Volume 43 Number 22 21 November 2014 Pages 7457-7956 Chem Soc Rev Chemical Society Reviews www.rsc.org/chemsocrev Themed issue:Catalysis for production of renewable energy ISSN0306-0012 ROYAL SOCIETY OF CHEMISTRY REVIEW ARTICLE Yong Wang et al. Catalytic fast pyrolysis of lignocellulosic biomass
Chem Soc Rev Chemical Society Reviews www.rsc.org/chemsocrev ISSN 0306-0012 REVIEW ARTICLE Yong Wang et al. Catalytic fast pyrolysis of lignocellulosic biomass Themed issue: Catalysis for production of renewable energy Volume 43 Number 22 21 November 2014 Pages 7457–7956
ROYAL SOCIETY OF CHEMISTRY Chem Soc Rev REVIEW ARTICLE View Article Online View Joumal View Issue Catalytic fast pyrolysis of lignocellulosic biomass Cite this:Chem.Soc.Rev.,2014, Changjun Liu,Huamin Wang,5 Ayman M.Karim,Junming Sun and Yong Wang*ab 43.7594 Increasing energy demand,especially in the transportation sector.and soaring COz emissions necessitate '8S:E:L0910/8I uo KusIan uooer yeyS Kq papeojumod the exploitation of renewable sources of energy.Despite the large variety of new energy carriers,liquid hydrocarbon still appears to be the most attractive and feasible form of transportation fuel taking into account the energy density.stability and existing infrastructure.Biomass is an abundant,renewable source of energy:however.utilizing it in a cost-effective way is still a substantial challenge.Lignocellulose is composed of three major biopolymers,namely cellulose.hemicellulose and lignin.Fast pyrolysis of biomass is recognized as an efficient and feasible process to selectively convert lignocellulose into a liquid fuel-bio-oil.However bio-oil from fast pyrolysis contains a large amount of oxygen,distributed in hundreds of oxygenates.These oxygenates are the cause of many negative properties,such as low heating value,high corrosiveness,high viscosity.and instability:they also greatly limit the application of bio-oil particularly as transportation fueL.Hydrocarbons derived from biomass are most attractive because of their high energy density and compatibility with the existing infrastructure.Thus,converting lignocellulose into transportation fuels via catalytic fast pyrolysis has attracted much attention.Many studies related to catalytic fast pyrolysis of biomass have been published.The main challenge of this process is the development of active and stable catalysts that can deal with a large variety of decomposition Received 15th November 2013 intermediates from lignocellulose.This review starts with the current understanding of the chemistry in 喜话02 D0:10.1039/c3cs60414d fast pyrolysis of lignocellulose and focuses on the development of catalysts in catalytic fast pyrolysis. Recent progress in the experimental studies on catalytic fast pyrolysis of biomass is also summarized www.rsc.org/csr with the emphasis on bio-oil yields and quality. 1.Introduction The Gene and Linda Voiland School of Chemical Engineering and Bioengineering. The world's total primary energy supply and consumption in 2010 was double that in 1971,as was the CO,emission.The Washington State University,Pullman,WA 99164,USA.E-mail:wang42@wsu.edu Institute for Integrated Catalysis,Pacific Northwest National Laboratory. worldwide delivered energy consumption is projected to increase Richland,WA 99352,USA continuously in the next two decades with an average annual Changjun Liu received his PhD in Dr Huamin Wang is currently Chemical Engineering from Sichuan a research engineer in Pacific University in 2010 (supervised by Northwest National Laboratory. Prof.Enze Min and Prof.Bin Liang), He received his PhD from Nankai and then worked as a postdoc University,China,and then did research associate with Prof Yong his postdoctoral research in ETH Wang in the Gene Linda Voiland Zurich and UC Berkeley.He has School of Chemical Engineering experience in heterogeneous cata- and Bioengineering,Washington lysis,inorganic material synthesis, State University,USA.His current hydroprocessing,and biomass con- research interests include biomass version.His current research comversion, bio-oil upgrading, involves thermochemical comer- Changjun Liu selective hydrogenation,acid-base Huamin Wang sion of biomass and fundamental catalysis,and two-phase flow. understanding of catalytic con- version of oxygenates. 7594|Chem.Soc.Rev.2014,43.7594-7623 This joumal is The Royal Society of Chemistry 2014
7594 | Chem. Soc. Rev., 2014, 43, 7594--7623 This journal is © The Royal Society of Chemistry 2014 Cite this: Chem. Soc. Rev., 2014, 43, 7594 Catalytic fast pyrolysis of lignocellulosic biomass Changjun Liu,a Huamin Wang,b Ayman M. Karim,b Junming Suna and Yong Wang*ab Increasing energy demand, especially in the transportation sector, and soaring CO2 emissions necessitate the exploitation of renewable sources of energy. Despite the large variety of new energy carriers, liquid hydrocarbon still appears to be the most attractive and feasible form of transportation fuel taking into account the energy density, stability and existing infrastructure. Biomass is an abundant, renewable source of energy; however, utilizing it in a cost-effective way is still a substantial challenge. Lignocellulose is composed of three major biopolymers, namely cellulose, hemicellulose and lignin. Fast pyrolysis of biomass is recognized as an efficient and feasible process to selectively convert lignocellulose into a liquid fuel—bio-oil. However bio-oil from fast pyrolysis contains a large amount of oxygen, distributed in hundreds of oxygenates. These oxygenates are the cause of many negative properties, such as low heating value, high corrosiveness, high viscosity, and instability; they also greatly limit the application of bio-oil particularly as transportation fuel. Hydrocarbons derived from biomass are most attractive because of their high energy density and compatibility with the existing infrastructure. Thus, converting lignocellulose into transportation fuels via catalytic fast pyrolysis has attracted much attention. Many studies related to catalytic fast pyrolysis of biomass have been published. The main challenge of this process is the development of active and stable catalysts that can deal with a large variety of decomposition intermediates from lignocellulose. This review starts with the current understanding of the chemistry in fast pyrolysis of lignocellulose and focuses on the development of catalysts in catalytic fast pyrolysis. Recent progress in the experimental studies on catalytic fast pyrolysis of biomass is also summarized with the emphasis on bio-oil yields and quality. 1. Introduction The world’s total primary energy supply and consumption in 2010 was double that in 1971, as was the CO2 emission.1 The worldwide delivered energy consumption is projected to increase continuously in the next two decades with an average annual a The Gene and Linda Voiland School of Chemical Engineering and Bioengineering, Washington State University, Pullman, WA 99164, USA. E-mail: wang42@wsu.edu b Institute for Integrated Catalysis, Pacific Northwest National Laboratory, Richland, WA 99352, USA Changjun Liu Changjun Liu received his PhD in Chemical Engineering from Sichuan University in 2010 (supervised by Prof. Enze Min and Prof. Bin Liang), and then worked as a postdoc research associate with Prof. Yong Wang in the Gene & Linda Voiland School of Chemical Engineering and Bioengineering, Washington State University, USA. His current research interests include biomass conversion, bio-oil upgrading, selective hydrogenation, acid–base catalysis, and two-phase flow. Huamin Wang Dr Huamin Wang is currently a research engineer in Pacific Northwest National Laboratory. He received his PhD from Nankai University, China, and then did his postdoctoral research in ETH Zurich and UC Berkeley. He has experience in heterogeneous catalysis, inorganic material synthesis, hydroprocessing, and biomass conversion. His current research involves thermochemical conversion of biomass and fundamental understanding of catalytic conversion of oxygenates. Received 15th November 2013 DOI: 10.1039/c3cs60414d www.rsc.org/csr Chem Soc Rev REVIEW ARTICLE Published on 07 May 2014. Downloaded by Shanghai Jiaotong University on 18/02/2016 07:32:58. View Article Online View Journal | View Issue
View Artide Online Review Article Chem Soc Rev growth of about 1.6%(Table D1 in the reference).2Despite the and therefore has been identified as scalable,economically large variety of new energy carriers,liquid hydrocarbon still viable,and potentially carbon neutral feedstock for the production appears to be the most attractive and feasible form of trans- of renewable biofuels via appropriate technologies.Biochemical portation fuel,including aviation fuel.3 The U.S.renewable fuels conversion methodologies proposed for lignocelluloses await cost- standard(RFS2)requires an increase in the domestic supply of effective technologies5 and can only process cellulosic and alternative fuels to 36 billion gallons by 2022,including 15 billion hemicellulosic portions of lignocellulosic biomass.However, gallons from corn-based ethanol and 21 billion gallons of the thermochemical conversion routes are more energy efficient,' advanced biofuels from lignocellulosic biomass.The U.S.Energy and more flexible in terms of feed and products.s Among the Information Administration projects that the production of primary thermochemical conversion routes (i.e.,gasification and liquid fuels from biomass will soar in the next 30 years irrespec- fast pyrolysis),fast pyrolysis is the most economically feasible way tive of whether the oil prices are low or high(Fig.59 in ref.4).to convert biomass into liquid fuels,and has therefore attracted New technologies must be developed for the efficient conversion a great deal of research over the past two decades.A techno- of biomass to fuels that have high energy density and compat- economic analysis of three conversion platforms (ie.,pyrolysis, ibility with the existing energy infrastructure.3 gasification,and biochemical)comparing capital and operating Lignocellulosic biomass(such as wood,grass,and agricul- costs for near-term biomass-to-liquid fuels technology scenarios tural waste)is the most abundant and cheapest carbon source was performed recently.The analysis showed that the stand- alone biomass-to-liquid fuel plants are expected to produce fuels with a product value in the range of $2.00-5.50 per gallon gasoline equivalent,with fast pyrolysis being the lowest,and Ayman M.Karim is currently a bio-chemical conversion the highest.Fast pyrolysis shows the senior research scientist at Pacific highest yield to liquid fuel products and retains most of the Northwest National Laboratory energy from feedstocks in the liquid products.-11 Biomass (PNNL).Prior to joining PNNL conversion via fast pyrolysis is also on the verge of commercia- he did a postdoctoral stay lization.2 For instance,Envergent (a joint venture between (2007-2008)with Prof.Dionisios UOP/Honeywell and Ensyn)has a pilot-scale demonstration G.Vlachos at the University of plant under construction in Hawaii for biomass conversion to Delaware.He obtained his PhD fuels via fast pyrolysis.3 in chemical engineering from the The primary liquid product of fast pyrolysis of biomass is University of New Mexico (2007) generally called bio-oil,which is obtained by immediately under the guidance of Prof.quenching the pyrolysis vapors.Bio-oils are composed of a Abhaya K.Datye.His current large variety of condensable chemicals derived from many Ayman M.Karim research interests include funda- simultaneous and sequential reactions during the pyrolysis of mental studies of colloidal nano-lignocellulosic biomass.Bio-oil is a highly complex mixture of particles synthesis mechanisms,in situ and in operando catalyst more than 300 oxygenated compounds.10.14.5 characterization by X-ray absorption spectroscopy and developing Typical bio-oil from fast pyrolysis of woody biomass has a novel catalytic materials for the synthesis of fuels and chemicals high oxygen content and a low H/C ratio compared to crude oil from biomass. (Table 1).The chemical composition classified by functional Junming Sun is an assistant Yong Wang joined Pacific research major professor in Northwest National Laboratory Prof.Yong Wang's group at (PNNL),USA,in 1994 and was Washington State University, promoted to Laboratory Fellow in USA.He received his PhD from 2005.In 2009,he assumed a joint Dalian Institute of Chemical position at Washington State Physics of Chinese Academy of University (WSU)and PNNL.In Science in 2007 (Prof.Xinhe Bao), this unique position,he continues after which he worked with Prof. to be a Laboratory Fellow at PNNL Bruce C.Gates at UC Davis and is the Voiland Distinguished (2007-2008)and then with Prof. Professor in Chemical Engineering Yong Wang at Pacific Northwest at WSU,a full professorship with Junming Sun National Laboratory(2008-2011) Yong Wang tenure.His research interests as a postdoc researcher.His include the development of novel current research interests include fundamental understanding and catalytic materials and reaction engineering for the conversion of rational design of acid-base/supported metal catalysts for biomass fossil and biomass feedstocks to fuels and chemicals. derived small oxygenates,bimetallic catalysis for hydrodeoxygenation. This joumnal is The Royal Society of Chemistry 2014 Chem.Soc.Rev.2014.43.7594-762317595
This journal is © The Royal Society of Chemistry 2014 Chem. Soc. Rev., 2014, 43, 7594--7623 | 7595 growth of about 1.6% (Table D1 in the reference).2 Despite the large variety of new energy carriers, liquid hydrocarbon still appears to be the most attractive and feasible form of transportation fuel, including aviation fuel.3 The U.S. renewable fuels standard (RFS2) requires an increase in the domestic supply of alternative fuels to 36 billion gallons by 2022, including 15 billion gallons from corn-based ethanol and 21 billion gallons of advanced biofuels from lignocellulosic biomass. The U.S. Energy Information Administration projects that the production of liquid fuels from biomass will soar in the next 30 years irrespective of whether the oil prices are low or high (Fig. 59 in ref. 4). New technologies must be developed for the efficient conversion of biomass to fuels that have high energy density and compatibility with the existing energy infrastructure.5 Lignocellulosic biomass (such as wood, grass, and agricultural waste) is the most abundant and cheapest carbon source and therefore has been identified as scalable, economically viable, and potentially carbon neutral feedstock for the production of renewable biofuels via appropriate technologies. Biochemical conversion methodologies proposed for lignocelluloses await costeffective technologies6 and can only process cellulosic and hemicellulosic portions of lignocellulosic biomass. However, the thermochemical conversion routes are more energy efficient,7 and more flexible in terms of feed and products.8 Among the primary thermochemical conversion routes (i.e., gasification and fast pyrolysis), fast pyrolysis is the most economically feasible way to convert biomass into liquid fuels,6 and has therefore attracted a great deal of research over the past two decades. A technoeconomic analysis of three conversion platforms (i.e., pyrolysis, gasification, and biochemical) comparing capital and operating costs for near-term biomass-to-liquid fuels technology scenarios was performed recently. The analysis showed that the standalone biomass-to-liquid fuel plants are expected to produce fuels with a product value in the range of $2.00–5.50 per gallon gasoline equivalent, with fast pyrolysis being the lowest, and bio-chemical conversion the highest.6 Fast pyrolysis shows the highest yield to liquid fuel products and retains most of the energy from feedstocks in the liquid products.9–11 Biomass conversion via fast pyrolysis is also on the verge of commercialization.12 For instance, Envergent (a joint venture between UOP/Honeywell and Ensyn) has a pilot-scale demonstration plant under construction in Hawaii for biomass conversion to fuels via fast pyrolysis.13 The primary liquid product of fast pyrolysis of biomass is generally called bio-oil, which is obtained by immediately quenching the pyrolysis vapors. Bio-oils are composed of a large variety of condensable chemicals derived from many simultaneous and sequential reactions during the pyrolysis of lignocellulosic biomass. Bio-oil is a highly complex mixture of more than 300 oxygenated compounds.10,14,15 Typical bio-oil from fast pyrolysis of woody biomass has a high oxygen content and a low H/C ratio compared to crude oil (Table 1). The chemical composition classified by functional Junming Sun Junming Sun is an assistant research & major professor in Prof. Yong Wang’s group at Washington State University, USA. He received his PhD from Dalian Institute of Chemical Physics of Chinese Academy of Science in 2007 (Prof. Xinhe Bao), after which he worked with Prof. Bruce C. Gates at UC Davis (2007–2008) and then with Prof. Yong Wang at Pacific Northwest National Laboratory (2008–2011) as a postdoc researcher. His current research interests include fundamental understanding and rational design of acid–base/supported metal catalysts for biomass derived small oxygenates, bimetallic catalysis for hydrodeoxygenation. Yong Wang Yong Wang joined Pacific Northwest National Laboratory (PNNL), USA, in 1994 and was promoted to Laboratory Fellow in 2005. In 2009, he assumed a joint position at Washington State University (WSU) and PNNL. In this unique position, he continues to be a Laboratory Fellow at PNNL and is the Voiland Distinguished Professor in Chemical Engineering at WSU, a full professorship with tenure. His research interests include the development of novel catalytic materials and reaction engineering for the conversion of fossil and biomass feedstocks to fuels and chemicals. Ayman M. Karim Ayman M. Karim is currently a senior research scientist at Pacific Northwest National Laboratory (PNNL). Prior to joining PNNL he did a postdoctoral stay (2007–2008) with Prof. Dionisios G. Vlachos at the University of Delaware. He obtained his PhD in chemical engineering from the University of New Mexico (2007) under the guidance of Prof. Abhaya K. Datye. His current research interests include fundamental studies of colloidal nanoparticles synthesis mechanisms, in situ and in operando catalyst characterization by X-ray absorption spectroscopy and developing novel catalytic materials for the synthesis of fuels and chemicals from biomass. Review Article Chem Soc Rev Published on 07 May 2014. Downloaded by Shanghai Jiaotong University on 18/02/2016 07:32:58. View Article Online
View Artide Online Chem Soc Rev Review Article Table 1 Typical elementary composition of bio-oil and crude oil. bio-oil,which is considered to be a primary issue among the (Adapted with permission from Dickerson et al,Energies,2013,6, 514-538.20 Copyright 2013 MDPL) differences between bio-oils and hydrocarbon fuels.18.24 The low heating value and flame temperature,greater ignition Composition Bio-oil Crude oil delay,and lower combustion rate of bio-oil are largely due to Water (wt%) 15-30 0.1 the high water content (15-30 wt%),although water could PH 2.8-3.8 reduce the viscosity and enhance the fluidity.25 The low pH Density (kg L-1) 1.05-1.25 0.86-0.94 of 2-3 of bio-oil is due to the significant amount of carboxylic Viscosity 50 C(cP) 40-100 180 HHV (MJ kg) 16-19 44 acids,mainly formic and acetic acids(Fig.1),and leads to its C (wt%) 55-65 83.86 corrosiveness to common construction materials such as carbon O(wt%) 28-40 1 steel and aluminum as well as some sealing materials.'The high H (wt%) 5-7 11-14 S(wt%) <0.05 <4 content of acidic components also makes the bio-oils extremely N(wt%) <0.4 <1 unstable.The physicochemical properties of bio-oils change as a Ash (wt%) <0.2 0.1 function of time under ordinary storage conditions.26 The visco- H/C 0.9-1.5 1.5-2.0 o/c 0.3-0.5 0 sity of bio-oil increases due to secondary condensation and polymerization of the high concentration of reactive compo- nents like aldehydes,ketones,and phenols.2 In distillation, 0 35-50 wt%of the primary bio-oil is left as a residue due to the Hemicellulose and Cellulose ☑Low wt% High wt% polymerization of reactive components and the substantial Acids Lignin amounts of nonvolatile sugars and oligomeric phenols.Highly Mise Oxy: 30 Acetic Glycolaldehyde oxygenated bio-oils are immiscible with hydrocarbon fuels, Propanoic Acctol DiOH-benzene which hinder their use as fuel additives. Dimeth-pnenol It is desirable and necessary to improve the quality of bio-oil Alcohols ydroglucose nol 20 toward properties similar to those of hydrocarbon fuel by certain Methanol Ethanol Furans Methyl guaiacol Ethylene Glycol upgrading techniques.Oxygen must be removed before the bio-oil can be used as a replacement for diesel and gasoline.10.23 Methyl formate Butyrolactone Ketor Bio-oil can be upgraded either off-line or during the fast pyrolysis Angelicalactone assisted by a catalyst,the so-called catalytic fast pyrolysis(CFP) process.Both cases require catalysts to efficiently remove oxygen. To date,catalytic deoxygenation has been extensively investigated for more than three decades and generally includes two approaches:catalytic cracking and hydrotreating.Catalytic Fig.1 Chemical composition of bio-oil from wood biomass and the most cracking creates products of lower oxygen content than the feed abundant molecules of each of the components.(Adapted with permission from Huber et al.Chem.Rev..2006.106.4044-4098.23 Copyright 2006 by solid acid catalysts,such as zeolites at atmospheric pressure American Chemical Society.) without the requirement of hydrogen.However,the process produces low grade products (benzene,toluene,and small chain alkanes),which require further refining,and has a low groups with relative abundance is shown in Fig.1.The main carbon yield because of significant coke formation,which components include three major families of compounds: results in a very short catalyst lifetime.34 Hydrotreating of bio- (i)small carbonyl compounds such as acetic acid,acetaldehyde, oil adopts the conventional fuel hydrotreating technologies and acetone,hydroxyaldehydes,hydroxyketones,and carboxylic acids;gives desired products by removing oxygen by hydrodeoxygena- (ii)sugar-derived compounds such as furfural,levoglucosan,tion and breaking the larger molecules in the presence of a anhydrosugars,furan/pyran ring-containing compounds;and pressurized hydrogen atmosphere and a catalyst such as sup- (iii)lignin-derived compounds,which are mainly phenols and ported molybdenum sulfide.3 Bio-oil hydrotreating has been guaiacols;oligomers of a molecular weight ranging from 900 to well developed and produces high grade products.There are 2500 are also found in significant amounts.The distribution excellent reviews that have summarized the historical develop- of these compounds mostly depends on the type of biomass ments,35 recent advances,36 and new focus on hydrogeoxygena- used and the process severity.4-2 Such a distribution also tion of lignin-derived bio-oils.However,because of bio-oil influences the physical properties of bio-oil. instability and the high oxygen content,hydrotreating suffers Some properties of bio-oil from fast pyrolysis of ligno-from high operating cost associated with significant catalyst cellulosic biomass significantly limit its direct utilization as deactivation,expensive catalysts used,and substantial hydro- transportation fuel in current systems.Generally,bio-oils are gen consumption.39 characterized by low vapor pressure,low heating value,high An alternative way is to use a catalyst to directly upgrade the acidity,high viscosity,and high reactivity.1819,23 Bio-oils show a pyrolysis vapors prior to quenching to produce bio-oil with wide range of boiling temperatures due to their complex com- improved quality,a process that is called catalytic fast pyrolysis positions.These adverse characteristics,particularly the instability (CFP).By instantly treating the hot pyrolysis vapor with a of bio-oil,are associated with the high oxygen content in the suitable catalyst,the pyrolysis intermediates are simultaneously 7596|Chem.Soc.Rev,2014.43.7594-7623 This joumal is The Royal Society of Chemistry 2014
7596 | Chem. Soc. Rev., 2014, 43, 7594--7623 This journal is © The Royal Society of Chemistry 2014 groups with relative abundance is shown in Fig. 1. The main components include three major families of compounds: (i) small carbonyl compounds such as acetic acid, acetaldehyde, acetone, hydroxyaldehydes, hydroxyketones, and carboxylic acids; (ii) sugar-derived compounds such as furfural, levoglucosan, anhydrosugars, furan/pyran ring-containing compounds; and (iii) lignin-derived compounds, which are mainly phenols and guaiacols; oligomers of a molecular weight ranging from 900 to 2500 are also found in significant amounts.16–18 The distribution of these compounds mostly depends on the type of biomass used and the process severity.14,18–22 Such a distribution also influences the physical properties of bio-oil. Some properties of bio-oil from fast pyrolysis of lignocellulosic biomass significantly limit its direct utilization as transportation fuel in current systems. Generally, bio-oils are characterized by low vapor pressure, low heating value, high acidity, high viscosity, and high reactivity.18,19,23 Bio-oils show a wide range of boiling temperatures due to their complex compositions. These adverse characteristics, particularly the instability of bio-oil, are associated with the high oxygen content in the bio-oil, which is considered to be a primary issue among the differences between bio-oils and hydrocarbon fuels.18,24 The low heating value and flame temperature, greater ignition delay, and lower combustion rate of bio-oil are largely due to the high water content (15–30 wt%), although water could reduce the viscosity and enhance the fluidity.18,25 The low pH of 2–3 of bio-oil is due to the significant amount of carboxylic acids, mainly formic and acetic acids (Fig. 1), and leads to its corrosiveness to common construction materials such as carbon steel and aluminum as well as some sealing materials.18 The high content of acidic components also makes the bio-oils extremely unstable. The physicochemical properties of bio-oils change as a function of time under ordinary storage conditions.26 The viscosity of bio-oil increases due to secondary condensation and polymerization of the high concentration of reactive components like aldehydes, ketones, and phenols.27 In distillation, 35–50 wt% of the primary bio-oil is left as a residue due to the polymerization of reactive components and the substantial amounts of nonvolatile sugars and oligomeric phenols. Highly oxygenated bio-oils are immiscible with hydrocarbon fuels, which hinder their use as fuel additives. It is desirable and necessary to improve the quality of bio-oil toward properties similar to those of hydrocarbon fuel by certain upgrading techniques.28–30 Oxygen must be removed before the bio-oil can be used as a replacement for diesel and gasoline.10,23 Bio-oil can be upgraded either off-line or during the fast pyrolysis assisted by a catalyst, the so-called catalytic fast pyrolysis (CFP) process. Both cases require catalysts to efficiently remove oxygen. To date, catalytic deoxygenation has been extensively investigated for more than three decades and generally includes two approaches: catalytic cracking and hydrotreating.31–33 Catalytic cracking creates products of lower oxygen content than the feed by solid acid catalysts, such as zeolites at atmospheric pressure without the requirement of hydrogen. However, the process produces low grade products (benzene, toluene, and small chain alkanes), which require further refining, and has a low carbon yield because of significant coke formation, which results in a very short catalyst lifetime.34 Hydrotreating of biooil adopts the conventional fuel hydrotreating technologies and gives desired products by removing oxygen by hydrodeoxygenation and breaking the larger molecules in the presence of a pressurized hydrogen atmosphere and a catalyst such as supported molybdenum sulfide.35,36 Bio-oil hydrotreating has been well developed and produces high grade products. There are excellent reviews that have summarized the historical developments,35 recent advances,36 and new focus on hydrogeoxygenation of lignin-derived bio-oils.37,38 However, because of bio-oil instability and the high oxygen content, hydrotreating suffers from high operating cost associated with significant catalyst deactivation, expensive catalysts used, and substantial hydrogen consumption.39 An alternative way is to use a catalyst to directly upgrade the pyrolysis vapors prior to quenching to produce bio-oil with improved quality, a process that is called catalytic fast pyrolysis (CFP). By instantly treating the hot pyrolysis vapor with a suitable catalyst, the pyrolysis intermediates are simultaneously Table 1 Typical elementary composition of bio-oil and crude oil. (Adapted with permission from Dickerson et al., Energies, 2013, 6, 514–538.20 Copyright 2013 MDPI.) Composition Bio-oil Crude oil Water (wt%) 15–30 0.1 pH 2.8–3.8 — Density (kg L1 ) 1.05–1.25 0.86–0.94 Viscosity 50 1C (cP) 40–100 180 HHV (MJ kg1 ) 16–19 44 C (wt%) 55–65 83.86 O (wt%) 28–40 o1 H (wt%) 5–7 11–14 S (wt%) o0.05 o4 N (wt%) o0.4 o1 Ash (wt%) o0.2 0.1 H/C 0.9–1.5 1.5–2.0 O/C 0.3–0.5 B0 Fig. 1 Chemical composition of bio-oil from wood biomass and the most abundant molecules of each of the components. (Adapted with permission from Huber et al., Chem. Rev., 2006, 106, 4044–4098.23 Copyright 2006 American Chemical Society.) Chem Soc Rev Review Article Published on 07 May 2014. Downloaded by Shanghai Jiaotong University on 18/02/2016 07:32:58. View Article Online
View Artice Online Review Article Chem Soc Rev cracked/upgraded into hydrocarbons as the biomass is pyro- development for CFP and related fundamental understanding lyzed.s The catalyst could be either directly mixed with biomass of the reaction mechanisms/routes in CFP for the sake of future feedstock or only mixed with the pyrolysis vapors.The process catalyst exploration and design. where the catalyst is mixed directly with the feedstock in the pyrolysis reactor is referred to as in situ catalytic fast pyrolysis (in situ CFP)40 while the process where the catalysts are only 2.Fast pyrolysis chemistry contacted with the pyrolysis vapors is referred to as ex situ catalytic fast pyrolysis(ex situ CFP).CFP has great potential to A fundamental understanding of the chemical properties of produce hydrocarbons directly from biomass or produce higher lignocellulosic biomass and the chemistry of the reactions quality bio-oils with improved stability lending them more taking place during the fast pyrolysis and CFP is essential to amenable for the subsequent upgrading process.The obvious rationally design more effective process and catalyst for fast advantage of CFP is the simplified process and avoided con- pyrolysis and CFP.In this section,we will summarize the recent densation and re-evaporation of the pyrolysis oil,41 since it is advancement in the chemistry of lignocellulosic biomass,the ZO/8 impossible to evaporate the bio-oils completely without degrad- fast pyrolysis of major composition of lignocellulosic biomass, ing once they have been condensed.42 The pyrolysis reaction and the catalytic fast pyrolysis of lignocellulosic biomass. pathways could be the same for both catalytic and non-catalytic Lignocellulosic biomass is a complex material,mainly com- fast pyrolysis of biomass since the bulk physical mixing of the posed of cellulose,hemicellulose,and lignin in addition to biomass and the catalyst will not be able to lead to molecular extractives (tannins,fatty acids,and resins)and inorganic level interaction.However,the presence of the catalyst could salts.34-57 The content of each component varies with the type of promote the secondary reactions of pyrolysis intermediates biomass;the woody biomass typically contains about 40-47 wt% toward certain products,and therefore considerably improve cellulose,25-35 wt%hemicellulose,and about 16-31 wt% the conversion and selectivity to desirable components in the lignin.19.58 Cellulose is a linear polymer of glucose connected produced bio-oil.43 It is known that bio-oil produced by fast by B-1,4-glycoside linkage,which forms the framework of the pyrolysis is a highly oxygenated mixture of carbonyls,carboxyls, biomass cell walls.54Cellulose is the most important element in phenolics,and water.44 In CFP,hydrocarbons are formed by biomass and has both crystalline and amorphous forms.39 removing oxygen from the pyrolysis-vapor intermediates in the Most of them are highly crystalline in nature with the polymeric form of CO2,CO,and H2O.The CFP process will lead to degree frequently in excess of 9000.60.61 Hemicellulose is struc- stabilized products and reduce the hydrogen demand in the turally amorphous and possesses a heterogeneous composition. 410e necessary hydrotreatment process that follows.The removal of It is formed by copolymers of five different Cs and Ce sugars, the most active oxygenates,such as carbonyl-and carboxyl- namely glucose,galactose,mannose,xylose,and arabinose.60 containing components,in CFP could also stabilize primary Unlike cellulose,hemicellulose is soluble in dilute alkali and bio-oils which are less prone to coke deposition and in turn consists of branched structures that vary considerably among improve the carbon yield to the final fuel products and long- biomass resources.62 Lignin is a complex three-dimensional term stability of the upgrading process.3545 CFP also provides polymer of propyl-phenol groups bound together by ether and the possibility of process intensification by means of multi-scale carbon-carbon bonds.The three basic phenol-containing com- integration and coupling of the reactions and reaction heats, ponents of lignin are p-coumaryl/p-hydroxylphenyl,coniferyl/ which reduce processing cost.Many factors affect the performance guaiacyl,and sinapyl/syringyl alcohol units.They are linked and economic feasibility of CFP of biomass.Catalysts,heating rate, with C-O(B-0-4,a-0-4,4-0-5 linking style)and C-C(B-5,5-5,B-1, residence time,and reaction temperature are the four pivotal B-B linking style)bonds.53 factors.The atmosphere in the reactor is also critical.43.46 Lin and Huber pointed out how critical the catalysis is in ligno- 2.1.Chemistry of non-catalytic fast pyrolysis of lignocellulosic cellulosic biomass conversion;47 a suitable catalyst is the key to biomass a successful CFP process.48 For instance,aromatic carbon yield 2.1.1.Fast pyrolysis process.Fast pyrolysis of ligno- as high as 30%was achieved by catalytic fast pyrolysis of cellulose proceeds by rapid heating of biomass to moderate glucose on ZSM-5,49 and this number can be further increased temperature in the absence of oxygen and immediate quench- to 40%on Ga/ZSM-5.30 Recently,Rezaei et al.reviewed the ing of the emerging pyrolysis vapors.Pyrolysis products are catalytic cracking of oxygenate compounds derived from bio- separated into char,gases,and bio-oil.Table 2 compares the mass pyrolysis with the emphasis on aromatic selectivity and process conditions and product distributions of three different olefin selectivity using zeolite catalysts.1 pyrolysis techniques.Fast pyrolysis gives the highest yield to CFP has attracted increasing attention in recent years,and bio-oil.Pyrolysis temperature,heating rate,residence time,and numerous studies have been reported over a variety of catalysts particle size all are important operation parameters affecting regarding the fundamental and practical aspects of CFP.Few bio-oil production.The optimum pyrolysis temperature was recent reviews have focused on CFp20.52 and other more general found to be about 500 C.64 Residence time greatly affects reviews have also highlighted the importance of CFP.12.43.53 the secondary reactions of pyrolysis vapors.Increasing the This review will start with the pyrolysis mechanism of ligno- residence time could either increase the gas phase cracking or the cellulose and mainly focus on the recent advances on catalyst secondary decomposition of pyrolysis vapors on the char surface. This joumnal is The Royal Society of Chemistry 2014 Chem.Soc.Rev.2014.43.7594-7623|7597
This journal is © The Royal Society of Chemistry 2014 Chem. Soc. Rev., 2014, 43, 7594--7623 | 7597 cracked/upgraded into hydrocarbons as the biomass is pyrolyzed.5 The catalyst could be either directly mixed with biomass feedstock or only mixed with the pyrolysis vapors. The process where the catalyst is mixed directly with the feedstock in the pyrolysis reactor is referred to as in situ catalytic fast pyrolysis (in situ CFP)40 while the process where the catalysts are only contacted with the pyrolysis vapors is referred to as ex situ catalytic fast pyrolysis (ex situ CFP).40 CFP has great potential to produce hydrocarbons directly from biomass or produce higher quality bio-oils with improved stability lending them more amenable for the subsequent upgrading process. The obvious advantage of CFP is the simplified process and avoided condensation and re-evaporation of the pyrolysis oil,41 since it is impossible to evaporate the bio-oils completely without degrading once they have been condensed.18,42 The pyrolysis reaction pathways could be the same for both catalytic and non-catalytic fast pyrolysis of biomass since the bulk physical mixing of the biomass and the catalyst will not be able to lead to molecular level interaction. However, the presence of the catalyst could promote the secondary reactions of pyrolysis intermediates toward certain products, and therefore considerably improve the conversion and selectivity to desirable components in the produced bio-oil.43 It is known that bio-oil produced by fast pyrolysis is a highly oxygenated mixture of carbonyls, carboxyls, phenolics, and water.44 In CFP, hydrocarbons are formed by removing oxygen from the pyrolysis-vapor intermediates in the form of CO2, CO, and H2O. The CFP process will lead to stabilized products and reduce the hydrogen demand in the necessary hydrotreatment process that follows. The removal of the most active oxygenates, such as carbonyl- and carboxylcontaining components, in CFP could also stabilize primary bio-oils which are less prone to coke deposition and in turn improve the carbon yield to the final fuel products and longterm stability of the upgrading process.35,45 CFP also provides the possibility of process intensification by means of multi-scale integration and coupling of the reactions and reaction heats, which reduce processing cost. Many factors affect the performance and economic feasibility of CFP of biomass. Catalysts, heating rate, residence time, and reaction temperature are the four pivotal factors. The atmosphere in the reactor is also critical.43,46 Lin and Huber pointed out how critical the catalysis is in lignocellulosic biomass conversion;47 a suitable catalyst is the key to a successful CFP process.48 For instance, aromatic carbon yield as high as 30% was achieved by catalytic fast pyrolysis of glucose on ZSM-5,49 and this number can be further increased to 40% on Ga/ZSM-5.50 Recently, Rezaei et al. reviewed the catalytic cracking of oxygenate compounds derived from biomass pyrolysis with the emphasis on aromatic selectivity and olefin selectivity using zeolite catalysts.51 CFP has attracted increasing attention in recent years, and numerous studies have been reported over a variety of catalysts regarding the fundamental and practical aspects of CFP. Few recent reviews have focused on CFP20,52 and other more general reviews have also highlighted the importance of CFP.12,43,53 This review will start with the pyrolysis mechanism of lignocellulose and mainly focus on the recent advances on catalyst development for CFP and related fundamental understanding of the reaction mechanisms/routes in CFP for the sake of future catalyst exploration and design. 2. Fast pyrolysis chemistry A fundamental understanding of the chemical properties of lignocellulosic biomass and the chemistry of the reactions taking place during the fast pyrolysis and CFP is essential to rationally design more effective process and catalyst for fast pyrolysis and CFP. In this section, we will summarize the recent advancement in the chemistry of lignocellulosic biomass, the fast pyrolysis of major composition of lignocellulosic biomass, and the catalytic fast pyrolysis of lignocellulosic biomass. Lignocellulosic biomass is a complex material, mainly composed of cellulose, hemicellulose, and lignin in addition to extractives (tannins, fatty acids, and resins) and inorganic salts.54–57 The content of each component varies with the type of biomass; the woody biomass typically contains about 40–47 wt% cellulose, 25–35 wt% hemicellulose, and about 16–31 wt% lignin.19,58 Cellulose is a linear polymer of glucose connected by b-1,4-glycoside linkage, which forms the framework of the biomass cell walls.54 Cellulose is the most important element in biomass and has both crystalline and amorphous forms.59 Most of them are highly crystalline in nature with the polymeric degree frequently in excess of 9000.60,61 Hemicellulose is structurally amorphous and possesses a heterogeneous composition. It is formed by copolymers of five different C5 and C6 sugars, namely glucose, galactose, mannose, xylose, and arabinose.60 Unlike cellulose, hemicellulose is soluble in dilute alkali and consists of branched structures that vary considerably among biomass resources.62 Lignin is a complex three-dimensional polymer of propyl-phenol groups bound together by ether and carbon–carbon bonds. The three basic phenol-containing components of lignin are p-coumaryl/p-hydroxylphenyl, coniferyl/ guaiacyl, and sinapyl/syringyl alcohol units. They are linked with C–O (b-O-4, a-O-4, 4-O-5 linking style) and C–C (b-5, 5-5, b-1, b–b linking style) bonds.63 2.1. Chemistry of non-catalytic fast pyrolysis of lignocellulosic biomass 2.1.1. Fast pyrolysis process. Fast pyrolysis of lignocellulose proceeds by rapid heating of biomass to moderate temperature in the absence of oxygen and immediate quenching of the emerging pyrolysis vapors. Pyrolysis products are separated into char, gases, and bio-oil. Table 2 compares the process conditions and product distributions of three different pyrolysis techniques. Fast pyrolysis gives the highest yield to bio-oil. Pyrolysis temperature, heating rate, residence time, and particle size all are important operation parameters affecting bio-oil production. The optimum pyrolysis temperature was found to be about 500 1C.64 Residence time greatly affects the secondary reactions of pyrolysis vapors. Increasing the residence time could either increase the gas phase cracking or the secondary decomposition of pyrolysis vapors on the char surface. Review Article Chem Soc Rev Published on 07 May 2014. Downloaded by Shanghai Jiaotong University on 18/02/2016 07:32:58. View Article Online