Energy Efficient Solvents for CO2 Capture by Gas ... - Springer Link

Green Energy and Technology

Wojciech M. Budzianowski Editor

Energy Efficient Solvents for CO2 Capture by Gas– Liquid Absorption Compounds, Blends and Advanced Solvent Systems

Green Energy and Technology

More information about this series at http://www.springer.com/series/8059

Wojciech M. Budzianowski Editor

Energy Efficient Solvents for CO2 Capture by Gas–Liquid Absorption Compounds, Blends and Advanced Solvent Systems

123

Editor Wojciech M. Budzianowski Consulting Services Wrocław Poland and Renewable Energy and Sustainable Development (RESD) Group Wrocław Poland

ISSN 1865-3529 Green Energy and Technology ISBN 978-3-319-47261-4 DOI 10.1007/978-3-319-47262-1

ISSN 1865-3537

(electronic)

ISBN 978-3-319-47262-1

(eBook)

Library of Congress Control Number: 2016956846 © Springer International Publishing AG 2017 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. Printed on acid-free paper This Springer imprint is published by Springer Nature The registered company is Springer International Publishing AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Preface

Greenhouse gas emissions limit the expansion of many industries and especially considering the whole life cycle it may be difficult and costly to remarkably reduce existing CO2 sources. It is therefore expected that CO2 separation from various gases may be required in the future in order to mitigate emissions and associated climate change. CO2 capture may be required not only in coal-fired power plants but also in many other industries. Besides, it may be needed not only for fossil fuel processing but also for biomass processing. In addition, by efficiently separating and concentrating CO2 a new carbon resource may be created for CO2 utilisation technologies. Consequently, CO2 capture is a very important process that will play a significant role in the future. Its high efficiency achieved by applying innovative techniques will greatly contribute to commercial success of many CO2 emitting and CO2 utilising industries. This edited book is dedicated to developments of solvents for CO2 capture by gas–liquid absorption. The entire process involving these solvents needs to be characterised by very high energy efficiency to enable implementation of this technology in actual industries. Book chapters put emphasis on compounds, their blends and advanced solvent-based capture processes (ASBCPs). Single compound solvents are usually simple and easy to handle. Solvent blends are used for enhancing CO2 capturing processes and are often characterised by reduced energy requirements. Finally, emerging innovative advanced solvent-based capture processes make use of novel phenomena and innovative approaches capable of creating disruptive innovations. The ASBCPs involve for instance two immiscible liquid phases or microencapsulated solvents which have very favourable operating characteristics. Therefore, ASBCPs may greatly contribute to the increase of energy efficiency of CO2 capture plants. The book includes a selection of Chapters dedicated to energy efficient solvents for CO2 capture by gas–liquid absorption. It also discusses the whole technical context of using CO2 capture solvents in practical applications. Chapter “Introduction to Carbon Dioxide Capture by Gas–liquid Absorption in Nature,

v

vi

Preface

Industry, and Perspectives for the Energy Sector and Beyond” introduces CO2 capture by gas–liquid absorption in nature, industry and discusses challenges in the energy sector. Chapter “Assessment of Thermodynamic Efficiency of Carbon Dioxide Separation in Capture Plants by Using Gas–liquid Absorption” assesses thermodynamic efficiency of a CO2 separation process relying on gas–liquid absorption. In Chapter “Process Implications of CO2 Capture Solvent Selection” the process implications in the selection of energy efficient solvents with the overall message that the process should not limit the choice of solvent but be designed to make best use of the advantages of the energy efficient solvent choice and minimise the disadvantages. Chapter “Useful Mechanisms, Energy Efficiency Benefits, and Challenges of Emerging Innovative Advanced Solvent Based Capture Processes” analyses advanced solvent-based capture processes for energy efficient carbon dioxide capture by gas–liquid absorption. Chapter “Phase Change Solvents for CO2 Capture Applications” investigates solvent systems that separate into two phases upon absorption of CO2 having significant potential as energy efficient solvents due to only the CO2 rich stream being heated in the regenerator resulting in reduced sensible and latent heating requirements. Chapter “Aqueous Amino Acid Salts and Their Blends as Efficient Absorbents for CO2 Capture” is dedicated to aqueous amino acid salts and their blends having high CO2 loading, high reaction rate, being less corrosive, less toxic, and requiring less regeneration energy compared to commercial amines. Further, Chapter “Ionic Liquids: Advanced Solvents for CO2 Capture” relates to ionic liquids (ILs) which have become more attractive for CO2 capture because of their excellent properties and potential energy saving efficiency. It reviews and analyses the research progress on CO2 capture with ILs including the absorption capacity, the absorption mechanism and the process simulation and assessment. Furthermore, Chapter “Amine-Blends Screening and Characterization for CO2 Post-combustion Capture” provides the CO2 loading and heat of absorption experimental data of amine blends for CO2 capture which can be useful for future industrial application in the selection of amine blend solvents. Chapter “Postcombustion Carbon Dioxide Capture with Aqueous (Piperazine + 2-Amino-2Methyl-1-Propanol) Blended Solvent: Performance Evaluation and Analysis of Energy Requirements” critically discusses the properties of the AMP+PZ blended solvent characterised by a regeneration energy demand of 2.9–3.7 GJ/tCO2. Chapter “Energy Efficient Absorbents for Industry Promising Carbon Dioxide Capture” reviews and analyses energy efficient absorbents for industrially relevant carbon dioxide capture systems. Chapter “The Absorption Kinetics of CO2 into Ionic Liquid—CO2 Binding Organic Liquid and Hybrid Solvents” is dedicated to absorption kinetics of CO2 into ionic liquids and hybrid solvents as measure to achieve higher efficiency energy utilisation in carbon capture. Finally, Chapter “Solubility of Carbon Dioxide in Aqueous Solutions of Linear Polyamines” provides information on the solubility of carbon dioxide in aqueous solutions of linear polyamines. Overall, this book provides a useful engineering resource for the

Preface

vii

development of energy efficient CO2 capture solvent systems involving gas–liquid absorption. The invited leading authors from different continents give their own perspectives on the associated problems and hence the whole picture is well balanced and up-to-date. Wrocław, Poland

Wojciech M. Budzianowski

Contents

Introduction to Carbon Dioxide Capture by Gas–Liquid Absorption in Nature, Industry, and Perspectives for the Energy Sector and Beyond . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Wojciech M. Budzianowski Assessment of Thermodynamic Efficiency of Carbon Dioxide Separation in Capture Plants by Using Gas–Liquid Absorption . . . . . . . Wojciech M. Budzianowski Process Implications of CO2 Capture Solvent Selection . . . . . . . . . . . . . . Leigh T. Wardhaugh and Ashleigh Cousins Useful Mechanisms, Energy Efficiency Benefits, and Challenges of Emerging Innovative Advanced Solvent Based Capture Processes . . . . . Wojciech M. Budzianowski Phase Change Solvents for CO2 Capture Applications. . . . . . . . . . . . . . . Kathryn A. Mumford, Kathryn H. Smith and Geoffrey W. Stevens

1

13 27

69 99

Aqueous Amino Acid Salts and Their Blends as Efficient Absorbents for CO2 Capture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 Azmi Mohd Shariff and Muhammad Shuaib Shaikh Ionic Liquids: Advanced Solvents for CO2 Capture . . . . . . . . . . . . . . . . . 153 Xiangping Zhang, Lu Bai, Shaojuan Zeng, Hongshuai Gao, Suojiang Zhang and Maohong Fan Amine-Blends Screening and Characterization for CO2 Post-combustion Capture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 Abdullah Al Hinai, Nabil El Hadri and Mohammad Abu Zahra Post-combustion Carbon Dioxide Capture with Aqueous (Piperazine + 2-Amino-2-Methyl-1-Propanol) Blended Solvent: Performance Evaluation and Analysis of Energy Requirements . . . . . . . 191 Sukanta K. Dash ix

x

Contents

Energy Efficient Absorbents for Industry Promising Carbon Dioxide Capture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217 Y.S. Yu, T.T. Zhang and Z.X. Zhang The Absorption Kinetics of CO2 into Ionic Liquid—CO2 Binding Organic Liquid and Hybrid Solvents . . . . . . . . . . . . . . . . . . . . . 241 Ozge Yuksel Orhan, Cyril Sunday Ume and Erdogan Alper Solubility of Carbon Dioxide in Aqueous Solutions of Linear Polyamines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263 Jian Chen, Ruilei Zhang, Zhongjie Du and Jianguo Mi

Introduction to Carbon Dioxide Capture by Gas–Liquid Absorption in Nature, Industry, and Perspectives for the Energy Sector and Beyond Wojciech M. Budzianowski

Abstract The study describes how the process of CO2 capture by gas–liquid absorption is used in nature, industry, and discusses perspectives for its use in the energy sector and beyond. The process is used in nature on a large scale greatly contributing to the global carbon cycle. CO2 is captured from air by oceans and stroma (aqueous fluid) of plant leaf cells during photosynthesis. Step-change biological developments are associated with the evolution of enzymes that catalyse CO2 dissolution in stroma and incorporate it in the cell’s Calvin cycle. In industry the first implementation of CO2 capture was in natural gas upgrading (in 1920s) followed by its use in Enhanced Oil Recovery (EOR) (in 1970s). Developments relied on using more efficient physical and chemical solvents and their blends. Only relatively recently CO2 capture has been proposed as a countermeasure to mitigate CO2 emissions. Perspectives for employing gas-liquid absorption for decarbonisation of the energy sector and modern industries are strengthened by recent solvent based research outcomes. Developments in gas–liquid absorption are essential for wide-scale deployment of CO2 capture and are therefore approached through all Chapters of this edited book. Promising developments relate to e.g. energy efficient solvents relying on innovative advanced solvent based capture processes.

1 Introduction CO2 capture by gas–liquid absorption is an old process widely employed in nature, industry, and in recent years considered for the mitigation of CO2 emissions from the energy sector and beyond. In nature CO2 capture by gas–liquid absorption is an important process of the global carbon cycle [1]. CO2 is absorbed from air by oceans and transformed there into carbohydrates (mainly cellulose and chitin) and W.M. Budzianowski Consulting Services, Wrocław, Poland W.M. Budzianowski (&) Renewable Energy and Sustainable Development (RESD) Group, Wrocław, Poland e-mail: [email protected] © Springer International Publishing AG 2017 W.M. Budzianowski (ed.), Energy Efficient Solvents for CO2 Capture by Gas–Liquid Absorption, Green Energy and Technology, DOI 10.1007/978-3-319-47262-1_1

1

2

W.M. Budzianowski

minerals (mainly calcium carbonate). The scale of CO2 absorption by oceans and other hydrosphere was in 2009 about 338 PgCO2/yr [1, 2]. On land CO2 is mainly absorbed by stroma (aqueous fluid) present in plant leaf cells via photosynthesis thus contributing to biomass productivity. The scale of CO2 absorption by biosphere and soils in 2009 amounted to about 450 PgCO2/yr [1, 2]. A remarkable part of these natural CO2 transfers from the atmosphere was achieved by CO2 capture by gas–liquid absorption meaning that this process is essential in natural conditions for capturing CO2 from air. In industry CO2 capture by gas–liquid absorption was historically applied to natural gas upgrading. Other mostly niche applications related e.g. to chemical and food industry. The scale of these processes was historically relatively small compared to CO2 removals by oceans and biosphere as well as to anthropogenic CO2 emissions. Only relatively recently CO2 capture has been proposed as a measure to counteract CO2 atmospheric emissions from fossil fuel combustion with the aim to mitigate climate change. The scale of CO2 emissions from fossil fuel combustion is big enough to seriously affect the global carbon cycle and in 2009 it amounted to 29 Pg CO2/yr [2, 3]. This newly emerged CO2 capture application has attracted significant global attention and R&D funding. CO2 capture by gas–liquid absorption is a relatively well understood process but cost reduction is required in order to apply CO2 capture on a large scale. In recent years many demonstration and pilot plant projects have been funded all over the world. Research efforts have led to remarkable development of this process by improving its fundamental understanding, optimising performance under conditions seen in the energy sector and beyond, and creating new innovative processes. Historical experience gained from applying CO2 capture by gas–liquid absorption processes in various natural and industrial processes can be employed to create new innovative processes that meet present-day challenges. The scale of CO2 capture required for applications in the energy sector and beyond is much greater than ever before in the industrial history of this process and it is therefore essential that the gas–liquid absorption process is much more efficient than it was previously. In some applications absorption is physical thus if not accompanied by high-pressure it is characterised by relatively small CO2 fluxes. In other applications absorption is chemical meaning that CO2 dissolved in the liquid phase undergoes one or more chemical reactions. For example, living plants utilise diluted CO2 from air. Since physical absorption from air would be very slow the plants evolved complex CO2 fixation mechanisms relying on chemical absorption that enables concentration of CO2 in the stroma present in living cells at atmospheric pressure and very low CO2 concentration in air. This natural process is driven by solar energy with relatively low efficiency (3–6%). In addition, it requires a lot of land per unit mass of fixed CO2. Industry today requires relatively intensive CO2 capture processes that with minimal energy penalty and without excessive land requirement would be able to absorb CO2 from a range of industrial gases, including flue gases from the energy sector. Conventionally, industrial CO2 capture by gas–liquid absorption has been

Introduction to Carbon Dioxide Capture by Gas–Liquid Absorption …

3

achieved by means of scrubbing/stripping columns using single solvents (physical or chemical) or blended solvents but these processes consume energy and de-rate power plants. Therefore, more energy efficient advanced solvent based capture processes are urgently required [4, 5]. This chapter is structured as follows. Sect. 2 discusses CO2 capture by gas-liquid absorption applied in nature while Sect. 3 in industry. Sect. 4 analyses requirements for process innovations to meet present-day challenges in the energy sector and beyond. Finally, Sect. 5 provides conclusions and outlook for the future.

2 CO2 Capture by Gas–Liquid Absorption in Nature CO2 capture achieved by gas–liquid absorption is a very old natural process. In nature it has been present at least since the creation of ocean and CO2 comprising atmosphere. Since then CO2 has been constantly captured from air by oceanic water. Dissolved CO2 is shifted to carbonate ions, mainly bicarbonate ions (HCO3 ). The absorbed CO2 mainly accumulates in the surface layer because it very slowly propagates through the oceanic water due to limited convective motions in deep ocean. CO2 utilisation by living photosynthesising organisms is to some extent possible at the ocean surface where sunlight is available but mineralisation processes (e.g. chitin and calcium carbonate formation) are not very much affected due to shallow penetration of dissolved CO2. Currently, together with the rise of CO2 content in the atmosphere above the pre-industrial average the rate of CO2 absorption by the ocean increased. One unwanted consequence is surface ocean acidification [6] that may eventually affect the stability of fragile marine ecosystems. The ocean CO2 capture process plays a significant role in removing CO2 from the atmosphere. It was estimated that oceans/hydrosphere captured 8.4 Pg CO2/yr in 2009, i.e. 29% of anthropogenic CO2 emissions [1, 2]. It also contributes to rock formation by precipitating minerals in ocean which naturally store CO2. In addition, CO2 is the main carbon source for marine organisms. Another very old natural process involving CO2 capture by gas–liquid absorption is the first step of photosynthesis. Terrestrial plants absorb CO2 from air through permeable cell membranes to stroma (aqueous fluid) of leaf cells. After permeating the cell wall CO2 dissolves physically in the stroma. Further, dissolved CO2 reacts with H+ ions and electrons. CO2 also reacts directly with abundant water via a reaction catalysed by an carbonic anhydrase enzyme [7, 8]. These reactions concentrate CO2 within the cells in the reacted forms. Further, the products of these reactions are shifted to organic compounds again by enzymatic catalysis completely removing them from the stroma. More specifically, plants using the C3 carbon fixation mechanism shift CO2-derived products by combining them with the enzyme ribulose 1,5-bisphosphate (RuBisCo) yielding two molecules of glycerate 3-phosphate. CO2-derived carbon is then incorporated into the plant’s Calvin cycle. Plants using the C4 carbon fixation approach employ an additional CO2

4

W.M. Budzianowski

concentration mechanism relying on reacting CO2 and its derivatives first with phosphoenolpyruvate (PEP) in a reaction catalysed by an enzyme PEP carboxylase yielding oxaloacetic acid. Oxaloacetic acid is subsequently translocated within the cell and decarboxylated releasing CO2 products which then immediately react with RuBisCo and as previously are incorporated into the plant’s Calvin cycle. Interestingly, although C4 plants represent only approximately 5% of global plant biomass, their contribution to CO2 uptake by all terrestrial plants is about 30% [9] meaning that C4 fixation is globally more effective in capturing CO2. C4 plants employed e.g. as energy crops could therefore enhance biological sequestration of atmospheric CO2 and simultaneously deliver bioenergy. The current rate of biomass production (about 130 TW [10]) is several times greater than global energy consumption (15.1 TW in 2012 [11]) and could be further increased by using C4 plant based energy cropping rotations. CO2 capture for use in photosynthesis is also important for removing CO2 from the atmosphere. It was estimated that soil and biosphere captured 10.3 Pg CO2/yr in 2009, i.e. 35% of anthropogenic CO2 emissions [1, 2]. Biological CO2 capture is a main source of carbon for terrestrial plants thus contributing to biomass production and biological solar energy harvesting. Two step-change developments include the evolution of (a) very efficient carbonic anhydrase (CA) enzyme that catalyses CO2 dissolution in stroma and (b) mechanisms for biochemical CO2 concentration in the stroma of living cells by C4 plants. CA has already been applied in carbon capture demonstration projects for CO2 removal from industrial flue gases [12]. Several other concepts of nature-inspired carbon fixation are explored for industrial application but most are fairly complex and are at an early stage of development. One example is biomimetic solar energy harvesting which requires some further fundamental developments before achieving commercialisation potential [13]. Biomimetic materials that attempt to implement biological mechanisms into CO2 capture and utilisation also attract attention. The benefits of biomimetic materials are associated with highly ordered architectures, lightweight, and capability to combine strength and toughness. They can also be chemically and thermally inert, non-toxic, multifunctional, capable to adapt biological self-repair and regeneration mechanisms, energy efficient, low cost, and characterised by rapid mass transfer [14].

3 CO2 Capture by Gas–Liquid Absorption in Industry Gas-liquid absorption is the most mature CO2 capture process in industry due to its high efficiency and lower cost [15]. The first application of CO2 capture in industry was CO2 removal from natural gas employed in the 1920s. The aim was to improve the quality of natural gas by raising its calorific value. In addition, by separating ballast CO2 less energy was required for gas compression and transportation via gas grids. The separated CO2, possibly after some minor treatment, was released to the atmosphere as it had no commercial value. The process incorporated gas–liquid absorption using early physical solvents. The first known CO2 capture process


5

patent was filed in 1927, and it was essentially a series of absorption towers [16]. In the 1930s physical solvents were used to separate CO2 and deliver it for food and chemicals productions from gases comprising up to 25% CO2. In addition, CO2 was separated from ambient air in cryogenic air separation plants producing N2, O2 and Ar in order to prevent the equipment fouling by dry ice formation [17]. In the early 1940s, physical solvents were used for CO2 separation from CO2 rich streams (up to 80%) at higher pressures (up to 10 MPa). In the 1950s CO2 capture was used to separate CO2 in H2 production in refineries. Most of the CO2 captured during these times was released to the atmosphere. First chemical solvents were also used during this time. In the early 1970s, CO2 captured from natural gas processing was for the first time utilised in Enhanced Oil Recovery (EOR). The first CO2 utilisation facility was erected in Texas (USA) where CO2 was transmitted to an oil field and injected thus boosting oil recovery. EOR has since proven very successful and added value to captured CO2. Following the success of EOR, Enhanced Gas Recovery (EGR) was developed a few years later. Millions of tonnes of CO2 have been sequestered by EOR and EGR since its creation. CO2 is now sourced not only from natural gas but occasionally also from various industrial processes. From the very beginning EOR/EGR included CO2 underground sequestration but it was not linked with CO2 emissions mitigation objectives during these early applications. However, the idea of capturing CO2 to prevent it from being released into the atmosphere to mitigate climate change was suggested a few years later, in 1977 [18]. In natural gas processing plants CO2 needs to be separated in order to raise the energy content of the natural gas and minimise its transportation costs. The separated CO2 often needs to be cleaned in order to obtain a marketable product. Natural gas processing facilities have to therefore capture and purify the CO2 before they produce natural gas and CO2 as useable commodities. The CO2 can be used in EOR/EGR projects achieving a value in monetary terms. In EOR/EGR the CO2 is partly extracted along with the oil/natural gas, separated and reused thus beneficially creating a CO2 loop [19]. Other applications include the separation of CO2 from hydrogen for the use in various hydrogenation processes and in producing beverage-grade CO2. Industrial technological developments relied on using more efficient solvents and their blends such as physical and later chemical solvents. Some process developments led to reduced energy penalty using physical solvents such as Selexol, Rectisol, Purisol, Jeffsol, and Morphysorb. The physical solvent processes are effective at high pressure where for example Purisol has much greater CO2 loading capacity than MEA. They all employ chemicals that are non-reactive towards CO2 but with high physical solubility. One process was commercialised that employs a combination of physical and chemical solvents (Sulfinol by Shell). After that reactive processes were developed and commercialised with particular focus on low pressure gases with lower concentrations of CO2. Among them several MEA and MDEA based processes were deployed (e.g. Fluor’s Econamine, Lummus MEA). A KM-CDR process employed hindered secondary amine KS-1. Besides, hot potassium carbonate processes were developed and used in industrial practice (e.g.

6

W.M. Budzianowski

Benfield, Catacarb, Flexsorb, Vetrocoke). In addition ammonia based processes were proposed on the market (Alstom’s Chilled Ammonia, ECO2). These processes employed ammonia solvent and due to ammonia volatility [20] absorption was realised at low temperatures (typically below 10 °C). In further developments the gas–liquid absorption technology of the single amines were gradually shifted to solvent blends [21, 22] such as MEA/MDEA and others (Aker Clean Carbon, Alstom Advanced Amine, Cansolv, HTC PureEnergy, Advanced PCCC). Additionally various reaction promoters such as piperazine were added to amine solvents [23]. Subsequently more and more sophisticated solvents were researched and demonstrated. For example Siemens proposed the POSTCAP process involving an amino-acid salt solvent. The process benefited from low environmental impact while technical parameters were comparable with Best Available Technologies (BATs) at that time. Further, a few companies proposed catalysed CO2 capture by carbonic anhydrase however the process has little commercial success due quick denaturation of CA under realistic industrial conditions. Other advanced solvent based capture processes (ASBCPs) included aminosilicones, chilled ammonia or potassium carbonate with phase change. More sophisticated solvent systems using new approaches such as phase change solvents have recently been researched or piloted [24] but as of 2015 were not commercially available [25].

4 Perspectives for the Energy Sector and Beyond The essential challenge that all commercial CO2 capture gas–liquid absorption processes have to overcome is insufficient energy efficiency. State-of-the-art CO2 capture processes are capable of separating CO2 from flue gases with a total equivalent work requirement of about 1 MJ/kgCO2 meaning that thermodynamic efficiency of separating CO2 is about 16% (see Chap. 4 of this book for further explanations on this topic). Therefore, the main challenge of CO2 separation is how to reduce these energy requirements and raise efficiency. Decarbonisation of the energy sector may be alternatively achieved by a shift from fossil fuel to renewable energy sources. Harvesting renewables to meet all global energy demands may however be challenging mainly due to high production cost, limited generation potential and high energy demands. In addition, renewable industries require very expensive infrastructures that are CO2 intensive by themselves (in the life cycle context). The increasing penetration of renewables will likely rely on harvesting lower quality sources than at the beginning which will have higher associated life cycle CO2 emissions. This might be a problem for some renewable industries that is today often overlooked. Regarding the potential of renewable energy, according to International Renewable Energy Agency (IRENA) with policies in place and under consideration in 2014, the global penetration of modern renewable energy (excluding traditional biomass) will reach 14% of total final energy consumption (TFEC) by 2030 [26]. If an extended policy package


7

proposed in IRENA’s REmap 2030 [26] is adopted and successfully implemented, the penetration of modern renewables (excluding traditional biomass) in the global energy mix may theoretically reach about 27% by 2030. This will be insufficient to decarbonise the whole energy sector. The further expansion of renewable energy industries may put pressure on growing life cycle CO2 emissions thus truly effective decarbonisation will be challenging. It also emphasises that decarbonisation of the energy sector cannot increase emissions in cooperating industries and decarbonisation beyond the energy sector is equally important. In recent long term projections it was suggested that under the scenario of limiting CO2 concentration in the atmosphere to 450 ppm by 2100 the share of fossil fuels in total primary energy supply (TPES) would decrease to about 44% while the share of renewables would increase to about 50% in 2100 [19]. However, to maintain this high share of fossil fuels coal based power plants might need up to 55% carbon capture utilisation and storage (CCUS) while natural gas fired power plants up to 25% CCUS in 2100. CO2 separations would need to be implemented throughout various niche industries that use fossil fuels or biomass and significant potential lies in highly concentrated CO2 streams for which CO2 capture is feasible at relatively low costs.

5 Technological Developments in Gas–Liquid Absorption The needs identified in the energy sector and other modern industries imply that there will be high potential for energy efficient CO2 capture technologies including energy efficient options relying on gas–liquid absorption. In this context advanced solvent based capture processes (ASBCPs) will need to be developed and employed. Energy efficient ASBCPs will require simultaneous developments in materials science, nanotechnology, catalysis, process engineering, systems research and environmental science. Figure 1 presents the multiple scales that a particular ASBCP for CO2 separation may entail. Most process innovations start with material development that needs to have optimal properties for mass transfer and stability

Fig. 1 Steps across advanced solvent based capture process design

8

W.M. Budzianowski

over multiple scrubbing/stripping cycles. The material may be a phase changing solvent, have catalytic properties or benefit from microencapsulation or polarity swing. The specific physico-chemical properties of materials may increase CO2 loading capacity by simultaneously binding CO2 chemically and dissolving physically (e.g. ionic liquids). Further, mass and heat transfers enable linkage of these favourable basic properties of the developed material with the requirement of the separation process. In general, CO2 transfer needs to be sufficiently fast while thermal effects should be limited to move the entire ASBCP closer to the reversible process. Finally, the ASBCP needs to integrate with the energy system to make use of synergies that are feasible due to the fact that energy is conserved. The ASBCP also needs to integrate with the environment to minimise adverse environmental impacts of the technology, in particular in the whole life cycle. All of these innovation steps need to be achieved simultaneously in order to fully account for interactions between all parts of the low-carbon energy system. Historically, developments in processes for CO2 capture by gas–liquid absorption have been driven by some specific natural and industrial drivers. Developments in nature were primarily driven by the need of sourcing CO2 from diluted air. Therefore, plants evolved catalysed absorption (using CA) allowing for process intensification by several orders making it effective despite the very low separation driving force. The second driver for innovation was associated with the need to colonise tropical arid regions by plants. In this way C4 plants capable of enriching CO2 in stroma by enzymatically catalysing CO2 reaction with water (first employing CA followed by PEP carboxylase) were much more effective in CO2 fixation. Due to temporal carbon fixation C4 plants were more active in capturing CO2 during colder nights and CO2 conversion during hot days. C4 plants achieved improved water efficiency. For example, at 30 °C C3 grasses lost approximately 833 molecules of H2O per one fixed molecule of CO2, whereas C4 grasses lost only 277 [27]. Due to the combination of intensified CO2 fixation and decreased water losts C4 plants conserved soil moisture and easily colonised tropical arid environments being more efficient than C3 plants. However, C3 plants colonised colder regions without water deficiency because the shortcoming of C4 plants relied on the need to fix every CO2 molecule twice (first by PEP and second by RuBisCO) implying that they used more energy than C3 plants (30 vs. 18 molecules of ATP per one molecule of glucose). Therefore, C3 plants were efficient in regions with lower solar illuminance while C4 plants were superior under high solar radiation intensity and associated water shortages. Historical industrial developments in CO2 capture by gas–liquid absorption were driven by techno-economic factors associated with the need to improve quality and hence value of natural gas. Later, along with EOR/EGR implementation, CO2 capture allowed an increase in natural gas production meaning that economic drivers appeared. Developments led to physical solvents being replaced with chemical solvents and solvent blends especially when low CO2 partial pressure were present. Current drivers in the energy sector are similar to those seen in industry in the past with the difference being that the goal of CO2 capture relates mainly to mitigating CO2 emissions and protecting climate. Capturing CO2 does not directly


9

contribute to the quality of energy products such as electricity from power plants or liquid fuels from refineries. Therefore, external economic drivers implemented by energy policy instruments are essential to make CO2 capture from industrial flue gases an attractive business line. CO2 capture by gas-liquid absorption is also applicable in the renewable energy sector, e.g. for biomethane production by CO2 separation from biogas [28–30]. It also has potential application in relation to various industrial CO2 rich sources, e.g. in the cement industry. Since renewable energy harvesting usually requires large backing industries (to provide construction materials and feedstocks) CO2 capture in industry will be very important in the near future in order to minimise life cycle CO2 footprint of renewables [31]. Drivers in these industries may be to some extent economical without external subsidies especially if CO2 can be valorised. But in industries cooperating with the renewable energy sector additional drivers must be implemented by policymaking.

6 Conclusions and Outlook CO2 capture by gas–liquid absorption has been used in nature for a very long time and in industry for almost a century. In nature CO2 is absorbed by oceans and by living plants. In 2009 the scale of these two natural absorption processes was as high as 788 PgCO2/year while the contribution to mitigating anthropogenic CO2 emissions was also high amounting to 64% (18.7 PgCO2/yr). Natural process developments for CO2 capture include catalysed absorption using CA in C3 plants and combined CA/PEP carboxylase in C4 plants. The C3 mechanism is more energy efficient in CO2 capture because it is a one step process while in the C4 mechanism the second step consumes additional energy. Industrial developments started from physical solvents and were followed by chemical solvents and solvent blends. Advanced solvent based capture processes are required to improve energy efficiency of the CO2 capture process and give promise for the implementation of CO2 capture to reduce emissions in the energy sector and beyond. Drivers for implementation of CO2 capture in industry are mainly economical. In the energy sector CO2 mitigation strategies need to carefully consider the whole life cycle and a balance between renewable, biomass and fossil fuel sources need to be found. Since the expanding renewable energy sector has large associated industries, it is important that CO2 emissions in these industries are minimised, potentially by involving CO2 capture. Technological developments are essential in achieving true decarbonisation throughout all economic sectors and therefore all chapters of this book provide insights into energy efficient processes for CO2 capture by gas–liquid absorption. Acknowledgments This study has been supported by the members of the Renewable Energy and Sustainable Development (RESD) Group (Poland) under the project RESD-RDG03/2016 which is gratefully acknowledged.

10

W.M. Budzianowski

References 1. GCP (Global Carbon Project) (2011) Carbon budget and trends 2010. www. globalcarbonproject.org/carbonbudget 2. Budzianowski WM (2013) Modelling of CO2 content in the atmosphere until 2300: Influence of energy intensity of gross domestic product and carbon intensity of energy. Int J Glob Warming 5(1):1–17 3. IEA (International Energy Agency) (2011) Key World Energy Statistics, Paris 4. Budzianowski WM (2016) Explorative analysis of advanced solvent processes for energy efficient carbon dioxide capture by gas-liquid absorption. Int J Greenhouse Gas Control 49:108-120. http://dx.doi.org/10.1016/j.ijggc.2016.02.028 5. Budzianowski WM (2015) Single solvents, solvent blends, and advanced solvent systems in CO2 capture by absorption: a review. Int J Glob Warming 7(2):184–225 6. Shanableh A, Merabtene T, Omar M, Imteaz M (2011) Impact of surface ocean acidification on the CO2 absorption rate. Int J Glob Warming 3(1–2):163–172 7. Li L, Fu ML, Zhao YH, Zhu YT (2012) Characterization of carbonic anhydrase II from Chlorella vulgaris in bio-CO2 capture. Environ Sci Pollut Res 19(9):4227–4232 8. Pendersen-van Elk NJMC, Derks PWJ, Fradette S, Veersteg GF (2012) Kinetics of absorption of carbon dioxide in aqueous MDEA solutions with carbonic anhydrase at 298 K. Int J Greenhouse Gas Control 9:385–392 9. Osborne CP, Beerling DJ (2006) Nature’s green revolution: The remarkable evolutionary rise of C4 plants. Philos Trans of the Royal Society B: Biol Sci 361(1465):173–194 10. Steger U, Achterberg W, Blok K, Bode H, Frenz W, Gather C, Hanekamp G, Imboden D, Jahnke M, Kost M, Kurz R, Nutzinger HG, Ziesemer T (2005) Sustainable development and innovation in the energy sector. Springer, Berlin 11. IEA (International Energy Agency) (2014) Key World Energy Statistics, Paris 12. Akermin Inc. (2013) Final Scientific/Technical Report−Advanced Low Energy Enzyme− Catalyzed Solvent for CO2 Capture. http://www.netl.doe.gov/File%20Library/Research/Coal/ carbon%20capture/post-combustion/FE0004228-Final-Report-01-07-2014.pdf 13. Boghossian AA, Ham MH, Choi JH, Strano MS (2011) Biomimetic strategies for solar energy conversion: a technical perspective. Energy Environ Sci 4(10):3834–3843 14. Kumar P, Kim K-H (2016) Recent progress and innovation in carbon capture and storage using bioinspired materials. Appl Energy 172:383–397 15. Leung DYC, Caramanna G, Maroto-Valer MM (2014) An overview of current status of carbon dioxide capture and storage technologies. Renew Sustain Energy Rev 39:426–443 16. Gaus W, Hochschwender K, Schunck W (1927) Process for extracting carbon dioxide from gaseous mixtures and forming alkaline carbonates. U.S. Patent 1897725 17. Greenwood K, Pearce M (1953) The removal of carbon dioxide from atmospheric air by scrubbing with caustic soda in packed towers. Trans Inst Chem Eng 31:201–207 18. IEA GHG. A brief history of CCS and current status. http://www.ieaghg.org/docs/General_ Docs/Publications/Information_Sheets_for_CCS_2.pdf. 2015-12 19. Budzianowski WM (2017) Implementing carbon capture, utilisation and storage in the circular economy. Int J Glob Warming 20. Budzianowski WM (2011) CO2 reactive absorption from flue gases into aqueous ammonia solutions: the NH3 slippage effect. Environ Prot Eng 37(4):5–19 21. Yu YS, Wang GX, Lu HF, Zhang ZX, Rudolph V (2013) Characterizing the transport properties of multi-amine solutions for CO2 capture by molecular dynamics simulation. J Chem Eng Data 58(6):1429–1439 22. Dash SK, Samanta AN, Bandyopadhyay SS (2014) Simulation and parametric study of post combustion CO2 capture process using (AMP + PZ) blended solvent. Int J Greenhouse Gas Control 21:130–139


11

23. Li H, Le Moullec Y, Lu J, Chen J, Valle Marcos JC, Chen G (2014) Solubility and energy analysis for CO2 absorption in piperazine derivatives and their mixtures. Int J Greenhouse Gas Control 31:25–32 24. Smith KH, Lee A, Mumford K, Li S, Indrawan I, Thanumurthy N, Temple N, Anderson C, Hooper B, Kentish S, Stevens G (2015) Pilot plant results for a precipitating potassium carbonate solvent absorption process promoted with glycine for enhanced CO2 capture. Fuel Process Technol 135:60–65 25. Cowan RM, Jensen MD, Pei P, Steadman EN, Harju JA (2011) Current status of CO2 capture technology development and application. National Energy Technology Laboratory U.S. Department of Energy 26. IRENA (International Renewable Energy Agency) (2014) REmap 2030: a renewable energy roadmap. IRENA, Abu Dhabi 27. Sage R, Russell M (1999) C4 Plant Biology 28. Budzianowski WM (2012) Benefits of biogas upgrading to biomethane by high-pressure reactive solvent scrubbing. Biofuels Bioprod Biorefin 6(1):12–20 29. Budzianowski WM (2012) Negative carbon intensity of renewable energy technologies involving biomass or carbon dioxide as inputs. Renew Sustain Energy Rev 16(9):6507–6521 30. Budzianowski WM (2012) Value-added carbon management technologies for low CO2 intensive carbon-based energy vectors. Energy 41(1):280–297 31. Budzianowski WM, Postawa K (2017) Renewable energy from biogas with reduced carbon dioxide footprint: implications of applying different plant configurations and operating pressures. Renew Sustain Energy Rev. Doi:10.1016/j.rser.2016.05.076

Assessment of Thermodynamic Efficiency of Carbon Dioxide Separation in Capture Plants by Using Gas–Liquid Absorption Wojciech M. Budzianowski

Abstract Typical carbon capture plants include CO2 separation and compression steps. CO2 separation from diluted flue gases may be achieved by using gas-liquid absorption. This process requires work input for e.g. separating CO2 from flue gases and regenerating the CO2 loaded solvent. Hence, CO2 capture plants involving gas-liquid absorption consume a remarkable part of power and thermal energy generated by power plants. By increasing the thermodynamic efficiency of capture plants one can increase the produced power and save fossil fuels. Therefore, this study provides a quantitative assessment of the thermodynamic efficiency of CO2 separation in capture plants. To this aim the minimum work required for CO2 separation and actual work input in realistic carbon capture plants are estimated. The results reveal that for the state-of-the-art MEA solvent the thermodynamic efficiency of the capture plant is about 16%, for state-of-the-art advanced solvent based capture process (ASBCP) is about 25%, while given the progress in developing ASBCPs in near future it may reach about 30%. Additional measures to reduce the energy requirement of the capture plant such as heat pumps are also discussed. This all means that CO2 separation by gas-liquid absorption is still a relatively inefficient process and remarkable potential for further improvements with step change innovations in gas-liquid absorption exist and may be beneficially used for optimising CO2 capture plants. Nomenclature ASBCP E G MEA n

Advanced solvent based capture process Energy, J Gibbs free energy, J Monoethanolamine Molar flow rate, kmol/s

W.M. Budzianowski Consulting Services, Wrocław, Poland W.M. Budzianowski (&) Renewable Energy and Sustainable Development (RESD) Group, Wrocław, Poland e-mail: [email protected] © Springer International Publishing AG 2017 W.M. Budzianowski (ed.), Energy Efficient Solvents for CO2 Capture by Gas–Liquid Absorption, Green Energy and Technology, DOI 10.1007/978-3-319-47262-1_2

13

14

NGCC pi p PC PCC Q R T TREF TSOURCE W Wact Wmin 2 yCO i iCO2 yi η ηturbine

W.M. Budzianowski

Natural gas combined cycle Partial pressure of the i-th gas, Pa Total pressure, Pa Pulverised coal Postcombustion capture Heat, J Ideal gas constant = 8.314 J/(mol K) Absolute temperature, K Reference temperature = 293.15 K Source temperature = 393.15 K Work, J Actual work, J Minimum work of separation, J Mole fraction of CO2 in the gas mixture i, – Mole fraction of non-CO2 remainder, − Thermodynamic efficiency, – Turbine efficiency = 90%

Subscripts and superscripts A B C i sep TOTAL

Stream A Stream B Stream C Index Separation Total

1 Introduction Energy efficient decarbonisation of the energy sector and major industries is essential for achieving sustainable development. Recent research efforts in CO2 capture led to the development of many new processes that may facilitate reduction of CO2 emissions. However, technological readiness levels (TRLs) of these processes are different and they use different necessary energetic inputs (e.g. waste heat, extracted stream, compressor work). It is thus difficult to compare the performance of such diverse processes [1]. Mature CO2 capture technologies may have slightly improved energy efficiencies compared to emerging technologies but at the same time they may offer much less potential for further improvements. Due to technology learning the future energy efficiency performance of various technological options currently at different stages of development is often unclear [2].

Assessment of Thermodynamic Efficiency of Carbon Dioxide …

15

CO2 capture by gas-liquid absorption undergoes rapid development and several new processes aimed at reducing its energy requirements have recently been proposed. Examples include solvent blends and advanced solvent based capture processes applying facilitated CO2 capture by employing phase changing solvents, catalysed absorption or microencapsulated solvents [3, 4]. These new techniques should be explored in order to assess their potential for getting closer to the thermodynamic limits of CO2 capture. Two laws of thermodynamics are essential tools for assessing the energy efficiency of CO2 capture plants. The first law reflects the principle of energy conservation and the second law accounts for process irreversibility. The second law efficiency shows how far a given capture process is from its thermodynamic limit. The methodology employed in this study uses the minimum work required for CO2 separation under idealised operating conditions which is calculated from the combined first and second law of thermodynamics. Further, actual work of CO2 separation is estimated through literature surveys and confirmed by calculations. The provided assessment of thermodynamic efficiencies offer some qualitative insights into the energy efficiency of CO2 capture by gas-liquid absorption and may quantify potentials for future improvements. The study begins with calculation of the minimum work of CO2 separation (Sect. 2) which is followed by the determination of actual work requirements from actual capture plant evidence and by calculation (Sect. 3). The thermodynamic efficiency of CO2 separation in capture plants involving gas-liquid absorption is calculated in Sect. 4. Section 5 discusses additional related measures capable of reducing the work requirements of CO2 separation. Finally, Sect. 6 summarises conclusions drawn from the study.

2 Minimum Thermodynamic Work of CO2 Separation The minimum work required for CO2 separation under idealised operating conditions is calculated from the combined first and second law of thermodynamics. It uses the flow rates and compositions of inlet and outlet streams and operating temperature. Figure 1 shows a CO2 capture plant along with a CO2 emitting plant and corresponding gas streams. Stream A represents a flue gas comprising CO2 while stream B is a CO2 rich stream, and stream C is the remainder of flue gases. The capture plant operates with a certain fixed capture rate and the purity level of CO2 rich stream is imposed.

Fig. 1 Schematic of the CO2 capture plant involving gas-liquid absorption

16

W.M. Budzianowski

The minimum work required for separating CO2 from a flue gas mixture for an isothermal and isobaric process equals to the negative of the difference in Gibbs free energy of the separated final states (streams B and C in Fig. 1) from the mixed initial state (stream A in Fig. 1). This is the negative of Gibbs free energy of mixing. For an ideal gas the Gibbs free energy change between stream A to streams B and C is: Wmin ¼ DGsep ¼ DGB þ DGC DGA

ð1Þ

For an ideal mixture, the partial molar Gibbs free energy for each gas is [5, 6]: @G pi 0 ¼ Gi þ RTln @ni p

ð2Þ

Therefore, the total Gibbs free energy of an ideal gas mixture can be expressed as: GTOTAL ¼

X i

ni

@G @ni

ð3Þ

The minimum work required to shift from state A to states B and C is associated with the free energy difference between the product and reactant states, which can be calculated by inserting Eqs. (2) to (3) resulting in: 2 CO2 ACO2 0 0 2 2 2 þ nACO GA ¼ nCO GACO2 þ RT nCO ln yA ln yACO A GCO2 þ nA A A A ð4AÞ 2 CO2 BCO2 0 0 2 2 2 þ nBCO GB ¼ nCO GBCO2 þ RT nCO ln yB ln yBCO B GCO2 þ nB B B B ð4BÞ 2 CO2 CCO2 0 0 2 2 2 GC ¼ nCO GCCO2 þ RT nCO ln yCCO þ nCCO C GCO2 þ nC C ln yC C C ð4CÞ This ideal mixing takes place at constant temperature and pressure. By substituting Eqs. (4A–C) to Eq. (1) the minimum work of separation is obtained: 3 2 CO2 2 2 nB yCO ln yB þ 1 yCO ln 1 yCO B B B 6 CO2 CO2 7 CO2 CO2 7 ¼ RT 6 5 4 þ nC yC ln yC þ 1 yC ln 1 yC CO2 CO2 CO2 CO2 þ 1 yA ln 1 yA nA yA ln yA 2

Wmin

ð5Þ

Equation (5) is subsequently used to calculate the minimum work of separation. Figure 2 presents the minimum work Wmin for CO2 separation as a function of the molar concentration of CO2 in the gas mixture (stream A in Fig. 1). Varied


17

Fig. 2 Impact of temperature, CO2 capture rate and CO2 stream purity on the minimum thermodynamic work of CO2 separation for gases with different CO2 contents. Parameters varied: T—(300–350 K), capture rate—(50–90%), CO2 stream purity—(90–98%)

parameters include T (300–350 K), capture rate (50–90%), and CO2 stream purity (90–98%). As it can be observed the minimum work required for CO2 separation increases with decreasing CO2 concentration. In contrast, Wmin increases with increasing temperature, capture rate and CO2 stream purity. More specifically, the minimum work per kg of CO2 captured in fixed temperature of 318 K, capture rate of 90% and CO2 stream purity of 98% at inlet gas comprising 10% CO2 is 174 kJ/kgCO2. When CO2 is separated from air (inlet gas contains 0.04% CO2) the minimum work is 509 kJ/kgCO2, which is three times greater. When CO2 is separated from biogas [7–9] combustion flue gases (inlet gas has 50% CO2) Wmin is 54 kJ/kgCO2, which is three times smaller. Effect of temperature is less pronounced: at 300 K Wmin is 164 kJ/kgCO2 (6% less) and at 350 K it is 192 kJ/kgCO2 (10% more). When capture rate is decreased to 50% Wmin is 149 kJ/kgCO2 (14% less). When CO2 purity is set at 90% Wmin is 159 kJ/kgCO2 (11% less).

3 Actual Work Requirement of CO2 Separation in Realistic Carbon Capture Plants The minimum work of separation reflects a theoretical idealised isothermal and isobaric CO2 capture process that cannot be achieved in the practice. In an actual CO2 capture plant employing gas-liquid absorption significantly more work will be

18

W.M. Budzianowski

required. For example, pumps are required to deliver the solvent to the scrubber and stripper which takes work. Similarly electrical blowers drive a flue gas through the scrubber. The solvent regeneration requires thermal energy to obtain the CO2 rich and regenerated solvent streams. Each of these processes have associated efficiencies based upon irreversibilities, such as friction, heat transfer, gas expansion, gas mixing, etc. Therefore, actual work requirement depends on the unit operations of the capture process and their individual efficiencies. Consequently, in addition to the minimum work which is insufficient to operate a CO2 capture plant some extra work is needed. This lost work or lost exergy reflects inefficiencies of a given process realisation and cannot be avoided in the practice. The actual work of CO2 separation can be estimated by using evidence from realistic carbon capture plants or by calculation of a model carbon capture plant. These two methods are employed below.

3.1

Evidence from Realistic Carbon Capture Plants

The actual work requirement of CO2 capture plants involving gas-liquid absorption operated under realistic conditions can be found in literature. Table 1 summarises such evidence taken from literature relevant to state-of-the-art CO2 capture solvents (especially MEA). It gives a breakdown of work requirement involving contributions from the flue gas blower, rich/lean solvent pumps, reboiler, and miscellaneous minor components. As seen from Table 1 work requirement of plant components varies across various CO2 capture plants reported in literature. It is associated with differences in

Table 1 Literature overview of breakdowns of actual work requirements of realistic carbon capture plants involving state-of-the-art gas-liquid absorption Capture plant item

Work requirement [kJ/kgCO2]

References

Flue gas blower

50 70 37 2 3.2 1.2 716–1110 (2867–4443) 497 (3956) 491 (4063) 964 (2892) 25

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

Rich/lean solvent pumps

Work equivalent of reboiler duty (actual thermal energy)

Miscellaneous (baseload—control valves, minor units)

[10]


19

Table 2 Representative actual work requirements of CO2 separation in capture plants involving state-of-the-art gas-liquid absorption Capture plant item


Flue gas blower Rich/lean solvent pumps Work equivalent of reboiler duty Miscellaneous (baseload—control valves, minor units) Total

70 10 900 25 1005

plant design, flue gas CO2 content, flue gas contaminants, capture rate, plant capacity, employed solvent etc. Table 2 summarises representative actual work requirements of CO2 separation by state-of-the-art gas-liquid absorption.

3.2

Calculations for a Model Carbon Capture Plant

Figure 3 displays a schematic of a model CO2 capture plant that is used to calculate work requirements and thus to improve the consistency of literature evidence provided in Sect. 3.1. It illustrates that a model CO2 capture plant consists of two major components: (i) CO2 separation and (ii) CO2 compression. In order to

Fig. 3 Schematic of a model CO2 capture plant involving state-of-the-art gas-liquid absorption used in the calculation of actual work requirements of CO2 separation

20

W.M. Budzianowski

calculate work required for separating CO2 only the former component needs to be accounted for. The methodology used in current calculations includes blowing work, pumping work, reboiler equivalent work and miscellaneous contributions. It has recently been applied for evaluating power requirements of CO2 separation from biogas by using gas-liquid absorption involving pressurised water scrubbing [16]. Here, heat released in the condenser is neglected and integration measures are not accounted for, except the cross heat exchanger.

3.2.1

Blowing Work

The work requirement for blowing flue gas is calculated from the gas flow rate (qG) and total pressure rise (Dp). The shaft work requirement (WB) is obtained by using the mechanical efficiency of the blower (ηB). WB ¼ qG Dp=gB

ð6Þ

The mechanical efficiency values of 0.78 and 0.75 are used for the blower and pump, respectively [17].

3.2.2

Pumping Work

The work requirements for pumping the rich and lean solvents as well as cooling water are calculated from the water density (qL), gravitational acceleration (g) and liquid flow rate (qL). In order to obtain shaft power (WP), the mechanical efficiency of a pump (ηP) is also accounted for. Wp ¼ qL gqL HT =gp

ð7Þ

The total pumping work includes contributions from pumping the rich and lean solvents as well as cooling water (WP-RS, WP-LS, WP-COOL). The total pressure head (HT) is needed to calculate work requirement for pumping and blowing. It is calculated as the sum of the pressure difference (HSCR − HATM), the static (HS) and dynamic (HD) heads as follows: HT ¼ ðHSCR HATM Þ þ HS þ HD

ð8Þ

The pressure difference HSCR − HATM is obtained by subtracting the pressure in the scrubber and the atmospheric pressure. The static head is taken equal to the height of columns (scrubber/stripper). The dynamic head is calculated from the Darcy-Weisbach equation assuming optimum gas velocity 20 m/s and liquid velocity 2.5 m/s:


HD ¼ f

21

L w2 D 2g

ð9Þ

Friction factor f is obtained explicitly from a relationship approximating the Colebrook-White equation [18]. f ¼

6:4

pffiffiffi2:4 lnðReÞ ln 1 þ 0:01Re De 1 þ 10 De

ð10Þ

CO2 scrubbing is more efficient at lower temperatures characterised by increased CO2 solubility. Therefore, lean solvent is cooled in a heat exchanger using water as a coolant. Cooling is designed to reduce lean solvent temperature to 5 K above ambient temperature. The flow rate of cooling water (qCOOL) required to cool the lean solvent is calculated from energy balance and subsequently inserted into Eq. (7) to calculate pumping work associated with lean solvent cooling. qCOOL ¼

3.2.3

qC qC cPC ðTCout TCin Þ qL cPL ðTLout TLin Þ

ð11Þ

Reboiler Equivalent Work

Qsens is estimated as heat that needs to be added to heat the solvent from temperature TINLET to TSTRIPPER: Qsens ¼ mS Cp ðTSTRIPPER TINLET Þ

ð12Þ

Qvap ¼ mS Hvap

ð13Þ

Qact ¼ Qsens þ Qvap

ð14Þ

Thermal energy is subsequently converted to equivalent work by using a conversion factor obtained from Carnot efficiency [19, 20]. In addition, turbine efficiency is used to convert the work into electricity. Wequivalent ¼ Qact gturbine

3.2.4

TSOURCE TREF TSOURCE

ð15Þ

Miscellaneous Contributions to Total Work

Miscellaneous contributions to total work include control valves and other minor units associated with CO2 separation.

22

W.M. Budzianowski

Table 3 Calculated work requirements of the model CO2 capture plant involving CO2 separation by state-of-the-art gas-liquid absorption Capture plant item


Flue gas blower Rich/lean solvent pumps Work equivalent of reboiler duty Miscellaneous (baseload—control valves, minor units) Total

40 20 890 25 975

3.2.5

Total Work Requirement of the Model CO2 Capture Plant

The calculations takes into account CO2 capture plant technical parameters from [21]. The results are presented in Table 3. As seen from comparison of Tables 2 and 3 total work requirement and breakdown are similar. Therefore these calculated actual work requirements will be used in the next Section.

4 Thermodynamic Efficiency of CO2 Separation Based on the minimum work of separation and actual work requirements the thermodynamic efficiency may be calculated as the ratio of minimum and actual work. g¼

Wmin Wact

ð16Þ

Wmin is the useful effect of separating flue gases and it is equal to the Gibbs free energy of CO2 separation. Wact is the work input required to run the realistic CO2 capture plant. Wact includes work consumed by pumps, blowers and compressors as well as work equivalent of thermal energy consumed by the stripper. This thermodynamic efficiency uses only work or work equivalent inputs/outputs and is therefore consistent with the second law of thermodynamics. Thermodynamic efficiencies of separation processes have been investigated for a variety of pollutant reduction techniques in a wide range of concentrations. In Fig. 4 the second law efficiency is plotted as a function of pollutants concentration (at separation rate 90%, PC power plant 500 MWel). Data are adopted from various real separation processes such as CO2 capture in coal PCC plants (by amine scrubbing) and NGCC plants (by amine scrubbing), as well as for separations of NOX (by selective catalytic reduction), SOX (by wet flue gas desulfurization), and Hg (by activated carbon injection). As it can be observed the second law efficiency increases with increasing pollutant concentration [22]. In view of these results the specific nature of CO2 is such that it has very high concentration compared to other


23

Fig. 4 Influence of pollutant concentration in flue gases on the second law efficiency of a flue gas cleaning process, data adapted from [23]

typical pollutants contained in flue gases, e.g. SOX, NOX or Hg. It means that CO2 separation may require different approaches than other highly diluted pollutants. Due to higher concentrations and flow rates the thermodynamic efficiency is more pronounced for separating CO2 than other minor contaminants. The CO2 capture plant separates relatively concentrated gases and therefore the second low efficiency of CO2 separation is higher than that of diluted pollutants, typically it is higher than 10%. The minimum work of separation (for 90% CO2 capture rate, 98% CO2 stream purity, temperature 318 K, and 13% CO2 content in flue gases) amounts to 158 kJ/kgCO2. Taking into account the actual work requirement of MEA based CO2 capture of about 975 kJ/kgCO2 the second law efficiency calculated from Eq. 16 is 16.2%. The iCAP project obtained a two immiscible liquid phases based ASBCP with the experimental energy requirement of 2400 kJ/kgCO2 for solvent regeneration. It translates to total work equivalent of about 630 kJ/kgCO2 and yields the second law efficiency of 25.1%. Given this progress and several other ongoing projects the second law efficiency of about 30% (527 kJ/kgCO2) may be achieved by CO2 capture systems employing innovative ASBCPs in the future. Figure 5 compares these calculations and expectations.

Fig. 5 Potential for the reduction of work requirement of CO2 capture by ASBCPs with technology development. Parameters: CO2 capture rate—90%, CO2 stream purity—98%, temperature—318 K, CO2 content in flue gases—13%

24

W.M. Budzianowski

It is worth to mention that a typical PC power plant having a CO2 emission intensity of 1000 kgCO2/MWhel (0.000278 kgCO2/kJel) produces 3600 kJel/kgCO2 emitted. Consequently, the state-of-the art MEA CO2 capture process that consumes 975 kJ/kgCO2 would use 27% of electricity delivered by such a power plant. It emphasises that the progress in reducing work requirement of CO2 separation is essential.

5 Examples of Additional Measures Capable of Reducing Work Requirement The actual work can be decreased by optimising one or all of the separation steps. The optimisation may rely on tuning existing CO2 capture processes or applying new ones bringing a step change in technology [e.g. advanced solvent based capture processes (ASBCPs)]. As shown in breakdown given in Table 3 the most significant contribution is associated with reboiler duty. Therefore, techniques that deliver thermal energy to the reboiler may greatly improve the performance of CO2 capture plants. For example, heat pumps are able deliver thermal energy taken from the environment or waste heat sources available in power plants [24]. Namely, using an absorption-driven heat pump for waste heat recovery the energy requirement of a conventional solvent based CO2 separation may be potentially reduced to 1670 kJ/kgCO2 [25]. If a heat pump uses renewable energy taken from the environment the CO2 capture plant may overcome thermodynamic limits of closed systems. The application of heat pumps may rely on cooling the scrubber which enhances absorption and heating the stripper which enhances solvent regeneration. The scrubber is fed with flue gases having relatively high temperature which ensures that this type of heat source is stable in capture plants. Another measure is associated with improved allocation of thermal energy within capture plants. It may also be achieved by a distributed cross heat exchanger. This device functions by delivering more of the heat to the bottom section of the stripper than to the top section. The stripper with lower temperature at the top has a lower solvent vapour fraction, which reduces useful heat lost to the condenser [19, 25]. Further, process intensification techniques for dedicated for post-combustion CO2 capture [26], if could be suitably integrated with ASBCPs can add additional energy efficiency benefits to the whole capture plant.


25

6 Conclusions This study analyses thermodynamic efficiency of CO2 separation by gas-liquid absorption. The minimum work required for CO2 separation under idealised operating conditions is provided. Insights in blowing work, pumping work, reboiler equivalent work and miscellaneous contributions are given based on realistic capture plants found in literature and calculations made for the model CO2 capture plant. Taking into account actual work required to separate CO2 the second law efficiency is calculated. For the MEA solvent it is about 16%, for the state-of-the-art ASBCP about 25%, and for future ASBCPs it may achieve about 30% or more. The second law efficiency is limited to closed systems. However, when a CO2 capture plant operates as an open system the thermodynamic limits may be overcome. This can be implemented by using renewable energy, e.g. through heat pumps utilisation. Heat pumps may also facilitate the distribution of thermal energy within the system thus leading to limited irreversibilities and exergy losses and consequently to improved thermodynamic efficiency. Heat management is particularly relevant since it is clearly the major contributor to total work requirement of CO2 separation, at least when thermal regeneration of solvent is employed. Due to the scale of CO2 capture plants all feasible measures need to be applied simultaneously in order to achieve the meaningful reduction of work requirement of CO2 separation. Acknowledgments This study has been supported by the members of the Renewable Energy and Sustainable Development (RESD) Group (Poland) under the project RESD-RDG03/2016 which is gratefully acknowledged.

References 1. Calbry-Muzyka S, Edwards CF (2014) Thermodynamic benchmarking of CO2 capture systems: exergy analysis methodology for adsorption processes. Energy Procedia 63:1–17 2. Rochedo PRR, Szklo A (2013) Designing learning curves for carbon capture based on chemical absorption according to the minimum work of separation. Appl Energy 108:383–391 3. Budzianowski WM (2016) Explorative analysis of advanced solvent processes for energy efficient carbon dioxide capture by gas-liquid absorption. Int J Greenhouse Gas Control 49:108–120 4. Budzianowski WM (2015) Single solvents, solvent blends, and advanced solvent systems in CO2 capture by absorption: a review. Int J Glob Warming 7(2):184–225 5. Gaskell D. (1995) Introduction to the thermodynamics of materials. Taylor & Francis. Washington D.C 6. House KZ, Harvey CF, Aziz MJ, Schrag DP (2009) The energy penalty of post-combustion CO2 capture & storage and its implications for retrofitting the U.S. installed base. Energy Environ Sci 2:193–205 7. Budzianowski WM (2012) Negative carbon intensity of renewable energy technologies involving biomass or carbon dioxide as inputs. Renew Sustain Energy Rev 16(9):6507–6521 8. Budzianowski WM (2012) Value-added carbon management technologies for low CO2 intensive carbon-based energy vectors. Energy 41(1):280–297

26

W.M. Budzianowski

9. Budzianowski WM (2012) Benefits of biogas upgrading to biomethane by high-pressure reactive solvent scrubbing. Biofuels, Bioprod Biorefin 6(1):12–20 10. Svendsen HF, Hessen ET, Mejdell T (2011) Carbon dioxide capture by absorption, challenges and possibilities. Chem Eng J 171(3):718–724 11. Zhang G, Yang Y, Xu G, Zhang K, Zhang D (2015) CO2 capture by chemical absorption in coal-fired power plants: energy-saving mechanism, proposed methods, and performance analysis. Int J Greenhouse Gas Control 39:449–462 12. Yu J, Wang S (2015) Modeling analysis of energy requirement in aqueous ammonia based CO2 capture process. Int J Greenhouse Gas Control 43:33–45 13. Geuzebroek FH, Schneiders LHJM, Kraaijveld GJC, Feron PHM (2004) Exergy analysis of alkanolamine-based CO2 removal unit with AspenPlus. Energy 29(9–10):1241–1248 14. Skorek-Osikowska A, Kotowicz J, Janusz-Szymańska K (2012) Comparison of the energy intensity of the selected CO2-capture methods applied in the ultra-supercritical coal power plants. Energy Fuels 26(11):6509–6517 15. Arias AM, Mores PL, Scenna NJ, Mussati SF (2016) Optimal design and sensitivity analysis of post-combustion CO2 capture process by chemical absorption with amines. J Clean Prod 115:315–331 16. Budzianowski WM, Wylock CE, Marciniak PA (2017) Power requirements of biogas upgrading by water scrubbing and biomethane compression: comparative analysis of various plant configurations. Energy Convers Manag. doi:10.1016/j.enconman.2016.03.018 17. Razi N, Svendsen HF, Bolland O (2013) Cost and energy sensitivity analysis of absorber design in CO2 capture with MEA. Int J Greenhouse Gas Control 19:331–339 18. Avci A, Karagoz I (2009) A novel explicit equation for friction factor in smooth and rough pipes. ASME J Fluids Eng 131:061203 19. Lin Y, Rochelle GT (2016) Approaching a reversible stripping process for CO2 capture. Chem Eng J 283:1033–1043 20. Kim H, Lee KS (2016) Design guidance for an energy-thrift absorption process for carbon capture: analysis of thermal energy consumption for a conventional process configuration. Int J Greenhouse Gas Control 47:291–302 21. Alhajaj A, Mac Dowell N, Shah N (2016) A techno-economic analysis of post-combustion CO2 capture and compression applied to a combined cycle gas turbine: Part I. A parametric study of the key technical performance indicators. Int J Greenhouse Gas Control 44:26–41 22. House KZ, Baclig AC, Ranjan M, van Nierop EA, Wilcox J, Herzog HJ (2011) Economic and energetic analysis of capturing CO2 from ambient air. Proc Natl Acad Sci USA 108 (51):20428–20433 23. Wilcox J. Carbon capture. 2012. Springer, New York 24. Wołowicz M, Milewski J, Futyma K, Bujalski W (2014) Boosting the efficiency of an 800 MW-class power plant through utilization of low temperature heat of flue gases. Appl Mech Mater 483:315–321 25. Higgins SJ, Liu YA (2015) CO2 capture modeling, energy savings, and heat pump integration. Ind Eng Chem Res 54(9):2526–2553 26. Wang M, Joel AS, Ramshaw C, Eimer D, Musa NM (2015) Process intensification for post-combustion CO2 capture with chemical absorption: a critical review. Appl Energy 158:275–291

Process Implications of CO2 Capture Solvent Selection Leigh T. Wardhaugh and Ashleigh Cousins

Abstract In the development of energy efficient solvents for advanced CO2 capture processes, it has often been the case that potentially excellent solvents are overlooked or set aside because of perceived process difficulties or extra costs due to the physical, chemical or thermodynamic properties of the solvent. This chapter considers the process implications of solvent selection with the objective of selecting and designing the process to suit the solvent rather than forcing the solvent into an existing process. The chapter discusses the design, modelling and costing of conventional amine processes insofar as advanced energy efficient solvents still rely on this process. Individual solvent properties, including reaction kinetics, thermodynamics and physical properties are outlined and their impact on design are discussed. Aspects such as degradation products, corrosivity and environmental considerations will impact the selection of process equipment and materials of construction. The impact of these properties of energy efficient solvents on individual unit operations are also discussed in some detail. Nomenclature and Acronyms Acronyms FDG HPLC HSS HTU IC LP MDEA MEA

Flue gas desulphurization High pressure liquid chromatography Heat stable salts Height of a transfer unit Ion chromatography Low pressure Methyldiethanolamine Monoethanolamine

L.T. Wardhaugh (&) CSIRO Energy, Mayfield West, NSW 2304, Australia e-mail: [email protected] A. Cousins CSIRO Energy, Pullenvale, QLD 4069, Australia © Springer International Publishing AG 2017 W.M. Budzianowski (ed.), Energy Efficient Solvents for CO2 Capture by Gas–Liquid Absorption, Green Energy and Technology, DOI 10.1007/978-3-319-47262-1_3

27

28

L.T. Wardhaugh and A. Cousins

MNPZ MP PCC PZ SCR VLE

Mono-nitrosopiperazine Medium pressure Post-combustion capture Piperazine Selective catalytic reduction Vapour-liquid-equilibrium

Symbols Cp E G H vap DHCO 2 I L m MW P Q T v w a D

Heat capacity (constant pressure) Efficiency; capture efficiency; Energy Mass flowrate of gas Height (of packing) Desorption enthalpy for CO2 Interest Mass flowrate of liquid Mass flowrate Molecular weight Pressure, Plant capacity Heat, Energy Temperature Volumetric fraction of a component in a gas Weight fraction CO2 loading (moles CO2/moles active solvent component) Difference

Subscripts 0 am cap CO2 cond cw DES FG in L out R sens tot

Unloaded solvent Amine or other active component of the capture solvent Capital (cost) Value for CO2 or its fraction of a stream Condenser, Condensate Cooling water Desorber (stripper) Flue Gas Inlet flow Lean (stripped) solvent Outlet flow Rich (containing captured CO2) solvent Sensible heat Total

Process Implications of CO2 Capture Solvent Selection

29

1 Introduction Processes to capture CO2 from combustion flue gas streams have developed from processes used to remove impurities (principally H2S, CO2) from natural gas streams that are usually at high pressure. Physical absorbents such as methanol or propylene carbonate, used in pressure swing processes are economically viable when the partial pressure of CO2 (or the target impurity) is above approximately 500 kPaabs [1], whereas the CO2 partial pressure in flue gas streams is in the range 5–20 kPaabs. The development of thermal swing absorption processes to capture CO2 using alkanolamines, first patented in 1930, made it economically feasible to remove impurities from gas streams with low partial pressures of CO2. The Gas/Spec FT-1 process developed by Dow Chemical and subsequently marketed by Fluor Daniel as the Economine FG process used monoethanolamine (MEA) together with proprietary additives and corrosion inhibitors to effectively remove CO2 from low pressure oxygen containing streams in the absence of reducing agents and with SO2 levels below 10 ppmv [2]. In the search for more efficient CO2 capture solvents, the absorption/desorption process using 30 wt% MEA has become the baseline solvent against which the new solvent developments are often compared. MEA has the advantage of having a high reaction rate but the disadvantages of high energy requirement (for the reverse reaction—desorption), high corrosivity, toxicity and susceptibility to degradation through reaction with other chemical species. Higher solvent concentrations would be preferred but are limited by considerations of corrosivity, degradation rate and viscosity. With the current world-wide focus on the need to reduce the emission of CO2 into the atmosphere, a wide range of absorption solvents (absorbents) have been investigated [3] to achieve the goal of low cost capture of CO2 from flue gases, hence the topic of this book. With the proposal of alternative absorbents, however, what is often forgotten is the process design, operability and capital cost implications of the proposed absorbent choice. This is the topic of this chapter. The following sections will put the choice of absorbent in the context of total plant design and costing; consider the thermodynamic and physical properties that impact design, operation and hence the total cost of the project, the design options and how they are impacted by absorbent choice, and a final word on operability, safety and other issues relating to absorbent choice. The overall message is that the process must be designed for the solvent, rather than trying to force fit an advanced solvent into a conventional process design. A new absorbent can only truly be evaluated in a process designed specifically and optimized for the absorbent in question.

30


2 The Design and Costing of Post-Combustion Capture Plants 2.1

Design Process Overview

The familiar layout of a typical Post-Combustion Capture (PCC) plant is shown in Fig. 1. Figure 1 is unusual in that the size of the icon for each unit operation is scaled to the capital cost of the item (costs based on Ramezan et al. [4], pro-rated on power or size). A similar figure could be generated illustrating the energy cost of each unit operation, with electrical energy pro-rated according to the power station efficiency. Such figures could in principle be created for each solvent option to provide a graphic illustration of the impact of solvent choice on cost. The scale of the process is determined by the flue gas inlet mass flowrate (Gin) and the required capture efficiency (ECO2: —usually set at 90%) while the flue gas source determines the flue gas composition, flue gas molecular weight (MWFG) and specifically the CO2 volume fraction ðvCO2 : Þ. MWCO2 is the molecular weight of CO2. CO2 captured ½kg=hr ¼ GCO2 ¼

ECO2 vCO2 : Gin MWCO2 MW FG

ð1Þ

Given a basic process using MEA with an (unloaded) alkanolamine concentration (wam ) of 30 wt%; a cyclic loading [the difference between lean and rich

Fig. 1 Typical PCC plant using 30 wt% MEA. The size of the icon for each unit operation is scaled to the capital cost of the item (costs based on Ramezan et al. [4], pro-rated on power or size; adapted from Cottrell et al. [5])


31

loadings (aR–aL)] of approximately 0.3 mol CO2/mol MEA, then for each tonne of CO2 captured, up to 16.4 tonnes of solvent must be circulated through the absorber and stripper setting the size of pumps, piping and heat exchangers. In general for an alkanolamine of molecular weight MWam, the required unloaded solvent circulation rate (L0) is given by:

ECO2 vCO2 Gin MWam ð/R /L Þ MWFG wam wam /L MWCO2 Lean solvent rate ½kg=hr ¼ LL ¼ Lo 1 þ MWam wam /R MWCO2 Rich solvent rate ½kg=hr ¼ LR ¼ Lo 1 þ MWam

Unloaded solvent rate ½kg=hr ¼ Lo ¼

ð2Þ ð3Þ ð4Þ

The rich and lean loadings cannot be independently set. It is desirable to have a high cyclic loading, but a high rich loading may be the consequence of an overdesigned absorber (increased capital cost) while a low value of lean loading may indicate a desorber column that is being run with too high a steam rate (increased operating cost). The minimum energy requirement (Qtot/CO2 captured— MJ/kg or GJ/t) is determined in a series of experiments (real or modelled) over a range of values of Liquid to Gas mass flowrate ratios (L/G) with the heat input adjusted to give a fixed value of Capture Efficiency ðECO2 Þ. It is presumed that the values of L/G chosen are not too far from the contactor optimum to avoid biasing the outcome of the experiment. This aspect is discussed in more detail in Sect. 4.2. Due to the formation of bicarbonate, tertiary amines tend to have higher absorption capacities than primary or secondary amines. CO2 solubility in an absorbent is often determined via various vapour-liquid-equilibrium measurements discussed in Sect. 3.2. These methods can be used to determine the equilibrium CO2 loading of the absorbent under absorber and stripping column conditions. The difference gives the loading capacity of the absorbent. Absorbents with a high cyclic capacity are often preferred, as less absorbent is required for a given CO2 removal. However, a high cyclic capacity is of little benefit if the mass transfer of CO2, and the rate of reaction is slow (taller absorber column required). Knowledge of the gas and liquid flowrates allows the design of the contactor column diameters to be determined (discussed further in Sect. 4.2) and a knowledge of the performance of the solvent (particularly the rate of reaction—Sect. 3.1; and diffusivity— Sect. 3.3) allows the column heights (or key parameters for other types of contactors—Sect. 5) to be determined. Inlet gas and liquid streams to CO2 absorber columns are typically maintained at temperatures in the range 40–60 °C. This temperature range typically provides a balance between the system thermodynamics, which favour lower temperatures, and reaction kinetics, which favour higher temperatures [6]. Higher temperatures will also improve viscosities and diffusion coefficients, enhancing overall mass

32


transfer rates. CO2 absorption into aqueous amines is an exothermic process, which leads to temperature increases in the column. This will lead to a temperature bulge in the absorber column, the location of which will be affected by the operating conditions of the plant (especially L/G). Peak temperatures of up to 70 °C have been observed in PCC pilot plants operating on 30 wt% MEA [7]. The total heat requirement for the PCC process can be determined directly from the steam consumption in the reboiler usually carried out by measuring the steam inlet pressure and the exiting condensate flowrate. This total heat measurement must be corrected for heat losses which in small scale equipment can be significant and variable leading to significant errors. Alternatively, the total heat requirement ðQtot Þ can be determined, using appropriate measurements from its components, namely: Qtot ðkWÞ ¼ Qsens þ QCO2 þ Qcond

ð5Þ

The components of the total heat requirement are as follows: • The sensible heat requirement (Qsens)

Qsens ðkWÞ ¼ LDESin CpL ðTDESout TDESin Þ

ð6Þ

where LDESin is the mass flow of absorption liquid entering the stripping column (kg/s), CpL is the specific heat capacity of the absorption liquid (kJ/kgK), and ðTDESout TDESin Þ is the temperature difference between the hot lean absorption liquid leaving the reboiler and the incoming (preheated) rich absorption liquid (K). • The energy required to reverse the CO2 absorption reactions QCO2

vap QCO2 ðkWÞ ¼ GCO2 DHCO 2

ð7Þ

vap where GCO2 GCO2 is the mass flow rate of CO2 produced/captured (kg/s), and DHCO 2 is the desorption enthalpy for CO2 (kJ/kg).

• The energy in the excess steam leaving the stripping column (Qcond)—approximated by the condenser duty, (kW)

Qcond ¼ mcw :CPcw :DTcw

ð8Þ

where mcw is the mass flow rate of cooling water to the stripping column condenser (kg/s), CPcw is the specific heat of the cooling water (kJ/kgK), and DTcw is the temperature difference of the cooling water into and out of the condenser (K).


33

The value obtained from Eq. (8) may have to be corrected for any sub-cooling of the condensate and, if significant, the effect of the pressure drop across the stripper. The total heat requirement for the PCC process ðQtot Þ can then be determined from the summation of the components as given in Eq. (5). Typically a less energy intensive solvent is also less reactive requiring taller absorber vessels. Therefore an energy cost is replaced with a capital cost. Much of the research into alternate solvents seeks to break this nexus between reactivity and desorption energy requirements [8].

2.2

Process Modelling for Different Solvent Choices

Considerable work has gone into the development of rigorous design procedures for PCC plants which in turn can be used to develop complete cost studies (summarized briefly in Sect. 2.3). The development of process models continues even for the system considered the baseline (30 wt% MEA) with work done by Plaza et al. [9], Zhang et al. [10], Freguia and Rochelle [11] and more recent contributions by von Harbou et al. [12] and Razi et al. [13]. Thorough validation has also been made against pilot scale results [14, 15]. Process models rely for their precision on 2 criteria: • A thorough and precise representation of the thermodynamics, chemical reaction kinetics and physical properties of the chemical species present and their various interactions. This can be readily tested prior to the development of the process model itself. For new solvents, their intermediate reaction products, products of degradation etc. may not be adequately covered in existing property databases so that appropriate property model parameters must be incorporated into the program. This may require the close liaison with the process software developer. Researchers such as Yu et al. [16] have also considered modelling the properties of solvent blends. • Correct allowance for operational matters such as foaming, back-mixing, inadequate distribution of gas and liquid streams in the contactors; heat loss or fouling in heat exchangers; multiphase flow in pipelines, valves and equipment etc. Appropriately designed experiments, in which the interchange of information between model and experiment can help elucidate each of these operational issues and thus strengthen the applicability of the models. Clearly from the above, process modelling has a key role to play in capture solvent evaluation but the particular pathway and outcomes will be strongly dependent on the choice of solvent.

34

2.3


Overview of Process Costing Relating to Solvent Choice

The costing of a particular solvent selection is the most important step but also the most contentious with a huge variation in the values quoted ($/t CO2 captured) for a completed PCC plant project. This is mainly due to differing costing assumptions. Efforts have been made to standardize costing assumptions and processes [17]. Key cost parameters are the assumptions concerning energy cost (CostEnergy —$/GJ) and interest on capital (Icap) taken as an average annual figure. As shown by Dave et al. [18] the relative cost of capital and energy prevailing in different locations will mean that the choice of solvent must be determined on a case by case basis with no clear winner based on energy cost alone. The purpose here is to illustrate the effect of solvent choice on design and cost and for this a simplified cost structure can illustrate the change in total cost (DCosttot—$/yr) of a solvent choice with some relatively simple calculations (Eq. 9) for a plant of capacity P ðtCO2 =yrÞ. Of course a full design and detailed cost analysis would be required to fully validate the choices made. DCosttot ¼ DCostcap Icap þ DEtot P CostEnergy

ð9Þ

where the change in capital cost DCostcap can be estimated by pro-rating the change in equipment size (e.g. column height) based on the change in the average value of parameters such as mass and heat transfer coefficients, pressure drop, etc. [19]; while the change in energy requirement ðDEtot Þ derives directly from the experimental determination of the optimum L/G ratio and/or the changing requirements of, for example, total pressure drop. The cost of the energy (steam and electrical) is based on the locally prevailing opportunity cost of electricity and the efficiency of the power plant in question.

3 Solvent Properties and Their Impact on Design In this section a focus is given to the solvent properties that most effect process selection, design and operation and discusses how the design may be modified to overcome potential problems and take advantage of the unique properties of particular solvents.

3.1

Reaction Kinetics

The most important property of a CO2 capture solvent formulation is its ability to reversibly react with CO2, the rate at which and extent to which this occurs is a function of solvent composition, temperature and CO2 loading. The reaction is


35

reversed by heating the CO2 loaded solution and by reducing the CO2 partial pressure (e.g. by generating excess steam). Measurements of the kinetics (discussed below) and phase equilibria (discussed in Sect. 3.2) dominate the literature concerning solvent choices. The rate of reaction determines the required liquid residence time and hence the size of the contactor equipment. Experiments and/or modelling determine the optimum conditions of temperature, pressure and liquid/gas (L/G) mass flowrate ratios to complete the design process. Slow reacting absorbents may require large contact areas and residence times to achieve the target level of CO2 removal, leading to large and expensive columns. Primary and secondary amines tend to react the fastest with CO2 as they form a stable carbamate. Tertiary and hindered amines catalyse the hydrolysis of CO2 to form a bicarbonate ion and protonated amine. The formation of bicarbonate is slower compared to carbamate formation, thus CO2 removal via tertiary and hindered amines tends to be slower than from primary or secondary amines [20]. CO2 mass transfer rates and reaction kinetics can be measured in the laboratory via a number of experimental techniques: • Wetted wall column—A wetted wall column is a simple gas/liquid contacting device that is used to determine the mass transfer of CO2 from the gas phase to a liquid absorbent under carefully controlled conditions. A thin liquid sheet of known surface area is bought into counter-current contact with the gas phase. Gas phase concentration measurements are used to determine the flux of CO2 taken up by the absorbent. Plots of the CO2 absorption flux against the applied driving force (difference between actual and equilibrium partial pressure) are then used to determine the overall mass transfer coefficient [21]. This is a useful measurement for initial screening of absorbents, and also for determining fundamental data required for modelling purposes. Absorbents with a high CO2 mass transfer rate are typically preferred. • Spray column—spray columns are sometimes used for absorbents with slow reaction rates that require larger surface areas for meaningful mass transfer determinations [22]. The method is similar to that used with the wetted wall experiments, except the surface area is provided by liquid droplets. • Stopped flow—Stopped flow spectrophotometers are rapid mixing devices used to study the kinetics of fast reactions in solution. For PCC applications, these are used to measure the reaction kinetics of CO2 with various absorbents. Syringes are used to force solutions into a mixing chamber where the flow is ‘stopped’ by an opposing piston. Measurements can be made on the level of a few milliseconds [21].

36

3.2


Phase Equilibria

The ultimate capacity of the absorbent for capturing CO2 is determined by the phase equilibria relationships between the absorbent and CO2 as a function of temperature, solvent composition, and CO2 partial pressure. Due to the formation of bicarbonate, tertiary amines tend to have higher absorption capacities than primary or secondary amines. CO2 solubility in an absorbent is often determined via vapour-liquid-equilibrium (VLE) measurements. There are various methods available for completing these measurements which are used to determine the distribution of CO2 between the gas and liquid phases for absorbents under various conditions. VLE of alkanolamine systems has been investigated by many researchers including Bishnoi and Rochelle [23], Dang and Rochelle [24], Jou et al. [25], Kumar et al. [26], Fan et al. [27], and Park et al. [28]. Most conventional capture solvents remain as a single phase liquid throughout the absorption and desorption processes. Measurements of phase equilibria are mostly concerned with the equilibrium content of CO2 that has reacted with the absorbent forming a mix of dissolved CO2, carbamates, bicarbonates and carbonates as a function of temperature, gas phase CO2 partial pressure and solvent composition. An overall CO2 content of the liquid is sufficient to carry out preliminary design, while advanced rate-based design techniques require a detailed knowledge of the speciation which can be determined by Stopped Flow reactor techniques [21] or NMR measurements [27]. The condition of perfect phase equilibria is seldom reached in any process equipment and the terms “approach to equilibrium” or “stage efficiency” are used to express this feature. A more rigorous approach is to consider the transfer of material through successive layers at the gas-liquid interface and the reactions that occur in those layers. Blended amines often incorporate a fast reacting species (termed a “promotor”) together with a higher capacity (but slow reacting) tertiary or hindered amine, thus gaining advantage of each specie’s best properties. The modelling of the reaction pathways in such systems can become quite complex [29, 30]. The vapour pressure of the absorbent species is often over-looked in early screening. A highly volatile absorbent, such as ammonia, can suffer from significant absorbent loss. This has implications not only on the environmental impact of the process, but also the cost. Care needs to be taken when assessing the possible vapour emissions of absorbent species. Early work with piperazine (PZ) suggested it would have a higher volatility than the standard MEA absorbent due to its lower boiling point (146 °C for PZ compared to 170 °C for MEA). However, CO2 loaded piperazine solutions were in fact found to have a lower volatility than the standard MEA [31]. Operation at a pilot plant treating coal power station flue gas saw no appreciable loss of piperazine through the vapour phase during operation [32]. A high vapour pressure does not necessarily rule out an absorbent for CO2 capture applications. Instead, additional engineering controls may be required. An example is the ammonia process where absorption is completed at low temperatures to


37

minimise evaporative losses. Careful design of downstream cleaning equipment (e.g. the absorber wash section) can also minimise absorbent and other emissions to the environment [33]. Increasingly there is an interest in solvents that undergo a phase change as a result of the reaction with CO2 to form 2 liquid phases [34] or a slurry phase. Processes can be developed to take advantage of this phase separation to reduce the energy demand by sending only the phase rich in reacted solvent to the stripper and recycling the unreacted (lean) phase back to the top of the absorber. In each case it is necessary to know the equilibrium between all the possible phases—VL1L2E (Vapour-Liquid1-Liquid2 Equilibria) or VLSE (Vapour-Liquid-Solid Equilibria). It is also necessary to know the rate and conditions of temperature and pressure at which the additional phases form and separate to determine equipment operating conditions while the physical nature of the phases (droplet and particle size distribution, surface tension, density difference) will determine the design and cost of the additional separation equipment. An aqueous solution of NH3 can be considered, in one respect, as being the ultimate amine, with an hydrogen atom attached to the NH2—functional group rather than an organic tail. Ammonia is cheap to produce (relative to other amines), does not degrade, reacts more rapidly and has lower energy demands (at comparable concentrations) and is environmentally benign (relative to other amines). Although ammonia is normally a vapour, the very high solubility in water makes aqueous ammonia a potential capture solvent. The process issues with aqueous ammonia include the loss of ammonia in the vapour streams (slip) due to its high vapour pressure, the formation of solid ammonium bicarbonate deposits in vapour streams leading to line blockages and the low reaction rates and high energy requirements which are a direct consequence of the very low concentrations and temperatures that must be applied to reduce the slip problem.

3.3

Solvent Viscosity and Diffusivity

The viscosity of the solvent typically increases with the increasing molecular weight of the amine, with increasing amine concentration and with decreasing temperature. Viscosity also typically increases with loading, and this should be determined experimentally as a function of the loading. The increase in viscosity has two effects on the performance of the solvent as a capture agent. Firstly, the diffusion coefficient is inversely proportional to viscosity, so that the mass transfer coefficient will decrease accordingly. Secondly, in a packed column, surface is generated by the flow of the liquid on the inclined surfaces of the packing under gravity. Increasing viscosity will lead to thicker fluid layers on the packing (lower specific surface area per unit liquid volume), greater hold-up and longer residence times. It is usually not recommended to use a packed bed with capture solvents with a viscosity greater than 5 mPas [35], although more recent work suggests that this is

38


not as serious a limit as originally thought. Viscosity also affects the liquid-side mass transfer coefficient through affecting turbulence in the liquid phase. Absorbents with a viscosity that is too low will not form suitable liquid films over column packing. Thus, for absorbents whose viscosity falls outside the suitable range for packed column applications, alternative mass transfer devices should be considered. Tray columns use gas pressure to generate surface area through the bubbling and spray formed as the gas passes through the cross-flowing liquid. Increasing viscosity severely reduces the formation of spray and increases the energy required to form bubbles. For a similar reason, spray columns will not form fine sprays above approximately 50 mPas. Rotating contactor equipment creates surface through the addition of centrifugal energy (at a cost) and is therefore applicable to a much wider range of solvent viscosities. The very high specific surface areas (per unit liquid volume) that are possible in rotating contactors and consequent thin liquid layers can compensate for the reduced diffusion rates by providing a shorter path length to the solvent reactant. This is discussed further in Sect. 5. Viscosity also plays a key role in the performance of heat exchanger equipment through its effect on Reynolds Number. This is of particular concern in the reboiler where laminar flow or poor flow distribution could lead to hot spots and more rapid thermal degradation of the amine. The performance of the cross heat exchanger will be determined by the difference in viscosities between the rich and lean solvents as discussed above. The viscosity of various loaded amine systems has been investigated by a number of researchers including Freeman and Rochelle [36], Amundsen et al. [37], Kumar et al. [38], Shokouhi et al. [39], Weiland et al. [40]. Diffusion and viscosity are closely related via an inverse relationship. Diffusion coefficients for CO2 into absorbents are often calculated via the N2O analogy. This states that the ratio of diffusion coefficients between N2O and CO2 in water is equal to that in an aqueous amine solution [29].

3.4

Surface Tension

The surface tension of a liquid seeks to minimise the surface area of the fluid, e.g. by forming droplets. This is a result of the unbalanced forces acting on liquid molecules at the surface. Molecules within a liquid are pulled in all directions by intermolecular forces. However, molecules at the surface have no intermolecular forces pulling them upward away from the surface. These intermolecular attractions thus tend to pull the surface molecules into the liquid and cause the surface to tighten like an elastic film [41]. Liquids that have strong intermolecular forces also have high surface tensions. The surface tension of a fluid is an important consideration in gas/liquid contacting devices as it affects the distribution of liquid over packing (or other contactor) surfaces (greater spreading associated with lower surface tension), liquid hold-up and


39

the formation and breakup of foams. Work completed on packed columns under CO2 absorption conditions determined that for high surface area packing materials, reduced surface tensions could increase the packing effective area, thought to be due to removal of capillary phenomena [42]. Lowering the surface tension of fluids via surfactant addition, however, can have deleterious effects. Sedelies et al. [43] investigated the effect of surface tension on packing effective area, however surfactant addition to achieve low surface tensions resulted in significant foam formation. Tsai et al. [42] similarly had to add an anti-foaming agent to their solution in order to perform low surface tension experiments. Surface tension is a measure of the amount of energy required to stretch or increase the surface area of a liquid by a unit area [41] and is difficult to measure precisely. As surface tension manifests itself in a number of phenomena, there are a number of different techniques to measure it (e.g. du Noüy ring, du Noüy-Padday, Wilhelmy plate, spinning drop, pendant drop, bubble pressure, drop volume, capillary rise etc.) with many methods using tensiometers or microbalances to measure the weight of the liquid meniscus. The surface tension of a fluid will decrease with an increase in temperature, becoming zero at its boiling point and disappearing at the critical temperature. The surface tension of a liquid will also be affected by impurities and degradation products. An example is the lowering of surface tension through the addition of surfactants.

3.5

Chemical Stability and Solvent Degradation

Amines undergo the following irreversible reactions that will consume the costly capture solvent, create deposition layers especially on the reboiler heat transfer surfaces that must be periodically cleaned, form by-products that accumulate in the solvent [e.g. heat stable salts (HSS)] and must be removed, or they may spontaneously leave with the gas streams increasing the emission of toxic materials. The nature and extent of chemical stability is determined by the nature of the solvent and the operating parameters and must be studied carefully. The impact of chemical stability and the possible emission of by-products may not be fully understood until the solvent in question is run for extended periods with full recycle including regeneration at pilot scale (e.g. Radgen et al. [44]) or at demonstration scale such as the SaskPower PCC demonstration plant at the Boundary Dam Power Station, Saskatchewan [45]. There are 4 main processes leading to the degradation of the active capture species all of which lead to a loss in cyclic CO2 loading and increased costs. • Thermal degradation—occurring mainly in the reboiler leading to more volatile by-products or free radicals that can subsequently polymerize forming the deposition layers on heat transfer surfaces. The presence of ammonia gas in the exit flue gas stream is the clearest indicator of thermal degradation. The process

40


consequence is the limitation placed on the reboiler tube surface temperature leading to the 120 °C limit for MEA. • Oxidative degradation—as oxygen is always present in flue gases (furnaces must be run with some degree of excess air to ensure complete combustion), a small amount will always be dissolved in the solvent and this is sufficient to form by-products such as formic acid, formaldehyde, etc. • Reaction with other acid gases. All amines react with the acid gases (SO2; SO3; NO2; N2O3) that are always present to some extent in flue gases by similar reaction pathways as CO2, however the products of reaction will not decompose back to the reactants in the stripper and are hence termed ‘heat stable salts’ (HSS). These are typically removed in a small heat exchanger, termed a Reclaimer, operated at sufficient temperature to boil off the amine returning it to the process. The residue from this vessel is bled off and should be considered as a hazardous waste. • NOx (nitrogen oxides) will readily react with secondary amines (R-NH-R’) to form nitrosamines which have been of particular concern until engineering solutions were found to mitigate the impact of these reactions [46–48].

3.6

Corrosivity

Unloaded aqueous amines are mildly corrosive to carbon steel, however, the corrosivity of solvents typically increases as they degrade and absorb acid gases from the flue gas [49]. The experience gained in natural gas processing plant provides some insight into likely corrosion under PCC conditions. One critical difference between natural gas processing conditions and PCC applications is the oxygen content of the gas. This is typically minimal under natural gas processing conditions, but can be in excess of 10 vol.% for power station flue gases being the feed-stock to PCC plants. Most amines will degrade in the presence of oxygen, and a number of the oxidative degradation products are corrosive. Thus corrosion under PCC conditions will likely be higher than that witnessed to date in natural gas processing plant. In addition, a number of the techniques used to minimise corrosion in natural gas plant, such as using an inert gas blanket to minimise oxygen ingress, will not be applicable under the higher oxygen conditions in PCC. Severe corrosion has been linked to amine degradation [1]. As a result, amines that are susceptible to oxidative degradation, such as diethanolamine (DEA), are less suitable for PCC applications. Degradation products formed through reactions with oxygen are noted to be particularly acidic [50]. These include oxalic, glycolic, formic and acetic acid salts. These are stronger acids than the carbonic acid formed through the absorption of CO2, and will build-up in amine solutions over time. Unlike the absorbed CO2, these salts cannot be thermally regenerated, and hence are known as Heat Stable


41

Salts (HSS). There are recommended upper limits for these salts in most commonly used amines [1]. When the salt content gets too high, a slip stream of the lean amine solution will generally be sent for reclamation via caustic addition and batch distillation. For DEA and methyl-diethanolamine (MDEA) solutions, caustic addition is recommended once HSS level reaches 0.5 wt% [1]. Different amines will have different levels of corrosivity due to the species formed when they react with CO2 and their chemical stability. Typically, primary amines tend to form more corrosive solutions, followed by secondary amines, whilst tertiary amines the least. The reason for this ranking is not well understood. One suggestion relates to the amines ability to form a stable carbamate [51, 52]. For this reason, tertiary amine absorption liquid solutions based around MDEA have become popular for natural gas processing. However, due to the low CO2 concentration and low flue gas pressure, PCC applications will likely require fast-reacting amines, such as primary amines or to use tertiary amines in combination with a ‘promoter’. As amines with high concentration are likely to form higher concentrations of acidic degradation products, there are maximum concentration recommendations for most common absorbents in order to limit corrosion. MEA for example has a recommended maximum concentration of 20 wt% under natural gas processing conditions [1]. However there are benefits in using higher concentration absorbents, such as reduced absorbent and energy requirements. As a result, many proprietary solutions use higher absorbent concentrations, but will also include corrosion inhibitors. An example is the Fluor Econamine FG Plus utilizing a higher concentration of MEA coupled with proprietary corrosion inhibitors [53]. As corrosion rates tend to increase with an increase in CO2 loading, highly loaded absorbents will suffer from greater corrosion affects. As a result, there are recommended upper limits on acid gas loading for most commonly used amines [1]. Under PCC applications, absorbent loading is often altered in order to minimise the regeneration energy requirement of the process. Kittel, Gonzalez [54] suggest optimum operating conditions for a 30 wt% MEA solution occur at a lean loading of 0.25 mol CO2/mol MEA. This is above the lean loading limit typically recommended. Thus operation under optimum energy conditions for PCC applications may result in higher levels of plant corrosion. Corrosion effects can be minimised through engineering design. Minimising acid gas flashing and turbulence through avoiding restrictions, tight corners, pump cavitation and minimising absorbent flow rates, are recommended [51]. Where engineering controls are not sufficient, corrosion inhibitors are often used. Inhibitor choice needs to be made carefully as inhibitors can affect plant operation (e.g. through increased foaming). Incorrect inhibitor choice can also increase oxidative amine degradation, potentially increasing corrosion rates [55]. Use of inhibitors can allow the use of lower grade, cheaper metals for construction. Where this is not sufficient, higher grade metals may be required. In natural gas processing plant, known corrosion prone areas include the base of the absorber column and high temperature areas such as the base of the stripping column [1]. As such, stainless steel is recommended for these sections, whilst other areas of the plant can be

42


constructed from lower grade carbon steel. Corrosion rates under PCC conditions are expected to be higher than those typically seen in natural gas processing. As a result, the use of cheaper carbon steels for plant components might not be possible, thus other low cost construction methods are being evaluated. The absorber columns at both the Test Centre Mongstad (TCM) [56] and the Boundary Dam CO2 capture plant [45] use ceramic lined concrete towers for their absorber columns. Novel absorbents being developed for CO2 capture applications often have limited information about their long term stability and corrosiveness under flue gas conditions. Significant research work is being conducted into corrosion rates and the formation of degradation products under PCC conditions. In addition, novel, low-cost construction materials are also being evaluated [57–59]. Advanced solvents such as ionic liquids or amino acids display little or no corrosivity giving these solvents a distinct advantage.

3.7

Material Compatibility

While corrosivity addresses the interaction between solvents and metal components, material compatibility deals with the interaction between solvents and non-metallic components such as seals, gaskets etc. The development of new solvents will inevitably raise questions of compatibility with the materials of plant construction and the components of equipment such as pumps, blowers, compressors and instrumentation. These complex chemical interactions must be established through laboratory and long term pilot or demonstration scale testing. Apart from the contactor vessels themselves, points of particular concern include the rich absorption liquid lines at the base of the absorber (containing the highest quantities of contaminant materials), rich absorption liquid exit from the lean/rich cross heat exchanger (due to the increase in temperature at this point), the hot lean absorption liquid exiting the stripping column [1] and the absorber wash (post-treatment) section [58]. While suitable metals (e.g. various grades of stainless steel) can be found to achieve corrosion resistance, this adds significantly to the cost of these enormous contactor columns. To reduce the cost of the absorber and pre-treatment columns, the SaskPower Boundary Dam project adopted ceramic tile lined concrete vessels [60]. Extended operation of this facility will provide a wealth of information on not only the chemical and physical resistance of the tiles, but also the adhesive and grouting materials. The Niederaussem pilot plant in Germany has been evaluating a plastic-lined concrete module [57], to achieve further reductions in construction cost.


3.8

43

Other Properties

When selecting an energy efficient absorbent for CO2 capture applications, items such as the absorbent’s reaction rate with CO2, its cyclic loading, and energy requirements are often used for initial screening. However, before an absorbent is used at pilot-scale, other factors need to be considered. The full evaluation of the performance of a new solvent formulation requires additional physical properties to be experimentally determined, as a function of temperature, pressure, solvent concentration, and loading, including the following: • Heat capacity. Necessary for heat balance and heat transfer calculations and hence the evaluation of heat exchanger performance and total heat requirement. Note that heat capacity can be a strong function of loading [61]. • Thermal conductivity. Necessary for heat transfer calculations and hence the evaluation of heat exchanger performance. • Density. A fundamental property affecting fluid flow, drainage rates and key to any multi-phase separation processes [37, 62]. • pH. The net outcome of the reactions occurring in the process. This is more likely to be measured in the process directly as an indication of the extent of reaction (loading). Refer to Sect. 6.5.

4 Process Options and Design Relating to Solvent Choice This section considers the impact of solvent choice on the selection and design of individual unit operations.

4.1

Gas-Liquid Contactors

The central elements of a PCC plant are the absorption column (absorber) and the desorption column (stripper). Conventional design of the PCC gas-liquid contactors at commercial scale foresees the use of multi-train packed columns with random packing or (increasingly) structured packing. Alternate contactor types are considered in Sect. 5. There are numerous design procedures in the literature with some degree of discrepancy between them, making pilot scale validation essential. It is in the validation and scale-up process that pit-falls relating to solvent choice occur as discussed below.

44

4.2


Packed Bed Contactors

The packed bed operates by the intimate contact of the liquid which flows under gravity and (ideally) spreads over the packing material surface with maximum coverage, while the gas travels upwards by pressure differential in the space left between the liquid coated surfaces (i.e. counter-current and co-continuous). CO2 gas (and other residual acid and inert gas impurities) diffuse into the liquid at the interface and react with the capture solvent. The reaction depletes the interface of CO2 encouraging more CO2 to pass through the interface. Physical absorption of the gas components is also occurring, limited by the diffusion rate and equilibrium capacity of the capture solution for each gas component. The ‘lean’ capture solution is distributed evenly across the top of each of the packing sections then trickles down under gravity over the packing surface which is designed to maintain as even and complete a gas-liquid interface as possible, becoming ‘enriched’ in CO2 as it travels the length of the column and collects in the column base. As there is a considerable variation in the flow stream compositions, temperatures and (to a lesser extent) pressure and flowrate along the column, a differential approach is usually taken and a rate based calculation procedure, utilizing one or more film layers, avoids unrealistic assumptions about equilibrium between phases. Computerized methods are commercially available which allow a choice of design procedures. General advice is to not mix design procedures and to check that the validation of the selected procedure covers the physical property range of the selected solvent (especially viscosity). The gas velocity in the contactor column is set to just beyond the ‘load point’ (below which the gas and liquid flows do not affect each other) and where there is a slight improvement in the mass transfer rate. At higher flowrates, of either liquid or gas, the pressure drop increases dramatically and/or becomes unstable and the mass transfer rate falls at which point the column is said to have ‘flooded’. The optimum operating point is therefore defined as a certain % flood (gas velocity ratio to the ‘flooding velocity’) or as a specific pressure drop per unit height of packing (DP/H) and this defines the diameter of the contactor columns and, at larger scales, the number of columns. At lower liquid rates, especially if liquid distribution across the packing is poor (distributer design or surface tension effects), gas may by-pass part or all of the liquid reducing the amount of CO2 that can be captured.

4.3

Absorption Column

The choice of capture solution determines the height of the absorber through the rate of reaction and the diffusion rate of CO2 into the solution. Detailed calculation procedures and automated design processes are available based on laboratory measurements of vapour-liquid equilibria (Sect. 3.2); reaction kinetics (Sect. 3.1) and physical properties, especially viscosity and diffusivity (Sect. 3.3).


45

The amount of CO2 that can be captured in the absorber is fixed by the difference between the rich loading (determined for the most part by the absorber design) and the lean loading (determined by the stripper design and operation). A taller absorber column will provide the residence time and interfacial area but this may be inefficient. Running the stripper at higher reboiler rates (more excess steam) to remove more CO2 (delivering a leaner return capture solution to the top of the absorber) may also be inefficient in terms of energy consumption. To determine the optimum operating point for a given capture solvent a series of experiments is carried out on pilot facilities in which the Liquid to Gas mass flowrate ratio (L/G) is varied while a fixed value of %capture is maintained to determine the minimum total energy input. Higher liquid rates (at fixed gas rate) will capture more CO2 but require more energy (per unit CO2 captured) due to the higher sensible heat requirement (Eq. 6). Lower liquid rates will require that the returning lean solvent be stripped to a greater degree in order to achieve the same % capture again requiring more energy per unit CO2 captured. It is clear from the previous discussion, however, that if the flows are varied, then the column has moved away from its optimum operating point (% flood or DP/H) which will also affect the results. Clearly it is impractical to run each L/G point for each solvent in a different optimally designed column, but this is one of the reasons that there is little agreement as to optimum L/G and minimum energy requirements for a given solvent. Some piloting facilities address this issue in part by having a variable liquid entry point thus varying the effective height of packing for the different capture solutions and L/G ratios being tested. Tests would be carried out, for example, by choosing liquid entry points that give equivalent rich loadings. A more definitive approach would be to use the pilot data to thoroughly validate process models to a high level of confidence and to use the models to determine operating points for a range of L/G ratios with each point representing an optimum column design. Such models would then be used to determine the energy minimum and (more cogently) the economic optimum and this used to compare the choice of capture solutions. Absorption is accompanied by a release of heat (the heat of reaction, equivalent to the heat of vaporization of CO2) some of which is transferred to the gas stream but most of which leads to an increase in temperature within the column (reducing the equilibrium loading) and increasing the rich solvent exit temperature. Control of this temperature bulge is desirable as this leads to an increase in the cyclic loading. As a result, a number of cooling methods have been proposed to minimise the effect of the temperature bulge, allowing the column to operate in a more isothermal manner [63, 64]. The most suitable location for inter-cooling on the absorber column is not necessarily the location of the temperature bulge. The CO2 concentration in the vapour phase will typically decrease as the flue gas moves up an absorber column. It is the difference between this gas phase CO2 partial pressure and the equilibrium CO2 partial pressure that provides the driving force for mass transfer to occur. When these two pressures approach the same value, the driving force for mass transfer approaches zero, and the column is said to be ‘pinched’. Inter-cooling can have a significant effect on the equilibrium CO2 partial pressure,

46


and the location of inter-cooling is best placed where it can maximise the difference between the equilibrium and operating CO2 partial pressures in the column. Under PCC conditions, this is typically found to be in the lower portion of the absorber column, where equilibrium CO2 partial pressures are higher [65], however, there are situations where intercooling towards the top of the absorber column is advantageous (e.g. to reduce the evaporation of a more volatile solvent component). Multiple inter-cooling steps may also be economically desirable.

4.4

Desorption Column

The desorption column consists of: • A large stripping section (most of the column—the reason that it is often referred to as the ‘stripper’) in which CO2 is stripped out of the capture solution by the action of heat and reduced partial pressure provided by the steam, generated in the reboiler, that is travelling up through the column packing and counter-current to the downflowing liquid; and • A small rectification section in which the returning reflux (all of the condensed excess steam) re-captures as much of the vaporized amine as possible. The amount of amine that enters the vapour stream is determined by the equilibrium vapour-liquid relationship which is a function of composition, temperature and loading as discussed in Sect. 3.2 and can be significant for certain choices of capture solution. The rectification section may also comprise a small section in which rich solvent, split off prior to entering the cross exchanger, enters the column above the normal feed point to retrieve some of the energy in the exiting steam (‘cold rich split’ process modification). The merits of this process modification also depend on the choice of capture solvent. The desorption column is usually smaller than the absorber in both height and diameter and may consist of trays or packing or a combination of the two in different sections. The column may also run at elevated pressures or different pressures in different sections [66] to minimize the column size (though not necessarily the cost); to minimize the loss of amine to the vapour stream; and to reduce the cost of the CO2 product compression by providing this product stream partially pressurized. Once again, the choices here are economic and to a large degree determined by the choice of capture solution. The major energy consumer of a PCC plant is the reboiler, however, energy savings can be achieved throughout the system and particularly in the desorber through more efficient design and energy saving modifications [6] the relative effect of each of which are also determined by the choice of capture solution.


4.5

47

Heat Exchangers

In terms of choice of solvent, the major impact on heat exchanger design and operation is through: • The steam temperature and hence pressure that must be provided to achieve the optimum degree of solvent stripping, • The sensitivity of the solvent to degradation in the reboiler, and • The change in solvent properties (especially density, viscosity, thermal conductivity and heat capacity) with the change in CO2 loading. In a power station retrofit design the steam supply rate and pressure will impact the operation of the power station steam circuit requiring modifications such as replacement power turbines [4]. The use of 30 wt% MEA as capture solvent requires a supply of low pressure (LP) steam from the power station steam circuit due to the reboiler limiting temperature of approximately 120 °C. At this temperature, an adequate degree of stripping can be achieved. For concentrated piperazine (PZ) as capture solvent, higher stripper temperatures are possible due to the lower degradation rates which may require that the steam supply to the reboiler be sourced from the Medium Pressure (MP) section of the power station steam circuit thus presenting a different design task [32, 67, 68]. It may be desirable to operate the stripper column at higher pressures to minimize solvent losses and reduce the CO2 product compression costs. Higher stripper operating pressures also require higher reboiler steam pressures (to maintain the temperature driving force) or possibly larger heat transfer areas (larger equipment). Regeneration at higher pressures will require the stripper column to be designed as a pressure vessel, leading to higher costs. The degree and type of degradation of the solvent increases with temperature and it is on the heat exchange surfaces of the reboiler that the highest temperatures will be experienced. The partial vaporization of solvent that occurs in the reboiler could lead to local hot-spots that in turn lead to higher rates of degradation than expected. Higher rates of circulation through the reboiler can alleviate this effect. The common types of reboiler systems include: • Forced circulation reboiler, in which a pump supplies lean solvent from the base of the stripper column to the reboiler. This type is more expensive but easier to control and vary as operating conditions change. • Thermosyphon reboiler, in which the boiling solvent in the reboiler presents a lower density than the solvent in the base of the stripper column with this density difference driving a natural circulation of solvent through the reboiler. For a thermosyphon reboiler to work effectively the solvent properties, especially density, viscosity, heat capacity and thermal conductivity must be well understood as a function of temperature, solvent concentration and loading. For some solvents the physical properties, especially viscosity and heat capacity are a strong function of the loading. This will affect primarily the operation of the

48


cross heat exchanger and the resulting outlet temperatures and overall system heat balance. Adequate attention in design will provide allowances for this effect especially during unsteady conditions.

4.6

Pumps

The pressures and temperatures of the conventional PCC plant are relatively moderate and the selection of suitable pumping equipment is quite straight-forward. Choice of capture solution will affect the selection of seals and pump components to handle the corrosivity and material compatibility of the capture solvent. Particular care should be taken of rubber components that could swell or degrade in the presence of capture solvent, impurities or degradation products. Compatibility tests must be carried out if specific compatibility information is not available, which is particularly applicable to newly developed capture solvents.

4.7

Flue Gas Blower

There is not sufficient pressure in the typical power station flue duct to drive the flue gas through the additional piping, columns and packing of the capture plant gas stream. Therefore additional flue gas pumping capacity must be provided. This is an axial fan or rotary blower usually placed at the outlet of the pre-treatment column where the presence of corrosive impurities (e.g. SO2), residual particulates and gas temperature are at a minimum. The flue gas is therefore drawn through the pre-treatment column then pumped through the absorber and post-treatment column sections in a single step. It may be more economical to place a second fan or blower at the inlet or outlet of the post-treatment column to provide additional capacity. The placement of the fan(s) or blower(s) affects the column pressure and hence the absorption driving force, so that upstream placement is preferred, while on the other hand, the total volume of gas is decreasing through the gas train, favouring downstream placement. The water vapour content of the gas stream also affects the design by changing the gas density. The capture solvent does not normally contact the flue gas blower but impacts the design and selection by its influence on the overall gas-side pressure drop. For example, a less reactive capture solution will require either a taller column with a more open packing (to maintain total pressure drop) and/or demand a higher pressure to be delivered by the fan or blower. Note that each type of fan or blower will deliver specific capacity and pressure limits with significant step changes in these values and in the cost, so that the design task is not a simple linear relationship [19].


49

5 Alternate Contactor Designs Although packed columns with structured packing are most likely to be the contactors in first generation PCC plants, alternate contactor types should be considered especially in light of the wide range of capture solvents under development.

5.1

Spray Columns

The simplest and oldest gas-liquid contactor is the spray column although this belies the vast research that has gone into perfecting the spray nozzle that is at the heart of the device to deliver the droplet size distribution that is best suited to the process. The spray column delivers the lowest gas phase pressure drop but has a very short residence time that is not suited to the reaction times typical of amines with CO2. The spray column does find application as the pre-treatment column which is required if flue gas desulphurization (FDG) has not been implemented or does not meet the specifications required by the selected CO2 capture solvent. Spray columns have been investigated for PCC applications by [69–71] each of whom show promising results. A different system has been proposed [72] and successfully tested [73] in which the gas passes between sheets of liquid adsorbent sprayed horizontally through the contactor chamber. Liquids with viscosities above 50 mPas do not readily form droplets (rotating contactors, discussed below, are better suited to such capture solutions). At lower solution viscosities, the average drop size is determined by both the viscosity and the energy input. Intuitively, a very fine droplet size would be ideal as this dramatically increases the interfacial area per unit volume of liquid, however this proves to be counterproductive as the finer droplets are easily entrained and require lower gas velocities and hence larger vessel diameters [74].

5.2

Tray Columns

Tray columns are the workhorse of distillation systems and processes. Liquid travels across a tray from a ‘downcomer’ in a thin continuous layer through which gas passes by means of holes (sieve tray designs); valve caps or bubble caps mounted in the tray, then flowing over a weir that maintains the liquid layer and into the downcomer of the next tray. Because of the backmixing that occurs due to (desired) foaming and spray formation, each tray is considered to be single equilibrium stage and (perhaps) many stages are required in a vertical column to achieve the desired separation. Despite extremely high interfacial areas generated by the formation of foam and spray on each tray, tray columns are less favoured for gas absorption processes because of higher gas phase pressure drops (the cumulative

50


pressure for the gas to force its way through multiple liquid layers); the extra height on each tray needed for spray disengagement (to avoid liquid being carried back up the column) thus adding to the total column height; and the extra diameter to allow for the liquid downcomer and to restrict spray formation by reducing gas velocities at the point of tray entry (holes or cap openings). Tray columns may find application with specific capture solution choices whose physical properties (especially viscosity and surface tension) may readily form foams but not readily form sprays and therefore should not be completely discounted.

5.3

Rotating Contactors

With the introduction of the ‘Higee’ rotating packed bed by ICI in the 1970s [75], the possibility of operating with a wider range of liquids in a more compact device has opened up. Since then a wide range of rotating devices has been explored. The application of rotating packed beds to CO2 capture from flue gases has been explored by Cheng and Tan [76] who showed that the height of a transfer unit (HTU) is significantly reduced, and Yu et al. [77] who showed that capture performance in the rotating packed bed could be maintained despite a 7-fold increase in the viscosity of a capture solvent chosen for its non-precipitating operational advantages. Jassim et al. [78] investigated a range of MEA solvent concentrations up to 100 wt% and showed that concentrations higher than the conventional 30 wt % could give effective operation in physically smaller devices. The liquid side volumetric mass transfer coefficient was increased by an order of magnitude in the application of a rotating packed bed contactor to ionic liquids [79]. Ionic liquids show promise as stable non-volatile capture solvents but are typically of higher viscosity and lower reaction rate. This is a classic example of a promising capture solvent looking for an appropriate process. Rotating internals other than packing have been proposed by Makarytchev et al. [80] who have investigated the performance of the inverted cone and Wang et al. [81] who have shown the improved performance of a modified rotating disc contactor for a range of applications including CO2 capture. A recent development [82] that has a centrally located rotating tube as the only internal with the liquid emitted as rotating continuous sheets taking the form of an auger or blades, is able to provide surface area matching that of a packed bed and is also able to pump the gas through the column, reducing or possibly even eliminating the need for a flue gas blower. The centrifugal force provided by rotation causes the liquid to spread more evenly and more thinly on the available surfaces increasing overall mass transfer rates and allowing capture solvents with higher viscosities (higher concentrations and/or higher molecular weights) to be effectively utilized whereas application of these solvents in a conventional packed bed would give inadequate performance due to the lack of a driving force for fluid flow. While the mechanical issues may for the moment preclude the use of rotating devices at the massive scale required


51

by, for example, power station capture facilities, the process intensification afforded by these devices may find a niche in smaller scale, space limited applications such as off-shore or ship-board capture processes. Such applications would require the specific development of a class of capture solvents to optimize the performance of such devices.

6 Operation and Safety Considerations 6.1

Environmental Considerations

Post combustion capture using aqueous amines is currently the most technologically advanced method for removing CO2 from power station flue streams. Under PCC conditions, release of amine and amine by-products to the environment could occur through carryover with the exiting flue gas, or through spills from the process. Thus consideration of the toxicity and potential environmental impact of new and existing amines and other potential CO2 capture absorbents warrants consideration. Most amines are considered hazardous substances and are classified as dangerous goods. They are corrosive, irritants, and can cause respiratory problems if inhaled. As such, various levels of protective equipment may be required for maintenance personnel working on CO2 capture plant where there is potential for exposure to the absorbent. The choice of absorbent can have a significant effect on the process requirements, and hence the level of personnel exposure. Precipitating absorbents for example may have increased cleaning requirements to deal with blockages. Highly corrosive absorbents may have a higher propensity for leaks. An absorbent’s rate of degradation, and the type of degradation products formed can also have a significant effect on its toxicity and safe handling requirements. Amino acids have received attention as a potential CO2 capture absorbent as they are considered environmentally friendly [83]. Eide-Haugmo et al. [84] evaluated the eco-toxicity and biodegradability in the marine environment for a number of absorbents being considered for PCC applications. Tertiary amines were found to have a low biodegradability, whilst the amino acids evaluated had low toxicity and high biodegradation potential. In addition to emissions of the absorbent, formation and emission of degradation and by-products should also be considered. Secondary amines are often preferred for CO2 capture applications due to their fast reaction rates. Piperazine in particular is already widely used as a rate promoter in blends with other amines (e.g. MDEA). When applied to CO2 removal from combustion flue gases however, other issues arise. All amines can form nitrosamines through reaction with nitrosating compounds (e.g. NO absorbed from combustion flue gases), however, only secondary amines will form a stable nitrosamine directly [85]. This is a concern as many nitrosamines are carcinogenic. The potential for nitrosamine formation requires

52


consideration for all amine systems as most amines will contain trace levels of other amines, and can form a number of amine products as they degrade, however, this will be a particular concern for systems using secondary amines. Currently, the Norwegian Climate and Pollution Agency has set limits for the sum of all N-nitrosamine and N-nitramine from exceeding 0.3 ng/m3 in air and 4 ng/L in waters downwind of the Mongstad CO2 capture facility [85, 86]. Piperazine, a secondary amine, was recently evaluated on a coal combustion flue gas [32]. During long term evaluation, the formation of mono-nitrosopiperazine (MNPZ) in the absorbent was monitored. MNPZ concentration was noted to increase with operation, reaching a steady-state level after approximately 200 h. Operating the stripping column at higher temperatures (up to 155 °C) was noted to reduce the level of MNPZ in the absorbent, thought to be due to thermal break-down of the nitrosamine. A recent study [87], quantified emissions from a CO2 capture pilot plant using an aged MEA solution to treat a coal combustion flue gas. Small concentrations of Nitrosodiethanolamine were found in the solvent liquor, but was not detected in the wash water or gaseous emissions from the plant. As mentioned in Sect. 3.2 above, one of the items to consider when selecting an absorbent is it’s volatility under CO2 capture conditions. This will not only affect the economics of the process, due to absorbent make-up requirements, but also the environmental implications of the process. Absorbents with a high volatility may still be applicable to CO2 capture applications, but may require more extensive downstream cleaning requirements. In addition to loss of absorbent, loss of volatile degradation products (such as NH3) and the potential for further reaction in the atmosphere requires consideration [88]. The capacity of most CO2 capture absorbents will likely decrease with continued operation. This will be a result of the degradation of the absorbent and build-up of by-products such as HSS. As such, commercial PCC processes will likely include some form of amine reclamation. A common reclamation method is caustic addition followed by batch distillation. This boils off the amine, which is then returned to the process, whilst the degradation products and other unwanted material remains behind as a waste sludge. Disposal of this waste requires careful consideration [89] as it will be hazardous in nature. The extent and frequency of reclamation will depend on the absorbent choice and operating conditions, with absorbents experiencing high rates of degradation producing higher rates of waste. One method for dealing with this reclaimer waste is to return it to the power station furnace where it can be burnt alongside the fuel. This requires careful consideration however as this could require the power station to be re-classified as a hazardous waste combustor in some jurisdictions [90].

6.2

Phase Separation—Planned and Unplanned

A number of promising absorbents for acid gas removal are often overlooked, or their use limited, due to operating difficulties such as phase separation when used in


53

a conventional absorption/desorption process. An absorbent that forms unwanted precipitates during operation, for example, can clog pipework or column internals, and increase corrosion rates through enhanced erosion. An example of one such absorbent is piperazine (PZ). Piperazine is a secondary amine that shows promise for post combustion CO2 removal applications due to its fast reaction rates, low corrosivity and good stability (compared to the standard 30 wt% aqueous MEA) [67, 91]. Its concentration, however, is typically restricted to below 10 wt% due to solubility issues. Aqueous piperazine will form precipitates under ambient conditions at concentrations above this. As such, use of piperazine is typically restricted to being a rate promoter (e.g. when blended with the tertiary amine N-methyldiethanolamine, MDEA). Recent work however has shown that concentrated piperazine solutions can be used for CO2 removal applications provided the CO2 loading of the absorbent is maintained between certain limits. Under typical operating conditions (i.e. temperatures above 40 °C) the CO2 loading of an 8 molal (40 wt%) solution of piperazine should be maintained between 0.1 and 0.4 mol CO2/mol alkalinity [67] to remain soluble. If the plant is shut down however, and lower temperatures are experienced, then a narrower range of CO2 loading is required to avoid solubility issues. When operating a plant, unplanned or emergency shut down of the process must be considered. At the Stanwell Corporation Ltd. owned Tarong CO2 capture pilot plant, heat tracing was applied to solvent lines to minimise precipitation issues resulting from unplanned plant outages when operating with an 8 molal piperazine solution [68]. This is an example of trying to force fit the absorbent to a standard absorption/desorption process. Precipitation from solution in unplanned shut-downs also affects advanced energy efficient solvents such as amino acids. Ammonia is another absorbent considered for CO2 capture applications where precipitation can be an issue. Precipitation can occur in the absorber, where low temperatures are used to minimise vapour losses, and also in the stripping column condenser [92]. Precipitation however is not necessarily a problem. Alstom have developed the Chilled Ammonia Process (CAP) which operates at absorber temperatures of 5–10 °C. Under these conditions precipitation of ammonium carbonate and bicarbonate will occur [93]. Alstom are currently demonstrating the chilled ammonia process at a number of pilot plants worldwide [94]. Precipitation experienced in the stripping column condenser can also be avoided or reduced through modifications to the standard process. Redirecting a portion of the cold rich solvent to the stripping column condenser section could remove precipitation issues and lower condenser cooling duties [95]. Phase separation can be beneficial to the CO2 removal process, particularly when suitable process choices are made. Phase separation can be used to separate CO2 loaded absorbent, lowering the amount of absorbent sent to regeneration, potentially lowering energy requirements. IFP Energies Nouvelles in France have developed the DMX absorbents [96, 97]. These absorbents undergo liquid-liquid phase separation as CO2 is absorbed (temperature dependant). The captured CO2 concentrates in one of the two liquid phases. This can be decanted and thus only a

54

L.T. Wardhaugh and A. Cousins CO2 lean flue gas

CO2 to compression

Lean amine Cooler Decanter Cooler

Lean amine Absorber

Stripper Rich amine

Blower

Lean/rich heat exchanger Reboiler

Flue gas Rich amine Lean amine

Fig. 2 Simplified flow diagram of the DMX process based on Raynal et al. [97]

portion of the total flow is sent to the stripper for regeneration. A simplified flow diagram of the DMX process is provided in Fig. 2. Another instance of useful phase separation is in the combined CO2 and SO2 capture process (CASPER) being developed by TNO in the Netherlands [98] and CSIRO in Australia [99]. Here, both SO2 and CO2 from a combustion flue gas are captured by an amino acid solution. The CO2 loaded absorbent is regenerated via standard thermal stripping. A slip stream of the lean absorbent is separated for removal of the captured SO2. This is achieved by allowing K2SO4 to crystallise, separating the sulphur compounds from the absorbent. The K2SO4 crystals are then removed via filtration, and clean, regenerated absorbent is recycled to the absorber. A flow diagram of the CASPER process is provided in Fig. 3. An absorbent’s propensity to form precipitates is often identified during routine laboratory analysis. Despite this, unexpected precipitation can still occur once an absorbent has been transferred to an operating plant. Slightly different operating conditions and the provision of nucleation sites often means precipitation can be more of an issue in an operating plant. When operating with NH3 at the Munmorah power station, dilute NH3 concentrations were used to minimise vapour losses. In addition, absorber column temperatures closer to ambient were used to minimise the likelihood of forming precipitates in this column. Despite these precautions, precipitation of solid ammonium bicarbonate still occurred in the stripping column condenser, causing shut-down of the facility [100]. CO2 lean flue gas emitted to the atmosphere will likely contain trace absorbent vapour. The conditions at the flue gas exit can sometimes be sufficient for localised precipitation to occur.


55

Recycle stream

K2CO3 Crystallisation unit

Solvent + solid K 2SO4

Wash liquid

To stack

K2SO4 Solids Slurry

Wash

Filtration unit

Split stream

Filtration unit

K2SO4 precipitation CO2 to compression

Filtrate

Cooler

Combined SO2/CO2 absorption

Cooler

Cooler

Absorber

Condenser Stripper Condenser pump

Blower

Lean/rich heat exchanger Reboiler

Flue gas

CO2 removal

Rich solvent pump Lean solvent pump

Fig. 3 Process flow diagram of the CASPER process (based on Misiak et al. [98] and Cousins et al. [99])

6.3

Pretreatment Requirements

When standard CO2 removal technologies are applied to combustion flue gases, significant pre-treatment of the flue gas may be required. This is because trace constituents in the flue gas (e.g. sulphur and nitrogen oxides) will react with most alkanolamines forming heat stable salts. These salts bind strongly with the absorbent, lowering its ability to capture CO2, and can also increase the corrosiveness of the absorbent solution. This removal of sulphur and nitrogen oxides from the flue gas may be required even if upstream cleaning technologies such as selective catalytic reduction (SCR) and flue gas desulphurisation (FGD) are in place. Current flue gas cleaning technologies are designed for existing regulations. The European directive for large combustion plants restricts SO2 emissions from power stations to below 200 mg/Nm3 for new plants above 100 MWth. This limit however is still above that recommended for most conventional amine technologies used for CO2 capture. Thus further treatment is often required before the flue gas enters the CO2 capture plant, either through an upgrade of the existing flue gas cleaning systems, or

56


through installing new systems that could be equivalent in size and cost to the CO2 absorber column. If upstream FGD and SCR are not in place, then the temperature of the flue gas may also require adjustment. Flue gas temperatures of up to 180 °C are possible, well above the typical operating temperature of most amine based absorbers (nearer 40 °C). In addition, depending on the flue gas source, removal of excess water or saturation of the flue gas may be required. This is because most conventional CO2 capture technologies use aqueous solutions of alkanolamines. Water entering the process with the flue gas, or leaving via evaporation, can significantly affect the concentration of the absorbent. Thus monitoring and maintenance of the water balance of the CO2 capture system is important. Pre-treatment systems can be relatively simple, such as a caustic wash installed upstream of the CO2 capture plant. However, due to the low pressure and large volume of flue gas to be treated, pre-treatment systems will be large equipment in a commercial CO2 capture plant. Rather than apply expensive pre-treatment to combustion flue gases, new research is looking into the combined capture of multiple pollutants, often removing the need for pre-treatment systems. This has the potential to reduce the number of process stages required for the overall separation, reducing cost. Examples include the Cansolv process currently in use at the Boundary Dam coal-fired power station, the CASPER process [98, 99], the CS-Cap process being developed by CSIRO in Australia [101], and the various processes using ammonia (chilled ammonia process, Alstom [94]; ECO2, Powerspan [102]). Again, fitting conventional technologies to combustion flue gases might not prove to be the most economic strategy in the future. This is particularly likely in countries such as Australia, where FGD and SCR are not currently used.

6.4

Foaming

An absorbent’s propensity to foam can have a significant impact on the effectiveness and operability of the CO2 removal process. Mechanical foaming can be caused by excessive gas velocities through the absorber [103]. Chemical foaming can be caused by contaminants such as suspended solids, organic acids, corrosion inhibitors, condensed hydrocarbons, grease, and degradation products [1, 50]. For post combustion capture at coal-fired power stations, the potential for trace fly ash to persist in the flue gas entering the CO2 capture plant will likely exacerbate foaming issues. Sudden changes in column pressure drop or liquid level can indicate foaming. Foaming can reduce the contact area between the gas and liquid phases, reducing plant efficiency [50], and can also lead to excessive absorbent loss [1]. Pressure and liquid level fluctuations can lead to unstable plant operation. The best method for dealing with foaming is to remove the cause, for example through avoiding or removing contaminants. Where this is not possible or sufficient, anti-foaming agents can be added to the absorbent. An indication of an absorbent’s foaming propensity, and the effectiveness of various anti-foaming agents, can be


57

determined via simple laboratory analysis [50]. Different amines will exhibit varying levels of foaming [104]. Whilst some absorbent characteristics can be indicative of foaming (such as hydrophilic head with a hydrophobic tail), there is currently no clear method for predicting quantitatively an amine’s propensity to foam based on absorbent properties. Foaming in PZ-MEA systems has been studied by Chen et al. [105] who noted the complex effects and interactions of degradation products; metal ions; hydrocarbons; additives such as oxidation inhibitors. Conventional antifoam agents will generally assist in reducing foaming tendency though the complex relationships and the impact of the antifoam agent on the mass transfer coefficient should be thoroughly studied in the laboratory and validated in pilot scale studies.

6.5

On-Line Solvent Property Determination

Once a plant is operational, the online measurement of various solvent properties can provide valuable information concerning the performance of the plant. Key performance indicators are the rich and lean loadings, the stability of the solvent concentration (water mass balance), solvent loss via exiting gas streams and solvent degradation. On-line analytical techniques are available or under development including online automatic titrations or spectroscopic methods [106] that are capable of measuring these parameters directly, but are usually expensive and complex to operate, maintain and interpret. Simpler and less expensive measurement techniques, as listed below, can provide at least qualitative information about changes in plant performance, for example, a change in density and viscosity, but not in pH, will indicate a change in the solvent concentration and indicate the action required to correct the water imbalance. The relationship between the physical properties and the key performance indicators is complex and requires detailed knowledge of these properties as a function of solvent concentration, CO2 loading and temperature. Online measurement techniques for gas streams include FT-IR spectroscopy [107], non-dispersive infrared spectroscopy, chemiluminescence (e.g. for NO), paramagnetic (e.g. for O2), and micro-gas chromatography [108]. For more detailed analysis of absorbent trace contaminants or degradation products, more sophisticated analysis measurements may be required [109]. Where manual sampling is required for subsequent laboratory analysis, special techniques must be developed when liquid droplets or particulates are present in the gas streams flow. • On-line gas analysis. This is the routine determination of the capture rate using IR or similar devices on the incoming and outgoing gas streams. These streams must be completely dry to avoid damaging the instruments but can be complemented by in-line humidity measurements as discussed below. Condensables are either removed prior to the instrument, or the inlet lines and the measuring device is maintained at high temperature to avoid condensation. More sophisticated analyses to detect solvent loss, contaminants (SOx, NOx, O2),

58


degradation products (especially ammonia, formic acid) are also possible and increasingly utilized. • On-line liquid analysis. The ideal situation is a direct measure of the CO2 loading, solvent composition (to determine solvent loss, water balance) and the presence of degradation products. In practice, these techniques are still under development and require considerable calibration effort for each solvent composition and loading. The National Carbon Capture Centre in the U.S.A. uses online automatic titration for analysis of absorbent concentration and CO2 loading. The following physical property measurements can be readily made on-line: • pH—provides an indication of CO2 loading as the solution will become less alkaline as CO2 is taken up by the capture solvent. pH is not a strong function of temperature or solvent concentration. • Density—online density measurements are possible using Coriolis meters (usually in conjunction with flow measurements). Correlations between density and liquid CO2 loading have been developed by Bui et al. [110] and Freeman [111]. • Humidity can be readily determined in gas inlet and outlet lines but is usually close to 100%. In this case it is possible that more water is present than detected, having condensed on the cooler pipe and vessel walls. • Viscosity can be determined directly using variants of the Coriolis meter, although a pressure drop and flow measurement in any section of line or across a restriction is an adequate measure of a change in viscosity and will provide a strong indication of solvent concentration change (water balance issues) if loading (capture rate) and temperature are relatively constant. • Conductivity is a direct measure of the ionic species present and can be tied to other measurements to indicate performance trends.

6.6

Analytical Requirements

Offline measurements to determine solvent and CO2 concentration include: • Total alkalinity via standard acid/base titration [107], • Specific absorbent concentration via gas or liquid chromatography [112]. • CO2 concentration is commonly determined by pH titration, or via the barium chloride method [112]. In addition to the online measurements recorded at post combustion capture plant, a number of off-line measurements will likely be required. Whilst work is progressing towards online determination of absorbent concentration and acid gas loading, as mentioned above, these techniques are complex and still under development. As such, absorbent samples may still need to be collected for analysis


59

offline of parameters such as absorbent concentration, acid gas loading, and extent of degradation. • Absorbent concentration. One of the key measurement requirements will be the absorbent concentration. Absorbent concentration can vary through losses (aerosol and evaporative), degradation, and lack of a water balance in the process. It is desirable to maintain absorbent concentration at its target level. Viscosity and corrosion will often increase with an increase in absorbent concentration, whilst energy requirements will increase with a decrease in concentration. A rough approximation of absorbent concentration can be made via standard acid/base titration [107]. However, as number of degradation products are also alkaline in nature, this method can over-estimate the actual absorbent concentration. More accurate concentration measurements can be made for the specific absorbent via gas or liquid chromatography [112]. Chromatographic measurements however usually require sophisticated analytical equipment. • Acid gas loading. Knowledge of the rich and lean acid gas loading of the absorbent can be used to verify absorption rates, identify the operating region of the process, and ensure high corrosion conditions are avoided. Absorbent CO2 concentration is commonly determined by pH titration, or via the barium chloride method [112]. • Extent of degradation. As mentioned previously, absorbents used for CO2 removal will degrade over time. Some of these degradation products are corrosive, and so there is generally a recommended maximum concentration for heat stable salts in aqueous amines used for CO2 removal. In addition, an absorbent that suffers from high degradation rates may prove a costly absorbent choice due to high absorbent make-up rate requirements. The formation of degradation products can be estimated via the loss in absorbent concentration. This will not be completely accurate as absorbent concentration can also be lost through other mechanisms (e.g. evaporative losses). Total alkalinity is not a good method to use for determining absorbent concentration for this purpose as some degradation products will also be alkaline in nature. Reduction in specific absorbent concentration (e.g. via high pressure liquid chromatographic (HPLC) determination) however can be used to provide this estimate. Where degradation products are known, individual analysis can also be done e.g. HSS determination via anion Ion Chromatography (IC) [113].

7 Conclusions This chapter has discussed the process implications of selecting a particular energy efficient CO2 capture solvent system and has explained how the unique physical and chemical properties of the selected solvent impact on design choices, costs and operability.

60


Individual solvent properties, including reaction kinetics, thermodynamics and physical properties have been outlined and their impact on design discussed. Aspects such as degradation products, corrosivity and environmental considerations impact the selection of process equipment and materials of construction which has a further impact on cost. The impact of these properties of energy efficient solvents on individual unit operations have also been discussed in some detail. Regardless of the nature of the advanced energy efficient solvent system, these fundamental aspects of design must always be taken into consideration. As all capture solvent systems have advantages and challenges, the goal of process design is to make the best use of the advantages and mitigate the impact of the challenges. The thorough evaluation of a particular solvent system should be carried out in a process design that has been optimized for that solvent. While this is not always possible, process modelling tools allow the process design impacts to be taken into account and scaled to a commercial scale design and costing.

References 1. Kohl AL, Nielsen R (1997) Gas purification, 5th edn. Gulf Publishing 2. Sander MT, Mariz CL (1992) The Fluor Daniel Econamine FG process—past experience and present-day focus. In: 1st International Conference on Carbon Dioxide Removal (ICCDR), Amsterdam, Netherlands, 04–06 Mar 1992, pp 341–348 3. Budzianowski WM (2015) Single solvents, solvent blends, and advanced solvent systems in CO2 capture by absorption: a review. Int J Glob Warm 7(2):184–225. Doi:10.1504/ijgw. 2015.067749 4. Ramezan M, Skone TJ, Nsakala NY, Liljedahl GN (2007) Carbon dioxide capture from existing coal-fired power plants. National Energy Technology Laboratory 5. Cottrell AJ, McGregor JM, Jansen J, Artanto Y, Dave N, Morgan S, Pearson P, Attalla MI, Wardhaugh L, Yu H, Allport A, Feron PHM (2009) Post-combustion capture R&D and pilot plant operation in Australia. In: Gale J, Herzog H, Braitsch J (eds) Greenhouse Gas Control Technologies 9, vol 1. Energy Procedia. pp 1003–1010. Doi:10.1016/j.egypro.2009.01.133 6. Cousins A, Wardhaugh LT, Feron PHM (2011) A survey of process flow sheet modifications for energy efficient CO2 capture from flue gases using chemical absorption. Int J Greenhouse Gas Control 5:605–619. Doi:10.1016/j.ijggc.2011.01.002 7. Cousins A, Cottrell A, Lawson A, Huang S, Feron PHM (2012) Model verification and evaluation of the rich-split process modification at an Australian-based post combustion CO2 capture pilot plant. Greenh Gases 2(5):329–345 8. Puxty G, Rowland R, Allport A, Yang Q, Bown M, Burns R, Maeder M, Attalla M (2009) Carbon dioxide postcombustion capture: a novel screening study of the carbon dioxide absorption performance of 76 amines. Environ Sci Technol 43(16):6427–6433 9. Plaza JM, Van Wagener D, Rochelle GT (2009) Modeling CO2 capture with aqueous monoethanolamine. In: Gale J, Herzog H, Braitsch J (eds) Greenhouse Gas Control Technologies 9, vol 1. Energy Procedia. pp 1171–1178. Doi:10.1016/j.egypro.2009.01.154 10. Zhang Y, Chen H, Chen CC, Plaza JM, Dugas R, Rochelle GT (2009) Rate-based process modeling study of CO2 capture with aqueous monoethanolamine solution. Ind Eng Chem Res 48(20):9233–9246. Doi:10.1021/ie900068k 11. Freguia S, Rochelle GT (2003) Modeling of CO2 capture by aqueous monoethanolamine. AIChE J 49(7):1676–1686


61

12. von Harbou I, Imle M, Hasse H (2014) Modeling and simulation of reactive absorption of CO2 with MEA: results for four different packings on two different scales. Chem Eng Sci 105:179–190. Doi:10.1016/j.ces.2013.11.005 13. Razi N, Svendsen HF, Bolland O (2014) Assessment of mass transfer correlations in rate-based modeling of a large-scale CO2 capture with MEA. Int J Greenhouse Gas Control 26:93–108. Doi:10.1016/j.figgc.2014.04.019 14. Dugas R, Alix P, Lemaire E, Broutin P, Rochelle G (2009) Absorber model for CO2 capture by monoethanolamine; application to CASTOR pilot results. Energy Procedia 1(1):103–107 15. Saimpert M, Puxty G, Qureshi S, Wardhaugh L, Cousins A (2013) A new rate based absorber and desorber modelling tool. Chem Eng Sci 96:10–25. Doi:10.1016/j.ces.2013.03.013 16. Yu YS, Lu HF, Wang GX, Zhang ZX, Rudolph V (2013) Characterizing the transport properties of multiamine solutions for CO2 capture by molecular dynamics simulation. J Chem Eng Data 58(6):1429–1439. Doi:10.1021/je3005547 17. Rubin ES, Short C, Booras G, Davison J, Ekstrom C, Matuszewski M, McCoy S (2013) A proposed methodology for CO2 capture and storage cost estimates. Int J Greenhouse Gas Control 17:488–503. Doi:10.1016/j.ijggc.2013.06.004 18. Dave N, Do T, Palfreyman D, Feron PHM, Xu S, Gao S, Liu L (2011) Post-combustion capture of CO2 from coal-fired power plants in China and Australia: an experience based cost comparison. In: Gale J, Hendriks C, Turkenberg W (eds) 10th international conference on Greenhouse Gas Control Technologies, vol 4. Energy Procedia, pp 1869–1877. Doi:10. 1016/j.egypro.2011.02.065 19. Peters MS, Timmerhaus KD, West RE (2003) Plant design and economics for chemical engineers, 5th edn. McGraw-Hill, Boston 20. Kierzkowska-Pawlak H, Chacuk A, Siemieniec M (2014) Reaction kinetics of CO2 in aqueous 2-(2-aminoethylamino)ethanol solutions using a stirred cell reactor. Int J Greenhouse Gas Control 24:106–114. Doi:10.1016/j.ijggc.2014.03.004 21. Conway W, Beyad Y, Maeder M, Burns R, Feron P, Puxty G (2014) CO2 absorption into aqueous solutions containing 3-piperidinemethanol: CO2 mass transfer, stopped-flow kinetics, H-1/C-13 NMR, and vapor-liquid equilibrium investigations. Ind Eng Chem Res 53(43):16715–16724. Doi:10.1021/ie503195x 22. Kunze A-K, Dojchinov G, Haritos VS, Lutze P (2015) Reactive absorption of CO2 into enzyme accelerated solvents: from laboratory to pilot scale. Appl Energy 156:676–685. Doi:10.1016/j.apenergy.2015.07.033 23. Bishnoi S, Rochelle GT (2000) Absorption of carbon dioxide into aqueous piperazine: reaction kinetics, mass transfer and solubility. Chem Eng Sci 55(22):5531–5543 24. Dang HY, Rochelle GT (2003) CO2 absorption rate and solubility in monoethanolamine/piperazine/water. Sep Sci Technol 38(2):337–357. Doi:10.1081/ss12016678 25. Jou FY, Otto FD, Mather AE (1994) Vapor-liquid-equilibrium of carbon-dioxide in aqueous mixtures of monoethanolamine and methyldiethanolamine. Ind Eng Chem Res 33(8):2002– 2005 26. Kumar PS, Hogendoorn JA, Feron PHM, Versteeg GF (2003) Equilibrium solubility of CO2 in aqueous potassium taurate solutions: Part 1. Crystallization in carbon dioxide loaded aqueous salt solutions of amino acids. Ind Eng Chem Res 42(12):2832–2840. Doi:10.1021/ ie0206002 27. Fan GJ, Wee AGH, Idem R, Tontiwachwuthikul P (2009) NMR studies of amine species in MEA-CO2-H2O system: modification of the model of vapor-liquid equilibrium (VLE). Ind Eng Chem Res 48(5):2717–2720. Doi:10.1021/ie8015895 28. Park SJ, Shin HY, Min BM, Cho A, Lee JS (2009) Vapor-liquid equilibria of water plus monoethanolamine system. Korean J Chem Eng 26(1):189–192 29. Puxty G, Rowland R (2011) Modeling CO2 mass transfer in amine mixtures: PZ-AMP and PZ-MDEA. Environ Sci Technol 45(6):2398–2405. Doi:10.1021/es1022784

62


30. Li H, Le Moullec Y, Lu JH, Chen J, Marcos JCV, Chen GF, Chopin F (2015) CO2 solubility measurement and thermodynamic modeling for 1-methylpiperazine/water/CO2. Fluid Phase Equilib 394:118–128. Doi:10.1016/j.fluid.2015.03.021 31. Freeman SA, Dugas R, Van Wagener D, Nguyen T, Rochelle GT (2009) Carbon dioxide capture with concentrated, aqueous piperazine. 10.1016/j.egypro.2009.01.195. In: Gale J, Herzog H, Braitsch J (eds) Greenhouse Gas Control Technologies 9, vol 1. Energy Procedia, pp 1489–1496 32. Cousins A, Nielsen P, Huang S, Cottrell A, Chen E, Rochelle GT, Feron PHM (2015) Pilot-scale evaluation of concentrated piperazine for CO2 capture at an Australian coal-fired power station: duration experiments. Greenhouse Gases 5(4):363–373. Doi:10.1002/ghg.1507 33. Moser P, Schmidt S, Stahl K, Vorberg G, Lozano GA, Stoffregen T, Roesler F (2014) Demonstrating emission reduction—results from the post-combustion capture pilot plant at Niederaussem. In: Dixon T, Herzog H, Twinning S (eds) 12th international Conference on Greenhouse Gas Control Technologies, Ghgt-12, vol 63. Energy Procedia. pp 902–910. Doi:10.1016/j.egypro.2014.11.100 34. Pinto DDD, Knuutila H, Fytianos G, Haugen G, Mejdell T, Svendsen HF (2014) CO2 post combustion capture with a phase change solvent. Pilot plant campaign. Int J Greenhouse Gas Control 31:153–164. Doi:10.1016/j.ijggc.2014.10.007 35. Song D, Seibert AF, Rochelle GT (2014) Effect of liquid viscosity on the liquid phase mass transfer coefficient of packing. In: Dixon T, Herzog H, Twinning S (eds) 12th international conference on Greenhouse Gas Control Technologies, Ghgt-12, vol 63. Energy Procedia. Elsevier Science Bv, Amsterdam, pp 1268–1286. Doi:10.1016/j.egypro.2014.11.136 36. Freeman SA, Rochelle GT (2011) Density and viscosity of aqueous (piperazine plus carbon dioxide) solutions. J Chem Eng Data 56(3):574–581. Doi:10.1021/je1012263 37. Amundsen TG, Oi LE, Eimer DA (2009) Density and viscosity of monoethanolamine plus water plus carbon dioxide from (25 to 80) degrees C. J Chem Eng Data 54(11):3096–3100. Doi:10.1021/je900188m 38. Kumar PS, Hogendoorn JA, Feron PHM, Versteeg GF (2001) Density, viscosity, solubility, and diffusivity of N2O in aqueous amino acid salt solutions. J Chem Eng Data 46(6):1357– 1361 39. Shokouhi M, Jalili AH, Samani F, Hosseini-Jenab M (2015) Experimental investigation of the density and viscosity of CO2-loaded aqueous alkanolamine solutions. Fluid Phase Equilib 404:96–108. Doi:10.1016/j.fluid.2015.06.034 40. Weiland RH, Dingman JC, Cronin DB, Browning GJ (1998) Density and viscosity of some partially carbonated aqueous alkanolamine solutions and their blends. J Chem Eng Data 43 (3):378–382 41. Chang R (1998) Chemistry, 6th edn. McGraw Hill, New York 42. Tsai RE, Schultheiss P, Kettner A, Lewis JC, Seibert AF, Eldridge RB, Rochelle GT (2008) Influence of surface tension on effective packing area. Ind Eng Chem Res 47(4):1253–1260. Doi:10.1021/ie0707801 43. Sedelies R, Steiff A, Weinspach P-M (1987) Mass transfer area in different gas-liquid reactors as a function of liquid properties. Chem Eng Technol 10:1–15 44. Radgen P, Rode H, Reddy S, Yonkoski J (2014) Lessons learned from the operation of a 70 tonne per day post combustion pilot plant at the coal fired power plant in Wilhelmshaven, Germany. In: Dixon T, Herzog H, Twinning S (eds) 12th international conference on Greenhouse Gas Control Technologies, Ghgt-12, vol 63. Energy Procedia. Elsevier Science Bv, Amsterdam, pp 1585–1594. Doi:10.1016/j.egypro.2014.11.168 45. Reitenbach G (2015) SaskPower’s Boundary Dam Carbon Capture Project Wins POWER’s Highest Award. Power 159(8):24 46. Jackson P, Attalla MI (2010) N-Nitrosopiperazines form at high pH in post-combustion capture solutions containing piperazine: a low-energy collisional behaviour study. Rapid Commun Mass Spectrom 24(24):3567–3577. Doi:10.1002/rcm.4815


63

47. Knuutila H, Svendsen HF, Asif N (2014) Decomposition of nitrosamines in aqueous monoethanolamine (MEA) and diethanolamine (DEA) solutions with UV-radiation. Int J Greenhouse Gas Control 31:182–191. Doi:10.1016/j.ijggc.2014.10.002 48. Fine NA, Rochelle GT (2013) Thermal decomposition of n-nitrosopiperazine. In: Dixon T, Yamaji K (eds) Ghgt-11, vol 37. Energy Procedia. Elsevier Science Bv, Amsterdam, pp 1678–1686. Doi:10.1016/j.egypro.2013.06.043 49. Pearson P, Cousins A (2015) Assessment of corrosion in amine-based post-combustion capture of carbon dioxide systems. In: Feron P (ed) Absorption-based post-combustion capture of Carbon Dioxide, in press edn. Elsevier (in press) 50. GPSA (2004) GPSA engineering data book. Tulsa, Oklahoma 51. Dupart MS, Bacon TR, Edwards DJ (1993) Understanding corrosion in alkanolamine gas treating plants. 1. Proper mechanism diagnosis optimizes amine operations. Hydrocarbon Process 72(4):75–80 52. Dupart MS, Bacon TR, Edwards DJ (1993) Understanding corrosion in alkanolamine gas treating plants. Hydrocarbon Process 72(5):89–94 53. Johnson DW, Reddy S, Brown JH (2009) Commercially available CO2 capture technology. Power 153(8):58–60 54. Kittel J, Gonzalez S (2014) Corrosion in CO2 post-combustion capture with alkanolamines— a review. Oil Gas Sci Technol 69(5):915–929. Doi:10.2516/ogst/2013161 55. Bello A, Idem RO (2006) Comprehensive study of the kinetics of the oxidative degradation of CO2 loaded and concentrated aqueous monoethanolamine (MEA) with and without sodium metavanadate during CO2 absorption from flue gases. Ind Eng Chem Res 45 (8):2569–2579. Doi:10.1021/ie050562x 56. de Koeijer G, Enge Y, Sanden K, Graff OF, Falk-Pedersen O, Amundsen T, Overa S (2011) CO2 Technology Centre Mongstad—design, functionality and emissions of the amine plant. In: Gale J, Hendriks C, Turkenberg W (eds) 10th international conference on Greenhouse Gas Control Technologies, vol 4. Energy Procedia, pp 1207–1213. Doi:10.1016/j.egypro. 2011.01.175 57. Moser P, Schmidt S, Uerlings R, Sieder G, Titz J-T, Hahn A, Stoffregen T (2011) Material testing for future commercial post-combustion capture plants—results of the testing programme conducted at the Niederaussem pilot plant. In: Gale J, Hendriks C, Turkenberg W (eds) 10th international conference on Greenhouse Gas Control Technologies, vol 4. Energy Procedia, pp 1317–1322. Doi:10.1016/j.egypro.2011.01.189 58. Cousins A, Ilyushechkin A, Pearson P, Cottrell A, Huang S, Feron PHM (2013) Corrosion coupon evaluation under pilot-scale CO2 capture conditions at an Australian coal-fired power station. Greenh Gases 3(3):169–184. Doi:10.1002/ghg.1341 59. Kittel J, Idem R, Gelowitz D, Tontiwachwuthikul P, Parrain G, Bonneau A (2009) Corrosion in MEA units for CO(2) capture: pilot plant studies. In: Gale J, Herzog H, Braitsch J (eds) Greenhouse Gas Control Technologies 9, vol 1. Energy Procedia, vol 1, pp 791–797. Doi:10.1016/j.egypro.2009.01.105 60. Monea M (2013) Boundary Dam—progress and status. Paper presented at the 7th Trondheim CCS conference, Trondheim 4–6 June 2013 61. Weiland RH, Dingman JC, Cronin DB (1997) Heat capacity of aqueous monoethanolamine, diethanolamine, N-methyldiethanolamine, and N-methyldiethanolamine-based blends with carbon dioxide. J Chem Eng Data 42(5):1004–1006 62. Shaikh MS, Shariff AM, Bustam MA, Murshid G (2014) Physicochemical properties of aqueous solutions of sodium l-prolinate as an absorbent for CO2 removal. J Chem Eng Data 59(2):362–368. Doi:10.1021/je400830w 63. Aroonwilas A, Veawab A (2007) Heat recovery gas absorption process 64. Leites IL, Sama DA, Lior N (2003) The theory and practice of energy saving in the chemical industry: some methods for reducing thermodynamic irreversibility in chemical technology processes. Energy 28(1):55–97

64


65. Cousins A, Wardhaugh LT, Feron PHM (2011) Preliminary analysis of process flow sheet modifications for energy efficient CO2 capture from flue gases using chemical absorption. Chem Eng Res Des 89(8A):1237–1251. Doi:10.1016/j.cherd.2011.02.008 66. Oyenekan BA, Rochelle GT (2007) Alternative stripper configurations for CO2 capture by aqueous amines. AIChE J 53(12):3144–3154. Doi:10.1002/aic.11316 67. Freeman SA, Dugas R, Van Wagener DH, Nguyen T, Rochelle GT (2010) Carbon dioxide capture with concentrated, aqueous piperazine. Int J Greenhouse Gas Control 4(2):119–124. Doi:10.1016/j.ijggc.2009.10.008 68. Cousins A, Huang S, Cottrell A, Feron PHM, Chen E, Rochelle GT (2015) Pilot-scale parametric evaluation of concentrated piperazine for CO2 capture at an Australian coal-fired power station. Greenhouse Gases 5(1):7–16. Doi:10.1002/ghg.1462 69. Kuntz J, Aroonwilas A (2008) Performance of spray column for CO2 capture application. Ind Eng Chem Res 47(1):145–153. Doi:10.1021/ie0617021 70. Seyboth O, Zimmermann S, Heidel B, Scheffknecht G (2014) Development of a spray scrubbing process for post combustion CO2 capture with amine based solvents. In: Dixon T, Herzog H, Twinning S (eds) 12th international conference on Greenhouse Gas Control Technologies, Ghgt-12, vol 63. Energy Procedia. Elsevier Science Bv, Amsterdam, pp 1667–1677. Doi:10.1016/j.egypro.2014.11.176 71. Kavoshi L, Rahimi A, Hatamipour MS (2015) CFD modeling and experimental study of carbon dioxide removal in a lab-scale spray dryer. Chem Eng Res Des 98:157–167. Doi:10. 1016/j.cherd.2015.04.023 72. Awtry AR, Brasseur JK, Henshaw TL, Hobbs KR, McDermott WE, Miller NJ, Neumann DK, Nizamov BR, Tobias JA, Anderson JL, Courtright JL (2009) Gas liquid contactor module, useful for e.g. removing gas phase molecules, comprises liquid inlet, gas inlet, gas outlet, array of nozzles in communication with the liquid inlet and gas inlet, gas liquid separator and liquid outlet. CN102217152-B 73. Awtry A (2013) Status of the Carbon Dioxide Absorber Retrofit Equipment (CARE) program. Paper presented at the 2013 CO2 Capture Technical Meeting 74. Wardhaugh L (2010) Toward significant capital cost reduction in post combustion capture plants. Paper presented at the Chemeca 2010, Adelaide, South Australia, 26–29 Sept 2010 75. Ramshaw C (1983) HIGEE Distillation—an example of process intensification. Chem Eng Lond 389:13–14 76. Cheng HH, Tan CS (2009) Carbon dioxide capture by blended alkanolamines in rotating packed bed. In: Gale J, Herzog H, Braitsch J (eds) Greenhouse Gas Control Technologies 9, vol 1. Energy Procedia, pp 925–932. Doi:10.1016/j.egypro.2009.01.123 77. Yu CH, Wu TW, Tan CS (2013) CO2 capture by piperazine mixed with non-aqueous solvent diethylene glycol in a rotating packed bed. Int J Greenhouse Gas Control 19:503–509. Doi:10.1016/j.ijggc.2013.10.014 78. Jassim MS, Rochelle G, Eimer D, Ramshaw C (2007) Carbon dioxide absorption and desorption in aqueous monoethanolamine solutions in a rotating packed bed. Ind Eng Chem Res 46(9):2823–2833. Doi:10.1021/ie051104r 79. Zhang L-L, Wang J-X, Liu Z-P, Lu Y, Chu G-W, Wang W-C, Chen J-F (2013) Efficient capture of carbon dioxide with novel mass-transfer intensification device using ionic liquids. AIChE J 59(8):2957–2965. Doi:10.1002/aic.14072 80. Makarytchev SV, Langrish TAG, Fletcher DF (2005) Exploration of spinning cone column capacity and mass transfer performance using CFD. Chem Eng Res Des 83(A12):1372– 1380. Doi:10.1205/cherd.03413 81. Wang YD, Fei WY, Sun JH, Wan YK (2002) Hydrodynamics and mass transfer performance of a modified rotating disc contactor (MRDC). Chem Eng Res Des 80(4):392– 400 82. Wardhaugh L, Allport A, Garland E, Solnordal C (2013) A novel gas-liquid contactor for PCC based on a rotating liquid sheet. Paper presented at the The 7th Trondheim CCS Conference (TCCS-7): CO2 Capture, Transport and Storage, Trondheim, Norway, 4–6 June 2013


65

83. Knuutila H, Aronu UE, Kvamsdal HM, Chikukwa A (2011) Post combustion CO2 capture with an amino acid salt. In: Gale J, Hendriks C, Turkenberg W (eds) 10th international conference on Greenhouse Gas Control Technologies, vol 4. Energy Procedia, pp 1550– 1557. Doi:10.1016/j.egypro.2011.02.024 84. Eide-Haugmo I, Brakstad OG, Hoff KA, Sorheim KR, da Silva EF, Svendsen HF (2009) Environmental impact of amines. In: Gale J, Herzog H, Braitsch J (eds) Greenhouse Gas Control Technologies 9, vol 1. Energy Procedia, vol 1, pp 1297–1304. Doi:10.1016/j. egypro.2009.01.170 85. Cousins A, Nielsen PT, Huang S, Rowland R, Edwards B, Cottrell A, Chen E, Rochelle GT, Feron PHM (2015) Pilot-scale evaluation of concentrated piperazine for CO2 capture at an Australian coal-fired power station: Nitrosamine measurements. Int J Greenhouse Gas Control 37:256–263. Doi:10.1016/j.ijggc.2015.03.007 86. Dai N, Shah AD, Hu L, Plewa MJ, McKague B, Mitch WA (2012) Measurement of nitrosamine and nitramine formation from NO reactions with amines during amine-based carbon dioxide capture for postcombustion carbon sequestration. Environ Sci Technol 46 (17):9793–9801. Doi:10.1021/es301867b 87. Azzi M, Tibbett A, Halliburton B, Element A, Artanto Y, Meuleman E, Feron P (2014) Assessing atmospheric emissions from amine-based CO2 post-combustion capture processes and their impacts on the environment—a case study, vol 1. Measurement of emissions from a monoethanolamine-based post-combustion CO2 capture pilot plant. 1 88. Azzi M, Angove D, Campbell I, Cope M, Emmerson K, Feron P, Patterson M, Tibbett A, White S (2014) Assessing atmospheric emissions from an amine-based CO2 post-combustion capture processes and their impacts on the environment—a case study, vol 2. Atmospheric chemistry of MEA and 3D air quality modelling of emissions from the Loy Yang PCC plant, Final report. Global Carbon Capture and Storage Institute 89. Nurrokhmah L, Mezher T, Abu-Zahra MRM (2013) Evaluation of handling and reuse approaches for the waste generated from MEA-based CO2 capture with the consideration of regulations in the UAE. Environ Sci Technol 47(23):13644–13651. Doi:10.1021/es4027198 90. Sexton A, Dombrowski K, Nielsen P, Rochelle G, Fisher K, Youngerman J, Chen E, Singh P, Davison J (2014) Evaluation of reclaimer sludge disposal from post-combustion CO2 capture. In: Dixon T, Herzog H, Twinning S (eds) 12th international conference on Greenhouse Gas Control Technologies, Ghgt-12, vol 63. Energy Procedia, pp 926–939. Doi:10.1016/j.egypro.2014.11.102 91. Rochelle G, Chen E, Freeman S, Van Wagener D, Xu Q, Voice A (2011) Aqueous piperazine as the new standard for CO2 capture technology. Chem Eng J 171(3):725–733. Doi:10.1016/j.cej.2011.02.011 92. Yang N, Yu H, Li LC, Xu DY, Han WF, Feron P (2014) Aqueous ammonia (NH3) based post combustion CO2 capture: a review. Oil Gas Sci Technol 69(5):931–945. Doi:10.2516/ ogst/2013160 93. Darde V, Thomsen K, van Well WJM, Stenby EH (2009) Chilled ammonia process for CO2 capture. In: Gale J, Herzog H, Braitsch J (eds) Greenhouse Gas Control Technologies 9, vol 1. Energy Procedia, pp 1035–1042. Doi:10.1016/j.egypro.2009.01.137 94. Telikapalli V, Kozak F, Francois J, Sherrick B, Black J, Muraskin D, Cage M, Hammond M, Spitznogle G (2011) CCS with the Alstom chilled ammonia process development program— field pilot results. In: Gale J, Hendriks C, Turkenberg W (eds) 10th international conference on Greenhouse Gas Control Technologies, vol 4. Energy Procedia, Elsevier Science Bv, Amsterdam, pp 273–281. Doi:10.1016/j.egypro.2011.01.052 95. Yu H, Qi G, Xiang Q, Wang S, Fang M, Yang Q, Wardhaugh L, Feron P (2013) Aqueous ammonia based post combustion capture: results from pilot plant operation, challenges and further opportunities. Paper presented at the the 11th Greenhouse Gas Control Technologies (GHGT) conference, Kyoto International Conference Center, Japan, 18–22 Nov 2012 96. Aleixo M, Prigent M, Gibert A, Porcheron F, Mokbel I, Jose J, Jacquin M (2011) Physical and chemical properties of DMX (TM) solvents. In: Gale J, Hendriks C, Turkenberg W

66

97.

98.

99.

100.

101.

102.

103. 104.

105.

106.

107.

108.

109. 110.

L.T. Wardhaugh and A. Cousins (eds) 10th international conference on Greenhouse Gas Control Technologies, vol 4. Energy Procedia. Elsevier Science Bv, Amsterdam, pp 148–155. Doi:10.1016/j.egypro.2011.01.035 Raynal L, Alix P, Bouillon PA, Gomez A, de Nailly ML, Jacquin M, Kittel J, di Lella A, Mougin P, Trapy J (2011) The DMX (TM) process: an original solution for lowering the cost of post-combustion carbon capture. In: Gale J, Hendriks C, Turkenberg W (eds) 10th international conference on Greenhouse Gas Control Technologies, vol 4. Energy Procedia. Elsevier Science Bv, Amsterdam, pp 779–786. Doi:10.1016/j.egypro.2011.01.119 Misiak K, Sanchez CS, van Os P, Goetheer E (2013) Next generation post-combustion capture: Combined CO2 and SO2 removal. In: Dixon T, Yamaji K (eds) Ghgt-11, vol 37. Energy Procedia. Elsevier Science Bv, Amsterdam, pp 1150–1159. Doi:10.1016/j.egypro. 2013.05.212 Cousins A, Artanto Y, Meuleman E, Pearson P, Puxty G, Jansen J, Conway W, Slater N, Curtis E, Monch A, Feron P, Verheyen V, Misiak K, Huizinga A, Sanchez Sanchez C, van Os P, Goetheer E, Castelow P (2014) Combined low-cost pre-treatment of flue gas and capture of CO2 from brown coal-fired power stations. Paper presented at the Low Rank Coal, Third International Industry Symposium, Melbourne, Australia, 28 Apr–1 May 2014 Yu H, Qi G, Wang S, Morgan S, Allport A, Cottrell A, Do T, McGregor J, Wardhaugh L, Feron P (2012) Results from trialling aqueous ammonia-based post-combustion capture in a pilot plant at Munmorah Power Station: gas purity and solid precipitation in the stripper. Int J Greenhouse Gas Control 10:15–25. Doi:10.1016/j.ijggc.2012.04.014 Puxty G, Wei SC-C, Feron P, Meuleman E, Beyad Y, Burns R, Maeder M (2014) A novel process concept for the capture of CO2 and SO2 using a single solvent and column. In: Dixon T, Herzog H, Twinning S (eds) 12th international conference on Greenhouse Gas Control Technologies, Ghgt-12, vol 63. Energy Procedia, pp 703–714. Doi:10.1016/j. egypro.2014.11.078 McLarnon CR, Duncan JL (2009) Testing of ammonia based CO(2) capture with multi-pollutant control technology. In: Gale J, Herzog H, Braitsch J (eds) Greenhouse Gas Control Technologies 9, vol 1. Energy Procedia, vol 1, pp 1027–1034. Doi:10.1016/j. egypro.2009.01.136 Manning FS, Thompson RE (1991) Oilfield processing of petroleum, vol 1: Natural gas, vol 1. Penwell Publishing Company, Tulsa Campbell KLS, Lapidot T, Williams DR (2015) Foaming of CO2-loaded amine solvents degraded thermally under stripper conditions. Ind Eng Chem Res 54(31):7751–7755. Doi:10.1021/acs.iecr.5b01935 Chen X, Freeman SA, Rochelle GT (2010) Foaming of aqueous piperazine and monoethanolamine for CO2 capture. Int J Greenhouse Gas Control 5(2):381–386. Doi:10. 1016/j.ijggc.2010.09.006 Richner G, Puxty G (2012) Assessing the chemical speciation during CO2 absorption by aqueous amines using in situ FTIR. Ind Eng Chem Res 51(44):14317–14324. Doi:10.1021/ ie302056f Artanto Y, Jansen J, Pearson P, Do T, Cottrell A, Meuleman E, Feron P (2012) Performance of MEA and amine-blends in the CSIRO PCC pilot plant at Loy Yang Power in Australia. Fuel 101:264–275. Doi:10.1016/j.fuel.2012.02.023 Azzi M, Day S, French S, Halliburton B, Jackson P, Lavrencic S, Riley K, Tibbett A (2010) CO2 Capture Mongstadt—Project A—Establishing sampling and analytical procedures for potentially harmful components from post-combustion amine based CO2 capture. Task 2: Procedures for manual sampling. EP 105456, CSIRO, Australia Gassnova (2015) http://www.gassnova.no/en/ccs-projects/full-scale-mongstad. Accessed July 2015 Bui M, Gunawan I, Verheyen TV, Meuleman E, Feron P (2014) Dynamic operation of post-combustion CO2 capture in Australian coal-fired power plants. In: Dixon T, Herzog H, Twinning S (eds) 12th international conference on Greenhouse Gas Control Technologies, Ghgt-12, vol 63. Energy Procedia, pp 1368–1375. Doi:10.1016/j.egypro.2014.11.146


67

111. Freeman SA (2011) Thermal degradation and oxidation of aqueous piperazine for carbon dioxide capture. Ph.D. thesis, University of Texas at Austin 112. Notz R, Mangalapally HP, Hasse H (2012) Post combustion CO2 capture by reactive absorption: Pilot plant description and results of systematic studies with MEA. Int J Greenhouse Gas Control 6:84–112. Doi:10.1016/j.ijggc.2011.11.004 113. Nielsen PT, Li L, Rochelle GT (2013) Piperazine degradation in pilot plants. In: Dixon T, Yamaji K (eds) Ghgt-11, vol 37. Energy Procedia, pp 1912–1923. Doi:10.1016/j.egypro. 2013.06.072

Useful Mechanisms, Energy Efficiency Benefits, and Challenges of Emerging Innovative Advanced Solvent Based Capture Processes Wojciech M. Budzianowski

Abstract Innovative advanced solvent based capture processes (ASBCPs) employing sophisticated separation mechanisms have emerged as essential solutions that may dramatically reduce high energy requirement of CO2 separation in capture plants. This study systematically characterises useful mechanisms of several such ASBCPs, being mostly at relatively low technology readiness levels. Based on improved understanding of these emerging ASBCPs it is shown how they can contribute to achieving energy efficiency benefits and thus increase CO2 capture efficiency. It is followed by an analysis of practical examples of all ASBCPs recently presented in academic and patent literature. Finally, major challenges of discussed emerging ASBCPs are identified showing potential directions for further research and development. Overall, innovative ASBCPs employ very different mechanisms translating to different energy efficiency benefits in various CO2 capture applications. The provided technological account along with the identified challenges may be useful for capture process developers which will be able to bring the most promising ASBCPs to higher technology readiness levels and finally they can be employed in the practice. Nomenclature 5OCB 7OCB AA AAS ASBCP BDA BTCA CA

4,4′-pentyloxy cyanobiphenyl 4,4′-heptyloxy cyanobiphenyl Aqueous ammonia Amino acid salt Advanced solvent based capture process 1,4-butanediamine Benzotriazole-5-carboxylic acid Carbonic anhydrase

W.M. Budzianowski (&) Consulting Services, Wrocław, Poland e-mail: [email protected] W.M. Budzianowski Renewable Energy and Sustainable Development (RESD) Group, Wrocław, Poland © Springer International Publishing AG 2017 W.M. Budzianowski (ed.), Energy Efficient Solvents for CO2 Capture by Gas–Liquid Absorption, Green Energy and Technology, DOI 10.1007/978-3-319-47262-1_4

69

70

CAP CAPEX CO2BOL DEEA DES DMCA DMF DPA DsBA EPD GAP-0 HA IL IPADM-2BOL MAPA MEA MOF OPEX PC PCH3 PCH7 TETA TMG TRL TSIL %wt

W.M. Budzianowski

Chilled ammonia process Capital expenditure CO2 binding organic liquid 2-(diethylamino) ethanol Deep eutectic solvent N,N-dimethylcyclohexylamine Dimethylformamide Dipropylamine Di-sec-butylamine N-ethyl piperidine 1,3-bis(3-aminopropyl)-1,1,3,3-tetramethyldsiloxane Hexylamine Ionic liquid 1-((1,3-dimethylimidazolidin-2-ylidene)amino)propan-2-ol 3-(methylamino)propylamine Monoethanolamine Metal organic framework Operating expenditure Pulverised coal 4,4′-propylcyclohexyl benzonitrile 4,4′-heptylcyclohexyl benzonitrile Triethylenetetramine 1,1,3,3-tetramethylguanidine Technology readiness level Task specific ionic liquid Percent by weight

1 Introduction Gas-liquid absorption is among the main technologies seriously considered for large scale CO2 capture plants, especially if they are operated in a post-combustion mode [1]. The technology has been developed in recent few years and is now becoming one of leading decarbonisation options. In particular developments of advanced solvent based capture processes (ASBCPs) create opportunities for dramatic reduction of the energy penalty associated with removing CO2 from flue gases. The potential of ASBCPs reaches however far beyond flue gases. ASBCPs may be used in various sectors from cement production through refineries to industries processing biomass and various carbonaceous materials. Due to their versatility ASBCPs have potential for decarbonisation of different CO2 intensive industries.

Useful Mechanisms, Energy Efficiency Benefits, and Challenges …

71

Innovative advanced solvent based capture processes employ various chemical and/or physical phenomena capable of enhancing a capture process [2] thus minimising energy requirements for the entire CO2 capture chain. Since reactive solvents are able to strongly bind with CO2 and thus efficiently separate it from diluted gases, gas-liquid chemical absorption is more efficient in post-combustion CO2 capture where CO2 is present in relatively low concentrations [3, 4]. CO2 capture by gas-liquid absorption requires temperatures in between solid sorbent and cryogenic captures. Such a moderate thermal regime usually alleviates capture energy penalty due to improved thermal compatibility with flue gas temperatures and thus reduced process irreversibility. It may further translate to thermodynamic benefits achieved in energy conversion cycles integrated with CO2 capture. Solid sorbents seem more suitable for pre-combustion and chemical looping applications while cryogenic capture cannot be efficiently integrated into thermal power plants [5]. Moreover, solid sorbents have limited regenerability and often degrade after a few cycles. The problem with membranes and cryogenic capture is their limited maturity for real scale applications. Some ASBCPs may be also highly environmentally benign. For example, by employing ASBCPs in which the solvent is not in direct physical contact with the capture installation and flue gases one can minimise solvent degradation, vaporisation and corrosion. Since real CO2 capture plants will process huge amounts of flue gases even minimal volatility would lead to high overall emissions. The physical separation of the solvent and infrastructure may be achieved by microencapsulation or solvent membranes. ASBCPs are also suitable for process intensification. For instance, CO2 absorption is intensified by precipitating CO2 comprising solids from the solution thus counteracting solvent saturation and maximising the absorption driving force. In addition, thermal regeneration may be also intensified e.g. by using a polarity swing phenomenon enabling CO2 release from the solvent at low temperatures. Figure 1 presents ASBCPs for CO2 capture by gas-liquid absorption investigated in this study. The emphasis within this current study is put on useful

Fig. 1 Advanced solvent based capture processes for CO2 separation by gas-liquid absorption investigated in this study

72

W.M. Budzianowski

mechanisms and energy efficiency benefits from applying ASBCPs followed by the identification of challenges that need to be addressed through future research and developments activities.

2 Precipitating Solvents 2.1

Useful Mechanisms

The useful mechanism employed by precipitating solvents relies on phase change upon contact with CO2 containing gases resulting in precipitation of solids from the solvent solution. The precipitate is usually a salt and forms in a scrubber as a CO2rich slurry of very small particles. The precipitated slurry, before it is sent to a stripper for regeneration and CO2 separation, is thickened by filtration or sedimentation. The obtained CO2-lean solvent fraction is recycled to the scrubber for continuous CO2 separation.

2.2

Energy Efficiency Benefits

By applying precipitating solvents the energy efficiency of the capture process is improved because the removal of dissolved CO2 by precipitation counteracts solvent saturation and hence maximises CO2 absorption flux. It reduces solvent flow and thus energy penalty per unit CO2 captured. In addition, the regeneration step can also be more energy efficient because the precipitate is decomposed at a higher pressure than liquid CO2-loaded solvents thereby producing pressurised CO2 stream requiring less energy for compression. Many precipitating solvents are anhydrous (e.g. GAP-0) and hence they require less energy for solvent regeneration because no water with high associated vaporisation enthalpy needs to be evaporated. In contrast to neat reactive solvents with strong CO2/solvent binding (the formed compound first needs to be evaporated and then thermally decomposed to yield the CO2 gas and the solvent which is very energy intensive [6]) most precipitating solvents may release CO2 without boiling the solvent which improves their energy efficiency.

2.3

Practical Examples

Numerous chemicals may serve as precipitating solvents but only a few have properties that may lead to developing an energy efficient capture process. A GAP-0 aminosilicone solvent is one of such candidates. It is a liquid characterised by a relatively high boiling point (265 °C) and low viscosity (4 cP at 25 °C). The


73

carbonation product is GAP-0 carbamate which relatively rapidly precipitates [7]. By increasing temperature the GAP-0 carbamate is decomposed into CO2 and the GAP-0 solvent suitable for reuse. GAP-0 is thermally stable up to 180 °C and poorly volatile. Higher operating temperatures lead to increased CO2 partial pressure and thus the cycling CO2 loading capacity is increased. A non-volatility property simplifies the CO2 stripping process (there is no need for evaporated solvent separation) thereby decreasing OPEX and CAPEX. The GAP-0 based process is able to reduce energy requirements by up to 50% compared to MEA (with concentration of 30%) [7, 8]. Since GAP-0 is less reactive towards CO2 than MEA, the energy requirements reduction is greater under concentrated CO2 gas stream conditions, i.e. the behaviour is similar as for physical vs. chemical solvents (physical solvents outperform chemical solvents at high CO2 concentrations). The reduced energy requirement also arises from the fact that the specific heat of MEA is approximately 60% greater than that of GAP-0 [3.7 vs. 2.3 kJ/(kg K)] which translates to reduced sensible heat in the stripping process involving solvent heating. Characteristics of the GAP-0 based precipitating solvent ASBCP are outlined as follows. First, CO2 is absorbed by using a spray column with little solvent. It is followed by separating the precipitated GAP-0 carbamate solids from flue gases by applying a cyclone. Then separated solids are screw conveyed from the scrubber operated at 0.1 MPa to the stripper operated at 0.5–2 MPa. Further, the GAP-0 carbamate precipitate is heated by steam injection to temperatures between 100 and 115 °C. Next, CO2 is stripped from the precipitate by applying two agitated vessels where operating pressure is reduced following from the first to the second vessel. Finally, the regenerated GAP-0 is recycled to the scrubber. Hydrous ammonia upon contact with CO2 under lower temperatures and higher concentrations precipitates ammonium salts [9] such as bicarbonate, carbonate, sesqui-carbonate, and carbamate. The precipitation is achieved in a crystalliser which separates the CO2-rich and CO2-lean solvent. Compared to hydrous ammonium without solid formation (chilled ammonia process), 40% lower mass flow rate need to be sent to the stripper due to CO2 enrichment, thus reducing the regeneration energy. The CO2-lean solvent is conveyed to bottom section of the scrubber while full regenerated solvent from the stripper to the top of the scrubber which optimises process performance [10]. One another chemical used as a precipitating solvent is an amino acid salt (AAS), e.g. taurine. The AAS solvent on contact with CO2, after reaching the solubility limit of the zwitterion, precipitates amino acids. The process benefits from increased pH (due to acid precipitation) which increases the CO2 absorption rate. It is achieved in a spray scrubber [11] because spray absorption is superior compared to packed beds due to lower pressure drops and acceptable CO2 absorption rates. The rate of absorption is not limited by stiff solvent droplets because the viscosity of AAS is low allowing for the fast renewal of droplet surface. The CO2 loaded solvent forms a slurry comprising about 15% of solids. The slurry is conveyed to a thermal stripper operated at about 75 °C where concentrated CO2 stream is obtained. The energy requirement of this ASBCP is about

74

W.M. Budzianowski

2400 kJ/kgCO2, however this number does not include low-grade heat consumed in order to dissolve the precipitated solids [12]. A DECAB Plus process employing K-TAU amino acid was claimed to reduce energy requirement by 65% compared to MEA benchmark (concentration 30%) [13]. However, for the CO2 capture applying another AAS (KSAR with concentration of 5 m) the recorded energy efficiency improvement was marginal [14]. The reason might be that not only solvent but also plant layout affects energy requirements thus an inefficient plant tends to increase overall energy requirements. Thus using energy efficiency techniques such as flash and lean vapour compression a further decrease of the energy requirement can be obtained, e.g. to 1900 kJ/kgCO2 as shown in [15]. A precipitating solvent based ASBCP may also include porous powder suspended in glycol. This system is exceptional in that it does not chemically react the CO2 and the solvent but the CO2 is only physically entrapped in the pores. Such a combined ASBCP makes use of the suitability of liquids for operation in large capture plants and potentially reduced energy requirements of solid sorbents since the system contains suspended solids. The system may consist of 2-methylimidazole glycol as a liquid and ZIF-8 (MOF) as a powder. The porosity characteristics of ZIF-8 is such that its pores of 3.4 Å are too small for entrapping the glycol (having a molecular size of 4.5 Å) while are sufficient for entrapping CO2 molecules. ZIF-8 is very well soluble in glycol and is characterised by chemical and thermal stability. CO2 can be stripped from the powder without evaporating the glycol leading to alleviated energy requirement. The whole system is therefore potentially suitable for energy efficient operation in multiple cycles. Oligomers, i.e. polymers highly soluble in solvents, which on contact with CO2 precipitate solids are another class of materials for use as precipitating solvents. Certain amino silicones in combination with anhydrous glycol have 50% higher CO2 loading capacity compared to MEA (concentration 30%) [16]. Due to precipitation and high loading capacity such an ASBCP achieves high CO2 absorption rate [17]. Nevertheless, new material candidates suitable for use in realistic capture conditions would have to be proposed or synthesised to make oligomers more important for capturing CO2. K2CO3 is a cheap and potentially promising precipitating solvent. It crystallises on contact with CO2 forming KHCO3 and subsequent crystal regeneration is obtained by applying increased pressure. The separation of KHCO3 is obtained by means of a hydrocyclone and the slurry is sent to the regenerator. The reported energy consumption is approximately 2500 kJ/kgCO2 [18]. In realistic capture applications 40 wt% K2CO3 + 10 wt% potassium glycine is applied, the latter acting as a promoter because it can enhance the CO2 regeneration rate by 6 fold [19]. The UNO MK 3 process employs a K2CO3 solvent and uses a hydrocyclone for rich solvent thickening as well as recycles exhaust gases to obtain a CO2-enriched stream. It achieves reduced energy requirement of about 3000 kJ/kgCO2 [20].


2.4

75

Challenges

The use of precipitating solvents in real capture conditions meets is however associated with several challenges. For instance the presence of non-CO2 gases such as SO2 and NOX in flue gases [21, 22] cause co- precipitation and often require different conditions for removal than CO2 thus complicating the regeneration step. If non-CO2 gases release from the precipitate they decrease CO2 stream purity. If they accumulate in the solvent, e.g. as solids, CO2 loading capacity is decreased over time. Accumulation of non-CO2 gases occurs if they bind stronger than CO2 with the solvent and cannot be fully separated in the regenerator. Another challenge typical of slurry systems is packed bed clogging by formed solids. The clogging increases column pressure drop thereby rising OPEX of the plant. The precipitate build-up on the packing may be slow but over time may fully clog the scrubber. Consequently, spray columns and wetted wall columns need to be considered because they do not clog. However, these contactors suffer from lower CO2 capture rates due to very low gas-liquid contact area and could be inefficient in large capture plants. For packed beds which usually achieve greater absorption rates new candidates for precipitating solvents would have to be implemented in order to overcome challenges associated with clogging.

3 Two Immiscible Liquid Phases 3.1

Useful Mechanisms

The property of two immiscible liquid phases may be used to create an effective ASBCP. For an effective system immiscibility of the involved liquids needs to be temperature dependent. More specifically, under scrubbing temperatures these solvents must be homogenous while under stripping temperatures must be heterogeneous. It is schematically displayed in the panel a of Fig. 2. Figure 2 in panel b displays two immiscible liquid phases with upper critical temperature which will form two phases in a scrubber if operated at lower temperature while panel c shows a solvent with two critical temperatures which would from two phases at moderate temperature. All present day two immiscible liquid phases ASBCPs are based on the mechanisms given in Fig. 2 panel a. An interesting ASBCP could be created if the immiscibility of involved liquids was CO2 concentration dependent. In this case a sufficient trigger to form two phases would be varied CO2 concentration and no heating would be need for solvent regeneration. However, again no such solvents have been found so far. In an ASBCP involving two immiscible liquid phases with lower critical temperature (Fig. 2 panel a) the two resulting liquid phases vary in terms of CO2 solubility and CO2 accumulates only in one phase. From the two phases formed (CO2-lean and CO2-rich) only the latter is conveyed to the stripper.

76

W.M. Budzianowski

Fig. 2 Phase behaviours of ASBCPs relying on two immiscible liquid phases. Panel a—lower critical temperature, panel b—upper critical temperature, panel c—upper and lower critical temperatures. Only panel a represents the temperature dependent case that has been employed in the practice for capturing CO2

For some already tested two immiscible liquid phases based ASBCPs the mechanism of their temperature dependent immiscibility is associated with the roles played by hydrophobic and hydrophilic functional groups comprised by lipophilic amines. Lipophilic amines have highly varied solubility in water which sharply decreases at higher temperatures leading to phase splitting. The known solvents split upon heating to about 80 °C [23]. The whole solvent as a single phase is used to scrub CO2. The CO2 loaded mixture is conveyed to a coalescence tank where two phases are formed. The CO2rich phase is sent to the stripper while the CO2-lean phase is recycled back to the scrubber, after mixing with the regenerated second phase.

3.2


The reason for energy efficiency benefits is that the volume of the regenerated CO2rich phase is much smaller than that of the CO2-lean phase. In anhydrous systems less energy is required for solvent heating and vaporisation (water has high sensible heat and heat of vaporisation [24]). As both involved liquids contribute to CO2 absorption in a scrubber thus mass transfer is enhanced and less solvent needs to be cycled between a scrubber and stripper. The CO2-rich phase has very high CO2 concentration and thanks to its lower volume the ASBCP requires very little solvent meaning also less required energy. The demonstration obtained in the iCAP project with a DEEA + MAPA based ASBCP revealed that the energy requirement of two immiscible liquid phases is about 2400 kJ/kgCO2 [25]. Similar results have been obtained by a DMX process showing energy consumption of about 2300 kJ/kgCO2 [26]. Reduced stripping temperature in this ASBCP facilitates the utilisation of low-grade heat for solvent recovery.


3.3

77

Practical Examples

Several liquid pairs have been proposed for capture applications. For instance, water and lipophilic amines such as HA, DMCA, DPA, DsBA, and EPD. Water forms a CO2-lean phase while lipophilic amines a CO2-rich phase. The CO2 loading capacity of water is marginal under practical capture operating conditions (it depends on pH and pressure) while for lipophilic amines it is approximately 1 kmol CO2/kmol lipophilic amine) [27]. These liquids are immiscible and CO2 concentrates in lipophilic amines which only need separation and regeneration. DEEA and MAPA liquids that form two immiscible phases were tested within the iCAP project. MAPA consisting of one primary and one secondary amine groups formed the heavier CO2-rich phase [25, 28]. DEEA being a tertiary alkanolamine formed the lighter CO2-lean phase. The DEEA + MAPA system operated at high pressure [25] which allowed to produce a pressurised CO2 stream. More recently the 5 M DEEA/2 M MAPA mixture was investigated in the Gløshaugen plant [29] achieving lower reboiler temperatures and duty (compared to MEA at 30% concentration). Xu et al. [30] explored the DEEA-BDA mixture having similar characteristics as the DEEA-MAPA but its validation under real flue gas conditions and scales are to be demonstrated. Ye et al. tested the TETA + DEEA blend (1:4) and found overall energy requirement reduction of 30% compared to MEA (concentration 30%) [31].

3.4

Challenges

Challenges include the need for unconventional solvent handling, e.g. through separation of two liquid phases in a coalescence tank. In addition, the selection of immiscible liquid pairs is complex and screening must be very detailed. No solvents exist representing temperature dependent cases displayed in Fig. 2 panels b, c. Also CO2 concentration dependent immiscibility has not been explored so far. Challenges also include solvent degradation and volatility. Especially volatility limits solvent selection opportunities and control to avoid solvent losses adds to the plant’s OPEX [29].

4 Catalysed Solvents 4.1

Useful Mechanisms

The rate of CO2 binding reaction may be increased by catalysts. A very fast natural catalyst is an enzyme carbonic anhydrase (CA). CA impacts primarily the rate of

78

W.M. Budzianowski

reaction (1)—CO2 hydration/HCO3− dehydration [32]—which is the real bottleneck of many hydrous reactive solvent based CO2 capturing processes: þ CO2 þ H2 O HCO 3 þH

ð1Þ

Reaction (1) is reactant diffusion controlled under low concentrations (catalysis is thus not helpful). However, at higher CO2 concentrations typical in CO2 capture reactants are abundantly available and CA efficiently catalyses the CO2 hydration to bicarbonate. The turnover rate of is up to 1.5 106 (mol CO2/(s mol CA) [33]. CA may be added to the solvent due to its water solubility but preferably it should be immobilised, e.g. on column packing to prevent washing to the high-temperature stripper where it may become denatured. However, if solvent regeneration is achieved at low temperatures (below 40–50 °C) CA may be employed to catalyse also the dehydration of CO2 by bicarbonate decomposition. The impact of CA catalysis is most pronounced for solvents with low reactivity such as K2CO3 while solvents with high reactivity such as MEA CA may give marginal benefits [34].

4.2


Major energy efficiency benefits arising from the use of CA are associated with increased CO2 reaction rates. This translates to compact designs and reduced solvent circulation leading to reduction in energy penalties. If CO2 stripping could be catalysed this could also potentially lead to a regeneration temperature reduction. CA may be also applied in various hybrid processes such membrane immobilisation and hence greater energy efficiency of these hybrids could be achieved.

4.3

Practical Examples

Due to varying catalytic mechanisms under different operating conditions the catalytic effect of CA depends on many parameters. For poorly reactive solvents such as K2CO3 benefits are usually significant. CA performance is improved if immobilised on the ceramic packing which minimises wash-out. CO2 absorption rate is increased by 20-fold. Typical energy requirement is 3500 kJ/kgCO2 [35]. For CA catalysed K2CO3 solvent flue gas flow rate may be increased by 7-fold compared to blank K2CO3. This reduces column size and CAPEX. In demonstrations by Akermin [35] some CA related problems have been overcome. For example, the process was tested on coal combustion derived flue gas achieving steady operation during 2800 h. Nevertheless, CA inactivation is still a problem since 54% CA deactivation has been observed after 1600 h of operation. In [36] a K2CO3 solvent promoted by a Zn complex containing a cyclic ligand of 1,4,7,10-tetracyclodode cane (Zn-cyclen) was used by applying a membrane


79

contactor. Zn-cyclen catalyst is more stable than CA, has a longer life time and its molecular weight (235 g/mol) is smaller than that of CA. This study found that in the Zn-cycle catalysed K2CO3 solvent the rate constant of CO2 hydration reaction was 10 fold increased compared to the non-catalysed solvent. Other catalysts such as oxoanions (e.g. phosphate, phosphite, arsenate, and arsenite) have been also proposed [37]. Although significant catalytic effects were noted these cheap materials require further research addressing toxicity and chemical stability of oxoanions.

4.4

Challenges

Challenges in using CA is mainly associated with the susceptibility of CA to denaturation at temperatures higher than 40–50 °C. It limits operating temperatures especially in the striper or requires CA separation from the solvent or immobilisation in the scrubber. Handling of CA through immobilisation or filtration (tested in HiPerCap [38]) adds complexity to the whole process. Besides, due to high liquid phase mass transfer resistance CA is only effective when acting precisely at the gas-liquid interface which necessitates the use of methods such as immobilisation that expose the CA for more direct gas contact [39].

5 Microencapsulated Solvents 5.1

Useful Mechanisms

In microencapsulated solvents solvent is entrapped within microcapsules. These microcapsules have highly CO2 permeable shells and the solvent is present in their cores. Figure 3 illustrates the mechanism of CO2 scrubbing and stripping by such Fig. 3 The mechanistic scheme of CO2 scrubbing and stripping involving microencapsulated solvents

80

W.M. Budzianowski

microcapsules. The scrubbing process starts with the diffusion of gaseous CO2 through the permeable shell which is followed by the chemical reaction between CO2 and the inner solvent comprising hydroxyl ions yielding bicarbonates. The stripping process involves heating of the microcapsules in order to decompose bicarbonate and the released CO2 permeates through the shell. Consequently, by physically separating scrubbing and stripping units a relatively pure CO2 stream is obtained.

5.2


Main energy efficiency benefits are associated with higher contact area meaning that less solvent is required and less pumping power is consumed. Solvent microencapsulation beneficially separates the solvent from the CO2 capture installation and flue gases which limits solvent degradation and infrastructure corrosion. Potentially less energy may be thus required in order to minimise unwanted emissions from solvent degradation and vaporisation by minimising downstream processing. Solvent vaporisation control by various additives [40–44] is thus no longer required with microencapsulated solvents minimising OPEX. The microcapsule structures enable to exploit solvents more efficiently. They make use of the loading capacity and selectivity of reactive solvents. Due to small dimensions of microcapsules the surface area is very high which enhance both scrubbing and stripping steps. Microencapsulation allows for the use of most advanced solvents including ILs (their viscosity is no longer a problem due to microencapsulation) or precipitating solvents (solids may precipitate and decompose within microcapsules [45]. This all suggests that microencapsulating may be a good solution towards developing truly energy efficient ASBCPs.

5.3

Practical Examples

Microencapsulated solvents as a relatively new ASBCP have limited demonstration. From what is available in the open literature it may be concluded that microcapsules may be produced by applying a microfluidic double-capillary device [46]. Typical microcapsules diameters are between 0.1 and 0.6 mm. Microcapsules consist of a few layers. The microcapsule core consists of the solvent (hydrous K2CO3 or Na2CO3). The rate of CO2 reaction with the solvent is increased by some 25 % by a catalyst (cyclen—Zn-1,4,7,10-tetraazacyclododecanen) placed in the microcapsule interior core. The overall increase of CO2 absorption rate is more than 5 fold compared to neat solvents due to contributions from catalysis and surface expansion. The external microcapsule layer consists of silicone material highly permeable for CO2 gas.


81

Flue gases may be passed through a fluidised bed of microcapsules to absorb CO2. CO2 loaded microcapsules are continuously removed and regenerated at an elevated temperature [47].

5.4

Challenges

Challenges associated with applying microcapsules especially in realistic capture conditions are associated with mass transfer resistance across the capsule shell. The shell may be contaminated by various constituents of flue gases reducing gas permeability. The contaminants will likely fill the pores of silicone shells reducing the lifetime of microcapsules. Microencapsulated solvents, in contrast to solvent membranes, need to be circulated between the scrubber and the stripper which may decrease lifetime of potentially fragile microcapsules. If direct contact with flue gases is considered, microcapsules need to be resistant to impurities such as SO2 or NOX.

6 Solvent Membranes 6.1

Useful Mechanisms

A solvent membrane ASBCP includes a porous membrane separating a solvent film from flue/sweep gases, Fig. 4. The porous membrane is in contact with pressurised flue gases enabling CO2 to permeate through the pores and react with the solvent in the inner solvent membrane. The CO2 loaded solvent molecules diffuse through the solvent membrane, decompose to release CO2 which passes the porous membrane and leaves the contactor as the CO2-rich permeate. The CO2 binding reaction needs to be rapid to intensify CO2 absorption rates by removing free CO2 from the porous membrane. The solvent membrane ASBCP may beneficially use volatile solvents

Fig. 4 Schematic of the solvent membrane ASBCP

82

W.M. Budzianowski

such as ammonia [48]. It is unsuitable for the use of precipitating solvents since it is difficult to deliver thermal energy to the solvent membrane in order to decompose precipitated solids.

6.2


This kind of ASBCP enables the use of more expensive solvents being highly reactive and selective towards CO2. The high reactivity and selectivity will minimise the need for pressure difference across the membrane and reduce associated energy requirements. Since the CO2 rich permeate is produced directly from the solvent membrane contactor, no additional thermal energy is needed to strip the CO2 from the loaded solvent which potentially reduces the overall energy requirement. Energy efficiency benefits are also gained by means of more compact scrubber design due to higher gas-liquid contact area [49]. Since the solvent is used as a solvent membrane virtually no solvent circulation is needed minimising associated pumping work. Since the solvent is separated from the gas phases by the porous membranes it does not vaporise and is kept clean which minimise solvent losses translating to lower life cycle energy needs associated with solvent production.

6.3

Practical Examples

Many solvent membrane ASBCPs have been investigated over recent years. Solvents implemented as CO2 capturing liquids inserted in membrane shells include amine acid salt [50], aqueous ammonia [51, 52] and ionic liquids. Some solvents achieve benefits while applied in solvent membrane contactors. For example, a shortcoming of ammonia vaporisation [53] can be overcome due to separating the solvent and flue gases by an additional membrane barrier. Ionic liquids such as [emim][Tf2 N], [emim][CF3SO3], [emim][dca] and [thtdp] [Cl] were tested and CO2 permeabilities in the range of 350–1000 barrers CO2/N2 and selectivities of 15–61 were reported [54]. The research on fluoroalkyl-functionalised imidazolium-based solvent membranes revealed CO2 permeabilities of 210–320 and selectivities of 16–27 (CO2/N2) and 13–19 (CO2/CH4) [55]. Unconventional IL-based solvent membrane tested in [56] showed selectivities 10–52 (CO2/N2), 5–13 (CO2/H2) and 5–23 (CO2/CH4). Tests on membrane systems involving two membrane supports with different hydrophobicity and imidazolium-based ILs as a liquid membrane revealed mixed-gas selectivities 20–32 (CO2/N2) and 98–200 (CO2/CH4). The hydrophobic membrane support was found to be more stable than the hydrophilic one [57]. Selectivities reported in [58] were 21.2 (CO2/N2) and 27 (CO2/CH4) for [emim][Tf2 N] and [emim][BF4] ILs, respectively. In addition amine-functionalised IL in a cross-linked Nylon 66 support were analysed [59] and selectivities greater than 15 and CO2 permeabilities ranging from 100 to 1000 barrers were obtained. With


83

IL-based solvent membranes in a Nafion matrix selectivity was 26 (CO2/CH4) [60]. Solvent membranes employing ammonium, imidazolium, pyridinium, pyrrolidinium and phosphonium revealed selectivity of 5–30 (CO2/CH4) [61]. Also aminefunctionalised IL-based solvent membranes enabled to achieve high selectivities [62] for CO2/CH4 and stable operation.

6.4

Challenges

The required pressure gradient across the solvent membrane is a significant contributor to OPEX and needs to be minimised. Also higher membrane selectivity will increase the purity of the obtained CO2 stream. Selectivity may be improved by applying solvent reactive only toward CO2 and exhibiting very low diffusivity of non-reacted species. The solvent membranes may have suitable applications in natural gas processing while for flue gases impurities may cause fouling and may accumulate in solvent membrane. For high CO2 content gas, multiple membranes may be required which are quite challenging to be implemented.

7 Ionic Liquids 7.1

Useful Mechanisms

Ionic liquids (ILs) may consist of either organic or inorganic salts that make use of both chemical and physical absorption. ILs reversibly react with CO2 and also have high physical CO2 loading capacity. The chemical CO2 loading capacity depends on the involved anions that attach CO2 [63, 64]. The physical CO2 loading capacity is mainly affected by free volume, size of an IL molecule and the IL’s chemical structure. The CO2 loaded IL is regenerated at slightly increased temperature enabling the release of physically as well as chemically bound CO2. They are recyclable and thermally stable.

7.2


Energy efficiency benefits are associated with high CO2 loading capacity of ILs which would reduce the flow rate of circulating liquid. In addition fast reaction rate exhibited by e.g. polyionic ILs has potential to further reduce required pumping work for solvent circulation. Besides, ILs have very low vapour pressures meaning that no (energy intensive) measures need to be applied to minimise solvent vaporisation. ILs are much less volatile than conventional solvents, which make them an energy-efficient solvent system. If ILs offer better selectivity, it can be more efficient.

84

7.3

W.M. Budzianowski

Practical Examples

ILs are viscous they are characterised by several times lower liquid mass transfer coefficients compared to MEA in the same conditions [65] which limits practical applications of ILs. Therefore, ILs need specific approaches which reduce existing liquid phase mass transfer limitations such as solvent membranes [66, 67].

7.4

Challenges

Challenges of IL solvents primarily involve high production cost due to complex chemical structures as well as high viscosity. The high viscosity is responsible for high pumping cost and mass transfer reduction due to slow diffusion of CO2 in ionic liquids. If ILs are in direct contact with flue gases they are quickly degraded which necessitates make-up of the expensive solvent. These are main obstacles for the use of ILs as neat solvents in large scale capture plants. It seems that ILs would have to be used only in the context of ASBCPs.

8 Polarity-Swing-Assisted Solvents 8.1

Useful Mechanisms

Polarity-swing-assisted solvents use organic liquids with switchable polarity. These materials, e.g. IPADM-2BOL, bind CO2 and the shift in polarity allows for more efficient CO2 stripping. The mechanism of switchable polarity relies on the use of a non-polar “antisolvent”, e.g. hexadecane. This additive is capable of destabilising the loaded solvent facilitating CO2 release. The stripping step can therefore be obtained under temperatures lower than required by neat solvents, e.g. 50–80 °C [68–70].

8.2


Due to a swing in solvent polarity the polarity-swing-assisted solvents may achieve alleviated energy requirement up to 65% compared to MEA benchmark (with concentration 30%) [71, 72]. Due to reversible CO2 binding [73] the stripping step is achieved through applying only modest heating or even bubbling with inert gas [74]. The thermodynamic properties of solvents are very important. For example, the 1,1,3,3-tetramethylguanidine (TMG) solvent is characterised by slightly lower reaction rate compared to MEA but has lower activation energy meaning that the process may be feasible at lower temperatures [75, 76]. If TMG is applied with


85

reaction promoters such as piperazine [77] the limited kinetics problem can be overcome. Overall, the process may be energy efficient. This expectation was confirmed by investigating the DBU/1-propanol system involving heptane as “antisolvent” [78]. This system achieved at 75 °C similar stripping rates and CO2 loadings as aqueous amines at 120 °C. Due to lower temperatures solvent degradation and evaporation were less pronounced [71, 72].

8.3

Practical Examples

One example of a polarity swing-assisted ASBCP is the IPADM-2BOL system comprising decane as “antisolvent”. The bench scale testing during the 4 months campaign of this ASBCP were described by Zheng et al. [79]. The process favourable characteristics included little effect of high solvent viscosities on mass transfer rates, no foaming, minimum solvent evaporation, no measurable solvent degradation. In addition, by applying a simple coalescence tank antisolvent was separated from the solvent by phase splitting. Antisolvent carryover was minimal. Water contamination up to 5 % did not degrade anhydrous solvent performance which suggested that this ASBCP could be employed for CO2 capturing from humid flue gases.

8.4

Challenges

These ASBCPs have relatively low TRL and sufficient technical characteristics under realistic capture conditions are not available. Materials selection is the main challenge, especially for large scale applications. The already demonstrated solvents required relatively high circulation rates implying the need for larger column sizing. High solvent viscosity implied increased pumping work [79]. In order to fully benefit from the polarity swing-assisted property further fundamental studies are therefore required.

9 Slurry Solvents 9.1

Useful Mechanisms

Small solid particles such as metal organic frameworks (MOFs) having remarkable capacity for CO2 sorption may be mixed with an appropriate liquid carrier to obtain a slurry with properties of a fluid and a solid [80], see Fig. 5. NMR measurements from [80] shows that the tested MOF significantly enhances bicarbonate/carbonate formation contributing to favourable CO2 loading capacity of the formed slurry.

86

W.M. Budzianowski

Fig. 5 The mechanism of CO2 scrubbing and stripping by slurry solvents

Thus slurries make benefit from high surface area of porous materials retaining benefits of liquid solvents.

9.2


Energy efficiency benefits mainly relate to the ease of solvent handling in a capture plant while benefiting from solid properties of the slurry. In addition, if efficient solid particles could be found, the resulting slurry solvent based ASBCP would consume little energy for solvent pumping while capture CO2 efficiently. The ease of slurry solvent recovery will also be important. They usually can be regenerated at low temperatures thus achieving reduced energy requirement compared to 30% aqueous MEA benchmark. MOFs can be usually regenerated below the boiling temperature of typical solvents thus reducing energy requirement.

9.3

Practical Examples

One of practical examples includes the metal organic framework complex {Zn[N [CH2(2-py)]3](l-OH)}2(NO3)2. It achieves CO2 loading capacity of around 2 molCO2/molMOF. However, due to its more than 8 times higher molecular weight it underperforms MEA solvents in terms of CO2 loading capacities on a mass basis. It has lower regeneration temperature compared to MEA, 80 °C is sufficient to


87

achieve complete regeneration including no carbamate accumulation which is a problem for MEA solutions at this temperature. If this MOF is dissolved in water the regeneration at 80 °C means operation below the solvent boiling temperature reducing energy requirement in stripping. The kinetics of CO2 absorption is similar to MEA [80]. In [81] a flexible and porous MOF consisting of Co3(OH)2(BTCA)2 with DMF as guest molecules exhibiting good CO2 uptake (223.7 mg/g at 273 K and 104.7 mg/g at 298 K) and better selectivity for CO2/N2 (up to 80) was investigated. It could be thus used as a slurry solvent.

9.4

Challenges

Main challenges are associated with difficulties to find small solid particles with sufficiently high CO2 loading capacity on a mass basis. These molecules are usually characterised by high molecular mass and hence even if have high CO2 loading capacity on a molar basis, their loading capacity on mass basis can be small relative to common solvents such as MEA. Also the stripping step of some slurry solvents may require harsh conditions increasing the overall energy requirement of the process.

10 10.1

Liquid Crystals Useful Mechanisms

Liquid crystals are molecules that form a structured crystalline phase and liquid phase. The former is referred to as a nematic phase and the latter as isotropic phase. The shift between the two phases is possible upon a trigger such as a (preferably small) temperature change. The isotropic phase is formed at higher temperatures. CO2 capture makes use of a difference between free volume of the solvent between the isotropic (higher free volume) and nematic (lower free volume) phases [82]. The nematic phase is characterised by orientational ordering but no positional ordering. The solubility of CO2 is higher in the isotropic phase than in the nematic phase making liquid crystals suitable for CO2 scrubbing and stripping cycles, see Fig. 6.

10.2


Heating the liquid crystal nematic phase just a few degrees is already sufficient to absorbed CO2. Given the very small temperature trigger this kind of ASBCP has potential to consume much less energy than required in conventional absorptions.

88

W.M. Budzianowski

Fig. 6 Mechanisms of the operation of liquid crystals based ASBCPs

Both phases have similar viscosity [83] meaning that work requirement for solvent pumping may be limited. Scrubbing/stripping cycles are different with liquid crystals than with conventional solvents since CO2 release is obtained at a lower temperature than CO2 scrubbing. This property may have implications for energy efficiency benefits however no work has addressed this process opportunity so far. Theoretically flue gases could be used to heat the scrubbing process while CO2 stripping could be subsequently achieved by cooling the solvent to at ambient temperature meaning no energy requirements for regeneration.

10.3

Practical Examples

Binary and ternary mixtures of chemicals forming liquid crystals are considered for CO2 capture applications. Researches carried out by de Groen et al. [83] show that the combination of PCH3 and PCH7 is more promising compared to 5OCB and 7OCB. The former binary system has higher phase transition enthalpy and associated CO2 concentration difference, however the latter has higher CO2 loading capacity. Further researches are required to find improved liquid crystals. Simultaneous experimental, theoretical and molecular simulation work is needed. PCH5 liquid crystals have CO2 loading capacity of about 5 %wt and scrubbing is obtained at 40 °C. The nematic phase is formed at 25 °C. CO2 stripping from the nematic phase requires enhancement, e.g. by bubbling with an inert gas [84] which complicates process feasibility.


10.4

89

Challenges

Major challenges are associated with low TRL and insufficient demonstration. It means that potential shortcomings has not yet been revealed. Among confirmed challenges is the low phase transition enthalpy (from nematic to isotropic) for tested liquid crystals. This leads to a small solubility difference of absorbed CO2 in the two phases. The small concentration difference may limit the effectiveness of scrubbing/stripping cycles [83].

11 11.1

Deep Eutectic Solvents Useful Mechanisms

Deep Eutectic Solvents (DESs) consist of a salt and a second compound as H2 donor forming a low melting point mixture. DESs have low melting temperatures due intermolecular H2 bonds [85]. Due to charge delocalization in DESs the melting points are even lower than that of individual components [86–88]. The H2 bonds play a remarkable role in more efficient binding of CO2 molecules by DESs compared to classical ILs.

11.2


Energy efficiency benefits are similar to those of ILs and largely depend on the specific DESs used. The role of a H2 donor molecules is essential in minimising energy needs in scrubbing/stripping cycles [89]. More process level studies are needed to more clearly demonstrate potential energy efficiency benefits associated with DESs under practical CO2 capture conditions.

11.3

Practical Examples

Practical examples of DESs include choline chloride salt and urea in molar ratio of 1:2 [90]. Choline chloride salt may also form DESs with carboxylic acids and polyols [85, 91]. Choline chloride may form DESs with levulinic acid/furfuryl alcohol for which detailed CO2 loading capacity data can be found in [92]. Several potential DESs have been reviewed by [93] including multiple halide slats and hydrogen bond donors but little process level information has been provided.

90

W.M. Budzianowski

Table 1 A summary of useful mechanisms, energy efficiency benefits and challenges of investigated ASBCPs Advanced solvent based capture process

Useful mechanisms

Energy efficiency benefits

Challenges

Precipitating solvents

Solids precipitate upon contact with CO2. The precipitated slurry is thickened and conveyed to a stripper

Complex slurry handling required Reactants are required

Two immiscible liquid phases

Temperature triggered formation of two immiscible liquid phases with one low-volume high CO2 concentration phase is conveyed to a stripper Carbonic anhydrase is used to catalyse CO2 reaction with water

Practically infinite CO2 loading capacity due to precipitation Stable and fast absorption rate associated with rapid precipitation of CO2 loaded compounds CO2 accumulation in one phase reduces energy requirement in solvent pumping and regeneration Low regeneration temperature enables to use low-grade thermal energy Reaction rate increase by up to 10 fold for high CO2 concentrations

Catalysed solvents

Microencapsulated solvents

The solvent is encapsulated in microcapsules which separates the solvent from flue gases and plant infrastructure

Solvent membranes

The porous membrane separates the solvent from flue gases. Flue gases (optionally pressurised) are

High surface area facilitates mass transfer Solvents have prolonged lifetime and more expensive solvents may be used Can handle precipitating solvents Capable of minimising solvent volatility facilitating the utilisation of volatile solvents such as ammonia

Unconventional solvent handling is required involving a coalescence tank Very specific chemicals are required to fit the operating window of CO2 capture A CA based ASBCP requires temperatures below 40–50 °C to prohibit protein denaturation CA handling is complex, e.g. it requires immobilisation on packing materials or filtration from the spent solvent Contaminated flue gases may gradually block the pores of the silicone shells May require circulation of microcapsules between the scrubber and stripper may reduce their lifetime Innovative membrane materials required The required pressure gradient (continued)


91

Table 1 (continued) Advanced solvent based capture process

Ionic liquids

Useful mechanisms


Challenges

conveyed to the channel comprising the porous membrane. CO2 permeates through the pores and reacts with the solvent in the inner space. The CO2 loaded solvent molecules diffuse through the solvent membrane to the sweep gas side, decompose to release CO2 which leaves the system as a permeate stream Chemical and physical absorption of CO2

The size of the absorption equipment reduced due to high surface to volume ratio Reactive liquids enable achieving very high selectivities Due to no direct contact between a solvent and flue gases, more expensive solvent materials may be employed, e.g. ILs or catalysed solvents High CO2 loading capacity due physicochemical nature of absorption Completely reversible reaction with CO2 facilitates stripping No thermal regeneration reduces overall energy requirement

across the solvent increases OPEX Membrane wettability and fouling

Polarity-swing-assisted solvents

Polarity shift due to “antisolvent” addition chemically destabilises the CO2 loaded solvent and facilitates stripping

Slurry solvents

Small solid particles mixed with a solvent may be operated like a fluid adding to operational flexibility of capture plants

Liquid crystals

Difference in CO2 solubility between isotropic and nematic phases that

Fluid nature of the slurry alleviates pumping work MOFs can be regenerated below the boiling temperature of typical solvents thus reducing energy requirement Small temperature trigger makes possible low heat demand

High solvent production cost limit application only to ASBCPs (too costly for use as a neat solvent)

Difficulties in finding suitable polarity swing materials for realistic CO2 capture conditions Very low TRL and lacking technical details Small solid particles with sufficiently high CO2 loading capacity on a mass basis needs to be found

Very small CO2 solubility difference between the two phases leads to (continued)

92

W.M. Budzianowski

Table 1 (continued) Advanced solvent based capture process

Deep eutectic solvents

11.4

Useful mechanisms


Challenges

shift upon a (small) temperature difference

Similar (possibly low) viscosity of both phases reduces solvent pumping work If scrubbing could use heat of flue gases and stripping could be achieved at ambient temperature, the process would require nearly no energy for solvent regeneration but such systems have not been found Released energies for CO2 capture are larger than those of classical ILs

inefficiencies of multiple scrubbing/stripping cycles

Consist of a salt and a second compound acting as hydrogen donor forming a low melting mixture The H2 bonds play a remarkable role in more efficient binding of CO2 molecules by DESs compared to classical ILs

Require high pressures and low temperatures in absorption

Challenges

From scarce literature information it is seen that in order to achieve meaningful CO2 loading capacity low temperatures and high pressures need to be applied. For example, for DES consisting of choline chloride and glycerol in a 1:2 molar ratio at 313 K CO2 loading capacity is 1.5 molCO2/kgDES at 30 bar and only about 0.15 molCO2/kgDES at 3 bar. It suggests that DESs operate like physical solvents and in the practice would require pressurised absorption in which high energy requirement is associated with flue gas compression if post combustion is to be used. Since there is almost no potential to reduce compression work (compression already approaches thermodynamic limits) little progress toward energy efficient CO2 capture with DESs requiring flue gas pressurisation can be expected. In general state-of-the-art DESs based ASBCPs lack demonstration activities that could provide reliable process related information for CO2 capture under realistic conditions.


12

93

Summary

Table 1 provides useful mechanisms, energy efficiency benefits and challenges of all discussed ASBCPs. The investigated ASBCPs are at different TRL levels. The ASBCPs that are at the lowest TRL require further fundamental studies to expound useful mechanisms and tune operating parameters [94]. The ASBCPs at higher TRL after demonstration can be qualified for real CO2 capture applications. By applying these qualified energy efficient ASBCPs the energy sector can be decarbonised [95, 96].

13

Conclusions and Outlook

The current study focused on energy efficient advanced solvent based capture processes (ASBCPs) for capturing CO2 by gas-liquid absorption. The analysis of 11 types of different ASBCPs was presented by explaining useful mechanisms, energy efficient benefits, practical examples and challenges. Several described ASBCPs demonstrated significant reductions in terms of energy requirement for CO2 capture compared to MEA as the solvent benchmark. However, many emerging ASBCPs cannot be employed in the practice because of incompatibility with harsh operating conditions and inability to overcome challenges associated with flue gas impurities. The identified challenges of ASBCPs suggest that further researches are needed for most of them. Overcoming these challenges will enable to achieve meaningful progress toward commercialisation and wide-scale adoption of ASBCPs. Candidates for ASBCPs need to be systematically screened, researched and demonstrated. This study highlights certain groups of promising ASBCPs. For instance precipitating solvents, two immiscible liquid phases, catalysed solvents and microencapsulated solvents as well as their combinations. Other discussed ASBCPs are interesting but need more fundamental insights in order to achieve progress toward potential commercialisation. Acknowledgments This study has been supported by the members of the Renewable Energy and Sustainable Development (RESD) Group (Poland) under the project RESD-RDG03/2016 which is gratefully acknowledged.

References 1. Abu-Zahra MRM, Abbas Z, Singh P, Feron P. (2013) Carbon dioxide post-combustion capture: solvent technologies overview, status and future directions. In Méndez-Vilas E (ed) Materials and processes for energy: communicating current research and technological developments, vol 1. Formatex Research Center, pp 923–934

94

W.M. Budzianowski

2. Guo BS, Jing GH, Zhou ZM (2015) Regeneration performance and absorption/desorption mechanism of tetramethylammonium glycinate aqueous solution for carbon dioxide capture. Int J Greenhouse Gas Control 34:31–38 3. Vasudevan S, Farooq S, Karimi IA, Saeys M, Quah MCG, Agrawal R (2016) Energy penalty estimates for CO2 capture: comparison between fuel types and capture-combustion modes. Energy 103:709–714 4. Budzianowski WM (2012) Benefits of biogas upgrading to biomethane by high-pressure reactive solvent scrubbing. Biofuels Bioprod Biorefin 6(1):12–20 5. Fu C, Gundersen T (2012) Using exergy analysis to reduce power consumption in air separation units for oxy-combustion processes. Energy 44(1):60–68 6. Liu H, Liu B, Lin LC, Chen G, Wu Y, Wang J, Gao X, Lv Y, Pan Y, Zhang X, Zhang X, Yang L, Sun C, Smit B, Wang W (2014) A hybrid absorption-adsorption method to efficiently capture carbon. Nat Commun 5:5147 7. Perry RJ, Wood BR, Genovese S, Obrien MJ, Westendorf T, Meketa ML, Farnum R, McDermott J, Sultanova I, Perry TM, Vipperla RK, Wichmann LA, Enick RM, Hong L, Tapriyal D (2012) CO2 capture using phase-changing sorbents. Energy Fuels 26(4): 2528–2538 8. Gonzalez-Salazar MA, Perry RJ, Vipperla RK, Hernandez-Nogales A, Nord LO, Michelassi V, Shisler R, Lissianski V (2012) Comparison of current and advanced postcombustion CO2 capture technologies for power plant applications. Energy Procedia 23:3–14 9. Yu J, Wang S, Yu H (2014) Modelling analysis of solid precipitation in an ammonia-based CO2 capture process. Int J Greenhouse Gas Control 30:133–139 10. Gazzani M, Sutter D, Mazzotti M (2014) Improving the efficiency of a chilled ammonia CO2 capture plant through solid formation: a thermodynamic analysis. Energy Procedia 63: 1084–1090 11. Fernandez ES, Goetheer ELV (2011) DECAB Process development of a phase change absorption process. Energy Procedia 4:868–875 12. Sanchez Fernandez E (2013) Novel process designs to improve the efficiency of postcombustion carbon dioxide capture. Ph.D. thesis, Delft University of Technology 13. Sanchez Fernandez E, Heffernan K, van der Ham L, Linders MJG, Eggink E, Schrama FNH, Brilman DWF, Goetheer ELV, Vlugt TJH (2013) Conceptual design of a novel CO2 capture process based on precipitating amino acid solvents. Ind Eng Chem Res 52(34):12223–12235 14. Aronu UE, Kim I, Haugen G (2014) Evaluation of energetic benefit for solid-liquid phase change CO2 absorbents. Energy Procedia 63:532–541 15. Sanchez-Fernandez E, Heffernan K, van der Ham L, Linders MJG, Brilman DWF, Goetheer ELV, Vlugt TJH (2014) Analysis of process configurations for CO2 capture by precipitating amino acid solvents. Ind Eng Chem Res 53(6):2348–2361 16. Perry R. (2010) Exploring more cost-effective methods for CO2 capture. http://ge.geglobal research.com/blog/exploring-more-cost-effective-methods-for-co2-capture/. Accessed Dec 2015 17. Perry R (2010) Phase changing solvents for CO2 capture. http://ge.geglobalresearch.com/ blog/phase-changing-solvents-for-co2-capture/. Accessed Dec 2015 18. Smith K, Xiao G, Mumford K, Gouw J, Indrawan I, Thanumurthy N, Quyn D, Cuthbertson R, Rayer A, Nicholas N, Lee A, da Silva G, Kentish S, Harkin T, Qader A, Anderson C, Hooper B, Stevens G (2014) Demonstration of a concentrated potassium carbonate process for CO2 capture. Energy Fuels 28(1):299–306 19. Smith K, Lee A, Mumford K, Li S, Indrawan I, Thanumurthy N, Temple N, Anderson C, Hooper B, Kentish S, Stevens G (2015) Pilot plant results for a precipitating potassium carbonate solvent absorption process promoted with glycine for enhanced CO2 capture. Fuel Process Technol 135:60–65 20. Anderson C, Hooper B, Qader A, Harkin T, Smith K, Mumford K, Pandit J, Ho M, Lee A, Nicholas N, Indrawan I, Gouw J, Xiao J, Thanumurthy N, Temple N, Stevens G, Wiley D (2014) Recent developments in the UNO MK 3 process—a low cost, environmentally benign precipitating process for CO2 capture. Energy Procedia 63:1773–1780


95

21. Lupiáñez C, Guedea I, Bolea I, Díez LI, Romeo LM (2013) Experimental study of SO2 and NOX emissions in fluidized bed oxy-fuel combustion. Fuel Process Technol 106:587–594 22. Asif M, Kim WS (2014) Modeling and simulation of the combined removal of SO2 and CO2 by aqueous ammonia. Greenhouse Gases Sci Technol 4(4):509–527 23. Zhang J, Qiao Y, Agar DW (2012) Intensification of low temperature thermomorphic biphasic amine solvent regeneration for CO2 capture. Chem Eng Res Des 90:743–749 24. Yu YS, Lu HF, Zhang TT, Zhang ZX, Wang GX, Rudolph V (2013) Determining the performance of an efficient non-aqueous CO2 capture process at desorption temperature below 373K. Ind Eng Chem Res 52(35):12622–12634 25. Monteiro JGM, Pinto DDD, Svendsen HF. (2011) Phase change solvents. In: EU-China Workshop on Innovative CCS Technologies, Beijing, 19–20 Sept 2011 26. Raynal L, Bouillon PA, Gomez A, Broutin P (2011) From MEA to demixing solvents and future steps, a roadmap for lowering the cost of post-combustion carbon capture. Chem Eng J 171(3):742–752 27. Zhang J, Nwani O, Tan Y, Agar DW (2011) Carbon dioxide absorption into biphasic amine solvent with solvent loss reduction. Chem Eng Res Des 89:1190–1196 28. Monteiro JGMS, Pinto DDD, Zaidy SAH, Hartono A, Svendsen HF (2013) VLE data and modelling of aqueous N, N-diethylethanolamine (DEEA) solutions. Int J Greenhouse Gas Control 19:432–440 29. Pinto DDD, Knuutila H, Fytianos G, Haugen G, Mejdell T, Svendsen HF (2014) CO2 post combustion capture with a phase change solvent. Pilot plant campaign. Int J Greenhouse Gas Control 31:153–164 30. Xu Z, Wang S, Qi G, Liu J, Zhao B, Chen C (2014) CO2 absorption by biphasic solvents: comparison with lower phase alone. Oil Gas Sci Technol 69(5):851–864 31. Ye Q, Wang X, Lu Y (2015) Screening and evaluation of novel biphasic solvents for energy-efficient post-combustion CO2 capture. Int J Greenhouse Gas Control 39:205–214 32. Pendersen-van Elk NJMC, Derks PWJ, Fradette S, Veersteg GF (2012) Kinetics of absorption of carbon dioxide in aqueous MDEA solutions with carbonic anhydrase at 298 K. Int J Greenhouse Gas Control 9:385–392 33. Li L, Fu ML, Zhao YH, Zhu YT (2012) Characterization of carbonic anhydrase II from Chlorella vulgaris in bio-CO2 capture. Environ Sci Pollut Res 19(9):4227–4232 34. Kunze AK, Dojchinov G, Haritos VS, Lutze P (2015) Reactive absorption of CO2 into enzyme accelerated solvents: from laboratory to pilot scale. Appl Energy 156:676–685 35. Akermin Inc. (2013) Final Scientific/Technical Report—Advanced Low Energy Enzyme— Catalyzed Solvent for CO2 Capture. http://www.netl.doe.gov/File%20Library/Research/Coal/ carbon%20capture/post-combustion/FE0004228-Final-Report-01-07-2014.pdf 36. Saeed M, Deng L (2016) Post-combustion CO2 membrane absorption promoted by mimic enzyme. J Membr Sci 499:36–46 37. Phan DT, Maeder M, Burns RC, Puxty G (2014) Catalysis of CO2 absorption in aqueous solution by inorganic oxoanions and their application to post combustion capture. Environ Sci Technol 48(8):4623–4629 38. Kvamsdal HM, Kim I, van Os P, Pevidac C, Hägg MB, Browne J, Robinson L, Feron P (2014) HiPerCap: a new FP7 project for development and assessment of novel and emerging post-combustion CO2 capture technologies. Energy Procedia 63:6166–6172 39. Yong JKJ, Stevens GW, Caruso F, Kentish SE (2015) The use of carbonic anhydrase to accelerate carbon dioxide capture processes. J Chem Technol Biotechnol 90:3–10 40. Li K, Yu H, Tade M, Feron P (2014) Theoretical and experimental study of NH3 suppression by addition of Me(II) ions (Ni, Cu and Zn) in an ammonia-based CO2 capture process. Int J Greenhouse Gas Control 24:54–63 41. Yu J, Wang S, Yu H (2013) Experimental studies on suppression of ammonia vaporization by additives. Greenhouse Gases Sci Technol 3(5):415–422 42. Ma S, Song H, Chen G, Wang Y, Zang B (2013) Cobalt(II) as an additive inhibiting ammonia escape in carbon capture using ammonia solution. Greenhouse Gases Sci Technol 3(5): 392–396

96

W.M. Budzianowski

43. Ma S, Song H, Zang B, Chen G (2013) Experimental study on additives inhibiting ammonia escape in carbon capture process using ammonia method. Chem Eng Res Des 91(12): 2775–2781 44. Ma S, Chen G, Han T, Zhu S, Yang J (2015) Experimental study on the effect of Ni(II) additive on ammonia escape in CO2 capture using ammonia solution. Int J Greenhouse Gas Control 37:249–255 45. Stolaroff J, Vericella J, Baker S, Spadaccini C, Duoss E, Aines R. Micro-encapsulation of advanced solvents for post-combustion carbon capture. In: 3rd Post Combustion Capture Conference (PCCC3) 46. Vericella JJ, Baker SE, Stolaroff JK, Duoss EB, Hardin JO, Lewicki J, Glogowski E, Floyd WC, Valdez CA, Smith WL, Satcher JH Jr, Bourcier WL, Spadaccini CM, Lewis JA, Aines RD (2015) Encapsulated liquid sorbents for carbon dioxide capture. Nat Commun 6:6124 47. Aines RD, Bourcier WL, Spadaccini CM, Stolaroff JK (2015) Polymer-encapsulated carbon capture liquids that tolerate precipitation of solids for increased capacity. US 8945279B2 48. Asif M, Kim WS (2015) Process simulation of ammonia-based CO2 capture and regeneration in packed column. Int J Glob Warming 8(3):401–415 49. Favre E, Svendsen HF (2012) Membrane contactors for intensified post-combustion carbon-dioxide capture by gas-liquid absorption processes. J Membr Sci 407–408:1–7 50. Lu JG, Ji Y, Zhang H, Chen MD (2010) CO2 capture using activated amine acid salt solutions in a membrane contactor. Sep Sci Technol 45:1240–1251 51. Budzianowski WM (2016) Explorative analysis of advanced solvent processes for energy efficient carbon dioxide capture by gas-liquid absorption. Int J Greenhouse Gas Control 49:108–120 52. Budzianowski WM (2015) Single solvents, solvent blends, and advanced solvent systems in CO2 capture by absorption: a review. Int J Glob Warming 7(2):184–225 53. Budzianowski WM (2011) CO2 reactive absorption from flue gases into aqueous ammonia solutions: the NH3 slippage effect. Environ Prot Eng 37(4):5–19 54. Scovazzo P (2009) Determination of the upper limits, benchmarks, and critical properties for gas separations using stabilized room temperature ionic liquid membranes (SILMs) for the purpose of guiding future research. J Membr Sci 343:199–211 55. Bara JE, Gabriel CJ, Carlisle TK, Camper DE, Finotello A, Gin DL, Noble RD (2009) Gas separations in fluoroalkyl-functionalized room-temperature ionic liquids using supported liquid membranes. Chem Eng J 147:43–50 56. Cserjesi P, Nemestothy N, Belafi-Bako K (2010) Gas separation properties of supported liquid membranes prepared with unconventional ionic liquids. J Membr Sci 349:6–11 57. Neves LA, Crespo JG, Coelhoso IM (2010) Gas permeation studies in supported ionic liquid membranes. J Membr Sci 357:160–170 58. Scovazzo P, Havard D, McShea M, Mixon S, Morgan D (2009) Long-term, continuous mixed-gas dry fed CO2/CH4 and CO2/N2 separation performance and selectivities for room temperature ionic liquid membranes. J Membr Sci 327:41–48 59. Myers C, Pennline H, Luebke D, Ilconich J, Dixon JK, Maginn EJ, Brennecke JF (2008) High temperature separation of carbon dioxide/hydrogen mixtures using facilitated supported ionic liquid membranes. J Membr Sci 322:28–31 60. Yoo S, Won J, Kang SW, Kang Y, Nagase S (2010) CO2 separation membranes using ionic liquids in a Nafion matrix. J Membr Sci 363:72–79 61. Iarikov DD, Hacarlioglu P, Oyama ST (2011) Supported room temperature ionic liquid membranes for separation. Chem Eng J 166:401–406 62. Hanioka S, Maruyama T, Sotani T, Teramoto M, Matsuyama H, Nakashima K, Hanaki M, Kubota F, Goto M (2008) CO2 separation facilitated by task-specific ionic liquids using a supported liquid membrane. J Membr Sci 314:1–4 63. Mumford KA, Wu Y, Smith KH, Stevens GW (2015) Review of solvent based carbon-dioxide capture technologies. Front Chem Sci Eng 9(2):125–141


97

64. Tang J, Tang H, Sun W, Plancher H, Radosz M, Shen Y (2005) Poly (ionic liquid) s: a new material with enhanced and fast CO2 absorption. Chem Commun 26:3325–3327 65. Ziobrowski Z, Krupiczka R, Rotkegel A (2016) Carbon dioxide absorption in a packed column using imidazolium based ionic liquids and MEA solution. Int J Greenhouse Gas Control 47:8–16 66. Ramdin M, de Loos TW, Vlugt TJH (2012) State-of-the-art of CO2 capture with ionic liquids. Ind Eng Chem Res 51(24):8149–8177 67. Zhang X, Zhang X, Dong H, Zhao Z, Zhang S, Huang Y (2012) Carbon capture with ionic liquids: overview and progress. Energy Environ Sci 5:6668–6681 68. Heldebrant DJ, Yonker CR, Jessop PG, Phan L (2009) CO2-binding organic liquids (CO2BOLs) for post-combustion CO2 capture. Energy Procedia 1(1):1187–1195 69. Heldebrant DJ, Koech PK, Ang MTC, Liang C, Rainbolt JE, Yonker CR, Jessop PG (2010) Reversible zwitterionic liquids, the reaction of alkanol guanidines, alkanol amidines, and diamines with CO2. Green Chem 12:713 70. Jessop PG, Heldebrant DJ, Li X, Eckert CA, Liotta CL (2005) Green chemistry: reversible nonpolar-to-polar solvent. Nature 436:1102 71. Koech PK, Zhang J, Kutnyakov IV, Cosimbescu L, Lee SJ, Bowden ME, Smurthwaite TD, Heldebrant DJ (2013) Low viscosity alkanolguanidine and alkanolamidine liquids for CO2 capture. RSC Adv 3(2):566–572 72. Mathias PM, Afshar K, Zheng F, Bearden MD, Freeman CJ, Andrea T, Koech PK, Kutnyakov I, Zwoster A, Smith AR, Jessop PG, Nik OG, Heldebrant DJ (2013) Improving the regeneration of CO2-binding organic liquids with a polarity change. Energy Environ Sci 6 (7):2233–2242 73. Heldebrant DJ, Yonker CR, Jessop PG, Phan L (2008) Organic liquid CO2 capture agents with high gravimetric CO2 capacity. Energy Environ Sci 1(4):487–493 74. Heldebrant DJ, Koech PK, Rainbolt JE, Zheng F, Smurthwaite T, Freeman CJ, Oss M, Leito I (2011) Performance of single-component CO2-binding organic liquids (CO2BOLs) for post combustion CO2 capture. Chem Eng J 171(3):794–800 75. Öztürk MÇ, Yüksel Orhan Ö, Alper E (2014) Kinetics of carbon dioxide binding by 1,1,3,3-tetramethylguanidine in 1-hexanol. Int J Greenhouse Gas Control 26:76–82 76. Yüksel Orhan Ö, Öztürk MÇ, Şeker A, Alper E (2015) Kinetics and performance studies of a switchable solvent TMG (1,1,3,3-tetramethylguanidine)/1-propanol/carbon dioxide system. Turk J Chem 39(1):13–24 77. Li L, Conway W, Puxty G, Burns R, Clifford S, Maeder M, Yu H (2015) The effect of piperazine (PZ) on CO2 absorption kinetics into aqueous ammonia solutions at 25.0 °C. Int J Greenhouse Gas Control 36:135–143 78. Zhang J, Kutnyakov I, Koech PK, Zwoster A, Howard C, Zheng F, Freeman CJ, Heldebrant DJ (2013) CO2-binding-organic-liquids-enhanced CO2 capture using polarity-swing-assisted regeneration. Energy Procedia 37:285–291 79. Zheng F, Heldebrant D, Mathias P, Koech P, Bhakta M, Freeman C, Bearden M, Zwoster A (2016) Bench scale testing and process performance projections of CO2 capture by CO2BOLs with and without polarity swing assisted regeneration. Energy Fuels 30(2):1192–1203 80. Heyn RH, Aronu UE, Vevelstad SJ, Hoff KA, Didriksen T, Arstad B, Blom R (2014) Use of metal-organics based solvents for CO2 capture. Energy Procedia 63:1805–1810 81. Ren H-Y, Zhang X-M (2016) Enhanced selective CO2 capture upon incorporation of dimethylformamide in the cobalt Metal-Organic Framework [Co3(OH)2(btca)2]. Energy Fuels 30(1):526–530 82. Oyarzun B, van Westen T, Vlugt TJH (2013) The phase behaviour of linear and partially flexible hard-sphere chain fluids and the solubility of hard spheres in hard-sphere chain fluids. J Chem Phys 138:204905 83. de Groen M, Vlugt TJH, de Loos T (2015) Binary and ternary mixtures of liquid crystals with CO2. AIChE J 61(9):2977–2984 84. Gross J, Jansens PJ. (2008) A method for the removal of a gas from a process gas stream by means of liquid crystals. Patent WO2008147181A1

98

W.M. Budzianowski

85. Zhang Q, de Oliveira Vigier K, Royer S, Jerome F (2012) Deep eutectic solvents: syntheses, properties and applications. Chem Soc Rev 41:7108–7146 86. Carriazo D, Serrano MC, Gutierrez MC, Ferrer ML, del Monte F (2012) Deep-eutectic solvents playing multiple roles in the synthesis of polymers and related materials. Chem Soc Rev 41:4996–5014 87. Paiva A, Craveiro R, Aroso I, Martins M, Reis RL, Duarte ARC (2014) Natural deep eutectic solvents—solvents for the 21st Century. ACS Sustain Chem Eng 2:1063–1071 88. Florindo C, Oliveira FS, Rebelo LPN, Fernandes AM, Marrucho IM (2014) Insights into the synthesis and properties of deep eutectic solvents based on cholinium chloride and carboxylic acids. ACS Sustain Chem Eng 2:2416–2425 89. García G, Atilhan M, Aparicio S (2015) A theoretical study on mitigation of CO2 through advanced deep eutectic solvents. Int J Greenhouse Gas Control 39:62–73 90. Abbott AP, Capper G, Davies DL, Rasheed RK, Tambyrajah V. (2003) Novel solvent properties of choline chloride/urea mixtures. Chem Commun 70–71 91. Wagle DV, Zhao H, Baker GA (2014) Deep eutectic solvents: sustainable media for nanoscale and functional materials. Acc Chem Res 47:2299–2308 92. Lu M, Han G, Jiang Y, Zhang X, Deng D, Ai N (2015) Solubilities of carbon dioxide in the eutectic mixture of levulinic acid (or furfuryl alcohol) and choline chloride. J Chem Thermodyn 88:72–77 93. García G, Aparicio S, Ullah R, Atilhan M (2015) Deep eutectic solvents: physicochemical properties and gas separation applications. Energy Fuels 29:2616–2644 94. Budzianowski WM (2016) A review of potential innovations for production, conditioning and utilization of biogas with multiple criteria assessment. Renew Sustain Energy Rev 54: 1148–1171 95. Budzianowski WM (2012) Negative carbon intensity of renewable energy technologies involving biomass or carbon dioxide as inputs. Renew Sustain Energy Rev 16(9):6507–6521 96. Budzianowski WM (2012) Value-added carbon management technologies for low CO2 intensive carbon-based energy vectors. Energy 41(1):280–297

Phase Change Solvents for CO2 Capture Applications Kathryn A. Mumford, Kathryn H. Smith and Geoffrey W. Stevens

Abstract Solvent systems that separate into two phases upon absorption of CO2, one rich and one lean in CO2, have significant potential to exhibit reduced energy requirements. This reduction stems primarily from the ability to separate the two phases such that only the stream containing CO2 is heated in the regenerator resulting in reduced sensible and latent heating requirements. Thus, this class of solvents are currently under intense investigation in universities and industries globally. This chapter presents recent developments in Solid-Liquid and Liquid-Liquid phase change solvents for CO2 capture, including laboratory, pilot scale and commercial system installations. Nomenclature A AAS ABS AMP COND DEA DEEA DMCA DPA HEX HP CO2 Kn LP CO2 MAPA MEA NMP

Amine (in Fig. 2) Amino acid salt Absorber 2-Amino-2-methyl-1-propanol Condenser Diethanolamine 2-(diethylamino)ethanol Dimethylcyclohexylamine Dipropylamine Heat exchanger High pressure carbon dioxide Equilibrium constant Low pressure carbon dioxide 3-(methylamino)propylamine Monoethanolamine N-methyl-2-pyrrolidone

K.A. Mumford (&) K.H. Smith G.W. Stevens The University of Melbourne, Melbourne, Australia e-mail: [email protected] © Springer International Publishing AG 2017 W.M. Budzianowski (ed.), Energy Efficient Solvents for CO2 Capture by Gas–Liquid Absorption, Green Energy and Technology, DOI 10.1007/978-3-319-47262-1_5

99

100

P REB SEP TBS TEGDME TEPA TETA

K.A. Mumford et al.

Pump Reboiler Separator Thermomorphic biphasic solvent Triethylene glycol dimethyl ether Tetraethylenepentamine Triethylenetetramine

1 Introduction High regeneration energy requirements for solvents used in CO2 capture processes are a key challenge in capture technology process development. Solvents that undergo a phase change upon CO2 absorption (such as precipitation or generation of two immiscible phases) have recently attracted attention as a promising option for reducing the regeneration energy requirement. It is known that certain components in solvent systems may form a precipitate upon absorption of CO2. This precipitation results in the formation of a slurry which may then be separated into two streams: a stream rich in CO2 and a stream lean in CO2. Potentially, the use of precipitating solvents may have several advantages over traditional, liquid based systems resulting from; their ability to maintain absorption at higher CO2 loadings resulting in higher cyclical loadings and lower solvent recirculation rates; increased kinetics and smaller contacting equipment; and reduced energy consumption during regeneration due to a reduction in latent heat requirements, as compared to traditional liquid amine based systems. A number of precipitating solvent systems are being investigated by different companies and research institutions including; chilled ammonia (Alstom [1]), amino acids (TNO [42], NTNU [35]) and potassium carbonate (UNO Technology [15], University of Melbourne [33, 34], Shell [23]), in addition to systems which utilise organic solvents as diluents to the reactive solute as opposed to water. Processes under development include both the precipitation of the carbon dioxide containing species and non-carbon dioxide containing species [8]. As yet, no precipitating processes are operating at an industrial scale for carbon dioxide capture, as the design and operation of suitable equipment for precipitating systems requires further development and optimisation. Another class of biphasic solvents results in two immisicible liquid phases following absorption of CO2. One liquid phase will have a low CO2 content while the other phase will be high in CO2. This resulting CO2 rich phase can be regenerated with a much lower energy penalty than traditional liquid processes. Commercial examples of these types of biphasic solvents include the DMX process, thermomorphic solvents and mixed amine (e.g. DEEA/MAPA) systems. This chapter will discuss recent advances in phase change solvents for CO2 capture, including solid-liquid and liquid-liquid processes.

Phase Change Solvents for CO2 Capture Applications

101

2 Solid-Liquid Phase Change Solvents for CO2 Capture A number of precipitating solvent processes have been developed for CO2 capture applications, including chilled ammonia, amino acid salts and carbonate solvents. All of these processes take advantage of the significant reduction in the energy penalty of the regeneration process when using a slurry or precipitating solvent. Other benefits of precipitating CO2 capture processes include a concentration step for the solids which leads to higher CO2 partial pressure and a lower solvent flow rate to the regenerator compared to competing processes [23].

2.1

Precipitating Chilled Ammonia Process

The use of chilled ammonia to capture carbon dioxide was first patented by Gal [10, 11] in 2006. This process is based upon the use of ammonia and other trace chemicals to absorb carbon dioxide from flue gases at low temperatures and pressures. Post-combustion flue gas streams emitted from power stations or other low pressure CO2 containing streams from industrial sources have been identified as the most likely market for this technology. Unlike traditional solvents, such as monoethanolamine (MEA), the ammonia solvent solution stability is not adversely impacted by oxygen or other trace contaminants, and operation at low temperatures allows use of waste heat for regeneration purposes and also decreases the ammonia slip in the absorber. Additionally, it is reported that the heat of absorption of CO2 by ammonia is significantly lower than for amines. In the chilled ammonia process, CO2, ammonia and water combine to form ammonium carbonate ððNH4 Þ2 CO3 Þ, ammonium bicarbonate ðNH4 HCO3 Þ and ammonium carbamate ðNH2 CO2 NH4 Þ all of which have the potential to form precipitates. The main reactions associated with this process are presented as reaction (1)–(6). CO2ðgÞ $ CO2ðlÞ

ð1Þ

2NH3ðaqÞ þ H2 OðlÞ þ CO2ðaqÞ $ ðNH4 Þ2 CO3ðaqÞ $ ðNH4 Þ2 CO3ðsÞ

ð2Þ

NH3ðaqÞ þ H2 OðlÞ þ CO2ðaqÞ $ NH4 HCO3ðaqÞ $ NH4 HCO3ðsÞ

ð3Þ

ðNH4 Þ2 CO3ðaqÞ þ CO2ðaqÞ þ H2 OðlÞ $ 2ðNH4 ÞHCO3ðaqÞ $ 2ðNH4 ÞHCO3ðsÞ ð4Þ ðNH4 Þ2 CO3ðaqÞ $ NH2 CO2 NH4ðaqÞ þ H2 OðlÞ $ NH2 CO2 NH4ðsÞ

ð5Þ

NH4 HCO3ðsÞ $ NH3ðaqÞ þ H2 OðlÞ þ CO2ðlÞ

ð6Þ

The overall capture process operates in a similar fashion to typical solvent absorption systems. First the flue gas passes through a direct contact cooler to

102

K.A. Mumford et al.

reduce the temperature to between 2 and 10 °C. Then the cooled gas enters the bottom of the absorber and lean solvent, comprising water, ammonia (approximately 29 wt%) and a low concentrations of CO2 enters from the top. The temperature of the absorber is kept low to prevent ammonia slip. CO2 in the gas phase reacts according to the reaction scheme outlined in reactions (1)–(6) and the CO2 lean flue gas exits through the top of the absorber after passing through an acid water wash system to remove any residual ammonia. The CO2 rich solvent stream exits at the bottom of the absorber. If sufficient CO2 has been absorbed the solubility limit of the product species can be reached and a precipitate formed. As the absorber is run at low temperatures the reaction rate of this system is lower than traditional amine systems and hence a larger than typical absorber is often required. The CO2 rich solvent stream that has exited the absorber passes through a heat exchanger to raise its temperature and then enters the regenerator at the top where it is heated to between 100 and 150 °C where the CO2 is released. The use of the higher stripping temperature allows the CO2 to be released at a higher pressure. Alstom’s chilled ammonia process has been trialled at a number of sites through their extensive partnerships [37], including; • 20 MWe pilot at AEP’s Mountaineer Power Plant [37] • 20 MW application at Statoil Hydro’s Mongstad Test Centre in Norway [18].

2.2

Precipitating Amino Acid Salts

Amino acids are a subset of amines whereby the side chain contains a carboxylic acid group i.e. R ¼ OOC R0 . This class of solvent has advantages over other amine types due to their higher resistance to oxidative degradation and negligible vapour pressure, both of which lead to a more environmentally benign capture process [12]. Primary and secondary amino acids react with CO2 as per the scheme presented in Fig. 1. As shown, in these systems 2 mol of amino acid ðRNH2 Þ react with one mole of CO2 to form a carbamate ðRNHCO 2 Þ and a zwitterion species þ ðRNH3 Þ. The carbamate and zwitterion species may then undergo hydrolysis to form amino acid, bicarbonate ðHCO 3 Þ and zwitterion. These species may then either form carbonate at high pH, or be regenerated to the starting amino acid and CO2 via the addition of heat. When using amino acid salts for CO2 absorption neutralization using equimolar potassium hydroxide is usually necessary. In recent Fig. 1 Schematic of carbamate, bicarbonate and carbonate formation using amines (primary or secondary)

RNH3+ RNHCO2carbamate 2RNH2 + CO2 amine

2RNH3+ CO32carbonate

H 2O pH

heat

+

-

RNH3 HCO3 + RNH2 bicarbonate


103

years, interest has grown in the potential use of different amino acid salts for post-combustion CO2 capture. In fact, the CO2 absorption characteristics for common amino acids and their properties have been under extensive investigation [16, 19]. Siemens Energy also brought commercial interest to Amino Acid salt based solvents in 2010 through the development of their PostCap technology that was tested at a 2.5 MW pilot facility in 2009 [27] and later at Polk Power Plant [28]. Siemens reports that the energy consumption of this process is approximately 73% of the conventional monoethanolamine (MEA) process. More recently, developments that use amino acids have moved towards inducing precipitation of either the zwitterion ðRNH3þ Þ, a non-carbon dioxide containing species, or bicarbonate, a carbon dioxide containing species, with the view to driving the CO2 capture reaction forward. The relative proportion of the precipitate product depends upon a number of parameters including; the amine concentration; solution pH, temperature and the chemical stability of the carbamate formed. Primary amino acids tend to form stable carbamates and hence at equilibrium a higher proportion of these species is formed, as is the case with taurine [8]. Conversely, secondary amino acids, which have a higher level of steric hindrance, tend to form both the zwitterion and bicarbonate, as in the case of be proline, sarcosine and b-alanine [20]. Tertiary amino acids, with the highest steric hindrance, are generally unable to form stable carbamates at all and so all of the reaction products for these species are bicarbonates [16, 20]. It may also be noted from Fig. 1 that the maximum absorption of CO2 is achieved when all of the absorbed CO2 exists as bicarbonate as only one mole of amine is required per mole of CO2 [14]. Also of importance is the regeneration of the amino acid. It has been found that some amino acid solutions, particularly glycine and alanine, fail to produce a sufficiently lean solution after desorption which means that they do not commence reabsorption at a favourable point in the absorption curve [14]. Therefore careful consideration of the regeneration process, as well as absorption process, is valuable. Systems under development that utilise precipitating amino acid salts include; • The DECAB Plus process—precipitation of the protonated amine • CASPER (CO2-capture And Sulphur Precipitation for Enhanced Removal)— CO2 and SO2 precipitation process. 2.2.1

DECAB and DECAB Plus Process

The DECAB and DECAB Plus Processes are based on the precipitation of a primary amino acid, such as taurine, as a zwitterion [9]. The DECAB process refers to a process whereby a spray tower is used to handle solids during absorption. The precipitates are re-dissolved before desorption, which takes place in a conventional stripper [8]. The DECAB Plus process refers to a process where the supernatant is separated from the slurry and recycled to the absorber. This acts to enhance the

104

K.A. Mumford et al.

release of CO2 in the remaining liquid inside the regenerator via an induced pH swing when the precipitate dissolves upon heating [31]. The inventors report that upon loading with carbon dioxide, the solubility limit of the zwitterion may be reached, resulting in its precipitation from solution. As a consequence of the precipitation of the acidic species, the pH rises, resulting in the ability of the solvent to absorb more CO2 from the gas stream. When the solvent exits the absorber, the supernatant is separated from the slurry, and returned to the absorber. The slurry stream, which now has a reduced K+ concentration compared to the amino acid, is heated so that the precipitate dissolves, resulting in a reduction in pH. The lower pH promotes the hydrolysis of the carbamate species, resulting in the formation of amino acid salt and bicarbonate, which reverts to the zwitterion species and carbon dioxide at lower pH, thereby resulting in a reduction in energy required for solvent regeneration. Fernandez [9] explored this concept in detail with 4 M aqueous potassium taurate, a primary amine. They found that this process reduces the energy penalty by approximately 35% compared to the MEA baseline [9]. Large scale implementation of this technology has been limited due to two main reasons: – Design and optimisation of appropriate contacting equipment for the precipitating system has not yet been finalised – Although the chemical equilibrium of this system has been investigated in detail, the reaction kinetics have not yet been fully investigated [31]. 2.2.2

CASPER (CO2-Capture and Sulphur Precipitation for Enhanced Removal)

CASPER is a CO2 capture and sulphur precipitation process for simultaneous CO2 and SO2 removal from flue gas [21]. It uses a potassium amino acid based aqueous solution and precipitates sulphur as K2SO4. It was tested by CSIRO in 2013 at Loy Yang Power Station, Victoria, Australia and it was found that the performance of 3 M potassium b-alanate was satisfactory as compared to MEA in the same pilot plant facility.

2.3

Precipitating Carbonate Solvent Systems

Carbon dioxide capture processes using potassium carbonate have been implemented in industry for many years. Potassium carbonate solvents have a number of advantages including; low vapour pressure, low toxicity and minimal oxidative degradation. The main challenge associated with potassium carbonate capture systems when applied to post-combustion capture streams is the slow reaction rate. However it has been found that the use of promoters which have an active N–H


105

group can accelerate the reaction rate significantly [6, 38, 39]. Amines and, their subset, amino acids have been successfully utilised to achieve this goal. Similar to precipitating amino acid salt (AAS) based solvent systems, precipitating potassium carbonate systems may further improve CO2 absorption efficiency and reduce regeneration energy requirements [24, 40]. Generally, in promoted potassium carbonate systems, as the concentration of the amino acid is kept low (10 wt% amino acid in 30–45 wt% potassium carbonate [38]) it does not reach its solubility limit and the bicarbonate only is precipitated. Fernandes and co-workers [7] have developed a conceptual diagram that describes the interplay between the important reactions associated with this process. Specifically it describes how the species composition changes with acidity, water and amino acid concentration. A representation of this diagram is shown in Fig. 2. Lee et al. [17] and Thee [38] have investigated the interactions between carbonate and amino acid promotors extensively including the reaction rate and vapour–liquid–solid equilibria. Their results corroborated with the model of Fernandes [7]. Specifically, they found that there was a reaction between the dissolved carbon dioxide and amino acid [38] (K7 and K8) which acted to increase the overall reaction rate and additionally that the amino acid does form a carbamate [17] but only the bicarbonate is of sufficient concentration to reach its solubility limit and form a precipitate at post-combustion capture conditions. Therefore, if the solids containing stream is concentrated, and separated from the supernatant there will be a lower liquid flow to the regenerator and thus a lower latent energy requirement.

K4

H2CO3

KCO2

K1

CO2 + H2O

+H+

CO2(aq) + OH-

K5

+H2O +RNH3

K9

K10

K8

CO32-

K2

K7

ACO2H

K3

HCO3-

ACO2-

RNH3+

K6 RNH2

Fig. 2 General reaction scheme for amine, CO2, carbonate in aqueous solution, where Kn represents an equilibrium constant and A—amine [7]

106

2.3.1

K.A. Mumford et al.

UNO MK3

The CO2CRC and the University of Melbourne have been investigating promoted potassium carbonate solvent systems for CO2 capture for a number of years. The most recent development has been a precipitating promoted potassium carbonate system termed UNO MK3 and this process is being commercialised by UNO Technology Pty Ltd. The process was developed in the laboratory at bench scale and then tested in pilot plants with a capacity of up to 1 tonne/day of CO2 using both synthetic and actual flue gas [33, 34]. In the precipitating system design, potassium bicarbonate is precipitated from a promoted potassium carbonate solvent following CO2 absorption and subsequent cooling. The precipitate can then be separated from the liquid phase via a hydrocyclone for selective regeneration of the KHCO3 species. A process flow diagram of the process can be found in Fig. 3. In this way, less water is passed to the regeneration stage and thus drives down the energy requirements from over 3 GJ/tonne CO2 for a liquid based system to less than 2.5 GJ/tonne for a precipitating based system [2]. The most widely reported system has been the use of 10 wt% potassium glycine, as a promotor, in a 40–45 wt% K2CO3 solution. The addition of glycine was found

Fig. 3 UNO MK 3 process flow diagram [33]


107

to improve the CO2 recovery rate by up to 6 times in a laboratory based pilot plant trial whilst also slightly increasing the pressure drop and holdup which was thought likely due to a reduction of the surface tension of the solvent [33]. A glycine promoted precipitating K2CO3 solvent process has also been demonstrated using a pilot plant that captured CO2 from flue gas generated from a brown coal based pulverised coal combustion power station in Victoria, Australia [32].

2.3.2

Shell Carbonate Slurry Process

Shell have been developing a carbonate slurry process that uses potassium carbonate solvent with a crystallisation and concentration step (see Fig. 4) following CO2 absorption [23, 41]. These studies have reported a potential *50% reduction in regeneration energy and the potential for lower nitrosamine emissions compared to traditional amine solvent processes [22]. The Shell carbonate process has been tested using a bench scale pilot plant that has a capacity of 25 kg/day CO2. An accelerator is used to enhance the mass transfer of CO2 to the liquid phase. The critical steps of this process were reported to be the crystal formation and solids handling equipment used to operate as a precipitating solvent process.

Fig. 4 Process flow diagram of the shell carbonate slurry process [23]

108

K.A. Mumford et al.

3 Non-aqueous and Liquid-Liquid Phase Change Solvents for CO2 Capture 3.1

Non-aqueous Solvents

Recently there has been focus on the development of non-aqueous solvent systems for CO2 capture applications. Typically these systems utilise well accepted solvents such as MEA or DEA, but instead of being dissolved in water, they are dissolved in organic solvents, such as ethanol or butanol. This change impacts the chemical reactivity of these solvents considerably and has the potential to provide significant benefits, including; • Reduced corrosivity • Increased reaction rates. – solubility of CO2 in alcohols such as ethanol is significantly higher than in water [46]. – alcohols such as ethanol can facilitate chemical reactions between solutes and CO2 [46]. • Decreased regeneration temperatures • As organic solvents are used, the formation of bicarbonate can’t occur and an unstable carbamate is formed instead. This species, if it is unstable, requires less energy to regenerate [46]. Additionally, as lower temperatures are required for regeneration, waste heat, from other sources may also be used for this duty. Recently, developments in these solvent systems have focused on loading the solvent with CO2 until a precipitate is formed. However, not all systems form crystal like precipitates as shown in Table 1. As indicated, alkanolamines such as MEA and DEA form gum structures rather than crystals upon absorption of CO2, whereas alkylamines such as TETA or TEPA almost always give crystals. It is postulated that the hydroxyl groups of MEA and DEA are likely causing this due to their ability to form strong hydrogen bonded networks. Therefore this aspect needs to be considered whilst selecting appropriate systems for use. Zheng et al. [46] performed an extensive study investigation into different solvent combinations, and in particular found that TETA in ethanol was a strong candidate. They found that TETA in ethanol exhibits a higher reaction rate and absorption capacity as compared to TETA in water solutions, additionally 81.8% of the absorbed CO2 in the solid phases was TETA carbamate, whereas in water no precipitate is observed. Additionally, the decomposition of TETA-carbamate, releasing CO2 commenced at approximately 90 °C. Other systems investigated include DEA in various ionic liquids (1-ethyl-3-mthylimidazolium bis(trifluoromethylsulfonyl)imide) [emim][Tf2N], 1-butyl-3-mthylimidazolium bis(trifluoromethylsulfonyl)imide) [bmim][Tf2N],

Ethanol 1-butanol 1-pentanol 1-hexanol Iso-octanol IL [emim][Tf2N] [bmim][Tf2N] [hmim][Tf2N]

Gum Gum Gum Gum Gum

[46] [46] [46] [46] [46]

Monoethanolamine (MEA) Gum [46] Gum [46] Gum [46] Gum [46] Gum [46] Powder [13]

Diethanolamine (DEA)

Table 1 Physical state of products of amine-CO2 reactions in various solvents

Powder Powder Powder Powder Powder

[46] [46] [46] [46] [46]

Triethylenetetramine (TETA)

Powder Powder Powder Powder Powder

[46] [46] [46] [46] [46]

Tetraethylenepentamine (TEPA)

Phase Change Solvents for CO2 Capture Applications 109

110

K.A. Mumford et al.

1-hexyl-3-mthylimidazolium bis(trifluoromethylsulfonyl)imide) [hmim][Tf2N]) [13] which upon loading with CO2 formed a DEA carbamate. It is also interesting to note that in the case of [emim][Tf2N] and [hmim][Tf2N], the precipitate rose to the surface quickly due to the hydrophobicity of the ionic liquid. Therefore it is postulated that in an industrial setting separation of the phases would somewhat simpler as compared to other potential solvent systems [13]. AMP/DEA in 1,2-propandiol/ethanol [5] formed an AMP-carbamate which was found to regenerate at 80 °C, and AMP in N-methyl-2-pyrrolidone (NMP) and triethylene glycol dimethyl ether (TEGDME) which also formed AMP carbamates was found to regenerate at 75 and 90 °C respectively [36]. It is worth noting that the TETA-carbamate regeneration temperature was higher than that of DEAcarbamate and AMP-carbamate, which suggests TETA-carbamate is more stable than DEA-carbamate and AMP-carbamate [46].

3.2

Liquid-Liquid Phase Change Solvents

A number of processes have been developed with liquid-liquid phase change solvent properties. When the solvent or solvent blend absorbs CO2, two immiscible liquid phases are formed; one that is rich in CO2 and another that is low in CO2. The CO2 rich phase can be separated by gravity and then regenerated with a much lower regeneration energy requirement. A number of different commercial processes and solvent blends have been proposed.

3.2.1

DMXTM Process (IFP Energies Nouvelles)

A demixing process for CO2 capture, known as DMXTM, was originally developed by IFP Energies noevelles. This process uses demixing solvents to form two immiscible liquid phases following CO2 absorption at specific CO2 loading and temperature conditions [30]. The apparent CO2 loading achievable with this process is comparable to the standard amine process however with the demixing solvent the salts are formed with CO2 concentrated in the resulting lower liquid phase. This results in the lower or heavy aqueous liquid phase having a high CO2 loading which can be separated via a decantation step and then only this CO2 rich phase needs to be sent to the regeneration column. The light amine phase, which has a very low CO2 content, is mixed with the regenerated solvent from the regeneration column and then returned to the absorber. Refer to Fig. 5 for a process flow diagram of the DMXTM process. The cyclic capacity of the process is increased and the mass of solvent to be regenerated is reduced resulting in a lower energy penalty. DMX-1 has been used as the demixing solvent due to its thermodynamic capacity as well as favourable degradation properties. The reported energy consumption for this process is 2.3 GJ/tCO2 with reports that this could be reduced to as low as 2.1 GJ/tCO2 following optimised heat integration, indicating significant operating cost


111

Fig. 5 Process flow diagram of the DMXTM process [30]

reductions. The capital cost or CAPEX of this process is estimated to be similar to a standard amine capture plant. Although there are reported capital cost savings from operating with a higher capacity solvent (such as a smaller diameter column, pumps and heat exchangers) this benefit is counterbalanced by the cost of the decanter equipment and higher column height from reduced reaction kinetics.

3.2.2

Thermomorphic Biphasic Solvent (TBS) Systems

Thermomorphic biphasic solvent (TBS) systems release absorbed CO2 in the regeneration column at much lower temperatures (e.g. 80 °C or lower) than standard amine solvent processes (which typically operate around 120–130 °C). These solvent systems also have a high CO2 capacity (e.g. 0.9 mol CO2/mol absorbant) [44]. The heating and agitation during the regeneration process results in two liquid phases being formed. When returned to the absorber the system reverts back to a single liquid phase. Lipophilic amines, such as N,N-dimethylcyclohexylamine (DMCA) and dipropylamine (DPA) are typically used in this process. When the organic phase is formed in the regenerator, it acts as the extractant which removes the amine from the aqueous phase resulting in dissociation of the bicarbonate and carbamate in the loaded aqueous phase. These lipophilic amines have high chemical stability and the use of waste heat can be considered as the regeneration temperature is low at around 80 °C. This concept has been tested in a bench-scale absorption column (2.5 cm diameter with structured packing) and stirred tank regenerator [45]. Three solvent combinations were studied with regeneration temperatures from 50 to 95 °C resulting in energy consumption reductions of more than 35% compared to the conventional MEA based solvent process (Fig. 6).

112

K.A. Mumford et al.

Fig. 6 Process flow diagram of a thermomorphic biphasic solvent (TBS) system [45]

3.2.3

Mixed Amines—DEEA/MAPA System

Amine blends and phase change solvents offer promising improvements for the CO2 absorption process [25]. When some amine solvents are mixed together they can form two phases after CO2 absorption and the cyclic loading can be much higher than the standard MEA process. An example of an amine blend with these properties occurs when 2-(diethylamino) ethanol (DEEA) and 3-(methylamino) propylamine (MAPA)are mixed together. DEEA is a tertiary alkanolamine and MAPA has two amine functional groups (primary and secondary). The amine mixture will lower the overall heat of absorption while the primary/secondary amine will improve the capture rate and the tertiary amine will improve the capture capacity of the solvent. A 5M DEEA/2M MAPA blend will form two phases upon loading with CO2 and after separation nearly all the CO2 is present in the lower phase which contains the MAPA. Only this heavy phase needs to be sent to the regenerator leading to lower operational costs for stripping CO2 from the solvent. Arshad et al. studied the performance of this solvent blend and reported that it has a lower regeneration energy at similar temperature compared to the standard MEA process [3, 4]. A pilot plant at the Gloshaugen (NTNU/SINITEF) facility has been used to test this solvent blend for post-combustion CO2 capture [29]. A process flow diagram of this pilot plant can be found in Fig. 7. The pilot plant performed well with no issues with the high viscosity or foaming. Due to the presence of the


113

Fig. 7 Process flow diagram of the Gloshaugen pilot plant testing the DEEA/MAPA solvent process [29]

tertiary amine, absorption of CO2 was relatively fast, CO2 stripping was easier and the reboiler was operated at a lower specific reboiler duty and temperature than the standard 30 wt% MEA process. Issues that need to be considered with these blended amine process are solvent degradation and solvent volatility in order to avoid further environmental issues and reduce solvent replacement requirements. Other blends of amines that result in phase change solvents have been proposed, including 2 M 1,4-butanediamine (BDA) and 4 M 2-(diethylamino)-ethanol (DEEA) by Xu et al. [43]. Another non-aqueous amine solvent process, called the “Self-Concentrating Absorbent CO2 Capture Process” is being developed by 3H Company [26]. When the proprietary amine solvent mixture reacts with CO2, the mixture separates into two distinct phases: a CO2-rich liquid phase and a dilute lean phase. The process is proposed to reduce the total regeneration energy by as much as 70%. The solvent volume required would be lower than an MEA system as the solvent has a high working capacity which also results in lower pumping requirements, lower auxiliary power demands, and reduced equipment size [26]. As the solvent is non-aqueous corrosion problems are also minimised. Demonstration of this process is proposed at an E-ON power plant in the United States as a next stage of commercialization development (Fig. 8).

114

K.A. Mumford et al.

Fig. 8 Self-concentrating amine absorbent process concept [26]

4 Conclusions This chapter provides an update on recent developments within a rapidly developing field, namely Phase Change Solvents for CO2 capture. This has included both solid-liquid and liquid-liquid system developments within academia and industry. It was shown that the benefit of these energy efficient solvent systems over traditional liquid only systems stems from their potential to separate the CO2 rich phase from the CO2 lean phase, thus improving solvent capacities, reaction kinetics and reducing energy penalties.

References 1. Alstom (2007) Chilled ammonia-based wet scrubbing for post-combustion CO2 capture. In: 401/021507 DNRN (ed) 2. Anderson C, Harkin T, Ho M et al (2013) Developments in the CO2CRC UNO MK 3 process: a multi-component solvent process for large scale CO2 Capture. In: Dixon T, Yamaji K (eds) Ghgt-11. Elsevier Science, Amsterdam, pp 225–232 3. Arshad MW, Fosbøl PL, Von Solms N et al (2013) Freezing point depressions of phase change CO2 solvents. J Chem Eng Data 58:1918–1926 4. Arshad MW, Von Solms N, Thomsen K et al (2013) Heat of absorption of CO2 in aqueous solutions of DEEA, MAPA and their mixture. Energy Procedia, 1532–1542 5. Barzagli F, Mani F, Peruzzini M (2013) Efficient CO2 absorption and low temperature desorption with non-aqueous solvents based on 2-amino-2-methyl-1-propanol (AMP). Int J Greenh Gas Control 16:217–223 6. Cullinane JT, Rochelle GT (2004) Carbon dioxide absorption with aqueous potassium carbonate promoted by piperazine. Chem Eng Sci 59:3619–3630 7. Fernandes D, Conway W, Burns R et al (2012) Investigations of primary and secondary amine carbamate stability by H-1 NMR spectroscopy for post combustion capture of carbon dioxide. J Chem Thermodyn 54:183–191


115

8. Fernandez ES, Goetheer ELV (2011) DECAB: process development of a phase change absorption process. 10th international conference on greenhouse gas control technologies. Vol 4, pp 868–875 9. Fernandez ES, Heffernan K, Van Der Ham LV et al (2013) Conceptual design of a novel CO2 capture process based on precipitating amino acid solvents. Ind Eng Chem Res 52: 12223–12235 10. Gal E (2008) ultra cleaning of combustion gas including the removal of Co2. In:Google Patents 11. Gal E, Jayaweera I (2010) Chilled ammonia based CO2 capture system with water wash system. In: Google Patents 12. Goetheer ELV, Nell L (2009) First pilot results from TNO’s solvent development workflow. Carbon Capture J 8:2–3 13. Hasib-Ur-Rahman M, Siaj M, Larachi F (2012) CO2 capture in alkanolamine/ room-temperature ionic liquid emulsions: a viable approach with carbamate crystallization and curbed corrosion behavior. Int J Greenh Gas Control 6:246–252 14. Hook RJ (1997) An investigation of some sterically hindered amines as potential carbon dioxide scrubbing compounds. Ind Eng Chem Res 36:1779–1790 15. Hooper B, Stevens G, Kentish S et al (2011) A process and plant for removing acid gases 16. Kumar PS, Hogendoorn JA, Feron PHM et al (2003) Equilibrium solubility of CO2 in aqueous potassium taurate solutions: part 1. crystallization in carbon dioxide loaded aqueous salt solutions of amino acids. Ind Eng Chem Res 42:2832–2840 17. Lee A, Wolf M, Kromer N et al (2015) A study of the vapour-liquid equilibrium of CO2 in mixed solutions of potassium carbonate and potassium glycinate. Int J Greenh Gas Control 36:27–33 18. Lombardo G, Agarwal R, Askander J (2014) Chilled ammonia process at technology center Mongstad—first results. Energy Procedia 51:31–39 19. Majchrowicz ME, Brilman DWF (2012) Solubility of CO2 in aqueous potassium L-prolinate solutions-absorber conditions. Chem Eng Sci 72:35–44 20. Majchrowicz ME, Brilman DWF, Groeneveld MJ (2009) Precipitation regime for selected amino acid salts for CO(2) capture from flue gases. Greenhouse Gas Control Technol 9(1):979–984 21. Misiak K, Sanchez CS, Van Os P et al (2013) Next generation post- combustion capture: combined CO2 and SO2 removal. Energy Procedia 37:1150–1159 22. Moene R (2013) CCS—innovation in shell (innovation is bringing an insightful ideal successfully to the market). In: CLIMIT SUMMIT 2013. Oslo 23. Moene R, Schoon L, Van Straelen J et al (2013) Precipitating carbonate process for energy efficient post-combustion CO2 capture. Energy Procedia 37:1881–1887 24. Moene R, Schoon L, Van Straelen J et al (2013) Precipitating carbonate process for energy efficient post-combustion CO2 capture. Ghgt-11 37:1881–1887 25. Monteiro JGMS, Pinto DDD, SaH Zaidy et al (2013) VLE data and modelling of aqueous N, N-diethylethanolamine (DEEA) solutions. Int J Greenhouse Gas Control 19:432–440 26. Mosser M (2012) Post-combustion CO2 capture for existing PC boilers by self-concentrating amine absorbent. In: National Energy Technology Laboratory 27. National Energy Technology Laboratory (2013) DOE/NETL advance carbon dioxide capture R&D program: technology update. In: Department of Energy, p 716 28. Pennenergy (2014) Clean coal: carbon capture pilot begins at Polk IGCC plant 29. Pinto DDD, Knuutila H, Fytianos G et al (2014) CO2 post combustion capture with a phase change solvent. Pilot plant campaign. Int J Greenhouse Gas Control 31:153–164 30. Raynal L, Bouillon PA, Gomez A et al (2011) From MEA to demixing solvents and future steps, a roadmap for lowering the cost of post-combustion carbon capture. Chem Eng J 171:742–752 31. Sanchez-Fernandez E, De Miguel Mercader F, Misiak K et al (2013) New process concepts for CO2 capture based on precipitating amino acids. Energy Procedia

116

K.A. Mumford et al.

32. Smith K, Harkin T, Mumford K et al (2016) Outcomes from a precipitating potassium carbonate solvent absorption pilot plant for CO2 capture from a brown coal fired power station in Australia. Fuel Processing Technology. doi:10.1016/j.fuproc.2016.08.008 33. Smith K, Lee A, Mumford K et al (2015) Pilot plant results for a precipitating potassium carbonate solvent absorption process promoted with glycine for enhanced CO2 capture. Fuel Process Technol 135:60–65 34. Smith K, Xiao G, Mumford K et al (2013) Demonstration of a concentrated potassium carbonate process for CO2 capture. Energy Fuels 28:299–306 35. Svendsen HF, Tobiesen FA, Mejdell T et al (2008) Method for capturing Co2 from exhaust gas. In: Google Patents 36. Svensson H, Hulteberg C, Karlsson HT (2014) Precipitation of AMP carbamate in CO2 absorption process. Energy Procedia 63:750–757 37. Telikapalli V, Kozak F, Francois J et al (2011) CCS with the Alstom chilled ammonia process development program—field pilot results. 10th international conference on greenhouse gas control technologies 4:273–281 38. Thee H, Nicholas NJ, Smith K et al (2014) A kinetic study of CO2 capture with potassium carbonate solutions promoted with various amino acids: glycine, sarcosine and proline. Int J Greenh Gas Control 20:212–222 39. Thee H, Suryaputradinata Y, Mumford KA et al (2012) A kinetic and process modeling study of CO2 capture with MEA-promoted potasium carbonate solutions. Chem Eng J 210 40. Van Straelen JPT (2011) Process for the removal of carbon dioxide from a gas. In: Google Patents 41. Van Straelen JPT (2015) Process for the removal of carbon dioxide from a gas. In: Google Patents 42. Versteeg GF, Kumar PS, Hogendoorn JA et al (2011) Method for absorption of acid gases. In: Google Patents 43. Xu Z, Wang S, Qi G et al (2014) CO2 absorption by biphasic solvents: comparison with lower phase alone. Oil Gas Sci Technol 69:851–864 44. Zhang J, Nwani O, Tan Y et al (2011) Carbon dioxide absorption into biphasic amine solvent with solvent loss reduction. Chem Eng Res Des 89:1190–1196 45. Zhang J, Qiao Y, Agar DW (2012) Improvement of lipophilic-amine-based thermomorphic biphasic solvent for energy-efficient carbon capture. Energy Procedia 23:92–101 46. Zheng SD, Tao MN, Liu Q et al (2014) Capturing CO2 into the precipitate of a phase-changing solvent after absorption. Environ Sci Technol 48:8905–8910

Aqueous Amino Acid Salts and Their Blends as Efficient Absorbents for CO2 Capture Azmi Mohd Shariff and Muhammad Shuaib Shaikh

Abstract The increase in global population and industrialization has led to an increase in global energy consumption exponentially. Over 85% of global energy is supplied by burning fossil fuel, which releases large volume of CO2 emissions in the atmosphere. Increasing of CO2 emissions is the major cause for the catastrophic climate change, which has led to increased demand for efficient and effective CO2 capture. CO2 absorption by chemical solvents is the most widely used technique commercially nowadays. Alkanolamine solvents such as monoethanolamine (MEA) and methyldiethanolamine (MDEA) are the most commonly used absorbents for CO2 removal from various gas streams. However, it is well known that these solvents suffer from variety of drawbacks such as limited CO2 loading capacity, equipment corrosion, toxic nature and highly volatile. Moreover, these absorbents are easily degradable, require high regeneration energy, and cause flooding problems in the operation. Therefore, better and efficient solvents should be searched for the removal of CO2 from exhaust gas streams. Aqueous amino acid salts and their blends are the promising solvents for CO2 capture as compared to alkanolamine. In this chapter, amino acid salts and their blends are introduced and their performance analysis as potential solvents for commercial possibilities are discussed. Based on the analysis, these absorbents show superior performance as an alternative to the conventional alkanolamines for CO2 capture. These solvents are environmental friendly with higher CO2 loading capacity, faster reaction kinetics and require less regeneration energy compares to the commercial amines. Besides, these solvents are non-volatile, less corrosive and oxidative stable. Moreover, aqueous amino acid salts are more effective by blending with additives such as piperazine.

A.M. Shariff (&) M.S. Shaikh Research Centre for CO2 Capture (RCCO2C), Department of Chemical Engineering, Universiti Teknologi PETRONAS, 31750 Tronoh, Perak, Malaysia e-mail: [email protected] © Springer International Publishing AG 2017 W.M. Budzianowski (ed.), Energy Efficient Solvents for CO2 Capture by Gas–Liquid Absorption, Green Energy and Technology, DOI 10.1007/978-3-319-47262-1_6

117

118

A.M. Shariff and M.S. Shaikh

Nomenclature Abbreviations %wt. AAAS AAS ALA AMP ARG ARG ASN ASP ASTM CAPEX CASPER CO2 CYS DEA DGA DIPA GHG GLN GLU GLY H 2S HIS ILU K-AABA K-ALA K-ASN K-BALA K-DiMGLY K-GLU K-GLY K-LYS kPa K-PRO K-SAR K-SER K-TAU K-THR LEU Li-PRO Li-SAR LYS

Percent by weight Aqueous amine amino acid salt Amino acid salt Alanine 2-amino-2-methyl-1-propanol Arginine Arginine Asparagine Aspartate American society for testing and materials Capital expenditure CO2 capture and sulfur precipitation for enhanced removal Carbon dioxide Cysteine Diethanolamine Diglycolamine Diisopropylamine Greenhouse gas Glutamine Glutamate Glycine Hydrogen sulfide Histidine Isoleucine Potassium salt of DL-a-amino butyric acid Potassium salt of alanine Potassium salt of L-asparagine/asparaginate Potassium salt of b-alanine Potassium salt of diethyl or dimethylglycine Potassium salt of glutamate Potassium salt of glycine Potassium salt of lysine Kilo pascal Potassium salt of proline Potassium salt of sarcosine Potassium salt of serine Potassium salt of taurine Potassium salt of threonine Leucine Lithium salt of proline Lithium salt of sarcosine Lysine

Aqueous Amino Acid Salts and Their Blends …

MDEA MEA MET MET Na-ALA NA-BALA Na-GLY Na-PH Na-PRO Na-SAR Na-SO3 Na-TAU Na-VO3 NH3 NOAA OPEX pH PHE PostCap ppm PRO PZ SARMAPA SER SO2 TEA THR TIPA TRP TYR VAL VLE

N-methyldiethanolamine Monoethanolamine Methionine Methionine Sodium salt of alanine Sodium salt of b-alanine Sodium salt of glycine Sodium phenolate Sodium salt of proline Sodium salt of sarcosine Sodium sulfite Sodium salt of taurine Sodium metavanadate Ammonia National Oceanographic and Atmospheric Administration Operating expenditure Power of hydrogen ion Phenylalanine Post combustion capture technology Parts per million Proline Piperazine Sarcosine with 3-(methylamino propylamine) Serine Sulphur dioxide Triethanolamine Threonine Tri-isopropanolamine Tryptophan Tyrosine Valine Vapor liquid equilibrium

Units and Symbols Ea k2 kov LD50 M T a

Activation energy (Kg/mol) Forward second order reaction rate (m3 mol-1S-1) Overall reaction rate constant (S-1) Lethal dose (mg/Kg) Molarity (mol/litre) Temperature (K/°C) Loading (mol/mol)

119

120


1 Introduction The energy demand is increasing worldwide after an industrial revolution. About 85% of global energy demand is fulfilled by burning fossil fuels to generate electricity in power plants [1, 2]. Therefore, the huge amount of greenhouse gases (GHG) are released in the atmosphere, causing a global environmental problem, termed as global warming. Carbon dioxide (CO2) is the main contributor to global warming [2, 3]. According to NOAA, 2014, the atmospheric CO2 concentration is 396.21 ppm, which exceeds the tolerable limit (350 ppm). It is expected that the CO2 concentration may rise beyond 400 ppm by the year 2016, if the suitable mitigation techniques are not established [4]. Based on the rising level of CO2 in the atmosphere, the average temperature may increase globally from 1.4 to 5.8 °C by the year 2100, according to the different climate prediction models. The harmful effects of global warming includes the rise of sea level, melting of glaciers and ice caps, climate change, and infrastructure destruction [5]. Hence, it has become a global panic to lessen the CO2 emissions by introducing the effective technologies [6]. In this connection, a lot of research work has been carried out worldwide. As the results, various technologies have been established such as absorption, adsorption, membrane and cryogenic processes [7–9]. Among these, the most extensively functioning technology is absorption by chemical solvents due to the great deal of research carried out on liquid solvents and their practical applicability [2, 10–14]. The conventional chemical solvents used commercially for CO2 absorption are alkanolamines (also called as amines) such as monoethanolamine (MEA), diethanolamine (DEA), diisopropanolamine (DIPA), triethanolamine (TEA), methyldiethanolamine (MDEA) [15–18]. Though these alkanolamines have been used widely for CO2 absorption, however, several shortcomings have been identified and reported [19, 20]. The problems concerned with these solvents include shorter life due to poor resistance to oxidative and thermal degradation. Corrosion of equipment and flow lines, loss of solvent due to high vapor pressure, higher energy of regeneration and toxicity are among the key problems allied with such solvents. Moreover, these solvents have inadequate cyclic CO2 loading capacity, except for tertiary amines such as MDEA, which has higher cyclic capacity compared to conventional MEA [18, 21–25]. These drawbacks are summarized in Table 1 with their effects. The shortcomings offered by these solvents confine their use for commercial applications [20, 26]. Therefore, it is mandatory for the scientists and researchers to discover the new solvent systems, which can compensate the shortcomings of the existing ones. The amino acid salt solutions have emerged as one of the promising solvents for CO2 capture as compared to amines. Amino acid salt solutions are environmental friendly with higher CO2 loading capacity, faster reaction kinetics and require less regeneration energy. Besides, these solvents are non-volatile, non-flammable, less


121

Table 1 Drawbacks of amines and their possible effects [15, 18, 21–25, 28–30, 32, 96, 116] Drawbacks of amines

Effects

Limited CO2 loading (Except tertiary amines i.e., MDEA, which has better loading than MEA, but still less than amino acid salts)

• High volume of solvent required • Increase the overall solvent cost • Increase the size of the plant equipment and accessories • Overall increase in CAPEX and OPEX • Leads to high circulation rate of the solvent • Increase the energy consumption • Increase the dimensions of the process equipment • Increase the CAPEX and OPEX • Increase the operating cost due to high energy requirements

Low cyclic capacity (Except MDEA)

High regeneration energy (Except MDEA which require less regeneration energy than MEA, but still higher than amino acid salts) High circulation rate

Thermal and oxidative degradation (Except MDEA, which has less degradation than MEA) Low absorption rate

Formation of heat stable salts Low surface tension Corrosive

Toxic High volatility and flammable

• Increase the dimensions of equipment (solvent heat exchangers, amine pumps, absorber etc.) • High energy consumption in plant operation • Increase solvent replacement costs • Shorter life of the solvent, thus increase the CAPEX • Increase the size of the absorber and other plant accessories • Require large volume of solvent • Increase the CAPEX and OPEX • Enhance the corrosion of equipment • Require extra cost for corrosion repair • Cause flooding and foaming • Lower the mass transfer efficiency • Decrease equipment and accessories life • Increase the cost required for restoration of corroded equipment • Increase maintenance cost due to unplanned downtime • Health and environmental risks • High vapor loss of the solvent • Increase the solvent replacement cost • Increase the overall CAPEX and OPEX • Fire and other related hazards due to flammable nature

corrosive and oxidative stable [27–31]. Several studies have been conducted to authenticate the performance of these solvents. In this chapter, the performance analysis of various aqueous amino acid salts and their blends have been carried out based on the literature data for evaluating their commercial possibilities.

122


2 Chemistry The amino acids are known as amphiprotic in nature, which can either donate or accept the proton. Thus, the amino acids can act as acid or a base, therefore, called as ampholytes. The structures of amino acids contain at least one basic carboxyl or sulphonyl group and one acid amino group. In an aqueous solution, amino acids exist as a zwitterion (form II in reaction 1) in the absence of the other solutes. The compounds exist as a zwitterion when they are positively and negatively charged due to presence of separate functional groups. When the acid is added to the amino acid solution, the zwitterion picks up the proton (form I in reaction 1), while the addition of base to amino acid solution removes a proton from the ammonium group and leaves the molecule with final negative charge (form III in reaction 1). The zwitterion and deprotonated form of amino acid is shown in reaction 1 [27, 32]. H þ

H þ

HO2 CRNH3þ $ O2 CRNH3þ $ O2 CRNH2 ðIÞ ðIIÞ ðIIIÞ

ð1Þ

At this stage, the amino acid solution contains deprotonated amino group, which is actually reactive with acid gases such as CO2. There are twenty types of amino acids, which are listed in Table 2 along with corresponding chemical formula and linear structure. Table 2 Types of amino acids [113] Amino acid

Chemical formula

Linear structure

Molecular weight (g/mol)

Alanine (ALA) Arginine (ARG)

C3H7NO2 C6H14N4O2

89.0935 174.2017

Asparagine (ASN) Aspartate (ASP) Cysteine (CYS) Glutamate (GLU) Glutamine (GLN) Glycine (GLY) Histidine (HIS)

C4H8N2O3 C4H7NO4 C3H7NO2S C5H9NO4 C5H10N2O3 C2H5NO2 C6H9N3O2

Isoleucine (ILE)

C6H13NO2

Leucine (LEU) Lysine (LYS) Methionine (MET)

C6H13NO2 C6H14N2O2 C5H11NO2S

CH3–CH(NH2)–COOH HN=C(NH2)–NH–(CH2)3–CH (NH2)–COOH H2N–CO–CH2–CH(NH2)–COOH HOOC–CH2–CH(NH2)–COOH HS–CH2–CH(NH2)–COOH HOOC–(CH2)2–CH(NH2)–COOH H2N–CO–(CH2)2–CH(NH2)–COOH NH2–CH2–COOH NH–CH=N–CH=C–CH2–CH (NH2)–COOH CH3–CH2–CH(CH3)–CH(NH2)– COOH (CH3)2–CH–CH2–CH(NH2)–COOH H2N–(CH2)4–CH(NH2)–COOH CH3–S–(CH2)2–CH(NH2)–COOH

132.1184 133.1032 121.1590 147.1299 146.1451 75.0669 155.1552 131.1736 131.1736 146.1882 149.2124 (continued)


123

Table 2 (continued) Amino acid

Chemical formula

Linear structure

Molecular weight (g/mol)

Phenylalanine (PHE) Proline (PRO) Serine (SER) Threonine (THR) Tryptophan (TRP)

C9H11NO2

Ph–CH2–CH(NH2)–COOH

165.1900

C5H9NO2 C3H7NO3 C4H9NO3 C11H12N2O2

115.1310 105.0930 119.1197 204.2262

Tyrosine (TYR) Valine (VAL)

C9H11NO3 C5H11NO2

NH–(CH2)3–CH–COOH HO–CH2–CH(NH2)–COOH CH3–CH(OH)–CH(NH2)–COOH Ph–NH–CH = C–CH2–CH(NH2)– COOH HO–Ph–CH2–CH(NH2)–COOH (CH3)2–CH–CH(NH2)–COOH

181.1894 117.1469

3 CO2 Capture Although the alkanolamines are very popular and widely used in industries for acid gas removal from variety of sour gas streams, however, the drawbacks listed in Table 1 restrict their applications for CO2 removal. The major downsides of alkanolamines such as high-energy consumption and high process cost stimulate the interest to search another type of solvents for CO2 capture that would minimize the energy usage and the cost. Amino acid salt solutions as effective substitutes have been investigated extensively for the removal of CO2. The reactivities of amino acid salts are similar to that of amines because of having identical functional groups in their molecules [27, 28, 30]. However, based on the various attractive and valuable characteristics identified by extensive research on these systems, amino acid salt solutions have been proposed as the potential alternative to amine-based systems [20]. They have the adequate life cycle because of possessing good ability to resist thermal and oxidative degradation [27–30]. The presence of ionic structure in amino acid salt systems lowers the volatility to almost negligible value, which prevent their loss even at high-temperature operations. Moreover, these solvents are able to regenerate, expected to be less corrosive, eco-friendly and easily available at the commercial level. Another benefit of amino acid salts is high surface tension, which makes such solvents appropriate for membrane gas absorption system [27, 29, 30, 33–35]. The summary of the benefits of amino acid salts as potential solvents for CO2 removal is given in Table 3 with their possible industrial impact. The CO2 reacts with the amino acid salt solutions through a zwitterion mechanism because of the similar functional group as amines [27]. The reaction of CO2 with amino acid salt solution is shown in Fig. 1.

124


Table 3 Benefits of amino acid salts and their potential impact [27–30, 32–35, 39–42] Benefits of amino acid salts

Potential impact

High CO2 loading

• • • • • • • • •

Improved cyclic capacity

Low circulation rate

Resistant to thermal and oxidative degradation High absorption rate

High surface tension Less corrosive

Less viscous

Environmental friendly Negligible volatility and not flammable

Less volume of solvent required Lower solvent circulation rates Reduce the equipment size Decrease the CAPEX and OPEX Reduce circulation rates of the solvent Decrease the energy consumption Reduce the equipment and accessories size Decrease the CAPEX and OPEX Reduce the dimensions of equipment (solvent heat exchangers, amine pumps, absorber etc.) • Less electricity required for pumps and other equipment • Overall cost reduction • Reduce solvent replacement costs • Increase solvent life and decrease the operating cost • Decrease the size of the absorber and other plant accessories • Require less volume of the solvent • Reduce the CAPEX and OPEX • Prevent flooding and foaming • Enhance mass transfer rates and efficiency • Increase equipment and accessories life • Reduce the cost required for restoration of corroded system • Reduce overall cost • Increase the pumping efficiency • Enhance mass transfer • Reduce pumping cost • Minimum health and environmental effects • No vapor loss and no fire hazards • Reduces the solvent replacement cost • Reduces over all CAPEX and OPEX

Fig. 1 Reaction of CO2 with aqueous amino acid salt [27]

In the reaction scheme shown in Fig. 1, initially, the zwitterion and carbamate are formed during the absorption reaction. Later, the carbamate undergoes hydrolysis, which results in the formation of deprotonated amino acid and bicarbonate/carbonate depending upon the pH of the solution. The deprotonated amino acid later can react with CO2. The extent of carbamate hydrolysis can be


125

determined by certain parameters such as concentration, carbamate stability and pH of the solution [36–38]. The reaction equilibria shown in Fig. 1 favors the formation of carbamate and bicarbonate at low temperature conditions. However, at high temperatures, this equilibria favors the liberation of amino acid and CO2.

4 Absorption Processes The first amino acid used for the removal for CO2 was glycine (GLY) in Giammarco-Vetrocoke process. In this process, GLY was used as an activator in the alkali carbonate solution to improve the CO2 removal performance. As a result, better absorption performance was found together with less amount of steam required for solvent regeneration as compared to the amines [15]. Similarly, another process based on amino acid salts known as Alkacid process, has been developed by BASF. There are three major process variations in this process, Alkacid “M”, Alkacid “dik” and Alkacid “S”. In these process variations, Alkacid “M” uses the sodium salt of alanine (Na-ALA) for the absorption of CO2 or H2S, Alkacid “dik” utilizes the potassium salts of diethyl or dimethylglycine (K-DiMGLY) for the selective removal of H2S from the gas streams containing CO2. Alkacid “S” uses sodium phenolate (Na-PH) solution as an absorbent for the removal of the contaminants other than CO2 and H2S such as ammonia, carbon disulfide, mercaptan, dust and tars. These absorbents showed better absorption capacity and required less steam for solvent regeneration as compared to alkanolamine (DEA, MDEA). Moreover, these solvents also showed better stability and less corrosiveness [15]. Another two processes were developed by TNO (The Netherlands) for separation of CO2 utilizing aqueous amino acid salts as reactive absorbents. In the first process, membrane gas absorption was used based on the commercial polypropylene hollow fiber membrane. The conventional solvents like alkanolamines were not compatible enough with these membranes. The new solvents (CORAL liquids) based on potassium salts of amino acids such as taurine (K-TAU) and glycine (K-GLY) have shown better operational performance with these membrane modules owing to the high surface tension of amino acid salts solutions compared to the amines. Besides, these new absorbents show better resistance towards thermal and oxidative degradation and negligible vapor pressure [43, 44]. Another proprietary post-combustion carbon capture technology (PostCap) has been developed by Siemens AG. The technology uses aqueous amino acid salt solutions, but the type of amino acid salt is undisclosed. The use of the solvent developed by Siemens AG showed that the process requires less energy with enhanced performance in the CO2 capture process. The amino acid based solvent (undisclosed) showed high chemical stability, environmentally friendly, and flexible operation with good economics of the process under a wide range of operating parameters. This process was also validated in a coal-fired pilot scale power plant in Frankfurt, Germany, while their second pilot plant is under the planning phase in Florida, USA [45]. Another process known as DECAB process has been developed by TNO (The Netherlands)

126


using alkaline salts of amino acids. The process takes an extra benefit of the fact that when the CO2 absorbs in aqueous salts of amino acids, the precipitation forms under certain conditions in the process such as solvent concentration, CO2 loading, temperature, etc. [43, 46]. These precipitations can be the amino acid itself according to Kumar et al. [47]. Due to the formation of precipitation, the CO2 equilibrium partial pressure will remain nearly constant at certain CO2 loading. Therefore, a possibility of higher loadings of the solvent may be achieved with the benefits for the solvent circulation and requirement of energy for regeneration. Moreover, the absorber size could essentially be reduced as the higher driving force is observed. However, it does not necessarily follow that the precipitation leads to an increase in the driving force for the reaction, as the liquid nevertheless remains saturated. The precipitation of the amino acid itself would represent a loss of reactant [47]. In this CO2 removal process, spray tower was used instead of the other contactors to handle the possible blockage of the equipment [43, 46]. Based on the above analysis, it can be said that, the formation of precipitation during CO2 capture process is one of the drawback of amino acid salts, if these are used in ordinary absorption columns/contactors. Moreover, the formation of precipitation could potentially block the equipment, pumps and other accessories in the absorption plant, which could lead to the plant shut down [43, 46]. Moreover, the formation of precipitation also depends on the type of amino acids and their respective concentration. The higher concentrations of amino acid salts cannot be used due to the formation of precipitation during CO2 absorption. This is one of the limitations of the amino acid salts [43, 51]. The summary of the CO2 absorption processes developed by various academic and commercial entities is given in Table 4.

Table 4 Processes for acid gas removal using amino acid salts [32] Name of the process

Amino acid salts used

Development stage

Refs.

Giammarco-Vetrocoke process Alkacid process developed by: BASF

• GLY as activator in alkali carbonates • Na-ALA • K-DiMGLY • Na-PH • Alkaline salts of TAU and GLY (CORAL liquids) • Potassium salts of amino acids (types: not disclosed) • CORAL liquids

Commercial

[15]

Commercial

[15]

Commercial

[43, 44]

Commercial

[45]

Development

[43, 46]

• Solvents under development

Development

[48]

• Aqueous K-BALA

Development

[49]

MGA process developed by: TNO Post Cap process developed by: Siemens AG DECAB process developed by: TNO DECAB plus process developed by: TNO CASPER process developed by: iCAP


127

In recent times, TNO has developed another process called as DECAB Plus which achieves an extra driving force for stripping of CO2. This is achieved by reduction in pH values due to the formation of precipitates [48]. Another process known as CASPER (CO2-capture And Sulphur Precipitation for Enhanced Removal) recently developed by iCAP consortium applies precipitation in CO2 saturated aqueous potassium salt of b-alanine (K-BALA) for the simultaneous removal of CO2 and SO2 from flue gas streams. The CASPER solvent achieved robust improvements in the removal of CO2 and SO2 from the flue gases [49].

5 Performance for CO2 Capture The performance of amino acid salts for CO2 capture is dependent on various factors. There are so many studies available in literature on amino acid salts for the removal of CO2 from various gas streams. These studies include the solubility of CO2 in amino acid salts, kinetic study, thermal degradation, corrosion study, toxicity analysis, energy and other considerations. It is very essential to assess the solvent performance based on the studies conducted by various researchers in order to evaluate their commercial possibilities.

5.1

Solubility of CO2

The solubility of CO2 in aqueous amino acid salt solutions provides the information on chemical solubility and loading capacity of acid gases (CO2/H2S). For the efficient design of CO2 removal process using new formulated solvents, it is very essential to assess the CO2 loading of the solvent at various temperatures and pressures for better process performance. CO2 loading data, therefore, helps in the selection of appropriate solvent for commercial applications [32, 47]. Various studies have been conducted on solubility of CO2 in aqueous amino acid salt solutions, as summarized in Table 5; however, there is a scarcity of the systematic information on comprehensive assessment and comparison of these solvents. In this section of the chapter, the solubility of CO2 in various amino acid salt solutions has been briefly reviewed and discussed systematically in order to make it easy to assess the solvent for commercial applications. Kumar et al. [47] reported the equilibrium solubility of CO2 in aqueous K-TAU solutions at 298 and 313 K in low CO2 partial pressure range (0.1–6.0 kPa). The range of concentration studied was 500–4000 mol m−3. The crystallization was observed during the CO2 absorption at a higher range of CO2 partial pressures and at high amino acid salt concentration (2 M and above). Furthermore, the influence of crystallization on the vapor liquid equilibria was also explained. It was found that even at a low range of CO2 loadings (0.2–0.5 mol of CO2/mol of AAS); the crystallization occurred, depending on the further amino acid salt’s concentration. It

128


Table 5 Various studies on CO2 solubility in amino acid salts Aqueous amino acid salts

Concentration

Temperature (K)

Pressure

Refs.

K-TAU Na-GLY Na-GLY K-GLY K-THR K-GLY K-SER SARMAPA K-PRO K-ALA K-PRO K-SAR K-GLY K-TAU K-PRO K-AABA K-LYS K-LYS K-PRO K-ASN K-GLU

500–4000 mol/m3 10–30 wt% 1–30 wt% 0.1–3.0 mol/dm−3 1.0 mol/dm−3 1–3 mol/L 14.3 mass% 1.0–5.0 M 0.5–3.0 mol/dm−3 2.5 M 2.5 M 4M 1.0–1.8 mol/kg 1.0–1.8 mol/kg 7.5–27.5 wt% 6.9–25.6 wt% 0.5–2.5 mol/dm−3 2.5 mol/L 2.5 mol/L 8.5–34% 9.2–36.8%

298–313 303.15–323.15 298.15–313.15 293–351 313.15 298 313.15–373.15 313–373 285–323 298–313 298–313 313.15–353.15 313–393 313–393 313.2–353.2 313.2–353.2 298–313 313.15–333.15 313.15–333.15 313.2–353.2 313.2–353.2

0.1–6.0 kPa 0.1–200 kPa 100–2500 kPa 6 104 Pa 6 104 Pa – 0.1–400 kPa 4.08–99.1 kPa 70 kPa Atmospheric Atmospheric 0–812.5 kPa 10–100 kPa 10–100 kPa 1000 kPa 1000 kPa 0–45 kPa 0.05–17.5 kPa 0.05–17.5 kPa Up to 950 kPa Up to 950 kPa

[47] [50] [51] [23] [23] [52] [54] [53] [58] [57] [57] [59] [60] [60] [61] [61] [62] [63] [63] [64] [64]

was reported that the crystallization product is the protonated amine or the zwitterionic form of the amino acid. The formation of crystallization was observed to have the positive effect on the CO2 absorption capacity [47]. Similarly, Song et al. [50] studied the solubility of CO2 in aqueous sodium glycinate (Na-GLY) at three different temperatures (303.15, 313.15, and 323.15 K) at partial pressure range of 0.1–200 kPa followed by different concentrations (10, 20, 30 wt%). It was observed that the solubility decreased with the rise of temperature and Na-GLY concentration; however, with increasing the pressure the solubility increased. It was concluded that the CO2 solubility in aqueous Na-GLY solution (10 wt%) is better than various aqueous alkanolamine solvents such as MEA, AMP, and TIPA [50]. Likewise, another study on CO2 solubility in aqueous Na-GLY was conducted by Harris et al. [51] over a wide range of concentrations at 298.15 and 313.15 K. The study was conducted at slightly higher pressures up to 2500 kPa. The finding of this study also shows that the Na-GLY has a better absorption capacity than aqueous MEA and other alkanolamine solvents [51]. Portugal et al. [23] measured the solubility of CO2 in aqueous K-GLY solutions at 293 to 353 K for the concentrations ranging from 0.1 to 3.0 mol dm−3. The study was conducted up to the partial pressure of 6 104 Pa. The solubility showed


129

increasing trend with an increase in pressure. However, the effect of temperature on solubility of CO2 in aqueous K-GLY solutions was not substantial from 293 to 323 K, which could be a limitation for the regeneration of the solvent. The noticeable effect of temperature on the absorption capacity was observed above 323 K, and the effect was more obvious at 353 K. Furthermore, the solubility data was also compared with the work of Song et al. [50] for aqueous Na-GLY, and it was concluded that there is no significant difference between the absorption capacity of aqueous Na-GLY and aqueous K-GLY at 313 and 323 at 1.06 mol dm−3. The effect of concentration was observed to be same as previous authors. As expected, the absorption capacity of aqueous solutions of K-GLY was found to be more than that of conventional MEA when compared [23]. Subsequently, in the same study, Portugal et al. [23] also investigated the solubility of CO2 in aqueous potassium threonate (K-THR) for one concentration (1 mol dm−3) at 313 K. It was concluded based on the results that aqueous K-GLY has better CO2 absorption capacity as compared to aqueous K-THR for the same temperature and concentration range [23]. Zhang et al. [52] conducted another study on CO2 absorption in aqueous K-GLY for the concentration range of 1 to 3 mol L−1 at 298 K; and the results were compared with aqueous MEA and MDEA. It was found that the CO2 absorption in aqueous K-GLY is higher than in aqueous MDEA but slightly lower than aqueous MEA. The later findings of the Zhang et al. [52] do not comply with the observations of Portugal et al. [23]. Additionally, the regeneration study of CO2 absorption in aqueous K-GLY was also carried out at 378 K and compared with aqueous MEA and MDEA. It was concluded that aqueous K-GLY has improved regeneration rate as compared to MEA but lesser than MDEA [52]. Aronu and co-researchers [53] have studied the VLE measurement of CO2 in aqueous amine amino acid salts (AAAS) solutions; 3-(methylamino) propylamine/sarcosine (SARMAPA) at 40 to 100 °C for the concentration range of 1–5 M, and a total pressure of 4.08–99.1 kPa. For 5 M loaded SARMAPA, the CO2 equilibrium was measured at 40–120 °C. Moreover, the reaction heat between aqueous SARMAPA (5 M) and CO2 was also estimated. The enthalpy of the present system was 79.2 kJ/mol of CO2 at loading of 0.2, which is slightly lower than 30 wt% MEA, 85.4 kJ/mol at 0.2 loading as reported by Lee et al. [55]. However, it is higher than 59.8 kJ/mol of MDEA at 0.1 loading for the concentration of 4.28 kmol m−3 [56]. Song et al. [54] have reported the solubility of CO2 in aqueous solutions of potassium serinate (K-SER) at three different temperatures (313.15, 343.15, and 353.15 K) for only one concentration, 14.3 mass% over a partial pressure range of 0.1–400 kPa. The solubility of K-SER was also compared with 15.3 mass% MEA. As a general trend, solubility of both solvents showed pressure dependent behavior, however, the values of solubility reduced with increasing the temperature. It was found that CO2 solubility in aqueous K-SER was much better than MEA at 313.15 K (absorber conditions) above partial pressure of 4.0 kPa. However, K-SER solution has a lower solubility at 373.15 K than that of MEA, which suggested that the cyclic capacity of CO2 in aqueous K-SER would be higher than that of MEA [54]. Lim et al. [57] have studied two types of amino acid salts, i.e., potassium salt of alanine (K-ALA) and proline (K-PRO) at 298 and 313 K for 2.5 M

130


concentration. The trend of solubility with respect of temperature is same as found by various researchers [50, 51]. The absorption capacity of the two amino acids in this study was better than MEA. Besides, the heat of absorption of these two amino acid salts was also determined and compared with MEA and DEA. It was found that the aqueous K-ALA and K-PRO have lower absorption heat than MEA and DEA. It was concluded that K-ALA is better than K-PRO in terms of capacity and heat of absorption [57]. Majchrowicz and Brilman [58] have also measured the solubility of CO2 in K-PRO for four different concentrations (0.5, 1.0, 2.0, and 3.0 mol dm−3) at 285 and 323 K and up to 70 kPa partial pressure. The effect of temperature, pressure and amino acid salt concentration on CO2 solubility followed the same trend as described previously. Moreover, the precipitation was also encountered at higher concentration (3-mol dm−3) at 285 K over 6-kPa pressure. The effect of this precipitation was also determined, which showed that the higher CO2 loading can be achieved at a lower partial pressure [58]. This finding also supported the work of Kumar et al. [47]. The detailed understanding on the benefits of precipitations has also been explained in DECAB process as mentioned in Sect. 4. The solubility of CO2 in aqueous 1 mol dm−3 was compared with various other amino acid salts and MEA. It was found that the aqueous K-TAU has lowest capacity while the K-THR has a loading similar to MEA. The loading capacity of K-PRO and K-GLY is 19% and 23% higher respectively than MEA. Moreover, the integral heat of absorption for K-PRO is of the same magnitude as that of Na-GLY and other amines, but it is less than MEA and slightly higher than MDEA. Thus, it was concluded that K-PRO is a potential solvent for CO2 capture from industrial gases [58]. Aldenkamp et al. [60] have reported the solubility of CO2 in K-TAU and K-GLY for two concentrations (1 and 1.8 mol kg−1) at absorber and desorber conditions (313, 333, 353, 373, and 393 K). The partial pressure range was studied as 10–100 kPa. The CO2 loading capacity of K-GLY was observed to be higher than K-TAU solutions at all range of temperatures [60]. In a very recent time, Chang et al. [61] reported the solubility of CO2 in aqueous K-PRO and DL-a-amino butyric acid (K-AABA) at various temperatures and 1000-kPa pressure. According to their findings, the solubility of both amino acid salts is higher and comparable to the commercial alkanolamines. Moreover, it is reported that K-AABA has higher loading capacity as compared to K-PRO [61]. Another recent work by Mazinani and his co-researchers [62] have reported the solubility of CO2 in potassium lysinate (K-LYS) solution. The effect of temperature, concentration and pressure is same as reported in various other research works. The results suggest that the amino acid salt studied can also be potential solvent for CO2 capture [62]. Another study on K-LYS and K-PRO reported that the solubility of CO2 in K-LYS and K-PRO is higher than that of MEA. Moreover, K-PRO has lower loading as compared to K-LYS, therefore, K-LYS (2.5 M) is proposed for CO2 absorption in place of 30 wt% MEA [63]. Chen et al. [64] have studied the amino acid salts such as potassium L-asparaginate (K-ASN) and L-glutaminate (K-GLU) up to 950-kPa pressure and at different temperatures. These amino acid salts have significant CO2 loading and comparable to alkanolamines [64]. Increase


131

in gas solubility in liquids with increasing the pressure is universal. The CO2 partial pressures in flue gases are generally less than 15 kPa-a, so the high-pressure studies are not particularly relevant to the problem of CO2 capture from flue gases. The high-pressure studies are significant for CO2 separation from natural gas. In addition, the reduction in solubility with increasing temperature is also universal, and essential for the thermal swing recovery process. A requirement for a high desorption temperature by a particular amino acid (requiring a high steam pressure) could be a distinct disadvantage to the power station supplying the steam. There are various types of amino acid salts for the absorption of CO2 from various gas streams reported by many groups of researchers. Even though, these amino acid salts offer better absorption characteristics as compared to amine solvents, but the improvements are still required to enhance the absorption characteristics of amino acid salt solutions. The small improvements in the CO2 absorption performance could potentially benefit the overall process cost. Therefore, the improvements are always required in the existing solvents to obtain the more appropriate solvent for CO2 capture. To achieve these goals, various researchers have proposed to use the solvent blends based on amino acid salts and other solvents, including alkanolamines to combine all the favorable characteristics in one solvent blend for sufficiently improving the absorption performance of the CO2 capture process [59, 65–67]. Although the amino acids blended with alkanolamines perform better, but it could also add to some extent the drawbacks mentioned in Table 1, depending upon the amount of the amine added to the amino acid salt solutions. Mazinani et al. [65] reported the solvent blend consists of aqueous Na-GLY and MEA for CO2 removal. The solubility of CO2 in Na-GLY + MEA is higher than MEA at a partial pressure above 20 kPa. The mixed solvent demonstrated the better potential compared to MEA alone; however, the addition of MEA could offer to some degree the limitations described above in Table 1 [65]. Lu et al. [67] proposed the solvents, which are blends of K-GLY + piperazine (PZ) and K-GLY + phosphate (K3PO4). The CO2 loading of K-GLY + PZ and K-GLY + K3PO4 blends is higher than a single K-GLY solvent. Kang et al. [59] investigated the potassium sarcosinate (K-SAR), K-ALA + PZ blends and K-SER + PZ blends. The results show that the blend of K-ALA + PZ has highest solubility than the other solvent blends at 313.15 and 353.15 K. While at temperature 353.15 K, the blends of K-ALA + PZ and K-SER + PZ showed better solubility than K-SAR absorbent [59]. Privalova et al. [66] investigated three different amino acid salts for 15 wt% concentration blended with 5 wt% PZ at 300 K over 1 bar pressure, as given in Table 6. According to the findings, addition of a small amount of PZ in aqueous solutions of amino acid salts remarkably increases the absorption performance. Park et al. [68] have also reported five different amino acid salts blended with PZ. Out of five, three solvent blends (1.5 M ALA + 1 M PZ, 1.5 M SER + 1 M PZ and 4 M SAR) were identified as potential CO2 absorbents especially for membrane contactors [68]. From the above studies, it is obvious that blending of amino acid salts and amines offer better CO2 absorption characteristics, and at the same time add to a some extent the drawbacks associated

132


Table 6 Various studies on CO2 solubility in amino acid salt blends Amino acid salt blends

Concentration

Temperature

Pressure

Refs.

Na-GLY + MEA

0.5–2 M SG 0.5–2 M MEA 1.5 M + 1 M 1.5 M + 1 M (15 + 5) wt% (15 + 5) wt% (15 + 5) wt% 4M+1M 2.5 M + 1.5 M 3M+1M 1.5 M + 1 M 1.5 M + 1 M

298–313 K

0–35 kPa

[65]

313.15–353.15 K 313.15–353.15 K 300 K 300 K 300 K 40 °C 40 °C 40 °C 40 °C 40 °C

0–665.5 kPa 0.2–1041.7 kPa 1.0 bar 1.0 bar 1.0 bar – – – – –

[59] [59] [66] [66] [66] [68] [68] [68] [68] [68]

K-SER + PZ K-ALA + PZ K-ALA + PZ K-BALA + PZ K-SAR + PZ K-GLY + PZ K-TAU + PZ K-SAR + PZ K-ALA + PZ K-SER + PZ

with amines. Therefore, it is required to carefully formulate the blends of amino acid salts and amines so that the goal of effective CO2 removal could be sustained. The solubility data available in the literature for different types of amino acid salts and their blends have been compared at 313 K as shown Fig. 2. It is evident form Fig. 2 that almost all the amino acid salt solutions and their blends have higher CO2 loading as compared to commercial amine (MEA). Among the amino acid salts studies, K-LYS shows highest solubility of CO2 as compared to various other

Fig. 2 CO2 loading of various amino acid salt solutions and their blends at 313 K


133

amino acid salt solutions. Moreover, the other amino acid salts have comparable CO2 loading at reported temperature for the given range of pressure, and concentration. Therefore, the amino acid salts are potential alternatives to alkanolamines for efficient absorption of CO2. This comprehensive comparison of CO2 loading of the various amino acid salts could be very helpful for the selection of appropriate solvent based on amino acid salts for CO2 capture from various streams of gas.

5.2

Kinetics of CO2 Absorption

Kinetics of CO2 absorption show that how fast the reaction between CO2 and amino acid salts occurs. Reaction kinetics is very important for the design and simulation of the acid gas removal process, especially for the absorber and desorber design [69, 70]. There are various studies available on the kinetics of CO2 absorption in aqueous amino acid salt solutions. These studies have been conducted to obtain the kinetic data useful for CO2 removal plant design. Table 7 presents the various studies conducted on reaction kinetics of CO2 absorption in aqueous amino acid salt solutions. In this section, the kinetics of CO2 absorption in various amino acid salt solutions have been briefly reviewed systematically in order to make it easy to assess the solvent for commercial applications.

Table 7 Kinetics studies of CO2 absorption in aqueous amino acid salts Aqueous amino acid salts

Concentration

Temperature (K)

Refs.

K-TAU K-GLY Na-GLY K-GLY K-THR K-SAR, Li-SAR K-PRO, Li-PRO K-SAR K-GLY K-SAR K-ALA K-PRO K-GLY K-ALA K-PRO Na-PRO Na-TAU Na-PRO K-PRO

200–300 mol m−3 200–300 mol m−3 1.0–3.5 kmol m−3 0.1–3.0 M 0.1–3.0 M 0.5–3.0 mol L−1 0.5–3.0 mol L−1 0.5–3.8 M 0.10–0.50 kmol m−3 1.0–4.0 kmol m−3 1.0–3.0 M 0.5–3.0 kmol m−3 0.5–2.0 M 0.5–2.0 M 0.5–3.0 mol dm−3 0.5–3.0 mol dm−3 5–50 mol m−3 4–12 mol m−3 1M

285–305 295 303.15–323.15 293–303 293–313.15 298 298 298–303 298–303 298.15–335.15 293.15–313.15 303–323 298.15–335.15 298.15 290–303 298 298–313 298–313 313

[69] [69] [71] [72] [74] [75] [75] [76] [77] [78] [79] [80] [81] [81] [82] [82] [83] [83] [84]

134


Kumar et al. [69] reported the kinetics of the CO2 absorption in aqueous K-TAU at 285 to 305 K, while K-GLY at 295 K. Unlike alkanolamines, aqueous K-TAU and K-GLY showed an increment in partial reaction order from one at molar concentration higher than 1000 mol m−3. Zwitterion and termolecular mechanism were studied for better understanding of the kinetics. It is reported that the value of k2 (forward second order reaction rate) for the zwitterion mechanism of CO2 absorption in aqueous K-TAU is considerably higher (12.60 m−3 mol−1 S−1) as compared to alkanolamines (4.94 m−3 mol−1 S−1) [70]. Moreover, it was determined that the reactivity of CO2 with K-GLY is higher than that of K-TAU because the K-GLY has higher basic strength [69]. Lee et al. [71] investigated the kinetics of CO2 absorption in aqueous Na-GLY. The second-order reaction rate constant are reported as 218, 576, and 1034 m3 kmol−1 S−1 respectively at 303.15, 313.15, and 323.15 K. Additionally, energy of activation for the reaction was reported to be 63.8 kJ/mol [70]. Portugal et al. [72] have reported the kinetics of CO2 absorption in aqueous K-GLY solutions. They have found out that aqueous K-GLY solution has faster reaction kinetics as compared to MEA. For MEA (1 M) at 298 K, the overall kinetic constant is reported as 5920 s−1 [73], however, at same conditions, the value is 13400 s−1 for aqueous K-GLY. Hence, the amino acid salt studied in this work showed significant potential to be used for absorption of CO2 [72]. In another study, Portugal et al. [74] reported the kinetics of CO2 absorption in aqueous K-THR solution. The results were also compared with that of alkanolamine (DEA) and K-GLY solution at 298 K for 1 M concentration. It is reported that K-THR has slower kinetics than the K-GLY; however, it is faster than DEA and comparable with other amine solutions [74]. Holst et al. [75] have reported the screening study of CO2 absorption kinetics in various amino acid salt solutions. The potassium salts of various amino acids such as 6-aminohexanoic acid, BALA, ARG, GLU, MET, PRO and SAR were used in the initial study at 298 K for 0.5 mol L−1. It is reported that the K-PRO and K-SAR has better absorption characteristics as compared to the rest of the amino acids studied because they combine with relatively high apparent rate constant with low values of pKa [75]. Simons et al. [76] investigated the kinetics of CO2 absorption in aqueous K-SAR and reported the influence of concentration, temperature and CO2 loading on the rate of reaction. Moreover, the temperature has the positive effect on the reaction rate constant because it increased with the rise in temperature. Similarly, the overall reaction rate constant increases with an increase in amine concentration. Furthermore, it was observed that the values of overall and apparent reaction rate constant decrease with increasing the CO2 loading of the amino acid salt solution. The reaction rate constant for K-SAR is higher than that of MEA [76]. Vaidya et al. [77] have studied the kinetics of CO2 absorption in two aqueous amino acid salt solutions, i.e., K-GLY and K-TAU. The second-order reaction rate constant of K-GLY at 303 K is 6.29 m3 mol−1 s−1. Additionally, K-GLY was also used as activator /promoter in DEEA to enhance the CO2 removal rate. It was found that K-GLY is an attractive promoter for enhancing the CO2 absorption rate [77]. Similarly, another research group Aronu et al. [78] carried out the kinetic study of CO2 absorption in K-SAR at few different temperatures (298.15–335.15 K) and


135

concentrations (1.0–4.0 kmol m−3). It was observed that the reaction rate constant increased with increasing the temperature and concentration, and this finding is similar to the one reported by Simons et al. [76]. Moreover, the value of reaction rate constant for K-SAR is higher than MEA, but still comparable [78]. Kim et al. [79] reported the kinetic study of CO2 absorption using K-ALA solution at 393.15 to 313.15 K for 1–3 M concentration. It is reported that K-ALA solution has lower absorption rate as compared to K-GLY solution. However, due to the steric hindrance effect, the loading capacity of K-ALA may be improved by increasing the charge density of N atom of the K-ALA solution. In case of the desorption, K-ALA has high desorption rate caused by the slow absorption rate. This could reduce the energy of regeneration [79]. Paul et al. [80] reported the kinetic data for the absorption of CO2 using K-PRO solution, and found that the studied solvent has higher second order reaction rate constants as compared to alkanolamines and few other amino acid salt solutions. The values of overall reaction rate constant (kov) of K-THR investigated by Portugal et al. [74] and K-TAU reported by Vaidya et al. [77] is lower than K-PRO studied in this work. However, the values reported for K-SAR and K-GLY by Portugal et al. [72] and Simons et al. [76] are very close to K-PRO at 303 K. Similarly, Majchrowicz et al. [82] have also studied the kinetics of CO2 absorption in K-PRO solution, and Na-PRO solution at 298 K. It is reported that the reactivity of K-PRO with CO2 is higher than Na-PRO. Moreover, the value of kov for K-PRO is higher than K-THR and K-TAU while it is comparable with MEA, K-SAR and K-GLY [82]. Another study on reaction kinetics of CO2 in aqueous Na-TAU and Na-PRO have been reported by Sodiq et al. [83]. It was found that the Na-PRO has faster reaction rates than Na-TAU and commercial MEA. However, Na-TAU showed slower rates than MEA at low concentrations [83]. Likewise, Fang et al. [84] have also reported the kinetics of absorption of CO2 in K-PRO solutions and influence of total pressure on absorption. It was found that high pressure has a positive effect on the rate of CO2 absorption. Moreover, K-PRO was found to be promising solvent for CO2 removal [84]. Various researchers have reported the overall reaction rate constant (kov) for some amino acid salt solutions. Figure 3 shows the comparison of kov for absorption of CO2 in various amino acid salt solutions at 303 K for various reported concentrations. As can be seen from Fig. 3 that K-PRO has the highest kov than other amino acid salt solutions. K-SAR and K-GLY has somewhat comparable values of kov. K-ALA and K-THR has the lowest values of kov. Therefore, it is apparent that the K-PRO is the most promising solvent for bulk removal of CO2 with faster reaction kinetics from other amino acid salt solutions. Moreover, the activation energy of various amino acid salt solutions reported by various researchers are given in Table 8, and the comparison between the activation energy of various amino acid salts solutions and amines is shown in Fig. 4. Activation energy determines the response of reaction rate to temperature, and it is one of the keys to determining the optimum operating conditions and loadings. It is clear from the Fig. 4 that Na-PRO has the lowest energy of activation than even K-PRO suggesting that the rate of reaction of CO2 with Na-PRO is very fast as also suggested by Sodiq et al. [83]. The K-PRO and K-SAR has the comparable energy

136


Fig. 3 Overall reaction rate constants for CO2 absorption in amino acid salt solutions

Table 8 Activation energy of aqueous amino acid salts and amines during CO2 absorption

Amino acid salts

Ea (KJ/mol)

References

K-TAU Na-GLY K-GLY K-SAR K-PRO K-PRO Na-TAU Na-PRO Amines

47.40 63.80 48.23 26.0 36.5 43.3 48.1 12.0

[69] [71] [72] [76] [80] [81] [83] [83]

MEA DEA MDEA AMP PZ TEA

41.2 53.1 47.9 41.7 33.6 35.8

[85] [85] [86] [87] [88] [89]

of activation. K-TAU and K-GLY have also the comparable activation energy but higher than K-PRO and K-SAR. Interestingly, Na-GLY has the highest activation energy than all the amino acid salts and amines; however, it is comparable with DEA and MDEA.


137

Fig. 4 Comparison of activation energy of amino acid salt solutions and amines

5.3

Degradation Tendency

Solvent degradation is a major concern in the CO2 absorption process because of the significant impact on operational cost and intention to use thermal compression from high temperature stripping to minimize the operational energy. The degradation of the CO2 capture solvent consume the amine/amino acid salt, as a result; the makeup rate of the solvent becomes higher. Besides, degradation of solvent also generates operational problems such as increasing the foaming and corrosion tendency during the absorption-regeneration process [15]. Therefore, the degradation performance evaluation is very important for the appropriate selection of solvent. Generally, there are two types of degradation in post-combustion CO2 removal process such as oxidative and thermal degradation. In an oxidation degradation, oxidized fragments of amine are formed. These include organic acids, ammonia, aldehydes, and amines. This degradation is caused in oxygen-rich environment typically during the flue gases treatment. Such degradation results in toxic products of degradation and the solvent loss. The other type is the thermal degradation, caused by either thermal decomposition that normally occurs above temperature 200 °C or by various reactions that are accelerated by high temperatures [90]. Amine based solvents are usually considered to have high oxidative and thermal degradation [15]. This is one of the drawbacks in post-combustion CO2 absorption because of the severe negative consequences on the process, and the effects of toxic degradation products on the environment. Due to these limitations of the amines, amino acid salts are introduced as alternatives to amines because they are considered as more stable towards degradation, especially in the presence of oxygen due

138


to the ionic nature [32, 39, 47]. There are very few studies on the degradation of the amino acid salts for CO2 capture. Epp et al. [90] studied the oxidative degradation of the amines (MEA, DGA) and amino acid salt, K-GLY. They have conducted the experiments in closed vessel containing oxygen as a gas phase. The K-GLY is not much oxidative stable (degradation: 0.0024 mol NH3/l CO2) as compared to MEA (degradation: 0.0029 mol NH3/l CO2), because the formation of degradation products, including ammonia and formaldehyde are almost same in case of MEA and K-GLY. The results advocated that the benefit of K-GLY in terms of better resistance to oxidative degradation is not significant as expected [90]. The more thorough investigation is required in real environment of flue gases for better and definite findings so that the appropriate solvent may be selected for commercial CO2 absorption process. Huang et al. [91] reported the thermal degradation of various amino acid salt solutions such as Na-GLY, Na-SAR, Na-ALA and Na-BALA at 125, 135 and 145 °C. It is reported that Na-GLY has more degradation rate as compared to MEA at the same amine concentration (5.0 M) and at all temperatures. These results are unexpected because the amino acid salts were considered more thermally stable compared to amines as reported in Portugal et al. [72], and at various places in the literature [32, 39] Moreover, the thermal degradation of other amino acids for 2.5 M concentration was also compared in the study at 135 °C. Na-GLY and MEA both at 2.5 M concentration were also used as reference. The results showed that all the studied amino acid salts have higher rates of amine loss as compared to MEA. Moreover, the thermal stability was found in the order of MEA > SAR > ALA > BALA. SAR showed the slowest degradation rate but still almost two times higher than MEA, while BALA has the fastest degradation at all studied temperatures [91]. From the above studies, it is clear that only few amino acid salts have been tested for thermal and oxidative degradation. The studied amino acid salts showed less thermal stability than commercial MEA, which contradicts with the information on benefits of the amino acid salts summarized in Table 3. Based on these observations, it is still not straightforward to say that all the amino acid salts are less or high thermally stable than the amine solvents because of the lack of studies available on degradation of amino acid salts. Therefore, a thorough study on degradation of amino acid salts is required for stronger claims in both the cases.

5.4

Corrosion Tendency

Corrosion in the CO2 absorption process is one of the most serious problems because it reduces the equipment life, causes unscheduled downtime, loss of production and sometimes even injury to the personnel [92]. The cost of the production losses due to the unscheduled downtime of the typical amine plant varies from $10,000 and $30,000 per day. Besides, the extra expenses are also incurred for the restoration of corroded systems and other management activities initiated for the mitigation of the corrosion [93]. The conventional alkanolamines are highly


139

corrosive towards the equipment and accessories used for CO2 absorption. Corrosion is one of the problems, which causes the high cost of the CO2 absorption by amines [94]. Amino acid salts’ solutions are most promising solvent for the absorption of CO2 due to various positive characteristics as mentioned in previous sections. A part from various CO2 absorption characteristics, it has been shown that the amino acid salt solutions are also less corrosive as compared to alkanolamines [32]. Therefore, the cost of the CO2 capture process due to corrosion loss could potentially be reduced by using the amino acid salt solutions instead of amines. Since the amino acid salts are still new and emerging solvents for CO2 capture, therefore, there are very few investigations available in the literature on the corrosion rate of these solvents. Ahn and co-researchers [94] have investigated the corrosion rate of two amino acid salts such as K-TAU and K-GLY by using the conventional weight loss method (ASTM E3-80). Corrosion tests were conducted for different concentrations of amino acid salts (1.5–5 M) and at various temperatures (313.15–353.15 K) using carbon steel grade 1018. This type of carbon steel is widely used for the construction and fabrication of equipment and accessories in the CO2 absorption plants. The effect of temperature, concentration and CO2 loading was determined for the studied amino acid salts. The results have been compared with the commercial MEA at 353.15 K and CO2 loading of 0.62 for MEA and amino acid salts. It is reported that the corrosion rate of MEA increases with an increase in the concentration because more amounts of molecules absorb more CO2. The CO2 dissolved in the form of bicarbonate ions cause iron dissolution. However, the corrosion rate decreased with an increase in the concentration of K-GLY. This is contrary to MEA because of the hydrolysis of the carbamate results in decreased bicarbonate ions, thereby reducing the corrosion. Moreover, K-TAU showed the same trend as of MEA. The corrosion rate increased with increasing the temperature and CO2 loading [94]. The corrosion rate of K-TAU is lower than that of MEA. It was suggested that K-GLY is beneficial at high concentration while K-TAU is beneficial at lower concentration [94]. Moreover, the addition of corrosion inhibitors has been proposed by Ahn et al. [94] to reduce the corrosion tendency of amino acid salts. In this regard, PZ, sodium metavanadate (NaVO3) and sodium sulfite (NaSO3) were suggested. Increasing the PZ concentration in amino acid salt further increased the corrosion of carbon steel because of the blended solution absorbed more CO2 thereby increasing the amount of bicarbonate ions and so does the corrosion [94]. However, addition of NaVO3 showed better inhibition performance (about 99.9%) than NaSO3 (93.7%). The inhibition performance increased with the rise of inhibitor concentration [94]. Although the corrosion inhibitors have potential to prevent corrosion, however, few operational issues can be induced with their use. For example, use of NaVO3 may increase the solvent degradation by accelerating the reaction rate such as observed by using it with MEA [95]. Mazinani et al. [65] investigated the corrosion tendency of blend of Na-GLY and MEA by electrochemical technique (ASTM G5-94) using potentiostat at various blend ratios without dissolved CO2 at 308 K. The results show that, with increasing

140


the Na-GLY concentration in the blend, the corrosion rate increases [65]. Recently, another study on the corrosion tendency of K-LYS was carried out by Mazinani et al. [62] following the same method as described in their previous investigation [65]. The results show that the corrosion rate increases with the rise in the concentration of K-LYS solution, and follows the same trend as observed previously. In both studies, the measurement could have shown a higher corrosion rate if CO2 was dissolved in the Na-GLY blends and aqueous solutions of K-LYS. Based on the analysis of the available literature, amino acid salts have been considered as less corrosive as compared to alkanolamines. However, the existing data and knowledge of the corrosion evaluation of amino acid salts are very scarce. More thorough investigation on corrosion is still required for amino acid salts to verify their less corrosivity. The data on recommended corrosion inhibitors to be used with amino acid salt solutions is also very limited; therefore, care must be exercised in the selection of a proper corrosion inhibitor, so that the solvent actual performance may not be affected.

5.5

Toxicity

The suitable absorption characteristics alone are not enough for the right selection of the solvent for CO2 capture, but the possible environmental and health risks associated with the selected solvents must be known prior to use or commercial application. Without the proper investigation on toxicity and risks assessment, it may not be viable to use any newly developed solvent; otherwise, the new environmental issues could be raised after solving the climate-change problems. The most widely used solvents for CO2 absorption are alkanolamines; however, apart from various operational limitations during CO2 removal, these solvents also induce various environmental and health risks [21–25]. The new emerging solvents such as amino acid salt solutions are considered as green and environmental friendly. They have no harmful effects on the environment because these are naturally present in the environment [25, 96]. Apart from various benefits of amino acid salts, the environmental friendly characteristics of these solvents also make them as an alternative to alkanolamine solvents. Shao et al. [97] have reported that the amino acid group show high biodegradability with lower toxicity than the amines [97]. The comprehensive toxicity analysis of amino acid salts after absorption of CO2 has not been reported yet in order to prove their environmental suitability. However, some of the toxicological information in terms of lethal dose (LD50) for various amino acids and amines is available, and given in Table 9. The data have been taken from the safety data sheets of manufacturers. LD50 is the amount of the chemical /toxic material, which is sufficient to kill the 50% of the test animals in the certain time. The higher values of LD50 shows the less toxicity while the lower values denotes the higher toxicity [114].

Aqueous Amino Acid Salts and Their Blends … Table 9 Toxicity of various amino acids and amines

141

Amino acid/amines

LD50 (mg/kg)

Refs.

Gly PRO TAU HIS MET GLU MEA DEA TEA PZ

7930 (oral—rat) >5110 (oral—rat) >5000 (oral—rat) 15,000 (oral—rat) 36,000 (oral—rat) 7500 (oral—rat) 700 (oral—mouse) 710 (oral—rat) 2200 (oral—rabbit) 600 (oral—mouse)

[98] [99] [100] [101] [102] [103] [104] [105] [106] [107]

The comparison between the toxicity of amino acids and amines is shown in Fig. 5. The higher values of LD50 indicate the lower toxicity of the chemical while lower values indicate the high toxicity as mentioned previously. As can be observed from Fig. 5, the values of LD50 are higher in case of all amino acids than the amines, suggesting that the amino acids are very less toxic as compared to various amines. From all the amino acids, MET is less toxic than other amino acids, while the PZ has the highest toxicity among all other amines. Thus, amino acids can be considered as eco-friendly solvent as compared to amines, and be used for CO2 absorption.

Fig. 5 Comparison of toxicity (LD50) of amino acids and amines

142

5.6


Energy Aspects

During the selection of appropriate solvent for CO2 absorption, it is essential to assess the energy required for the CO2 absorption process because it is directly associated with the process efficiency and cost. Since CO2 capture is an energy demanding process, therefore, it is always important to optimize the energy efficiency of the capture process by developing the new solvents and processes. Some estimations show that about 30% of the energy produced by power plants would have to be dedicated to the operation of CO2 capture plant. This makes the cost of electricity generation to almost double. The doubling of the electricity cost by the addition of a carbon capture plant is also due to the overall effects of capital and operating cost. Moreover, these costs vary with location. The major portion of the energy penalty is due to the high energy required for solvent regeneration, and the energy required for compression of CO2 for pipelining and sequestration [39, 108]. The commercial alkanolamines solvents require high energy for the absorption of CO2 apart from their various other operational drawbacks. Therefore, the cost of the energy required and the cost incurred due to various operational drawbacks make up the overall process cost at maximum [39]. To overcome these issues, it is always required to search for the candidate solvent that would save the cost in terms of minimum energy required and by various other means. Amino acid salt solutions have been considered as efficient solvents in terms of various benefits described in previous sections, as well as in terms of the energy requirements. Tobias et al. [109] reported the process developed by Siemens utilizing the amino acid salt solution, and it is demonstrated that the amino acid salt solutions have a low absorption enthalpy that can reduce the energy demand. Siemens efficiently developed the process for CO2 capture employing the amino acid salt solution with the advantages of a low energy requirement for regeneration of solvent. Weiland et al. [110] reported that Na-GLY requires almost same reboiler duty as that of MEA. However, 85% of the CO2 recovery can be achieved with 1250 m3/h of the solvent as compared to 1500 m3/h for MEA for a typically configured CO2 capture plant (3000 tons/day CO2) [110]. This reduction in the flow rate could save the energy required for process operation. Moreover, some solvent blends based on 30 wt% KDiMGLY blended with 15 wt% MEA were also tested. The simulated results showed that 10% of the energy could be saved with 85% of the CO2 recovery as compared to 30 wt% MEA. However, addition of 15 wt% MEA could arise the issues linked with amines. In addition to this, another solvent blend ratio consists of 40 wt% KDiMGLY and 5 wt% PZ achieved 20% reduction of the energy with 85% recovery of CO2 and reduced solvent flow rate of 1250 m3/h. The heat of absorption (integral values from 0 to 0.3 loading) of MEA, MDEA, PZ, Na-GLY and KDiMGLY are 84, 58, 76, 85, and 55 kJ/gmol−1 respectively at 25 °C [110]. It is apparent that KDiMGLY has lowest heat of absorption, thereby may require less regeneration energy as compared to other solvents studied. Lim et al. [57] have reported the heat of absorption of K-ALA and K-PRO at 298 K. The values show that K-ALA has lowest heat of absorption (53.26 kJ/mol) than K-PRO


143

(90.20 kJ/mol), MEA (81.77 kJ/mol) and DEA (67.06 kJ/mol) respectively. These results suggest that amino acid salt, especially; K-ALA is expected to achieve better regeneration performance because of low heat of absorption. Similarly, Majchrowicz and Brilman [58] have reported the integral enthalpy of CO2 absorption in K-PRO solution as 54.3 kJ/mol at loading of 0.80 and temperature range of 298.37–313.83 K. This value is lower than Na-GLY (72.5 and 59.5 kJ/mol at loading, 0.90 and 0.86 over temperature range of 313.15–323.15 K) and MEA [111]. The results obtained by Majchrowicz and Brilman [58] and Salazar et al. [111] indicate that Na-GLY and K-PRO are efficient absorbent in terms of energy required for regeneration. K-PRO may be more cost savings than Na-GLY based on the results of the enthalpies. Few discrepancies are observed, such that the value of enthalpy of Na-GLY reported by Weiland et al. [110] is higher than that of Salazar et al. [111]. Similarly, the enthalpy of K-PRO reported by Lim et al. [57] is quite higher than the one obtained by Majchrowicz and Brilman [58]. Sanchez-Fernandez et al. [48] have summarized the results of an energy-efficient process known as DECAB, DECAB Plus and pH swing. In this process, a precipitating amino acid such as K-TAU was used. The results demonstrated that the process requires a reduced amount of energy than MEA. DECAB Plus process was recognized as more energy efficient (66% of the conventional MEA) [48]. Although the design of this process requires somewhat larger equipment, especially absorber, the maintenance cost would be significantly lower [48]. Some studies suggested that the amino acid salts are energy efficient solvents, while the others indicated that these may require moderate energy during CO2 absorption as compared to alkanolamines. Since, there are very limited studies available in literature on the energy performance of amino acid salt solutions; therefore, more research should be devoted to energy analysis of amino acid salts to completely endorse their use. For comprehensive evaluation of energy analysis of the process employing the amino acid salts, the study of relation between heat of absorption, overall regeneration heat duty and process parameters is required [112].

5.7

Other Considerations

There are few considerations important for the solvents to be used for CO2 absorption, such as vapor pressure and flammability. The vapor pressure of the solvent plays an important role in their selection for CO2 absorption. The solvents that have lower vapor pressure are desirable because it minimize the vapor loss. While the high vapor pressure can cause the solvent loss by evaporation together with the clean gas. The loss of the solvent due to higher vapor pressure would incur an extra process cost [39]. Therefore, it is always required to search, and develop a solvent with lower or negligible vapor pressure. Commercially used alkanolamines have higher vapor pressure; therefore, the urge to find the new solvent has increased, and consequently; a new class of solvents such as amino acid salt solutions have been introduced. These solvents have near-zero or negligible vapor

144


pressure because they are salts with ionic structure. Such typical characteristic of the amino acid salts makes them a prospective alternative to alkanolamines. The solvent refill cost can be reduced potentially due to the minimum vapor loss during CO2 absorption [32, 109]. Another benefit of amino acid salt is that they are not flammable and are less sensitive to oxygen as compared to alkanolamine. Therefore, these solvents are considered as non-hazardous [109]. Based on the analysis of these characteristics of amino acid salt solutions, it can be concluded that these are potential, cost saving, and energy efficient solvents for CO2 absorption.

6 Perspective and Conclusion The energy and cost efficient CO2 absorption from flue gases is a main technical challenge in terms of the solvent selection. Conventional alkanolamine solvents have many operational limitations as detailed out in the above sections. The new solvents based on amino acid salts are promising in terms of energy and cost. There have been several individual studies over the recent years on such solvent systems because of their versatile and promising characteristics. However, in this chapter, the overall performance analysis of aqueous amino acid salts and their blends was carried out for evaluating their commercial possibilities. Amino acid salt solutions and their blends demonstrated a significant potential for CO2 absorption; however, it is not straightforward to select the potential solvent. There are numerous parameters to be considered, which affect the overall cost of the process. Therefore, the sensible and holistic approach is required for choosing the most appropriate solvent. The performance analysis of amino acid salt solutions was carried out, and some parameters were identified to have substantial influence. These parameters include but not limited to CO2 absorption performance (CO2 loading), kinetics of absorption, degradation, corrosion, toxicity, energy aspects and various other considerations. Each of the parameters were analyzed and compared with conventional solvents. The performance analysis was somewhat difficult because of limited studies available on amino acid salts, and their blends. Besides, few discrepancies were also found between various studies, which made it challenging to streamline, and normalize the performance analysis results. Usually, the solvent is selected based on the comparative analysis. Therefore, a possible comparative analysis was carried out to meet the objective. However, a comprehensive comparison between various amino acid salts, and their blends was relatively challenging because of scarcity of data. Therefore, a complete performance and comparative analysis would require more research on amino acid salts, and their blends. From the performance analysis carried out in this chapter, it was found that almost all the amino acid salts showed good absorption characteristics than alkanolamines. Amino acid salts blended with promoters, especially PZ showed increased CO2 loading than various amino acid salts alone. Moreover, amino acid


145

salt such as K-PRO and Na-PRO showed fast reaction kinetics as compared to other amino acids. The activation energy of few amino acid salts such as K-TAU, Na-TAU and Na-GLY was higher than MEA and PZ, suggesting that these amino acids have slow reaction kinetics. Na-PRO and K-SAR have the lowest activation energy than all other amino acid salts and alkanolamines. The degradation rates of few of the amino acid salts were higher than MEA, while the corrosion resistance found to be lower. The toxicity of the amino acids was very less as compared to amines. From the energy analysis, it was found that various amino acid salts and blends such as K-ALA, KDiMGLY + PZ, and K-PRO require less regeneration energy than commercial MEA. The results of a new process such as DECAB Plus showed that the large amount of energy could be saved by using the precipitating amino acid (K-TAU). Moreover, the amino acid salts have negligible (near-zero) vapor pressure, and they are nonflammable. From the above evidence, it can be concluded that amino acid salts could be used as an effective alternative to alkanolamines for CO2 capture. In summary, based on the comprehensive analysis and assessment, these solvents are environmental friendly and energy efficient with higher CO2 loading capacity, faster reaction kinetics and require less regeneration energy. Besides, these solvents are non-volatile, non-flammable, less corrosive and oxidative stable to some extent. Moreover, these solvents can be made more effective by blending various additives with aqueous amino acid salts. Some of the amino acid salts have a comparatively significant potential than the others. Therefore, these solvents should be tested for pilot or commercial-scale demonstration. For the others, additional fundamental and systematic research is required, especially on thermal degradation, corrosion, kinetic study, and energy requirement assessment for comprehensive performance analysis. Acknowledgments The authors are grateful to Research Centre for CO2 Capture (RCCO2C), Department of Chemical Engineering, Universiti Teknologi PETRONAS for supporting this work.

References 1. Abu-Zahra MRM, Niederer JPM, Feron PHM, Versteeg GF (2007) CO2 capture from power plants: part I. A parametric study of the technical performance based on monoethanolamine. Int J Greenhouse Gas Control 1(2):135–142 2. Dongwoo K, Sangwon P, Hoyong J, Jaehong M, Jinwon P (2013) Solubility of CO2 in amino-acid-based solutions of (potassium sarcosinate), (potassium alaninate + piperazine), and (potassium serinate + piperazine). J Chem Eng Data 58(6):1787–1791 3. Yang HQ, Xu ZH, Fan MH, Gupta R, Slimane RB, Bland AE, Wright I (2008) Progress in carbon dioxide separation and capture: a review. J Environ Sci 20(1):14–27 4. NOAA (2014) Recent global monthly mean CO2. National Oceanographic and Atmospheric Administration: (NOAA). Silver Springs, MD 5. Williams M (2002) Climate change: information kit. United Nations Environment Programme (UNEP) and the United Nations Framework Convention on Climate Change (UNFCCC), Geneva

146


6. IPPC (Intergovernmental Panel on Climate Change) (2007) IPCC Report. IPCC, Geneva 7. Rubin ES, Rao AB (2002) A technical, economic and environmental assessment of amine-based CO2 capture technology for power plant greenhouse gas control. Environ Sci Technol 36(20):4467–4475 8. Khan FM, Krishnamoorthi V, Mahmud T (2011) Modelling reactive absorption of CO2 in packed columns for post-combustion carbon capture applications. Chem Eng Res Des 89 (9):1600–1608 9. Shaikh MS, Shariff AM, Bustam MA, Murshid G (2015) Measurement and prediction of physical properties of aqueous sodium L-prolinate and piperazine as a solvent for CO2 removal. Chem Eng Res Des 102:378–388 10. Lucquiaud M, Gibbins J (2011) On the integration of CO2 capture with coal-fired power plants: a methodology to assess and optimize solvent-based post-combustion capture systems. Chem Eng Res Des 89(9):1553–1571 11. Mores P, Scenna N, Mussati S (2011) Post-combustion CO2 capture process: equilibrium stage mathematical model of the chemical absorption of CO2 into monoethanolamine (MEA) aqueous solution. Chem Eng Res Des 89(9):1587–1599 12. Diamantonis NI, Boulougouris GC, Tsangaris DM, Kadi MJE, Saadawi H, Negahban S, Economou IG (2013) Thermodynamic and transport property models for carbon capture and sequestration (CCS) processes with emphasis on CO2 transport. Chem Eng Res Des 91 (10):1793–1806 13. Posch S, Haider M (2013) Dynamic modeling of CO2 absorption from coal-fired power plants into an aqueous monoethanolamine solution. Chem Eng Res Des 91(6):977–987 14. Sofia D, Coca Llano P, Giuliano A, Iborra Hernández M, García Peña F, Barletta D (2014) Co-gasification of coal-petcoke and biomass in the puertollano IGCC power plant. Chem Eng Res Des 92(8):1428–1440 15. Kohl AL, Nielsen R (1997) Gas Purification, 5th edn. Gulf Publishing Company, Houston 16. Hoff KA, Juliussen O, Falk-Pedersen O, Svendsen HF (2004) Modeling and experimental study of carbon dioxide absorption in aqueous alkanoamines solutions using a membrane contactor. Ind Eng Chem Res 43(16):4908–4921 17. Rochelle GT (2009) Amine scrubbing of CO2 capture. Science 325:1652–1654 18. Moioli S, Pellegrini LA (2013) Regeneration section of CO2 capture plant by MEA scrubbing with a rate-based model. Chem Eng Trans 32:1849–1854 19. Lepaumier H, Picq D, Carrette PL (2009) New amines for CO2 capture: II. Oxidative degradation mechanisms. Ind Eng Chem Res 48(20):9068–9075 20. Shaikh MS, Shariff AM, Bustam MA, Murshid G (2014) Physicochemical properties of aqueous solutions of sodium l-prolinate as an absorbent for CO2 removal. J Chem Eng Data 59(2):362–368 21. Dawodu OF, Meisen A (1996) Degradation of alkanoamines blends by carbon dioxide. Can J Chem Eng 74(12):960–966 22. Jamal A, Meisen A (2001) Kinetics of CO2 induced degradation of aqueous diethanolamine. Chem Eng Sci 56(23):6743–6760 23. Portugal AF, Sousa JM, Magalhaẽs FD, Mendes A (2009) Solubility of carbon dioxide in aqueous solutions of amino acid salts. Chem Eng Sci 64(9):1993–2002 24. Aronu UE, Hartono A, Svendsen HF (2011) Kinetics of carbon dioxide absorption into aqueous amine amino acid salt:3-(methylamino) propylamine/sarcosine solution. Chem Eng Sci 66(23):6109–6119 25. Shuaib Shaikh M, Shariff AM, Bustam MA, Murshid G (2014) Physical properties of aqueous solutions of potassium carbonate + glycine as a solvent for carbon dioxide removal. J Serb Chem Soc 79(6):719–727 26. Shaikh MS, Shariff AM, Bustam MA, Murshid G (2015) Physicochemical properties of aqueous solutions of sodium glycinate in the non-precipitation regime from 298.15 to 343.15 K. Chin J Chem Eng 23(3):536–540 27. Hook RJ (1997) An investigation of some sterically hindered amines as potential carbon dioxide scrubbing compounds. Ind Eng Chem Res 36(5):1779–1790


147

28. Simons K, Brilman WDWF, Mengers H, Nijmeijer K, Wessling M (2010) Kinetics of CO2 absorption in aqueous sarcosine salt solutions: influence of concentration, temperature, and CO2 loading. Ind Eng Chem Res 49(20):9693–9702 29. Hartono A, Aronu UE, Svendsen HF (2011) Liquid speciation study in amine amino acid salts for CO2 absorbent with 13C-NMR. Energy Procedia 4:209–215 30. Song H-J, Sangwon P, Hyuntae K, Ankur G, Park J-W, Lee S-J (2012) Carbon dioxide absorption characteristics of aqueous amino acid salt solutions. Int J Greenhouse Gas Control 11:64–72 31. Shaikh MS, Shariff AM, Bustam MA, Murshid G (2013) Physical properties of aqueous blends of sodium glycinate (SG) and piperazine (PZ) as a solvent for CO2 capture. J Chem Eng Data 58(3):634–638 32. Majchrowicz ME (2014) Amino acid salt solutions for carbon dioxide capture. Ph.D. dissertation, University of Twente, The Netherlands 33. Goan JC, Miller RR, Piatt VR. (1960) Alkazid M as a Regenerative Carbon Dioxide Absorbent. NRL Report 5465. Naval Research Laboratory, Washington DC 34. Holst JV, Versteeg GF, Brilman DWF, Hogendoorn JA (2009) Kinetic study of CO2 with various amino acid salts in aqueous solution. Chem Eng Sci 64(1):59–68 35. Nuchitprasittichai A, Cremaschi S (2011) Optimization of CO2 capture process with aqueous amines using response surface methodology. Comput Chem Eng 35(8):1521–1531 36. Caplow M (1980) Kinetics of carbamate formation and breakdown. J Am Chem Soc 90 (24):6795–6803 37. Ewing SP, Lockshon D, Jencks WP (1980) Mechanism of cleavage of carbamate anions. J Am Chem Soc 102(9):3072–3084 38. Chakraborty AK, Bischoff KB, Astarita G, Damewood JR (1988) Molecular orbital approach to substituent effects in amine-CO2 interactions. J Am Chem Soc 110(21):6947–6954 39. Lerche BM (2012) CO2 capture from flue gas using amino acid salt solutions. Ph.D. thesis, Technical University of Denmark 40. Gabrielsen J (2007) CO2 capture from coal fired power plants, IVCSEP. Ph.D. thesis, Technical University of Denmark 41. Da Silva EF (2005) Computational chemistry study of solvents for carbon dioxide absorption. Ph.D. Thesis, Norwegian University of Science and Technology, NO-7491, Trondheim 42. Darde V (2011) CO2 capture using aqueous ammonia. Ph.D. Thesis, Technical University of Denmark, Frydenberg, Copenhagen 43. Feron PHM, Asbroek NAM. (2004) New solvents based on amino-acid salts for CO2 capture from flue gases. GHGT-7, Vancouver, Canada 44. Goetheer ELV, Nell L (2009) First pilot plant results from TNO’s solvent development workflow. Carbon Capture J 8:2–3 45. Schneider R, Schramm H (2011) Environmental friendly and economic carbon capture from power plant flue gases: The SIEMENS PostCap technology. First Post Combustion Capture Conference, Abu Dhabi 46. Fernandez ES, Goetheer ELV (2011) DECAB process development of a phase change absorption process. Energy Procedia 4:868–875 47. Kumar PS, Hogendoorn JA, Timmer JS, Feron PHM (2003) Equilibrium solubility of CO2 in aqueous potassium taurate solutions: part 2. Experimental VLE data and model. Ind Eng Chem Res 42(12):2841–2852 48. Sanchez Fernandez E, Heffernan K, van der Ham L, Linders MJG, Eggink E, Schrama FNH, Brilman DWF, Goetheer ELV, Vlugt TJH (2013) Conceptual design of a novel CO2 capture process based on precipitating amino acid solvents. Ind Eng Chem Res 52(34):12223–12235 49. Misiak K, Sanchez Sanchez C, van Os P, Goetheer E (2013) Next generation post-combustion capture: combined CO2 and SO2 removal. Energy Procedia 37:1150–1159 50. Song H-J, Lee S, Maken S, Park JJ, Park JW (2006) Solubilities of carbon dioxide in aqueous solutions of sodium glycinate. Fluid Phase Equilib 246(1–2):1–5

148


51. Harris F, Kurnia KA, Mutalib MIA, Murugesan T (2009) Solubilites of carbon dioxide and densities of aqueous sodium glycinate solutions before and after CO2 absorption. J Chem Eng Data 54(1):144–147 52. Zhang W, Wang Q, Fang M, Luo Z (2010) Experimental Study on CO2 absorption and regeneration of aqueous solutions of potassium glycinate. 978-1-4244-4713-8/10/$25.00 ©2010 IEEE 53. Aronu UE, Hoff KA, Svendsen HF (2011) Vapor–liquid equilibrium in aqueous amine amino acids salt solution: 3-(methylamino) propylamine/sarcosine. Chem Eng Sci 66 (17):3859–3867 54. Song HJ, Lee MG, Kim H, Gaur A, Park JW (2011) Density, viscosity, heat capacity, surface tension, and solubility of CO2 in aqueous solutions of potassium serinate. J Chem Eng Data 56(4):1371–1377 55. Lee JI, Otto FD, Mather AE (1974) The solubility of H2S and CO2 in aqueous monoethanolamine solutions. Can J Chem Eng 52(6):803–805 56. Jou FY, Mather AE, Otto FD (1982) Solubility of hydrogen sulfide and carbon dioxide in aqueous methyldiethanolamine solutions. Ind Eng Chem Process Design Develop 21 (4):539–544 57. Lim J, Kim DH, Yoon Y, Jeong SK, Park KT, Nam SC (2012) Absorption of CO2 into aqueous potassium salt solutions of L-alanine and L-proline. Energy Fuels 26(6):3910–3918 58. Majchrowicz ME, Brilman DWF (2012) Solubility of CO2 in aqueous potassium L-prolinate solutions-absorber conditions. Chem Eng Sci 72:35–44 59. Kang D, Park S, Jo H, Min J, Park J (2013) Solubility of CO2 in Amino-Acid-Based Solutions of (potassium sarcosinate), (potassium alaninate + piperazine), and (potassium serinate + piperazine). J Chem Eng Data 58(6):1787–1791 60. Aldenkamp N, Huttenhuis P, Penders-van Elk N, Hamborg ES, Versteeg GF (2014) Solubility of carbon dioxide in aqueous potassium salts of glycine and taurine at absorber and desorber conditions. J Chem Eng Data 59(11):3397–3406 61. Chang YT, Leron RB, Li MH (2015) Carbon dioxide solubility in aqueous potassium salt solutions of L-proline and DL-a-aminobutyric acid at high pressures. J Chem Thermodyn 83:110–116 62. Mazinani S, Ramazani R, Samsami A, Jahanmiri A, Bruggen BV, Darvishmanesh S (2015) Equilibrium solubility, density, viscosity and corrosion rate of carbon dioxide in potassium lysinate solution. Fluid Phase Equilib 396:28–34 63. Shen S, Yang Y, Wang Y, Ren S, Han J, Chen A (2015) CO2 absorption into aqueous potassium salts of lysine and proline: Density, viscosity and solubility of CO2. Fluid Phase Equilib 399:40–49 64. Chen ZW, Leron RB, Li MH (2015) Equilibrium solubility of carbon dioxide in aqueous potassium L-asparaginate and potassium L-glutaminate solutions. Fluid Phase Equilib 400:20–26 65. Mazinani S, Samsami A, Jahanmiri A (2011) Solubility (at low partial pressures), density, viscosity, and corrosion rate of carbon dioxide in blend solutions of monoethanolamine (MEA) and sodium glycinate (SG). J Chem Eng Data 56(7):3163–3168 66. Privalova E, Rasi S, Maki-Arvela P, Eranen K, Rintala J, Murzin DY, Mikkola JP (2013) CO2 capture from biogas: absorbent selection. RSC Advances 3:2979–2994 67. Lu JG, Fan F, Lui C, Ji Y, Zhang H (2011) Performance evaluation of amino acid salt-based complex absorbents for CO2 capture. Environ Eng Manage J 10(2):297–304 68. Park S, Song HJ, Park J (2014) Selection of suitable aqueous potassium amino acid salts: CH4 recovery in coal bed methane via CO2 removal. Fuel Process Technol 120:48–53 69. Kumar PS, Hogendoorn JA, Versteeg GF (2003) Kinetics of the reaction of CO2 with aqueous potassium salt of taurine and glycine. AIChE J 49(1):203–213 70. Versteeg GF, Van Dijck LAJ, Van Swaaij WPM (1996) On the kinetics between CO2 and alkanolamines both in aqueous and non-aqueous solutions: an overview. Chem Eng Commun 144(1):113–158


149

71. Lee S, Song HJ, Maken S, Park JW (2007) Kinetics of CO2 absorption in aqueous sodium glycinate solutions. Ind Eng Chem Res 46(5):1578–1583 72. Portugal AF, Derks PWJ, Versteeg GF, Magalhãesa FD, Mendesa A (2007) Characterization of potassium glycinate for carbon dioxide absorption purposes. Chem Eng Sci 62(23):6534– 6547 73. Glasscock DA, Critchfield JE, Rochelle GT (1991) CO2 absorption desorption in mixtures of methyldiethanolamine with monoethanolamine or diethanolamine. Chem Eng Sci 46 (11):2829–2845 74. Portugal AF, Magalhães FD, Mendes A (2008) Carbon dioxide absorption kinetics in potassium threonate. Chem Eng Sci 63(13):3493–3503 75. Holst JV, Versteeg GF, Brilman DWF, Hogendoorn JA (2009) Kinetic study of CO2 with various amino acid salts in aqueous solution. Chem Eng Sci 64(1):59–68 76. Simons K, Brilman WDWF, Mengers H, Nijmeijer K, Wessling M (2010) Kinetics of CO2 absorption in aqueous sarcosine salt solutions: influence of concentration, temperature, and CO2 loading. Ind Eng Chem Res 49(20):9693–9702 77. Vaidya PD, Konduru P, Vaidyanathan M (2010) Kinetics of carbon dioxide removal by aqueous alkaline amino acid salts. Ind Eng Chem Res 49(21):11067–11072 78. Aronu UE, Hartono A, Hoff KA, Svendsen HF (2011) Kinetics of carbon dioxide absorption into aqueous amino acid salt: potassium salt of sarcosine solution. Ind Eng Chem Res 50 (18):10465–10475 79. Kim M, Song HJ, Lee MG, Jo HY, Park JW (2012) Kinetics and steric hindrance effects of carbon dioxide absorption into aqueous potassium alaninate solutions. Ind Eng Chem Res 51 (6):2570–2577 80. Paul S, Thomsen K (2012) Kinetics of absorption of carbon dioxide into aqueous potassium salt of proline. Int J Greenhouse Gas Control 8:169–179 81. Majchrowicz ME, Kersten S, Brilman W (2014) Reactive absorption of carbon dioxide in L-prolinate salt solutions. Ind Eng Chem Res 53(28):11460–11467 82. Guo D, Thee H, Tan CY, Chen J, Fei W, Kentish S, Stevens GW, Da Silva G (2013) Amino acids as carbon capture solvents: chemical kinetics and mechanism of the glycine + CO2 reaction. Energy Fuels 27(7):3898–3904 83. Sodiq A, Rayer AV, Olanrewaju AA, Abu-Zahra MRM (2014) Reaction kinetics of carbon dioxide (CO2) absorption in sodium salts of taurine and proline using a stopped-flow technique. Int J Chem Kinet 46:730–745 84. Fang M, Zhou X, Xiang Q, Cai D, Luo Z (2015) Kinetics of CO2 absorption in aqueous potassium L-prolinate solutions at elevated total pressure. Energy Procedia 75:2293–2298 85. Hikita H, Asai S, Ishikawa H, Honda M (1977) The kinetics of reactions of carbon dioxide with monoethanolamine, diethanolamine and triethanolamine by a rapid mixing method. Chem Eng J 13(1):7–12 86. Rangwala HA, Morrell BR, Mather AE, Otto FD (1992) Absorption of CO2 into aqueous tertiary amine/MEA solutions. Can J Chem Eng 70(3):482–490 87. Alpe E (1990) Reaction mechanism and kinetics of aqueous solutions of 2-amino-2-methyl-1-propanol and carbon dioxide. Ind Eng Chem Res 29(8):1725–1728 88. Bishnoi S, Rochelle GT (2002) Absorption of carbon dioxide in aqueous piperazine/methyldiethanolamine. AIChE J 48:2788–2799 89. Littel R, Van Swaaij W, Versteeg G (1990) Kinetics of carbon dioxide with tertiary amines in aqueous solution. AIChE J 36:1633–1640 90. Epp B, Hans F, Monika V (2011) Degradation of solutions of monoethanolamine, diglycolamine and potassium glycinate in view of tail-end CO2 absorption. Energy Procedia 4:75–80 91. Huang Q, Saloni B, Joseph ER, John P, Selegue KL (2013) Thermal degradation of amino acid salts in CO2 capture. Int J Greenhouse Gas Control 19:243–250 92. DuPart MS, Bacon TR, Edwards DJ (1993) Hydrocarbon Processing 72(5):89 93. Hawkes EN, Mago BF (1971) Hydrocarbon Process 50(8):109

150


94. Ahn S, Song HJ, Park JW, Lee JH, Lee IY, Jang KR (2010) Characterization of metal corrosion by aqueous amino acid salts for capture of CO2. Korean J Chem Eng 27(5):1576–1580 95. Bello A, Idem RO (2006) Comprehensive study of the kinetics of the oxidative degradation of CO2 loaded and concentrated aqueous monoethanolamine (MEA) with and without sodium metavanadate during CO2 absorption from flue gases. Ind Eng Chem Res 45 (8):2569–2579 96. Budzianowski WM (2015) Single solvents, solvent blends, and advanced solvent systems in CO2 capture by absorption: a review. Int J Global Warming 7(2):184–225 97. Shao R, Stangeland A (2009) Amines used in CO2 capture-health and environmental impacts. Bellona Report, September 2009 98. Glycine; MSDS Cat. code. SLG1972, SLG2191; [online]; Science Lab.Com, Inc 14025 Smith Road, Houston, Texas 77396USA. http://www.sciencelab.com/msds.php?msdsId= 9927179. Accessed 16 Dec 15 99. L-proline; MSDS Cat code. SLP5488, SLP2662, SLP4471, Science Lab.Com, Inc 14025 Smith Road, Houston, Texas 77396 USA. http://www.sciencelab.com/msds.php?msdsId= 9927237. Accessed 16 Dec 15 100. Taurine; MSDS Cat code. SLT2014, Science Lab.Com, Inc 14025 Smith Road, Houston, Texas 77396 USA. http://www.sciencelab.com/msds.php?msdsId=9925166. Accessed 16 Dec 15 101. L-Histidine; MSDS Cat code. SLH2317, SLH3163, SLH1843, Science Lab.Com, Inc 14025 Smith Road, Houston, Texas 77396 USA. http://www.sciencelab.com/msds.php?msdsId= 9927189. Accessed 16 Dec 15 102. L-Methionine; MSDS Cat code. SLM1987, SLM3361, SLM1216, Science Lab.Com, Inc 14025 Smith Road, Houston, Texas 77396 USA. http://www.sciencelab.com/msds.php? msdsId=9927225. Accessed 16 Dec 15 103. L-Glutamine; MSDS Cat code. SLG1462, SLG1963, Science Lab.Com, Inc 14025 Smith Road, Houston, Texas 77396 USA. http://www.sciencelab.com/msds.php?msdsId=9924160. Accessed 16 Dec 15 104. Monoethanolamine; MSDS Cat code. SLA4792, SLA2452, SLA3955, Science Lab.Com, Inc 14025 Smith Road, Houston, Texas 77396 USA. http://www.sciencelab.com/msds.php? msdsId=9922885. Accessed 16 Dec 15 105. Diethanolamine; MSDS Cat code. SLD1834, SLD3199, Science Lab.Com, Inc 14025 Smith Road, Houston, Texas 77396 USA. http://www.sciencelab.com/msds.php?msdsId=9923743. Accessed 16 Dec 15 106. Triethanolamine; MSDS Cat code. SLT3841, SLT1822, Science Lab.Com, Inc 14025 Smith Road, Houston, Texas 77396 USA. http://www.sciencelab.com/msds.php?msdsId=9927306. Accessed 16 Dec 15 107. Piperazine; MSDS Cat code. SLP4681, Science Lab.Com, Inc 14025 Smith Road, Houston, Texas 77396 USA. http://www.sciencelab.com/msds.php?msdsId=9926575. Accessed 16 Dec 15 108. Gabrielsen J (2007) CO2 capture from coal fired power plants, IVCSEP. Ph.D. Thesis, Technical University of Denmark 109. Jockenhoevel T, Schneider R, Rode H (2008) Development of an economic post-combustion carbon capture process, GHGT-9-Washington D.C, USA 110. Weiland RH, Hatcher NA, Nava JL (2010) Post-combustion CO2 capture with amino-acid salts, Optimized Gas Treating, Inc. Clarita, 74535, USA 111. Salazar V, Sanchez-Vincente Y, Pando C, Renuncio JAR, Cabanas A (2010) Enthalpies of absorption of carbon dioxide in aqueous sodium glycinate solutions at temperatures of (313.15 and 323.15) K. J Chem Eng Data 55(3):1215–1218 112. Oexmann J, Kather A (2010) Minimizing the regeneration heat duty of post-combustion CO2 capture by wet chemical absorption: the misguided focus on low heat of absorption solvents. Int J Greenhouse Gas Control 4(1):36–43 113. Amino acids formulas and molecular weight. Accessed 19 Dec 15


151

114. Walum E (1998) Acute oral toxicity. Environ Health Perspect 106(2):497–503 115. Shen KP, Li MH (1992) Solubility of carbon dioxide in aqueous mixtures of monoethanolamine with methyldiethanolamine. J Chem Eng Data 37(1):96–100 116. Tan LS, Shariff AM, Lau KK, Bustam MA (2012) Factors affecting CO2 absorption efficiency in packed column: a review. J Ind Eng Chem 18(6):1874–1883

Ionic Liquids: Advanced Solvents for CO2 Capture Xiangping Zhang, Lu Bai, Shaojuan Zeng, Hongshuai Gao, Suojiang Zhang and Maohong Fan

Abstract As one of promising advanced solvents, ionic liquids (ILs) have become more attractive for CO2 capture due to their unique properties, special structures and potential energy saving efficiency. This chapter mainly reviews the research progress on CO2 capture with ILs, focusing on the CO2 absorption capacity of conventional ILs, task-specific ILs and ILs based mixtures as well as the comparison and analysis. The influence of cations, anions and functional groups of ILs on the CO2 absorption was analyzed and the mechanisms of physisorption and chemisorption were revealed using experimental test and molecular simulation results. Especially considering the real applications of the new ILs-based capture technologies, the research on process simulation and assessment of CO2 capture processes was also reviewed. Finally, we discussed the challenges and opportunities of transferring the lab-scale research results to practical industries processes, and also present some perspectives of ILs based novel technologies.

1 Introduction A large amount of greenhouse gases in atmosphere, especially CO2 emitted from the fossil fuel combustion in the industrial processes, has contributed to serious global warming problems, which will lead to climate warming, sea level rising and aggravated disasters. CO2 capture and storage (CCS) has been regarded as one of the most promising options to reduce CO2 emissions. In CCS, CO2 separation is one of the most critical parts [1, 2]. Additionally, in the field of energy gas X. Zhang (&) L. Bai S. Zeng H. Gao S. Zhang State Beijing Key Laboratory of Ionic Liquids Clean Process, State Key Laboratory of Multiphase Complex Systems, Institute of Process Engineering, Chinese Academy of Sciences, Beijing, China e-mail: [email protected] M. Fan Department of Chemical and Petroleum Engineering, School of Energy Resources, University of Wyoming, Laramie, WY, USA © Springer International Publishing AG 2017 W.M. Budzianowski (ed.), Energy Efficient Solvents for CO2 Capture by Gas–Liquid Absorption, Green Energy and Technology, DOI 10.1007/978-3-319-47262-1_7

153

154

X. Zhang et al.

resources, like natural gas, coal bed gas, shale gas, biogas, and so on, CO2 as impurities in these gases should be removed before utilizations because CO2 components in these gases might decrease the heating values, demand high energy consumption for transport and cause pipeline corrosions [3]. Thus, CO2 separation has been received growing attentions due to its importance in environmental protection and industrial application. At present, there are already many technologies for CO2 capture, like physical or chemical solvent scrubbing, cryogenic distillation, pressure/temperature swing adsorption (P/TSA), gas membrane separation and so on. Although these technologies have their own advantages, most of them still suffer from some drawbacks. For examples, amine scrubbing by chemical reaction is the widely used method for CO2 capture in industry because of the excellent absorption performances even under low partial pressure of CO2, but there are some inherent shortcomings, such as the loss of solvent due to thermal or chemical degradation, equipment corrosion and high energy demands during the solvent regeneration [4, 5]. Cryogenic distillation requires substantial energy of refrigeration in real applications for CO2 separation [2]. Solid adsorbents used in the process of adsorption, like activated carbons, zeolites, metal organic frameworks, microporous organic polymers etc., show some advantages for CO2 capture, but the adsorption capacity, selectivities, stabilities, recycling, etc. of adsorbents still need to be improved for industrial applications. The membrane gas separation is another potential technology for CO2 capture because of less equipment space, lower energy requirement and absence of potentially hazardous chemicals; nevertheless, the compromise between permeability and selectivity of membranes limits its large-scale applications [2], as well as their higher costs. Thus, developing novel solvents/materials and corresponding new technologies for CO2 capture is an enduring R&D topic and highly desirable. Ionic liquids (ILs) as advanced solvents have been regarded as prospective candidates for CO2 capture because of their excellent properties and potential energy saving efficiency, and being paid more and more attentions in recent years. ILs are well known as fluid at room temperature and composed entirely of organic cations and organic/inorganic anions. They have some unique properties, including negligible vapor pressures, low melted points, high thermal stabilities and tunable structures, which make them be widely used in many areas, such as organic synthesis, catalysis, electrochemistry, biochemistry, gas separation, especially for CO2 capture. When capturing CO2 with ILs, the negligible vapour pressure of ILs makes no contamination in gas stream and negligible losses of ILs. More importantly, due to the negligible vapour pressure of ILs, stripping CO2 from rich solvent could be carried out by flash operation instead of distillation tower used in amine technologies, which needs much less energy consumption and lower equipment investment. For example, it was evaluated that CO2 capture with [bmim][Ac] IL can reduce the energy losses by 16 % and the economic investment will be 11 % lower compared to a MEA-based process [6]. It was also investigated by simulations that the process of IL-amine hybrid solvents ([Bpy][BF4]-MEA) for CO2 capture can save about 15 % regeneration heat duty compared to a MEA process [7]. Additionally, ILs are designable solvents, thus the physicochemical properties

Ionic Liquids: Advanced Solvents for CO2 Capture

155

could be tuned for aimed applications through structural modifications of cations and anions and/or changing the combination of cations and anions. Therefore, ILs as designable solvents according to the real applications is potential option for energy and cost efficient capturing CO2. Since it was firstly reported that CO2 could be efficiently dissolved in ILs [8], a series of conventional ILs with physical absorption and task-specific ILs with chemical absorption were reported in succession [9–12]. ILs captured CO2 by physical interaction as result of the special internal microstructure and force of ILs. According to ATR-IR spectrum, the high CO2 solubility was achieved by the Lewis acid-base interaction between the anion of IL and CO2 [13, 14], as well as the hydrogen-bonded interactions between the H proton of C2 atom on the imidazole ring and CO2 molecule [15]. Quantum chemistry and Molecular dynamics simulation have indicated that the cations and anions form the hydrogen bonding network in the imidazolium-based ILs and the molecules of CO2 can fill in the network formed by the cations of ILs so that the CO2 dissolves in ILs with high solubility [16]. Generally, CO2 solubility of ILs is mainly depended on the anion, hence the hydrogen-bonded interactions induce to a high CO2 solubility. Yu et al. [17] found that more charge-localized character of [TMG][L], especially the C1 carbocation on [TMG]+, and the intermolecular-NH2-associated hydrogen bonds can substantially increase the cation-anion interaction, and the interaction energy of [TMG]L is 65.3– 109.3 kJ/mol higher than some [bmim]+ based ILs, thus the CO2 solubility in [TMG]L is higher than that in [bmim][PF6]. In order to further improve CO2 absorption capacity, task-specific ILs were designed and developed subsequently [18–21]. Among these ILs, amino-functionalized ILs were well investigated, for example, the solubility in [aemmim][Tau] reached 0.9 mol CO2/mol IL [20]. Although significant progress has been made, the gravimetric capacities of these ILs which is much more interested in real industries, are usually lower than 0.1 g CO2/g IL [9], so later researchers focused on the improvement of CO2 gravimetric capacities in ILs. Wang et al. [22] synthesized a series of superbase-derived protic ILs, and the gravimetric capacities of [MTBDH][Im] reached 0.205 g CO2/g IL, but the thermal stability of ILs need to be improved. Our research group also developed a new dual amino-functionalized and the gravimetric capacities can reach 18 % [23]. Besides the CO2 capacities, the viscosity of IL solvents is another important property. Generally, the viscosity of task-specific ILs is relatively high and the viscosity increases after absorption due to the chemical reaction, for example, the viscosity of [aP4443][AA] increased nearly threefold after absorption of CO2 [21], and the viscosity of [P66614][Isoleucinate] increased over 200-fold when exposed to 1 bar of CO2 [24]. Therefore, the development of new ILs with low viscosity is extremely desirable. It was reported that the viscosity of [P66614][2-CNpyr] is less than 100 cP at 50 °C before exposure to CO2 and changes slightly after absorption of CO2, which is superior to the amino-functionalized ILs [24]. Additionally, mixing ILs with other solvents is an efficient way to offset the inherently high viscosities of ILs and also reduce the energy requirement. For instance, by mixing ILs with organic amine solvent, which only needs to break the binding energy between the solvent and CO2, and takes few energy to vapor water, so it is regarded

156

X. Zhang et al.

as one of the promising technology [25]. It was also found that the absorption capacities had nearly no change when ILs was mixed with solvents, while the viscosity decreased significantly and the whole energy consumption decreased about 20 % [10]. In summary, this chapter aims to introduce the research progress on CO2 capture with ILs, including conventional ILs, task-specific ILs and ILs based mixtures. The CO2 absorption capacity is summarized and compared, meanwhile, the relationship between structures and properties of ILs and CO2 absorption mechanism with experimental characteristic and molecular simulation is discussed in detail. The process simulation and assessment of CO2 capture with ILs was further reviewed for the practical application. Some challenges and perspectives of ILs based technology for CO2 are also presented.

2 Conventional Ionic Liquids for CO2 Separation The conventional IL [bmim][PF6] for CO2 capture was firstly reported and it can efficiently and physically absorb 0.75 mol CO2/mole IL at 25 °C and 8.3 MPa. [8]. An interesting result was found that a large quantity of CO2 dissolved in the IL phase, while less IL existed in the CO2 phase [26], showing that ILs has potential ability for CO2 separation and receive extensive attentions in recent years. The conventional ILs usually absorb CO2 through the physical interaction between the cations/anions and CO2, e.g. electrostatic interaction, van der Waals forces, hydrogen bonds. Therefore, the structures of cations and anions of ILs play an important role in CO2 solubility. The CO2 absorption capacities of ILs with physisorption have been concluded and tabulated in our previous review (Energy Environ. Sci., 2012, 5, 6668–6681, Table 1).

2.1

The Effect of Cations on CO2 Absorption

Imidazolium-based IL is most widely investigated and reported in literature for CO2 capture [8, 13, 27–29]. Kazarian et al. [13] thought that the high solubility of CO2 in imidazolium-based ILs is mostly attributed to the hydrogen bonding between the acidic hydrogen on the imidazolium and CO2. In order to understand the effect of the hydrogen in the C2 position of imidazolium on CO2 solubility, Brenneck et al. [27] studied three pairs of ILs: 1-n-butyl-3-methylimidazolium hexafluorophosphate ([bmim][PF6]) and 1-n-butyl-2,3-dimethylimidazolium hexafluorophosphate ([bmmim][PF6]);1-n-butyl-3-methylimidazolium tetrafluoroborate ([bmim][BF4]) and 1-n-butyl-2,3-dimethylimidazolium tetrafluoroborate ([bmmim][BF4]); and 1-ethyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide ([emim][Tf2N]) and 1-ethyl-2,3-dimethylimidazolium bis(trifluoromethylsulfonyl)imide ([emmim] [Tf2N]). As shown in Fig. 1, there is some decreased solubility for the


157

Fig. 1 Solubility of CO2 in [emim][Tf2N] (filled square); [emmim][Tf2N] (open square);[bmim][PF6] (filled circle); [bmmim][PF6] (open circle); [bmim][BF4] (filled triangle); [bmmim][BF4] (open triangle) at 25 °C [27]

methyl-substituted IL relative to the hydrogen-substituted IL in each pair, implying the weak influence of the cations on CO2 solubility. Furthermore, CO2 solubility in [bmim][Tf2N], [hmim][Tf2N] and [omim][Tf2N] was measured, and the influence of the length of alkyl chains on CO2 solubility was investigated by Aki et al. [30]. The results indicated that CO2 solubility marginally increases with an increase of the alkyl chain length on the cation. The similar results were obtained from the Yunus’s work [31]. Huang et al. [28] found that CO2 perhaps takes up free space from cavities already available in the rigid and intricate topography of ILs when CO2 physically dissolves in ILs, but there are no sufficiently large cavities to accommodate CO2 in neat ILs, so subtle rearrangements of ILs take place to create enough spaces, which results in no obvious volume expansion during the addition of CO2 and a slight increase of the densities of ILs (Fig. 2). Therefore, the increase of the side chain length on the cation can result in the greater free volume, which can increase CO2 solubility in ILs. In addition, some functional groups are introduced into the cations of ILs to increase CO2 solubility. Almantariotis et al. [29] studied the effect of fluorination on CO2 solubility in ILs, and found that the higher solubility of CO2 in the fluorine-substituted IL [C8H4F13mim][Tf2N] than the normal IL [C8mim][Tf2N] is due to the larger free volume of the fluorine-substituted IL and the stronger interaction between CO2 and the fluorinated alkyl chains (Fig. 3). In fact, the [Cnmim]based ILs do not possess both high CO2 solubility and ideal solubility selectivities. If the cations of ILs contain functional groups such as ethers, nitriles, the CO2 solubility in these ILs was similar to that in their analogues, but the solubilities of N2 and CH4 were much lower in these ILs. Therefore, the higher CO2/N2 and CO2/ CH4 selectivity could be obtained by using these ILs [25, 32, 33].

2.2

The Effect of Anions on CO2 Absorption

A large number of experimental data and simulation calculations demonstrated that the anions of ILs play a greater role in CO2 solubility in conventional ILs than

158

X. Zhang et al.

cations. Anthony et al. [34] investigated systematically the influence of cations (such as imidazolium, ammonium, pyrrolidinium, and phosphonium) and anions (Tf2N, PF6 and BF4) on CO2 solubility. When the anion is the same, the types of cations have a slight influence on CO2 solubility. In contrast, the influence of anions is greater, and the IL with the [Tf2N] anion has a considerably higher affinity to CO2 than either of the [BF4] IL and [PF6] IL, implying that the dominant role of the anion in CO2 dissolution. Brennecke et al. investigated CO2 solubility in the ILs with the same cation ([bmim]) and several different anions. The results indicated that the solubility of CO2 in [bmim] cation based ILs increases in the following order: [NO3] < [DCA] < [BF4] < [PF6] < [TfO] < [Tf2N] < [Methide] at 25 °C [30, 35]. Maiti and Sistla et al. also confirmed the results by the COSMO-RS method [13, 14]. Sergei et al. studied the mechanism of CO2 dissolution in ILs with the [BF4] and [PF6] anions and CO2 by in situ ATR-IR spectroscopy. The results indicated that the interaction is a Lewis acid-base type where the anion serves as a Lewis base, while CO2 acts as a Lewis acid, and the interaction between CO2 and [BF4] anion should be stronger than that for [PF6] anion, since [BF4] is a stronger base. However, experimental data showed a higher CO2 solubility in [bmim][PF6]

Fig. 2 Structural rearrangement of ILs due to CO2 dissolution [28] Fig. 3 Effect of fluorination groups on CO2 solubility [29]


159

rather than [bmim][BF4]. Thus, CO2 solubility in ILs could not solely be explained by anion-CO2 interactions [13]. In addition, CO2 solubility is higher in the ILs with fluorine groups in the anion than that in the ILs with nonfluorinated anions. It is to say that the CO2 solubility will be increased with the number of fluorine groups in the anion increases. For example, CO2 solubility increases in order of the anions: [BF4] < [PF6] < [Tf2N] < [Methide] < [eFAP] < [bFAP] [30, 35]. Zhang et al. [36] using COSMO-RS further predicted that the longer fluoroalkyl chain in the anion (e.g. FAP anion), the higher CO2 solubility obtained. Bhargava et al. [37] further studied the interaction of CO2 molecules with various anions of ILs using density functional theory. As shown in Fig. 4, the optimized structures of anion-CO2 complexes appeared to be dominated by Lewis acid-base interactions, but the larger anions such as Tf2N possesses two or three preferred sites for binding CO2 that could be related to their ability for CO2 dissolution, implying that the free volume of ILs also plays a significant role in dissolving CO2. The experimentally obtained data of CO2 solubility in ILs suggests a relationship between the molar volume of the anion and the solubility [38].

3 Task-Specific Ionic Liquids for CO2 Separation Although the emerging of ILs provides a new way for CO2 capture, CO2 physical solubility in conventional ILs is much lower than current commercially available solvents. Therefore, the design and development of the task-specific ILs for highly efficient capture of CO2 becomes a hot topic. The task-specific ILs also called functionalized ILs, include single amino ILs, dual amino ILs and non-amino ILs, which absorb CO2 through chemical interaction between the basic groups and CO2. The CO2 absorption capacities of ILs with chemisorption have been concluded and tabulated in our previous review (Energy Environ. Sci., 2012, 5, 6668–6681, Table 2). Except for these, some new ILs for CO2 capture with chemisorption are also reviewed in this part.

3.1

Single Amino ILs

Bates et al. [39] firstly reported the single amino-functionalized IL containing an amine group on the cation, which is capable of reversibly capture CO2 with a high capacity of nearly 0.5 mol CO2/mol IL. This is due to the IL could react with CO2 and form carbamate (Fig. 5), the stoichiometry of IL is similar to that of the aqueous MEA system (one CO2 molecule reacts with two amines). Subsequently, Zhang et al. [18] reported a series of single amino-functionalized ILs containing an amine group on the anions, and loaded them on porous silica gel for CO2 capture due to their high viscosity, and then 0.5 mol of CO2 per mol IL could be captured. The possible mechanism of chemical absorption of CO2 is that the CO2 is attacked

160

X. Zhang et al.

Fig. 4 The interaction between CO2 and anions: a NO3−, b BF4−, c N(CN)−, d CH3COO−, e PF6− and f Tf2N− [37]

by the free electron pair of the N atom on the NH2 group and thus forms a hydrogen bond with another NH2 group. Gurkan et al. [40] also reported two single amino-functionalized ILs containing an amine group on the anion. In their work, the ILs showed very high capacity of 0.9 mol of CO2 per mole of IL (Fig. 6). The introduction of NH2 groups into the imidazolium ring can effectively improve CO2 capacity, but not for all the cations with NH2 groups. Zhang et al. [41] studied CO2 absorption performances in the IL 1,1,3,3-tetramethylguanidinium


161

lactate ([TMG]L) with NH2 groups on the cation. [TMG]L has very low CO2 physical solubility of only 0.25 wt%, which does not comply with the absorption molar ratio of 1:2 between CO2 and NH2 group if it follows the same mechanism as [NH2p-bim][BF4] [42]. The underlying reason is the large FMO energy gap (9.53 eV) between HOMO-5 of [TMG]L and LUMO of CO2, which is much larger than the energy gap (6.07 eV) between HOMO of [NH2p-bim][BF4] and LUMO of CO2 as shown in Fig. 7. It is the carbocation that lowers the HOMO-5 energy of [TMG]L and weakens its nucleophilicity; as a result, [TMG]L cannot effectively interact with CO2 [43].

3.2

Dual Amino ILs

In order to enhance CO2 absorption capacity of amino-functionalized ILs, dual amino groups are tethered to the cations or anions of ILs. Zhang et al. [19] developed a series of dual amino-functionalized phosphonium ILs containing two amine groups on the anion and cation, respectively. Xue et al. [20] also reported a dual amino IL with amino-functionalized imidazolium cation and taurine anion. The CO2 absorption capacities of them were found to approach 1 mol of CO2 per mole of IL due to the reaction of both the cation and the anion with CO2. When the amine is tethered to the cation of the ILs, the amine will react with CO2 according to 1:2 mechanism. When the amine is tethered to the anion of the ILs, the amine will

Fig. 5 Proposed reaction mechanism of CO2 with [NH2p-bim][BF4] [39]

Fig. 6 Proposed reaction mechanism of CO2 with [P66614][Pro] [40]

162

X. Zhang et al.

Fig. 7 The HOMO and LUMO energies for [NH2p-bim][BF4], [TMG]L and CO2 [43]

react with CO2 according to 1:1 mechanism. Therefore, for the dual amino-functionalized ILs containing two amine groups on the anion and cation, the theoretical absorption capacity should be about 1.5 mol of CO2 per mole of IL. Xue et al. [20] thought the reason is that the R-N+H2COO− formed by the amine tethered anion reacts with CO2 is not stable, it can react with another amine to form R-NHCOO− (Fig. 8). Zhang et al. [23] reported a dual amino-functionalized IL containing two amine groups tethered to the cation of IL. The CO2 absorption capacities of this IL could be up to 1.05 mol of CO2 per mole of IL. FTIR and NMR spectra proved that the IL could react with CO2 and form carbamate on the basis of 1:2 stoichiometry (Fig. 9).

3.3

Non Amino ILs

Although the amino-functionalized ILs could improve the CO2 absorption capacities by the reaction of amine in the cation or the anion with CO2, the viscosity of the amino-functionalized ILs is usually too high, which restricts their further applications. Moreover, the strong interaction between ILs and CO2 increases the difficulty of the regeneration. Wang et al. [22, 44–47] proposed a novel strategy for equimolar CO2 capture by a series of non-amino functionalized ILs. Firstly, Wang et al. [22] synthesized several superbased-derived protic ILs by neutralization of a superbase with weak proton donors, such as fluorinate alcohols, imidazoles, pyridines. The results showed that [MTBDH][TFE] and [MTBDH][Im] has very high absorption capacity of 1.13 and 1.03 mol CO2/mol IL at 23 °C and 0.1 MPa,

Ionic Liquids: Advanced Solvents for CO2 Capture NH2 2 O3S

+ CO2

163

NHCOO O3S

SO3

H3N

Fig. 8 Proposed reaction mechanism of CO2 with the anion [Tau] [20]

Fig. 9 Proposed reaction mechanism of CO2 with DAIL-Br [23]

respectively. Due to the weak proton donor was deprotonated, the ILs will have a thermodynamic driving force for the reaction with CO2. The mechanism of CO2 capture by these ILs was also proved by spectroscopic investigation and quantum chemical calculations (Fig. 10). Subsequently, Wang et al. [46] designed and prepared phenolic ILs for the efficient and reversible capture of CO2 from phosphonium hydroxide and substituted phenols. The results showed that [P66614][4-Me-PhO] and [P66614][4-Cl-PhO] exhibited the high absorption capacity of 0.91 and 0.95 mol CO2/mol IL. Recently, Wang et al. [47] also proposed another strategy for improving CO2 capture by new anion-functionalized ILs making use of multiple site cooperative interactions. An extremely high capacity of CO2 is up to 1.60 mol CO2/mol IL and excellent reversibility is achieved. The plausible mechanism of CO2 absorption is shown in Fig. 11.

4 Ionic Liquid Mixed Solvents for CO2 Separation Although ILs are proposed as potential solvents for CO2 separation due to their unique properties, there are also some other problems which may limit their industrial applications, such as higher viscosity of ILs than that of other molecular solvents. It is reported that the viscosity of ILs could decrease greatly by adding a small amount of water or organic solvent [48]. Therefore, the mixture of ILs with water or organic solvents as a new kind of absorbent for CO2 separation has paid much attention by researchers. At present, the research on ILs mixed solvents for CO2 capture mainly focused on IL-water solutions, IL-alkanolamine blends, IL-organic solvents and IL-IL mixtures. The CO2 absorption capacities of some different IL mixtures are listed in Table 1.

164

X. Zhang et al.

Fig. 10 CO2 absorption by the anions of superbased-derived protic ionic liquids [22]

Fig. 11 CO2 reaction mechanism by [P66614][2-Op] through multiple-site interactions [47]

4.1

IL-Water Systems

The presence of water in ILs could significantly affect the physical properties of ILs especially viscosity, even a very small amount of water could result in sharp decrease of viscosity. The first research about the influence of water on the solubility of CO2 in two hygroscopic ILs butylmethylimidazolium nitrate ([Bmim] [NO3]) and hydroxypropylmethylimidazolium nitrate ([Hopmim][NO3]) was reported by Bermejo et al. [49]. It was found that when the water concentration is low, the CO2 solubility in [Bmim][NO3]-water solution is slightly higher than that in water-free [Bmim][NO3]. However, when the water concentration is high, it will lead to lower CO2 solubility. It is mainly explained by the positive excess molar volumes of the aqueous solutions at the working conditions. Later, the investigation of other IL-water systems for CO2 separation also showed the existed water in ILs could extremely reduces the viscosity while cause a slight decrease of CO2 absorption capacity [50, 51, 61]. Ventura et al. [50] found that the solubility of CO2 in the aqueous (tri-iso-butyl(rnethyl)phosphonium tosylate) [iBu3MeP][TOS] system increase with the increase of IL molar composition revealing a salting-in effect promoted by the IL. Goodrich et al. [61] also reported that with 14 wt% water, the CO2 capacity of trihexyl(tetradecyl)phosphonium prolinate ([P66614][Pro]) is reduced by approximately 0.2 mol of CO2 per mole of IL at 0.25 bar and by 0.1 mol of CO2 per mole of IL at 1 bar. In addition, Wang et al. [51] found that triethylbutylammonium acetate ([N2224][CH3COO]) has a little larger absorption capacity than its water complexes due to the different reaction mechanisms.


165

Table 1 The CO2 capacities of ILs mixturesa Ionic liquids mixture

[Bmim][NO3] + H2O (Rmo = 98.01:1.99) [Hopmim][NO3] + H2O (Rmo = 95.89:4.11) [iBu3MeP][TOS] + H2O (Rmo = 4:96) [N2224][CH3COO] + H2O (Rmo = 1:2) [P66614][2-CNPyr] + H2O (Rma = 95.5:4.5) [Hmim][NTf2] + MEA (Rmo = 1:1) [N1111][Gly] + MDEA (Rma = 1:1) [N2222][Gly] + MDEA (Rma = 1:1) [N1111][Lys] + MDEA (Rma = 1:1) [N2222][Lys] + MDEA (Rma = 1:1) [C3OHmim][Cl] + MEA (Rmo = 1:2) [C2OHmim][DCA] + MEA (Rma = 1:3) [Bmim][DCA] + MEA (Rma = 1:3) [Choline][Pro] + PEG 200 (Rma = 1:1) [Emim][Ac] + [Emim][TFA] (Rmo = 49.98:50.02) [Emim][BF4] + [Omim][NTf2] (Rma = 1:1) [Bmim][BF4] + [Omim][NTf2] (Rma = 1:1) [Emim][EtSO4] + [Emim][Ac] (Rmo = 1:1) a Rma mass ratio, Rmo mole ratio

CO2 absorption capacity molCO2/mol IL Lit. data

Conditions bar/K

References

*0.1

1.53/323

[49]

*0.099

2.33/318

[49]

*0.030 0.197 0.9

4.5/289 0.1/298 0.1/295

[50] [51] [52]

0.5 0.56 0.64 0.69 0.74 0.396 0.638 0.652 0.520 0.124

0.1/313 0.097/298 0.097/298 0.097/298 0.097/298 0.1/308 0.185/313 0.233/313 0.107/338 0.1/323

[53] [54] [54] [54] [54] [55] [56] [56] [57] [58]

0.32

2.18/313

[59]

0.34

2.22/313

[59]

0.19

3.93/298

[60]

Without water, [N2224][CH3COO] can only absorb CO2 via a Lewis acid and base reaction, while reactions involving [N2224][CH3COO]-nH2O could easily reach the CO2 saturation because of the formation of a stable acetic acid-H2O compound. Recently, enhanced CO2 capture in a system of 1-alkyl-3-methylimidazolium tricyanomethanide ([CnC1im][TCM]) and H2O was reported by Romanos [62]. It was found that the lower water concentrations lead to a reduction of the solubility of CO2 in the hybrid solvent. However, an increment of water content resulted in an enhancement of CO2 absorption in the mixture of [CnC1im][TCM] + H2O system compared to the dry ILs, in contrast to the detrimental influence of water on the CO2 solubility for most ILs. A molecular exchange mechanism between CO2 in the gas phase and H2O in the liquid phase was used to explain the enhanced CO2 absorption in the hybrid solvents. When the concentration of water reaches a critical value, the interaction between IL and H2O molecules will be broken and H2O will

166

X. Zhang et al.

be replaced by CO2. A series of trihexyl(tetradecyl)phosphonium 2-cyano-pyrrolide ([P66614][2-CNPyr]) + H2O mixtures below the saturation limit was studied for CO2 absorption and showed that an increase of H2O content leads to a slightly increased solubility of CO2 and the shape of the isotherm was changed dramatically [52]. The enhanced CO2 solubility could be ascribed to the changes in the activity of the IL-CO2 complex upon addition of water. The mixture of ILs based on carboxylate anions formulated with water with a high chemical absorption for CO2 were investigated [63]. The authors thought that the high solubility of CO2 is attributed to the basicity of the ILs anions activates the reaction between water and CO2 to form the hydrogencarbonate anion and the conjugate acid of the anion. Adding a certain amount of water into the ILs will slightly decrease the solubility of CO2 for most of ILs, but will slightly increase the solubility of CO2 for certain functionalized ILs. The affect of water on the absorption mechanism of IL + water should be investigated deeply in the future. Additionally, the long term stability of the IL + water should be considered for industrial application because the water could lead to the hydrolysis of ILs at high temperature.

4.2

IL-Alkanolamine Systems

Due to the fact that CO2 solubility in conventional ILs is not satisfactory, thus functionalized ILs with high CO2 capacity have been developed. However, because of the higher viscosity and complicated synthetic and purification steps compared to conventional ILs, the functionalized ILs do not appear to be viable for industrial process. In 2008, the idea of mixing ILs and alkanolamines for CO2 capture was put forward by Camper et al. [53]. In their study, the 1-hexyl-3-methyl-imidazolium bis (trifluoromethylsulfonyl)imide ([Hmim][NTf2]) solutions containing 50 mol% are able to rapidly and reversibly capture 1 mol of CO2 per 2 mol monoethanolamine (MEA) to give an insoluble MEA-carbamate precipitate that helps to drive the capture reaction. The desirable properties of ILs and high performance of alkanolamines for CO2 may be incorporated and energy can be saved during the regeneration process without affecting the absorption performance. Ahmady et al. [64] studied the solubility of CO2 in the aqueous mixture of N-methyldiethanolamine (MDEA) with 1-butyl-3-methyl-imidazolium tetrafluoroborate ([Bmim][BF4]). The results showed that the presence of a low concentration of [Bmim][BF4] in aqueous MDEA has no significant effect on the mixture loading capacity, but increased the initial absorption rate. The CO2 loading decreased with increasing [Bmim][BF4] concentration in the mixture due to a lack of water at high concentrations of [Bmim] [BF4]. They also investigated the solubility of CO2 in the aqueous mixture of MDEA with other two types of ILs, 1-butyl-3-methyl-imidazolium acetate ([Bmim] [CH3COO]) and 1-butyl-3-methyl-imidazolium dicyanamide ([Bmim][DCA]), and found that CO2 loading decreased significantly as the ILs concentration increased [65]. Four functionalized amino acid ILs: tetramethylammonium glycinate ([N1111] [Gly]), tetraethylammonium glycinate ([N2222][Gly]), tetramethylammonium


167

lysinate ([N1111][Lys]) and tetraethylammonium lysinate ([N2222][Lys]) were mixed with MDEA aqueous solutions to capture CO2 by Zhang et al. [54, 66, 67]. The results suggested that ILs could greatly enhance both the absorption and absorption rate of CO2 in MDEA aqueous solutions, because the ILs could capture CO2 through chemical absorption. Huang et al. [55] found that the chloride ions could significantly enhance the capacity and thermal stability of CO2 captured by MEA in hydroxyl imidazolium based ILs. This phenomenon may be caused by the hydrogen bonding and electrostatic attraction between the chloride and the cationic species. Sairi et al. [68] investigated the systems of guanidinium trifluoromethanesulfonate ([gua][OTf]) and MDEA for absorption of CO2., which indicated that the addition of [gua][OTf] to MDEA solution leads to a slight decrease for CO2 solubility. Yang et al. [69] used the mixed IL-amine solution (30 wt% MEA + 40 wt% [Bmim] [BF4] + 30 wt% H2O) to capture CO2, and showed that the energy consumption of the hybrid system for absorbent regeneration was 37.2 % lower than that of aqueous MEA solution. The MEA loss per ton of captured CO2 for the mixed solution was 1.16 kg, which is much lower than that of 3.55 kg for the aqueous amine solution. Our group also developed a series of IL-alkanolamine mixtures to capture CO2 and obtained some good results. A series of novel ILs, 2-aminoethanol tetrafluoroborate ([MEA][BF4]), 2-[2-hydroxyethyl(methyl)-amino] ethanol tetrafluoroborate ([MDEA][BF4]), 2-[2-hydroxyethyl(methyl)amino] ethanol chloride ([MDEA][Cl]), 2-[2-hydroxyethyl(methyl)amino] ethanol phosphate ([MDEA][PO4]), and 2[2-hydroxyethyl(methyl)amino] ethanol sulfate ([MDEA][SO4]), were synthesized and blended with amines and H2O to capture CO2. Among these absorbents, the MDEA + [MDEA][Cl] + H2O + piperazine system shows the best performance on CO2 capture [64]. Two low viscous ILs, 1-(2-hydroxyethyl)-3-methyl-imidazolium dicyanamide ([C2OHmim][DCA]) and [Bmim][DCA] were also mixed with aqueous 30 wt% MEA for CO2 absorption. The results indicated that the CO2 loading decreased significantly as the ILs concentration increased [56]. Another three kinds of ILs [Bmim][BF4], [Bmim][NO3] and [Bmim][Cl] were selected to blend with MDEA + PZ aqueous solution to capture CO2. The results showed that the influence of ILs on the CO2 cyclic capacity following the order of MDEA + PZ + [Bmim] [BF4] > MDEA + PZ + [Bmim][Cl] MDEA + PZ > MDEA + PZ + [Bmim] [NO3], indicating that an addition of [Bmim][BF4] decreased the sensible heat [70]. Because of certain amount of water replacing by ILs, the energy consumption of IL-alkanolamine mixtures would be lower than that of aqueous alkanolamine solution. For the functionalized ILs, the solubility of CO2 in IL-alkanolamine mixtures increases as the ILs concentration increased. Therefore, IL-alkanolamine mixtures may be a potentail solvent to capture CO2 in an industrial scale.

4.3

IL-Organic Solvent Systems

The viscosity of ILs could decrease significantly by adding a small amount of organic solvent besides water. Therefore, the mixture of ILs and organic solvents

168

X. Zhang et al.

are investigated as other candidates for CO2 capture. Hong et al. [71] measured the solubility of CO2 in the mixture of 1-ethyl-3- methylimidazolium bis (trifluoromethanesulfonyl)amide ([Emim][NTf2]) + acetonitrile (CH3CN) at 0.1 MPa and temperature between 290 and 335 K. The results indicated that the solubility of CO2 decrease about 50 % with the mole fraction of acetonitrile increasing from 0 to 0.77. Li et al. [57] investigated the absorption of CO2 using the mixture of (2-hydroxyethyl)-trimethyl-ammonium (S)-2-pyrrolidine-carboxylic acid salt ([Choline][Pro]) and PEG 200 at the temperature from 308.15 to 353.15 K under the ambient pressure. The results showed that the molar ratio of CO2 to the IL could exceed 0.5 slightly, which is the theoretical maximum for the absorption of CO2 chemically, indicating that both chemical and physical absorption exists. Addition of PEG200 in the IL could improve the rates of absorption and desorption of CO2. Ahnn et al. [72] measured the solubility of CO2 in the mixtures of dimethyl carbonate (DMC) + [Hmim][NTf2] at several temperatures between 303.15 K and 333.15 K and at pressures up to about 7 MPa, showing that the CO2 solubility decreased as the IL content increased at the same pressure and temperature. Wang et al. [73] investigated the absorption performance of CO2 by 1-butyl-3-methylimidazolium bis(trifluoromethanesulfonyl)amide ([Bmim] [NTf2]) + 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) and [Bmim][NTf2] + 1,3,4,6,7,8-hexahydro-1-methyl-2H-pyrimido[1,2-a]pyrimidine (MTBD), and the results suggested that the CO2 absorption capacity is about 1 mol per mole of IL, which is superior to traditional ILs, indicating that both chemical and physical absorption occurred simultaneously. Lei et al. [59, 74] determined the solubility of CO2 in the mixtures of propanone + 1-ethyl-3-methylimidazolium tetrafluoroborate ([Emim][BF4]) or acetone + [Bmim][BF4] and found that the mixtures could be applied as promising solvents for capturing CO2 since they combine the advantages of organic solvents and ILs. Lei et al. [75] also measured the solubility of CO2 in the mixtures of methanol + 1-octyl-3-methylimidazolium bis(trifluoromethanesulfonyl)amide ([Omim][NTf2]) at 273.2, 258.2, 243.2, 228.2 K and pressures up to 3.0 MPa. The process simulation reveals that the volatile loss of methanol decreases in some degrees by adding [Omim][NTf2] into methanol. The mixtures of IL and organic solvent show potentials for CO2 capture, but the stability and loss of the solvent should be considered when being applied for real industries.

4.4

IL-IL Systems

The mixtures of ILs themselves are also reported to capture CO2, which could provide an additional degree of tailoring over the intrinsic tunable properties of “single” ILs, while totally maintaining the IL property. Finotello et al. [49] found that the selectivity for CO2 with N2 and CH4 in the 90 and 95 mol% mixtures of [Emim][BF4] in [Emim][NTf2] was higher than in both pure components. Shiflett et al. [58] measured the solubility of CO2 in the IL mixture containing


169

equimolar amounts of 1-ethyl-3-methylimidazolium acetate ([Emim][Ac]), 1-ethyl-3-methylimidazolium trifluoroacetate ([Emim][TFA]) at three temperatures (298.1, 323.1, and 348.1 K) and pressures up to about 2 MPa. The results indicated that the mixture could absorb CO2 by chemical and physical interaction. Lei et al. [76] determined the solubility of CO2 in binary mixtures of [Emim] [BF4] + [Omim][NTf2] and [Bmim][BF4] + [Omim][NTf2] at high pressure up to 6.0 MPa for physical absorption. The Henry’s law constants decreased with increasing [Omim][NTf2] content in [Emim][BF4] or [Bmim][BF4], and were consistent with the COSMO-RS calculation. Pinto et al. [4] investigated two binary mixtures of [Emim][NTf2] + 1-ethyl-3-methylimidazolium ethylsulfate ([Emim] [EtSO4]) and [Emim][NTf2] + 1-butyl-3-methylimidazolium ethylsulfate ([Bmim] [EtSO4]) to absorb CO2 at 298 K and pressures up to 1.6 MPa. Although the absorption capacity of the mixtures was not higher than that of [Emim][NTf2], the cost and toxicity of the IL mixtures could be tuned by composition. They also measured the CO2 solubility in the mixture of [Emim][Ac] + [Emim][EtSO4] at 298.2 and 353.2 K, and at pressures up to 1.7 MPa. The results suggested that the addition of [Emim][EtSO4] to [Emim][Ac] prevent the solidification of the product resulting from the chemical reaction between CO2 and [Emim][Ac] [60]. Wang et al. [77] studied the CO2 absorption performance using the binary ILs of 1-butyl-3-propylamineimidazolium tetrafluoroborate ([NH2p-bim][BF4]) + [Emim] [BF4] + [Bmim][BF4]. The results showed that when the mole fraction of [NH2e-mim][BF4] was 0.4, the CO2 absorption performance and cost of the mixtures were the best. The CO2 absorption capacity and CO2 absorption rate decreased with the increase of the absorption temperature. IL-IL mixtures, in particular of those that combine physical and chemical absorption of CO2, may be better than simple IL for the development of CO2 capture processes. This method could balance the absorption capacity, the thermo-physical properties, cost and toxicity of the ILs.

5 Simulation and Assessment of Ionic Liquids Process for CO2 Separation System integration and assessment is very indispensable for developing new processes. As a new kind of solvents, a systematic assessment of CO2 capture process with ILs is rarely studied, so a comprehensive simulation and assessment is very necessary for CO2 separation with ILs in order to quickly compare different strategies, analyse process sensitivity, estimate the whole operating and equipment cost, understanding the energy consumption and the conversion efficiency, and then finally find the optimum route. Many researchers have performed experiments for CO2 capture with various ILs, as ILs possess good properties of non-volatile, good stability and relative low viscosity. However, since the shortage of rigorous thermodynamic data for complicated system, IL-based process simulation was rarely reported [78]. Shiflett et al.

170

X. Zhang et al.

[6] simulated a CO2 capture process using pure [Bmim][Ac]. The whole process consisted of two parts: absorption and solvent regeneration. Compared with the commercial MEA-based process, their IL-based process separated about 90 % of CO2 from the flue gas slip stream and the annual capacity was 47,000 metric tons, the IL-based process showed a higher recovery of 91.3 % and achieved higher CO2 purity of 98.7 %. Moreover, the IL-based system reduced 16 % energy consumption and the assessment result showed that the investment for the IL process was 11 % lower than the MEA-based process and also 12 % lower in equipment footprint, respectively [6]. Basha et al. [79] developed a process using three flash drums as regenerators to recycle the [hmim][Tf2N] for CO2 capture. Through three different pressure adiabatic flash drums (the pressure is 20, 10, 1 bar respectively), the CO2 gas, containing some H2 and H2O vapor, can be separated from the IL. Then the IL can be recycled into the absorbers. The results show that at the absorber conditions (2.4 m inner diameter, 30 m high), the [hmim][Tf2N] IL could achieve a CO2 recovery about 97.27 %. Eisinger et al. [80] employed an IL-[P2222][BnIm] to capture CO2 from the flue gas, which showed a phase change from solid to liquid upon reaction with CO2. Then the process can make a good use of the heat generated by this phase change to reduce parasitic power consumption, which is 55 % lower than that of the MEA process. And assessment indicated that the cost of electricity is 28 % lower but the capital cost is higher than the conventional MEA process. Not only the pure ionic liquid, researchers also found that the IL-amine hybrid solvents showed more effective aspect for CO2 capture, which can combine both the advantages of IL and the traditional solvents. Huang et al. [7] studied three ILs ([Bmim][DCA], [Bmim][BF4] and [Bpy][BF4]) with MEA aqueous solution and compared with the process simulation method. The results showed that compared with the conventional MEA process, the [Bpy][BF4]-MEA process can reduce 15 % in regeneration heat duty, 12 % in the whole energy penalty and 13.5 % in capture cost. As a consequence, this combined solvent process can be regarded as energy-saving and cost-efficient carbon capture process. These previous work opens a door of using IL solvents for CO2 separation, which can achieve the target of high CO2 recovery and low energy consumption compared to traditional processes. As is known, it is an important approach to simulate and assess the separation process using IL-based solvents. However, since the simulation work is different from the actual condition which much loss of the energy and substances can cause. Then many assumptions should be established and different conditions should be considered.

6 Vision Due to the unique properties of ILs, they have been proposed as one of promising CO2 technologies from the economic and environmental viewpoints. In this chapter, the research progress on the CO2 capture of conventional ILs, task-specific ILs and ILs based mixtures have been reviewed and discussed. The CO2 solubility


171

of conventional ILs is lower because of the weak physical interaction between ILs and CO2, while the solubility of task-specific ILs, especially with amine groups is much higher due to the chemical reaction with CO2. The addition of other substances to form ILs mixed solvents could significantly decrease the viscosity of ILs while the desired properties of ILs for CO2 are remained. It was described and explained through the experimental and simulation method that the cations, anions and functional groups have influences on the CO2 absorption. The process simulation of ILs/ILs mixtures for CO2 capture presented that IL system is an energy-saving and cost efficient way for separating CO2. Although CO2 capture with ILs/ILs based mixtures has been well developed, there are still some barriers which need to be solved in future works for the industrialization of ILs technology. Firstly, the high viscosity and moderate capacity of ILs is a great obstacle for industrial applications. So designing new ILs/ILs based mixtures to improve the CO2 gravimetric capacity and selectivity as well as decreasing the viscosity is extremely important. To this end, it is critical to deeply study and understand the relationship of structures, properties and absorption mechanism of ILs, and then directing the design and synthesis of ILs with certain characters for efficient CO2 separation. Secondly, the high price of ILs themselves has significantly affected the investment of the new ILs synthesis processes. Most ILs are prepared in laboratory scale with complex synthesis and purification steps, which leads to the high cost of ILs. Thus the improvement of synthetic method and the production on a large scale to reduce the price of ILs at a certain extent should be focused on. Thirdly, the impacts on the environment of IL process are not clear due to the unknown safety and health of ILs, so the toxicity and degradation of ILs should be studied thoroughly before commercializing the IL process. Except for these, the effect of other gas components on CO2 capture should be systematically studied in future work, because the practical separation processes involving gas mixtures, including CO2 and other gases, like oxygen, water, SO2, and so on, which has influence on the CO2 absorption with ILs. In addition, a systematic assessment of the new processes based on a thermodynamic model for ILs is very scarce. Thus, rigorous thermodynamic of IL complex system and reasonable simulation assumptions should be established, and also different conditions should be considered in future works, in order to assess the feasibility and reasonability of the IL process before accepted by industrial. Finally, before the lab research results are transferred to large scaled processes, building a pilot plant test is helpful to find and clearly understand the problems existed in ILs technology and hence realize successful industrialization. Currently, ILs have also been combined with other technologies like absorption and membrane gas separation to develop new technologies for CO2 capture, in which ILs based membrane for gas separation is regarded as future research trends of ILs. Making ILs immobiled to form ILs based membranes could significantly improve the permeability and selectivity, reduce the energy consumption and the membrane area, meanwhile, it could overcome the high viscosity and cost of ILs. Some works about combining ILs with membrane gas separation have been developed, including supported ionic liquids membranes, poly(ionic liquid)

172

X. Zhang et al.

membranes, ionic liquids based composite membranes, etc. Although ILs based membranes connecting the advantages of ILs and membranes, it is still a big challenge to develop novel ILs based membranes with higher permeability and selectivity, thermal and mechanical stability and thinner membranes in order to meet the requirement of industrial applications. Moreover, the separation mechanism of membranes is not cleared to date, so the influence of membrane structures, interaction between ILs and polymers on the membrane performance should be investigated deeply in order to explain the gas separation mechanism, hence guiding the designing of membranes and future industrial applications. Acknowledgments This work was supported by the National Natural Science Fund for Distinguished Young Scholars (No. 21425625), the National Natural Science Foundation of China (No. 51574215, 21506219), the International S&T Cooperation Program of China (No. 2014DFA61670), and the External Cooperation Program of BIC, Chinese Academy of Sciences (No. l22111KYS820150017).

References 1. Service RF (2004) Choosing a CO2 separation technology. Science 305(5686):963 2. Kenarsari SD, Yang D, Jiang G, Zhang S, Wang J, Russell AG, Wei Q, Fan M (2013) Review of recent advances in carbon dioxide separation and capture. Rsc Adv 3(45):22739–22773 3. Huang K, Zhang X-M, Xu Y, Wu Y-T, Hu X-B, Xu Y (2014) Protic ionic liquids for the selective absorption of H2S from CO2: thermodynamic analysis. AIChE J 60(12):4232–4240 4. Pinto AM, Rodriguez H, Colon YJ, Arce A Jr, Arce A, Soto A (2013) Absorption of carbon dioxide in two binary mixtures of ionic liquids. Ind Eng Chem Res 52(17):5975–5984 5. Sharma P, Park SD, Baek IH, Park KT, Yoon YI, Jeong SK (2012) Effects of anions on absorption capacity of carbon dioxide in acid functionalized ionic liquids. Fuel Process Technol 100:55–62 6. Shiflett MB, Drew DW, Cantini RA, Yokozeki A (2010) Carbon dioxide capture using ionic liquid 1-butyl-3-methylimidazolium acetate. Energy Fuels 24:5781–5789 7. Huang Y, Zhang X, Zhang X, Dong H, Zhang S (2014) Thermodynamic modeling and assessment of ionic liquid-based CO2 capture processes. Ind Eng Chem Res 53(29):11805– 11817 8. Blanchard LA, Hancu D, Beckman EJ, Brennecke JF (1999) Green processing using ionic liquids and CO2. Nature 399(6731):28–29 9. Zhang X, Zhang X, Dong H, Zhao Z, Zhang S, Huang Y (2012) Carbon capture with ionic liquids: overview and progress. Energy Environ Sci 5(5):6668–6681 10. Zhao Y, Zhang X, Zhen Y, Dong H, Zhao G, Zeng S, Tian X, Zhang S (2011) Novel alcamines ionic liquids based solvents: Preparation, characterization and applications in carbon dioxide capture. Int J Greenhouse Gas Control 5(2):367–373 11. Wappel D, Gronald G, Kalb R, Draxler J (2010) Ionic liquids for post-combustion CO2 absorption. Int J Greenhouse Gas Control 4(3):486–494 12. Vega LF, Vilaseca O, Llovell F, Andreu JS (2010) Modeling ionic liquids and the solubility of gases in them: recent advances and perspectives. Fluid Phase Equilib 294(1–2):15–30 13. Kazarian SG, Briscoe BJ, Welton T (2000) Combining ionic liquids and supercritical fluids: in situ ATR-IR study of CO2 dissolved in two ionic liquids at high pressures. Chem Commun 20:2047–2048 14. Cammarata L, Kazarian SG, Salter PA, Welton T (2001) Molecular states of water in room temperature ionic liquids. Physical Chemistry Chemical Physics 3(23):5192–5200


173

15. Crowhurst L, Mawdsley PR, Perez-Arlandis JM, Salter PA, Welton T (2003) Solvent-solute interactions in ionic liquids. Phys Chem Chem Phys 5(13):2790–2794 16. Dong K, Zhang S, Wang D, Yao X (2006) Hydrogen bonds in imidazolium ionic liquids. J Phys Chem A 110(31):9775–9782 17. Yu G, Zhang S (2007) Insight into the cation-anion interaction in 1,1,3,3-tetramethylguanidinium lactate ionic liquid. Fluid Phase Equilib 255(1):86–92 18. Zhang J, Zhang S, Dong K, Zhang Y, Shen Y, Lv X (2006) Supported absorption of CO2 by tetrabutylphosphonium amino acid ionic liquids. Chem Eur J 12(15):4021–4026 19. Zhang Y, Zhang S, Lu X, Zhou Q, Fan W, Zhang X (2009) Dual amino-functionalised phosphonium ionic liquids for CO2 capture. Chem Eur J 15(12):3003–3011 20. Xue Z, Zhang Z, Han J, Chen Y, Mu T (2011) Carbon dioxide capture by a dual amino ionic liquid with amino-functionalized imidazolium cation and taurine anion. Int J Greenhouse Gas Control 5(4):628–633 21. Liu X, Zhou G, Zhang S, Yao X (2009) Molecular dynamics simulation of dual amino-functionalized imidazolium-based ionic liquids. Fluid Phase Equilib 284(1):44–49 22. Wang C, Luo H, D-e Jiang, Li H, Dai S (2010) Carbon dioxide capture by superbase-derived protic ionic liquids. Angewandte Chemie-International Edition 49(34):5978–5981 23. Zhang J, Jia C, Dong H, Wang J, Zhang X, Zhang S (2013) A novel dual amino-functionalized cation-tethered ionic liquid for CO2 capture. Ind Eng Chem Res 52 (17):5835–5841 24. Gurkan B, Goodrich BF, Mindrup EM, Ficke LE, Massel M, Seo S, Senftle TP, Wu H, Glaser MF, Shah JK, Maginn EJ, Brennecke JF, Schneider WF (2010) Molecular design of high capacity, low viscosity, chemically tunable ionic liquids for CO2 capture. J Phys Chem Lett 1(24):3494–3499 25. Carlisle TK, Bara JE, Gabriel CJ, Noble RD, Gin DL (2008) Interpretation of CO2 solubility and selectivity in nitrile-functionalized room-temperature ionic liquids using a group contribution approach. Ind Eng Chem Res 47(18):7005–7012 26. Blanchard LA, Gu ZY, Brennecke JF (2001) High-pressure phase behavior of ionic liquid/CO2 systems. J Phys Chem B 105(12):2437–2444 27. Cadena C, Anthony JL, Shah JK, Morrow TI, Brennecke JF, Maginn EJ (2004) Why is CO2 so soluble in imidazolium-based ionic liquids? J Am Chem Soc 126(16):5300–5308 28. Huang XH, Margulis CJ, Li YH, Berne BJ (2005) Why is the partial molar volume of CO2 so small when dissolved in a room temperature ionic liquid? Structure and dynamics of CO2 dissolved in bmim+pf6−. J Am Chem Soc 127(50):17842–17851 29. Almantariotis D, Gefflaut T, Padua AAH, Coxam JY, Gomes MFC (2010) Effect of fluorination and size of the alkyl side-chain on the solubility of carbon dioxide in 1-alkyl-3-methylimidazolium bis(trifluoromethylsulfonyl)amide ionic liquids. J Phys Chem B 114(10):3608–3617 30. Aki S, Mellein BR, Saurer EM, Brennecke JF (2004) High-pressure phase behavior of carbon dioxide with imidazolium-based ionic liquids. J Phys Chem B 108(52):20355–20365 31. Yunus NM, Mutalib MIA, Man Z, Bustam MA, Murugesan T (2012) Solubility of CO2 in pyridinium based ionic liquids. Chem Eng J 189:94–100 32. Bara JE, Gabriel CJ, Lessmann S, Carlisle TK, Finotello A, Gin DL, Noble RD (2007) Enhanced CO2 separation selectivity in oligo(ethylene glycol) functionalized room-temperature ionic liquids. Ind Eng Chem Res 46(16):5380–5386 33. Makino T, Kanakubo M, Umecky T (2014) CO2 solubilities in ammonium bis (trifluoromethanesulfonyl)amide ionic liquids: Effects of ester and ether groups. J Chem Eng Data 59(5):1435–1440 34. Anthony JL, Anderson JL, Maginn EJ, Brennecke JF (2005) Anion effects on gas solubility in ionic liquids. J Phys Chem B 109(13):6366–6374 35. Muldoon MJ, Aki SNVK, Anderson JL, Dixon JK, Brennecke JF (2007) Improving carbon dioxide solubility in ionic liquids. J Phys Chem B 111(30):9001–9009 36. Zhang X, Liu Z, Wang W (2008) Screening of ionic liquids to capture CO2 by COSMO-RS and experiments. AIChE J 54(10):2717–2728

174

X. Zhang et al.

37. Bhargava BL, Balasubramanian S (2007) Probing anion-carbon dioxide interactions in room temperature ionic liquids: gas phase cluster calculations. Chem Phys Lett 444(4–6):242–246 38. Seki T, Grunwaldt J-D, Baiker A (2009) In situ attenuated total reflection infrared spectroscopy of imidazolium-based room-temperature ionic liquids under “supercritical” CO2. J Phys Chem B 113(1):114–122 39. Bates ED, Mayton RD, Ntai I, Davis JH (2002) CO2 capture by a task-specific ionic liquid. J Am Chem Soc 124(6):926–927 40. Gurkan BE, de la Fuente JC, Mindrup EM, Ficke LE, Goodrich BF, Price EA, Schneider WF, Brennecke JF (2010) Equimolar CO2 absorption by anion-functionalized ionic liquids. J Am Chem Soc 132(7):2116–2117 41. Zhang SJ, Yuan XL, Chen YH, Zhang XP (2005) Solubilities of CO2 in 1-butyl-3-methylimidazolium hexafluorophosphate and 1,1,3,3-tetramethylguanidium lactate at elevated pressures. J Chem Eng Data 50(5):1582–1585 42. Yu G, Zhang S, Zhou G, Liu X, Chen X (2007) Structure, interaction and property of amino-functionalized imidazolium ILs by molecular dynamics simulation and ab initio calculation. AIChE J 53(12):3210–3221 43. Zhang SJ, Chen YH, Li FW, Lu XM, Dai WB, Mori R (2006) Fixation and conversion of CO2 using ionic liquids. Catal Today 115(1–4):61–69 44. Wang C, Mahurin SM, Luo H, Baker GA, Li H, Dai S (2010) Reversible and robust CO2 capture by equimolar task-specific ionic liquid-superbase mixtures. Green Chem 12 (5):870–874 45. Wang C, Luo X, Luo H, D-e Jiang, Li H, Dai S (2011) Tuning the basicity of ionic liquids for equimolar CO2 capture. Angewandte Chemie-Intl Ed 50(21):4918–4922 46. Wang C, Luo H, Li H, Zhu X, Yu B, Dai S (2012) Tuning the physicochemical properties of diverse phenolic ionic liquids for equimolar CO2 capture by the substituent on the anion. Chem Eur J 18(7):2153–2160 47. Luo X, Guo Y, Ding F, Zhao H, Cui G, Li H, Wang C (2014) Significant improvements in CO2 capture by pyridine-containing anion-functionalized ionic liquids through multiple-site cooperative interactions. Angewandte Chemie-Intl Ed 53(27):7053–7057 48. Seddon KR, Stark A, Torres MJ (2000) Influence of chloride, water, and organic solvents on the physical properties of ionic liquids. Pure Appl Chem 72(12):2275–2287 49. Bermejo MD, Montero M, Saez E, Florusse LJ, Kotlewska AJ, Cocero MJ, van Rantwijk F, Peters CJ (2008) Liquid-vapor equilibrium of the systems butylmethylimidazolium nitrate-CO2 and hydroxypropylmethylimidazolium nitrate-CO2 at high pressure: Influence of water on the phase behavior. J Phys Chem B 112(43):13532–13541 50. Ventura SPM, Pauly J, Daridon JL, da Silva JAL, Marrucho IM, Dias AMA, Coutinho JAP (2008) High pressure solubility data of carbon dioxide in (tri-iso-butyl(methyl)phosphonium tosylate plus water) systems. J Chem Thermodyn 40(8):1187–1192 51. Wang GN, Hou WL, Xiao F, Geng JA, Wu YT, Zhang ZB (2011) Low-viscosity triethylbutylammonium acetate as a task-specific ionic liquid for reversible CO2 absorption. J Chem Eng Data 56(4):1125–1133 52. Seo S, Quiroz-Guzman M, DeSilva MA, Lee TB, Huang Y, Goodrich BF, Schneider WF, Brennecke JF (2014) Chemically tunable ionic liquids with aprotic heterocyclic anion (AHA) for CO2 capture. J Phys Chem B 118(21):5740–5751 53. Camper D, Bara JE, Gin DL, Noble RD (2008) Room-temperature ionic liquid-amine solutions: tunable solvents for efficient and reversible capture of CO2. Ind Eng Chem Res 47 (21):8496–8498 54. Zhang F, Fang CG, Wu YT, Wang YT, Li AM, Zhang ZB (2010) Absorption of CO2 in the aqueous solutions of functionalized ionic liquids and mdea. Chem Eng J 160(2):691–697 55. Huang Q, Li Y, Jin XB, Zhao D, Chen GZ (2011) Chloride ion enhanced thermal stability of carbon dioxide captured by monoethanolamine in hydroxyl imidazolium based ionic liquids. Energy Environ Sci 4(6):2125–2133


175

56. Xu F, Gao HS, Dong HF, Wang ZL, Zhang XP, Ren BZ, Zhang SJ (2014) Solubility of CO2 in aqueous mixtures of monoethanolamine and dicyanamide-based ionic liquids. Fluid Phase Equilib 365:80–87 57. Li XY, Hou MQ, Zhang ZF, Han BX, Yang GY, Wang XL, Zou LZ (2008) Absorption of CO2 by ionic liquid/polyethylene glycol mixture and the thermodynamic parameters. Green Chem 10(8):879–884 58. Shiflett MB, Yokozeki A (2009) Phase behavior of carbon dioxide in ionic liquids: [emim] [acetate], [emim][trifluoroacetate], and [emim][acetate] plus [emim][trifluoroacetate] mixtures. J Chem Eng Data 54(1):108–114 59. Lei ZG, Han JL, Zhang BF, Li QS, Zhu JQ, Chen BH (2012) Solubility of CO2 in binary mixtures of room-temperature ionic liquids at high pressures. J Chem Eng Data 57 (8):2153–2159 60. Pinto AM, Rodriguez H, Arce A, Soto A (2014) Combined physical and chemical absorption of carbon dioxide in a mixture of ionic liquids. J Chem Thermodyn 77:197–205 61. Goodrich BF, de la Fuente JC, Gurkan BE, Lopez ZK, Price EA, Huang Y, Brennecke JF (2011) Effect of water and temperature on absorption of CO2 by amine-functionalized anion-tethered ionic liquids. J Phys Chem B 115(29):9140–9150 62. Romanos GE, Zubeir LF, Likodimos V, Falaras P, Kroon MC, Iiev B, Adamova G, Schubert TJS (2013) Enhanced CO2 capture in binary mixtures of 1-alkyl-3-methylimidazolium tricyanomethanide ionic liquids with water. J Phys Chem B 117(40):12234–12251 63. Anderson K, Atkins MP, Estager J, Kuah Y, Ng S, Oliferenko AA, Plechkova NV, Puga AV, Seddon KR, Wassell DF (2015) Carbon dioxide uptake from natural gas by binary ionic liquid-water mixtures. Green Chem 17(8):4340–4354 64. Ahmady A, Hashim MA, Aroua MK (2010) Experimental investigation on the solubility and initial rate of absorption of CO2 in aqueous mixtures of methyldiethanolamine with the ionic liquid 1-butyl-3-methylimidazolium tetrafluoroborate. J Chem Eng Data 55(12):5733–5738 65. Ahmady A, Hashim MA, Aroua MK (2011) Absorption of carbon dioxide in the aqueous mixtures of methyldiethanolamine with three types of imidazolium-based ionic liquids. Fluid Phase Equilib 309(1):76–82 66. Gao Y, Zhang F, Huang K, Ma JW, Wu YT, Zhang ZB (2013) Absorption of CO2 in amino acid ionic liquid (AAIL) activated MDEA solutions. Int J Greenhouse Gas Control 19:379– 386 67. Zhang F, Gao Y, Wu XK, Ma JW, Wu YT, Zhang ZB (2013) Regeneration performance of amino acid ionic liquid (AAIL) activated mdea solutions for CO2 capture. Chem Eng J 223:371–378 68. Sairi NA, Yusoff R, Alias Y, Aroua MK (2011) Solubilities of CO2 in aqueous N-methyldiethanolamine and guanidinium trifluoromethanesulfonate ionic liquid systems at elevated pressures. Fluid Phase Equilib 300(1–2):89–94 69. Yang J, Yu XH, Yan JY, Tu ST (2014) CO2 capture using amine solution mixed with ionic liquid. Ind Eng Chem Res 53(7):2790–2799 70. Gao JB, Cao LD, Dong HF, Zhang XP, Zhang SJ (2015) Ionic liquids tailored amine aqueous solution for pre-combustion CO2 capture: role of imidazolium-based ionic liquids. Appl Energy 154:771–780 71. Hong G, Jacquemin J, Husson P, Gomes MFC, Deetlefs M, Nieuwenhuyzen M, Sheppard O, Hardacre C (2006) Effect of acetonitrile on the solubility of carbon dioxide in 1-ethyl-3-methylimidazolium bis(trifluoromethylsulfonyl) amide. Ind Eng Chem Res 45 (24):8180–8188 72. Ahn JY, Lee BC, Lim JS, Yoo KP, Kang JW (2010) High-pressure phase behavior of binary and ternary mixtures containing ionic liquid [C6mim][Tf2N], dimethyl carbonate and carbon dioxide. Fluid Phase Equilib 290(1–2):75–79 73. Wang CM, Luo HM, Luo XY, Li HR, Dai S (2010) Equimolar CO2 capture by imidazolium-based ionic liquids and superbase systems. Green Chem 12(11):2019–2023

176

X. Zhang et al.

74. Zhao YS, Zhang XP, Zeng SJ, Zhou Q, Dong HF, Tian XA, Zhang SJ (2010) Density, viscosity, and performances of carbon dioxide capture in 16 absorbents of amine plus ionic liquid + H2O, ionic liquid + H2O, and amine + H2O systems. J Chem Eng Data 55 (9):3513–3519 75. Dai CN, Wei WJ, Lei ZG, Li CX, Chen BH (2015) Absorption of CO2 with methanol and ionic liquid mixture at low temperatures. Fluid Phase Equilib 391:9–17 76. Lei ZG, Qi XX, Zhu JQ, Li QS, Chen BH (2012) Solubility of CO2 in acetone, 1-butyl-3-methylimidazolium tetrafluoroborate, and their mixtures. J Chem Eng Data 57 (12):3458–3466 77. Wang M, Zhang LQ, Gao LX, Pi KW, Zhang JY, Zheng CG (2013) Improvement of the CO2 absorption performance using ionic liquid [NH2emim][BF4] and [emim][BF4]/[bmim][BF4] mixtures. Energy Fuels 27(1):461–466 78. Kumar S, Cho JH, Moon I (2014) Ionic liquid-amine blends and CO2BOLs: Prospective solvents for natural gas sweetening and CO2 capture technology—a review. Int J Greenhouse Gas Control 20:87–116 79. Basha OM, Keller MJ, Luebke DR, Resnik KP, Morsi BI (2013) Development of a conceptual process for selective CO2 capture from fuel gas streams using [hmim][Tf2N] ionic liquid as a physical solvent. Energy Fuels 27(7):3905–3917 80. Eisinger RS, Keller GE II (2014) Process for CO2 capture using ionic liquid that exhibits phase change. Energy Fuels 28(11):7070–7078

Amine-Blends Screening and Characterization for CO2 Post-combustion Capture Abdullah Al Hinai, Nabil El Hadri and Mohammad Abu Zahra

Abstract Amines have been identified as one of the most promising agents to capture CO2 released from industrial and energy sources to the atmosphere. The primary aim of this chapter is to present experimental results showing the two major factors considered in the selection of suitable amine solvents for carbon capture in post-combustion technology. Ten amines were selected from varying classes and structure types, and were investigated on the basis of CO2 loading and heat of absorption. These solvents were selected based on the preliminary experimental results and the effect of blending the best reacting solvents was further investigated. Details of the experimental procedure and results are discussed as well as the relevant conclusions.

1 Introduction CCS (carbon capture and storage) post-combustion technology can be categorized as one of the most suitable technology process to decrease the CO2 released to the atmosphere because of the ability to install this technology directly into existing power plants [1] and this technology can be applied for CO2 separation from highly diluted stream of flue gas using the chemical absorption technique. The three basic classes of alkanolamines: Monoethanolamine (MEA)—primary amine, diethanolamine (DEA)—secondary amine, and methyl diethanolamine (MDEA)—tertiary amine, are the most widely used [2]. The primary and secondary alkanolamines have a high rate reaction with CO2 when first there is a formation of a zwitterion (reaction 1), which then transfers a proton to an amine resulting in the formation of a carbamate ion (reaction 2). Their maximum CO2 loading is 0.5 mol of CO2/mole of amine (two molecules of amines is needed to react with one molecule of CO2). At a high CO2 pressure, the carbamate could be hydrolyzed to A. Al Hinai N. El Hadri M. Abu Zahra (&) Department of Chemical and Environmental Engineering, Masdar Institute of Science and Technology, P.O. Box 54224, Masdar City, United Arab Emirates e-mail: [email protected] © Springer International Publishing AG 2017 W.M. Budzianowski (ed.), Energy Efficient Solvents for CO2 Capture by Gas–Liquid Absorption, Green Energy and Technology, DOI 10.1007/978-3-319-47262-1_8

177

178

A. Al Hinai et al.

form a free amine and bicarbonates and then the free amine will again react with CO2 (reaction 3). R1 R2 NH þ CO2 ðaqÞ $ R1 R2 NH þ COO

ð1Þ

R1 R2 NH þ R1 R2 NH þ COO $ R1 R2 NH þ þ R1 R2 NCOO

ð2Þ

R1 R2 NCOO þ H2 O $ R1 R2 NH þ HCO 3

ð3Þ

The maximum CO2 loading which can be obtained from tertiary or hindered amines is 1 mol of CO2 per mole of amine (one molecule of amine will react with one molecule of CO2) with the formation of bicarbonates (reaction 4). However, despite the higher achievable CO2 loading, tertiary amines have low reactivity with CO2 in comparison with the primary or secondary amines. R1 R2 R3 N þ CO2 ðaqÞ þ H2 O $ R1 R2 R3 NH þ þ HCO 3

ð4Þ

A typical conventional CO2 post-combustion capture process based on aqueous MEA solution (30 wt%) has been used as a reference case in many studies. MEA has been chosen because of its high reaction rate with CO2 and reasonable CO2 absorption capacity and low cost [3]. However, the regeneration energy to regenerate the MEA solution after CO2 absorption is high and this hinders industrial application of the conventional post-combustion process [4, 5]. Recently, great efforts has been made to find alternative amines with high CO2 absorption capacity, high absorption rate and low heat of absorption in comparison with MEA. The main challenge is to find new amine solvents performance in order to reduce the cost of CO2 capture process. Several studies have been carried out in order to characterize the CO2-reaction enthalpy of aqueous amines solution [6–8]. Understanding this property is essential in order to identify the quantity of heat which is needed in the regeneration column on the process. The heat of absorption is associated with the quantity of steam necessary to regenerate the aqueous amine solution. The heat of absorption is very high for primary and secondary amines and increase the regeneration energy of the solvent. In opposite, the heat of absorption of the tertiary amines is low and the energy required to regenerate the solvent is reduced [9]. The determination of the heat of absorption could be done using the Gibbs-Helmholtz equation from the CO2 solubility data or by direct calorimetric measurements. The Gibbs-Helmholtz equation was used by Kim et al. for the calculation of the heat of absorption of MEA and MDEA from the equilibrium constant of each reaction occurred between the reactions of CO2 with the respective aqueous amine solution [10]. The direct calorimetric measurements have the ability to give accurate data for the heat of absorption because it incorporates the heat due to physical dissolution of CO2 in the aqueous amine solution and the chemical reaction between CO2 and aqueous amine solution. Different equipment are published in various works. Mathonat et al.

Amine-Blends Screening and Characterization for CO2 …

179

measured the CO2-reaction enthalpy for aqueous solutions of MEA and MDEA 30 wt% at 40, 80 and 120 °C using a flow calorimeter [11]. The objective of this work is to investigate the improvement of thermodynamic properties such as absorption capacity and heat of absorption by blending amines.

2 Research Methodology and Experiments 2.1

CO2 Loading in Amines

Solubility of CO2 by aqueous amine solutions have been looked into in several studies in order to find promising molecules with a high CO2 loading [12–17]. Bonenfant et al. studied the effect of different amine structures (amine, diamines and polyamines) on the absorption and desorption of CO2 at a concentration of amine of 5 wt%. The work was based on eight amines and the results show that N-(2-aminoethyl)-1,3-propanediamine (AEPDNH2, polyamines with two primary and one secondary amine), 2-(2-aminoethylamino)ethanol (AEE, diamine with one primary and one secondary amine) have the promising characteristics for CO2 removal. Singh et al. also studied different amines (linear, cyclic, polyamines etc.) in order to establish a relation between the structure of the amine and CO2 loading and absorption rate [18–20]. The results show the potential of the structure variation and the authors found 1,6-hexanediamine, N,N-dimethyl to have a good CO2 solubility in comparison with all amines studied. The solvent screening setup (S.S.S) which is one of the most available techniques to obtain aqueous amine solution loaded with CO2 is used for the experiments. A simple solvent screening setup (S.S.S.) equipment constitutes of six glass reactors (V = 250 mL ± 0.5) which can be operated independently in the temperature range from [298.15–423.15 K (±1 K)] and the pressure range of [0–6 bars (±0.01 bar)] (Fig. 1). A magnetic stirrer [speed max = 1500 rpm (±1 rpm)] is used to ensure a homogeneous contact between the solution and CO2 by creation of a vortex. An aqueous amine solution 30 wt% (150 g) was prepared and introduced into the glass reactors. The tests were carried out at 313.15 K with a mixing speed of 500 rpm. To simulate the gas flow, a blend of 15 vol.% CO2 and 85 vol.% N2 is initially fed to a make-up vessel to reach a pressure of 2 bar and allowed into the reactors at 15 L/h with the aid of a mass flow controller. The pressure inside each reactor was kept at 1 bar during the all CO2 absorption experiment. The reaction of CO2 with the aqueous amine solution is considered complete when equilibrium is reached i.e. when the flow of CO2 into the set-up is equal to the flow of CO2 out of the set-up. The experiments were performed at a partial pressure PCO2 ¼ 15 kPa and temperature T = 40 °C which are representative conditions for the absorber in the post-combustion capture process. The screening results obtained were further studied as a basis for initial selection of the aqueous amine solution performance for CO2 absorption and to understand the impact of amine structure on CO2 loading.

180

A. Al Hinai et al.

Fig. 1 Schematic of the solvents screening setup (S.S.S.)

2.2

Heat of Absorption

The heat required in order to regenerate the aqueous amine solution in the stripper column could be approximated as the contribution of three types of energy: the heat of desorption of CO2 from the solution, the sensible heat to elevate the temperature in the stripper and the latent heat contained in the water vapor leaving the stripper with the CO2-product [5]. The heat of absorption is associated to the quantity of steam necessary to regenerate the aqueous amine solution. It is assumed that desorption of CO2 from the amine solution in the stripper is the inverse reaction of absorption of CO2 by the aqueous amine solution that happens in the absorber. In other words, the heat of absorption CO2 is considered nearly equal to the heat of desorption of CO2. Moreover, study on the heat of CO2 absorption by an aqueous amine solution shows that its absorption heat depends strongly on the CO2 absorption capacity, the structure of amine and temperature but not significantly depend on pressure and amine concentration [21]. To evaluate the heat of absorption of the aqueous amine solution, a flow micro-calorimeter supplied by Thermal Hazard Technology (UK) controlled by a URC control software was used (Fig. 2). The equipment can be operated from 298.15 to 353.15 K (±1 K) and flow gas can be adjusted from 0 to 20 ml/min (±0.1 ml/min). The micro reaction calorimeter is used to determine the heat of absorption of all the amine samples under study. This type of calorimeter is selected for this study due to its numerous advantages over the other methods such as reliability, energy efficiency, reaction yield and safety. For each experiment, the cell containing


181

Fig. 2 Schematic of flow Micro Reaction Calorimeter (URC)

sample was placed into the calorimeter at 313.15 K and atmospheric pressure. At the initial stage of the test, there is a variation of power (mW) with time (s) which was recorded by the software. Next, CO2 gas flows through a desiccant column in order to remove moisture before entering into the sample cell at a rate of 0.5 ml/min (±0.01 ml/min). Due to the exothermic nature of the reaction between CO2 and the aqueous amine solution, the power signal initially increases and then decreases before it finally becomes constant when the reaction reaches equilibrium. The difference between the (cell + sample) weight is calculated before and after the CO2 absorption to obtain the mass of CO2 absorbed during the test. The CO2 loading is determined by using the mass of CO2 and the mass of aqueous amine solution as in Eq. (5). CO2 loading ¼

ðm1 m0 Þ=44 C1 C2 m0 ðM þM Þ 1 2

ð5Þ

Then, with the software URC, the integral heat Q (in kJ) was determined. Finally, the heat of absorption DH (−kJ/mole of CO2) is calculated by the division of the integral heat Q by the mole of CO2 absorbed. The experiments are performed at PCO2 ¼ 15 kPa and T = 40 °C which are the same conditions as in the absorber in the post-combustion capture process. For this study, the heat of absorption limit is taken to be below 70 kJ/mole.

3 Results This section presents results of both single and blended amines. Ten structures of stand-alone amines which represents various classes, structure, configuration and chemical group of amines were investigated. Among this 10 structures, two were conventional amines: N-methyldiethanolamine (MDEA) and diethanolamine (DEA). The remaining compounds were selected to investigate the impact of the amine structure on the CO2 loading. 2MAE is obtained from the substitution of one

182

A. Al Hinai et al.

Table 1 Stand-alone amine structures studied Name

Structure

CAS

1

N-methyldiethanolamine (MDEA)

105-59-9

2 3

Diethanolamine (DEA) 2-(methylamino)ethanol (2MAE)

4

2-(ethylamino)ethanol (2EAE)

110-73-6

5

2-(butylamino)ethanol (2BAE)

111-75-1

6

2-(dimethylamino)ethanol (2DMAE)

108-01-0

7

1-dimethylamino-2-propanol (1DMA2P)

108-16-7

8

N,N-diethylethanolamine (DEEA)

100-37-8

9

3-dimethylamino-1-propanol (3DMA1P)

3179-63-3

10

N,N,N′,N′-Tetramethyl-1,3-propanediamine (TMDAP)

110-95-2

HO

NH

OH

111-42-2 109-83-1

hydrogen by methyl or ethyl group in MEA (primary amine) structure. From the 2MAE, the methyl group is changed to ethyl for 2EAE and butyl for 2BAE. The tertiary amine, 2DMAE, is also obtained from MEA to illustrate the effect of replacing the 2 methyl groups with 2 ethyl groups (DEEA). The effect of the increase of the carbon length from 2DMAE to 3DMA1P (2 carbons to 3 carbons) is also considered by using polyamines such as TMDAP. These structures are listed in Table 1. Based on the results of the stand-alone amines, eight amine blends were further studied under three different concentration weight percentage ratio. The blends included conventional amines to represent various classes and configuration as well as chemical groups. The blends and the different concentration weight percent studied are shown in Table 2.

3.1

Stand Alone Amines

3.1.1

CO2 Loading Results

The values obtained for the CO2 loading of the aqueous amine solution at 30 wt% at 40 °C are listed in Table 3. The results indicate that TMDAP with two amino groups in its structure has the highest CO2 loading as 1.16 mol CO2/mole amine. Among all amines (with one amino group), DEEA and 3DMA1P have the highest CO2 absorption capacity ( 0.80 mol CO2/mole amine). MDEA and DEA have the


183

Table 2 Blend amine structure studied at 30 wt% total of aqueous amines Blends with 2EAE

Blends with 2MAE

Concentration (wt%)

2EAE 2EAE 2EAE 2EAE 2EAE 2EAE 2EAE 2EAE 2EAE 2EAE 2EAE 2EAE 2EAE 2EAE 2EAE

2MAE + MDEA 2MAE + 1DMA2P 2MAE + 2DMAE

5 + 25 5 + 25 5 + 25 5 + 25 5 + 25 10 + 20 10 + 20 10 + 20 10 + 20 10 + 20 15 + 15 15 + 15 15 + 15 15 + 15 15 + 15

+ + + + + + + + + + + + + + +

MDEA 1DMA2P 2DMAE 3DMA1P TMPDA MDEA 1DMA2P 2DMAE 3DMA1P TMPDA MDEA 1DMA2P 2DMAE 3DMA1P TMPDA

2MAE 2MAE 2MAE 2MAE

+ + + +

TMPDA MDEA 1DMA2P 2DMAE

2MAE 2MAE 2MAE 2MAE

+ + + +

TMPDA MDEA 1DMA2P 2DMAE

2MAE + TMPDA

Table 3 Absorption capacity of CO2 in aqueous amine solution 30 wt% at 40 °C and 15 kPa CO2

1 2 3 4 5 6 7 8 9 10

Amine solution

Concentration wt%

Absorption capacity mol CO2/mol amine

N-methyldiethanolamine (MDEA) Diethanolamine (DEA) 2-(methylamino)ethanol (2MAE) 2-(ethylamino)ethanol (2EAE) 2-(butylamino)ethanol (2BAE) 2-(dimethylamino)ethanol (2DMAE) 1-dimethylamino-2-propanol (1DMA2P) N,N-diethylethanolamine (DEEA) 3-dimethylamino-1-propanol (3DMA1P) Tetramethyl-1,3-diaminopropane (TMDAP)

30 30 30 30 30 30 30 30 30 30

0.52 0.53 0.56 0.67 0.69 0.73 0.72 0.90 0.89 1.16

lowest CO2 absorption among all tested amines with values of 0.52 and 0.53 mol CO2/mole amine, respectively. The screening of the CO2 absorption capacity results acquired with the S.S.S. equipment shows that amines which contain an alkyl group (methyl, ethyl or t-butyl) attached to the nitrogen or near it have a high CO2 loading. It is also observed that alcohol group near the amino group have a negative effect on the CO2 absorption. Additionally, the polyamines studied in this work showed the highest CO2 loading among all amines studied. These results are in accordance with work from Singh et al. where the impact of the alkyl and number of amino groups in the amine molecule enhance the absorption capacity of the amine based solvents [15, 16, 22].

184

3.1.2

A. Al Hinai et al.

Heat of Absorption

The heat of absorption was measured at 313.15 K and atmospheric pressure and the results are presented in Table 4. The experiments for all aqueous amines solutions were conducted at least 3 times to ensure the consistency of values obtained. Using the heat of absorption limit as 70 kJ/mole of CO2, there are four amines with heat of absorption which falls above 70 kJ/mole and they include DEA, 2MAE, 2BAE and DEEA while the other amines are below 70 kJ/mole of CO2, From these results, we assume that the potential amines for CO2 capture application will have a CO2 loading up to 0.60 and a heat of absorption below 70 kJ/mole of CO2. The amines considered are MDEA, 1DMA2P, 3DMA1P, TMDPA, 2DMAE, 2MAE, 2BAE 2EAE. In order to evaluate the operational validity of the calorimeter equipment used in this work, the heat of CO2 absorption of MEA, MDEA, DEA and AMP 30 wt% at 313.15 K and atmospheric pressure was measured and compared with the available literature data [11, 23, 24]. A value of −52.51 kJ/mole of CO2 (aCO2 = 0.74) for MDEA was determined, and 85.13 kJ/mole of CO2 (aCO2 = 0.59) for MEA. The results for DEA and AMP also has been found to be −74.24 kJ/mole of CO2 (aCO2 = 0.61) and −80.91 kJ/mole of CO2 (aCO2 = 0.78) respectively [25]. The values show a good agreement with the literature with 2% uncertainty, which implies that the calorimeter have been used has a capability and consistency to be worked on.

Table 4 Heat of absorption of CO2 in aqueous amine solution 30 wt% at 40 °C

1 2 3 4 5 6 7 8 9 10

Amine solution

Concentration wt%

CO2 loading

Heat of absorption DH (kJ/mol of CO2)

N-methyldiethanolamine (MDEA) Diethanolamine (DEA) 2-(methylamino)ethanol (2MAE) 2-(ethylamino)ethanol (2EAE) 2-(butylamino)ethanol (2BAE) 2-(dimethylamino)ethanol (2DMAE) 1-dimethylamino-2-propanol (1DMA2P) N,N-diethylethanolamine (DEEA) 3-dimethylamino-1-propanol (3DMA1P) Tetramethyl-1,3-diaminopropane (TMDAP)

30 30 30 30 30 30

0.74 0.61 0.67 0.71 0.73 0.77

−52.5 −74.2 −73.8 −69.0 −74.4 −63.3

30

0.83

−60.7

30 30

0.83 0.85

−73.2 −54.6

30

1.32

−59.9


3.2

185

Amine Blends

Tertiary amines have favourable thermodynamics characteristics. However, it has drawback due to its low reactivity compared to MEA and other conventional amines. This problem can be overcome by blending them with highly reactive amines that have good thermodynamics characteristics such as high absorption capacity and low heat of absorption. Potential candidates for this are 2EAE, 2MAE and 2BAE. Two compounds 2EAE and 2MAE were selected as the basis for the amine blends system. For the purpose of clarification in this chapter, system A will consist of 2EAE blended with MDEA, 1DMA2P, 2DMA2P, 3DMA2P and TMPAD while system B will consist of 2 MAE blended with MDEA, 1DMA2P, 2DMA2P and TMPAD. In both systems, the blends concentrations are 5, 10 and 15 wt% of 2EAE or 2MAE.

3.2.1

CO2 Loading Result

For system A, the results shown in Table 5 indicate that the lowest CO2 loading is 0.71 (2EAE 5 wt%/MDEA 25 wt%) while among all the three different

Table 5 Heat of absorption of CO2 in aqueous 2EAE blends amine solution 30 wt% at 40 °C Blends of System A 2EAE 5 wt% 10 wt% 15 wt% 2EAE 5 wt% 10 wt% 15 wt% 2EAE 5 wt% 10 wt% 15 wt% 2EAE 5 wt% 10 wt% 15 wt% 2EAE 5 wt% 10 wt% 15 wt%

MDEA 25 wt% 20 wt% 15 wt% 1DMA2P 25 wt% 20 wt% 15 wt% 2DMA2P 25 wt% 20 wt% 15 wt% 3DMA2P 25 wt% 20 wt% 15 wt% TMDAP 25 wt% 20 wt% 15 wt%

CO2 loading (mol CO2/mol of amine)

Heat of absorption (kJ/mol of CO2)

0.71 0.74 0.75

−59.5 −62.8 −71.5

0.82 0.84 0.76

−63.9 −71.4 −75.2

0.83 0.8 0.78

−61.5 −67.8 −71.4

0.83 0.81 0.72

−63.9 −70.6 −75.6

1.25 1.14 1.02

−62.3 −67.8 −69.2

186

A. Al Hinai et al.

concentrations, the 2EAE + TMDPA blend has the highest CO2 loading. 2EAE 5 wt%/TMDPA 25 wt 2EAE 10 wt%/TMDPA 20 wt%—2EAE 15 wt%/TMDPA 15 wt% having 1.25, 1.14 and 1.02 respectively. Comparison was made between the blend containing 2DMAE (tertiary amine where the number of carbon between the nitrogen and the alcohol is two) with that containing 3DMA1P (tertiary amine where the number of carbon between the nitrogen and the alcohol is three). The increase in the carbon length does not have significant influence on the absorption of CO2 at the different concentrations except for the 15% weight concentration of 2EAE where the 2DMAE and 3DMA1P blends have a CO2 loading which are 0.78 and 0.72, respectively. Except for the blends 2EAE/TMPDA, all other blends have a CO2 absorption calculated from calorimeter equipment between 0.7 and 0.9. 2EAE/TMPDA blends have the highest CO2 loading because it contains a polyamine with two nitrogen atoms in its structure. However, it can be seen from the results in Table 5 that the CO2 loading of 2EAE/TMPDA reduces as the concentration weight percentage of 2EAE increases in the blend. For system B, the result from blending 2MAE with 2DMAE, 3DMA1P, 1DMA2P, MDEA and TMPAD at 40 °C are presented in Table 6. The results show that the lowest loading obtained is 0.54 from blending (2MAE 5 wt%/MDEA 25 wt%) while the highest loading is obtained from the blend of 2MAE with TMPAD as 2MAE 5 wt%/TMPAD 25 wt% (aCO2 = 1.35), 2MAE 10 wt%/ TMPAD 20 wt% (aCO2 = 1.22), and for 2MAE 15 wt%/TMPAD 15 wt% (aCO2 = 1.04).

Table 6 Heat of absorption of CO2 in aqueous 2MAE blends amine solution 30 wt% at 40 °C Blends of System B 2MAE 5 wt% 10 wt% 15 wt% 2MAE 5 wt% 10 wt% 15 wt% 2MAE 5 wt% 10 wt% 15 wt% 2MAE 5 wt% 10 wt% 15 wt%

MDEA 25 wt% 20 wt% 15 wt% 1DMA2P 25 wt% 20 wt% 15 wt% 2DMA2P 25 wt% 20 wt% 15 wt% TMDAP 25 wt% 20 wt% 15 wt%

CO2 loading (mol CO2/mol of amine)

Heat of absorption (kJ/mol of CO2)

0.54 0.56 0.57

−58.6 −61.0 −64.6

0.68 0.65 0.63

−62.4 −65.3 −69.5

0.69 0.65 0.63

−65.7 −67.3 −70.2

1.35 1.22 1.04

−60.4 −64.5 −66.5


187

Comparing 1DMA2P with 2DMAE which are MEA derivatives where one hydrogen has been substituted with methyl group for 1DMA2P and one hydrogen has been substituted with ethyl group for 2DMAE, the results show that there are no significate difference in terms of loadings. The loading for 2MAE 5 wt% + 1DMA2P 25 wt% and 2MAE 5 wt% + 2DMA2P were identical (aCO2 = 0.56). Also for 2MAE 10 wt% + 1DMA2P 20 wt% and 2MAE 10 wt% + 2DMA2P 20 wt%, the CO2 loading is 0.63.

3.2.2

Heat of Absorption

Similar to the stand alone amines experiments, values for the heat of absorption for this blends were obtained at 313.15 K and atmospheric pressure. These results are presented in Tables 5 and 6. The results from system A show in general that the blends have a much lower heat of absorption than the stand alone amines with the highest heat of absorption being −75.25 kJ/mole of CO2 for 2EAE 15 wt%/3DMAIP 15 wt% while the lowest is −59.53 kJ/mole of CO2 for 2EAE 5 wt%/MDEA 25 wt%. It is obvious from this study that as the amine concentration of 2EAE increases in the blends 30 wt% total (from 5 to 15 wt%), the heat of absorption increases. For 2EAE 5 wt%, the heat of absorption is between 59.53 and 63.91 kJ/mole of CO2 while for the 2EAE 15 wt%, the heat of absorption is between 69.15–75.25 kJ/mole of CO2 which is the highest value. The blends which have a good heat of absorption with a value below 70 kJ/mole of CO2 are 2EAE 5 wt%/MDEA 25 wt%—2EAE 10 wt%/MDEA 20 wt%—2EAE 5 wt%/1DMA2P 25 wt%—2EAE 5 wt%/3DMA1P 25 wt%—2EAE 10 wt%/ 2DMAE 20 wt%—2EAE 5 wt%/2DMAE 25 wt%—2EAE 5 wt%/TMDPA 25 wt%— 2EAE 10 wt%/TMDPA 20 wt%—2EAE 15 wt%/TMDPA 15 wt%. Our results show similar values of CO2 loading and heat of absorption for the 2EAE-1DMA2P AND 3DMA1P blends. For 2EAE 5, 10 and 15 wt%, the CO2 loading for 1DMA2P/3DMA1P are 0.82/0.83, 0.84/0.81 and 0.76/0.72 respectively while the heat of absorption are 63.91/63.85, 71.39/70.64 and 75.16/75.25, respectively. For the three groups of amine concentration of 2EAE (5 and 10%), the 2EAE/MDEA have the lowest heat of absorption (−59.53 and −62.82 kJ/mole of CO2), respectively. These blends also have the lowest CO2 loading of 0.71, 0.74 and 0.75 for 2EAE 5 wt%/MDEA 25 wt%, 2EAE 10 wt%/MDEA 20 wt% and 2EAE 15 wt%/MDEA 15 wt% respectively. Out of all the studied blends, this blend can be considered for CO2 capture application. This study shows that the blend with 2EAE 5 wt% having the highest CO2 loading of 1.25 has the third-lowest heat of absorption of 62.29 kJ/mole of CO2 and from this, we assume that the most promising amine blend for CO2 capture application will be 2EAE 5 wt%/TMDPA 25 wt%. This blend concentration fulfills the two major concerns in carbon capture which is a high CO2 loading and low heat of absorption.

188

A. Al Hinai et al.

On the other hand, results from system B show in general that blends have a higher heat of absorption than the stand alone amines and system A. The highest heat of absorption is −84.86 kJ/mole of CO2 for 2MAE 15 wt%/MDEA 15 wt% while the lowest is −65.13 kJ/mole of CO2 for 2MAE 5 wt%/1DMA2P 25 wt%. Also, the system B illustrates similar to system A, that for all the blends, a higher concentration of 2MAE from 5 to 15 wt%, will have a higher heat of absorption. The blends which have a good heat of absorption, i.e. with a value below 70 kJ/mole of CO2, are 2MAE 5 wt%/MDEA 25 wt%, 2MAE 5 wt%/1DMA2P 25 wt%, 2MAE 5 wt%/2DMA2P 25 wt%, 2MAE 5 wt%/TMDPA 25 wt%, 2MAE 10 wt%/TMDPA 20 wt%. Studying the five blends with low heat of absorption, results show that three of the five blends have a CO2 loading less than 0.70-2MAE 5 wt%/MDEA 25 wt% (0.54 mol CO2/mol of amine), 2MAE 5 wt%/1DMA2P 25 wt% (0.68 mol CO2/mol of amine) and 2MAE 5 wt%/2DMA2P 25 wt% (0.69 mol CO2/mol of amine). 2MAE 5 wt%/TMDPA 25 wt% and 2MAE 10 wt%/TMDPA 20 wt% have a high CO2 loading of 1.35 and 1.22 mol CO2/mol of amine respectively and thus have the potential to be used in CO2 capture application. In general, for both systems, TMDPA 25 wt% is a good blend combination with 5 wt% of 2EAE or 2MAE.

4 Conclusions Following the results from this study, it can be concluded that CO2 loading is the highest and in the range of 1.04–1.45 by blending either 2EAE or 2MAE with TMDAP. Increasing the concentration of 2EAE or 2MAE in the blend causes increase in the heat of absorption. Blends containing 5 wt% of 2MAE or 2EAE in general shows the lowest heat of absorption but due to the low CO2 loading observed, further investigation is not advised to be carried out on MDEA, 1DMA2P, 2DMA2P and 3DMA2P. It is important to follow up this work by investigating the kinetics of blends of 2MAE or 2EAE with TMDAP.

References 1. Wang M, Lawal A, Stephenson P, Sidders J, Ramshaw C (2011) Post-combustion CO2 capture with chemical absorption: a state-of-the-art review. Chem Eng Res Des 89:1609–1624 2. Samanta A, Zhao A, Shimizu GKH, Sarkar P, Gupta R (2012) Post-combustion CO2 capture using solid sorbents: a review. Ind Eng Chem Res 51:1438–1463 3. Jou F-Y, Mather AE, Otto FD (1995) The solubility of CO2 in a 30 mass percent monoethanolamine solution. Can J Chem Eng 73:140–147 4. Supplement 5. Quang DV, Rabindran AV, El Hadri N, Abu-Zahra MR (2013) Reduction in the regeneration energy of CO2 capture process by impregnating amine solvent onto precipitated silica. Eur Sci J 9


189

6. Rodier L, Ballerat-Busserolles K, Coxam J-Y (2010) Enthalpy of absorption and limit of solubility of CO2 in aqueous solutions of 2-amino-2-hydroxymethyl-1,3-propanediol, 2-[2(dimethyl-amino)ethoxy] ethanol, and 3-dimethyl-amino-1-propanol at T = (313.15 and 353.15) K and pressures up to 2 MPa. J Chem Thermodyn 42:773–780 7. Arshad MW, von Solms N, Thomsen K, Svendsen HF (2013) Heat of absorption of CO2 in aqueous solutions of DEEA, MAPA and their mixture. Energy Procedia 37:1532–1542 8. Arcis H, Ballerat-Busserolles K, Rodier L, Coxam J-Y (2012) Enthalpy of solution of carbon dioxide in aqueous solutions of triethanolamine at temperatures of 322.5 K and 372.9 K and pressures up to 5 MPa. J Chem Eng Data 57:3587–3597 9. Carson JK, Marsh KN, Mather AE (2000) Enthalpy of solution of carbon dioxide in (water monoethanolamine, or diethanolamine, orN-methyldiethanolamine) and (water + monoethanolamine + N-methyldiethanolamine) atT = 298.15 K. J Chem Thermodyn 32: 1285–1296 10. Kim I, Hoff KA, Hessen ET, Haug-Warberg T, Svendsen HF (2009) Enthalpy of absorption of CO2 with alkanolamine solutions predicted from reaction equilibrium constants. Chem Eng Sci 64:2027–2038 11. Mathonat C, Majer V, Mather AE, Grolier JPE (1997) Enthalpies of absorption and solubility of CO2 in aqueous solutions of methyldiethanolamine. Fluid Phase Equilib 140:171–182 12. Chowdhury FA, Okabe H, Yamada H, Onoda M, Fujioka Y (2011) Synthesis and selection of hindered new amine absorbents for CO2 capture. Energy Procedia 4:201–208 13. Chowdhury FA, Yamada H, Higashii T, Goto K, Onoda M (2013) CO2 capture by tertiary amine absorbents: a performance comparison study. Ind Eng Chem Res 52:8323–8331 14. Ma’mun S, Jakobsen JP, Svendsen HF, Juliussen O (2006) Experimental and modeling study of the solubility of carbon dioxide in aqueous 30 mass% 2-((2aminoethyl)amino)ethanol solution. Ind Eng Chem Res 45:2505–2512 15. Singh P, Niederer JPM, Versteeg GF (2009) Structure and activity relationships for amine-based CO2 absorbents-II. Chem Eng Res Des 87:135–144 16. Singh P, Brilman DWF, Groeneveld MJ (2011) Evaluation of CO2 solubility in potential aqueous amine-based solvents at low CO2 partial pressure. Int J Greenhouse Gas Control 5:61–68 17. Singh P, Niederer JPM, Versteeg GF (2007) Structure and activity relationships for amine based CO2 absorbents—I. Int J Greenhouse Gas Control 1:5–10 18. Arcis H, Rodier L, Ballerat-Busserolles K, Coxam J-Y (2008) Enthalpy of solution of CO2 in aqueous solutions of methyldiethanolamine at T = 322.5 K and pressure up to 5 MPa. J Chem Thermodyn 40:1022–1029 19. Arcis H, Ballerat-Busserolles K, Rodier L, Coxam J-Y (2011) Enthalpy of solution of carbon dioxide in aqueous solutions of monoethanolamine at temperatures of 322.5 K and 372.9 K and pressures up to 5 MPa. J Chem Eng Data 56:3351–3362 20. Kim I, Svendsen HF (2007) Heat of Absorption of carbon dioxide (CO2) in monoethanolamine (MEA) and 2-(aminoethyl)ethanolamine (AEEA) solutions. Ind Eng Chem Res 46(2007/08/01):5803–5809 21. Kim I, Svendsen HF (2011) Comparative study of the heats of absorption of post-combustion CO2 absorbents. Int J Greenhouse Gas Control 5:390–395 22. Singh P, Versteeg GF (2008) Structure and activity relationships for CO2 regeneration from aqueous amine-based absorbents. Process Saf Environ Prot 86:347–359 23. Sodiq A, Rayer AV, Olanrewaju AA, Abu Zahra MRM (2014) Reaction kinetics of carbon dioxide (CO2) absorption in sodium salts of taurine and proline using a stopped-flow technique. Int J Chem Kinet 46:730–745 24. Caplow M (1968) Kinetics of carbamate formation and breakdown. J Am Chem Soc 90 (1968/11/01):6795–6803 25. Chowdhury FA, Okabe H, Yamada H, Onoda M, Fujioka Y (2011) Synthesis and selection of hindered new amine absorbents for CO2 capture. Energy Procedia 4:201–208

Post-combustion Carbon Dioxide Capture with Aqueous (Piperazine + 2-Amino-2-Methyl-1-Propanol) Blended Solvent: Performance Evaluation and Analysis of Energy Requirements Sukanta K. Dash

Abstract Post-combustion CO2 capture (PCC) and its sequestration has been found to be a viable option for reducing CO2 in the earth’s atmosphere. There are many technological options for separation of CO2 from a post combustion gas stream. However, regenerative chemical absorption process is considered to be a near-term feasible solution for this. In regenerative chemical absorption, the key component is the solvent, which plays a major role in the process efficiency and economics. There are many conventional and newer commercial solvents with patented technologies available for this process. In this chapter, the suitability of aqueous AMP along with PZ as an energy efficient mixed solvent for the PCC process have been presented by critically analyzing the absorption rate, equilibrium thermodynamics, reaction kinetics as well as regeneration energy requirement. Energy analysis from bench scale and pilot scale studies, and modelling and simulation work have been investigated and compared with the bench marked solvent MEA. The role of important solvent properties for this application, i.e., density, viscosity, physical gas solubility, reaction mechanism and kinetics, equilibrium solubility and heat of absorption are found to be suitable for the CO2 capture by AMP + PZ solvent. Besides, it is also found that the negative impact such as, corrosion, thermal and oxidative degradation, possible amine and nitrosamine emission from the capture plant have less impact to the environment. Heat energy requirements of this process are found to be in the range of 2.9–3.7 GJ/tCO2 for different conditions such as, %CO2 capture, etc., and from different study. This energy requirement is about 20% less than that of the bench marked MEA solvent. All this performance indicators show that the AMP + PZ blended solvent is a competitive energy efficient alternative one for CO2 capture by chemical absorption.

S.K. Dash (&) Department of Chemical Engineering, Pandit Deendayal Petroleum University, Gandhinagar 382007, Gujarat, India e-mail: [email protected] © Springer International Publishing AG 2017 W.M. Budzianowski (ed.), Energy Efficient Solvents for CO2 Capture by Gas–Liquid Absorption, Green Energy and Technology, DOI 10.1007/978-3-319-47262-1_9

191

192

S.K. Dash

Nomenclature AMP AMPH+ CCS CCTs CH4 CO2‒ 3 CO2 DAE DRS ECO2 eNRTL model GC HCO‒3 He H+PZCOO‒ k L/G MDEA MEA N2 N 2O NOx O2 pamine PCC pCO2 pH 2 O PZ PZCOO− PZ(COO−)2 PZH+ R RA SOx T VLE aCO2 apCO2;lean apCO2;rich DHabs c u dni dCO2

Amino-2-methyl-1-propanol Protonated AMP Carbon capture and storage Clean coal technologies Methane Carbonate Carbon dioxide Differential and algebraic equations Data regression system Enhancement factor Electrolyte non-random two-liquid model Gas chromatography Bicarbonate Helium Protonated PZ carbamate Reaction rate constant Liquid to gas ratio Methyldiethanolamine Monoethanolamine Nitrogen Nitrous oxide Nitrogen oxide Oxygen Partial pressure of amine Post-combustion CO2 capture Partial pressure of CO2 Partial pressure of H2O Piperazine PZ carbamate PZ dicarbamate Protonated PZ Ideal gas constant Rate of absorption Sulfur oxide Absolute temperature Vapor-liquid equilibrium CO2 loading (mol CO2/mol amine) CO2 loading at regenerator outlet pressure CO2 loading at absorber inlet pressure Heat of absorption Activity coefficient Fugacity coefficient Mol change of species with mol change of CO2 in reaction

Post-combustion Carbon Dioxide Capture with Aqueous …

193

1 Introduction Fossil fuels based power plants are the largest contributor to anthropogenic CO2 emission in this planet. At present coal combustion generates about 40% of the world’s electrical energy and coal is likely to maintain this major role for the coming decades. However, a central issue in this will be the increasing use of Carbon Capture and Storage (CCS) and Clean Coal Technology (CCT) initiatives to mitigate the anthropogenic greenhouse gas concentration in the atmosphere. Both will be important pathways in maintaining the use of coal as a power generating fossil fuel in the future. Out of the amine based chemical absorption processes, the MEA based one is considered the bench mark process for CO2 capture. This process has been practised for several years now for CO2 removal from natural gas streams and for synthesis gas processing. It is easier to separate CO2 from a high pressure gas stream but difficult to separate it from the flue gas stream of a power plant, which is essentially at atmospheric pressure. Besides, large volumetric flow rates of flue gas from coal combustion necessitate the requirement of a large diameter absorption column. Again, apart from CO2 and N2, the flue gas often contains SOx, NOx and significant amount of O2. Presence of these components put up further resistance for implementation of chemical process for CO2 capture from atmospheric flue gas streams, as these components can degrade the amine solvent and lower its capacity for CO2 intake. Hence, the major R&D thrust for CO2 capture is on the development of high reactivity, high equilibrium capacity, and higher degradation resistance solvents with lower regeneration energy requirement to make the process economically viable. Advanced solvents can enhance the technology for providing lower energy consumption with equivalent or better mass transfer rates and lower degradation/corrosion impact than those of the bench marked MEA process. The energy requirement can be reduced by using activated solvents based on tertiary amines or sterically hindered amines (SHA). One most commercially used tertiary amine is MDEA and SHA is AMP. As the reaction rates of these amines with CO2 are relatively lower, activators/promoters are necessary to maintain equivalent rates of reaction and mass transfer in the newer solvent. Diamines such as, PZ shows very high CO2 absorption capacity and reaction rate as single solvent for CO2 capture as well as when blended with other amines such as MDEA or AMP [1, 2]. Use of mildly hindered amines such as AMP may also lead to a lower heat of absorption. When the amine is not too greatly hindered, it will provide equivalent mass transfer rates at high loading because the reactive amine is not significantly depleted by reaction with CO2. AMP is characterized by a moderate CO2 absorption rate and a high CO2 stoichiometric loading capacity. But, since AMP is the hindered form of MEA, it absorbs CO2 at a slower rate than MEA, and the low pCO2 of flue gas prevents AMP from realizing the one mole per mole CO2 loading which is stoichiometrically possible for non-carbamate forming amines. While, on the other hand, AMP based solvents offer good absorption and regeneration efficiencies

194

S.K. Dash

towards CO2 as compared to those of MEA [3]. Again, from a bench scale absorber pilot plant study by Hairul et al. [4] shows that the mass transfer performance of AMP + PZ blended solvent has superior mass transfer characteristics over the single AMP solvent. In view of these facts (AMP + PZ) solvent has been chosen for the present work. On going through the resent literature, it is also found that there are lot of R&D focus on this energy efficient solvent.

2 Energy Requirement The objective of the energy efficient solvent is to reduce both the Capex and Opex in a CO2 capture plant attached to a fossil fuel based power plant. The solvent should have appropriate equilibrium solubility and fast reaction kinetics. Furthermore, stability, corrosion and environmental aspects are important. A complete survey of reported energy demand of the PCC process for different patented processes and solvents can be found in Budzianowski [5]. In addition to heat energy, a CO2 capture plant may require electrical energy to drive machineries. The total electrical energy requirement of a 400 MW coal fired power plant with MEA based solvent for CO2 capture is *32.5 MW which is for the inlet CO2 blower to the absorber, recompression of the stripped CO2 and miscellaneous electrical energy needed for the solvent pumps etc., [6]. Overall this electrical energy is a small fraction of the total regeneration energy requirement for the process. As the scope of this chapter is to focus energy efficient solvent for CO2 capture by absorption process, the regeneration energy demand, of few patented process are discussed here. Kansai Electric and Mitsubishi Heavy Industries (MHI), commercially employ PCC solvents based on aq. solution of a sterically hindered amine known as “KS-1TM”. The organization claims that the regeneration energy requirement for the KS-1TM solvent is *3 GJ/tCO2 with the use of their process and proprietary equipments [7]. This energy demand is about 20% lower than that of MEA solvent, which is *3.7 GJ/tCO2 for CO2 capture at coal-fired power plants [8]. MHI also claims that the KS-1TM process have lower degradation and higher corrosion resistance as compared to the MEA process. Siemens and E.ON, use another solvent based on aq. amino acid salt known as “Siemens AAS” [9]. They claim that ‘Siemens AAS’ solvent demands 2.7 GJ/tCO2 for regeneration which operated in a full-scale capture plant. It is realized that in many cases the reported numbers of energy demand are just claimed by the vendors or developers. These are just indicatives, applicable in the specified conditions and should not be compared with each other. Since the chemical absorption of CO2 still remains an energy intensive process in spite of the developments achieved so far, the R&D focus is on advanced and newer energy efficient solvents and processes for CO2 capture. The Chapter begins with a discussion on the regeneration energy demand of commercial solvents claimed by their developers and from literature review in


195

Sect. 2. A brief review of physico-chemical properties, chemical and aqueous phase thermodynamics, amine-CO2 reaction kinetics, as well as, a summary of the previous work in respect of CO2 + AMP + H2O, CO2 + PZ + H2O and CO2 + AMP + PZ + H2O systems are presented in Sects. (3–5). Experimental and theoretical studies of ternary systems such as vapour-liquid equilibrium (VLE) data from the equilibrium cell measurements and the electrolyte non-random two-liquid (eNRTL) modelling of VLE to predict the equilibrium solubility and speciation of these solvents are presented in Sects. 4 and 5. VLE data of CO2 in aq. (AMP + PZ) in a wide range of compositions, temperatures and CO2 partial pressures, ðpCO2 Þ have also been discussed along with the estimation of heats of absorption of CO2 in these solvents. The absorption rate measurements of CO2 in this blended solvent using wetted wall contactor in our laboratory and results of absorption measurements from other literature sources are presented as well. The experimental results and modeling work of CO2 absorption in aq. (AMP + PZ) and the physico-chemical properties necessary for the rate model along with simulation study of absorber-regenerator have been described. Amine degradation and corrosion have been explained in Sect. 6. Energy requirement analysis obtained from a parametric study and bench scale and pilot tests information from the literature are described in Sect. 7. Finally general conclusions and recommendations in this energy efficient process have been discussed.

3 Physico-Chemical Properties of AMP + PZ Solvent Knowledge of physico-chemical properties such as viscosity and density of the solvent is needed for the selection, operation and energy requirement analysis of process equipments such as, the lean-rich heat exchanger, pumps, and for the hydrodynamic calculations of absorber and stripper. Again, these data are required for estimating the liquid phase diffusivity by modified Stokes-Einstein equation. Viscosity, density and physical gas solubility (Henry’s constant) of the mixed solvent AMP + PZ at different temperatures is also required in mass transfer and kinetics rate modelling in the absorber and regenerator as these quantities are influencing the liquid phase mass transfer coefficients. The viscosity and density of aq. (AMP + PZ) has been reported at different temperatures and relative concentration of AMP and PZ. The reader may refer Paul and Mandal [10], Sun et al. [11], Samanta and Bandyopadhyay [12], Dash et al. [13] for a detail discussion about the experimental work as well as development of correlations for viscosity, density of unloaded solvents and physical gas solubility. It is reported by these authors that the density and viscosity decreases with increasing temperature and decreasing PZ concentration in the solution. Viscosity and density data for the CO2 loaded solution could be useful for plant design but these data are scarce as it is difficult to maintain the CO2 loading in the solvent and temperature simultaneously while

196

S.K. Dash

measuring these properties at different temperatures. Any future research on this topic is highly desirable.

4 Thermodynamics of Quaternary Systems: CO2 + AMP + PZ + H2O Thermodynamic study includes the understanding of CO2 equilibrium solubility, pH of loaded solution, heat of absorption, distribution of chemical species in the loaded solvent (speciation) and amine vaporization losses. The CO2 equilibrium solubility in liquid phase has contributions of both physical solubility and solubility by chemical reaction. The VLE data at different temperatures, solvent concentrations, and CO2 partial pressures have significant role in the design and optimization of absorber-regenerator as it is necessary to calculate the driving force for these equilibrium governed processes. Normally, absorber model is represented mathematically by a set of differential and algebraic equations (DAE). Information on physical and chemical equilibrium acts as boundary conditions for the DAE. For a detail discussion on this, the reader may refer Dash et al. [13] for the development of DAE for the CO2 + AMP + PZ + H2O system. The equilibrium solubility of CO2 in the liquid phase also determines the solvent recirculation rate in the absorber-regenerator set up. Besides, it also fixes the CO2 concentration in the outgoing gas stream from the absorber and regenerator. Appropriate mathematical model is necessary to represent the VLE of CO2 in the solvents. Among the different models used to represent VLE of CO2 in amine solvents, the frequently used model considers the activity coefficient-fugacity coefficient model for the phase equilibrium, known as the c-u approach. One such popular model is electrolyte non-random two-liquid (eNRTL) model used by Dash et al. [14–16] to model VLE of CO2 in aqueous PZ, aqueous AMP and aqueous (AMP + PZ) solvents.

4.1

Experimental Techniques for the Measurement of VLE

The various experimental techniques used to measure the vapor (CO2) and liquid (solvent) equilibrium are classified in two categories, (i) static method and (ii) dynamic method. In the static method, the total pressure ðpCO2 þ pH2 O þ pamine Þ is measured with a pressure measuring device. The mixing of vapor and liquid phases is carried out by agitating the liquid phase by a shaft mounted with impellers or by using a magnetic stirrer. The pCO2 is then obtained by subtracting the water vapour pressure from the total pressure thus neglecting the amine vapour pressure. It is known that, this method is highly productive to generate equilibrium data at high temperature and pressure for CO2-amine-H2O systems. In the dynamic method the


197

vapour composition can be analyzed by using gas chromatography (GC), infrared analyzer, and mass spectrometer. To maintain a relative high pressure in the system, N2, CH4, or He is usually used as the inert gas. Liquid samples can be analyzed by titration. However, for mixed or blended amine solvents, liquid analysis by titration is difficult. A complete discussion of the different experimental methods for VLE measurement of CO2 in AMP + PZ solvent and its modelling, the reader may refer Bruder et al. [17], Dash et al. [14, 15], Tong et al. [18], Hartono et al. [19], Haghtalab et al. [20] and Halim et al. [21].

4.2

VLE of CO2 in Aqueous PZ

Dash et al. [14, 15, 22] have obtained the VLE data of CO2 in aq. PZ, aq. AMP and aq. (AMP + PZ) using the static method explained earlier. Incorporating the eNRTL equation for VLE modelling, they determined the interaction parameters by data regression using the Data Regression System (DRS) available in Aspen Plus® simulation software. As VLE of CO2 in single amine such as PZ and AMP are necessary to formulate a model to represent the VLE of CO2 in blended amines such as AMP + PZ, it is worthwhile to report here some of the results of VLE of CO2 in aq. PZ and aq. AMP. Figure 1 shows a typical VLE plot of CO2 over 0.6 mol.dm3 PZ at various temperatures. This VLE is presented by plotting the pCO2 , kPa) verses CO2 loading in liquid phase (aCO2 mol CO2/mol amine). The model results have been obtained using the interaction parameters reported by Dash et al. [14]. The behaviour of the equilibrium curves can be explained by observing the trend in the experimental data and model predictions. This trend is, pCO2 increases with

100

100

T = 343 K

CO 2 Partial Pressure [kPa]

Fig. 1 Equilibrium pCO2 over aqueous 0.6 mol/dem3 aq. PZ at 313 and 343 K. Experimental result from reference [23, 24]

10

10

1

1

Expt. Data: [PZ]=0.6 M

0.1

0.1

-Bishnoi, T= 313 K -Bishnoi; T= 343 K -Derks; T= 343 K 0.01 -Derks; T= 313 K -eNRTL model -eNRTL model

0.01

T = 313 K 1E-3

1E-3

0.0

0.2

0.4

0.6

0.8

Loading (mol CO2/mol PZ)

1.0

1.2

198

S.K. Dash

increase in CO2 loading in liquid phase (expressed in mol CO2/mol PZ) and temperature. One important information obtained from the thermodynamic modelling are the solution chemistry and liquid phase speciation. After estimating the c values of the species in liquid phase, the concentration of the species can be calculated with respect to the extent of reaction with CO2. Normally, the extent of reaction is expressed in terms of liquid phase CO2 loading and can be predicted up to the stoichiometric limit of loading of CO2 in the solvent which is 2 mol of CO2/mol of PZ theoretically. Figure 2 shows the reaction stoichiometry i.e., the relation among change of moles of any species in the liquid phase with respect to change in number of moles of CO2 verses CO2 loading. Figure 3 shows the speciation of CO2 + PZ + H2O system predicted using eNRTL model [14]. As usual PZ is

Fig. 2 Model predicted reaction stoichiometry in 3.2 mol/l PZ solution at 318 K

6

PZH

+

+

H PZCOO

-

HCO 3

-

4

CO 2

2 δni

0

δ CO 2

2-

-2

CO 3

PZ

-

-4

PZCOO

-6 0.0

0.2

0.4

0.6

PZ(COO )2

-

[PZ] = 3.2 M T = 318 k 0.8

1.0

α CO2 (mol CO 2 / mol PZ)

1.0

Fraction of PZ species in liquid phase

Fig. 3 Liquid phase speciation (xi) in CO2 loaded 0.8 mol/l aqueous PZ at 328 K

[PZ] = 0.8 M T = 328 K

PZ

0.8

+

-

H PZCOO

+

PZH

0.6

-

HCO3

0.4

-

PZCOO

-

PZ(COO)2

0.2

CO2 2-

CO3 0.0 0.2

0.4

0.6

0.8

1.0

Loading mol (CO2/mol PZ)

1.2

1.4


199

consumed gradually with CO2 loading forming protonated PZ carbamate and PZ dicarbamate at lower loading. At higher loading, PZ carbamate and PZ dicarbamate are converted to protonated PZ carbamate and at this concentration most of the PZ exists in the protonated form (PZH+). This indicates that the activity of PZ is more in these conditions of CO2 loading. Availability of PZ activity also supports the action of PZ as a good rate promoter for CO2 reaction. The heat of absorption is an essential data for predicting the performance of the solvent as it is a major component of the total energy requirement for solvent regeneration. The heat of absorption can be estimated from the VLE data using the well known Gibbs- Helmholtz equation given in Eq. (1): DHabs d ln pCO2 ¼ R dð1=TÞ

ð1Þ

Figure 4 compares experimental heat of absorption of CO2 in 2.4 m (mol/kg water) aq. PZ at 313 K taking data from Hillard [25] and our own results using the model predicted equilibrium pCO2 . As expected, heat of absorption of CO2 in aq. PZ is comparable to that of MEA, which is *80 kJ/mol CO2. Amine volatility is important information needed for choosing the solvent for CO2 capture application. Figure 5 compares the experimental and model results of PZ volatility expressed in partial pressure of PZ over the CO2 loaded solvent.

4.3

VLE of CO2 in Aqueous AMP

Equilibrium concentration of CO2 in aq. AMP and the pCO2 over the solution is also an important data for the thermodynamic modelling of CO2 + AMP + H2O and CO2 + AMP + PZ + H2O systems. Many authors have reported this important VLE data and a good review on this topic can be found in Bougie and Iliuta [26]. Fig. 4 −ΔHabs (kJ/mol CO2) of a 2.4 mol/(kg water) aqueous PZ at 313 K

100 [PZ] = 2.4 m, T=313 K

-ΔHabs (kJ/ mol CO 2)

80

60

40

20

- model predicted Experimental results

- Hilliard (2008) 0 0.0

0.2

0.4

0.6

0.8

αCO2 (mol CO2 / mol PZ)

1.0

200

S.K. Dash

Fig. 5 Amine volatility for a 2 mol/(kg water) PZ solution at 313 and 343 K loaded with CO2

-1

10

[PZ = 2 m] 343 K -2

pPZ , [kPa]

10

-3

10

313 K

- model predicted Experimental results - Hilliard (313 K) - Hilliard (343 K)

-4

10

-5

10

0.2

0.4

0.6

0.8

1.0

αCO2 (mol CO2/ mol PZ)

Here, we compare few literature data with our own modelled results. Figure 6 represents the experimental solubility of CO2 in 3 mol/dm3 aq. AMP from Tontiwachwuthikul et al. [27] and our own work [12]. Figure 7 represents solubility of CO2 in 5 mol.dm−3AMP from Teng and Mather [28]. Figures 8 and 9 present the solubility data CO2 over aq. AMP and VLE modelling results at different temperatures considering the model parameters from Dash [29]. In this case also, the equilibrium pCO2 is low at lower CO2 loading because at this condition CO2 is almost consumed in the solvent by chemical reaction. But pCO2 increases at high CO2 loading which can be due to the dominance of physical absorption. A detail discussion of other properties i.e., model predicted activity coefficients, liquid phase speciation, reaction stoichiometry, heat of absorption and amine volatility, can be found in Dash et al. [15] and Tong et al. [28]. The −ΔHabs of CO2 into aqueous AMP solvent is estimated from the eNRTL model using

1000

Fig. 6 Solubility of CO2 over 3 mol/l AMP at different temperatures CO2 partial pressure [kPa]

[AMP] =3 M 100

10

Experimental results Tontlwachwuthikul

1

- 293 K - 313 K - 333 K - 353 K - lines model predicted

0.1

0.01 0.2

0.4

0.6

0.8

Loading (mol CO2/ mol AMP)

1.0

1.2


201

10000

Fig. 7 Solubility of CO2 over 5 mol/(kg water) AMP at 323 K

Y1 Axis Title

1000

100

Experimental Teng & Mater 5. 0m T = 50 C - Model

10

1

0.1 0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.3

X Axis Title

pCO2, [kPa]

Fig. 8 Equilibrium pCO2 over 22 mass% aqueous AMP solution

10

3

10

2

10

1

10

0

Experimental results This work, T = 298 K T = 308 K T = 318 K T = 328 K - (lines) model predicted

[AMP] = 22 mass % 10

-1

0.2

0.4

0.6

0.8

1.0

1.2

α CO2 (mol CO 2 /mol AMP)

pCO2, [kPa]

Fig. 9 Equilibrium pCO2 over 30 mass% aqueous AMP solution

10

3

10

2

10

1

10

0

10

-1

10

-2

Experimental results T = 298 K T = 308 K T = 318 K T = 328 K - (Lines) model predicted

[AMP]= 30 mass %

0.2

0.4

0.6

0.8

αCO2 (mol CO2/ mol AMP)

1.0

202

S.K. Dash

Eq. (1). It is found to be about 65 kJ/mol CO2 by taking an average value between 313 and 373 K.

4.4

VLE of CO2 in Aqueous (AMP + PZ)

Since this chapter discusses the suitability of AMP + PZ solvent for PCC process, we focus more on the VLE of CO2 in AMP + PZ solvent and the energy requirement. The solubility of CO2 in aqueous 2 and 3 mol/dm3 AMP + 0.5, 1.0 and 1.5 mol/dm3 PZ and 25 and 20 wt%, AMP + 5 and 10 wt% PZ have been presented by Yang et al. [30] and Tong et al. [18], respectively. Both the workers modelled the VLE using Kent- Eisenberg approach. Tong et al. [18] compared the results with the solubility of CO2 in 30 wt% AMP and reported that the stoichiometric loading of this blend is comparable to that of 30 wt% MEA. The solubility of CO2 in aq. (3 mol/dm3 AMP + 1.5 mol/dm3 PZ) has been presented by Bruder et al. [17] after screening some other compositions of AMP and PZ. They found that this composition has shown the highest cyclic capacity and can be used commercially without the problem of solid precipitation upon CO2 loading. Dash et al. [16, 22] presented experimental results of VLE of CO2 in aqueous 30 wt% (AMP + PZ: 22 + 8, 25 + 5, and 28 + 2), 40 wt% (AMP + PZ: 40 + 0, 32 + 8, 35 + 5, 32 + 8) and 50 wt% (AMP + PZ: 50 + 0, 42 + 8, 45 + 5, 48 + 2) total amine. The VLE of CO2 in these solvents have been represented by modelling it using eNRTL equation for the ternary system AMP + H2O + CO2 and PZ + H2O + CO2 and also by regressing the VLE data of the quaternary system AMP + PZ + H2O + CO2. Figure 10 compares typical CO2 solubility data at different temperature and the smoothed curves obtained using e-NRTL model of this work.

pCO2 /kPa

Fig. 10 Equilibrium pCO2 over aqueous (25 wt% AMP + 5 wt% PZ) at 298– 328 K

10

3

10

2

10

1

10

0

10

-1

10

-2

Experimental results, this work [AMP + PZ] = (25 + 5) mass % T: 298 K T: 308 K T: 318 K T: 328 K (Lines): Model

0.2

0.4

0.6

0.8

αCO2, mole CO 2/mole amines

1.0


203

It is observed that more PZ in AMP + PZ solvent leads to increase in CO2 loading as the VLE curves exhibit. From these VLE curves the solvent capacity can be calculated using Eq. (2). The assumptions in this calculation are (a) the liquid phase CO2 loading in the rich solvent outlet stream apCO2 ;rich from the absorber is

in equilibrium with the pCO2 in the inlet gasstream and (b) the liquid phase CO2 loading in the lean solvent stream apCO2 ;lean entering to the absorber is in equilibrium with the pCO2 in the outlet gas stream. capacity

mole CO2 mole ðAMP + PZÞ ¼ apCO2 ;rich apCO ;lean ½AMP þ PZ 2 kg solvent kg solvent ð2Þ

The solvent capacity of aq. (35 wt% AMP + 5 wt% PZ) and (25 wt% AMP + 5 wt% PZ) are 0.902 and 0.623, respectively, at 313 K and at the lean-rich pCO2 of 1 and 10 kPa respectively, as reported by Dash et al. [16]. They also reported that increasing PZ concentration in the aqueous AMP solvent and increasing the total amine concentration in the solvent both increased the solvent capacity. Higher capacity can be achieved if the pCO2 is more in the flue gas rather than decreasing the lean loading of the solvent at the outlet of the regenerator. This behaviour also observed in case of aq. PZ, for example the cyclic capacity of 4.5 mol/dm3 PZ solvent is 40% more than the capacity of 3.2 mol/dm3 PZ when analyzed at 1 kPa lean pCO2 and 10 kPa rich pCO2 . The eNRTL model can predict the liquid phase concentration of various species formed by CO2 reaction such as protonated AMP (AMP+), Protonated PZ, PZCOO−, PZ (COO−)2, H+PZCOO−, carbonate (CO32−) and bicarbonate (HCO3−) ions. This speciation is presented in Figs. 11 and 12 for 22 wt% AMP + 8 wt% PZ and 35 wt% AMP + 5 wt% PZ, respectively, at 313 K. It is evident that there are 4.0

-3

3.5

Constration of species, mole.dm

Fig. 11 Model predicted speciation in CO2 loaded aqueous (22 wt% AMP + 8 wt% PZ) at 318 K

-

HCO3

3.0

AMP

AMPH

2.5

+

PZ 2.0 -

PZCOO

1.5

PZH

-

PZ(COO )2

1.0

+ 2-

CO2

CO 3 +

-

H PZCOO

0.5

OH

-

0.0

0.0

0.2

0.4

0.6

0.8

αCO2 , mole CO2/ mole amines

1.0

1.2

204 0.6 -3

AMPH

Concentration of species, kmol.m

Fig. 12 Liquid phase PZ species of a CO2 loaded aqueous (35 wt% AMP + 5 wt% PZ) at 313 K

S.K. Dash

+

AMP

0.5

2-

CO3

0.4

PZ

-

HCO3 +

H PZCOO

-

PZ(COO )2

0.3

-

0.2 -

PZCOO 0.1

OH

PZH

-

+

CO2

0.0 0.0

0.2

0.4

0.6

0.8

1.0

αCO 2(mol CO2/ mol amine)

differences between the speciation of aq. (AMP + PZ) and aq. AMP or aq. PZ. Both AMPH+ and PZH+ are the predominant products in the carbonated solution of AMP + PZ favouring rapid CO2 absorption. The fast formation of AMPH+ and its concentration delays the protonation of PZ towards higher CO2 loading making more PZ activity and reactive PZ available in the solution. This characteristics favours fast kinetics of PZ-CO2 reaction and thus make the solvent suitable for CO2 capture. As discussed earlier, the heat of absorption of CO2 in AMP + PZ solvent is an important information for the estimation of the energy demand in the regenerator. Using a differential reaction calorimeter, Xie et al. [31] measured the heat of CO2 absorption in 3 mol/dm3 AMP + 0.5 mol/dm3 PZ, 4 mol/dm3 AMP + 1 mol/dm3 PZ and 3 mol/dm3 AMP + 1.5 mol/dm3 PZ at 313–373 K. It is reported by them that heat of absorption is high and it slowly decreases in the low CO2 loading region. This is due to the fact that at the low CO2 loading region, PZ is converted to PZ carbamate which contributes about 30.07 kJ/mol energy as heat of reaction. Their finding revealed that ΔHabs of the three different blends of AMP + PZ are comparable and at low loading this value is about 75 kJ/mol. Dash et al. [14] reported the eNRTL model predicted average ΔHabs between 313 and 343 K. It is about 70 kJ/mol of CO2 for aqueous (AMP + PZ) solvent which is about 7% lower than the value found out experimentally by Xie et al. [31]. The low heat of absorption value qualifies this AMP + PZ solvent as an energy efficient one. The information on amine volatility expressed as vapour pressure of AMP at the absorber outlet temperature is needed to quantify the amount of amine loss with the outgoing flue gas. As excessive volatility of the solvent should be avoided to reduce amine loss to environment, a water wash packed section should be provided at the top portion of the column. Nguyen et al. [32] presented the amine volatility of 5 m AMP solvent at 313–333 K as data on volatility is important at the conditions of the outlet gas from the absorber. They reported that AMP has a slight higher volatility


205

of 112 ppm at this condition of temperature and pCO2 as AMP does not form non-volatile carbamates in the carbonated solution. Dash et al. [16] showed that the amine volatility of a mixed solvent of aqueous (AMP + PZ) loaded with CO2 is less than that of CO2 loaded single amine solvent i.e., aq. AMP or aq. PZ. Here, addition of PZ in AMP shows a positive effect of lowering the volatility of both the amines due to highly non ideal phase of the carbonated solution. For a detailed discussion of non-ideality and activity coefficients, amine volatility, heat of absorption, and other properties of CO2-(AMP + PZ) system Dash et al. [16] may be referred. Khakharia et al. [33] reported amine emission to environment for a AMP + PZ CO2 capture process. Their experimental study reviled that the amount of amine emission to environment depends on lean solvent temperature. Amine emission is reduced with increasing lean solvent temperature and aerosol emissions are found only when lean solvent have high pH value. The measured emission of AMP is in the range of 1500–3000 mg/Nm3 and PZ in the range of 200–400 mg/Nm3. This emission can be controlled by providing a water wash section in the absorber that fall within the environmental regulations.

5 Absorption of CO2 into AMP + PZ Solvent One of the most essential data for CO2 absorption in amine solvents is the kinetic data. Both high equilibrium capacity and high absorption rate are desirable for the CO2 capture solvent. For AMP and AMP + PZ solvents, knowledge about reaction mechanism and kinetic constants are important. The CO2 absorption rates into various compositions of aq. (AMP + PZ) have been measured by various workers [9, 13, 34] at different temperatures and at different CO2 partial pressures using wetted wall column. For the purpose of calculation of kinetic parameters using the coupled mass transfer kinetics model, they considered the following reactions of CO2 with aq. (AMP + PZ) represented in reactions R1–R11. K1 ;k21

CO2 þ AMP þ H2 O $ AMPH þ þ HCO 3 K2 ;k22

CO2 þ PZ þ H2 O $ PZCOO þ H3 O þ K3 ;k23

CO2 þ AMP þ PZ $ PZCOO þ AMPH þ K4 ;k24

CO2 þ PZCOO þ H2 O $ PZ(COO Þ2 þ H3 O þ K5 ;k25

CO2 þ AMP þ PZCOO $ PZ(COO Þ2 þ AMPH þ

ðR:1Þ ðR:2Þ ðR:3Þ ðR:4Þ ðR:5Þ

206

S.K. Dash K6 ;k26

þ CO2 þ 2H2 O $ HCO 3 þ H3 O

ðR:6Þ

K7

ðR:7Þ

K8

ðR:8Þ

K9

ðR:9Þ

2 þ HCO 3 þ H2 O $ CO3 þ H3 O

PZ þ H3 O þ $ PZH þ þ H2 O PZCOO þ H3 O þ $ H þ PZCOO þ H2 O K10

ðR:10Þ

2H2 O $ H3 O þ þ OH

ðR:11Þ

AMPH þ þ H2 O $ AMP þ H3 O þ K11

Samanta and Bandyopadhyay [34] reported the values of kinetic parameters (k23 and k25) for the CO2 reaction with aq. total 30 wt% (AMP + PZ) in the range of 298–313 K. These values are presented in Eqs. (R.3) and (R.4).

k23

7:2 104 1 1 ¼ 3:516 10 exp T 298 R 4

k25

7:6 104 1 1 ¼ 1:836 10 exp T 298 R

ð3Þ

4

ð4Þ

Dash et al. [2] worked with more concentrated solvents having total 40 wt% (AMP + PZ) and total 50 wt% (AMP + PZ) keeping PZ concentration within the range of 2–8 wt%. They reported the absorption measurements at 303–333 K and pCO2 in the range of 4.9–15.1 kPa. In that work, for the first time they incorporated the eNRTL model to estimate the initial liquid phase concentrations for the coupled mass transfer- kinetics model to predict the kinetic parameters. The kinetic parameters obtained for total 50 mass% (AMP + PZ) are given by Eqs. (R.5) and (R.6). 65225 1 1 k23 ¼ 3:516 104 exp R T 298 70550 1 1 k25 ¼ 1:836 104 exp R T 298

ð5Þ ð6Þ

The Arrhenius plots of these kinetic parameters are represented in Fig. 13. The effect of temperature on specific rate of absorption (RA) of CO2 into aq. (45 wt% AMP + 5 wt% PZ) at different pCO2 have been presented in Fig. 5.13. It is evident that the rate of CO2 absorption increases with the increase of temperature and pCO2 . Considering the experimental results of CO2 absorption rate in aq. (45 wt% AMP + 5 wt% PZ) solvent from Dash et al. [2], we have analysed the variation of

Post-combustion Carbon Dioxide Capture with Aqueous … 10

8

10

7

10

6

10

6

10

5

10

4

-

_

kPZCOO (m6.kmol-2.s-1)

eq. (5.55) eq. (5.56)

k2PZ(COO )2(m6.kmol-2.s-1)

Fig. 13 Arrhenius plot for the reaction rate constant, Kpzcoo–(k23) and kPZ(COO–) 2 for CO2– (AMP + PZ + H2O) 50 wt% (AMP + PZ)

207

10

5

10

4

2.9

3.0

3.1

3.2

3.3

1000/T(K)

rate of absorption with temperature and the trend of enhancement factor with CO2 loading. Figure 14 shows the variations of the measured rates of absorption with pCO2 at different temperatures. It shows that the rates of absorption increase with pCO2 . Similarly, Fig. 15 shows the variation of the measured enhancement factor Fig. 14 Effect of temperature on specific rate of absorption (RA) of CO2 into aq. (45 wt% AMP +5 wt% PZ) at different CO2 partial pressure

35

RAx106(kmol.m-2.s-1)

30

pCO2=15 kPa

25 pCO2=10 kPa 20 pCO2= 5 kPa

15

10 320

324

328

332

336

T(K)

35 323 K 328 K 333 K

30

6

-2

-1

RAx10 (kmol.m .s )

Fig. 15 Effect of partial pressure on specific rate of CO2 absorption into aqueous (45 wt% AMP + 5 wt% PZ) at different temperatures

25

20

15

10 4

6

8

10

pCO2(kPa)

12

14

16

208

S.K. Dash

Fig. 16 Effect of CO2 loading on the enhancement of CO2 absorption rate into aqueous solution of 45 wt% AMP + 5 wt% PZ at different temperatures

420 323 K 328 K 333 K

410 400

ECO2

390 380 370 360 350 0.02

0.03

αCO2

0.04

0.05

ðECO2 Þ with CO2 loading. It shows that the enhancement factor ðECO2 Þ sharply decreases with increase in the loading of CO2 until a loading of about 0.035. Thereafter, enhancement factor remains essentially constant (Fig. 16). All these findings indicate that rapid reaction occur at low CO2 loading corresponding to low pCO2 indicating the suitability of the AMP + PZ solvent for CO2 capture from flue gas.

6 Amine Degradation and Corrosion Usually CO2 capture plants operate at near ambient to high temperature. The regenerator operates at elevated temperature of about 380–400 K. At this temperature the solvent may degrade producing components of its homologous series. Another difficulty in flue gas treating is the fact that it contains O2 which may result in oxidizing the solvent species and accelerating the production of degradation products. Thus, high a temperature and presence of O2 environment both lead to higher solvent degradation. These amine degradation products are sometimes responsible for corrosion in the stripper, reboiler and the piping. Again, corrosion may be there due to carbonated solution and it may be vary localized where temperature is high and concentration of O2 is relatively more. Corrosion and corrosion inhibition of aqueous AMP has been studied by Veawab et al. [35, 36] using static weight loss method. They reported that aq. solutions of AMP were more corrosion resistant to carbon steel as compared to aq. MEA solution under similar environment of both pure CO2 and a mixture of CO2 + N2 + O2 having about 10% O2 in it. Freeman et al. [37] reported that both AMP and PZ are more stable than other alkanolamines. They also reported that both AMP and PZ are slower in oxidative degradation than MEA and much more resistant to thermal degradation than MEA. Although AMP is relatively stable, it is susceptible to oxazolidinone formation in concentrated O2 environment. Due to presence of NOx in flue gas, it may act as a source of nitrosamine formation in the degraded products. A complete degradation


209

study of AMP and PZ based solvents, the degradation mechanism and effect of degradation products, can be found in Wang and Jens [38, 39], Mazari et al. [40]. In his report, Wang and Jens [39] mentioned that mononitrosopiperazine may be present in the degraded product of AMP + PZ solvent which may emit to the environment but can be avoided by adding water wash section. All these study reveals that overall, with respect to corrosion, thermal and oxidative degradation, AMP + PZ solvent yet have an advantage over the bench marked MEA.

7 Energy Demand Analysis from Pilot Plant Study and Process Simulation As discussed earlier in the Sect. 5.2, there are many vendors developing commercial solvents for PCC processes. Few examples are described in the energy analysis section and a detailed review of the energy requirements for the commercial solvents developed by power generation companies can be found from Budzianowski [5]. A comparison of energy requirement for MEA and AMP + PZ process is described here. Mangalapally and Hasse [41] performed pilot-plant study of CO2 absorption using both MEA and AMP + PZ blended solvents. They found that for the gas fired power plant case (pCO2 = 5.5 kPa in their flue gas), the energy demand for MEA in the pilot-plant was around 7.2 GJ/t CO2. For coal fired power plant case (pCO2 = 11.20 kPa), it was around 5.5 GJ/t CO2 to reach 90% removal, which is relatively high compared to the standard energy demand of around 4 GJ/t CO2. Notz et al. [42] conducted systematic pilot-plant study of CO2 absorption from natural gas-fueled power plant using MEA absorption process. They reported that for a loading range of 0.096–0.36 mol CO2/mol MEA, the specific energy demand was on the lower side of 3.7 GJ/ton of CO2 and on the higher side of 10.2 GJ/ton of CO2. In order to compare the performance of different solvents, it is necessary to perform pilot plant study and process simulation on a consistent basis and perform a process analysis of the system. Concentrated aqueous blends of AMP and PZ have been proposed to have improved rate kinetics and batter capacity solvent for CO2 capture with lower energy requirement. The regeneration energy studies have been presented. The simplified PCC process using amine absorption is presented in Fig. 17. In this process, the treaded flue gas (cooled and desulfurized) is fed to an absorption column where gas phase CO2 reacts with the aq. amine solvent and transferred to solvent phase. This CO2-rich solvent is sent to the regenerator column where heat is used to strip the CO2 from the solvent. The process flow sheet described in Fig. 17 is a simple one which has been practised for several years. But, for flue gas treating, alternative process configurations should be considered to lower the energy requirement. Performance of various alternative process configurations have been studied by Karim et al. [43,

210

S.K. Dash

CO2OUT

GASOUT

LEANIN ABSORBER

RICHIN

STRIPPER

GASIN LRHX

RICHOUT LEANOUT

Fig. 17 Simplified flow sheet of absorption-regeneration unit

44]. Based on their simulation study they suggested a number of process alternatives and conditions to be followed. One such consideration is temperature profile of absorber and regenerator. Since CO2 absorption in aqueous amines is an exothermic process, it increases the temperature of the liquid phase and reduces the driving force. On the other hand increased temperature also increases the rate of reaction. Hence, there should be a trade-off between this two effects and the optimum temperature profile should be adopted. One configuration in view of energy requirement is the split flow process. In this process a fraction of the partially lean and partially rich solvent from the middle of the regenerator and absorber column respectively, can be withdrawn and reconnected at the bottom of the respective columns. Cousins et al. [45] studied these alternatives i.e., inter stage cooling for temperature control and the split flow process and found that there could be increase in capture efficiency. The other alternative arrangements are rich split and vapour recompression in the regenerator and heat integration. In rich split configuration, vapour liberated from the hot rich solvent transfers heat to the cold rich solvent entering to the regenerator. In vapour recompression, striping steam is removed from a suitable location of the regenerator, compressed and reintroduced into the regenerator to supply additional heat. The recompression of the vapour is done by applying mechanical energy which is utilized to provide additional stripping steam. Vapour recompression is often suggested for reducing regeneration energy requirement but it is achieved by providing additional energy for compression. The extensions of vapour recompression are multi-pressure stripping and matrix stripping reported by Cousins et al. [45]. Heat integration aims at utilizing some of the waste heat from the PCC process and hence reducing the heat loss. The pinch analysis which analyzes approach to equilibrium in the absorber and stripper determines the rich and lean loadings and the capacity of the solvent which also helps energy minimization and efficient plant operation. The rich loading specifically determines the minimum vapor rate at the top of the stripper. There are


211

important trade-offs of absorber and stripper packing height, solvent rate, and steam rate. Any change in contactor, solvent, process configuration etc., should be followed by a careful optimization of the rich and lean loadings. Cousins et al. [46] and Karimi et al. [43] compared five alternative stripper configurations with respect to energy consumption and capital investment. Process simulation tools UniSim and ProTreat were used by them to investigate the alternative stripper configurations. Of the five alternative stripper configurations considered by Karimi et al. [43], the vapor compression stripper configuration was found to be the best by them. The vapour recompression technique can be adopted for this AMP + PZ process to gain some energy advantage.

8 Regeneration Energy Requirement for AMP + PZ Solvent Prior to commercialization of CO2 capture process, several pilot plant studies are necessary to gain experience in this process. Mangalapally and Hasse [41] presented pilot plant study of CO2 capture process. They used AMP + PZ mixed solvent with an objective of commercialization of the AMP + PZ process. They reported that AMP + PZ is a suitable solvent for CO2 capture from fossil fuel based power plants. Based on their pilot plant study, they also found, CESAR-1 solvent 28 wt% AMP + 17 wt% PZ has an expected reduction of about 20% in the regeneration energy demand as compared to that of MEA solvent. Also, AMP + PZ solvent can reduce about 45% in the solvent circulation rate over those of the MEA process. Artanto et al. [46] conducted pilot plant study of CO2 capture from flue gas using 25 wt% AMP + 5 wt% PZ solvent and compared its performance with that of 30 wt% MEA solvent. They reported that AMP + PZ solvent shows better performance over MEA when the pilot plant study is conducted for the same %CO2 removal and at the same L/G ratio. Their modelling work also revealed that the regeneration energy could be lower by using concentrated solvent since the cyclic capacity of the solvent would be higher. They found that the condenser heat duty and the sensible heat duty are the major contributions to the total energy demand. The energy efficiency of the aq. (18 wt% AMP + 17 wt% PZ) as a solvent for CO2 absorption by chemical process has been studied by Dash et al. [47]. A parametric study of this process based on regenerative CO2 absorption with AMP + PZ solvent has been done by them using rate based model. They compared their simulation results with the reported pilot plant data for CO2 capture using aq. AMP and aq. (AMP + PZ) [41] and reported good agreements. From the modeling and simulation work of Dash et al. [47], It has been found that the optimum regeneration temperature of AMP + PZ solvent is about 393–398 K at the bottom and 381–384 K at the top of the regenerator column. As the reboiler heat duty represents the major portion of the total operating cost in a PCC plant, they analyzed the reboiler heat duty by splitting it into three components viz., energy to be

212

S.K. Dash

supplied (i) for providing the heat of absorption or heat of desorption of CO2, (ii) to generate additional stripping steam, and (iii) for rich solvent heating and to provide the sensible heat. By making an energy balance as described by Artanto et al. [46], they estimated all these quantities. These findings of the individual energy components are illustrated in Figs. 18 and 19 with respect to various %CO2 capture, rich loading at the absorber and lean loading at the regenerator, respectively. To have a sensitivity analysis on the lean loading and the reboilor duty, it is found that with the increase of %CO2 capture, the reboiler duty increases as well. Due to higher sensible heat requirement, there is a steep increase in reboilor duty once the capture approaches 95%. It is estimated that the total energy demand is 3.9 GJ/ton of CO2 at an L/G of 2.75 and a lean loading of 0.129. However, the energy demand is 3.75 GJ/ton-CO2 at an L/G of 2.9 and a lean loading of 0.155 for the same conditions. Hence, the energy demand is less when the lean loading is lowered to achieve the desired %CO2 capture rather than by increasing the L/G ratio. 4600 0.50 Rich loading

Reboiler duty (kJ/kg-CO 2)

4400

0.40 (18 wt.% AMP + 17.5 wt.% PZ) Column ht.= 10 m

4000

0.35 0.30

L/G= 2.9

3800

0.25 3600

Lean loading 3400

0.20 0.15

3200 0.84

0.10 0.86

0.88

0.90 0.92 %-CO2 capture

0.94

0.96

4200

3.0 L/G

Reboiler duty (kJ/kg-CO 2)

Fig. 19 Effect of CO2 capture on reboiler duty (kJ/kg CO2) for a constant lean loading with a temperature approach of 5 K in the lean-rich heat exchanger. Reproduced from Dash et al. [47] with permission

0.45

4200

Loading (mol CO2/mol amine)

Fig. 18 Effect of % CO2 capture on reboiler duty (kJ/kg CO2) for a constant L/G with a temperature approach of 5 K in the lean-rich heat exchanger. (Reproduced from Dash et al. [47] with permission)

2.5

4100

2.0

(18 wt.% AMP + 17.5 wt%. PZ) Lean loading = 0.12 Column ht.= 10 m

4000

1.5

3900 1.0 3800

0.5 Rich loading

3700

0.0 84

86

88

90

92

%-CO2 capture

94

96


213

It is found that for 90% CO2 removal, about 1.56 GJ/tCO2 is required towards heat of desorption of CO2. The heat of desorption for AMP + PZ solvent is independent of CO2 loading up to a value of about 0.5 mol CO2/mol amine. About 1.45 GJ/tCO2 goes towards generation of stripping steam which again increases with the requirement of increase in % CO2 capture. Only about 0.64 GJ/tCO2 is needed to meet the sensible heat requirement, which decreases to some extent with the increase in stripping steam. The total energy requirement for the AMP + PZ solvent is *3.65 GJ/tCO2 and *3.2 for 90 and 85% CO2 capture respectively. This range is nearly equal to the value reported by Mangalapally and Hasse [41], but slightly lower than the value obtained by Artanto et al. [46] from their pilot plant study. The reason could be Artanto et al. [46] used lower concentration of 25 wt% AMP + 5 wt% PZ solvent in their pilot plant study and found the energy demand to be *4 GJ/tCO2. This value could be still lower if they would have used higher amine concentration such as, 20% AMP + 15% PZ. On the contrary Khan et al. [3] reported that the regeneration energy demand increases with the increase in PZ concentration as PZ has a heat of absorption *80 kJ/mol of CO2. They also found that the energy demand for the 30 wt% total AMP + PZ solvent is in the range of 3.4–4.2 GJ/t CO2 whereas for 30 wt% MEA it is 3.6–4.5 GJ/tCO2 when both the solvents tested in the same bench scale pilot plant. Spek et al. [48] conducted simulation study of a full scale CO2 capture plant and reported a regeneration heat duty of 2.9 GJ/tCO2 for AMP + PZ solvent as compared to 3.6 GJ/tCO2 for MEA solvent. They also verified that all the desirable properties of AMP + PZ solvent are technical performance indicators for the capture plant. As PZ has a solubility limit in aq. AMP, the final solvent composition should be selected based on solvent circulation and the energy requirement as well as the maximum solid (PZ) solubility so that, there are no precipitate formation in the colder parts of the plant and equipments. This higher solubility limit could be 15 wt % of PZ as reported by Bruder et al. [17], Dash et al. [47]. Based on all these findings, we suggest that the optimum solvent concentration should be 20–25 wt% AMP + 10–15 wt% PZ) for an energy efficient CO2 capture process.

9 Conclusions In this study, analysis of various important properties of CO2 capture solvent AMP + PZ has been presented. From the critical analysis of the present study and comprehensive literature review on this CO2 capture process, it is confirmed that this solvent has all the desirable properties such as moderate density and viscosity, relatively fast kinetics, high equilibrium solubility and better regeneration characteristics, all of which lead to energy efficient CO2 capture. Besides, energy efficiency is also related to the other essential properties of this AMP + PZ solvent system such as low or moderate corrosiveness, higher degradation resistance, lower undesirable environmental emissions and relatively lower regeneration energy requirement. It is evident from the forgoing discussion that this solvent is suitable

214

S.K. Dash

for CO2 capture from low partial pressures gas streams. From the various pilot plant tests and the simulation data, the regeneration energy requirement been found to be *2.9–3.7 GJ/tCO2 for a solvent composition of aq. (18–25 wt% AMP + 10–15 wt % PZ) for *85–90% CO2 capture. This energy requirement is about 20–30% less than the energy requirement of the MEA process. Although pilot plant study and simulation work provide good information on the heat duty, it is desirable to check the energy demand of this PCC process with large‐ scale pilot plant test and demonstration plant test. The amount of actual energy saving in various process configurations such as vapour recompression should be confirmed by tests in a demonstration plant with efficient random or structured packing in the column with applicable column stage efficiencies. A comprehensive study of combination of one model for the power plant including coal combustion, steam cycle etc., with the another one for CO2 capture should be useful for estimation of total energy demand as well as for cost optimization utilizing the low pressure steam of the power to supply the required energy for the regenerator.

References 1. Dash SK, Bandyopadhyay SS (2016) Studies on the effect of addition of piperazine and sulfolane into aqueous solution of N-methyldiethanolamine for CO2 capture and VLE modelling using enrtl equation. Int J Greenhouse Gas Control 44:227–237 2. Dash SK, Bandyopadhyay SS (2013) Carbon dioxide capture: absorption of carbon dioxide in piperazine activated concentrated aqueous 2-amino-2-methyl-1-propanol. J Clean Energy Technol 1(3):184–188 3. Khan AA, Haldera GN, Saha AK (2016) Experimental investigation of sorption characteristics of capturing carbon dioxide into piperazine activated aqueous 2-amino-2-methyl-1-propanol solution in a packed column. Int J Greenhouse Gas Control 44:217–226 4. Hairul NAH, Shariff AM, Bustam MA (2016) Mass transfer performance of 2-amino-2-methyl-1-propanol and piperazine promoted 2-amino-2-methyl-1-propanol blended solvent in high pressure CO2 absorption. Int J Greenhouse Gas Control 49:121–127 5. Budzianowski WM (2015) Single solvents, solvent blends, and advanced solvent systems in CO2 capture by absorption: a review. Int. J. Global Warming 7(2) 6. Svendsen, H (2010) CO2 capture by solvents; possibilities and challenges. In: Proceedings of post-combustion CO2 capture workshop. Tufts European Center Talloires, France, 11–13 July 2010 7. Kishimoto S, Hirata T, Iijima M, Ohishi T, Higaki K, Mitchell R (2009) Current status of MHI’s CO2 recovery technology and optimization of CO2 recovery plant with a PC fired power plant. Energy Procedia 1(1):1091–1098 8. Knudsen JN, Jensen JN, Vilhelmsen PJ, Biede O (2009) Experience with CO2 capture from coal flue gas in pilot-scale: testing of different amine solvents. Energy Procedia 1:783–790 9. Jockenhoevel T, Schneider R, Rode H (2010) Validation of a second-generation post-combustion capture technology—results from POSTCAP pilot plant operation. In: Powergen Europe, 8–10 June 2010 10. Paul S, Mandal B (2006) Density and viscosity of aqueous solutions of (N-Methyldiethanolamine + Piperazine) and (2-Amino-2-methyl-1-propanol + Piperazine) from (288 to 333) K. J Chem Eng Data 51:808–1810


215

11. Sun WC, Yong CB, Li MH (2005) Kinetics of absorption of carbon dioxide into mixed aqueous solutions of 2-amino-2-methyl-1-propanol and piperazine. Chem Eng Sci 60:503– 516 12. Samanta A, Bandhyopadhyay SS (2006) Density and viscosity of aqueous solutions of piperazine and (2-Amino-2-methyl-1-propanol + Piperazine) from 298 to 333 K. J Chem Eng Data 51:467–470 13. Dash SK, Samanta A, Samanta AN, Bandyopadhyay SS (2011) Absorption of carbon dioxide in piperazine activated concentrated aqueous 2-amino-2-methyl-1-propanol solvent. Chem Eng Sci 66:3223–3233 14. Dash SK, Samanta A, Samanta AN, Bandyopadhyay SS (2011) Vapour liquid equilibria of carbon dioxide in dilute and concentrated aqueous solutions of piperazine at low to high pressure. Fluid Phase Equilib 300:145–154 15. Dash SK, Samanta AN, Bandyopadhyay SS (2011) (Vapour + liquid) equilibria (VLE) of CO2 in aqueous solutions of 2-amino-2-methyl-1-propanol: New data and modelling using enrtl-equation. J Chem Thermodyn 43:1278–1285 16. Dash SK, Samanta AN, Bandyopadhyay SS (2012) Experimental and theoretical investigation of solubility of carbon dioxide in concentrated aqueous solution of 2-amino-2-methyl-1-propanol and piperazine. J Chem Thermodyn 51:120–125 17. Bruder P, Grimstvedt A, Mejdell T, Svendsen HF (2011) CO2 capture into aqueous solutions of piperazine activated 2-amino-2-methyl-1-peopanol. Chem Eng Sci 66:6193–6198 18. Tong D, Maitland GC, Trusler M, Fennell PS (2013) Solubility of carbon dioxide in aqueous blends of 2-amino-2-methyl-1-propanol and piperazine. Chem Eng Sci. http://dx.doi.org/10. 1016.ces.2013.05.034 19. Hartono A, Saeed M, Ciftja AF, Svendsen HF (2013) Binary and ternary VLE of the 2-amino-2-methyl-1-propanol (AMP)/piperazine (Pz)/water system. Chem Eng Sci 91:151– 161 20. Haghtalab A, Eghbali H, Shojaeian A (2014) Experiment and modelling solubility of CO2 in aqueous solutions of Diisopropanolamine + 2-amino-2-methyl-1-propanol + Piperazine at high pressures. J Chem Thermodyn 71:71–83 21. Halim HNA, Shariff AM, Bustam MA (2015) High pressure CO2 absorption from natural gas using piperazine promoted 2-amino-2-methyl-1-propanol in a packed absorption column. Sep Purif Technol 152:87–93 22. Dash SK, Samanta AN, Bandyopadhyay SS (2011) Solubility of carbon dioxide in aqueous solution of 2-amino-2-methyl-1-propanol and piperazine. Fluid Phase Equilib 307:166–174 23. Bishnoi S, Rochelle GT (2000) Absorption of carbon dioxide into aqueous piperazine: reaction kinetics, mass transfer and solubility. Chem Eng Sci 55:5531–5543 24. Derks PWJ, Dijkstra HBS, Hogendoorn JA, Versteeg GF (2005) Solubility of carbon dioxide in aqueous piperazine solutions. AIChE J 51:2311–2327 25. Hillard M (2008) A predictive thermodynamics model for an aqueous blend of potassium carbonate, piperazine, and monoethanolamine for carbon dioxide. Ph.D. Thesis, The University of Texas at Austin 26. Bougie F, Iliuta MC (2012) Strerically hindered amine-based absorbents for the removal of CO2 from gas streams. J Chem Eng Eng Data 57:635–669 27. Tontiwachwuthikul P, Meisen A, choon JL (1991) Solubility of carbon dioxide in 2-amino-2-methyl-1-propanol solutions. J Chem Eng Data 36:130–133 28. Teng TT, Mather AE (1989) Solubility of H2S, CO2 and their mixtures in an AMP solution. Can J Chem Eng 67:846–850 29. Dash SK (2012) Carbon dioxide capture by Absorption in piperazine activated 2-Amino-2-methyl-1-propanol solvent. Ph.D. Thesis. Indian Institute of Technology, Kharagpur 30. Yang ZY, Soriano AN, Caparanga AR, Li MH (2010) Equilibrium solubility of carbon dioxide in (2-amino-2-methyl-1-peopanol + piperazine + water). J Chem Thermodyn 42:659–665

216

S.K. Dash

31. Xie Q, Aroonwilas A, Veawab A (2013) Measurement of Heat of CO2 Absorption into 2-Amino-2-methyl-1-propanol (AMP)/piperazine (PZ) blends using differential reaction calorimeter. Energy Procedia 37:826–833 32. Nguyen T, Hilliard M, Rochelle GT (2010) Amine Volatility in CO2 capture. Int J Greenhouse Gas Control 4:707–715 33. Khakharia P, Brachert L, Mertensc J, Anderlohr C, Huizinga A, Fernandez ES, Schallert B, Schaber K, Vlugt TJH, Goetheer E (2016) nderstanding aerosol based emissions in a post combustion CO2 capture process: Parameter testing and mechanisms. Int J Greenhouse Gas Control 34:63–74 34. Samanta A, Bandyopadhyay SS (2009) Absorption of carbon dioxide into aqueous solutions of piperazine activated 2-amino-2-methyl-1-propanol. Chem Eng Sci 64:1185–1194 35. Veawab A, Tontiwachwuthikul P, Bhole SD (1996) Corrosivity in 2-amino-2-methyl-1-peopanol (AMP)-CO2 system. Chem Eng Commun 144:65–71 36. Veawab A, Tontiwachwuthikul P, Bhole SD (1997) Studies of corrosion and corrosion control in a CO2-2-amino-2-methyl-1-peopanol (AMP) environment. Ind Eng Chem Res 36:264–269 37. Freeman SA, Dugas R, Wagener DV, Nguyen T, Rochelle GT (2010) Carbon dioxide capture with concentrated, aqueous piperazine. Int J Greenhouse Gas Control 4:119–124 38. Wang T, Jens K (2012) Oxidative degradation of aqueous 2-Amino-2-methyl-1-propanol solvent for post combustion CO2 capture. Ind Eng Chem Res 51:6529–6536 39. Wang T, Jens K-J (2014) Oxidative degradation of aqueous PZ solution and AMP/PZ for post-combustion carbon dioxide capture. Int J Greenhouse Gas Control 24:98–105 40. Mazari SA, Ali B, Jan BM, Saeed IM (2014) Degradation study of piperazine, its blends and structural analogs for CO2 capture: a review. Int J Greenhouse Gas Control 31:214–228 41. Mangalapally HP, Hasse H (2011) Pilot plant study of two new solvents for post combustion carbon dioxide capture by reactive absorption and comparison to monoethanolamine. Chem Eng Sci 66:5512–5522 42. Notz R, Mangalapally HP, Hasse H (2012) Post combustion CO2 capture by reactive absorption: pilot plant description and results of systematic studies with MEA. Int J Greenhouse Gas Control 6:84–112 43. Karimi M, Hillestad M, Svendsen HF (2011) Capital costs and energy considerations of different alternative stripper configurations for post combustion CO2 capture. Chem Eng Res Des 89:1229–1236 44. Karimi M, Hillestad M, Svendsen HF (2012) Investigation of the dynamic behavior of different stripper configurations for post-combustion CO2 capture. Int J Greenhouse Gas Control 7:230–239 45. Cousins A, Wardhaugh L, Feron P (2011) Preliminary analysis of process flow sheet modifications for energy efficient CO2 capture from flue gases using chemical absorption. Chem Eng Res Des 89:1237–1251 46. Artanto Y, Jansen J, Pearson P, Puxty G, Cottrell A, Meuleman E, Feron P (2014) Pilot-scale evaluation of AMP/PZ to capture CO2 from flue gas of an Australian brown coal-fired power station. Int J Greenhouse Gas Control 20:189–195 47. Dash SK, Samanta AN, Bandyopadhyay SS (2014) Simulation and parametric study of the post combustion CO2 capture using aqueous 2-amino-2-methyl-1-propanol and piperazine. Int J Greenhouse Gas Control 21:130–139 48. van der Spek M, Arendsen R, Ramirez A, Faaij A (2016) Model development and process simulation of postcombustion carbon capture technology with aqueous AMP/PZ solvent. Int J Greenhouse Gas Control 47:176–199

Energy Efficient Absorbents for Industry Promising Carbon Dioxide Capture Y.S. Yu, T.T. Zhang and Z.X. Zhang

Abstract There are growing concerns that carbon dioxide (CO2) emissions are contributing to global climate change. CO2 capture and sequestration system is an effective way to alleviate this phenomenon. Chemical absorption of CO2 is a mature and efficient way to capture CO2 from industrial flue gas. Amines are the most commonly discussed solvent, such as NH3 as inorganic solvent and monoethanolamine (MEA) as the typical alkanolamine. MEA aqueous solutions have been widely analyzed and achieved good performance. However, the energy cost for the amine regeneration remains great which barricades its widely industrial application. Several energy efficient absorbents are currently discussed from the perspective of the reaction kinetics, desorption efficiency and sensible heat consumption. These discussions show that the absorbents with higher reaction rate and mass transfer coefficient and advanced process can reduce the sensible heat (less consumption of absorbent) and reaction heat, which thus make them as the energy efficient absorbents. Nomenclature C D, Dm G H HS h I k km

Concentration of the amine, kmol/m3 Diffusion coefficient, m2/s Inert gas flow rate, kmol/m2/s Henry’s constant, MPa m3/kmol Henry’s constant of the solvent, MPa m3/kmol Van Krevelen coefficient, m3/kmol Ionic strength of the solution, kmol/m3 Reaction rate constant, m3/kmol/s Reaction kinetics, dimensionless

Y.S. Yu T.T. Zhang Z.X. Zhang School of Chemical Engineering and Technology, Xi’an Jiaotong University, No. 28 Xianning West Road, 710049 Xi’an, People’s Republic of China T.T. Zhang Z.X. Zhang (&) State Key Laboratory of Multiphase Flow in Power Engineering, Xi’an Jiaotong University, 710049 Xi’an, People’s Republic of China e-mail: [email protected] © Springer International Publishing AG 2017 W.M. Budzianowski (ed.), Energy Efficient Solvents for CO2 Capture by Gas–Liquid Absorption, Green Energy and Technology, DOI 10.1007/978-3-319-47262-1_10

217

218

KGaV MSN P Re R T t yA,G YA,G Z a

Y.S. Yu et al.

Overall mass transfer coefficient, kmol/m2/s/MPa Molecular synergy number, dimensionless Operating pressure, MPa Reynolds number, dimensionless Gas constant, 8.314 J/mol/K Temperature, K Time, s Gas phase concentration, mol/mol Mole ratio value Column height, m Conversion rate, dimensionless

1 Introduction With the increasing amount of greenhouse gas emission in the atmosphere, the average temperature of the world rises during the past century, which has a serious impact on the environment of the earth. Therefore, it is becoming urgent to take effective measures to reduce carbon dioxide (CO2) emissions from its intensive emission source, such as coal fired power plants and cement plants, since CO2 is recognized as a typical greenhouse gas [1, 2]. Among all the methods to capture CO2, chemical absorption is considered to be an effective way to cover the large reduction of the greenhouse gas emission. During the chemical absorption, amines (alkanolamines) are the typical absorbents to capture CO2. However, there are still some problems for amines absorption of CO2, such as low CO2 absorption capacity and high energy consumption for CO2 desorption [3]. In order to solve these problems, improvements on the absorbent are proposed as one of the efficient methods. After reviewing the recent improvements on the absorbents for CO2 capturing, several kinds of absorbents are analyzed below, including mixed aqueous solutions, non-aqueous solutions and alternative amines. These categories are determined as the energy efficient absorbents, which are quite important for improving the performance of amine absorption of CO2. The provided detail information of these absorbents is believed to be helpful for the theoretical research and industrial application.

2 Blended Aqueous Energy Efficient Solutions The absorption of acid gases in mixed amines has outstanding advantages over the use of single amines. The addition of a small amount of primary amine to conventional tertiary amines can increase the rate of absorption of CO2 to a large extent without appreciably affecting the stripping characteristics [4, 5]. By varying the

Energy Efficient Absorbents for Industry …

219

relative concentrations of the amines, an optimum absorption system can, in principle, be designed for a specific application. Blends of primary and tertiary amines, such as MDEA + MEA + H2O, have been suggested for CO2 removal [5]. Compared to MDEA, AMP does not only have the same high CO2 loading capacity but also has a higher reaction rate constant for the reaction with CO2 [5–7]. Thus, AMP + MEA + H2O may be an attractive new solvent in addition to MDEA + MEA + H2O for the acid gases treating process [8]. Due to the increasing importance of blended-amine systems in acid-gas treating processes, it is necessary to understand the kinetic phenomena in mixed amine systems. The reaction kinetics normally determines the mass transfer coefficient and further determines the solution consumption amount, which has great impacts on the sensible heat consumption in CO2 desorption.

2.1

Reaction Types of CO2 in Aqueous Solutions

In aqueous amines solutions, the reactions between amines and CO2 mainly consist of two parts. The first reaction to be considered is the chemically process between H2O and CO2, þ CO2 + H2 O $ HCO 3 + H

ð1Þ

This reaction rate is very slow and may be usually neglected. The second reaction is the bicarbonate formation: CO2 + OH $ HCO 3

ð2Þ

This reaction rate is fast and can enhance mass transfer even when the concentration of hydroxyl ion is low. The second part is the reactions of CO2 with alkanolamines. Here, primary and secondary amines reacting with CO2 have the same reaction process, but the tertiary amines are different. The carbamate formation reaction occurs when CO2 reacts with primary and secondary alkanolamines, CO2 + 2R1 R2 NH $ R1 R2 NCOO + R1 R2 NH2þ

ð3Þ

where R1 is an alkyl and R2 is H for primary amines and an alkyl for secondary amines. The zwitterion mechanism is generally accepted as the reaction mechanism for the carbamate formation between CO2 with primary and secondary alkanolamines. The zwitterion mechanism has been used successfully in aqueous alkanolamine solutions as well as in some organic and viscous solutions [9, 10]. Besides the primary and secondary alkanolamines, the zwitterion mechanism was also found to

220

Y.S. Yu et al.

be suitable for modeling the absorption of CO2 into aqueous AMP solutions and into 1-propanol-AMP solvents [11, 12]. The reaction steps successively involve the formation of a zwitterion, CO2 + R1 R2 NH $ R1 R2 NH þ COO

ð4Þ

R1 R2 NH þ COO + B $ R1 R2 NCOO + BH þ

ð5Þ

where B is a base that could be an amines, OH−, or H2O, and the corresponding reactions are as follows, R1 R2 NH þ COO + R1 R2 NH $ R1 R2 NCOO + R1 R2 NH2þ

ð6Þ

R1 R2 NH þ COO + OH $ R1 R2 NCOO + H2 O

ð7Þ

R1 R2 NH þ COO + H2 O $ R1 R2 NCOO + H3 O þ

ð8Þ

For the reaction of CO2 with tertiary alkanolamines(R3N) in aqueous solution, as no N–H occurs in the amine structure, tertiary amines cannot react directly with CO2 [7, 13]. The following reaction mechanism was proposed [14], CO2 + R3 N + H2 O $ R3 NH þ + HCO 3

ð9Þ

which is a based-catalyzed hydration of CO2. In multi-amine solutions, with the different mixing rules, the reaction process could be a combination of all the involved possible reactions.

2.2 2.2.1

Energy Efficient Amine Solvents Binary Amine Solutions

The binary amine solutions are the most discussed mixtures for CO2 absorption, such as MEA, DEA based amines solutions mixed with MDEA, AMP. In these binary amine solutions, MEA and DEA were usually used as promoters to enhance the absorption rate of CO2 while AMP and MDEA were used to reduce the energy consumption for solvent regeneration [15–18]. In the researches above, the basic Henry constant H is defined as log10

H HS

¼ hI

ð10Þ


221

The diffusion coefficient D is simply offered as the function of the temperature, which is D ¼ expð13:275 2198:3=T 0:078142CÞ

ð11Þ

By utilizing the parameters above, the fundamental absorption performance can be assessed correctly with the help of reaction kinetics and viscosity parameters [6]. The performance of blended solutions was ever examined at a total alkanolamine concentration of 3.0 kmol/m3, a mixing ratio of 1:1 mol/mol and three different CO2 loadings (0, 0.25 and 0.4 mol/mol) [19]. The results show apparently that the absorption performances of blended solutions are generally between the performances of their separate component. When CO2 loading increased, the CO2 concentration of the blended solutions shifted upward, indicating a lower CO2 absorption efficiency. For MEA-MDEA and DEA-MDEA blend solutions, the performance of blended solutions approached the performance of the rate promoters (MEA and DEA) at a low CO2 concentration. With the increasing of CO2 loading, the performance of blended MEA-MDEA and DEA-MDEA solutions moves to the performance of MDEA. These results show that at low CO2 loading, the rate promoters (MEA and DEA) played a dominant role because they reacted with CO2 faster than MDEA to form very stable carbamate compounds. As CO2 loading was increased, more MEA or DEA absorbed CO2 to form carbamate, leading to the increase in the ratio of unreacted MDEA to unreacted promoter. Subsequently, the unreacted MDEA amount became the dominant factor for the CO2 absorption rate. Besides MEA and DEA, piperazine (PZ), diglycolamine (DGA), aminoethylethanolamine (AEEA), 3-methylaminopropylamine (MAPA), diethylenetriamine (DETA), triethylenetetramine (TETA), and tetraethylenepentamine (TEPA) are also used as activators to improve CO2 absorption performance of aqueous MDEA solutions [20–24]. Take the last four alkyl amines with multiple amine groups for example, their blends with MDEA have higher CO2 absorption capacities compared with that of aqueous solutions of MEA [30% (w/w)] and MDEA [30% (w/w)]. MDEA/TEPA shows the highest CO2 loading amount of 0.753 mol/mol at 313 K. Also, these MDEA blends show high cyclic capacities (0.241−0.330 mol/mol), which are about 3 times higher than that of MEA. All MDEA blends show higher absorption fluxes than MDEA. Some blends (MDEA/MAPA) showed about 8 times higher overall mass transfer coefficient than that of MDEA and even higher than that of MEA. In addition, the absorption heats of the MDEA blends are about 30% higher than that of MDEA and about 30% lower than that of MEA [22]. However, scarce studies are performed for triethanolamine (TEA) blended amines. As known to all, TEA is a kind of tertiary amine, which has advantages in absorption capacity and desorption efficiency compared with MEA and DEA. Thus, TEA blended amine is considered as an energy efficient solvent for CO2 capture. Its drawback is the low absorption rate. Based on the principle of balancing the absorption rate and desorption energy, MEA, PZ, AEEA, AEP (N-aminoethyl piperazine) with high absorption rates, are here used to mix with TEA respectively to

222

Y.S. Yu et al.

produce new binary solutions. This will probably offer an alternative energy efficient solvent for CO2 capture. In order to assess the TEA blended amines, absorption and desorption experiments are performed. The experiment is set up in Figs. 1 and 2. As shown in Fig. 1, flue gas is firstly sent to the bottom of the absorption column at a constant flow rate. After the countercurrent absorption process, the rich solution gathered from the bottom is pumped to the top of the absorption column. The CO2 volume fraction in exhaust gas is monitored by an infrared gas analyzer and recorded every minute till the outlet concentration becomes constant. The liquid flow rate is 0.575 L/min. The absorption column diameter is 80 mm and the packing is plastic Raschig rings and the packing height is 280 mm. The accuracy of the infrared gas analyzer is 0.1%. Desorption apparatus is shown in Fig. 2. After absorption, the produced rich solution is desorbed in a conical flask atmospherically in an oil bath. CO2 generated after desorption is dried and then measured by Drainage method. At the same time, a certain amount of absorbed rich liquid is desorbed by strong acid and then measured by Drainage method. The results are compared to calculate the desorption

1. Simulated flue gas 2. Pressure release valve 3. Flow meter 4. Packed column 5. Pump 6. Infrared gas analyzer Fig. 1 CO2 absorption setup

Fig. 2 CO2 desorption setup

1. Thermostatic oil bath 3. Wild-mouth bottle

2. Drying equipment 4. Cylinder


223

ratio. In general, the CO2 fraction in industrial flue gas is 10–13%. Here, it is set as 13% and the flue gas is made up of N2 and CO2 only, which is defined as the simulated flue gas. By keeping the total amine concentration as 3 mol/L, the absorption performances of the mixed solutions such as 3 mol/L TEA, 2 mol/L TEA+1 mol/L MEA, 2 mol/L TEA+1 mol/L AEP, 2 mol/L TEA+1 mol/L AEEA and 2 mol/L TEA+1 mol/L PZ are discussed and compared. The absorption rate of the mixed solutions changes with time in Fig. 3 and the CO2 loading varies with time in Fig. 4.

Fig. 3 Absorption rates of different mixed solutions

Fig. 4 Solution loadings of different mixed solutions

224

Y.S. Yu et al.

As shown in Fig. 3, the absorption rates of the TEA blended amines show great differences. At the beginning, the absorption rate of the blended TEA amines are 3 to 5 times of TEA solution, which verifies that adding additional amine does greatly improve TEA absorption rate. This is due to the fact that the primary amine groups, secondary amine groups have a better absorption property than tertiary amine groups. Among these additives, AEP is the most effective one, which helps TEA to obtain a high absorption rate for a long time. For the other discussed amines, the absorption rate follows the order of PZ > AEEA > MEA. TEA blended with MEA solution firstly absorbs CO2 at a high rate. In 33 min, the absorption rate becomes lower than TEA solution. 50 min later, the absorption rate remains almost constant. The initial absorption rate of TEA blended with AEEA is better than that blended with MEA and PZ. Adding PZ into TEA solution achieves a great absorption rate initially, which however decreases rapidly versus time. Even though the absorption rate of the blended TEA amines decreases versus time, it is still higher than single MEA and single TEA solutions. Figure 4 shows that, with the increasing absorption time, the loading of TEA blended amines increases accordingly and shows a significant difference. At the first 15 min, the loadings of the four kinds of TEA blended amines are quite close, which are much higher than that of the 3 mol/L TEA solution. Thereafter, with the decrease of the absorption rate, TEA and MEA blended amines tends to saturate and thus no further increases in the loading. However, the loadings of the other TEA blended amines still increase, corresponding to the high absorption rates. The loading of TEA blended with AEP is the highest one due to fast absorption occurring. The loading of the TEA blended with AEEA solution is inferior to TEA blended with AEP. More importantly, after adding the AEP, the loading of the TEA solutions has been improved significantly. After the absorption experiment, a portion of the rich solution was heated to 120° in the oil bath until CO2 is totally desorbed. Then, the 50 mL rich solution and 50 mL lean solution are desorbed by using a strong acid. The corresponding product is measured by Drainage method. The desorption results are offered in Table 1. As can be seen in Table 1, among the TEA blended amine solutions, the TEA blended with AEP shows the 12% higher desorption efficiency than the TEA solutions. However, the other TEA blended amines shows lower desorption efficiency comparing with TEA solution. This is because the AEP itself having tertiary Table 1 Desorption of TEA based solutions Composition

Desorption amount in 50 mL rich solution/mL

Desorption amount in 50 mL poor solution/mL

Desorption efficiency/%

3 mol/L TEA TEA/MEA = 2:1 TEA/AEP = 2:1 TEA/AEEA = 2:1 TEA/PZ = 2:1

414 910 1280 1083 820

120 370 260 380 350

71.01 59.34 79.69 64.91 57.32


225

amine groups and thus it is easier to be desorbed than the rest. Therefore, considering absorption rate and desorption efficiency, TEA + AEP blended amine is taken as a promising energy efficient solvent for CO2 capture. It is necessary to determine the optimal composition for TEA + AEP blended amine. By keeping the total amine concentration as 3 mol/L and changing the TEA and AEP concentration, the absorption performance of the TEA + AEP blended amine solutions such as 3 mol/L AEP, 1 mol/L TEA+2 mol/L AEP, 1.5 mol/L TEA+1.5 mol/L AEP, 2 mol/L TEA+1 mol/L AEP and 2.5 mol/L TEA +0.5 mol/L AEP are compared. The absorption rates results are presented in Fig. 5 and the CO2 loadings results are offered in Fig. 6. As clearly observed in Figs. 5 and 6, with the increasing amount of AEP from 0.5 mol/L to 3 mol/L, the absorption rate of the TEA + AEP blended amine has shown a decrease tendency. CO2 loading reaches a maximum at 1.5 mol/L. For the desorption of CO2 + TEA + AEP, the rich solution is heated in the oil bath at 120 °C. Simultaneously, the 50 mL rich solution and 50 mL lean solution are desorbed by using a strong acid. The results are summarized in Table 2. As clearly observed in Table 2, TEA + AEP blended amine with TEA/AEP ratio of 1:1 shows the largest desorption efficiency of 82.26%, which shows great promise to reduce the energy consumption for CO2 capture. All the results show that binary amines solutions can be used as alternatives to MEA, achieving high CO2 loading and high reaction kinetics with low heat of reaction. These would allow reducing the solvent flow rate and solvent regeneration energy in the CO2 capture process, which strongly proves that binary amines are the

Fig. 5 Absorption rates of the TEA + AEP blended amine

226

Y.S. Yu et al.

Fig. 6 Solution loadings of the TEA + AEP blended amine

Table 2 Desorption of TEA + AEP blended amine Composition TEA/AEP TEA/AEP TEA/AEP TEA/AEP TEA/AEP

= = = = =

2:1 1:1 5:1 1:2 0:3

Desorption amount in 50 mLrich solution/mL

Desorption amount in 50 mL poor solution/mL

Desorption efficiency/%

1280 1240 1060 1090 950

260 220 240 354 331

79.69 82.26 77.36 67.52 65.16

energy efficient absorbents. However, the high viscosity and high corrosiveness of amines with four and five amino groups such as TETA and TEPA have to be considered seriously.

2.2.2

Ternary and More Energy Efficient Amine Solutions

Upon the binary amine mixing solutions, few works were done on ternary or more kinds of amine solutions [25–29]. Adding AMP into a mixture of MDEA and DEA is an attempt to produce ternary amine solution for CO2 capture. For ternary or more complicated mixture amine solutions, experimental method is difficult to know the complex physical and chemical properties due to high uncertainties under multi-variable condition. Molecular dynamics (MD), as a mature computational


227

technology, could be used instead as an alternative method after validated models are developed properly. Considering the integration of field synergy theory and molecular dynamics theory, a synergy molecular dynamics (SMD) model was developed to describe multiple amine mixtures. The SMD model is able to provide the transport properties of complicated mixture amines for CO2 capture. In the SMD model, the molecular synergy number (MSN) was used to quantify the interactions between diffusion and molecule motion, which are usually hard to be obtained by experiments. In the SMD model, the route is to deduce the transport properties by the key parameter of total energy. Here, the total energy of the amines and CO2 system is considered as the sum of potential energy and kinetic energy. The potential energy is comprised of four parts, i.e., bond stretching, angle bending, out of plane bending, and dihedral motion potentials. Moreover, Lennard-Jones potentials are determined to characterize the van der Waals interactions and electrostatic interactions. The kinetic energies are calculated by the classical velocity computational method. After potential and kinetic energies are obtained correctly, the diffusivity is achieved by Green Kubo equation. The transfer characteristic of viscosity is thus induced by its correlation to pressure tensors. In order to achieve the optimization index for mixing amines, the molecular synergy number (MSN) is developed by referencing the synergy number ever used in the multi-field synergy analysis, which is given by MSN =

Dm 2km Dm0 15Re4=3

ð12Þ

More detailed information is given in Yu’s work [29]. The MSN is able to help with selection of the energy efficient solvent since the smaller MSN corresponding to the less energy consumption for desorption. The sensible heat consumption in the desorption process is affected by the CO2 loading and mass transfer coefficient. The higher CO2 loading and mass transfer coefficient will normally requires less solution flux and thus produce less sensible heat consumption in desorption. According to the mass transfer coefficient results of the blended amines, the sensible heat consumption of the blended amines are assessed accordingly, which helps to determine which blended amines are energy efficient or not to some extent. The sensible heat consumption amount for CO2-AMP desorption is set as the baseline, which is represented by E0. The ratio of the sensible heat consumption amount of CO2 and blended amines over CO2-AMP is presented by E/E0. Here, MEA-DEA-AMP, MEA-DEA-TEA, MDEA-DEA-AMP and MDEADEA-TEA with different weight fraction ratios are selected as the discussed ternary systems as shown in Figs. 7, 8 and 9. With the analysis of molecular synergy number (MSN), it is clearly seen that that the mixture of MDEA, DEA and TEA with a MSN being 3.89 (the smallest one) is the best synergy one of the discussed combinations.

228

Y.S. Yu et al.

Fig. 7 Field synergy effects and sensible heat consumption of ternary amines mixtures

Fig. 8 Field synergy effects and sensible heat consumption of quaternary amine mixtures

This mixture achieves the best synergy between diffusion and molecule motion and less drag force with mass transfer coefficient being 33.33% higher than that of AMP. For the quaternary amines comprising of MEA-MDEA-DEA-AMP and MEA-MDEA-DEA-TEA and quintuple amine solution of MEA-MDEADEA-AMP-TEA, the mass transfer coefficient are all higher than that of referenced AMP system. This proves that mixed amines improve the CO2 absorption


229

Fig. 9 Field synergy effects and sensible heat consumption of quintuple amine mixtures

performance from the mass transfer view of point. Also, the overall synergy in ternary amine systems has been found better than that in quaternary and quintuple amine systems. All the results could provide a benchmark for the amine mixing solution design with better CO2 capture. As shown in Fig. 7, the MEA-DEA-TEA is an energy efficient solvent since its sensible heat consumption amount is 40% lower than that of the AMP baseline. This is due to the fact that higher mass transfer coefficient of MEA-DEA-TEA produces much less solvent consumption amount by absorbing equal CO2. For the quaternary amine, the similar results are found that the sensible heat consumption for desorption of CO2-MEA-MDEA-DEA-TEA is 16–27% lower than that of the AMP baseline in Fig. 8. For the quintuple amine MEA-MDEA-DEA-AMP-TEA and CO2 system, the sensible heat consumption amount decreases to 68–79% as shown in Fig. 9. All the results prove that blended amines are energy efficient solvents for CO2 capture. The sensible heat consumption can be qualitatively assessed by the MSN since they have similar variation tendency versus the amine weight fraction as shown in Figs. 7, 8 and 9. According to the discussion above, the mixed amine solutions have the potential to perform better than the single alkanolamines as long as mixing species and weight ratios are well-designed. The higher mass transfer coefficient of the mixed amine solutions offers the less consumption of the absorbent amount, which could reduce the sensible heat consumption in the desorption process. Hence, the mixed amine solutions are energy efficient absorbents for CO2 capture.

230

Y.S. Yu et al.

3 Non-aqueous Energy Efficient Solutions Apart from aqueous alkanolamine solutions for CO2 removal from gas produced by burning fossil fuels, non-aqueous systems comprising methanol solutions of alkanolamines have been employed commercially for the absorption of acid gases (CO2, H2S, COS) for the high solubility and capacity, low corrosiveness, and low energy consumption during generation of the used liquor. Methanol is widely used as a physical solvent for CO2 removal from gas streams, and the solubility of CO2 in aqueous monoethanolamine can be enhanced by the presence of methanol, which was 25% higher in a MEA and methanol mixture than in an aqueous MEA solution of equivalent MEA concentration. Mixed solvents (chemical and physical) are expected to have a higher capacity for the acid gases (CO2) over a wide range of partial pressures than the separate physical or chemical solvent. Meanwhile, non-aqueous solvent such as methanol has lower boiling point and smaller latent heat which leads to lower operating temperature in the regeneration process and thus requires a lower regeneration energy amount compared with aqueous environment. Hence, the non-aqueous alkanolamines have potentials to improve the performance of CO2 absorption.

3.1

Reaction Mechanism

The Eqs. (3), (4) and (5) show the overall reaction between CO2 and alkanolamines of primary and secondary in aqueous solution [30–33]. In case of non-aqueous solvents, only amine is considered as the base in the proton removal step. CO2 + R1 R2 NH $ R1 R2 NH þ COO

ð13Þ

R1 R2 NH þ COO + R1 R2 NH $ R1 R2 NCOO þ R1 R2 NH2þ

ð14Þ

For the tertiary amines, since they do not form carbamates and act as homogeneous base catalysts for carbon dioxide hydrolysis, the reaction Eq. (4) cannot be used to explain the reaction mechanism between CO2 and TEA/MDEA in non-aqueous solution [9, 34]. It was proposed that in non-aqueous solutions of tertiary amine the dissolved carbon dioxide will react with solvated tertiary amine to form anion pair as follows: R3 NH HSOL þ CO2 ! R3 NH þ CO2 SOL

ð15Þ

where HSOL represents the solvent. Experiment works on reaction kinetics of CO2 absorption in non-aqueous alkanolamines, such as monoethanolamine (MEA) and triethanolamine (TEA) in methanol, n-propanol and ethylene glycol with different weight mixing ratio were done at 298 K. The reaction orders and rate constants were achieved, and the


231

Table 3 The reaction orders and rate constants of CO2 in the reaction of CO2 with non-aqueous MEA at 298 K [34] Solvent

n

K1

K2/K3

water methanol ethanol n-propanol n-butanol Ethylene glycol Propylene glycol Propylene carbonate

1.53 1.79 (1.62) 1.85 (1.62) 1.85 (1.90) 1.86 1.82 1.85 1.83

6312 (7740) 5631 (8330) 5479 5361 5408 5731 5536 5621

1.73 (1.28) 7.41 (1.28) 9.65 (2.91) 11.23 (3.66) 11.88 7.24 10.14 8.79

experimental results of MEA in non-aqueous solutions absorbing CO2 are shown in Table 3[34–37]. The reaction orders of the solvents depended on the polarity of the solvent, which have some relationship with the dielectric constant of the solvent. For the reaction rate constant, linear relationship with the solubility parameter of the solvent was obtained in several experiments. The magnitude of the rate constants may be a function of the degree to which the solvent is able to stabilize the zwitterionic intermediate. Thus, the reaction of CO2 with MEA in polar solvents was identified as the reaction scheme above such as the zwitterionic mechanism. Besides experiment, as computational calculation developed, ab initio calculations were also employed to investigate the reaction mechanism of alkanolamines absorbing CO2 in gas phase and in aqueous and non-aqueous solvation effects [38, 39]. With different calculating methods and basis sets, the reaction energies were a little different from each other. The reaction pathways are almost consistent with the zwitterion mechanism, which are two steps reaction processes with zwitterion as intermediate, followed by a proton transfer process with another base (amine, H2O molecule). According to different bases, some of the second proton transfer processes are almost energy-free and one-step and three-molecular reaction process was also proposed. The reaction energy barriers show that the highest energy barrier appears in the water solvation effect. All the results above prove that non-aqueous solvents show great mass transfer coefficient from the reaction kinetics data and ab initio calculation data, which produces less solvent consumption in absorbing equal amount of CO2 compared with the MEA baseline. Thus, the non-aqueous solvent is another kind of energy efficient solvent in CO2 capture, which has great potential in industrial application.

3.2

Absorption Performance of the MEA-Methanol System

To compare the absorption performances between MEA aqueous and non-aqueous solutions, studies on CO2 absorption into three solvent solutions, 5 M MEA-methanol

232

Y.S. Yu et al.

solution, 5 M MEA in 1:1 water-methanol (volume ratio) and 5 M MEA aqueous solutions, were investigated in a packed column, offering overall mass transfer coefficient (KGav) and mass flux of CO2 absorption over ranges of methanol composition, CO2 loading, liquid flow rate and inert gas flow rate. In the film theory, by combing mass flux and material balance equations at steady-state condition, the overall mass transfer coefficient is defined as Eq. (16) [40, 41], K G av ¼

G PyA;G

!

dYA;G dZ

¼

G 1 + YA;G dYA;G P YCO2 dZ

ð16Þ

The results show that the mass transfer performance of 5 M MEA in methanol was higher than those of 5 M MEA in 1:1 water-methanol and 5 M MEA aqueous solutions, respectively, as the methanol has higher physical solubility and physical diffusivity of CO2 than water. The higher methanol composition leads to the higher physical solubility and physical diffusivity of CO2. For the CO2 loading, as CO2 loading increases in the range of 0.05–0.3 mol/mol, the mass transfer performance in terms of KGav decreases due to that the amount of active amine decreases as CO2 loading increases. Increasing liquid flow rate can lead to greater degree of wetted packing surface and increasing of mass transfer performance. However, this result is controversial since the bubbles will be produced at high liquid flow rates, which directly affects the active surface area between CO2 and absorbent. Thus, the optimum operating flow rate differs from the actual operating conditions. As the CO2 absorption process is liquid film control, the inert gas flow rate has no significant effect on mass transfer performance. Above all, the mass transfer performance was higher in MEA methanol solutions than aqueous solutions.

3.3

Multi-stage Energy Efficient Process for CO2MEA-Methanol Regeneration System

As methanol can enhance the mass transfer of MEA absorbing CO2 process and show the properties different from water, a new regeneration operation process specially designed for MEA-methanol system was proposed. It’s a multi-stage regeneration process which can be operated below 373 K [33]. For the MEA-methanol regeneration system, the main reaction considered and reaction rate constant at 343–363 K are shown as follows. MEAH þ + MEACOO ! 2MEA + CO2 k = exp( 84:478 þ

28642:44 Þ T

ð17Þ ð18Þ


233

In the operation process, to demonstrate the CO2 desorption performance in the absence of water, three cases using N2, ethanol vapor and steam as the purge gases were analyzed for the same amount of lean solution. The results show that the operating conditions (temperature, pressure), lean solvent loadings, gas/liquid ratio, packing and internal all influence the regenerating performance. A parametric analysis found that the energy consumption could be reduced by 7−24%. A minimum energy consumption of 2.28 GJ/t was identified in the non-aqueous process. It was clearly found that increasing the temperature, pressure and lean solvent loading could reduce the energy consumption by 21, 22 and 20%, respectively. Regarding the effect of the packing, it is noted that a random packing was found to consume less energy than a structured packing. The CMR-2 packing produced the lowest energy consumption regardless of the desorption conditions. The internals placed in the stripper intensified the desorption process, with energy consumption being reduced by about 23%. The non-aqueous solvent could improve the desorption efficiency by 10% compared to that obtained in a typical aqueous solution desorption. Apart from the typical non-aqueous solution of MEA in methanol, other discussed non-aqueous solutions for CO2 capture include MDEA/diisopropanolamine (DIPA)/triethanolamine (TEA) with polar organic solvents, MEA/triethylene glycol (TEG), AMP and some AMP–alkanolamine blends (IPMEA, TBMEA) in non-aqueous solvents (triethylene glycol (TEG), diethylene glycol (DEG), ethylene glycol and 1-propanol), 2-(2-aminoethylamine)ethanol (AEEA)+benzylalcohol and so on [36, 42–44]. These non-aqueous capture systems were also under discussion for their absorbing characteristics, which show the good performance being operated at low stripping temperature (under 373 K), which have the potential to reduce the regeneration energies. However, more works needed to be done to determine the optimum operating conditions, and to avoid the corrosion and degradations of the proposed solutions. Above all, compared with traditional aqueous alkanolamine solutions, the non-aqueous solutions could do a better work on CO2 absorption. For the regeneration process, they can lower the regeneration temperature, which could greatly lead to the reduction of energy cost in the regeneration process. Therefore, non-aqueous solutions are energy efficient absorbents to capture CO2.

4 Alternative Energy Efficient Solvent In order to find out the alternatives to traditional alkanolamines, several other new solvents or special mixtures are developed in capturing CO2. The developed ones offer the large capacity and low energy consumption and thus show a wide application potential. Recently, tetramethylammonium hydroxide (TAMH) has been already identified as an alternative solvent to capture CO2 for the advantages of using its CO2 absorption product for photoresist and etch residue removal. Hence, it is interesting to do research on using TAMH to solve large CO2 emission

234

Y.S. Yu et al.

and utilization problem to some extent. However, the limitation of using TAMH is that some tetramethylammonium carbamates are easy to be precipitated which barricades the residue removal. To avoid the shortcoming, several physical solvents are proposed to dissolve the tetramethylammonium carbamates. Here, the typical physical solvent is selected as the tetramethylene sulfone (TMS), which shows great solubility for amine and CO2. Additionally, ethylene glycol (EG) is considered as another typical solvent for physical absorption of CO2 [45–47]. More importantly, TMS and EG show great ability to provide good solubility for tetramethylammonium carbamates, which affords to solve the precipitation problem for TAMH absorption of CO2. Thus, TAMH-TMS-EG aqueous solution was suggested as a new solvent mixture for CO2 capture [48].

4.1

Reaction Mechanism

As TAMH reacts with CO2 to form carbonate and TMS-EG provides the strong dissolution of tetramethylammonium carbonate, it is difficult for the tetramethylammonium carbonate to exist in the TAMH-TMS-ETG-CO2 system. Therefore, the main reaction type is reasonably summarized as the formation of carbonate between TAMH and CO2, shown as follows. TAMH + CO2 ! TAM þ HCO 3

ð19Þ

Based on the reaction equation, the reaction kinetics is accurately determined as da ¼ 8:975 1035 e200590=RT ð1aÞ0:91 dt

ð20Þ

The precise reaction kinetics determines the reaction rate of CO2TAMH-TMS-EG system, which is greater than the reaction rates of typical aqueous CO2-MEA and CO2-MDEA [49]. Thus, greater reaction rate of CO2TAMH-TMS-EG provides the strong base for increasing the mass transfer coefficient, which suggests that TAMH-TMS-EG is suitable for CO2 absorption.

4.2

Absorption Performance and Sensible Heat Consumption

Since the TAMH-TMS-EG is a new solvent, its CO2 absorption performance is determined precisely. CO2 solubility in the TAMH-TMS-EG solutions, as the basic mass transfer parameter, is carefully tested and discussed. It is concluded that the 2–25 kPa partial pressure of CO2 produces the corresponding CO2 loading of 0.15–0.453 mol/molTAMH respectively at 313.15 K. This CO2 loading is


235

comparable to that in the typical MEA system, which proves that the solubility of CO2 in TAMH-TMS-EG is high enough for CO2 absorption. At 373.15 K, the partial pressure of CO2 varies from 100 to 675 kPa, which offers the CO2 loading of 0.18–0.5 mol/molTAMH. This average CO2 partial pressure is a little less than the partial pressure of CO2 in MDEA-TMS aqueous system and suggests that it is supposed to strip CO2 more easily in TAMH-TMS-EG solution. This is quite useful to achieve the low cost CO2 capture process. During the CO2 absorption operation assessment, CO2 loading is the key parameter to achieve the continuous run. CO2 fraction, solvent composition (TAMH fraction and TMS fraction) and pressure normally influence CO2 loading in the solvent. Therefore, the effects of these factors were discussed in detail here. In order to explain the energy efficient advantage of the TAMH-TMS-EG solutions, the sensible heat consumption amount of CO2-MEA is set as the baseline. The ratio of the sensible heat consumption amount of CO2-TAMH-TMS-EG desorption over CO2-AMP desorption is presented by E/E0. CO2 fraction in flue gas is normally hard to keep constant with the dynamic operating conditions in industrial process. Here, the CO2 mol fraction is set by 5–25%, which are the common states in the industrial flue gas. The results in Fig. 10 provide that the sensible heat consumption ratio E/E0 apparently decreases as the CO2 mol fraction increases at 303.15, 308.15 and 313.15 K, respectively. This offers that the sensible heat consumption of CO2-TAMH-TMS-EG desorption

Fig. 10 Effects of CO2 fraction on sensible heat consumption

236

Y.S. Yu et al.

is 25–45% lower than that of the CO2-MEA desorption in the typical 10–15% CO2 mol fraction in the power plant, which strongly supports that TAMH-TMS-EG is an energy efficient solvent. This is attributed to the fact that TAMH-TMS-EG absorbs more CO2 than MEA under equal solution flux conditions. Taking the solvent composition into account, the weight fractions of TAMH and TMS is supposed to influence CO2 loading, which are depicted in Figs. 11 and 12. It is shown that the sensible heat consumption ratio E/E0 decreases from 0.85 to 0.45 as TAMH weight fraction increases from 5 to 25%. This interesting conclusion is due to the fact that TAMH is involved in the chemical reaction with CO2, which results in that the redundant TAMH naturally absorbs more CO2. This will certainly consume much less solvent compared with MEA case and thus produce less sensible heat consumption amount. In this sense, the sensible heat consumption ratio E/E0 is decreased by 15% when the temperature increases from 303.15 to 313.15 K. This is partially due to the exothermic heat occurring in the TAMH absorption of CO2 process. Another composite TMS, one of the physical solvents, also has great influence on the sensible heat consumption ratio E/E0. Its effects on sensible heat consumption ratio E/E0 is provided in Fig. 12. As expected, sensible heat consumption ratio E/E0 increases from 0.36 to 0.43 as TMS weight fraction increases from 55 to 75%. The meaningful result is attributed to the fact that the absorption capacity of the physical solvent TMS is not so strong as that of chemical solvent of TAMH.

Fig. 11 Effects of TAMH fraction on sensible heat consumption


237

Fig. 12 Effects of TMS fraction on sensible heat consumption

This is further validated by that the CO2 loading is averagely decreased by 24% as temperature ascends from 303.15 to 313.15 K. After the chemical reaction and dissolution are considered to affect the TAMH-TMS-ETG absorption of CO2, diffusion is believed to be another important factor that plays great role in the CO2 absorption. Thus, the mass transfer resistance able to achieve the goal of the further solvent composition optimization is studied as follows. According to the previously correct findings, the Hatta number is here used as a variable to determine the solvent chemical property. Also, a dimensionless solubility is employed to indicate the ratio of the CO2 concentration between liquid phase and gas phase at equilibrium state. It is quite clear that the Hatta number is below 3 and agrees with the literature results [49] for CO2 absorption in the solvents. The most interesting one is that the minimum mass transfer resistance corresponds to the TMS weight fraction from 40 to 80%. The TMS concentration range produces the higher Hatta number accordingly. Additionally, this TMS weight fraction range offers the higher CO2 solubility above 1. All the results may help well with the future TAMH-TMS-ETG solvent optimization as it is used in industry to mitigate the CO2 emission. Since the energy consumption for desorption of the solutions and CO2 determines the CO2 capture cost, energy consumption comparisons with typical MEA-CO2-H2O and MEA-CO2-TMS are made to determine the overall CO2 capture performance of TAMH-TMS-EG. It is suggested that the average CO2

238

Y.S. Yu et al.

loading is 25% higher in TAMH-TMS-EG than that in MEA-CO2-H2O and MEA-CO2-TMS. Moreover, the energy consumption is determined as 1.11–1.34 GJ/t for TAMH-TMS-EG absorption of CO2. This energy cost is much less than the typical 3.0–4.5 GJ/t in MEA solution absorption of CO2. It is believed that the molecules of TMS have a higher dielectric constant (about 82.56 at 303.15 K) compared with water. Thus, TMS becomes a more active hydrogen atom donor and accept or rather than keep intermolecular relations. All in all, the TAMH-TMS-EG was proven as an energy efficient solvent candidate for effective CO2 capture.

5 Conclusions In order to reduce the energy consumption, energy efficient solvents are summarized and analyzed in details. Reaction kinetics, desorption efficiency and sensible heat consumption are discussed since these parameters determine the energy consumption amount directly. The proper composition and operating conditions for binary amines, ternary amines and quaternary amines, non-aqueous amine and alternative solvents being the energy efficient solvents are determined. The higher CO2 loading and mass transfer coefficient of these solvents help to obtain much less sensible heat consumption compared with typical MEA baseline, which strongly support their energy efficient characteristic and shows great economical advantages in the industrial application for CO2 capture. Acknowledgments Financial support of National Natural Science Foundation of China (no. 51276141) is gratefully acknowledged. This work is also supported by the Natural Science Basic Research Plan in Shaanxi Province of China (No. 2015JQ5192) and “Fundamental Research Funds for the Central Universities”.

References 1. Mathias PM, Zheng F, Heldebrant DJ, Zwoster A, Whyatt G, Freeman CM et al (2015) Measuring the absorption rate of CO2 in nonaqueous CO2-binding organic liquid solvents with a wetted-wall apparatus. Chem Sus Chem 8:3617–3625 2. Walters MS, Edgar TF, Rochelle GT (2016) Regulatory control of amine scrubbing for CO2 capture from power plants. Ind Eng Chem Res 55:4646–4657 3. Singto S, Supap T, Idem R, Tontiwachwuthikul P, Tantayanon S, Al-Marri MJ et al (2016) Synthesis of new amines for enhanced carbon dioxide (CO2) capture performance: The effect of chemical structure on equilibrium solubility, cyclic capacity, kinetics of absorption and regeneration, and heats of absorption and regeneration. Sep Purif Technol 167:97–107 4. Abu-Zahra MR, Abbas Z, Singh P, Feron P (2013) Carbon dioxide post-combustion capture: solvent technologies overview, status and future directions. Materials and processes for energy: communicating current research and technological developments. Formatex Research Center, Badajoz, pp 923–34 5. Chakravarty T, Phukan U, Weilund R (1985) Reaction of acid gases with mixtures of amines. Chem Eng Prog (U.S) 81:32–36


239

6. Budzianowski WM (2015) Single solvents, solvent blends, and advanced solvent systems in CO2 capture by absorption: a review. Int J Glob Warming 7:184–225 7. Crooks JE, Donnellan JP (1990) Kinetics of the reaction between carbon dioxide and tertiary amines. J Organic Chem 55:1372–1374 8. Li M-H, Chang B-C (1994) Solubilities of carbon dioxide in Water + Monoethanol amine + 2-Amino-2-methyl-1-propanol. J Chem Eng Data 39:448–452 9. Versteeg G, Van Dijck L, Van Swaaij W (1996) On the kinetics between CO2 and alkanolamines both in aqueous and non-aqueous solutions. An overview. Chem Eng Commun 144:113–158 10. Versteeg G, Van Swaaij W (1988) On the kinetics between CO2 and alkanolamines both in aqueous and non-aqueous solutions—I. Primary and secondary amines. Chem Eng Sci 43:573–585 11. Bosch H, Versteeg G, Van Swaaij W (1990) Kinetics of the reaction of CO2 with the sterically hindered amine 2-amino-2-methylpropanol at 298 K. Chem Eng Sci 45:1167–1173 12. Xu S, Wang Y-W, Otto FD, Mather AE (1996) Kinetics of the reaction of carbon dioxide with 2-amino-2-methyl-1-propanol solutions. Chem Eng Sci 51:841–850 13. Rangwala H, Morrell B, Mather A, Otto F (1992) Absorption of CO2 into aqueous tertiary amine/MEA solutions. Can J Chem Eng 70:482–490 14. Donaldson TL, Nguyen YN (1980) Carbon dioxide reaction kinetics and transport in aqueous amine membranes. Ind Eng Chem Fundam 19:260–266 15. Dash SK, Samanta AN, Bandyopadhyay SS (2014) Simulation and parametric study of post combustion CO2 capture process using (AMP + PZ) blended solvent. Int J Greenhouse Gas Control 21:130–139 16. Li H, Le Moullec Y, Lu J, Chen J, Marcos JCV, Chen G (2014) Solubility and energy analysis for CO2 absorption in piperazine derivatives and their mixtures. Int J Greenhouse Gas Control 31:25–32 17. Smith K, Lee A, Mumford K, Li S, Thanumurthy N, Temple N et al (2015) Pilot plant results for a precipitating potassium carbonate solvent absorption process promoted with glycine for enhanced CO2 capture. Fuel Process Technol 135:60–65 18. Xiao J, Li C-W, Li M-H (2000) Kinetics of absorption of carbon dioxide into aqueous solutions of 2-amino-2-methyl-1-propanol+ monoethanolamine. Chem Eng Sci 55:161–175 19. Aroonwilas A, Veawab A (2004) Characterization and comparison of the CO2 absorption performance into single and blended alkanolamines in a packed column. Ind Eng Chem Res 43:2228–2237 20. Al-Juaied M, Rochelle GT (2006) Absorption of CO2 in aqueous diglycolamine. Ind Eng Chem Res 45:2473–2482 21. Bishnoi S, Rochelle GT (2002) Absorption of carbon dioxide in aqueous piperazine/methyl diethanolamine. AIChE J 48:2788–2799 22. Choi SY, Nam SC, Yoon YI, Park KT, Park S-J (2014) Carbon Dioxide Absorption into Aqueous Blends of Methyldiethanolamine (MDEA) and alkyl amines containing multiple amino groups. Ind Eng Chem Res 53:14451–14461 23. Zhang X, Zhang C-F, Qin S-J, Zheng Z-S (2001) A kinetics study on the absorption of carbon dioxide into a mixed aqueous solution of methyldiethanolamine and piperazine. Ind Eng Chem Res 40:3785–3791 24. Zoghi AT, Feyzi F, Zarrinpashneh S (2012) Experimental investigation on the effect of addition of amine activators to aqueous solutions of N-methyldiethanolamine on the rate of carbon dioxide absorption. Int J Greenhouse Gas Control 7:12–19 25. Adeosun A, Abu-Zahra MR (2013) Evaluation of amine-blend solvent systems for CO2 post-combustion capture applications. Energy Procedia. 37:211–218 26. Huang Y, Zhang X, Zhang X, Dong H, Zhang S (2014) Thermodynamic modeling and assessment of ionic liquid-based CO2 capture processes. Ind Eng Chem Res 53:11805–11817 27. Mumford KA, Smith KH, Anderson CJ, Shen S, Tao W, Suryaputradinata YA et al (2011) Post-combustion capture of CO2: results from the solvent absorption capture plant at Hazelwood power station using potassium carbonate solvent. Energy Fuels 26:138–146

240

Y.S. Yu et al.

28. Thee H, Nicholas NJ, Smith KH, da Silva G, Kentish SE, Stevens GW (2014) A kinetic study of CO2 capture with potassium carbonate solutions promoted with various amino acids: Glycine, sarcosine and proline. Int J Greenhouse Gas Control 20:212–222 29. Yu Y, Lu H, Wang G, Zhang Z, Rudolph V (2013) Characterizing the Transport Properties of Multiamine Solutions for CO2 Capture by Molecular Dynamics Simulation. J Chem Eng Data 58:1429–1439 30. Beyad Y, Puxty G, Wei S, Yang N, Xu D, Maeder M, et al (2014) An SO2 tolerant process for CO2 capture|NOVA. The University of Newcastle’s Digital Repository 31. Budzianowski WM (2011) Mitigating NH3 vaporization from an aqueous ammonia process for CO2 capture. Int J Chem Reactor Eng 9:1–27 32. Cousins A, Wardhaugh LT, Feron PH (2011) Preliminary analysis of process flow sheet modifications for energy efficient CO2 capture from flue gases using chemical absorption. Chem Eng Res Des 89:1237–1251 33. Yu Y, Lu H, Zhang T, Zhang Z, Wang G, Rudolph V (2013) Determining the performance of an efficient nonaqueous CO2 capture process at desorption temperatures below 373 K. Ind Eng Chem Res 52:12622–12634 34. Park S-W, Choi B-S, Lee J-W (2006) Chemical absorption of carbon dioxide with triethanolamine in non-aqueous solutions. Korean J Chem Eng 23:138–143 35. Chen S, Chen S, Zhang Y, Qin L, Guo C, Chen J (2016) Species distribution of CO2 absorption/desorption in aqueous and non-aqueous N-ethylmonoethanolamine solutions. Int J Greenhouse Gas Control 47:151–158 36. Guo C, Chen S, Zhang Y, Wang G (2014) Solubility of CO2 in Nonaqueous Absorption System of 2-(2-Aminoethylamine) ethanol+ Benzyl Alcohol. J Chem Eng Data 59:1796–1801 37. Park S-W, Lee J-W, Choi B-S, Lee J-W (2005) Kinetics of absorption of carbon dioxide in monoethanolamine solutions of polar organic solvents. J Ind Eng Chem 11:202–209 38. Da Silva EF, Svendsen HF (2007) Computational chemistry study of reactions, equilibrium and kinetics of chemical CO2 absorption. Int J Greenhouse Gas Control 1:151–157 39. Zhang T, Zhang Z (2014) Computational study of CO2 absorption in Aqueous and non-aqueous solutions using MEA. Energy Procedia. 63:1347–1353 40. Naami A, Edali M, Sema T, Idem R, Tontiwachwuthikul P (2012) Mass transfer performance of CO2 absorption into aqueous solutions of 4-diethylamino-2-butanol, monoethanolamine, and N-methyldiethanolamine. Ind Eng Chem Res 51:6470–6479 41. Sema T, Naami A, Fu K, Edali M, Liu H, Shi H et al (2012) Comprehensive mass transfer and reaction kinetics studies of CO2 absorption into aqueous solutions of blended MDEA–MEA. Chem Eng J 209:501–512 42. Li J, You C, Chen L, Ye Y, Qi Z, Sundmacher K (2012) Dynamics of CO2 absorption and desorption processes in alkanolamine with cosolvent polyethylene glycol. Ind Eng Chem Res 51:12081–12088 43. Tan J, Shao H, Xu J, Du L, Luo G (2011) Mixture absorption system of monoethanolamine− triethylene glycol for CO2 capture. Ind Eng Chem Res 50:3966–3976 44. Zheng C, Tan J, Wang Y, Luo G (2012) CO2 Solubility in a mixture absorption system of 2-Amino-2-methyl-1-propanol with Glycol. Ind Eng Chem Res 51:11236–11244 45. Galvão AC, Francesconi AZ (2010) Solubility of methane and carbon dioxide in ethylene glycol at pressures up to 14 MPa and temperatures ranging from (303 to 423) K. J Chem Thermodyn 42:684–688 46. Gui X, Tang Z, Fei W (2011) Solubility of CO2 in alcohols, glycols, ethers, and ketones at high pressures from (288.15 to 318.15) K. J Chem Eng Data 56:2420–2429 47. W-m Qian, Y-g Li, Mather AE (1995) Correlation and prediction of the solubility of CO2 and H2S in an aqueous solution of methyldiethanolamine and sulfolane. Ind Eng Chem Res 34:2545–2550 48. Yu YS, Zhang TT, Wu XM, Mu DL, Zhang ZX, Wang GGX (2015) Exploiting an alternative CO2 absorption process by efficient solvent mixture. Ind Eng Chem Res 54:6165–6174 49. Derks P, Versteeg G (2009) Kinetics of absorption of carbon dioxide in aqueous ammonia solutions. Energy Procedia. 1:1139–1146

The Absorption Kinetics of CO2 into Ionic Liquid—CO2 Binding Organic Liquid and Hybrid Solvents Ozge Yuksel Orhan, Cyril Sunday Ume and Erdogan Alper

Abstract Carbon dioxide (CO2) capture is a global concern because of its effect on climate change especially as regards to global warming. Among greenhouse gases, CO2 is the most abundant with high concentration released from post combustion processes into the atmosphere. For instance, the volume of CO2 emission from thermal power plants, petroleum refineries, petrochemical plants, hydrogen and cement factories has become one of the top important global concerns nowadays. In order to capture post-combustion CO2 and securely store it way or to produce useful products from it, it requires separation of CO2 from flue gas stream. Industrially, it is generally accepted that the most appropriate method that can be applied commercially to capture CO2 involves absorbing it with a reversible reaction from gas streams into aqueous amine especially monoethanolamine. Although, CO2-aqueous amine process is accepted as a mature technology but its absorption/desorption systems are the subject of several studies as the process is energy-intensive among other issues. In view of the shortfalls of CO2-aqueous amine systems and the greater societal concern to control the amount of CO2 released to the environment from industrial sources to abate its effects, research for alternative viable solvent systems becomes of high interest to researchers as well as industrialists. Hence, this chapter is mainly to focus on highlighting and discussing relevant advanced solvent systems for CO2 capture. Among the novel solvents or technology worthy of discussion here include use of organic solvents consisting of an amidine or a guanidine and a linear alcohol, such as 1-hexanol, instead of aqueous amines. In this case, CO2 loaded solvent could be regenerated at 90–100 °C which is much lower than the boiling point of the solvent and as a result, sufficient drop in energy requirements could be O.Y. Orhan (&) E. Alper Department of Chemical Engineering, Hacettepe University, Ankara, Turkey e-mail: [email protected] E. Alper e-mail: [email protected] C.S. Ume Department of Chemical and Petroleum Engineering, Federal University Ndufu-Alike Ikwo (FUNAI), Abakaliki, Nigeria e-mail: [email protected] © Springer International Publishing AG 2017 W.M. Budzianowski (ed.), Energy Efficient Solvents for CO2 Capture by Gas–Liquid Absorption, Green Energy and Technology, DOI 10.1007/978-3-319-47262-1_11

241

242

O.Y. Orhan et al.

achieved. In order to be applicable, CO2 binding organic liquid systems (“CO2BOLs”) will react with CO2 at sufficient rates. In case, the reaction rate is not sufficient it can be upgraded with piperazine derivatives. Another potential novel method is the capture of CO2 by special liquids which have negligible vapor pressure and high thermal stability known as ionic liquids (ILs). The ILs also have favorable CO2 solubility and a wide liquid state temperature with tunable physicochemical characteristics. Further advanced approach can be to develop a blended solvent which can also react with CO2 more efficiently. A blended system is a hybrid that possesses combined benefits of the amines mixture components thereby providing a better alternative than using single component. Other novel solvents for post combustion CO2 capture are noted and discussed here.

Nomenclature—Symbols and Acronyms AMP [BF4−] [BR−] CO2 CO2BOLs CO + H2 CH4 [CI−] CFCs [(CF3SO2)2 N−] [CF3SO3−] [CH3CO2−] [CF3CO2−] [(CN)2 N−] CS2 DBU DEA GHG GHGs HFCs H 2S ILs [I−] ko MDEA MEA N2 N 2O NO

2-amino-2-methylpropanol Tetrafluoroborate Bromide Carbon dioxide Carbon dioxide binding organic liquids Synthetic gas Methane Chloride Chlorofluorocarbons Bis (trifluoromethylsulfonyl) imide Triflate Acetate Trifluoroacetate Dicyanamide Carbon disulphide 1,8-diazabicyclo[5,4,0] undec-7-ene Diethanolamine Greenhouse gas Greenhouse gases Hydrofluorocarbon Hydrogen sulphide Ionic liquids Iodide Observed reaction rate constant Methyldiethanolamine Monoethanolamine Nitrogen Nitrogen (i) oxide Nitrogen (ii) oxide

The Absorption Kinetics of CO2 into Ionic Liquid—CO2 …

NO2 [NO3−] PEG 200 [PF6−] ROH RTILs SHAs SO2 SO3

243

Nitrogen (iv) oxide Nitrate Polyethylene glycol 200 Hexafluorophosphate Alkanol Room temperature ionic liquids Sterically hindered amines Sulphur (iv) oxide Sulphur (vi) oxide

1 Introduction The removal of carbon dioxide (CO2) from process or flue gas streams is an important step in many industrial processes for a number of reasons especially as it raises environmental concern. CO2 capture from flue gas streams of process industries is part of environmental issues of major concern in the world today as it is the most important greenhouse gas (GHG) released in abundant quantity to the atmosphere which contributes to global warming. The perceived consequences of global warming have created alarming environmental worry over the reduction of GHGs emission from industrial sources. Though the patent for application of amines to industrially capture CO2 has been granted to Bottoms as early as 1930 and notwithstanding the improvements made on the technology till date, the process still has a great setback due to high energy input and cost. In view of the global outcry for effective control of GHGs emission to mitigate its global warming effect there is still high need to search and obtain alternative CO2 capture system that will be efficient and cost effective as CO2 is one of the main target gases. Therefore, this book chapter provides a judicious practical application of this core subject. High emphasis is placed on the aspects of chemical kinetics in relation to its practical application in analyzing and solving real problems that form the foundation for the practice of a chemical engineer and other related fields. Section two of this chapter discusses briefly the term global warming, greenhouse gases and sources of CO2 emission. Section three highlights CO2 capture systems with concise explanation of pre-combustion, oxy-combustion and post combustion CO2 capture processes. In addition, solvent scrubbing techniques or CO2 capture technology were presented in section four amines—CO2 capture process were explicitly discussed with additional information on sterically hindered amines and blends of amines. Application of CO2 binding organic liquids, ionic liquids and hybrid solvents for CO2 capture are treated. Finally, reaction mechanisms of amines are explicitly discussed and kinetic equations of the various processes and model are formulated.

244

O.Y. Orhan et al.

2 Global Warming and Sources of CO2 Emission Global warming is real and it is fast affecting our planet with several evidences around us. The major factor causing increased global warming comes from carbon dioxide emission. Global warming doesn’t just mean that the earth gets hotter but the whole climate is changing. It should be noted that naturally the average surface temperature of the earth is a function of solar energy obtained from sun as a primary source. Energy from the sun is sourced as heat and light to earth during the day time. Some of the sun’s rays get ‘trapped’ in the atmosphere making the atmosphere serve as heat store, this heat warms the earth at night which makes our planet warm enough to live on. Some of the solar energy gets reflected back into space and the earth also radiates heat into space, which cools it down. Thus, when the difference in amount of energy absorbed from sun is much greater to the one reflected or radiated into space from the earth surface, then the earth heats up and the climate will change leading to global warming.

2.1

Greenhouse Gases

The gases in earth’s atmosphere that generate greenhouse effect are known as greenhouse gases. Greenhouse effect is attributed to some gases present in the atmosphere which causes absorption and emission of infrared radiation leading to warming of the atmosphere. Examples of such gases include methane (CH4), carbon dioxide (CO2), nitrous oxide (N2O) and halogens namely chlorofluorocarbons (CFCs) and hydrofluorocarbons (HFCs). The contribution of all the gases to the overall greenhouse effect is based on its emission volume as well as their individual greenhouse potentials.

2.2

Sources of CO2 Emission

The sources of CO2 emission comes from different categories ranging from low concentration to high concentration by volume or mass of CO2 into the environment. High concentration emission sources are of greater interest to CO2 capture researchers as it is a worry to the society. Among the high concentration sources of carbon dioxide and other greenhouse gases released into the atmosphere include emissions obtained from fossil fuels combustion, burning of agricultural areas or forest. Other concentrated sources come from human activities such as burning of coal, natural gas, and oil for generation of electricity.


245

3 CO2 Capture Systems It should be noted that the established ways or processes of capturing CO2 are termed CO2 capture systems. The three basic processes in use involve pre-combustion, oxyand post-combustion capture. Fossil fuels combustion takes place as a set of reactions between oxygen and hydrocarbons with CO2 as one of the products. The exhaust gas that is produced resembles air but with a higher concentration of CO2. The total amount of exhaust gas produced is very large and to store all of it is not an option therefore, CO2 must be separated from other exhaust gas components.

3.1

Pre-combustion Process

Using pre-combustion carbon capture process, combustible fuel is first transformed into synthetic gas (CO + H2) by gasification or partial oxidation. The synthetic gas is furthermore reformed with steam to produce hydrogen and carbon dioxide. The resultant product stream is then separated into a CO2 gas stream, and a stream of hydrogen. Due to the high CO2 concentration, it is removed normally using physical solvents.

3.2

Oxy-Combustion Capture

The application of oxy-combustion process to capture CO2 involves generation of high concentrated CO2 flue gas stream by using pure oxygen instead of air mixture for combustion of the primary fuel. The combustion products are mainly water vapour and CO2. Cooling and compressing the gas stream will remove water fraction of the product stream while further purification may be needed to remove air pollutants and non-condensed gases such as nitrogen from the flue gas before the CO2 is sent to storage.

3.3

Post-combustion Capture

Post-combustion capture refers to CO2 capture from flue gas streams resulting from fuel combustion using aerial oxygen. Depending on the fuel used, the flue gas contains mainly N2, steam, CO2, NO2, NO, SO2, SO3. The absorption using chemical or physical solvent is in fact most well developed technology for post combustion CO2 removal. The absorption processes are carried out by dissolving CO2 present in a flue gas stream into organic solvents or simultaneous absorption and reaction into aqueous base solutions (designated chemical absorption). The process selection

246

O.Y. Orhan et al.

depends mainly on the inlet CO2 partial pressure, degree of removal and also the energy requirement. The energy requirements are dependent mainly on circulation rate, temperature difference between the absorber and the regenerator and degree of heat exchange and solvent chemistry.

4 Solvent Scrubbing Technology for CO2 Capture There are some proved solvents that are applicable for solvent scrubbing technology for CO2 capture. In this section, the focus will be on amines, CO2 binding organic liquids (CO2BOLs), ionic liquids and hybrid solvents.

4.1

Amines

Amine technology is the most mature technology used to industrially capture CO2. The basic process covering the application involves acid gas absorption from a gas stream into an aqueous solution of an (alkanol) amine, which was patented as early as 1930 by Bottoms [1]. The conventional amine solvents mostly used for CO2 capture are single amines. In the absorption process, CO2 reacts with amines in aqueous form. Thus, the chemical solvents do react with CO2 forming non-volatile ionic species; which creates two main advantages over the physical solvents. It increases reaction kinetics of CO2 and hence, enhances its solubility in water. Figure 1 illustrates the

Fig. 1 Basic flow scheme for CO2 absorption using chemical solvent [2]


247

basic flow scheme and apparatus commonly used industrially for CO2 absorption with chemical solvent [2]. Industrially, CO2 capture using a chemical solvent (amine) is commonly conducted by passing the warm exhaust gas stream into the bottom of absorption column. From base of the column, gas rises through it meeting a stream of absorbing solution passed counter-current to the gas to absorb CO2. The CO2 absorbs and reacts with components in the solution, and the gas stream gradually loses its CO2 while moving up the column. At the top, gas with low CO2 content is released into the atmosphere. CO2 content of the solution increases as solution moves down the column. The liquid stream is typically at 90–95% of equilibrium with incoming exhaust gas at bottom of the column. At the bottom, rich concentrated CO2 solution is taken out and is channeled to a second column termed stripper (desorber). In stripper, temperature and/or pressure are set so that the chemical equilibrium in the liquid are reversed and the CO2 is released into gas phase. Pressure release is very common in natural gas applications whereas changing the temperature is the most common approach for exhaust gas treatment. Change in temperature is usually achieved by adding heat as steam in reboiler below the stripper column. A gas phase consisting only of CO2 and steam is taken out at top of the column. The steam is separated from CO2 in the overhead condenser and CO2 can be compressed to facilitate its transportation and storage. Regenerated chemical solvent from the stripper has low CO2 concentration and is again recycled for CO2 absorption. The amine solvent recycled keeps circulating between absorber and stripper columns, transporting CO2 between the columns. In an industrial process, the absorber will often be operated at temperatures around 40–55 °C while the stripper will be operating at around 120 °C. Economically, operating conventional amine technology for CO2 capture is still at a high cost notwithstanding the improvements made on the process since inception till date. This is due to the associated practical problems like high energy demand for its solvent regeneration, absorbent losses, and high corrosion rate limiting its use at high amine concentration to increase reactivity. These setbacks result in high operating costs. Hence, to reduce the operating cost, there is need to obtain and use better solvents in the CO2 separation process. Many experimental works have been published on various aspects of CO2 capture processes and therefore a brief overview is given here on the nature of published data. A number of studies deal with gas-liquid equilibrium experiments. The CO2 partial pressure and CO2 concentration is determined where the system is taken to be in a pre-fixed pressure and temperature values. Obtained experimental results are then presented as plots of CO2 partial pressure (kPa) versus CO2 uptake (loading) so as to determine CO2 solubility in a given amine. Another important form of experiment is the study of kinetics. The experimental set-up for kinetics studies varies significantly but the general approach is to measure the rate of CO2 uptake in a liquid at a given set of conditions. The conditions set are usually temperature, pressure and liquid composition.

248

O.Y. Orhan et al.

The CO2 uptake does not necessarily reflect a single rate of reaction. Some analysis work is usually required to extract reaction kinetics data from the experimental results. The importance of chemical kinetics study with respect to reducing energy demand by carbon capture process cannot be overemphasized. Reaction kinetics gives us the insight of reaction rates of various solvents proposed for application in carbon capture. It provides knowledge of affects of various variables in the system thereby enabling one to determine the optimum condition for a targeted output with minimum waste. Kinetic studies helps in many ways to determine energy efficiency of carbon capture. Specifically, it provides details of reaction rates of various solvents and mediums which enables one to make a choice of most efficient system to be applied in the carbon capture process. The kinetic studies also relate effect of variables, intermediates reaction with overall reaction mechanism that can generate mathematical models that will describe chemical reactions. This makes it easier for one to make choice of the most efficiency solvent option that can be applied for CO2 capture with minimum energy input in CO2 absorption, stripping and regeneration of the solvent in use there by improving the economic efficiency of the process. In kinetics study, calorimetric experiments can also be used to obtain information on the enthalpy of CO2 absorption. The same kind of measurements can also be used to determine heat capacities. Many pure amines and amine-water systems have available data of their measured physical properties. The extent of such data is however more limited than for common organic molecules. The primary amine, MEA is commonly used industrially for CO2 separation. MEA solution reacts fairly fast and absorbs CO2 and H2S simultaneously. It reacts also with COS, CS2 and mercaptans. The high vapour pressure property of monoethanolamine makes it a good option for CO2 removal especially in flue gas stream with very low H2S concentrations and without COS or CS2 [3]. The kinetic results for reaction of MEA with CO2 in aqueous solution using a direct stopped flow technique compare favourably with those obtained using indirect methods. Ali et al. [4], experimentally studied kinetics of MEA with CO2 using stopped flow method, and the results they obtained were in good agreement with similar experimental results published by other Authors [5, 6]. Secondary amine like DEA helps overcome the limitation of MEA, and is used in the presence of COS and CS2. DEA is less basic and its reaction rate does not equal the performance of MEA, but it is easier to regenerate. Thus, it is still an industrially accepted amine [7]. It should be noted that tertiary amines can result in several processing advantages over the use of MEA and DEA. First of all, it can selectively remove H2S from gas streams, because the reaction rate with CO2 is finite and slow [8]. Tertiary amines do not form carbamate but contribute to the formation of bicarbonate making it possible for its equilibrium to be easily reversed in the stripper. Because amine-CO2 stoichiometry with respect to bicarbonate formation is 1 to 1, tertiary amines do also have the potential to absorb higher amounts of CO2. However, they tend to have low reaction rates. For natural gas treatment the tertiary amine


249

N-methyldiethanolamine (MDEA) is widely used [9]. In industrial flue gases, concentration of CO2 in the gas phase is lower thus MDEA is thought to have too low reactivity to be efficiently applied in such a case. Tertiary amines are often combined with promoters in order to take advantage of the shuttle-effect [10, 11].

4.2

Sterically Hindered Amines

Research interest in sterically hindered amines (SHAs) stimulates on its ability to form unstable carbamates. These cabamate ions formed as SHAs react with CO2 are unstable and so lead to formation of biocabonates and free amines. 2-amino-2-methylpropanol (AMP) is an example of SHAs in frequent use. It offers a good alternative to conventional amines for CO2 capture as it has ability of higher CO2 absorption capacity, favourable selectivity of CO2 in presence of gas mixtures and degradation resistance advantages [12]. Reaction rate of aqueous AHPD—CO2 system has been reported though there is discrepancy between the rates from other authors [13] but all the published data gave low reaction rates indicating that sterically hindered AHPD will not be suitable alone. It was therefore, noted that AHPD is a potential SHA as CO2 absorbent if the reaction rate is enhanced by an additional amine such as piperazine. AHPD offers a higher absorption capacity, as well as lower regeneration heat over conventional amines [13].

4.3

Blends of Amines

In recent years, the usage of suitable amine mixtures over selection of single amines has become a popular approach in gas treating processes. This is mainly due to the high capacity coupled with better regeneration economy of blended amines over single ones especially when blend involves tertiary amine or sterically hindered amine with high reaction rates primary or secondary amines [14]. Alkanolamines are primary, secondary or tertiary amines containing one or more hydroxyl molecule(s). The amino group provides the basicity to absorb acid gases by a bronsted type acid-base reaction, while the OH functional group brings about decrease in vapour pressure of the alkanolamine leading to its higher solubility in water. Stoichiometrically, the ratio of 2:1 absorption capacity is achieved for a reaction between moles of primary or secondary amines and that of CO2 respectively, while a loading of 1 or higher can be obtained in the case of tertiary amines. Blends of primary or secondary alkanolamines with tertiary alkanolamines have become common to utilize the high absorption rates of the former along with the high loading capacity of the latter. There is a vast range of possible amines that can be blended to achieve desired properties but suitability of blending amines and its relative composition has to be

250

O.Y. Orhan et al.

determined experimentally to obtain the optimum blending ratio of two or more solvents. It should also be noted that though blended amines give observed reaction rate constant, ko which might be higher in value compared to addition of rate constants, ko of the respective pure amines in the blend at its equivalent concentrations under the same conditions but generally, this is not true for all blends. Notwithstanding the earlier challenges or problems associated with amine technology, it is worthy to note that the work on ultrasonic desorption of CO2 indicates that the use of monoethanolamine system for CO2 capture can still stand a test of time. They noted that the use of ultrasonic treatment of aqueous CO2-containing amine solutions, in particular ethanol amines and blends of amines, generally leads to an accelerated degassing of CO2 [15]. It revealed further that thermal degradation is minor when the reboiler temperature is held below 110 °C. Based on this claim, if amine solvents can be regenerated within temperature range of 60–80 °C from the present 100–120 °C in practice, it implies significant energy savings and reduction in thermal degradation effect. This will be an outstanding advantage for amine solvents.

4.4

CO2 Binding Organic Liquids (CO2BOLs)

Recently, a completely different approach was proposed by Jessop et al. in order to handle disadvantages of aqueous systems by switching to an organic base and high-boiling liquid compounds [16]. CO2-binding organic liquids (CO2BOLs), which are also known as switchable solvents, comprised of strong amidine/guanidine base and alcohols that capture CO2 to form amidinium or guanidinium alkyl carbonate salts. CO2BOLs can reversibly switch from a non-polar form to a polar form when exposed to carbon dioxide and causes a dramatic change in polarity (Fig. 2) [17]. Then, the high-polarity solvent can revert back to its non-ionic form by the removal of CO2 from solution [18]. The removal of CO2 is achieved by heating the solution below its boiling point or by sweeping with an inert gas such as nitrogen. The main advantages of using all-organic solvents are their lower cost of energy due to the elimination of the vaporization of water, lower stripper reboiler temperature, and their tunable physicochemical properties. Therefore, they have considerable potential to be an efficient CO2 capture solvent due to high CO2 binding capacities, low heat capacities, less solvent loss during CO2 stripping and lower energy requirement for regeneration than the traditional aqueous amine systems [19]. Consequently, there

Fig. 2 First generation of CO2BOL reacting with CO2. R = (CH2)nCH3 [17]


251

are several studies examining the absorption performance and reversibility of CO2BOL systems [16, 17, 19]. However, intrinsic reaction kinetics of CO2BOL composed of a mixture of 1,8-diazabicyclo [5,4,0] undec-7-ene (DBU) amidine base in 1-hexanol or 1-propanol were investigated by Ozturk et al. [20]. Further still, Ozturk et al. [20], studied the mechanism and kinetics of CO2 absorption in CO2-binding organic liquids (CO2BOLs) system. They noted that 1,8-Diazabicyclo[5.4.0]undec-7-ene (DBU):1-Alkanol mixtures are promising CO2 absorption systems with beneficial properties for industrial processes, such as lower specific heats and almost complete regeneration with simple heating well below the boiling point of the mixture [21]. They also stated that second order rate constants of DBU system and its observed reaction order is very much competitive with commonly used industrial CO2 absorption aqueous solvents such as AMP and comparable to MEA and DGA. Therefore, with further advancement, CO2BOLs are believed to be of future potential CO2 capture technology [22].

4.5

Ionic Liquids

Another alternative is the use of ionic liquids instead of the aqueous solution [23, 24]. Ionic liquids are molten salts composed of a large organic cation such as 1-alkyl-3-alkyl imidazolium and a small inorganic anion such as tetrafluoroborate. The melting point of most of the ionic liquids is below the room temperature. Therefore, these liquids are generally known as room temperature ionic liquids (RTILs). Ionic liquids seem to have a certain potential in separation processes due to their low vapour pressure and selective solubility for CO2. Conceptually, CO2 can be selectively absorbed from the gas mixture at high pressure and low temperature, and then can desorb at low pressure and high temperature. In such an innovative process, as there is no need to CO2 dehydration, there is no loss of the solution due to negligible vapour pressure. It is not energy-intensive for it did not require evaporation. However, generally only physical absorption will exist with existing ionic liquids, mass transfer (absorption) rate will be slow and solution capacity will be limited by the physical solubility. The absorption capacity can be enhanced by functional amine groups. However, this increases the viscosity and can cause difficulties in operation [25]. Although, variety of cations and anions are infinite that can create ionic liquids, commonly used anions and cations are shown in Table 1. Some of the possible cations; imidazolium, pyridinium, pyrrolidinium, phosphonium, ammonium or sulfonium. Possible anions; hexafluorophosphate [PF6−], tetrafluoroborate [BF4−], bis (trifluoromethylsulfonyl) imide [(CF3SO2)2 N−], triflate [CF3SO3−], acetate [CH3CO2−], trifluoroacetate [CF3CO2−], dicyanamide [(CN)2 N−], nitrate [NO3−], chloride [CI−], bromide [BR−] or iodide [I−]. The chemical structure of [EMIm]+ [Tf2N] is shown in Fig. 3.

Tosylate

Methanesulfonate

Pyridinium

Pyrrolidinium

Sulfonium

Ammonium

Phosphonium

Anion (organic) Alkylsulfate

Cation (organic) Imidazolium

Table 1 Some anions and cations constituting the ionic liquids

Halide

Tetrafluoro-borate

Hexafluoro-phosphate

Anion (inorganic) Bis(trifluoromethylsulfonyl) imide

252 O.Y. Orhan et al.


253

Fig. 3 Chemical structure of [EMIm]+ [Tf2N]−

There is ongoing interest in ionic liquids because of their advantages such as high thermal stability, negligible vapour pressure, adjustable physicochemical characteristics and high CO2 loading capacity [24, 26]. The major disadvantage of ionic liquids is the high viscosity as previously mentioned. The viscosity of the ionic liquids is determined by Van der Waals forces and hydrogen bonds. However, the viscosity can be adjusted within an acceptable range, such as 50–10.000 cP, by selecting a suitable combination of cations and anions [25]. For imidazolium-based cations, viscosity of ionic liquids depends on the length of the alkyl chain as well as upon the nature of the anion. An increase in viscosity is clearly stated in Table 2 with increasing alkyl chain lengths, because of the increased probability of van der Waals interactions between cations. Experimental and simulation studies have shown that CO2 is more soluble in alkyl imidazolium-based ionic liquids. The reason for this high solubility is that anion dominates the interactions of CO2 in ionic liquids and cation plays secondary role [26]. The length of alkyl-side chain of imidazolium based cations also affects the CO2 solubility (Table 1). Fluoro-substituted side chains increase CO2 uptake greatly rather than fluorine-nonsubstituted side chains but substantially this causes Table 2 Viscosity of various ionic liquids Ionic liquid

Viscosity (cP)

Temperature (°C)

Source

[C2mim][BF4] [C2mim][PF6] [C2mim][Tf2N] [C2mim][OTf] [C4mim][PF6] [C4mim]Cl [C4mim][BF4] [C4mim][I] [C4mim][Tf2N] [C6mim][PF6] [C6mim]Cl [C6mim][BF4] [C8mim]Cl [C8mim][PF6]

43 23.4 32.6 50 450 1534 219 1110 69 585 716 314 337 682

25 70 25 20 25 50 25 25 25 25 25 20 25 25

[27] [28] [27] [28] [29] [28] [29] [29] [29] [29] [29] [28] [29] [29]

254

O.Y. Orhan et al.

Table 3 Henry constant at different temperatures for carbon dioxide in ionic liquids [33] Ionic liquid

HCO2 (bar) 10 °C

20 °C

25 °C

30 °C

40 °C

50 °C

[bmim][Tf2N] [bmim][BF4]

28 ± 2 41.9 ± 0.2

30.7 ± 0.3 52 ± 2

34.3 ± 0.8 56 ± 2

42 ± 2 63 ± 2

45 ± 3 73 ± 1

51 ± 2 84 ± 4

increase in the viscosity [30–32]. Anion appears to be of more powerful effect on the gas solubility than cations. Imidazolium-based ionic liquids containing [Tf2N] anions show higher CO2 solubility (Table 3). Presence of long alkyl chains on the cation of the ionic liquid can create steric hindrance between the CO2-cation interactions. When the length of the alkyl side chain on the cation is increased, cation-anion interactions decrease and an increase in CO2 solubility generally results because of the increased free volume available for CO2 (Tables 3 and 4) [33, 34]. Table 4 Henry constant for carbon dioxide in different ionic liquids Ionic liquid

Abbreviation

HCO2 (bar)

Source

1-ethyl-3-methylimidazolium bis (trifluoromethanesulfonyl)imide 1-butyl-3-methylimidazolium bis (trifluoromethanesulfonyl)imide 1-hexyl-3-methylimidazolium bis (trifluoromethanesulfonyl)imide 1-butyl-3-methylimidazolium hexafluorophosphate 1-ethyl-3-methylimidazolium tetrafluoroborate 1-butyl-3-methylimidazolium tetrafluoroborate 1-methyl-3-(3,3,4,4,5,5,6,6,6-nonafluorohekzil) imidazolium bis(trifluoromethanesulfonyl)imide 1-methyl-3(3,3,4,4,5,5,6,6,7,7,8,8,8-tridecafluorooctyl) imidazolium bis(trifluoromethanesulfonyl)imide 1-hexyl-3-methylimidazolium bis (trifluoromethanesulfonyl)imide 1-hexyl-3-methylimidazolium tris(pentafluoroetil) trifluorofosfat 1-hexyl-3-methylimidazolium tris(heptafluoropropil) trifluorophosphate 1-butyl-2,3-methylimidazolium tetrafluoroborate

[emim][Tf2N]

35.6

[26]

[bmim][Tf2N]

33.0

[26]

[hmim][Tf2N]

31.6

[26]

[bmim][PF6] [emim][BF4] [bmim][BF4] [C6H4F9mim] [Tf2N] [C8H4F13mim] [Tf2N]

53.4 80.0 59.0 28.4

[35] [36] [35] [26]

27.3

[26]

[hmpy][Tf2N]

32.8

[26]

[hmim][eFAP]

25.2

[26]

[hmim] [pFAP] [bmmim] [BF4] [emmim] [Tf2N] [emim][OTf]

21.6

[26]

61.0

[26]

39.6

[26]

73

[37]

1-ethyl-2,3-methylimidazolium bis (trifluoromethanesulfonyl)imide 1-ethyl-3-methylimidazolium trifluoromethanephosphate


255

Viscosity of the commonly used ionic liquids is very high at room temperature. At the same temperature (33 °C), the viscosity of [bmim] [BF4] (79.5 cP) is 40 times higher than that of the 30% mass MEA solution [38]. To cope with this limitation of viscosity of ionic liquids, they can be mixed with water or ordinary organic solvents [39]. However, the addition of such liquids causes the reduction of gas capture ability. Gas solubility reduces with increasing amount of organic liquid [eg. polyethylene glycol 200 (PEG)]. This phenomenon can be explained by the low solubility of CO2 in PEG 200. It was determined that by adding an appropriate amount of PEG 200, both desorption rate and absorption can be increased. This case can be explained by the PEG 200 solvent properties or decrease in viscosity [40]. Another more viable option is the combination of non-volatile and stable ionic liquids with alkanolamine systems. Thus, negligible vapour pressure, high thermal stability and low heat capacity of ionic liquids can be combined with fast absorption kinetics of alkanolamines in the amine functional ionic liquids [41]. CO2 absorption capacity of the ionic liquids functionalized with suitable amino groups (“Task specific”) are three times higher than room temperature ionic liquids (RTILs) [39, 42]. Chemical and physical absorption increase with using task specific ionic liquid while working at high pressure. The task specific ionic liquids when exposed to CO2 for 3 h at room temperature and pressure, the mass gain was 7.4% which corresponds to 0.5 molar uptake of CO2. They remain stable even after five gas absorption/desorption cycles without any detectable loss in efficiency [25].

4.6

Hybrid Solvents

Separation of carbon dioxide from gas mixture by absorption is most widely used method. Conventional aqueous alkanolamine solutions used in “absorber-stripper” system have some problems such as low capacity (leading to high circulation rate) and high cost of energy in “boiler”. Capacity problems can be overcome by using hindered amines which have unstable carbamate ions but reversible reaction requires temperatures such 120–130 °C which leads to evaporation losses and high energy demand. The previous studies were to improve absorption rate and regeneration capacity of widely used alkanolamine solution. For this purpose, the advantages of the various amine reaction mechanisms were combined with activators, such as piperazine, imidazolium were added to the amine solution. Low production costs and high reaction rates of primary and secondary amines; high absorption capacity and less energy demand during regeneration due to their low heat of reaction of tertiary the amine are advantageous parameters to mix them. Because of these reasons, use of ionic liquids which remain liquid at room temperature and have high boiling point (at least 200 °C) has been raised. Commercial ionic liquids pass through several synthesis and purification steps so they are not as cheap as amines. Another limitation is their high viscosity. Solvents with high viscosity often result in high energy consumption during absorption. Various disadvantages of this kind of solvents used in carbon dioxide capture lead

256

O.Y. Orhan et al.

to high interest for new hybrid solvents that can combine the advantages of different amine systems. Therefore, a new solvent formulation which is superior in terms of energy requirement has been developed. The addition of ionic liquid to CO2BOLs is expected to bring high thermal stability and reduce the effect of volatile alcohol (eg. Hexanol) in the CO2BOL. The preparation of hybrid solutions are more practical and quick than adding functional amine groups to ionic liquids and they do not require high synthesis costs. Furthermore, the existing viscosity problems of task specific ionic liquids (TSIL) will not be encountered [43]. In this chapter, CO2-binding organic liquid—ionic liquid hybrid solvents have been proposed as good alternatives to existing alkanolamine solutions [44]. Hybrid solvents composed of such advantageous of CO2BOLs (high CO2 loading capacities, low specific heat capacities and a less energy intensive solvent regeneration) and ionic liquids (have negligible vapour pressures, high thermal stabilities, favorable CO2 solubilities and can remain liquid over a large temperature range with tunable physicochemical properties) provide to increase the absorption capacity and reaction rate. Although, there have been several studies which showed the synergetic effect on CO2 capture of ionic liquids–amine solutions there is basically no information about the reaction kinetics of ionic liquid—CO2BOL systems [45–49].

5 Reaction Mechanisms Generally, the reaction of carbon dioxide with amines can be described by two established mechanisms. These mechanisms are known as zwitterion mechanism and termolecular mechanism. The zwitterion mechanism has originally proposed by Caplow [50] and then reintroduced by Danckwerts [51] while termolecular mechanism has proposed by Crooks and Donnellan [52] and supported by Alper [53]. The zwitterion mechanism consists of a two-step process. In the first step, a zwitterion intermediate is formed rather than one-step carbamate formation. In the next step, deprotonation of the zwitterion takes place to produce carbamate ion and a protonated base. Zwitterion formation for a primary amine is as follows: CO2 þ RNH2

k2 ! RN þ H2 COO k1

ð1Þ

Deprotonation by any base (B) present in the solution RN þ H2 COO þ B ! kB RNHCOO þ BH þ

ð2Þ

In this reaction, an amine or an alcohol can act as the base depending on whether amine or alcohol is carrying out the deporotonation. The resulting net reaction is given in Eq. (3).


CO2 þ 2 RNH2 RNHCOO RNH3þ

257

ð3Þ

In a termolecular reaction mechanism, a molecule of amine (R2NH) reacts with a molecule of carbon dioxide and a molecule of a base (B) to form a loosely-bound encounter complex in a single step as represented by Eq. (4). CO2 þ RNH2 B RNHCOO BH þ

ð4Þ

In this mechanism the bonding between amine and CO2, and the proton transfer take place simultaneously (Fig. 4). While a portion of the weakly bound intermediates is transforming to the reactants (carbon dioxide and amine), even a small portion is reacted with one more amine molecule and ionic products form. The observed rate equation with respect to CO2 absorption is given in Eq. (5). robs ¼ ko ½CO2

ð5Þ

According to termolecular reaction mechanism, Eq. (6) is valid for the pseudo-first-order rate constants of hybrid systems consisting of ionic liquid (IL) and amidine (or guanidine)/linear alcohol ((B)/n-hexanol). ko ¼ kOH ½OH ½IL þ k0OH ½OH½B þ kIL ½IL½IL þ kB ½B½B þ kILB ½IL½B ð6Þ Considering that the alcohol is in excess and nearly at constant concentration, k = kOH [OH] and k ¼ k0OH [OH] are assumed so Eq. (7) is obtained. ko ¼ ðk þ kIL ½ILÞ½IL þ ðk þ kB ½BÞ½B þ kILB ½IL½B

ð7Þ

In the experiments carried out with stopped-flow technique, [IL] is kept constant and [B] is changed. When [IL] is kept constant at [IL]0, the following equations are obtained: ko ¼ k þ kIL ½IL0 ½IL0 þ ðk þ kB ½BÞ½B þ kILB ½IL0 ½B

ð8Þ

ko ¼ k4 þ k3 ½B þ kA ½B½B

ð9Þ

Fig. 4 Single-step reaction mechanism [52]

258

O.Y. Orhan et al.

where, k4 = (k + kIL [IL]0) [IL]0 and k3 = k* + kIL-B[IL]0, are fixed under experimental conditions. The kinetic equations valid for the amine mixtures given by Eqs. (6)–(9) are original and have not been previously proposed in the literature. In this study, the rate constants for hybrid mixtures were obtained using Eq. (9). If the degree of the reaction according to the amine is 1.00 under constant ionic liquid concentration, based on termolecular reaction mechanism, pseudo first order observed rate constant is given by Eq. (10). ko ¼ k½B

ð10Þ

According to Eq. (10), if ko (s−1) is plotted versus [B], the rate constant k (m3/kmol.s) is obtained from the slope by linear regression. If the reaction order is fractional between 1.00 and 2.00, according to termolecular reaction mechanism, affecting base can be [B] or 1-hexanol and observed pseudo first order rate constant can be defined as: ko ¼ k½B þ kB ½B2

ð11Þ

According to Eq. (11), if ko (s−1) is plotted versus [B], the rate constants k (m3/kmol.s) and kB (m6/kmol2.s) are obtained from the slope by polynomial regression. If the degree of reaction by the amine is accepted 2.00, according to termoleculer and zwitterion mechanisms the base which affected the reaction is considered probably of that amine. The effect of hexanol as a base to the reaction is negligible beside amines. The obtained pseudo first order reaction rate constant equation simplifies to: ko ¼ kB ½B2

ð12Þ

According to Eq. (12), if ko (s−1) is plotted versus ([B]2), the rate constant kB (m /kmol2.s) is obtained from the slope by linear regression. The zwitterion mechanism becomes equivalent to the termolecular mechanism and gives rise to similar expressions for reaction kinetics when the lifetime of the zwitterions intermediate approaches zero. 6

References 1. Astarita G (1983) Gas treating with chemical solvents. Wiley-Interscience Publication, New York 2. Kohl A, Nielsen R (1997) Gas purification, 5th edn. Gulf Publishing Co., Texas 3. Li J. (2006) Kinetic study of CO2 and alkanolamines both in aqueous and non-aqueous solutions using the stopped flow technique. Master’s Thesis, University of Regina, Canada


259

4. Ali SH, Merchant SQ, Fahim MA (2002) Reaction kinetics of some secondary alkanolamines with carbon dioxide in aqueous solutions by stopped flow technique. Sep Purif Technol 27:121–136 5. Alper E (1990) Kinetics of reactions of carbon dioxide with diglycolamine and morpholine. Chem Eng J 44:107–111 6. Mimura T, Suda T, Iwaki I, Honda A, Kumazawa H (1998) Kinetics of reaction between carbon dioxide and sterically hindered amines for carbon dioxide recovery from power plant flue gases. Chem Eng Commun 170:245–260 7. Ali SH (2004) Kinetic study of the reaction of diethanolamine with carbon dioxide in aqueous and mixed solvent systems: application to acid gas cleaning. Sep Purif Technol 38:281–296 8. Versteeg GF, van Swaaij WPM (1988) On the kinetics between CO2 and alkanolamines both in aqueous and non-aqueous solutions-I. Primary and secondary amines. Chem Eng Sci 43:573–585 9. de Koeijer G, Solbraa E. (2004) High pressure gas sweetening with amines for reducing CO2 emissions. 7th international conference on greenhouse gas control technologies. Vancouver, Canada 10. Bishnoi S, Rochelle GT (2002) Thermodynamics of piperazine/methyldiethanolamine/ water/CO2. Ind Eng Chem Res 41:604–612 11. Zhang X, Wang J, Zhang CF, Yang YH, Xu JJ (2003) Absorption rate into a MDEA aqueous solution blended with piperazine under a high CO2 partial pressure. Ind Eng Chem Res 42:118–122 12. Sartori G, Savage DW (1983) Sterically hindered amines for CO2 removal from gases. Ind Eng Chem Fundam 22:239–249 13. Ume CS, Alper E (2012) Reaction kinetics of carbon dioxide and 2-amino-2-hydroxymethyl-1,3-propanediol in aqueous solution obtained from stopped flow method. Turk J Chem 36:427–435 14. Ali SH (2005) Kinetic of the reaction of carbon dioxide with blends of amines in aqueous media using the stopped-flow technique. Int J Chem Kinet 37(7):391–405 15. Gantert S, Möller D (2012) Ultrasonic desorption of CO2: a new technology to save energy and prevent solvent degradation. Chem Eng Technol 35(3):576–578 16. Jessop PG, Heldebrant DJ, Li X, Eckert CA, Liotta CL (2005) Green chemistry: reversible nonpolar-to-polar solvent. Nature 436:1102 17. Jessop PG, Mercer SM, Heldebrant DJ (2012) CO2-triggered switchable solvents, surfactants, and other materials. Energy Environ Sci 5(6):7240–7253 18. Yu T, Gaviola GC, Weiss RG (2008) Carbon dioxide and molecular nitrogen as switches between ionic and uncharged room-temperature liquids comprised of amidines and chiral amino alcohols. Chem Mater 20:5337–5344 19. Mathias PM, Afshar K, Zheng F, Bearden MD, Freeman CJ, Andrea T, Koech PK, Kutnyakov I, Zwoster A, Smith AR, Jessop PG, Nik OG, Heldebrant DJ (2013) Improving the regeneration of CO2-binding organic liquids with a polarity change. Energy Environ Sci 6 (7):2233–2242 20. Ozturk MC, Ume CS, Alper E (2012) Reaction mechanism and kinetics of 1,8-Diazabicyclo [5.4.0]undec-7-ene and carbon dioxide in alkanol solutions. Chem Eng Technol 35(12): 2093–2098 21. Heldebrant DJ, Koech PK, Rainbolt JE, Zheng FR (2011) CO2-binding organic liquids: An integrated acid gas capture system. Energy Procedia 4:216–223 22. Yüksel Orhan Ö, Ozturk MC, Seker A, Alper E (2015) Kinetics and performance studies of a switchable solvent TMG (1,1,3,3-tetramethylguanidine)/1-propanol/carbon dioxide system. Turk J Chem 39:13–24 23. Anthony JL, Maginn EJ, Brennecke JF (2002) Solubilities and thermodynamic properties of gases in the ionic liquid 1-n-butyl-3-methylimidazolium hexafluorophosphate. J Phys Chem B 106:7315–7320 24. Muldoon MJ, Aki SN, Anderson VK, Dixon JL, Brennecke JK (2007) Improving carbon dioxide solubility in ionic liquids. J Phys Chem B 111:9001–9009

260

O.Y. Orhan et al.

25. Hasib-ur-Rahman M, Siaj M, Larachi F (2010) Ionic liquids for CO2 capture-development and progress. Chem Eng Process 49(4):313–322 26. Cadena C, Anthony JL, Shah JK, Morrow TI (2004) Why is CO2 so soluble in imidazoliumbased ionic liquids. J Am Chem Soc 126:5300–5308 27. McEwen AB, Ngo HL, LeCompte K, Goldman JL (1999) Electrochemical properties of imidazolium salt electrolytes for electrochemical capacitor applications. J Electrochem Soc 146:1687–1695 28. Seddon KR, Stark A, Torres MJ (2000) Influence of chloride, water, and organic solvents on the physical properties of ionic solids. Pure Appl Chem 72:2275–2287 29. Huddleston JG, Visser AE, Reichert WM, Willauer HD, Broker GA, Rogers RD (2001) Characterization and comparison of hydrophilic an hydrophobic room temperature ionic liquids incorporating the imidazolium cation. Green Chem 3:156–164 30. Shin EK, Lee BC (2008) High-pressure phase behavior of carbon dioxide with ionic liquids, 1-alkyl-3-methylimidazolium trifluoromethanesulfonate. J Chem Eng Data 53:2728–2734 31. Palgunadi J, Kang JE, Nguyen DQ, Kim JH, Min BK, Lee SD, Kim H, Kim HS (2009) Solubility of CO2 in dialkylimidazolium dialkylphosphate ionic liquids. Thermochim Acta 494:94–98 32. Baltus RE, Culbertson BH, Dai S, Luo H, DePaoli DW (2004) Low-Pressure solubility of carbon dioxide in room-temperature ionic liquids measured with a quartz crystal microbalance. J Phys Chem B 108:721–727 33. Hou Y, Baltus RE (2007) Experimental measurement of the solubility and diffusivity of CO2 in room-temperature ionic liquids using a transient thin-liquid-film method. Ind Eng Chem Res 46:8166–8175 34. Shiflett MB, Yokozeki A (2005) Solubilities and diffusivities of carbon dioxide in ionic liquids: [Bmim][PF6] and [Bmim][BF4]. Ind Eng Chem Res 44:4453–4464 35. Anthony JL, Anderson JL, Maginn EJ, Brennecke JF (2005) Anion effects on gas solubility in ionic liquids. J Phys Chem B 109:6366–6374 36. Camper D, Becker C, Koval C, Noble R (2005) Low pressure hydrocarbon solubility in room temperature Ils containing imidazolium rings interpreted using regular solution theory. Ind Eng Chem Res 44:1928–1933 37. Scovazzo P, Camper D, Kieft J, Poshusta J, Koval C, Noble R (2004) Regular solution theory and CO2 gas solubility in room temperature ionic liquids. Ind Eng Chem Res 43:6855–6860 38. Ion IH, Muhammad R, Faiçal L (2014) CO2 absorption in diethanolamine/ionic liquid emulsions—chemical kinetics and mass transfer study. Chem Eng J 240:16–23 39. Carvalho PJ, Alvarez VH, Schröder B, Gil AM, Marrucho IM, Aznar M, Santos MNBF, Countinho JAP (2009) Specific solvation interactions of CO2 on acetate and trifluoroacetate imidazolium based ionic liquids at high pressures. J Phys Chem B 113:6803–6812 40. Li X, Hou M, Zhang Z, Han B, Yang G, Wang X, Zou L (2008) Absorption of CO2 By ionic liquid/polyethylene glycol mixture and the thermodynamic parameters. Green Chem 10: 879–884 41. Camper D, Bara JE, Gin DL, Noble RD (2008) Room-temperature ionic liquid-amine solutions: tunable solvents for efficient and reversible capture of CO2. Ind Eng Chem Res 47:8496–8498 42. Carvalho PJ, Alvarez VH, Marrucho I, Aznar M, Coutinho JAP (2009) High pressure phase behavior of carbon dioxide in 1-butyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide and 1-butyl-3-methylimidazolium dicyanamide ionic liquids. J Supercrit Fluids 50:105–111 43. Ahmady A, Hasmin A, Aroua M (2010) Experimental investigation on the solubility and initial rate of absorption of CO2 in aqueous mixtures of methyldiethanolamine with the ionic liquid 1-butyl-3-methylimidazolium tetrafluoroborate. J Chem Eng Data 55(12):5733–5738 44. Arslan B, Şeker A, Alper E (2012) İyonik Sıvı—Karbon Dioksit Tutan Organik Sıvı Hibrit Sistemleri ile Karbon Dioksit Yakalanması. Ulusal Kimya Mühendisliği Kongresi. 3-5 Eylül, Türkiye


261

45. Ahmady A, Hashim MA, Aroua MK (2011) Absorption of carbon dioxide in the aqueous mixtures of methyldiethanolamine with three types of imidazolium-based ionic liquids. Fluid Phase Equilibr 309(1):76–82 46. Ahmady A, Hashim MA, Aroua MK (2012) Kinetics of carbon dioxide absorption into aqueous MDEA+[bmim][BF4] solutions from 303 to 333K. Chem Eng J 200:317–328 47. Lu BH, Jin JJ, Zhang L, Li W (2012) Absorption of carbon dioxide into aqueous blend of monoethanolamine and 1-butyl-3-methylimidazolium tetrafluoroborate. Int J Greenhouse Gas Control 11:152–157 48. Lv B, Sun C, Liu N, Li W, Li S (2015) Mass transfer and kinetics of CO2 absorption into aqueous monoethano)lamine/1-hydroxyethy-3-methyl imidazolium glycinate solution. Chem Eng J 280:695–702 49. Iliuta I, Hasib-ur-Rahman M, Larachi F (2014) CO2 absorption in diethanolamine/ionic liquid emulsions—chemical kinetics and mass transfer study. Chem Eng J 240:16–23 50. Caplow M (1968) Kinetics of carbamate formation and breakdown. J Am Chem Soc 90:6795–6803 51. Danckwerts PV, Kennedy AM (1954) Kinetics of liquid-film processes in gas absorption, Part II: measurements of transient absorption rates. Trans Inst Chem Eng 32:53–59 52. Crooks JE, Donnellan JP (1989) Kinetics and mechanism of the reaction between carbon dioxide and amines in aqueous solution. J Chem Soc Perkin Trans 1989:331–333 53. Alper E (1990) Reaction-mechanism and kinetics of aqueous-solutions of 2-amino-2-methyl1-propanol and carbon-dioxide. Ind Eng Chem Res 29:1725–1728

Solubility of Carbon Dioxide in Aqueous Solutions of Linear Polyamines Jian Chen, Ruilei Zhang, Zhongjie Du and Jianguo Mi

Abstract For the analysis of energy consumption for carbon dioxide capture processes from flue gases, CO2 solubility in aqueous amine solutions of various amines at different temperatures and pressures are crucial. In this work, solubility of CO2 in aqueous solutions of five linear polyamines were determined at 313.15 and 393.15 K and CO2 partial pressure of about 1–500 kPa using the constant-volume method combined with gas chromatography analysis. The amines are diethylenetriamine, dipropylenetriamine, trimethylenediamine, tetraethylenepentamine, triethylene-tetramine. The relationship between molecular structure of these polyamines and capture performance is discussed. The results show that the capture performance is affected by the species and number of amino groups, the carbon number between the amino groups, and the chain length. The corresponding values of CO2 absorption reaction heat were estimated employing the Gibbs-Helmholtz equation and were also discussed with the molecular conformations. Compared with original solvents, polyamines are more energy efficient solvents.

1 Introduction Global warming and climate change associated with increased emissions of CO2 from anthropogenic sources have become one of the most critical worldwide issues of the current age. A multitude of technological advances have been developed to reduce CO2 emissions from combustion exhaust gases [1–3]. Nowadays amine-based aqueous solutions are the most commonly used absorbents for the absorption process [4]. In this case, finding new solvents that offer highly cyclic absorption capacity for CO2 and less energy consumption for solvent regeneration J. Chen (&) R. Zhang State Key Laboratory of Chemical Engineering, Tsinghua University, Beijing 100084, China e-mail: [email protected] R. Zhang Z. Du J. Mi State Key Laboratory of Organic-Inorganic Composites, Beijing University of Chemical Technology, Beijing 100029, China © Springer International Publishing AG 2017 W.M. Budzianowski (ed.), Energy Efficient Solvents for CO2 Capture by Gas–Liquid Absorption, Green Energy and Technology, DOI 10.1007/978-3-319-47262-1_12

263

264

J. Chen et al.

is important. For the selection of better energy efficient solvents, CO2 solubility data measurements in new solvents are crucial. The first used amine solvent is methanolamine (MEA). MEA is a primary amine and a strong base. For CO2 absorption, it has advantages as high absorption ability, high reaction rate and high absorption ratio. But it also has disadvantages as high regeneration heat, high degradation rate and serious facility corrosion. The study on CO2 solubility in aqueous MEA solutions show that CO2 loading (mole CO2/mole MEA) is about 0.5–0.6 at the conditions for absorption, while at the conditions of regeneration the CO2 loading is about 0.15–0.25 [5–7]. So the circle loading is about 0.35. Cyclic absorption capacity is the solubility difference between absorption and desorption. Another kind of solvents are secondary amines, as diethanolamine (DEA) [8], diglycolamine (DGA) [9] and diisoproanolamine (DIPA) [10]. Compared with MEA, these secondary amines have advantages as high boiling temperature and less degradation, with the similar absorption ability with MEA. Their disadvantage is higher molecular weights, which leads more heat capacity. And tertiary amines have advantages as high CO2 solubility at high CO2 partial pressure and less degradation. The typical solvent is N-methyl-diethanolamine (MDEA), which is widely used in acid gas removal from high pressure gases. The circle loading at high CO2 partial pressure can be higher as 0.65 [11, 12], which is higher than those of primary and secondary amines. Nowadays another kind of amine solvents are steric-hindred amines, as 2-amino-2-methyl-1-propanol (AMP), with its CO2 absorption mechanism is mainly bicarbonate over carbarmate [13]. So its CO2 loading is close to 0.8, and circle loading is as more as 0.6. Chakraborty et al. [14], Sartori and Savage [15], and Singh et al. [16] studied the steric-hindered effects, and pointed out that these amines have a-substituted groups, and because of hydrolysis there are about 1 mol of free amines liberated [17], leads higher concentration of free amines to react with more CO2. Based on AMP, for increase of solvent boiling points, and decrease of solvent evaporation lose, amino-methyl-propanediol and amino-ethyl-propanediol are used to absorb CO2, but the second hydroxyl group leads lower CO2 solubility [18, 19]. Cycle amines are always the research direction, and the typical one is piperazine (PZ). The CO2 solubility in aqueous PZ solution has been reported [20], and CO2 loading at usual CO2 partial pressure is as more as 0.8, the circle loading is about 0.5. At high CO2 partial pressure, CO2 loading is more than 1.0 [21]. In multi-amino solvents, amino-ethyl- ethanolamine (AEEA) is the amine studied frequently [22]. Compared with MEA, it has advantages of high boiling temperature, and low solvent evaporation loss, with CO2 solubility about 20 % higher than in MEA. In order to find absorption solvent with low energy consumption, relationship of molecular structure with CO2 absorption ability has been studied, e.g., chain length of alkanolamines, chain lengths of alkylamines, chain length of diamines, and also effects of side chain number and amino numbers [16]. It is also found that increase of chain length between hydroxyl group and amino group can increase the CO2 solubility in solvents [23]. The key point was pointed out that the electronegativity

Solubility of Carbon Dioxide in Aqueous Solutions …

265

of the nitrogen atom in alkanolamines is the most important effects for molecular structure on CO2 absorption ability [24]. For different requirements of CO2 absorption amounts, kinetic rates, mixtures of amines have been used and CO2 solubility measured in these mixed amines, including MEA-MDEA [25, 26], MEA-AMP [27], DEA-AMP [28, 29], AMP-PZ [30], MDEA-PZ [31–33] and AMP-MDEA [34]. These mixed amine solutions are mainly mixtures of primary or secondary amines with tertiary or steric-hindered amines, for overall consideration of solubility and kinetic rate. Compared with monoamine solvents, polyamines [35] solvents are expected to have lower volatility along with higher CO2 loading capacity and mass transfer rate. For example, diethylenetriamine (DETA) entails a higher mass transfer rate [36], higher cyclic capacity [37–39], and significant lower heat of absorption [40] than monoethanolamine (MEA). tetraethylenepentamine (TEPA) shows outstanding potential for CO2 absorption, and it can maintain a high absorption rate as well as cyclic absorption capacity [41]. 1.0 mol TEPA removes three times more CO2 per cycle than 1.0 mol MEA. These investigations reveal that polyamines make a good prospect for chemical absorption. Singh et al. showed that an increase in chain length between the amine and different functional groups, result in a decrease of absorption rate whereas, the absorption capacity was increased in most absorbents [42, 43]. Machida et al. observed that alkyl chain length between two amines has an important role in CO2 solubility [44]. Although some fragment solubility data for CO2 in polyamines have been reported in previous literature, the solubility data for CO2 in polyamines in different structures are insufficient. In this work, the vapor–liquid equilibrium data of CO2 in ten polyamine solutions were measured at 313.15 K (absorption process) and 393.15 K (desorption process). Based on the solubility data, the corresponding absorption heats (DHabs ) were calculated to analyze the energy consumptions for absorbent regeneration. These data could be used for detailed analysis on the performance of these solvents and on possible energy consumption for CO2 capture.

2 Measurement 2.1

Materials

CO2 and N2 with a volume fraction of 0.99999 were supplied by Millennium City Gas. The solvents used are listed in Table 1, and their molecular structures are shown in Table 2. The solution was prepared using ultrapure water, which was taken from the Center 120 FV-S Ultrapure Water Machine. The resistivity of ultrapure water is 18.2 MXcm at 298.15 K. All components were used without further purification.

266

J. Chen et al.

Table 1 Chemicals used in this work Component

Abbreviation

Molecular formula

CAS number

Purity (%)

Source

Diethylenetriamine Dipropylenetriamine Trimethylenediamine Tetraethylenepentamine

DETA DPTA TMDA TEPA

C4H13N3 C6H17N3 C3H10N2 C8H23N5

111-40-0 56-18-8 109-76-2 112-57-2

99 >98 98 >90

Triethylenetetramine

TETA

C6H18N4

112-24-3

>70

J&KScientific TCI Shanghai Alfa Aesar Sinopharm chemical Sinopharm chemical

2.2

Apparatus and Procedure

The constant-volume method, combined with gas chromatography analysis was used to measure the differing levels of solubility. A detailed description of the vapor–liquid equilibrium apparatus can be obtained from previous literature [45, 46]. It is comprised of two stainless steel tanks for buffer and reaction. Both tanks are equipped with temperature transducers (PT-100, Kunlunhaian Co.) and pressure transducers (JYB-KO-HAG, Kunlunhaian Co.). The accuracy of the temperature and pressure transducers is ±0.1 K and ±0.5 %, respectively. In accordance with the Peng-Robinson (PR) equation, the precise amount of CO2 in the gas phase was determined using its volume, pressure, and temperature [47]. The principle of this method is to accomplish a known volume of gas with the known-volume polyamine solutions. The amount of CO2 gas introduced into the reaction tank was determined by the change in the buffer tank, before and after injection. After equilibrium was achieved at a constant temperature, the amount of CO2 gas absorbed in the solutions was equal to total amount of CO2 subtracted by the amount of CO2 in the vapor phase. The CO2 solubility, described as CO2 loading, in the liquid phase was defined as the mole amount of CO2 in the liquid phase divided by the mole number or mass of amine. The experimental error of CO2 loading is estimated to be ±8 %. At low partial pressure (