Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics [1 ed.] 9781616682149, 9781607418412

Syngas is the name given to a gas mixture that contains varying amounts of carbon monoxide and hydrogen. Examples of pro

226 32 9MB

English Pages 472 Year 2009

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics [1 ed.]
 9781616682149, 9781607418412

Citation preview

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved. Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved. Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

ENVIRONMENTAL SCIENCE, ENGINEERING AND TECHNOLOGY SERIES

SYNGAS: PRODUCTION METHODS, POST TREATMENT AND ECONOMICS

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

No part of this digital document may be reproduced, stored in a retrieval system or transmitted in any form or by any means. The publisher has taken reasonable care in the preparation of this digital document, but makes no expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No liability is assumed for incidental or consequential damages in connection with or arising out of information contained herein. This digital document is sold with the clear understanding that the publisher is not engaged in rendering legal, medical or any other professional services.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

ENVIRONMENTAL SCIENCE, ENGINEERING AND TECHNOLOGY SERIES Nitrous Oxide Emissions Research Progress Adam I. Sheldon and Edward P. Barnhart (Editors) 2009. ISBN: 978-1-60692-267-5 Fundamentals and Applications of Biosorption Isotherms, Kinetics and Thermodynamics Yu Liu and Jianlong Wang (Editors) 2009. ISBN: 978-1-60741-169-7 Environmental Effects of Off-Highway Vehicles Douglas S. Ouren, Christopher Haas, Cynthia P. Melcher, Susan C. Stewart, Phadrea D. Ponds, Natalie R. Sexton Lucy Burris, Tammy Fancher and Zachary H. Bowen 2009. ISBN: 978-1-60692-936-0 Agricultural Runoff, Coastal Engineering and Flooding Christopher A. Hudspeth and Timothy E. Reeve (Editors) 2009. ISBN: 978-1-60741-097-3 Agricultural Runoff, Coastal Engineering and Flooding Christopher A. Hudspeth and Timothy E. Reeve (Editors) 2009. ISBN: 978-1-60876-608-6 (Online book)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Conservation of Natural Resources Nikolas J. Kudrow (Editor) 2009. ISBN: 978-1-60741-178-9 Conservation of Natural Resources Nikolas J. Kudrow (Editor) 2009. ISBN: 978-1-60876-642-6 (Online book) Directory of Conservation Funding Sources for Developing Countries: Conservation Biology, Education and Training, Fellowships and Scholarships Alfred O. Owino and Joseph O. Oyugi 2009. ISBN: 978-1-60741-367-7

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Forest Canopies: Forest Production, Ecosystem Health and Climate Conditions Jason D. Creighton and Paul J. Roney(Editors) 2009. ISBN: 978-1-60741-457-5 Soil Fertility Derek P. Lucero and Joseph E. Boggs (Editors) 2009. ISBN: 978-1-60741-466-7 The Amazon Gold Rush and Environmental Mercury Contamination Daniel Marcos Bonotto and Ene Glória da Silveira 2009. ISBN: 978-1-60741-609-8

Process Engineering in Plant-Based Products Hongzhang Chen 2009. ISBN: 978-1-60741-962-4 Buildings and the Environment Jonas Nemecek and Patrik Schulz (Editors) 2009. ISBN: 978-1-60876-128-9 Tree Growth: Influences, Layers and Types Wesley P. Karam (Editor) 2009. ISBN: 978-1-60741-784-2

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Syngas: Production Methods, Post Treatment and Economics Adorjan Kurucz and Izsak Bencik (Editors) 2009. ISBN: 978-1-60741-841-2

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved. Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

ENVIRONMENTAL SCIENCE, ENGINEERING AND TECHNOLOGY SERIES

SYNGAS: PRODUCTION METHODS, POST TREATMENT AND ECONOMICS

ADORJAN KURUCZ AND

IZSAK BENCIK

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

EDITORS

Nova Science Publishers, Inc. New York

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009 by Nova Science Publishers, Inc. All rights reserved. No part of this book may be reproduced, stored in a retrieval system or transmitted in any form or by any means: electronic, electrostatic, magnetic, tape, mechanical photocopying, recording or otherwise without the written permission of the Publisher. For permission to use material from this book please contact us: Telephone 631-231-7269; Fax 631-231-8175 Web Site: http://www.novapublishers.com NOTICE TO THE READER The Publisher has taken reasonable care in the preparation of this book, but makes no expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No liability is assumed for incidental or consequential damages in connection with or arising out of information contained in this book. The Publisher shall not be liable for any special, consequential, or exemplary damages resulting, in whole or in part, from the readers’ use of, or reliance upon, this material. Any parts of this book based on government reports are so indicated and copyright is claimed for those parts to the extent applicable to compilations of such works. Independent verification should be sought for any data, advice or recommendations contained in this book. In addition, no responsibility is assumed by the publisher for any injury and/or damage to persons or property arising from any methods, products, instructions, ideas or otherwise contained in this publication. This publication is designed to provide accurate and authoritative information with regard to the subject matter covered herein. It is sold with the clear understanding that the Publisher is not engaged in rendering legal or any other professional services. If legal or any other expert assistance is required, the services of a competent person should be sought. FROM A DECLARATION OF PARTICIPANTS JOINTLY ADOPTED BY A COMMITTEE OF THE AMERICAN BAR ASSOCIATION AND A COMMITTEE OF PUBLISHERS. LIBRARY OF CONGRESS CATALOGING-IN-PUBLICATION DATA Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

ISBN: 978-1-61668-214-9 (E-Book) Available upon request

Published by Nova Science Publishers, Inc. Ô New York

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

CONTENTS Preface Chapter 1

Chapter 2

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Chapter 3

ix Stage of Deployment of Syngas Cleaning Technologies and RD&D Needs to Accomplish the New Challenges of Syngas Utilization Filomena Pinto, Rui Neto André and I. Gulyurtlu Syngas Generation from Hydrocarbons and Oxygenates with Structured Catalysts V. Sadykov, L. Bobrova, S. Pavlova, V. Simagina, L. Makarshin, V. Parmon, J. R. H. Ross, C. Mirodatos, A. C. Van Veen and A. P. Khristolyubov Preparation of Thin-Film Pd Membranes for H2 Separation from Synthesis Gas and Detailed Design of a Permeability Testing Unit M. Bientinesi and L. Petarca

1

53

141

Chapter 4

Fischer-Tropsch Synthesis with Fe-Based Catalysts M. Ojeda, T. Herranz, F. J. Pérez-Alonso, J. M. González-Carballo, S. Rojas and J. L. G. Fierro

191

Chapter 5

Syngas Production in Membrane Reactors Fausto Gallucci and Angelo Basile

227

Chapter 6

Reformer and Membrane Modules Plant to Optimize Natural Gas Conversion to Hydrogen M. De Falco, G. Iaquaniello, B. Cucchiella and L. Marrelli

Chapter 7

Chapter 8

Recent Developments of Fischer-Tropsch Synthesis Catalysts Preparation and Characterization Naoto Koizumi, Takehisa Mochizuki, Daichi Hongo and Muneyoshi Yamada High Temperature Electrolysis of Steam and CO2 for Syngas Production Carl M. Stoots, James E. O’Brien, J. Stephen Herring and Joseph J. Hartvigsen

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

263

313

345

viii Chapter 9

Chapter 10

Contents Chemical Looping Reforming for Syngas Generation from Natural Gas –Based on Results from a 120 KW Fuel Power Installation Johannes Bolhàr-Nordenkampf, Tobias Pröll, Philipp Kolbitsch and Hermann Hofbauer Partial Oxidation of Methane over Zirconia- and MagnesiaSupported Ruthenium and Rhodium Catalysts Maria do Carmo Rangel, Marluce Oliveira da Guarda Souza, Dino de Jesus Sodré, André Leopoldo Macêdo da Silva, Márcia Souza Ramos and José Mansur Assaf

375

395

Chapter 11

Technologies of Syngas Production from Biomass Generated Gases Simone Albertazzi, Francesco Basile, Patricia Benito Martin, Giuseppe Fornasari, Ferruccio Trifirò and Angelo Vaccari

409

Chapter 12

Process for Conversion of Coal to Substitute Natural Gas (SNG) Meyer Steinberg

417

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Index

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

435

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

PREFACE Syngas is the name given to a gas mixture that contains varying amounts of carbon monoxide and hydrogen. Examples of production methods include steam reforming of natural gas or liquid hydrocarbons to produce hydrogen, the gasification of coal and in some types of waste-to-energy gasification facilities. Syngas is also used as an intermediate in producing synthetic petroleum for use as a fuel or lubricant via Fischer-Tropsch synthesis and previously the Mobil methanol to gasoline process. Syngas consists primarily of hydrogen, carbon monoxide, and very often some carbon dioxide, and has less than half the energy density of natural gas. It is combustible and often used as a fuel source or as an intermediate for the production of other chemicals. This new book gathers the latest research from around the globe in this dynamic field covering topics such as syngas production from biomass generated gases, recent developments of Fischer-Tropsch synthesis catalysts, syngas cleaning technologies, and new syngas utilizations at different stages of deployment. Chapter 1 - Syngas is normally produced from gasification of coal or wastes. More recently syngas production from co-gasification of different types of wastes mixed with fossil fuels, mainly coal but also heavy fractions from petroleum refining industry, like petcoke has been investigated. The diversification of feedstocks for syngas productions presents several challenges, as syngas characteristics and quality is much dependent on feedstock composition. The gasification of wastes with considerable amounts of S, Cl and N is expected to produce a syngas with several undesirable S, Cl and N compounds, which need to be controlled and reduced, as they may compromise some syngas end-uses. Otherwise, syngas will need complicated and expensive cleaning technologies to achieve the suitable properties prior to its utilization. Several syngas cleaning processes will be analyzed with the aim of removing particulate dust, tar and S, Cl and N undesirable compounds. The enrichment in hydrogen content will be also discussed to enlarge the range of applications. Special consideration will be given to hot syngas cleaning technologies, including thermal catalytic reforming of tar, hydrocarbons abatement and water gas shift reaction. The most known syngas utilization is for power and heat generation. Due to the low calorific value of syngas obtained from air gasification processes, co-firing of syngas in pulverized coal installations has been successfully used. Syngas may also be used in higher efficient heat and power generators like gas turbines or fuel cells, however, as these equipments are more exigent towards syngas quality, special care needs to be taken in syngas cleaning processes. Other new syngas utilizations will be also analyzed, such as: synthesis of

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

x

Adorjan Kurucz and Izsak Bencik

methanol, ethanol and dimethyl ether, Fischer-Tropsch synthesis, hydrogen production and production of bio-based products from syngas fermentation. New syngas utilizations are at different stages of deployment, but all of them need further research and development, especially concerning the development of new catalysts, supporting materials and sorbents to improve the efficiency and selectivity of syngas cleaning technologies and utilization processes. Chapter 2 - This paper presents results of research of syngas generation by oxidative or steam reforming of hydrocarbons and oxygenates in structured catalytic reactors. Development of structured catalysts included design of active components stable to coking and sintering while efficiently transforming different types of fuels into syngas. It was based upon constructing nanocomposites comprised of components efficient in fuel molecules activation (Ni, Cu, precious metals, etc) and ceria-containing mixed oxides which provide a high rate of oxidants (O2, H2O, CO2) activation and supply of oxygen-containing species to the perimeter of metal particles to consume activated C-H-O fragments. The catalyst development was linked to reactor design. To ensure efficient heat and mass –transfer management required for high performance of structured catalytic reactors, several types of heat-conducting monolithic substrates comprised of refractory alloys and cermets (including microchannel structures) were developed. Procedures for supporting protective layers and active components on these substrates were successfully elaborated. Several types of pilot-scale reactors (with the radial or the axial flow direction) equipped with unique liquid fuel evaporation and mixing units and internal heat exchangers were designed and manufactured. Extended tests of these reactors fed by fuels from C1 to gasoline, mineral and sunflower oil have been carried out with a broad variation of experimental parameters including stability tests up to 1000 h. Performance analysis has been made with a due regard for equilibrium restrictions on the operational parameters. Transient behavior of the monolith reactor during start-up (ignition) of the methane partial oxidation to synthesis gas was studied and analyzed via mathematical modeling based upon detailed elementary step mechanism. This provides required bases for theoretical optimization of the catalyst bed configuration and process parameters. Chapter 3 - Hydrogen is one of the most important chemical products and it is used in a wide variety of industrial fields including chemical, petrochemical, metallurgic and energy applications. Hydrogen demand in all of these fields is continuously growing. At present, hydrogen is mainly produced from fossil fuels via processes such as steam reforming, partial oxidation and gasification. All of these processes lead to the production of an H2-containing gas stream (synthesis gas) from which hydrogen has to be separated. Thin-film palladium membranes are one of the most promising technologies for the separation of hydrogen from synthesis gas. It involves some advantages over traditional separation methods like pressure swing adsorption (PSA), and over other membrane materials (polymeric, porous or dense ceramic), among which are 100% separation efficiency, high permeability and operating conditions compatible with upstream fuel conversion processes. In this work, palladium films were deposited above stainless steel porous supports using the electroless plating (ELP) technique. Two different geometries (disc sheet and tubes) were chosen for the supports, which both have a 0.1 μm filter grade. Surface morphology and cross-section were observed through scanning electron microscopy (SEM). For each membrane, film thickness was estimated both by weight gain and by cross-section observation with SEM. A good agreement was found between the two values. The evolution

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Preface

xi

of film thickness and morphology for increasing ELP cycle number was studied, as well as the influence of a previous phase of thermal oxidation of the metallic substrate. Membranes were tested in an appropriate set up for nitrogen tightness in order to individuate local defects. The results showed that after only 4 cycles of deposition a uniform dense film of palladium with a thickness of about 10 μm is obtained, but even after 6 cycles a small number of defects still subsist, probably due to support of local morphological discontinuities. Finally, the detailed design of a permeability testing unit is reported, including the fluid dynamic, thermal and mechanical dimensioning, the selection of materials and equipment, and some safety considerations. The sketches of the two permeation cells (for the two different membrane geometries) are also reported. Chapter 4 - The world demand of light and middle distillate petroleum products (mainly gasoline and diesel) is increasing dramatically, especially within the transportation sector. Eventually, such requirement will surpass refineries capacity to supply fuels and hydrocarbon feedstock. This situation has encouraged the exploration of other sources of petroleum products rather than merely sticking to conventional oil sources. Within this scenario, natural gas, coal and biomass appear as suitable carbon sources to be converted into the highly demanded hydrocarbon feedstock through a demonstrated two-step technology: formation of synthesis gas (carbon source → H2 + CO) and Fischer-Tropsch synthesis (FTS), which consists in the CO hydrogenation reaction to form hydrocarbons (H2 + CO → hydrocarbons). The FTS provides thus an alternative route for the production of clean transportation fuels and high molecular weight hydrocarbons from natural gas (gas-to-liquids, GTL), coal (coal-toliquids, CTL) and/or biomass (biomass-to-liquids, BTL). This process takes place at high temperatures and moderate pressures in fixed bed or slurry reactors. Synthesis gas mixtures can be converted into useful fuels and petrochemicals with Fe- and Co-based catalysts. In particular, iron-based catalysts are interesting for this process because of their low cost, high activity, flexible product distribution, and possibility of using coal- and biomass-derived synthesis gas with low H2/CO ratios. Moreover, Fe-based catalysts offer the possibility of using synthesis gas containing high amounts of CO2. The catalytically active Fe phase has been the focus of intense research. It is mostly accepted that Fe carbides play an active role in the FTS, although the fact that iron suffers a reconstruction process from the oxide phase to a mixture of phases (metallic iron, iron oxides and iron carbides) under reaction conditions hinders the unequivocal identification of the real active sites. There is a lack of effective characterization techniques under typical reaction conditions in the FTS (1-4 MPa, 450-570 K). Therefore, most of the characterization studies have focused on catalytic precursors and used catalysts. The situation is still more complicated considering that the presence of chemical promoters and supports affects the Fe phase and structure during the reaction. In this chapter, the FTS technology, namely reactor types, reaction conditions and typical catalysts, will be briefly reviewed. The core of the chapter will be devoted to the description of active sites and reaction intermediates with Fe-based catalysts. Chapter 5 - Syngas is an important feedstock for the production of higher hydrocarbons or methanol. It can be produced via conversion of methane and the most extensively used process for this conversion is the methane steam reforming reaction carried out in large furnaces. The steam reforming reaction is a highly endothermic reaction which is industrially operated under severe conditions resulting in several undesirable consequences: sintering of the catalyst, very high carbon deposition and the use of high-temperature resisting materials. These drawbacks for methane steam reforming can be overcome by using membrane reactors,

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

xii

Adorjan Kurucz and Izsak Bencik

systems able to combine the separation properties of membrane with the typical characteristics of catalytic reactions. By using for example Pd-based membrane reactors, the hydrogen produced can be continuously withdrawn from the reaction system circumventing the thermodynamic limitations and making the methane steam reforming feasible at lower temperatures than the traditional systems. A potential alternative technique to steam reforming processes for producing syngas is the partial oxidation of methane with oxygen, having the disadvantage (economical and technological) that pure oxygen is required. Utilisation of air instead of pure oxygen is beneficial only if it can be performed by using a membrane reactor in which the membrane is perm-selective to oxygen. Another possible route for the partial oxidation of methane is the use of catalytic membrane reactors in which the membrane acts as both separation layer and reaction media. Finally, an interesting and promising route for syngas production is the methane reforming with CO2 (dry reforming) which is also a reaction system that can be used for the CO2 emissions mitigation. The methane dry reforming can be effectively performed in Pdbased membrane reactors as well as in catalytic membrane reactors. In this chapter, different examples of syngas production in both Pd-based membrane reactors and in catalytic membrane reactors will be presented and compared with traditional reaction systems. Chapter 6 - Membrane technology may play a crucial role in the efficient production of hydrogen from natural gas and heavy hydrocarbons. The chapter assesses the performance of hydrogen production plants in which Pd-based selective membranes are integrated. Two different configurations have been proposed and are evaluated:

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

• Reformer and Membrane Modules (RMM), by which the hydrogen produced in reaction units is separated by Pd-based membrane modules assembled downstream each reaction step. • Membrane Reformer (MR), which combines the hydrogen separation through the selective membrane and the steam reforming reaction into one unit and separates hydrogen immediately after it was formed. Both the configurations allow a reduction of operating temperature (< 650°C for RMM, < 550°C for MR) with benefits in terms of process energy efficiency. Moreover, lower operating temperatures allow location of the modules downstream of a gas turbine, achieving an efficient hybrid system producing electric power and hydrogen. The chapter will be composed by the following sections: • The concept of membrane integration in steam reforming process: benefits and drawbacks. • Hydrogen selective membranes state-of-the-art and performance. • RMM plant configuration: modeling and simulations. • MR plant configuration: modeling and simulations. • Comparison between RMM/MR technologies with conventional steam reformer plants: economical evaluation of the process layout.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Preface

xiii

Chapter 7 - Since Fischer and Tropsch discovered the Fischer-Tropsch synthesis (FTS) reaction in 1923, in which the mixture of CO and H2 (so called syngas) derived from coal is converted into hydrocarbon, there have been continuous efforts on improving the understanding of both fundamental and technological aspects of this catalytic reaction. More recently, FTS reaction has attracted attentions as a synthetic method of clean transportation fuels from various carbon resources such as natural gas, coal and biomass. In this review, recent developments in both catalyst preparation and catalyst characterization technique for Fe and Co FTS catalysts have been reviewed, including our recent results on novel preparation method of Co FTS catalyst with chelating agents. Chapter 8 - Presented is a synopsis of recent experiments on simultaneous hightemperature electrolysis (coelectrolysis) of steam and carbon dioxide using solid-oxide electrolysis cells to produce syngas. Coelectrolysis is complicated by the fact that the reverse shift reaction occurs concurrently with the electrolytic reduction reactions. All reactions must be properly accounted for when evaluating results. Electrochemical performance of the button cells and stacks were evaluated over a range of temperatures, compositions, and flow rates. The apparatus used for these tests is heavily instrumented, with precision mass-flow controllers, on-line dewpoint and CO2 sensors, and numerous pressure and temperature measurement stations. It also includes a gas chromatograph for analyzing outlet gas compositions. Comparisons of measured compositions to predictions obtained from a chemical equilibrium coelectrolysis model are presented, along with corresponding polarization curves. Results indicate excellent agreement between predicted and measured outlet compositions. Cell area-specific resistance values were found to be similar for steam electrolysis and coelectrolysis. Coelectrolysis significantly increases the yield of syngas over the reverse water gas shift reaction equilibrium composition. The process appears to be a promising technique for large-scale syngas production. Most of this discussion was previously published in the Journal of Fuel Cell Science and Technology [1] and is used here with permission from the American Society of Mechanical Engineers. Chapter 9 - In this chapter chemical looping systems are investigated as a synthesis gas generation process. Two different atmospheric chemical looping process configurations are discussed. The chemical looping autothermal reforming represents a chemical looping system operated at a global air to fuel ratio below one. The second configuration represents a tubular steam reformer utilized with heat from a chemical looping combustor. Further synthesis upgrading steps (shift reactor, CO2 separation, etc.) are considered for both applications. To underline the presented process application operating results of the dual circulating fluidized bed chemical looping pilot rig at Vienna University of technology are shown. Dependencies of reactor temperature on the gas conversion, fuel and air respectively, are discussed. It can be concluded, that chemical looping systems for synthesis production may be an attractive competitor to standard reforming technology in the near future. Chapter 10 - Hydrogen and syngas are key feedstocks for several industrial processes such as hydrotreating and hydrocracking in addition methanol, ammonia and Fischer-Tropsch synthesis. For these applications, different H2/CO molar ratios of syngas are required. Among the various routes for obtaining hydrogen and syngas, steam reforming is by far the most commonly-used one in commercial practice. However, this reaction has some drawbacks, such as the high endothermicity and catalyst deactivation. The partial oxidation of methane, however, is an exothermic reaction, making it an attractive option for obtaining hydrogen and syngas. In order to find alternative catalysts to the reaction, ruthenium- and rhodium-based

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

xiv

Adorjan Kurucz and Izsak Bencik

catalysts were compared, using zirconia and magnesia as supports; also, the effect of the addition of small amounts of magnesia (Mg/Zr (molar)= 0.1) on zirconia-based catalysts was studied. The supports were prepared by precipitation and then impregnated with ruthenium or rhodium chloride to get solids with 1% of metal. Samples were characterized by differential thermal analysis, thermogravimetry, X-ray diffraction, thermoprogrammed reduction and nitrogen adsorption. The catalysts were evaluated in the partial oxidation of methane carried out at 1 atm and in the range of 450 to 750°C. Monoclinic and tetragonal phases were detected in the zirconia-based solid, while magnesium-doped zirconia showed only the tetragonal phase. Magnesia showed the cubic phase typical of magnesium oxide. Magnesium decreased the specific surface area of zirconia-based catalysts, regardless of the kind of metal; it also made the rhodium and ruthenium reduction more difficult. Zirconia was found to be a more suitable support for the catalysts, in the range of 450 to 750°C, than magnesia and magnesium-doped zirconia, probably due to its ability in favoring well-dispersed metal particles. In a general tendency, ruthenium produces more active and selective catalysts than rhodium and thus the most promising catalyst was zirconia-supported ruthenium. All catalysts produced carbon dioxide, showing that the reaction occurred through indirect oxidation, the amount depending on the metal, the support and on the reaction temperature. Most of the catalysts produced H2/CO molar ratio values of 2 above 550°C, being suitable for methanol and Fischer-Tropsch synthesis; on the other hand, the rhodium and magnesium-containing catalyst is more suitable for producing hydrogen. Therefore, rhodium and ruthenium catalysts can be tailored to produce different H2/CO molar ratios for several applications. Chapter 11 - Advanced biomass gasification will play a crucial role in at least reducing, if not eliminating the need for imported oil and the negative effects of greenhouse gases generated from the combustion of fossil fuels. The biomass generated gas mainly contains CO, CH4, H2, and CO2 together with contaminants and catalyst poisons (tar, fly-ash, H2S, NH3). Therefore these species have to be removed and the syngas upgraded before the conversion to liquid fuels. The process required to convert residual light hydrocarbons into syngas can be chosen between Auto Thermal Reforming (ATR) and Partial Oxidation (POX). Operating with ATR will allow higher yield and higher overall process efficiency (12 %). Moreover, less expensive refractory lining is required, due to the lower operating temperature. On the other hand, POX is a robust process, insensitive to both poisoning (since catalyst is not present) and mixing (homogenous phase reaction). Therefore, ATR should be preferred to POX if a catalyst can be found to handle both the contaminants present in the produced gas and the thermal sintering. At this purpose, a commercially available Ni catalyst has been tested in a bench-scale ATR in presence of H2S to evaluate the possibility to use it in a real process. Chapter 12 - This report covers a review and process comparison of the three major processes for conversion of coal to substitute natural gas (SNG): (1) Steam – Oxygen gasification; (2) Catalytic gasification; and (3) Hydrogasification. In addition, two of these three processes are compared for underground coal gasification application at depleted coal bedded methane (CBM) sites. The process chemistry, flowsheets, and mass and energy balances are presented. An economic analysis, including capital investment and production cost estimates is given. A critical comparative evaluative analysis of the coal to SNG process is made. The information presented indicates that for above ground conversion of coal to SNG, hydrogasification is the most thermally efficient process, reaching 80%, has the least CO2

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Preface

xv

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

emission and is the most economical process, producing SNG for $4.61/MSCF. For underground depleted CBM well conversion of the unminable coal seam, the hydrogasification process, applied to multiple wells using the latest capital investment for methane reforming for the hydrogen production, can produce SNG for as low as $1.44/MSCF. This value yields a large profit margin considering current natural gas market price is reaching $9.00/MSCF. A limited field test is recommended at a depleted CBM site and is described to prove out the system at a cost of about $1 million.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved. Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

In: Syngas Production Methods, Post Treatment… Editors: Adorjan Kurucz and Izsak Bencik

ISBN: 978-1-60741-841-2 © 2009 Nova Science Publishers, Inc.

Chapter 1

STAGE OF DEPLOYMENT OF SYNGAS CLEANING TECHNOLOGIES AND RD&D NEEDS TO ACCOMPLISH THE NEW CHALLENGES OF SYNGAS UTILIZATION Filomena Pinto*, Rui Neto André and I. Gulyurtlu INETI, Estrada do Paço do Lumiar, 22, 1649-038 Lisboa, Portugal

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

ABSTRACT Syngas is normally produced from gasification of coal or wastes. More recently syngas production from co-gasification of different types of wastes mixed with fossil fuels, mainly coal but also heavy fractions from petroleum refining industry, like petcoke has been investigated. The diversification of feedstocks for syngas productions presents several challenges, as syngas characteristics and quality is much dependent on feedstock composition. The gasification of wastes with considerable amounts of S, Cl and N is expected to produce a syngas with several undesirable S, Cl and N compounds, which need to be controlled and reduced, as they may compromise some syngas end-uses. Otherwise, syngas will need complicated and expensive cleaning technologies to achieve the suitable properties prior to its utilization. Several syngas cleaning processes will be analyzed with the aim of removing particulate dust, tar and S, Cl and N undesirable compounds. The enrichment in hydrogen content will be also discussed to enlarge the range of applications. Special consideration will be given to hot syngas cleaning technologies, including thermal catalytic reforming of tar, hydrocarbons abatement and water gas shift reaction. The most known syngas utilization is for power and heat generation. Due to the low calorific value of syngas obtained from air gasification processes, co-firing of syngas in pulverized coal installations has been successfully used. Syngas may also be used in higher efficient heat and power generators like gas turbines or fuel cells, however, as these equipments are more exigent towards syngas quality, special care needs to be taken in syngas cleaning processes. Other new syngas utilizations will be also analyzed, such

*

Corresponding author: E-mail: [email protected], Tel: 351 21 092 4787, Fax: 351 21 716 6569

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

2

Filomena Pinto, Rui Neto André and I. Gulyurtlu as: synthesis of methanol, ethanol and dimethyl ether, Fischer-Tropsch synthesis, hydrogen production and production of bio-based products from syngas fermentation. New syngas utilizations are at different stages of deployment, but all of them need further research and development, especially concerning the development of new catalysts, supporting materials and sorbents to improve the efficiency and selectivity of syngas cleaning technologies and utilization processes.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

1. INTRODUCTION For some decades, the deposition of wastes in disposal areas have been substituted by wastes incineration or combustion, which has the great advantage of taking profit of wastes energy content and at the same time to decrease the amount of wastes disposed in land fills. Due to wastes different types and compositions, pollutants emissions of different kinds than those produced during coal combustion are expected to occur. On the other hand, due to the increasing environmental concern, new legislations all over the world have been imposing reductions on the emission limits of incineration and combustion units, which require the use of expensive technologies for flue gas treatment, mechanical pre-treatment and/or separation of some waste types or components. Alternatively, new termochemical processes have been applied and developed for taking profit of wastes energy content, like gasification and pyrolysis. Nowadays, pyrolysis main products are liquids of several types and with different compositions, depending on waste type and pyrolysis technology. Gasification of coal, biomass and/or other organic wastes produces syngas, whose main components are: CO, CO2, H2, CH4 and other gaseous hydrocarbons. Gasification occurs in presence of steam and/or oxygen (or air). The use of either oxygen or air supplies the energy necessary to the process, through partial combustion reactions. The use of oxygen instead of air, is also technically advantageous, as the syngas produced will not be diluted by nitrogen, however oxygen separation units are economically prohibitive for small-scale plants. The use of steam as gasification medium has the disadvantage of increasing operational costs, since the heat needed for the process has to be supplied by external sources. Gasification reactor systems may be based on a fixed bed, an entrained flow or a fluidized bed reactor, having the fluidized bed reactor usually a better performance. A low cost catalyst may be added to the gasifier to decrease the formation of tar. When feedstocks with high contents of sulfur and chlorine are gasified, a specific sorbent may be also added to the bed to retain these elements compounds inside the bed. Waste gasification studies started some decades ago, but most of these early efforts were technologically irrelevant, as hardly any one of the early processes still remains at commercial scale. Several aspects were responsible for this initial failure. Waste gasification is a complex chemical process, whose fundamental knowledge was not available, the heterogeneity and complexity of wastes led to different important mechanical problems, including wastes feeding, sintering and ash removal, which led to units shut-down. Processes developed at lab-scale with small reactors were converted into much larger scales, without considering the problems of fast scaling-up capacities. Though most of the early gasification processes produced large quantities of tar, they did not consider gas cleaning, thus tar carried out by syngas led to tar condensation and clogging in pipes and equipments.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Stage of Deployment of Syngas Cleaning Technologies…

3

Stiegel and Maxwel, 2001 analyzed the historical growth in gasification capacity since 1970. During this decade the most important gasification installations were in South Africa with Lurgi gasifiers operating at Sasol. During the 1980s the great increase in gasification capacity was associated with Sasol II and III and to the commissioning of 14 Lurgi gasifiers at the Dakota Gasification plant in Buelah, USA. Gasification capacity remained stable till around 1993, but then it increased around 50% during the following years. Later developments on waste gasification has tried to solve some of the previously mentioned drawbacks trying to overcome problems related to pollutant emissions, inert solid residues, higher product flexibility, higher efficiency to power, etc. Several processes have been tested on pilot scale and other are under development, either for fuel gas cleaning (LURGI, TPS and DANECO) and flue gas cleaning (Siemens-Kiener and ScanArc). According to Stiegel and Maxwel, 2001 it is expected that this trend will continue to rise, due to the necessity of using the resources available in a more sustainable way. Nowadays, gasification is widely deployed all over the world, being most of these facilities located in Western Europe, the Pacific Rim, Africa, and North America. The most used feedstocks are coal and petroleum residuals, with a small fraction from petroleum coke (petcoke). During the next years, more gasifiers are expected to be constructed adding further MWth of synthesis gas capacity, probably using as feedstocks different types of carbonaceous materials, previously considered as wastes. However, the effectivity of syngas cleaning technologies needs to be fully demonstrated, otherwise some of syngas utilizations would be compromised and even the viability of the gasification processes. Syngas may be used in dryers, boilers, ovens, limekilns, brick kilns, metallurgical furnaces, etc. Syngas after advanced gas cleaning processes may be also used for power generation in engines, gas turbines (combined cycles) or fuel cells. However, as these equipments are exigent towards syngas quality, special care needs to be taken in syngas cleaning processes, such as: S, N and Cl compounds abatement, tar cracking, steam reforming, etc. Other new syngas utilizations include: synthesis of methanol and dimethyl ether, FischerTropsch synthesis, hydrogen production and production of bio-based products from syngas fermentation. Chemical synthesis obliges the right H2/CO ratios, which can be obtained by water-shift reaction. This process is also important for hydrogen production, as CO may be converted into H2 and CO2. As the resulting CO2 concentration is much higher than in syngas, the energy effort for CO2 separation is much lower. As physical absorption processes for CO2 removal have great disadvantages, shift gas membranes with high H2 permeability are under development. The remaining CO2 rich stream are sent to storage, whilst H2-rich fuel can thereafter be used in a gas turbine combined cycle or in fuel cells to produce electricity. Therefore, gasification has an important role in the development of zero emissions technologies (ZET).

2. THE EFFECT OF GASIFICATION CONDITIONS ON SYNGAS PROPERTIES AND COMPOSITION Syngas results from a wide range of chemical reactions that occur when carbonaceous materials reach a temperature generally higher than 700ºC (973 K), in which participates solid

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Filomena Pinto, Rui Neto André and I. Gulyurtlu

4

carbon and the several components of the gasification medium, usually oxygen, steam or a mixture of both and the gaseous compounds formed by gasification reactions. Hundreds of different reactions may occur, but in Table 1 are only summarized the most important ones. Oxidation reactions (1) and (2) supply the heat necessary to the gasification process and release CO and CO2 that participate in several gasification reactions, as for instance Boudouard reaction (3) between solid carbon and CO2 to form more CO. Boudouard reaction occurs mainly at temperatures higher than 1000 K. Water gas reactions, (4) and (5), involve solid carbon and water vapor, are endothermic and favored by higher temperatures and lower pressure. Methanation or hydrogasification, reaction (6), occurs between carbon and hydrogen, it is usually very slow, but favored at higher pressure. Water gas shift reaction (7), occurs in presence of water vapor and convert CO into H2 and CO2, therefore it is very important to change syngas CO/H2 ratio. Besides, being converted by cracking reactions, hydrocarbons, including CH4 present on the gas phase, may suffer several reforming reactions, either with steam, reactions (8) to (11) or with CO2, all leading to the increase of syngas CO, H2 or CH4 concentrations. CH4 may be also formed by hydrogen reforming reaction (15), which occurs between H2 and CO. This reaction usually takes place at a low degree, except at higher pressure or in presence of suitable catalysts. Table 1 Most important gasification reactions that leads to syngas formation (Smoot, L.D. and Smith, P.J., 1985).

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Designation Oxidation

∆H (kJ/mol) -392,5

(1)

C(s) + ½ O2 ⇆ CO

-110,5

(2)

Boudouard

C(s) + CO2 ⇆ 2 CO

172,0

(3)

Water Gas: primary secondary

C(s) + H2O ⇆ CO + H2

131,4

(4)

C(s) + 2 H2O ⇆ CO2 + 2 H2

90,4

(5)

Methanation

C(s) + 2 H2 ⇆ CH4

-74,6

(6)

Water-gas shift

CO + H2O ⇆ CO2 + H2

-41,0

(7)

Steam Reforming

CH4 + H2O ⇆ CO + 3 H2

205,9

(8)

CH4 + 2 H2O ⇆ CO2 + 4 H2

164,7

(9)

CnHm + n H2O ⇆ n CO + (n + m/2) H2

210,1

(10)

CnHm + n/2 H2O ⇆ n/2 CO + (m-n) H2 + n/2 CH4

4,2

(11)

CH4 + CO2 ⇆ 2 CO + 2 H2

247,0

(12)

CnHm + n CO2 ⇆ 2n CO + m/2 H2

292,4

(13)

CnHm + n/4 CO2 ⇆ n/2 CO + (m-3n/2) H2 + (3n/4) CH4

45,3

(14)

CO + 3 H2 ⇆ CH4 + H2O

-205,9

(15)

CO2 Reforming

H2 Reforming

Mechanism C(s) + O2 ⇆ CO2

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Stage of Deployment of Syngas Cleaning Technologies…

5

During gasification also occurs the formation of precursor’s pollutants compounds containing N, S and halogens. Due to gasification reduction conditions, these compounds are usually formed as NH3, H2S and HCl. When the gasification gas is combusted, NH3 form nitrogen oxides (NOx) and H2S originate sulfur oxides (SOx), which are difficult to remove and are precursors to acid rains. Therefore, it is of most importance to decrease NH3 and H2S concentrations before syngas end-use, as even the most efficient low NOx and low SOx combustion methods may not be enough to keep the final NOx and SOx levels acceptable. The formation of these undesirable compounds during gasification may be controlled by adjusting gasification operating parameters and by using specific catalysts or sorbents. However, there are many different phenomena occurring during gasification and some of them act in opposite directions, thus making difficult the complete understanding of the complex reactions that may occur and thus syngas composition. Syngas properties and composition is most affected by gasification experimental conditions and gasification technology. Pinto et al., 2008a analyzed and discussed the effect of experimental parameters on gasification performance and on syngas composition. The most important parameter on syngas composition is feedstock composition.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

2.1. Feedstock Composition Pinto et al., 2008a have co-gasification coal mixed with different types of wastes, such as: sewage sludge (SS), municipal solid waste (MSW), refuse derived fuel (RDF), olive oil bagasse, pine, cardoon, edible oil wastes, petcoke and PE (polyethylene). Most studies have been made with higher amounts of coal than wastes, because as coal is the most gasified material, it makes easier the control and stabilization of co-gasification processes and therefore, coal gasification installations already available could be easily converted to coprocess wastes by substituting a fraction of coal by wastes. By this way the seasonable problem of some wastes or the lack of quantity could also be overtaken. On the other hand, coal and wastes blends could facilitate gasification of low-grade coals, which are usually more difficult to gasify. Some negative characteristics of low grade coals like high ashes and sulfur contents could be counterbalanced by mixing them with some wastes with low ashes content and high volatile matter. The presence of wastes during coal gasification may be advantageous, as it increases gas yield, gas energy content and energy conversion, mainly because the gas produced has greater hydrocarbons content, but depending on waste type, the release of tar and of S, N and Cl compounds may be also favored. Gasification gas composition is influenced by the amount of waste content in the feedstock and also by the type of waste. Hydrocarbons and tar concentrations increase with the rise of some wastes tested, especially plastics, sewage sludge, MSW and RDF, which may limit some of syngas end-use applications. The adjustment of experimental conditions, like the rise of temperature and ER (equivalent ratio which represents the amount of oxygen in the input feed), allowed decreasing hydrocarbons, char and tars contents, but in presence of 20% (w/w) of plastics the release of tar and gaseous hydrocarbons was quite high, even at 900ºC and at an ER vale of 0.4, Pinto et al., 2008a. The increasing interest of using plastic wastes (scrap from cars, boats, computers etc.), spent rubber tyres and other forms of rubber would require the use of more complex syngas cleaning processes.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

6

Filomena Pinto, Rui Neto André and I. Gulyurtlu

Some biomass wastes like those from forest and wood byproducts usually contain low contents of sulfur, but coal may contain high levels of sulfur. Coal may present different sulfur-containing minerals, which undergo several transformations during coal gasification, thus releasing some sulfur as H2S, while other is retained the unburned carbon and released with the ash. The use of wastes with high contents of N, S and Cl, like sewage sludge and MSW, led to the release of high contents of H2S, HCl and NH3, even after increasing temperature and ER. Further reduction of these compounds can only be achieved with the addition of sorbents with elements like Ca, Fe, Mg and Si that by reacting with sulfur, and chlorine, promote theirs retention in the solid phase that stays inside the gasifier. Another possibility is the separation of wastes into fractions in order to gasify only the fractions with the suitable characteristics. The main challenges are the determination if separation is advantageous, considering both technical and economical benefits, and the definition of which type of separation is appropriate. Besides the cost associated with wastes separation and pre-treatment, these procedures may be advantageous, due to material recovery, such as metals, glass and paper, for instance, municipal solid waste (MSW) separation and pre-treatment of residue derived fuel (RDF), either when separation occurs before collection or when mechanical separation is done after collection. Clean RDF fractions with low contents of chlorine and high heating value could also compete with the biomass fuel market. The production of such fractions will be more viable in the years to come, due to the increasing requirements of recycling and separation.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

2.2. Gasification Temperature Gasification temperature is also a very important parameter for syngas composition. Several authors studied the effect of temperature: Gil et al. 1997, Pan et al. 1999, Pinto et al. 2003, Zhao et al. 2003, and André et al. 2005. In general the rise of temperature favors gasification reactions and thus increases the hydrogen release and reduces hydrocarbons formation. Gil et al. 1997 observed an increase in H2 release and a decrease in hydrocarbons contents with the rise of temperature during gasification of pine wood chips in a pilot scale installation. Hofbauer et al. 2001 also obtained the same trends for the FICFB gasification steam process. Pinto et al., 2008a reported that the rise of temperature was quite important to decrease the release of hydrocarbons, especially when plastic wastes were co-gasified, for instance the rise of temperature till 900ºC during co-gasification of 20% PE and 20% (w/w) of pine wastes in coal blends led to an increase in hydrogen concentration of about 60% and to reductions of around 30 and 63% in methane and heavier hydrocarbons concentrations, respectively. It was also observed a reduction in CO2 concentration, probably due to CO2 consumption dry reforming reactions of light hydrocarbons and tars, which are favored by temperature rise. The increase of temperature also reduces the formation of tars, a temperature higher than 900ºC should be used to decreases the risk of clogging in the reactor and in the pipes outside it. Tar contains several risk components and if syngas with high tar contents is burned some tar contaminants may be released to the atmosphere. On the other hand, at higher temperatures, above 1 000ºC, ashes may melt, sinter and cause damages inside the reactor. Co-gasification temperature of coal mixed with wastes, especially with high contents of S, N and Cl, also affects the release of precursors of pollutants like: H2S, NH3 and HCl. The

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Stage of Deployment of Syngas Cleaning Technologies…

7

formation of these compounds is a complex issue that depends on many parameters, besides temperature, therefore, there is some controversy about temperature effect. Kuramochi et al. 2005 predicted that the rise of temperature from 750ºC to 850ºC during biomass gasification favored the production of H2S. On the other hand, Pinto et al., 2009 observed that rise of temperature till 900ºC decreased the release of H2S. The rise of temperature also decreases NH3 content in syngas, as described, for instance by Paterson et al. 2002, Liu et al. 2003, Tian et al. 2002, Pinto et al., 2009. As the thermodynamic decomposition of NH3 is an endothermic reaction, it is favored by temperature rise. High temperature processes also favors NOx formation, increases operational costs and reduces efficiency. In relation to the influence of temperature on HCL release there is also some controversy. Li et al. 2005 stated that the rise of temperature led to higher chlorine release in the gaseous phase, independently of the coal type or chlorine form. Kuramochi, 2005 predicted that for temperatures higher than 550ºC, HCl release with temperature also depended on the presence of elements, such as: K, Al, Na and Si, and the competition among them to react with chlorine. Wei et al. 2005 studies about sewage sludge gasification showed that HCl did not vary much with temperature, when Na content was greater than that of K, due to the formation of silicates that appeared to form at high temperatures. Therefore, the analysis of the opposite effects of temperature on syngas composition and on gasification performance leads to the selection of gasification temperature in the range of 850 to 900ºC.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

2.3. Oxygen Flow Rate As gasification medium may be used air, or oxygen and/or steam. The use of air or oxygen has the advantage of getting the energy needed for gasification reactions through the partial combustion of the feedstock. Due to the air high nitrogen content, syngas produced in presence of air is nitrogen diluted and has a low heating value. The effect of air, or oxygen flow rate is usually analyzed by the equivalent ratio (ER), defined as the ratio between the amount of oxygen used and the stoichiometric oxygen needed for combustion of the feedstock. When oxygen is used instead of air, in oxy-gasification, the diluting effect of nitrogen is avoided, but as the production of oxygen has a high cost, gasification operating costs are also higher. Because of this, some gasification processes have been developed, using steam as gasification medium. But these processes have the disadvantage of also increasing operational costs, as the energy needed for the gasification processes has to be supplied by external sources. Pinto et al., 2009 analyzed the effect of rising ER during co-gasification of different types of feedstocks, either coals or wastes or blends of coal and wastes. The results obtained showed that the rise of ER allowed decreasing the release of tar and gaseous hydrocarbons, but hydrogen concentration in syngas also decreased, while CO and CO2 rose, due to partial combustion of the feedstock. Therefore, though gas yield increased, its heating value decreased and in the limit combustion prevailed against gasification. Narvaéz et al., 1996, studied the effect of increasing ER during biomass gasification at 800ºC and verified a reduction on tar emissions, through the increased availability of oxygen

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

8

Filomena Pinto, Rui Neto André and I. Gulyurtlu

on the pyrolysis zone. A decrease on hydrocarbons, H2 and CO contents, a consequent decrease on gas heating value and an increase of CO2 were also observed. The presence of steam in gasification medium is also important, as it increases the production of H2 through water gas reactions (4) and (5), water-gas shift reaction (7) and steam reforming reactions (8) to (11). Therefore, oxygen and steam mixtures are a good choice for gasification medium, as oxygen promotes partial combustion and supply the required energy for gasification process and steam favors gasification reactions and the formation of H2. Gil et al, 1997 suggested a H2O/O2 molar ratio around 3.0 and (steam+O2)/biomass mass ratio between 0.8 and 1.2 for a temperature between 800 and 860ºC. The results obtained by these authors and also those reported by Narvaéz et al., 1996 and Herguido et al., 1992 showed that the use of only steam favored tar formation, while the use of only air gave minimum tar amounts. Intermediate values were obtained by the use of steam and O2 mixtures, as gasification agent. Several researchers reported that the increase of ER allowed decreasing the release of NH3 in syngas, probably because NOx formation from fuel-N was favored, which decreased the nitrogen available to form NH3. Liu et al., 2003 stated that the rise of ER from 0.2 to 0.4 led to a reduction in NH3 released in syngas. Pinto et al., 2008a also observed that though the formation of NH3 decreased with the rise of ER, there was an increase of NOx, especially when ER values higher than 0.4 were used, remaining constant the total amount of nitrogen released to the syngas. However, Zhou et al., 2000 reported that during biomass gasification an increase of ER from 0.17 to 0.25 led to lower NH3 contents in the syngas and the release of NOx stayed bellow 10 ppmv, therefore, fuel-N released to the gaseous phase could mostly be converted to N2 in the presence of char, H2 and CO. The different statements reported by these authors are probably due to the effect of different parameters and processes used by each author, as in fact formation and destruction is a complicated issue that depends on many parameters, some of them acting in opposite directions. On the other hand, Tian et al., 2005 showed that an increase of steam in the gasification medium of several biomass types led to higher NH3, through hydrolysis of nitrogen compounds. Pinto et al., 2008a studied the effect of rising ER on H2S formation during co-gasification of coal with different kinds of wastes. In general the increase of ER led to a decrease in H2S content in syngas, probably because higher oxygen contents favored SOx formation from fuelS. However, in presence of some wastes, like cardoon and RDF, the rise of ER till values around 0.2 led to an increase in H2S content, which decreased for higher ER values. The first increase of H2S could be due to further conversions of char, which could release more H2S. The further increase of ER could have supplied the necessary oxygen to convert fuel-S into SOx through oxidation reaction. Nichols et al., 1989 also studied the effect of ER on the formation of H2S during coal gasification, and observed that the increase of ER above 0.5 led to a reduction in H2S and CS2 in the syngas, which was followed by higher SO2 and COS formation. In general the increase of ER led to a decrease of H2S and NH3 contents, but to higher formation of SOx or NOx, therefore, no great changes were observed in the percentage of sulfur and nitrogen released to the gas phase. On the other hand, as the syngas also has lower hydrogen contents and lower heating values with the rise of ER values, the selection of this parameter should be maintained around 0.2.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Stage of Deployment of Syngas Cleaning Technologies…

9

2.4. Gasifier Type Syngas yield and composition is also dependent on the type of gasifier used and on gasification technology. There are several types of gasifiers: fixed beds (BGL, Lurgi, EPIC), moving beds, either fluidized beds (Southern Co, KRW) or circulating beds and entrained flows (Shell, GE (Texaco), Conoco-Phillips (Dow/Destec)). Entrained flow gasifiers use very small particles and therefore need extensive and costly pre-treatments. As they operate at high temperatures, around 1300°C, the gas obtained has little or no hydrocarbons. Fixed beds are simple and reliable reactors, but have low and non-uniform heat and mass transfer and thus produce large amounts of char, tar and particulates. They usually have higher operational problems than fluidized beds. Depending on fluxes directions, fixed beds reactors may be co-current or counter-current, updraft or down-draft. Counter-current (updraft) reactors create large amounts of tar in syngas, because there is a larger amount of tar formed during fuel pyrolysis that is released without being decomposed. As stated by Bridgwater, 1994, fuel conversion to tar can be up to 10%. In downdraft reactors gas usually has lower tar amounts, because the gas goes through a hotter gasifier zone were tar decompose, hence final tar contents are lower than 1%, Bridgwater, 1994. Co-current and down-draft reactors are usually limited to very small sizes (1 MW), therefore, it is unlikely that such small plants can justify the cost of the emission control equipment needed for some wastes like MSW and/or RDF. In general, co-current and downdraft reactors require less gas cleaning although it is still necessary. Fluidized beds allows overtaking these advantages, as due to the presence of hot and inert bed materials, normally silica sand, they have uniform and high heat and mass transfer and may operate at lower temperatures than those needed by entrained flow gasifiers. Fluidized beds use temperatures around 800-900°C, therefore, syngas contains higher tar and hydrocarbons contents than those of entrained flows and needs catalytic upgrading. However, they are easier to design, built, operate and to control temperature.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

2.5. Presence of Catalyst The addition of catalysts directly into the gasification reactor usually improves syngas quality, as they reduce the release of tar and of heavier gaseous hydrocarbons, initially formed by pyrolysis and incomplete gasification reactions. Catalysts affect cracking, methanation and reforming reactions. Catalyst selection must consider not only its effectiveness for tar and hydrocarbons conversion, but also its resistance to deactivation, the possibility of regeneration, mechanical resistance to erosion and cost. The catalysts commonly used on gasification studies can be grouped in different types, namely: natural minerals (dolomite, olivine and other), nickel based catalysts and more expensive catalysts containing many other metals. Though, all these catalysts have been used inside the gasification reactor for research purposes, due to the adverse conditions to which they are subject within the gasifier, their use at these conditions is not advisable, unless their cost is very small. Due to the formation of sulfur compounds and of coke deposition on catalyst surface, it is expected a fast catalyst deactivation or poisoning. Therefore, only low cost minerals, which do not need to be regenerated for reutilization, are advisable to be used inside the gasifier. However, the use of different types of catalysts inside the gasifier is

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

10

Filomena Pinto, Rui Neto André and I. Gulyurtlu

reviewed, as the results obtained have given important information for the selection of the right catalysts to be used for syngas cleaning processes.

2.5.1. Use of dolomite as catalyst

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Dolomite (MgO•CaO) is one of the most studied catalysts, due to its low cost and proven activity on tar decomposition reactions, as it is referred by Sutton et al., 2001 and Devi et al., 2003. Dolomite is usually used after calcination, to release CO2 and to improve the catalytic effect. Corella et al., 2004a compared the catalytic activity of dolomite with other minerals, such as: calcined calcite (CaO) and calcined magnesite (MgO) and concluded that the highest tar reductions were obtained with calcined dolomite. Several authors have reported that dolomite was effective for tar reduction, though it has a low effect on the conversion of gaseous hydrocarbons, especially methane. Vassilatos et al., 1992, observed that the use of dolomite at a temperature in the range of 800-900ºC, was not enough for a complete destruction of naphthalene, one of the most abundant tar species, therefore, dolomite was not capable to guarantee the complete conversion of tar. The amount of tar reduction promoted by dolomite depends on the experimental conditions, but also on dolomite origin, mineral composition, surface areas and pore distribution sizes. Narvaez et al., 1996 reported tar reduction of around 40%, while Corella et al., 1999b stated tar reductions up to 80% at optimized experimental conditions. Orio et al., 1997, Delgado et al., 1997, and Vassilatos et al., 1992, stated that the increase of gasification temperature, improved the catalytic action of dolomite on tar reduction. Orio et al., 1997 compared the catalytic action of several Spanish dolomites with different compositions and origins. It was observed that the presence of higher Fe2O3 amounts and larger pore sizes increased tar conversions and produced higher gas yields. To increase dolomite catalytic activity on tar conversion, Pinto et al., 2007 prepared and tested a dolomite catalyst enhanced with Nickel, using the procedures described by Wang T. et al., 2005a. The results obtained showed that this catalyst had a better performance on tar reduction that dolomite, but worst than a Ni-based catalyst. However, this catalyst presented a better sulfur retention capacity than Ni-based catalyst, though worst than that of dolomite. After calcination, dolomite main component was CaO, which according to reaction (16), M being Ca, could have reacted with H2S to form CaS that remained in the gasification bed. MO + H2S ⇆ MS + H2O

(16)

Several authors have reported the effectiveness of dolomite in sulfur retention. Xu et al. 2005 reported that in the presence of CaO, higher retention of sulfur in the solid phase could be achieved, probably due to the formation of CaS. This was also proven by Sage and Welford 1997, as during co-gasification of sludge with coal, the sulfur retention reached 96% in the presence of dolomite. Similar results were obtained by Paterson et al. 2002, who achieved around 87% of sulfur retention in the solid phase with the addition of dolomite. Khan, 1989 also reported that the addition of CaO significantly reduced the release of H2S into the gaseous phase; however, the use of CaCO3 had relatively low influence, because due to the reduction conditions of gasification, significant calcination of CaCO3 to CaO did not occur. However, as reported by Park et al. 2005, sulfides (MS) produced from the H2S

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Stage of Deployment of Syngas Cleaning Technologies…

11

reaction with metal oxides, could further react either with CO, which is a major component of syngas, to produce COS or with H2O to produce SO2, which decreased sulfur retention inside the gasifier. Tijmensen et al. 2000 reported that sulfur could be also retained due to iron oxides presence, according to (17) to (19). However, these reactions could be reversed by the presence of steam and CO2 in hot gases and then affect sulfur retention. Because of this, zinc ferrites (ZnO.xFe2O3) and zinc titanates (ZnO.xTiO2) have shown to be more strong sorbents for desulfurisation, as reported by Zevenhoven et al. 2004. Fe3O4 + H2S ⇆ FeS + Fe2O3 + H2O

(17)

2Fe2O3 + H2S + 2H2 ⇆ FeS + Fe3O4 + 2H2O

(18)

2Fe2O3 + H2S + 2CO ⇆ FeS + Fe3O4 + 2CO2

(19)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Dolomite main compounds, CaO and Ca(OH)2, may also react with HCl, by reactions (20) and (21) and thus retain HCl in the solid phase, as CaCl2. CaO + 2 HCl ⇆ CaCl2 + H2O

(20)

Ca(OH)2 + 2 HCl ⇆ CaCl2 + 2 H2O

(21)

These reactions may be reversed due to the water steam that exists in the gasification medium. Reaction (20) and (21) are also affected by the presence of elements such as: Si, Al, alkaline earth metals and heavy metals, as they might compete with CaO to react with HCl. For instance, the presence of K in the feedstock could produce gaseous KCl instead of HCl. On the other hand, CaCl2 starts to decompose at 740 ºC, which is a temperature lower than that commonly used during gasification, thus some CaCl2 may decompose and release some HCl, which would decrease chlorine retention in the solid phase. Besides, CaC12 has a melting point at 772 ºC, and the lowest liquid temperature for the CaC12-CaO system is 750 ºC with a eutectic composition of about 6 % (mol) CaO. Consequently, a liquid phase of CaCl2 saturated with CaO might be formed at gasification temperatures, which could have reduced the sorption capacity of CaO. Moreover, the sorption capacity of CaO might also be decreased by the reversibility of reactions (20) and (21) at high temperature, causing the release of HCl, as stated by Weinell et al. 1992. Corella et al. 2004a found that though there was some NH3 reduction when dolomite was used during gasification, NH3 content in the fuel gas was higher with dolomite than with olivine, probably because dolomite iron content was lower than that of olivine and iron oxides can catalyze NH3 decomposition into N2 and H2. These authors also defended that dolomite was more active for tar cracking than olivine, hence dolomite would crack more N-tar than olivine, hence producing more NH3. Pinto et al., 2008a stated that NH3 reductions in presence of either dolomite or olivine were low, thus other catalysts should be used for effective NH3 abatement. Dolomite iron content was lower than that of olivine and though iron oxides can catalyze the reaction of N2 and H2, the results obtained seem to show that iron oxide content did not

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

12

Filomena Pinto, Rui Neto André and I. Gulyurtlu

seem to be a decisive issue in the NH3 amount released to the gasification gas. According to Corella et al. 2004a, as dolomite was more active for tar cracking than olivine, dolomite would crack more N-tar than olivine, hence producing more NH3. This hypothesis also allowed explaining why dolomite was more efficient in reducing NH3 contents in presence of pine wastes and less capable of that in presence of sewage sludge, as higher tar contents were also produced with this last waste. The main disadvantage of dolomite is deactivation, especially inside fluidized beds, due to the loss of mechanical resistance and coke deposition, as reported by Vassilatos et al., 1992, and Sutton et al., 2001. The low mechanical resistance is responsible for particles erosion, resulting in an increased entrainment of catalyst fine particles that must be removed before a further treatment of syngas. Coke deposition may be decreased by using steam as gasification agent, because it promotes coke reforming reactions, or by higher air flow rates, as it favors coke oxidation at the surface or in the inner pores, decreasing dolomite deactivation. The presence of earth-alkaline oxides resulting from the dolomite calcination also catalyzed the conversion of coke.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

2.5.2. Use of olivine as catalyst Olivine, (Mg, Fe)•2SiO4, with a Mg/Fe ratio of 9/1, is another natural mineral that has been used to promote tar reduction. It is not clear if olivine presents a higher catalytic activity than dolomite, as different authors report opposite conclusions, probably due different origins, pore size distribution and chemical composition. Rapagnà et al., 2000, reported that olivine presents a tar reduction activity similar to calcined dolomite, leading to reduction up to 90%. On the other hand, Corella et al., 2004a, stated that olivine was 1.4 times more effective on tar conversion, but dolomite produced 4 to 6 times more fines during gasification than olivine. These authors also found similar gas compositions when dolomite or olivine where used or even without any of those catalyst, depending on the right choice of experimental conditions. Another controversy is about olivine calcination. Calcination may affect olivine catalytic activity, by changing both the chemical composition and the physical properties. Corella et al., 2004a, reported that calcination destructed the porous structure of olivine, reducing its activity on tar decomposition. On the other hand, Devi et al., 2005 referred that, after calcination at 900ºC during 10h, olivine presented a higher catalytic activity, leading to naphthalene conversions of about 80%, which were significantly higher than those obtained with natural olivine. Though olivine contains iron and Najjar et al. 1995 stated that iron compounds promoted sulfur retention inside the gasifier. However, Pinto et al., 2008b reported that olivine did not lead to great reductions in H2S content in the fuel gas, probably, because olivine porosity is not high enough and/or the iron content in olivine was not high enough to achieve further sulfur capture. Calcined olivine should be more efficient for H2S reductions, because through calcination different metallic compounds are converted into their oxides, which could favor sulfur retention, as metallic sulfides, MS, according to reaction (16), M being mainly Ca, Fe or Mg. However, the performance of calcined olivine in sulfur was not always better than that of natural olivine, depending on the type of feedstock to be gasified, sometimes it was even worst, as reported by Pinto et al., 2008b. Olivine may also reduce the release of HCl, due to the presence of FeO(OH), which may react with HCl, by reaction (22). During olivine calcination, FeO(OH) may be converted into

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Stage of Deployment of Syngas Cleaning Technologies…

13

iron oxide and further lead to the formation of silicates, which could explain the lower chlorine retentions obtained with calcined olivine, as reported by Pinto et al., 2008a. FeO(OH) + 3 HCl ⇆ FeCl3 + 2 H2O

(22)

Pinto et al., 2008a observed that gasification of pine, sewage sludge and bagasse in presence of dolomite led to lower NH3 reductions than those obtained when olivine were used, probably due to iron content of olivine, either natural or calcinated. The results obtained by Dou et al. 2002 also showed that the Fe-based catalyst removed about 35% of NH3. However, NH3 reductions obtained by Pinto et al., 2008a when either dolomite or olivine was used were very low and showed more effective catalysts should be used for NH3 abatement. Olivine is less abundant than dolomite and do not exist in many countries, which is another drawback of olivine, as the operation costs of gasification units would increase if they were dependent on olivine imports. To increase olivine catalytic activity on tar conversion, some authors have enriched olivine with nickel. Corella et al., 2004a, used a nickel enriched olivine with a mass fraction of 3.7% of Ni and reported that this catalyst had a fast deactivation, requiring periodical regeneration. Though the catalyst showed a good activity on tar conversion, the best results were obtained only on steam gasification, with a fluidized bed of 100% of olivine. On the other hand, during air gasification with a silica sand fluidized bed the catalyst activity was severely reduced.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

2.5.3. Use of other minerals as catalysts or sorbents As reported by Devi et al., 2003, limestone was one of the first gasification catalysts, due to its benefit of avoiding bed agglomeration. Delgado et al., 1996, studied the catalytic action of different calcium containing minerals on tar reduction and compared it with the catalytic action of commercial catalysts. The results obtained showed that the highest tar reductions were obtained with commercial catalysts and the performance of calcium containing minerals was the following: calcined dolomite > calcined calcite > calcined magnesite. The same trend was also observed for gas yields, however, these three catalysts led to similar gas compositions. Delgado et al., 1997 also reported that catalytic activity was very dependent on surface area BET, which was affected by experimental conditions used during calcination like: temperature, time and atmosphere used. This, together with different origins and mineral composition could explain results differences obtained by different authors. Sutton et al, 2001 also reported that catalysts with different actions on tar reduction and on gas yields led to similar gas compositions. The catalytic action of different minerals on biomass gasification was the following: potassium carbonate > sodium carbonate > trona– Na3H(CO3)2·2H2O > borax–Na2B4O7·10H2O. In spite of the highest activity of potassium carbonate, the highest agglomeration of the particles inside the reactor was also produced in presence of potassium carbonate, thus leading to the highest erosion of the reactor wall. To decrease particle agglomeration, Sutton et al, 2001 suggested catalyst impregnation on biomass, instead of using a simple mixture, as it could also reduce the deactivation associated with coke deposition on catalyst surface.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

14

Filomena Pinto, Rui Neto André and I. Gulyurtlu

Encinar et al., 1998 studied different materials: LiCl, NaCl, KCl, KCO3, AlCl2·6H2O and ZnCl2 on gasification of wastes from wine production industry. All these materials with the exception of KCl led to an increase on gas and char yields. The highest gas yields were obtained with KCO3. The presence of zinc favored H2 production, however, the increase of zinc concentration led to higher char productions and lower liquids and gas yields, as also reported by Sutton et al, 2001. The use of ZnO has proven to be effective for chlorine retention in the solid phase, as it may react with HCl by reaction (23), which is affected by the steam present in the gasification medium. ZnO + 2HCl ⇆ ZnCl2 + H2O

(23)

Besides, some of the ZnCl2 formed by reaction (23) may also react with H2S, through reaction (24), due to H2S presence in the reaction medium. However, Gibbs free energy and equilibrium constant values of reaction (24) at various temperatures, calculated by Gupta et al., 2000, showed that reaction (24) is strongly favored at temperatures lower than 650 °C. ZnCl2 + H2S ⇆ ZnS + 2 HCl

(24)

On the other hand, at the gasification temperature ZnCl2 is liquid and due to its significant vapor pressure, it may escape from the sorbent. Additionally, ZnCl2 being in a liquid phase, could favor the reaction with H2S than with solid ZnO, thus leading to further reductions of H2S concentration. ZnO + H2 → Zn + H2O

(25)

ZnO + CO → ZnC + O2

(26)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

According to Park et al. 2006, ZnO could also react with H2 and CO from the fuel gas through reactions (25) and (26). Hence, the metallic zinc gradually diffused to the surface and vaporized, for this reason and to improve ZnO performance, some inorganic oxides, like TiO2, have been added to increase ZnO stability, mechanical strength and capacity of retention.

2.5.4. Use of nickel based catalysts As the reforming reactions may be catalyzed by elements from groups 8, 9 and 10 (formerly group VIII) of the periodic table, Yoshiniri et al., 1984 compared the catalytic activity of different metallic oxides like V2O5, Cr2O3, Mn2O3, Fe2O3, CoO, NiO, CuO, MoO3 supported in Al2O3. Though these oxides increased gas yield, the highest catalytic activity was presented by NiO. According to Corella et al., 2004a, nickel promotes the dissociation of H2O into OH• radicals, which act on the opening of aromatic and poly-aromatic rings which are the main constituents of tar, thus decreasing its content. Several authors have also tested the catalytic effect of different nickel compounds on the gasification of biomass and/or coal. Baker et al., 1987, studied the catalytic effect of several

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Stage of Deployment of Syngas Cleaning Technologies…

15

commercial catalysts containing nickel and found that all the catalysts were effective for tar reduction, but some of them were more easily deactivated by carbon deposition. Aznar et al., 1993, also studied the catalytic effect of several commercial catalysts. The most effective catalysts for heavier hydrocarbons decomposition were also those more adequate for tar abatement. Kinoshita et al., 1995, observed that catalyst activity was improved by using higher temperatures, longer residence times and higher ER values. Baker et al., 1987, found that some catalysts were more easily deactivated by carbon deposition than others and that the regeneration of some catalysts proved to be difficult, as some materials have sinterization tendency, which leads to activity loss. Yamaguchi et al., 1986 tested nickel catalyst, supported in alumina and also verified that it presented a fast activity loss, due to carbon deposition and catalyst sinterization. Other authors prepared nickel based catalysts enriched with other elements. Richardson and Gray, 1997 tested nickel and molybdenum catalysts enriched with alkaline metals, added as KNO3, KOH, NaOH and LiOH during biomass gasification with the aim of increasing the catalyst activity and reducing coke deposition. KNO3 proved to be useless, while the other chemicals showed to be effective on acidity reduction but did not have effect on coke deposition. Garcia et al., 2002, synthesized several nickel and magnesium catalysts supported in alumina and tested them during biomass gasification. The best tar reduction was obtained with NiMgAl2O5, being followed by NiMgAl4O8, the former catalyst also presented the higher stability and the highest initial activity. Bangala et al., 1998 impregnated a nickel catalyst over an alumina support with MgO, TiO2 or La2O3 and observed that gas yield increased for values of nickel up to 15%, decreasing afterwards, though coke deposition increased with higher nickel contents. The highest gas yields and the largest reduction on coke deposition were obtained with La2O3. Several authors, for instance Wang W. et al., 1999, Devi et al., 2003, Dou et al. 2002 and Pinto et al., 2008a, stated that nickel based catalysts suitable for tar abatement were also effective for NH3 reduction. Pinto et al., 2008a tested different types of catalysts and observed that the highest NH3 reductions were obtained in presence of a Ni-Mg catalyst. Reductions between 45% and 55% were obtained depending on the types of waste gasified: pine, bagasse, sewage sludge and straw pellets. Dou et al., 2002 also reported that the Ni-based catalysts had higher activity, and more than 88% of NH3 present in the feed gas was removed. The main disadvantage of nickel catalysts is the loss of activity, due to coke deposition and H2S poisoning. These problems can be overtaken by either increasing operating temperature or rather by the use of these catalysts in a secondary reactor after syngas cleaning for sulfur removal, as suggested by Baker et al., 1987, Devi et al., 2003 and Corella et al., 2004a.

2.5.5. Use of other catalysts Several other catalysts have been prepared and tested at small-scale laboratory gasifiers. Rapagnà et al., 2002, prepared and tested a tri-metal catalyst, LaNi0.3Fe0.7O3, with a strong interaction between nickel and the oxide phase that led the formation of small nickel particles on the catalyst surface. This catalyst decreased coke deposition and enabled catalyst regeneration by calcination. It led to higher gas yields and hydrogen concentrations and to a marked decrease on tar formation and lower methane contents on gasification gas.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

16

Filomena Pinto, Rui Neto André and I. Gulyurtlu

Martinez et al., 2003, also prepared 3 catalysts with a nickel mass fraction of 33% and different amounts of La and Al. All of them were adequate to be used in fluidized bed gasifiers, without significant elutriation or erosion and lower coke deposits than those obtained with a Ni-Al catalyst. Ni-Al-La catalysts produced a gas with higher amounts of CO, CO2, CH4, C2 and higher gas yield than those obtained with a Ni-Al catalyst, though no significant changes were observed on H2 concentration. Other catalysts were prepared from chemicals that are too expensive to allow catalyst use at a pilot or industrial scale gasification installations, despite the good results reported. Asadullah et al., 2003 tested a catalyst containing Rh/CeO2/SiO2, which allowed obtaining higher conversions of fuel carbon to gas and higher CO, H2 and CH4 contents. Asadullah et al., 2004 compared the activity of this catalyst with that of dolomite and of a commercial steam reforming catalyst and observed that the performance of the commercial catalyst on tar reduction was better than that of dolomite, but only in presence of Rh/CeO2/SiO2 no tar was detected in syngas, thus allowing syngas use on gas turbines, even when temperatures of 700ºC (973 K) were used. As almost all the initial carbon was converted into syngas phase, there was nearly no coke deposition and so the catalyst was not deactivated.

3. HOT SYNGAS CLEANING PROCESSES Syngas cleaning processes are often referred as gas conditioning, which means the removal of unwanted impurities and involves multi-step and integrated processes. Three main strategies may be used for syngas cleaning: hot gas conditioning, wet scrubbing, or dry/wetdry scrubbing, as shown in Figure 1. These processes were reviewed by Milne et al., 1998, some of them were used in demonstration plants, but few are commonly accepted by technoeconomic criteria.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Dry/Wet-dry Scrubbing : Particle Cyclones Cooling Wet-Dry Contacters Dry/Wet-dry Scrubbing : Particle Cyclones Cooling Wet-Dry Contacters Absortion and/or Adsortion in Solids Wet Electrostatic Precipitators

Syngas Cleaning Processes

Wet Scrubbing: Particle Cyclones Cooling Towers Venturi Wet Cyclones demisters, granular Filers, Wet Electrostatic Precipitators Hot Gas Conditioning: Particle Cyclones Hot Filters Thermal or Catalytic Cracking Catalytic Reforming Water-gas Sfit

Figure 1. Syngas cleaning processes.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Stage of Deployment of Syngas Cleaning Technologies…

17

When the syngas is used at atmospheric and temperature conditions wet scrubbing is a possibility. In wet scrubbing processes, after the cyclone, syngas goes into cooling or scrubbing towers for heavy tars condensation. Venturi scrubbers are usually the second wet scrubbing units. Other tar separation units include demisters, granular filers and wet electrostatic precipitators (ESP), which are more expensive than other tar removal systems. The main disadvantages of wet scrubbing processes are tar disposal and the formation and accumulation of wastewater with organics, inorganic acids, NH3 and metals. Hot gas cleaning processes are advisable when syngas is to be used at high temperature, such as in cracking or reforming units or when it goes to a shift reactor, because these processes have high inlet temperatures. Cooling the gas in wet scrubbing units and reheating it afterwards would result in lower energy efficiency and higher operational costs. The first steps of hot gas cleaning processes are filters and separation devices in which the high temperature of the syngas can partly be maintained. Hot gas filtration with fabric, ceramic, or metallic filters to remove near-dry condensing tar particles is also possible and is usually combined with catalytic reforming.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

3.1. Particulates Removal Technologies Syngas utilization in boilers is the most permissive regarding particulates levels, as concentrations as high as 1000 mg/Nm3 and particle size around 10 µm are accepted. However, most of other syngas applications require low levels of particulates, as shown in Table 1. Therefore, it is very important to reduce particulates content in syngas, especially when fluidized beds are used, due to the production of smaller particle sizes. In fact, the representative range of particulate content for fluidized beds and circulating beds are 2-20 and 10-35 g/Nm3, respectively, while for fixed bed, either downdraft or updraft, are usually in the range of 0.1-1.0 g/Nm3. Particulates may be removed from syngas by several mechanical methods such as: cyclones, bag filters, baffle filters, ceramic filters, fabric filters, rotating particle separators, wet electrostatic precipitators and water scrubbers. However, depending on syngas utilization and syngas characteristics, including particulate and tar type and contents, some of these options may not be adequate. Cyclones are the most used devices for particulates removal, because they are commercially available at relatively low prices, are easy to operate, can work over a wide range of temperatures and are quite efficient in removing particulates with larger particle sizes, above about 5 µm. If used in series, cyclones can remove more than 90% of particulates above about 5 µm in diameter at negligible pressure drops. Cyclones can also remove some particulates with particle sizes in the range of 1-5 µm. However, they are not appropriate for sub-micron particulates removal. Some tars and alkali material are also reduced from the gas stream by condensation in cyclones, however, the vaporized forms of those constituents remain in the gas stream. Another important advantage of cyclones is the possibility of working at high temperatures, retaining the sensible heat in the syngas, which is especial advantageous when syngas is going to be further cleaned by thermal processes. As cyclones are not enough to remove sub-micron particulates, other devices are still needed, which may be placed after cyclones.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Filomena Pinto, Rui Neto André and I. Gulyurtlu

18

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Table 1. Limits permitted for syngas composition as a function of its application. Impurity

Boiler

Gas Engines

Gas Turbines

Fuel cells

Particulate (mg/Nm3) Particle size (µm)

1 000

< 50

< 15

< 0.1

10

< 10

Fe > Co > Rh > Ni > Ir > Pt > Pd. Only ruthenium, iron, cobalt, and nickel have catalytic characteristics to be used as FT synthesis catalysts, but nickel produces too much methane, ruthenium is too expensive and its worldwide reserves are insufficient for large-scale industry, therefore, cobalt and iron are the most suitable catalysts to be used in FT synthesis. Iron catalysts are cheaper, more tolerant to sulfur compounds in the syngas, catalyst lifetime is short and produce more olefin products and alcohols. Cobalt catalysts have the advantage of longer life, higher conversions, they are more reactive for hydrogenation and produce less unsaturated hydrocarbons (olefins) and alcohols than iron catalysts. The type of catalyst used also affects products selectivity and operating conditions. Selectivity depends on catalyst type of metal, iron or cobalt, catalyst support, preparation, pre-conditioning and age. It also depends on syngas H2/CO ratio, temperature, pressure and reactor type. According to Tijmensen et al., 2002 higher partial pressures of H2 and CO lead to higher selectivity towards hydrocarbons with more than five carbons atoms (C5+), while lower partial pressures of H2 and CO reduced the selectivity to these hydrocarbons. Fischer-Tropsch operation conditions that have to be used also depend on catalyst type and performance. Both operating conditions and catalysts can be adjusted to favor the production of mainly low or higher molecular weight hydrocarbons. FT synthesis may be done at high temperature (300 to 350ºC) in a fluidized bed reactor in the presence of ironbased catalyst to produce hydrocarbons in the range C1-C15, which may be used as liquid fuels or to produce valuable chemicals, like R-olefins, alcohols, acetic acid, and ketones including acetone, methyl ethyl ketone and methyl isobutyl ketone. In the low temperature (200 to 250ºC) FT synthesis either iron and cobalt catalysts can be used to produce linear long chain hydrocarbon waxes and paraffins and sulfur-free diesel fuels. According to Khodakov et al., 2007 the low temperature process with cobalt catalyst is a better option due to the suitable performance and characteristics of this catalyst. These authors reviewed different methods for preparing and analyzing cobalt catalyst to be used in FT synthesis. Khodakov, 2008 also reviewed the influence of cobalt particle size and cobalt phase composition, catalyst support and support texture and promotion with noble metals on Fischer-Tropsch reaction rates, hydrocarbon selectivity and catalyst stability. This author concluded that it is of most importance to produce a significant concentration of stable cobalt metal surface sites, cobalt metal particles in the activated catalysts should not be smaller than an optimum size (60–80Å) and that the repartition of cobalt surface sites in a catalyst grain could be essential in attaining high yield of hydrocarbons. Catalyst deactivation and production of undesired mixed compounds could also be controlled by adjusting the experimental parameters of catalyst synthesis, pretreatments and FT reaction. Cobalt supported catalysts are a good choice for FT synthesis, but significant amounts of this expensive catalyst are required, therefore, the cost of overall FT could be decreased by the use of more efficient catalysts. Boerrigter et al., 2004 also studied the production of liquid

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Stage of Deployment of Syngas Cleaning Technologies…

33

fuels from biomass, through Fischer-Tropsch synthesis, using a cobalt-based catalyst at a temperature of 227°C and pressure of 30 bar and obtained C5+ selectivity around 95% (liquid and wax), when H2/CO ratios in the range of 0.8-2.1 were used. The other 5% are gaseous C1C4 products, whose energy can be used to generate electricity. The heavier FT products can be selectively converted into fuels in a hydro cracking step with 98% efficiency. By modifying the catalytic properties product selectivity may be increased, for instance by coupling the Co or Fe catalyst with a ZSM-5 zeolite catalyst it is possible to produce high octane gasoline by FT synthesis, as the longer hydrocarbons chains will be cracked, thus producing gasoline range fuel with C5 to C12 paraffins and aromatics. According to Martinez et al., 2005 Fe catalysts supported on ZSM-5 had higher alkane carbon distributions for gasoline range fuel than conventional Fe supported catalysts. Fischer–Tropsch (FT) synthesis is operated commercially at Sasol in South Africa using coal-derived syngas and at Shell in Malaysia using natural gas-derived syngas. Sasol uses iron catalysts and operates several types of reactors, being the slurry bubble column reactor the most versatile. In Malaysia, Shell operates the SMDS (Shell Middle Distillate Synthesis) process, which uses a cobalt-based catalyst in multi-tubular fixed bed reactors and produces heavy waxes. In 2003 Shell started the engineering for a 75,000 bbpd SMDS plant in Qatar, while Sasol has a 30,000 bbpd cobalt-based SSPD plant under construction in Qatar, Boerrigter et al., 2004. However, FT synthesis from biomass-derived syngas has received little attention so far, though liquid fuels produced from biomass are carbon neutral transportation fuels. There are three main kinds of FT reactors: fixed bed, fluidized bed reactor and the slurry phase reactor. According to many authors fixed beds and the slurry reactors are the most promising. The main disadvantage of the slurry reactors is the need for catalyst separation from wax, which is complex and they are more sensitive to sulfur poisoning, requiring thorough syngas cleaning. But, fixed beds need higher maintenance and intensive labor due to periodical catalyst replacement, Tijmensen et al., 2002. Recently several authors have studied FT synthesis from bio syngas, with the aim of increasing process efficiency and decreasing process costs. Wang Z. X. et al., 2007 studied FT synthesis in fixed-bed flow reactor and found that the Fe/Cu/Al/K catalyst with 100:6:16:6 wt ratio was a promising one for direct synthesis of liquid bio-fuel from bio-oil-syngas. The optimum experimental conditions were 280-300 °C, 1.0-2.0 MPa, contact time higher than 12.5gcathmol-1 and CO2/(CO+CO2) ratio below < 0.5. Therefore, CO2 content had to be decreased through methane co-reforming, CO2 removal and reverse water-gas shift reaction. Visconti et al., 2009 also found the importance of decreasing CO2 content when studying FT synthesis, using a cobalt-based catalyst at 220ºC, 20 bar and H2/COx ratio between 2.45 and 4.9 mol/mol. Though CO2 showed a higher reactivity than CO, it led to products with over 90% of methane. CO2 hydrogenation inhibited the chain growth, thus favoring the methanation reaction, while in presence of CO and due to the competition between CO and CO2 for the adsorption on the catalyst active sites, CO2 was hardly hydrogenated and behaved as an inert species. On the other hand, Pirola et al., 2009 prepared a Iron-based catalysts supported on silica with high amounts of metal (10–50%wt) and found that CO conversion and yield to hydrocarbons increased with iron content in the catalyst and also with the presence of promoters like K and Cu. The increase of the reaction temperature rose CO conversion, but the selectivity towards undesired products (CO2 and CH4) also increased significantly.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

34

Filomena Pinto, Rui Neto André and I. Gulyurtlu

The FT products are totally free of sulfur and nitrogen and contain very few aromatics, which are all normally found in mineral oil products. Different types of hydrocarbons (paraffins and olefins) with a wide range of molecular weights and oxygenated compounds like alcohols are obtained, according to reactions (27) to (29). Therefore, products need to be separated in different fractions and up-graded, before their utilization as fuels. Usually FT products require hydrocracking, through which double bonds are removed and afterwards hydrogen promotes chemical bonds cracking, depending on experimental conditions, mainly diesel or kerosene are produced. As Fischer-Tropsch facilities require large capital investments and according to Boerrigter et al., 2004 the resulting product is two to three times more expensive than mineral diesel, large fuel production capacities of more than 1000 MWth are required to make the products competitive with other alternative renewable transportation fuels, such as bioethanol and bio-diesel. The energy content of FT flue gas, containing unconverted syngas and C1 to C4 products, could be converted into electricity, by using this flue gas in gas and steam turbines for energy production. Streams of the whole process, including the exhaust gas from the gas turbine could be used for superheated steam production to be introduced in a steam turbine to generate electricity. Tijmensen et al., 2002 also studied the technical and economical feasibility of biomass integrated gasification–Fischer Tropsch (BIG-FT) processes and they also suggested the use of the combined cycle. Though the production costs of Ficher-Tropsch fuels are currently higher than those of diesel, these fuels contain little or no contaminants, such as sulfur and aromatics in opposition of what happens with gasoline and diesel, Wang L. et al., 2008. On the other hand, with largescale production, costs are expected to decrease, Tijmensen et al., 2002. Viability of more advanced BG-FT biofuels units would depend on market support mechanisms like lower biomass cost and incentives for the production of biomass-derived transportation fuels to be used in cars, trucks, and buses. Therefore, the main challenges for Fischer Tropsch synthesis are: the development of large-scale pressurized biomass gasification systems, improvements in syngas cleaning processes and the development of selective catalysts towards gasoline production at milder temperature and pressure conditions.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

4.3. Synthesis of Methanol and Dimethyl ether Methanol synthesis interest has increased lately, because of its potential as a fuel and also because methanol may be used for the production of other chemicals. Methanol can be used directly or blended with other petroleum products and can be an alternative fuel to gasoline and diesel, as it is a clean liquid fuel, without little or no contaminants such as: sulfur, nitrogen or chlorine compounds. Methanol can be synthesized from syngas, by converting CO, CO2 and H2 into methanol through reactions (30) and (31). The right molar ratios of H2/CO and CO2/CO should be used in presence of a suitable Cu-Zn or Cu–ZnO/Cr2O3catalyst, at 220 to 300 °C and 40 to 60 MPa. Cu catalysts for methanol synthesis undergo slow deactivation by sintering and poisoning by sulphur. The presence of chlorine in syngas causes sintering of the Cu catalyst. CO + 2H2 ⇆ CH3OH

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

(30)

Stage of Deployment of Syngas Cleaning Technologies… CO2 + 3H2 ⇆ CH3OH + H2O

35 (31)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Methanol production is higher when H2-CO-CO2 mixtures are used, due to activity and selectivity reasons, the required stoichometric ratio for (H2-CO2)/(CO+CO2) should be slightly above 2. Methanol synthesis also produces by-products such as: methane, dimethyl ether, methyl formate, higher alcohols and acetone. Methanol is used to produce a range of other chemicals and fuels including olefins, gasoline, dimethyl ether, methyl tert-butyl ether, acetic acid and formaldehyde. After purification methanol may be used to produce gasoline by Mobil Methanol to Gasoline process - MTG, according to reactions (32) to (35). Mobil also developed another process, Mobil Olefin to Gasoline/Distillate – MOGD, Radtke et al., 2006. While the products obtained by Fischer-Tropsch synthesis usually have higher contents of paraffins and olefins, hydrocarbons produced, from the previously mentioned synthesis, usually present high content of aromatic compounds, a distillation range similar to those of gasolines and a high octane number, which are good characteristics to be used as fuels. Different types of compounds may be formed, through reactions (32) to (35), depending on operating conditions and catalysts used, usually zeolites. MTG process uses ZSM-5 at 350 °C and 20 atm and around 85% of the products is gasoline, while the other 15% is light petroleum gas. The gasoline produced contains around 40% of aromatic compounds. 2 CH3OH ⇆ ( CH3 )2 O + H2O

(32)

2 CH3OH + ( CH3 )2 O ⇆ light olefins + H2O

(33)

light olefins ⇆ heavy olefins

(34)

heavy olefins ⇆ aromatics, paraffins, cycleparaffins

(35)

One of the main challenges in methanol synthesis is the development of lower cost and more efficient catalysts to optimize the reaction process. The properties of catalysts used in the synthesis of methanol have also been extensively studied, concerning the role of active sites involved in the catalysts, the effect of addition of various promoters, and the reaction mechanism. Cu is an important active catalyst component. Some studies showed that Cu/ZnO-based catalysts could be active and stable for a long period in a continuous methanol synthesis operation from CO2-rich syngas. New catalysts based on nickel, copper, and alloys and also ultrafine particle catalysts have been proposed. Other aspect that should be further investigated is the effect of the feed composition on the nature of the catalyst surface and on the catalyst activity. Another important issue is the carbon source for methanol synthesis The use of biomass as an energy resource reduces greenhouse gas emissions, therefore, methanol synthesis from biomass syngas is an environmentally friendly method of biomass utilization. Several biomass to methanol demonstration projects have been developed, such as the Hynol project in the United States, the BioMeet and Bio-Fuels projects in Sweden, and the BGMSS project in Japan. The SVZ Recycling Project in Schwarze Pumpe, in Germany, has used low or negative value carbon sources including coal, petroleum coke, high sulfur fuel oil, and waste organic

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

36

Filomena Pinto, Rui Neto André and I. Gulyurtlu

materials to produce syngas to be used partially in methanol synthesis and in part for a gas turbine combined cycle power generation system. As described in BTG, 2000 this installation produces 68 MWe and has an annual production of 130,000 tons of methanol, equivalent to a total of 250 MWe. Methanol and dimethyl ether (DME) can be alternative fuels to gasoline and diesel, as they are cleaner fuels. Methanol may also be used to synthesize DME through reaction (36) and in presence of a suitable catalyst such as Cu/ZnO/Al2O3+Y-Al2O3 as referred by Ng et al., 1999.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

2CH3OH ⇆ CH3OCH3 + H2O

(36)

Recent research about DME production involves the development of bi-functional catalysts to produce DME in a single gas-phase step. Ng et al., 1999 defended the combination of methanol synthesis with the DME production reaction, as by effective removal of the products from the methanol synthesis reaction, it was possible to minimized the reverse reaction and increase the overall methanol yield. These authors used a 2/1 ratio of the commercial CuO/ZnO/Al2O3 catalyst to the үalumina. The rise of hydrogen feed concentration enhanced the selectivity to methanol, but decreased that of DME. CO2 content feed led to strong synergy in total methanol production. Tao J.-L. et al., 2001 studied the performance of other catalysts like Cu/ZnO/Al2O3/Cr2O3 and H-ZSM-5. Tao J.-L. et al., 2001 combined the stability of ZSM-5 catalysts with Cu-ZnObased methanol synthesis catalysts and developed stable hybrid catalyst, Cu-ZnO-Al2O3Cr2O3 + H-ZSM-5 (SiO2/Al2O3 = 80). The activity of this catalyst was satisfactory, as the total yield of DME and methanol was higher than 26% with over 90% of DME selectivity. Methanol may be produced in large quantities by the conversion of natural gas or petroleum into a synthesis gas. These installations should have very large sizes to take advantage of the economies of scale. It is possible to produce clean synthesis gases from biomass using similar processes to those used in coal-based systems, however, the smaller biomass facilities have not been economically competitive, Stevens, 2001. These processes usually need to be large, because of the dominant economy-of-scale effect in biofuels synthesis and upgrading. However, large plant size means that higher amounts of feedstocks have to be transported from longer distances, thus higher feedstock costs are used. This problem may be diminished by biomass pre-treatment near the geographical origin of the biomass that converts the initial feedstock into a material easier to transport. The main challenge is to develop integrated systems that can be economical on a scale compatible with the availability of biomass feedstocks. If petroleum and natural gas prices increases and reaches high values, these economic evaluations will be changed and the production of liquid fuels from biomass may have an opportunity.

4.4. Synthesis of Ethanol Most of the ethanol produced nowadays is through fermentation, however, the production of fuel-grade ethanol is expensive, energy-inefficient, due to the energy spent in distillation processes and the microorganisms used in fermentation process are not suitable to metabolize

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Stage of Deployment of Syngas Cleaning Technologies…

37

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

completely the 5-carbon pentose sugars derived from lignocellulose or wood, Subramani et al., 2008. New fermentation processes to convert 5-carbon and 6-carbon sugars into ethanol are being developed. Another option for ethanol production is through the conversion of syngas, which can be converted into ethanol and higher alcohols, either directly or via methanol as an intermediate. Different methods may be used: a) direct conversion of syngas by reaction (37) and (38), conversion of methanol either by b) reductive carbonylation, reaction (39) or by c) methanol synthesis and carbonylation to acetic acid followed by acetic acid hydrogenation to ethanol reactions (30), (40) and (41). 2CO + 4H2 ⇆ C2H5OH + H2O

(37)

2CO2 + 6H2 ⇆ C2H5OH + 3H2O

(38)

CH3OH + CO + 2H2 ⇆ C2H5OH + H2O

(39)

CH3OH + CO ⇆ CH3COOH

(40)

CH3COOH + 2H2 ⇆ C2H5OH + H2O

(41)

The direct conversion of syngas to ethanol has been much studied and as reported by Spivey et al., 2007 the methane formation should be limited, by using the right experimental conditions (temperatures below roughly 350 ºC and pressure around 30 bar) and catalysts, as otherwise ethanol yield is very low. Though processes b) and c) have been developed to pilot scale, none of them was commercially developed. All processes are accompanied by secondary reactions that lead to a mixture of different products, including methane, by reaction (15), C2–C5 alkanes and olefins, ketones, aldehydes, esters, and acetic acid. The water–gas shift reaction (8) also occurs, because it is catalyzed by most of the catalysts typically used in syngas conversion to alcohols. To increase the ethanol yield and selectivity, the catalyst type and experimental conditions need to be optimized. Another important parameter is H2/CO ratio, for direct ethanol synthesis a ratio around 2 is required, lower ratios can lead to catalyst deactivation or modification of the active sites through carbon deposition or carbide formation. A wide range of homogeneous and heterogeneous catalysts have been studied by several authors and different processes, which were reviewed by Subramani et al., 2008. Homogeneous catalysts contain Co, Ru, or Rh metal complexes and convert syngas into ethanol and C2 oxygenates. The selectivity to ethanol is improved by the higher hydrogenation capability of Ru in the Co–Ru bimetallic complexes. Homogeneous catalytic processes are relatively more selective for ethanol, but they also have some disadvantages, which make them unsuitable for commercial applications, such as: expensive catalysts, high operating pressure and catalyst separation and recycling by difficult and long procedures. However, Rathke et al., 2006 reported a new process in which syngas was converted into methanol by a commercial heterogeneous Cu/ZnO catalyst, followed by hydrogenation of methanol by reaction (42) using a HFe(CO)4 complex as a homogeneous catalyst. This process needs a reaction temperature between 180 and 220 °C and pressures up to 300 atm. This process has the advantage of producing ethanol, instead of a mixture of ethanol and water produced by conventional processes and uses a non-noble metal-based

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

38

Filomena Pinto, Rui Neto André and I. Gulyurtlu

catalyst, which could be cost effective. However, it also has some drawbacks, namely the toxicity of Fe(CO4) complex and the need of high pressure (300 atm), which may compromise the commercial viability of this process, Subramani et al., 2008.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

CH3OH(g) + 2CO(g) + H2(g) ⇆ C2H5OH(g) + CO2(g)

(42)

Heterogeneous catalytic processes for converting syngas to ethanol have low yields and low ethanol selectivity, due to slow kinetics of the initial C–C bond formation and fast chain growth of the C2 intermediate. Some researchers have developed and tested several catalysts and proposed different processes, which were reviewed by Subramani et al., 2008. Heterogeneous catalysts may be noble metals-based or non-noble metals-based. The first ones are primarily supported rhodium (Rh) catalysts and produce mostly ethanol and other C2 oxygenates. Several noble metals-based catalysts containing Rh, Ru, and Re supported on various oxides, such as: SiO2, Al2O3, CeO2, ZrO2, MgO, have been studied, with the aim of determining the effect of nature of promoters and supports on catalysts activity and ethanol selectivity. These studies have shown that, though the Rh-based heterogeneous catalysts promoted by Fe or Mn are more selective to ethanol than to other alcohols, Subramani et al., 2008, they are unsuitable for commercial application, due to the limited availability and high cost of Rh, together with the insufficient ethanol yield. The non-noble metals-based catalysts are: a) alkali-modified, low-temperature methanol synthesis catalysts based on Cu–ZnO/Al2O3, b) alkali-modified CuCo-based modified FT catalysts or c) alkali-modified MoS2-based catalysts. Cu–ZnO/Al2O3 catalyst shows the lowest ethanol yield, but is more selective to alcohols than to hydrocarbons. CuCo-based catalysts have a high ethanol yield, but low selectivity. The MoS2-based catalysts have higher ethanol selectivity, but lower ethanol yield. These catalysts produce a mixture of alcohols ranging from C1 to C6. They lead to lower ethanol yields and lower total alcohol production than those attained by methanol synthesis, which shows that further development in catalysts is still needed, for instance by adding alkali metal promoters and by improving the ways to load promoters, Subramani et al., 2008. Most reactions of syngas conversion to ethanol are exothermic, but to guarantee high activity, selectivity, and longer catalyst lifetime, it is of most importance to remove the heat from the reactional zone, thus the design of a suitable reactor is a key issue. Most of the laboratory, bench-scale and pilot-plant studies have been made in fixed-bed reactors, while only few studies have used slurry reactors (CSTR) and slurry bubble column reactors (SBCR), as reported by Subramani et al., 2008. However, according to these authors SBCR are probably a better option, due to efficient heat removal, temperature control, and backmixing, thus leading to higher product yield and selectivity. Syngas may be also used to produce higher alcohols, including ethanol, 2-propanol, and butanol in presence of catalysts containing Cu, Zn, Mo, or Cr, promoted with alkali metals. The main reactions that usually occur are methanol synthesis, WGS reaction, CO beta addition, ethanol homologation, higher alcohol homologation, condensation, dehydration, DME formation, branched iso-alcohols and methyl ester synthesis. Commercial processes for production of mixed higher alcohols have been developed by Snamprogetti-Topsoe, Lurgi, Dow and IFP-Idemitsu, Huber et al., 2006.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Stage of Deployment of Syngas Cleaning Technologies…

39

To convert syngas into ethanol with high product yields and ethanol selectivity is essential to further develop catalysts, as the available catalysts still give moderate ethanol selectivities, high methane formation and high methanol selectivities.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

4.5. Hydrogen Production Hydrogen produced through gasification may have different applications for energy production, it may be used in fuel cells or in combustion systems. Solid oxide fuel cells are more tolerant to H2 impurities and can operate with a mixture of H2 and CO and because of that, their integration with biomass gasifier is easier. After syngas cleaning processes, syngas can be further processed to increase its composition in hydrogen, through the water gas-shift reaction (7), which converts CO into CO2 and H2. The addition of steam is important to minimize undesirable side reactions that compete with the water gas-shift reaction, such as the reverse of Boudouard reaction (3) that lead to the formation of carbon. The conversion of water gas-shift reaction may also be improved by removing hydrogen, to shift the reaction equilibrium to the right side of the reaction. Therefore, water gas-shift reaction may be carried out in two stages or in only one when the water gas-shift reactor also incorporates a hydrogen separation unit. Water gas-shift reaction and the formation of hydrogen is usually increased by the presence of suitable catalysts, whose performance may be affected by the presence of different pollutants that can play a key role on the catalyst life. Different catalysts may be used Fe-oxide-based catalysts that need higher temperature (350-500 °C ) or Cu based catalysts that usually operate at lower temperatures (around 200 °C), Huber et al., 2006. Some Ni-based catalysts have also proven to be suitable to catalyze water gas shift reaction, but as their activity is reduced by sulfur compounds poisoning, it is important to guarantee low content of these compounds in syngas. Some water gas shift catalysts have been developed, for instance by C&CS – Catalysts & chemical Specialties with the aim of also converting tar and hydrocarbons and retaining sulfur, due to the presence of CaO in the catalyst. After water gas shift reaction, syngas main compounds are H2 and CO2, and H2 has to be separated from the CO2 stream. Pinto et al., 2009 reviewed several different processes for H2 and CO2 separation, which include: pressure swing adsorption (PSA), absorption after contact with solvents, adsorption on activated carbon or other materials, cryogenic separation and membrane separation. Physical absorption with solvents is a well known process, but it needs large equipment for solvent circulation and heat exchange, have high degradation rates and have great energy consumption, which leads to low efficiency. PSA needs high-pressure, usually higher than 10MPa, and adsorption in suitable materials, such as activated carbon or zeolites, being the process dependent on temperature, partial pressures, surface forces, and adsorbent pore sizes. In the cryogenic distillation process, the gas mixture is frozen and separated due to the difference in the boiling points of the gases. Both PSA and cryogenic separation need high amounts of energy. Membrane separation is a promising process, because it has several advantages: higher energy efficiency than other processes, cost effectiveness for smaller units, simplicity in operation and it is environmentally friendly. However, its application on the industrial scale is so far limited, while PSA and cryogenic distillation are more mature technologies, Shao L. et

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

40

Filomena Pinto, Rui Neto André and I. Gulyurtlu

al., 2009, and there is not much information about long-term performance and stability of membranes under real industrial gas streams conditions. Membranes for H2/CO2 separation can be polymeric (organic), inorganic (metallic and non-metallic), supported-liquid or facilitated transport membranes and mixed-matrix (hybrid). Polymeric membranes may be made by glassy polymers or rubbery polymers. Inorganic membranes may be metallic, silica, ceramic, carbon, zeolite and others, being the separation process affected by the membrane material and structure. Usually inorganic membranes have good gas separation performance, good chemical resistance and thermal stability, however, they are relatively fragile, more difficult to manufacture and more expensive. Supportedliquid or facilitated transport membranes have a polymeric support impregnated with suitable carriers that interact with one gas component. They are highly permeable to CO2 and have a high CO2/H2 selectivity, however, they are not long-term stable, as they may change, due to carrier saturation. Mixed-matrix membranes try to join the good features of polymers and inorganic membranes. Shao L. et al., 2009 reviewed different types of membranes for CO2/H2 separation analyzing not only materials, but also membrane configuration and module design. Gupta et al., 2007 developed a new process to produce hydrogen from syngas, named as syngas redox (SGR) process. In this process, syngas promotes the reduction of a metal oxide to metallic form by reaction (43). After H2O condensation, CO2 stream may be sequestered by a suitable sequestration method. MO + CO/H2 ⇆ M + CO2/H2O

(43)

Afterwards, the metal oxide is regenerated with steam to release hydrogen by reaction (44) in a cyclic operation. When the metal oxide can not be completely regenerated by the reaction with steam, it may be necessary an additional oxidation with oxygen.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

M + H2O ⇆ MO + H2

(44)

Gupta et al., 2007 tested different metal oxides of Ni, Cu, Cd, Co, Mn, Sn, and Fe. Fe2O3 led to the maximum conversions for both reduction and oxidation reactions and showed good recycling capacity over a number of redox cycles. According to the authors, the efficiency of the system was about 74.2% HHV, which is higher than that of the water gas-shift process. Hanaoka et al., 2005 proposed a method for producing H2 from steam gasification of biomass, followed by water gas shift reaction and CO2 absorption using CaO as a CO2 sorbent, by reaction (45). When all these processes are integrated in a single reactor, H2 production occurs through reaction (46) and CaO may play different roles, it may catalyze biomass gasification, retain sulfur and chloride and act as a CO2 sorbent. The relase of CO2 and the regeneration of CaO occurred by reaction (47), by treating CaCO3 with an HCl solution. CaO + CO2 ⇆ CaCO3

(45)

C(Carbonaceous materials) + CaO + 2H2O ⇆ 2H2 + CaCO3

(46)

CaCO3 + 2HCl ⇆ CaCl2 + H2O + CO2

(47)

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Stage of Deployment of Syngas Cleaning Technologies…

41

Table 3. Biological products obtained by different bacteria types. Bacteria Type Butyribacterium methylotrophicum Clostridial bacteria

Products Butanol, ethanol, butyric acid and acetic acid ethanol

photosynthetic bacteria Rhodospirillus rubrum

poly-3-hydroxybutyrate for H2 and polyesters

Main Reference Worden et al., 1991 Vega et al., 1989 Datar et al., 2004 Maness et al., 1994 Brown et al., 2003

Hanaoka et al., 2005 observed that a Ca/C ratio higher than one was enough to absorb CO2. The conversion of carbonaceous material to gas and H2 yield increased with the rise of Ca/C ratio from 0 and 2, decreasing for values higher than 2. The conversion of carbonaceous material to gas and H2 yield also increased with temperature and were dependent on reaction pressure, reaching H2 yield a maximum value at 0.6 MPa. This process was only tested at a very small lab-scale and as the CO2 was released from the CaCO3 by an aqueous solution with HCl, this may bring some environmental problems at a large scale installation. Though the good results reported by the authors that developed some of these new processes to obtain H2, they are at an early stage and a long way still needs to be followed before these new processes are tested at larger scale.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

4.6. Production of Bio-based Products from Syngas Fermentation Syngas can also be used for anaerobic fermentation in presence of specific bacteria that convert syngas into bio-based products including organic acids, alcohols and polyesters. Several bacteria have been studied and in Table 3 are presented the main biological products obtained. In biological processes, syngas is used as a source of carbon and energy for bacteria growth and biological production of a wide range of products. Worden et al., 1991 stated that by adjusting the fermentation pH, it was possible to control the relative proportions of butanol, ethanol, butyric acid and acetic acid. Two fermentation processes were developed, one-stage and two-stage. In the single-stage process fermentation conditions to favor alcohol formation were used, while in the two-stage process conditions to favor the production of butyric and acetic acid was used. These acids were then reduced to the corresponding alcohols by C. acetobutylicum in a second fermentation. Biological processes do not need high temperature and pressure conditions and bacteria are sulfur tolerant and are not sensitive to CO/H2 ratio. Thus, syngas biological conversion has several advantages in relation to chemical processes. Both CO and H2/CO2 mixtures can be used, though bacteria normally prefer CO to H2, but they are not negatively affected by a specific CO/H2 ratio. The Clostridial bacteria also produces acetic acid, besides the main product, ethanol, but as mentioned by Huber et al., 2006 the ethanol yields for syngas fermentation are similar to those for direct fermentation of corn-derived starches. As bacteria are not as exigent as chemical processes in relation to syngas characteristics, expensive syngas cleaning processes are not needed. In addition, product specificity of biocatalysts is higher than that of chemical catalysts. Nevertheless, biological conversion of syngas has some disadvantages in relation to chemical processes, being the main barriers to

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Filomena Pinto, Rui Neto André and I. Gulyurtlu

42

their commercialization due to difficulties in maintaining anaerobic conditions and mass transfer between gas phase and liquid phase, another difficulty is due to the relatively low rates of bacteria growth and production.

5. THE ROLE OF SYNGAS TOWARDS THE ZET (ZERO EMISSIONS TECHNOLOGIES) Extensive use of fossil fuels for energy needs have had great negative impact on environment, due to pollutants emissions, which are responsible for climate change. CO2 molecule is the least potent greenhouse gas, however due to its abundance and increasing emissions, is one of the most concerning greenhouse gases. An important challenge of nowadays societies is to achieve energy production to guarantee economic growth and a better life quality, without putting at risk the environment and life as we know it. Many countries have compromised to accomplish these objectives by signing Kyoto Protocol. As at present, fossil fuels like coal, oil, and natural gas have the greatest contribution to the supply of world’s energy, accounting for around 85% of today’s needs. Therefore, it is of most importance to develop technologies that allow energy production from fossil fuels with the lowest CO2 emissions. Carbon capture and storage (CCS) is technically feasible by different technologies, which may be grouped as: •

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

• •

pre-combustion, (separation of CO2 from fossil fuel, by gasification and pyrolysis processes) oxy-combustion (combustion with pure O2 and recycled flue gas) and post-combustion (removal of CO2 from combustion flue gases).

In post-combustion processes, CO2 is removed from combustion flue gases, prior to its emission to the atmosphere. CO2 separation from flue gas includes different methods: absorption by solvents, adsorption on activated carbon or other materials, cryogenic separation and membrane separation. After separation, it is necessary to produce a concentrated stream of CO2 at high pressure that can readily be transported to a storage site, where it is injected underground. To reduce the energy costs and the environmental impact associated with these processes it is necessary to develop 1) new and less energy-intensive materials, such as adsorbents and membranes, which exhibit stable performance in flue gases, 2) methods to reduce impact of flue gas contaminants (particulates, sulphur- and nitrous oxides) on the process, 3) conceptual studies, including power plant integration, 4) process development to demonstrate technical feasibility of the process or a particular component on a small scale and/or in a realistic environment In oxy-combustion processes, oxygen is used instead of air to produce very concentrated CO2 streams to easier subsequent capture and CO2 removal from fossil fuel power stations. As combustion conditions and gas physical and chemistry characteristics are quite different from conventional combustion processes, further research is needed to optimize combustion conditions in large-scale units.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Stage of Deployment of Syngas Cleaning Technologies…

43

In pre-combustion processes, carbon content of the fuel is removed prior to combustion to produce a hydrogen rich fuel and a CO2 by-product stream. Coal and/or wastes are first gasified or pyrolysed to produce synthesis gas, consisting mainly of CO and H2. Synthesis gas can also be produced from natural gas through steam reforming or partial oxidation. The CO is then further reacted with steam in a catalytic reactor, according to the exothermal watershift reaction to form more H2 and CO2. As the resulting CO2 concentration is much higher than in a flue gas, the energy effort for CO2 separation is much lower. CO2 can be separated by pressure swing adsorption (PSA), membrane or cryogenic separation. PSA and cryogenic separation have some disadvantages and require much energy, as mentioned before, and shift gas membranes are under development. The remaining CO2 rich stream is sent to storage, whilst H2-rich fuel can thereafter be used in a gas turbine combined cycle or in fuel cells to produce electricity. Pre-combustion processes need further research and development to increase processes efficiency and reduce cost, the demonstration of technically viable technologies is also necessary prior to the full use of gasification to achieve CCS. It should be noted that plants operating with CO2 capture will have higher cost of electricity generation than similar units without CO2 capture. However, to assure a sustainable development it is of most importance to achieve effective CCS and syngas produced from gasification may play an important role in accomplishing zero emissions.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

6. CONCLUSIONS The production of biofuels through gasification of biomass and other wastes may lead to the production of fuels from low cost wastes, using wide range of wastes like: biomass forest wastes, wood byproducts, urban yard waste, agro-industrial wastes, and also municipal solid waste. On the other hand, production of ethanol from corn grain or biodiesel from soybeans compete for food crop resources and may lead to the rise of food prices, as it happened during 2008. Biofuels should only be produced from nonedible fractions of biomass. Though, the largest fraction of biomass is nonedible lignocellulose materials such as straw, grasses, corn stover, wood, forest products, etc., due to the chemical structure of edible biomass, its conversion into fuels is easier. Ethanol production by hydrolysis and fermentation of cellulosic material would also compete with clean cellulosic biomass feedstocks usually used for other proposes. The use of hydrogen as a fuel for transportation sector would require completely different infrastructures: systems capable of storing and pumping liquid hydrogen under high pressure and vehicles with engines or fuel cells capable of using hydrogen efficiently. The use of ethanol as a fuel would also need some infrastructure changes, though at a lower scale, like vehicles more flexible towards fuel and modified transport and storage systems to handle with the hygroscopic nature of pure ethanol. In contrast, the use of liquid fuels produced by FT synthesis or MTG process, using syngas, would not need any important changes in existing fuel and transportation infrastructures, as the new fuels are compatible with the existing ones. The production of biofuels from biomass gasification has the advantage of decreasing CO2 emissions, as these emissions due to biofuels combustion will be compensated by the CO2 capture by photosynthesis process, during the growing of biomass used for biofuel

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

44

Filomena Pinto, Rui Neto André and I. Gulyurtlu

production. Some carbon will be also retained in soil by roots and leaf fall from growing biomass. As biomass has lower sulfur contents than conventional fuels, oil or coal, biofuels produced via biomass gasification have low sulfur compounds, even without biogas cleaning processes, thus SO2 emission problems will be easily solved. Though, it is technically possible to convert biomass and organic wastes into biofuels, the available technologies need to be demonstrated at a commercial scale. Apart from this, it is also possible to reduce CO2 emissions when coal is gasified, instead of biomass, as one possibility of achieving CCS is through gasification followed by syngas cleaning water gas-shift reaction and CO2 capture and sequestration. Even SO2 emissions that could result from coal gasification can be largely reduced by the sulfur removing processes from syngas, described previously, even if this means a higher cost production. At present, fuels production through coal gasification is only economical on large scale units, which require large investments, Dietenberger et al., 2007, which may be a drawback to the spread of these processes. So far production of liquid fuels from biomass gasification has not been economical competitive with production of liquid fuels from petroleum, but the situation may change if petroleum prices rise or if syngas-derived biofuels are given tax exemptions, Huber et al., 2006. Mankind should not be limited to the production of liquid and gaseous fuels by conventional processes, new technologies need to be developed and demonstrated at commercial scale. Sustainable policies are needed to decrease the dependency on fossil fuels and to support renewable energy sources, such as: biomass and wastes, wind, solar, hydroelectric and nuclear power. The main challenges for the near future is the achievement of higher efficiency energy conversion, both for known technologies and for new ones and also develop lifestyles that conserve energy and defend the environment, as only then it will be possible to guarantee a sustainable development.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

REFERENCES Alderliesten, P. T. (1990). “System study high temperature gas cleanup at IGCC systems” Joint system study performed by ECN, KEMA, Stork Boilers and TNO, Report 90-310, NOVEM, Netherlands Agency for Energy and the Environment. André, Rui Neto; Pinto, Filomena; Franco, Carlos; Dias, M.; Gulyurtlu, I., Matos, M. A. A. & Cabrita, I. (2005). Effect of used edible oils in coal fluidised bed gasification, Fuel, 84, 1635-1644. Asadullah Mohammad, Miyazawa Tomohisa, Ito Shin-ichi, Kunimori Kimio, Koyama Shuntarou, Tomishige Keiichi, (2004). A comparison of Rh/CeO2/SiO2 catalysts with steam reforming catalysts, dolomite and inert materials as bed materials in low throughput fluidized bed gasification systems”, Biomass and Bioenergy, 26, 269-279. Asadullah Mohammad, Miyazawa Tomohisa, Ito Shin-ichi, Kunimori Kimio, Tomishige Keiichi, (2003). Catalyst performance of Rh/CeO2/SiO2 in the pyrogasification of biomass, Energy & Fuels, 17, 4, 842-849. Atakul, H. & Wakker, J. P. (1995). Removal of H2S from fuel gases at high temperatures using MnO/y-A12O3, Fuel, 74, 187-191.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Stage of Deployment of Syngas Cleaning Technologies…

45

Aznar, M. P., Corella, J., Delgado, J. & Lahoz, J. (1993). Improved steam gasification of lignocellulosic residues in a fluidized bed with commercial steam reforming catalysts, Ind. Eng. Chem. Res., 32, 1-10. Baker Eddie, G., Mudge Lyle, K. & Brown Michael, D. (1987). Steam gasification of biomass with nickel secondary catalysts, Ind. Eng. Chem. Res., 26, 1335-1339. Bangala, D. N., Abatzoglou, N. & Chornet, E. (1998). Steam reforming of naphthalene on NiCr/Al2O3 catalysts doped with MgO, TiO2 and La2O3, AIChE Journal, 44, 927-936. Bangala, D. N., Abatzoglou, N., Martin, J. P. & Chornet, E. (1997). Catalytic gas conditioning: Application to biomass and waste gasification, Industrial & Engineering Chemistry Research, 36, 10, 4184-4192. Boerrigter, H., Calis, H. P., Slort, D. J., Bodenstaff, H., Kaandorp, A. J., Uil, H. & den Rabou, L. P. L. M. (2004). Gas Cleaning for Integrated Biomass Gasification (BG) and Fischer-Tropsch (FT) Systems Experimental demonstration of two BG-FT systems (“Proof-of-Principle”), ECN Biomass of the Energy research Centre of the Netherlands, ECN-C-04-056, December. Brown, R. C. (2003). Biorenewable resources: engineering new products from agricultural. Ames, IA: Blackwell Publishing processing—an overview, In: B. Kamm, P. R. Gruber, & M. Kamm, (Eds.), Biorefineries—industiral processes and products. Status Quo and future directions, vol. 1, Weinheim: Wiley-VCH Verlag GmbH & Co., 2006, 227-51. BTG - Biomass Technology Group, 2000. Gasifier Inventory, U. Twente, Netherlands, www.btgworld.com/gi/. Caballero, M. A., Corella José, Aznar M-Pilar, Gil Javier, (2000). Biomass gasification with air in fluidized bed. Hot gas cleanup with selected commercial and full-size nickel-based catalysts, Ind. Eng. Chem. Res., 39, 1143–1154. Cal, M. P., Strickler, B. W., Lizzio, A. A. & Gangwal, S. K. (2000). High temperature hydrogen sulfide adsorption on activated carbon II. Effects of gas temperature, gas pressure and sorbent Regeneration, Carbon, 38, 13, 1767-1774. Christou, C., Hadjipaschalis, I. & Poullikkas, A. (2008). Assessment of integrated gasification combined cycle technology competitiveness, Renewable and Sustainable Energy Reviews, 12, 2459-2471. Clark, P. D., Dowling, N. I. & Huang, M. (2004). Production of H2 from catalytic partial oxidation of H2S in a short-contact-time reactor, Catalysis Communications, 5, 12, 743747. Corella Jose, Toledo Jose, M. & Molina Gregorio, (2008). Performance of CaO and MgO for the hot gas clean up in gasification of a chlorine-containing (RDF) feedstock, Bioresource Technology, 99, 7539-7544. Corella Jose, Toledo Jose, M. & Padilla Rita, (2004a). Olivine or dolomite as in-bed additive in biomass gasification with air in a fluidized bed: Which is better?, Energy & Fuels, 18, 713-720. Corella José, Toledo José, M. & Padilla Rita, (2004b). Catalytic hot gas cleaning with monoliths in biomass gasification in fluidized beds. 1. Their effectiveness for tar elimination, Ind. Eng. Chem. Res., 43, 10, 2433 -2445. Corella José, Toledo José, M. & Padilla Rita, (2005). Catalytic hot gas cleaning with monoliths in biomass gasification in fluidized beds. 3. 3. Their effectiveness for ammonia elimination, Ind. Eng. Chem. Res., 44, 7, 2036 -2045.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

46

Filomena Pinto, Rui Neto André and I. Gulyurtlu

Corella Jose´, Aznar Marıa-Pilar, Gil Javier & Caballero Miguel, A. (1999a). Biomass gasification in fluidized bed: Where to locate the dolomite to improve gasification?, Energy & Fuels, 13, 1122-1127. Corella Jose´, Orío Alberto & Aznar Pilar, (1998). Biomass gasification with air in fluidized bed: Reforming of the gas composition with commercial steam reforming catalysts, Ind. Eng. Chem. Res., 37, 4617-4624. Corella Jose´, Orıo Alberto & Toledo Jose-Manuel, (1999b). Biomass Gasification with Air in a Fluidized Bed: Exhaustive Tar Elimination with Commercial Steam Reforming Catalysts, Energy & Fuels, 13, 702-709. Datar, R. P., Shenkman, R. M., Cateni, B. G., Huhnke, R. L. & Lewis, R. S. (2004). Fermentation of biomass-generated producer gas to ethanol, Biotechnology & Bioengineering, 86, 587-94. Delgado, J., Aznar M-Pilar & Corella, J. (1996). Calcined dolomite, magnesite, and calcite for cleaning hot gas from a fluidized bed biomass gasifier with steam: Life and usefulness, Ind. Eng. Chem. Res., 35, 10, 3637-3643. Delgado Jesus, Aznar M-Pilar & Corella, J. (1997). Biomass gasification with steam in fluidised bed: effectiveness of CaO, MgO, Cao-MgO for hot gas cleaning, Ind. Eng. Chem. Res., 36, 5, 1535-1543. Devi, L., Ptasinski, K. J. & Janssen, F. J. J. G. (2003). A review of the primary measures for tar elimination in biomass gasification processes, Biomass & Bioenergy, 24, 125-40. Devi Lopamudra, Craje Menno, Thüne Peter, Ptasinski Krzysztof, J. & Janssen Frans, J. J. G. (2005). Olivine as tar removal catalyst for biomass gasifiers: Catalyst characterization, Appl. Catalysis A, 294, 1, 68-79. Dietenberger, M. A. & Anderson, M. (2007). vision of the U.S. biofuel future: A case for hydrogen-enriched biomass gasification, Ind. Eng. Chem. Res., 46, 8863-8874. Dou Binlin, Mingchuan Zhang, Gao Jinsheng, Shen Wenqin & Sha Xingzhong (2002). Hightemperature removal of NH3, organic sulfur, HCl, and Tar component from coal-derived gas, Ind. Eng. Chem. Res., 41, 17, 4195-4200. Encinar, J. M., González Juan, F., Rodríguez Juan, J. & Ramiro María, J. (2001). Catalysed and an catalysed steam gasification eucaliptus char: Influence of variables and kinetic study, Fuel, 80, 2025-2036. Encinar José, M., Beltran, F. J., Ramiro, A. & González Juan, F. (1998). Pyrolysis gasification of agricultural residues by carbon dioxide in the presence of different additives: influence of variables, Fuel Process. Techn., 55, 3, 219-233. Franco, A. & Diaz Ana, R. (2008). The future challenges for ‘‘clean coal technologies’’: Joining efficiency increase and pollutante mission control, Energy, in press, doi:10.1016/j.energy.2008.09.012. Frey, H. C. & Zhu, Y. (2006). Improved system integration for integrated gasification combined cycle (IGCC) systems, Environ. Sci. Technol., 405, 1693-1699. Gangwal, S. K., Gupta, R. P., Portzer, J. W. & Turk, B. S. (1998). Simultaneous Removal of H2S and NH3 Using Metal Oxide Sorbents, Research Triangle Institute K. Jothimurugesan, Department of Chemical Engineering Hampton University, Advanced Coal-Based Power & Environmental Systems ‘98 Conference, Conference Proceedings. Gangwal, S. K., Gupta, R., Krishnan, G. N. & Hung, S. L. (1993) NH3/H2S Advances, Proceedings of the Coal-Fired Power Systems ’93 - Advances in IGCC and PFBC Review Meeting, p. 399, DOE/METC - 93/6131 (DE93000289).

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Stage of Deployment of Syngas Cleaning Technologies…

47

Garcia, L., Benedicto, A., Romeo, E., Salvador, M. L., Arauzo, J. & Bilbao R. (2002). Hydrogen production by steam gasification of biomass using Ni−Al coprecipitated catalysts promoted with magnesium, Energy & Fuels, 16, 5, 1222-1230. Gibbins, J. & Chalmers, H. (2008). Carbon captureandstorage, Energy Policy, 36, 4317-4322. Gil, J., Aznar, M. P., Caballero, M. A., Francés, E. & Corella, J. (1997). Biomass gasification in flidised bed at pilot scale with steam-oxygen mixtures. Product distribution for very different operating conditions, Energy & Fuels, 11, 6, 1109-1118. Gupta, P., Velazquez-Vargas, L. G. & Fan, L. S. (2007). Syngas Redox (SGR) process to produce hydrogen from coal derived syngas, Energy & Fuels, 21, 2900-2908. Gupta, R. P. & O'Brien William, S. (2000). Desulfurization of hot syngas containing hydrogen chloride vapors using zinc titanate sorbents, Ind. & Eng. Chem. Res., 39, 610-619. Hanaoka, T., Yoshida, T., Fujimoto, S., Kamei, K., Harada, M. & Suzuki, Y. (2005). Hydrogen production from woody biomass by steam gasification using a CO2 sorbent, Biomass & Bioenergy, 28, 63-8. Herguido, J., Corella, J. & González-Saiz, J. (1992). Steam gasification of lignocellulosic residues in a fluidized bed at a small pilot scale. Effect of the type of feedstock, Ind. Eng. Chem. Res., 31, 1274-1282. Hofbauer, H. & Rauch, R. (2001). in "Progress in Thermochemical Biomass Conversion", In: A. V. Bridgwater, & D. G. B. Boocok (Eds.), Blackie Academic & Professional, 199-208. Huber, G. W., Iborra, S. & Corma, A. (2006). Synthesis of Transportation Fuels from Biomass: Chemistry, Catalysts, and Engineering, Chem. Rev., 106, 4044-4098. Khan, M. R. (1989). Prediction of sulphur distribution in products during low temperature coal pyrolysis and gasification, Fuel, 68, 1439-1449. Khodakov Andrei, Y. (2008). Fischer-Tropsch synthesis: Relations between structure of cobalt catalysts and their catalytic performance, Catalysis Today, article in press, doi:10.1016/j.cattod.2008.10.036. Khodakov Andrei, Y., Chu, W. & Fongarland, P. (2007). Advances in the Development of Novel Cobalt Fischer-Tropsch Catalysts for Synthesis of Long-Chain Hydrocarbons and Clean Fuels, Chem. Rev., 107, 1692-1744. Kinoshita, C. M., Wang Yue & Zhou Jiachun (1995). Effect of reformer conditions on catalytic reforming of biomass-gasification tars, Ind. Eng. Chem. Res., 34, 2949-2954. Kuramochi, H., Wu, W. & Kawamoto, K. (2005). Prediction of the behaviors of H2S and HCl during gasification of selected residual biomass fuels by equilibrium calculation, Fuel, 84, 377-387. Leppälahti, J. & Koljonen, T. (1995). Nitrogen evolution from coal, peat and wood during gasification: literatyre review, Fuel Proc. Technol., 43, 1-45. Leppälahti, J., Koljonen, T., Hupa, M. & Kilpinen, P. (1997). Selective catalytic oxidation of NH3 in gasification gas. 2. Oxidation on aluminum oxides and aluminum silicates, Energy & Fuels, 11, 1, 39-45. Leppälahti, J., Simell, P. & Kurkela, E. (1991). Catalytic conversion of nitrogen compounds in gasification gas, Fuel Proc. Technol., 29, 43-56. Li, Y., Nishioka, M. & Sadakata, M. (1999). High Calcium Utilization and Gypsum Formation for Dry Desulfurization Process, Energy & Fuels, 13, 1015-1020.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

48

Filomena Pinto, Rui Neto André and I. Gulyurtlu

Li, W., Lu, H., Chen, H. & Li, B. (2005). The volatilization behavior of chlorine in coal during its pyrolysis and CO2-gasification in a fluidized bed reactor, Fuel, 84, 1874-1878. Liu, H. & Gibbs, B. M. (1988). The influence of limestone addition at different positions on gaseous emissions from a coal-fired circulating fluidized bed combustor, Fuel, 77, 15691577. Liu, H. & Gibbs, B. M. (2003). Modeling NH3 and HCN emissions from biomass circulating fluidized bed gasifiers, Fuel, 82, 1591-1604. Logdberg Sara, Tristantini Dewi, Borg Øyvind, Ilver Lars, Gevert Borje, Jaras Sven, Blekkan Edd, A. & Holmen Anders, (2008). Hydrocarbon production via Fischer-Tropsch synthesis from H2-poor syngas over different Fe-Co/ү-Al2O3 bimetallic catalysts, Applied Catalysis B, Environmental, article in press, doi:10.1016/j.apcatb.2008.11.037. Maness, P. C. & Weaver, P. F. (1994). Production of poly-3-hydroxyalkanoates from CO and H2 by a novel photosynthetic bacterium, Applied Biochemistry and Biotechnology, 45-46, 395-406. Martinez, A. & Lopez, C. (2005). The influence of ZSM-5 zeolite composition and crystal size on the in situ conversion of Fischer–Tropsch products over hybrid catalysts, Appl. Catal. A, 294, 251-259. Martínez, R., Romero, E., García, L. & Bilbao, R. (2003). The effect of lanthanum on Ni–Al catalyst for catalytic steam gasification of pine sawdust, Fuel Processing Technology, 85, 201-214. Milne, T. A., Evans, R. J. & Abatzoglou, N. (1998). Biomass Gasifier Tars: Their Nature, Formation and ConVersion; Report No. NREL/TP-570-25357; National Renewable Energy Laboratory: Golden, CO, USA, http://www.osti.gov/bridge. Najjar Mitri, S. & Jung Dick, Y. (1995). High temperature desulfurization of synthesis gas with iron compounds, Fuel Processirpg Technology, 44, 1-3, 173-180. Narváez Ian, Orio Alberto, Aznar M-Pilar & Corella, J. (1996). Biomass gasification with air in an atmospheric bubbling fluidized bed. Effect of six operational variables on the quality of the produced raw gas, Ind. Eng. Chem. Res., 35, 2110-2120. Ng, K. L., Chadwick, D. & Toseland, B. A. (1999). Kinetics and modeling of dimethyl ether synthesis from synthesis gas, Chemical Engineering Science, 54, 3587-3592. Nichols, K. M., Hedman, P. O., Smoot, L. D. & Blackman, U. A. (1989). Fate of coal-sulphur in a laboratory-scale coal gasifier, Fuel, 68, 243-248. Norman, J. S., Pourkashanian, M. & Williams, A. (1995). Ammonia and its control in clean coal technology systems, Proc. of the 12th Annual Pittsburgh Coal Conf., In: S. H-. Shiang (Ed.), Pitssburgh (PA) 193-198. Orío, A., Corella, J. & Narváez, I. (1997). Performance of different dolomites on hot raw gas cleaning from biomass gasification with air, Ind. Eng. Chem. Res., 36, 9, 3800-3808. Pan, Y. G., Roca, X., Velo, E. & Puigjaner, L. (1999). Removal of tar by secondary air in fluidised bed gasification of residual biomass and coal, Fuel, 78, 1703-1709. Park No-Kuk, Lee Dong-Hwal, Jun Jin Hyuk, Lee Jong Dae, Ryu Si Ok, Lee Tae Jin, Kim Jae-Chang & Chang Chih Hung (2006). Two-stage desulfurization process for hot gas ultra cleanup in IGCC, Fuel, 85, 227-234. Park, N., Lee, J. D., Lee, T. J., Ryu, S. O. & Chang, C. H. (2005). The preparation of a high surface area metal oxide prepared by a matrix-assisted method for hot gas desulphurization, Fuel, 84, 2165-2171.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Stage of Deployment of Syngas Cleaning Technologies…

49

Paterson, N., Zhuo, Y., Dugwell, D. R. & Kandiyoti, R. (2002). Investigation of ammonia formation during gasification in an air blown spouted bed, ASME Turbo Expo 2002 – Land, Sea & Air, (2002) The Netherlands, June, GT2002-530012. Pérez, P., Aznar, M. P., Caballero, M. A., Gil, J., Martín, J. A. & Corella, J. (1997). Hot Gas Cleaning and Upgrading with a Calcined Dolomite Located Downstream a Biomass Fluidized Bed Gasifier Operating with Steam−Oxygen Mixtures, Energy & Fuels, 11, 1194-1203. Pfeifer, C. & Hofbauer, H. (2008). Development of catalytic tar decomposition downstream from a dual fluidized bed biomass steam gasifier, Powder Technology, 180(1-2), 9-16. Pineda, M., Palacios, J. M., Alonso, L., Garcia, E. & Moliner, R. (2000). Performance of zinc oxide based sorbents for hot coal gas desulfurization in multicycle tests in a fixed-bed reactor, Fuel, 79, 885-895. Pinto, F., André, R. N., Abelha, P., Crujeira, A. T. & Gulyurtlu, I. (2009). “Control of Flue Gases Produced by New Coal Utilization Technologies: Combustion and Gasification” in Flue Gases: Research, Technology and Economics, Nova Science Publishers, Inc. N.Y., ISBN: 978-1-60692-449-5. Pinto, F., Lopes, H., André, R. N., Gulyurtlu, I. & Cabrita, I. (2008b). Effect of catalysts in the quality of syngas and by-products obtained by co- gasification of coal and wastes. 2. Heavy Metals, Sulphur and Halogen Compounds Abatement, Fuel, 87, 7, 1050-1062. Pinto, F., Lopes, H., André, R. N., Dias, M. & Gulyurtlu, I. (2008a). “Energetic valorization of wastes through gasification”, in “Alternative processes for waste management”, Nova Science Publishers, Inc. N.Y., ISBN: 978-1-60456-256-9. Pinto, F., Lopes, H., André, R. N., Gulyurtlu, I. & Cabrita, I. (2007). Effect of catalysts in the quality of syngas and by-products obtained by co- gasification of coal and wastes. 1. Tars and nitrogen compounds abatement, Fuel, 86, 14, 2052-2063. Pinto Filomena, André Rui Neto, Franco Carlos, Gulyurtlu, I. & Cabrita, I. (2003). Effect of Experimental conditions on co-gasification of coal, biomass and plastic wastes with air/steam mixtures in a fluidised bed, Fuel, 82, 15-17, 1967-1976. Pirola, C., Bianchi, C. L., Michele, A. Di, Vitali, S. & Ragaini, V. (2009). Fischer Tropsch and Water Gas Shift chemical regimes on supported iron-based catalysts at high metal loading, Catalysis Communications, 10, 823-827. Qadar, S. A., Qadar, Q. A. & Muzio, L. J. (1996). Decomposition of Ammonia in IGCC Fuel Gas Streams, DOE/ER/81964-97/C0732, Advanced Coal-Fired Power Systems '96 Review Meeting, Morgantown, West Virginia, July 16-18. Radtke, K. R., Heinritz-Adrian, M. & Marsico, C. (2006). New Wave of Coal-to-liquids, VGB PowerTech 5/2006. Rapagnà, S. J. N., Kiennemann, A. & Foscolo, P. U. (2000). Steam-gasi cation of biomass in a Iuidized-bed of olivine particles, Biomass and Bioenergy, 19, 187-197. Rapagnà, S., Provendier, H., Petit, C., Kiennemann, A. & Foscolo, P. U. (2002). Development of catalysts suitable for hydrogen or syn-gas production from biomass gasification, Biomass and Bioenergy, 22, 377-388. Richardson, S. M. & Gray, M. R. (1997). Enhancement of residue hydroprocessing catalysts by doping with alkali metals, Energy & Fuels, 11, 6, 1119-1126. Sage, P. W. & Welford, G. B. (1997). ASME Turbo Expo 1997 – Land, Sea and Air, vol. 2, 1295-1302.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

50

Filomena Pinto, Rui Neto André and I. Gulyurtlu

Shao L., Low, B. T., Chung, T. S. & Greenberg, Al. R. (2009). Polymeric membranes for the hydrogen economy: Contemporary approaches and prospects for the future, Journal of Membrane Science, doi:10.1016/j.memsci.2008.11.019. Simell, P., Kurkela, E., Stålberg, P. & Hepola, J. (1996). Catalytic hot gas cleaning of gasification gas, Catalysis Today, 27, 55-62. Spivey, J. J. & Egbebi, A. (2007). Heterogeneous catalytic synthesis of ethanol from biomassderived syngas, Chem. Soc. Rev., 36, 1514-1528. Sridhar, G., Paul, P. J. H. S. & Mukunda, H. S. (2001). Biomass derived producer gas as a reciprocating engine fuel - an experimental analysis, Biomass and Bioenergy, 21, 61-72. Stevens, D. J. (2001). Hot Gas Conditioning: Recent Progress With Larger-Scale Biomass Gasification Systems, NREL/SR-510-29952. Subramani, V. & Gangwal, S. K. (2008). A review of recent literature to search for an efficient catalytic process for the conversion of syngas to ethanol, Energy Fuels, 22, 814-839. Sudiro, M., Bertucco, A., Ruggeri, F. & Fontana, M. (2008). Improving Process Performances in Coal Gasification for Power and Synfuel Production, Energy & Fuels, 22, 6, 3894-3901. Sutton David, Kelleher Brian & Ross Julian, R. H. (2001). Review of literature on catalysts for biomass gasification, Fuel Processing Technology, 73, 155-173. Tao, J. L., Jun, K. W. & Lee, K. W. (2001). Co-production of dimethyl ether and methanol from CO2 hydrogenation: development of a stable hybrid catalyst, Applied Organometallic Chemistry, 15, 105-8. Tian Fu-Jun, Li Bao-Qing, Chen Yong & Li Chun-Zhu (2002). Formation of NOx precursors during the pyrolysis of coal and biomass. Part V. Pyrolysis of a sewage sludge, Fuel, 81, 2203-2208. Tian Fu-Jun, Yu, J., Mckenzie, L. J., Hayashi, J., Chiba, T. & Li, C. (2005). Formation of NOx precursors during the pyrolysis of coal and biomass. Part VII. Pyrolysis and gasification of cane trash with steam, Fuel, 84, 371-376. Tijmensen, M. J. A., Faaij, A. P. C., Hamelinck, C. N. & van Hardeveld, M. R. M. (2002). Exploration of the possibilities for production of Fischer Tropsch liquids and power via biomass gasification, Biomass & Bioenergy, 23, 129-52. Tijmensen, M. J. A., (2000). The Production of Fischer Tropsch liquids and power through biomass gasification, Utrecht University, November. Toledo, J. M., Corella, J. & Molina, G. (2006). Catalytic hot gas cleaning with monoliths in biomass gasification in fluidized beds. 4. Performance of an advanced, secondgeneration, two-layers-based monolithic reactor, Ind. Eng Chem. Res., 45(4), 1389-1396. Tsakiridis, P. E., Agatzini-Leonardou, S. & Oustadakis, P. (2004). Red mud addition in the raw meal for the production of Portland cement clinker, Journal of Hazardous Materials, 116, 1-2, 103–110. Vassilatos, V., Taralas, G., Sjostrom, K. & Bjornbom, E. (1992). Catalytic cracking of tar in biomass pyrolysis gas in the presence of calcined dolomite, Can. J. Chem. Eng., 70, 1008. Vega, J. L., Prieto, S., Elmore, B. B., Clausen, E. C. & Gaddy, J. L. (1989). The biological production of ethanol from synthesis gas, Applied Biochemistry and Biotechnology, 20, 21, 781-97.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Stage of Deployment of Syngas Cleaning Technologies…

51

Visconti, C. G., Lietti, L., Tronconi, E., Forzatti, P., Zennaro, R. & Finocchio, E. (2009). Fischer–Tropsch synthesis on a Co/Al2O3 catalyst with CO2 containing syngas, Applied Catalysis A: General, 355, 61-68. Wang, L., Weller, C. L., Jones, D. D. & Hanna, M. A. (2008). Contemporary issues in thermal gasification of biomass and its application to electricity and fuel production, Biomass and Bioenergy, 32, 573-581. Wang, T., Chang, J. & Lv Pengmei, (2005b). Novel catalyst for cracking of biomass tar, Energy & Fuels, 19, 22-27. Wang, T., Chang, J., Wu, C. Z., Fu, Y. & Chen, Y. (2005a). The steam reforming of naphthalene over a nickel–dolomite cracking catalyst, Biomass and Bioenergy, 28, 5, 508-514. Wang, W., Padban, N., Ye, Z., Olofsson, G., Andersson, A. & Bjerle, I. (2000). Catalytic hot gas cleaning of fuel gas from an air-blown pressurized fluidized-bed gasifier, Ind. Eng. Chem. Res., 39, 4075-4081. Wang, W., Padban Nader, Ye Zhicheng, Andersson Arne & Bjerle Ingemar (1999). Kinetics of ammonia decomposition in hot gas cleaning, Ind. Eng. Chem. Res., 38, 4175-4182. Wang, Z. X., Dong, T., Yuan, L. X., Kan, T., Zhu, X. F., Torimoto, Y., Sadakata, M. & Li, Q. X. (2007). Characteristics of Bio-Oil-Syngas and Its Utilization in Fischer-Tropsch Synthesis, Energy & Fuels, 21, 2421-2432. Wei, X., Schnell, U. & Hein, K. R. G. (2005). Behaviour of gaseous chlorine and alkali metals during biomass thermal utilization, Fuel, 84, 841-848. Weinell Claus, E., Jensen Peter, I., Dam-Johansen Kim & Livbjerg Hans (1992). Hydrogen chloride reaction with lime and limestone: kinetics and sorption capacity, Ind. Eng. Chem. Res., 31, 164-171. Worden, R. M., Grethlein, A. J., Jain, M. K. & Datta, R. (1991). Production of butanol and ethanol from synthesis gas via fermentation, Fuel, 70, 5, 615-619. Wu, C., Yin, X., Ma, L., Zhou, Z. & Chen, H. (2008). Design and operation of a 5.5 MWe biomass integrated gasification combined cycle demonstration plant, Energy & Fuels, 22, 6, 4259-4264. Xu, C., Tsubouchi, N., Hashimoto, H. & Ohtsuka, Y. (2005). Catalytic decomposition of ammonia gas with metal cations present naturally in low rank coals, Fuel, 84, 1957-1967. Yamaguchi, T., Yamasaki, K., Yoshida, O., Kanai, Y., Ueno, A. & Kotera, Y. (1986). Deactivation and regeneration of catalyst for steam gasification of wood to methanol synthesis gas, Ind. Eng. Chem. Prod. Res. Dev., 25, 239-243. Yoshinori, T., Yamaguchi, T., Yamasaki, K., Ueno, A. & Kotera, Y. (1984). Catalyst for steam gasifcation of wood to methanol synthesis gas, Indust. Engineering Chem. Prod. Res. Develop., 23, 225-229. Zevenhoven, R. & Kilpinen, P. (2004). Control of pollutants in flue gases and fuel gases (Sulfur chapter), The Nordic Energy Research Programme, Helsinki University of Technology, February 2004. Zhao, Z., Li, W., Qiu, J. & Li, B. (2003). Effect of Na, Ca and Fe on the evolution of nitrogen species during pyrolysis and combustion of model chars, Fuel, 82, 1839-1844. Zhao, Z., Li, W., Qiu, J. & Li, B. (2003). Effect of Na, Ca and Fe on the evolution of nitrogen species during pyrolysis and combustion of model chars, Fuel, 82, 1839-1844. Zhou, J., Masutani, S. M., Ishimura, D. M., Turn, S. Q. & Kinoshita, C. M. (2000). Release of fuel-bound nitrogen during biomass gasification, Ind. Engng. Chem., 39, 626-34.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved. Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

In: Syngas Production Methods, Post Treatment… Editors: Adorjan Kurucz and Izsak Bencik

ISBN: 978-1-60741-841-2 © 2009 Nova Science Publishers, Inc.

Chapter 2

SYNGAS GENERATION FROM HYDROCARBONS AND OXYGENATES WITH STRUCTURED CATALYSTS V. Sadykov1,2, L. Bobrova1, S. Pavlova1, V. Simagina1, L. Makarshin1, V. Parmon1,2, J. R. H. Ross3, C. Mirodatos4, A. C. Van Veen5 and A. P. Khristolyubov6 1

Boreskov Institute of Catalysis 5, Pr. Ak. Lavrentieva, Novosibirsk, 630090, Russia 2 Novosibirsk State University, Novosibirsk, 630090, Russia 3 Centre of Environmental Research,University of Limerick, Limerick, Ireland 4 Institut de Recherches sur la Catalyse et l’Environnement de Lyon, Villeurbanne, France 5 Ruhr University Bochum, Bochum, Germany 6 VNIIEF, Sarov, Russia

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

ABSTRACT This paper presents results of research of syngas generation by oxidative or steam reforming of hydrocarbons and oxygenates in structured catalytic reactors. Development of structured catalysts included design of active components stable to coking and sintering while efficiently transforming different types of fuels into syngas. It was based upon constructing nanocomposites comprised of components efficient in fuel molecules activation (Ni, Cu, precious metals, etc) and ceria-containing mixed oxides which provide a high rate of oxidants (O2, H2O, CO2) activation and supply of oxygen-containing species to the perimeter of metal particles to consume activated C-H-O fragments. The catalyst development was linked to reactor design. To ensure efficient heat and mass –transfer management required for high performance of structured catalytic reactors, several types of heat-conducting monolithic substrates comprised of refractory alloys and cermets (including microchannel structures) were developed. Procedures for supporting protective layers and active components on these substrates were successfully elaborated. Several types of pilot-scale reactors (with the radial or the axial flow direction) equipped with unique liquid fuel evaporation and mixing units and internal heat exchangers were designed and manufactured. Extended tests of these reactors fed by fuels from C1 to gasoline, mineral and sunflower oil have been carried out with a broad

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

54

V. Sadykov, L. Bobrova, S. Pavlova et al. variation of experimental parameters including stability tests up to 1000 h. Performance analysis has been made with a due regard for equilibrium restrictions on the operational parameters. Transient behavior of the monolith reactor during start-up (ignition) of the methane partial oxidation to synthesis gas was studied and analyzed via mathematical modeling based upon detailed elementary step mechanism. This provides required bases for theoretical optimization of the catalyst bed configuration and process parameters.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

1. INTRODUCTION Most of the current energy supply system, in which fossil fuels are used, causes many environmental problems: air pollution, acid and greenhouse gases emissions. However, the modern life is strongly related to mass consumption of energy for electricity generation and transportation sector. These sobering trends justify an urgency with respect to the need for more efficient, as well as alternative, energy conversion technologies and devices. The features, which are required from alternative energy conversion technologies, include high energy conversion efficiency, environmental friendliness, compatibility with both conventional fuels and renewable energy sources and sustainability [1-3]. Hydrogen’s unchallenged potential to reduce greenhouse gases and atmospheric pollutants makes it a major candidate to meet the world energy demand, especially for smallscale applications. This simplest element and the most plentiful gas in the universe does not exist alone in nature. It always combines with other elements such as oxygen and carbon, and is mostly present in water, biomass, and fossil hydrocarbons. Conversion of fuels into hydrogen-rich product streams is commonly referred to as fuel processing. The fossil fuels, petroleum, coal and natural gas are high energy density materials and, thus, amenable to centralized, large-scale processing plants. Industry generates ca 50 millions metric tons of hydrogen globally each year from fossil fuels [2-6]. It is expected that significant development of a hydrogen transportation infrastructure will not occur within the next decade. The lack of safe, efficient and cost effective hydrogen storage system is the major obstacle for hydrogen application as an energy carrier [6]. The most promising approach for distributed small/middle-scale applications is the small -scale processing of fuels near the point of usage. Natural gas and liquid fuels such as alcohols, gasoline, jet fuel, kerosene and diesel, are favored as the most suitable hydrogen carriers because they are easily transportable and have existing infrastructures for distribution [7-9]. Nowadays, the modern biomass-based transportation fuels (liquid or gaseous) such as bioethanol, fatty acid (m)ethylester, biomethanol, biodiesel as well as diesel produced from biomass by Fischer–Tropsch synthesis are generally considered as offering many advantages including sustainability, reduction of greenhouse gas emissions, regional development, social structure and agriculture, security of supply [10]. One of the ways to generate hydrogen-rich gases is catalytic reforming of carbonaceous energy carriers to produce syngas. In its simplest form, syngas is composed of hydrogen and carbon monoxide. Syngas is considered as an alternative to conventional fuels in all its applications. It can be used like natural gas, as a source of hydrogen for fuel cells, or reformed into other hydrocarbon fuels. In principle, syngas can be produced by reformation from any fossil fuels, including natural gas, naphtha, residual oil, petroleum coke, coal, and biofuels. The process is carried out in the presence of a catalyst which controls the product

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Syngas Generation from Hydrocarbons and Oxygenates with Structured Catalysts

55

composition. The catalytic reactions take place at the solid-gas or solid-liquid interfaces. The catalyst plays a major role in enhancing the hydrogen yield and reducing the formation of undesired compounds. New catalytic processes holds the greatest promise for increasing the efficiency and environmental benefits of energy related processes [11, 12]. Currently, one of the limitations of hydrogen generation is that it is based on fossil fuels, leading to a net production of greenhouse gases. Thus, the use of biomass as an alternative hydrogen source, with CO2-neutral impact, may become important in the future. Biofuels result from conversion of biomass, the recent remains of organisms or their wastes. Biofuels are renewable. An obstacle to converting biomass into syngas, at an economical scale, has been the decentralized nature of biomass operations. Bio-oil produced from fast pyrolysis of biomass represents a type of high-energy density chemical and, in addition, ease of transport and generation of hydrogen via steam reforming of this bio-oil on-site, where the hydrogen is needed, can be achieved [12]. Since the production of bio-oils is considered now to be a mature technology [13], their catalytic treatment to produce hydrogen can be the promising technology for a clean and renewable hydrogen generation. The bio-oils could be treated directly as a whole or using specific fractions. Nevertheless, direct feeding of raw bio-oil into the reformer reactor is not easy, since it is only partly soluble in water and highly unstable upon heating, polymerizing at temperatures as low as 80oC. Fast pyrolysis liquids are complex mixtures of oxygenated compounds, which include alcohols, acetone, and aldehydes, emulsified with water. Such bio-fuels may contain acids, aldehydes, ketones, alcohols, glycols, esters, ethers, phenols and derivatives, as well as carbohydrates, and a large proportion (20–30 wt.%) of lignin-derived oligomers [14, 15]. Recently, a lot of research have been carried out in the field of steam reforming of some oxygenated molecules, such as methanol, acetic acid, ethanol, acetone, phenol, cresol etc used as model compounds of biooils [16-20]. From thermodynamic point of view, syngas production from hydrocarbons or oxygenates may be categorized into two basically different types of processes. One is endothermic-steam reforming, the other one is exothermic partial oxidation, where the feed reacts directly with air, or enriched air with carefully balanced oxygen to fuel ratios. The gross oxidative reaction of syngas formation from a general fuel (hydrocarbon or oxygenate) can be described by the following expression:

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

СnНmOp + х О2 + (n-2х-p) Н2О = n СО + (n-2х-p+(½ m)) Н2 Criteria for operating modes can be estimated with thermodynamic calculations. Nevertheless, the risk of residues and soot formation increases with the complexity of the fuel being reformed. For a practical consideration of hydrogen-rich gas production from the particular fuel, the operational characteristics should be chosen on the bases of analysis of many factors including a range of throughputs, design of reactor and configuration of catalytic bed, as well as catalytic activity. The operational parameters have to be worked out to perform the process safely and escape all the possible problems arisen. The type of a catalyst determines the design of a reactor and feed preparation/supply system for the small-scale application. There has been a significant interest since the first publications [21-28] in the new catalytic processes, which operate at high temperatures and under short contact times over structured catalysts. Structured catalysts are ceramic and metallic configurations which constitute both the catalyst support and the reactor. The most

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

V. Sadykov, L. Bobrova, S. Pavlova et al.

56

important structured reactors are based on gauzes, foams and monoliths. The catalyst bed may also be arranged as either arrays of particles or superimposed sheets, to allow cross-flow. They are compact reactors with excellent performance in activity and selectivity. In general, monoliths are the most satisfactory structured reactors. Traditionally, monolithic catalysts are continuous unitary structures, which contain very small, mostly parallel passages. The catalytically active material is dispersed in the form of a thin layer (washcoat) over the ceramic or metallic substrate. In gas-phase applications the structured reactors are often preferred due to their favourable properties with respect to selectivity, pressure drop, and robustness. Their millisecond characteristics have potential in syngas production and selective conversions into valuable products. There is a growing need to extend our catalyst and reaction engineering knowledge base to the structured reactor category, to have better understanding of flow contacting, multispecies transports and detailed reaction kinetics and their interactions in the reactors, to have improved understanding of the catalyst structure impact and materials (washcoat) influence on process performance [29-33]. Catalytic monoliths with a high thermal stability and thermal shock resistance are also required. Preheat temperature, flow rate, oxygen-to-carbon and water-to carbon ratios in the feed affect the temperature distribution in the structured catalyst. Hence, composition of the active components, configuration of the catalyst bed and the process performance as a whole, require optimization for the particular case of the fuel being reformed. This review considers the important aspects of efficient oerfomance of syngas generation from a variety of fuels on structured catalysts with a due regard for results obtained within represented by authors international broad-scale collaboration as well as related world-wide research published in available sources.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

2. REFORMING CHEMISTRY The operation mode for the reformer can be very different, with wide implications on the composition of the reformed effluent and the energy demand, necessary to generate the hydrogen-rich feed. As for the reforming chemistry, there are three possible operational modes of the small-scale fuel reforming: catalytic steam reforming (SR), direct partial oxidation (PO) and so called indirect partial oxidation, i.e. combination of partial oxidation and steam reforming in stand-alone catalytic system, viewed as autothermal reforming (ATR) [27, 34-38].

2.1. Steam Reforming Steam reforming involves the reaction of steam with the fuel in the presence of a catalyst, as illustrated in Eq. (1) for methane and Eq. (2) for generic hydrocarbon fuel. This process is the most highly developed and cost effective approach for generating hydrogen and is also the most efficient, giving conversion rates of 70 to 80 per cent. Reaction (1)

CH4 + H2O ' CO + 3H2

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Syngas Generation from Hydrocarbons and Oxygenates with Structured Catalysts

57

ΔHo298 = +206.2 kJ mol-1 Reaction (2)

CnHm + nH2O  nCO + (n+m/2)H2

For a generic oxygenated molecule, the steam reforming reaction proceeds according to the following equation: Reaction (3)

CnHmOk + (n-k)H2O  nCO + (n+m/2-k)H2

The steam reforming reactions (Eqs. 1- 3) are followed by the exothermic water gas shift reaction (Eq. 4) and methanation reaction (Eq. 5): Reaction (4)

CO + H2O ' CO2 + H2 ΔHo298 = -41.2 kJ mol-1

Reaction (5)

CO + 3H2 CH4 + H2O ΔHo298 = -206.2 kJ mol-1

When the steam-to-carbon ratios are close to the stoichiometric value (Eqs. 1-3), coke produced by thermal cracking of hydrocarbons (Eq. 6) or by the Bouduard reaction (Eq. 7) may lead to the catalyst deactivation. Reaction (6)

CH4 ' C + 2H2 ΔHo298 = +75 kJ/mol

Reaction (7)

2CO ' C + CO2 ΔHo298 =- 172 kJ/mol

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

To minimize coke formation, excess steam is used to ensure that any carbon formed is gasified Reaction (8)

C + H2O ' CO + H2.

For methane, a steam-to-carbon ratio of ~ 2.5 is sufficient to avoid coking. The steam-tocarbon ratio and operating temperatures depend on the type of fuel. Heavier feeds require a higher steam-to-carbon ratio and a higher operating temperature. For higher hydrocarbons, a steam-to-carbon ratio of 6-10 is common [5, 39]. While bio-oils are more reactive than petroleum fuels, high temperature is needed in the reactor to gasify the formed coke deposits. High ratios of steam to carbon (greater than 7) are necessary to avoid catalyst deactivation by coking [14]. The thermal decomposition (Eq. 9) has also to be considered for most of the hydrocarbon and oxygenated fuels [40-42]. Steam reforming of bio-fuels is more complicated since some

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

V. Sadykov, L. Bobrova, S. Pavlova et al.

58

bio-fuels components are thermally unstable and decompose upon heating. In this case, the carbon is essentially a soot-like deposit. Reaction (9) CnHmOk CxHyOz+carbon+ (H2, CO, CO2, CH4, CxHx+2). The formed CxHy fragments become also favorable for coking via cyclization processes. Deactivation of the catalysts due to coking is one of the major problems, and bio-oils have more deactivation problems than do petroleum-derived feedstocks. In fact, steam reforming of bio-oils in fixed bed reactors requires a catalyst regeneration step after 3-4 h of time-onstream [14]. The main disadvantage of this method for hydrogen or syngas production is that it is endothermic and, hence, requires external heating and overheated steam. This reduces the overall efficiency of the system from the theoretical 100 per cent and also means that there is a delay before the system is ready to operate. As a result, this method is not particularly suitable for portable applications as consumers expect devices to start immediately [37].

2.2. Partial Oxidation Partial oxidation (also called direct partial oxidation) involves the reaction of the hydrocarbon with oxygen to liberate hydrogen as presented by Eqs. 10, 12: Reaction (10)

CH4 + ½ O2 ' CO + 2H2 ΔHo298 = - 35.7 kJ mol-1

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Reaction (11)

CnHm+n/2 O2nCO+m/2 H2

Partial oxidation can be used for converting methane and higher hydrocarbons into syngas but is rarely used for oxygenates [43-50]. The catalytic partial oxidation is the mild exothermic reaction. The oxygen-to-fuel ratio determines the heat of reaction and the hydrogen yield. The partial oxidation processes are attractive alternatives because they avoid the need for large amounts of expensive superheated steam. Direct partial oxidation catalytic reaction in monolithic reactors proceeds at millisecond contact times, which are at least 2 orders of magnitude shorter than those for traditional steam reforming process (~ 1 s), and a high conversion is achieved. However, less hydrogen is produced for the same amount of fuel than in the case of steam reforming. One more advantage of partial oxidation is that formation of olefins via endothermic catalytic dehydrogenation and cracking is excluded due to a high conversion at short residence time. A low coking was attributed to the fact that carbonproducing reactions (olefin cracking, CO disproportionation, and reverse-steam reforming) do not approach equilibrium and that the presence of CO2 and H2O favor reforming of the carbon formed [43].

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Syngas Generation from Hydrocarbons and Oxygenates with Structured Catalysts

59

2.3. Autothermal Reforming Autothermal reforming (so called indirect partial oxidation) combines the endothermic steam reforming process with the exothermic partial oxidation reaction (Eqs. 12, 13) in standalone catalytic system, therefore balancing the heat flow into and out of the reactor [7, 9, 34, 37, 51-54]. No external heating source is required. Reaction (12)

2 CH4 + ½ O2 + H2O ' 2CO + 5H2 ΔHo298 = -18.4 kJ mol-1

The general formula for ATR, using air as the oxygen source is Reaction (13) СnНmOp +х(О2 + 3.76N2)+(n-2х-p) Н2О = nСО + (n-2х-p+(½ m)) Н2 +3.76xN2. The fuel gas contains a mixture of H2, CO, CO2 with the relative concentrations being determined by the water-gas shift reaction (Eq. 4), if thermodynamic equilibrium is achieved. A lower operating temperature of the catalytic ATR compared to PO has a several advantages for small-scale applications [46]: –

– –

less complicated reactor design and lower reactor weights, because less thermal integration (i.e. heat exchange between incoming reactants and hot products) is required; a wider choice of materials required; lower fuel consumption during startup, because for a given reactor mass, the energy required to heat the reformer to its operating temperature is proportional to its operation temperature.

The amount and concentration of hydrogen generated from a given amount of hydrocarbon fuel, and the quality of the raw reformate (CO, CO2, CH4 and possibly other species, H2O and nitrogen contents) are determined by reforming conditions.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

2.4. Coke Deposition Coke deposition takes place where polymerisation, thermal decomposition of fuel and other reactions occur, leading to blockage of catalyst pores and, in extreme cases, to complete failure of the reactor. The reaction conditions must be maintained such that no graphitic or amorphous carbon is formed in the reformation process. Depending on the catalyst used, the reforming reaction may require temperatures of ~1000-1300 K, or even higher. At these temperatures, most of the carbon in the fuel is converted to CO and CO2, with possible formation of relatively small amounts of CH4 and other hydrocarbons. Carbon deposition in the reactor may also take place by the direct decomposition of methane and by the Boudouard reaction (reactions 6 and 7). Carbon formation is a problem in reforming of hydrocarbon fuels, particularly for hydrocarbon fuels with two or more carbon atoms in the main chain. More aromatic fuels have a higher tendency of carbon formation. Heavier hydrocarbons in the

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

60

V. Sadykov, L. Bobrova, S. Pavlova et al.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

jet fuels and diesel fuels can form carbon deposits at relatively low temperatures such as 450oC due to fuel pyrolysis [10, 55, 56]. A catalyst applied may initiate carbon formation as well. For example, carbon formation catalyzed by Ni usually takes place in the form of whiskers with a Ni particle at the top of a fibre [57]. Steam reforming of oxygenated hydrocarbons is thermodynamically favored at lower temperatures than that of hydrocarbons [58], so the steam reforming of oxygenates can take place at lower temperatures because it is less endothermic [59]. Two main reactions, steam reforming and pyrolysis, occur in a high temperature steam/bio-fuel mixture. However, many reactions occur simultaneously in the reformer including many side reactions. Side reactions provide a higher thermodynamic stability of the overall process operation by coupling exothermic and endothermic reactions. It was found that to prevent carbon deposition, the oxygen/carbon ratio (O/C) in the reaction system have to satisfy the following requirements: in pyrolysis O/C ≤ 1, gasification O/C < 2 and combustion O/C > 2 [60]. An equilibrium phase diagram (Figure 1) taken from Slinn et al [60, 61] represents graphically gasification and reforming systems. The phase diagram shows that, above the carbon deposition boundary (dashed line), solid carbon particles exist in equilibrium with the gaseous components. Below this carbon deposition boundary (shaded section) carbon is present as CO, CO2 and CH4. To avoid additional carbon formation either oxygen or H2 must be added to shift the point below the defined carbon deposition boundary. Further addition of oxygen will shift the equilibrium position over the line of complete carbon combustion where free oxygen is present [60, 61]. Fossil fuels have a lower oxygen content within their molecular structure, and they would be placed above the carbon boundary in equilibrium with solid carbon. Such oxygenate as glycerol contains a higher oxygen content and has an oxygen/carbon ratio of 1, so it is already at the carbon boundary and does not need any extra oxygen or hydrogen. However, this phase diagram assumes thermodynamic equilibrium, which might not always be the case. Therefore, there may be more carbon than predicted.

Figure 1. Carbon–hydrogen–oxygen equilibrium phase diagram. Adapted with permission from ref60. Copyright 2008 Elsevier B.D. [61].

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Syngas Generation from Hydrocarbons and Oxygenates with Structured Catalysts

61

Actually, there are two limits to be considered for the coke deposition in the hydrocarbons reforming: thermodynamic and kinetic. By thermodynamic criteria, coke is formed if the gas composition shows affinity for carbon. From kinetic point of view, carbon formation in the reactor occurs in conditions where hydrocarbon can decompose yielding carbon, even if thermodynamics predicts no carbon formation after the equilibrium has been reached. Carbon formation is then a question of kinetics, local process conditions and reactor (reformer) design [57, 61-63].

3. CATALYSTS FOR REFORMING HYDROCARBON FUELS AND OXYGENATES The small- scale processing of hydrocarbon fuels and oxygenates to hydrogen-rich gas requires that the reforming catalyst exhibit a higher activity and better thermal and mechanical stability than the reforming catalysts currently used in the production of H2 for large-scale industrial processes.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

3.1. Catalysts for Syngas Generation from Hydrocarbon Fuels Catalyst are typically group VIII metals, such as rhodium, platinum, palladium, ruthenium, cobalt, nickel, and iridium [64-71], which are either supported on oxide substrates [64, 65], or used unsupported, as metal wires and gauzes [66] for partial oxidation of hydrocarbons to hydrogen-rich gas. Au et. al. [67] carried out a theoretical study on the comparison of single component catalysts such as Rh, Ru, Ir, Os, Pd, and Pt, and concluded that the most efficient catalyst for methane dissociation is Rh, which results in high CH4 conversion. However, at stoichiometric fuel-to-oxygen ratio, a high and stable performance was demonstrated only for catalysts with a high (5-10 wt.%) loading of expensive Rh on corundum foam substrates, while less expensive supported Pt/alumina catalysts were less active and selective [22-26, 45-48] deactivating with time-on –stream due to coking [68-70]. Moreover, in the short contact time monolithic reactors all oxygen in the feed is consumed in a narrow (1-2 mm) inlet part of the catalytic layer, where high (up to 1200oC) temperatures are developed due to oxidation reactions [22]. As the result, evaporation and loss of precious metals followed by continuous moving of the high-temperature zone along the monolith length may occur up to complete deactivation of the catalyst. From the work of M. Prettre et al. [71], which can be considered as a historic benchmark for research on the catalytic partial oxidation of methane, the family of the inexpensive catalysts on the base of reduced Ni supported on refractory Mg-Al spinel is widely used in industry [57] due to their high stability and activity in synthesis gas production. The study by Vermeiren et al [72] on the catalytic activity of Ni supported on Al2O3, SiO2-Al2O3, SiO2ZrO2, and zeolite Y shows that complete oxidation followed by steam or CO2 (or both) reforming reactions could occur on all these types of catalysts. The Ni- supported group of catalysts on the bases of rare-earth (La, Ce, Pr, Nd, Sm, Eu, Gd, Dy, Er, Yb) oxides show a similar activity and selectivity for the oxidation reforming reaction [73]. The tendency to coke deposition is generally considered to be the main drawback in the application of supported Ni

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

62

V. Sadykov, L. Bobrova, S. Pavlova et al.

catalysts. Among the catalyst group mentioned, NiO-La2O3 catalyst was the best one. The addition of rare earth metal oxide or alkaline metal oxide to alumina or the use of rare-earth oxides as support can somewhat restrict carbon deposition. The promotion by the rare earth oxide addition to the catalyst support is probably due to its capability to act as oxygen or oxygen - containing compounds storage, which can help in oxidizing the deposited coke or its precursors. Moreover, decoration of the surface of Ni particles by hydroxocarbonate species could prevent carbon nucleation due to ensemble dilution effect. It is also believed that the presence of rare earth oxide such as CeO2 can stabilize the support and prevent it from sintering during the high-temperature reaction. Cerium is supposed to weaken a strong bond of Ni with aluminum oxide. This prevents Ni diffusion into the alumina lattice [74, 75]. Perovskites, the composite oxides of alkali-earth (rare-earth) and transition metals with a general formula АВО3, are the vast class of catalytic systems for oxidation reforming of hydrocarbons. Cation of transition metal B is in the octahedral oxygen surrounding. A big cation A is surrounded by the 12 oxygen ions. A dense packing is provided by АО3 layers comprised of oxygen anions and A cations. Lattice defects are formed, if A or B cations are replaced by cations with different charge. Activity of perovskites in different catalytic reactions has been intensively investigated since 1970. Catalytic activity seemed to correlate with the density of different lattice defects [76]. In the case of nickelates as catalysts of steam reforming or partial oxidation reactions, metallic Ni formed under reducing reaction feed effect is capable to provide the homolytic rupture of С-Н bond followed by coke formation from activated CHx fragments if their subsequent transformation into syngas is a slow step. In this case, triple-charged La and Cr cations in the lattice of complex perovskites improve the oxygen anions mobility in the lattice and stabilize Ni in the state of 1+ or 2+ [77, 78]. The most promising type of the active component is a nanocomposite containing precious metals combined with mixed oxides possessing a good oxygen mobility and reactivity. Thus, catalyst developed at Argonne National Laboratory and produced by Süd-Chemie Inc. contains a transition metal supported on an oxide-ion-conducting substrate, such as ceria, zirconia or lanthanum gallate doped with a small amount of non-reducible element, such as gadolinium, samarium etc [34, 37]. Various transition metals supported on doped ceria exhibited excellent isooctane reforming activity between 500 and 800oC with high fuel conversion and H2 selectivity. Among the metals investigated (Ni, Co, Ru. Pd, Fe, Cu, Ag), all metals, except for Ag, exhibit conversion of > 95 % at T > 600oC , and all metals exhibit 100 % conversion at 700oC. At temperatures 60%) than the first-raw transition metals at > 650oC. At temperatures < 600oC, the H2 selectivity decreases to 1. Thermodynamic analysis of ethanol processor systems [202] has shown that the most important parameter which affects efficiency in hydrogen production, is the water to ethanol molar ratio in the feed. The values higher than stoichiometry result in the reduced efficiency, because of increased enthalpy needs for water evaporation. The thermodynamics of steam assisted high-pressure conversions of model components of bio-oil – isopropyl alcohol, lactic acid and phenol – to synthesis gas (H2 + CO) have been investigated in [203] to understand the effects of process variables such as temperature and inlet steam-to-fuel ratio on the product distribution. It proved possible to adjust the H2/CO ratios and the amounts of CH4 and CO2 in synthesis gas by changing the steam-to-fuel ratio, the value depending on temperature and the fuel type. Thus, by using the thermodynamic calculations we can establish a correlation between the inlet and outlet characteristics of an adiabatic reactor at steady state. Complex interaction between the reactive flow and reactions on the catalytic surface occurs in the structured catalytic layers of different forms such as foam or extruded monoliths, wire gauzes, or sintered spheres, in a reforming reactor. Quite often the catalytic processes, which can be run nearly autothermally and adiabatically, exhibit an extremely fast variation of temperature, velocity, and transport coefficients of the reactive mixture, especially near the catalyst entrance. Spatio-temporal profiles developed in the reactors are results of interplay among kinetics, hydrodynamics, and the heat transfer. Generally, for a practical consideration, the operational characteristics should be chosen on the base of analysis of many factors including a range of throughputs, design of the reactor and configuration of monoliths, as well as catalytic activity. Other issues such as catalyst kinetics or safety aspects appear to play an important role in determining the operating conditions of the reforming reactor. The understanding of the details of the reactor behavior demands a better agreement between experimental and modeled flow field without neglecting the complex chemistry. Quite often, a direct experimental investigation of a reaction mechanism is difficult if possible at all under realistic conditions. Simulation studies are very helpful in examining physical-chemical processes in the reactor in detail. To achieve this, elementary-step kinetic models should be coupled with appropriate reactor models, and the simulation results should then be tested against available experimental data such as conversions and selectivities for the specific reactor configurations. Particularly, transient experiments and dynamic simulations, such as ignition and extinction of reactions, can often yield important insights into reaction

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

V. Sadykov, L. Bobrova, S. Pavlova et al.

88

mechanisms also due to their high sensitivity to the specific reaction path taken during the transient excursion of the reaction system [204-210]. Table 3. Surface reaction mechanism for oxidation of methane over Pt/Ce-Zr-La/α-Al2O3 [211].

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

No

Reaction

kf

Ef

,

kb

Eb ,

Δr H ,

kcal/ mole

kcal/ mole

Preexponential, s-1 or sticking coeff.

kcal/ mole

Preexponential, s-1

1

ОН ∗ + ∗ ↔ Н ∗ + О ∗

5.60 × 1011

18.3

1.70 × 1010

13.4

4.9

2

Н 2 О ∗ + ∗ ↔ Н ∗ + ОН ∗

1.20 × 1010

39.1

3.50 × 1011

0.0

39.4

3

Н 2 О ∗ + О ∗ ↔ 2ОН ∗

1.00 × 1011

34.1

1.00 × 1011

0.0

34.1

4

Н 2 + 2∗ ↔ 2Н ∗

0.09

0.0

3.33 ×1012

20.0

-16.0

5

О2 + 2 ∗ ↔ 2О ∗

0.03

0.0

1.00 ×1011

19.0

-15.0

6

Н 2О + ∗ ↔ Н 2О ∗

1.00

0.0

5.33 ×1012

10.0

-10.0

7

ОН + ∗ ↔ ОН ∗

1.00

0.0

1.00 × 1013

30.0

-30.0

8

Н +∗ ↔ Н ∗

1.00

0.0

1.00 × 1013

60.2

-60.2

9

О +∗ ↔ О∗

1.00

0.0

1.00 × 1013

67.0

-67.0

10

СН 4 + 2 ∗ ↔ СН 3 + Н ∗

0.68

12.0

3.97 ×1010

5.5

6.5

11

СН 3 + ∗ ↔ СН 2 + Н ∗

1.32 ×1013

25.8

4.04 ×1010

6.1

19.7

12

СН 2 ∗ + ∗ ↔ СН ∗ + Н ∗

1.00 × 1011

25.0

1.00 × 1011

12.2

12.8

13

СН ∗ + ∗ ↔ С ∗ + Н ∗

1.00 × 1011

5.4

1.00 × 1011

37.6

-31.4

14

СН 3∗ + О ∗ ↔ СН 2 ∗ + ОН ∗

1.00 × 1011

17.7

1.00 × 1011

3.1

14.6

15

СН ∗ + ОН ∗ ↔ СН 2 ∗ + О ∗

1.00 × 1011

13.2

1.00 × 1011

20.5

-7.3

16

С ∗ + ОН ∗ ↔ СН ∗ + О ∗

1.00 × 1011

38.2

1.00 × 1011

1.5

36.7

17

СН 2∗ + Н 2О ∗ ↔ СН 3∗ + ОН ∗

1.00 × 1011

19.5

1.00 × 1011

0.0

19.5

18

СН ∗ + Н 2 О ∗ ↔ СН 2 ∗ + ОН ∗

1.00 × 1011

26.7

1.00 × 1011

0.0

26.7

19

С ∗ + Н 2 О ∗ ↔ СН ∗ + ОН ∗

1.00 × 1011

70.9

1.00 × 1011

0.0

70.9

20

СО ∗ + ∗ ↔ С ∗ + О ∗

1.00 × 1011

74.2

1.00 × 1011

0.0

75.

21

СО2 ∗ + ∗ ↔ СО ∗ + О ∗

1.00 × 1011

43.1

1.00 × 1011

0.0

43.1

22

СО + ∗ ↔ СО ∗

0.0

1.21×1013

34.0

-34.0

23

СО2 + ∗ ↔ СО2 ∗

0.71 0.7

0.0

1.46 ×1012

17.0

-17.0

24

СО2 ∗ + Н ∗ ↔ СО ∗ + ОН ∗

1.00 × 1011

38.2

1.00 × 1011

0.0

38.2

25

СО ∗ + Н ∗ ↔ СН ∗ + О ∗

1.00 × 1011

106.0

1.00 × 1011

0.0

106.0

26

СО ∗ + Н ∗ ↔ С ∗ + ОН ∗

1.00 × 1011

69.2

1.00 × 1011

0.0

69.2







Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Syngas Generation from Hydrocarbons and Oxygenates with Structured Catalysts

89

Table 3. (Continued) No

Reaction

kb

Eb ,

Δr H ,

kcal/ mole

Preexponential, s-1

kcal/ mole

kcal/ mole

0.0

1.00 × 1013

38.0

-38.0

kf

Ef

Preexponential, s-1 or sticking coeff. 1.00

,

27

СН 3 + ∗ ↔ СН 3

28

СН 2 + ∗ ↔ СН 2 ∗

1.00

0.0

1.00 × 1013

68.0

-68.0

29

СН + ∗ ↔ СН ∗

1.00

0.0

1.00 × 1013

97.0

-97.0

1.00

0.0

13

1.00 × 10

150.0

-149.0

31.0

4.17 × 10

0.0

31.9

1.4

0.2 ×103

1.4

0.0



30

С+ ↔С

31

2СО ∗ ↔ С ∗ + СО2 ∗

2.40 ×10

32

O∗ + Z ↔∗ + ZO

0.2 ×103





12

-5

9

2

The site densities were assumed: for the Pt - 1.65x10 mol/m , for ZO – 1.65x10-2

In the next section the details of reaction behavior inside monolithic catalyst has been examined by mathematical modeling of the methane partial oxidation on the base of detailed chemistry.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

5. NUMERICAL STUDY OF THE PARTIAL OXIDATION OF METHANE The basic reactor and methane partial oxidation reaction model for the Pt/Ce-Zr-La/αAl2O3 honeycomb monolith has been presented in detail in a separate paper [211]. A dynamic one-dimensional two-phase reactor model of the processes with accounting for both transport limitations in the boundary layer of a fluid near the catalyst surface and detailed molecular unsteady-state kinetic model for surface reactions has been developed and verified with the transient experiments data. A mechanism proposed by P. Aghalayam et.al. [212] was taken as a basis for the unsteady-state kinetic model to be implemented in the dynamic reactor model after some modification. The Pt/Ce-Zr-(La)-O/α-Al2O3 honeycomb monolith is a complex catalytic system where washcoat (fluorite-like nano-crystallites of solid solution Ce-Zr-La-O) strongly interacts with the active component -Pt. Study of detailed kinetic and mechanism of the partial oxidation reaction using the step response technique has made it clear that active oxygen of the bulk quickly re-oxidizes the reduced platinum [116]. The simplified reaction scheme illustrates the reaction mechanism over Pt/CeO2-ZrO2 catalyst:

The unsteady state kinetic model of the partial oxidation of methane over the catalyst account for the interaction of washcoat (Z, ZO) with the active catalyst sites (Pt, PtO)

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

V. Sadykov, L. Bobrova, S. Pavlova et al.

90

(reaction 32 in Table 3). The implemented detailed mechanism contains 32 elementary steps of methane oxidation, 14 gaseous compounds and 13 intermediates on the catalyst surface with corresponding kinetic parameters. The reactor model coupled with the detailed elementary step unsteady-state kinetic model was applied to test the simulation results against available transient experimental data of the reaction ignition during start-up. Due to their high sensitivity to the specific reaction path taken during the transient excursion of the reaction system, good correlations in such comparison approve reliability of an applied mathematical model [213, 214]. The experimental data with the full-size hexagonal corundum extruded monoliths with triangular shape channels (Figure 12) and active component comprised of 0.4 wt.% Pt/10 wt.% La-Ce-Zr-O were taken for the comparison with the numerical results. The numerically predicted gas-phase axial temperature profiles as the functions of time (curves) in comparison with the measurements (symbols) are given in Figure 25.

0

Temperature, C

1200 1000

40s 30s

800

20s

600

10s 5s

400 0.00

0.01

0.02

0.03

0.04

0.05

Length, m

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 25. Gas temperature evolution in extruded monoliths with triangular shape channels during startup. Feed mixture: 24 % CH4 in air, T=420oC, P=1atm, uo=0.5 m/s (STP). Lines - modeling predictions; symbols - experimental data. Adapted with permission from ref 211. Copyright 2007 Elsevier B.V.

Generally, the simulation is shown to be in good agreement with the experimental results, yet for the first 20 seconds the predicted temperature in the front part of the monolith increases slightly faster than the experimentally measured temperature. During this time, the hottest part was found to be at the length of 0.01m in the experiment and at 0.005m in the simulations. The error bar of the measurements certainly existed, but the slower temperature rise in the experiments in comparison with the 2-D modeling data was also noted earlier by Schwiedernoch et. al. [213]. The numerically predicted concentrations of components in the product gas (Figure 26a) are compared with the experimentally derived data (Figure 26b). Again, a fair qualitative agreement between the measured and simulated species profiles is observed. During light-off, carbon dioxide, as a product of complete oxidation of methane, appears initially. Then, synthesis gas selectivity slowly increases with rising temperature. On the contrary to the data for Rh -loaded monolith [213], where CO formation starts before hydrogen formation, hydrogen is detected first during the ignition of Pt/Ce-Zr-La/α-Al2O3 monolith. And again, the simulated gas phase composition reaches steady state faster than in the real transient experiment. This fact was also observed by Schwiedernoch et. al. [213]. A

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Syngas Generation from Hydrocarbons and Oxygenates with Structured Catalysts

91

special study should be undertaken to clarify the matter discussed. However, the detailed reactor and reaction models can be applied in studying non-stationary/non-equilibrium area to obtain fine details of chemical processes. Detailed simulations allow insights into processes which could not be gained through experiments alone, particularly when extreme reaction conditions (such as very high temperatures) are involved. 0.4

0.4

Xi , mole fraction

0.3 H2

CH4 0.2

CO

out

Xiout, mole fraction

H2

0.1

0

30

60

CO 0.1

0.0

90

120

150

180

Time, sec

a)

0.2

CO2 CH4

CO2 0.0

0.3

0

30

60

90

120

150

180

Time, sec

b)

Figure 26. Time dependencies of main components in the product gas. Feed mixture: 26.7 %. CH4 in air, To=400oC, P=1atm, uo=0.34 m (STP)/s. a) - experimental data, b) - modeling results. Adapted with permission from ref 211. Copyright 2007 Elsevier B.V.

1400

1400

6

o

Solid phase temperature, C

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

o

Gas phase temperature, C

Simulation of the thermal behavior of the ceramometal monolithic catalyst (Figure 27) shows the transient development of the gas temperature (left) and the solid temperature (right) profiles during the light-off process occurred on a time scale of seconds. A large difference between the gas and solid phase temperatures is observed. This agrees with results of IR thermography and thermocouple measurements performed by Basini et. al. [215, 216] which revealed a large temperature difference between the surface and the gas phase, the surface temperature even exceeding adiabatic gas temperatures.

1200 1000 800 600 400 0.0

6 5 4 3 2 1

0.2

0.4

0.6

0.8

dimensionless length

1.0

1200

5

1000 800

4 3 2

600 400

0.0

1 0.2

0.4

0.6

0.8

1.0

dimensionless length

Figure 27. The temperature evolution in ceramometal monolith (30 PPI) during light-off: 1-in 0.8s, 22.4s, 3-4s, 5-5.6s, 5-7.2s, 6-12s. Feed mixture: 25 mol.%. CH4 in air, To=4000C, P=1atm, uo=0.5 m (STP)/s.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

V. Sadykov, L. Bobrova, S. Pavlova et al. 3

1.0

2

1

0.8

4 5 0.4 0.2

6 0.2

Н2

0.30

0.6

0.0 0.0

0.35

Gas species, mol.fr.

Fraction of oxygen covered sites

92

0.4

0.6

0.8

0.25 0.20 0.15 СО

0.10 0.05 О2 0.00

1.0

0.0

СО 2

СН4 Н2 О

0.2

0.4

0.6

0.8

1.0

dimensionless length

dimensionless length

Figure 28. Fraction of oxygen-covered surface sites for the simulated light-off in ceramometal monolith (left): 1-in 0.8 s, 2-2.4 s, 3-4 s, 5-5.6 s, 5-7.2 s, 6-12 s, axial distribution of the gas phase species in steady state (right).

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Complex dynamic behavior of the surface species is revealed by the simulation data. In spite of the fact that the reactants’ residence time in the operational conditions is of the order of milliseconds, the vacancies coverage reaches steady state rather slowly (Figure 28, left) due to the multi-step heterogeneous reaction mechanism. Calculated molar gas-phase concentrations at the steady state of methane oxidation in the monolith (Figure 28, right) show that the main chemical transformations occurred within 10% of length at the monolith entrance. Besides the fact that the high surface temperature is a major threat to the stability of supported noble metal catalyst, it is possible that the catalyst generates radicals involved in the chain reaction propagation on the surface, in the film at the gas-solid interface or even in the gas phase [23, 217]. It is generally believed that at the atmospheric pressure homogeneous reaction paths play a minor role in the methane partial oxidation. To estimate the occurrence of gas phase chemistry at the operational conditions typical for the methane partial oxidation, we applied a detail kinetic model (see Table 4) based on a free radical mechanism proposed by Berger and Marin [218] in the reactor model [211]. The gas-phase mechanism contains 40 reversible elementary free-radical reactions with 13 molecules and 10 radicals. General features of the reaction mechanism do not depend on the total pressure. Effect of pressure was taken into account by the variation of kinetic parameters of unimolecular reactions. The curve of the pseudo-first order rate coefficient ku for an unimolecular reaction

A + M → A1 + A2 + M with r = ku C A , against pressure is called the fall-off curve of the unimolecular reaction. The value of a collision partner M for species A, known as the thirdbody concentration CM (“bath gas”), consists of all the species in the reaction mixture. The third-body concentration CM is commonly used in spite of the term “pressure” falloff [72]. The value for CM was taken as the weighed sum of concentrations of all the molecular species. The weight factor wi takes into account the relative collision efficiencies (assumed to be independent of temperature and reaction) of third bodies [218]:

CM = ∑ wi Ci

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Syngas Generation from Hydrocarbons and Oxygenates with Structured Catalysts

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Table 4. Model for the gas-phase partial oxidation of methane to synthesis gas in the absence of a catalyst. №

Reactions

k0

1

CH 4 + M ↔ CH 3• + H • + M

0.24 ×1017

438.98

-6.3

2

CH 4 + O2 ↔ CH 3• + H 2O •

0.398 ×108

223.03

0.1

3

CH 4 + H • ↔ CH 3• + H 2

0.473 ×108

50.31

2.0

4

CH 4 + O • ↔ CH 3• + OH •

0.173 ×109

49.12

12.0

5

CH 4 + OH • ↔ CH 3• + H 2O

0.659 ×108

34.54

4.8

6

CH 4 + HO2• ↔ CH 3• + H 2O2

0.128 ×108

88.18

5.7

7

CH 3• + O2 ↔ CH 2O + OH •

0.396 ×106

54.29

37.8

8

CH 3• + O2 ↔ CH 3O • + O •

0.102 ×1010

151.30

11.7

9

CH 3• + HO2 • ↔ CH 3O • + OH •

0.255 ×108

0.00

23.6

10

2CH 3• + M ↔ C2 H 6 + M

0.329 ×107

-11.34

6.9

11

CH 3O • + M ↔ CH 2O + H • + M

0.383 ×109

81.12

7.8

12

CH 2O + O2 ↔ HO2• + CHO •

0.282 ×109

184.27

8.7

13

CH 2O + OH • ↔ CHO • + H 2O

0.951×109

7.74

13.5

14

CH 2O + HO2• ↔ CHO • + H 2O2

0.461×107

43.62

14.3

15

CH 2O + CH 3• ↔ CHO • + CH 4

0.266 ×107

13.39

8.6

16

CHO • + M ↔ CO + H • + M

0.835 ×108

47.07

3.8

17

CHO • + O2 ↔ CO + HO2•

0.305 ×108

13.74

10.2

18

CO + HO2• ↔ CO2 + OH •

0.474 ×108

73.95

33.5

19

C 2 H 6 + H • ↔ C2 H 5 • + H 2

0.223 ×109

44.10

4.5

20

C2 H 6 + OH • ↔ C2 H 5• + H 2O

0.230 ×109

18.60

7.4

21

C2 H 6 + CH 3• ↔ C2 H 5• + CH 4

0.874 ×109

97.64

2.5

22

C2 H 5 • + M ↔ C2 H 4 + H • + M

0.317 ×1015

195.98

0.8

23

C2 H 5• + O2 ↔ C2 H 4 + HO2•

0.377 ×105

-1.56

7.2

24

C 2 H 4 + H • ↔ C 2 H 3• + H 2

0.542 ×109

62.36

4.9

25

C2 H 4 + OH • ↔ C2 H 3• + H 2O

0.205 ×108

24.86

7.8

26

C2 H 4 + CH 3• ↔ C2 H 3• + CH 4

0.416 ×107

46.56

2.9

27

C2 H 3• + M ↔ C2 H 2 + H • + M

0.200 ×1015

166.28

0.1

28

C2 H 3• + O2 ↔ C2 H 2 + HO2•

0.121×106

0.00

6.5

29

C2 H 3• + O2 ↔ CH 2O + CHO •

0.542 ×107

0.00

44.8

30

C2 H 5• + CH 3• ↔ C3 H 8

0.105 ×109

0.00

3.8

a

Ea , kJ/mol

A / RT

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

93

V. Sadykov, L. Bobrova, S. Pavlova et al.

94

Table 4. (Continued)

a

Ea , kJ/mol

A / RT

0.109 ×107

35.64

-0.4

0.238 ×1010

30.44

4.9

0.433 ×10

157.69

1.0

0.119 ×107

11.01

7.3

0.728 ×10

9

77.91

14.3

0.150 ×103

-6.98

6.4

14.92

5.6



Reactions

k0

31

C2 H 4 + CH 3• ↔ C3 H 7 •

32

C3 H 8 + H • ↔ C3 H 7 • + H 2

33

C3 H 7 ↔ C3 H 6 + H

34

C3 H 7 • + O2 ↔ C3 H 6 + HO2 •





15

35

O2 + H ↔ OH + O

36

O2 + H • + M ↔ HO2• + M





a



37

HO2 + HO2 ↔ O2 + H 2O2

0.121×10

38

H 2O2 + M ↔ OH • + OH • + M

0.967 ×1010

159.66

14.2

0.304 ×10

29.08

2.9

0.506 ×109

3.66

26.2





39

OH + H 2 ↔ H 2O + H

40

HO2• + H • ↔ OH • + OH •





7

8

k0 is expressed in s-1 or m3mol-1s-1 or m6mol -2s-1. The relation between the affinity A , the forward G H G H (r ) and backward reaction rate (r ) is ln(r / r ) = A / RT . Conventional two-parametric Arrhenius dependency was used to express the rate constants.

The values of the weight factors required for the calculation of the concentration of third bodies were H2O - 6.5, CH4 - 6.5, CO2 - 1.5, CO - 0.75, O2 - 0.4, N2 - 0.4, relative to hydrogen. Other weight factors were equal to unity [219]. The first order approximation is assumed for ku :

ku =

kCM k∞ 1 + kCM

where k∞ is the rate coefficient for unimolecular reactions at a high-pressure limit. At

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

sufficiently high pressure, that is, when kCM>1, ku is independent of the total pressures, and approaches k∞ - the rate coefficient for unimolecular reactions at a high-pressure limit. At low pressure, that is, when kCM0.65, syngas yield (first of all, hydrogen) decreased due to combustion. However, for all that higher oxygen content in the feed has more pronounced effect on the concentration of hydrogen then CO. Behavior of the other major products indicates that the deep oxidation route is getting more visible at further increasing oxygen –to carbon ratio. However, the feed flow rate varied in wide limits affect rather the catalyst temperature than product gas composition. Thus, decrease in the contact time in 3 times (from 12 ms to 4 ms) resulted in decrease of synthesis gas yield only by 7%. Maximal concentrations of desired products, H2 and CO, were detected at the lower limit of flow rate (1.8 m3/h).

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

118

V. Sadykov, L. Bobrova, S. Pavlova et al.

Figure 59. Structured catalytic package comprised of sheets and gauzes.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 60. View of a reactor with the catalytic package for studies of n-decane partial oxidation at contact times 0.27-2 ms.

The catalyst temperature increases with the flow rate as expected (Figure 62) which helps to maintain a high syngas yield even at very short contact times. As contact time increased, the backside catalyst temperature declined steadily (Fig 62, b), thus indicating that contribution of endothermic routes became more important. The temperature in front of the catalytic monolith is an intricate function of the interplay of chemistry and heat-mass transfer. In partial oxidation of decane at very short (0.27-2 ms) contact times on package comprised of stacked fechraloy sheets and gauzes, to ensure a stable performance in runs with a low exothermicity (at O2/C ratio 1000

>1000

oxygenates. It should be noted that this order is based on thermodynamic calculations without considering the effects of the reaction kinetics control. The kinetic limitations, which may determine a catalytic system, can have important consequences on the order of formation probability predicted from thermodynamics.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Fischer-Tropsch Synthesis with Fe-Based Catalysts

205

5. OPERATING CONDITIONS AND REACTORS Typical reaction conditions in FTS are temperatures comprised between 473 and 623 K and total pressures in the range 2-6 MPa. The FTS typically operates in one of the two following temperature regimes [45]: (i) Low temperature Fischer-Tropsch (LTFT). This process operates with temperatures ranging between 470 and 530 K with either Fe or Co catalysts to form predominantly high molecular weight hydrocarbons (waxes). (ii) High temperature Fischer-Tropsch (HTFT). This process uses a Fe catalyst between 570 and 630 K to form mainly gaseous hydrocarbons.

Δ Go/n (kJ/mol)

+100

CH3OH

C2H5OH C4H10 C2H4

+50

C2H6 CH4

0

300

1000

Temp. (K)

-50

-100

-150

Figure 5. Change of the Gibbs free energy (∆G0) for the formation of different FTS reaction products

Table 3. Change of the Gibbs free energy (∆G0) for the formation of C1-C3 FTS products at 500 K Carbon number 1

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

2

3

Compound CH4 CH3OH C2H6 C2H4 C2H5OH C3H8 C3H6 C3H7OH

∆G0/n (kJ mol-1) -92.0 +22.2 -61.0 -23.0 -14.2 -51.8 -32.2 -33.4

In principle, different reactor technologies are suitable for performing the highly exothermic FTS (enthalpy of reaction 170 kJ mol-1). Each type of process operates with different reactors. Moving bed reactors are typically employed in the HTFT, whereas fixed bed multitubular or slurry bubble column reactors found application in LTFT. These reactors are considered as the most advanced option and are widely used in industrial application. Thus, fixed-bed multi-tubular rectors are used in the Shell Middle Distillate Synthesis (SMDS) process (tubular reactors with a diameter of 2.54 cm and a length of 12.2 m). These

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

206

M. Ojeda, T. Herranz, F.J. Pérez-Alonso et al.

reactors are easy to manipulate and to design because the parallel tubes behave very similarly. This reactor presents, however, several disadvantages, such us high pressure drop, low catalyst utilization and insufficient heat removal. The pressure drop can be minimized by decreasing the catalyst pellets size (1-3 mm), although this leads to a negative effect on products selectivity. The formation of hot zones inside the reactor can be partially avoided by adding liquid products at the reactor inlet, operating thus in the trickle-flow regime. However, severe mass transfer limitations limit this approach. Another possibility for heat removal in fixed bed reactors is the application of sufficiently high gas recycles with external heat removal under adiabatic reactor operation. The other type of reactors, slurry bubble column rectors wits suspended catalyst (developed by Sasol), manage reasonable well heat transfer issues. Moreover, these reactors use catalyst powders with dimensions of 10 to 200 μm. Thus, the influence of internal mass transfer resistances are negligible and optimal activity and selectivity can be achieved. The slurry reactor appears to be the most efficient system for production of light olefins and gasoline and diesel production. Nevertheless, the slurry bubble reactors present important problems related to the catalyst separation. Moreover, the scaling-up could be a serious drawback to this type of reactors, which otherwise, found broad acceptance for low temperature FTS. Advantages and disadvantages of fixed bed and bubble column reactors are summarized in Table 4 (adapted from reference [45]). Table 4. Advantages (+) and disadvantages (-) of FTS reactors (adapted from [45])

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Pore diffusion Amount of catalyst Mass transfer Heat removal Catalyst exchange Catalyst attrition Scale up Cost

Fixed Bed + + + + -

Bubble Column + + + +

Nowadays, different approaches are under consideration in order to improve reactor performances in FTS. Thus, the use of alternative catalysts geometry is a preferred option. Honeycomb monolith catalysts (large number of identical, parallel channels with a high cell density) present several advantages, such us high external surface area, easy scale up and low pressure drop. Furthermore, they improve very significantly mass transfer between gas, liquid and the catalyst phase because of the high surface area. However, liquid recirculation is necessary to maintain high liquid flow rates required to maintain the slug-flow or Taylor-flow regime inside the capillaries. The application of foams for FTS appears also as a promising alternative because of satisfactory axial and radial heat and mass transfer combined with a low pressure drop. Other approaches involve the use of novel reactors; the so- called microstructured reactors show adequate features because the presence of a large number of parallel channels leads to

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Fischer-Tropsch Synthesis with Fe-Based Catalysts

207

superior heat transfer and mass transfer properties. In fact, even for high exothermic reactions such as the FTS, isothermal operation is possible with microstructured reactors. Finally, membrane reactors also appear as an option to be investigated in the future. These reactors distribute the reactants through a membrane, which is reflected in an improved temperature control. Furthermore, the reaction selectivity can be manipulated because it depends on the H2/CO ratio. Thus, a distribution of H2 in a stream of CO can lead to a decrease in methane selectivity and to a higher yield to long chain hydrocarbons. N2

Fe2+ solution

CO2

Temperature Controller

Fe2+ solution

NaHCO3 Precipitation

NH4OH

Figure 6. Preparation of FTS Fe catalysts by conventional (co)precipitation with a pH-state (left) or a chemical buffer (right)

6. IRON CATALYSTS FOR FISCHER-TROPSCH SYNTHESIS

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

6.1. Preparation of Fe-Based Catalysts Many different methods have been extensively studied for the preparation of Fe-based Fischer-Tropsch Synthesis catalysts. The earliest catalysts, prepared by Fischer and Tropsch, consisted of fused iron treated with alkali [5]. These catalysts were prepared by melting iron ore in the presence of one or more promoters. The resulting catalyst precursor, composed predominantly of magnetite (Fe3O4), possesses a very low surface area (1-15 m2 g-1). Although this type of catalyst is inexpensive, the observed fast deactivation hinders its use in a commercial scale. An alternative approach aiming to prepare catalyst with high surface area consists in the pyrolysis of a Fe precursor, for instance, iron carbonyl compounds. These materials show surface area values as high as 300 m2 g-1 with a medium particle size of ca. 3 μm. The material produced in this manner is formed by highly dispersed particles. Nevertheless, some particles are agglomerated during initial FTS turnovers and the high surface area is decreased in some extent. Currently, the most common method used to prepare Fe-based FTS catalysts is via aqueous precipitation (Figure 6, left) or hydrolysis of Fe2+ and/or Fe3+ salts (nitrates, chlorides, acetates, etc.) [46,47]. Typically, the iron solution is treated with an aqueous solution of NH4OH or (NH4)2CO3, leading to the precipitation of the Fe ions as

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

M. Ojeda, T. Herranz, F.J. Pérez-Alonso et al.

208

oxyhydroxides (FeOxHy). The resulting precipitated material is often washed with deionized water, although alcohols, which have lower surface tension than water, have been also used in order to minimize pore pinching during subsequent drying, forming thus solids with higher surface area values [48]. The dried iron precursor is subsequently decomposed to hematite (Fe2O3) by treating in air at high temperature (600-800 K). The physical and chemical properties of the solids prepared by the aqueous precipitation methodology are influenced by a high number of experimental parameters, including precipitating agent, solution concentration, precipitation temperature, pH, pretreatment temperature, ageing and drying conditions, etc. When monometallic catalysts are prepared, the precipitation of iron ions is simple and fast. However, the situation becomes more complicated when bi- or multi-metallic catalytic precursors need to be synthesized. In this case, the precipitation conditions need to be controlled more carefully in order to obtain a homogeneous distribution of the different components. For instance, pH must be adjusted and controlled using a pH-state or a chemical buffer (Figure 6, right). Thus, the carbonate/bicarbonate buffer can be used by adding a solution of sodium bicarbonate (NaHCO3) and bubbling CO2 through the solution in order to keep constant the pH around 8.5 when iron nitrates are employed as precursors [49,50].

Oil phase

Micelle

Microemulsion water-in-oil (w/o)

Aqueous phase

Oil phase

Micelle Microemulsion oil-in-water (o/w)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 7. Microscopic structure of water-in-oil and oil-in-water microemulsions

A controlled precipitation or coprecipitation method occurring inside aqueous micelles (microemulsion) has been recently developed to prepare Fe-based catalysts [51]. A microemulsion is an optically transparent and thermodynamically stable mixture constituted by an organic phase and an aqueous solution stabilized by a surfactant [52]. The microemulsion is called water-in-oil (w/o) when the aqueous solution is the minority phase (Figure 7). In contrast, the oil phase is the minor phase in the oil-in-water (o/w) microemulsions. The microscopic structure of a microemulsion w/o consists in micelles of aqueous phase surrounded and stabilized by the polar head of surfactant molecules. The use of microemulsions as nanoreactors has been generalized in order to synthesize nanoparticles of controlled size [53]. These nanoparticles can be afterwards deposited on a high area inorganic support [54]. The use of the microemulsion technology is an ideal technique for the

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Fischer-Tropsch Synthesis with Fe-Based Catalysts

209

preparation of materials containing two (or more) metallic or oxide phases because the different species are homogeneously mixed within the micelles, rendering therefore solids with high internal homogeneity and an optimal interaction between its constituents [53]. Other advantages of the materials prepared by microemulsions are the high surface area and good stability [55,56]. This methodology has been proved to yield more active and selective catalysts compared to more conventional procedures because of the resulting higher surface area of the samples and the higher extent of interaction between components in the case of bimetallic catalysts [57,58]. In contrast to other metals, Fe-based catalysts for FTS are usually used as unsupported materials. One of the major drawbacks related to the utilization of massive unsupported catalysts is the physical degradation, producing catalyst fines as a consequence of the volume changes that occur during the FTS reaction. These fines either will plug the fixed-bed reactors, generating a large pressure drop, or will difficult catalyst separation in slurry reactors [59,60]. The development of appropriate supports and/or binders, which increase attrition resistance of the samples, improves the catalyst life time. High surface are oxides like silica (SiO2), alumina (Al2O3), titania (TiO2), magnesia (MgO), manganese oxide (MnO) and zirconia (ZrO2), are among the supports most frequently used. Several advantages of supported iron catalysts, such as an improved catalytic stability and lower deactivation rate, have been reported [61]. Nevertheless, catalysts containing a binder or support usually display lower activity than the unsupported counterpart. This has been attributed to the development of strong metal-support interactions that affect the reducibility of the iron phases during the pretreatment step. Thus, it has been proposed that wustite (FeO, metastable phase) is formed in the case of supported catalysts. FeO is inactive in FTS because it cannot be transformed into iron carbides, which is thought to be the active species in FTS. This results in lower catalytic activities when using supported Fe materials [62]. The wustite Fe phase has been also detected in samples containing doping agents that difficult iron reduction [63]. The use of the adequate preparation conditions is of critical importance to synthesize Fe catalysts with high dispersions and optimal reducibility [64]. More advanced and sophisticated preparation methods aiming to obtain high surface areas and well-crystallized iron oxide nanostructures have been also proposed by forming solid solutions at high temperature solution following methodologies similar to those used for preparing semiconductor quantum dots. Thus, maghemite (γ−Fe2O3) has been prepared by thermal decomposition of an iron complex in octylamine [65]. Oleic acid and trimethylamine oxide were used during the synthesis to control the growth of the particles and to provide a protective surfactant capping layer [66]. Similarly, magnetite (Fe3O4) has been synthesized using iron acetylacetonate as precursor in a mixed solution of oleic acid, 1,2-hexadecanediol, oleylamine and phenyl ether [67,68]. Other preparation techniques described in the literature are based on ultrasound and laser heating [69]. Rice et al. [70] utilized a laser to induce iron carbonyl present in a stream of ethene to form iron carbide catalysts. This preparation method is, however, difficult to scale to a commercial level because the production cost is not competitive at all with those materials prepared by aqueous precipitation approaches. The solvated metal atom impregnation (SMAI) method allows the preparation of Febased FTS catalysts with precise control of the size of crystallites in the nanoscale range [71]. This method consists in the decomposition of an iron complex over a high area support in inert atmosphere. The reactivity of Fe nanoparticles prepared in this way is, however, similar to that shown by supported iron catalysts prepared by conventional impregnation [72].

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

210

M. Ojeda, T. Herranz, F.J. Pérez-Alonso et al.

Moreover, a rapid deactivation process occurs when contacted with the reactants stream, probably due to the sintering of the small Fe crystallites. Model iron-based catalysts prepared by metal evaporation in ultrahigh vacuum have been also synthesized in order to perform basic research. For example, Yubero et al. [73] prepared thin films of hematite grown on silicon wafers by either Fe evaporation in an oxygen atmosphere or ion beam induced chemical vapor deposition (IBICVD). In the first case, metallic iron was evaporated (rate ~1 Å material min-1) by joule heating through a tantalum wire where a high purity (99.5%) Fe filament was grasped. In the IBICVD method, Fe(CO)5 vapor was passed over a silicon substrate which is bombarded by O2+ ions that decompose the iron compound [74].

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

6.2. Catalytic Promoters The addition of different promoters to Fe-based catalysts is a common approach for the development of materials displaying an improved catalytic performance. Thus, the presence of different chemical elements can enhance the activity and/or selectivity of the active sites. The exact role of the promoters is usually discussed in terms of electronic and structural effects. In general, structural promoters increase the number of active sites of the catalyst, while electronic promoters increase the intrinsic reactivity (turnover frequency, TOF) of these active sites [75]. Typical elements used as promoters for Fe-based catalysts include alkaline, alkaline-earths, Cu, Mn, SiO2, Al2O3, etc. Alkaline elements (potassium is the most studied) act as chemical promoters by modifying the adsorption energetics of the reactants (H2 and CO) onto the active sites. The observed effect of these elements on the catalytic behaviour of Fe catalysts has been explained as a consequence of the Fe ability to withdraw electronic density from potassium, resulting in a strengthening of the Fe-CO bond [76]. The corresponding higher surface coverage of the adsorbed CO species (θCO*) results in a lower relative coverage of hydrogen species (θH*). As a consequence, the probability of chain termination pathways by hydrogen addition to form paraffins decreases, being reflected in a higher selectivity towards olefins, high molecular products and the depletion of methane formation. However, it is still unclear the mechanism involved in the interaction between the alkaline elements and the Fe active phase (carbide iron species) in order to produce the atomic contact required for these electronic effects. Alkaline elements also affect the FTS reaction selectivity through secondary pathways. Branched paraffins are mainly formed via isomerization reactions of linear hydrocarbons catalyzed by the acid sites present on the support surface. Potassium (or other alkaline element) titrates these acid sites, decreasing thus the formation of non-linear FTS products [39]. Moreover, alkaline promoters increase the selectivity to CO2 via water-gas shift reactions. The amount of alkali added to the Fe catalyst determines significantly the extent and the consequences of the promoter effect. Although contradictory reports can be found in the literature, catalyst activity increases generally with increasing promoter loadings, reaches a maximum at a certain content, and then declines with further additions [26,77]. The experimental conditions also affect considerably the observed effect when promoters are added. Raje et al. [78] carried out a systematic study about the effect of K promoter on the

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Fischer-Tropsch Synthesis with Fe-Based Catalysts

211

FTS activity with Fe catalysts. They found that the relationship between potassium loading and FTS rate depends on the space velocity, and consequently, on synthesis gas conversion level. Thus, the FTS activity decreases with potassium loading at low conversions. At intermediate conversions, a maximum in FTS activity with potassium loading is observed, while potassium slightly enhances the reaction rate at high CO conversions levels. These effects are explained considering the influence of WGS reaction in all conversion regimes. The overall FTS activity is independent of the WGS reaction at low synthesis gas conversions. In this situation, K acts as a catalyst poison and FTS activity decreases with promoter loading. As the synthesis gas conversion increases, the partial pressure of H2 is significantly decreased in the reactor. Under these conditions, the participation and contribution of the H2 formed in situ by WGS pathways becomes more important. The extent of WGS reaction increases with potassium loading. In summary, the maximum FTS activity is found either: (i) with intermediate potassium loadings and synthesis gas conversion values; or (ii) high potassium loadings and high conversions. Nevertheless, the optimum amount of potassium promoter needs to be determined for each catalyst and FTS conditions. Copper addition to Fe-based Fischer-Tropsch catalysts has been also extensively used because of its ability to improve the reducibility of Fe oxides. When copper oxide is reduced to metallic Cu, the crystallites formed provide H2 dissociation sites, which in turn lead to reactive hydrogen species. Iglesia et al. [25] have shown that these species increase the rates of reduction and carburization of Fe precursors and of nucleation of reduced Fe-containing phases. These faster nucleation rates reflect a larger number of nuclei, which ultimately lead to higher active surface areas. These authors observed the same effect when potassium is added to catalyst. The previously reported decrease in FTS rates at low conversions with increasing K content was not observed in this case. In a doubly promoted sample with Cu and K, the effects are nearly additive, suggesting an almost independent effect of each promoter. A promoter surface density of 1 and 2 atoms nm-2 for Cu and K, respectively (normalized by the precursor oxide surface area) was found to yield Fe catalysts displaying the highest FTS rates. Some attempts to improve the density of active sites and to decrease the selectivity to CO2 have led to the replacement of Cu by Ru [25]. The addition of Ru instead of Cu increases significantly the hydrocarbon synthesis rate while decreasing slightly the selectivity to carbon dioxide. Similarly to Cu, the presence of Ru is reflected in higher reduction and carburization rates compared to the unpromoted samples. However, Ru forms smaller crystallites of reduced Fe species than Cu, which yields catalysts with higher FTS rates. Zr and Cr promoters increase the catalytic activity of Fe-based catalysts by increasing the number of active surface intermediates on the iron surface, leading thus to higher yields to hydrocarbon products [79]. Manganese has been reported to act as a chemical and structural promoter of Fe catalysts. The catalytic behaviour of mixed iron-manganese oxides is found to be influenced by the preparation technique and the structural properties of the catalytic precursors [63,80]. The preparation of these systems by the microemulsion technology results in an improved interaction between the two components, rendering more homogeneous samples than the traditional coprecipitation route and enhancing the Fe-Mn interaction [57]. Moreover, the catalytic performance of the samples depends on the Fe/Mn atomic ratio, being necessary a small amount of manganese (only 5 wt.%) to enhance the catalytic activity and improving the hydrocarbons formation rate. Furthermore, it is significant that Mn promotion leads to a

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

212

M. Ojeda, T. Herranz, F.J. Pérez-Alonso et al.

higher olefin-to-paraffin ratio [58]. A very similar effect has been detected when using Ce as a promoter. In this case, the activity of the catalyst is enhanced when Fe-Ce interactions are developed, which can be achieved preparing Fe-Ce solid solutions by the coprecipitation technique. The specific identity of the Fe-Ce structure obtained depends on the relative concentration of Fe and Ce. Thus, Fe and Ce atoms are dissolved within the ceria or hematite structure, respectively [49]. Further studies have shown that, irrespective of preparation method, a solid solution is formed after treating the material at high temperature when a microscopic contact between Fe and Ce cations in the precursor is present [50]. This was evidenced by several characterization techniques, including X-ray diffraction, and Raman and Mössbauer spectroscopic techniques [49,50]. An important consequence of the formation of Fe-Ce mixed oxides is that the solid surface area is significantly increased. As a result, those catalysts displaying a higher degree of iron-cerium interactions show a better catalytic performance. The promotion of Fe catalysts with Ce can be directly related to the formation of Fe-O-Ce bridges in the calcined solids. Thus, it was proposed that the Ce promoter effect is consequence of Fe0-Ce(III) ensembles arising from Fe-O-Ce bridges [50]. Precipitated iron catalysts typically contain a structural promoter to prevent total collapse of the highly porous precipitated iron oxide/hydroxide upon calcination and reduction [26]. Structural promoters with a good activity are silica, alumina and zinc oxide [25]. Another approach to obtain catalysts with attrition resistance in FTS is the use of Fe supported catalysts. In general, these materials display lower FTS rates than the unsupported catalysts. Nevertheless, the use of alternative preparation methodologies (microemulsion) has led to Fe/SiO2 catalysts with higher catalytic rates (normalized per mass of Fe) than the unsupported reference [64]. In summary, a typical Fe-based catalyst for FTS consists of bulk Fe oxide promoted several elements. Copper is used to increase the reduction and carburization rates (forming thus a higher number of active sites); an alkaline metal, preferentially K, acts as electronic and textural promoter to improve the rate and selectivity towards large hydrocarbons and olefins. Finally, a structural promoter is required to provide a high surface area and to improve attrition resistance.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

6.3. Activation of Fe-Based Catalysts Co-based catalysts are typically activated in H2 stream at 473-723 K to transform cobalt oxide into metallic Co atoms, which are the active sites in FTS [39]. In contrast, the activation protocol is much more complicated for Fe-based catalysts. The activation of Fe catalysts can be carried out using CO [81,82], H2 or H2/CO mixtures [83,84]. During the pretreatment step, and even during the FTS reaction, Fe species develop several phase transformations. The identity of the Fe phase formed after the pretreatment step influences significantly the catalytic performance of the catalyst, especially at the early stages of the reaction. Nevertheless, the Fe phase initially formed evolves with time as the reaction proceeds. During the activation step, the iron oxide precursor (usually hematite, α-Fe2O3) is transformed into magnetite (Fe3O4), irrespective of the activation gas used for the pretreatment. From this point, different iron phases can be formed depending on the activating atmosphere. When H2 is used, Fe3O4 is converted into metallic Fe (α-Fe0). In contrast, the use of CO or H2/CO mixtures results in the formation of metallic Fe and

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Fischer-Tropsch Synthesis with Fe-Based Catalysts

213

different Fe carbides species, including metastable O-carbides (ε-Fe2C, ε’-Fe2.2C and FeCx,) and TP-carbides (χ-Fe2.5C and θ-Fe3C) [85]. Fe carbides are formed when carbon atoms from CO dissociation are dissolved into the α-Fe0 lattice. The different Fe carbides can be easily identified and characterized by X-ray diffraction techniques because of the changes in lattice parameters of the crystal; Mössbauer spectroscopy is also useful to detect different Fe carbides. In contrast, characterization techniques insensitive to the coordination of the atoms (as X-ray photoelectron spectroscopy, XPS) can detect the presence of Fe carbides, but cannot provide information about the specific type of Fe carbide. As mentioned above, Fe precursors treated in H2 yield exclusively metallic Fe, but this structure evolves into Fe carbides during FTS reaction [86]. These general relationships between pretreatment atmospheres (H2, CO or H2/CO) and Fe phase formed are also influenced by the other factors. Thus, the identity of Fe structure formed during the activation step will also depend on the time of exposure to the reactant feed, the composition of this feed, the reactor system and the activation conditions (temperature and pressure) [87]. Although the exact identity of the active Fe phase (Fe0, FeCx or FeOy) in FTS is still controversial [88-90], a correlation between the carbide content and the rate of hydrocarbons formation has been widely observed [91,92]. Moreover, several carbonaceous species are also formed on the catalyst surface during the activation process in CO or H2/CO. These carbon species are of great importance in FTS catalysis because they are representative of the surface reaction intermediates. Therefore, the identification and the establishment of a structure– activity relationship between Fe phases and surface carbon species is a critical point to improve the FTS rate and tune the reaction selectivity. Temperature-programmed surface reaction with H2 (TPSR-H2 or TPH) has been used to identify the Fe carbides and surface carbon species formed during pretreatment and/or FTS reaction. Bartholomew et al. [93,94] have identified the following species using this technique: Cα (atomic carbonaceous species resulting from CO dissociation), Cβ (polymeric surface species with 2-3 carbon atoms), Cγ (iron carbides, mainly, θ-Fe3C and χ-Fe2.5), and Cδ (graphite-like species). The temperature required to hydrogenate these carbon species to methane (the only hydrogenation product obtained at ambient pressure) is indicative of the degree of their reactivity. Experimentally, it is found that the reactivity decreases in the following order: Cα > Cβ > Cγ > Cδ. By combining in situ temperature programmed treatments using H2 (TPH) and inert gases (TPD), ex situ chemical and structural characterization (XRD, Mössbauer spectroscopy, Raman spectroscopy) after passivation, and measurement of the catalytic activity, the identification of the surface carbonaceous and Fe phases (formed after different pretreatments and stages of the FTS reaction) has been accomplished [95]. The influence of the pretreatment atmosphere (CO, CO/Ar, H2/CO or H2) has been investigated and correlated with the presence and abundance of surface carbon and Fe species, as well as with the changes and evolution of such species during the FTS reaction, establishing thus a relationship between the nature of the Fe phase and surface carbon species and their abundance during the different kinetics episodes of the reaction. In this scenario, the effect of Ce and Mn addition on the identity of the Fe species has been extensively investigated by our group. Figure 8 summarizes the catalysts, reacting atmospheres and characterization techniques used in these studies.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

214

M. Ojeda, T. Herranz, F.J. Pérez-Alonso et al.

As mentioned above, the temperature at which methane is formed in TPSR-H2 experiments is indicative of the reactivity of the surface carbon species. Thus, methane formation from Cα species is observed at ~770 K when a pure iron oxide (hematite) solid is pretreated in CO. In contrast, these species are not detected when the pretreatment is carried out with H2/CO or when catalytic promoters (Ce or Mn) are added. The temperature required to form methane from Cβ species (polymeric surface species with 2-3 carbon atoms) varies between 870 and 950 K, depending again on the catalyst pretreatment and composition. Methane formation via hydrogenation of iron carbide species (θ-Fe3C or χ-Fe2.5) occurs at temperatures above 1000 K. Hydrocarbons

H2

FT S

FTS O H 2/C

FTS

S FT

CO

H/ 2 CO

H2/CO CO/A r

S FT

FTS CO

CO

FeOx

χ-Fe2.5C

Mn(Ce)-FeOx

Fe0

θ-Fe3C

Carbon species

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 8. Transformation routes of iron oxide during pretreatment and FTS reaction

A correlation between the surface carbonaceous species and the identity of the iron carbides formed after the pretreatment step has been found by combining TPSR-H2 experiments with characterization data obtained from X-ray diffraction, and Raman and Mössbauer spectroscopy of the Fe samples after the activation step. Previously to the characterization experiments, the samples were passivated by flowing an 1 vol.% O2/He mixture at room temperature for 1 h, according to the procedure reported previously in the literature [96]. The treatment of Fe catalysts with H2/CO mixtures causes the transformation of Fe2O3 precursor into the Hägg carbide (χ-Fe2.5C), evidenced by XRD and Mössbauer spectroscopy. TPSR-H2 experiments of unpromoted and promoted Fe samples pretreated with H2/CO revealed the exclusive formation of Cβ species (polymeric surface species). On the contrary, the pretreatment of the unpromoted Fe2O3 precursor with CO originates predominantly cementite (θ-Fe3C). Moreover, the TPSR-H2 experiments show that both Cα and Cβ species are formed over the cementite-type carbide. When Ce and Mn promoters are added to the Fe2O3 precursor, the treatment with CO renders the formation of cementite carbide (θ-Fe3C) with Cα and Cβ carbon species, although the relative concentration of Cβ species is higher in this case. This clearly reveals that the catalyst composition and

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Fischer-Tropsch Synthesis with Fe-Based Catalysts

215

pretreatment procedure affects the nature of the surface species. Obviously, the pretreatment with H2 form α-Fe0 in both the unpromoted and promoted Fe samples. Unpromoted and Ce- and Mn-promoted Fe catalyst were tested in the FTS reaction at 573 K and 1.01 MPa for ~170 h after the activation pretreatment with different reactive atmospheres (H2, CO, CO/Ar and H2/CO). As a general trend, the pretreatment with H2/CO mixtures yields the most active catalysts, whereas the use of H2 leads to catalysts with low activity. Those samples pretreated with CO (either pure or diluted with Ar) require Ce or Mn promotion to show relatively high reaction rates. The CO conversion rate (normalized by the number of Fe atoms) of the unpromoted and Ce- and Mn-promoted samples is very similar when the pretreatment is performed in H2/CO mixtures. In contrast, the activation in CO reveals significant differences in reaction rates, following the order: FeCe > FeMn > Fe. A common feature observed for all the catalysts is the increase of the catalytic activity after at certain time-on-stream period (ca. 15 h). This behaviour has been ascribed to a surface reconstruction of the catalyst [86]. Although the structure of the catalyst surface evolves during the reaction, the product selectivity does not change with the reaction time, indicating that more active sites are formed but without changing the catalytic properties. By combining the catalytic activity data with the TPSR-H2 experiments, it can be concluded that in general, those samples containing high concentrations of Cβ species display high FTS rates. In addition, the unequivocal identification of the different type of Fe carbides by XRD and Mössbauer spectroscopy allowed the correlation of iron carbides and surface carbon species. Thus, it is found that Cβ species are preferentially formed over the Hägg carbide (χ-Fe2.5C). In the case of the Ce- and Mn-promoted samples activated in CO, although the cementite-type carbide (θ-Fe3C) is the most abundant species, a higher amount of Cβ species is stabilized compared with the unpromoted Fe sample. Interestingly, the total amount of iron carbides formed during the pretreatment step is not as critical as its nature. This feature is clearly demonstrated from the experiments consisting in treating the catalyst with pure or diluted CO. When the catalyst is treated in pure CO, iron oxide hematite (α-Fe2O3) evolves predominantly to cementite (θ-Fe3C). However, when the treatment is carried out in diluted carbon monoxide (5 vol.% CO/Ar), most of the iron remains in the form of oxides (hematite and magnetite) and only a small amount of cementite is found by XRD. Surprisingly, the treating in CO/Ar leads to more active catalysts in FTS, despite the fact that a higher amount of iron carbide species are formed when the treatment is carried out in pure CO. Mössbauer analysis after FTS reaction of the iron catalysts treated with diluted CO showed that the iron oxide present after the pretreatment has evolved to Hägg carbide (χ-Fe2.5C). In the case of H2 activated samples, Hägg carbide was also found after reaction using Mössbauer spectroscopy. This observation strongly suggests that irrespective of the gas used for the pretreatment, the Hägg carbide (χ-Fe2.5C) is formed under the FTS reaction conditions. Therefore, Hägg-carbide seems to be the active phase responsible for the formation of hydrocarbons from H2/CO mixtures. This is further supported by the fact that the catalytic activity of the samples pretreated with CO increased at some point after the beginning of the reaction [95]. Mössbauer spectroscopy analysis of the COactivated samples after reaching the steady state in reaction reveals that the cementite (θFe3C) present after the pretreatment evolves partially to form Hägg carbide (χ-Fe2.5C). However, the activity of these samples is not as high as the samples pretreated with H2/CO because the extent of this transformation is limited. This result evidences a transformation of

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

216

M. Ojeda, T. Herranz, F.J. Pérez-Alonso et al.

cementite into Hägg carbide during reaction concomitant with an increment of the catalytic activity. The pretreatment with H2 leads to less active catalysts compares to the use of H2/CO mixtures, despite the fact that all the metallic iron atoms evolve to Hägg carbide (χ-Fe2.5C) during reaction. The reason is that these bulk Fe samples sinter during the reduction in hydrogen. This causes the decrease of both the surface area and the catalytic activity compared to the samples activated in synthesis gas. Regarding the Mn- and Ce-promoted samples, the situation is slightly different when the activation is carried out with CO. Although most of the iron is transformed into cementite during the pretreatment, the amount of carbonaceous Cβ species detected by TPSR-H2 is higher than in the unpromoted iron sample and comparable to the amount found in the samples activated in syngas (H2/CO). This explains the higher activity of the promoted samples, even though the type of carbide found after treatment (cementite) is not the most suitable to stabilize the active carbonaceous intermediates (Cβ). In summary, when iron oxide (hematite) is pretreated with H2, CO and H2/CO, different iron phases are formed. Only metallic iron is found when the pretreatment is carried out with hydrogen. Iron carbide, mainly cementite (θ-Fe3C), is formed with CO (pure or diluted) pretreatments. On the other hand, Hägg carbide (χ-Fe2.5C) is formed when the pretreatment is performed with syngas (H2+CO). Surface carbonaceous species are stabilized on both types of iron carbides: Cα species are stabilized on cementite (θ-Fe3C), whereas Cβ species are stabilized on the Hägg carbide (χ-Fe2.5C). When θ-Fe3C species are present, the activity in the FTS is lower, although it can evolve in the reaction medium (H2/CO at ~573 K) into the more active species, the Hägg carbide, where the Cβ intermediates are stabilized. The extent of this transformation is not complete, and therefore, some cementite species remain on the catalyst surface, explaining thus the lower activity of the catalysts activated in CO. Mn and Ce promotion is only effective when the samples are activated with CO, favoring thus the stabilization of Cβ species.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

6.4. Dynamic Character of the Catalytically Active Fe Phases The composition of the bulk and surface of Fe-based catalysts suffers severe changes during pretreatment steps and the FTS reaction. These modifications of the Fe phases are responsible for the typical different kinetic episodes that are observed experimentally. Hence, Schulz et al. [86] reported that the approach to the steady-state with Fe-based catalysts can be separated into several episodes of distinct kinetic regimes (induction periods) when the catalysts are activated with H2. These episodes consist in carbon deposition and Fe carburization processes until the totality of the α-Fe (the main phase after H2 pretreatment) is transformed into χ−Fe2.5C. At this point, the steady-state is reached and the formation of the true FTS catalyst is supposed to occur. However, some authors have not found a correlation between the amount of bulk iron carbide and the activity of the catalyst [97]. They proposed that the bulk iron carbide serves merely as a support of the active surface species. Goodwin et al. [98] investigated the differences between unpromoted and Mn- and Kpromoted Fe catalysts concerning the induction period required to reach the steady-state in FTS. These authors determined the turnover frequency of the catalysts using steady-state

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Fischer-Tropsch Synthesis with Fe-Based Catalysts

217

isotopic transient kinetic analysis (SSITKA). The specific reaction rates were similar for all catalysts. The addition of Mn and/or K to the Fe catalyst leads to shorter induction periods and a higher number of active sites. However, the nature and the evolution of the active Fe phase were not reported in this study. As already described, Bartholomew et al. [94] have identified and quantified different Fe carbide and carbonaceous species on the surface of unsupported and supported Fe catalysts by TPSR-H2. A correlation between the coverage of atomic surface carbon species and the catalyst activity was found in both cases. Herranz et al. [58] have also identified different carbonaceous and Fe carbide species on Mn- and Ce-promoted Fe catalysts after different activation pretreatments by using the same TPSR technique. A straightforward correlation between the concentration of the Hägg carbide (χ-Fe2.5C) species and the catalytic activity was found. Furthermore, the higher activity of the catalysts was directly related to a higher coverage of polymeric surface carbon species (Cβ). Hitherto, the actual nature of the Fe active phase(s), the surface species and their evolution in the different kinetic regimes has not been unequivocally determined with these studies. Unfortunately, these changes cannot be followed with in situ characterization techniques due to the technical complications derived from the stringent reaction conditions (high temperature and pressure). However, a successful experimental strategy to shed light on the nature of the Fe phase transformations occurring during FTS reaction has been recently developed [24]. The first step in this approach involves pre-treating the catalysts with a H2/N2 mixture (1:2 ratio) at 673 K to form α-Fe0 that evolves during the reaction. In this way, the different kinetic episodes can be easily followed. The catalysts are tested in the FTS for different time-on-stream (TOS) values and the surface and bulk composition is studied using several ex situ characterization techniques (TPSR-H2, N2 adsorption-desorption isotherms, XRD and Mössbauer spectroscopy). This procedure requires the use of a suitable passivation protocol to avoid undesirable changes in the catalyst structure as a consequence of contact with ambient air when transferring the samples from the reactor to the characterization instruments. These studies have been conducted over both unpromoted (referred to as Fe) and two 5 at.% Ce-promoted Fe catalysts. The Ce-promoted samples were prepared by impregnation of either iron oxyhydroxide (referred to as FeCe-I) or iron oxide (referred to as FeCe-IC) [24]. It has been observed that the rate of hydrocarbon formation depends on the catalyst structure and composition. Moreover, the reaction rate is also a function of time-onstream. The unpromoted Fe catalyst displays a higher initial FTS rate compared to the promoted catalysts. However, the rate decreases monotonically as the reaction occurs. An inflection point is reached at a certain point (ca. 60 h on stream) and then the rate increases. The initial reaction rates of the Ce-promoted samples are lower compared to the unpromoted catalyst. In fact, the FeCe-IC sample presents a lower reaction rate in the whole range of TOS values because Fe-Ce interactions are not fully developed over this sample and that the addition of Ce actually results in a decreasing number of iron active sites. On the other hand, the catalytic performance of the sample FeCe-I, over which a chemical Fe-O-Ce interaction is effectively developed [49], exceeds the reaction rate of the unpromoted catalyst after ca. 15 h on stream, despite the fact that the initial activity of this sample is the lowest among all the catalysts studied here. The BET surface area of the different catalysts noticeably changes in parallel with the hydrocarbon formation rate as the reaction proceeds. Figures 9 collects the BET and hydrocarbon formation rate vs. time on stream for those catalysts. We have not

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

M. Ojeda, T. Herranz, F.J. Pérez-Alonso et al.

218

1.5

4

40

-1

-1

Catalyst: Fe

50

2

-1

BET surface area (m g )

60

HC formation rate x 10 (gCH2 s gcat )

found a straightforward correlation between the BET surface area and catalytic activity at the early stages of the reaction. However, this is detected as the reaction proceeds. Therefore, the reaction rate is determined by a combination of several factors, not only by catalyst surface area. This is clearly seen in the case of the FeCe-I sample, which records the highest surface area among all the materials yet the lowest reaction rate. Other features should therefore be called upon to explain both the initial activity of the catalysts and the evolution of the reaction rate as the FTS proceeds.

1.0

30 20

0.5

10 0

0

20

40

60

80

100

120

0.0

-1

Catalyst: FeCe-I

-1

60 50

1.5

40

4

2

-1

BET surface area (m g )

HC formation rate x 10 (gCH2 s gcat )

Time-on-stream (h)

1.0

30 20

0.5

10 0

0

20

40

0.0

60

80

-1

4

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

40

-1

1.5

2

-1

BET surface area (m g )

Catalyst: FeCe-IC

50

1.0

30 20

0.5

10 0

HC formation rate x 10 (gCH2 s gcat )

Time-on-stream (h) 60

0

20

40

60

80

100

120

0.0

Time-on-stream (h)

Figure 9. Evolution of the BET surface area and hydrocarbon (HC) formation rate during time-onstream observed with the unpromoted and Ce-promoted Fe samples

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Fischer-Tropsch Synthesis with Fe-Based Catalysts

70

50

CO conversion (%)

CO conversion (%)

60

60

Sample: FeCe-I

0

Fe χ-Fe2.5C θ-Fe3C

Fe3O4

40 30 20

χ-Fe2.5C

10

θ-Fe3C

0

Fe3O4 (traces) 0

10

20

30

40

50

Time-on-stream (h)

60

70

χ-Fe2.5C

50

θ-Fe3C

40

Fe3O4

219

Sample: Fe

30 20 χ-Fe2.5C

10 0

Fe3O4 (traces) 0

20

40

60

80

100

120

Time-on-stream (h)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 10. Relationship between the Fe phases detected (XRD and Mössbauer spectroscopy) and the FTS kinetic episodes with FeCe-I and Fe catalysts.

X-ray diffraction patterns showed that the carburization process of the Ce-containing samples is retarded compared to the unpromoted Fe catalyst. For the Ce-promoted solids, the presence of metallic iron (α-Fe), magnetite (Fe3O4) and iron carbides (χ-Fe2.5C and θ-Fe3C) during the early stages of the reaction was evidenced. By contrast, magnetite and iron carbides were observed for the unpromoted samples. Thus, the higher activity of the Ce-free sample during the early stages of the FTS is explained in terms of its higher carburization degree as compared to the Ce-containing samples. The XRD analysis of this unpromoted catalyst used in the FTS for 40 h on stream indicates that the Hägg carbide becomes the predominant Fe phase at this point, being thus responsible for the increasing rate as the reaction proceeds. However, the situation for the Ce-promoted catalysts is slightly different. In this case, the rate of hydrocarbon formation increases even before the θ-Fe3C phase evolves to χ-Fe2.5C, although χ-Fe2.5C is the predominant carbide phase on the FeCe catalysts, as shown in Figure 10. These observations have been further corroborated by Mössbauer spectroscopy. Both characterization techniques, Mössbauer and XRD, show a lower carburization extent of the Ce-promoted samples. The amount of the Hägg carbide observed in the unpromoted Fe catalyst increases at the expense of the cementite as the FTS proceeds, and eventually, the χFe2.5C structure is the only carbide phase detected. By contrast, both iron carbide phases (cementite and Hägg) are always detected in the Ce-containing samples, although a continuous transformation of cementite into the Hägg carbide along the FTS reaction is observed. The Ce-promoted catalysts develop more active materials than the unpromoted sample once the catalytic precursor is totally transformed into iron carbides (either as cementite or Hägg carbide). It should also be noted that TPSR-H2 characterization found a direct correlation between the concentration of Cβ on the catalyst surface and the catalytic activity for all catalysts. This work evidences the reconstruction process that Fe-based catalysts undergo during the FTS. As it was pointed out above, a direct correlation between the concentration of Cβ species (partially polymerized carbon species) on the catalyst surface and the catalytic activity of all the samples for time-on-stream values higher than 7 h was found. In addition, the increase of the concentration of this type of species was parallel with the increase of the surface area of the solids. In the case of the Ce-free iron sample, the increase of the catalytic

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

220

M. Ojeda, T. Herranz, F.J. Pérez-Alonso et al.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

activity was also accompanied by an increase of the concentration of χ-Fe2.5C species and by a decrease of the θ-Fe3C concentration. According to the XRD and Mössbauer spectroscopy results, this evolution was observed continuously during the reaction. Finally, the θ-Fe3C phase could not be detected after 120 h on stream. The Ce-free iron catalyst underwent the increase of the catalytic activity after 40 h on stream, which coincides with the total transformation of the θ-Fe3C phase into χ-Fe2.5C and with the stabilization of the graphitic deposits concentration on the surface. In summary, a good correlation between χ-Fe2.5C and FT activity was found for the Ce-free iron catalyst. In the case of the Ce-loaded iron catalysts, the transformation of θ-Fe3C species into χFe2.5C is also observed, although the θ-Fe3C species did not evolved totally. Considering that the FeCe catalysts displayed a shorter induction process, Ce addition would be responsible of the formation of Cβ species, independently of the graphitic deposits formation. The effect of Ce addition could be also related with the increase of the surface area. Despite the fact that the θ-Fe3C did not evolve totally into χ-Fe2.5C in the case of the Ce-loaded iron catalysts, Ce addition originated a higher number of active sites, and consequently, the effect of graphitic deposits is less significant in absolute terms. With regard to the promoter effects, it was speculated that Ce-promoted catalysts show higher reaction rates because of the formation on the catalyst surface of tilted CO species, which are easier to dissociate. Ce (III) species are needed in the framework of the catalysts in order to form such tilted CO species. This is supported by the X-ray photoelectron spectroscopy (XPS) results obtained after treating the samples in H2/CO mixtures, showing the presence of these Ce(III) species [49] . In summary, the nature of the surface species and bulk composition of iron-based catalyst is dynamic and depends on the time-on-stream. Several iron phases, including iron oxides, metallic iron and different iron carbides are observed initially but the increase of the catalytic activity of iron based catalysts is directly related with the evolution of iron carbide species to Hägg carbide species (χ-Fe2.5C) and the consequently formation of Cβ species on the catalyst surface. The addition of certain additives as Ce to iron oxide FT catalysts increases the catalytic activity in terms of a faster activation and a higher catalytic activity level in the steady-state. This is due to the stabilization of higher amounts of active reaction intermediates (Cβ).

6.5. Fischer-Tropsch Synthesis with CO2-Containing Syngas The increasing global energy demand during the past decades has caused an enormous raise of the man-made CO2 emissions. Many efforts have been devoted in order to reduce such CO2 emissions as a consequence of the Third Conference of the Parties (COP-3) in Kyoto in 1997 [99]. Furthermore, the improvement of the efficiency of energy conversion or utilization processes strategies requires other secondary approaches, including capture, storage, and fixation of the carbon dioxide produced [100]. In this context, the production of clean fuels via CO2 hydrogenation appears as a very interesting and attractive option from an industrial point of view. Hydrogenation of carbon dioxide has been traditionally carried out with catalysts that have been demonstrated to be active and selective for the FTS reaction, mainly because CO2

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Fischer-Tropsch Synthesis with Fe-Based Catalysts

221

may be a significant component in the synthesis gas fed to the FTS plants [92]. Thus, CO2 is an important component in the synthesis gas obtained via natural gas partial oxidation or steam reforming, in the syngas obtained from biomass gasification and also in the feed to the industrial FTS reactors that recycle unconverted reactants. Moreover, CO2 may be present in the synthesis gas produced through the carbon dioxide reaction with methane, a route of great interest in recent years because it constitutes a promising alternative to utilize the CO2 originated in various anthropogenic processes. The removal of CO2 from H2/CO mixtures is a very expensive process. Therefore, the development of catalysts active in CO2 hydrogenation to hydrocarbons is a key step to increase the feasibility of FTS overall processes [42,101]. K-promoted Fe-based catalysts have exhibited the most promising results in the CO2 hydrogenation to form long-chain hydrocarbons [102,103], although promotion with other elements (Cr, Mo, Mn and Zn) have been also addressed [57,104,105]. These studies have indicated that the reaction proceeds through the reverse water-gas shift (RWGS) reaction (CO2 + H2 → CO + H2O). According to this mechanism, CO2 is first converted into CO, which is subsequently hydrogenated to form different hydrocarbons. Consequently, the direct CO2 hydrogenation to hydrocarbons is irrelevant, and the reaction always takes place through the formation of CO intermediates [57,102]. Therefore, an important requirement for the CO2 hydrogenation catalysts in order to be active is that they must be able to catalyze the RWGS reaction. Iron-based materials are active catalysts for both RWGS and FTS reactions, and as a result, they can be used to obtain hydrocarbons from H2/CO2 mixtures. Fe catalysts attain the steady state activity in FTS reaction through in situ activation as a consequence of the catalyst reconstruction that involves the formation of different types of Fe carbides [24,92]. Riedel et al. [103] showed that the reconstruction of Fe catalyst is similar when using either H2/CO or H2/CO2 mixtures, although the kinetic regimes in CO2 hydrogenation are up to 10 times slower than in H2/CO reaction because a much lower concentration of CO is available to form the Fe carbides. It is also remarkable that the product selectivity was found to be similar at steady-state in both CO and CO2 hydrogenation reactions. This suggests common active sites for both reactions at the steady-state. The hydrogenation of CO2 with unpromoted and Ce-promoted Fe catalysts has been studied recently [106]. Cerium is known to promote both the WGS and RWGS reactions [101], which is a critical requirement in order to activate the carbon dioxide molecules. Ce addition to Fe-based catalysts leads to shorter induction periods in the formation of hydrocarbons from CO2 hydrogenation, which is due to a higher carburization rate of the Cepromoted Fe catalyst. In contrast, a similar product distribution is found with the unpromoted and Ce-promoted catalysts. The hydrogenation of CO2 to hydrocarbons is limited by the approach to the equilibrium at high conversions. This limitation reduces the CO available to form Fe carbides, which is reflected in longer induction periods. This can be partially avoided using higher operation temperatures. In this way, a higher concentration of CO is formed, but possesses negative effects on the selectivity of CO hydrogenation. Therefore, more research about process conditions and catalyst composition is needed to obtain high yields to hydrocarbons in CO2 hydrogenation reactions.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

222

M. Ojeda, T. Herranz, F.J. Pérez-Alonso et al.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

7. ECONOMICS AND FUTURE PROSPECTS Fuels and chemical derived from FTS compete with traditional oil, and therefore, the economic feasibility of producing these fuels and chemicals via FTS would be ultimately dominated by the price of oil. Moreover, different factors, such us reducing liquid fuel dependence, obtaining liquid fuels from coal or strained natural gas, and environmental benefits derived from the BTL process, could play a major role in obtaining tax exemptions for synthetic FTS fuels. Currently, FTS fuels cannot compete with traditional fossil fuels without these incentives. Although somehow scarce, there are open public studies dealing with the feasibility of obtaining liquid fuels from FTS via biomass gasification [10]. These documents coincide in that biomass gasification is not yet a mature technology and improvements in the gasification step, mainly by using pressurized gasifiers, should make FTS from biomass more competitive. Two independent studies reached similar conclusions [107,108] . Thus, the price of FT-liquid fuels would be around US$ 16/GJ (9.1-16.7 €2002/GJ) in the short term. In 2002, oil reference price was roughly 2.6-7.0 €2002/GJ. Since then, oil price have significantly increased up to US$2008150/barrel (US$2008 24.5/GJ) in the first part of 2008, although oil prices plummet from $150 to $40 per barrel in the second part of that year. Obviously, the FTS becomes competitive with traditional oil in the price scenario of years 2007 or early 2008. However, due to the large investments required to construct and operate a FTS plant, volatile market in oil price is an important drawback. Total capital investment for a FTS plant depends on various factors, including the total capacity or syngas source. In this sense, the investment needed for a FTS plant using natural gas as synthesis gas source is lower than that using biomass as a carbon source. This accounts for the pre-treatment and more complex gasification steps, as well as for the cleaning and water-gas shift units. The capital investment needed to construct a biomassderived FTS operating with current technology is calculated to be around 286 M€ for a 400 MWth,HHV (thermal watts, high-heating value) input plant. FT diesel production price at such plant would be around 16.1 €/GJ. A major influence on the diesel price is the scale. Thus, FT diesel price could drop to ca. 14 €/GJ in a 2000 MWth input plant. There exist some strategies to the economics of the FTS plants. Serious attempts to do this should focus first and foremost in decreasing the price of the pre-treatment, gasification, cleaning and conditioning (shift) units. This part of the process accounts for ca. 75 % of the total capital investment of the FTS process. It is foreseen that in the long term, further drops of biomass price (2€/GJ) along with technological developments (oxygen production, catalytic tar cracking) could reduce capital costs around 15 % or 9 % in the FTS cost. For instance, developing more selective catalyst to the diesel fraction could avoid the upgrading unit. In summary, the economic feasibility of GTL, CTL and/or BTL plants depends significantly on the production scale. Thus, large scale plants are required to produce synthetic fuels competitive with those obtained from crude oil.

8. CONCLUSIONS The production of hydrocarbons via FTS is a sustainable route to liquid transportation fuels. The BTL-process is one of the more attractive options because the possibility of using the large reserves of biomass. Fe-based catalysts are especially suitable because of the low

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Fischer-Tropsch Synthesis with Fe-Based Catalysts

223

cost and the ability of converting syngas mixtures with low H2/CO ratios to different hydrocarbons. Nevertheless, adequate catalyst preparation methods and addition of catalytic and structural promoters is required to maximize the yield to hydrocarbons and to tune the product distribution.

REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9]

[10] [11] [12] [13] [14] [15] [16] [17]

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

[18] [19] [20] [21] [22] [23] [24] [25] [26] [27]

Huber, GW; Iborra, S; Corma, A. Chem. Rev., 2006, 106, 4044-4098. Annual Energy Outlook , Early Release Overview, Energy Information Administration, United States, http://www.eia.doe.gov/oiaf/aeo/pdf/earlyrelease.pdf, 2009. International Energy Outlook, Energy Information Administration, United States, http://www.eia.doe.gov/oiaf/ieo/pdf/0484(2008).pdf, 2008. Sabatier, P; Senderens, JB. J. Chem. Soc., Faraday Trans., 1902, 82, 333-337. Fischer, F; Tropsch, H. Brennst. Chem., 1923, 4, 276. http://www.choren.com/en/ biomass_to_energy/ 2009. Wilhelm, DJ; Simbeck, DR; Karp, AD; Dickenson, RL. Fuel Procces.Technol., 2001, 71, 139-148. Steynberg, AP; Nel, HG. Fuel, 2004, 83, 765-770. Navarro, RM; Peña, MA; Fierro, JL. G. Chem. Rev., 2007, 107, 3952-3991; and references therein. Tijmensen, MJA; Faaij, APC; Hamelinck, CN; van Hardeveld, MRM. Biomass Bioenergy, 2002, 23, 129-152. Bergman, PCA; van Paasen, SVB; Boerrigter, H. Pyrolysis and Gasification of Biomass and Waste Meeting, Strasbourg, France, 2002. Khodakov, AY; Chu, W; Fongarland, P. Chem. Rev., 2007, 107, 1692-1744. van der Laan, G. P; Beenackers, A. A. C. M. Catal. Rev. Sci. Eng. 1999, 41, 255-318. van Steen, E; Claeys, M. Chem. Eng. Technol., 2008, 31, 655-666. Iglesia, E; Reyes, SC; Madon, RJ; Soled, SL.Adv. Catal., 1993, 39, 221-302. Iglesia, E. Appl. Catal. A: Gen. 1997, 161, 59-78. Bezemer, GL; Bitter, JH; Kuipers, HP; CE; Oosterbeek, H; Holewijn, JE; Xu, X; Kapteijn, F; van Dillen, AJ; de Jong, KP. J. Am. Chem. Soc., 2006, 126, 3956-3964. Jacobs, G; Das, TK; Zhang, Y; Li, J; Racoillet, G; Davis, BH. Appl. Catal. A: Gen., 2002, 233, 263-281. Martinez, A; Prieto, G. J. Catal., 2007, 245, 470-476. Sun, S; Fujimoto, K; Yoneyama, Y; Tsubaki, N. Fuel, 2002, 81, 1583-1591. Morales, F; de Smit, E; de Groot, FMF; Visser, T; Weckhuysen, BM. J. Catal., 2007, 246, 91-99. Moradi, GR; Basir, MM; Taeb, A; Kiennemann, A. Catal. Commun., 2003, 4, 27-32. Li, S; Meitzner, GD; Iglesia, E. J. Phys. Chem. B, 2001, 105, 5743-5750. Pérez-Alonso, FJ; Herranz, T; Rojas, S; Ojeda, M; López Granados, M; Terreros, P; Fierro, JL G; Gracia, M; Gancedo, JR. Green Chem., 2007, 9, 663-670. Li, S; Li, A; Krishnamoorthy, S; Iglesia, E. Catal. Lett., 2002, 77, 197-205. Bukur, DB; Mukesh, D; Patel, SA. Ind. Eng. Chem. Res., 1990, 29, 194-204. Koizumi, N; Murai, K; Ozaki, T; Yamada, M. Catal.Today, 2004, 89, 465-478.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

224 [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49]

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

[50] [51] [52] [53] [54] [55] [56]

[57] [58]

M. Ojeda, T. Herranz, F.J. Pérez-Alonso et al. Hindermann, JP; Hutchings, GJ; Kiennemann, A. Catal. Rev. Sci. Eng. 1993, 35, 1-127. Dry, M. E. Appl. Catal. A: Gen., 1996, 138, 319-344. Adesina, AA. Appl. Catal. A: Gen., 1996, 138, 345-367. Kobori, Y; Yamasaki, H; Naito, S; Onishi, T; Tamaru, K. J. Chem. Soc., Faraday Trans., 1982, 78, 1473-1490. Biloen, P; Helle, JN; Sachtler, WMH. J. Catal., 1979, 58, 95-107. Biloen, P; Helle, JN; van der Berg, FGA; Sachtler, WMH. J. Catal. 1983, 81, 450-463. Bianchi, D; Tau, LM; Borcar, S; Bennett, CO. J. Catal., 1983, 84, 358-374. Ojeda, M; Ishikawa, A; Iglesia, E; Nabar, R; Mavrikakis, M. AIChE Annual Meeting Conference Proceedings, Houston, USA, 2006. Storch, HH; Golumbic, N; Anderson, RB. The Fischer-Tropsch and related syntheses, Wiley, New York, 1951. Inderwildi, OR; Jenkins, SJ; King, DA. J. Phys. Chem. C, 2008, 112, 1305-1307. Storsaeter, S; Chen, D; Holmen, A. Surf. Sci., 2006, 600, 2051-2063. Anderson, RB. The Fischer-Tropsch synthesis, Academic Press Inc., Orlando, FL, 1956, p. 2. van Dijk, H. A. J. The Fischer-Tropsch synthesis: a mechanistic study using transient isotopic tracing, Ph.D. Dissertation, Technische Universiteit Eindhoven, 2001. Pichler, H; Schulz, H. Brennst. Chem., 1967, 3, 78. Dry, ME. J. Chem. Technol. Biotechnol., 2001, 77, 43-50. Madon, RJ; Iglesia, E. J. Catal., 1993, 139, 576-590. Falbe, J. New syntheses with carbon monoxide, Springer-Verlag, Berlin, 1980. Guettel, R; Kunz, U; Turek, T. Chem. Eng. Technol., 2008, 31, 746-754. Cornell, R. M; Schwertmann, U. The iron oxides, VCH Publishers, Weinheim, 2003. Ziolo, RF; Ginnelis, EP; Weinstein, BA; O'Horo, MP; Ganguly, BN; Mehrotra, V; Russell, MW; Huffman, DR. Science, 1992, 257, 219-223. Li, S; Krishnamoorthy, S; Li, A; Meitzner, GD; Iglesia, E. J. Catal., 2002, 206, 202-217. Pérez-Alonso, FJ; López Granados, M; Ojeda, M; Terreros, P; Rojas, S; Herranz, T; Fierro, JL. G. Chem. Mater., 2005, 17, 2329-2339. Pérez-Alonso, FJ; López Granados, M; Ojeda, M; Herranz, T; Rojas, S; Terreros, P; Fierro, JLG; Gracia, M; Gancedo, JR. J. Phys. Chem. B, 2006, 110, 23870-23880. López-Quintela, MA. Curr. Opin. Colloid Interface Sci. 2003, 8, 137-144. Schwuger, J; Stickdorn, K. Chem. Rev., 1995, 95, 849-864. Eriksson, S; Nylén, U; Rojas, S; Boutonnet, M. Appl. Catal. A: Gen., 2004, 265, 207-219. Kishida, M; Umakoshi, K; Ishiyama, J; Nagata, H Wakabayashi, K. Catal. Today 1996, 29, 355-359. Capek, I. Adv. Colloid. Interf. Sci., 2004, 110, 49-74. Rojas, S; Eriksson, S; Boutonnet, M. Microemulsion: an alternative route to preparing supported catalysts, Catalysis, Spivey, J. J., Ed., Royal Society of Chemistry. Catalysis, Oxford, vol. 17, 2004. Herranz, T; Rojas, S; Ojeda, M; Pérez-Alonso, FJ; Terreros, P; Pirota, K; Fierro, JLG. Chem. Mater., 2006, 18, 2364-2375. Herranz, T; Rojas, S; Pérez-Alonso, FJ; Ojeda, M; Terreros, P; Fierro, JL. G. Appl. Catal. A: Gen., 2006, 311, 66-75.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Fischer-Tropsch Synthesis with Fe-Based Catalysts

225

[59] Jager, B; Espinoza, R. Catal. Today, 1995, 23, 17-28. [60] O'Brien, RJ; Xu, L; Bao, S; Raje, A; Davis, BH. Appl. Catal. A: Gen., 2000, 196, 173-178. [61] Pham, HN; Nowicki, L; Xu, J; Datye, AK; Bukur, DB; Bartholomew, CH. Ind. Eng. Chem. Res., 2003, 42, 4001-4008. [62] Chen, K; Yan, Q. Appl. Catal. A: Gen. 1997, 158, 215-223. [63] Leith, IR; Howden, MG. Appl. Catal. 1988, 37, 75-92. [64] Herranz, T; Rojas, S; Pérez-Alonso, FJ; Ojeda, M; Terreros, P; Fierro, JL. G. Appl. Catal. A: Gen., 2006, 308, 19-30. [65] Rockenberg, J; Scher, EC; Alivisatos, AP. J. Am. Chem. Soc., 1999, 121, 11595-11596. [66] Hyeon, T; Lee, SS; Park, J; Chung, Y; Na, HB. J. Am. Chem. Soc., 2001, 123, 12798-12801. [67] Sun, S; Zeng, H. J. Am. Chem. Soc., 2002, 124, 8204-8205. [68] Sun, S; Zeng, H; Robinson, DB; Raoux, S; Rice, PM; Wang, SX; Li, G. J. Am. Chem. Soc., 2004, 126, 273-279. [69] Kumar, RV; Koltypin, Y; Xu, XN; Yeshurun, Y; Gedanken, A; Felner, I. J. Appl. Phys., 2001, 89, 6324-6328. [70] Rice, GW; Fiato, RA; Soled, SL., US Patent, 4,659,681, 1987. [71] Wu, SH; Zheng, XC; Wang, SR; Han, DZ; Huang, WP; Zhang, SM. Catal. Lett., 2004, 97, 17-23. [72] Yu, L; Zhang, X; Du, Z; Wang, D; Wang, S; Wu, S. J. Natural Gas Chem., 2007, 16, 46-52. [73] Yubero, F; Gonzalez-Elipe, AR; Tougaard, S., Surf. Sci., 2000, 457, 24-36. [74] Yubero, F; Ocaña, M; Justo, A; Contreras, L; González-Elipe, A. R. J. Vacuum Sci. Technol. A: Vacuum, Surf. Films 2000, 18, 2244-2248. [75] Hansen, TW; Wagner, JB; Hansen, PL; Dahl, S; Topsoe, H; Jacobsen, CH. Science 2001, 294, 1508-1510. [76] Yang, Y; Xiang, HW; Xu, YY; Bai, L; Li, YW. Appl. Catal. A: Gen., 2004, 266, 181-194. [77] Arakawa, H; Bell, AT. Ind. Eng. Chem. Proc. Des. Dev., 1983, 22, 97-103. [78] Raje, AP; O'Brien, RJ; Davis, BH. J. Catal., 1998, 180, 36-43. [79] Lohitharn, N; Goodwin, JG. Jr. J. Catal., 2008, 257, 142-151. [80] Hutchings, GJ; Boeyens, JCA. J. Catal., 1986, 100, 507-511. [81] Bukur, DB; Koranne, M; Lang, X; Rao, KRP. M; Huffman, GP. Appl. Catal. A: Gen., 1995, 126, 85-113. [82] O'Brien, RJ; Xu, L; Spicer, RL; Davis, BH. Energy Fuels, 1996, 10, 921-926. [83] Bukur, DB; Lang, XS; Ding, YJ. Appl. Catal. A: Gen., 1999, 186, 255-275. [84] Kölbel, H; Ralek, M. Catal. Rev. Sci. Eng., 1980, 21, 225-274. [85] Hofer, LJE. Crystallite phases and their relationship to Fischer-Tropsch catalysis, Emmet, PH., Ed., 1956, p. 407. [86] Schulz, H; Schaub, G; Claeys, M; Riedel, T. Appl. Catal. A: Gen., 1999, 186, 215-227. [87] Shroff, MD; Kalakkad, DS; Coulter, KE; Kholer, S; Harrington, MS; Jackson, NB; Sault, AG; Abhaya, KD. J. Catal., 1995, 156, 185-207. [88] Blanchard, F; Reymond, JP; Pommier, B; Teichner, SJ. J. Molec. Catal., 1982, 17, 171-181. [89] Dictor, RA; Bell, AT. J. Catal., 1986, 97, 121-136.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

226

M. Ojeda, T. Herranz, F.J. Pérez-Alonso et al.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

[90] Kuivila, CS; Stair, PC; Butt, JB. J. Catal., 1989, 118, 299-311. [91] Amelse, JA; Butt, JB; Scwartz, LH. J. Phys. Chem., 1978, 82, 558-563. [92] Riedel, T; Schulz, H; Schaub, G; Jun, KW; Hwang, JS; Lee, KW. Topics Catal., 2003, 26, 41-54. [93] Eliason, SA; Bartholomew, CH. Stud. Surf. Sci. Catal., 1997, 111, 517-526. [94] Xu, J; Bartholomew, C. H. J. Phys. Chem. B 2005, 109, 2392. [95] Herranz, T; Rojas, S; Pérez-Alonso, FJ; Ojeda, M; Terreros, P; Fierro, JLG. J. Catal., 2006, 243, 199-211. [96] Shroff, MD; Datye, AK. Catal. Lett., 1996, 37, 101. [97] O'Brien, RJ; Xu, L; Milburn, DR; Li, YX; Klabunde, KJ; Davis, BH. Topics Catal., 1995, 2, 1-15. [98] Lohitharn, N; Goodwin, JG. Jr. J. Catal., 2008, 260, 7-16. [99] Malin, CB. Oil Gas J., 1998, 96, 33-35. [100] Eliasson, B; Riemer, P. Greenhouse gas control technologies, Elsevier Science Ltd., Oxford, 1999. [101] Hilaire, S; Wang, X; Luo, T; Gorte, RJ; Wagner, J. Appl. Catal. A: Gen., 2001, 215, 271-278. [102] Krishnamoorthy, S; Li, A; Iglesia, E. Catal. Lett., 2002, 80, 77-86. [103] Riedel, T; Schaub, G; Jun, KW; Lee, KW. Ind. Eng. Chem. Res., 2001, 40, 1355-1363. [104] Lee, JF; Chern, WS; Lee, MD. Can. J. Chem. Eng. 1992, 70, 511. [105] Lee, MD; Lee, JF; Chang, CS; Dong, TY. Appl. Catal., 1991, 72, 267-281. [106] Pérez-Alonso, F. J; Ojeda, M; Herranz, T; Rojas, S; González-Carballo, JM; Terreros, P; Fierro, JLG. Catal. Commun., 2008, 9, 1945-1948. [107] Olofsson, I; Nordin, A; Söderlind, U. http://biofuelregion.se/dokument/5_95.pdf, 2005. [108] Boerrigter, H; Calis, HP; Slort, DJ; Bodenstaff, H. http://ecn.nl/docs/library/report/ 2004/rx04041.pdf, 2004.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

In: Syngas Production Methods, Post Treatment… Editors: Adorjan Kurucz and Izsak Bencik

ISBN: 978-1-60741-841-2 © 2009 Nova Science Publishers, Inc.

Chapter 5

SYNGAS PRODUCTION IN MEMBRANE REACTORS Fausto Gallucci1* and Angelo Basile2 1

Fundamentals of Chemical Reaction Engineering, Faculty of Science and Technology, IMPACT, University of Twente, Enschede, The Netherlands 2 CNR-ITM, c/o University of Calabria, Via Pietro Bucci, Cubo 17C, 87030 Rende (CS), Italy

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

ABSTRACT Syngas is an important feedstock for the production of higher hydrocarbons or methanol. It can be produced via conversion of methane and the most extensively used process for this conversion is the methane steam reforming reaction carried out in large furnaces. The steam reforming reaction is a highly endothermic reaction which is industrially operated under severe conditions resulting in several undesirable consequences: sintering of the catalyst, very high carbon deposition and the use of high-temperature resisting materials. These drawbacks for methane steam reforming can be overcome by using membrane reactors, systems able to combine the separation properties of membrane with the typical characteristics of catalytic reactions. By using for example Pd-based membrane reactors, the hydrogen produced can be continuously withdrawn from the reaction system circumventing the thermodynamic limitations and making the methane steam reforming feasible at lower temperatures than the traditional systems. A potential alternative technique to steam reforming processes for producing syngas is the partial oxidation of methane with oxygen, having the disadvantage (economical and technological) that pure oxygen is required. Utilisation of air instead of pure oxygen is beneficial only if it can be performed by using a membrane reactor in which the membrane is perm-selective to oxygen. Another possible route for the partial oxidation of methane is the use of catalytic membrane reactors in which the membrane acts as both separation layer and reaction media. Finally, an interesting and promising route for syngas production is the methane reforming with CO2 (dry reforming) which is also a reaction system that can be used for

*

[email protected]

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

228

Fausto Gallucci and Angelo Basile the CO2 emissions mitigation. The methane dry reforming can be effectively performed in Pd-based membrane reactors as well as in catalytic membrane reactors. In this chapter, different examples of syngas production in both Pd-based membrane reactors and in catalytic membrane reactors will be presented and compared with traditional reaction systems.

SYNGAS PRODUCTION IN MEMBRANE REACTORS – A SHORT REVIEW

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

An attractive route for the utilization of the large reserves of natural gas, which consists of a gas mixture containing mainly methane, is their conversion to synthesis gas, a mixture of H2 and CO, from which a wide variety of valuable hydrocarbons and oxygenates can be synthesized. In fact, synthesis gas (or syngas) is an important starting material for several important industrial processes and laboratory researches, such as methanol and higher alcohols synthesis [1,2], Fischer–Tropsch process [3], ammonia and aldehydes [4,5], acetylene [6], dimethyl ether [7], aromatic and aliphatic hydrocarbons [8], and so on. In Figure 1 a scheme for the synthesis of different products from natural gas via syngas is reported. Syngas can be converted into liquid fuels such as gasoline and methanol, which are more convenient for transportation and storage than natural gas. For example, the production of synthetic liquid fuels using syngas in a hydrogasification reactor directly coupled to a reforming process have been successfully demonstrated by Norbeck et al. [9] using several carbonaceous feed stocks.

Figure 1. Scheme for synthesis of different products from natural gas via syngas.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Syngas Production in Membrane Reactors

229

The molar ratio of hydrogen to carbon monoxide (the so-called syngas ratio: H2/CO) may vary depending on the desired product and fuel processing technology. Depending on the input gas composition, there are several commercial technologies available for syngas production: each of them results in a specific syngas ratio. For example, in the case of the production of Fischer–Tropsch diesel, fuel requires a syngas ratio of about 2 (from less than one to over two depending on the catalyst and specific technology used). Vice versa, oxosynthesis and other processes, such as dimethyl ether synthesis, generally require a syngas ratio = 1. Up to now, steam reforming of methane is the dominant process for producing syngas from natural gas. Alternatively, coal gasification may be also used as a source of syngas production. However, this latter process, already commercialized in South Africa, is both more complex and expansive than the one that uses natural gas. Starting from methane (or natural gas), the four possible reactions concerning the production of syngas from methane and their reaction heats are shown below: Steam reforming: methane + steam → syngas Dry reforming: methane + carbon dioxide → syngas Partial oxidation: methane + oxygen → syngas Autothermal reforming: methane + steam + oxygen → syngas

ΔH°298 K = + 206.1 kJ/mol ΔH°298 K = + 246.9 kJ/mol ΔH°298 K = - 36 kJ/mol ΔH°298 K = + 170.5 kJ/mol

The first and the second are endothermic reactions (i.e. they need heat), while the third one is exothermic and methane conversion is always close to thermodynamic limitations. Generally, the syngas composition from syngas production technologies is not suited for direct use in the downstream fuel process. In fact:

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

steam reforming of methane produces a syngas ratio of more than 3; dry reforming of methane produces a syngas ratio of 1; partial oxidation gasification produces a syngas ratio of 1; autothermal reforming produces a syngas ratio of 2. So, downstream shift reactors, membrane separators or pressure swing adsorption are employed to meet the syngas ratio requirement. Cost and complexity to the overall process are added, of course. Over the past several years, extensive efforts have focused on both approaches to valueadded products, particularly easily transportable fuels [10, 11]. In the direct reaction scheme, methane is directly oxidized via partial oxidation of methane to methanol, formaldehyde, or olefins. It is the more difficult approach because the desired products of reaction are more reactive than the starting methane reactant, which then leads to deep oxidation and high selectivity of CO2 and H2O [12]. Indirect approaches for methane conversion require oxidation of methane to syngas in a first stage. As already said, the syngas produced is afterwards converted into upgraded

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Fausto Gallucci and Angelo Basile

230

products in a second stage by Fischer-Tropsch technology [13] or methanol synthesis [14, 15] or others. At present, steam reforming is considered for the conversion of methane to syngas, whereas partial oxidation of methane into syngas is presently considered as a potential alternative to the steam reforming process. The catalysts mainly used are Ru/TiO2 [16,17] as well as Rh- or Pt-loaded monolith catalysts at short residence times [18,19]. Being the steam reforming a highly endothermic reaction, the indirect approaches are usually very energy- and capital-intensive and operate at high pressures and temperatures. Generally, the cost of syngas production by steam reforming accounts for at least 70% of the integrated cost of the overall plant. For example, a potential alternative to steam-reforming processes is the partial oxidation of methane with air as oxygen source. Nevertheless, downstream processing requirements cannot tolerate nitrogen (recycling with cryogenic separations is required) and therefore pure oxygen is required. Thus, the most significant cost associated with conventional reactors where the partial oxidation of methane to syngas is carried out, is that of the oxygen separation plant. In the following, the four possible reactions concerning the production of syngas from methane are considered in details for both conventional and membrane reactors.

Steam Reforming of Methane Conventional technology for syngas production is catalytic steam reforming of methane. However, the process has different disadvantages such as high endothermicity (energyintensive), catalyst-coking and requirements of high temperature (> 700 °C) and pressure (>20 atm). Moreover, the syngas ratio in the product stream is not optimal for most applications. On the whole, steam reforming is highly capital intensive accounting for about 70% of total investment and operating costs in methanol production based on syngas obtained from this reaction [20].

a) Conventional reactors

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

The most important commercial route of syngas manufacturing is the catalytic steam reforming of methane. It is a well-known commercially available process for both syngas and hydrogen production [21]. The methane steam reforming reaction CH4 + H2O ↔ CO + 3 H2

ΔH°298 K = + 206.1 kJ/mol

is endothermic and requires external heat input, generally provided by the combustion of a fraction of the feedstock (up to 25%) or from burning waste gases. Economics favor reactor operation at high pressures (> 20 atm) and temperatures (700 850 °C). Methane and steam react in catalyst filled tubes: the mass ratio of steam to carbon is about 3 or more to avoid "coking" or carbon build-up on the catalysts. On the other hand, most of the research efforts on the steam reforming reaction are related to the pure hydrogen production, thus also CO needs to be further converted. In this respect, after the reforming process, the produced syngas is fed to the water-gas shift reactors, where the hydrogen output is increased:

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Syngas Production in Membrane Reactors CO + H2O

CO2 + H2

231

ΔH°298 K = - 41.15 kJ/mol

Depending on the catalyst used, this reaction is favored in the temperatures range of about 200 – 600 °C, and moreover, depending of the catalyst used, it can take place as low as 200 °C. The gas exiting the shift reactors contains mostly H2 (70 - 80%), CO2, CH4, H2O and small quantities of CO. In case of pure hydrogen production, the shift reaction is often accomplished in two stages. A high temperature shift reactor (working at 350-475 °C) accomplishes much of the conversion, followed by a lower temperature (200-250°C) one. At the end, the CO concentration is reduced down to a few percent (by volume) or less. Compact steam methane reformers able to reform natural gas for closely coupled fuel cells have developed by various companies, such as Haldor-Topsoe, International Fuel Cells, Ballard Power Systems, Sanyo Electric, and Osaka Gas Company. Generally, these technologies are newly commercialized and show both the possibility of reducing capital costs, as compared to conventional small-scale reformers, and compactness. To give only some examples: Based on this type of reformer, Praxair commercialized a small stand-alone hydrogen production system [22]. Researchers at the Fraunhofer Institute for Solar Energy Systems both designed a more compact multi-tube steam methane reformer with a catalytic heater instead of a burner and built methane reformers to make H2 for use with Energy Partners vehicle [23]. Dais-Analytic Corporation is building a residential PEMFC power system with its own reformer (Dais-Analytic Corporation website) [24]. Sanyo Electric Co. is building a residential PEMFC power system with a multi-tube steam reformer [24]. IdaTech is building a residential PEMFC power system with a multi-fuel reformer able to produce 99.9% pure hydrogen [25]. IFC is in a joint venture with Toshiba is using this type of reformer for the development of stationary PEMFCs [24].

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

b) Membrane reactors In the last decades, there was an increasing interest in industrial R&D activity on membrane technologies for both syngas and pure hydrogen production. Interest by major energy companies in applying membrane technology to large-scale syngas and hydrogen production may have significant spin-offs. In fact, the number of patents related to membrane technologies used in the steam reforming reaction to produce pure hydrogen or syngas is dramatically increasing every year. Nevertheless, apart from the case of the membrane reformer commercialized by the Tokyo Gas Company, membrane reactor steam reformers are still undergoing laboratory R&D as well [26-28]. This highly promising technology is represented by the Membrane Reactor (MR) concept, where steam reforming, water gas shift and gas selective permeation steps all take place in a single equipment. In particular, in a dense inorganic MR, pure hydrogen is produced; whereas, in a porous inorganic MR syngas is produced. In the first case (dense membrane), on one side of the MR there is a high selectivity inorganic (generally Pd-based) membrane that is selectively permeable only to hydrogen. As the steam reforming reaction

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

232

Fausto Gallucci and Angelo Basile

proceeds, the hydrogen selectively permeates through the membrane due to its driving force (i.e. the hydrogen square root pressure difference between the two sides of the membrane). Depending on some parameters, such as temperature, pressure and reactor length, methane is completely converted, producing highly pure hydrogen, which is removed from the reaction system by the selective membrane as the reaction proceeds. With respect to a conventional reactor, this allows lower temperature operation, and consequently lower cost materials. In fact, a MR represents a considerable system simplification of the process design and capital cost reduction, because fewer process vessels will be needed. Deeper considerations on such aspects could be found on various good reviews, see for example on ref. [29-33]. Small-scale membrane reformers for producing pure hydrogen are commercialized by Johnson Matthey and Tokyo Gas Co. In particular, Johnson Matthey was working on the development of hydrogen separation membranes to be used in membrane reactors [34]; whereas in 2004, Tokyo Gas Co. developed a prototype system of a Membrane Reformer which utilizes Pd-based membranes able to produce 40 Nm3·h-1 of 99.999% pure H2 [35, 36]. The system is still under investigation with respect the determination of both the effect of co-existing gases on H2 permeability at high temperatures [37], and the surface reaction of H2 under co-existence of other gases (such as H2O, CO, CO2 or CH4) [38]. To the best of our knowledge, in the specialised literature only one paper related to the syngas production via steam reforming of methane in a membrane reactor has been published. It is reported in the following discussion. Rather than the equilibrium-limited reaction in the conventional reactors, the methane steam reforming reaction carried out in a membrane reactor is a transfer-limited reaction mainly related to membrane porosity and gas diffusivity. This technology, permitting the control of the syngas ratio, seems to be more suitable to be used in gas-to-liquid processes for production of syngas. The sensitivity analysis recently performed by Fernandes and Soares [39] by simulating a membrane reactor model, indicates that the thickness of the membrane plays a major role both in methane conversion and syngas ratio control. In particular, working at moderate temperatures, membranes having a thickness less than 1 micron increases methane conversion by 22.7% (with respect to the traditional one), also reducing of the syngas ratio by 30.4%, A 0.5 micron thick membrane produces a methane conversion of 96% at 850 K, that represents an increase of 69.6% over conventional reactors (able to achieve this value only at T > 1000 K but with a higher syngas ratio). As reported by these authors, the syngas ratio control is very important for gas-to-liquid processes where the optimum ratio varies from 0.7 up to 3.0, and influences hydrocarbon product distribution. Concerning the membrane thickness, its decreases generally favours the syngas ratio application in gas-to-liquid processes. Another important aspect concerns the interaction between the catalytic performance with respect the ones of the membrane, in order both to improve the separation efficiency of membrane and/or the activity of the catalyst. Thus, renewed interest in developing catalysts suitable for the methane steam reforming reaction at low temperatures is recently gaining particular attention. In other words, catalysts suitable for the methane steam reforming reaction in a membrane reactor, especially having good activity at low temperatures, are under study. For instance, Kusakabe et al. [41] recently developed catalysts suitable for carrying out the steam reforming of methane reaction in a membrane reactor over Ce–ZrO2supported Ni, Pt, Ru and Rh catalysts at 500–800 °C. In particular, from their work, the following conclusions can be drawn:

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Syngas Production in Membrane Reactors

233

Ce0.15Zr0.85O2 with 10 wt.% Ni loading support shows the highest CH4 conversion at 500 –600 °C and a high syngas ratio; moreover, the catalytic activity of Ce1_xZrxO2 supports was improved by the loading other metals (Pt, Ru, Rh). Rh/Ce0.15Zr0.85O2 catalyst shows the highest activity observed: the methane conversion is 28.1% at 500 °C. The work reported by Chen et al. [40] investigates the catalytic performances of a nickelbased catalyst pre-reduced at high temperature in membrane reactor for the methane steam reforming reaction. These authors reported an increase of methane conversion in membrane reactor of 260–400% compared with equilibrium values. For instance, at 900 kPa and 723 K, methane conversion is 65.0% at 723 K, versus the equilibrium value of only 13.2%. Thus, “in contrast with most previous investigations” [as reported by the same authors], their nickel-based catalyst prepared by coprecipitation–deposition and pre-reduced at higher temperature is able to show high activity for this reaction at low reaction temperatures.

Dry Reforming of Methane Being performed at high temperature and high pressure, the methane steam reforming is a costly chemical process. Moreover, syngas ratio is around 3-4, and thus not suitable, for example, for synthesizing hydrocarbons or methanol by Fischer–Tropsch reaction (where a syngas ratio of 2 is more suitable). Therefore, in case the concentration of hydrogen exceeds the ideal one for synthesizing a useful compound, other methane reforming reactions are considered to me more useful. Another possible reaction for syngas production is catalytic CO2 reforming of methane. It is also a highly endothermic reaction requiring large energy input. Despite this drawback, a resurgence of interest in CO2 reforming is observed mainly due to the realization of a possible positive impact of the large-scale application of the process on global CO2 emissions [42].

a) Conventional reactors

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

During the past two decades, another interesting approach for the syngas production has attracted much attention from both industrial and environmental sectors. It is the catalytic reforming of methane using carbon dioxide, known as dry reforming, an endothermic reaction able to give a 1:1 mixture of H2 and CO: CH4 + CO2 = 2CO + 2H2

ΔH°298 K = + 246.9 kJ/mol

Although this reaction is more endothermic than the methane steam reforming, an industrial application of the methane dry reforming is of practical interest because it could reduce the amount of undesirable greenhouse gases (both gases are cheap and abundant) present in the atmosphere as well as could produce value-added products. In fact, the disadvantage due to the reaction cost (in terms of both energy input and the cost of materials) is balanced by the benefit deriving from the CO2 consumption. Concerning the reaction mechanism, as reported by Hou and Hughes [43], during the dry reforming reaction, the CO2 is shifted to CO and H2O through the reverse water–gas shift reaction and then H2O is reacted with CH4 to produce syngas.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

234

Fausto Gallucci and Angelo Basile

For this reaction, a variety of catalysts have been developed: highly promising are considered catalysts of supported noble metals, such as Rh [44, 45], Ru [46, 47], Pd [48], although the high costs of the precious metals could be a limiting factor for their application. From the commercial standpoint, a nickel-based catalyst is more profitable. Nevertheless, the main problem with the nickel-based catalysts is its deactivation due to coking, a direct consequence of nickel sintering and carbon deposition. Carbon deposition can lead to cover active sites of the catalyst with a consequent progressive decrease of its performance during the reaction. Hence, in recent years, various studies on resistance to carbon deposition on the catalyst surface have been reported (see, for example, ref. [49]). Recently, Choudhary et al. [50] reported an interesting study on nickel-based catalysts that with perovskite structure showed a good resistance to coking. Moreover, Bolt et al. [51] observed that the extent of nickel sintering was reduced over the NiAl2O4 catalyst with discontinuous interfacial layers. Recent advances in the development of carbide-based non-coking catalysts for methane reforming solidified the industrial feasibility of this process [52].

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

b) Membrane reactors Another important limitation for the methane dry reforming is related to its thermodynamic, which represents a rigid and unbridgeable limit for the conversion of reactants in case of traditional reactors are used. On the other hand, an increase of both methane and carbon dioxide conversion can be obtained if one or both the products of the reaction are selectively removed from the reaction side. This fact was demonstrated, for example, by Paturzo et al. [53], which carried out this reaction in both a TR and a MR. The membrane consists of a commercial ceramic tube in which an internal two-layers Ru nanoparticles was deposited. Probably, the most important problem to be solved with this reaction when carried out in a membrane reactor is the carbon deposition. Being the ruthenium exhibiting the highest catalytic activity towards the CO2 reforming reaction [54], Paturzo et al. [53] used a Ru-based MR and suggested a possible way to realize a catalytic membrane able to depress the carbon formation as well as to remove hydrogen and improve conversion. Using a dense Pd-based MR for the production of syngas from methane, Galuszka et al. [55] observed the destruction of the membrane due to the generation of filamentous carbon. A different membrane was used by Liu and Au [56]. Using zeolite membranes to carry out the dry reforming of methane to syngas, these authors obtained ideal separation factors for H2/CH4 and CO/CO2 higher than the corresponding Knudsen diffusion values. They also observed CH4 conversion values of 21.4 mol% higher than the ones obtained in a traditional fixed bed reactor. Their main result was the demonstration that the use of a zeolite MR, instead of a dense Pd-based one, is significantly advantageous for producing syngas via CO2 reforming of methane. The same authors, is a successive work, prepared a NaA (thickness of about 20 micron, defect-free) and La2NiO4/NaA composite membranes by means of in situ hydrothermal synthesis used for CH4/CO2 reforming to syngas [57]. The conversions of CH4 and CO2, selectivities of CO and H2 obtained using these composite membranes were remarkably significantly higher than the ones obtained using a fixed-bed reactor. The NaA zeolite layer was able to promote the dispersion of nickel particles (Ni/La = 0.33) and reduced the amount of carbon deposited on the surface of the catalytic membrane.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Syngas Production in Membrane Reactors

235

Recently, Gallucci et al. [58] showed the application of both porous and dense membrane reactors for the dry reforming of methane and the results are shown in the case study below. It was found that using a dense Pd-based membrane reactor, the carbon deposition on the catalyst is drastically reduced. A similar situation was found in the steam reforming of methane in a Pd-based MR, where up to 100% CH4 conversion is obtained at 500 °C with no carbon deposition [59].

Mixed Reforming Reaction As already reported, the major disadvantage of the methane dry reforming is the tendency for carbon deposition leading to catalyst deactivation. However, this problem can be mitigated by the addition of steam to the process [60-62]. The simultaneous steam and carbon dioxide reforming of methane is known as the mixed reforming reaction and allows some control of the syngas ratio while reducing carbon deposition. Interesting researches using conventional reactors have been performed in this direction by Li et al. [60], Abashar [63] and Froment et al. [64, 65]. These researches can be summarized by showing that steam reforming of methane in the presence of carbon dioxide is able to: enhance the conversion of methane; influence the syngas ratio in order to obtain a final syngas ratio in the range 1-2; avoid carbon deposition. Other parameters such as temperature, the steam to methane ratio and the type of catalyst can also be controlled in order to influence the syngas ratio [66-68].

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Partial Oxidation of Methane As already said, the commercially used steam reforming process needs to operate at high temperatures and pressures with a consequent very high increase of the cost of syngas production. At the moment, catalytic partial oxidation of methane is receiving an increasing attention. This reaction is predominantly considered as an alternative method for the production of syngas [69]. Moreover, it is also used on a more fundamental level for direct methane coupling using dense ion-conducting mixed-oxide membranes [70].

a) Conventional reactors This reaction can greatly speed up the production of syngas since it is operated at very high space velocities of (1.0–5.0) × 105 h-1. The main reactions involving the partial oxidation of methane (or some other hydrocarbon feedstock such as oil) are: total combustion of methane:

CH4 + 2O2 = CO2 + 2H2O

ΔH°298 K = - 801 kJ/mol

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

236

Fausto Gallucci and Angelo Basile dry reforming of methane:

ΔH°298 K = + 246.9 kJ/mol

CH4 + CO2 = 2CO + 2H2

steam reforming reaction:

CH4 + H2O = CO + 3H2

ΔH°298 K = + 206 kJ/mol

Methane combustion is a highly exothermic reaction, while the reforming reactions are strongly endothermic. For example, the exothermicity of the first reaction could create severe problems (heat management, safety and stability). Moreover, in case of modeling works, it must be taken into account also for these other reactions: steam reforming reaction: water gas shift reaction: Boudouard reaction: methane cracking: carbon gasification by steam: carbon gasification by O2:

CH4 + 2H2O = CO2 + 4H2 CO + H2O = CO2 + H2 2CO = C + CO2 CH4 = C + 2H2 C + H2O = CO + H2 C + O2 = CO2

According to the direct reaction scheme, syngas can be also formed through the following direct oxidation of methane reaction:

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

CH4 + 1/2O2 = CO + 2H2

ΔH°298 K = - 36 kJ/mol

This reaction has the advantage to be only slightly exothermic, reducing or eliminating the above mentioned problems. Indirect heat exchanger is not needed. Because of the high temperature, catalysts are not required. However, the hydrogen yield per mole of methane input (and the system efficiency) can be significantly enhanced by use of catalysts [24]. Using many active catalysts, high CH4 conversion, high CO selectivity and proper ratio of H2 to CO have been obtained [71-74]. Large-scale partial oxidation systems have already been used commercially to produce hydrogen from hydrocarbons such as residual oil, for applications such as refineries [24]. These systems usually incorporate a pure oxygen production plant. Of course, the operation carried out with pure oxygen, rather than air, reduces the size and cost of the reformers. Concerning small-scale partial oxidation reformers, they able to handle a variety of fuels, including methane, ethanol, methanol, and gasoline, and are also commercially available, but still are undergoing intensive R&D [24, 74, 75]. Generally, the partial oxidation reactor is more compact than a steam reformer, especially because heat must not be added via an external heat exchanger. It is interesting to observe that the efficiency of the partial oxidation unit is relatively high (70-80%). However, due to both higher operating temperatures (which increases heat losses) and the problem of heat recovery, these systems are typically less energy efficient than the steam reformers. On the other hand, being more compact, and without indirect heat exchange, partial oxidation systems are as expansive as steam reformer systems. In particular, while the partial oxidation reactor could be less expensive than a steam reformer one, the downstream shift and purification stages seems to be more expensive [24].

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Syngas Production in Membrane Reactors

237

In fact, the use of pure oxygen for feeding the partial oxidation reaction needs high capital costs for small-scale oxygen production, but eliminates the need to deal with nitrogen downstream. Oxygen enrichment of incoming air is another way of reducing, but not eliminating, the amount of nitrogen. Innovative ion transport membrane technology is under investigation by Air Products and may allow lower cost oxygen for partial oxidation reactors [76, 77], and by Praxair and partners [78]. Carrying out this reaction system for producing syngas, there are at least two important problems to be considered: The first one is related to the hot-spot problem, which might result in reactor runway. In order to efficiently remove the additional heat, it must be carefully considered in the reactor design. For instance, in the cyclic reactor proposed by Blanks et al. [79], the flow direction was periodically reversed for heat exchange, whereas Piga et al. [80] designed an advanced type of reactor configuration, a heat-integrated wall reactor, able to reduce the magnitude of hot spots by controlling the temperature in the combustion zone. The second problem is that air is usually used as oxidant. In fact, the downstream upgrading process requirements cannot tolerate nitrogen, i.e. the syngas is needed to be free from both N2 and the by-production of NOx. In other words, pure oxygen is required for the development of this reaction. One solution consists in separating air (for example using membranes or conventional cryogenic oxygen plants) into its main components before being fed to the reactor. Upstream oxygen separation from air is more favorable than the costly downstream purification, but the conventional cryogenic oxygen plant is very costly in operation combined with the partial oxidation process. To avoid this problem, novel catalytic membrane reactors are today under study. Large research efforts are recently being focused on this reaction [72, 81], especially for what concerns the different kinds of catalysts investigated [71, 72].

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

b) Membrane reactors As already reported, an important limitation in realizing this reaction commercially viable for traditional reactors is Thermodynamics. In particular, the pressure increase gives a decrease in equilibrium methane conversions [82]. Vice versa, a membrane reactor permits to overcome the rigid thermodynamic barrier of traditional reactors, giving the possibility to obtain a high methane conversion at low temperature. In literature, this reaction was carried out in both composite and Pd-based membrane reactors from both experimental and simulation points of view. Some interesting results [83] in terms of methane conversion, compared with both the ones obtained using membrane and traditional reactors as well as the thermodynamic equilibrium values, are summarised in the following: a.

at a fixed temperature methane conversion is remarkably higher in membrane reactors, b. Pd-based MR shows the highest methane conversion; c. methane conversion exceeds equilibrium values. Syngas can be produced from methane also using oxygen permeable membrane reactors. In particular, oxygen can be provided by transport through a mixed conducting membrane

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Fausto Gallucci and Angelo Basile

238

[82-92], or through an ionic conductor with electronic short-circuiting or assisted by electrochemical pumping [93]. Mixed conducting membrane technology having also a potential for energy and cost and NOx emissions reduction are still under study. Dense oxygen ceramic membranes, used in a membrane reactor for selectively separating pure oxygen from air at high temperatures, work as follows. Air is fed in one side of the membrane reactor, which consists of a mechanical (generally macroporous) support and a dense oxide layer through which oxygen selectively permeates, whereas methane is sent on the other side. The selective permeation through the oxide lattice provides a controlled amount of oxygen to the side where the catalyst is located. The catalyst could be either the oxide layer or another material previously deposited on the top layer of the membrane. Methane is converted by directly using the oxygen permeated through the membrane: oxygen reacts with methane forming a partly oxidized compound and undesirable carbon oxides. If the oxygen is supplied by properly balancing its rate of consumption then the process enables a sustained oxidation of methane without mixing the methane itself and air. The process does not require any expensive cryogenic oxygen production apparatus and the reaction heat of the partial oxidation of methane is sufficient for the membrane reactor process to be selfstanding. To be considered that under suitable operating conditions one this reaction takes place: •

CH4 + 0.5 O2 = CO + 2H2

Somewhat excessive oxygen partial pressures are necessary to attain high conversion of methane, with the risk of further oxidation of the components of the syngas mixture to water vapor and/or carbon dioxide: • •

H2 + 0.5 O2 = H2O CO + 0.5 O2 = CO2

The effects of methane cracking on metallic surfaces (e.g. Ni or Fe) should also be consider:

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.



CH4 → C + 2H2

In this way, the main complication is carbon deposition, which is able to block the methane conversion systems, and indeed decreasing the efficiency of the conversion. Methane cracking may be reduced by avoiding the use of metallic surfaces at high temperatures, i.e. by using all ceramic systems, such as mixed conducting ceramic membranes, and by ensuring conditions when partial oxidation and steam reforming of methane are more favorable. Ceramic mixed conducting membranes are expected to be more promising than ionic conductors with metallic electrodes for electrochemical pumping or short-circuited. The combination of the mixed conducting membranes with partial oxidation of methane reaction makes the separation of oxygen and catalytic oxidation in only one process that simplifies the methane – syngas conversion operation process and reduces the production cost about 20–30% [94, 95].

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Syngas Production in Membrane Reactors

239

Recently, extensive efforts had been addressed on the development of high oxygen permeability membranes. Among them mostly are perovskite-like ceramics, such as Ba0.5Sr0.5Co0.8Fe0.2O3-δ and La0.2Sr0.8Co0.8Fe0.2O3- δ [96-98]. However, it must be said that there is still a disparity between the performance of these ceramic membranes and the application requirements. Therefore, it is necessary to investigate new kinds of selective oxygen permeation materials to be used under the partial oxidation reaction conditions. Balachandran et al. [99], for example, are developing dense membranes for separating hydrogen from product streams that are generated during methane partial oxidation, water-gas shift reactions, and coal gasification mixed gases at commercially significant fluxes. Being their membranes able to separate pure hydrogen operating nongalvanically (i.e. without using electrodes or an external power supply), materials exhibiting suitable electronic and protonic conductivities as well as high hydrogen permeability are necessary. Their early study was focused on the mixed proton/electron conductor BaCe0.8Y0.2O3-δ, obtaining an electronic conductivity insufficient to support a high nongalvanic hydrogen flux [100-101]. Later, for increasing both the electronic conductivity and the hydrogen flux, they developed various ceramic-metal composite membranes, in which a suitable metal powder was dispersed in a ceramic matrix [102]. In these membranes, the metal is able to enhance the hydrogen permeability of the ceramic phase by increasing the electronic conductivity of the composite one. Recently, their results suggest the possibility of using two types of dense ceramic membranes to convert methane into pure hydrogen [103]. With this approach the need for the energy- and capital- intensive steam reforming or cryogenic oxygen plant is eliminated. So far it has been demonstrated the capability of oxygen permeable and hydrogen permeable membranes to perform the individual steps of producing hydrogen from methane. Ion transport membrane technology for oxygen separation and syngas production has been recently reported by Dyer et al. [103]. This oxygen process, in which membranes are used to separate high-purity oxygen from air, has interesting advantages when integrated with power generation cycles. In fact, by combining air separation and high-temperature syngas generation processes into a single compact ceramic membrane reactor, this syngas process make possible reducing the capital investment for gas-to-liquid plants and for distributed hydrogen. Air Products and other partners (U.S. Department of Energy, Ceramatec, etc.) developed, scaled-up and commercialized this technology. Details regarding the different stages of development the development, the industrial applications, and the technical hurdles to be overcome before successful commercialization are well described by Dyer et al. [103]. An experimental study of multilayered composite palladium membrane reactors for partial oxidation of methane to syngas has been performed by Basile and Paturzo [84]. They carried out the partial oxidation reaction of methane to syngas via partial oxidation reaction in four different membrane reactors as well as in a traditional reactor. Their main conclusions are the following: • • •

conversion of methane is higher in membrane reactors than in the traditional reactor; applying membrane reactors, conversions exceeding equilibrium are achieved; according to literature data, no carbon was found on catalyst under the conditions investigated.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Fausto Gallucci and Angelo Basile

240

In other two experimental works of Basile et al. [104, 105] a membrane reactor using composite Pd-based membranes operating in a Knudsen regime and a traditional reactor were studied the effect of reaction temperature on methane and oxygen conversion. They obtained a methane conversion of 93% at 500°C working at the time factor of 0.48 gcatmin/cm3 with a feed composition CH4/O2/gas-carrier of 2/1/14 and using a composite Pd membrane reactor prepared by an electroless plating technique. In all the analysed cases, membrane reactors allowed higher conversions of methane with respect to traditional reactors. For the same reaction, complete conversion and depression of coke deposition have been obtained by Kikuchi and Chen [106]. On the same reaction in a Pd membrane reactor, later Basile et al. [107], reported some experimental studies showing that it is possible to increase methane conversion values above the traditional equilibrium ones. Moreover, in order to better interpret their experimental results, a mathematical model for both dynamic simulation and kinetic simulation was also developed. .

Autothermal Reforming a) Conventional reactors

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

In contrast to steam reforming of methane (a strongly endothermic reaction which requires that heat must be supplied from an external source) and partial oxidation of methane (a weakly exothermic reaction), autothermal reforming is able to combine some of the best features of steam reforming and partial oxidation systems. It is a thermally neutral reaction and uses air or pure oxygen as oxidant. When air is used, the syngas produced contains about 50% N2, which requires a substantial amount of heat [108]. On the other hand, the ideal oxidant pure oxygen is highly expensive. This explains why in the specialized literature an increasing interest in developing oxygen permeable dense membranes able to produce pure oxygen from air was recently observed. Several companies are developing small autothermal reformers for converting liquid hydrocarbon fuels to hydrogen in fuel cell systems [24]. In an autothermal reaction system, a hydrocarbon (methane or a liquid fuel) feed reacts with both steam and air to produce syngas via both the steam reforming and partial oxidation reactions. For example, with methane the following reactions take place: • •

CO + 3 H2 CH4 + H2O CH4 + 1/2 O2 CO + 2 H2

ΔH°298 K = +206.1 kJ/mol ΔH°298 K = - 36 kJ/mol

By correctly dosing the mixture of input (fuel, air and steam), the reaction is able to supply all the heat needed to drive the catalytic steam reforming reaction. In other words, the autothermal reformer do not require external heat source and also indirect heat exchangers are not needed. As a consequence, these reformers are simpler and more compact than steam reformers, and with a lower capital cost. In fact, in the autothermal reactor all the heat generated by the partial oxidation reaction is utilized to carry out the steam reforming

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Syngas Production in Membrane Reactors

241

reaction. Thus, an autothermal reactor offers higher system efficiency than partial oxidation systems, where excess heat must recovered using heat exchangers. Recently, the thermodynamic performance of both autothermal and CO2 reforming of methane have been studied by Gibbs free minimization by Li et al. [108]. They found the optimal conditions of both reaction systems and also the ways for avoiding the coke deposition. In particular, their results show that coke reduction (or elimination) is obtained by: a) an increase of the reaction temperature in CO2 reforming; and b) an increase of the steam in the feed of the steam reforming reaction. Moreover, for the oxidative steam reforming, the following ratios O2/CH4 > 0.4 or H2O/CH4 > 1.2, in both cases at T > 700 °C, give the best results in terms of decreasing the coke deposition.

b) Membrane reactors As above reported, a recent development in syngas production technology is the use of oxygen-permeable dense ceramic membranes integrating the oxygen separation and partial oxidation of methane processes in a single space. For example, using this technology, the capital cost could be reduced by 30%. An example of Ba0.5Sr0.5Co0.8Fe0.2O3 dense oxygen permeation membrane having high oxygen permeability (11.5 ml/cm2 min) at reaction conditions was developed by Xiong et al. [89, 109]. They also developed an excellent LiLaNiO/γ-Al2O3 catalyst for both partial oxidation of hydrocarbons to syngas and steam reforming and autothermal reforming [110-112]. Unfortunately, no papers on the autothermal reforming for syngas production carried out in membrane reactors are published in the specialised literature.

SYNGAS PRODUCTION IN MEMBRANE REACTORS – CASE STUDY

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Case Study: Dry Reforming of Methane in a Membrane Reactor The problems related to CO2 greenhouse effects pushed in last years the scientific community towards a useful route the CO2 consumption in order to decrease its emission into the atmosphere. An interesting reaction is the conversion of CO2 into syngas. With this aim, the most promising reaction is the dry reforming of methane, an endothermic reaction producing a gaseous mixture with a low H2/CO molar ratio. As already said before, different kinds of catalyst such as Ni, Ru, Rh, Pd, and Pt-based ones were proposed for this reaction system [44-48]. Methane dry reforming is affected by thermodynamic constraints, which limit reactants conversion. In fact, the main reactions involved in this process are as follows: CH4 + CO2 ↔ 2CO + 2H2 CO2 + H2 ↔ CO + H2O (reverse water gas shift), CO2 + 4H2 ↔ CH4 + 2H2O (methanation),

ΔH298K = +247.4 kJ/mol ΔH298K = +41.1 kJ/mol ΔH298K = −164.6 kJ/mol

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

(1) (2) (3)

Fausto Gallucci and Angelo Basile

242

However, other two undesired reactions also occur, producing carbon deposition: CH4 → 2H2 + C, 2CO ↔ CO2 + C,

ΔH298K = +75.0 kJ/mol ΔH298K = −173.0 kJ/mol

(4) (5)

Following the Le Chatelier principle, a membrane reactor, allowing a product to permeate out from the reaction system, is able to shift the equilibrium towards products: in this way, it is possible to achieve either higher conversions than a traditional process at a fixed temperature, or the same conversion but at lower temperature. Applying this kind of membrane reactor for the dry reforming of methane, a higher conversion of CO2 and CH4 towards syngas is obtained. On the other hand, pure hydrogen is also recovered via the dense membrane and it is available for adjusting the syngas ratio as requested for further processing. In this section, the use of dense self supported Pd-Ag membrane reactor will be used for carrying out the dry reforming reaction and a combination of dry reforming and steam reforming reaction.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Experimental plant The scheme of the experimental plant (also used for all the other reaction systems, with small changes, especially in the membrane type and in the feed section) is shown in Figure 2. Reactants (CH4 and CO2/O2) and an inert gas (N2) are fed by means of mass-flow controllers furnished by Bronkhorst. H2O is fed by means of a HPLC pump. Nitrogen has been used as internal standard for the Gas Chromatograph (GC) analysis and as sweep gas for the dense Pd–Ag MR. It was also used as carrier when the reactor temperature was changed. The composition of outlet is detected by a GC, type HP 6890 with a TCD (Thermal Conductivity Detector) at 250 °C and Ar as carrier gas. The GC is equipped by two packed columns: Carboxen 1000 (15 ft×1/8 in.) and Molecular Sieve 5°A (6 ft×1/8 in.); a 10-way valve was used to optimize the total time of the analysis, which was about 13 min. This apparatus is driven by software furnished by Hewlett-Packard. Regarding the MR, since it has two outlet streams (permeate and retentate), two TCD detectors were simultaneously used for measuring their composition at the same time. The internal standard procedure for the GC was used. Pressures are measured by means of pressure transducers. The pressure was regulated by means of back-pressure regulators. The reactor (TR or MR) is placed in a temperature controlled P.I.D. (Proportional + Integral + Derivative Control) oven. A four-point thermocouple is placed into the right side of the reactor, to monitor the temperature profile inside both TR and MR. The same experimental apparatus has also been used for permeation tests of the Pd–Ag membranes. The TR consists of a stainless steel tube, useful length 150 mm, internal diameter 6.7 mm, wall thickness 12.5 mm. The useful length (150 mm) is the part of the TR length (250 mm) containing the packed catalyst. The MR consists of a stainless steel module containing the Pd–Ag membrane, with wall thickness of 50 μm.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Syngas Production in Membrane Reactors

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 2. Experimental set-up for the dry-steam reforming of methane in membrane reactors

Figure 3. Arrhenius plot and Sieverts plot for dense Pd-Ag membrane [58].

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

243

244

Fausto Gallucci and Angelo Basile

The membrane is a pine-hole free Pd–Ag thin wall membrane tube permeable only to hydrogen, having thickness 50 μm, o.d. 10 mm, length 150 mm. The membrane is joined to two stainless steel tube ends, but it is plugged from one side (finger like configuration) and reactants are fed by means of a stainless steel tube (o.d. 1.6 mm, i.d. 0.6 mm) placed inside the membrane lumen; the end of this internal tube is at a distance of 10 mm from the plug of the membrane tube. In this way, reactants are fed at the beginning of the catalyst bed, and both un-reacted and un-permeated species exit from the membrane tube lumen. Each gas was used with purity percentage > 99.995%. The catalyst used is Ni/Al2O3 (Ni E-5256 3/64” furnished by Engelhard). Before reaction tests, the catalyst has been activated with a N2 flow (1.02×10−3 mol/min) at 450 °C for 6 h, followed by an H2 flow (1.02×10−3 mol/min) at the same temperature for 3 h. Carbon deposition on the catalyst surface has been measured by feeding pure oxygen (160 mL/min) at 450 °C on the catalyst bed (reactor pressure 2000 Pa rel): the resulting carbonaceous gases have been detected by the Gas Chromatograph during a certain time, depending on the amount of deposited carbon. Finally, with the trapezoid integration method, the total amount of deposited carbon has been calculated. The C atom mass balance generally closed to within ±2% in all experimental data reported. After each test the catalyst was activated with the procedure reported above.

Permeation tests The dense membrane was tested for permeation of pure gases: it shows an infinite H2/(other gases) perm-selectivity. Figure 3 reports the Arrhenius’ plot and the Sieverts’ plot for the dense Pd–Ag membrane. The figure shows that the hydrogen flux increases by increasing both the temperature and the difference of the square root of the hydrogen partial pressure between the lumen side and the shell side of the reactor, that is to say that the H2 diffusion through the dense membrane follows the well-known Richardson equation:

Ji = Pe ⋅ e

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

0

− Ea RT

[(

⋅ p Hlumen 2

)

0.5

(

− p Hshell 2

)

0.5

]

where Ea is the apparent activation energy equal to 14.62 kJ/mol and Pe0 is the preexponential factor equal to 1.00151×10−6 mol/(m2 s kPa0.5). The data reported above are in good agreement with the literature data for the same kind of membrane [113]. When the MR is operating with this dense membrane, N2 as sweep gas was used and, nearby the reactants conversion, the hydrogen production was the main objective.

Reaction tests The reaction tests for the dry reforming of methane were carried out at 400 and 450 °C. Moreover, the CH4/CO2 feed flow ratio was changed between 1/1 and 1/6 by keeping constant the total flow rate.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Syngas Production in Membrane Reactors

245

Table 1. CH4 and CO2 conversion, H2 recovery (defined as the pure hydrogen permeated through the membrane over the total hydrogen produced) at different pressures and temperatures. Dry reforming reaction from [58]

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

P, bar 1.2 1.6 2.0 1.2 1.6 2.0

T, °C 400 400 400 450 450 450

CH4 conversion, % 8.0 9.0 9.8 18.0 16.5 17.0

CO2 conversion % 4.5 4.0 4.5 13.8 11.5 12.0

H2 recovery % 4.0 4.5 5.0 11.0 16.0 23.0

In our previous work [58], it was found that it is possible to use the dense Pd–Ag membrane for increasing the CH4 conversion to syngas obtaining also a pure hydrogen stream. With this aim the following experiments were performed with particular attention to the pure hydrogen recovery. The effect of reaction pressure on the conversion and on the pure hydrogen recovery is shown in Table 1. Both CO2 and CH4 conversion are quite constant with the pressure at both the temperatures considered, while the hydrogen recovery increases by increasing the pressure in the lumen side. In particular, the recovery increases from 11% at 1.2 bar to 23% at 2 bar. Being the effect of the pressure on the hydrogen recovery simple to be explained by considering the Richardson equation, the effect of the pressure on the reactants conversion in the dense membrane can be explained by considering two opposite effects. The first effect is related to the stoichiometric aspect of the reaction system: in this case by increasing the reaction pressure the reactant conversion could slowly decrease being the reaction system proceeding with a volume increase. The second effect is related to the hydrogen permeation through the dense palladium membrane. In fact, as already stated, the hydrogen flux increases by increasing the hydrogen partial pressure square roots difference between lumen side and shell side of the reactor. In this case, by increasing the reaction pressure, the hydrogen flux through the membrane increases resulting in a high conversion and a high hydrogen recovery. A problem affecting the dry reforming reaction is the amount of carbon deposited during the reaction as indicated by the reaction system proposed above. Concerning the carbon deposition, at 400 °C it ranges between 0.010g (at 1.2 bar) and 0.093g (at 2 bar) while, at 450 °C the carbon deposition ranges between 0.130g (at 1.2 bar) and 0.210g (at 2 bar). The following Table 2 gives an overview of the main results reported in the open literature for the methane dry reforming. Even though our results are in good agreement with the results found in literature, a direct comparison is not possible due to the different conditions used in the different papers. However, a general consideration can be drown: it seems that if the dry reforming is view as a carbon dioxide consumption method, the best way to proceed is to use Ni/CaAlO catalyst, the Pd/Al2O3 catalyst or Ni/Al2O3 in the TR or in a porous membrane reactor. Vice versa, if the dry reforming is view as a hydrogen or syngas production method, the best way is to operate with (1% Rh)/TiO2, Co/Al2O3 or Ni/Al2O3 catalysts in a dense palladium membrane reactor. In particular, by using the dense palladium membrane reactor, the carbon deposition on the catalyst is drastically reduced and a syngas with a higher H2/CO ratio can be obtained.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

246

Fausto Gallucci and Angelo Basile

Case Study: Dry Reforming of Methane in a Catalytic Membrane Reactor A catalytic membrane reactor is a device in which the membrane acts as both separation media and catalytic material. In this case the membrane can be intrinsically catalytic (zeolites, metallic membranes) or can be made catalytic by dispersion, inclusion, or impregnation of catalytic particles/complexes on the membrane structure. In this section the use of catalytic membrane reactors will be shown for the methane dry reforming reaction. In particular, the reaction is carried out in a Ru-based ceramic tubular membrane reactor, in which two Ru depositions have been performed using the co-condensation technique. The results in terms of CH4 and CO2 conversion versus temperature during time are presented and compared with the results of a traditional reactor. Experimental evidence points out a good catalyst activity for the methane dry reforming reaction, confirming the potentiality of a catalytic membrane applied to the reaction system.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Membrane preparation The membrane was prepared on a porous support with the so-called co-condensation technique already discussed in our previous work [119]. The starting tubular support was a commercial alumina ceramic tube (SCT, France) vitrified at both ends to assure a good sealing with the membrane module. The catalytic activity is achieved by depositing two Ru layers on the support. The deposition of both layers on the inner surface of the support has been performed by decomposition at room temperature under hydrogen of the complex Ru(η6-cycloocta1,3,5-triene)(η4-cycloocta-1,5- diene), hereinafter called Ru-complex. More in details, the starting support has been filled with the yellow solution resulting from the dissolution of 0.1 g of Ru-complex in 10 ml of dry mesitylene, and the support was plugged at both ends. The system was placed in a Schlenk tube (a glass box) equipped with a side tape; in this box a pressure of 0.1mm Hg was realized, and hydrogen was fed until the atmospheric pressure was reached. The box was left to rotate for one night. Then, the resulting uncolored solution was removed and the ruthenium deposited on the internal side of the membrane was washed with a solution of pentane and dried under an Ar stream. Similarly the second layer of Ru was deposited starting from 0.072 g of Ru-complex. Using pentane as solvent, the diffusion of the solution from the internal (lumen) to the external side of the membrane was observed, with a resulting external deposition of the metal on the tubular support. With this procedure a total amount of 55 mg of Ru has been deposited into the support.

Reaction tests The reaction tests, especially the comparison with traditional systems, described in more details in our previous work [53], are resumed in Table 3.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Table 2. Dry Reforming of Methane: comparison among literature data. Authors

p [105 Pa]

T [°C]

CH4/CO2/dil. gas

Catalyst type [% wt.]

W/F Catal. [gcat·min/molCH4] weight [g]

CH4 conv. [%]

CO2 conv. [%]

Reactor type

Gallucci et al [58] Gallucci et al [58] Gallucci et al [58] Gallucci et al [58] Chang et al. [114]

1

400 450 400 450 400 450 400 450 700 800

1/1/0.45 (N2)

Ni/Al2O3

1792

4

Ni/Al2O3

1792

4

1/1/0.45 (N2)

Ni/Al2O3

1792

4

1/6/1.52 (N2)

Ni/Al2O3

6200

4

1/1/2.2 (N2)

(5.3%

-

-

14.22 14.02 15.9 20.6 4.32 13.15 4.35 8.11 78 90

TR

1/1/0.45 (N2)

5.61 17.41 2.1 8.4 7.9 17.80 12.17 26.87 79 -

Wang and Lu [115]

1

650 700

1/1

KNiCa)/ZSI Ni/γ-Al2O3

-

0.2

48 68

50 70

TR

Ji et al. [116]

1

700 750

1.13/1/1 (Ar)

Co/γ-Al2O3

-

0.1

90 98

90 98

TR

Choudary et al. [117] Galuszka et al. [55] Galuszka et al. [55]

1

850 900 550 600 550

1/1

NiO/MgO/SA

-

0.3

Pd/γ-Al2O3

800

1

1.2/1/(40%N2)

Pd/γ-Al2O3

800

1

64 78 24.6 56.6 51

TR

1.2/1/(40%N2)

48 60 17.2 40.9 37.5 48.6

63

54

-

1 2 2 1

1 1

KNiCa)/ZSI (5.3%

600 Kikuchi [118]

1

500

1/1

Pt/CeO2-Al2O3

4480

-

MR, porous MR, dense MR, dense

TR

TR MR, dense Pd/α-Al2O3 δ = 10-15 μm MR, dense Pd/Ag δ = 4.5-13 μm

248

Fausto Gallucci and Angelo Basile

Table 3. Comparison among different reactors. T=400 °C, p=1.2 bar, feed ratio=1 Reactor type

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Membrane Reactor Tradidional reactor 5%Ru Tradidional reactor 0.5%Ru

Ru content, mg

CH4 conversion, %

CO2 conversion, %

CH4 conversion/ time factor

CO2 conversion/ time factor

55

15.23

20.8

1.9·104

2.6·104

200

2.17

15.9

7.6·102

5.6·103

20

0.35

10.8

1.2·103

3.8·104

Even though the Ru content in the membrane reactor is lower than the traditional systems loaded with 5%Ru and 0.5% catalysts, both CH4 and CO2 conversion decrease when changing from MR to TR (5% Ru) and then to TR (0.5% Ru). An interesting parameter to be considered is the conversion divided by time factor, because it included both the conversion and the effects of load/catalyst amount. As far as the CH4 conversion per unit of time factor is concerned, the MR still presents the best performances, being the results at least one order of magnitude higher than the corresponding performances of both traditional systems. However, when CO2 conversion is considered the performances of the traditional system loaded with the 0.5%Ru catalyst became closer to the performance of the MR. These results confirm the good catalytic activity of the Ru deposited inside the ceramic membrane mainly due to the finer dispersion of the catalytic particles on the membrane surface. Other considerations can be drown by considering the results in Table 3. As long as the dry reforming is performed in the MR, the gap between CH4 and CO2 conversion is 5.6% while this difference is about 13.7% for traditional system with 5%Ru catalyst and about 10.5% for traditional system with 0.5%Ru catalyst. This experimental evidence might be due to the different catalyst distribution present in the MR and in the traditional systems. Even if detection of the Ru dispersion was not done to evaluate active surface of catalyst, such a difference could lead to make unfavorable the chemical side-reactions involving CO2. This justifies the lower gap between CH4 and CO2 conversion in MR. These aspects can be viewed as an improvement of the catalyst activity in terms of reactants conversion, due to the mentioned Ru dispersion technique over the ceramic support. The conclusions can be summarized considering that a special and deeper investigation is needed, in order to tune the structure of the Ru deposited inside the tubular ceramic membrane, and realizing a very active metal layer for the reaction of interest in the direction of both increasing reactants conversion and decreasing carbon deposition during the reaction. The dry reforming of methane seems to be a feasible technology, and membrane systems could penetrate this target in the future, not only from a separation viewpoint as indicated in the previous case study, but also from an integrated reaction–separation device.

Case Study: Partial Oxidation of Methane in a Membrane Reactor As already said before, among the different reactions for syngas production, the partial oxidation of methane (POM) reaction is quite interesting being a mildly exothermic reaction

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Syngas Production in Membrane Reactors

249

that would be more energy efficient than steam reforming reactions. In fact POM, having a H2/CO mole ratio of 2.0, would be a viable alternative reaction to the commonly methane steam reforming reaction for the syngas generation. The main reaction considered for POM reaction, as already said is the following: ΔH° 298 = −36 kJ/mol

CH4 + 0.5 O2 = CO + 2H2

Of course, the use of membrane reactors for this reaction is straightforward, in fact first at all a Pd-based membrane can be used to withdraw the hydrogen from the reaction system allowing the thermodynamic limitations to be circumvented. A higher methane conversion at less severe temperatures can be obtained. In this section the results of both a membrane reactor with self-supported membrane and a membrane reactor with a supported membrane will be discussed and compared with traditional reactors operated at the same conditions.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Self supported membrane Generally speaking, a self supported Pd-membrane can be obtained via rolling of Pdalloy foil up to the desired thickness. In this way a pinhole free membrane can be obtained which presents an infinite H2 perm-selectivity. The membrane can then be welded in tubular shape or used as flat membrane. In this section a 70 micron thick, 14 cm long Pd-Ag self supported tubular membrane has been used. The results of the partial oxidation reaction in the self-supported Pd-based membrane reactor have been reported in a previous paper [107], in which the behavior of hydrogen permeation through the membrane was shown. The membrane behaves like the Pd-based membrane reported in the previous section, the permeation results of which are reported above in Figure 3. In the next Table 4 the comparison between the performances in terms of methane conversion for both traditional reactor and membrane reactor is presented. The same table also included the methane conversion calculated for the thermodynamic equilibrium (which applies to the TR only). It is quite clear that membrane reactor allows attaining higher conversions with respect to a traditional reactor working at the same temperature. Moreover, the membrane reactor also helps overcoming the thermodynamic limitations, while the conversions in traditional systems are always lower than the corresponding equilibrium conversion. The difference between the conversion in membrane reactor and the equilibrium one increases by increasing the temperature. This because the higher the temperature is the greater the flux through the membrane is. Table 4. Methane conversion in traditional reactor and membrane reactor as a function of temperature. CH4/O2/N2 = 2/1/14, p = 1.6 bar Temperature, °C

CH4 conversion in TR, %

350 400 450 500

10.67 18.10 25.34 34.09

CH4 conversion in MR, % 33.90 45.72 56.19 70.28

CH4 conversion thermodynamic equilibrium, % 29.75 33.47 38.60 46.51

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Fausto Gallucci and Angelo Basile

250

Table 5. Methane conversion in traditional reactor and membrane reactor as a function of time factor. T = 500°C. Time factor, gcat min/molCH4 1400 2100 3060 4288

CH4 conversion in TR, % 47.91 52.32 48.83 54.71

CH4 conversion in MR, % 75.33 77.91 83.07 83.98

CH4 conversion thermodynamic equilibrium, % 56 56 56 56

Table 6. Palladium thickness as a function of number of depositions (after [84]). Number of depositions 1 2 3 4 5 6 7

Pd-layer Thickness [μm] 10 12 13 17 18 19 21

A way to increase the methane conversion in the traditional reactor up to the equilibrium conversion is the increase of the time factor (i.e. decreasing the gas load on the catalyst) as indicated in the following Table 5. By increasing the time factor, the methane conversion increases in the membrane reactor too, reaching a maximum value (similar to a plateau) of 84% at 4288 gcat min/molCH4 with an increase on the methane conversion attained in the traditional reactor of 53.5%. It is then clear that the use of a Pd-based dense membrane reactor can give a great improvement in terms of methane conversion to syngas resulting in a big energy and material saving if applied at industrial level.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Porous Pd-based membrane In order to produce Pd-based membranes industrially interesting, the Pd layer should be decreased with the aim of increasing the hydrogen flux and decreasing the material costs. The way to achieve this goal is to use a different membrane production technique than the rolling one described above. The most used technique is the deposition of Pd layers on a support via the so-called electroless-plating technique. In this section the use of a Pd-based supported membrane to the partial oxidation of methane is discussed. The first step is the production of the Pd layer on the support. A commercial TiO2 tubular support has been chosen. The inner surface of the support presents a 56 μm layer with 5 nm pore diameter. The deposition via electroless plating was repeated several times in order to obtain a Pd-layer as dense as possible. After each deposition the membrane thickness has been evaluated and the membrane as been used for the partial oxidation reaction [84,120]. The Pd-layer thickness increased with increasing the number of depositions as indicated in the following Table 6.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Syngas Production in Membrane Reactors

251

Table 7. Methane conversion as a function of the number of Pd layers for the MR at 400 and 450 °C Number of Pd layers 0 2 4 6 7

CH4 conversion, 400 °C - (TR) 15.67 33.51 32.43 37.66

CH4 conversion, 450 °C 18.02 (TR) 33.69 48.28 44.14 46.88

With increasing the Pd-layer the selectivity of the membrane was also increased as indicated in previous works [84,120]. The application of this membrane to the partial oxidation reaction indicates that the membrane with the higher thickness gives better results in terms of methane conversion. This because, the higher the thickness the denser the Pd-layer is and the higher the perm-selectivity results. The results in terms of conversion are reported in Table 7. As it can be seen from the table, the membrane reactor is able to give always higher conversions than the traditional reactor working at the same conditions, and the difference between the two systems increases by increasing the number of Pd-layers. Also in this case the membrane reactor can be seen as a good candidate for producing synthesis gas via partial oxidation of methane, allowing attaining a higher conversion than a traditional system working at the same conditions of temperature and pressure.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Case Study: Partial Oxidation of Methane in a Catalytic Membrane Reactor The Ru-based catalytic membrane, the preparation of which has been already discussed for the case study of methane dry reforming, has been applied for the methane partial oxidation and the catalytic performances will be discussed in this short section. In the case of methane partial oxidation, the catalytic membrane reactor was used in three different configurations as indicated in Figure 4. The results obtained with the catalytic membrane reactor are plotted in Figure 5, which plots the behavior of the catalytic membrane reactor in terms of methane conversion versus temperature. In particular, a comparison of the experimental results between one and twolayer Ru membrane reactor is reported in the figure. The retentate closed configuration is considered (configuration B) so that only the permeate outlet stream was present. The thermodynamic equilibrium curve is also plotted in this figure. In the retentate closed configuration, the reactant gases keep full contact with the Ru catalyst deposited inside the membrane. For both kinds of membranes, an increasing trend was observed, according to the endothermicity of the overall reaction system. In particular, considering the one-layer Ru membrane, methane conversion was about 23% at 350 °C and 54% at 500 °C. Nevertheless, methane conversion was about 15% at 300 °C and about 55% at 500 °C for the two-layer Ru membrane. The two layer Ru membrane seems to give a lower methane conversion with respect to the one-layer Ru membrane, and the maximum difference is achieved at 450 °C. At this temperature, the maximum methane conversion is about 40% for the one-layer Ru membrane versus a methane conversion of about 23% related to the two layer Ru membrane.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

252

Fausto Gallucci and Angelo Basile

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

This experimental evidence can be explained considering two opposite effects on methane conversion, due to the second Ru deposition. The situation can be discussed by considering the scheme shown in Figure 6. In particular, with regard to the catalytic effect, an increase in the Ru amount should give an increase in methane conversion. On the contrary, the second Ru deposition should reduce the pore size, increasing the gas velocity throughout the membrane (permeation effect): in this condition, the contact time with the Ru catalyst decreases, so methane conversion is negatively affected. In our experimental condition, the second effect seems to prevail. Considering the atom mass balance closure, the same conversion is approached when the temperature reaches 500 °C.

Figure 4. Catalytic membrane reactor configurations.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Syngas Production in Membrane Reactors

253

70 350 °C ==> 500 °C (1st Ru Layer) 500 °C ==> 350 °C (1st Ru Layer) 350 °C ==> 500 °C (2nd Ru Layer) 500 °C ==> 350 °C (2nd Ru Layer) Thermodynamic equilibrium

CH4 conversion, %

60

50

40

30

20

10 300

350

400

450

500

Temperature, °C

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 5. Methane conversion versus temperature for catalytic membrane reactor containing one and two layers of deposited Ru membranes, retentate closed configuration, p= 0.2 bar, H4/O2/N2 = 2/1/14, and CH4,feed 2 ⋅10-3 mol/min (after [121])

Figure 6. Effect of the Second Ru Deposition on the Methane Conversion in catalytic membrane reactor, Retentate Closed Configuration, under a Constant Mass Flow Rate of Reactants (after [121])

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

254

Fausto Gallucci and Angelo Basile

In conclusion, the Ru deposition performed with the described technique produces a catalytic membrane that gives methane conversion values near the thermodynamic predictions, at each temperature investigated. Although, in principle, the Ru catalyst needs doped supports in order to exhibit high catalytic activity, in our work the starting tubular support did not suffer any pretreatment, and nevertheless the deposited Ru exhibited an interesting catalytic activity, such that thermodynamic equilibrium conversion was approached.

CONCLUSIONS In this chapter the syngas production in membrane reactor has been considered as an energy efficient alternative to traditional systems. In the first part of the chapter, a review concerning the different routes for syngas production has been presented, while in the second part different case study about membrane reactors for syngas production have been shown. Membranes can be used for syngas production in different ways. The most promising way is to use catalytic membrane reactors able to achieve higher conversion than traditional systems with lower catalyst amount and even with lower temperatures. On the other hand, Pd-based membranes can be used for syngas production resulting in pure hydrogen stream production, syngas ratio conditioning and less severe conditions for the reaction systems if compared with traditional reactors.

REFERENCES [1]

[2]

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

[3] [4] [5] [6]

[7]

Lachowska, M; Skrzypek, J. Synthesis of higher alcohols. Enhancement by the addition of methanol or ethanol to the syngas, in Studies in Surface Science and Catalysis, A. Parmaliana et al. (Eds.), Elsevier, 1998, 119, 473-478. Apesteguia, CR; Miseo, S; De Rites, B; Soled, S; Isoalcohol synthesis from CO/H2 feedstocks, in Studies in Surface Science and Catalysis, A. Parmaliana et al. (Eds.), Elsevier, 1998, 119, 485-490. Jager, B. Developments in Fischer-Tropsch technology, in Studies in Surface Science and Catalysis, A. Parmaliana et al. (Eds.), Elsevier, 1998 , 119, 25-34. Ross, JRH; Van Keulen, ANJ; Hegarty, MES; Seshan, K. The catalytic conversion of natural gas to useful products, Catal. Today, 1996, 30, 193-199. Vannice, MA; The catalytic synthesis of hydrocarbons from carbon monoxide and hydrogen, Catal. Rev. Sci. Eng., 1976, 14, 153-191. Schmidt, LD; Hohm, KL; Davis, MB. Catalytic partial oxidation of methane at extremely short contact times: production of acetylene, in Studies in Surface Science and Catalysis, A. Parmaliana et al. (Eds.), Elsevier, 1998, 119, 397-402. Shikada, T; Ohno, Y; Ogawa, T; Ono, M; Mizuguchi, M; Tomura, K; Fujimoto, K. Direct synthesis of dimethyl ether from synthesis gas, in Studies in Surface Science and Catalysis, A. Parmaliana et al. (Eds.), Elsevier, 1998, 119, 515-520.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Syngas Production in Membrane Reactors [8]

[9]

[10] [11] [12] [13]

[14] [15] [16] [17]

[18] [19] [20] [21]

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

[22] [23]

[24]

[25] [26]

[27]

255

Mysov, VM; Ione, KG; Toktarev, AV. The investigation of the processes of organic products synthesis from natural gas via syngas, in Studies in Surface Science and Catalysis, A. Parmaliana et al. (Eds.), Elsevier, 1998, 119, 533-538. Norbeck, M; Hackett, CE. United States patent number 7,208,530 B2; C.S. Park, S.P. Singh, J.M. Norbeck, Steam hydrogasification of carbonaceous matter to liquid fuels, Proc. 24th Annual Int. Pittsburgh Coal Conf., Johannesburg (South Africa), 2007. Spencer, ND; Pereiva, CJ. V2O5-SiO2 catalyzed methane partial oxidation with molecular oxygen, J. Catal., 1989, 116, 399-406. Gesser, HD; Hunter, NR; Prakash, CB. The direct conversion of methane to methanol by controlled oxidation, Chem. Rev., 1985, 85 235-244. Amenomiya, Y; Buss, VI; Goledzinszka, M; Galuszka, J; Sanger, AR. Conversion of methane by oxidative coupling, Catal. Rev.-Sci. Eng., 1990, 32, 163-227. Henrici-Olive, G; Olive, S. The Fischer-Tropsch synthesis: molecular weight distribution of primary products and reaction mechanism, Angew. Chem. Int. Ed. Engl., 1976, 15, 136-141. Kung, HH. Methanol synthesis, Catal. Rev., 1980, 22, 235-259. Gradassi, MJ. Economics of gas to liquids manufacture, in Studies in Surface Science and Catalysis, A. Parmaliana et al. (Eds.), Elsevier, 1998, 119, 35-44. Boucouvalas, Y; Zhang, ZL; Verykios, XE. Partial oxidation of methane to synthesis gas via the direct reaction scheme over Ru/TiO2, Catal. Lett., 1996, 40, 189-195. Boucouvalas, Y; Zhang, ZL; Efstathiou, AM; Verykios, XE. Partial oxidation of methane to synthesis gas over Ru/TiO2 catalysts in Studies in Surface Science and Catalysis, Vol. 101, L.W. Hightower et al. (Eds), Elsevier, Amsterdam, 1996, p. 443-452. Hickman, DA; Haupfear, EA; Schmidt, LD. Synthesis gas formation by direct oxidation of methane over Rh monoliths, Catal. Lett., 1993, 17, 223-237. Torniainen, PM; Chu, X; Schmidt, D. Comparison of monolith-supported metals for the direct oxidation of methane to syngas, J. Catal., 1994, 146, 1-10. Rostrup-Nielsen, JR. Production of synthesis gas, Catal. Today, 1993, 18, 305-324. Rostrup-Nielsen, JR. Catalytic Steam Reforming, Springer-Verlag, New York, NY, 1984. World Car Conference, 19-22 January, 1997, Riverside, CA, pp. 331-338. Vogel, B; Schaumberg, G; Schuler, A; Henizel, A. Hydrogen Generation Technologies for PEM Fuel Cells, 1998 Fuel Cell Seminar Abstracts, Nov., 1998,16-19, Palm Springs, CA, pp. 364-367. Ogden, JM. Review of small stationary reformers for hydrogen production, A report for the International Energy Agency Agreement on the Production and Utilization of Hydrogen Task 16, Hydrogen from Carbon-Containing Materials, 2001. Edlund, D; Pledger, WA; Dickman, A. Field Testing Residential Fuel Cell Systems, 2000 Fuel Cell Seminar Abstracts, Oct. 30-Nov. 2, 2000, Portland (OR), pp. 496-499. Oklany, JS; Hou, K; Hughes, R. A simulative comparison of dense and microporous membrane reactors for the steam reforming of methane, Applied Catal. A: General, 1998, 170 13-22. Kikuchi, E. Membrane reactor application to hydrogen production, Catal. Today, 2000, 56, 97-101.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

256

Fausto Gallucci and Angelo Basile

[28] Lin, Y; Rei, M. Process development for generating high purity hydrogen by using supported palladium membrane reactor as steam reformer, Int. J. Hydrogen En., 2000, 25 211-219. [29] Armor, J. Applications of catalytic inorganic membrane reactors to refinery products, J. Membrane Sci., 1998, 147, 217-233. [30] Julbe, A; Ferrusseng, D; Guizard, C. Porous ceramic membranes for catalytic reactors – overview and new ideas, J. Membrane Sci., 2001, 181, 3-20. [31] Dittmeyer, R; Hollein, V; Daub, K. Membrane reactors of hydrogenation and dehydrogenation processes based on supported palladium, J. Molecular Catal., 2001, 173, 135-184. [32] Deshmukh, SARK; Heinrich, S; Morl, L; Van Sint Annaland, M; Kuipers, J.A. Membrane assisted fluidixzed bed reactors: potential and hurdles, Chem. Eng. Sci., 2007, 62, 416-436. [33] Smith, B; Shantha, MS. Membrane reactor based hydrogen separation from biomass gas – A review of technical advancements and prospects, Int. J. Chem. Reactor Eng., 2007, 5, Art. A84. [34] Booth, J; Doyle, M; Gee, S; Miller, J; Scholtz, LA; Walker, P. Advanced Separation via Thin Supported Pd Membranes, Proc. 11th World Hydrogen Energy Conference, Stuttgart (Germany), June, 1996, 23-28, [35] Yasuda, I; Shirasaki, Y; Tsuneki, T; Asakura, T; Kataoka A; Shinkai, H. Proc. of the 15th world hydrogen energy conference (CD-ROM), Yokohama (Japan), 2004, 27 June–2 July, [36] Shirasaki, Y; Tsuneki, T; Ota, Y; Yasuda, I. Shinya Tachibana, Hiroshi Nakajima and K. Kobayashi, WHTC2007, Montecatini Terme (Italy), 2007, Nov. 5. [37] Unemoto, A; Kaimai, A; Sato, K; Otake, T; Yashiro, K; Mizusaki, J; Kawada, T; Tsuneki, T; Shirasakic, Y; Yasuda, I. The effect of co-existing gases from the process of steam reforming reaction on hydrogen permeability of palladium alloy membrane at high temperatures, Int. J. Hydrogen En., 2007, 32(14), 2881-2887. [38] Unemoto, A; Kaimai, A; Sato, K; Otake, T; Yashiro, K; Mizusaki, J; Kawada, T; Tsuneki, T; Shirasakic, Y; Yasuda, I. Surface reaction of hydrogen on a palladium alloy membrane under co-existence of H2O, CO,CO2 or CH4, Int. J. Hydrogen En., 2007, 32(14), 4023-4029. [39] Fernandes, FAN; Soares Jr, AB. Methane steam reforming modeling in a palladium membrane reactor, Fuel, 2006, 85, 569–573. [40] Chen, Y; Wang, Y; Xu, H; Xiong, G. Efficient production of hydrogen from natural gas steam reforming in palladium membrane reactor, Appl. Catal. B: Env., 2008, 81, 283–294. [41] Kusakabe, K; Sotowa, KI; Eda, T; Iwamoto, Y. Methane steam reforming over Ce– ZrO2-supported noble metal catalysts at low temperature, Fuel Proc. Technol., 2004 86, 319–326. [42] Edwards, JH. Potential sources of CO2 and the options for its large-scale utilisation now and in the future, Catal. Today, 1995, 23 59-66. [43] Hou, K; Hughes, R. The kinetics of methane steam reforming over a Ni/α-Al2O catalyst, Chem. Eng. J., 2001, 82, 311–328. [44] Wang, HY; Au, C.T. Carbon dioxide reforming of methane to syngas over SiO2supported rhodium catalysts, Appl. Catal. A, 1997, 155, 239-252.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Syngas Production in Membrane Reactors

257

[45] Erdöhelyi, A; Cserényi, J; Solymosi, F. Activation of CH4 and its reaction with CO2 over supported Rh catalysts, J. Catal. 1993, 141, 287-299. [46] Basini, L; Sanfilippo, D. Molecular aspects in syn-gas production: the CO2-reforming reaction case, J. Catal., 1995, 157 162-178. [47] Mark, MF; Maier, WF; CO2-reforming of methane on supported Rh and Ir catalysts, J. Catal. 1996, 164, 122-130. [48] Rostrup- JR; Nielsen, Hansen, JHB. CO2-Reforming of methane over transition metals, J. Catal., 144 (1993) 38-49. [49] Shamsi, A; Johnson, CD. Effect of pressure on the carbon deposition route in O2 reforming of 13CH4, Catal. Today, 2003, 84, 17-25. [50] Choudhary, VR; Uphade, BS; Balhekar, AA. Oxidative conversion of methane to syngas over LaNiO3 perovskite with or without simultaneous steam and CO2 reforming reactions: influence of partial substitution of La and Ni, J. Catal., 1996,163 312-318. [51] Bolt, PH; Habraken, FHPM; Geus, JW. On the role of a NiAl2O4 intermediate layer in the sintering behavior of Ni/α-Al2O3, J. Catal., 1995, 15, 300-306. [52] York, APE; Claridge, JB; Brungs, AJ; Tsang, SC; Green, MLH. Molybdenum and tungsten carbides as catalysts for the conversion of methane to synthesis gas using stoichiometric feedstocks, Chem. Comm., 1997, 39-40. [53] Paturzo, L; Gallucci, F; Basile, A; Vitulli, G; Pertici, P. A Ru-based catalytic membrane reactor for dry reforming of methane—its catalytic performance compared with tubular packed bed reactors, Catal. Today, 2003, 82, 57-65. [54] Sakurai, H; Akita, T; Tsubota, S; Ciuchi, M; Haruta, M. Low-temperature activity of Au/CeO2 for water gas shift reaction, and characterization by ADF-STEM, temperature-programmed reaction, and pulse reaction, Appl. Catal. A: Gen., 291 (2005) 179-187 [55] Galuszka, J; Pandey, RN; Ahmed, S. Methane conversion to syngas in a palladium membrane reactor, Catal. Today, 1998, 46, 83-89 [56] Liu, BS; Au, CT. A La2NiO4-zeolite membrane reactor for the CO2 reforming of methane to syngas, Catal. Letters, 2001, 77(1–3), 67-74 [57] Liu, BS; Gao, LZ; Au, CT. Preparation, characterization and application of a catalytic NaA membrane for CH4/CO2 reforming to syngas, Applied Catal. A: General, 2002, 235, 193–206 [58] Gallucci, F; Tosti, S; Basile, A. Pd–Ag tubular membrane reactors for methane dry reforming: A reactive method for CO2 consumption and H2 production, J. Membrane Sci., 2008, 317, 96-105 [59] Jorgensen, SL; Nielsen, PEH; Lehrmann, P. Steam reforming of methane in a membrane reactor, Catal. Today, 1995, 25, 303-307. [60] Li, C; Fu, Y; Bian, G; Xie, Y; Hu, T; Zhang, J. Effect of steam in CO2 reforming of CH4 over a Ni/CeO2–ZrO2–Al2O3 catalyst, Kinetics & Catal., 2004, 45, 679–683. [61] Qin, D; Lapszewicz, J; Jiang, X. Comparison of partial oxidation and steam-CO2 mixed reforming of CH4 to syngas on MgO-supported metals, J. Catal., 1996, 159, 140–149. [62] Tsipouriari, VA; Verykios, XE. Kinetic study of the catalytic reforming of methane with carbon dioxide to synthesis gas over Ni/La2O3 catalyst, Catal.Today, 2001, 64, 83–90. [63] Abashar, MEE. Coupling of steam and dry reforming of methane in catalytic & fluidized bed membrane reactors, I. J. Hydrogen En., 2004, 29, 799–808

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

258

Fausto Gallucci and Angelo Basile

[64] Froment, GF. Production of synthesis gas by steam- and CO2-reforming of natural gas, J. Molecular Cat. A: Chemical, 2000, 163, 147–156. [65] Snoeck, JW; Froment, G; Fowles, M. Kinetic evaluation of carbon formation in steam/CO2-natural gas reformers. Influence of the catalyst activity and alkalinity, Int. J. Chem. Reactor Eng., 2003, 1, 1–16. [66] Effendi, A; Zhang, ZG; Hellgardt, K; Honda, K; Yoshida, T. Steam reforming of a clean model biogas over Ni/Al2O3 in fluidized- and fixed-bed reactors, Catal. Today, 2002, 77, 181–189. [67] Choudhary, VR; Rajput, AM. Simultaneous carbon dioxide and steam reforming of methane to syngas over NiO–CaO catalyst, Ind. & Eng. Chem. Res., 1996, 35, 3934– 3939. [68] Lemonidou, AA; Vasalos, IA. Carbon dioxide reforming of methane over 5 wt.% Ni/CaO–Al2O3 catalyst, Appl. Catal. A: General, 2002, 228, 227–235. [69] Tsang, SC; Claridge, JB; Green, MLH. Recent advances in the conversion of methane to synthesis gas, Catal. Today, 1995, 23, 3-15. [70] Lu, YP; Dixon, AG; Moser, WR; Ma, YH; Oxygen-permeable, UB, dense membrane reactor for the oxidative coupling of methane, J. Membrane Sci., 2000, 170, 27-34. [71] Aschcroft, AT; Cheetham, AK; Foord, JS; Green, MLH; Grey, CP; Murrell, AJ; Vernon, PDF. Selective oxidation of methane to synthesis gas using transition metal catalysts, Nature, 1990, 344, 319-321. [72] Vermeiren, WJM; Blomsma, E; Jacobs, PA. Catalytic and thermodynamic approach of the oxyreforming reaction of methane, Catal. Today, 1992, 13, 427-436. [73] Hickman, DA; Schmidt, LD. Production of syngas by direct catalytic oxidation of methane, Science, 1993, 259, 343-346. [74] Mitchell, W; Thijssen, J; Bentley, JM. Development of a catalytic partial oxidation ethanol reformer for fuel cell applications, Society of Automotive Eng., 1995, 9527611. [75] Cross, J., Mitchell, WL; Chintawar, P; Hagan, M; Thompson, C; Swavely, D. PEM Fuel Cell Power Technology, 2000 Fuel Cell Seminar Abstracts, Oct.30-Nov. 2, 2000, Portland (OR), pp. 260-263. [76] Dyer, PN. Ion Transport Membranes, Presentation at the U.S. Dept. of Energy Hydrogen Program 1999 Annual Peer Review Meeting, Lakewood (CO), May 4-6, 1999, cited in ref. [24]. [77] Dyer, PN; Chen, CM. ITM Syngas and ITM H2, Proc. of the 2000 Hydrogen Program Review, NREL/CP-570-28890, 2000, cited in ref. [24]. [78] Shah, M; Drnevich, RF; Balachandran, U. Integrated Ceramic Membrane System for Hydrogen Production, Proc. of the 2000 Hydrogen Program Review, NREL/CP-57028890, 2000, cited in ref. [24]. [79] Blanks, F; Wittrin, TS; Peterson, DA. Bidirectional adiabatic synthesis gas generator, Chem. Eng. Sci., 1990, 45, 2407-2413 [80] Piga, A; Ioannides, T; Verykios, XE. Synthesis gas formation by catalytic partial oxidation of methane in a heat-integrated wall reactor, in Studies in Surface Science and Catalysis, A. Parmaliana et al. (Eds), Elsevier, Amsterdam, 1998, 119, 411-416. [81] Dissanayake, D; Rosynek, MP; Kharas, KCC; Lunsford, JH. Partial oxidation of methane to carbon monoxide and hydrogen over a Ni/Al2O3 catalyst, J. Catal., 132 (1991) 117-127.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Syngas Production in Membrane Reactors

259

[82] Hickman, DA; Schmidt, LD. Steps in CH4 oxidation on Pt and Rh surfaces - Hightemperature reactor simulations, AIChE J. 1993, 39, 1164-1177. [83] Ostrowski, T; Girior-Fendler, A; Mirodatos, C; Mleczko, L. Comparative study of the catalytic partial oxidation of methane to synthesis gas in fixed-bed and fluidized-bed membrane reactors: Part I: A modeling approach, Catal. Today, 1998, 40, 181-190. [84] Basile, A; Paturzo, L. An experimental study of multilayered composite palladium membrane reactors for partial oxidation of methane to syngas, Catal. Today, 67 (2001) 55-64. [85] Balachandran, U; Dusek, JT; Maiya, PS; Ma, B; Mieville, RL; Kleefisch, MS; Udovich, CA. Ceramic membrane reactor for converting methane to syngas, Catal. Today, 1997, 36, 265-272 [86] Tsai, CY; Dixon, AG; Moser, WR; Ma, YH. AIChE J. 1997, 43 (11A), 2741. [87] Tsai, CY; Dixon, AG; Ma, YH; Moser, WR; Pascucci, MR. Dense Perovskite, La1xA'xFe1-yCoyO3-δ (A'= Ba, Sr, Ca), Membrane Synthesis, Applications, and Characterization, J. Am. Ceram. Soc., 1998, 81 (6), 1437-1444. [88] Jin, W; Li, S; Huang, P; Xu, N; Shi, J; Lin, YS. Tubular lanthanum cobaltite perovskite-type membrane reactors for partial oxidation of methane to syngas, J. Membrane Sci., 2000, 166, 13-22. [89] H. Dong, Z. Shao, G. Xiong, J. Tong, S. Sheng,W. Yang, Investigation on POM reaction in a new perovskite membrane reactor, Catal. Today, 2001, 67, 3-13. [90] Shao, Z; Xiong, G; Dong, H; Yang, W; Liu, L. Synthesis, oxygen permeation study and membrane performance of a Ba0.5Sr0.5Co0.8Fe0.2O3−δ oxygen-permeable dense ceramic reactor for partial oxidation of methane to syngas, Sep. Purif. Technol., 2001, 25, 97116. [91] Ishihara, T; Tsuruta, Y; Todaka, T; Nishiguchi, H; Takita, Y. Fe doped LaGaO3 perovskite oxide as an oxygen separating membrane for CH4 partial oxidation, Solid State Ionics, 2002, 152, 709-714. [92] Zhu, DC; Xu, XY; Feng, S; Liu, W; Chen, CS. La2NiO4 tubular membrane reactor for conversion of methane to syngas, Catal. Today, 2003, 82, 151-156. [93] Wang, H; Cong, Y; Yang, W. Investigation on the partial oxidation of methane to syngas in a tubular Ba0.5Sr0.5Co0.8Fe0.2O3−δ membrane reactor, Catal. Today, 2003, 82, 157-166. [94] Ritchie, JT; Richardson, JT; Luss, D. Ceramic membrane reactor for synthesis gas production, AIChE J., 2001, 47 (9), 2092-2101. [95] Kharton, VV; Sobyanin, VA; Belyaev, VD; Semin, GL; Veniaminov, SA; Tsipis, EV; Yaremchenko, AA; Valente, AA; Marozau, IP; Frade, JR; Rocha, J. Methane oxidation on the surface of mixed-conducting La0.3Sr0.7Co0.8Ga0.2O3-δ, Catal. Commun., 2004, 5, 311-316. [96] Bouwmeester, HJM. Dense ceramic membranes for methane conversion, Catal. Today, 82 (2003) 141-150. [97] Ikeguchi, M; Yoshino, Y; Kanie, K; Nomura, M; Kikuchi, E; Matsukata, M. Effects of preparation method on oxygen permeation properties of SrFeCo0.5Ox membrane, Sep.. Purif. Technol., 2003, 32, 313-318. [98] Jin, W; Li, S; Huang, P; Xu, N; Shi, J. Fabrication of La0.2Sr0.8Co0.8Fe0.2O3−δ mesoporous membranes on porous supports from polymeric precursors, J. Membrane Sci., 2000, 170, 9-17.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

260

Fausto Gallucci and Angelo Basile

[99] Balachandran, U; Ma, B; Lee, TH; Picciolo, JJ; Dorris, SE. Proc. Int. Hydrogen En. Congress & Exhibition, Istanbul (Turkey), July 13–15, 2005. [100] Guan, J; Dorris, SE; Balachandran, U; Liu, M. Transport properties of BaCe Y O

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

0.95

0.05

3-∂

mixed conductors for hydrogen separation, Solid State Ionics, 1997, 100, 45-52. [101] Guan, J; Dorris, SE; Balachandran, U; Liu: M. The effects of dopants and A:B site nonstoichiometry on properties of perovskite-type proton conductors. J. Electrochem. Soc., 1998, 145, 1780-1786. [102] Guan, J; Dorris, SE; Balachandran, U; Liu, M. Development of mixed-conducting ceramic membranes for hydrogen separation. Ceram. Trans., 1998, 92, 1. [103] Dyer, PN; Richards, R.E; Russek, L; Taylor, DM. Ion transport membrane technology for oxygen separation and syngas production, Solid State Ionics, 2000, 134, 21–33. [104] Basile, A; Fasson, S; Vitulli, G; Drioli, E. An experiemental study of the partial oxidation of methane in a membrane reactor, in Studies in Surface Science and Catalysis, A. Parmaliana et al. (Eds.), Elsevier, 1998, 119, 453-458. [105] Basile, A; Fasson, S. Progresses on the partial oxidation of methane to syngas in a membrane reactor, in Studies in Surface Science and Catalysis, A. Parmaliana et al. (Eds.), Elsevier, 1998, 119, 459-464. [106] Kikuchi, E; Chen, Y. Syngas formation by partial oxidation of methane in palladium membrane reactor, in Studies in Surface Science and Catalysis, A. Parmaliana et al. (Eds.), Elsevier, 1998, 119, 441-446. [107] Basile, A; Paturzo, L; Laganà, F. The partial oxidation of methane to syngas in a palladium membrane reactor: simulation and experimental studies, Catal. Today, 2001, 67, 65–75. [108] Lange, JP; Perspectives for manufacturing methanol at fuel value, Ind. Eng. Chem. Res., 1997, 36, 4282-4290. [109] Shao, ZP; Dong, H; Xiong, GX; Cong, Y; Yang, W. Performance of a mixedconducting ceramic membrane reactor with high oxygen permeability for methane conversion, J. Membrane Sci., 2001, 183, 181-192. [110] Liu, S; Xiong, G; Yang, W; Xu, L; Xiong, G; Li, C. Partial oxidation of ethane to syngas over nickel-based catalysts modified by alkali metal oxide and rare earth metal oxide, Catal. Lett., 1999, 63, 167-171. [111] Liu, S; Xiong, G; Sheng, S; Yang, W; Yang, W. Partial oxidation of methane and ethane to synthesis gas over a LiLaNiO/g–Al2O3 catalyst, Appl. Catal: A, 2000, 198, 261-266. [112] Ran, R; Xiong, GX; Sheng, SS; Yang, W; Stroh, N; Brunner, H. Catalytic partial oxidation of n-heptane for hydrogen production, Catal. Lett., 2003, 88 (1–2), 55-59. [113] Gallucci, F; Paturzo, L; Basile, A. Hydrogen recovery from methanol steam reforming in a dense membrane reactor: simulation study, Ind. Eng. Chem. Res., 2004, 43(10) 2420–2432. [114] Chang, JS; Park, SE; Chon, H. Catalytic activity and coke resistance in the carbon dioxide reforming of methane to synthesis gas over zeolite-supported Ni catalysts, Appl. Catal. A: Gen., 1996, 145, 111-124. [115] Wang, S; Lu, GQ. Reforming of methane with carbon dioxide over Ni/Al2O3 catalysts: Effect of nickel precursor, Appl. Catal. A: Gen., 1998, 169, 271-280.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Syngas Production in Membrane Reactors

261

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

[116] Ji, L; Tang, S; Zeng, HC; et al., CO2 reforming of methane to synthesis gas over solgel-made Co/γ-Al2O3 catalysts from organometallic precursors, Appl. Catal. A: Gen., 207 (2001) 247-255. [117] Choudhary, VR; Uphade, BS; Mamman, AS. Simultaneous steam and CO2 reforming of methane to syngas over NiO/MgO/SA-5205 in presence and absence of oxygen, Appl. Catal. A: Gen., 1998, 168, 33-46. [118] Kikuchi, E; Palladium/ceramic membranes for selective hydrogen permeation and their application to membrane reactor, Catal. Today, 1995, 25, 333-337. [119] Gallucci, F; Basile, A. Pd-based membranes synthesis and their application in membrane reactors, Chapter 1 in: Handbook of Membrane Research: Properties, Performance and Applications, Editor: Stephan V. Gorley, Nova Science. Pub. New York, ISBN: 978-1-60741-638-8. [120] Paturzo, L; Basile, A. Methane Conversion to Syngas in a Composite Palladium Membrane Reactor with Increasing Number of Pd Layers , Ind. Eng. Chem. Res., 2002, 41, 1703-1710. [121] Paturzo, L; Gallucci, F; Basile, A; Pertici, P; Scalera, N; Vitulli, G. Partial Oxidation of Methane in a Catalytic Ruthenium Membrane Reactor, Ind. Eng. Chem. Res., 2003, 42, 2968-2974.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved. Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

In: Syngas Production Methods, Post Treatment… Editors: Adorjan Kurucz and Izsak Bencik

ISBN: 978-1-60741-841-2 © 2009 Nova Science Publishers, Inc.

Chapter 6

REFORMER AND MEMBRANE MODULES PLANT TO OPTIMIZE NATURAL GAS CONVERSION TO HYDROGEN M. De Falco, G. Iaquaniello, B. Cucchiella and L. Marrelli University of Rome, Rome, Italy

ABSTRACT Membrane technology may play a crucial role in the efficient production of hydrogen from natural gas and heavy hydrocarbons. The chapter assesses the performance of hydrogen production plants in which Pd-based selective membranes are integrated. Two different configurations have been proposed and are evaluated:



Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.



Reformer and Membrane Modules (RMM), by which the hydrogen produced in reaction units is separated by Pd-based membrane modules assembled downstream each reaction step. Membrane Reformer (MR), which combines the hydrogen separation through the selective membrane and the steam reforming reaction into one unit and separates hydrogen immediately after it was formed.

Both the configurations allow a reduction of operating temperature (< 650°C for RMM, < 550°C for MR) with benefits in terms of process energy efficiency. Moreover, lower operating temperatures allow location of the modules downstream of a gas turbine, achieving an efficient hybrid system producing electric power and hydrogen. The chapter will be composed by the following sections:

• • • •

The concept of membrane integration in steam reforming process: benefits and drawbacks. Hydrogen selective membranes state-of-the-art and performance. RMM plant configuration: modeling and simulations. MR plant configuration: modeling and simulations.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

M. De Falco, G. Iaquaniello, B. Cucchiella et al.

264 •

Comparison between RMM/MR technologies with conventional steam reformer plants: economical evaluation of the process layout.

1. INTRODUCTION: MEMBRANE INTEGRATION IN STEAM REFORMING PROCESS 1.1. Steam Reforming Process The today’s total hydrogen produced (50 million of tonnes per year) is used in: •

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.



chemical industry, as reactant for the ammonia and methanol synthesis, for the hydrogenation of vegetal oil and as reductant to produce metals from their oxides; in the refinery processes for hydro-desulphuration of sulphur compounds and for hydrocracking process. The deteriorating quality of crude oils, more stringent petroleum product specifications and environmental problems is leading to larger need of hydrogen in hydro-processing. Hydrogen can be obtained from different sources (Figure 1.1): fossil fuels (natural gas reforming, partial oxidation, coal gasification), renewable fuels (biomass gasification), algae and vegetables (biological production) or water (electrolysis and thermochemical cycles). Many different energy forms can be used in most of these processes: heat from nuclear reactors, electricity from several sources, solar energy.

Figure 1.1. H2 production feedstocks and process alternatives

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Reformer and Membrane Modules Plant to Optimize Natural…

265

Figure 1.2. Feedstocks to produce hydrogen

Nowadays, 48% of the hydrogen is produced from the natural gas steam reforming (NG), as reported in the pie-chart below (Figure 1.2). The reactions involved in the steam reforming process are: kJ mol

(1.1)

kJ mol

(1.2)

CH 4 + H 2 O ⇔ CO + 3H 2 ΔH 298 K = 206.1

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

CO + H 2 O ⇔ CO 2 + H 2 ΔH 298 K = −41.15

The assumption of a natural gas composed by only methane is made. However, usually other hydrocarbon (ethane, propane, etc.) are cracked as the feedstock goes into the reactor. The first reaction is the steam reforming reaction (SR), which is an endothermic reaction, thermodynamically promoted by high temperatures and by low pressures. The second reaction is the water-gas shift reaction (WGS), an exothermic reaction favoured at low temperature and not affected by operating pressure. Globally the steam reforming process should be supported by high temperatures and low pressures. However, usually the reactions are conducted at 25-40 bar in industrial plants, in order to reduce the total volume of the devices and to favour heat transfer. High heat duty is required by the process since high methane conversion (90%) are reached at temperature within the range 850-950°C. Two configurations are possible to supply reaction heat duty: 1. Tubular fired reforming; 2. Autothermal reforming.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

M. De Falco, G. Iaquaniello, B. Cucchiella et al.

266

In the first process scheme, the heat is supplied by an external source, usually by natural gas burners. The catalytic tubular reformers are placed in a furnace and gas mixture is heated up by a heat flux from hot tube wall to packed bed. The heat transport mechanism can be divided in the following contributions: • • • • •

Radiant heat exchange between the burners flames/combustion gas and the external tube wall; Convective contribution from the combustion gas to the external tube wall; Conduction from the external to the internal tube wall; Convection/Conduction from the internal tube wall to the first reactor gas mixture layer near the wall; Heat transport (convection, conduction and thermal radiation) in the packed bed.

The heat flux has to overcome a series of thermal resistance and axial and radial temperature gradients are generated. The strategy to supply heat to the packed bed is completely different for autothermal reformer scheme: a part of the natural gas feedstock is directly burned in the first section of the reactor by adding oxygen or enriched air. Therefore, the heat duty is not supplied by external source. The reformer is composed by 2 sections: first one is the burner where methane and water steam are mixed with oxygen and the combustion of a part of the inlet methane allows the high temperature needed by the endothermic reaction to be achieved; in the second part of the reformer the catalytic bed is packed and the reactions take place. The autothermal process is competitive with the tubular fired reforming only if oxygen or enriched air are available at low cost: in fact, the use of natural air for the methane combustion would not allow to reach the high temperature required. In the present work, only the first process scheme (tubular fired reforming) is considered.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

1.1.1. NG Steam reforming process flow diagram A typical flowsheet of the tubular fired steam reforming process is shown in Figure 1.26a meanwhile figure 1.3b shows a large size plant with the typical top-firing reformer. Natural gas is mixed with steam and recycled hydrogen before entering the reformer reactor. The recycle of some of the H2 produced to feedstock is necessary to keep the catalyst in the reduced (active) state. Reactions (1.1) and (1.2) occur in parallel in the steam reformer: the high temperature of the reactor, placed in a furnace, supports the steam reforming reaction to the detriment of the WGS. The hot syngas produced, composed by CH4, H2O, H2, CO and CO2, is used to generate high pressure (HP) steam for mixing with the feed (internal use) and for “export” outside the unit. In the design and operation of reformers, generation of HP steam for export is given almost the same importance in industry as the H2 production.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Reformer and Membrane Modules Plant to Optimize Natural…

267

Figure 1.26a. NG steam reforming process flow diagram

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 1.3b. Large size plant with the typical top-firing reformer

The cooled syngas is sent to a 2 steps water gas shift converter: in the first step (high temperature) the operating temperature is about 400-450°C, in order to support the reaction kinetics over the iron oxides catalyst; the second step (low temperature) is carried out at 200250°C in order to promote the exothermic reaction thermodynamic. At the exit of the WGS converter, the CO content is 1% about. The H2-rich exit stream from the shift converter is cooled to condensate the steam, producing the water recycled into the water boilers. Then, the H2 is separated from the off-gases (CH4, CO, CO2) in a pressure swing adsorption (PSA) section, which involves the adsorption of impurities onto a fixed bed of adsorbents at high pressure. The impurities are subsequently desorbed at low pressure into an off-gas stream. This operation allows an extremely pure hydrogen to be obtained: H2 purities in excess of 99.999% can be achieved.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

M. De Falco, G. Iaquaniello, B. Cucchiella et al.

268

The off-gases, mixed with an additional fuel from the external, are sent to the burners to supply the process heat duty. CO2 sequestration is required for environmental matter and a MDEA (Methyl-DiEthanol-Amine) unit has to be inserted to separate the carbon dioxide from the other off-gas mixture components. The global efficiency of the traditional steam reformer plants, calculated as the ratio between the net heat value of the hydrogen stream produced and the total process heat duty requirement (reactor heat duty, steam generation, pre-heating of the reactant mixture, PSA, MDEA) is typically within the range 65 – 85%, depending on the size of the plant.

1.1.2. Hydrogen cost Figures 1.4-7 reports hydrogen cost (US$/GJ) versus feedstock cost for NG steam reforming, coal gasification, biomass gasification and water electrolysis. The two solid lines of each graph represent the hydrogen cost including or not the carbon cost. It has to be noticed that: • •



NG steam reforming produces the most cost-competitive hydrogen if the natural gas price is not too high (5 US$/GJ about); the cost of hydrogen produced by NG steam reforming is much dependent on the feedstock price, since the NG is both the reactant and the fuel of the process (see paragraph 1.3); obviously, a carbon tax will influence the cost of the hydrogen produced using NG and carbon.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

A prospective study conducted by IEA [1] about H2 production cost in the 2020-2030 is reported in Figure 1.8: it is foreseen that the steam methane reforming (SMR) will be costcompetitive both in large and small-scale.

Figure 1.4. Cost of steam reforming H2 production

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Reformer and Membrane Modules Plant to Optimize Natural…

Figure 1.5. Cost of carbon gasification H2 production

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 1.6. Cost of biomass gasification H2 production

Figure 1.7. Cost of electrolytic H2 production

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

269

270

M. De Falco, G. Iaquaniello, B. Cucchiella et al.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 1.8. Projections of H2 production cost (US$/GJ) for 2020-2030 [1]

Figure 1.9. Coal and NG price, historical and projected

Therefore, the NG steam reforming seems to be the most promising process to produce large amounts of hydrogen at competitive cost and to promote the hydrogen technologies in the next years. However, an uncertainty comes from the natural gas feedstock, since natural gas price is very volatile (spot prices can double or triple in a short period of time) and strongly influences the total production cost. Although the Department of Energy’s Energy Information Agency forecasts natural gas prices to rise slowly through 2025 (see Figure 1.9), the future is even less certain after 2025.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Reformer and Membrane Modules Plant to Optimize Natural…

271

It’s self-evident that the strongest technological effort to improve the NG steam reforming performance has to be directed towards the reduction of the hydrogen cost dependence on the NG price, i.e. towards improving the global process energy efficiency in order to reduce the NG needed. The technology proposed in this work should lead to an improvement of the process efficiency and a reduction of the operating temperature, which could allow different thermal sources to be used in the place of methane burners.

1.2. Membrane Steam Reforming Concept

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

As said above, steam reforming is a strongly endothermic reaction very fast over Nibased catalyst, so that equilibrium conditions are quickly reached; therefore, a significant hydrogen yield is achieved only at high temperatures (850–900°C) and the process requires a high heat flux towards the reactor, which can be supplied only by a furnace. A part of methane feedstock has to be burned in the furnace, reducing the process global efficiency, increasing the greenhouse gas (GHG) emissions and strengthening the dependence of hydrogen cost on the natural gas cost. Moreover, the need of inserting the reformer in a furnace leads to difficulties in scaling-down steam reforming plants, so that the efficiency of small-scale plants is much lower, leading to a centralization of the hydrogen production. The integration of hydrogen selective membranes inside the steam reforming reaction environment appears to be a promising way to enhance hydrogen yield at lower temperatures, because the selective removal of hydrogen from the reaction zone enables the equilibrium conversion to be not achieved. Therefore, the continuous removal of the hydrogen from reaction environment should allow the reaction to be supported at lower temperature. The selective membrane can be integrated in steam reforming process by means of two potential configurations: 1. the hydrogen selective membrane is assembled in separation modules applied downstream to reaction units. This configuration, called Reformer and Membrane Module (RMM), is composed by a series of reaction-separation modules: the reformer feed is sent to a convective steam reformer where it is partially converted into hydrogen; then hydrogen is recovered through a Pd alloy membrane separation module, while the retentate is sent to the next step or recycled to the first module. The operating temperature can be reduced thanks to the high temperature (450°C about) hydrogen separation between reaction steps, and it is possible to replicate the RMM until the desired natural gas conversion is achieved. Figure 1.10 reports a conceptual layout of the process scheme. 2. The selective membrane is assembled directly inside the reaction environment, so that the hydrogen produced by the reactions is immediately removed. A draft of the membrane reactor (MR) is shown in figure 1.11: the steam reformer is composed by two concentric tubes, the catalyst pellets are packed in the annular zone (reaction zone), while the membrane is the internal tube itself (permeation zone). The heat flux is supplied from the external and a sweeping gas is sent in the inner section to carry out the hydrogen permeated: the sweeping gas could be water vapour to make easier the recovery of hydrogen at the reactor outlet (by a simple condensation process).

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

272

M. De Falco, G. Iaquaniello, B. Cucchiella et al.

Figure 1.10. Two Reformer and Membrane Modules process scheme

Figure 1.11. Steam reforming membrane reactor draft

Recently, several papers have been devoted to experimental studies on membrane reactors performance with various feedstocks (methane, ethanol, methanol) [2-10] and on RMM configuration [11, 12] with very promising results.

1.2.1. Selective membrane integration benefits

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

The application of hydrogen selective membranes in steam reforming process leads to the following main benefits: Strong reduction of the reaction temperature. Avoiding the equilibrium condition to be achieved allows the promotion of the reactions at lower temperatures (500650°C); due to the lower thermal level involved, a heat exchanger with a heating fluid could be used in the place of the furnace, with the following advantages: a. efficiency of the heat transfer from the external source to the reactor; b. lower exergy of the heating fluid in comparison with the high temperature combustion gas used in the furnace, which means a lower heating cost; c. the possibility to use different heating fluids, depending on their availability; d. simple integrability of the reformer in industrial plants, wherever a hot fluid is available or thermal recovery is sought; a high value added fuel as hydrogen is produced (plug-in and retrofit concepts);

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Reformer and Membrane Modules Plant to Optimize Natural… e.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

f.

273

the easy scalability (scale-up or scale-down) of the system and therefore its applicability in many fields (small-medium-large scale); use of cheaper alloy steel, since the tubular reactor is less stressed.

Process efficiency increase. The lower temperature results in an increase of overall process efficiency, since the heat supplied is better exploited. It’s foreseen that the global process efficiency should increase from the 65-80% of today’s technology up to 85% and more for all the plant sizes. Large methane saving. Reduction of reaction temperature means that the heat duty requirement is lower than for the traditional process. The heat flux from the external source to the catalytic bed should be about 30-40 kW/m2 instead of 80 kW/m2 and more of the traditional process [13]. Therefore, a smaller amount of methane has to be burned. The lower thermal level could allow the coupling of the reformer with a different and clean energy source, i.e. solar energy, nullifying the fraction of natural gas to be burned for process heat requirements. Reduction of CO2 emissions. The methane saving leads to a reduction of the greenhouse gas emissions, since less or no carbon dioxide is produced by the methane combustion. In a traditional process, the ratio (CO2 released)/(H2 produced) is 8 - 12 kgCO2/kgH2, depending on the process efficiency. An increase of the efficiency could lead to a reduction of GHG emissions within the range 20-55%, up to 5.5 kgCO2/kgH2 if a renewable energy source is used for process heat duty. Easier CO2 purification. Membrane integration into the reaction environment ensures a first substantial hydrogen separation step (up to 90% of the hydrogen produced can be removed); as for CO2 separation, because of the higher carbon dioxide partial pressure in the reformer outlet stream, due to the hydrogen removal, physical separation methods could be used to separate CO2 rather than the chemical adsorption in mono-di-ethanol ammine (MDEA), which is a very expensive separation process. Reduced dependence on natural gas cost. Increasing the reaction efficiency and reducing the amounts of methane to be burned to supply the process heat duty requirements lead to a reduction of the total amount of methane required for producing a mass unit of hydrogen. Therefore, although a higher plant cost has to be supported because of the increasing reactor complexity, the hydrogen price would be less dependent on the natural gas. Obviously, the total hydrogen cost will be influenced on the Pd-based membranes, which are nowadays expensive. However, a large industrial production of this component surely will reduce the price, making the membrane steam reforming process more competitive.

1.2.2. RMM and MR benefits and drawbacks Table 1.1 reports the main benefits and drawbacks of both the membrane steam reforming process configurations assessed. In the next paragraph, the technological state-of-the-art of hydrogen selective membranes is reported. Then the performance of RMM and MR configurations are assessed via simulations and compared and economical issues are tackled.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

M. De Falco, G. Iaquaniello, B. Cucchiella et al.

274

Table 1.12 – Benefits and drawbacks for RMM and MR configurations

RMM configuration

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

MR configuration

Benefits -) Possibility to decouple reaction and separation operating conditions: the reforming and separation module temperatures can be optimized independently, both increasing methane conversion for each reaction step and membrane stability and lifetime. -) Simplification of the mechanical design of membrane tubes relative to those embedded in catalyst tubes; a simple “shell and tube” geometry can be selected for the tubular separation module. -) Simplification of membrane modules maintenance and of catalyst replacement. -) Compactness of the process: the integration in a single device of steam reforming reaction and of hydrogen separation step allows a more compact reforming plant. -) Easiness in scalability: scale-up and scale-down of the membrane reformer are very easy through an increase or a decrease of the number of parallel tubular reactors or of the length of the single membrane reformer. -) Less useless catalyst: in the traditional process the catalyst pellets placed in the central zone of reformers usually don’t work since the temperature is too low for promoting the reactions for the large radial temperature gradient. In membrane reformers, the central zone of the reactor does not contain catalyst but the membrane tube devoted to collect the hydrogen permeated.

Drawbacks

-) Compactness of the process: RMM configuration is composed by reactors, separation modules, heat exchangers. -) High membrane surfaces required. -) High cost, mainly for selective membranes surface required.

-) Technological problem in designing the reactor: a sensible component as the selective membrane has to be inserted in a critical environment. -) Coupling between reaction and separation operating conditions. -) No easy maintenance of the reactor.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Reformer and Membrane Modules Plant to Optimize Natural…

275

2. HYDROGEN SELECTIVE MEMBRANES STATE-OF-THE-ART 2.1 Introduction Different types of hydrogen selective membranes (table 2.1) are actually available for hydrogen separation from gaseous mixtures. Each membrane has its own operating ranges, in terms of temperatures and flow compositions. The properties of the flow to be separated are therefore a starting point to select a suitable membrane. Table 2.1. hydrogen separation membranes

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Dense polymer Temperature range H2 selectivity H2 flux (103 mol/m2s) at ∆p = 0.1 bar Materials

< 100°C

Micro porous ceramic 200-600°C

Dense metallic

Dense ceramic

300-600°C

600-900°C

low

5-139

>1000

>1000

low

60-300

60-300

6-80

polymers

Palladium alloy

Proton conducting ceramics

Transport mechanism

solution/diffusion

solution/diffusion

solution/diffusion (proton conduction)

Development status

commercial by Air Products, Linde, BOC, Air Liquide

silica zirconia alumina titania molecular sieving and/or Knudsen diffusion prototype tubular silica membranes available

commercial by Johnson Matthey, MRT purifiers

small samples available for testing from CanMet

prototype tubular Pd alloy membranes available for testing from ECN, NGK, Acktar

In general, three membrane types that differ on the basis of structure and hydrogen transport mechanisms can be identified: dense, porous or composite membranes. Dense membranes are membranes without pores of microscopic dimensions and mass transport occurs with a solution-diffusion mechanism. With this type of membrane it is possible to obtain high selectivity but limited flows. Differently, transport of the substances through porous membranes, that are membranes with pores of large dimensions compared to molecular dimensions (0.005-20 μm), occurs

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

276

M. De Falco, G. Iaquaniello, B. Cucchiella et al.

principally in the pores and selectivity is determined by the mutual dimensions of the species and the pores. In composite membranes, formed of a thin layer of selective material covering a thicker layer which lends the structure the necessary mechanical resistance, H2 is separated by the selective layer usually composed of a Pd alloy whereas the support can be either dense or porous.

2.2. Membranes Separation Mechanisms Porous membrane separation mechanism: Knudsen diffusion Knudsen diffusion is a free molecular diffusion. Diffusion happens for high Knudsen number (Kn), a dimensionless number defined as the ratio of the mean free path of the gas molecules λ (average distance between collisions) and a representative physical length scale L (e.g., the pore radius).

If the pore radius is used as representative physical length scale the mean free path lengths are substantially higher than pore radiuses at Knudsen numbers larger than 10. The result is that mainly the lighter molecules permeate through the pores and selectivity of the membrane grew up. Otherwise with Knudsen numbers < 1 the dominant transport mechanism is viscous flow, which is non-selective. In some cases the molecules move using different combined methods (a mechanisms scheme is in Figure 2.1):

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

- Knudsen diffusion - surface diffusion - capillary condensation - molecular sieving Surface diffusion (see Figure 2.1 b) can occur in parallel with Knudsen diffusion. Gas molecules are adsorbed on the pore walls of the membrane and migrate along the surface reducing the effective pore diameter and increasing selectivity. Capillary condensation (see Figure 2.1 c) occurs if a condensed phase (partially) fills the membrane pores. When the pores are completely filled with condensed phase, only the species soluble in the condensed phase can permeate through the membrane. If pore sizes become sufficiently small (3.0-5.2 A) molecules that differ in kinetic diameter are separated by molecular sieving (see Figure 2.1 d)

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Reformer and Membrane Modules Plant to Optimize Natural…

277

Dense membrane separation mechanism: solution/diffusion mechanism The solution-diffusion mechanism is the physical model that describes gas transport through dense membranes.

Figure 2.1. transport mechanism in porous membranes: a) Knudsen diffusion, b) surface diffusion, c) capillary condensation, d) molecular sieving

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

It is a complex process consisting of adsorption and dissociation of molecules, followed by diffusion of atoms through the membrane material lattice, recombination of atoms at the low pressure side and desorption of molecules, according to the following 7-steps activated mechanism: 1. 2. 3. 4. 5. 6.

adsorption of molecules on the high pressure side of the membrane surface; dissociation of molecules on the same surface; dissociation of the chemisorbed molecules into atoms species; dissolution of the atoms into the lattice of the membrane material; diffusion of the atoms through the lattice; re-combination of protons and electrons and re-association of atoms with formation of molecules at the low pressure membrane surface; 7. desorption of molecules from the low pressure membrane side to the bulk.

Hydrogen Transport through Composite Pd-Based Membranes Hydrogen transport through Pd-based membranes is correctly expressed by the Sieverts’ law (eq. 2.1) which results from hydrogen transport mechanism through the Pd layer:

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

278

M. De Falco, G. Iaquaniello, B. Cucchiella et al.

hydrogen adsorption and dissociation onto Pd surface and subsequent atomic hydrogen diffusion in the metallic lattice.

N H2

(

0.5 0.5 DH K S pH 2 Ret − pH 2 Per = 2 δ

)

(2.1)

In Sieverts’ law NH2 represents the H2 flux, DH is the average diffusivity of atomic hydrogen in the Pd-Ag alloy, pH2 is the partial pressure of hydrogen at the retentate (Ret) or permeate (Per) side, δ is the membrane thickness and KS is the Sievert’s constant which represents the equilibrium constant of the adsorption dissociation reaction considered above. Consistently with the solution diffusion model for transport in dense membrane the hydrogen permeability, PH 2 , [mol.m-1.s-1.bar-0.5

or Nl.m-1.s-1.bar-0.5], and the hydrogen

permeance, K H 2 , [mol.m-2.s-1.bar-0.5 or Nl.m-2.s-1.bar-0.5], through the Pd based membrane can be defined as : PH 2 =

N H2 δ DH K S = 0.5 2 pH 2 Ret − pH0.52 Per

K H2 =

(

PH 2

δ

=

(p

N H2 0.5 H 2 Ret

− pH0.52 Per

)

(2.2)

(2.3)

)

In order to evaluate the real resistance to transport of the composite membrane, the support has to be considered apart from the dense Pd layer. In the porous support gas can move through Knudsen diffusion, viscous flow or a combination of the two, depending on the average pore dimensions, and to resistance in the gas phase. In many cases, due to the different driving pressure dependence of the transport mechanisms considered, the presence of such additional resistances can be considered modifying Sievert’s law as follows:

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

N H2 = P H2

(p

n H 2 Ret

− pHn 2 Per

)

δ

(2.4)

where n is an empirical exponent which usually takes values between 0.5 and 1 depending on the dominant resistance in the system.

2.3. Hydrogen Selective Membranes Stability Ceramic membranes The thermal and hydrothermal stability (hydrothermal conditions) of silica membranes has been analysed by Nijmeijer [14] for steam reforming process. The objective of the work was the development of a multi-layered ceramic membrane with a flow of H2 equal to 1*10-5

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Reformer and Membrane Modules Plant to Optimize Natural…

279

mol/m2.s.Pa and with a separation factor > 50 regarding other gases (CH4, CO and CO2), that could operate for 1000 hours at the temperature of 600°C in steam reforming simulating atmosphere: total pressure 30 bars, CH4:H2O = 1:3. The greater expected problem was the stability in steam atmosphere. The author points out that observing selectivity trend with time, it is possible to estimate also the membrane stability. However the work of Nijmeijer doesn’t introduce tests of exposure in steam reforming simulating conditions. The author asserts that "especially the development of steam stable membranes may be to large step forward in the development of ceramic membranes" and that improvements will have to regard the support material of the membrane as well as the adherence between various layers of the membrane. Others important factors of destabilization of silica membranes, that have to be consider in adding to the humid atmosphere, are: • •

alterations caused by carbonic depositing (Coking) alterations caused by reactions with components of the catalyst

For the author a program of stability tests would have to consider the effects of: • • • •

Temperature Pressure Steam/methane ratio Hydrogen concentration

The attended problems in the ceramic membrane fabrication are connected with the insufficient adhesion, or debonding, between the layers of the structure that, in the extreme case, can carry to the demolition of the membrane with a pathological increase of permeability to the detriment of selectivity.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Pd membranes The main technological problem in the use of massive palladium membranes seems to be the insufficient mechanical stability, consequence of structural modifications of the reticulum of Palladium due to the presence of one gap of miscibility between two crystallographic phases (α and β) of Pd (diagram of phase H-Pd DOE 1989, Johnson Matthey 2004); the region where α and β phases are in equilibrium extends around the atom ratio H/Pd = 0,27 and has critical temperature to approximately 549 K, as shown in figure 2.3.1 (DOE 1989). The β phase molar volume is greater than that of α phase and, during hydrogen absorption on palladium, the β phase is compressed and α phase suffers plastic deformation; the desorption of H(D) leads to the transformation from β to α phase and therefore stresses of traction in the structure and fractures of the reticulum. Analogously, tensions and fractures in the structure can be caused by the separation of the two phases when lessening of the temperature happens, entering in the two-phase field. Formation of these phases is the natural consequence of the solubility of hydrogen in palladium, solubility that, in its turn, assures the permeation of the metal towards hydrogen; therefore starting phases of the membrane reactor (when loading of Palladium with hydrogen

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

M. De Falco, G. Iaquaniello, B. Cucchiella et al.

280

happens) and interruption of the operations (when desorption of hydrogen occurs) represent two operations potentially very critic for integrity of the membrane. This leads to the necessity to extract hydrogen from the reticulum of palladium (extraction in inert gas stream) before the cooling of the membrane [15].

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 2.2 H – Pd hydrogen phase diagram

Figure 2.3 – Hydrogen permeability in Pd-alloy membranes at 350°C and 2.2 MPa

The α-β transition phase can be suppressed drugging palladium with other metals. Alloys widely used for the hydrogen selective membranes are: • •

Ag 20 – 30 wt%, surely the most used; Cu 40%.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Reformer and Membrane Modules Plant to Optimize Natural…

281

Figure 2.3 shows the hydrogen permeability through some Pd-alloy membranes, based on the weight percentage of the added component. Some experimental results on Pd-Cu alloy membranes (temperatures of test between 623 and 1173 K) are reported by Howard et al. [16]. Authors describe the effects of phase transformations due to thermal cycles on permeability. It’s interesting that also Pd-Cu alloys show biphasic structures with miscibility gaps and therefore they can suffer the same effects described for Pd-H system. However the chemical stability of Palladium membranes constitutes a crucial problem; they seem much sensitive towards sulphur compounds [17], towards chlorine and CO [18]. It’s possible that such problems can be reduced choosing the suited palladium alloy each time.

2.4. Pd-Alloy Membranes Synthesis Methods Up to now many methods have been developed to deposit Pd or Pd alloy selective layers on porous supports to form membranes, including:

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

physical vapor deposition (PVD) electroless plating electrodeposition chemical vapor deposition (CVD) micro-emulsion pyrolysis pore-plugging by liquid impregnation sputtering. Among the mentioned techniques, the most used methods are electroless plating, CVD and sputtering. Under properly controlled conditions all three methods produce good quality thin Pd/Ag membranes, with hydrogen to nitrogen selectivity over 3000 at temperatures above 300 °C. The two chemical methods (electroless plating and CVD) have the advantage of easy scale-up and the flexibility to coat the metal film on supports of different geometry. However, the main disadvantage is the difficulty to control the composition of the alloy. Sputtering has several advantages like: 1. 2. 3. 4.

synthesis of ultrathin films with minimal impurity; easily controllable process parameters; flexibility for synthesizing alloys; the ability to generate nanostructured films.

The last two points are very important in membrane preparation for hydrogen separation because fabricating membrane alloys helps to overcome the problem of hydrogen embrittlement, while the nanostructured films may have unique size-dependent properties, e.g. high hydrogen permeation.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

M. De Falco, G. Iaquaniello, B. Cucchiella et al.

282

Table 2.2. Hydrogen selective membranes fabrication methods Deposition method

Membrane

Temperature (°C)

CVD (TEOS) CVD (Pd) Electroless PVD PVD

Al2O3 Pd/ Al2O3 Pd/SS Pd-Ag/ Al2O3 Pd-Ag/ γ-Al2O3

600 500 450 300 150-200

H2 Permeance 10-8 mol m-2 s1 Pa-1 2.2 300 250 0.1-0.2 3

H2 Selectivity

Reference

H2/N2 =1000 H2/N2>500 H2/N2 =3000 H2/He=20 H2/He=4000

[19] [20] [21] [22] [23]

Due to the mentioned advantages, sputtering (PVD) has frequently been used to fabricate (sub) micron thick Pd– Ag alloy layers.

Electroless Deposition

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Electroless deposition is a technique based on the controlled autocatalytic reduction of a dissolved metallic salt by reducing agents at a substrate interface. Among other techniques, electroless deposition provides strong advantages such as uniformity of deposits even on very complex shapes, very simple equipment and low cost. Either non-conductive surfaces or conductive surfaces can be coated easily by using electroless deposition. The deposit follows exactly all the contours of the work-piece without build-up at edges or corners. Many factors including the support quality, activation process and the bath chemistry, contribute to the thickness, stability and selectivity of the membranes. In addition, researchers also tried to modify the plating procedures to make thinner membranes. Co-deposition behavior of palladium and silver on porous stainless steel (PSS) was first investigated by Shu et al. [24]. EDX and XRD analysis were performed to determine the composition of the deposited layer as a function of Ag content of the plating bath, keeping the same total metal concentration. The baths containing more than 20 at.% of Ag resulted in the preferential deposition of silver, which passivated further Pd deposition. An effective activation of the substrate with Pd pre-deposition was developed to reduce the Ag enrichment of the support. No permeation data or thickness was reported. Table 13.3. Standard composition of hydrazine-based Pd/Ag plating bath (Cheng, Yeung 1999).

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Reformer and Membrane Modules Plant to Optimize Natural…

283

Later, Cheng and Yeung [25] investigated the plating kinetics, film microstructure and composition during co-deposition of Pd and Ag on Vycor glass and developed a mathematical model to estimate the plating rate, thickness, and composition of the layer as functions of reaction time and concentrations of the reacting species. Plating solutions with higher hydrazine to metal ratios yielded membranes with higher palladium content. Increasing the plating temperature also leads to higher palladium composition. The average plating rate decreases with the higher Ag content confirming that the presence of silver inhibits the Pd deposition. Most recently, Akis [26] investigated the effect of sequence of Ag electroless deposition and annealing on the performance of composite Pd/Ag alloy/porous stainless steel membranes. Two Pd/Ag membranes were synthesized by using different procedures. The first membrane was prepared by sequential Pd and Ag plating with intermediate annealing after each Ag deposition (coating and diffusion technique). The resulting membrane was 42 µm thick and had 18 wt% Ag. The second membrane consisted of several sandwiched layers of Pd and Ag with no intermediate heat treatment. The membrane was finally annealed after it was dense. The final thickness of the membrane was about 15 µm and contained around 10 wt % Ag. The practical application for these membranes was illustrated by the hydrogen permeation data. The increase in the He flux of Pd-Ag second membrane for both temperatures 350500°C, corresponds to a leak development during annealing, yielding a very low H2/He selectivity due to the formation of pinholes or cracks. On the other hand, the first Pd-Ag membrane had a stable hydrogen permeance of 6 m3/m2h atm0.5 at 500°C and showed a complete selectivity H2/He.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Chemical Vapor Deposition (CVD) Chemical vapor deposition (CVD) results from the chemical reaction of gaseous precursor(s) at a heated substrate to yield a fully dense deposit. Gaseous compounds of the materials are transported to a substrate where the thermal reaction/deposition occurs. The CVD process allows depositing either metallic or dielectric materials in submicron scale but it’s not easy to control alloys composition. By definition, metal-organic chemical vapor deposition (MOCVD) differs from the conventional CVD in its precursor used, wich are restricted in organometallic compounds. The deposition of Pd and Pd-Ag membranes with CVD of PdCl2 on a porous support was first studied by Xomeritakis et al. [22-23] incorporaring Ag sputtering. The gas transport properties of the thin metallic membranes were determined by multicomponent permeation experiments with He, H2 and Ar at 25–300°C and 1 atm total pressure. The H2 permeance was in the range 1.0–2.0 × 10−7 mol m−2 s−1 Pa−1 meanwhile H2: He selectivity and 30–200 at 300°C. The selectivity was fairly low due to the difficulties in the adherence with the ceramic substrate. Jun and Lee [27] prepared thin Pd and Pd–Ni alloy membranes by the MOCVD. Pd(C3H5)(C5H5) and Ni(C3H5)(C5H5) were decomposed into densely aggregated metal crystallites that were to plug mesoporous nickel-stainless steel (Ni-SUS) or g-Al2O3/a-Al2O3

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

284

M. De Falco, G. Iaquaniello, B. Cucchiella et al.

supports. The use of highly volatile organometallic precursors enabled continuous and controlled deposition mode, which resulted in reproducible formation of thin impervious Pd or Pd–Ni alloy membranes. The H2 permeance of the Pd/Ni-SUS membrane was 2.0–5.0_10−2 cm3 cm−2 cmHg−1 s−1 (723 K); the H2/N2 selectivity was 1600. The H2 permeance of the Pd/Al2O3 was 1.5_10−2 cm3 cm−2 cmHg−1 s−1 (723 K); the H2/N2 selectivity was >1000. The membranes showed a continuos hydrogen flux reduction at 500°C due to the intermetallic diffusion.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Physical Vapor Deposition (PVD) PVD is the process of depositing solid material on a substrate from vapor phase. The source of the vapor phase is generally a solid “chunk” (target) of the desired film material. The desired material is either heated until evaporation (thermal evaporation) or sputtered by ions (sputtering) on the target surface. During sputtering, a low pressure gas with large number of ions and free electrons, referred as plasma, is created via high energy field. Bombardment of solid (target) by high energy chemically inert (e.g. Ar) ions extracted from plasma causes ejection of atoms from the target which are then re-deposited on the surface of the substrate with the help of magnetic field generated by permanent magnets. During deposition the substrate is already heated enhancing the inter diffusion of metal ions, which are in close contact. Therefore no heat treatment or low annealing temperatures is needed for alloying. While sputtering an alloy film from a corresponding target, in the beginning there is a tendency to eject one atomic species in excess of the other species. But after sometime, the atoms are ejected in the same ratio as the original ratio in the target. The time required for this process is called the equilibration time. Jayaraman and Lin [28] observed that substrate temperature strongly affected the membrane gas tightness. Increasing the substrate temperature provided energy to the arriving metal atoms to orient in a certain lattice position on the growing film. Zhao et al. [29] made similar studies on the influence of the substrate temperature on the compactness of the Pd-Ag membranes. The columnar structures of the Pd-Ag alloy films were gradually modified and disappeared when the substrate temperature was higher than 350°C. At low substrate temperatures, higher points on a growing surface received more vapor atoms than valleys, resulting in a columnar structure, i.e. shadow effect. With increasing substrate temperature, diffusion of metal atoms became increasingly stronger leading to more compact Pd-Ag alloy composite membranes. In 2005 Tong and Berg [30] realized a thin Pd-Ag membrane by dual-sputtering technique. The deposition parameters listed in the following table allowed to deposit a very thick film with low roughness:

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Reformer and Membrane Modules Plant to Optimize Natural…

285

3. REFORMERS AND MEMBRANES PLANT CONFIGURATION 3.1. Process Configuration As described in paragraph 1, the selective membrane can be integrated in steam reforming process by means of two potential configurations: the hydrogen selective membrane can be assembled in separation modules applied downstream to reaction units (Reformer and Membrane Modules – RMM plant) or can be placed directly inside the reaction environment (Membrane Reactor – MR plant). In the present paragraph, the RMM configuration is analyzed, modeled and assessed via simulations. The process layout is shown in Figure 1.10. The first reactor is fed by a CH4 (for sake of simplicity, it is assumed that natural gas is composed only by methane) and vaporized water stream, which is partially converted in CO, CO2 and hydrogen. Then, the operating temperature is lowered at 450°C and the stream is sent to a separation unit composed by Pdbased membranes modules. The system reactor + separator is a single module. A series of modules is assembled in order to reach the methane conversion or the hydrogen production required. The separation unit is tubes-and-shell shaped: inside the membrane tubes a sweeping gas (steam) is sent to carry out the hydrogen permeated through the selective layer (figure 3.1). In the following, a mathematical model of the RMM plant is proposed and explained. Then, the behaviour of the system varying the number of RMMs assembled in the plant and the operating conditions as reaction temperature, pressure and feedstock composition is evaluated.

3.2. Mathematical Modelling The single Reformer + Separator step is modeled by means of mass, energy and momentum balances, which leads to a PDEs set. The reaction scheme is composed by:

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

CH 4 + H 2O ⇔ CO + 3H 2

Figure 3.1. Tubes-and shell membrane separator

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

(3.1)

M. De Falco, G. Iaquaniello, B. Cucchiella et al.

286

CO + H 2O ⇔ CO2 + H 2

(3.2)

CH 4 + 2 H 2O ⇔ CO2 + 4 H 2

(3.3)

Secondary reactions, as carbon coke formation, are neglected. The model is based on the following assumptions: •

.For the reformer model:

steady-state conditions; negligible axial dispersion and radial convection; ideal gas behavior; a single tube representative of any other tube; pseudo-effectiveness factors η1, η2 and η3 for the reactions (3.1-3.3) independent of local conditions and fixed at 0.02 as an average value of those reported in the literature [31, 32]; •

For the separator model:

steady-state conditions; plug flow hypothesis, leading to a one-dimensional nature of the model; ideal gas behavior; a single tube representative of any other tube; negligible pressure drop; co-current configuration between process stream and sweeping gas; hydrogen selectivity assumed infinite. In the following, the equations composing the model are reported.

REFORMER MODEL

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Mass balances dp ⋅L ∂ (u~z c~i ) ∂ 2 (u~z c~i ) 1 ∂ (u~z c~i ) η ⋅ ρ cat ⋅ (1 − ε ) ⋅ L )− ⋅ ri = ⋅( ~2 + ~ ⋅ 2 z r r u z , 0 cCH 4 , 0 ∂~ ∂~ ∂r Pemr ⋅ Ri

(3.4)

i = CH 4 , H 2O, H 2 , CO, CO2

z and ~ r are the where R i and L are the internal radius and the length of the reformer, ~

~ and u axial and radial dimensionless coordinates, u z z , 0 are the dimensionless gas velocity

~ and c and its inlet value, c i CH 4 , 0 the dimensionless component concentration and the inlet

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Reformer and Membrane Modules Plant to Optimize Natural…

287

methane concentration, r i is the reaction rate of the component i according to Xu-Froment kinetics model [33], η is the effectiveness factor,

ρ cat is the catalyst density , ε is the bed

void fraction, d p is the catalyst particle diameter.

Pe mr =

uz d p Der

is the mass effective radial Pèclet number calculated according to [34].

Energy balance 3

~ ~ ~ λer ⋅ L ∂TR ∂ 2TR 1 ∂TR = ( ⋅ + ⋅ )+ 2 ∂~z ∂~ r2 ~ r ∂~ r (u z ctot ) ⋅ c p , mix ⋅ Ri

ρ cat ⋅ (1 − ε ) ⋅ L ⋅ ∑η j ⋅ (− ΔH j ) ⋅ r j j =1

(u z ctot ) ⋅ c p , mix ⋅ TR , 0

(3.5)

~

where T R and T R , 0 are the dimensionless and the inlet reactor temperature, c p , mix is the gas mixture specific heat,

η j , (−ΔH j) and r j are the effectiveness factor, the enthalpy

and rate of the reaction j . The effective radial thermal conductivity

λ er is calculated according to [35] that,

although concerning pseudo-homogeneous models, results widely suitable for the simulation of packed bed reactor heat transport.

Momentum balance

~

where P

~ 2 dP f ⋅ G ⋅ μ g ⋅ L (1 − ε ) ⋅ = d~ z ε3 ρ g ⋅ d p 2 ⋅ P0

(3.6)

and P 0 are the dimensionless and the inlet pressure in the reaction zone. The

friction factor f is calculated by the well-known Ergun equation. The one-dimensional

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

nature of the momentum balance depends on plug-flow assumption that imposes the same gas velocity in every point of the reactor section.

SEPARATOR MODEL Only mass balance is applied for modeling of membrane separator, since there are not thermal effects and pressure drop can be considered negligible.

Mass balances High pressure zone

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

M. De Falco, G. Iaquaniello, B. Cucchiella et al.

288

d (u~z c~H 2 ,up ) 2π ⋅ nmem Rmem ⋅ Lmem = − J H2 ⋅ ~ dz uc

(3.7)

d (u~z c~H 2 , down ) 2π ⋅ nmem Rmem ⋅ Lmem = JH2 ⋅ ~ dz uc

(3.8)

z H 2 ,in

Low pressure zone

z H 2 , in

where nmem , Rmem and Lmem are the number of membranes assembled in the separator unit, the radius and the length of each membrane, cH 2 ,in is the molar concentration of hydrogen in

~ the separator feedstock (equal to the output of reactors), c~H 2 ,up and c H 2 , down are hydrogen

dimensionless concentration in the upstream (high pressure) and downstream (low pressure) mixtures. The term J H 2 is the hydrogen flux permeated through the membrane and calculated according to the Sieverts’ law:

J H2 =

BH

δ

(

⋅ p H0.52 ,up − p H0.52 , down

)

(3.9)

where BH is the hydrogen permeability, calculated according to Shu et al. [36] for Pd-Ag membranes,

δ is the membrane thickness (20 μm in the following simulations), pH

2 , up

and

pH 2 , down are the hydrogen partial pressures in high pressure and low pressure zones, respectively. The boundary conditions imposed are listed below:

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

REFORMER MODEL ~z = 0, ∀~ r

(3.10)

u~z c~CH 4 = 1 for the first reactor

(u z cCH 4 ) sep ,out for other reactors u~z c~CH 4 = u z , 0 cCH 4 , 0 u~z c~i =

u z ci , 0 u z , 0 cCH 4 , 0

( i = H 2 O, H 2 , CO, CO2 ) for the first reactor

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Reformer and Membrane Modules Plant to Optimize Natural…

289

(u z ci ) sep ,out ( i = H 2 O, H 2 , CO, CO2 ) for other reactors u~z c~i = u z , 0 cCH 4 , 0 ~ TR = 1 ~ P =1 ~ r = 1, ∀~ z

(3.11)

∂ (u~z c~i ) =0 ∂~ r ~ ∂TR U ⋅ Ri λer ~ = (TW − TR| R ) i ∂r TR ,0 ~ r = 0, ∀~ z

(3.12)

∂ (u~z c~i ) =0 ∂~ r ~ ∂TR =0 ∂~ r The coefficient U is the overall heat transfer coefficient, calculated according to [35],

TW is the reformer hot wall temperature.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

SEPARATOR MODEL For co-current configuration, the boundary conditions are imposed both on the initial section: ~ z =0 (3.13)

u~z c~H 2 ,up = 1 u~z c~H 2 ,down = 0 The set of PDEs is solved discretizing the radial coordinate by means of central secondorder differences: the resulting ODE set is solved using a Runge-Kutta method with variable step. The hot wall temperature is a modeling parameter and it is assumed equal to heating fluid temperature.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

M. De Falco, G. Iaquaniello, B. Cucchiella et al.

290

For a more detailed analysis of the mathematical model, refer to [37,38]. Table 3.14. Reformer parameters. Length Internal Diameter Inlet temperature

6m 0.11 m 773 K

Pressure Inlet methane flowrate Steam to carbon ratio

15 – 20 – 25 bar 4-8-12 kmol/h 3

TW

773 – 923 K

ε

0.5

Table 3.15 – Separator parameters Length 5m Downstream pressure Diameter 0.04 m Separator temperature Number of membrane 8 δ tubes for each reformer* Sweep steam flowrate 1 kmol/h for each separator * For each reformer a system of 8 permeators tube is assembled downstream.

1 bar 723 K 20 µm

3.3. Plant Performance Simulations are performed fixing geometrical parameters and operating conditions shown in Table 3.14 and 3.15. Reformers sizes are typical of industrial application [39] with the length of the single reformer equal to an half of the typical length of an industrial reformer. Separators sizes are assumed. The total hydrogen production is imposed equal to 10,000 Nm3/h: a single reformer is simulated and then the number of reformers to be assembled in parallel is calculated as 10,000/hydrogen produced. The hot wall temperature TW , the operating inlet reactor pressure and the inlet methane

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

flowrate are varied to assess their effect on plant performance. The outputs are: • •

Total methane conversion Kilograms of catalyst needed for the production of 10,000 Nm3/h of pure hydrogen ( Wcat )



The hydrogen separated for unit of membrane surface ( H 2 / Supmem )

• • • •

Total membrane surface required Total steam generation for reactors and separators (sweeping gas) Total methane feedstock Heat duty required by each reactor.

Simulations are made imposing 2 and 3 RMM steps.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Table 3.3. Plant performance versus hot reformer wall temperature (operating pressure = 20 bar).

RMM steps =2

RMM steps =3

(Nm /m h)

Total membrane surface (m2)

Total steam generation (Nm3/h)

Total methane feedstock (Nm3/h)

13.61

8.3

1205.3

72559

21499

0.26

9.59

11.77

849.5

51141

15153

873

0.34

7.37

15.31

653

39313

11648

923

0.44

6

18.76

533

32089

9508

973

0.55

5.13

22

454.7

27374

8111

773

0.228

14.2

7.94

1258.4

50504

14964

823

0.32

10.13

11.15

897

35996

10666

873

0.42

7.83

14.41

693.6

27836

8247.8

923

0.53

6.42

17.6

568.3

22806

6757.3

973

0.64

5.5

20.56

486.4

19520

5783.8

TW (K)

X CH 4

Wcat (tonn)

773

0.178

823

H 2 / Supmem 3

2

Reactors heat flux (kcal/m2h) Reactor 1: 14462 Reactor 2: 7255 Reactor 1: 29198 Reactor 2: 17167 Reactor 1: 46115 Reactor 2: 27501 Reactor 1: 65403 Reactor 2: 38251 Reactor 1: 87265 Reactor 2: 49396 Reactor 1: 14462 Reactor 2: 7255 Reactor 3: 5983 Reactor 1: 29198 Reactor 2: 17167 Reactor 3: 14851 Reactor 1: 46115 Reactor 2: 27501 Reactor 3: 23598 Reactor 1: 65403 Reactor 2: 38251 Reactor 3: 31970 Reactor 1: 87265 Reactor 2: 49396 Reactor 3: 39716

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Table 3.4 – Plant performance versus operating pressure ( TW = 923 K)

RMM steps =2

RMM steps =3

(Nm /m h)

Total membrane surface (m2)

Total steam generation (Nm3/h)

Total methane feedstock (Nm3/h)

7.5

15

664.4

39999

11852

0.44

6

18.76

533

32089

9508

25

0.42

5.43

20.8

480.9

28950

8578

15

0.54

8.22

13.7

728.5

29237

8663

20

0.53

6.42

17.6

568.3

22806

6757.3

25

0.52

5.77

19.5

511.6

20532

6083.6

P (bar)

X CH 4

Wcat (tonn)

15

0.47

20

H 2 / Supmem 3

2

Reactor heat duty (kcal/m2h) Reactor 1: 69773 Reactor 2: 38133 Reactor 1: 65403 Reactor 2: 38251 Reactor 1: 62389 Reactor 2: 38596 Reactor 1: 69773 Reactor 2: 38133 Reactor 3: 29849 Reactor 1: 65403 Reactor 2: 38251 Reactor 3: 31970 Reactor 1: 62389 Reactor 2: 38596 Reactor 3: 33260

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Table 3.5 - Plant performance versus inlet methane flowrate ( TW = 923 K, operating pressure = 20 bar)

RMM steps =2

RMM steps =3

(Nm /m h)

Total membrane surface (m2)

Total steam generation (Nm3/h)

Total methane feedstock (Nm3/h)

7.36

15.34

651.7

21796

5812.4

0.44

6

18.76

533

32089

9508

12

0.414

6.88

16.4

609.9

53034

16318

4

0.64

8.18

13.8

724.5

16153

4307.5

8

0.53

6.42

17.6

568.3

22806

6757.3

12

0.47

7.7

14.67

681.8

39525

12161

FCH 4 , 0 (kmol/h)

X CH 4

Wcat (tonn)

4

0.51

8

H 2 / Supmem 3

2

Reactor heat duty (kcal/m2h) Reactor 1: 34975 Reactor 2: 22210 Reactor 1: 65403 Reactor 2: 38251 Reactor 1: 93880 Reactor 2: 53975 Reactor 1: 34975 Reactor 2: 22210 Reactor 3: 18569 Reactor 1: 65403 Reactor 2: 38251 Reactor 3: 31970 Reactor 1: 93880 Reactor 2: 53975 Reactor 3: 42408

294

M. De Falco, G. Iaquaniello, B. Cucchiella et al.

Wall temperature effect The operating pressure is imposed equal to 20 bar, the inlet methane flowrate is 8 kmol/h. Results of model simulations are reported in Table 3.3. It is a worth assessment that: •



Plant performance improves increasing reactors temperature: methane conversion and hydrogen produced and permeated for unit of membrane surface increase, while catalyst weight, total steam and total methane flowrate required for producing 10,000 Nm3/h of H2 decrease by increasing reformers temperature. If 3 RMM are assembled the methane conversion is greater and, consequently, the total steam and methane requirements are lower. On the other hand the catalyst weight is higher and the H2 to membrane surface is lower. This is due to the fact that the third step is lower efficient since the third reactor feedstock is composed by high contents of CO2 and in the third separator inlet stream the hydrogen content, and therefore the upstream hydrogen pressure, is lower. In order to clarify these aspects, methane conversion profiles are shown in Figures 3.2 for a 3 steps RMM plant. Obviously, the first reactor has the highest methane conversion, since the feedstock is clean (at TW =873 K, X CH 4 ,1° reactor = 0.24; X CH 4 , 2° reactor = 0.133; X CH 4 ,3° reactor =

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

0.118). The recovery ratios, i.e. the ratio between hydrogen permeated and hydrogen inlet in the separator, are 47.9% for the first, 46.6% for the second and 45.7% for the third separator.

Figure 3.2a – Methane conversion profiles for the three reactors ( TW =873 K)

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Reformer and Membrane Modules Plant to Optimize Natural…

295

Figure 3.2b – Methane conversion profiles for the three reactors ( TW =923 K)

Pressure effect The wall temperature for the reformers is imposed equal to 923 K, the inlet methane flowrate is 8 kmol/h. Results of model simulations are reported in Table 3.4. It has to be noticed that operating pressure has a double effect on plant performance: •

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.



Increasing pressure has a negative effect on reactor performance since the reactions occur with an increasing of gas volume; Pressure has a positive effect on separation since increases the permeation driving force.

Globally, an operating pressure of 20 bar seems to be optimal for the operating conditions imposed.

Residence time effect The effect of gas mixture residence time is evaluated fixing the reformers and membranes geometry and the composition of feedstock and varying the inlet methane flowrate. The wall temperature for the reformers and the operating pressure are imposed equal to 923 K and 20 bar. Results of model simulations are reported in Table 3.5.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

296

M. De Falco, G. Iaquaniello, B. Cucchiella et al.

The methane conversion is positively affected by the residence time, since the gas mixture comes out the reformers at higher temperatures and the hydrogen has more time to permeate through the membranes in the separation units. On the other hand, it has to be noticed that the system productivity has a maximum (minimum values of catalyst amount and membrane surface required) at FCH 4 , 0 = 8 kmol/h. This is due to the fact that reducing the residence time has normally a positive effect on reactors productivity but, if the inlet feedstock rate is too high, the greater pressure drops lead to a reduction of separation capabilities.

4. MEMBRANE REACTOR PLANT CONFIGURATION 4.1 Process Configuration

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

In the present paragraph, the membrane reactor technology is presented and assessed. If the hydrogen selective membrane is placed directly inside the reaction environment (see Figure 1.11), part of the hydrogen produced by means of the reaction scheme (eqs. 3.13.3) is removed from the reaction environment. By this way, the composition of gas mixture is always far from the equilibrium and high methane conversions are achieved at lower operating temperatures [40-42]. Membrane reactor is composed by two concentric tubes: the catalyst pellets are packed in the annular zone (reaction zone), while the membrane is the internal tube itself (permeation zone). The heat flux is supplied from the external and a sweeping gas is fed in the inner section to carry out the hydrogen permeated: the sweeping gas is water steam in analogy with RMM plant. Since the reaction and separation steps are carried out in one single device, the system is more compact. The membrane assembled in the reactor is composed by a Pd-Ag selective layer and a SS support [36]. A temperature threshold for the selective membrane has to be imposed, since at temperature > 500°C the thermal stresses between selective layer and support can lead to the loss of adherence and the creation of preferential way and the lost of selectivity properties.

4.2. Mathematical Modelling The membrane reactor is modeled by means of mass, energy and momentum balances, which leads to a PDEs set. The reaction scheme is the same imposed for RMM plant (eqs. 3.1 – 3.3). The model is based on the following assumptions: • • • •

steady-state conditions; negligible axial dispersion and radial convection; ideal gas behavior; a single reactor representative of any other reactor;

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Reformer and Membrane Modules Plant to Optimize Natural…

297

Reactant mixture Sweeping steam Reactant mixture

Figure 4.1. Co-current membrane reactor configuration



• •

pseudo-effectiveness factors η1, η2 and η3 for the reactions (3.1-3.3) independent of local conditions and fixed at 0.02 as an average value of those reported in the literature [31,32]; co-current configuration between reaction and permeation zones (see figure 4.1). hydrogen selectivity assumed infinite.

In the following, the equations composing the model are reported.

REACTION ZONE Mass balances dp ⋅L ∂ (u~z c~i ) ∂ 2 (u~z c~i ) 1 ∂ (u~z c~i ) η ⋅ ρ cat ⋅ (1 − ε ) ⋅ L = ⋅ +~⋅ ⋅ ri )− ( 2 ∂~z ∂~ r2 r ∂~ r u z , 0 cCH 4 , 0 Pemr ⋅ Ri

(4.1)

i = CH 4 , H 2O, H 2 , CO, CO2 Energy balance 3

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

ρ cat ⋅ (1 − ε ) ⋅ L ⋅ ∑η j ⋅ (− ΔH j ) ⋅ r j ~ ~ ~ λer ⋅ L ∂TR ∂ 2TR 1 ∂TR j =1 (4.2) ⋅( ~2 + ~ ⋅ ~ ) + = 2 ∂~z ∂ ∂ r r r ( u c ) ⋅ c (u z ctot ) ⋅ c p , mix ⋅ Ri z tot p , mix ⋅ TR , 0 Momentum balance

~ 2 dP f ⋅ G ⋅ μ g ⋅ L (1 − ε ) = ⋅ d~ z ε3 ρ g ⋅ d p 2 ⋅ P0

(4.3)

It has to be noticed that these equations are the same ones of the reformer model of RMM plant. As it is reported in the following, the boundary conditions are different.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

M. De Falco, G. Iaquaniello, B. Cucchiella et al.

298

PERMEATION ZONE Mass balances

dYH 2 J H 2 ⋅ 2π ⋅ Rmem ⋅ L = d~ z u z cCH 4 0

(4.4)

where:

YH 2 =

u z , permcH 2 , perm

(4.5)

u z cCH 4 0

and the term J H 2 is calculated as reported in eq. 3.8, where pH 2 , up is the hydrogen partial pressure in reaction zone and pH 2 , down is the hydrogen partial pressure in permeation zone. Energy balance ~ dTP L = ⋅ U1 ⋅ 2π ⋅ Rmem ⋅ (TR − TP ) + J H 2 ⋅ π ⋅ Rmem ⋅ hH 2 , reac − hH 2 , perm (4.6) ~ dz u z , permctot , perm ⋅ c p , perm

[

(

)]

where u z , permctot , perm is the total molar flux in permeation zone, c p , perm is the gas mixture heat capacity, U1 is the global heat transfer coefficient between reaction and permeation zone, hH 2 , reac and hH 2 , perm are hydrogen enthalpies of reaction zone and permeation zone, respectively. It is assumed that pressure drop is negligible in permeation zone. The boundary conditions imposed are listed below:

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

REACTION ZONE ~z = 0, ∀~ r

(4.7)

u~z c~CH 4 = 1 u~z c~i =

u z ci , 0 u z , 0 cCH 4 , 0

( i = H 2 O, H 2 , CO, CO2 )

~ TR = 1 ~ P =1

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Reformer and Membrane Modules Plant to Optimize Natural…

299

~ r = 1, ∀~ z

(4.8)

∂ (u~z c~i ) =0 ∂r~ ~ ∂TR U ⋅ Ri λer ~ = (TW − TR| R ) i ∂r TR ,0 ~ r = Rmem / Ri , ∀~ z

dp Pemr



(4.9)

Ri ∂ (u~z c~H 2 ) = J H2 ~ u z , 0cCH 4 ,0 ∂r

∂ (u~z c~i ) = 0 ( i = CH 4 , H 2O, CO, CO2 ) ∂r~ ~ ~ ~ ⋅ ~ = U1 ⋅ (TR| Ri − TP ) Ri ∂r

λer ∂TR

PERMEATION ZONE For co-current configuration, the boundary conditions are imposed both on the initial section:

~ z =0

(4.10)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

YH 2 = 0

~ TP = TP , 0 / TR , 0 The set of PDEs is solved discretizing the radial coordinate by means of central secondorder differences: the resulting ODE system is solved using a Runge-Kutta method with variable step. The hot wall temperature is a modeling parameter and it is assumed equal to heating fluid temperature. For a more detailed analysis of the mathematical model, refer to [37,38].

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

300

M. De Falco, G. Iaquaniello, B. Cucchiella et al.

4.3. Plant Performance Since a comparison with the RMM plant has to be made, the same operating conditions have to be fixed. Membrane reactor is 6 m long and the radius is calculated so that the annulus surface is equal to the section of reformers in RMM plant:

Ri = R 2 i , RMM + R 2 mem

(4.11)

where Rmem is imposed. Moreover, in order to impose the same apparent residence time, the methane inlet flowrate is divided for 3 for comparison with 3 steps RMM plant. It means that a system of 3 MR in parallel is imposed to make comparison with the 3 RMM steps in series. The wall temperature is limited in order to avoid overheating of selective membrane. The total hydrogen production is imposed equal to 10,000 Nm3/h: a single MR system is simulated and then the number of systems to be assembled in parallel is calculated as 10,000/hydrogen produced.

Wall temperature effect The operating pressure is imposed equal to 20 bar, the inlet methane flowrate is 8 kmol/h, therefore 2.67 kmol/h for each MR. Results of model simulations are reported in Table 4.2.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Table 4.1. Membrane reformer parameters. Length

6m

773 K

0.06 m

Sweep steam flowrate Permeation zone temperature Inlet methane flowrate

Diameter

Eq. 4.11

Membrane diameter

3 kmol/h

TW

723 – 873 K

Steam to carbon ratio

3

Reaction zone pressure Permeation zone pressure

15 – 20 – 25 bar 1 bar

ε δ

0.5 20 µm

4 – 8- 12 kmol/h

Table 4.2. Plant performance versus hot reformer wall temperature.

TW

X CH 4 Wcat

(K)

MR

723 773 823 873

0.18 0.26 0.356 0.46

H 2 / Supmem

(tonn)

(Nm3/m2h)

17.6 12.2 9.1 7.1

6.4 9.22 12.4 15.9

Total membrane surface (m2) 1561.2 1084 805 629.6

Total steam generation (Nm3/h) 76576 53164 39478 30882

Total methane feedstock (Nm3/h) 18564 12888 9570.5 7486.5

Reactor heat duty (kcal/m2h) 67295 100830 138930 180410

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Reformer and Membrane Modules Plant to Optimize Natural…

301

It is a worth assessment that: •



MR plant performance improves increasing wall temperature: methane conversion and hydrogen produced and permeated for unit of membrane surface increase, while catalyst weight, total steam and total methane flowrate required for producing 10,000 Nm3/h of H2 decrease by increasing reformers temperature. Imposing the same wall temperature, MR performance is better than RMM one. Considering a TW = 823 K, MR conversion is 0.36 vs. 0.32 of 3 steps RMM plant, catalyst required is 9.1 vs. 10.13 tonn, H 2 / Supmem is 12.4 vs. 11.15 Nm3/m2h.



On the other hand, at TW = 873 K, the membrane is heated up to 815 K, which is a value too high. Therefore, the wall temperature has to be limited at 823 K. This is the main drawback of membrane reactor technology: the membrane temperature threshold leads to limit reaction operating temperature. It is a worth assessment that in RMM plant there is not this threshold, since the operating conditions of reformers can be imposed independently and the reaction can be promoted thanks to higher reformer temperatures.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 4.2 reports the MR methane conversion and the conversion for a 1, 2 and 3 RMM steps plant: MR represents an asintotic value of conversion, which means that infinite RMM steps are required to reach MR methane conversion, even if already after 3 RMM steps the final conversion of RMM plant is only 11% about lower than MR plant conversion. Moreover, wall temperature is not limited at 823 K for the RMM plant, by which methane conversion equal to 0.53 (at TW =923 K) and 0.64 (at TW =973 K) can be reached.

Figure 4.2 – Methane conversion for MR and RMM ( TW = 823 K)

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

302

M. De Falco, G. Iaquaniello, B. Cucchiella et al.

Pressure effect The wall temperature for the reformers is imposed equal to 823 K and the inlet methane flowrate is 2.67 kmol/h for each MR. Results of model simulations are reported in Table 4.3. Table 4.3 – Plant performance versus operating pressure ( TW = 823 K) mem X CH 4 Wcat (tonn) H 2 3/ Sup 2

P

(bar) 15 MR 20 25

(Nm /m h)

0.358 0.356 0.355

9.73 9.1 8.7

11.6 12.4 13

Total Total membrane steam surface (m2) generation (Nm3/h) 862.3 42298 805 39478 770.4 37790

Total methane feedstock (Nm3/h) 10254 9570.5 9161

Reactor heat duty (kcal/m2h) 130930 138930 144280

As for RMM plant, operating pressure has a double opposite effect on plant performance, negative for thermodynamics, positive for the hydrogen separation. Globally, MR performance slightly increases at higher operating pressures, since a greater amount of hydrogen is recovered (862.3 m2 of membrane required at 15 bar vs. 770.4 3 m2 at 25 bar).

Residence time effect Simulation results are reported in Table 4.4. Table 4.4 - Plant performance versus inlet methane flowrate ( TW = 823 K, reaction zone pressure = 20 bar).

FCH 4 , 0 for each MR (kmol/h) 1.33 MR 2.67 Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

4

X CH 4

Wcat (tonn)

H 2 / Supmem (Nm3/m2h)

Total membrane surface (m2)

Total steam generation (Nm3/h)

Total methane feedstock (Nm3/h)

Reactor heat duty (kcal/m2h)

0.474

11.78

9.6

1043.5

32570

6203.8

104950

0.356

9.1

12.4

805

39478

9570.5

138930

0.3

8.25

13.7

730.4

48857

13029

156250

Reducing the residence time (increasing the methane flowrate) leads to a reduction of methane conversion, since the hydrogen has less time for permeating and the composition of the mixture is nearer to the equilibrium one, slowing down reaction kinetics. In fact, the percentage of hydrogen permeated to total hydrogen produced is 65.5% at 12 kmol/h and 86.7% at 4 kmol/h. On the other hand, with shorter residence time the amount of catalyst and membrane surface required for the production of 10,000 Nm3/h of hydrogen are lower, attesting a higher productivity of the reactor.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Reformer and Membrane Modules Plant to Optimize Natural…

303

5. KEY ECONOMIC SUPPOSITIONS AND PARAMETERS In this section the new membrane concept with hydrogen-selective, dense palladiumsilver membranes is compared to the conventional steam reformer coupled with a gas fired cogeneration plant to evaluate the production cost for a 10,000 Nm3/h hydrogen unit. Such a cost has been estimated on the basis of the variable cost, the maintenance and depreciation costs. The RMM technology was selected against the MR geometry because under the actual membranes performance and temperature limits, the RMM geometry performs better and much easier to implement form a mechanical point of view. Economic parameters used to estimate the production costs for H2 are listed in Table 5.1. A simple approach for cost analysis has been used in order to compare RMM technology with conventional technologies. Hydrogen production by natural gas steam reforming and a stand-alone natural gas-fired cogeneration plant producing about 22.3 MW of power have been considered. The capital cost for the conventional co-generation configuration has been estimated at €1200 per kW, with overall efficiency set at 45% and medium pressure steam extraction. Table 5.1 – Economic parameter for the estimation of hydrogen production costs

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Natural gas price Co-product electricity price MP Steam product price Capacity factor H2 Plant capacity Depreciation Maintenance materials & labor (H2 section) Maintenance materials & labor (power section) Maintenance materials & labor (membrane module) Post-combustion CO2 capture cost [ …]

0.20 €/ Nm3 (LHV= 8,700 Kcal /Nm3) 0.085 €/kWh 13 €/ton 95% 10,000 Nm3/h 10% / y of investment 2.5% of the investment 1.5 % of the investment Replacement of membrane module every three years (0.50 MM € per year) 25% higher than pre-combustion capture,

5.1. Variable operating costs (VOC) Operating costs include the consumption of feed, fuel and electricity. Setting export steam to zero simplifies the economic comparison and this was achieved by supplying such a steam to the stripper of the CO2 removal unit in both cases. In order to have a proper comparison in terms of overall CO2 removal, in the conventional unit, part of the CO2 is also removed from the flue gases. Table 5.2 provides a comparison for the two schemes based on the operating and economic parameters listed section 3 and Table 5.1. Due to the positive impact of power production, variable operating production costs (VOC) for the hybrid scheme are nearly 15% lower than conventional H2 production technology coupled with a gas-fired co-generation plant.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

304

M. De Falco, G. Iaquaniello, B. Cucchiella et al. Table 5.2. Comparison between RMM and convention plants

TECHNOLOGY NG FEED + FUEL [MMkcal/h]: NG FEED to gas turbine [MMkcal/h] Duty reformer [MMkcal/h] Steam to carbon ratio (S/C) T inlet reformer [°C] T outlet reformer [°C] P hydrogen battery limits [kPa] Net Power Output [MW] System efficiency [kcal/Nm3 H2] Net System Efficiency [kcal/Nm3 H2] Hydrogen Plant capacity [Nm3/h] Retentate recycle ratio Steam output to CO2 removal [ton /h] Sweeping steam rate [%] Hydrogen purity [%]

Hybrid MEMBRANE (RMM) 45,0 17,5 8,5 (b) 3 530 650/600/550 (d) 1600 10,8 6250 3900 (a) 10,000 0.6 7,1 (c) 30 99.999

SMR+ cogen unit 40,31 25.25 10.04 3 560 850 1600 10.8 6,555 4,233 (1) 10,000 / 16,0 (c) / 99.999

a) (( feed+fuel)-(power exported)/0,4)/ H2 production b) the value is obtained by adding up the duties of 3 steps c) Includes steam generation for CO2 removal d) to optimise the heat recovery from the gas turbine exausts

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

5.2. Investment costs for a 10,000 Nm3/h hydrogen plant The investment has been estimated using Q-estimate, an estimation tool, and TechnipKTI’s long-standing experience in building hydrogen production plants. When calculating investments, it has been assumed that membranes up-dated cost is €2000 per m2 [43, 44] and gas turbine €500 per installed kW. The cost of the gas turbine includes retrofitting of the machines’ combustion chamber. Table 5.3a gives the estimated costs of equipment for the hydrogen section of the hybrid configuration. It is important to note that the cost of rotating machines is now becoming much more important than for conventional technology, reaching one-third of the total. The membrane module costs more or less the same amount as the reformer. Table 5.3a. Estimated costs of equipment Millions of euros Convective reformer + WHB + Structure 1,56 Heat exchangers, vessels and reactors 1.04 PSA 1.0 Rotating machines 2,65 Separation modules 1,50 (1) Other incl. catalysts 0.30 Total 8,05 (1) the overall estimated membrane surface for the two modules is 750 m2

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Reformer and Membrane Modules Plant to Optimize Natural…

305

Table 5.3b. Overall investment costs for the delivered equipment costs Millions of euros 8,05 3,22

% of delivered equipment cost 100 40

Equipment costs (delivered) Bulk materials (piping, instrumentation, electrical etc.) Total direct costs 11,27 140 Engineering + supervision 3,62 45 Construction 4,03 50 Total direct+ indirect costs 18,92 235 Contractor profit and project 2,42 30 contingency Total investment costs 21,34 265 (1) the overall estimated membrane surface for the two modules is 750 m2

4500

15000

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

5500

CSR-01

CSR-02

CC-01

CC-02

CC-03

CC-04

CC-05

Figure 5.1 - Layout of an unfired “convective reformer”

Table 5.3b lists the overall investment costs for the delivered equipment costs. All the items are estimated as a percentage of the delivered equipment costs [45]. This method is commonly used for preliminary estimates. It has been confirmed in our recently-constructed hydrogen plants that the accuracy of estimates is in the range of +/- 20 %. CO2 removal equipment for the conventional unit was estimated on the basis of an investment cost per ton of CO2 removed [46]. The investment cost of the separation module is taken as directly proportional to the required membrane surfaces. The higher investment costs for the hybrid scheme are mainly related to final recompression of the low-pressure hydrogen product (up to 15 barg) before being exported at battery limits. Conversely, the reformer has been reduced by 50% due to the use of stainless steel 347H instead of exotic and expensive

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

306

M. De Falco, G. Iaquaniello, B. Cucchiella et al.

material such as the centrifugally cast 20-25 Cr-Ni, as a result of a lower operating temperature (200 °C less). Figure 5.1 offers a schematic representation of an unfired “convective reformer” (coils CSR-01 and CSR-02) integrated in a conventional waste heat boiler. Such a design represents a very compact and economic solution compared with conventional commercial reformers consisting in a number of tubes placed inside a radiantfired chamber where heat is supplied by burners. The remaining coils located downstream are used for steam superheating, steam generation and BFW preheating.

5.3. Hydrogen production costs Production costs include Variable Operating Costs (VOC), Operating &Maintenance costs (O&M) and the depreciation rate (Table 5.4). Table 5.4 – Hydrogen production costs

3,11

37,0

20,0

13,0

3,85

36.86

3,70

44,0

Depreciation per 1000 Nm3 H2 € !000 per Nm3

31,07

Millions €

3,63

Yearly depreciation

Total investment

6,10

Millions €

Investment CO2 section Millions €

Hybrid technology Convention al technology + power generation

Millions €

21,34

Investment H2 section

Investment power section Mllions €

Case

gas turbine only , being the WHB included in the H2 part assuming €200,000 per ton of CO2 removed in pre-combustion step and € 250,000 per ton of CO2 removed in post combustion step (the values have been updated compared to the year of publication);

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

The net hydrogen production costs in terms of euros per Nm3 are given in Table 5.5. Table 5.5. net hydrogen production costs in terms of euros per Nm3 Case

VOC, € per 1000 Nm3 H2

Hybrid technology Conventional technology +power generation

51 59

O&M per 1000 Nm3 H2 10,4 (1) 7

Depreciation rate per 1000 Nm3 H2 37 44

Total per 1000 Nm3 H2 98,4 110,0

Membrane replacement are included

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Reformer and Membrane Modules Plant to Optimize Natural…

307

The membrane modules have been taken as requiring replacement every three years. A membrane lifetime of several (2 – 5) years (at temperature ≤ 400 °C) was recently assessed based on long-term stability tests over 150 days [47]. Therefore, O&M costs for the membrane-based hybrid technology are the highest. Nevertheless it seems quite competitive compared to traditional steam reforming technology. The hydrogen production costs for the hybrid are more than 10% lower than those of the conventional H2 scheme coupled with the cogeneration unit. This cost differential will increase as hydrogen membrane technology improves from its infancy. It is important to note that the high pressure of the retentate makes CO2 removal through the MDEA system easier compared to the traditional scheme. This may result in another advantage of proposed architecture. It is important to note that with conventional technology minus power cogeneration, about 54 of the production costs are related to variable costs. Higher natural gas prices will have a greater impact on conventional technology. On the other hand, the situation is slightly different in the hybrid system. Variable costs are only 52% of production costs. Such a difference implies that if accelerated depreciation is applied, production costs will be more heavily affected than the hybrid system. The analysis shows O&M costs for the hybrid scheme are an important part of the production costs (almost 11% compared to 6,5% for conventional schemes). This reflects the current state-of-the-art status of membranes. It is likely that advances in membrane technology and production will reduce said figure, thus increasing the cost differential between the two technologies .

6. CONCLUSIONS In this chapter, innovative technologies for the production of hydrogen through natural gas steam reforming, based on the integration of hydrogen selective membranes, have been presented and assessed from a technical and economical point of view. Two plant configurations have been hypothesized: •

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.



Reformer and Membrane Modules (RMM) plant. Reaction and hydrogen separation are performed in two different units, placed in series. Membrane Reformer (MR), which combines the hydrogen separation through the selective membrane and the steam reforming reaction into one unit and separates hydrogen immediately after it was formed.

The main outcomes are: 1. At the same operating conditions (wall temperature), MR leads to better performance. A membrane reactor can be considered as a infinite series of reactor + separator modules. 2. On the other hand, membrane temperature threshold leads to limit the wall temperature ( TW < 823 K) and therefore imposes a limit to reaction conditions as well. It is a worth assessment that in RMM plant there is not this threshold, operating

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

308

M. De Falco, G. Iaquaniello, B. Cucchiella et al. conditions of reformers can be imposed independently and the reaction can be promoted thanks to higher reformer temperatures. 3. Globally, the hydrogen production costs for the RMM technology are more than 10% lower than those of the conventional H2 scheme coupled with the cogeneration unit.

Surely, the Pd based membrane technological development is crucial for a wide diffusion of the membrane reactors. The R&D efforts have to be focused mainly on the increase of membrane temperature threshold, improving the adherence between the active layer and the support and working on the welding techniques, and on the improvement of membrane reliability in order to produce objects suitable for commercial applications. A future improvement of membrane performance would surely promote the applications of MRs, but the actual membrane state-of-the-art leads to the conclusion that RMM plant seems to be more suitable for industrial applications.

REFERENCES

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

[1] [2]

www.iea.org/Textbase/techno/essentials.htm - April 2007. Dittmeyer, R; Höllein, V; Daub, K. Membrane reactors for hydrogenation and dehydrogenation processes based on supported palladium, Journal of Molecular Catalysis A: Chemical, 2001, 173, 135–184. [3] Howard, BH; Killmeyer, RP; Rothenberger, KS; Cugioni, AV; Morreale, BD; Enick, RM; Bustamante, F. Hydrogen permeance of palladium-copper alloy membranes over a wide range of temperatures and pressures, Journal of Membrane Science, 2004, 241, 207-218. [4] Peachey, NM; Snow, RC; Dye, RC. Composite Pd/Ta metal membranes for hydrogen separation, Journal of Membranes Science, 1996, 123-133. [5] Ozaki, T; Zhang, Y; Komaki, M; Nishimura, C. Preparation of palladium-coated V and V-15Ni membranes for hydrogen purification by electroless plating technique, International Journal of Hydrogen Energy, 2003, 28, 297-302. [6] Tosti, S; Bettinali, L; Castelli, S; Sarto, F; Scaglione, S; Violante, V. Sputtered, electroless and rolled palladium-ceramic membranes, Journal of Membrane Science, 2002, 196, 241-249. [7] Tosti, S; Basile, A; Bettinali, L; Borgognoni, F; Chiaravalloti, F; Gallucci, F; Longterm tests of Pd–Ag thin wall permeator tube, Journal of Membrane Science, 2006, 284, 393-397. [8] Marigliano, G; Barbieri, G; Drioli, E. Effect of energy transport on a palladium-based membrane reactor for methane steam reforming process, Catalysis Today, 2001, 67, 8599. [9] Shu, J; Grandjean, BPA; Van Neste, A; Kaliaguine, S. Catalytic palladium-based membrane reactors: a review, The Canadian Journal of Chemical Engineering, 1991, 69 (5) 1036-1060. [10] Itoh, N; Xu, WC. Selective hydrogenation of phenol to cyclohexanone using palladium-based membranes as catalysts, Applied Catalysis A: General, 1993, 107, 83– 100.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Reformer and Membrane Modules Plant to Optimize Natural…

309

[11] De Falco, M; Barba, D; Cosenza, S; Iaquaniello, G; Marrelli, L Reformer and membrane modules plant powered by a nuclear reactor or by a solar heated molten salts: assessment of the design variables and production cost evaluation, International Journal of Hydrogen Energy, 2008, 33, pp. 5326-5334. [12] De Falco, M; Barba, D; Cosenza, S; Iaquaniello, G; Farace, A; Giacobbe, FG Reformer and membrane modules plant to optimize natural gas conversion to hydrogen, Special Issue of Asia-Pacific Journal of Chemical Engineering on Membrane Reactors, in press. [13] Dybkjaer, I Tubular reforming and autothermal reforming of natural gas – an overview of available processes, Fuel Processing Technology, 1995, 42, 85-107. [14] Nijmejier, A. Hydrogen selective silica membranes for use in membrane steam reforming Thesis, University of Twente, 1999. [15] Morreale, BD; Ciocco, MV; Howard, BH; Killmeyer, RP; Cugini, AV; Enick, RM. Effect of Hydrogen Sulphide on the hydrogen permeance of palladium copper alloys at elevated temperatures Journal of . Membrane Science, 2004, 241, 219 – 224. [16] Howard, BH; Killmeyer, RP; Rothenberger, KS; Ciocco, MV; Morreale, BD; Enick, RM; Bustamante F.: Hydrogen permeability of Palladium – Copper alloy composite membranes over a wide range of temperatures and pressures http://www.netl.doe.gov/ coal/fuels/refshelf/papers/hydrogen/Hydrogen%20Permeability%20of%20PDCu%20Alloy%20Composite%20Membranes.pdf [17] Morreale, BD; Enick, RM Morsi, BI; Howard, BH; Rothenberger, KS. Evaluation of high pressure, high temperature inorganic hydrogen membranes http://www.netl.doe. gov/osta/techpapers/2001-477.PDF [18] Kluiters SCA. Status review on membrane systems for hydrogen separation, Intermediate report EU project MYGREID NNES - 2001 – 670, December 2004 http://www.ecn.nl/docs/library/report/2004/c04102.pdf [19] Morooka, S; Kim, SS; Yan, S; Kusakabe, K; Watanabe, M. Preparation of SiC Membrane on Porous Alumina Support Tube by Chemical Vapor Deposition of Tripropylsilane and Its Elevated-Temperature Gas Permselectivity, Int. J. Hydrogen Energy, 1996, 21, 183-188. [20] Morooka, S;Yan, S; Kusakabe K. Akiyama Y.: Morphology and gas permeance of ZSM-5-type zeolite membrane formed on a porous α-alumina support tube, J. Membrane Sci., 101 (1995) 8998. [21] Nam, SE; Lee, KH. A study on the palladium/nickel composite membrane by vacuum electrodeposition Journal of Membrane Science, 2000, 170, 91–99 [22] Xomeritakis G; Lin YS. Fabrication of thin metallic membranes by MOCVD and sputtering J. Membr. Sci., 1997, 133, 217-230. [23] Xomeritakis, G; McCool, B. Composition control and hydrogen permeation characteristics of sputter deposited palladium silver membranes Journal of Membrane Science, 1999, 161, 67–76. [24] Shu, J; Grandjean, BPA; Van Neste, A; Kaliaguine, S. Simultaneous Deposition of Pd and Ag on Porous Stainless Steel by Electroless Plating. Journal of Membrane Science, 1993, 77, 181-195. [25] Cheng, YS; Yeung, KL. Palladium-silver composite membranes by electroless plating technique Journal of Membrane Science, 1999, 158, 127-141

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

310

M. De Falco, G. Iaquaniello, B. Cucchiella et al.

[26] Akis, BC. Preparation of Pd-Ag/PSS Composite Membranes for Hydrogen Separation Thesis Submitted to the Faculty of the WORCESTER POLYTECHNIC INSTITUTE, 2004. [27] Jun, CS; Lee, KH. Palladium and palladium alloy composite membranes prepared by metal-organic chemical vapor deposition method (cold-wall) Journal of Membrane Science, 2000, 176, 121–130. [28] Jayaraman, V; Lin, YS. Synthesis and hydrogen permeation properties of ultrathin palladium-silver alloy membranes Journal of Membrane Science, 1995, 104, 251-262, [29] Zhao, HB; Xiong, GX; Baron, GV. Preparation and characterization of palladium-based composite membranes by electroless plating and magnetron sputtering Catalysis Today, 2000. 56, 89–96 [30] Tong, HD; vanden Berg, AHJ; Gardeniers, JGE; Jansen, HV; Gielens, FC; Elwenspoek, MC. Preparation of palladium–silver alloy films by a dual-sputtering technique and its application in hydrogen separation membrane Thin Solid Films, 2005, 479, 89– 94 [31] Rostrup-Nielsen, JR. Production of synthesis gas, Catalysis Today 18 (1993) 305-324. [32] Xu, J; Froment, G. Methane steam reforming II: Diffusional limitations and reactor simulation, AIChE Journal, 1989, 35 (1), 97-103. [33] Xu, J; Froment, G. Methane steam reforming, methanation and water-gas shift: I. Intrinsic kinetics, AIChE Journal, 1989, 35 (1) 8-96. [34] Kulkarni, BD; Doraiswamy, LK. Estimation of effective transport properties in packed bed reactors, Catalysis Reviews: Science and Engineering, 1980, 22 (3), 431-483. [35] De Wasch, AP; Froment, GF. Heat transfer in packed beds, Chemical Engineering Science, 1972, 27, 567–76. [36] Shu, J; Grandjean, B; Kaliaguine, S. Methane steam reforming in asymmetric Pd and Pd-Ag porous SS membrane reactors, Applied Catalysis A: General, 1994, 119, 305325. [37] De Falco, M; Nardella, P; Marrelli, L; Di Paola, L; Basile, A; Gallucci, F. The effect of heat flux profile and of other geometric and operating variables in designing industrial membrane steam reformers, Chemical Engineering Journal, 2008, 138, 1-3 pp. 442451. [38] De Falco, M; Di Paola, L; Marrelli, L; Nardella, P. Simulation of large-scale membrane reformers by a two-dimensional model, Chemical Engineering Journal 2007, 128, pp. 115-125. [39] Elnashaie, S; Elshishini, S. Modelling, simulation and optimization of industrial fixed bed catalytic reactors, Vol. 7 of Topics in Chemical Engineering, Gordon and Breach Science Publisher, 1993. [40] Lin, Y; Liu, S; Chuang, C; Chu, Y. Effect of incipient removal of hydrogen through palladium membrane on the conversion of methane steam reforming: experimental and modeling, Catal. Today, 2003, 82, 127-139. [41] Chai, M; Machida, M; Eguchi, K; Arai, H. Promotion of hydrogen permeation on a metal-dispersed alumina membrane and its application to a membrane reactor for steam reforming, Appl Catal A: General, 1994, 110, 239-250. [42] Tong, J; Matsumura, Y. Pure hydrogen production by methane steam reforming with hydrogen-permeable membrane reactor, Catalysis Today, 2006, 111, 147-152.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Reformer and Membrane Modules Plant to Optimize Natural…

311

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

[43] Praxair inc, Integrated ceramic membrane system for H2 production, US DOE Annual Merit Review meeting, May, 2005, 24. [44] Arps, J. Cost-Effective Method for Producing Self-Supporting Pd Alloy Membrane for Use in the Efficient Production of Coal-derived Hydrogen, DOE Hydrogen Program 2006. [45] Max, S. Peters, KD; Timmerhaus, Plant Design and Economics for Chemical Engineers, Third Edition, McGraw-Hill Book Company 1980. [46] Kreutz, T. Co-production of hydrogen, electricity and CO2 from coal with commercially ready technology. Part B Economic analysis, International Journal of Hydrogen Energy, 2005, 30, 769-784. [47] Bredesen, R. Thin Pd-23w%Ag membranes for hydrogen separation, CASTOR/ CACHET/DYNAMIS/ENCAP workshop, Lyon, 2008, 22nd – 24th January.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved. Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

In: Syngas Production Methods, Post Treatment... Editors: Adorjan Kurucz and Izsak Bencik

ISBN: 978-1-60741-841-2 © 2009 Nova Science Publishers, Inc.

Chapter 7

RECENT DEVELOPMENTS OF FISCHER-TROPSCH SYNTHESIS CATALYSTS - PREPARATION AND CHARACTERIZATION Naoto Koizumi, Takehisa Mochizuki, Daichi Hongo and Muneyoshi Yamada* Department of Applied Chemistry, Graduate School of Engineering, Tohoku University, Aoba 6-6-07, Aramaki, Aoba-ku, Sendai 980-8579, Japan

ABSTRACT

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Since Fischer and Tropsch discovered the Fischer-Tropsch synthesis (FTS) reaction in 1923, in which the mixture of CO and H2 (so called syngas) derived from coal is converted into hydrocarbon, there have been continuous efforts on improving the understanding of both fundamental and technological aspects of this catalytic reaction. More recently, FTS reaction has attracted attentions as a synthetic method of clean transportation fuels from various carbon resources such as natural gas, coal and biomass. In this review, recent developments in both catalyst preparation and catalyst characterization technique for Fe and Co FTS catalysts have been reviewed, including our recent results on novel preparation method of Co FTS catalyst with chelating agents.

INTRODUCTION Recently, efficient use of energy with low environmental impact has become very important concept. From this point of view, diesel engine is promising, because its thermal efficiency is higher than that of gasoline engine. However, its exhaust gas is responsible for the automobile origin NOx and the automobile origin particulate matters (PM). Thus, development of high-quality diesel fuel, which reduces NOx and PM emissions, is important to make the best use of the higher thermal efficiency of diesel engine. FTS is one of versatile process for the synthesis of transportation fuels, because syngas can be produced from, in principle, any carbon resources such as natural gas (GTL), coal

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

314

Naoto Koizumi, Takehisa Mochizuki, Daichi Hongo et al.

(CTL), biomass (BTL) and/or coal bed methane. Furthermore, it has recently become known that the tail gas from the combustion of FT oil in diesel engine contains much less NOx and PM than that of the diesel fuel from petroleum (i.e. gas oil). Therefore, by using FT oil instead of gas oil as diesel fuel, the simultaneous reduction of CO2, NOx, PM and SO2 is expected. However, production cost of FT oil is expensive. For the reduction of production cost of FT oil, development of next-generation FTS catalyst is a key technology. So far, several reviews of FTS reaction have been published. Those reviews have mainly dealt with the reaction mechanism and/or process development [1-6]. In this paper, recent developments in both catalyst preparation and catalyst characterization techniques for Fe and Co FTS catalysts have been reviewed. Novel preparation method for Co FTS catalyst with chelating agents developed in our laboratory has also been introduced here.

CATALYST TYPE AND REACTION MECHANISM

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Catalyst Type Various transition metals such as Fe, Co, Ni, Cu, Mo, Ru, Rh and Pd have catalytic activity for the hydrogenation of CO. It is well known that CO hydrogenation with Fe, Co and Ru mainly gives hydrocarbons. The chain growth probability of hydrocarbon increases in the order; Fe < Co < Ru. Fe also has a relatively higher activity for the synthesis of oxygenates compared with Co and Ru. In contrast, Cu, Rh and Pd have higher methanol and ethanol synthesis activities rather than hydrocarbon synthesis (FTS) activity. Such the difference in CO hydrogenation activity and selectivity could be related with adsorption properties of these metal surfaces for CO at room temperature. CO adsorbs molecularly on Cu, Rh and Pd surfaces [7], which is suitable for the formation of methanol and ethanol from syngas. On the contrary, Fe has the ability for dissociative adsorption of CO even at room temperature [7]. The dissociative adsorption of CO is thought as the first elemental step for FTS reaction in surface carbide model (see below). On the other hand, CO hydrogenation activity and selectivity also change depending on types of support and promoters used. Co catalyst modified with Mo shows a higher selectivity for alcohol under high-pressure conditions (1-7 MPa) [8]. Instead, Rh catalyst gives exclusively methane when Al2O3 and SiO2 are used as support [9]. These results imply that the adsorption properties of metal surfaces and surface reaction mechanism are influenced by various factors, including the presence of promoter, metal-support interaction, size of metallic cluster, etc. Systematic investigations on these factors are necessary to control the activity and selectivity for CO hydrogenation with these metallic catalysts. Besides the metallic catalysts, Fe also shows FTS activity in nitride form [10] although controversy still exists regarding the composition of the catalytically active Fe phase. Metal sulfides have been regarded as an inactive form for FTS reaction, since metallic catalysts are severely deactivated in the presence of small amount of H2S [11-15]. Mo sulfide is an exception, which shows weak, but constant activity for the synthesis of light hydrocarbon even in the presence of H2S [16]. Alcohol selectivity greatly increased when it is promoted with alkali metals [17]. However, CO hydrogenation activity and selectivity of other metal sulfides were unknown at that time. The authors prepared various metal sulfides, and

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Recent Developments of Fischer-Tropsch Synthesis Catalysts...

315

investigated their CO hydrogenation activity and selectivity. In consequence, it was found that Rh and Pd sulfides show higher methanol synthesis activity than Mo sulfide promoted with potassium in the absence and presence of H2S [18-21]. Both Rh and Pd sulfides are novel sulfur tolerant catalysts.

Anderson-Schulz-Flory Distribution

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

FTS reaction is normally conducted at 473-573 K with 1.0-2.0 MPa total pressure. Under these reaction conditions, thermodynamic limitations are absent. Thermodynamic calculation shows that both carbon deposition and methane formation are favorable under these conditions [1]. However, product distribution obtained from practical FTS reaction is far from calculated one [1]. FTS reaction provides hydrocarbon and small amount of oxygenate with wide range of carbon numbers. It is suggested that a series of sequential carbon-carbon bond formation occurs on the catalyst surface as depicted in Figure 1. If we assume that the chain growth probability, α, is independent of carbon number of adsorbed species, carbon number distribution of hydrocarbon is a function of α. This is so-called Anderson-Schulz-Flory (ASF) distribution. Carbon number distributions obtained with Fe, Co and Ru catalysts are reported to behave in accordance with this distribution (Figure 2) [22]. ASF distribution predicts that it is impossible to obtain hydrocarbon with specific carbon number selectively except for methane and wax. For example, maximum selectivity for C10-C20 hydrocarbon (diesel fraction) is no more than 38 C-mol%. This is a serious defect in the synthesis of kerosene and gas oil by using FTS reaction. To overcome this defect, wax hydrocarbon obtained from FTS reaction is subsequently hydrocracked into diesel fraction in Shell Middle Distillate Synthesis process in Bintulu, Malaysia. 70% or higher diesel selectivity is reported for this process [2,3].

Figure 1. Carbon-carbon bond formation model on FTS catalysts.

Surface Reaction Mechanism It is not doubtful that FTS reaction is a kind of polymerization reaction over catalyst surface. So far, several models have been proposed for detailed surface reaction mechanism. First model is a surface carbide model originally proposed by Fischer, in which CO is dissociatively adsorbed on metal surface to form surface carbide species. This species is then hydrogenated into adsorbed CH2 or CH3 species. Sequential insertion of CH2 species into metal-CH3 bond is thought as a carbon-carbon bond formation step. Spectroscopic investigations verify that CO is dissociatively adsorbed on metal surface such as Fe even at

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

316

Naoto Koizumi, Takehisa Mochizuki, Daichi Hongo et al.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

room temperature [7]. Chain growth mechanism of hydrocarbon was suggested by the experiment by Brady and Pettit [23] using the reaction of CH2N2 on transition metal surface. They found that the reaction of CH2N2 over several metal surfaces (except for Cu) gives C1C6 hydrocarbon in the presence of H2. Their carbon number distribution was similar to that obtained from syngas. In the absence of H2, C2H4 was exclusively formed. These results indicate that the reduction of CH2 species to CH3 species followed by sequential insertion of CH2 species into metal-CH3 bond is a polymerization mechanism as shown in Figure 3. However, surface carbide model does not explain the formation of oxygenate from synags. Emmet and his coworkers [24-27] found that 14C is involved in hydrocarbon product when 14 C-labeled olefin and/or alcohol are added to syngas over Fe-based catalysts. Based on these results, they proposed that C-C bond formation occurs through dehydroation-polymerization of adsorbed hydroxycarbene, known as enol model (Figure 4) [28]. In contrast to surface carbide model, enol model reasonably explains oxygenate formation during FTS reaction. Alternatively, CO-insertion model also explains the formation of oxygenate (Figure 5) [28]. As mentioned above, it is known that the product type (paraffin, olefin, oxygenate, etc.) depend on the type of catalysts used. Davis [28] suggested that, for Fe catalyst, oxygenate mechanism is more appropriate because alcohol is formed over this catalyst, and alcohol added to syngas is involved in the product. In contrast, lower productivity and reactivity of alcohol over Co catalyst suggest that surface carbide model is more probable for this catalyst. Although much more investigations are necessary to make clear the difference and similarity in the chemistry of the FTS reaction with Fe and Co catalysts, this point is important for developing more efficient FTS catalysts.

Figure 2. Typical carbon number distribution with Fe, Co and Ru catalysts (from Ref. [22]).

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Recent Developments of Fischer-Tropsch Synthesis Catalysts...

317

Figure 3. Surface carbide model

Figure 4. Enol model

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 5. CO insertion model

FE CATALYSTS Syngas is usually produced by a steam reforming or partial oxidation of carbon resources such as natural gas, coal and biomass. The syngas derived from coal is rich in CO and has the H2 to CO molar ratio from 0.5 to 1.0, whereas the syngas derived from biomass is rich in H2 (H2/CO molar ratio: 2–3). Because Fe catalysts have water–gas shift activity, they are more suitable for the FTS reaction using the CO-rich syngas than Co and Ru catalysts. Recently, attentions have been paid to the utilization of the biomass-derived syngas, BTL, from the “carbon neutral” point of view. Since Fe is low-cost compared with Co and/or Ru, Fe catalyst is more suitable for the conversion of the biomass-derived syngas. In the followings, recent developments of Fe-based catalysts are reviewed in view of effects of promoter addition and

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

318

Naoto Koizumi, Takehisa Mochizuki, Daichi Hongo et al.

pretreatment conditions. Results for CO2 and CO+CO2 hydrogenation with Fe-based catalysts are also summarized here because the syngas derived from biomass contains CO2.

FTS Active Fe Phase Fe catalysts for FTS have been known for over 80 years. However, controversy still exists regarding the composition of the catalytically active Fe phase. The difficulty arises from the fact that various Fe phases, magnetite, α-iron, or iron carbides, coexist under typical FTS reaction conditions. Besides, the distribution of these phases is a sensitive function of the FTS reaction conditions and the time-on stream of the syngas, which often results in a lack of correlation between the formation of specific Fe phase and the FTS activity and selectivity [1,29]. So far, Fe3O4 [30-33], metallic iron [34-36] or iron carbides [37-40] have been proposed as FTS active Fe phases by different researchers. In followings, their experimental evidences are summarized.

Fe3O4 as active phase Teicher and co-workers [31,32] observed the absence of a delay in the onset of synthesis activity upon contact of unreduced α-Fe2O3 with syngas. The unreduced α-Fe2O3 was in part converted into Fe carbide after 1.5 h of FTS reaction, but the greater fraction of the catalyst was converted into Fe3O4. It was only after 15 h of FTS reaction that iron carbide was the most abundant phase. Because the catalytic activity (presented by CH4) decreased when the formation of carbide in the solid phase was increasing, it was suggested that iron carbide is not the active phase in the FTS reaction. The initial increase of the FTS activity could be correlated with the conversion of α-Fe2O3 into Fe3O4 [31].

Metallic iron as active phase In the so-called competition model [34,35], iron atoms at the surface are considered as the active sites. Because it is difficult to remain the catalyst surface without carbon during the FTS reaction, this assignment cannot be proved.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Iron carbides as active phase Davis [41] investigated active Fe phase by using different pretreatment procedures. Syngas-pretreated 100Fe/3.7Si/0.7K showed low (20%) conversion during about 100 h on stream. The catalysts withdrawn during the 100 h period were composed of mainly Fe3O4 phase. After this synthesis period, hydrogen flow was terminated and the catalyst was exposed to CO for 24 h. After CO exposure, the catalyst contained about 35% carbide phase, the remainder being Fe3O4. CO conversion after the CO exposure, following a short induction period, was greater than 90%. The same catalyst exhibited CO conversion greater than 90% when it was initially pretreated with CO. The activity of the carbide phase is at least five times greater than that of Fe3O4.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Recent Developments of Fischer-Tropsch Synthesis Catalysts...

319

Other researchers [39,42,43] lead to the same conclusion as Davis [41] in view of the FTS activity of magnetite and iron carbide. There is a linear relation between the amount of bulk carbides and the rate of hydrocarbon formation [38]. However, on the basis of Mössbauer spectroscopy experiments, the activity of iron catalysts is not related to the amount of bulk iron carbide [44].

Transient kinetic study using time-resolved in-situ Fe K-edge XANES spectroscopy To make clear the structure-activity relationship directly, Li et al. [45] recently conducted transient kinetic study combined with time-resolved Fe K-edge XANES spectroscopy. During FTS reaction with Fe2O3 promoted with copper, CH4 was formed after short induction period (ca. 60 s) after exposure to syngas at 523 K. In contrast, H2O and CO2 formations were observed without induction period, suggesting that CH4 formation requires some oxygen removal from Fe2O3 surface by H2 and CO. Time-resolved in-situ Fe K-edge XANES spectroscopy showed that the concentration of Fe2O3 phase started to decrease immediately after exposure to syngas, which accompanied with simultaneous formation of Fe3O4 and FeCx phases. They also found that CH4 formation is observed without induction period when Fe3O4-Cu and/or FeCx-Cu prepared form Fe2O3-Cu were exposed to syngas. Steady state CH4 formation rates over Fe3O4-Cu and FeCx-Cu were comparable. Oxygen removal by H2O and CO2 were much smaller than that observed for the FTS with Fe2O3-Cu. Based on these results, they suggested that Fe3O4 and FeCx surfaces are converted into active structure very rapidly (within order of turnover rate) after exposure to syngas. They concluded that it is inaccurate and misleading to describe bulk Fe carbides or oxides as the active phases in Fischer-Tropsch synthesis.

Promoter Effect

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

K and Cu promoters Potassium and copper are usually added to iron catalysts as promoters. Potassium has influence both on the FTS activity and selectivity of the catalyst. The addition of alkali metals other than potassium also affects the product selectivity. The chain growth probability of hydrocarbon increases with alkali metal promotion in the order of Li, Na, K and Rb. Because of the high market price of Rb, potassium is beneficial for practical use [1]. The addition of potassium to Fe2O3 suppresses the rates of olefin hydrogenation and isomerization, and the rate of methanol formation, but enhances the formation rates of C2+ hydrocarbons, branched hydrocarbons, and aldehydes [29]. Furthermore, potassium addition increases carbon deposition on catalysts, which would accelerate the rate of catalyst deactivation [46,47]. Copper addition significantly shortens the time-on stream for attaining the maximum activity. However, its impact on other catalytic properties is still surrounded by controversy. O’Brien et al. [48] reported that the FTS rate increased with increasing copper loading (0–2 atomic ratio per 100 Fe). Copper had a similar effect on product selectivity as potassium. Methane selectivity decreased and products heavier than C11 increased with increasing copper loading, whereas copper addition did not affect olefin selectivity or the isomerization of 1olefins to 2-olefins. In contrast, Bukur et al. [46] found that copper is more effective promoter

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

320

Naoto Koizumi, Takehisa Mochizuki, Daichi Hongo et al.

than potassium in increasing the rate of FTS reaction. In the presence of copper, the hydrocarbon distribution shifted toward higher molecular weight products, but its effect was weaker than that observed for the catalysts promoted with potassium. Copper addition enhanced slightly the secondary reactions (olefin hydrogenation and isomerization). For the catalysts doubly promoted with copper and potassium, the selectivity was strongly influenced by the potassium content [46]. Potassium is known to increase the rate of water gas shift (WGS) reaction [46,49-51], whereas Cu does not catalyze WGS reaction [51]. As concerned with role of potassium, Dry et al. [52] investigated heats of chemisorption of CO and H2 on promoted iron surfaces. Promotion with K2O increases the heat of CO adsorption at low coverage, whereas it decreased the initial heats of hydrogen adsorption. This is explained by the fact that potassium donates electrons to iron, facilitating CO chemisorption since CO tends to accept electrons from iron. On the other hand, hydrogen at higher surface coverage donates electrons to iron and the presence of electron-donating alkali metals would be expected to weaken the iron-hydrogen bond46). On the contrary, Li et al. [53] recently found that the initial reduction and carburization rates of Fe oxides in Fe2O3-Cu are promoted by potassium addition by means of time-resolved in-situ Fe K-edge XANES spectroscopy. Furthermore, BET surface area and CO chemisorption capacity of potassium promoted Fe2O3-Cu (after 1 h FTS reaction) were greater than those of Fe2O3 and Fe2O3-Cu. Based on these results they suggested that potassium promoter creates multiple nucleation sites for carburization of Fe oxide surface, which leads to higher carburization rate and to smaller Fe carbide crystallites, that is, higher active site density. They mentioned that potassium (and also copper) works as structural promoter rather than chemical promoter.

Zn promoters

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Zn has been investigated only little as a promoter of Fe FTS catalysts [51,54-56]. It was suggested that Zn acts just as textural promoter [56] because the FTS selectivity (CH4, C5+ selectivity and 1-pentene/n-pentane ratio) remains unchanged by the addition of Zn. It was found that high surface area Fe/Zn catalyst promoted with both potassium and copper shows a hydrocarbon production rate comparable with Co-Re/TiO2 catalyst even under low temperature reaction conditions typically used for Co-based catalyst (473 K, 2.0 MPa) [55]. Methane selectivity of this catalyst (2%, CO2-free basis) was as low as that of Co-Re/TiO2 catalyst, whereas CO2 selectivity was much greater.

Mn promoter The addition of MnO on Fe has a strong impact on its FTS activity and selectivity. The Fe/MnO catalyst shows a higher selectivity for C2–C4 olefin than the unpromoted catalyst [57,58]. Since C2–C4 olefin is useful for the chemical feedstock, effects of the Fe/Mn ratio [57,58], the activation procedures and the reaction conditions [59] were investigated. Besides these studies, Chaffee et al. [60] found that the Fe/MnO catalyst shows a superior sulfur resistance than the Fe/Cu/K catalyst in both the FTS reactions using the H2-rich and the COrich syngas. Most of previous works dealing with this type of catalyst focused on its catalytic performance. MnO addition on the surface fine structure (not bulk structure) and the adsorption property of the surface species over the Fe catalyst has never been investigated yet.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Recent Developments of Fischer-Tropsch Synthesis Catalysts...

321

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

To make clear the role of MnO, the authors [61] investigated adsorption properties of surface Fe species by diffuse reflectance FT-IR (DRIFT) spectroscopy using CO as the surface probe molecule. The precipitated Fe and Fe/MnO catalysts were pretreated in a syngas stream at 573 K and 1.2 MPa for 6 h and then the catalyst was immediately cooled to room temperature. The syngas adsorption was performed using the high-pressure syngas stream at room temperature. DRIFT spectra of adsorbed CO were measured at different adsorption times. Figure 6 (A) shows the DRIFT spectra of CO adsorbed on the precipitated Fe catalyst from the syngas adsorption. During the adsorption, weak bands can be observed at 2025 and 1896 cm-1. With increasing the adsorption time, their intensities increase slightly. Different from the Fe catalyst, a large number of distinct bands of adsorbed species are observed at 2010, 1913, 1856 and 1780 cm-1 when the Fe/MnO (1/6) catalyst was subjected to the DRIFT measurement (Figure 6 (B)). Several weak bands can be also observed at 1978, 1945, 1938 and 1896 cm-1. The bands with the frequencies of 2130–2000 cm-1 have been assigned to the linearly adsorbed CO on tops of Fe0 sites and the bands of 2000–1880 cm-1 to the two-fold bridge-bonded CO and those of 1880–1650 cm-1 to the multiply bridge-bonded CO on deep hollow sites [62-66]. From the results mentioned above, mainly three kinds of adsorbed CO are identified on the Fe catalyst, i.e. two linear (2043 and 2025 cm-1) and one bridged (1896 cm-1). These bands are almost consistent with those of C–O vibrations in an iron carbonyl [67]. Furthermore, no bands of CO adsorbed on the deep hollow sites are observed. It may reveal that the formed Fe0 particles are composed of very small Fe0 clusters. In contrast, the specific feature for the Fe/MnO (1/6) catalyst is the large number of wellresolved bands occurring upon CO adsorption. The appearance of the bands arising from the multiply bridge-bonded CO on the deep hollow site of the Fe0 (1778, 1720 and 1662 cm-1) indicates the size of the Fe0 particles is relatively larger. A similar phenomenon was also observed in a Co–MnO system [68]. Because it is reasonable to assume that the increase in the size of the Fe0 particles decreases its reactivity toward H2S, this may be one of reasons for the superior sulfur resistance of the Fe/MnO catalyst.

Figure 6. DRIFT spectra recorded as a function of time on-stream from the syngas adsorption on the Fe (A) and Fe/MnO (B) samples pretreated with the syngas at 573 K and 1.2 MPa for 6 h (from Ref. [61]).

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

322

Naoto Koizumi, Takehisa Mochizuki, Daichi Hongo et al.

Effects of Gas Composition during Catalyst Pretreatment

Gas composition during catalyst pretreatment has impacts on the FTS activity and selectivity of Fe catalysts. Pennline et al. [59] investigated the FTS activity of Fe/Mn/Cu/K pretreated with CO, syngas (H2/CO = 1) or H2. Higher FTS activity was obtained when the catalyst was pretreated with CO. FTS activity and selectivity was also depend upon the temperature and pressure during catalyst pretreatment. Bukur et al. [69,70] found that pretreatment of Fe/Cu/K catalyst with CO or syngas (H2/CO = 0.7 or 2.0) leads to higher initial activity and selectivity for heavier product than the pretreatment with hydrogen; however, hydrogen pretreatments led to more stable catalysts during on stream. It was also reported that, for an ultrafine iron oxide, the FTS activity of carbon monoxide pretreated catalyst was superior to that pretreated with hydrogen or synthesis gas (H2/CO=1.0) at 533k and 0.79 MPa [71]. In the case of pretreatment with syngas, the syngas pressure affects subsequent FTS activity [39,59]. Both 100Fe/3.6Si/0.71K and 100Fe/4.4Si/1.0K catalysts were treated with syngas (H2/CO=0.7) at 3.1 NL·h-1·(g of Fe)-1 and 543 K for 24 h. A dramatic increase in the FTS activity was observed when the syngas pretreatment was conducted at 0.10 MPa, as compared with when the syngas pretreatment was conducted at 1.31 MPa. Such the pressure dependent phenomena were not observed for the catalyst pretreated with CO. It was also reported that initial CO conversion was approximately 65% when 100Fe/3.6Si/0.71K catalyst was pretreated with syngas having low H2/CO ratio (0.1), which was intermediate value between the pretreatment with CO and with syngas having higher H2/CO ratio (0.70). These results indicate that the pressure dependent phenomenon during the syngas pretreatment is related with H2 partial pressure of syngas. Higher H2 partial pressure during the syngas pretreatment would facilitate the reduction of Fe oxide, leading to higher H2O partial pressure. Higher partial pressure of H2O may suppress the formation of Fe carbide(s) during the syngas pretreatment.

Utilization of CO2 as Carbon Source with Fe Catalysts

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

The syngas derived from biomass contains large amount of CO2 [72,73]. Therefore, the utilization of CO2 as carbon source is important for BTL process. In the following, results for CO2 and CO+CO2 hydrogenation with Fe catalysts are summarized.

Fe catalyst vs. Co catalyst Riedel et al. [74] compared the hydrogenation activity of Fe and Co catalysts for the syngas containing CO2. With increasing CO2 in the feed gas, the product composition shifts from an FTS type (mainly higher hydrocarbons) to almost exclusively methane over Co catalyst. Zhang et al. [75] also found that the CO2 hydrogenation products contained about 70% or more methane for supported cobalt. In contrast Fe-based catalyst synthesizes the same hydrocarbon products from CO2/H2 as from CO/H2 [74]. These differences are partly attributed to the fact that Fe catalyst is active for WGS reaction, whereas Co catalyst has little activity for this reaction. Over Fe catalyst, it is assumed that CO2 is hydrogenated into FTS products by two sequential steps: (1) CO is converted from CO2 by the reverse WGS reaction

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Recent Developments of Fischer-Tropsch Synthesis Catalysts...

323

(2) formed CO is then hydrogenated to hydrocarbon [76-79]. Pijolat et al. [80] also suggested that CO is formed not only by the reverse WGS reaction, but also by the partial dissociation of CO2.

CO2 hydrogenation For CO2 hydrogenation, it was suggested that Fe carbides are active Fe phases for the formation of olefins and long-chain hydrocarbons in CO2 hydrogenation [81]. However, Ando et al. [78] found that the major surface phases of the Fe-Cu catalysts were FeO and/or FeCO3 after CO2 hydrogenation. Suo et al. [82] investigated CO2 hydrogenation activity of TiO2-, ZrO2- and Al2O3-supported catalysts. The catalyst with the optimum ratio of iron cations and zero-valent iron gave good catalytic activity and selectivity in the synthesis of C2+ hydrocarbons from CO2 and H2. The different views about the active Fe phase for CO2 hydrogenation indicate that further investigations are necessary to make clear this point. As concerned with promoter effects, H2 is adsorbed on only Fe [83,84] but CO2 adsorbed on both Fe and K [52,83,84]. Higher K content is thus beneficial for CO2 conversion to hydrocarbon [74].

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

CO+CO2 hydrogenation Jun et al. [77] reported effect of H2 partial pressure in feed gas on the conversion of CO and CO2 to hydrocarbon. The reaction with H2-deficient feed (CO/CO2=0.33, H2/(2CO+3CO2)=0.44) showed that only CO was converted to hydrocarbon, whereas, in H2rich feed (CO/CO2=0.33, H2/(2CO+3CO2)=1), CO2 was converted to hydrocarbon as well as CO. Higher concentration of H2 was thought to promote the conversion of CO2 to CO by reverse WGS reaction, followed by FTS reaction, in which CO was further hydrogenated to hydrocarbon. Figure 7 depicts the fraction of 13C involved in CO, CO2, hydrocarbon and oxygenate during the reaction of 13CO2 added H2/12CO feed (508 K, 0.8MPa, H2/CO = 2) [51]. No 13C is detected in CO, suggesting that the reverse WGS reaction is negligible under the reaction conditions. The hydrocarbon products have a negligible 13C content, indicating that CO2 is much less reactive than CO towards chain initiation and growth. Similarly, the addition of 14 CO2 (1.4 mol%) to H2/12CO (1:1) did not lead to detectable 14C contents in CO and hydrocarbons on Fe catalysts (513 K, 0.1 MPa, H2/CO=1) [25]. However, Xu et al. [85] detected almost identical radioactivity per mole in CO2 and in hydrocarbon obtained with FeSi-K catalyst (543 K, 0.8 MPa, H2/CO=0.7), suggesting that each hydrocarbon molecule contained one 14C from 14CO2. Besides, it is reported that CO2 addition to syngas affects the selectivity of FTS products such as C5+ selectivity [51], olefin/paraffin ratio [51,74]. Krishnamoorthy et al. revealed that these phenomena are resulted from the decrease of H2 concentration in the feed gas by the reverse WGS reaction [51]. The effects of CO2 addition to the syngas on the selectivities are much smaller at 508 K than at 543 K.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

324

Naoto Koizumi, Takehisa Mochizuki, Daichi Hongo et al.

Figure 7. 13C fraction in CO CO2, paraffins, olefins and oxygenates during 13CO2 addition runs on the FeZn0.1K0.04Cu0.02 catalyst at 508 K, 0.8 MPa, H2/CO=2, 0.1 MPa 13CO2 [from Ref. [51]).

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

CO CATALYSTS Co catalyst was originally prepared by precipitation method as well as Fe catalyst. Precipitated Co catalysts showed higher FTS activity and, had superior stability during onstream at the reaction temperature higher than 473 K (P ≥ 0.6 MPa) [86]. ThO2 and MgO were frequently used as promoters. However, because of higher market price of Co, precipitated catalysts were not suitable for commercial use. In early 1980’s, Co catalysts supported on SiO2 was found to have higher FTS activity and selectivity at 523 K with a medium syngas pressure (≥ 1.4 MPa) [87]. More recently, the importance of supported Co catalysts has grown up as the catalyst for the conversion of syngas derived from natural gas into liquid hydrocarbon, GTL process. In followings, recent developments of the preparation of supported catalysts are reviewed, which includes metallic Co cluster size effect and effects of Co precursor, support and promoter. FTS activity and selectivity of various Co catalysts reported so far are summarized in Table 1. Finally, our recent results on novel preparation method for Co/SiO2 FTS catalyst using chelating agents are presented.

Metallic Co Cluster Size Effect

Cluster size versus turnover frequency Supported Co catalysts are usually prepared by impregnating high surface area support material with Co precursor solution followed by drying and calcination. Before FTS reaction, the calcined catalyst is subjected to H2 reduced for the activation. Aqueous Co nitrate solution is frequently used as the precursor. It is suggested that Co nitrate is decomposed and oxidized

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

325

Recent Developments of Fischer-Tropsch Synthesis Catalysts...

to Co3O4 during drying and calcinations steps. During subsequent H2 reduction step, Co3O4 is reduced to metallic Co species via CoO. The size of metallic species is reported to be in rage of 20-40 nm on Co/SiO2 [88-91], and 5-10 nm on Co/Al2O3 [92,93]. Coordinatively unsaturated Co atoms (or their ensemble) on the surface of metallic clusters are generally regarded as FTS active sites, although their catalysis is not well understood yet. When we assume spherical and/or hemispherical morphologies of the metallic cluster, the number of coordinatively unsaturated Co atoms (per g-Co) increases with decreasing the cluster size. FTS activity (per g-Co) also increases with decreasing the cluster size unless the FTS activity per coordinatively unsaturated sites (turnover frequency; TOF) is dependent of the cluster size. This assumption is, however, invalid for some catalytic reaction such as dehydrocyclization with Pt catalysts [94]; TOF strongly depends on the cluster size (structure-sensitive reaction). This means that the strategy for catalyst development is dependent on TOF vs. the cluster size relationship. Thus, it is important to understand the cluster size effect in the FTS reaction. Important researches by Iglesia [6,95] and Bezemer et al. [96] concerning the cluster size effect are firstly introduced in this paragraph. Table 1. FTS activity and selectivity of Co catalysts reported in literatures

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Ref. No.

Catalyst composition

Co precursor 88 Co(10)/SiO2a 101 Co(10)/SiO2b 130 NTA-Co(5)/SiO2c Co(20)/CyDTA/Si 137 O2 d Support 108 Co(29)/SBA-15 108 ReCo(20)/SBA-15 110 Co(10)/MCM-41 96 Co(7.5)/CNF 109 Co(20)/SBA-15g Co(12)/ 111 CNF(platelet) 89, 112 Co(10)/SiO2-SiO2h 89, 112 Co(10)/ZrO2-SiO2h Co(10)/Al2O390 SiO2i Promoters

Reaction conditions

CO conv. /%

TOF /10-3 s-1

STY /g kg-cat-1 h-1

Selectivity /%

α

513 K, 1.0 MPa 513 K, 1.0 MPa 503 K, 1.1 MPa

43 67 53

127 66 85

– – 115 (C10-20)

– – 72 (C5+)

0.86 0.93 0.82

503 K, 1.1 MPa

60



815 (C10-20)

77 (C5+)

0.81

493 K, 2.0 MPa 493 K, 2.0 MPa 503 K, 1.0 MPa 523 K, 3.5 MPa 503 K, 2.0 MPa

33.1 43.0 27.1 84 84

– – – 35 –

63 (C5+)e 74 (C5+) 74 (C5+) 74 (C5+) 72 (C5+)

0.86 0.87 – – 0.93

483 K, 2.0 MPa

3.3f

45

82 (C5+)



513 K, 1.0 MPa 513 K, 1.0 MPa

33 86

134 44

– – – 723 (CH2) 350 (C10-20) ca. 480 (CH2) – –

85 (C5+) 83 (C5+)

0.87 0.86

513 K, 1.0 MPa

38.9

121





0.87

100-150 – – (C10-20) 114 Co(5)/Re/TiO2 573 K, 2.1 MPa 81 – – – – 121 Co(10)Mn/CNFj 493 K, 2.0 MPa ca. 60% 60 – 77 (C5+) 0.89 122 Co(7.5)/Mn/TiO2 493 K, 1.8 MPa 4.0k – – 59 (C5+) 0.83 124 Co(10)/Zn/TiO2 493 K, 0.8 MPa 54.6 – – 72 (C5+) – 91 Co(8.5)-Zr/SiO2 463 K, 0.5 MPa 15.9 55 – 58 (C5+) – a Prepared from the solution containing Co acetate and Co nitrate , b Prepared by impregnating SiO2 with Co nitrate solution followed by acetic acid, c W/F=5 g h mol-1, d W/F=1.25 g h mol-1, e hydrocarbon distribution, f 102 mol-Co g-cat-1 h-1, g W/F=2.5 g h mol-1, h Bimodal catalyst, i Bimodal catalyst, j CNF=Carbon nanofiber, k 10-6 mol-Co g-Co-1 s-1 113

Co(20)/Pt/Al2O3

593 K, 2.0 MPa

87



Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

326

Naoto Koizumi, Takehisa Mochizuki, Daichi Hongo et al.

Figure 8. Metallic Co cluster size effect on TOF of Co-based catalysts (■; from Ref. [6,95], ○; from Ref. [96], □; our results).

(1) dCo0 ≥ 10 nm Iglesia [6,95] investigated the FTS activities of Co/Al2O3, Co/TiO2 and Co/SiO2 with various Co loadings at 483 K and 2 MPa. The size of metallic Co clusters determined by hydrogen chemisorption was in 10-250 nm range (11-1% dispersion). It was found that the TOF over these catalysts is constant irrespective of the size of the metallic cluster. This means that CO conversion rate (per g-Co) increases with decreasing the cluster size in this range (Figure 8).

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

(2) dCo0 ≤ 8 nm Until recently, the relationship between TOF and the cluster size was not systematically investigated when the cluster size is smaller than 10 nm. This problem links with general thought that the formation of fine metallic Co cluster on metal oxide supports leads to simultaneous formation of irreducible Co species like Co silicate and Co aluminate species during catalyst preparation and/or FTS reaction due to strong Co-support interaction, which may obscure TOF vs. cluster size relationship. To solve this problem, Bezemer et al. [96] utilized carbon nanofiber (CNF) as catalyst support, and showed that 2-27 nm-sized fine metallic clusters (determined by XPS) are successfully synthesized on this support. These clusters were stable under high-pressure FTS conditions. From the catalystic activity test at 483 K and 3.5 MPa, it was found that TOF increases with increasing the cluster size when the cluster size is below 8 nm (Figure 8). When the cluster size is larger than 8 nm, TOF was constant, and identical to those reported by Iglesia. It was also found that C5+ selectivity increases with the cluster size. According to their results, CO conversion rate (per g-Co) shows a maximum around 6 nm when plotted against the cluster size. The formation of too small clusters is not suitable for obtaining a higher CO conversion rate. They claimed that the cluster size dependency found in their study

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Recent Developments of Fischer-Tropsch Synthesis Catalysts...

327

is different from those observed for the structure sensitive reaction catalyzed by fine metallic clusters. Followings are mechanistic aspect for observed phenomena suggested by Bezemer et al.; Wilson and de Groot [97] observed significant reconstruction on the surface of Co(0001) single crystals after the exposure to the syngas at medium pressure by means of in-situ STM. These authors suggested that such the reconstruction is caused by the formation of certain kinds of mobile Co species such as Co subcarbonyl-like species through the chemisorption of CO during the exposure. When we consider the STM picture of Co(0001) single crystals after the exposure, it is easily imagined that this reconstruction creates various types of surface metallic Co sites with different coordinative unsaturations. On the other hand, FTS reaction over the surface of metallic Co species may require different types of coordinatively unsaturated sites because this reaction is composed of different kinds of elementary step such as the dissociation of H-H and C-O bonds and/or the C-C bond formation. Therefore, the above-mentioned reconstruction seems to be essential for the formation of long-chain hydrocarbons over the surface of metallic Co species. Based on this, it is suggested that the formation of ensemble of coordinatively unsaturated sites required for hydrocarbon formation is suppressed when the cluster size is small. This may be one of reasons for the lower TOF of fine metallic Co clusters. However, such the TOF vs. cluster size relationship has been found only for Co/CNF. TOF over Co/Al2O3, Co/TiO2 and/or Co/SiO2 is still unknown yet when the size of the metallic cluster is smaller than 10 nm. It is probable that different types of Co-support interaction cause different types of cluster size effects. This point is discussed again (see 5.5.).

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Determination of cluster size of fine Co0 particle Size determination of metallic Co clusters, especially fine clusters, is essential to study the cluster size effect in Fischer-Tropsch catalysis. Several characterization techniques have been utilized so far to evaluate the size of the metallic clusters, such as XRD, XPS, TEM, EXAFS and/or hydrogen chemisorption. Hydrogen chemisoprtion is frequently used to evaluate the cluster size of metallic Co cluster. However, little information about the stoichiometry for hydrogen adsorption on metal surface makes it difficult to obtain quantitative results. In the case of TEM observation, it is not so easy to distinguish fine metallic Co form Co oxide. Chemavskii et al. [98] recently tried in-situ characterization of fine metallic Co clusters supported on SiO2 by Foner magnetic method, which can selectively probe metallic clusters. They found fine superparamagnetic, single-domain and multi domain ferromagnetic metallic clusters on reduced catalysts. Superparamagnetic clusters had smaller sizes than single-domain and multi domain ferromagnetic particles. They were found in the catalysts prepared from Co acetate precursor, and/or the catalyst from Co nitrate precursor containing sucrose. The size of superparamagnetic clusters was calculated at 3-6 nm, whereas no definitive information about particle size was provided for these catalysts after calcination by XRD measurements. This method is also suited for in-situ investigation of the formation of fine metallic clusters during H2 reduction.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

328

Naoto Koizumi, Takehisa Mochizuki, Daichi Hongo et al.

Effect of Co Precursor Effect of Co precursor on the FTS activity FTS activity of supported Co catalysts depends on the type of Co precursor used. The use of conventional Co nitrate solution leads to the formation of Co3O4 species after calcination. This specie is almost completely reduced to form metallic Co cluster with lower dispersion. In contrast, the use of Co acetate for the preparation of Co/SiO2 results in the formation of irreducible Co species with higher dispersion after calcination [88,99,100]. The lower reducibility of this Co species is suggested to arise from strong Co-SiO2 interaction like Co silicate. FTS activity of the catalyst from Co acetate is much lower than that of the catalyst from Co nitrate. On the other hand, Sun et al. [88] found that the FTS activity of Co/SiO2 is improved when Co nitrate is co-impregnated with Co acetate. The catalyst prepared with both Co nitrate and Co acetate shows CO conversion ca. 1.4 times higher than the catalyst with Co nitrate alone (42.5 vs. 29.8%). Since Co acetate is known to form highly dispersed, but irreducible Co species as mentioned above, the higher activity induced by co-impregnation is interesting. Later. Liu et al. [101] reported that Co conversion is improved by a factor of 1.3 when SiO2 was impregnated with Co nitrate solution followed by acetic acid solution. Besides, Kraum and Baerns [102] reported FTS activity of Co/TiO2 catalysts prepared from Co-ethylenediamine-N,N,N',N'-tetraacetic acid (EDTA) complex, Co acetylacetonate and Co oxalate. They found that the use of Co oxalate leads to two times higher CO conversion compared with the catalyst from conventional Co nitrate.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Factors determining Co precursor effect So far, several studies have been devoted to make clear Co precursor effect on the structure and catalytic performance of Co-based FTS catalysts. Steen et al. [103] investigated reducibility of variously prepared Co/SiO2 catalysts by temperature programmed reduction (TPR). The reducibility of the catalyst was influenced by various preparation factors such as type of Co precursor, surface area of SiO2 support and/or polarity of solvent. They described interaction between Co ions and silica gel surface during impregnation according to the concept of interfacial coordination chemistry. Ming and Baker [104] also showed that pH pf the impregnation solution affect the structure of Co species on Co/SiO2 catalyst. They explained their results in terms of electrostatic interaction between Co ions and SiO2 surface during impregnation. When pH of the solution is lower than point of zero charge of SiO2 surface (ca. 1-2), SiO2 surface is positively charged. Thus the adsorption of positively charged Co ions is slower, leading to lower dispersion of Co. In contrast, when pH of the solution is higher than point of zero charge such as Co acetate solution, the adsorption of Co ion is favored, leading to higher dispersion. They also suggested that SiO2 surface was partially dissolved at pH higher than 5, which may result in substitution of Si atoms by Co ions at SiO2 surface. More recently, Girardon et al. [105] investigated the structure of Co species during impregnation, drying, calcinations and reduction by spectroscopic techniques such as XANES/EXAFS and XPS. It was shown that coordination structure around Co is almost identical to that of bulk Co acetate after the impregnation of Co acetate solution. Coordination structure was changed to α-Co2SiO4-like structure during calcination (above 443 K). Thus

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Recent Developments of Fischer-Tropsch Synthesis Catalysts...

329

calcination step rather than impregnating step is more important for the speciation of Co after calcination. Because exothermic peak was observed during calcination of the catalyst prepared from Co acetate by DSC-TGA, they suggested that strong Co-SiO2 interaction is caused by the exothermic decomposition of Co acetate complex during calcination. Although the detailed mechanism of Co precursor effect is not fully understood yet, previous studies mentioned above suggest that Co precursor effect links with Co-support interaction during the catalyst preparation (impregnation or calcinations step). This further implies that the use of novel type of Co precursor leads to novel type of Co-support interaction, which may lead to improve the FTS activity and selectivity.

Support Effect

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Recently, various types of mesoporous materials with uniform pore structures, such as MCM-41, MCM-48 and SBA-15, are utilized to support material for Co FTS catalysts by Khodakov et al. [106,107], Martinez et al. [108], Ohtsuka et al. [109] and Li et al. [110] Martinez et al. [108] found that CO conversion over 20% Co/SBA is ca. 1.5 times greater than those on conventional 20% Co/SiO2. The use of SBA-15 led to slightly higher dispersion of metallic Co cluster (9.1 to 11.2%). On the other hand, Ohtsuka et al. [109] reported that 20% Co/SBA-15 yields 350 g kg-cat-1 h-1 of C10-C20 hydrocarbon at 503 K and 2.0 MPa, which is ca. 3 times greater than that reported for commercial catalysts (see Table 1). In their paper, following factors are suggested to be crucial for higher C10-C20 hydrocarbon yield mentioned above; (1) Higher surface area of SBA-15 leads to higher dispersion of metallic Co cluster even at higher Co loading. (2) Large pore diameter of SBA-15 is suitable for attaining higher diffusion rates of syngas and hydrocarbon. Other than mesoporous SiO2, Bezemer et al. [96] and Yu et al. [111] reported the FTS activity and selectivity of Co supported CNF. Apart form above studies, Tsubaki et al. found that modification of large pore SiO2 with ZrO2 [89,112], SiO2 [112] and Al2O3 [90] improves the FTS activity of Co/SiO2 catalyst. Modification with ZrO2 was most effective among them, and led to the FTS activity 5 times greater than original Co/SiO2. They suggested that modification of SiO2 creates smaller pores inside the large pore of original SiO2, which provides porous structure for the dispersion of metallic Co clusters.

Promoter Effect Platinum group metal promoters Platinum group metals are commonly used to promote Co FTS catalyst. Patents for the preparation and use of Co catalysts promoted with Re and Pt are reported by Sasolburg [113] and Exxon Research and Engineering [114]. The addition of platinum group metals greatly improves the rate of CO conversion. However, effect of these metals on the product selectivity is somewhat complex. For example, the addition of Re and Pt to Co/Al2O3 prepared from Co nitrate hardly affects the product selectivity [115,116]. In contrast, Ru and Re addition to Co/Al2O3 prepared from Co acetate suppresses the chain growth of

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

330

Naoto Koizumi, Takehisa Mochizuki, Daichi Hongo et al.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

hydrocarbons [117]. The latter also holds true for Co/SiO2 when Co nitrate is co-impregnated with Co acetate [118]. Platinum group metals are suggested to promote the reduction of Co oxide during H2 reduction by hydrogen spillover mechanism. However, the location of these metals under working state is still unknown yet. Deep understanding of their role in FTS is of importance to improve the FTS activity of the catalyst with smaller amount of rare and expensive platinum group metals. Davis and his co-workers [115,119] showed that the addition of Pt to 15% Co/Al2O3 catalyst lowered the reduction temperatures for both the 1st (Co3O4 to CoO) and 2nd (CoO to metallic Co) reduction steps, whereas Re addition lowers only the temperature for the 2nd reduction step. This is because Re oxide is reduced at higher temperature than Pt and Ru oxides, and approximately the same temperature as the 1st reduction step of Co oxide species. The overall reduction degree of Co oxide species was improved by a factor of 2 to 2.5 by the addition of Pt and Re promoters. The cluster size of metallic Co was also decreased from 5.9 to 5.1 nm by the addition of Re promoter, whereas Pt promoter addition hardly affects the cluster size. TPR results suggest the hydrogen spillover mechanism in these metal promotions, because the reduction of metals is necessary before the reduction of Co oxide species by spillover hydrogen occurs. Fourier transform of in-situ Pt (Re) LIII edge EXAFS spectra of reduced 0.5% Pt (Re)-15% Co/Al2O3 catalyst showed a single peak with ca. 0.06 nm shorter distance than that of Pt-Pt (Re-Re) bond in metallic Pt (Re) reference (Figure 9). The authors assigned this peak to Pt (Re)-Co bond rather than Pt-Pt (Re-Re) bond. It was suggested that isolated metal atoms are located on the surface of metallic cluster in view of hydrogen spillover mechanism.

Figure 9. Re reference (top) and reduced 0.5% Re–15% Co/Al2O3 catalyst k3-weighted (bottom) Fourier transform magnitude spectra. Catalyst reduced at 623 K in situ with hydrogen and cooled under hydrogen to liquid nitrogen temperatures (from Ref. [115]).

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Recent Developments of Fischer-Tropsch Synthesis Catalysts...

331

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Shannon et al. [120] conducted aberration-corrected scanning TEM (STEM) with 0.1 nm resolution or higher and EELS measurements on Co/Al2O3 catalysts promoted with Ru, Ir, Pt and Re to obtain atomic images of these promoter atoms. Aberration-corrected STEM with high-angle annular dark-field (HAADF) detector could obtain atomic image platinum group metal promoters such as Re, Pt and Ir in both calcined and reduced-passivated samples. Some images of Pt and Ir-promoted catalysts are shown in Figure 10. These promoter metals were suggested to sit on or in the surface of particles, because images tend to change in repeated scans. Visualized platinum group metals promoters by HAADF aberration-corrected STEM imaging, however, were much less than expected. EELS image mapping of reducedpassivated sample showed that some metallic Co clusters are associated with Ru clusters. Rufree metallic clusters were also detected. It is of importance that Ru clusters are not observed on Al2O3 surface. This is also true for heavier Re, Ir and Pt atoms. It can be said, even though promoter-free metallic cluster are found, these promoters effectively decrease the reduction temperature of all Co species present. Therefore they also suggested the hydrogen spillover mechanism. Although these authors suggest hydrogen spillover mechanism in platinum group metal promotion, intimate contact of promoter metal atoms with metallic Co clusters implies that alternative mechanism works in the promotion of these metals.

Figure 10. Aberration-corrected STEM/HAADF images of reduced and passivated samples of 0.3% Ptpromoted 20% Co on γ-Al2O3 (a–c) and 0.3% Ir-promoted 20% Co on θ-Al2O3 catalyst (d–f) (from Ref. [120]).

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

332

Naoto Koizumi, Takehisa Mochizuki, Daichi Hongo et al.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Other promoters Besides platinum group metals, B [121], Mn [122-124], Fe [125], Zn [126] and Zr [91,127,128] have been studied as promoters for Co FTS catalyst. Among them, Mn, Zn or Zr addition is reported to improve the FTS activity. The chain growth probability of hydrocarbons is also improved by the addition of Mn or Zr. Morales et al. [124] showed that small amount of Mn addition to Co/TiO2 decreases the size of metallic Co cluster. Their DRIFT measurements combined with CO adsorption suggested that Mn oxide is intimate contact with metallic clusters, and withdraw electron density from metallic Co sites. This may result in suppressing the hydrogenation ability, leading to higher olefin content and higher C5+ selectivity. In contrast, Steen et al. [91] found that Zr addition to Co/SiO2 leads to the formation of metallic Co cluster with lower dispersion and higher reducibility. It was suggested that the formation of ZrO2-SiO2 interaction weakens Co-SiO2 interaction. They also claimed that the formation of larger metallic cluster is beneficial for suppressing the oxidation of metallic cluster by H2O, i.e. catalyst deactivation. Thus higher steady-state activity is obtained. As concerned with method for Zr addition, Seki [128] found that thin ZrO2 layer deposited on SiO2 surface by liquid phase deposition method effectively promotes Co/SiO2 even at lower reaction temperature, which is suitable for the chain growth of hydrocarbon. Chain growth probability of hydrocarbons was reported to reach to 0.93 over this catalyst. Besides these promoters, modification of α-Al2O3 support with Ni was also found to show weak promoting effect for Co/Re system. The use of α-Al2O3 support for Co/Re catalyst instead of γ- Al2O3 slightly increased long chain hydrocarbon selectivity at the expense of mechanical strength and the catalytic activity. Modification of α-Al2O3 support with Ni improved the mechanical strength, and most of the lost activity was regained. Ni was identified as NiAl2O4 by XRD. Arslan et al. [129] investigated the role of Ni (NiAl2O4) by means of STEM tomography. STEM tomography of 20% Co/0.5% Re/γ-Al2O3 catalyst showed that Co oxide agglomerates and fills the alumina pores, forming interlocking oxide/support clusters. In the STEM tomography of 20% Co/0.5% Re/NiAl2O4/α-Al2O3 catalyst, the α-Al2O3 is predominantly composed of large nonporous particles, on which NiAl2O4 forms porous structure. Co oxide selectively wets on the porous NiAl2O4 surface, forming cage-like structure. The surface area of this local porous structure was calculated to 64±5 m2 g-1, which is much greater than the bulk surface area determined by N2 physisorption (11 m2 g-cat-1). STEM tomography clearly showed that NiAl2O4 provides porous structure for the dispersion of Co oxide species.

Preparation of Highly Active Co/SiO2 Catalyst with Chelating Agents

Based on these previous studies on the preparation of Co-based FTS catalysts, it is of importance to make clear the cluster size effect on conventional support such as SiO2 and Al2O3, which is closely related with strategy for catalyst development. Our hypothesis is to control the Co-support interaction during the preparation steps using novel types of Co precursor for the formation of fine metallic clusters. Co complex with chelating agents were chosen as precursors, because the coordination structure and the stability of Co precursor can

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Recent Developments of Fischer-Tropsch Synthesis Catalysts...

333

be changed systematically by using various chelating agents. Followings are summary of our recent results.

Effect of complex formation [130] Co/SiO2 catalyst was prepared using aqueous solution containing Co nitrate and chelating agent or organic acid (L) as the impregnating solution (Co2+ to L molar ratio was unity). In this study, total 7 chelating agents and organic acids with various complex formation constants were used for the preparation to investigate effect of complex formation [131-134] (Table 2). It is noted that Co loading of these catalysts is limited to maximum of 5 mass%, because of lower solubility of the chelating agent. Impregnated sample was then dried and calcined in static air. Their FTS activity and selectivity were investigated at 503 K and 1.1 MPa using fixed bed reactor after H2 reduction at 773 K and 0.1 MPa. Table 2. Logarithmic complex formation constants of organic acids and chelating agents with Co2+ (log KCo) [131-134] used for the preparation of Co/SiO2 catalyst

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Compound (Notation) Organic acid Glycine (Glycine) Citric acid (Citric) L-Asparatic acid (Asparatic) Chelating agents Nitrilotriacetic acid (NTA) Ethylenediamine-N,N,N',N'-tetraacetic acid (EDTA) Trans-1,2-diaminocyclohexane-N,N,N',N'-tetraacetic acid (CyDTA) Triethylenetetramine-N,N,N',N",N"',N"'-hexaacetic acid (TTHA)

log KCo 0.6 5.0 5.9 10.4 16.3 18.9 28.8

In Figure 11, CO conversion (at 20 h on stream) over the reduced catalysts is plotted against logarithmic complex formation constants (log KCo) of the chelating agents and/or organic acids examined here. This figure shows that the conversion shows the maximum around the complex formation constant of around 10 (NTA). Preparation with the chelating agents and/or organic acids having much smaller and/or larger complex formation constants shows negligible and/or negative effects. In other words, the complex formation with moderate stability is important for obtaining the higher FTS activity. Figure 12 depicts the yield of C10-20 hydrocarbons (diesel fraction) obtained with the catalysts prepared using chelating agents in comparison with conventional Co/Pt/Al2O3 catalyst (20 mass%-Co). This figure clearly shows that the yield of C10-20 hydrocarbons obtained with NTA-Co/SiO2 is greater than that with the conventional catalyst, even though Co loading of the former is only one-fourth of that of the later.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

334

Naoto Koizumi, Takehisa Mochizuki, Daichi Hongo et al.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 11. Relationship between CO conversion over reduced catalysts prepared using organic acids and/or chelating agents and their logarithmic complex formation constants with Co2+.

Figure 12. Promoting effect of NTA on the space-time yield (STY) of C10-20 hydrocarbon with Co/SiO2 catalyst (Co/Pt/Al2O3 from ref. [113]).

Metallic Co cluster size effect [130,135] Effect of chelating agents on the formation of metallic cluster was investigated to make clear the role of chelating agents. Surface metallic Co sites were titrated by H2 chemisorption measurements. The amount of H2 chemisorption on the reduced catalysts is summarized in Table 3. From this table, we can see that the preparation of the catalyst using chelating agents and organic acids (except for glycine and TTHA) increases the amount of H2 chemisorption on reduced catalyst. The amount of H2 chemisorption of these catalysts correlates well with their FTS activity shown in Figure 11, showing that the use of the chelating agents and/or

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Recent Developments of Fischer-Tropsch Synthesis Catalysts...

335

organic acids increases the metallic Co surface area, leading to the higher FTS activity. We further calculated the size of metallic Co cluster assuming hemispherical morphology (Table 3). The cluster size of metallic Co is smaller when the catalyst is prepared with chelating agents and/or organic acids except for glycine and TTHA. The complex formation with NTA, EDTA and/or CyDTA leads to the formation of fine metallic Co clusters (1-5 nm size). It is noted that CyDTA-Co/SiO2 shows lower FTS activity than NTA-Co/SiO2, although the preparation using CyDTA leads to most fine metallic clusters. This is due to lower reducibility of Co on CyDTA-Co/SiO2, which leads to lower surface area of metallic Co. Table 3. Effects of organic acids and/or chelating agents on the amount of chemisorbed hydrogen and Co0 cluster size of reduced catalysts.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

a

Co/SiO2 Glycine-Co/SiO2 Citric-Co/SiO2 Asparatic-Co/SiO2 NTA-Co/SiO2 EDTA-Co/SiO2 CyDTA-Co/SiO2 TTHA-Co/SiO2

H2 uptake /μmol-H2 g-cat-1 20.3 20.9 39.4 49.5 65.0 37.9 35.6 8.7

%Co reductiona 85 21 51 49 85 12 8 5

Percentage of Co reduced during H2 reduction at 773 K, 6 h, morphology

b

Co0 cluster sizeb /nm 17.1 4.1 5.3 4.1 5.4 1.8 1.3 2.4

Calculated assuming hemispherical

The formation of fine metallic clusters using chelating agents is important in view of the cluster size effect. Thus Co cluster size effect for reduced L-Co/SiO2 was then investigated, and compared with those reported previously. Figure 8 summarizes the relationship between the size of metallic Co cluster and TOF of L-Co/SiO2, in comparison with those reported by Iglesia [6,95] and Bezemer et al. [96] In general, TOF of our catalysts is higher than those reported by Iglesia and Bezemer et al. This is due to the fact that higher reaction temperature was employed in the present study (503 K) compared with those by Iglesia and Bezemer et al. (483 K). Sun et al. [88] reported that TOF of Co/SiO2 is ca 80 × 10-3 s-1 at 513 K and 1.1 MPa, which is comparable with that of our Co/SiO2 (86 × 10-3 s-1). It is also noted that TOF of our catalysts is independent of the cluster size of metallic Co in 1-17 nm range. Furthermore, TOF of these catalysts is independent of the cluster size of metallic Co in 1-17 nm range. This relationship is evidently different from that for Co/CNF [96]. It is suggested that Co-support interaction is different between SiO2 and CNF, leading to different metallic cluster size effect. The present finding clearly shows that the rate of CO conversion normalized to total weight of Co increases with decreasing the cluster size of metallic Co when all of Co could be reduced to metallic state.

Role of chelating agents [135,136] Role of chelating agents during the catalyst preparation steps was then investigated using various characterization techniques. DRIFT measurement of impregnated and dried catalysts

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

336

Naoto Koizumi, Takehisa Mochizuki, Daichi Hongo et al.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

showed that both NTA-Co and CyDTA-Co complexes are interacted with OH group on SiO2 surface. These complexes were preserved after drying with keeping interaction with OH group. In the following calcination step, these complexes were decomposed. Exothermic peaks were observed with DTA during calcination of the catalysts prepared using NTA and CyDTA, whereas weak endothermic peak was observed in the case of the catalyst without chelating agents. Peak temperature in DTA profiles increased in the order; Co/SiO2 0.8. The air/fuel ratio is varied down to approximately 0.52, where operation is autothermal, i.e. no reactor cooling is required. This ratio is determined by the heat loss of the pilot rig and by the heat withdrawn with the hot gas streams. All these quantities depend on the operating temperature. At an air/fuel ratio of 0.5 and a S/C ratio of approximately 0.4 (considering all loop seal steam from the lower loop seal to enter the FR, what is a worst case scenario and it is likely that the actual S/C ratio in the FR is much lower than 0.4), the H2/CO ratio in the synthesis gas reaches a value of 2.0. To maximize the ratio of H2/CO temperature can be reduced. The energy balance of the reactor system has at any time to be fulfilled which determines some limiting constraints, such as air/fuel ratio, reaction velocity and coke formation. Figure 10und show the gas species H2, CO, CO2, CH4 in the fuel reactor exhaust gas (i.e. in the raw synthesis gas) at two lower temperatures of 798°C and 747°C, respectively. Again, methane conversion is incomplete at higher values of the air/fuel ratio. The other graphs run along the equilibrium lines. In some operating points the thermodynamic equilibrium is reached. At the lowest air/fuel ratio of 0.5 at a temperature of 798°C and a steam/carbon ratio of approximately 0.4, the ratio of H2/CO is increased to 2.4. A further increase in the H2/CO ratio can be seen by lowering the temperature further to approximately 747°C. At an air/fuel ratio of 0.51 and a steam/carbon ratio of 0.4 another 10% increase to 2.6 can be observed. At the lowest

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Chemical Looping Reforming for Syngas Generation from Natural Gas…

387

temperature energy balance and heat loss respectively allow to lower the air to fuel ratio down to 0.46. No significant increase in H2/CO ratio can be seen.

gas concentration [vol%]

75

75 TFR ~903°C

60 45

60 45

H2

30

30

H2O CO

15

15

CO2

0 0.4

CH4 0.6

0.8

1.0

0 1.2

global air to fuel ratio [-]

Figure 9. Wet-gas concentration in the FR exhaust gas at 903°C 140 kW natural gas and S/C =0.4

gas concentration [vol%]

75

75 TFR ~798°C

60

60

45 H2

45

30 H O 2

30

15

CO

15

CO2

CH4 0 0.4

0.6

0.8

1.0

0 1.2

global air to fuel ratio [-]

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 10. Wet-gas concentration in the FR exhaust gas at 798°C 140 kW natural gas and S/C =0.4

It is important to notice that no coke formation has been observed at any of the operating points shown in the graphs. From measurement of CO2 in the air reactor exhaust gas, exact detection of carbon loss to the air reactor is possible. These measurements reveal the following: 1. No gas leakage occurs at any time from the fuel reactor to the air reactor (even though system pressure in the bottom region of the fuel reactor is higher than in the bottom region of the air reactor) 2. The onset of coke formation has been found at global air/fuel ratios as low as 0.4 for practically all the temperatures investigated.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

388

Johannes Bolhàr-Nordenkampf, Tobias Pröll, Philipp Kolbitsch et al. 60

60

gas concentration [vol%]

TFR ~747°C

H2O

45

45

30

30 CO2 H2

15

0 0.4

15

CO CH4 0.6

0.8

1.0

0 1.2

global air to fuel ratio [-]

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 11. Wet-gas concentration in the FR exhaust gas at 747°C 140 kW natural gas and S/C =0.4

The second point represents an unexpected result with respect to the fact that no additional steam is added to the fuel at all. Only the steam from fluidization of the loop seals is present in the fuel reactor. Not only the FR composition but also the oxygen absorption in the AR is an important parameter. It defines the oxygen content transported with the oxygen carrier to the FR. In other words, if in CLR mode oxygen leaves the AR with the gas stream the provided oxygen to the FR does not represent the global air to fuel ratio as defined in the introduction. To investigate this behavior the oxygen content in the AR exhaust gas is plotted in Figure 12 for three different FR temperatures. The reactor temperatures represent mean values over the reactor height. With increasing FR temperature a significant increase in the oxygen absorption is observed. The oxygen content reaches the equilibrium oxygen concentration, i.e. no oxygen in the exhaust gas at an AR temperature of approximately 900°C and an air to fuel ratio below 1. Only where the oxygen is completely consumed in the AR the global air to fuel ratio describes the actually provided oxygen to the FR. In all other cases less oxygen is actually available in the FR than supplied with the air stream. In Figure 13 the AR temperature is shown for the three different constant FR temperatures. The global solids circulation is a strong function of the AR gas velocity, i.e. of the air flow rate introduced in the AR. A temperature difference between AR and FR is necessary to supply the required heat to the FR. Therefore, only one reactor temperature can be kept constant. The AR temperature continuously increases with decreasing air to fuel ratio. With increasing reactor temperature not only the chemical equilibrium changes but also the reactivity of the oxygen carrier. It is interesting to notice that, independently of the FR temperature, complete oxygen absorption is reached at an AR temperature of approximately 900°C. As a consequence, this temperature value can be regarded as the optimum for full oxygen absorption in the AR with this specific oxygen carrier at an air to fuel ratio below 1. However, oxygen carrier particles may be further optimized in order to improve oxygen absorption at lower temperatures. Another point of particle optimization is to decrease the air to fuel ratio where carbon formation starts to be a problem. A comparison of two different Ni-

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Chemical Looping Reforming for Syngas Generation from Natural Gas…

389

based oxygen carriers operated at the 120 kW pilot rig has been performed by Kolbitsch et al. [16].

O2 concentration [vol%]

8

8

TFR 747 [°C] TFR 798 [°C]

6

6

TFR 903 [°C] TFR

4

4

2

0 0.4

2

0.6

0.8

1.0

0 1.2

global air to fuel ratio [-]

Figure 12. Oxygen concentration in the AR exhaust gas at three different FR temperatures

reactor temperature [°C]

1100

1100 AR Temp. [°C] FR Temp. [°C]

1000

1000

900

900

800

800

700 0.4

0.6

0.8

1.0

700 1.2

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

global air to fuel ratio [-]

Figure 13. AR temperature curves at three different FR temperatures

CONCLUSION The 120 kW chemical looping system at Vienna University of Technology has successfully demonstrated chemical looping autothermal reforming at atmospheric pressure. The dual circulating fluidized bed system (DCFB) design allows variation of air/fuel ratio over a wide range without adaptations. For all experimental results high methane conversion is observed. Methane conversion reaches thermodynamic equilibrium for all FR temperatures at a global air to fuel ratio below 0.7. All other species run along the equilibrium curves with significant deviations only where methane conversion is incomplete. The maximum H2/CO

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

390

Johannes Bolhàr-Nordenkampf, Tobias Pröll, Philipp Kolbitsch et al.

ratio in the synthesis gas of 2.6 can be seen at a FR temperature 747°C. Even though no steam has been added to the fuel apart from the loop seal fluidization (what would result in an S/C ratio of 0.4 for the worst case assumption that all steam from the lower loop seal enters the fuel reactor), no coke formation has been observed for global air to fuel ratios above 0.46. Furthermore the oxygen absorption in the AR was investigated. At an AR mean temperature of approximately 900°C the oxygen absorption reaches equilibrium, below this temperature unreacted oxygen leaves the AR with the exhaust gas. Even though CLR was invented already in the 1950-ies, it could obviously not compete with fixed bed reforming systems up to now. The major drawbacks are for sure the necessary dust removal from the product streams, difficulties in making dual fluidized bed systems work at attractive conditions (increased pressure) and the need for attrition-resistant catalyst particles. Today, significant progress has been made in basically all of these fields. There are bag filter systems available for operation at increased temperatures up to 250°C, dual fluidized bed systems have been demonstrated at industrial scale [10] and pressurized circulating fluidized bed combustion systems have been put in operation [7]. The lifetime of the catalyst particle used within this study, manufactured from commercially available raw materials, has been calculated to more than 33000 hours in fluidized bed conditions [19]. Considering these achievements together with the advantages of CLR especially with respect to increased operating temperatures in the reformer and reduced reactor volume, CLR may be an attractive competitor to standard reforming technology in the near future.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

NOMENCLATURE AET-Lib AR CFB CL CLC CLC_SR CLR DCFB FBH-StR FR OC TFR S/C

advanced energy technology library air reactor circulating fluidized bed chemical looping chemical looping combustion Chemical looping combustion steam reforming chemical looping reforming dual circulating fluidized bed fluidized bed heated steam reformer fuel reactor oxygen carrier fuel reactor temperature steam to (organic) carbon

ACKNOWLEDGMENTS This work was part of the EU financed project CACHET (FP6 Contract No. 019972), coordinated by BP. The project is also part of Phase II of the CO2 Capture Project (CCP). The oxygen carrier has been produced by the Flamish Institute for Research and Technology

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Chemical Looping Reforming for Syngas Generation from Natural Gas…

391

(VITO), Belgium under the guidance of Chalmers University of Technology, Sweden in the context of the EU financed project CLC GAS POWER (FP6 Contract No. 019800).

REFERENCES [1]

[2] [3]

[4]

[5]

[6] [7]

[8]

[9]

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

[10]

[11] [12]

[13]

Abad, A., Adanez, J., Garcia-Labiano, F., De Diego, L. F., Gayan, P. & Celaya, J. (2007). Mapping of the range of operational conditions for Cu-, Fe-, and Ni-based oxygen carriers in chemical-looping combustion. Chemical Engineering Science, 62, 1-2, 533–549. Abad, A., Mattisson, T., Lyngfelt, A. & Johansson, M. (2007). The use of iron oxide as oxygen carrier in a chemical-looping reactor. Fuel, 86, 7-8, 1021–1035. Andrus, H. (2007). Chemical looping combustion - r&d efforts by alstom. In IEA GHG 2nd Workshop of the International Oxy-Combustion Research Network (Windsor, USA). Andrus, H. E., Chiu, J. H., Liljedahl, G. N., Stromberg, P. T., Thibeault, P. R. & Jain, S. C. (September 25 - 28 2006). Alstoms hybrid combustion-gasification chemical looping technology development - phase ii. In 23rd Annual International Pittsburgh Coal Conference (Pittasburgh, PA, USA), pp. 20–. Bolhar-Nordenkampf, J., Pröll, T., Kolbitsch, P. & Hofbauer, H. (2009). Comprehensive modeling tool for chemical looping bas processes. Chemical Engineering and Technology, 32, 3, 410–417. Corbella, B. M. & Palacios, J. M. (2007). Titania-supported iron oxide as oxygen carrier for chemical-looping combustion of methane. Fuel, 86, 1-2, 113–122. Dodd, A. M., Dryden, R. J. & Morehead, H. T. (April 1997). McIntosh Unit 4 PCFB Demonstration Project. In American Power Conference (Chicago, Illinois, USA), pp. 562–567. Dybkjaer, I. (1995). Tubular reforming and autothermal reforming of natural gas – an overview of available processes. Fuel Processing Technology, 42, 2-3, 85–107. Trends in Natural Gas Utilisation. Garcia-Labiano, F., Adanez, J., de Diego, L. F., Gayan, P. & Abad, A. (2006). Effect of pressure on the behavior of copper-, iron-, and nickel-based oxygen carriers for chemical-looping combustion. Energy & Fuels, 20, 1, 26–33. Hofbauer, H., Rauch, R., Loeffler, G., Kaiser, S., Fercher, E. & Tremmel, H. (17.-21. June 2002). Six years experience with the FICFB-GASIFICATION process. In 12th European Conference on Biomass and Bioenergy (Amsterdam, The Neatherlands), Eigenverlag, p. 4. Ishida, M. & Jin, H. (1994). A new advanced power-generation system using chemicallooping combustion. Energy, 19, 4, 415–422. Ishida, M., Zheng, D. & Akehata, T. (1987). Evaluation of a chemical-loopingcombustion power-generation system by graphic exergy analysis. Energy, 12, 2, 147– 154. Jerndal, E., Thijs, I., Snijkers, F., Mattisson, T. & Lyngfelt, A. (2008). NiO particles with Ca and Mg based additives produced by spray-drying as oxygen carriers for chemical-looping combustion. accepted for publication in Energy Procedia.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

392

Johannes Bolhàr-Nordenkampf, Tobias Pröll, Philipp Kolbitsch et al.

[14] Johansson, M., Mattisson, T. & Lyngfelt, A. (2006). Use of NiO/NiAl2O4 particles in a 10 kW chemical-looping combustor. Industrial & Engineering Chemistry Research, 45, 17, 5911–5919. [15] Jukkola, G., Liljedahl, G., Nsakala, N. Y., Morin, J. & Andrus, H. (May 22-25 2005). An alstom vision of future cfb technology based power plant concepts. In 18th International Conference on Fluidized Bed Combustion (Toronto, ON, Canada), ASME, pp. 109–120. [16] Kolbitsch, P., Bolhar-Nordenkampf, J., Pröll, T. & Hofbauer, H. (2009). Comparison of two ni-based oxygen carriers for chemical looping combustion of natural gas in 140kw continuous looping operation. submitted to Industrial & Engineering Chemistry Research. [17] Kolbitsch, P., Bolhar-Nordenkampf, J., Pröll, T. & Hofbauer, H. (2009). Design of a chemical looping combustor using a dual circulating fluidized bed (DCFB) reactor system. Chemical Engineering and Technology, 32, 3, 398–403. [18] Lewis, W. K. & Gilliland, E. R. (1954.). Production of pure carbon dioxide. U.S. Patent Office, Number 2,665,972. [19] Linderholm, C., Mattisson, T. & Lyngfelt, A. (2009). Long-term integrity testing of spray-dried particles in a 10-kw chemical-looping combustor using natural gas as fuel. Fuel In Press, Uncorrected Proof. [20] Lyngfelt, A., Leckner, B. & Mattisson, T. A (2001). fluidized-bed combustion process with inherent CO2 separation; application of chemical-looping combustion. Chemical Engineering Science, 56, 10, 3101–3113. [21] Mattisson, T., Johansson, M. & Lyngfelt, A. (2006). The use of NiO as an oxygen carrier in chemical-looping combustion. Fuel, 85, 5-6, 736–747. [22] Mattisson, T. & Lyngfelt, A. (2001). Applications of chemical-looping combustion with capture of CO2. In Second Nordic Minisymposium on Carbon Dioxide Capture and Storage (Goeteborg, Sweden), pp. –. [23] Paisley, M., Farris, M., Black, J., Irving, J. & R. P. O. (2000). Preliminary operating results from the battelle/ferco gasification demonstration plant in burlington, vermont, usa. In 1st World Conference on Biomass for Energy and Industry (Sevilla, Spain). [24] Perz, E. (1991). A computer method for thermal power cycle calculation. Journal of Engineering for Gas Turbines and Power, 113, 2, 184–189. [25] Pröll, T. & Hofbauer, H. (2008). Development and application of a simulation tool for biomass gasification based processes. International Journal of Chemical Reactor Engineering, 6. [26] Pröll, T., Rupanovits, K., Kolbitsch, P., Bolhar-Nordenkampf, J. & Hofbauer, H. (2009). Cold flow model study on a dual circulating fluidized bed (DCFB) system for chemical looping processes. Chemical Engineering and Technology, 32, 3, 418–424. [27] Richter, H. J. & Knoche, K. F. (1983). Reversibility of combustion processes. ACS Symposium Series 235, 71–85. [28] Ryden, M. & Lyngfelt, A. (2004). Hydrogen and power production with integrated carbon dioxide capture by chemical-looping reforming. In 7th Conference on Greenhouse Gas Control Technologies (Vancouver, Canada). [29] Ryden, M. & Lyngfelt, A. (2006). Using steam reforming to produce hydrogen with carbon dioxide capture by chemical-looping combustion. International Journal of Hydrogen Energy, 31, 1271–1283.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Chemical Looping Reforming for Syngas Generation from Natural Gas…

393

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

[30] Ryu, H., Bae, D. & Jin, G. (2006). Chemical-looping combustion process with inherent CO2 separation; reaction kinetics of oxygen carrier particles and 50kwth reactor design. In The World Congress of Korean and Korean Ethnic Scientists and Engineers, Seoul, Korea, pp. 738–743. [31] Son, S. R. & Kim, S. D. (2006). Chemical-looping combustion with NiO and Fe2O3 in a thermobalance and circulating fluidized bed reactor with double loops. Industrial & Engineering Chemistry Research, 45, 8, 2689–2696. [32] Van Der Drift, B., Van Ree, R., Boerrigter, H. & Hemmes, K. (May 2004). Bio-syngas: Key intermediate for large scale production of green fuels and chemicals. In 2nd World Conference on Biomass for Energy, Industry and Climate Protection (Rome, Italy), vol. Vol. II, pp. pp. 2155–2157.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved. Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

In: Syngas Production Methods, Post Treatment… Editors: Adorjan Kurucz and Izsak Bencik

ISBN: 978-1-60741-841-2 © 2009 Nova Science Publishers, Inc.

Chapter 10

PARTIAL OXIDATION OF METHANE OVER ZIRCONIA- AND MAGNESIA-SUPPORTED RUTHENIUM AND RHODIUM CATALYSTS Maria do Carmo Rangel1*, Marluce Oliveira da Guarda Souza2, Dino de Jesus Sodré1, André Leopoldo Macêdo da Silva3, Márcia Souza Ramos1 and José Mansur Assaf3 1

GECCAT,Instituto de Química, Universidade Federal da Bahia, Campus Universitário de Ondina, Federação 40 170-280, Salvador, Bahia, Brazil 2 Departamento de Ciências Exatas e da Terra-Campus I–Universidade do Estado da Bahia, Brazil 3 Laboratório de Catálise. Departamento de Engenharia Química, Universidade Federal de São Carlos, São Paulo, Brazil

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

ABSTRACT Hydrogen and syngas are key feedstocks for several industrial processes such as hydrotreating and hydrocracking in addition methanol, ammonia and Fischer-Tropsch synthesis. For these applications, different H2/CO molar ratios of syngas are required. Among the various routes for obtaining hydrogen and syngas, steam reforming is by far the most commonly-used one in commercial practice. However, this reaction has some drawbacks, such as the high endothermicity and catalyst deactivation. The partial oxidation of methane, however, is an exothermic reaction, making it an attractive option for obtaining hydrogen and syngas. In order to find alternative catalysts to the reaction, ruthenium- and rhodium-based catalysts were compared, using zirconia and magnesia as supports; also, the effect of the addition of small amounts of magnesia (Mg/Zr (molar)= 0.1) on zirconia-based catalysts was studied. The supports were prepared by precipitation and then impregnated with ruthenium or rhodium chloride to get solids with 1% of metal. Samples were characterized by differential thermal analysis, thermogravimetry, X-ray diffraction, thermoprogrammed reduction and nitrogen adsorption. The catalysts were *

Corresponding author: E-mail: [email protected]

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

396

Maria do Carmo Rangel, Marluce Oliveira da Guarda Souza et al. evaluated in the partial oxidation of methane carried out at 1 atm and in the range of 450 to 750°C. Monoclinic and tetragonal phases were detected in the zirconia-based solid, while magnesium-doped zirconia showed only the tetragonal phase. Magnesia showed the cubic phase typical of magnesium oxide. Magnesium decreased the specific surface area of zirconia-based catalysts, regardless of the kind of metal; it also made the rhodium and ruthenium reduction more difficult. Zirconia was found to be a more suitable support for the catalysts, in the range of 450 to 750°C, than magnesia and magnesium-doped zirconia, probably due to its ability in favoring well-dispersed metal particles. In a general tendency, ruthenium produces more active and selective catalysts than rhodium and thus the most promising catalyst was zirconia-supported ruthenium. All catalysts produced carbon dioxide, showing that the reaction occurred through indirect oxidation, the amount depending on the metal, the support and on the reaction temperature. Most of the catalysts produced H2/CO molar ratio values of 2 above 550°C, being suitable for methanol and Fischer-Tropsch synthesis; on the other hand, the rhodium and magnesiumcontaining catalyst is more suitable for producing hydrogen. Therefore, rhodium and ruthenium catalysts can be tailored to produce different H2/CO molar ratios for several applications.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

INTRODUCTION Hydrogen is expected to play an important role in the future, powering high-efficiency electricity generating systems and thus reducing greenhouse gas emissions. In addition, its use in petroleum refining has been growing rapidly because of several factors, including those related to changes in environmental regulations. They impose limits of sulfur in diesel and of NOx and SOx in off-gas emissions to the atmosphere and aromatic and light hydrocarbon concentration in the gasoline, among others [1, 2]. Therefore, hydrogen plants had been driven to supply the demand for ever more stringent desulphurization and fewer aromatics in fuels. On the other hand, syngas (a mixture of hydrogen and carbon monoxide) is an important feed for Fischer-Tropsch, methanol, ammonia and dimethyl ether synthesis. Syngas production is essential for commercial Fischer-Tropsch synthesis and is responsible for around 60% of the investments [3, 4]. The technologies for the production of both hydrogen and syngas are generally based on hydrocarbon feedstock through three processes: steam reforming, partial oxidation and autothermal reforming. Among them, the oldest and most economical route to produce hydrogen and syngas is the steam reforming of methane (Equation 1), which covers a wide capacity range (10,000 to 100,000 Nm3.h-1) [2, 3, 5]. However, this process still has some drawbacks since the reaction is highly exothermic and suffers from heat transport limitation, requiring contact times on the order of seconds with large energy input at high temperature [4]. This implies high energy and capital costs for maintaining the reaction conditions of superheated steam and high temperature and pressure [5]. Furthermore, the water gas shift reaction produces significant concentrations of carbon dioxide in the product gas, and the H2/CO ratio is higher than the optimum value of 2 required for the downstream synthesis gas conversion to hydrocarbons [5, 6]. The deactivation of the nickel-based catalysts by coke and by sintering of the metal particles is another problem [6]. CH4 + H2O

CO + 3H2 ΔH°298 K = + 226 kJ.mol-1

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

(1)

Partial Oxidation of Methane over Zirconia- and Magnesia-Supported …

397

One of the alternatives to methane steam reforming is the catalytic dry reforming of methane with carbon dioxide (Equation 2), which produces syngas with a lower H2/CO ratio and has several advantages such as the reuse of carbon dioxide and a possible means of energy storage and transmission, besides the possibility of using two abundant concomitant gases in higher value products more easily transported [7, 8]. However, this route showed the inconvenience of the rapid deactivation of the catalyst, due to carbon deposition [4]. 2CO + 2H2 ΔH°298 K = +261 kJ.mol-1

CH4 + CO2

(2)

A more advantageous alternative to produce syngas is the catalytic partial oxidation of methane (Equation 3), a mild exothermic reaction that can produce a gas with an H2/CO ratio of 2 [5, 8, 9]. Due to the exothermicity of the reaction, the energy required is around 10–15% less and the capital investment can be reduced by 25–30%, as compared to steam reforming [9, 10]. Furthermore, the oxidation reactions are also much faster than the reforming reactions, suggesting that a single stage process can be used [10]. In this case, high methane conversion and hydrogen selectivity with short contact times of around 10-3 s can be achieved in single and smaller reactor, without external heating [11, 12]. However, this process has some drawbacks that should be overcome before it can be developed in industrial scale. These problems include the co-feeding of methane and oxygen under explosive conditions and the production of hot spots in the catalyst and in the reactor, as well as carbon deposition [10, 13]. Therefore, the solutions for these engineering problems as well as the improvement of the catalyst are much needed before the process can be commercialized. CO + 2H2 ΔH°298 K = - 44 kJ.mol-1

CH4 + ½O2

(3)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

The reaction pathway of methane partial oxidation is still debated and two mechanisms are accepted: (i) the direct oxidation (Equation 3), by which hydrogen is directly produced from the dissociation of methane and carbon monoxide is formed by the reaction between surface oxygen and surface carbon species coming from methane decomposition and (ii) the indirect oxidation involving the methane combustion (Equation 4) to produce carbon oxides and water, followed by steam or carbon dioxide reforming (Equations 1 and 2) and the water gas shift reaction (Equation 5) to give the synthesis gas [5, 14]. CH4 + 2O2

CO2 + 2H2O

ΔH°298 K = - 802.3 kJ.mol-1

(4)

CO + H2O

CO2 + H2

ΔH°298 K = -41.2 J.mol-1

(5)

However, the typical reaction systems involves more complex reaction paths and a critical factor for succeeding is to kinetically inhibit reactions, such as combustion (Equation 4), methanation (Equation 6), cracking (Equation 7), the Boudouard reaction (Equation 8) and the water gas shift reaction (Equation 5) [15, 16]. CO + 3H2 CH4 2CO

CH4 + H2O C + H2 C + CO

ΔH°298 K = -206.2 kJ.mol-1

(6)

ΔH°298 K = -74.9 kJ.mol-1

(7)

ΔH°298 K = -172.4 kJ.mol-1

(8)

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

398

Maria do Carmo Rangel, Marluce Oliveira da Guarda Souza et al.

Several catalysts have been employed in the partial oxidation of methane, mainly platinum group metals on various supports, such as rhodium, ruthenium, palladium, platinum and iridium [12, 17-20], besides nickel and cobalt [15, 21, 22]. Some perovskite type oxides [23, 24] have also been tested. Among these catalysts, noble metals are more active and have been widely used [16]; however, the high price of such systems limits their widespread industrial application. Nickel is an inexpensive metal but shows a tendency of sintering and needs to be stabilized [16]; in addition, it rapidly goes on deactivation by coke deposition [4, 21]. Various strategies have been proposed to overcome these drawbacks. Many additives were successfully used to decrease carbon deposition such as lithium, lanthanum, potassium and sodium [25]. The use of different supports with basic sites, like calcium oxide, silica and magnesia [26] or reducible supports [27] was also reported. Pure supports, such as ceria and zirconia, have been shown to be especially efficient for decreasing coke deposition by carbon gasification due to their oxygen storage capacity [28]. Supported noble metals catalysts are mainly used in the partial oxidation of methane, including rhodium and ruthenium [12, 17, 29, 30]. It has been recognized that the support is an important factor in determining the activity, selectivity and stability of the catalysts. In the present work, ruthenium and rhodium-based catalysts were compared, using zirconia and magnesia as supports. The effect of the addition of small amounts of magnesia on zirconiabased catalysts was also studied.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

EXPERIMENT Magnesium-doped zirconia (Mg/Zr (molar) = 0.1) was prepared by adding an ammonium hydroxide solution (1.22 mol.L-1) to a solution containing zirconium oxychloride and magnesium nitrate (1.79 and 0.32 mol.L-1, respectively) at a constant flow rate, under stirring, at 50°C. After the complete addition of the reactants, the pH value was adjusted to 10 by adding an ammonium hydroxide solution (4.59 mol.L-1). The sol produced was kept at this temperature for 4 h, under stirring. Then, the gel was isolated by centrifugation, rinsed with water and centrifuged again. The oxide precursor was dried at 120°C for 12 h and then calcined under air flow (100 mL.min-1), at a rate heating of 5°C.min-1 from room temperature up to 200, 400 and 600°C successively; the solid was kept for 2 h at each temperature. The procedure was repeated to get pure magnesia and pure zirconia. The solids thus obtained were then impregnated using ruthenium chloride or rhodium chloride solutions. In the first case, 3.5 mL of an aqueous solution (0.14 mol.L-1) were used for 5 g of support in order to get 1% (w/w) of metal in the catalyst; in the other one, 8 mL of an aqueous solution (0.06 mol.L-1) were used. The supports were dispersed in the metal precursor solution and kept at 80°C under stirring, for 2 h, in order to evaporate the excess of impregnating solution. Then, the solids were dried at 120°C, for 12 h. The catalyst precursors were calcined under air flow (100 mL.min-1) at a rate heating of 5°C.min-1 from room temperature up to 400, 600 and 750°C; the solid was kept for 2 h at each temperature. The support precursors were characterized by differential thermal analysis (DTA) and thermogravimetry (TG) and the supports were analyzed also by X-ray diffraction (XRD) and nitrogen porosimetry. The catalysts were also characterized by temperature programmed reduction (TPR).

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

399

Partial Oxidation of Methane over Zirconia- and Magnesia-Supported …

For the differential thermal analysis and thermogravimetry experiments, the sample (0.005 g) was heated at 10°C.min-1, under air flow from 30 to 1000°C, in a Metter Toledo model TGA/SDTA851 equipment. X-ray diffractograms were obtained using a model Shimadzu XRD600 equipment with a nickel filter. The sample was exposed to CuKα radiation (λ = 1.5420 A) and then a scanning (2 degrees.min-1) in the range of 2θ of 10 to 80° was carried out. The specific surface area (BET method) and porosity measurements were performed in a model ASAP 2020 Micromeritics apparatus on 0.30 g of sample, which was heated under vacuum (10 µmHg) at 200°C for 30 min and then to 1µmHg, before the nitrogen adsorption. Temperature programmed reduction profiles were obtained using a Micromeritics model TPD/TPR 2900 equipment. In the analyses, the catalyst (0.30 g) was placed into a quartz cell and heated under nitrogen flow (30 ml.min-1) up to 160°C, at a rate of 10°C.min-1 and kept at this temperature for 30 min. The sample was then cooled to room temperature and heated again at a rate of 10°C.min-1 up to 1000 ºC, under a 5% H2/N2 flow mixture (60 ml.min-1). The partial oxidation of methane was carried out using a fixed bed continuous flow reactor loaded with 0.1 g of catalyst and a constant gas flow rate of methane (40 mL.min-1) and oxygen (20 mL.min-1). The experiments were performed at temperatures of 750, 650, 550 and 450°C in this order and at 1 atm. In order to study the catalysts stability, samples were evaluated at 750°C, for 6 h. Prior to the experiments, the samples were reduced in situ at 600°C, for 2 h under hydrogen flow (60 mL.min-1). 100

M

Weight loss (a.u.)

ΔT (a.u.)

Z

ZM

90 ZM Z

80 70 60

M 50

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

200

400

600

Temperatur e (°C)

800 1000

200

400 600 800 1000 Temperatur e (°C)

(a)

(b)

Figure 1. Curves of (a) differential thermal analysis and (b) thermogravimetry for the support precursors. Z sample: zirconium-based solid; M sample: magnesium-based solid and ZM sample: zirconium and magnesium-based solid.

RESULTS AND DISCUSSION The curves of differential thermal analysis of the catalyst precursors are illustrated in Figure 1(a). We can see, in all cases, a peak below 200°C associated with the first weight loss

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

400

Maria do Carmo Rangel, Marluce Oliveira da Guarda Souza et al.

in the thermogravimetric curves shown in Figure 1(b), which is assigned to the loss of volatiles adsorbed on the solids. Above this temperature, the peaks can be related to oxide formation. For the magnesium-based sample, the endothermic peak is associated to magnesia production, which is followed by the loss of water [31]. On the other hand, the zirconiumbased sample showed an exothermic peak, centered at 420°C, due to zirconia formation [32], this transformation is followed by a slight weight loss. This peak was shifted to a higher temperature for the magnesium and zirconium-based samples, indicating that magnesium made the zirconia production more difficult.

Intensity (a.u.)

Z

M

ZM 10

20

30

40 50 60 2 θ (degrees)

70

80

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 2. X-ray diffractograms for the catalyst supports. Z= zirconia; M= magnesia; ZM= zirconia with Mg/Zr= 0.1.

From the X-ray diffractograms of the supports (Figure 2), it can be observed that magnesia showed the cubic phase typical of magnesium oxide, MgO (JCPDS 87-0652). Zirconia presented monoclinic (JCPDS 86-1451) and tetragonal (JCPDS 88-1007) phases, while magnesium-doped zirconia showed only the tetragonal one. Therefore, magnesium stabilizes the tetragonal phase of zirconia, as found in previous works for several dopants [3239]. It is well-known [34, 35] that zirconia exhibits three polymorphs, monoclinic (m-ZrO2), tetragonal (t-ZrO2) and cubic (c-ZrO2), only the monoclinic phase is thermodynamically stable at room temperature while the other phases are stable above 1170 and 2370°C, respectively. However, the high-temperature polymorphs of zirconia may exist in either of the following states: metastable (often sustained by the co-existence of a true phase), stabilized (by foreign agents such as Ce4+, Ca2+, Mg2+, Y3+, Sc3+, sulfate, phosphate and others) and strained (due to high temperature or while being transformed) state [36]. Several different explanations for this metastability have been proposed and include considerations of surface and strain energy [37, 38], chemical [38] and kinetics effects [39]. One explanation widely accepted was given by Garvie [37], who proposed a “crystal size theory”. According to this view, the stability of a single crystallite of zirconia is determined by the sum of the free energy from bulk, surface and strain effects. From his analysis, microcrystals below a critical size of about 30 nm are stabilized against the tetragonal to monoclinic transformation because

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

401

Partial Oxidation of Methane over Zirconia- and Magnesia-Supported …

of the lower surface energy of the tetragonal phase and then this phase is stable at room temperature For the crystals above this critical size, the surface energy is not enough to compensate the thermodynamic driving force to form the monoclinic phase, these particles are then susceptible to phase change. The argument can be extended to the chemical effect as a result of introducing dopants which alter the bulk free energy term. Therefore, if cationic dopants (Ca2+, Mg2+, Y3+) are introduced into zirconia the stability of the tetragonal phase is increased. Other explanation considers that this stabilization is due to the introduction of oxygen vacancies, which provide stability for tetragonal and cubic structures; they are generated by the dopants replacing zirconium atoms [29]. After the metal impregnation, no change was noted for the diffractograms and no ruthenium or rhodium-containing phase was noted, probably due to the small amounts of these metals in solids. The addition of small amounts of magnesium did not affect the specific surface areas of zirconia, as shown in Table 1. As found previously [40], the surface energy increase, related to the stabilization of the tetragonal phase, may occur without an increase in surface area. In fact, it has been proposed that small amorphous particles of zirconia crystallize and coalesce into a larger zirconia particle with all, at least the dominant fraction, of the small particles forming individual distinct crystals. Therefore, zirconia undergoes this transformation to produce a large particle comprised of numerous tetragonal-phase domains. For the magnesium-doped sample, in the present work, the stabilization of the tetragonal phase is not accompanied by an increase of specific surface area, suggesting that the particles are made off of small domains of the tetragonal phase. On the other hand, the metal addition (ruthenium or rhodium) led a decrease in specific surface area of the catalysts, in all cases, a fact which can be related to the blockage of pores by the metals. The presence of magnesium decreased the specific surface area of zirconia-based catalysts, regardless the kind of metal. Table 1. Specific surface area (Sg) of the supports and of the catalysts. Z= zirconia; M= magnesia and ZM= zirconia with Mg/Zr= 0.1. Sample

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Z ZM M

Sg (m2.g-1) 110 106 32

Sample RhZ RhZM RhM

Sg (m2.g-1) 51 36 27

Sample RuZ RuZM RuM

Sg (m2.g-1) 50 34 19

The nitrogen adsorption/desorption isotherms for different samples at 77 K are shown in Figure 3. Magnesia displayed typical isotherms of macroporous materials (Type II) with an insignificant hysteresis loop while zirconia showed curves typical of mesoporous materials with a hysteresis loop (Type IV) [41]. It can be noted that the addition of small amounts of magnesium changed the profile of the isotherm, generating a profile typical of a less porous solid in which less nitrogen is adsorbed. The pore size distributions of the samples are shown in Figure 4. Magnesia displayed a bimodal distribution related to mesopores and macropores with two maximum values at around 4 and 60 nm, respectively. On the other hand, the zirconia-based solids showed unimodal distributions; for pure zirconia the maximum occurred at around 7 nm while for magnesium-doped zirconia it occurred at 4 nm.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

402

Maria do Carmo Rangel, Marluce Oliveira da Guarda Souza et al.

The TPR curves of ruthenium-based catalysts displayed a peak around 100-150°C in all cases (Figure 5a), which can be assigned to the reduction of well dispersed RuOx species [12]. The reduction profiles of the zirconia-containing samples also showed a peak around 300°C, attributed to the reduction of ruthenium oxide (RuO2) particles [12]. These peaks were shifted to higher temperatures due to small amounts of magnesium, indicating that magnesium, in zirconia matrix, made the ruthenium reduction more difficult. In addition, a third peak appeared for the zirconium and magnesium-containing sample indicating a stronger interaction of the metal with the support.

VN2 (cm3 STP.g-1)

200 160 120 80 40 0.0

0.2

0.4

1600 1200 800 400 0

0.0

0.8

1.0

1200 VN2 (cm3 STP.g-1)

VN2 (cm3 STP.g-1)

2000

0.6 P/Po (a)

0.2

0.4 0.6 P/Po (b)

0.8

1.0

800 400 0 0.0

0.2

0.4 0.6 P/Po

0.8

1.0

(c)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 3. Nitrogen isotherms for the catalyst supports. (a) M= magnesia; (b) Z= zirconia and ZM= zirconia with Mg/Zr= 0.1.

Concerning the rhodium-based catalysts (Figure 5b), magnesia-supported rhodium showed a single peak at around 300°C and a broad peak at higher temperatures, in agreement with previous works [30]. The first one is assigned to the reduction of rhodium oxide in weak interaction with the support while the other is due to the reduction of rhodium in strong interaction with the support. On the other hand, the rhodium-supported zirconia showed a different reduction profile, with three peaks, indicating several kinds of metal-support interactions. By adding small amounts of magnesium to zirconia (RhZM sample), the first two peaks overlapped and were shifted to higher temperatures, indicating that magnesium, in zirconia matrix, made the rhodium reduction more difficult. The catalytic evaluation showed that all samples were active in the partial oxidation of methane (Figure 6) in the range of 450 to 750°C. The methane conversion over rutheniumbased catalysts increased continuously with temperature, but this increase was much slower in the case of magnesium-doped zirconia one. On the other hand, the methane conversion over

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

403

Partial Oxidation of Methane over Zirconia- and Magnesia-Supported …

0.25 0.20 0.15 0.10 0.05 0.00

100 10 Pore Diameter (nm) (a)

1

0.8 0.6 0.4 0.2 0.0 1

10 Pore Diameter (nm)

100

Pore Volume (cm3.g-1 )

Pore Volume (cm3.g-1 )

Pore Volume (cm3.g-1 )

the rhodium-based catalysts did not show any effect of temperature until 550°C and increased sharply at higher temperatures.

0.6 0.4 0.2 0.0 1

(b)

10 Pore Diameter (nm) (c)

100

RuZ RuM

RuZM

-50000

25

200 400 600 800 1000 Temperature ( °C) (a)

Hydrogen consumption (a.u.)

Hydrogen consumption (a.u.)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 4. Pore volume distribution for the catalyst supports. (a) M = magnesia; (b) Z = zirconia and ZM = zirconia with Mg/Zr = 0.1.

100 00025

RhZ RhM

RhZM

200 400 600 800 1000 Temperatu re (°C) (b)

Figure 5. TPR curves of (a) ruthenium and (b) rhodium-based catalysts. Z = zirconia; M = magnesia and ZM = zirconia with Mg/Zr = 0.1.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

404

Maria do Carmo Rangel, Marluce Oliveira da Guarda Souza et al.

90

CH4 conversion (%)

CH4 conversion (%)

For the ruthenium-based catalysts, zirconia led to the highest methane conversion followed by magnesia, the addition of small amounts of magnesium to zirconia decreased these values even more. This behavior can be related to the effect of magnesium in affecting the ruthenium reduction. As found by TPR, the zirconia-supported sample showed the most dispersed particles followed by the magnesia-supported one. A different profile was noted for the rhodium-based samples, whose effect of the support showed a dependency with the reaction temperature. In all cases, the methane conversion increased with temperature and no catalyst deactivation was found in this temperature range. The stability of the catalysts for 6 h of reaction time at 750°C is displayed in Figure 7. It can be observed that all catalysts were stable under this condition leading to similar conversions, except the zirconia-supported rhodium which showed the lowest value.

70 50 30 10

90 70 50 30 10

450

550 650 750 Temperatur e (ºC)

450

550 650 Temperatur e (ºC)

(a)

750

(b)

90 85



p

80



p

• p• p• p•

p

p

p

• •

p

CH4 Conversion (%)

CH4 Conversion (%)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 6. Methane conversion as a function of temperature during the partial oxidation of methane over (a) ruthenium and (b) rhodium-based catalysts. Z (S) = zirconia; M ( ) = magnesia and ZM (•) = zirconia with Mg/Zr = 0.1.

p

• •n

75 0

100

200 300 Time (min) (a)

400

90 85

• •

• • • • • • • •

80 p

p

p

p p

p

p

p

p

p

75

0

100 200 300 Time (min) (b)

400

Figure 7. Methane conversion as a function of temperature during the partial oxidation of methane over (a) ruthenium and (b) rhodium-based catalysts. Z (S) = zirconia; M ( ) = magnesia and ZM (•) = zirconia with Mg/Zr = 0.1.

Table 2 shows the molar composition of the reactor effluent. It can be noted that the ruthenium-based catalysts produce more hydrogen than the rhodium-based ones, regardless the support. The same behavior was noted for the production of carbon monoxide, except for

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

405

Partial Oxidation of Methane over Zirconia- and Magnesia-Supported …

the magnesium and zirconium-containing samples. All catalysts produced carbon dioxide, indicating that the partial oxidation of methane occurred according to the mechanism of indirect oxidation. However, these amounts largely varied with the metal and the support and with the reaction temperature, showing different activities of the catalysts towards the water gas shift reaction. The majority of the catalysts produced similar H2/CO molar ratio of 2 above 550°C, showing that they are suitable for methanol and Fischer-Tropsch synthesis. Only the rhodium and magnesium-containing catalysts showed values higher than 2 showing that they are more suitable to produce hydrogen. Table 2. Amount (% molar) of hydrogen, carbon monoxide and carbon dioxide and H2/CO in the reactor effluent for the catalysts in the partial oxidation of methane as a function of temperature. Z = zirconia, M = magnesia and ZM= zirconia with Mg/Zr= 0.1

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Sample 450°C H2 550°C 650°C 750°C 450°C CO 550°C 650°C 750°C 450°C CO2 550°C 650°C 750°C 450°C H2/CO 550°C 650°C 750°C

RuZ 21 27 73 83 45 58 33 37 0.63 1.9 4.5 6.6 3 2 2 2

RuM 15 21 66 79 37 49 29 34 7.2 5.6 2.7 0.82 3 2 2 2

RuZM 9.0 11 39 43 29 33 13 16 7.5 6.6 5.7 4.0 3 2 2 2

RhZ 18 22 52 81 43 50 27 35 7.1 6.1 4.2 1.4 3 2 2 2

RhM 12 16 57 86 34 39 26 38 3.6 5.2 14 26 3 3 2 2

RhZM 11 20 68 86 31 42 32 39 0.76 3.2 6.8 8.0 5 3 2 2

CONCLUSION The addition of small amounts of magnesium (Mg/Zr (molar) = 0.1) to zirconia stabilizes the tetragonal phase but does not alter the specific surface area of the solid, which is believed to be made of small domains of tetragonal phase. On the other hand, the impregnation of ruthenium and rhodium decreases the specific surface area of this support, as well as of pure zirconia or magnesia, but does not alter the type of phases. Magnesium also makes rhodium and ruthenium less reducible in zirconia matrix. Zirconia has been demonstrated to be a more suitable support to the catalysts for the partial oxidation of methane in the range of 450 to 750°C as compared to magnesia and magnesium-doped zirconia, a fact that was assigned to its ability in promoting well-dispersed metal particles. As a whole, ruthenium produces more active and selective catalysts than

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

406

Maria do Carmo Rangel, Marluce Oliveira da Guarda Souza et al.

rhodium and thus the most promising catalyst was zirconia-supported ruthenium. All catalysts produced carbon dioxide, showing that the reaction occurred through indirect oxidation, the amounts depending on the metal, the support and on the reaction temperature. Most of the catalysts produced H2/CO molar ratio values of 2 above 550°C, showing that they are suitable for methanol and Fischer-Tropsch synthesis. Only the rhodium and magnesium-containing catalyst showed a value higher than 3, showing that it is more suitable for producing hydrogen. All catalysts were stable in the range of 450 to 750°C. It can be concluded that the rhodium and ruthenium catalysts can be tailored to obtain different H2/CO molar ratios for several applications. The addition of small amounts of magnesium to zirconia increased this ratio, especially for the rhodium-based solid.

REFERENCES [1] [2] [3] [4] [5] [6] [7]

[8]

[9]

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

[10] [11] [12]

[13]

[14]

Ramachandran, R., Menon, R. K. (1998). An overview of industrial uses of hydrogen, Int. J. Hydrogen Energy, 23, 593-598. Rostrup-Nielsen, T. (2005). Manufacture of hydrogen, Catal. Today, 106, 293-296. Rostrup-Nielsen, J. R. (2001). Syngas in perspective, Catal. Today, 71, 243-247. Peña, M. A., Gómez, J. P., Fierro, J. L. G. (1006). New catalytic routes for syngas and hydrogen production, Appl. Catal. A: Gen. 144, 7-57. Tsang, S. C., Claridge, J. B. & Green, M. L. H. (1006). Rcent Advances in the conversion of methane to synthesis gas, Catal. Today 23, 3-15. Disckin, A. M. & Ormerod, R. M. (2000). Partial oxidation of methane over supported nickel catalysts, Stud. Surf. Sci. Catal. 130, 3519-3524. Cimino, S., Landi, G., Lisi, L. & Russo, G. (2005). Development of a dual functional structured catalyst for partial oxidation of methane to syngas, Catal. Today, 105, 718-723. Zhang, Z., Verykios, X. E., MacDonald, S. M. & Affrossman, S. (1996). Comparative Study of Carbon Dioxide Reforming of Methane to Synthesis Gas over Ni/La2O3 and Conventional Nickel-Based Catalysts, J. Phys. Chem. 100, 744–754. Zhu, J., Zhang, D. & King, K. D. (2001). Reforming of CH4 by partial oxidation: thermodynamic and kinetic analyses, Fuel 80, 899-905. Bharadwaj, S. S. & Schmidt, L. D. (1995). Catalytic partial oxidation of natural gas to syngas, Fuel Processing Technology, 42, 109-127. Schmidt, L. D., Huff, M. & Bharadwaj, S. S. (1994). Catalytic partial oxidation reactions and reactors, Chem. Eng. Sci., 49, 3981-3994. Yan, Q. G., Wu, T. H., Weng, W. Z., Toghiani, H., Toghiani, R. K., Wan, H. L. & Pittman Jr., C. U. (2004). Partial oxidation of methane to H2 and CO over Rh/SiO2 and Ru/SiO2 catalysts, J. Catal. 226, 247-259. Ruckenstein, E. & Hu, Y. H. (1998). Combination of CO2 Reforming and Partial Oxidation of Methane over NiO/MgO Solid Solution Catalysts, Ind. Eng. Chem. Res., 37, 1744–1747. York, A. P. E., Xiao, T. & Green, M. L. H. (2003). Brief Overview of the Partial Oxidation of Methane to Synthesis Gas, Top. Catal., 22, 345-358.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Partial Oxidation of Methane over Zirconia- and Magnesia-Supported …

407

[15] Lødeng, R., Bjørgum, E. & Enger, B. C. (2007). Catalytic partial oxidation of CH4 to H2 over cobalt catalysts at moderate temperatures, Appl. Catal. A:Gen., 333, 11-23. [16] Freni, S., Calogero, G. & Cavallaro, S. (2000). Hydrogen production from methane through catalytic partial oxidation reactions, J. Power Sources, 87, 28-38. [17] Lanza, R., Järas, S. G. & Canu, P. (2007). Partial oxidation of methane over supported ruthenium catalysts, Appl. Catal. A:Gen., 325, 57-67. [18] Bradford, M. C. J. & Vannice, M. A. (1999). The role of metal–support interactions in CO2 reforming of CH4, Catal. Today, 50, 87-96. [19] Torniainen, P. M., Chu, X. & Schmidt, L. D. (1994). Comparison of monolithsupported metals for the direct oxidation of methane to syngas, J. Catal., 146, 1-10. [20] Fernandes Junior, L. C. P., De Miguel, S., Fierro, J. L. G. & Rangel, M. C. (2007). Evaluation of Pd/La2O3 for dry reforming of methane. Stud. Surf. Sci. Catal., 167, 499-504. [21] Pompeo, F., Gazzoli, D. & Nichio, N. N. (2009). Stability improvements of Ni/α-Al2O3 catalysts to obtain hydrogen from methane reforming, Int. J. Hydrogen Energy, 34, 2260-2268. [22] Sokolovski, V. D., Jeannot, J. C., Coville, N. J., Glasser, D., Hildebrandt, D. & Makoa, M. (1997). High yield syngas formation by partial oxidation of methane over Co-alumina catalysts, Stu. Surf. Sci. Catal., 107, 461-465. [23] Araújo, G. C., Lima, S., Rangel, M. C., Pagola, V., Peña, M. A. & Fierro, J. L. G. (2005). Characterization of precursors and reactivity of LaNi1-XCOxO3 for the partial oxidation of methane. Catal. Today, 107, 906-912. [24] Guo, C., Zhang, X., Zhang, J. & Wang, Y. (2007). Preparation of La2NiO4 catalyst and catalytic perfomance for partial oxidation of methane, J. Mol. Catal. A: Chem., 269, 254-259. [25] Miao, Q., Xiong, G., Sheng, S., Cui, W., Xu, L. & Guo, X. (1997). Partial oxidation of methane to syngas over nickel-based catalysts modified by alkali metal oxide and rare earth metal oxide, Appl. Catal. A:Gen., 154, 17-27. [26] Tang, S., Lin, J. & Tan, K. L. (1998). Partial oxidation of methane to syngas over Ni/MgO, Ni/CaO and Ni/CeO2, Catal. Lett., 51, 169-175. [27] Pantu, P. & Gavalas, G. R. (2002). Methane partial oxidation on Pt/CeO2 and Pt/Al2O3 catalysts, Appl. Catal. A:Gen., 223, 253-260. [28] Roh, H. S., Potdar, H. S., Jun, K. W., (2004). Carbon dioxide reforming of methane over co-precipitated Ni–CeO2, Ni–ZrO2 and Ni–Ce–ZrO2 catalysts, Catal. Today, 93, 39-44. [29] Li, P., Chen, I. & Penner-Hahn, J. E. (1994). Effect of dopants on zirconia stabilizationan X-ray absorption study: I, Trivalents dopants, J. Am. Ceram. Soc., 77, 118-128. [30] Ruckenstein, E. & Wang, H. Y. (2000). Partial oxidation of methane to synthesis gas over MgO-supported Rh catalysts: the effect of precursor of MgO, Appl. Catal. A, 198, 33-41. [31] Ivanova, A. S. (2005). Structure, Texture and acid-base properties of alkaline earth oxides, rare oxides and binary oxides systems, Kinet. Catal. 46, 620-633. [32] Picquart, M., López, T., Gómez, R., Torres, E., Moreno, A. & Garcia, J. J. (2004). Dehydration and crystallization process in sol-gel zirconia, Thermal and spectroscopic study. J. Therm. Anal. Cal., 76, 755-761.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

408

Maria do Carmo Rangel, Marluce Oliveira da Guarda Souza et al.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

[33] Querino, P. S, Bispo, J. R. C. & Rangel, M. C. (2005). The effect of cerium on the properties of Pt/ZrO2 catalysts in the WGSR. Catal. Today, 108, 920 - 925. [34] Stevens, R. (2000). Zirconia and Zirconia Ceramics. Magnesium Electron Ltda., Twickenham. [35] Rojas, Cervantes, M. L., Matin-Arnada, R. M. & Lopez-Peinado, A. J. (1994). LopezGonzalez, J. de D. ZrO2 obtained by the sol-gel method: influence of synthesis parameters on physical and structural characteristics, J. Mater. Sci., 29, 3743-3748. [36] Al, A. A. M. & Zaki, M. I. (2002). HT-XRD, IR and Raman characterization studies of metastable phases emerging in the thermal genesis course of monoclinic zirconia via amorphous zirconium hydroxide: impacts of sulfate and phosphate additives, Thermochim. Acta, 387, 29-38. [37] Garvie, R. C. (1965). The occurrence of metastable tetragonal zirconia as a crystallite size effect, J. Phys. Chem., 69, 1238-1243. [38] Clearfield, A. (v). Crystalline hydrous zirconia, Inorg. Chem., 3, 146-148. [39] Srinivasan, R., Taulbee, D. & Davis, B. H. (1991). The effect of sulfate on the crystal structure of zirconia, Cat. Lett., 9, 1. [40] Srinivasan, R., Hubbard, C. R., Cavin, O. B. & Davis, B. H. (1993). Factors determining the crystal phases of zirconia powders: a new outlook, Chem. Mater., 5, 27-31. [41] Webb, P. A. & Orr, C. (1993). Analytical Methods in Fine Particle Technology, Micromeritics Instruments Corporation, Norcross.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

In: Syngas Production Methods, Post Treatment… Editors: Adorjan Kurucz and Izsak Bencik

ISBN: 978-1-60741-841-2 © 2009 Nova Science Publishers, Inc.

Chapter 11

TECHNOLOGIES OF SYNGAS PRODUCTION FROM BIOMASS GENERATED GASES Simone Albertazzi*, Francesco Basile, Patricia Benito Martin, Giuseppe Fornasari, Ferruccio Trifirò and Angelo Vaccari Dept. Chimica Industriale e dei Materiali, University of Bologna, Viale Risorgimento 4, 40136

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

ABSTRACT Advanced biomass gasification will play a crucial role in at least reducing, if not eliminating the need for imported oil and the negative effects of greenhouse gases generated from the combustion of fossil fuels. The biomass generated gas mainly contains CO, CH4, H2, and CO2 together with contaminants and catalyst poisons (tar, flyash, H2S, NH3). Therefore these species have to be removed and the syngas upgraded before the conversion to liquid fuels. The process required to convert residual light hydrocarbons into syngas can be chosen between Auto Thermal Reforming (ATR) and Partial Oxidation (POX). Operating with ATR will allow higher yield and higher overall process efficiency (12 %). Moreover, less expensive refractory lining is required, due to the lower operating temperature. On the other hand, POX is a robust process, insensitive to both poisoning (since catalyst is not present) and mixing (homogenous phase reaction). Therefore, ATR should be preferred to POX if a catalyst can be found to handle both the contaminants present in the produced gas and the thermal sintering. At this purpose, a commercially available Ni catalyst has been tested in a bench-scale ATR in presence of H2S to evaluate the possibility to use it in a real process.

1. INTRODUCTION The high dependence of most modern transport, agricultural and industrial systems on the relative low cost and high availability of oil caused its production decline and severe increases in prices in the last years. It is now evident that this critical commodity heads into *

Corresponding author: e-mail: [email protected]

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

410

Simone Albertazzi, Francesco Basile, Patricia Benito Martin et al.

decline due to natural depletion. At the moment biomass is the most promising renewable energy source, already furnishing almost a 10 % of the World Total Primary Energy Supply (Figure 1, [1]). There are several ways to produce fuel and power from biomass [2, 3]. Among them, lignocellulosic biomass (wood and dedicated energy crops) and residues (bark, husks, straw, paper mill wastes) can be converted by thermal processes (pyrolysis, gasification) to liquid fuels such as Fischer-Tropsch (FT) diesel blends or dimethyl ether (a clean-burning alternative to diesel). The productivity in terms of yield and process time is very high, since thermal conversion takes place in very short reaction times (typically seconds or minutes). Pressurized oxygen circulating fluidized bed (CFB) gasification is the best largescale technology for the clean, sustainable use of biomass for fuel generation [4, 5]. The biomass generated gas mainly contains CH4, light hydrocarbons, CO, CO2, H2 and it needs to be upgraded before its conversion to liquid fuel. The process has to overcome a number of technical and non-technical barriers before industry will implement its commercialization [6]. The exit gas from the gasifier needs to be improved to synthesis gas in order to produce fuels. Nowadays, the consolidated technology for syngas generation is the steam reforming of natural gas (SR), in which methane and steam are catalytically and endothermically converted to hydrogen and carbon monoxide [7]. 2006 WTPES (%) 2,2

10,1

0,6

6,2

34,4

20,5

26

Oil Natural Gas Hydro Solar - Wind- Geothermal

Coal

Nuclear Biomass

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 1. 2006 World Total Primary Energy Supply.

CH4 + H2O ' CO + 3H2 (ΔH0298= 250 kJ/mol) A different approach is autothermal reforming (ATR, [8]). The process is "autothermal" in that the endothermic reforming reactions proceed with the assistance of the internal combustion (or oxidation) of a portion of the feed hydrocarbons, in contrast to the external combustion of fuel characteristic of conventional tubular reforming. Plants based on oxygenblown autothermal reforming at low steam to carbon (S/C) ratios are the preferred option for large-scale applications [9]. Some parameters such as H2O/CH4, O2/CH4 and CO2/CH4 ratios can be considered the key for thermodynamic evaluation of ATR process. Substantially, variations of these parameters strongly influence thermodynamic equilibrium of the chemical reactions. Typical operating conditions for ATR are 850-1000°C and 20-40 atm. ATR is more

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Technologies of Syngas Production from Biomass Generated Gases

411

flexible than tubular reforming because the higher allowable operating temperature can compensate any increase in methane slip which higher pressure would cause otherwise. Furthermore, the higher operating temperature can also compensate a wider equilibrium approach (difference between actual and equilibrium temperature at the reactor outlet), which may be caused by mild sulphur poisoning of the catalyst. The maximum operating temperature is not limited by the tube material, as in tubular reformers, but only by the stability of the catalysts and the refractory lining of the reactor. Selection of the oxidizing agent (oxygen or air) depends on whether the presence of nitrogen in the reformed gas is desirable (for example, the synthesis of ammonia). Introduction of air or oxygen into the process gas produces the highly exothermic reaction between hydrogen and oxygen, and this is partially balanced by a continuation of the endothermic steam reforming reaction. Nevertheless, temperatures of about 1000°C are achieved, and the major requirements of the reformer are the good mixing of the gases and the burner design, otherwise resulting in local temperatures much higher than 1000°C. A layer of refractory fused alumina chips is often put on the top of the bed to protect the catalyst and to promote good mixing before the gas enters the catalyst bed. The commercial plants commonly use supported nickel catalysts [10, 11]. These materials contain 15–25 wt. % nickel oxide on a mineral carrier (α-Al2O3, aluminosilicates, magnesia and MgAl spinel). Before start up, nickel oxide must be reduced to metallic nickel with hydrogen but also with the feedgas itself at high temperature (above 600°C, depending of the reducing stream). Required properties of the carriers are relatively high specific surface area, low pressure drop and high mechanical resistance at temperatures up to 1000°C. The main poison for Ni based catalysts is sulphur and concentrations as low as 50 ppm can completely deactivate the catalysts [12]. During biomass thermal conversion and SR/ATR, all S compounds are converted to H2S that chemisorbs on metal sites forming NiS, in according to the equilibrium:

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Ni + H2S ' NiS + H2 The low melting point and high surface mobility of NiS also accelerate the sintering process of Ni crystallites. Since the formation of NiS is an exothermic process, activity loss can be partially recovered by raising the reaction temperature, however accelerating the thermal degradation of the catalyst and increasing the carbon formation by cracking reactions. In the specific case of biomass gasification, a number of alkaline salts and heavy metals and metal oxides particles may act as additional poisons by enhancing the sintering of the Ni crystallites or being adsorbed on the Ni sites [13, 14]. While acid supports such as alumina react with alkali to form crystalline phases, basic supports (like MgO) do not react directly with them, however alkali cause coverage of the surface and plugging of the pores. Another cause of activity loss is the carbon deposition that can be avoided if a high S/C ratio is employed [15, 16]. The presence of tars in the reforming reactor enhances coking and it is the main cause of carbon formation in Biomass to Liquid (BTL) processes. Therefore, an optimal hot gas cleaning stage and the development of more tolerant catalysts are key points for the feasibility of the gas upgrading step. Many researchers have been studying alternative active phases to Ni based reforming catalyst in the last years [17-19]. These studies claim that Pt and Rh based catalysts were the most active in model reforming, but their performances have to be still confirmed by processing a real biomass generated gas.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Simone Albertazzi, Francesco Basile, Patricia Benito Martin et al. Inlet 100 kmol total + added O2 and H2O

POX

ATR

0.4

1200

1100

Vol%

0.3

Temperature C

412

1000

0.2 900

H2O CO 0.1 CO2) N2 CH4 H2

800

0.0 3

4

5

6

7

Added oxygen % of inlet

Figure 2. Calculated compositions of syngas as function of oxygen fed.

A great deal of interest has grown in the last few years in partial oxidation (POX) of methane to syngas [20-22]: CH4 + ½O2 ' 2H2 + CO (ΔH0298= - 36 kJ/mol) In comparison to previous technologies, the POX shows advantageous process conditions:

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

i) It is a exothermic reaction, so there is no need for external heating. ii) It can operate at low contact times (10-2 – 10-4 s), allowing the use of small reactors. iii) A higher efficiency can be achieved by introducing an active and stable catalyst, making the process more economically attractive. Small plants using air and natural gas are quite diffuse, expecially home-made reactors used in foundries to produce in situ syngas for annealing treatments. Only in Sarawak, Malaysia, Shell has been successfully operating a high-scale process for the production of synthesis gas at high temperatures (typically > 1200ºC) and pressures of around 50-70 atm as a part of the middle distillate synthesis process [23]. On the other hand, catalyzed partial oxidation (CPO) has become more and more interesting in the last years [24]. The process works under small pressure (1.5-2 atm) to favour mass-transfers, and operates at an average temperature of 900°C and GHSV = 10000 h-1 (1 l of catalyst for 10 Nm3/h of syngas produced). The residence time at which the reaction is carried out is relatively high (360 ms) as compared to the processes claimed for the transformation of natural gas in laboratory plants (few milliseconds, [25]), due to the fact that a higher residence time would mean less heat production, thus less affecting the catalyst and process stability. Pt impregnated on γAl2O3 has been designed as catalyst for this process [26-28]. However, a few constraints limit the diffusion of this process. In particular, the main concerns are:

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Technologies of Syngas Production from Biomass Generated Gases

413

i)

The difficult control of the reaction temperature, due to the presence of hot spots along the catalytic bed. ii) The high stability required by the catalyst at operating temperature. iii) The low H2/CO ratio (< 2) obtained in the process does not fit fuel cells and FT applications.

The presented research paper will focus on the application of both technologies (ATR an POX) in the production of syngas from a raw biomass generated gas.

2. ATR VS POX The reformer unit will be chosen between ATR and POX. SR has been discarded, since it presents the same drawbacks of ATR but lacking of an equally convenient scale economy. The syngas compositions, calculated as function of the added oxygen by means of the software HSC Chemistry version 5.0 [29], are shown in Table 1. ATR works with up to about 5 % of added oxygen, while POX at slightly higher concentrations, thus resulting in lower yield of hydrogen. In the specific case of reforming the product gas from a CFB O2/H2O gasifier, a syngas of superior quality has been obtained by ATR (Table 1). Since ammonia and hydrogen sulphide are not supposed to be converted in both solutions, these contaminants have to be removed before the final utilization of the upgraded gas. Operating with ATR will allow higher yield and higher overall process efficiency (12 %). Moreover, less expensive refractory lining is required, due to the lower operating temperature. On the other hand, POX is a robust process, insensitive to both poisoning (since catalyst is not present) and mixing (homogenous phase reaction). Therefore, ATR should be preferred to POX if a catalyst can be found to handle both the contaminants present in the produced gas and the thermal sintering. Table 1. Calculated compositions after oxygen/steam blown gasifier, supposing an ATR working at 1000°C and a POX working at 1300°C. After ATR 1000 C (vol%)

After POX 1300 C (vol%)

Inlet O2

7

10

Inlet Temperature C

800

800

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Component

After gasifier (vol%)

C2-hydrocarbons

1.6

-

-

CH4

8.2

-

-

CO

11.9

23.8

24.3

CO2

27.9

19.8

19.2

H2

11.8

23.0

16.1

H2 O

37.7

33.4

39.7

NH3

0.3

0.2

0.2

H2S

0.01

0.01

0.01

Tars

0.3

-

-

LHV MJ/kg

6.6

5.6 (85 % of inlet)

4.8 (73 % of inlet)

LHV MJ/Nm3

7.3

5.4

4.8

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

414

Simone Albertazzi, Francesco Basile, Patricia Benito Martin et al.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure. 3. Methane conversion % (Wcat= 40 g; GHSV = 500 Nl/h; S/C= 3.0 v/v; Tin= 800°C at the beginning; analysis in dry gases; addition of 50 ppm vol H2S).

At this purpose, catalytic tests have been carried out on a commercially available Ni/MgAl(O) sample in a bench scale reformer [30]. The tests have been started up feeding only methane and water, and then 50 ppm of H2S wad added (Figure 3). On feeding H2S, the catalytic activity (defined as methane conversion) in the reformer decreased. The activity was firstly restored by increasing the inlet temperature (on increasing the heat supply) and then by adding O2 to the feed, thus increasing the temperature in the catalytic bed by oxidation reactions and simulating ATR conditions. Therefore, upgrading the product gas with an ATR would be possible even if few ppm of sulphur are present in the feed. Moreover, it is important to point out that sulphur poisoning may be partially reversible. Indeed, if the sulphur is removed from the feed, the catalyst will be regenerated by desorption of H2S from the surface of the nickel crystallites. However, transporting the sulphur from the bulk of the metal to the surface is a slow process and the released H2S up-stream the reactor will effect the catalyst down-stream the reactor. Therefore, shut down of the plant is required to try to regenerate bulky NiS. Laboratory tests performed on the spent catalysts put in evidence that regeneration of Ni active sites is possible by use of steam at high temperature (800°C): NiS + H2O ' NiO + H2S Also the hydrogen sulphide and the Ni metal will be oxidized by the water: H2S + 2 H2O ' SO2 + 3 H2 Ni + H2O ' NiO + H2 Finally, the simultaneous presence of SO2 and H2S will involve the Claus reaction:

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Technologies of Syngas Production from Biomass Generated Gases

415

2 H2S + SO2 ' 3/n Sn + 2 H2O The sulphur dioxide is ventilated of the bed and the catalyst is regenerated, but in oxidised form, therefore it must be reduced again before reaction.

3. CONCLUSION For low amount of sulphur in the inlet gas, ATR is suitable for the upgrading of biomass generated gases, otherwise there will be the need to work at higher temperatures or with higher amount of oxygen, causing problems of metal sintering, thus letting more convenient working with a POX unit. On the other hand, developing more tolerant catalysts by adding traces of noble metal to Ni is under investigation at a laboratory scale [31]. When these catalysts will be scaled-up to be commercially available, the ATR will be used in any case.

REFERENCES [1] [2] [3] [4]

[5] [6] [7] [8]

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

[9] [10] [11] [12] [13] [14] [15] [16]

http://www.iea.org/textbase/nppdf/free/2008/key_stats_2008.pdf. Accessed October 21th 2008. Demirbaş, A. Progress in Energy and Combustion Science, 2007, 33, 1. Bridgwater, T. Journal of the Science of Food and Agriculture, 2006, 86, 1755. Albertazzi, S; Basile, F; Trifirò, F. Gasification of biomass to produce hydrogen, in: Renewable Resources and Renewable Energy: A Global Challenge, Graziano M. and Fornasiero P. (Ed.s), Taylor & Francis, 2007, 197. Stiegel, GJ; Maxwell, RC. Fuel Processing Technology, 2001, 71, 79. Albertazzi, S; Basile, F; Brandin, J; Fornasari, G; Rosetti, V; Sanati, M; Trifirò, F; Vaccari, A. Catalysis Today, 2005, 106, 297. Wilhelm, DJ; Simbeck, DR; Karp, AD; Dickenson, RL. Fuel Processing Technology, 2001, 71, 139. Aasberg-Petersen K; Bak Hansen, JH; Christensen, TS; Dybkjær, I. Seier Christensen, P; Stub Nielsen, C; Winter Madsen, S. E. L; Rostrup-Nielsen, JR. Applied Catalysis A: General, 2001, 221, 379. Aasberg-Petersen, K; Christensen, TS; Stub Nielsen, C; Dybkjær, I. Fuel Processing Technology, 2003, 83, 253. Arpe, HJ. (Ed.). Ullmann’s Encyclopedia of Industrial Chemistry 5th Ed., 1989, A12, 238. Twigg, MV. (Ed.). Catalyst Handbook 2nd Ed., Wolfe Publ., London, 1996. Albertazzi, S; Basile, F; Brandin, J; Einvall, J; Fornasari, G; Hulteberg, C; Sanati, M; Trifirò, F; Vaccari, A. Biomass and Bioenergy, 2008, 32, 345. Choi, JS; Kwon, HH; Lim, TH; Hong, SA; Lee, HI. Catalysis Today, 2004, 95, 553. Einvall, J; Albertazzi, S; Hulteberg, C; Malik, A; Basile, F; Larsson, AC; Brandin, J; Sanati, M. Energy & Fuels, 2007, 21, 2481. Rostrup-Nielsen, JR. Catalysis Today, 1997, 37, 225. Trimm, DL. Catalysis Today, 1997, 37, 233.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

416

Simone Albertazzi, Francesco Basile, Patricia Benito Martin et al.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

[17] Rioche, C; Kulkarni, S; Meunier, FC; Breen, JP; Burch, R. Applied Catalysis B: Environmental, 2005, 61, 130. [18] Salge, JR; Deluga, GA; Schmidt, LD. Journal of Catalysis, 2005, 235, 69. [19] Ross, JHR. Catalysis Today, 2005, 100, 151. [20] Bharadwaj, SS; LD. Schimdt, LD. Fuel Processing Technology, 2005, 42, 109. [21] Rostrup-Nielsen, JR. Catalysis Today, 1993, 19, 305. [22] Hickman, DH; Schimdt, LD. Science, 1993, 259, 343. [23] Eilers, J; Posthuma, SA; Sie, ST. Catalysis Letters, 1990, 7, 253. [24] Peña, MA; Gomez, JP; Fierro, JLG. Applied Catalysis A: General, 1996, 144, 7. [25] Hickman, DH; Schimdt, LD. Journal of Catalysis, 1992, 138, 267. [26] Foulds, GA; Lapszewicz, JA; Spivey, JJ; Karval, SK. Catalysis, 1994, 11, 413. [27] Mitchell, PJ; Oh, SH; Siewert, RM. Journal of Catalysis, 1991, 132, 287. [28] Claridge, JB; Green, MLH; Tsang, SC. Catalysis Today, 23, 1995, 3. [29] http://www.outotec.com/pages/Page____21783.aspx?epslanguage=EN. Accessed th October 22 2008. [30] Albertazzi, S; Basile, F; Brandin, J; Einvall, J; Fornasari, G; Hulteberg, C; Sanati, M; Trifirò, F; Vaccari, A. Biomass and Bioenergy, 2008, 32, 345. [31] Nurunnabi, M; Li, B; Kunimori, K; Suzuki, K; Fujimoto, K; Tomishige, K. Applied Catalysis A: General, 2005, 292, 272.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

In: Syngas Production Methods, Post Treatment… Editors: Adorjan Kurucz and Izsak Bencik

ISBN: 978-1-60741-841-2 © 2009 Nova Science Publishers, Inc.

Chapter 12

PROCESS FOR CONVERSION OF COAL TO SUBSTITUTE NATURAL GAS (SNG) Meyer Steinberg* HCE, LLC, Melville, New York 10747, USA

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

SUMMARY This report covers a review and process comparison of the three major processes for conversion of coal to substitute natural gas (SNG): (1) Steam – Oxygen gasification; (2) Catalytic gasification; and (3) Hydrogasification. In addition, two of these three processes are compared for underground coal gasification application at depleted coal bedded methane (CBM) sites. The process chemistry, flowsheets, and mass and energy balances are presented. An economic analysis, including capital investment and production cost estimates is given. A critical comparative evaluative analysis of the coal to SNG process is made. The information presented indicates that for above ground conversion of coal to SNG, hydrogasification is the most thermally efficient process, reaching 80%, has the least CO2 emission and is the most economical process, producing SNG for $4.61/MSCF. For underground depleted CBM well conversion of the unminable coal seam, the hydrogasification process, applied to multiple wells using the latest capital investment for methane reforming for the hydrogen production, can produce SNG for as low as $1.44/MSCF. This value yields a large profit margin considering current natural gas market price is reaching $9.00/MSCF. A limited field test is recommended at a depleted CBM site and is described to prove out the system at a cost of about $1 million.

INTRODUCTION Because the increasing demand for natural gas (methane) in the United States and the limited domestic supply, foreign natural gas imports have grown and the cost has risen to *

Corresponding author: www.hceco.com

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

418

Meyer Steinberg

current values of $6 and $9 / MSCF. Unconventional sources such as coal bedded methane (CBM) are increasing in supply importance. A singularly large indigenous energy resource in the United States is coal. It therefore becomes prudent to examine the technology and economics of processes for conversion of coal to substitute natural gas (SNG), which would open another source of supply for methane. Table 0 lists important reasons for converting U.S. coal to SNG. Table 0. Why Convert Coal to Substitute Natural Gas (SNG)? 1. Coal is Largest Indigenous Energy Resource in the United States. 2. Natural Gas is the Most Convenient and Cleanest Consumer Fuel For Heat, Power & Automotive Use. 3. Transportation and Distribution of Natural Gas by Pipeline are Widely Available and Economical. 4. Natural Gas Resource in the U.S. is Limited. 5. Natural Gas Demand is Increasing – Resulting in Reliance on Imports (Gas from Canada and LNG from Overseas). 6. Natural Gas has Significantly Increased in Cost. 7. Processes for Conversion of Coal to SNG are Increasing in Competitiveness. 8. Coal Bedded Methane (CBM) Production is increasing in Supply.

PROCESS DESCRIPTION There are at least 5 process methods for conversion of coal to SNG. 1. 2. 3. 4. 5.

Steam-Oxygen Gasification Catalytic Steam Gasification Hydrogasification Underground Steam-Oxygen Gasification Underground Hydrogasification

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

1. Steam-Oxygen Gasification Figure 1 shows a process flow sheet and Table 1 gives the process chemistry, mass balance and energy balance for the steam-oxygen gasification process. This process is demonstrated in the North Dakota Gasification Plant in Beulah, North Dakota, where approximately 20,000 T/D of lignite is converted to 150 x 106 SCF of methane (SNG). (S. Stelter, “The New Synfuels Energy Pioneers,” published by Dakota Gasification Co., Beulah, North Dakota (2001)). The calculated thermal efficiency based on data in the Table 1 indicates a thermal efficiency of 61.9% for conversion of the heating value of lignite to the heating value of the methane produced. The capital investment for the plant is high because of the need for an air liquefaction plant, a steam-oxygen coal gasifier, and a catalytic methanator.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Process for Conversion of Coal to Substitute Natural Gas (SNG)

419

Table 1. Steam-Oxygen Coal Gasification for Production of Substitute Natural Gas (SNG) Unit Operations and Process Chemistry Basis: Lignite Coal

1) Steam-Oxygen Gasification of Lignite CH0.8O0.2 + 0.8 H2O = 1.2H2 + CO ∆H = +38.2 Kcal/mol lignite – endothermic - 90% lignite conversion Lignite Combustion - CH0.8O0.2 + 1.1 O2 = 0.4 H2O + CO2 ∆H = 110.3 Kcal / mol Moles at 80% efficiency = 38.2 / 0.8 (110.3) = 0.432 2) Oxygen Plant – Using an Electricity Powered Cryogenic Process Energy = 11.4 Kcal / g-mol lignite 3) Hot Gas Cleanup – Remove Sulfur and Nitrogen Compounds 4) Water Gas Shift for H2 Production 0.45 CO+ 0.45 H2O = 0.45 CO2 + 0.45 H2 ∆H = 0 – energy neutral 5) CO2 Separation – Pressure Swing Adsorption - PSA 6) Methanation 0.55 CO+ 1.65 H2 = 0.55 CH4 + 0.55 H2O ∆H = - 33.0 Kcal – exothermic for steam production Overall Stoichiometry CH0.8O0.2 + 0.7 H2O = 0.55CH4 + 0.45 CO2 Thermal Efficiency =

=

0.55 CH 4 x 0.9 (1.0 + 0.432) CH 08O0.2 + O2 Plant Energy 0.55 x 0.9 x 212 X 100 = 61.9% 1.432 x 110.3 + 11.4

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

It is estimated that the capital investment is of the order of $6,250 / MSCF/D of methane produced, determined by updating the North Dakota plant investment. For estimating the production cost, the financial factors used previously (HCEI-11-04-2 PCM for SNG Production, Cost Estimate (November 25, 2004)) are adopted here. Production Cost based on $12 / ton lignite = $0.73 / MMBTU and Thermal Efficiency = 61.9% is calculated as follows: Factor

Calculation

$ / MSCF

Lignite

= 0.73 / 0.619 =

1.18

Fixed Charges

= (0.20 x 6250) / (0.8 x 365) =

4.28

O&M

= 0.15 x 4.28 =

0.64

Production Cost

6.10

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

420

Meyer Steinberg S, N & Particulates for Disposal

Steam

COAL - LIGNITE

H2, CO, H2S, NH3, particulates

COAL GASIFIER ~ 800oC & 70 atm

Ash

O2

Steam

WATER GAS SHIFT REACTOR ~ 250oC & 70 atm

HOT GAS CLEANUP

H2 + CO ~800oC & 70 atm

H2 + CO ~250oC & 70 atm HEAT EXCHANGER

Air

CO2 Removal

H2 + CO + CO2

Steam

GAS SEPARATOR (PSA or CRYO)

H2 + CO

METHANATOR

AIR SEPARATION PLANT Coolant Water N2

Substitute Natural Gas Production

Coolant Water H2O

REACTION CHEMISTRY -Calculated Thermal Efficiency = 61.9% (1) Steam Gasification of Coal Lignite: CH0.8 O0.2 + 0.8 H2O = 1.2 H2 + CO --- Delta H = + 47.7 Kcal / g-mol lignite - endothermic (2) Lignite Combustion: CH0.8 O0.2 + 1.1 O2 = 0.4 H2O + CO2 --- Delta H = - 110.3 Kcal / g-mol (3) Hot Gas Cleanup removes N, S & particulates (4) Water Gas Shift 0.45 CO + 0.45 H2O = 0.45CO2 + 0.45 H2 (5) Gas Separator, e.g. Differential Pressure Swing Adsorption or Cryogenic Separation, takes out CO2 (6) Methanation: 0.55 CO+ 1.65 H2 = 0.55 CH4 + 0.55 H2O --- Delta H = - 33.0 Kcal – exothermic for steam production (7) Overall Mass Balance: CH0.8 O0.2 + 0.7 H2O = 0.55 CH4 + 0.45 CO2

Figure 1. Steam-oxygen coal gasification for substiture natural gas (SNG) production

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 2. Cataylic steamagasification for substitute natural gas (SNG) production

2. Catalytic Steam Gasification Figure 2 shows a process flow sheet and table 2 gives the process chemistry and mass and energy balance for catalytic steam gasification of lignite. The process was originally developed by Exxon in the 1970s. (R.F. Probstein and R.E. Hicks, Synthetic Fuels, pp. 195201, pH Press, Cambridge, MA (1990). The catalyst is potassium carbonate used in large quantities, amounting to about 20% by weight of the feedstock, which combines with the coal ash, and has to be separated and recovered from the alumina and silica in the ash. It is estimated that the energy requirement for the recovery process is equivalent to 0.05 moles CH4 per mole of lignite. Catalytic gasification requires less energy input to the gasifier than steam-oxygen gasification and the methane is produced directly. There is no requirement for

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Process for Conversion of Coal to Substitute Natural Gas (SNG)

421

an oxygen plant and a methanator. The capital investment would, therefore, be about 75% of the investment in the steam-oxygen gasification plant, which results in a capital investment of $4,688 / MSCF per day. The catalyst cost assumes that 1% of the weight of the coal carrying 20% catalyst is lost and has to be replace at $500 / ton of K2CO3. Production Cost Based On $12 per ton lignite = $0.73 / MMBTU and thermal efficiency = 71.4% Factor

Calculation

$ / MSCF

Lignite

= 0.73 / 0.714 =

1.02

Fixed Charges

= (0.20 x 4688) / (0.8 x 365) =

3.21

O&M

= 0.15 x 3.21 =

0.48

Cost of Catalyst

=

0.41

Production Cost

5.12

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

3. Hydrogasification Figure 3 shows the process flow sheet and Table 3 gives the mass balance and energy balance for the hydrogasification of lignite to produce substitute natural gas (SNG). Much laboratory data is available on coal hydrogasification and a 10 T/hr pilot plant has been operated in Germany in the early 1980s. (S. Lambertz, et al., “Recent Operational Results of the High Temperature Winkler (HTW) and Hydrogasification (HKV) Process,” presented at Second EPRI Conference on Synthetic Fuels in San Francisco, CA (April 1985)). The main feature of this process is that the hydrogasification is exothermic, which makes the process thermally energy efficient. The main problem is the necessity of making up for the deficiency of hydrogen by reforming part of the methane produced in the hydrogasifier. The thermal efficiency of the process is 79.6%, which is 30% higher than the steamoxygen gasification. It is estimated that the capital investment for this plant is 75% of that of the steam-oxygen plant - $4,688 / MSCF per day, about the same as the catalytic gasification process. Production cost estimate is based on $12 / ton lignite = $0.73 / MMBTU: Factor

Calculation

$ / MSCF

Lignite

= 0.73 / 0.796 =

0.92

Fixed Charges

= (0.20 x 4688) / (0.8 x 365) =

3.21

O&M

= 0.15 x 3.21 =

0.48

Production Cost

4.61

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

422

Meyer Steinberg Table 2. Catalytic Steam Gasification of Coal for Production of Substitute Natural Gas (SNG)

Unit Operations and Process Chemistry Basis: Lignite Coal 1) Catalytic Steam Gasification of Lignite CH0.8O0.2 + 0.7 H2O = 0.55 CH4 + 0.45 CO2 ∆H = +5.2 Kcal/mol lignite – endothermic In presence of K2CO3 catalyst -- 90% Conversion Catalyst Content is 20% of lignite feedstock by weight 2) Separation and Recovery of K2CO3 from ash containing silica and alumina Takes 0.05 g-mol CH4 of equivalent energy for process per g-mol lignite 3) Pressure Swing Adsorption or cryogenic separation of CH4 and CO2 from H2 and CO 4) Recycling H2 and CO and preheating in a methane-fired furnace takes 0.08 mols CH4 / gmol lignite Net CH4 produced = 0.55 - 0.05 - 0.08 = 0.42 g-mol Thermal Efficiency = [ (0.9 x 0.42 x 212) / 110.3 ] x 100 = 72.7%

The estimated production cost is 25% lower than the steam-oxygen gasification plant and 10% lower than for the catalytic gasification plant.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 3. Lignite coal hydro gasification for substitute natural gas (SNG) production

4. Underground Steam-Oxygen Gasification of Coal (UCG) This process is the same as the above-ground steam-oxygen gasification of coal with the exception that two boreholes are drilled into a coal seam: one is an injection borehole and the other is an extraction borehole. (C.R.F. Probstein and R.E. Hicks, Synthetic Fuels, pp 202208, pH Press, Cambridge, MA (1990)). Fracturing the coal seam between the boreholes is accomplished by explosives or directional drilling to provide a path for the steam and oxygen between the injection and extraction boreholes. The oxygen permits burning the coal, which creates the temperature and pressure and provides the energy for the steam to endothermically

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Process for Conversion of Coal to Substitute Natural Gas (SNG)

423

react with the coal in the seam. Oxygen instead of air avoids dilution of the gases with nitrogen. Table 3. Coal Hydrogasification for Production of Substitute Natural Gas (SNG) Unit Operations and Process Chemistry Basis: Lignite Coal 1) Hydrogasification of Lignite: CH0.8O0.2 + 1.2 H2 = 0.8 CH4 + 0.2 CO ∆H = - 9.8 Kcal/g-mol lignite – exothermic -- 90% Conversion 2) Hot Gas Cleanup. Remove Sulfur and Nitrogen Compounds. 3) Water Gas Shift : 0.2 CO+ 0.2 H2O = 0.2 CO2 + 0.2 H2 ∆H = 0 – energy neutral 4) Pressure Swing Adsorption or Cryogenic Separation of CH4 and CO2 from H2 and CO (5) Steam Reforming of Methane for H2 Makeup Production 0.25 CH4 + 0.25 H2O = 0.25 CO + 0.75 H2 ∆H = +15.0 Kcal - endothermic (6) Methane Combustion: CH4 + 2 O2 = CO2 + 2 H2O ∆H = 212 Kcal / g-mol – at 80% eff. CH4 = 15 / (0.8 x 212) = 0.09 g-mol (7) Water Gas Shift for H2 production: 0.25 CO + 0.25 H2O = 0.25 CO2 + 0.25 H2 (8) CO2 Separation – PSA (9) Overall Stoichiometry: CH0.8O0.2 + 0.7 H2O = 0.55 CH4 + 0.45 CO2 (10) Net CH4 produced = 0.55 - 0.09 = 0.46 g-mol (11) Thermal Efficiency = [ (0.46 x 212 x 0.9) / 110.3 ] x 100 = 79.6%

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

The gasification reaction produces carbon monoxide and hydrogen synthesis gas. The sulfur and nitrogen in the coal are converted to H2S and NH3, which are extracted with the synthesis gas. Above ground, the sulfur and nitrogen compounds and any entrained coal or ash particulates are removed using hot gas cleaning operations.

Figure 4. Underground steam0oxygen coal gasification for substitute natural gas (SNG) production

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

424

Meyer Steinberg

The hydrogen to carbon monoxide ratio in the extracted reaction gas is adjusted by water gas shift to provide a 3 to 1 ratio of hydrogen to carbon monoxide. This ratio is needed to covert the gas to methane in a catalytic methanator. The methane reaction is exothermic and the heat generates steam for the process. The water produced in the methanator is condensed to produce a concentrated substitute natural gas (SNG) product for pipelining. The thermal efficiency for this process is 61.9%. Figure 4 is a schematic of the underground steam-oxygen gasification of coal process. By eliminating the mining of the coal, but including underground site preparation, it is estimated that the capital investment for steam-oxygen gasification is reduced to $6095 / MSCF/D of methane. Continuous operation, gas storage and redundant equipment can provided a high capacity factor. The production cost is calculated as follows: Factor

Calculation

$ / MSCF

Fixed Charges

= (0.20 x 6095) / 365 =

3.34

O&M

= 0.15 x 3.34 =

0.50

Production Cost

3.84

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

5. Underground Hydrogasification of Coal (aka Pumped Carbon Mining, PCM) The underground hydrogasification of coal for SNG production is similar to the above ground process with the exception that the hydrogasification takes place underground. This process is especially useful for unminable coal seams and where methane is produced from coal bedded methane (CBM) in these seams. (HCE, LLC Report, HCEI-10-04-3rl, “Pumped Carbon Mining (PCM) for Substitute Natural Gas Production (October 4, 2004)). The coal seam is accessed by two vertical boreholes spaced a distance apart; one is the intended injection borehole and the other is the intended extraction borehole. A flow connection is then established between the boreholes. This can be accomplished by a number of means, one of which is by horizontal drilling between the holes. The existing methane resource in the coal seam is removed through the extraction borehole using established coal bedded methane extraction procedures. In this process, the water in the seam is also removed and this is beneficial to the subsequent hydrogasification process. The hydrogasification process then begins with the injection into the coal seam of heated and pressurized hydrogen. Under these conditions, the hydrogen exothermically reacts with the coal, producing methane and carbon monoxide. Some of the nitrogen and sulfur in the coal is converted to ammonia and hydrogen sulfide. An excess of hydrogen is used to convert the carbon to methane under equilibrium conditions (see HCEI 10-04-3). The reaction gas flowing out of the extraction borehole is subjected to hot gas cleanup, which removes most of the unwanted contaminant gases and particulates and leaves a methane-rich stream containing hydrogen and carbon monoxide. The methane is separated

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Process for Conversion of Coal to Substitute Natural Gas (SNG)

425

from the other gases by pressure swing adsorption (PSA) or cryogenically. Since there is insufficient hydrogen in the coal to combine with the carbon in the coal to form methane, the hydrogen must be produced by reacting part of the methane produced with water in a steamreforming operation. The net methane produced results in a thermal efficiency of 79.6% for conversion of the energy in lignite to the energy in the methane. When connecting the hydrogasification to coal bedded methane (CBM) operations, the underground site preparation cost is borne by the CBM operation and there is no oxygen or methanator investment. This reduces the capital investment to about $4,571 / MSCF/D (HCEI-11-04-2). The production cost is estimated as follows: Factor

Calculation

$ / MSCF

Fixed Charges

= (0.20 x 4571) / 365 =

2.50

O&M

= 0.15 x 2.50 =

0.38

Production Cost

2.88

QUALITATIVE COMPARATIVE ANALYSIS OF PROCESSES FOR PRODUCTION OF SUBSTITUTE NATURAL GAS (SNG) FROM COAL RESOURCES There are at least five processes for conversion of coal to substitute natural gas (SNG) as described previously. The following is a critical qualitative analysis of these processes.

1) Steam Oxygen Gasification of Coal

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

The steam-oxygen gasification of coal is a well-know process, which has been practiced since the 1940s. Various types of gasifiers have been developed employing steam and oxygen with coal feedstock. The following is a list of the drawbacks in using this process for conversion of coal to SNG: 1. The steam-oxygen reaction with coal to form synthesis gas (CO and H2) is highly endothermic. 2. An oxygen plant is required. 3. A methanator is required. 4. The thermal efficiency is low, about 60%. 5. Capital investment is high.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

426

Meyer Steinberg

2) Catalytic Gasification of Coal Exxon developed this process during the 1970s, operating it in a pilot plant. A full scale production plant was never built because it was not competitive. The features of the process compared to steam-oxygen gasification are as follows: 1. The catalytic steam reaction is endothermic, but much less so than the steam-oxygen process. 2. There is no need for an oxygen plant. 3. There is a huge requirement for catalyst, amounting to as much as 20% of the coal feedstock. Recovery of catalyst, K2CO3 from coal ash is costly. 4. There is no requirement for a methanator. 5. The thermal efficiency is higher than steam-oxygen gasification, reaching into the 70%. 6. The capital investment is lower than for steam-oxygen gasification.

3) Hydrogasification of Coal Hydrogasification of coal for production of methane (SNG) was pilot planted in Germany in the 1970’s, but the process was never put into practice at full scale. The features of this process are as follows: 1. The hydrogasification of coal is exothermic, thus requiring no oxygen or steam addition. 2. There is no requirement for a methanator. 3. The process has a high thermal conversion efficiency reaching into the 80%. 4. It is necessary to convert part of the methane back to hydrogen by reforming with steam. 5. The capital investment is lower than for steam-oxygen gasification. 6. There is no requirement for a catalyst.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

4) Underground Steam-Oxygen Coal Gasification (UCG) There has been much research and development on underground coal gasification in the United States and Russia during the 1970s. The features of the process are as follows: 1. Coal mining and preparation for above ground processing is eliminated. 2. Steam-oxygen injection is required underground, which may have safety problems due to incomplete reaction and production of explosive gaseous mixtures in confined spaces. 3. Difficult to control problems with fissures and crossovers to keep inflow and outflow paths separated. 4. An oxygen plant and methanation reactor are required. 5. Thermal conversion efficiency is lower than hydrogasification.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

427

Process for Conversion of Coal to Substitute Natural Gas (SNG)

5) Underground Coal Hydrogasification (Pumped Carbon Mining, PCM) The underground hydrogasification of coal has been proposed in the 1980s, but was never tested. It has been recently proposed in conjunction with coal bedded methane extraction. The hydrogasification is beneficial in configuration with the extraction of methane from unminable coal deposits. The features of the process are as follows: 1. After the coal bedded methane is extracted, the hydrogasification of the remaining coal would increase the production of methane from the coal seam by a factor of 20 or more times that produced from the coal bedded methane recovery operation alone (HCEI-11-04-2). 2. There is no mining or handling of coal for above ground processing. 3. The additional hydrogen needed is produced by steam reforming part of the methane directly produced by hydrogasification. 4. No oxygen or methanator plants are needed, so that the capital investment is lower than steam-oxygen processing. 5. The cost of the preparation of the mine for underground hydrogasification is borne by the coal bedded methane operations. It should be noted that underground catalytic gasification of coal as developed by Exxon is not feasible because of the large loss of catalyst underground even if it were possible to inject catalyst underground. Table 4 gives a summary comparison of the various factors for conversion of coal to SNG. Underground hydrogasification of coal (PCM) for SNG appears to have the highest thermal efficiency and lowest production cost. Table 4. Comparison of Processes for Conversion of Coal to Substitute Natural Gas (SNG)

Item

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Coal Feedstock Oxygen Plant Methanator Thermal Efficiency Catalyst Requirement CO2 Emission Lbs CO2/MSCF % Reduction CO2 From S-O gasif. Capital Investment $ per MSCF per day Production Cost $ per MSCF

No No 79.6%

Underground Fracturing of Coal Seam Yes Yes 61.9%

Underground Hydrogasification Pumped Carbon Mining (PCM) Following Coal Bedded Methane Extraction No No 79.6%

Yes

No

No

No

123

78

51

123

51

0

37

59

0

59

6250

4688

4688

6095

4571

6.10

5.12

4.61

3.84

2.88

SteamOxygen Gasification

Catalytic Steam Gasification

Mined, Crushed, Transported Yes Yes 61.9%

Mined, Crushed, Transported No No 71.9%

No

Hydrogasification

Mined, Crushed, Transported

Underground Steam-Oxygen Gasification

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

428

Meyer Steinberg

CO2 EMISSIONS Increasing efficiency also has an impact on carbon dioxide emissions. All of these processes essentially emit concentrated carbon dioxide streams, which can be captured and sequestered. However, the higher the thermal efficiency, the lower the carbon dioxide volume and thus the lower sequestration requirements. Thus, the hydrogasification process at 79.6% thermal efficiency emits 59% less carbon dioxide than the steam-oxygen gasification process at a thermal efficiency of 61.9%, as shown in Table 4. It is also possible to sell and sequester the CO2 in depleted oil wells for enhanced oil recovery (EOR). A large part of the CO2 is emitted at almost 100% concentration, which can be sequestered directly.

PUMPED CARBON MINING (PCM) COST ESTIMATE BASED ON LARGE CENTRAL PLANT SERVICING A NUMBER OF DEPLETED COAL BEDDED METHANE (CBM) WELLS

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

As noted in the cost estimate report, HCEI 11-04-2, it is possible to optimize the size of the hydrogen production plant so that it can supply hydrogen to a number of underground hydrogasification wells. The optimized hydrogen plant would collect the gas from a number of extraction boreholes of the wells and deliver hydrogen to the injection boreholes. Insulated piping to and from centralized processing would be utilized to deliver the high temperature and high pressure hydrogen and carbon monoxide to the wells and carry back the methane rich gas to the central processing plant. The central processing plant would contain the hot gas cleanup, heat exchangers, water gas shift and methane and carbon dioxide gas separation and pumping equipment as well as the steam reformer for the hydrogen production. The concept of a central processing plant is important because each well may have variable production. It would become expensive to locate all processing equipment at each such well especially for those wells that had relatively small production capacity. A rough example of an optimized central processing plant is as follows: A methane reforming plant producing 100,000 MSCF / D of hydrogen should cost today approximately $100 million (HCEI-11-04-2). This amount of hydrogen can service the following capacity of methane produced by the Pumped Carbon Mining or lignite gasification process following extraction of the coal bedded methane. Net Methane Production Methane to Reformer Methane to Furnace

= = =

46,000 MSCF / D 25,000 MSCF / D 9,000 MSCF / D

Total Well Methane Produced

=

80,000 MSCF / D

If the average CBM capacity per well is only 200 MSCF / D (Mike Gatens, Oil & Gas , pp. 41-43, Dec. 13, 2004) and the production of hydrogasified coal is 20 times that of the CBM capacity, then the central plant can handle gases from about 20 wells, or if the CBM wells produce an average of 400 MSCF / D, then the central plant can process gas from about 10 wells.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Process for Conversion of Coal to Substitute Natural Gas (SNG)

429

A rough capital and production cost estimate for this capacity is as follows: The flow sheet of Figure 5 is followed scaled up to the 46,000 MSCF/D methane production and where only one central methane reforming plant and one water gas shift reactor is used to connect the CO to H2 with steam. The highest cost element is the methane reformer. The other equipment is roughly estimated relative to the reformer.

Figure 5. Pumped carbon mining: substiture natural gas production

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Capital Investment Unit Hot Gas Cleanup Heat Exchanger Water Gas Shift (Included in the H2 plant costs) Gas Separation (PSA or Cryo) Circulator Pumping Methane Reforming for H2 Total Capital Investment

$ Millions 10 10

--20 10 100 150

Production Cost: (using same financial factors as in HCEI-11-04-2) Factor Calculation The underground lignite well preparation Fixed Charge = (0.2 x 150 x 106 ) / (365 D / yr ) Op. & Maint. = 0.15 x 82,190

Total Production Cost Unit production Cost of Methane = 94,520 / 46,000

= =

$/D charged to CBM 82,190 12,330

= =

94,520 $2.05 / MSCF

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

430

Meyer Steinberg

Thus, it is shown that by centralizing the processing and collecting the gas from a number of wells, the unit methane production cost can be reduced significantly, resulting in a higher rate of return.

NEW COST INFORMATION

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Recently, a paper appeared in the Hydrocarbon Engineering Journal (February, 2004) by C. A. Boyce, M.A. Crews and R. Ritter of Chicago Bridge and Iron Company (CB&I) entitled, “Time for a New Hydrogen Plant?” In it, a comparison is made between old style and a modern hydrogen production plant. The process deals with a plant producing 90 million SCFD of hydrogen based on the steam reforming of natural gas is estimated to cost $50 million. The new capital investment for the PCM plant based on 100 million SCFD of hydrogen is thus estimated to be $55 million. In contrast, the old capital investment number assumed in the above PCM plant for a 100 million SCFD hydrogen plant was assumed to be $100 million. The reduction in new capital investment is credited to higher efficiency and the use of lower cost gas separation equipment and improved water gas shift units. A revised PCM production cost estimate based on this new capital investment estimate, which assumes a capital investment of $55 million for 100 million SCFD hydrogen production is as follows: The PCM central hydrogen plant services 10 to 20 CBM wells, as shown in Figure 6. The capital investment includes heat exchangers and gas cleanup at the wells and methane separation at the central plant.

Figure 6. Pumped carbon mining – For substitute natural gas (SNG) production

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Process for Conversion of Coal to Substitute Natural Gas (SNG)

431

Capital Investment Unit Hot Gas Cleanup Heat Exchangers (located at each well) Water Gas Shift (Included in the H2 plant costs) Gas Separation (PSA or Cryo) Circulator Pump and Piping Methane Reforming for H2 Total Capital Investment

$ Millions 10 10

--20 10 55 105

Production Cost: (using same financial factors as in HCEI-11-04-2), 20% fixed charges on investment and 15% for operation and maintenance. Factor Calculation Underground lignite well completion Fixed Charge = (0.20 x 105 x 106) / (365 D / yr) Op. & Maint. = 0.15 x 57,530 Total Production Cost

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Unit production Cost of Methane = 66,160 / 46,000

$/D charged to CBM account = 57,530 = 8,630 =

66,160

=

$1.44 / MSCF

Thus a modern methane steam reforming plant to provide the additional hydrogen for hydrogasification of the lignite coal provides a larger margin of profit than the older plant. More definitive design and application of local financing factors can refine and validate the above estimates. A summary comparison of process parameters for conversion of Coal to substitute natural gas (SNG) is shown in Table 4. It is noted for the three above ground processes, hydrogasification has the highest thermal efficiency at about 80% and at $4.61/MSCF indicates the lowest production cost for SNG production. For underground processing, hydrogasification at $2.88/MSCF is less than the steam-oxygen gasification. A summary of Production rates and production cost base on various assumptions for underground PCM processing for SNG production is given in Table 5. A single well, which had the highest CBM production rate of 700 MSCF/D is compared to multiple well production at an average per well that produced 200 MSCF/D during its lifetime. An optimized 100,000 MSCF/D hydrogen steam reforming plant can process 20 wells that had the average CBM production rate. By applying the new modern cost of hydrogen plant, the production cost can be reduced from $2.05/MSCF to $1.44/MSCF, thus providing a wide margin of profit from the current market price of natural gas reaching $9.00/MSCF.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

432

Meyer Steinberg Table 5. Pumped Carbon Mining (PCM) For SNG Production (Underground Hydrogasification in Depleted CBM wells in Unminable Coal Seams Summary of Production Rates and Production Cost)

NUMBER OF WELLS CBM Methane Production MSCF/D/Well Net Methane PCM Production MSCF/D Methane to Reformer MSCF/D Methane to Reformer Furnace MSCF/D Plant Cost Millions Unit Production Cost $/MSCF

SINGLE - 1

MULTIPLE - 20

MULTIPLE - 20

700

200

200

14,000

46,000

46,000

7,600

25,000

25,000

2,700

9,000

9,000

64*

150*

105**

2.88

2.05

1.44

*Scaled based on conventional methane reforming plant producing 100,000 MSCF/D of hydrogen costing $100 million. **Scaled based on recent capital cost of a new modern methane reforming plant for hydrogen production, producing 90,000 MSCF/D of hydrogen costing $50 million.

LIMITED FIELD TEST OF PUMPED CARBON MINING (PCM) FOR METHANE (SNG) PRODUCTION

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

As shown above, Pumped Carbon Mining (PCM) of depleted coal bedded methane (CBM) wells for substitute natural gas production (SNG) appears very lucrative in today’s natural gas market. To prove out this process, it is recommended that a limited field test be run (HCE, LLC Report, HCEI-3-05-002, “Limited Field Test of Pumped Carbon Mining (PCM) for Methane (SNG) Production,” (February, 2005). The test consists of pumping hydrogen through a depleted CBM well to obtain the following information: 1. determine the optimum concentration of methane in the gas produced as a function of process conditions; 2. loss of hydrogen, if any underground; and, 3. determine the quantity of residual water and methane produced during ambient and warm hydrogen flushing of the depleted well. The preliminary estimated cost of conducting the test is between $600,000 and $1,000,000, depending on whether a once thru hydrogen system is used or recirculating compressor and condenser is installed for the hydrogen flow. A schematic of the test setup is shown in Figure 7.

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Process for Conversion of Coal to Substitute Natural Gas (SNG)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 7. Punped carbon mining: substitute natural gas production

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

433

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved. Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

INDEX

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

A abatement, ix, 1, 3, 11, 13, 15, 19, 21, 22, 23, 24, 25, 26, 27, 49, 143 absorption, 3, 19, 39, 40, 42, 196, 279, 388, 390 accounting, 42, 89, 230, 355, 357 accuracy, 305, 362 acetaldehyde, 67 acetate, 67, 68, 151, 164, 166, 325, 327, 328, 329, 369 acetic acid, 32, 35, 37, 41, 55, 68, 71, 87, 325, 328 acetone, 32, 35, 55, 67, 68, 69, 70, 71, 74, 87, 113, 114, 166, 169 acetylene, 228, 254 achievement, 44 acid, 5, 23, 32, 35, 37, 41, 54, 55, 68, 71, 87, 158, 161, 166, 169, 209, 210, 325, 328, 333, 407, 411 acidity, 15, 19 ACS, 392 activated carbon, 39, 42, 45, 148 activation, x, 53, 63, 65, 67, 69, 70, 95, 154, 156, 166, 170, 171, 172, 179, 198, 212, 213, 214, 215, 216, 217, 220, 221, 244, 282, 320, 324, 364 activation energy, 154, 244 active oxygen, 89, 385 active site, xi, 24, 33, 35, 37, 67, 191, 197, 198, 210, 211, 212, 215, 217, 220, 221, 234, 318, 320, 325, 338, 414 activity level, 220 additives, 19, 46, 119, 220, 391, 398, 408 adhesion, 157, 179, 180, 279 adiabatic, 76, 81, 82, 83, 84, 85, 86, 87, 91, 99, 100, 102, 126, 148, 206, 258, 355, 356, 358 adjustment, 5, 151, 194 adsorption, x, xiv, 23, 33, 39, 42, 43, 45, 69, 106, 141, 147, 148, 151, 152, 153, 156, 158, 200, 201, 204, 210, 217, 229, 267, 273, 277, 278, 314, 320, 321, 327, 328, 332, 395, 399, 401, 425

aerosol, 164 AFM, 168, 169 Africa, 3, 31, 33, 193, 229, 255 Ag, 62, 152, 154, 161, 162, 168, 242, 243, 244, 245, 247, 249, 257, 278, 280, 281, 282, 283, 284, 288, 296, 308, 309, 310, 311 age, 32 ageing, 208 agent, 8, 12, 167, 168, 208, 333, 336, 382, 411 agents, 73, 166, 167, 209, 282, 333, 334, 336, 338, 400 agricultural, 29, 45, 46, 409 agricultural residue, 46 agriculture, 54 air pollution, 54 alcohol, 38, 41, 69, 87, 314, 316, 368 alcohols, 32, 34, 35, 37, 38, 41, 54, 55, 69, 71, 200, 201, 204, 208, 228, 254 aldehydes, 37, 55, 71, 201, 228, 319 algae, 264 algorithm, 29 alkali, 17, 21, 22, 38, 49, 51, 62, 67, 192, 199, 207, 210, 260, 314, 319, 320, 407, 411 alkaline, 11, 15, 31, 62, 166, 167, 210, 212, 407, 411 alkane, 33, 204 alkanes, 37, 67, 82, 122, 197, 204 alloys, x, 35, 53, 63, 72, 73, 130, 154, 160, 161, 162, 163, 168, 281, 283, 309 alternative, xi, xii, xiii, 30, 34, 36, 54, 55, 68, 69, 72, 98, 191, 192, 203, 206, 207, 212, 221, 224, 227, 230, 235, 249, 254, 331, 395, 397, 410, 411 alternatives, 58, 264, 346, 397 aluminosilicates, 411 aluminum, 23, 47, 62, 73, 75, 127, 128 aluminum oxide, 24, 47, 62 ambient air, 217 ambient pressure, 213 amine, 32, 167 amines, 19

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

436

Index

ammonia, xiii, 24, 45, 49, 51, 144, 151, 167, 170, 182, 194, 196, 199, 228, 264, 375, 395, 396, 411, 413, 424 ammonium, 167, 398 ammonium hydroxide, 167, 398 amorphous, 59, 401, 408 Amsterdam, 255, 258, 391 anaerobic, 41, 42 anatase, 129 annealing, 129, 167, 186, 283, 284, 412 anode, 63, 67, 69, 165, 346, 349, 352, 356, 358 anthropogenic, 221, 346 APC, 223 application, xiii, xiv, 18, 27, 38, 39, 51, 54, 55, 61, 70, 74, 98, 122, 150, 159, 164, 165, 205, 206, 232, 233, 234, 235, 239, 251, 255, 257, 261, 272, 283, 310, 375, 392, 413, 417, 431 aqueous solution, 41, 164, 165, 166, 182, 197, 198, 207, 208, 333, 398 argon, 165 argument, 401 aromatic compounds, 35, 195 aromatic hydrocarbons, 119 aromatic rings, 14 aromatics, 33, 34, 35, 76, 83, 123, 144, 396 Arrhenius law, 94, 156 ash, xiv, 2, 6, 22, 196, 409, 420, 422, 423, 426 Asia, 309 assessment, 294, 301, 307, 309 assignment, 318 assumptions, 100, 286, 296, 431 Athens, 140 atmosphere, 6, 13, 42, 167, 182, 209, 210, 212, 213, 233, 241, 279, 396 atmospheric pressure, 31, 64, 87, 92, 148, 149, 192, 246, 348, 380, 382, 389 atomic force, 168 atomic force microscopy (AFM), 168 atoms, 32, 59, 63, 76, 78, 157, 158, 163, 164, 165, 166, 175, 197, 198, 202, 211, 212, 213, 214, 215, 216, 277, 284, 318, 325, 328, 330, 331, 401 attachment, 350 attacks, 19 Australia, 373 Austria, 375 automotive applications, 143 autothermal, xiii, 56, 63, 67, 68, 69, 86, 99, 102, 114, 115, 117, 122, 123, 229, 240, 241, 266, 309, 375, 376, 380, 381, 385, 386, 389, 391, 396, 410 availability, 7, 21, 36, 38, 143, 272, 409 averaging, 365

B bacteria, 41 bacterium, 48 barrier, 18, 196, 237 barriers, 41, 95, 410 basic research, 210 baths, 170, 282 batteries, 145 battery, 304, 305 behavior, x, 48, 54, 71, 87, 89, 91, 92, 95, 115, 249, 251, 257, 282, 286, 296, 388, 391, 404 Belarus, 131 Belgium, 131, 385, 391 benchmark, 61 benefits, xii, 6, 30, 55, 149, 222, 263, 272, 273 benzene, 195 bicarbonate, 208 binary oxides, 407 biocatalysts, 41 biodiesel, 43, 54, 67, 68, 69 bioethanol, 54, 69 biofuels, 27, 34, 36, 43, 44, 46, 54, 69, 70, 192 biogas, 44, 258 biological processes, 41 biomass, ix, xi, xiii, xiv, 2, 6, 7, 8, 13, 14, 15, 22, 23, 27, 28, 29, 33, 34, 35, 36, 39, 40, 43, 44, 45, 46, 47, 48, 49, 50, 51, 54, 55, 67, 68, 71, 87, 119, 143, 146, 147, 191, 192, 193, 194, 195, 196, 198, 199, 221, 222, 223, 256, 264, 268, 269, 313, 314, 317, 322, 346, 392, 409, 410, 411, 413, 415 blends, 5, 6, 7, 73, 410 blocks, 166 boats, 5 boilers, 3, 17, 19, 20, 267, 384 boiling, 39, 67, 146 bonds, 67, 201, 327 boreholes, 422, 424, 428 boundary conditions, 288, 289, 297, 298, 299 brass, 129, 130 Brazil, 395 Brussels, 131 bubble, 33, 38, 167, 171, 172, 174, 180, 205, 206 buffer, 207, 208 building blocks, 200 buildings, 144 burning, 230, 346, 410, 422 buses, 34 butane, 204 butyl ether, 35 butyric, 41 bypass, 114 by-products, 35, 49, 69, 80, 200

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Index

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

C Ca2+, 400 calcium, 13, 398 Canada, 392, 418 candidates, 63 capillary, 144, 154, 156, 276, 277 capital cost, 30, 222, 231, 232, 237, 240, 241, 303, 396, 432 capital intensive, 230 caps, 172 carbenes, 201 carbide, xi, 37, 191, 198, 200, 201, 209, 210, 213, 214, 215, 216, 217, 219, 220, 234, 314, 315, 316, 317, 318, 319, 320, 322, 323 carbohydrates, 55 Carbon, 20, 24, 42, 45, 47, 59, 60, 61, 64, 71, 192, 205, 234, 244, 255, 256, 258, 315, 322, 325, 371, 373, 392, 406, 407, 424, 427, 428, 432 carbon atoms, 59, 198, 202, 213, 214 carbon dioxide, ix, xiii, xiv, 19, 28, 46, 64, 84, 90, 129, 142, 143, 147, 152, 182, 183, 197, 211, 220, 221, 229, 233, 234, 235, 238, 245, 257, 258, 260, 268, 273, 345, 346, 348, 353, 371, 372, 373, 392, 396, 397, 405, 406, 428 carbon monoxide, ix, 54, 78, 80, 84, 99, 145, 146, 181, 200, 201, 215, 224, 229, 254, 258, 322, 346, 396, 397, 404, 405, 410, 423, 424, 428 carbonates, 199 carburization, 211, 212, 216, 219, 221, 320 carrier, 40, 54, 142, 143, 144, 240, 242, 346, 348, 349, 356, 377, 379, 388, 390, 391, 392, 393, 411 case study, 235, 248, 251, 254 cast, 306 casting, 163, 166 catalysis, 126, 198, 203, 213, 225, 325, 327 catalyst deactivation, xiii, 9, 37, 57, 67, 129, 235, 319, 332, 395, 404 catalytic activity, 10, 12, 13, 14, 55, 61, 71, 87, 211, 213, 215, 216, 217, 218, 219, 220, 233, 234, 246, 248, 254, 314, 318, 323, 332, 380, 414 catalytic C, 233 catalytic effect, 10, 14, 176, 252 catalytic fixed bed reactor, 26 catalytic properties, 33, 199, 215, 319 catalytic system, 56, 59, 62, 71, 72, 89, 96, 126, 204 catechol, 71 cathode, 165, 346, 349, 352, 356, 358 cation, 49, 62, 157 cavities, 127 C-C, 68, 201, 316, 327 cell, 69, 72, 126, 128, 143, 145, 160, 165, 169, 172, 174, 176, 182, 184, 185, 206, 240, 258, 346, 347,

437

348, 349, 350, 351, 353, 354, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 369, 370, 371, 372, 399 cell surface, 347 cellulose, 151 cellulosic, 43 cement, 50 ceramic, x, 17, 18, 40, 55, 72, 74, 75, 100, 130, 141, 151, 154, 156, 157, 158, 160, 162, 163, 187, 234, 238, 239, 241, 246, 248, 256, 259, 260, 261, 275, 278, 279, 283, 308, 311, 371 ceramic microporous membranes, 151 ceramics, 130, 160, 239, 275 cerium, 212, 408 CFD, 347, 373 CH3COOH, 37 chain propagation, 201 chain termination, 200, 203, 210 channels, 65, 69, 73, 83, 90, 114, 128, 206, 350, 351 charcoal, 18 chelating agents, xiii, 313, 314, 324, 332, 333, 334, 335, 336, 337, 338 chemical bonds, 34 chemical composition, 12, 70, 76, 119 chemical energy, 28 chemical industry, 144, 264 chemical properties, 208 chemical reactions, 3, 146, 165, 410 chemical reactor, 114 chemical stability, 281 chemical vapor deposition, 210, 281, 283, 310 chemical vapour, 157, 164 chemical vapour deposition, 157, 164 chemicals, ix, 15, 16, 27, 31, 32, 34, 35, 193, 222, 393 chemisorption, 198, 199, 320, 326, 327, 334 China, 29 chloride, xiv, 40, 47, 51, 171, 395, 398 chlorine, 2, 6, 7, 11, 13, 14, 19, 22, 26, 34, 45, 48, 51, 281 CHP, 142 chromatography, 101, 123 chromium, 71 Chrysler, 145 circulation, 30, 39, 388 classes, 71, 158, 200 classical, 382 classification, 150 clean energy, 273 cleaning, ix, x, 1, 2, 3, 5, 9, 10, 15, 16, 17, 20, 21, 23, 24, 25, 26, 27, 28, 31, 33, 34, 39, 41, 44, 45, 46, 48, 50, 51, 147, 166, 169, 171, 195, 196, 222, 411, 423

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

438

Index

cleanup, 20, 44, 45, 48, 424, 428, 430 climate change, 42 closure, 252 clusters, 63, 66, 70, 197, 198, 321, 325, 326, 327, 329, 331, 332, 335 coal, ix, xi, xiii, xiv, 1, 2, 3, 5, 6, 7, 8, 10, 14, 27, 28, 29, 30, 31, 33, 35, 36, 42, 44, 46, 47, 48, 49, 50, 54, 142, 145, 146, 147, 191, 192, 193, 194, 195, 196, 198, 222, 229, 239, 264, 268, 309, 311, 313, 317, 346, 375, 417, 418, 420, 421, 422, 423, 424, 425, 426, 427, 428, 431, 432 cobalt, 32, 33, 47, 61, 67, 193, 197, 198, 212, 322, 371, 398, 407 co-existence, 232, 256, 400 coil, 171 coke, 3, 9, 12, 13, 15, 16, 22, 23, 35, 54, 57, 61, 62, 67, 68, 70, 80, 83, 87, 114, 195, 240, 241, 260, 286, 382, 386, 387, 390, 396, 398 Coke, 12, 59 coke formation, 22, 57, 62, 67, 80, 83, 87, 286, 382, 386, 387, 390 collaboration, 56 collisions, 276 combustion, xiv, 2, 5, 7, 8, 23, 27, 28, 29, 39, 42, 43, 51, 60, 65, 66, 68, 70, 103, 117, 142, 143, 145, 146, 194, 230, 235, 236, 237, 266, 272, 273, 303, 304, 306, 314, 370, 376, 377, 379, 380, 382, 383, 390, 391, 392, 393, 397, 409, 410 combustion chamber, 28, 304 combustion processes, 42, 43, 376, 392 commercialization, 42, 239, 410 commodity, 409 community, 241 compaction, 160 compatibility, 54, 157 competition, 7, 33, 318 competitiveness, 45 competitor, xiii, 375, 390 complexity, 2, 55, 100, 229, 273 complications, 217 components, x, 2, 4, 6, 53, 56, 58, 60, 62, 63, 69, 70, 71, 73, 74, 76, 77, 78, 80, 87, 90, 91, 95, 98, 103, 113, 118, 130, 145, 147, 148, 156, 167, 194, 195, 208, 209, 211, 237, 238, 268, 279, 348, 351, 352, 353, 354, 356, 372 composites, 69, 74 compounds, ix, 1, 2, 3, 4, 5, 6, 7, 9, 11, 12, 14, 19, 21, 22, 23, 24, 25, 26, 28, 30, 32, 34, 35, 39, 44, 48, 55, 62, 76, 87, 90, 119, 123, 148, 151, 154, 158, 160, 161, 168, 182, 194, 195, 196, 198, 199, 207, 264, 281, 283, 369, 411 computational fluid dynamics, 347

concentration, 3, 6, 7, 14, 16, 21, 23, 32, 36, 43, 59, 63, 66, 78, 81, 85, 86, 92, 94, 95, 99, 107, 108, 109, 110, 111, 112, 113, 117, 125, 129, 130, 153, 155, 158, 160, 167, 170, 171, 195, 196, 198, 208, 212, 214, 217, 219, 221, 231, 233, 279, 282, 286, 288, 319, 323, 348, 349, 359, 364, 367, 370, 372, 386, 387, 388, 389, 396, 428, 432 condensation, 2, 17, 18, 31, 38, 40, 123, 154, 156, 196, 201, 246, 271, 276, 277, 349, 376, 382, 384 conditioning, 16, 32, 45, 197, 222, 254 conduction, 72, 98, 157, 158, 160, 266, 275 conductive, 97, 165, 282, 352 conductivity, 72, 101, 104, 129, 130, 158, 239, 287, 363 conductor, 238, 239 configuration, x, xii, xiii, 26, 28, 29, 30, 31, 40, 54, 55, 56, 87, 115, 171, 172, 237, 244, 251, 253, 263, 271, 272, 274, 285, 286, 289, 297, 299, 303, 304, 375, 376, 382, 384, 427 Congress, 260, 373, 393 conjugation, 130 consciousness, 192 consent, 148 constraints, 241, 386, 412 construction, 33, 352 consumers, 58 consumption, 6, 28, 39, 54, 59, 69, 103, 111, 126, 144, 164, 200, 233, 238, 241, 245, 257, 303, 338, 346, 375, 382 contact time, 33, 55, 58, 61, 63, 64, 66, 67, 69, 71, 73, 83, 97, 99, 101, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 117, 118, 119, 120, 121, 123, 125, 126, 130, 252, 254, 396, 397, 412 contaminant, 169, 424 contaminants, xiv, 6, 18, 19, 34, 42, 164, 166, 197, 409, 413 contingency, 305 control, 5, 9, 26, 29, 38, 41, 46, 48, 127, 164, 165, 167, 168, 182, 183, 204, 207, 209, 226, 232, 235, 281, 283, 309, 314, 332, 348, 350, 413, 426 convection, 266, 286, 296 convective, 271, 305, 306 conversion rate, 56, 215, 326 cooling, 17, 105, 116, 172, 182, 280, 378, 379, 381, 386 COP, 220 copper, 20, 35, 70, 129, 154, 161, 165, 168, 171, 184, 186, 211, 308, 309, 319, 320, 391 copper oxide, 211 corn, 41, 43, 368 correlation, 76, 87, 173, 213, 214, 215, 216, 217, 218, 219, 318, 353, 355 correlations, 90, 95

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Index corrosion, 19 corrosive, 184 cost effectiveness, 39 costs, 2, 7, 13, 17, 30, 33, 34, 36, 42, 144, 145, 166, 196, 222, 230, 231, 234, 237, 250, 303, 304, 305, 306, 307, 308, 379, 396, 429, 431 Coulomb, 370 coupling, 33, 60, 95, 126, 235, 255, 258, 273 covering, ix, 177, 276 crack, 11, 12 cracking, 3, 4, 9, 11, 12, 17, 20, 21, 23, 24, 33, 34, 50, 51, 57, 58, 67, 68, 71, 82, 86, 114, 119, 123, 154, 196, 203, 222, 236, 238, 264, 379, 397, 411 Crete, 140 critical temperature, 151, 161, 279 crops, 410 crude oil, 68, 119, 192, 222, 264 cryogenic, 39, 42, 43, 144, 194, 230, 237, 238, 239, 422 crystal phases, 408 crystalline, 411 crystallites, 89, 199, 209, 211, 283, 320, 411, 414 crystallization, 407 crystals, 327, 401 CSR, 306 CVD, 164, 281, 282, 283 cycles, xi, 3, 20, 23, 40, 141, 154, 163, 176, 177, 178, 179, 180, 185, 239, 281 cyclohexanone, 308 cyclone, 17, 18, 21, 196, 379

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

D database, 356 dating, 145 decane, 76, 80, 82, 84, 85, 117, 118, 119, 120, 121 decomposition, 7, 10, 11, 12, 15, 23, 24, 49, 51, 57, 59, 63, 94, 164, 167, 194, 209, 246, 329, 336, 397 decomposition reactions, 10 decomposition temperature, 164, 336 defect formation, 165 defects, xi, 62, 63, 141, 164, 166, 167, 174, 180, 185 deficiency, 421 definition, 6, 195, 283 deformation, 279 degradation, 19, 39, 209, 411 degradation rate, 39 dehydration, 38 dehydrocyclization, 325 dehydrogenation, 58, 67, 123, 256, 308 delocalization, 143 Demonstration Project, 391 Denmark, 373

439

density, 62, 67, 73, 144, 162, 171, 175, 183, 206, 210, 211, 287, 320, 332, 353, 356, 357, 358, 359, 360, 371 density values, 360 Department of Energy, 188, 239, 270, 373 deposition, xi, 2, 9, 12, 13, 15, 16, 20, 22, 23, 37, 59, 60, 61, 67, 68, 71, 80, 118, 142, 154, 157, 163, 164, 165, 166, 167, 168, 171, 172, 173, 174, 176, 177, 178, 179, 180, 185, 186, 210, 216, 227, 233, 234, 235, 238, 240, 241, 242, 244, 245, 246, 248, 250, 252, 254, 257, 281, 282, 283, 284, 310, 315, 319, 332, 397, 398, 411 deposition rate, 68 deposits, 16, 20, 57, 60, 220, 282, 427 depreciation, 303, 306, 307 depression, 240, 365 derivatives, 55 desorption, 23, 148, 151, 152, 158, 168, 200, 201, 203, 217, 277, 279, 280, 401, 414 destruction, 8, 10, 21, 23, 24, 27, 160, 196, 234 detachment, 118, 165 detection, 248, 387 deviation, 116 dew, 20, 26, 196, 349, 352, 363, 365 dielectric materials, 283 diesel, xi, 28, 31, 32, 34, 36, 54, 60, 143, 144, 191, 193, 206, 222, 229, 313, 314, 315, 333, 338, 346, 396, 410 diffuse reflectance, 321 diffusion, 62, 128, 151, 152, 153, 154, 155, 156, 160, 163, 170, 206, 234, 244, 246, 275, 276, 277, 278, 283, 284, 308, 329, 412 diffusion rates, 329 diffusivity, 151, 156, 161, 232, 278, 372 direct costs, 305 discontinuity, 180 discs, 162, 163, 169, 171, 176 dispersion, 23, 128, 197, 234, 246, 248, 286, 296, 326, 328, 329, 332, 336 disposal areas, 2 dissociation, 14, 61, 66, 69, 153, 159, 200, 201, 211, 213, 277, 278, 323, 327, 397 distillation, 35, 36, 39, 76, 78 distillation processes, 36 distribution, xi, 10, 12, 47, 54, 56, 71, 73, 76, 80, 84, 85, 86, 87, 92, 95, 100, 101, 108, 123, 126, 144, 145, 156, 163, 174, 191, 192, 198, 202, 203, 207, 208, 221, 223, 232, 248, 255, 315, 316, 318, 320, 325, 349, 351, 401, 403 diversification, ix, 1 dopants, 260, 400, 407 doped, xiv, 45, 62, 63, 64, 65, 67, 69, 70, 71, 122, 158, 254, 259, 349, 396, 398, 400, 401

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

440

Index

doping, 49, 70, 209 dosing, 240 double bonds, 34 draft, 9, 271, 272 DRIFT, 321, 332, 335 drying, 128, 169, 194, 208, 324, 328, 336, 391 DSC, 329 durability, 97 dust, ix, 1, 28, 32, 147, 195, 196, 390 duties, 304, 371

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

E earth, 12, 61, 62, 260, 352, 407 economic growth, 42 economics, 222, 418 economies of scale, 36 effluent, 56, 368, 369, 404, 405 electric charge, 18 electric current, 72 electric energy, 370, 371 electric power, xii, 31, 263 electricity, 3, 27, 28, 31, 33, 34, 43, 51, 54, 142, 264, 303, 311, 396 electrochemical reaction, 370 electrodeposition, 165, 166, 281, 309 electrodes, 18, 19, 238, 239, 349 electroless deposition, 282, 283 electrolysis, xiii, 143, 264, 268, 345, 346, 347, 348, 349, 350, 352, 353, 354, 356, 358, 359, 360, 361, 364, 365, 366, 367, 369, 370, 371, 372, 373 electrolyte, 68, 347, 349, 352, 363 electron, 157, 158, 167, 168, 239, 320, 332 electron density, 332 electron microscopy, x, 141, 168 electronic structure, 198 electron-probe microanalysis, 168 electrons, 157, 158, 167, 277, 284, 320 electroplating, 164 e-mail, 409 emission, xv, 2, 9, 42, 44, 142, 147, 193, 241, 338, 417 endothermic, xi, 4, 7, 25, 55, 58, 59, 60, 68, 72, 80, 86, 98, 104, 107, 111, 118, 145, 195, 227, 229, 230, 233, 236, 240, 241, 265, 266, 271, 336, 355, 370, 371, 376, 377, 381, 400, 410, 419, 422, 423, 425, 426 energy consumption, 39, 338, 375, 382 energy density, ix, 54, 55, 100, 127, 144 energy efficiency, xii, 17, 28, 39, 263, 271 Energy Efficiency and Renewable Energy, 188 Energy Information Administration, 223 energy supply, 54

energy transfer, 142 engines, 3, 20, 23, 28, 29, 43, 143, 145 environment, 30, 42, 44, 76, 131, 169, 170, 271, 273, 274, 285, 296, 368, 385 environmental impact, 27, 28, 42, 192, 313 environmental issues, 192 environmental regulations, 396 equilibrium, x, xiii, 14, 23, 39, 47, 54, 58, 60, 61, 70, 76, 77, 79, 81, 86, 91, 100, 108, 116, 153, 170, 204, 221, 232, 233, 237, 239, 240, 242, 249, 250, 271, 272, 278, 279, 296, 302, 345, 347, 353, 354, 355, 356, 357, 358, 359, 361, 362, 365, 366, 367, 368, 370, 372, 376, 378, 384, 386, 388, 389, 411, 424 estimating, 419 etching, 75, 127 ethane, 204, 214, 260, 265, 319 ethanol, x, 2, 34, 36, 37, 38, 39, 41, 43, 46, 50, 51, 55, 67, 68, 69, 70, 71, 73, 87, 108, 111, 112, 113, 114, 128, 204, 236, 254, 258, 272, 273, 314, 369 Ethanol, 36, 43, 70, 268, 369, 370 ethers, 55 ethyl acetate, 369 ethylene, 87, 143, 182 ethylene glycol, 87, 182 ethylenediamine, 328 Europe, 28, 144 European Commission, 131 European Community, 131 European Union, 131 evaporation, x, 53, 61, 73, 108, 164, 210, 284 evolution, x, 47, 51, 90, 91, 103, 141, 167, 174, 176, 177, 213, 217, 220 EXAFS, 327, 328, 330, 336 experimental condition, 5, 10, 12, 13, 19, 26, 28, 33, 34, 37, 116, 124, 210, 252, 365 explosions, 95 explosives, 422 exposure, 72, 186, 213, 279, 318, 319, 327 external environment, 184 extinction, 87 extraction, 30, 280, 303, 379, 422, 424, 427, 428 extrusion, 74 Exxon, 329, 420, 426, 427

F fabric, 17 fabricate, 282 fabrication, 169, 279, 282 failure, 2, 59 family, 61, 157 fatty acid, 54

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

441

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Index FCC, 24 February, 51, 188, 373, 430, 432 feedback, 348, 350 feeding, 2, 55, 146, 148, 149, 150, 169, 183, 237, 244, 397, 414 feedstock, ix, xi, 1, 5, 7, 11, 12, 22, 23, 26, 30, 36, 45, 47, 68, 191, 227, 230, 235, 265, 266, 268, 270, 271, 285, 288, 290, 291, 292, 293, 294, 295, 296, 300, 302, 320, 346, 371, 372, 396, 420, 422, 425, 426 FeMn, 215 fermentation, x, 2, 3, 36, 41, 43, 51, 68, 368, 369, 370 ferromagnetic, 327 fertilizer, 31, 375 fibers, 73, 101, 150, 152 162 filament, 210 fillers, 73 film, x, 92, 141, 154, 157, 159, 161, 162, 164, 165, 166, 167, 168, 169, 171, 173, 175, 176, 177, 178, 179, 180, 185, 186, 281, 283, 284, 310 film formation, 164, 179 film thickness, x, 141, 169, 171, 176, 177, 178, 179, 180, 185, 186 filters, 17, 18, 21, 22, 196 filtration, 17, 18, 20, 22 financial support, 186 financing, 431 fines, 12, 209 fire, 48 Fischer-Tropsch synthesis, ix, x, xi, xiii, 2, 3, 25, 31, 32, 33, 35, 47, 48, 191, 197, 224, 255, 313, 319, 341, 395, 396, 405, 406 fixation, 220 fixed bed reactors, 26, 33, 58, 206 flame, 164 flammability, 144 flexibility, 3, 281 flow, x, xiii, 2, 7, 9, 12, 18, 20, 21, 25, 26, 33, 53, 56, 59, 69, 73, 87, 98, 99, 100, 101, 104, 105, 106, 109, 114, 115, 116, 117, 118, 126, 128, 146, 153, 154, 155, 159, 163, 164, 173, 174, 181, 182, 183, 206, 237, 242, 244, 266, 267, 275, 276, 278, 286, 287, 318, 345, 348, 350, 351, 353, 354, 355, 356, 358, 359, 360, 362, 363, 364, 367, 369, 371, 379, 385, 388, 392, 398, 399, 418, 420, 421, 424, 429, 432 flow field, 87, 351 flow rate, xiii, 7, 12, 26, 56, 69, 98, 99, 100, 101, 104, 105, 106, 109, 115, 116, 117, 118, 126, 173, 206, 244, 345, 348, 349, 353, 354, 355, 356, 358, 359, 360, 362, 363, 367, 369, 388, 398, 399 fluctuations, 182, 386

flue gas, 2, 3, 34, 42, 43, 51, 303 fluid, xi, 89, 115, 142, 164, 172, 184, 272, 289, 299, 347, 379 fluidization, 379, 385, 388, 390 fluidized bed, xiii, 2, 9, 12, 13, 16, 17, 21, 22, 24, 32, 33, 44, 45, 46, 47, 48, 49, 50, 146, 193, 194, 257, 375, 378, 379, 380, 382, 383, 384, 389, 390, 392, 393, 410 flushing, 18, 432 FMF, 223, 342 foams, 56, 72, 206 focusing, 159 foils, 72 food, 43, 144 Ford, 145 formaldehyde, 35, 168, 229 fossil, ix, x, xiv, 1, 28, 31, 42, 44, 54, 55, 141, 142, 143, 145, 147, 149, 185, 188, 222, 264, 346, 375, 409 fossil fuels, ix, x, xiv, 1, 28, 31, 42, 44, 54, 55, 141, 142, 143, 145, 147, 149, 222, 264, 346, 409 fouling, 21, 152, 196, 198 Fourier, 330 fractional composition, 82 fractures, 279 France, viii, 53, 132, 223, 246, 409 free energy, 14, 76, 204, 205, 400 free-radical, 92 friction, 287 FTIR, 182 FT-IR, 321 FTS, xi, xiii, 191, 192, 193, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 209, 210, 211, 212, 213, 214, 215, 216, 217, 219, 220, 221, 222, 313, 314, 315, 316, 317, 318, 319, 320, 322, 323, 324, 325, 326, 328, 329, 330, 332, 333, 334, 336, 338 fuel cell, ix, 1, 3, 19, 39, 43, 54, 63, 68, 69, 126, 128, 142, 143, 145, 154, 231, 240, 258, 349, 359, 360, 413 fuel flow rate, 98 fuel type, 87 furnaces, xi, 3, 227, 349

G gadolinium, 62 gas, ix, x, xi, xii, xiii, xiv, xv, 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 33, 34, 35, 36, 37, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 54, 55, 56, 59, 61, 67, 68, 71, 73, 74, 76, 78, 79,

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

442

Index

gas chromatograph, xiii, 101, 123, 345, 347, 348, 358, 361, 366, 368, 372 gas phase, 4, 8, 42, 71, 90, 91, 92, 95, 123, 153, 278, 371 gas separation, 40, 150, 428, 430 gas turbine, ix, xii, 1, 3, 16, 20, 28, 29, 30, 31, 34, 36, 43, 107, 263, 304, 306 gases, ix, 11, 20, 31, 36, 39, 42, 44, 51, 54, 147, 149, 151, 152, 156, 184, 186, 193, 195, 196, 213, 230, 232, 233, 239, 244, 251, 256, 267, 268, 279, 348, 356, 358, 379, 385, 397, 411, 414, 415, 423, 424, 428 gasification, ix, x, xiv, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 19, 20, 21, 22, 23, 26, 27, 28, 29, 34, 39, 40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 60, 66, 67, 71, 119, 122, 141, 142, 143, 146, 147, 188, 193, 194, 221, 222, 229, 236, 239, 264, 268, 269, 346, 375, 385, 391, 392, 398, 409, 410, 411, 417, 418, 420, 421, 422, 423, 424, 425, 426, 427, 428, 431 gasifier, 2, 6, 9, 11, 12, 21, 23, 29, 30, 31, 39, 46, 48, 49, 51, 410, 413, 418, 420 gasoline, ix, x, xi, 33, 34, 35, 36, 53, 54, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 98, 99, 100, 102, 191, 206, 228, 236, 313, 346, 396 GCC, 28 gel, 328, 398 generation, ix, x, xiii, 1, 3, 20, 28, 30, 36, 43, 50, 53, 54, 55, 56, 63, 66, 67, 105, 114, 130, 142, 145, 165, 234, 239, 249, 266, 268, 290, 291, 292, 293, 300, 302, 303, 304, 306, 314, 375, 376, 391, 410 generators, ix, 1, 130 geometrical parameters, 290 Germany, 31, 35, 53, 192, 193, 256, 421, 426 GHG, 273, 391 Gibbs, 14, 48, 76, 80, 204, 205, 241, 386 Gibbs energy, 80 Gibbs free energy, 14, 76, 204, 205 glass, 6, 144, 157, 162, 163, 172, 174, 176, 246, 283, 349 glassy polymers, 40 global warming, 192, 346 glycerol, 60, 69 glycine, 334 glycol, 19 goals, 346 gold, 161 grain, 32, 43, 129, 168, 169, 177, 178, 179 grains, 166, 176, 177, 179, 185, 368 graph, 268 graphite, 213 grasses, 43 Greece, 140

greenhouse, xiv, 30, 35, 42, 54, 55, 143, 147, 149, 233, 241, 271, 273, 346, 396, 409 Greenhouse, 226, 392 greenhouse gas (GHG) xiv, 30, 35, 42, 54, 55, 147, 149, 233, 271, 273, 346, 396, 409 greenhouse gases, xiv, 42, 54, 55, 233, 346, 409 groups, 14, 32, 71, 200 growth, 3, 23, 33, 38, 41, 42, 200, 201, 202, 203, 209, 314, 315, 316, 319, 323, 329, 332, 337, 338 growth mechanism, 316 guidance, 385, 391

H halogens, 5, 19, 26 handling, 427 hands, 158 Hawaii, 373 heat capacity, 298 heat conductivity, 72, 129 Heat exchangers, 304, 371, 431 heat loss, 25, 76, 116, 118, 236, 358, 386, 387 heat release, 377, 378 heat removal, 38, 206 heat transfer, 71, 87, 105, 116, 126, 171, 206, 207, 265, 272, 289, 298, 356, 358 heating, 6, 7, 8, 28, 55, 58, 59, 72, 76, 78, 145, 170, 171, 182, 209, 210, 222, 268, 272, 289, 299, 355, 370, 372, 381, 397, 398, 412, 418 heating rate, 170 heavy metals, 11, 411 height, 73, 351, 388 hematite, 198, 208, 210, 212, 214, 215, 216 heptane, 260 heterogeneity, 2 heterogeneous, 37, 38, 92, 166 high pressure, 31, 38, 42, 43, 94, 108, 148, 152, 206, 230, 233, 266, 267, 277, 288, 307, 309, 347, 428 high temperature, xi, 7, 9, 11, 17, 20, 22, 32, 41, 44, 55, 57, 60, 63, 69, 70, 72, 80, 83, 91, 95, 104, 107, 127, 153, 157, 163, 164, 165, 166, 169, 181, 184, 191, 194, 197, 198, 208, 209, 212, 217, 230, 231, 232, 233, 235, 236, 238, 256, 265, 266, 267, 271, 272, 309, 347, 348, 368, 371, 372, 396, 400, 411, 412, 414, 428 homogeneity, 209 homogeneous catalyst, 37 homogenous, xiv, 409, 413 homolytic, 62, 63 Honda, 145, 258 hot spots, 72, 237, 397, 413 HPLC, 182, 183, 242 HSC, 355, 373, 413

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Index hybrid, xii, 36, 40, 48, 50, 263, 303, 304, 305, 307, 391 hydrazine, 167, 170, 171, 282, 283 hydride, 181 hydrides, 144 hydro, ix, x, xi, xii, xiv, 1, 2, 4, 5, 6, 7, 8, 9, 10, 15, 19, 21, 22, 24, 25, 26, 31, 32, 33, 34, 35, 38, 39, 53, 54, 55, 57, 58, 59, 60, 61, 62, 63, 68, 72, 76, 80, 95, 119, 123, 130, 146, 154, 191, 192, 193, 195, 196, 197, 199, 200, 201, 202, 204, 205, 207, 210, 211, 212, 213, 215, 221, 222, 227, 228, 233, 236, 241, 254, 263, 264, 314, 319, 322, 323, 327, 330, 332, 333, 337, 338, 346, 396, 409, 410, 422 hydrocarbon fuels, 54, 59, 61, 62, 74, 76, 80, 96, 126, 240, 346, 376, 380 hydrocarbons, ix, x, xi, xii, xiv, 1, 2, 4, 5, 6, 7, 8, 9, 10, 15, 19, 21, 22, 24, 25, 26, 31, 32, 33, 34, 35, 38, 39, 53, 54, 55, 57, 58, 59, 60, 61, 62, 63, 72, 76, 80, 95, 123, 130, 146, 154, 191, 192, 193, 195, 196, 197, 199, 200, 201, 202, 204, 205, 207, 210, 211, 212, 213, 215, 221, 222, 227, 228, 233, 236, 241, 254, 263, 314, 319, 322, 323, 327, 330, 332, 333, 337, 338, 346, 396, 409, 410 hydrocracking, xiii, 34, 395 hydrodynamics, 87, 371, 384 hydrogen abstraction, 200 hydrogen atoms, 63, 76, 78 hydrogen chloride, 47 hydrogen gas, 348 hydrogen peroxide, 166 hydrogen sulfide, 45, 424 hydrogenation, xi, 32, 33, 37, 50, 191, 193, 200, 201, 203, 213, 214, 220, 221, 256, 264, 308, 314, 318, 319, 320, 322, 323, 332 hydrolysis, 8, 43, 196, 207 hydrophilic, 68 hydrothermal, 63, 73, 112, 234, 278 hydrothermal synthesis, 234 hydroxide, 167, 198, 212, 398, 408 hypothesis, 12, 153, 286, 332, 365 hysteresis, 401 hysteresis loop, 401 Hyundai, 145

I IB, 161 id, 401 Idaho, 345, 368, 373 identification, xi, 76, 191, 213, 215 identity, 198, 212, 213, 214 IEA, 187, 268, 391 ignition energy, 143

443

Illinois, 391 images, 73, 74, 118, 121, 331 immersion, 170, 180 implementation, 193 imports, 13, 417 impregnation, 13, 70, 73, 74, 197, 209, 217, 246, 281, 328, 329, 336, 337, 401, 405 impurities, 16, 28, 31, 39, 148, 154, 194, 195, 196, 267, 368, 369, 382 in situ, 48, 198, 200, 211, 213, 217, 221, 234, 330, 399, 412 inactive, 209, 314 incentives, 34, 222 incineration, 2 inclusion, 26, 30, 246 independence, 346, 347, 372 Indiana, 28 indication, 169 Indigenous, 418 indirect measure, 198 induction, 166, 216, 220, 221, 318, 319 induction period, 216, 221, 318, 319 induction time, 166 industrial, x, xiii, 16, 28, 39, 43, 61, 70, 122, 141, 143, 159, 181, 186, 193, 205, 220, 221, 228, 231, 233, 234, 239, 250, 265, 272, 273, 290, 308, 310, 368, 390, 395, 397, 398, 406, 409 industrial application, 143, 159, 181, 186, 205, 233, 239, 290, 308, 398 industrial production, 193, 273 industrial wastes, 43 industry, ix, 1, 14, 31, 32, 61, 144, 154, 193, 264, 266, 410 inert, 3, 9, 30, 31, 33, 44, 150, 164, 165, 196, 209, 213, 242, 280, 284, 348, 355, 356, 380, 385 infancy, 307 infinite, 154, 155, 158, 160, 244, 249, 286, 297, 301, 307 Information Age, 270 infrared, 349 infrastructure, 43, 54, 346 inhibition, 199 initiation, 200, 203, 323 injection, 30, 422, 424, 426, 428 inorganic, 14, 17, 20, 40, 119, 164, 166, 195, 208, 231, 256, 309 inorganic salts, 164 insertion, 200, 201, 203, 315, 317 instability, 160 instruments, 168, 217 integration, xii, 28, 29, 30, 39, 42, 46, 59, 244, 263, 271, 272, 273, 274, 307, 357, 358, 384 integrity, 199, 280, 392

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

444

Index

interaction, 87, 89, 209, 210, 211, 217, 232, 314, 326, 327, 328, 329, 332, 335, 336, 338, 402 interactions, 56, 67, 76, 197, 209, 212, 217, 402, 407 interface, 69, 92, 153, 166, 282, 385 interfacial layer, 234 internal combustion, 143, 145, 410 International Energy Agency, 255 interphase, 106 interstate, 187 interstitials, 157 intrinsic, 72, 98, 151, 197, 210 investment, xiv, xv, 195, 222, 230, 239, 303, 304, 305, 306, 397, 417, 418, 419, 421, 424, 425, 426, 427, 430, 431 ion beam, 210 ion transport, 237 ionic, 238, 355, 356, 357, 363, 364 ions, 62, 165, 166, 167, 207, 208, 210, 284, 328 Ireland, 53 iridium, 24, 61, 398 iron, xi, 11, 12, 13, 20, 21, 23, 32, 33, 48, 49, 191, 192, 197, 198, 199, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 219, 220, 224, 267, 318, 319, 320, 321, 322, 323, 371, 376, 391 iron-manganese, 211 isomerization, 203, 210, 319 isothermal, 79, 80, 81, 84, 100, 207, 355, 358, 370 isotherms, 217, 401, 402 Italy, 141, 186, 227, 256, 263, 393 iterative solution, 357, 358

J

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Japan, 28, 35, 256, 313, 338, 341 jet fuel, 54, 60 Jun, 48, 50, 134, 226, 283, 310, 323, 341, 407 Jung, 48

K kerosene, 34, 54, 315 ketones, 32, 37, 55, 71 kinetic model, 87, 89, 90, 92 kinetic parameters, 90, 92 kinetics, 38, 51, 56, 61, 66, 87, 95, 200, 202, 203, 204, 213, 256, 267, 283, 287, 302, 310, 347, 358, 371, 393, 400 King, 224, 406 KOH, 15, 21, 22 Korean, 393 Kyoto Protocol, 42

L labor, 33, 303 lactic acid, 87 land, 2 lanthanum, 48, 62, 259, 398 large-scale, xiii, 18, 32, 34, 42, 54, 61, 160, 194, 197, 231, 233, 256, 310, 345, 346, 410 laser, 75, 127, 209 lattice, 62, 63, 66, 69, 70, 130, 152, 153, 154, 157, 158, 160, 161, 213, 238, 277, 278, 284, 336 lattice parameters, 213 law, 152, 153, 154, 156, 277, 278, 288 laws, 186 LDP, 25 leakage, 184, 364, 371, 387 legislation, 144 life quality, 4 lifestyles, 44, 192 lifetime, 32, 38, 67, 97, 274, 307, 390, 431 lignin, 55 lignocellulose, 37, 43 likelihood, 347 limestones, 24 limitation, 221, 234, 237, 396 limitations, xii, 29, 55, 89, 128, 181, 186, 196, 204, 206, 227, 229, 249, 310, 315 linear, 32, 126, 163, 197, 198, 203, 210, 319, 321 linear dependence, 126 links, 326, 329 liquefaction, 418 liquid fuels, xiv, 32, 33, 36, 43, 44, 54, 73, 95, 98, 100, 192, 193, 195, 222, 228, 255, 409, 410 liquid hydrogen, 43 liquid interfaces, 55 liquid nitrogen, 330 liquid oxygen, 108 liquid phase, 11, 14, 18, 42, 332 liquids, xi, 2, 14, 49, 50, 55, 108, 123, 191, 192, 193, 255, 371 lithium, 398 lithography, 75 LNG, 418 localization, 174, 180 London, 134, 187, 343, 415 long period, 35, 186 losses, 25, 28, 126, 142, 236, 353 Low cost, 21 low molecular weight, 156, 196 low power, 144 low risk, 384 low temperatures, 60, 67, 197, 232 LPG, 31

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Index lubricating oil, 19, 119 lumen, 244, 245, 246

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

M machines, 304 macropores, 401 maghemite, 209 magnesium, xiv, 15, 22, 24, 47, 396, 398, 399, 400, 401, 402, 404, 405, 406 magnetic, 171, 284, 327 magnetite, 198, 207, 209, 212, 215, 219, 318, 319 magnetron, 310 magnetron sputtering, 310 magnets, 284 maintenance, 33, 274, 303, 431 Maintenance, 303, 306 Malaysia, 33, 315, 412 Malta, 140 management, x, 28, 49, 53, 236 manganese, 20, 209, 211 Manganese, 211 manifold, 349, 350 man-made, 220 manufacturer, 183 manufacturing, 72, 230, 260 mapping, 331 market, xv, 6, 34, 144, 145, 193, 222, 319, 324, 417, 431, 432 markets, 31 Mass Flow, 253 mass spectrometry, 122, 123 mass transfer, 9, 42, 72, 97, 171, 172, 204, 206 materials science, 107 matrix, 40, 48, 63, 73, 239, 402, 405 measurement, xiii, 183, 184, 186, 213, 321, 335, 345, 348, 349, 353, 385, 387 measures, 46 mechanical energy, 28 mechanical treatments, 163 media, xii, 173, 180, 227, 246 melt, 6 melting, 11, 20, 22, 207, 411 melts, 19 membranes, x, xii, 3, 40, 42, 43, 50, 141, 148, 149, 150, 151, 152, 154, 155, 156, 157, 158, 159, 160, 162, 163, 165, 168, 170, 174, 176, 177, 178, 179, 180, 182, 183, 185, 186, 187, 188, 232, 234, 235, 237, 238, 239, 240, 241, 242, 246, 250, 251, 253, 254, 256, 259, 260, 261, 263, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 285, 288, 295, 296, 303, 304, 307, 308, 309, 310, 311

445

mercury, 174 MES, 254 mesitylene, 246 mesoporous materials, 329, 401 metal content, 159 metal ions, 167, 284 metal oxide, 11, 40, 48, 62, 157, 166, 170, 260, 326, 376, 377, 378, 379, 381, 407, 411 metallic catalysts, 314 metallurgy, 144 metals, x, 6, 9, 11, 15, 17, 21, 22, 23, 26, 31, 32, 38, 49, 51, 53, 61, 62, 63, 67, 68, 69, 72, 73, 112, 119, 130, 152, 160, 161, 162, 163, 165, 168, 195, 196, 197, 198, 199, 201, 209, 233, 234, 255, 257, 264, 280, 314, 319, 320, 329, 330, 331, 332, 380, 398, 401, 407 methane oxidation, 90, 92 methanol, ix, x, xi, xiii, 2, 3, 18, 19, 23, 26, 34, 35, 36, 37, 38, 39, 50, 51, 55, 69, 70, 71, 73, 127, 128, 129, 130, 144, 151, 204, 227, 228, 229, 230, 233, 236, 254, 255, 260, 264, 272, 314, 315, 319, 376, 395, 396, 405, 406 methylene, 200 metric, 54 Mg2+, 400 micelles, 208 microchannel reactors, 126 microemulsion, 198, 208, 211, 212 microemulsions, 208 microorganisms, 36 microparticles, 129 microscopy, 168 microspheres, 144 microstructure, 283 migration, 66, 163, 166, 167, 336 million barrels per day, 192 minerals, 6, 9, 10, 13, 19, 20, 21 mining, 424, 426, 427, 429, 430, 433 minority, 208 misleading, 319 missions, 2, 7, 28, 30, 31, 42, 43, 44, 54, 142, 192, 220, 271, 273, 346, 428 mixing, x, xiv, 5, 23, 53, 73, 158, 238, 266, 379, 409, 411, 413 mobile phone, 145 mobility, 62, 63, 66, 70, 130, 411 MOCVD, 164, 166, 283, 309 modeling, x, xii, 30, 48, 54, 89, 90, 91, 236, 256, 259, 263, 287, 289, 299, 310, 347, 391 models, 87, 91, 287, 315, 385 modern society, 192 modules, xii, 149, 150, 152, 263, 271, 274, 285, 304, 305, 307, 309

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

446

Index

moisture, 147, 163, 184 moisture content, 147 molar ratio, xiii, 8, 34, 80, 83, 84, 85, 86, 87, 98, 99, 100, 117, 128, 229, 241, 317, 333, 346, 395, 405, 406 molar ratios, xiii, 34, 86, 395, 406 molar volume, 279 mole, 88, 89, 99, 203, 236, 249, 323, 349, 353, 354, 355, 357, 360, 361, 367, 370, 420 molecular oxygen, 255 molecular structure, 60 molecular weight, xi, 32, 34, 119, 122, 155, 156, 191, 196, 199, 202, 204, 205, 255, 320 molecular weight distribution, 255 molecules, x, 53, 55, 67, 69, 70, 71, 92, 95, 151, 155, 156, 194, 200, 208, 221, 276, 277 molybdenum, 15 momentum, 285, 287, 296 monomers, 200, 201 morphological, xi, 142, 186 morphology, x, 141, 154, 167, 168, 169, 175, 177, 179, 186, 335 Moscow, 140 Mössbauer, 212, 213, 214, 215, 217, 219, 220, 319 MSC, 157 MSW, 5, 6, 9 MTBE, 143 municipal solid waste (MSW), 5, 6, 43

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

N NaCl, 14 nanocomposites, x, 53 nanocrystalline, 63, 112 nanometers, 166 nanoparticles, 208, 209, 234 nanoreactors, 208 nanostructures, 209 nanotube, 144 naphthalene, 10, 12, 45, 51 natural, ix, xi, xii, xiii, xiv, xv, 9, 12, 13, 19, 20, 21, 23, 27, 28, 29, 30, 33, 36, 42, 43, 54, 73, 101, 103, 107, 114, 115, 116, 117, 144, 145, 146, 147, 191, 192, 193, 194, 195, 198, 221, 222, 228, 229, 231, 254, 255, 256, 258, 263, 264, 265, 266, 268, 270, 271, 273, 279, 285, 303, 307, 309, 313, 317, 324, 346, 372, 376, 379, 385, 387, 388, 391, 392, 406, 410, 412, 417, 420, 421, 422, 423, 424, 425, 429, 430, 431, 432, 433 natural gas, ix, xi, xii, xiii, xiv, xv, 27, 28, 29, 30, 33, 36, 42, 43, 54, 73, 101, 103, 107, 114, 115, 116, 117, 144, 145, 146, 147, 191, 192, 193, 194, 195, 198, 221, 222, 228, 229, 231, 254, 255, 256, 258,

263, 264, 265, 266, 268, 270, 271, 273, 285, 303, 307, 309, 313, 317, 324, 346, 372, 376, 379, 385, 387, 388, 391, 392, 406, 410, 412, 417, 420, 421, 422, 423, 424, 425, 429, 430, 431, 432, 433 Nb, 152 Nd, 61 Netherlands, 28, 44, 45, 49, 227 network, 145, 199 New York, v, vi, 132, 186, 187, 188, 189, 224, 255, 261, 343, 373 New Zealand, 193 Ni, x, xiv, 10, 13, 15, 16, 21, 22, 23, 24, 25, 32, 39, 40, 45, 47, 48, 53, 60, 61, 62, 67, 69, 70, 72, 118, 121, 122, 123, 130, 131, 152, 192, 195, 196, 197, 200, 232, 233, 234, 238, 241, 244, 245, 247, 256, 257, 258, 260, 271, 283, 284, 306, 314, 332, 347, 376, 377, 380, 388, 391, 406, 407, 409, 411, 414, 415 nickel, 9, 13, 14, 15, 16, 21, 22, 23, 24, 26, 32, 35, 45, 51, 61, 63, 67, 75, 127, 128, 129, 165, 197, 233, 234, 260, 283, 309, 348, 349, 351, 352, 358, 366, 391, 396, 398, 399, 406, 407, 411, 414 nickel foam, 129 nickel oxide, 21, 411 Nielsen, 131, 133, 138, 255, 257, 310, 406, 415, 416 NiO, 14, 62, 63, 69, 70, 73, 74, 247, 258, 261, 377, 382, 385, 391, 392, 393, 406, 414 nitrate, 164, 168, 324, 325, 327, 328, 329, 333, 336, 398 nitrates, 207, 208 nitride, 166, 314 nitrogen, xi, xiv, 2, 5, 7, 8, 23, 24, 26, 28, 29, 30, 31, 34, 47, 49, 51, 59, 141, 144, 146, 154, 160, 166, 167, 174, 179, 180, 181, 182, 183, 185, 194, 195, 230, 237, 281, 330, 348, 353, 356, 357, 381, 382, 395, 398, 399, 401, 411, 423, 424 nitrogen compounds, 8, 24, 26, 47, 49, 423 nitrogen gas, 174, 182 nitrogen oxides, 5, 194 nitrous oxide, 42 noble metals, 21, 32, 38, 63, 68, 112, 234, 398 nodes, 63 non-noble, 37, 38 non-renewable, 346 non-uniform, 9, 163 normal, 76, 158 North America, 3 Norway, 193 novelty, 385 nuclear, 44, 264, 309, 346, 347, 372 nuclear power, 44, 346 nuclear reactor, 264, 309 nucleation, 62, 211, 320

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

447

Index nuclei, 167, 211

P

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

O observations, 77, 82, 100, 174, 179, 185, 219 OCs, 379, 385 octane, 33, 35, 370 octane number, 35 oil, x, xi, xiv, 5, 19, 29, 31, 33, 34, 35, 42, 44, 53, 54, 55, 67, 68, 70, 71, 87, 113, 119, 122, 123, 125, 127, 129, 144, 145, 191, 192, 196, 208, 222, 235, 236, 264, 314, 315, 346, 351, 376, 409, 428 oil production, 346 oil recovery, 428 oil refining, 376 oils, 44, 55, 57, 58 olefins, 32, 34, 35, 37, 58, 69, 76, 80, 123, 198, 199, 200, 201, 202, 203, 206, 210, 212, 229, 319, 323, 324 Olefins, 200 oligomers, 55, 67 olive, 5 olive oil, 5 online, 345, 347, 352, 372 opposition, 34 optimization, x, 29, 54, OR, 224, 255, 258, 428 organ, 261, 283, 284 organic, 2, 35, 40, 41, 44, 46, 71, 73, 78, 119, 124, 147, 164, 166, 196, 208, 255, 283, 310, 333, 334, 335, 368, 382, 390 organic compounds, 78, 196 organic fibers, 73 Organometallic, 50 261, 283, 284 orientation, 169 osmotic, 167 osmotic pressure, 167 oxalate, 328 oxidants, x, 53 oxidative, x, 53, 55, 62, 68, 98, 100, 241, 255, 258 oxidative reaction, 55 oxide, xi, xiii, xiv, 11, 13, 15, 20, 21, 23, 39, 40, 49, 61, 62, 63, 64, 65, 66, 67, 69, 122, 157, 175, 191, 197, 198, 199, 209, 211, 212, 214, 215, 216, 217, 220, 235, 238, 259, 260, 320, 322, 327, 330, 332, 345, 346, 347, 348, 352, 353, 368, 372, 373, 391, 396, 398, 400, 402, 407, 411 oxide clusters, 197 Oxygen, xiv, 7, 30, 49, 146, 237, 258, 319, 350, 379, 385, 389, 417, 418, 419, 422, 423, 425, 426, 427 oxygen absorption, 388, 390 oxygen consumption, 103 oxygen saturation, 64 oxyhydroxides, 208

Pacific, 3, 309 palladium, x, 61, 141, 152, 154, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 170, 171, 172, 173, 174, 176, 177, 178, 179, 180, 181, 183, 185, 186, 239, 245, 256, 257, 259, 260, 279, 280, 281, 282, 283, 303, 308, 309, 310, 398 paraffins, 32, 33, 34, 35, 76, 198, 200, 201, 203, 204, 210, 324 parameter, 5, 6, 8, 37, 76, 87, 100, 204, 248, 289, 299, 303, 353, 388 particles, x, xiv, 9, 12, 13, 15, 17, 18, 21, 32, 49, 53, 56, 60, 62, 63, 67, 129, 130, 177, 179, 197, 198, 207, 209, 234, 246, 248, 321, 327, 331, 332, 379, 385, 388, 390, 391, 392, 393, 396, 401, 402, 404, 405, 411 particulate matter, 313 passivation, 213, 217 patents, 149, 159, 231 pathways, 77, 95, 200, 201, 210, 211 PCA, 223 PCM, 419, 424, 427, 428, 430, 431, 432 PDEs, 285, 289, 296, 299 peat, 47 PEMFC, 231 penalties, 30 penalty, 382 pentane, 246, 320 periodic, 14, 32, 161 periodic table, 14, 32, 161 permeability, x, xi, 3, 141, 142, 150, 151, 152, 153, 154, 156, 158, 159, 160, 161, 168, 174, 179, 180, 183, 185, 186, 232, 239, 241, 256, 260, 278, 279, 280, 281, 288, 309 permeable membrane, 237, 239, 310 permeation, xi, 142, 149, 152, 153, 160, 161, 169, 174, 179, 181, 182, 183, 184, 231, 238, 239, 241, 242, 244, 245, 249, 252, 259, 261, 271, 279, 281, 282, 283, 295, 296, 297, 298, 309, 310 permit, 353 perovskite, 72, 130, 157, 158, 234, 239, 257, 259, 260, 352, 398 perovskite oxide, 259 perovskites, 62, 158 petrochemical, x, 141, 143, 149, 151 petroleum, ix, xi, 1, 3, 34, 35, 36, 44, 54, 57, 58, 191, 264, 314, 396 petroleum products, xi, 34, 191 pH, 41, 167, 170, 207, 208, 328, 398, 420, 422 phase diagram, 60, 160, 280 phase transformation, 212, 217, 281 phenol, 55, 71, 87, 308

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

448

Index

phosphate, 400, 408 photoelectron spectroscopy, 213 photosynthesis, 43 photosynthetic, 41, 48 physical properties, 12 pinhole, 249 pipelines, 144 pipelining, 424 planar, 347, 362, 371, 372 plants, xii, 2, 9, 16, 28, 43, 54, 137, 143, 148, 151, 160, 188, 193, 194, 196, 221, 222, 237, 239, 263, 264, 265, 268, 271, 272, 304, 305, 396, 411, 412, 427 plasma, 165, 168, 284, 352 plastic, 5, 6, 49, 74, 127, 279 plastic deformation, 279 platelets, 73, 325 platinum, 61, 62, 63, 72, 89, 165, 329, 330, 331, 332, 349, 358, 398 play, xi, xii, xiv, 31, 39, 40, 43, 72, 87, 92, 191, 222, 263, 396, 409 plug-in, 272 PMA, 168 poison, 154, 195, 197, 211, 411 poisoning, xiv, 9, 15, 19, 20, 21, 26, 33, 34, 39, 154, 161, 181, 195, 196, 409, 411, 413, 414 poisonous, 186 poisons, xiv, 168, 199, 409, 411 polarity, 18, 328 polarization, xiii, 155, 345, 364 pollutants, 2, 3, 5, 6, 19, 28, 30, 39, 42, 51, 54 pollution, 54 polyesters, 41, 63 polyethylene, 5, 19 polyimide, 151 polymer, 63, 68, 151, 275 polymer structure, 151 polymeric membranes, 150, 151 polymerization, 67, 200, 202, 315 polymerization mechanism, 316 polymers, 40, 70, 160, 275 poor, 48, 198 pore, 10, 12, 18, 39, 67, 72, 73, 155, 156, 157, 162, 163, 174, 180, 208, 250, 252, 276, 278, 281, 329, 371, 401 pores, 12, 59, 67, 155, 156, 163, 164, 173, 174, 175, 177, 186, 275, 276, 329, 332, 401, 411 porosity, 12, 68, 72, 73, 128, 156, 162, 163, 169, 232, 399 porous, x, 12, 18, 21, 23, 72, 73, 75, 119, 127, 128, 141, 152, 154, 156, 157, 159, 162, 163, 164, 165, 166, 169, 171, 172, 173, 175, 177, 180, 185, 212,

231, 235, 245, 246, 247, 259, 275, 276, 277, 278, 281, 282, 283, 309, 310, 329, 332, 371, 401 porous materials, 18, 162, 163 potassium, 13, 19, 22, 210, 211, 315, 319, 320, 398, 420 potatoes, 368 powders, 73, 128, 163, 206, 368, 408 power, ix, xii, 1, 3, 20, 28, 29, 30, 31, 36, 42, 44, 50, 142, 144, 145, 231, 239, 263, 303, 304, 306, 307, 346, 349, 350, 351, 358, 377, 379, 385, 386, 391, 392, 410 power generation, 3, 20, 28, 36, 145, 239, 306 power plant, 28, 42, 142, 377, 392 power plants, 142, 377 power stations, 42 PPI, 91 precipitation, xiv, 167, 198, 207, 208, 209, 324, 395 preference, 22 prevention, 167 prices, 17, 36, 43, 44, 143, 192, 222, 270, 307, 409 primary products, 203, 255 printing, 352 probability, 202, 203, 204, 210, 314, 315, 319, 332, 337, 338 probe, 64, 65, 321, 327 process gas, 356, 361, 363, 411, 428 producers, 154 production costs, 34, 303, 306, 307, 308 production technology, 241, 303 productivity, 67, 126, 130, 296, 302, 316, 338, 410 profit, xv, 2, 305, 417, 431 program, 279, 347, 373, 385 promoter, 71, 199, 210, 211, 212, 220, 314, 317, 319, 320, 323, 324, 330, 331 propagation, 92 propane, 265 property, 320, 371 propulsion, 144 protocol, 212, 217 protons, 158, 277 prototype, 232, 275 PSA, x, 39, 43, 141, 147, 148, 149, 267, 268, 304, 384, 419, 423, 425, 429, 431 pseudo, 92, 94, 286, 287, 297 PSS, 282, 310 pulse, 64, 65, 257 pumping, 43, 238, 428, 432 pumps, 109, 164 purification, 35, 148, 149, 157, 236, 237, 273, 308 PVC, 172, 173, 176 pyrogasification, 44 pyrolysis, 2, 8, 9, 42, 47, 48, 50, 51, 55, 60, 64, 66, 71, 87, 119, 164, 194, 207, 281, 410

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Index

Q Qatar, 33, 193 quadrupole, 101 quantum dots, 209 quartz, 399 quasi-linear, 185

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

R R&D, 25, 26, 231, 236, 308 radiation, 266, 399 radical mechanism, 92 radical reactions, 92 radius, 155, 276, 286, 288, 300 Raman, 186, 212, 213, 214, 408 Raman spectroscopy, 213 range, ix, xiii, xiv, 1, 3, 7, 10, 17, 18, 20, 24, 26, 27, 31, 32, 33, 34, 35, 37, 41, 43, 55, 69, 80, 83, 84, 86, 87, 94, 100, 101, 108, 109, 115, 117, 118, 121, 146, 147, 148, 150, 156, 181, 182, 183, 186, 200, 202, 203, 204, 205, 209, 217, 231, 235, 265, 268, 273, 275, 283, 305, 308, 309, 315, 326, 335, 345, 350, 356, 368, 371, 379, 389, 391, 396, 399, 402, 404, 405 rare earth, 62, 260, 407 rat, 70 rate of return, 430 raw material, 193, 194, 346, 390 raw materials, 390 reactant, 21, 200, 213, 229, 245, 251, 264, 268, 356, 370, 371 reactants, 59, 92, 106, 192, 198, 199, 207, 210, 221, 234, 241, 244, 245, 248, 347, 398 reaction mechanism, 35, 70, 87, 88, 89, 92, 106, 200, 233, 255, 314, 315 reaction medium, 14, 216 reaction rate, 32, 94, 195, 198, 199, 211, 215, 217, 220, 287 reaction temperature, xiv, 20, 33, 37, 76, 79, 80, 82, 83, 87, 120, 121, 128, 204, 233, 240, 241, 272, 273, 285, 324, 332, 335, 396, 404, 405, 406, 411, 413 reaction time, 215, 283, 404, 410 reaction zone, 271, 287, 296, 298, 302, 376, 377, 379 reactivity, 24, 33, 62, 70, 71, 78, 82, 86, 204, 209, 210, 213, 214, 316, 321, 388, 407 reading, 365 reagent, 70 reagents, 146, 171 recombination, 277 reconciliation, 385, 386

449

reconstruction, xi, 106, 191, 215, 219, 221, 327 recovery, 6, 25, 31, 148, 149, 151, 236, 245, 260, 271, 272, 294, 304, 420, 427, 428 recycling, 6, 37, 40, 230 Redox, 40, 63, 6947 refineries, xi, 143, 144, 148, 149, 191, 236 refining, ix, 1, 119, 144, 376, 396 refractory, x, xiv, 53, 61, 72, 74, 130, 409, 411, 413 regenerate, 21, 414 regeneration, 9, 13, 15, 20, 25, 26, 40, 51, 58, 67, 68, 148, 414 regulators, 182, 183, 242 rejection, 169 relationship, 211, 213, 225, 319, 325, 326, 327, 335 relaxation, 64, 66 relevance, 76 reliability, 77, 90, 308 renewable energy, 44, 54, 273, 346, 410 renewable resource, 142 research and development, x, 2, 25, 43, 373, 426 reserves, 32, 192, 222, 228, 346 residential, 231 residuals, 3 residues, 3, 45, 46, 47, 55, 410 resistance, xiii, 9, 12, 21, 22, 24, 40, 56, 63, 150, 152, 154, 158, 159, 161, 163, 181, 184, 186, 198, 199, 209, 212, 234, 260, 266, 276, 278, 320, 321, 345, 347, 353, 357, 359, 364, 411 resistive, 118 resolution, 331 resources, xiii, 3, 43, 45, 142, 193, 313, 317, 338, 347, 372 retention, 6, 10, 11, 12, 14, 20, 26, 114 reticulum, 279, 280 rhodium, xiii, 38, 61, 62, 67, 68, 119, 256, 395, 398, 401, 402, 403, 404, 405, 406 rice, 29 rice husk, 29 risk, 6, 19, 42, 55, 150, 238 Rita, 45 robotic, 70 robustness, 56 rods, 73, 350 rolling, 154, 249, 250 Rome, 263, 393 room temperature, 181, 182, 214, 246, 314, 316, 321, 358, 366, 398, 399, 400 roughness, 163, 175, 180, 185, 284 Royal Society, 224 rubber, 5, 172 Russian, 53, 66, 131, 135, 140, 426 Russian Academy of Sciences, 131

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

450

Index

ruthenium, xiii, 23, 32, 61, 67, 197, 234, 246, 395, 398, 401, 402, 403, 404, 405, 406, 407

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

S safety, xi, 72, 87, 101, 142, 144, 181, 182, 236, 426 salt, 164, 167, 282 salts, 22, 164, 168, 207, 309, 411 samarium, 62 sample, 66, 70, 73, 211, 215, 216, 217, 219, 331, 333, 366, 368, 369, 399, 400, 401, 402, 404, 414 sampling, 366 sand, 9, 13, 21 SAS, 193 saturation, 40, 64 sawdust, 18, 22, 48 SBA, 325, 329 scalability, 273, 274, 379 scaling, 2, 130, 206, 271 scanning electron microscopy, x, 141, 168 scientific community, 241 SCRs, 349 SCT, 246 seals, 379, 388 search, 50 second generation, 192 security, 54, 192 segregation, 169 selecting, 20, 371, 372 selectivity, x, 2, 32, 33, 35, 36, 37, 38, 39, 40, 56, 61, 62, 67, 68, 69, 72, 90, 98, 105, 109, 110, 116, 117, 121, 126, 127, 148, 150, 151, 152, 154, 155, 156, 157, 158, 159, 160, 168, 180, 181, 186, 198, 200, 202, 203, 206, 207, 210, 211, 212, 213, 215, 221, 229, 231, 236, 244, 249, 251, 275, 276, 279, 281, 282, 283, 284, 286, 296, 297, 314, 315, 318, 319, 320, 322, 323, 324, 325, 326, 329, 332, 333, 337, 338, 397, 398 Self, 249, 311 SEM, x, 118, 121, 141, 168, 169, 174, 175, 177, 178, 179, 185, 186 SEM micrographs, 175, 178, 179, 185 semiconductor, 154, 209 sensitivity, 88, 90, 154, 232 sensitization, 166, 170 sensors, xiii, 345, 350, 352 separation, x, xii, xiii, 2, 3, 6, 17, 29, 30, 33, 37, 39, 40, 42, 43, 114, 141, 142, 148, 149, 150, 152, 154, 155, 156, 157, 159, 160, 165, 168, 182, 184, 185, 187, 188, 194, 206, 209, 227, 230, 232, 234, 237, 238, 239, 241, 246, 248, 256, 260, 263, 271, 273, 274, 275, 276, 277, 279, 281, 285, 295, 296,

302, 305, 307, 308, 309, 310, 311, 372, 375, 376, 382, 384, 392, 393, 422, 428, 430 services, 430 sewage, 5, 6, 7, 12, 13, 15, 50 shape, 83, 90, 185, 249 Shell, 9, 33, 193, 205, 315, 412 shock, 56, 101 short period, 270 Siemens, 3 sign, 28 silica, 9, 13, 33, 40, 101, 157, 160, 162, 163, 199, 209, 212, 275, 278, 279, 309, 328, 398, 420, 422 silica glass, 162, 163 silicate, 7, 13, 23, 47, 326, 328, 336 silicon, 171, 210 silver, 154, 161, 163, 166, 168, 183, 186, 282, 283, 303, 309, 310 similarity, 77, 316 simulation, 30, 87, 90, 92, 237, 240, 260, 287, 310, 384, 385, 392 simulations, xii, 87, 90, 95, 259, 263, 273, 285, 288, 294, 295, 300, 302 single crystals, 327 sintering, x, xi, xiv, 2, 21, 34, 53, 62, 63, 67, 71, 118, 129, 163, 210, 227, 234, 257, 385, 396, 398, 409, 411, 413, 415 SiO2, 16, 36, 38, 44, 61, 157, 163, 197, 200, 209, 210, 212, 255, 256, 314, 324, 325, 326, 327, 328, 329, 330, 332, 333, 334, 335, 336, 337, 338, 406 SiO2 surface, 328, 332, 336 sites, xi, xiv, 24, 32, 33, 35, 37, 67, 89, 92, 97, 103, 114, 158, 191, 197, 198, 200, 204, 210, 211, 212, 215, 217, 220, 221, 234, 318, 320, 321, 325, 327, 332, 334, 398, 411, 414, 417 skin, 152 sludge, 5, 6, 7, 10, 12, 13, 15, 50 Sm, 61, 63, 70 SME, 392 smoothing, 98 smoothness, 179, 186 SMR, 129, 195, 268, 304 SO2, 8, 11, 44, 314, 414, 415 social conflicts, 192 social structure, 54 sodium, 13, 20, 22, 167, 208, 398 SOFC, 131, 373 software, 29, 78, 242, 384, 385, 413 soil, 44 solar, 44, 264, 273, 309 solar energy, 264, 273 sol-gel, 157, 163, 166, 261, 407, 408 solid oxide fuel cells, 63, 69 solid phase, 6, 10, 11, 14, 91, 318

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Index solid solutions, 63, 69, 112, 209, 212 solid waste, 5, 6, 28, 43 solubility, 151, 152, 161, 279, 333, 336 solvent, 39, 42, 171, 246, 328 soot, 20, 55, 58, 195, 347 sorbents, x, 2, 5, 6, 11, 13, 19, 20, 47, 49 sorption, 11, 51, 69 South Africa, 3, 31, 33, 193, 229, 255 soybeans, 43 space shuttle, 144 space-time, 334, 337 Spain, 28, 191, 392 spatial, 95, 153 speciation, 329, 336 species, x, xiv, 10, 33, 51, 53, 59, 62, 66, 67, 68, 69, 70, 71, 78, 90, 92, 95, 96, 100, 114, 116, 146, 149, 150, 151, 155, 156, 200, 201, 203, 209, 210, 211, 212, 213, 214, 215, 216, 217, 219, 220, 244, 276, 277, 283, 284, 315, 320, 321, 325, 326, 327, 328, 330, 331, 332, 336, 346, 354, 356, 370, 371, 386, 389, 397, 402, 409 specific heat, 287, 371 specific surface, xiv, 72, 396, 399, 401, 405, 411 specificity, 41 spectroscopy, 101, 168, 177, 213, 214, 215, 217, 219, 220, 319, 320, 321 spectrum, 122, 124, 125, 201, 204 speed, 235 spheres, 18, 87, 114 spin, 231 sputtering, 281, 282, 283, 284, 309, 310 stability, x, 14, 15, 20, 32, 36, 40, 54, 56, 60, 61, 67, 70, 72, 92, 98, 112, 115, 129, 130, 152, 154, 158, 159, 161, 197, 198, 209, 236, 274, 278, 279, 281, 282, 307, 324, 332, 333, 336, 343, 379, 398, 399, 400, 404, 411, 412, 413 stabilization, 5, 71, 199, 216, 220, 401, 407 stabilize, 20, 62, 216 stages, ix, x, 2, 39, 157, 212, 213, 218, 219, 231, 236, 239 stainless steel, x, 75, 127, 128, 141, 157, 169, 172, 177, 180, 182, 185, 242, 244, 282, 283, 305, 351 standard model, 385 starches, 41 starvation, 353, 357, 360 STD, 99, 102 steady state, 87, 90, 92, 96, 97, 99, 104, 106, 215, 221, 365 steel, x, 75, 101, 108, 127, 128, 141, 144, 157, 163, 169, 172, 177, 179, 180, 182, 185, 242, 244, 273, 282, 283, 305, 349, 351 STEM, 257, 331, 332 STM, 327

451

stoichiometry, 78, 87, 99, 200, 327 storage, 3, 28, 30, 42, 43, 54, 62, 63, 144, 147, 220, 228, 382, 397, 398, 424 strain, 400 strategies, 16, 27, 220, 222, 398 streams, 40, 42, 54, 67, 114, 149, 159, 182, 239, 242, 356, 372, 377, 385, 386, 390, 428 strength, 14, 20, 23, 98, 152, 154, 157, 159, 160, 161, 163, 197, 198, 332, 380 strong interaction, 15, 70, 402 strontium, 352 structural changes, 150 structural characteristics, 408 structural modifications, 279 styrene, 143 substances, 199, 275, 385 substitution, 143, 257, 328 substrates, x, 53, 61, 71, 72, 74, 75, 101, 104, 108, 112, 116, 117, 123, 130 sucrose, 167, 327 sugars, 37, 71 sulfate, 400, 408 sulfidation, 20 sulfur, 2, 5, 6, 8, 9, 10, 11, 12, 15, 19, 20, 25, 26, 31, 32, 33, 34, 35, 39, 40, 41, 44, 46, 194, 199, 315, 320, 321, 396, 423, 424 sulfur oxides, 5 sulphur, 22, 34, 42, 47, 48, 144, 154, 161, 181, 195, 264, 281, 411, 414, 415 Sun, 223, 225, 328, 335, 341, 342 sunflower, x, 53, 113, 114 superheated steam, 34, 58, 379, 396 supervision, 305 supply, x, xi, 4, 8, 42, 53, 54, 55, 114, 191, 192, 239, 240, 265, 266, 268, 273, 348, 383, 388, 396, 414, 417, 428 surface area, 10, 13, 22, 48, 67, 72, 74, 152, 173, 177, 199, 206, 207, 208, 209, 211, 212, 216, 217, 218, 219, 220, 320, 324, 328, 329, 332, 335, 367, 401, 405 surface diffusion, 155, 156, 276, 277 surface energy, 401 surface reactions, 89, 95 surface roughness, 163, 175, 180, 185 surface tension, 208 surfactant, 208, 209 sustainability, 54 sustainable development, 43, 44 SVZ, 35 Sweden, 35, 193, 385, 391, 392 swelling, 160 switching, 65, 66 Switzerland, 373

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

452

Index

symbols, 90, 97, 99, 360, 361, 366 synthesis, ix, x, xi, xiii, xiv, 2, 3, 19, 20, 23, 25, 27, 31, 32, 33, 34, 35, 36, 37, 38, 43, 47, 48, 50, 51, 54, 61, 67, 68, 70, 74, 78, 87, 90, 93, 114, 117, 141, 142, 144, 145, 173, 191, 192, 193, 194, 195, 197, 198, 199, 200, 209, 211, 216, 221, 222, 224, 228, 229, 230, 234, 251, 254, 255, 257, 258, 259, 260, 261, 264, 281, 310, 313, 314, 315, 318, 319, 322, 323, 338, 339, 341, 372, 375, 376, 380, 382, 384, 386, 390, 396, 397, 405, 406, 407, 408, 410, 411, 412, 423, 425 synthetic fuels, 31, 192, 193, 222, 372

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

T tanks, 368 tantalum, 210 tar, ix, xiv, 1, 2, 3, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 26, 27, 28, 39, 45, 46, 48, 49, 50, 51, 71, 77, 119, 147, 194, 196, 222, 409 tar removal, 17, 20, 21, 23, 46 targets, 373 tax exemptions, 44, 222 technological developments, 222 teflon, 172 TEM, 70, 168, 169, 327, 331 temperature dependence, 358 temperature gradient, 72, 97, 98, 105, 266, 274, 371 temporal, 87 TEOS, 282 Texas, 373 TGA, 329, 399 thermal activation, 197 thermal analysis, xiv, 395, 398, 399 Thermal Conductivity, 242 thermal decomposition, 57, 59, 194, 209 thermal degradation, 411 thermal efficiency, 147, 313, 418, 421, 424, 425, 426, 427, 428, 431 thermal evaporation, 284 thermal expansion, 162, 163 thermal oxidation, xi, 141 thermal resistance, 163, 266 thermal stability, 40, 56, 70, 98, 152, 154, 158, 197 thermal treatment, 166, 167, 168 thermochemical cycle, 264 thermodynamic, xii, 7, 55, 59, 60, 61, 76, 77, 80, 82, 83, 87, 98, 99, 100, 116, 130, 167, 204, 227, 229, 234, 237, 241, 249, 250, 251, 254, 258, 267, 315, 371, 386, 389, 401, 406, 410 thermodynamic calculations, 55, 77, 83, 87, 100, 204

thermodynamic equilibrium, 59, 60, 76, 98, 237, 249, 250, 251, 254, 386, 389, 410 thermodynamics, 23, 61, 87, 116, 204, 302 thermogravimetry, xiv, 395, 398, 399, 400 thin film, 19, 160, 163, 164, 165, 210 threat, 72, 92 threshold, 296, 301, 307, 308 tin, 166, 170 TiO2, 14, 15, 20, 24, 45, 129, 130, 157, 163, 197, 209, 230, 245, 250, 255, 320, 323, 325, 326, 327, 328, 332, 380 titanates, 11, 20 titania, 23, 157, 160, 209, 275 titanium, 74 Tokyo, 231, 232, 341 toluene, 195 Toshiba, 231 total product, 270 toxic, 181, 184 toxic gases, 181 toxicity, 38 Toyota, 145 TPH, 213 traction, 145, 279 trans, 372 transfer, x, 53, 70, 71, 87, 97, 98, 101, 105, 116, 118, 126, 130, 142, 167, 171, 172, 206, 207, 232, 265, 272, 289, 298, 310, 356, 358 transformation, 6, 62, 63, 70, 192, 203, 214, 215, 216, 219, 220, 279, 400, 401, 412 transition, 62, 152, 154, 160, 168, 257, 258, 280, 314, 316 transition metal, 62, 257, 258, 314, 316 transmembrane, 152, 179 transmission, 144, 168, 397 transmission electron microscopy, 168 transparent, 208 transport, 36, 40, 43, 55, 73, 87, 89, 106, 126, 144, 145, 150, 151, 152, 154, 155, 156, 158, 202, 204, 237, 239, 260, 266, 275, 276, 277, 278, 283, 287, 308, 310, 377, 379, 396, 409 transportation, xi, xiii, 27, 30, 33, 34, 43, 54, 127, 144, 191, 192, 222, 228, 313, 338 transportation infrastructure, 43, 54 traps, 123 trichloroethylene, 166 trimethylamine, 209 trucks, 34 tubular, xiii, 33, 101, 108, 128, 150, 154, 164, 169, 170, 171, 172, 173, 174, 175, 176, 179, 180, 182, 184, 185, 186, 205, 246, 248, 249, 250, 254, 257, 259, 266, 273, 274, 275, 375, 376, 380, 382, 383, 410

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

453

Index tungsten carbide, 73, 257 turbulent, 379 Turkey, 260 turnover, 197, 210, 216, 319, 324, 325 two-dimensional, 310

U ultrasound, 209 uncertainty, 270 uniform, xi, 9, 77, 126, 142, 162, 163, 164, 168, 172, 173, 177, 179, 329, 351 United States, 28, 35, 144, 223, 255, 346, 417, 418, 426 universe, 54 updating, 419 urban centres, 143 urea, 70

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

V vacancies, 66, 92, 157, 158, 401 vacuum, 65, 148, 164, 165, 167, 173, 179, 210, 309, 399 valence, 158 validation, 385 values, x, xiii, xiv, 8, 14, 15, 23, 24, 28, 36, 41, 77, 83, 87, 94, 99, 141, 153, 155, 184, 197, 203, 204, 207, 208, 211, 217, 219, 233, 234, 237, 240, 254, 278, 296, 306, 338, 345, 354, 356, 358, 359, 360, 361, 362, 365, 367, 370, 371, 372, 380, 386, 388, 396, 401, 404, 405, 406, 418 vapor, 4, 14, 210, 238, 281, 283, 284, 310, 349, 353 variables, 46, 48, 87, 307, 309, 310 variation, x, 54, 83, 87, 92, 98, 100, 105, 106, 115, 130, 162, 348, 354, 357, 359, 386, 389 vegetables, 264 vehicles, 43 velocity, 71, 87, 99, 100, 102, 115, 116, 126, 127, 211, 252, 286, 287, 386, 388 vessels, 184, 232, 304 viscosity, 173 visible, 117, 177, 350 vision, 46, 188, 392 voiding, 272 volatility, 63, 144 volatilization, 48 Volkswagen, 145

W wall temperature, 101, 289, 290, 291, 295, 299, 300, 301, 302, 307 wastes, ix, 1, 2, 5, 6, 7, 9, 12, 14, 23, 26, 29, 43, 44, 49 wastewater, 17 water evaporation, 87, 108, 164 water gas shift reaction, ix, xiii, 1, 25, 39, 40, 57, 70, 111, 128, 145, 146, 147, 159, 236, 257, 345, 368, 372, 376, 381, 396, 397, 405 water sorption, 69 water vapor, 4, 238, 349, 353 water vapour, 195, 271 water-soluble, 125 waxes, 32, 33, 193, 205 weak interaction, 402 weight gain, x, 141, 169, 170, 171, 175, 176, 177, 185 weight loss, 399 welding, 73, 180, 185, 308 wells, xv, 417, 428, 430, 431, 432 Western Europe, 3 wet-dry, 16 WGSR, 408 wheat, 368 wind, 44 wine, 14 wires, 61, 72, 349, 350 wood, 6, 37, 43, 47, 51, 71, 119, 410 workers, 318, 330 workstation, 70 World War, 193, 346

X XANES, 319, 320, 328 XPS, 168, 169, 213, 220, 326, 327, 328, 336 X-ray absorption, 407 X-ray analysis, 168 X-ray diffraction, xiv, 168, 212, 213, 214, 219, 395, 398 X-ray photoelectron spectroscopy (XPS), 168, 220 XRD, 70, 168, 169, 213, 214, 215, 217, 219, 220, 282, 327, 332, 398, 408 xylenes, 195

Y yield, xiii, xiv, 5, 7, 9, 14, 15, 16, 20, 22, 32, 33, 36, 37, 38, 41, 55, 58, 68, 71, 76, 80, 87, 97, 105, 106, 114, 117, 118, 127, 128, 142, 145, 146, 147,

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook

454

Index

197, 200, 207, 209, 211, 213, 223, 236, 271, 283, 329, 333, 334, 337, 345, 367, 372, 376, 407, 409, 410, 413 YSZ, 67, 69, 73, 74, 380 yttria-stabilized zirconia, 380

Z

Zinc, 20 zinc oxide, 49, 212 zirconia, xiv, 23, 62, 63, 64, 65, 67, 69, 72, 74, 110, 112, 160, 209, 275, 349, 350, 352, 380, 395, 398, 400, 401, 402, 403, 404, 405, 406, 407, 408 zirconium, 63, 398, 399, 400, 401, 402, 405, 408 ZnO, 11, 14, 20, 34, 35, 36, 37, 38, 70, 127, 129, 196, 197, 199

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

zeolites, 20, 23, 35, 39, 246 zinc, 11, 14, 20, 24, 47, 49, 129, 212

Syngas: Production Methods, Post Treatment and Economics : Production Methods, Post Treatment and Economics, Nova Science Publishers, Incorporated, 2009. ProQuest Ebook