Review of Biomass Pyrolysis Oil Properties and Upgrading Research

Energy Conversion and Management 48 (2007) 87–92 www.elsevier.com/locate/enconman

Review of biomass pyrolysis oil properties and upgrading research Zhang Qi   a,b,*, Chang Jie a, Wang Tiejun a, Xu Ying a

a

Guangzhou Institute of Energy Conversion, Chinese Academy of Sciences, Guangzhou 510640, PR China b University of Science and Technology of China, Hefei 230027, PR China

Received 9 November 2005; accepted 10 May 2006 Available online 22 June 2006

Abstract

Biomass fast pyrolysis liquefaction has aroused great attention and interests both at home and abroad extensively in recent years. This paper reviews the physicochemical properties and discusses the characteristics of the components and compositions of biomass pyrolysis oil. Furthermore, the problems and focuses were summarized with some suggestions presented on upgrading and applications of bio-oil in the decades.   2006

Elsevier Ltd. All rights reserved.

Keywords:   Biomass; Pyrolysis; Bio-oil; Properties; Upgrading

1. Introduction

Considering the fact that energy consumption is increasing ing and and limit limited ed fossi fossill fuels fuels are are nearly nearly exha exhaust usted ed,, with with incr increa easin singg popu popula latio tions ns and and econ econom omic ic deve develo lopm pmen ents ts,, renewable energy should be widely explored in order to renovate the energy sources structure and keep sustainable develo developme pment nt safe. safe. Biomass Biomass is clean clean for it has negligibl negligiblee conten contentt of sulphu sulphur, r, nitroge nitrogen n and ash, which which give give lower lower emissions of SO2, NO  and soot than conventional fossil fuels. Zero net emission of CO2   can be achieved because CO2  released from biomass will be recycled into the plants by photosynthesis quantitatively. The energy crisis and fuel tension made biomass fast pyrolysis liquefaction a more important area of research and development  [1–3]  [1–3].. Pyrolysis has been supported with the highest amount of funds provided by the European Union for liquid bio-fuel technologies [4] [4].. Bio-oil from biomass fast pyrolysis is mainly produced from biomass residues in the absence of air at atmosp atmospher heric ic pressu pressure, re, a low temperat temperature ure (45 (450–5 0–550 50   C), 3 4 high high heating heating rate (10  –10 K/s) and short short gas residence residence time to crack into short chain molecules and be cooled to

liquid rapidly. Fast pyrolysis, an effective biomass conversion with high liquid yield, as much as 70–80% and a high ratio of fuel to feed, is regarded as one of the reasonable and promising technologies to compete with and eventually replace replace non-ren non-renewab ewable le fossil fossil fuel fuel resour resources ces [5] [5].. Rece Recent nt resear research ch on the physic physicoch ochemic emical al proper properties ties of biomas biomasss pyroly pyrolysis sis oil was reviewe reviewed d and some some recomme recommenda ndation tionss were put forward based on the upgrading upgrading status and application demands for future improvements. improvements.

x

2. Bio-oil

The liquid product from biomass pyrolysis is known as biomass biomass pyroly pyrolysis sis oil, oil, and bio-oi bio-oil, l, pyroly pyrolysis sis oil, or bio-cru bio-crude de for short. Bio-oil is not a product of thermodynamic equilibrium during pyrolysis but is produced with short reactor times and rapid cooling or quenching from the pyrolysis temperatures. This produces a condensate that is also not at thermo thermodyn dynamic amic equilib equilibriu rium m at storage storage temper temperatu atures res.. The chemical composition of the bio-oil tends to change toward thermodynamic equilibrium during storage.  2.1. Composition and physicochemical properties

*

Corresponding author. Tel.: +86 20 87057760; fax: +86 20 87057789. E-mail addresses: addresses:   [email protected]   (Q. Zhang),   changjie@ms.

giec.ac.cn (J. giec.ac.cn  (J. Chang). 0196-8904/$ - see front matter   2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.enconman.2006.05.010

BioBio-oi oils ls are are mult multii-co comp mpon onen entt mixt mixtur ures es of diffe differrent size molecu molecules les derive derived d from from depoly depolymeri merizat zation ion and

88

Q. Zhang et al. / Energy Conversion Conversion and Management Management 48 (2007) 87–92

Table 1 Typical properties of wood pyrolysis bio-oil and of heavy fuel oil [6] oil [6] Physical property

Bio-oil

Heavy fuel oil

Moisture content (wt%) pH Specific gravity Elemental composition (wt%) C H O N Ash HHV (MJ/kg) Viscosity (at 50  C) (cP) Solids (wt%) Distillation residue (wt%)

15–30 2 .5 1.2

–

54–58 5.5–7.0 35–40 0–0.2 0–0.2 16–19 40–100 0.2–1 up to 50

85 11 1.0 0.3 0.1 40 180 1 1

0.1   0.94

fragmentation of cellulose, hemicellulose and lignin. Therefore, the elemental composition of bio-oil and petroleum derived fuel is different, and the basic data are shown in Table 1. 1.  2.1.1. Water

Bio-oil has a content of water as high as 15–30 wt% derived from the original moisture in the feedstock and the product of dehydration during the pyrolysis reaction and storage. The presence of water lowers the heating value and and flame flame temp tempera eratu ture, re, but but on the the othe otherr hand hand,, wate waterr reduces the viscosity and enhances the fluidity, which is good good for the atomiza atomizatio tion n and combustio combustion n of bio-oi bio-oill in the engine. Shihadeh and Hochgreb  [ [7] 7] compared  compared the biooils of NREL (National Renewable Energy Laboratory, US) to those of ENSYN (Ensyn Technologies, Inc., CA) and found that additional thermal cracking improved its chemical and vaporization characteristics. The better performance and better ignition of NREL oil derived from its lower water content and lower molecular weight.  2.1.2. Oxygen

The oxyge oxygen n conten contentt of bio-o bio-oils ils is usuall usuallyy 35– 35–40% 40% [6,8] [6,8],, disdistributed in more than 300 compounds depending on the resource resource of biomass biomass and severity of the pyrolytic processes processes (temperatu (temperature, re, residence residence time and heating heating rate). rate). The presence presence of oxygen oxygen creates creates the primar primaryy issue issue for the differe difference ncess between bio-oils and hydrocarbon fuels. The high oxygen content leads to the lower energy density than the conventional fuel by 50% and immiscibility with hydrocarbon fuels also. In addition, addition, the strong acidity of bio-oils bio-oils makes them extremely unstable. Because of their complex compositions, bio-oils show a wide range of boiling point temperature. During During the distill distillati ation, on, the slow slow heatin heatingg induce inducess the polyme polymerrization of some reactive components, and bio-oils start boiling below 100   C, while stopping at 250–280   C, leaving 35–  50 wt% as solid residues. Therefore, bio-oils cannot be used in the instance of complete evaporation before combustion.

Bio-oils [9] [9]   produced from   Pterocarpus indicus and  Fraxinus mand mandshuric shurica a  had a kinetic viscosity of 70–350 mPa s and 10– 10–70 70 mPa s, respec respectiv tively, ely, and that that from from rice straw straw had a minimum kinetic viscosity of about 5–10 mPa s for its high water content. Sipilae`   et al.   [10]   investigate investigated d the bio-oils from hardwood, softwood and straw by flash pyrolysis in an atmospheri sphericc fluidiz fluidized ed bed. bed. It was found found that that the viscosit viscosities ies were reduced reduced in the bio-oils bio-oils with higher higher water water conten contentt and less water insoluble components. Viscosity was also affected by alcohols: an addition of 5 wt% methanol into hard hardwo wood od pyro pyroly lysi siss oil oil with with low low meth methan anol ol cont conten entt decreased its viscosity by 35%. The straw oil is less viscous and had the highest methanol content of 4 wt%. The research   [11]   of NREL showed that the viscosity increased only from 20 to 22 cP over a 4 month period when stored at 20   C with 10% methanol addition to the bio-oil. This would extrapolate to a viscosity of 30 cP after storage for 12 months. Ethanol at 20% had a similar stabilizing lizing effect. effect. With With 10v 10visco iscosity sity at 40  C rose from about 13 cP to an interpolated 15 cP after preheating for 12 h at 90   C, e.g. to reduce the viscosity for ease of atomization. Boucher et al.   [12] tested [12]  tested bio-oil performance with the addition of methanol regarding its use as a fuel for gas turbine applications. The methanol reduced the density and viscosity and increased the stability with the limitation of  a lowered flash point in the blend.  2.1.4. Acidity

Bio-oil Bio-oilss compris comprisee substa substantia ntiall amount amountss of carbox carboxyli ylicc acids, such as acetic and formic acids, which leads to low pH values of 2–3. The bio-oil of pine had a pH of 2.6, while that of hardwood was 2.8  [10]  [10].. Acidity makes bio-oil very corros corrosive ive and extrem extremely ely severe severe at elevat elevated ed tempera temperature ture,, which imposes more requirements requirements on construction construction materials of the vessels and the upgrading process before using bio-oil in transport fuels.  2.1.5. Heating value

The properties of bio-oils depend on factors, such as biomass feedstocks, production processes, reaction conditions and collecting efficiency. Usually the bio-oils of oil plants have a higher heating value compared with those of straw, wood or agricultural agricultural residues. Beis et al. [13] al. [13] con conducted pyrolysis experiments on a sample of safflower seed and obtained bio-oil with a heating value of 41.0 MJ/kg and a maximum yield of 44%. Ozcimen and Karaosmanoglu [14] glu  [14] produced  produced bio-oil from rapeseed cake in a fixed bed with a heating value of 36.4 MJ/kg and a yield of 59.7%. Howeve However, r, taking taking wood wood and agricul agricultur tural al residu residues es as raw mate ma teria rials, ls, the the biobio-oi oils ls have have a heat heating ing valu valuee of abou aboutt 20 MJ/kg and a yield up to 70–80%.  2.1.6. Ash

 2.1.3. Viscosity

Depend Depending ing on the biomas biomasss feedsto feedstocks cks and pyroly pyrolytic tic processes, the viscosities of bio-oils vary in a large range.

The presence of ash in bio-oil can cause erosion, corrosion and kicking problems in the engines and the valves and even deterioration when the ash content is higher than

Q. Zhang et al. / Energy Conversion and Management 48 (2007) 87–92

0.1 wt%. However, alkali metals are problematic components nents of the ash. ash. More More specifi specificall cally, y, sodium, sodium, potass potassium ium and vanadium are responsible for high temperature corrosion and deposition, while calcium is responsible for hard deposits. The H50 bio-oil was found to contain 2 ppm K, 6 ppm Na and 13 ppm Ca [12] Ca  [12].. The best job of hot gas filtering to date at NREL   [15]  resulted in less than 2 ppm alkali metals and 2 ppm alkaline earth metals in the bio-oil.  2.2. Compositions of bio-oil 

The 99.7% of bio-oil, bio-oil, a complex complex mixture containing carbon, hydrogen and oxygen, is composed of acids, alcohols, aldehy aldehydes des,, esters esters,, ketone ketones, s, sugars sugars,, phenol phenols, s, guaiac guaiacols, ols, syringols, furans, lignin derived phenols and extractible terpene with multi-functio multi-functional nal groups groups  [16]  [16].. Wang et al.  [17]   analyzed the compositions of bio-oils from   F. mand pyrolysis by gas chromatogra chromatographyphymandshuric shurica a   pyrolysis mass spectroscopy (GC–MS) and illustrated the similarities of the main contents, that is that the fragments, such as furfural, dimethyl phenol, 2-methoxy-4-methyl phenol, eugenol, cedrol, furanone, etc., are a large proportion in every bio-oil. Most of the components identified are the phenols with ketones and aldehydes groups attached, and nearly all the functi functiona onall groups groups showed showed the extens extensive ive existen existence ce of  oxygen. On the other hand, the analysis proved that the abundant aldehydes and ketones make bio-oils especially hydrophilic and highly hydrated, which leads to the water being difficult to eliminate. The bio-oil from  P. indicus [9] [9]   is mainly comprised of  levogl levogluco ucosan san,, furfur furfural, al, phenol phenolss (with (with methyl methyl,, methox methoxyy

89

and/or and/or propenyl propenyl groups), groups), aldehydes aldehydes (including (including benzaldebenzaldehyde with methyl and/or hydroxyl) and vanillin according to the GC–MS analysis. Table analysis.  Table 2 shows 2  shows the distribution of  some detected compounds in the organic composition of  bio-oil bio-oil from P. indicus. The The ma main in compo compone nent ntss of the the biobio-oi oill obtai obtaine ned d from from spruce wood in Adam’s  [18]  [18] study  study exhibited many similar components with those in  Table 2. 2. Zhang et al.   [19]  separated the bio-oil into four fractions: aliphatic, aromatic, polar and non-volatile fragments by using solvent solvent extraction extraction and liquid chromatograph chromatographyy on an aluminum column. Identification revealed the high contents tents of acetic acetic acids acids and hydrox hydroxyac yaceto etones nes in the water water phase and more polar components and less aliphatic and aromatic hydrocarbons in the oil phase. In conclusion, bio-oils are a complex mixture, highlyoxygenated with a great amount of large size molecules, which nearly involve all species of oxygenated organics, such as esters, ethers, aldehydes, aldehydes, ketones, ketones, phenols, carboxylic acids and alcohols [20] alcohols  [20].. 3. Upgrading of bio-oil

The delete deleterio rious us proper properties ties of high high viscosi viscosity, ty, therma thermall instab instabilit ilityy and corros corrosive iveness ness presen presentt many many obstac obstacles les to the substitution of fossil derived fuels by bio-oils. So, an upgr upgrad adin ingg proc proces esss by redu reduci cing ng the the oxyg oxygen en cont conten entt is required required before its application. application. The recent upgrading upgrading techniques are described as follows. 3.1. Hydrodeoxygenation

Table 2 Relative content of main compounds in organic composition of bio-oil produced from  P. indicus Compound

Relative content (%)

Furfural Acetoxyacetone, 1-hydroxyl Furfural, 5-methyl Phenol 2-Cyclopentane-1-one, 3-methyl Benzaldehyde, 2-hydroxyl Phenol, 2-methyl Phenol, 4-methyl Phenol, 2-methoxyl Phenol, 2,4-dimethyl Phenol, 4-ethyl Phenol, 2-methoxy-5-methyl Phenol, 2-methoxy-4-methyl Benzene, 1,2,4-trimethoxyl Pheno henol, l, 2,62,6-di dime meth thyyl-4l-4-(1 (1-p -pro rope peny nyl) l) 1,2-Be 1,2-Benze nzened nedica icarbox rboxylic ylic acid, acid, diisooc diisooctyl tyl ester ester 2-Furanone Levoglucosan Phenol, ol, 2,6-d 6-dimethox hoxy-4 y-4-prop ropenyl Furanone, 5-methyl Acet Acetoph ophen enone one,, 1-(41-(4-hy hydro droxy xy-3 -3-m -met etho hoxy) xy) Vanillin Benz Benzal alde dehy hyde de,, 3,53,5-di dime meth thyl yl-4 -4-h -hyd ydro roxy xyll Cinnamic Cinnamic aldehyde, aldehyde, 3,5-demetho 3,5-demethoxy-4-hy xy-4-hydroxyl droxyl

9.06 1.21 1.82 2.55 1.58 2.70 5.04 0.51 0.27 9.62 2.18 4.15 0.55 3.80 4.25 4.25 1.80 5.70 6.75 3.14 .14 0.49 2.94 2.94 6.35 4.54 4.54 2.19

The hydro-process is performed in hydrogen providing solven solvents ts activa activated ted by the cataly catalysts sts of Co–Mo, Co–Mo, Ni–Mo Ni–Mo and their oxides or loaded on Al 2O3 under pressurized conditions of hydrogen and/or CO. Oxygen is removed as H 2O and CO2, and then the energy density is elevated. Pindoria [21,22]   hydrotr hydrotreat eated ed the volati volatiles les from from fast fast pyroly pyrolysis sis of  eucalyptus in a two stage reactor. Hydro-cracking without cata cataly lysts sts was was opera operate ted d in the the first first stage, stage, and and cata catalyt lytic ic hydrot hydrotrea reatme tment nt was operat operative ive in the second second stage stage with lower temperature and the same pressure compared with that in the first stage. The analysis indicated that the deactivation of the catalyst did not result from carbon deposition. tion. Instead Instead,, the embodim embodiment entss of volati volatile le compon component entss blocke blocked d the activated activated sites sites in the zeolite zeolite cataly catalyst. st. This This hydro-process produced much water and complicated the bio-oil with many impurities. Zhang et al.   [23]   separated the bio-oil with a yield of  70% 70% into into water water and oil oil phas phases es,, and and the the oil oil phas phasee was was hydr hydrot otre reat ated ed and and cata cataly lyze zed d by sulp sulphi hide ded d Co–M Co–Mo– o–P/ P/ Al2O3. The reaction was operated in an autoclave filled with tetralin (as a hydrogen donor solvent) under the optimum conditions of 360   C and 2 MPa of cold hydrogen pressure. The oxygen content was reduced from 41.8% of  the crude oil to 3% of the upgraded one, and besides, the

90

Q. Zhang et al. / Energy Conversion Conversion and Management Management 48 (2007) 87–92

crude oil was methanol soluble while the upgraded one was oil soluble for the dehydroxy. Elloit and Neuenschwand Neuenschwander er   [24]   examined the hydrocataly catalytic tic reacti reactions ons of bio-oil bio-oilss from from NREL, NREL, UF (Union (Union Electric Electricaa Fenosa Fenosa)) and RPT3 RPT3 (Ensyn (Ensyn Techno Technolog logies, ies, Inc.) Inc.) in a continuous feed fixed bed reactor. Higher conversion was obtained in the downflow configuration compared to the upflow one. The conversion was doubled on NiMo/ cAl2O3   comp compaared red to the the CoM CoMo/sp o/spin inel el at elev elevat ateed temperature. Senol et al. [25] al.  [25] eliminated  eliminated oxygen in carboxylic groups with model compounds of methyl heptanoate and methyl hexano hexanoate ate on sulphi sulphided ded NiMo/ NiMo/ c-Al2O3   and CoMo CoMo//cAl2O3 in a flow reactor to discern the reaction schemes. Aliphatic methyl esters produced hydrocarbons via three main pathwa pathways: ys: the first first pathwa pathwayy gav gavee alcoho alcohols ls follow followed ed by dehydration to hydrocarbons. The de-esterification yielded an alcohol and a carboxylic acid in the second pathway. Carbox Carboxyli ylicc acid acid was furthe furtherr conver converted ted to hydroc hydrocarb arbons ons either directly or with an alcohol intermediate. Apparen Apparently tly the hydro-t hydro-trea reatin tingg proces processs needs needs complicomplicated cated equipm equipment ents, s, superio superiorr techni technique quess and excess excess cost cost and usually is halted by catalyst deactivation and reactor clogging. 3.2. Catalytic cracking of pyrolysis vapors

Oxygen containing bio-oils are catalytically decomposed to hydrocarbons with the removal of oxygen as H 2O, CO2 or CO. Nokkosmaki et al.  [26]  [26] proved  proved ZnO to be a mild cataly catalyst st on the compos compositio ition n and stabil stability ity of bio-oil bio-oilss in the conversion of pyrolysis vapors, and the liquid yields were not found to be substantially reduced. Although it had had no effec effectt on the the wate waterr inso insolub luble le frac fractio tion n (lign (lignin in derived derived), ), it decomp decompose osed d the diethy diethyll insolu insoluble ble fracti fraction on (water soluble anhydrosugars and polysaccharides). After heating at 80   C for 24 h, the increase in viscosity was significantly lowered for the ZnO-treated oil (55% increase in viscosity) compared to the reference oil without any catalyst (12 (129% 9% increas increasee in viscos viscosity) ity).. Despite Despite the indicat indicated ed deactivation of the catalyst, the improvement in the stability of the ZnO treated oil was clearly observed. Adam et al. [18] al.  [18] illustrated  illustrated the effects and catalytic properties of Al-MCM-41, Cu/Al-MCM-41 and Al-MCM-41 with pores pores enlarge enlarged d on bio-oil bio-oil upgrad upgrading ing.. The resultin resultingg compositions of vapors were changed through the catalysts layer. Levoglucosan Levoglucosan was completely completely eliminated, which was typical for each catalyst. While the catalysis increased the yields of acetic acid, furfural and furans, it reduced those of large molecular phenols among the cellulose pyrolysis products of spruce wood. The pore size enlargement and incorporation of catalysts reduced the yield of acetic acid and water. Adjaye and Bakhshi [27,28] Bakhshi  [27,28] studied  studied the catalytic performance of the different catalysts for upgrading of bio-oil. Among the five catalysts studied, HZSM-5 was the most effec effectiv tivee cata cataly lyst st for for prod produc ucin ingg the the orga organi nicc dist distill illat atee

fraction, overall hydrocarbons and aromatic hydrocarbons and the least coke formation. Reaction pathways were postulated tulated that that bio-oil bio-oil conversio conversion n procee proceeded ded as a result result of  therma thermall effects effects followe followed d by thermoc thermocata atalyti lyticc effects effects.. The thermal effects produced separation of the bio-oil into light organics and heavy organics and polymerization of the biooil to char. The thermocatalytic effects produced coke, tar, gas, water and the desired organic distillate fraction. Guo et al. [29] al.  [29] reviewed  reviewed various catalysts used in bio-oil upgrad upgrading ing in detail detail and believ believed ed that that althou although gh catalyt catalytic ic cracki cracking ng is a predom predominan inantt techni technique que,, the catalys catalystt with with good performance of high conversion and little coking tendency is demanding much effort. Althoug Although h cataly catalytic tic crackin crackingg is regard regarded ed as a cheape cheaperr route by converting oxygenated feedstocks to lighter fractions, the results seem not promising due to high coking (8–25 wt%) and poor quality of the fuels obtained. 3.3. Emulsification

The simplest way to use bio-oil bio-oil as a transport fuel seems to be to combine it with Diesel directly. Although the biooils are immiscible immiscible with hydrocarbon hydrocarbons, s, they can be emulsified by the aid of a surfactant. Chiaramonti   [30,31]   prepare pared d emul emulsi sifie fied d biobio-oi oill by the the rati ratios os of 25, 25, 50 and and 75 wt% and found the emulsions were more stable than the original ones. The higher the bio-oil content, the higher the viscosity of the emulsion. The optimal range of emulsifier to provide acceptable viscosity was between 0.5% and 2%. In particular, the effect of the long term use of emulsions sions on the stainl stainless ess steel steel engine engine and its subass subassemb emblies lies should be estimated. Ikura et al.   [32]   obtained light fractions of bio-oil by cent centrif rifug ugati ation on and and emul emulsi sifie fied d them them in No. No. 2 Diese Diesell by CANMET surfactant with ratios of 10, 20 and 30 wt% separat aratel ely. y. The The cost cost of prod produc ucin ingg stab stable le emul emulsi sion onss were were 2.6 cents/L for 10% emulsion, 3.4 cents/L for 20% emulsion and 4.1 cents/L cents/L for 30% emulsio emulsion n separat separately ely.. The bio-oil was determined to have a cetane number of 5.6, which will decrease by 0.4 for each 10% concentration augmentat mentation ion.. The viscosi viscosity ty of 10– 10–20 20 wt% emulsio emulsions ns was much lower than that of pure bio-oil, and their corrosiveness was about half that of bio-oil alone. Emulsifi Emulsificat cation ion does does not demand demand redund redundant ant chemica chemicall transformations, but the high cost and energy consumption input cannot be neglected. The accompanying corrosiveness ness to the the engin enginee and and the the suba subass ssemb emblie liess is inev inevit itab ably ly serious. 3.4. Steam reforming 

Hydrogen is a clean energy resource and very important in the chemical industry, and the rising focus on reforming the water fraction of bio-oil looks promising. Production of hydrog hydrogen en from from reform reforming ing bio-oil bio-oil was invest investiga igated ted by NREL [33,34] NREL [33,34] extensively,  extensively, including the reactions in a fixed bed bed or a fluid fluidiz ized ed bed bed and and stud studie iess of the the refo reform rmin ingg

Q. Zhang et al. / Energy Conversion and Management 48 (2007) 87–92

mechanisms on model compounds. The fixed bed used in the conventional reforming of natural gas or naphtha is not suitable in the lignin derived fraction of bio-oil because of its tendency to decompose thermally and form carbon deposits on the upper layer of the catalyst and in the reactor freeboard. Czernik et al. [35] al.  [35] obtained  obtained hydrogen in a fluidized bed reactor reactor from the carbohyd carbohydrate rate derived fraction of wood pyrolysis oil with a yield of about 80% of theoretical, which corres correspon ponds ds to approx approxima imatel telyy 6 kg of hydrog hydrogen en from from 100 kg of wood used for pyrolysis pyrolysis.. Commercial Commercial nickel catalysts showed good activity in processing biomass derived liquids and was readily regenerated (20 min to 2 h) by steam or CO2  gasification after deactivation, which occurred during reform reforming ing.. The commer commercia ciall cataly catalysts sts,, design designed ed for fixed fixed bed applic applicati ations ons,, were suscep susceptib tible le to attrit attrition ion in the fluidiz fluidized ed bed. Consequently, they were entrained at a rate of 5%/day. The development of a fluidizable catalyst that has both high activity and mechanical strength at the conditions of the steam reforming process is needed and is being pursued. Garcia et al.   [36]   chose magnesium and lanthanum as support modifiers to enhance steam adsorption that facilitates the carbon gasification, while cobalt and chromium additives were applied to alleviate the coke formation reactions, which modified the metal sites forming alloys with nickel nickel and possib possibly ly reducin reducingg the crystal crystallite lite size. size. More More hydrophilic hydrophilic sites surrounding surrounding nickel crystallites crystallites were effective in extending the duration of the catalyst’s activity. The catalys catalystt deactiv deactivati ation on upon upon steam steam reformin reformingg of complex complex bio-oils occurs much faster than with natural gas or naphtha, but the catalyst can be efficiently regenerated by steam or CO2  gasification. Takanabe et al.   [37]   completely converted acetic acid, the model compound, by steam reforming over Pt/ZrO 2 catalysts and found a hydrogen yield close to thermodynamic equilibrium. Analysis showed that Pt was essential for the steam reforming to proceed, and ZrO2  was needed to activate the steam, which was also active for oligomer precur precursor sor format formation ion under under the condit condition ionss investi investigat gated. ed. The results illustrated that steam reforming took place at the Pt–ZrO Pt–ZrO2   boundar boundary, y, and that that deacti deactivat vation ion occurr occurred ed when this boundary is blocked by oligomers. 3.5. Chemicals extracted from the bio-oils

Hundreds of the components of bio-oil are determined, and reclaiming one or more kinds of small and valuable chemicals chemicals arouses great interests among scholars and businessmen. There are many substances that can be extracted from bio-oil, such as phenols used in the resins industry, volatile organic acids in formation of de-icers, levoglucosan, hydroxyacetalde hydroxyacetaldehyde hyde and some additives applied in the pharmaceutical, fiber synthesizing or fertilizing industry and flavoring agents in food products  [1,38]  [1,38].. Commercialization cialization of special chemicals chemicals from bio-oils requires more devotio devotion n to develo developing ping reliable reliable low cost cost separa separation tion and refining techniques.

91

4. Conclusions and recommendations for future work

Bio-oils have received extensive recognition from international energy organizations around the world for their characteristics as fuels used in combustors, engines or gas turbines and resources in chemical industries. Some problems blocking its industrializa industrialization tion processes and recommendations are described as follows: Since bio-oils bio-oils are complex complex and chemically chemically unstable mixtures, biomass fast pyrolysis should corresponded with the characteristics of the biomass feedstocks, products uses, suitable reactors and processes for the products applications. •   Much Much more work work is needed needed on the stabilizat stabilization ion and upgrading upgrading of bio-oils bio-oils with some modifications to equipment configuration before applying them in generating heat or power. It is important to learn more about the mech mechan anism ismss invo involv lved ed in addi adding ng solv solven entt to slow slow the the apparent aging period. • Hydrod Hydrodeox eoxyge ygenat nation ion,, cataly catalytic tic cracki cracking ng and steam steam reforming of bio-oils are so complicated techniques that stea steady dy,, depe depend ndab able le,, full fullyy deve develo lope ped d reac reacto tors rs are are urgently desirable. •  According to the main components and reasonable reactions, it is meaningful to envision adding some solvents to reduce polymerization polymerization by esterification esterification,, acetalization acetalization and phenol phenol/fo /formal rmaldeh dehyde yde reactio reactions, ns, antiox antioxida idants nts to reduce reduce olefin polymerization polymerization reactions reactions and emulsifiers to prevent phase separation problems. • Bioma Biomass ss can can be an econ econom omica ically lly viab viable le rene renewa wabl blee resource for efficient energy utilities if used in an integrated grated proces processs that that also genera generates tes other other market marketabl ablee coproducts to maintain sustainable developments. •

Acknowledgements

The supports from National National natural natural science science foundation foundation of China (50476090) and Guangdong natural science foundation (04000378) are greatly appreciated. References [1] Bridgwater AV, Peacocke GVC. Fast pyrolysis processes for biomass. Sustain Renew Energy Rev 2000;4:1–73. 2000;4:1–73. [2] Scott DS, Piskorz J, Radlein D. Liquid products from the continuous flash pyrolysis of biomass. Ind Eng Chem Process Des Dev 1985; 24:581–6. [3] Czernik SR, Bridgwater AV. Overview of applications of biomass fast pyrolysis pyrolysis oil. Energy Fuels 2004;18(2):59 2004;18(2):590–8. 0–8. [4] Moore A. Environmentally sound production of liquid biofuels: the view of the European Commission. Proceedings Proceedings of the International International worksho workshop p on environ environme mental ntal aspects aspects of energy energy crop crop product production, ion, Brasimo Brasimone, ne, Italy, Italy, October October 9–10, 9–10, 1997. 1997. Wiesel Wieselbur burg: g: BLT; 1998 1998.. p. 43–67. [5] Demirb Demirbas as A. Biomass Biomass resourc resourcee facilit facilities ies and biomass biomass conver conversion sion processing for fuels and chemicals. Energy Convers Manage 2001;42: 1357–78.

92

Q. Zhang et al. / Energy Conversion Conversion and Management Management 48 (2007) 87–92

[6] Oasmaa A, Czernik Czernik S. Fuel oil quality of biomass pyrolysis pyrolysis oils-state of the art for the end-users. Energy Fuels 1999;13:914–21. [7] Shihadeh A, Hochgreb S. Impact of biomass pyrolysis oil process conditions on ignition delay in compression ignition engines. Energy Fuels 2002;16:552–61. [8] Scholze B, Meier D. Characteriz Characterization ation of the water-insoluble water-insoluble fraction fraction from pyrolysis oil (pyrolytic lignin) Part I. PY-GC/MS, FTIR, and functional functional groups. J Anal Appl Pyrol 2001;60:41–54. 2001;60:41–54. [9] Luo Luo ZY, ZY, Wang Wang SR, SR, Liao Liao YF, YF, et al. Rese Resear arch ch on bioma biomass ss fast fast pyrolysis for liquid fuel. Biomass Bioenergy Bioenergy 2004;26:455–62 2004;26:455–62.. ` s L, et al. Charac [10] Sipilae`   K, Kuoppal Kuoppalaa E, Fag Fagern ernae ae Character teriza ization tion of  biomass–based biomass–based flash pyrolysis oils. Biomass Bioenergy Bioenergy 1998;14(2): 1998;14(2): 103–13. [11] Diebold JP. A review of the chemical and physical mechanisms of the storage storage stabilit stabilityy of fast pyrolys pyrolysis is bio-oil. bio-oil. Availab Available le from: from:   http:// webdev.its.iastate.edu/webnews/data/site_biorenew_reading/19/webnewsfilefield_file/ReviewOfMechanisms.pdf  [2005-08-20].   [2005-08-20]. [12] [12] Bouch Boucher er ME, ME, Chaa Chaala la A, Roy Roy C. Bio-oi Bio-oils ls obtai obtaine ned d by vacu vacuum um pyrolysis of softwood bark as a liquid fuel for gas turbines. Part I: Properties of bio-oil and its blends with methanol and a pyrolytic aqueous phase. Biomass Bioenergy 2000;19:337–50. [13] Beis SH, Onay O, Kockar OM. Fixed-bed Fixed-bed pyrolysis of safflower safflower seed: influence of pyrolysis parameters on product yields and compositions. Renew Energy 2002;26:21–32. [14] Ozcimen D, Karaosmanoglu F. Production and characterization of  bio-oil bio-oil and biochar biochar from from rapesee rapeseed d cake. cake. Renew Renew Energy Energy 2004 2004;29: ;29: 779–87. [15] Scahill J, Diebold JP, Feik C. Removal of residual char fines from pyrolysis vapors by hot gasification. In: Bridgewater AV, Bock DGB, editors. editors. Developmen Developments ts in thermochem thermochemical ical biomass conversion. conversion. London: Blackie Academic Academic and Professional; Professional; 1996. p. 253–66. [16] Guo Y, Wang Y, Wei F, et al. Research progress in biomass flash pyrolysis technology for liquids production. Chem Ind Eng Progr 2001;8:13–7. [17] Wang SR, Luo ZY, Tan H, et al. The analyses analyses of characteristics characteristics biooil produced from biomass flash pyrolysis. J Eng Thermophys 2004; 25(6):1049–52. [18] Adam J, Blazso´   M, Me´ sza´ ros E, et al. Pyrolysis of biomass in the presen presence ce of Al-MCM Al-MCM-41 -41 type catalys catalysts. ts. Fuel Fuel 2005 2005;84( ;84(12– 12–13) 13):: 1494–502. [19] [19] Zhang Zhang SP, SP, Yan Yan YJ, YJ, Ren Ren ZW, ZW, et al. Analy Analysis sis of liqui liquid d produ product ct obtained by the fast pyrolysis of biomass. J Chin Sci Technol 2001; 27:666–8. [20] Peng WM, Wu QY. Production of fuels from biomass by pyrolysis. New Energy Sour 2000;22(11):39 2000;22(11):39–44. –44. [21] Pindoria RV, Lim J-Y, Hawkes Hawkes JE, et al. Structural Structural characterizatio characterization n of biomass pyrolysis tars/oils from eucalyptus wood wastes: effect of  H2   pressure and samples configuration. Fuel 1997;76(11):1013–23. [22] Pindoria RV, Megaritis A, Herod AA, et al. A two-stage fixed-bed reactor for direct hydrotreatment of volatiles from the hydropyrolysis

of biomass: biomass: effect effect of cataly catalyst st temper temperatur ature, e, pressu pressure re and cataly catalyst st ageing time on product characteristics. Fuel 1998;77(15):1715–26. [23] Zhang SP, Yan Yongjie, Li T, et al. Upgrading of liquid fuel from the pyrolysis of biomass. biomass. Bioresour Bioresour Technol Technol 2005;96:545–50 2005;96:545–50.. [24] Elliott Elliott DC, Neuens Neuensch chwan wander der GG. Liquid Liquid fuel fuel by low-se low-sever verity ity hydrotreating of biocrude. Developments in thermochemical biomass conversion. London: Blackie Academic and Professional; 1996. p. 611–21. [25] Senol OI, Viljava TR, Krause AOI. Hydrodeoxygenation of methyl esters esters on sulphide sulphided d NiMo/ NiMo/c-Al2O3   and CoMo/ CoMo/c-Al2O3   catalysts. Catal Today 2005;100(3–4):331–5. [26] Nokkosm Nokkosmaki aki MI, Kuoppal Kuoppalaa ET, Leppam Leppamaki aki EA, et al. Catalyt Catalytic ic conversion of biomass pyrolysis vapours with zinc oxide. J Anal Appl Pyrol 2000;55:119–31. [27] Adjaye JD, Bakhshi NN. Production of hydrocarbons by catalytic upgrading of a fast pyrolysis bio-oil. part I: conversion over various catalysts. Fuel Process Technol 1995;45(3):161–83. [28] Adjaye JD, Bakhshi NN. Production of hydrocarbons by catalytic upgrading of a fast pyrolysis bio-oil. Part II: Comparative catalyst performance performance and reaction reaction pathways. pathways. Fuel Process Technol 1995; 45(3):185–202. [29] Guo XY, Yan YJ, Ren ZW. The using and forecast forecast of catalyst in biooil upgrading. Acta Energiae Solaris Sin 2003;124(12):20 2003;124(12):206–12. 6–12. [30] [30] Chia Chiara ramo monti nti D, Bonini Bonini M, Frati Fratini ni E, et al. al. Deve Develop lopme ment nt of  emulsions from biomass pyrolysis liquid and Diesel and their use in engines—Part 1: Emulsion production. Biomass Bioenergy 2003;25: 85–99. [31] [31] Chia Chiara ramo monti nti D, Bonini Bonini M, Frati Fratini ni E, et al. al. Deve Develop lopme ment nt of  emulsions from biomass pyrolysis liquid and Diesel and their use in engines—Part 2: Tests in Diesel engines. Biomass Bioenergy 2003;25: 101–11. [32] Ikura Ikura M, Stanciu Stanciules lescu cu M, Hogan Hogan E. Emulsifi Emulsificat cation ion of pyrolysi pyrolysiss derived derived bio-oil in Diesel Diesel fuel. Biomass Bioenergy 2003;24:221–32. 2003;24:221–32. [33] Wang D, Czernik S, Montane´   D, et al. Biomass Biomass to hydroge hydrogen n via pyrolysis and catalytic steam reforming of the pyrolysis oil and its fractions. fractions. Ind Eng Chem Res 1997;36:1507–1 1997;36:1507–18. 8. [34] Wang Wang D, Czermi Czermik k S, Chorne Chornett E. Produc Productio tion n of hydroge hydrogen n from biomass by catalytic steam reforming of fast pyrolytic oils. Energy Fuels 1998;12:19–24. [35] Czernik S, French R, Feik C, et al. Hydrogen by catalytic steam reformi reforming ng of liquid liquid byproduc byproducts ts from biomass biomass thermo thermocon conver version sion processes. processes. Ind Eng Chem Res 2002;41:4209–1 2002;41:4209–15. 5. [36] Garcia L, French R, Czernik S, et al. Catalytic steam reforming of  bio-oils for the production of hydrogen: effects of catalyst composition. Appl Catal A: Gen 2000;201:225–39. [37] Takanabe K, Aika K, Seshan K, et al. Sustainable hydrogen from bio-oil–steam reforming of acetic acid as a model oxygenate. J Catal 2004;227:101–8. [38] Bridgwater AV, Meier D, Radlein D. An overview of fast pyrolysis of  biomass. Org Geochem 1999;30:1479–93.