basis (DB) a number of questions arise that relate to the conceptual design of the process as well as the overall installation of a plant. Be it the capacity of the whole plant or operational data for relevant process steps and its equipments, the operating mode (batch, continuous), the availability, constraints or requirements for utilities (heating, cooling, compressed air, etc.), the raw material and product specification, etc. All these definitions and information shall be documented in the DB among other information like site conditions, environmental data, etc. The clarity provided through the DB to the team working on the project will reduce ambiguity and improve the quality of each disciplines’ output. The ideal tool in Basic Engineering (BE) is a process simulation which should reflect the logic of the Process Flow Diagram (PFD). This simulation is forming the

230

6 Safety and Control in High Pressure Plant Design and Operation

basis for the next engineering step which is to elaborate proper equipment specifications and a complete plant design. The output of such simulation, often covering different cases of operation, is the mass and energy balance and gives sufficient initial information to work out the equipment specifications and a first plant layout. Another deliverable of the Basic Engineering is the Piping and Instrumentation Diagram (P&ID) that shows all equipments of the plant e.g. apparatus, pumps, compressors, heat exchangers, piping, valves, instruments, etc. and indicates clear numbers attached to the items (tag numbers). All lines between equipments—which represent the connecting pipe work in principle—carry identifications for the pipe classes incl. nominal diameter, pressure, medium, material, insulation, etc. In addition the P&ID contains information about position and function of instruments (e.g. measuring devices, valves incl. function), control loops, safety devices, etc. which also carry an identification code. Further mechanical devices indicated may be strainers and filters, traps, sight glasses, etc. which are important elements of the plant function as well. For P&IDs again certain codes and standards exist in different countries of the world to use a common language between the different engineering disciplines contributing to a plant project [examples for codes are EN ISO 10628:2015 or ANSI/ISA 5.1-2022] (American National Standard 2009; Beuth publishing DIN 2015). The P&ID is also used at the end of the Basic Engineering to check for the safety of the plant and its operability in a theoretical investigation nowadays named HAZOP study—hazard and operability study (International Electrotechnical Commission 2016a). The HAZOP technique was initially developed at the famous chemical company ICI in the United Kingdom in the 1960th and gained wider acceptance in other chemical companies and further industries after the tragic Flixborough disaster on June 1, 1974. A strong driver for plant safety, hazard analyses and HAZOP studies was Trevor Kletz who was the safety advisor of ICI from 1968 to 1982 and refined the methodology of HAZOP throughout his long professional life. A lot of literature can be found from himself as well as about HAZOP (e.g. International Electrotechnical Commission (2016a)), thus only a few general remarks shall give a quick insight into the principle: Firstly, HAZOP is a team effort. Typically the team consists of an independent team leader knowing the technique and being supported by a secretary documenting the outcome of the sessions. Other participants should be members of the engineering team (design, project, process, instrument/controls, commissioning manager, R&D in case a new chemistry is involved). The size of the team should be appropriate (sufficient know-how, not too big) and an open-minded, frank and alert discussion must be achieved under the guidance of the chairman. Secondly, the basis for the HAZOP study should be formed by deliverables from the basic engineering, mainly the P&ID and the nodes within the diagram are in particular focus to discuss the desired functionality or design intent. However, particular emphasis should also be put on transient states i.e. start-up, shut-down or changes between different steady state operating modes. The learning from practical operations’ statistics is that emergency resulting from unforeseen transient events fortunately only stands for very little overall utilization time of a plant. However, it

6.1 Safety Considerations for High Pressure Equipment and Plants

231

causes more than 50% of the incidents. Thus, clear and complete practical instructions—which are thought through until a safe status of the asset is achieved—are paramount. Thirdly, a firm principle will be followed during HAZOP which is structured by a list of guide words theoretically examining the parameters of the process, resp. their deviations from the intended values, directions, sequence, etc. The outcome should be a comprehensive collection of qualitative parameter deviations from the intended values or function, their causes and effects resp. consequences and finally clear instructions for required actions. These instructions shall also be incorporated into the installation, operating and maintenance manuals to make sure the learnings of the HAZOP study are documented and carried on to the relevant stages of the project. A final advice should be mentioned based on a quote from Trevor Kletz: “People say that accidents are due to human error, which is like saying falls are due to gravity”. It is essential on every single level of an organization to apply the best skill and diligence. Organizational error on managerial level, engineering, planning or equipment material failure as well as non-conformance in installation, operation or maintenance can and must be avoided. Since organizations per se do not have a memory it is essential to care for proper documentation and training of the personnel to avoid any kind of repetition and turn failure into success by learning from it. This was achieved in the high pressure arena very often so far and will make future new applications also more safe.

6.1.1.3

Considerations for the Design of High Pressure Equipment

The remarks in the previous sub-chapter are valid for any kind of plant used in the process industries, while the design of the core equipment for high pressure plants has to be handled special. A number of parameters has to be taken into account and viewed differently. The main factor for dimensioning of the wall thickness is a combination of design pressure, equipment diameter (size), material properties (yield strength) and safety factor. It is eminent that the wall can withstand the static force and a fracture of the material is definitely avoided under any operational process condition. On the other hand, the number of cycles during the lifetime of the equipment i.e. the changes in process pressure due to batch operation, oscillation of pressure e.g. of a piston in a pump body, etc. also have to be mastered. Such operation can cause fatigue which is a slowly progressing failure pattern that has to be considered during design and engineering as well. Typical designs of low pressure vessels allow for many different shapes or even changing the shape along different process sections of one equipment. Further, large openings, side ports, etc. are possible for their designs. In contrast, under high pressure conditions the usual shape is a uniform cylinder and the diameter should be considered carefully. The inner diameter—in combination with the pressure—is strongly influencing the wall thickness and thus the weight of the equipment and obviously its associated material cost. As openings into the vessel (and connections

232

6 Safety and Control in High Pressure Plant Design and Operation

to pipe work) the front ends have to be employed which is often a challenge when processing solid matter under high pressure. E.g. in supercritical fluid plants which are often operating in a batch process with baskets, it is essential to utilize the largest possible portion of the inner diameter to keep the processing volume large and avoid lost circumferential space. This circumferential volume has to be built, filled with fluid under high pressure for every batch and thus reduces the economics of such process. In case the SFE can be better handled without a basket (e.g. coffee, rice) the autoclaves are designed with a higher L/D ratio in order to keep the wall thickness lower. Side ports should be avoided as they show stress peaks that can cause equipment failure later on. This is increasingly valid the higher the process pressure is in operation. Another strong differentiator between thin-walled low pressure vessels and thickwalled high pressure equipment is the stress distribution within the wall material itself (see Fig. 6.4). While the tangential stress is the highest in both cases, its distribution over the wall is more uniform in low pressure applications as the wall is much thinner. In contrast, the tangential stress in a high pressure vessel is the highest at the inside of the wall and reduces towards the outer wall. This basic principle forms the foundation of many practical design details when going higher in process pressure. It is eminent to control the stress level at the inner wall by the design which will be discussed later in this chapter. Further it is important to control the surface quality at the inner surface as little defects (marks, scratches, etc.) might act as starters for cracks, grow through the wall and thus damage the equipment. For a deeper understanding of high pressure vessel design the following Fig. 6.5 illustrates a generic differentiation of principle design areas. It should be noted that different models can be applied and for a detailed design the relevant codes have to be followed. Nevertheless, the diagram can be used to explain the design principles that prevail under various conditions. The diagram shows a practical approach to judge the degree of material utilization which can be derived from the ratio of the maximum pressure load of a vessel and the yield strength of the material used. On

Fig. 6.4 Distribution of tangential stress under low pressure (left) and high pressure (right) (Lütge 2011)

6.1 Safety Considerations for High Pressure Equipment and Plants

233

Fig. 6.5 Design ranges for high pressure vessels visualized by material utilization versus diameter ratio (Lütge 2011)

the other axis the diagram uses the diameter ratio between the outside diameter and the inner diameter of the part. The lower curve indicates the border of a fully elastic design which means that yielding starts at higher pressure values (or lower yield strength) following the condition of Tresca and Guest. The upper curve indicates the border of a fully plastic design beyond which the material will not withstand the operation acc. to the von Mises criterion and will be destroyed very soon (Buchter 1967). The diagram shows three different regimes of the diameter ratio. In Zone I—up to a diameter ratio of ca. 1.6—no yielding occurs and the elastic limit is always slightly above the plastic limit. The very left end close to a diameter ratio of ca. 1 can be read as a thin-walled vessel carrying a membrane stress only. Beyond the diameter ratio of 1.6 up to 4.1, i.e. in Zone II, the material must be pre-stressed in order to avoid yielding of the material. Thus, this area is mentioned as operating range for high pressure applications requiring certain design techniques which will be explained later in this chapter. Beyond a diameter ratio of 4.1, i.e. in Zone III, no practical design or operation is possible. The stress range grows above 2 times the yield value. However, pre-stressing technologies can theoretically (i.e. neglecting further practical effects) only compress the material up to the yield strength from the outside. Thus, no further improvement by pre-stressing can be achieved and this has

234

6 Safety and Control in High Pressure Plant Design and Operation

to be understood as the design limit. A further increase of the wall thickness will not prevent the equipment from rapid destruction. A question arising from the viewpoint of a chemical engineer out of Fig. 6.5 might be to visualize current high pressure processes used in various industries and their relative position in the diagram which is obviously dominated by mechanical theory. This exercise can be found in Fig. 6.6. We can realize that practical designs are nicely following the considerations explained earlier and the different processes are utilizing the chosen materials very well. The equipment designs for the processes mentioned remain below the plastic limit load in any case and those with a higher diameter ratio than 1.6 are all within the operating window oriented to the high side. This is economic design through high material utilization. Processes using liquid fluids such as water jet cutting (WJC, #4 and 4’) or high pressure processing (HPP #5) can be found at higher process pressures. In contrary, processes using a supercritical fluid (e.g. CO2 #1 or #2; ethylene #3) are found to the left side of the diagram at smaller diameter ratios. Finally it is interesting that the operating range for supercritical fluid processes is dominated by the production of LDPE, whereas SFE processes remain at lower operating pressures (and lower temperatures). Conventional SFE plants are mainly designed for pressures of 50 MPa and only a few extraction vessels have been designed and commissioned for higher pressures so far. Nevertheless, higher pressure can accelerate the process, reduce the equipment size at a given plant capacity and can be the appropriate tool to deeply extract further valuable substances from raw materials. This was practically shown for hops or algae extracting highly valuable ingredients for health improvement or even pharmaceutical applications. Therefore, a wider range of ultra-high pressure applications might be considered in new processes. From the standpoint of a mechanical engineer the technical limits are far beyond what was realized so far. Thus, the high pressure R&D community is encouraged to be courageous and leverage the benefit of higher pressures in process developments. It could be seen from Fig. 6.4 that the highest stress in a vessel wall operated under high pressure will be found at the inner bore. It is obvious that this stress level has to be primarily controlled to arrive within the operating window indicated in Fig. 6.5. In case a “normal equipment” is under no internal pressure, no stress pattern prevails within the wall and the stress at the inner bore will be zero. Therefore, in case fabrication techniques could allow for a negative stress level at the inner wall in case the vessel is without internal pressure, the principle of superposition would allow to reduce the stress peak under operating pressure. Such technologies applied to impose a compressive stress to the inner wall are called pre-stressing techniques and will be explained herewith. The principle examples are demonstrated in Fig. 6.7 and show the initial stress distribution within the wall without internal pressure. It is obvious that the negative stress (compressive stress) prevailing at the inner bore supports to reduce the stress peak under operating conditions and thus allows to operate at higher pressure compared to no pre-stressing. Another advantage of pre-stressing is the fact that starters of cracks within the layer of compressive stress will have a much reduced tendency to grow and propagate. Therefore, pre-stressing is also a means to improve the inherent equipment safety.

6.1 Safety Considerations for High Pressure Equipment and Plants

235

Fig. 6.6 Practical applications of industrial processes within the design ranges for high pressure vessels (Lütge 2011)

Fig. 6.7 Pre-stressing techniques for high pressure vessels (Lütge 2011)

The first technique for pre-stressing is the autofrettage. It can be accomplished by pressurizing the vessel beyond the yield limit in a controlled manner e.g. using a hydraulic medium inside. Upon pressurization of the medium a plastic deformation from the inside out will occur and can be registered at the outer wall similar to a tensioning test. The inside pressure is the variable defining the stress level within the wall and the strain or elongation of the outside dimension is following as the resulting value or indicator. For a certain period of pressurization the strain follows the

236

6 Safety and Control in High Pressure Plant Design and Operation

linear elastic behavior according to Hook’s law. Upon reaching the yield strength the material will start to deform plastically and the strain follows a curve, thus allowing to impose a defined residual elongation onto the material after depressurization. At the end of this fabrication procedure the outside wall section will be—practically speaking—“too small” for the inner section and build-up a compressive stress. Since the stress pattern is achieved within a single layer of material the process is called auto-frettage which translates into self-shrinking. The process is widely applied. The next technique possible is the so-called shrink fit. At least two different cylinders are fitted into each other with a slightly larger outside diameter of the inner piece compared to the bore of the next larger one. Fabrication is mainly done by temperature difference cooling the inner piece down and heating the outer part up. One advantage is that different materials can be applied for the different layers, i.e. a corrosion resistant inner piece can be combined with another high strength carbon steel. However, for larger high pressure vessels a number of disadvantages during fabrication have to be considered. The forgings are typically non-standard items and require a long lead time at the steel mill. Thereafter, very precise machining to very accurate geometric dimensioning and tolerances and to a good surface quality is required both at the outside of the inner part and the inside of the outer part. In case a long vessel has to be assembled (e.g. 3 m final length) the cooled inner piece has to be put into a slightly larger bore of the hot outer piece. The heavy part has to be maneuvered very precisely from the double height compared to the final length. In case the cooled inner part gets in contact to the heated outer metal, the pieces may get stuck too early and the process has to be aborted with the parts lost. Therefore, mostly smaller/shorter parts are being pre-stressed by shrink-fit e.g. for pump heads or compressor parts. The final pre-stressing method is a multi-layer design composed of wire-winding (or plate winding). Only the inner layer is a precisely machined cylinder onto which a wire or corrugated band will be wound. The pre-tensioning forces can be controlled by the tension in the wire while winding it to the cylinder and also the inner diameter of the cylinder is a good indicator of the pre-tensioning achieved. Wire winding is a time consuming effort, however, the wire is more a standard item to be purchased. Therefore, different vessel volumes for a type series can be obtained by varying the diameter and the length of the cylinder and winding the same wire to the outside. Thus, lead times of such assembly may be shorter than for shrink-fit vessels. All three pre-stressing techniques explained before are common for the processes 3–5 mentioned in Fig. 6.6. Modern high pressure equipment for supercritical fluid extraction applications or process steps at similar pressure regimes (refer to #1 or 2 in Fig. 6.6) is mostly fabricated as welded designs from an assembly of hollowforged pieces or from high-strength steel avoiding welding completely. Through this fabrication principle welding is reduced to high quality narrow gap welding on the circumference which can be automated due to clearly defined geometry. Longitudinal welding—which normally reduces the lifetime due to fatigue issues—should be avoided for equipment at elevated pressures nowadays. Previous designs also used welded assemblies from rolled steel plates with a longitudinal weld or layered plate vessels with several longitudinal welding seams. However, due to reduced fatigue

6.1 Safety Considerations for High Pressure Equipment and Plants

237

characteristics and difficult inspection by non-destructive examination these designs are uncommon in modern plants which enhances the process and equipment safety. Modern design for all manufactured equipment or plants also includes paperwork and documentation from the engineering, design and upfront theoretical considerations. By the stringent requirements of the applied codes and standards a high safety level is safeguarded and maintained. These engineering deliverables are the first documents in the manufacturing sequence being registered in a comprehensive Quality Control plan (QC plan). Through these documents a clear practice is applied to prove the steps executed during design e.g. in form of a manufacturer’s design report. Part of this report is e.g. a specification, a detailed drawing and the calculation and dimensioning of the vessel which is often performed as a stress analysis by FEM (finite element analysis). Figure 6.8 may be understood as an example for such FEM analysis carried out. The picture shows a bottom part of an extraction vessel with the lower lid held to the vessel body by a clamp closure for quick emptying of the material filled inside the vessel. Acc. to the code requirements all documents have to be compiled and submitted for approval to a third party inspector prior to procurement of materials to build the equipment. This hold point is important to make sure changes required from inspecting the documentation will be incorporated. Only after the engineering documentation is fully acknowledged by signature and stamp obtained from the notified body or third party inspector it is released for fabrication. Therefore, already during design a high degree of diligence and strict methodology is making sure, high pressure equipment is fabricated to the highest safety standard. Fig. 6.8 Example for an FEM-analysis of a vessel section forming part of the design documentation submitted for approval to third party inspection (Lütge 2011)

238

6.1.1.4

6 Safety and Control in High Pressure Plant Design and Operation

Considerations for Fabrication and Testing of High Pressure Equipment

Upon written approval of the design documented through drawings, parts lists and stress calculations the fabrication of high pressure equipment commences. It is paramount to specify the materials being needed for fabrication in high detail and incorporate necessary QA measures, hold points or third party inspections. It is advised to check the fabrication of raw materials in the mill or the forging shop and also have detailed documentation of heat treatment procedures done. Further, destructive tests and non-destructive tests e.g. through visual inspection accompany the manufacturing sequence (also refer to Table 6.1). The material incl. its individually hard marked identification will be finally documented in the mill certificate. Upon receipt of the raw material piece and the accompanying material certificate in the fabrication workshop a positive material identification (PMI) should be carried out to independently witness that material and mill certificate are associated. Thereafter, production engineering shall define the subsequent fabrication details which may consist of mechanical manufacturing (i.e. turning, milling, drilling, etc.) as well as welding processes in case weldable material is being used. For all welding strictly defined procedures apply which are elaborated and supervised by a qualified and certified welding engineer and proper welding supervision. Important shop documents are Welding Procedure Specifications (WPS) giving detailed written instructions for the joint to be produced. Also instructions for post-weld operations such as heat treatment are given. A complementing office document is the Procedure Qualification Records (PQR) which is a record of a test weld performed and tested to ensure the weld can be expected as being fit for the intended application. Furthermore, all welders practically contributing to the manufacturing sequence have to undergo Welder Qualification Tests resp. Welder Performance Qualifications (WQT/ WPQ) acc. to a firm qualification protocol and documentation. The quality of the welding seams has to be inspected, verified by non-destructive testing, and must be documented. Prior to or during assembly of the parts making up a pressure vessel (e.g. body, lids, clamps, bolts, nuts, etc.) it has to be checked that the parts are acc. to the drawings (i.e. material double-check, dimensional check, surface quality check, etc.), in good alignment and can be maneuvered smoothly. Under the exposure of a commercial operation the high pressure equipment has to prove its robustness which can be assessed early on. As a final step of fabrication the pressure test is carried out. This step is defined by the code and requires a certain over-pressure compared to the design pressure to make sure the equipment will withstand the operating conditions. Upon the pressure test passed a hard stamping applies to the name plate of the vessel indicating the relevant code and an individual number of the equipment that will be registered with the responsible authority. As a practical remark it is advised to avoid any scratches or marks on the inside walls of a high pressure equipment. As explained earlier the inner wall is carrying the highest stress level and thus a very good surface finish is the best support of a long equipment life-time. Further, moving parts shall not chafe or scratch on another

6.1 Safety Considerations for High Pressure Equipment and Plants

239

part as fretting also will damage the parts and reduce life-time in an unnecessary manner.

6.1.1.5

Considerations for Lay-Out and Assembly of High Pressure Plants

Most supercritical fluid extraction plants built so far are operated to process solid materials. These plants follow the general lay-out as mentioned in Chap. 5. Nevertheless, a few additional safety considerations are helpful to ensure smooth operation and longevity of the asset already incorporated from the planning and engineering phase. The lay-out of the plant should consider different sections into which the operation can be divided. Such sections may be: • the main process section (e.g. extraction, separation), • the fluids supply section (i.e. supercritical fluid, entrainers), • the utility section (i.e. heating and cooling supply towards the main process section), • raw material handling area (i.e. machinery for batch preparation such as sorting, grinding, compacting, etc.) and storage, • product handling area (intermediate storage, standardization, dosing, packaging, etc.), • workshop area to handle small mechanical tasks on equipment apart from the plant. The break-down into such plant sections should reflect that the operation of the high pressure plant core should not be influenced by other process steps e.g. handling raw materials or products. Further, it can be foreseen to install the different sections acc. to different safety requirements such as explosion proof design, clean room, etc. depending on the particular industry needs. Sufficient space around and clearance above high pressure equipment should be foreseen to be able to handle raw materials (e.g. big bags, baskets, etc.), to execute maintenance at the high pressure equipment or the piping and to elevate materials from one plant level to another. All such operations should not endanger the high pressure equipment or the pipe-work to collisions, twists or deformations of any kind. Also operating personnel should have sufficient space to be working in the plant without too many constraints. Further, it needs to be considered that the preparation of the raw material may cause noise, dust, etc. that should not influence the high pressure plant and keep the operation area clean. In terms of the assembly of high pressure equipment and its interconnecting piping it is a general principle that all parts need to be connected with the best workmanship and without deformation or outside stress. Forcing or squeezing parts of the plant into their position must be avoided by all means. It is a dangerous act itself and may cause additional forces or stress to the assembly under operation (using elevated pressure and temperature) and be a precursor for consequential damage.

240

6 Safety and Control in High Pressure Plant Design and Operation

Some raw materials tend to change their characteristics under a process using supercritical fluids that may cause difficulty with or danger to the equipment. Such property changes might be a compacting of the material when pressurized from one side only. To avoid such compacting it may be advised to install a by-pass at the extractors that allows to pressurize the vessel volumes from both ends. During startup and pressurization the by-pass may be closed slowly and in a controlled manner, however, a slight elongation of the high pressure process may be accepted. In case no by-pass is installed, channeling may occur or—in the worst case—the basket carrying the material inside the extractor may be damaged possibly causing also damage to the inside wall of the equipment. Another feature could be a density change of the raw material while being exposed to a fluid under high pressure, i.e. a swelling and thus a reduction of bulk density and an increase of material volume. In case the basket is fully loaded with bulky material the swelling may build up additional stress to the equipment and cause deformation. Any such additional stress to the equipment must also be avoided. For large industrial high pressure systems which are operated at even higher process pressure than conventional SFE conditions further parameters have to be considered, engineered and converted into proper design of the plant. Such influences may be a pulsation and/or vibration induced to the high pressure system by large compressors elevating the fluid pressure and feeding the fluid into the system. Such machines are typically piston compressors that deliver the fluid “batch-wise” and induce pulsation within the fluid as well as a vibration from the rotation of machine parts of the compressor. All such influences can be handled through proper vibration analysis and can be incorporated within the engineering phase. A sufficiently large number of high pressure systems (e.g. for LDPE production) can prove the concept successful and be taken as safe references. The same may apply for large systems that do not necessarily allow for or require indoor installation. Most SFE plants are operated to process sensitive natural raw materials and are built inside an appropriate building; these are typically small installations compared to many other chemical or petrochemical plants. However, it may also be considered to design parts of the plant for outdoor installation which is common in many industries and foresee the section operated under high pressure being installed in a confined space or bunker area. Thus, it can be ensured that in case of a safety incident the surrounding area should not be affected too much and damage to personnel, assets around or the environment is avoided. As soon as a high pressure plant is assembled from individual, pressure tested equipments and the piping is completed incl. non-destructive testing of welding seams in the pipe-work, another test is required prior to commissioning and start-up. This test (or series of tests) reflects the integrity of the assembly and focuses on the tightness of the complete plant. Depending on the plant sections and pressure levels applied it may be advised to perform the tightness test step-wise and investigate and record the findings. In case of deficiencies it is not recommended to improve the status while the equipment is still pressurized. Rather, it is another basic principle to work on high pressure equipment only as long as it is without internal pressure.

6.1 Safety Considerations for High Pressure Equipment and Plants

241

6.1.2 Diligence in Operation and Care for Longevity As soon as a high pressure plant is tested for tightness the commissioning can commence with cleaning/flushing the inside, loop checks, functional tests and startup procedures. Typically, these details are compiled and comprehensively described in the operating manual. This manual has to be used as a fundamental document for training of operators, for operating routines and also as reference for any maintenance activity. Only a few remarks shall be given for a safe and a long-life operation of a high pressure asset. These general remarks will only highlight some important topics and shall not be understood as a complete list. In any case the advice of the process owner, equipment manufacturer or other experienced advisors shall be sought: • Cleanliness around and within the plant is an essential element for safety and longevity—this relates particularly to SFE plants which are operated to process solid raw materials. Dust and dirt are to be avoided at functional surfaces (e.g. sealings, instruments). The same applies to any sign of corrosion either from the outside or from the inside. As corrosion (e.g. pitting, stress corrosion cracking, etc.) will impact the wall thickness, the integrity of the equipment will be compromised which shall be avoided. • Instruments have to be kept clean and intact, in particular those indicating pressure. A redundant installation may be advised to double-check internal pressure prior to opening any high pressure equipment. By design or instrument choice clogging of connecting lines to instruments must be avoided to ensure correct and precise indication of the internal status. • Filters should be installed at sufficient places—in particular when natural products are being processed or when solids may be carried within the fluid circuit and be transported into machinery (pumps, compressors). Filters have to be checked regularly and emptied/cleaned to keep the process in function and control the pressure drop. • Opening of equipment (e.g. for batch-change of a basket, removal of extract, etc.) shall only be performed when the fluid pressure is fully released and the equipment is definitely at ambient pressure. It shall never be allowed to open a high pressure equipment with high external or even with brute force. Either the device supplied by the experienced high pressure equipment manufacturer or the installed pneumatic/hydraulic systems shall be used. In case of an unclear plant status no maneuver is recommended or allowed. • Safety devices have to be kept intact and shall be checked for their proper function regularly. They are essential elements for the safety of the plant, the operators and the environment and shall never be compromised. A controlled release of the fluid inventory within the plant is absolutely essential and shall be led to the outside, resp. into a flare (in such case it is essential to have the piping designed acc. to the peak flow rate of a flare). • Operational records shall be made in addition to the automatic recordings done within the control system. It is advised to record the pressure cycles and document

242

6 Safety and Control in High Pressure Plant Design and Operation

the utilization e.g. through the batches. Pressure vessels are often rated for a certain number of pressurizations and need to undergo a repetitive inspection as soon as the allowed number of cycles is consumed or exhausted. • Spare parts need to be readily available, be it during assembly, commissioning or the commercial operation. Status of the spare parts should be checked prior to installation. During operation of a high pressure plant any unusual noise, vibration, or new/ abnormal plant behavior should be taken as a reason to check the installation in more detail. A vibration at high pressure equipment can cause failure through fatigue and thus reduce the life-time of equipment drastically. Also noise can be understood as an indicator for vibration and should be checked thoroughly. Finally, the operating manual also contains a section giving instructions for maintenance or repair. However, when a broader repair or even a change of the plant set-up is required and performed, it is strictly advised to have a proper documentation of such repair or changes following the prevailing codes and standards. The strict rules and regulations for high pressure apply in any case and may require a new visitation and inspection by the notified body prior to re-start of the plant.

6.1.3 Externally Induced Error or Defects Although the modern, high safety standards with respect to material choice, diligent engineering, care in fabrication and testing etc. allow for a wide variety of high pressure applications in many industries, a couple of things remain important, shall be mentioned in this context and must be taken into practice very seriously. The personnel involved in engineering, fabrication or operation of any high pressure plant must never lose the respect for the high pressure itself and its inherent power. Fortunately, today we do only very seldom see or hear about reported incidents with high pressure plants. However, this cannot be understood as an indicator that high pressure is easy or simple to handle and must never end in a superficial attitude in any activity around high pressure processes or its applications. Rather, it is the result of long-lasting and consequent learning and improvements of the rules and regulations that apply to high pressure. This statement is generally valid in any scale of high pressure application—be it small in an R&D lab or in a large commercial plant. High pressure should only be handled in properly designed and fabricated equipment and operated with highest possible care and diligence. Further, any impairment to a single high pressure equipment or a plant needs to be recorded and inspected in detail. E.g. a dent from an outside collision of a heavy part with a pressure equipment needs to be taken seriously and be checked with appropriate expertise. In case such collision happens under pressurized operation, a shut-down of the plant is advised and economical considerations must not supersede such decision.

6.2 Control of High Pressure Plants

243

6.2 Control of High Pressure Plants Before turning to a more particular focus of instrumentation and control of high pressure plants some high level principles of functional safety need to be mentioned.

6.2.1 Functional Safety To achieve the target of functional plant safety from the disciplines of electrical and instrumentation engineering mainly the following internationally applied standards are relevant: IEC 61508 is an international standard published by the International Electrotechnical Commission consisting of methods on how to apply, design, deploy and maintain automatic protection systems called safety-related systems. IEC 61508 is a generic functional safety standard applicable to all kinds of industries. It defines functional safety as: “part of the overall safety relating to the Equipment Under Control (EUC) and the EUC control system which depends on the correct functioning of the safety-related systems, other technology safety-related systems and external risk reduction facilities”. The fundamental concept is that any safety-related system must work correctly or fail in a predictable (safe) way. The safety life cycle has 16 phases which roughly can be divided into three groups as are Analysis (Part 1–5), Realization (Part 6–13), and Operation (Part 14–16). Central to the standard are the concepts of probabilistic risk for each safety function. The risk is a function of frequency (or likelihood) of the hazardous event and the event consequence severity. The risk is reduced to a tolerable level by applying safety functions which may consist of electrical, electronic, programmable electronic safety-related systems (E/E/PES), associated mechanical devices, or other technologies. Many requirements apply to all technologies, but there is strong emphasis on programmable electronics. In addition to the aforementioned generic standard for functional safety the standard IEC 61511 is a technical standard which sets out good engineering practices that ensure the safety of an industrial process through the use of instrumentation. Such systems are referred to as Safety Instrumented Systems (SIS; see below). The title of the standard is “Functional safety—Safety instrumented systems for the process industry sector” and clearly focuses on the applications in the process industries. This standard is complemented in the US by ANSI/ISA 84.00.01-2004 which primarily mirrors IEC 61511 in content. It should be noted, however, that IEC 61511 is not harmonized under any directive of the European Commission. Thus, local regulations may apply. IEC 61511 covers the design and management requirements for SISs throughout the entire safety life cycle. Its scope includes: initial concept, design, implementation, operation, and maintenance through to decommissioning. It starts in the earliest

244

6 Safety and Control in High Pressure Plant Design and Operation

phase of a project and continues through start-up. It contains sections that cover modifications that come along later, along with maintenance activities and the eventual decommissioning activities. A Safety Instrumented System (SIS) is composed of a separate and independent combination of sensors, logic solvers, final elements, and support systems that are designed and managed to achieve a specified Safety Integrity Level (SIL). A SIS may implement one or more safety instrumented functions (SIFs), which are designed and implemented to address a specific process hazard or hazardous event. The SIS management system should define how an owner/operator intends to assess, design, engineer, verify, install, commission, validate, operate, maintain, and continuously improve their SIS. The essential roles of the various personnel assigned responsibility for the SIS should be defined and procedures developed, as necessary, to support the consistent execution of their responsibilities. IEC 61511 references IEC 61508 (the master standard) for many items such as manufacturers of hardware and instruments and thus IEC 61511 cannot be fully implemented without reference to IEC 61508. IEC 61511 rather is to be understood as the process industry implementation of IEC 61508 (International Electrotechnical Commission 2010; International Electrotechnical Commission 2016b).

6.2.2 Considerations for Engineering of Instrumentation and Control System As a result of the basic engineering, which is to be complemented by the HAZOP study, the detailed engineering of instrumentation hardware and the control system commences. From the practical view of plant safety some general considerations shall be discussed within this chapter. The most important parameters that need to be monitored in high pressure plants are the pressure and—due to the importance of the material properties which reduce with increasing temperature—also the temperature inside the pressure bearing equipment. Since many high pressure plants are operated in a batch process and high pressure equipment needs to be opened and closed regularly, the safety of operators around such vessels is paramount. No opening shall be possible as long as there is residual pressure inside the equipment. Most high pressure plants are equipped with highly sophisticated process control systems. However, at low pressures near the ambient conditions the electrical pressure transducer that is used to monitor the high process pressure might be an inadequate means to judge whether the vessel is non-pressurized. A big advantage in terms of safety are mechanical gauges which will also operate and indicate the pressure of the system section in case no electrical supply is available. Therefore, it is for safety reasons highly recommended to install mechanical pressure gauges at relevant places within the plant to indicate the pressure within the system. These places should also include heat exchangers or piping

6.2 Control of High Pressure Plants

245

sections that might suffer clogging and allow to read and understand the status of the inside at all times (i.e. under operation and during stand-still). It is a very common design principle for a pressure vessel that needs to be opened regularly, to use the internal pressure as a support to lock the complete closure system. I.e. the cover or lid of the pressure vessel will lift slightly when pressurized and will mechanically lock the lid to the closure device. As long as the vessel is under residual pressure the lid will not return into the initial position and the closure system cannot be moved. In addition, it is often realized to interlock the closure system within the overall control system. Thus it can be ensured that the plant will only be started-up as soon as all elements of the closure system are fully in place which is double-checked by proximity switches. In smaller pilot systems such interlock can be realized by safety valves incorporated into the manual closure devices, i.e. the internal pressure can only be built-up after properly closing the system. It is needless to say that all instruments used in a high pressure plant shall carry a design approval and it should also be considered during detail engineering at which points redundant measurements or duplicate installation is required. As soon as inflammable or explosive components are used under high pressure it is also a strict requirement to design the instruments and the electrical system in ex proof version. To protect the plant against over-pressure during regular operation, various solutions are utilized and practically proven. On the fluid supply side interlocks for the entrance of high pressure medium can be implemented allowing pumps or compressors only to operate under pre-defined conditions. Further, on the fluid discharge a controlled protection against over-pressure is mostly realized through the high pressure control valve. For fast going chemical reactions it may be advised to install additional interlocks with safety valves to ensure very quick depressurization of plant sections. In such case a sophisticated control system is recommended to allow for extremely fast response times and very quick or instant interaction with the emergency shut-down (ESD). The important safety devices for quick pressure relief are either burst discs or safety valves. Burst discs are typically metal discs breaking open at a defined pressure and allow the over-pressure to be released through a header system to the environment. Such system is simple and easy to install and very often used in smaller high pressure systems such as lab-scale or pilot plants. However, also in big industrial systems bursting discs may be found. As soon as a bursting disc broke open the holder needs to be dis-assembled, inspected and cleaned and a new metal disc be installed to get the device back into function. Also the root cause for the bursting of the disc should be analyzed in detail, be fully understood and through appropriate measures avoided in future. Another pressure relief device is a safety valve that typically has a spring loaded opening part and will only open as soon as a pre-defined pressure level is reached. It should be mentioned that the design of safety valves needs to sufficiently reflect the real gas behavior for applications of supercritical fluids under high pressure. Most sizing literature assumes ideal gas characteristics and thus may result in smaller designs of the opening port than required in reality (Beuth publishing DIN 2016). Therefore, high diligence for sizing of safety valves

246

6 Safety and Control in High Pressure Plant Design and Operation

is essential and the professional advice of qualified suppliers should be sought. Also for safety valves it is essential to maintain them properly acc. to the supplier manual and clean them after they were activated to make sure they will safeguard the plant to their best functionality. Last but not least it should be mentioned that gas detection systems should be foreseen to monitor for leakages or for accumulation of gas within the plant area. In case of light flammable gas leaking out of the high pressure system the overall plant safety may be endangered. Hot surfaces and oxygen may trigger rapid reactions and thus need to be monitored and avoided. On the other hand, in plants using CO2 (or similar components heavier than air) as fluid for SFE, the accumulation of CO2 must also be avoided by a proper ventilation system. Operators entering a plant section with accumulated CO2 may suffer suffocation which also has to be prevented by diligence in engineering and plant design.

6.3 Engineering Ethics A final and concluding remark shall be made to this important chapter of plant safety in view of engineering practice in general. As engineers we are all too often working to improve equipments or processes, we are going beyond practically proven limits and we are discovering new technical ground. These working principles are essential elements of engineering, nurtured by curiosity and creativity of engineers and propelled the modern industry and society to where we stand today. Nevertheless, engineers are often lacking complete or detailed enough information, must decide on limited information available and thus manage the unknown. Therefore, ethical behavior also has to be considered on the individual level of an engineer, on team or organizational level or even on society level in case engineers’ opinion is turned into a decision for the society. More detailed literature is published on this topic and only a few short remarks may be given here (Fleddermann 2012). The most important duty of an engineering practitioner is to protect the safety and well-being of the public. Therefore, professional engineering societies incorporated in their own codes also a code of ethics. This concerns to all engineering disciplines involved in a project as mentioned before, i.e. chemical engineers, civil engineers, electrical engineers, mechanical engineers, etc. and can be found within the codes of all relevant engineering societies. This approach is comparable to the ethical conduct of physicians that is firmly anchored within medical practice and widely known as Hippocratic Oath. In the context of plant safety any and all engineering tasks have to be executed to make the product (i.e. process, equipment, plant) as safe as reasonably possible and make safety an integral part of the engineering effort. However, some remaining risk needs to be handled and sufficient diligence has to be used to manage that risk. As a

References

247

basic approach to handling the risk four criteria may be used helping to ensure safe engineering for high pressure applications: 1. 2. 3. 4.

Is it compliant with applicable law (as a minimum requirement)? Does it follow accepted engineering practice? Is the design the safest available design (i.e. did we explore alternatives)? Which potential misuses can be foreseen and must be avoided by design (i.e. refer to the principles of a HAZOP study)?

Finally and as explained earlier, a rigorous testing even under harsh conditions in the final working environment shall be undergone for any high pressure process or equipment as long as no reference case is available. Only by going through such practical examination is it possible to understand causes of failure, failure modes and improve the product prior to a broader application of a new high pressure technology. As the final word for sincere and earnest consideration the Code of Ethics for Engineers from the preamble of the National Society of Professional Engineers shall be quoted: Engineering is an important and learned profession. As a member of this profession, engineers are expected to exhibit the highest standards of honesty and integrity. Engineering has a direct and vital impact on the quality of life for all people. Accordingly, the services provided by engineers require honesty, impartiality, fairness, and equity and must be dedicated to the protection of the public health, safety, and welfare. Engineers must perform under a standard of professional behavior that requires adherence to the highest principles of ethical conduct.

References American National Standard (2009) ANSI/ISA-5.1-2009 instrumentation symbols and identification Beuth publishing DIN (2015) DIN EN ISO 10628-1:2015-04, Diagrams for the chemical and petrochemical industry—Part 1. Beuth Verlag GmbH Beuth publishing DIN (2016) DIN EN ISO 4126-1:2016–12, Safety devices for protection against excessive pressure—Part 1. Beuth Verlag GmbH Buchter HH (1967) Apparate und Armaturen der chemischen Hochdrucktechnik. Soc Chemical Industry 14 Belgrave Square London SW1X 8PS, England, London Ellis ER (2005) Polythene came from Cheshire: the discovery, early development and production by the high-pressure process, 1933–1971. E.R. Ellis, Chester EU-OSHA (2014) Directive 2014/68/EU: Pressure Equipment Directive for the European Union European Standard (2021) EN 13445-1:2021—Unfired pressure vessels—Part 1: General European Standard (2012) EN 13480-1:2012—metallic industrial piping—Part 1: General Fleddermann CB (2012) Engineering ethics (4e). Pearson Education Inc

248

6 Safety and Control in High Pressure Plant Design and Operation

International Electrotechnical Commission (2010) IEC 61508-1:2010 Functional safety of electrical/electronic/programmable electronic safety-related systems—Part 1–7. https://webstore. iec.ch/publication/5515. Accessed 25 Oct 2022 International Electrotechnical Commission (2016a) IEC 61882:2016a—Hazard and operability studies (HAZOP studies)—application guide. In: ITeh Stand. Store. https://standards.iteh. ai/catalog/standards/iec/a77032db-bbf0-4270-9eb3-4ee5863317ee/iec-61882-2016a. Accessed 25 Oct 2022 International Electrotechnical Commission (2016b) IEC 61511–1:2016/AMD1:2017 Functional safety—Safety instrumented systems for the process industry sector. https://webstore.iec.ch/ publication/32443. Accessed 25 Oct 2022 Lütge C (2011) Design and manufacturing of high pressure equipment, Annual assembly of the high pressure chapter of GVC (German Association of Chemical Engineers) Peters PDT, Ritter ER (2019) ASME boiler and pressure vessel code Section VIII, Division 3: example problem manual. American Society of Mechanical Engineers Times S to TNY (1905) Grover Shoe Factory disaster. N. Y. Times Wikipedia UA (2022) Accident at Darlington on 2nd February 1850. Steam locomotive boiler explosion, York Newcastle & Berwick Railway, Darlington, 2nd February 1850. http://www.rai lwaysarchive.co.uk/eventsummary.php?eventID=1730

Chapter 7

Conclusion and Future Perspectives

Abstract In this chapter summary on history, thermodynamic and mass transfer data, early applications, advantages of application of supercritical fluids on industrial scale, extraction processes of solids and liquids on industrial scale, production of polymer particles and foams, application of supercritical fluids as reactants and as solvent for chemical and biochemical reaction are presented. Future expected developments in the use of supercritical fluids as processing media for extraction of solids and liquids, for impregnation of different substrates and as solvents and reactants in chemical and biochemical reaction are discussed. Advantages using CO2 as environmental friendly and the second cheapest solvent with the new challenges and opportunities are presented. Keywords High pressure · Supercritical · Subcritical · Fluid · Application · Industrial · Future direction

Sub- and supercritical fluids are in nature present forever. The pressure in our nearest star—the Sun—is 1 × 1011 bars (Asimov 1977) while the temperature at the surface of the Sun is about 5600 °C. The temperature rises from the surface of the Sun inward towards the very hot center of the Sun where it reaches about 1.5 × 107 °C (IPAC 2020). These values are by far beyond the critical values. Supercritical fluids are present in the depth of oceans, as well as in the atmosphere of several planets. More examples could be found. In the history of man, the use of high pressure technology was invented relatively early. In Roman times the wooden pumps were used to supply water from 100 m deep fountains. Industrial application of high pressure started several hundred years later when high pressure steam engines were developed by James Watt around 1785. Through this high pressure invention and its applications for transportation, for heavy working machinery as well as for heating also the use of electricity was pushed forward making our lives better. Nearly 100 years later—in year 1869 the thermodynamic fundamentals of supercritical fluids were reported by Thomas Andrews.

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Ž. Knez and C. Lütge, Product, Process and Plant Design Using Subcritical and Supercritical Fluids for Industrial Application, https://doi.org/10.1007/978-3-031-34636-1_7

249

250

7 Conclusion and Future Perspectives

One of the most important inventions for the application of high pressure in industry was the synthesis of ammonia by Haber and Bosch in 1918. Through large scale industrial production of fertilizer hunger crisis could be reduced since then and the nourishment of almost 10 billion people on earth is secured today. Another large scale industrial application of a reaction within a SCF—developed ca. 30 years later—is the polymerization of low density polyethylene (LDPE) from ethylene under pressures of 1500 bar (autoclave process) or 3000 bar (tubular process). The unique properties of LDPE made the product a highly valuable material which still enjoys a robust growth in the polymers industry. Later, as a special process of high pressure technologies, using subcritical and supercritical fluids gave the possibility to develop new products with special characteristics or to design new processes, which are environmentally friendly and sustainable. By use of sub- or supercritical fluids as a processing media the legal limitations for lower solvent residues and restrictions on use of conventional solvents in several processes can be achieved. Thus, modern products with obvious health benefits can be manufactured. These developments started as early as in the 1980s when green chemistry or sustainability were no words commonly used to describe the particular characteristics of gentle processing under high pressure. To date we are finding high pressure applications in a wide array of process industries and many other processes have been developed so far which could break into further industries that might also benefit from high pressure applications. As shown in previous chapters one of the major process benefits is derived from the thermo-physical properties of SCFs, high diffusivity, low viscosity, high density, dielectric constant of SCF, which can be fine-tuned by changes of operating pressure and/or temperature. Nowadays supercritical fluids are used in several processes which are developed to industrial/commercial scale in chemical and biochemical, pharmaceuticals, nutraceuticals, food & beverage, or textile industries or for cleaning of sensitive parts used in high-tech machinery. One very promising application of high pressure technology is the extraction of valuable compounds from plant materials and their formulation “in situ” into products with specific, customer designed properties. Particle formation using supercritical fluids can help to overcome the drawbacks of conventional particle size reduction processes. Because of their unique thermo-dynamic and fluiddynamic properties, dense gases in subcritical or supercritical state can also be used for impregnation of solid particles, particle coating, foaming etc. Some biochemical and chemical reactions performed in supercritical fluids have already been implemented to industrial scale to obtain products with high added value, while the use of supercritical fluids as heat carriers is a newly emerging field.

7.1 Future Directions/Perspectives

251

7.1 Future Directions/Perspectives 7.1.1 Extraction Processing of products of natural origin with supercritical fluids has been an extensive area of research during the past decades. CO2 and water have been the SCFs of choice so far as they offer a number of advantages. Nowadays, the trends shift towards the use of unconventional subcritical or supercritical fluids such as sulfur hexafluoride and argon. Propane, as a very selective solvent for subcritical fluid extraction has recently been intensively applied. The disadvantage is that propane is highly flammable and this is its main drawback in comparison to the use of CO2 as processing media. However, comparing the process conditions from LDPE polymerization at 3000 bar and 300 °C, the way forward seems clear and simple, while the critical point of raw propane is much lower, with a critical pressure of 42.5 bar and a critical temperature of 96.74 °C. Since industrial LDPE plants have been in operation for more than 70 years, plant safety has advanced considerably, and the chemical reactivity of propane is lower than that of ethylene. It is therefore essential to bear in mind that propane is a medium for industrial high-pressure processes. Subcritical and supercritical fluid based technologies offer important advantages over organic solvent technology, such as ecological friendliness and simple product fractionation. Extraction of hop components, decaffeination of tea and coffee are the highest volume scale extraction processes using sub- or supercritical solvents, which are realized on industrial scale. Also for the extraction of spices for food industry and natural substances for use in cosmetics several industrial plants are in operation. Less industrial units for separation of components from liquid mixtures using subor supercritical fluids are in operation worldwide. The main advantages of using supercritical fluids for isolation of natural products are solvent free products, relatively easy fractionation of components in extract, no co-products, low temperature in separation process. In addition, the processes can easily be linked with direct micronization and crystallization from supercritical fluid. Another important advantage of the use of supercritical fluids as extraction solvent is the selective extraction of components or fractionation of total extracts. This is possible by changing the process parameters—pressure and temperature—in the extraction process or stepwise separation of components from the total extract obtained at process parameters where all desired substances are extracted. In the separation stage with different operating pressures and temperatures in individual separators very efficient fractionation of total extracts can be achieved. The other possibility for selective extraction is the use of different gases for isolation/fractionation of components, where beside the mostly used gas for sub- or supercritical extraction, i.e. carbon dioxide, also sub- or supercritical solvents with different properties such as e.g. polarity are used. Sub- and supercritical CO2 and supercritical H2 O

252

7 Conclusion and Future Perspectives

are non-carcinogenic, non-toxic, non-mutagenic, non-flammable and thermodynamically stable. In addition, CO2 does not usually oxidize substrates and products, allowing the process to be operated at low temperatures. Water is at the moment the cheapest solvent and several substances are highly soluble in water. Therefore, more and more research on the use of sub- or supercritical water for isolation and fractionation of substances is under investigation. Even along all benefits of using subcritical or supercritical fluids as extraction media, there is a limitation of further application of extracts obtained by high pressure processes, which is the price of the product in comparison with conventionally obtained products. Processing cost may be higher and thus limit a wider use of sub- or supercritical fluid processes. The legal limitations of solvent residues and isolation/ fractionation of special components from total extracts in combination with different formulation may propel future applications as the requirements of sustainability also demand for new technologies. Development of SFE plant capacities by region or the number of compressed fluids extraction plants by region show that the number of extraction units’ increase with time and the most “dense” areas are Europe and Asia where increased number of installed new SCF extraction units also in recent time could be observed (Lütge and Schuetz 2007). It is also evident that there is no clear trend about the pressure of extraction units installed in the world – one direction is towards ultra-high pressure and the other one in direction of lower pressure units. The operating pressure of extraction units depends on the nature of the material to be extracted. Many product applications processed with sub- or supercritical technologies are small in worldwide annual demand. Nevertheless, such products can be industrialized by making use of existing high pressure plant capacities and tolling the production in a facility that is already existing. This gives the advantage that an early investment into a production plant can be postponed to a later stage of market introduction, when demand and customer expectation is fully understood. Thus, a much more precise business plan can be elaborated increasing the viability of market success.

7.1.2 Micronisation Particle formation of substances using sub- or supercritical fluid is currently a subject of intensive research. There are some industrial scale plants for micronisation in operation with supercritical fluids; compared to more than hundred commercial highpressure extraction units, the number of micronization units is still relatively low. There is certain time needed before applied research is converted into industrial scale application, and based on developments in the area of high-pressure extraction, we could suppose that the number of high-pressure micronization units will increase in the near future. The main advantage using sub- or supercritical fluids for production of fine particles are again their unique thermodynamics and fluid dynamic solvent properties.

7.1 Future Directions/Perspectives

253

These properties of sub- or supercritical fluids can be used for formation of products with unique customer-designed properties in different applications. Particle formation processes could be easily connected to sub- or supercritical extraction processes, or to a downstream processing of products of chemical or biochemical synthesis in sub- or supercritical fluids. Using CO2 also prevents oxidation of products during processing steps, processing of substances using PGSS™ process could be performed even below their melting point. One can be sure that advances in the field of micronization and formulation processes using sub- and supercritical fluids will open up new ways for substances produced at an industrial scale in the near future.

7.1.3 Impregnation with SC Fluids Supercritical impregnation is a promising and alternative technique for the impregnation of poorly water soluble drugs. Through literature survey, it was discovered that impregnation of different drugs into different aerogels resulted in different loadings. Furthermore, final loading has been shown to be dependent on the solubility of drugs in supercritical fluids, characteristics of aerogels such as specific surface area and pore characteristics and also the conditions under which the drugs were impregnated such as pressure, temperature, time and flow rate. The interaction between the drug and the aerogel in most cases was shown to play a key role. Nowadays, most attention is paid to polysaccharide aerogels when it comes to impregnation of drugs, because of their biocompatibility and biodegradability. These properties are highly desired in the pharmaceuticals industry. When comparing the loading of e.g. Ketoprofen, as the most commonly used drug for impregnation, it is evident that much higher loadings were achieved with silica aerogels compared to any of the polysaccharide aerogels. The highest loading with Ketoprofen in the case of silica aerogels was 96%, while for polysaccharide aerogels it was 22% only. That is a significant difference in favor of silica aerogels. However, looking from pharmaceutical point of view, polysaccharide aerogels are more advantageous compared to silica. Polysaccharides are biocompatible, biodegradable, nontoxic and naturally occurring. Silica aerogels, on the other hand, are not biodegradable, but are biocompatible, which is a huge limitation for pharmaceutical or food applications. The supercritical fluid deposition technique overcomes many of the drawbacks of conventional techniques such as particle dimensions, particle size and distribution. The parameters influencing the final drug loading in the case of supercritical impregnation techniques are the same as the parameters influencing the final metal loading in the case of supercritical deposition technique. It is evident that further research on ternary systems comprising CO2 + impregnates + solid substrates is required for a detailed understanding of phase behavior, mass transfer and diffusion in the substrate and of influences on the properties of

254

7 Conclusion and Future Perspectives

the bulk material as crystallinity, morphology, anisotropy, reactivity. For optimizing expansion processes nucleation, bubble formation, and diffusion of gas from the solid polymers in the presence of liquid solvents should be investigated. For some processes of industrial relevance such as wood impregnation or dyeing of textiles there are broad experimental data bases and extensive knowledge that result in rather good models for describing the processes. Therefore, both supercritical impregnation and supercritical fluid deposition can be seen as competitive and alternative new CO2 -intensified impregnations in industrial applications in the coming decade compared to conventional techniques.

7.1.4 Chemical and Biochemical Reactions One of the main advantage of using dense gases as solvent for chemical and biochemical reactions is the tunability of solvent properties. For biochemical reactions the process parameters influence the activity, stability and selectivity of several enzymes. Another process advantages for conducting chemical and biochemical reactions in sub- or supercritical fluids is the ease of substrate and product fractionation and what is more and more important in the present time, integration of reaction and separation units of production process. Chemical or biochemical synthesis processes can be simply linked with direct micronisation and crystallization from SC CO2 by fluid expansion. An important fact is that CO2 is an inert solvent and therefore does not oxidize substrates and products. Additional research on new types of reactors (micro-reactors, membrane reactors) for continuous chemical or biochemical synthesis in dense fluids will reduce the size of equipment (and thus reduce investment costs) and will employ continuous processing with minimized flowrate of gas and eliminate large pressure drops (lowering the operating costs). Most of the research on chemical and biochemical reactions in subcritical and supercritical fluids was made by use of carbon dioxide as solvent. There are quite a few limitations using CO2 as reaction media like low solubility of reactants in dense CO2 , for bio-catalyzed reactions change of enzyme activity or deactivation of enzymes due to carbamate formation or acidification of reaction media, deactivation of enzymes or enzyme preparation due to pressure and/or temperature and due to cyclic pressure changes in batch processes. However, CO2 also offers extremely favorable attributes for its application as reaction media such as inflammability, non-toxicity and low costs. Thus, the main advantages of SCFs applications refer to their specific thermodynamic and heat transfer properties, the possibility of avoiding the use of organic solvent and obtaining products with high purity, and the lower energy consumption due to low process temperatures. As it was described before for bio-catalyzed reaction systems, although many enzyme species are stable in subcritical and supercritical fluids, for each substrate/ enzyme/fluid system the correct process conditions (mainly pressure, temperature,

7.1 Future Directions/Perspectives

255

pressure reduction in batch processes) must be experimentally determined. Based on the literature data the most stable enzymes are hydrolases (lipases and esterases), where pressure effect is lower than temperature deactivation. Despite the low research interest in the investigations of bio-catalyzed reactions in subcritical and supercritical fluids, based on the advantages these processes promise, we could be sure that sooner or later they will play an important role in industrial production of several substances. Another challenge and probably one of the major applications of biocatalyst in dense fluids could be the resolution of racemic mixtures to single enantiomers for use in pharmaceutical and agrochemical industry. In most cases only one of the two enantiomers has the desired bioactivity, whereas no activity or even undesirable side effects reside in the other enantiomer. The stereo-specificity of an enzyme depends largely on the structure of the substrate, interaction at the active site and on the process conditions. New enzyme species, which are “active” at unconventional process conditions (high pressure and/or high temperature) could be isolated from extremophiles or produced by new biotransformation or gene modified techniques. The use of whole cells or cell debris as biocatalysts in subcritical or supercritical fluids gave new challenges to researchers and engineers. From the ecological point of view supercritical fluid based technologies offer important advantages over conventional organic solvent technologies. Since water is the cheapest solvent and also offers high solubility of several substrates in water and for biocatalysed reactions, it allows for high activity of enzymes in water. However, there are no enzyme biotransformations performed in subcritical or supercritical fluids on industrial scale till now. Nevertheless, due to high environmental concerns and due to increasing organic solvent prices the demand for using new solvents increases. CO2 as environmental friendly and the second cheapest solvent (after water) and with the new challenges and opportunities will overcome the current gaps and will in the future open new pathways for production of several organic substances by chemical or biochemical synthesis in subcritical or supercritical fluids on an industrial scale. Therefore, we conclude that the advances in the field of high pressure technologies have opened up new pathways for substances and products obtained with cheap and environmentally friendly methods.

256

7 Conclusion and Future Perspectives

References Asimov I (1977) The collapsing universe. Walker & Co, New York IPAC (2020) Cool cosmos. https://coolcosmos.ipac.caltech.edu/ Lütge C, Schuetz E (2007) Market trends and technical developments in high pressure technology. In: Proceedings of the 5th international symposium on high pressure process technology and chemical engineering. European Federation of Chemical Engineering, Segovia, pp 24–27