Fractional characterization of a bio-oil derived from rice husk

Fractional characterization of a bio-oil derived from rice husk

b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 6 7 1 e6 7 8 Available at www.sciencedirect.com http://www.elsevier.com/locate/biombioe Fract...

691KB Sizes 3 Downloads 29 Views

b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 6 7 1 e6 7 8

Available at www.sciencedirect.com

http://www.elsevier.com/locate/biombioe

Fractional characterization of a bio-oil derived from rice husk Rui Lu a, Guo-Ping Sheng a,*, Yan-Yun Hu b, Ping Zheng b, Hong Jiang a, Yong Tang a, Han-Qing Yu a a b

Department of Chemistry, University of Science and Technology of China, 96 Jinzhai Road, Hefei 230026, China Anhui Entry-Exit Inspection and Quarantine Bureau, Hefei 230061, China

article info

abstract

Article history:

Bio-oils usually contain many types of compounds with various chemical properties.

Received 15 February 2010

A bio-oil sample derived from rice husk through rapid pyrolysis was fractioned using

Received in revised form

solvent- or solid-extraction techniques based on their various properties. Ultra-

15 September 2010

violetevisible spectroscopy, three-dimensional excitationeemission matrix (EEM) fluores-

Accepted 18 October 2010

cence spectroscopy and Fourier transform infrared spectroscopy were used to characterize

Available online 11 November 2010

their various spectral properties for further understanding the characteristics of the bio-oil. Bio-oil mostly contains many aromatic ring components, acidic polar fractions, few weak-

Keywords:

and non-polar components. The results all show that the main compounds and functional

Rice husk

groups in the various bio-oil fractions were different and depended on the fractionation

Pyrolysis

methods. The compositions of the bio-oil fractions were also analyzed with a gas chro-

Bio-oil

matography/mass spectrometry (GC/MS) method. The consistency of the results obtained

Fluorescence spectroscopy

from the spectrometric methods with the GC/MS method indicates that the spectrometric

FTIR

methods have a good potential for rapid and effective characterization of bio-oils. ª 2010 Elsevier Ltd. All rights reserved.

GC/MS

1.

Introduction

Various technologies, including pyrolysis [1,2], gasification [3,4], and biochemical methods [5], have been used to convert biomass to energy. Rapid pyrolysis of biomass has attracted growing interests, as it is able to offer significant logistical and hence economic advantages over other thermal conversion processes. The liquid product from pyrolysis, also named as biooil, can be stored until required or readily transported to where it can be most effectively utilized [6,7]. It can be used both as an energy source and a feedstock for chemical production [8]. The design and operation of the rapid pyrolysis process for biomass have been well studied. Yields of bio-oil from wood, paper and other biomass depended on the relative amount of cellulose and lignin in the raw materials [9]. The bio-oils contained numerous oxygenated organic compounds, including acids, alcohols, aldehydes, ketones, substituted phenols, and

other complex oxygenates derived from carbohydrates and lignin in biomass [9]. The bio-oil compositions are usually measured using gas chromatography/mass spectrometry (GC/ MS) [10], and/or pyrolysis-GC/MS [11]. Because of the complex compositions of bio-oils, various separation methods have been used to obtain bio-oil fractions with different characteristics. The common used method for such a fractionation is solvent-fractionated by a column elution process based on the polarity [12,13] and solubility [14]. The bio-oils are also distilled based on the boiling point to obtain different fractions [15]. The conventional methods for bio-oil analysis include column chromatography, elemental analysis, 1H NMR, gas chromatography [16]. Most of these methods are time-consuming, or need special pretreatments. Thus, effective and sensitive methods should be proposed for the rapid bio-oil analysis. Since spectroscopic techniques are simple, fast, non-destructive, reproducible, high sensitive and

* Corresponding author. Tel.: þ86 551 3607453; fax: þ86 551 3601592. E-mail address: [email protected] (G.-P. Sheng). 0961-9534/$ e see front matter ª 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.biombioe.2010.10.017

672

b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 6 7 1 e6 7 8

accurate, they are appropriate for characterizing complex samples. However, the spectral characteristics of bio-oil fractions have not been well elucidated yet, although the biooil has been characterized using Fourier transform infrared (FTIR) spectroscopy. Bio-oil contains large quantities of aromatic compounds, which can fluorescence after excited at some excitation wavelengths. The overall fluorescence characteristics of the various fluorescent components in the mixed bio-oil samples can be acquired by three-dimensional excitationeemission matrix (EEM) fluorescence spectroscopy. EEM fluorescence spectroscopy is a rapid, selective, sensitive and non-destructive technique. Its advantage is that information about the fluorescence characteristics can be acquired entirely [17]. EEM fluorescence spectral characteristics can be regarded as an overall “fingerprint” of bio-oil samples, which covers different fluorophores in bio-oils. It can provide information about the concentration of aromatic ring systems in oligomeric structures. It can also be employed to distinguish the fluorescence compounds present in the complex mixtures from various origins. In addition, this spectroscopy can be used to explore the structural features of tar and other coal/biomass-derived liquids. In spite of its well-known quantitative natures, the EEM fluorescence spectroscopy has not been used for charactering the chemical components in bio-oils. In this work, the characteristics of the bio-oil derived from rice husk were quantified. Several spectroscopic analytical techniques, such as EEM, ultravioletevisible (UVeVis) and FTIR techniques were used to characterize the bio-oil fractioned by various separation methods. In addition, the GCeMS was used to confirm the spectroscopic analytical results. A rapid and entire characterization of the bio-oil fractions with various properties would be useful to further understand the bio-oil compositions and characteristics, and to develop an effective monitoring method from the biomass pyrolysis process.

2.

Materials and methods

2.1.

Bio-oil production and fraction

The bio-oil samples derived from rice husk were produced by rapid pyrolysis in a laboratory-scale auto thermal fluidizedbed reactor with a capacity of 120 kg/h. The reactor mainly consists of a hopper, two screw feeders, an electric heater, a fluidized-bed reactor, two cyclones, a condenser, and an oil pump, as well as some thermocouples and pressure meters. The detailed information about the reactor has been described by Zheng et al. [18]. The bio-oil samples used in this study contained 25.2% of water, and the elemental analysis showed that it contained 41.7% C, 7.7% H, 50.3% O, 0.3% N and 0.2% S in weight percentages. The bio-oil sample contained an average of 2 wt % of solid residue, which was mainly composed of charcoal fines. Before the fractionation of bio-oil using solvent- [19] or solidextraction [20] techniques, the fine particles were removed using a 0.45 mm membrane. After filtration, water was removed by adding sufficient amount of anhydrous sodium sulfate. After that, the bio-oil was fractioned using solvent- or

solid-extraction techniques. The main steps for bio-oil fraction are illustrated in Fig. 1. The bio-oil was extracted using three solvents with various polarities, e.g., methanol, ethyl acetate and acetone. This resulted in three bio-oil fractions, which were defined as Samples 1, 2 and 3, respectively. The compounds in the bio-oil were also fractioned according to their acidic or basic characteristics using a HydrophileeLipophile Balance (HLB) solid phase extraction column in series. The Oasis HLB column (60 mg, 3 mL; Waters Inc., USA) was initially conditioned with 1 mL of methanol and 1 mL of ultrapure water, then 0.5 mL of bio-oil was loaded on an HLB column, and then washed with 10 mL of 5% acetic acid/ 10% methanol/water solution (pH ¼ 2.23). The eluate was collected and extracted using 5 mL of ethyl acetate to obtain the alkaline bio-oil fraction (Sample 4). The HLB column was then eluted using 10 mL of 5% ammonia/10% methanol/water solution (pH ¼ 11.47). The eluate was also collected and extracted using ethyl acetate to obtain the acidic bio-oil fraction (Sample 5). At last, 5 ml of ethyl acetate was used to elute the HLB column to obtain the neutral fraction (Sample 6). The compounds in the bio-oil were also fractioned based on their polarities using C18 solid phase extraction columns in parallel. A C18 column (500 mg, 3 mL; Waters Inc., USA) was initially conditioned with 10 mL of methanol and equilibration with 30 mL of ultrapure water. After the C18 column condition, the bio-oil was loaded on the columns in parallel. Thereafter, the columns were eluted respectively with methanol, ethyl acetate and cyclohexane, to collect the bio-oil fractions with various polarities, which were defined as Samples 7e9, respectively.

2.2.

UVeVis, EEM and FTIR spectroscopy

The UVeVis spectra of the bio-oil fractions were recorded using UV-2450 spectrometer (Shimadzu Co., Japan). All measurements were conducted using a 1 cm path length cuvette at ambient temperatures using corresponding solvent as blanks. All EEM spectra were measured using a luminescence spectrometry (LS-55, PerkineElmer Co., USA). The EEM spectra were collected with subsequent scanning emission spectra from 250 to 500 nm at 0.5 nm increments by varying the excitation wavelength from 250 to 450 nm at 10 nm increments. Excitation and emission slits were both maintained at

Dissolve

Bio-oil

HLB

C18

Methanol

(Sample 1)

Ethyl acetate

(Sample 2)

Acetone

(Sample 3)

A eluted

Ethyl acetate extraction

(Sample 4)

A eluted

B eluted

Ethyl acetate extraction (Sample 5)

A eluted

B eluted

Ethyl acetate eluted

(Sample 6)

Methanol eluted

(Sample 7)

Ethyl acetate eluted

(Sample 8)

Cyclohexane eluted

(Sample 9)

Fig. 1 e Fractionation procedures for the bio-oil. (A: 5% acetic acid/10% methanol/water solution; B: 5% ammonia/ 10% methanol/water solution).

673

b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 6 7 1 e6 7 8

3.

Results and discussion

3.1.

UVeVis spectra

The UVeVis spectra of the various bio-oil fractions are shown in Fig. 2. The absorption at 205 nm is the characteristic E2 absorption strip of benzene ring, while the absorption at 270 nm is the B strip, suggesting that there are many aromatic ring components in the bio-oil. Samples 1e3 were the solutions of the bio-oil dissolved in methanol, ethyl acetate and acetone, respectively. Comparison among their UVeVis spectra indicates that no difference among peak positions was observed, but that the peak intensities were significantly different. The peak intensity of Sample 1 was the highest, while that of Sample 2 was the lowest. This indicates that most of the bio-oil compounds could be dissolved in methanol, while that little was dissolved in ethyl acetate. Since the solvents polarity was in order of methanol > acetone > ethyl acetate, this implies that most compounds in the bio-oil were polar ones. Samples 4e6 were the alkaline, acidic, neutral fractions of the bio-oil. Compared among their UVeVis spectra demonstrates that Sample 4 had a strong absorption at 270 nm, while Sample 5 almost had no absorption at this position. Samples 7e9 were the bio-oil fractions with various polarities. Sample 8 had a strong absorption peak than Sample 7, and their positions were similar to each other. This indicates that Sample 8 contained more polar and medium polar components. Sample 9 was much different, mainly attributed

Sample 2

3

Sample 3 2

1

GCeMS analysis

0 4

b

3

A (a.u.)

The compositions of bio-oil fractions were analyzed using GCeMS. The separation was made using a DB5-MS fused silica capillary column (30 m  0.25 mm) with a film thickness of 0.25 mm by an Agilent 6890 GC. The GC oven temperature was held at 50  C for 2 min and then programmed to attain a temperature of 270  C at 5  C/min. The injector temperature was 280  C with split ratio 10:1. Helium was used as the carrier gas with a flow rate of 1 cm3 min1. The end of the column was directly introduced into the ion source of a HewlettePackard model 5973 series mass selective detector that was operated in an electron impact ionization mode. Typical mass spectrometry operating conditions were as follows: temperature of transfer line, 280  C; temperature of ion source, 250  C; and electron energy, 70 eV. Data acquisition was performed with a computer-based G1034C Chemstation software and an NBS mass spectra laboratory database. Computerized matches were manually evaluated through comparing mass spectra.

Sample 1

a

Sample 4 Sample 5 Sample 6

2

1

0 4

c

3

A (a.u.)

2.3.

4

A (a.u.)

10 nm, and the scanning speed was set at 1200 nm/min for all measurements. The software Matlab 7.1 (MathWorks Inc., USA) was employed for handling EEM data. EEM spectra are illustrated as the elliptical shape of contours. The bio-oil FTIR spectra were recorded on a Bruker VERTEX 70 FTIR spectrophotometer at the wave number of 4000e400 cm1 with samples coated on KRS-5 crystal. The resolution was 4 cm1. Each sample was scanned 32 times to get an average spectrum.

Sample 7 Sample 8 Sample 9

2

1

0 200

300

400

Wavelength (nm) Fig. 2 e UVeVis absorption profiles of the different bio-oil fractions as suggested in Fig. 1.

to the fact only non-polar compounds could be eluted by cyclohexane from C18 column, whose amount was limited in the bio-oil samples.

3.2.

EEM fluorescence spectra

The EEM spectra of various bio-oil fractions dissolved respectively in methanol, ethyl acetate and acetone are illustrated in Fig. 3. The peak with the highest intensity in the EEM spectrum was at excitation/emission (Ex/Em) of 320/375 nm, 300/360 nm and 310/370 nm for the bio-oil fraction dissolved in methanol, ethyl acetate and acetone, respectively. As the solvent polarity increased, the excitation and emission wavelength increased simultaneously. This is because different fractions dissolved in the three solvents mainly contained different compounds. As shown in Fig. 4, the EEM spectrum of the alkaline fractions had a strongest Ex/Em peak at 260/350 nm, while the

674

b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 6 7 1 e6 7 8

strongest Ex/Em peak was at 280/320 nm and 290/360 nm for the acidic and neutral fractions, respectively. The emission wavelength in the EEM spectra of the alkaline bio-oil fraction was larger than that of the acidic one. The intensity of the peak for the acidic bio-oil fraction (280/320 nm) was the strongest, while that for the alkaline one (260/350 nm) was the weakest. This suggests that the bio-oil samples mainly contained acidic fractions. Fig. 5 shows the EEM spectra of the bio-oil fractions with various polarities eluted with different polar solvents from C18 columns. The EEM spectrum of the fraction eluted with methanol had a strongest peak at 310/360 nm; while at 310/

360 nm for the fraction eluted with ethyl acetate and at 280/ 320 nm for the fraction eluted with cyclohexane, respectively. The polar components in the bio-oil could be eluted with methanol. The polar and middle-polar components could be eluted with ethyl acetate. Fluorescence peak intensity of the EEM spectrum for the ethyl acetate-eluting fraction was slightly stronger than that of the fraction eluted with methanol, indicating that the main components of the bio-oil were polar components. The bio-oil sample contained few weakand non-polar components, which could be eluted with cyclohexane. 450

900

a Ex (nm)

600 500

350

700 600

Ex (nm)

700

800

400

800

400

900

a

450

500

350

400 300

400

300

200

300

300

100

200

250 250

100

250 250

0

300

350

400

450

500

450

500

350

Ex (nm)

Ex (nm)

600

500

550 900 800 700 600 500

350

400 300

400

300

300

300

250 250

200 100

200 100

250 250

0

300

350

400

450

500

550

0

300

350

400

450

500

550

450

450

c

500

350

Ex (nm)

600

800

400

800 700

900

c

900

400

Ex (nm)

450

400

800 700

400

b

900

b

350

450

550

400

0

300

700 600 500

350

400 300

400 300

300

200 100

200 100

250 250

300

0

300

350

400

450

500

550

Em (nm) Fig. 3 e EEM fluorescence spectra of the bio-oils dissolved in (a) methanol, (b) ethyl acetate and (c) acetone.

250 250

0

300

350

400

450

500

550

Em (nm) Fig. 4 e EEM fluorescence spectra of the different bio-oil fractions washed with 5% of acetic acid in (a) 5% of methanol/water, (b) 5% of ammonia and (c) 5% of methanol/ water and ethyl acetate.

675

b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 6 7 1 e6 7 8

3.3.

FTIR spectra

The FTIR spectra of the bio-oil samples are shown in Fig. 6, and the peaks of the functional groups in the bio-oil sample are listed in Table 1. There were broad OeH stretching vibrations between 3600 and 3400 cm1 and the OeH bending vibrations between 1350 and 1260 cm1. The aromatic CeH stretching vibrations between 3040 and 3000 cm1 and C]C stretching vibrations at 1600, 1500, 1580 and 1450 cm1 indicate the aromatic structure in the bio-oil samples. The symmetric and asymmetric CeH stretching vibrations between 2980 and

450 900

a

800

400

700

Ex (nm)

600 500

350

400 300

300

200 100

250 250

0

300

350

400

450

500

550

450 900

b

800

400

700

2850 cm1 could be attributed to the aliphatic eCH3 and eCH2 groups. Moreover, the CeH bending vibrations of aliphatic CH3 and CH2 groups at 1465, 1450 and 1380 cm1 could also been found in the FTIR spectra of the bio-oil samples. The C]O stretching vibrations at 1770e1700 cm1 also suggest the presence of ketones and aldehydes in the bio-oil. The absorbance between 1625 and 1590 cm1, which represents the C]C stretching vibrations, implies the presence of alkenes and aromatics. The spectra of various bio-oil fractions in Fig. 6 are compared. The FTIR spectra of the bio-oil fractions dissolved in various solvents (Samples 1e3) were similar, while their peak absorbances were different. Compared with the FTIR spectra of the alkaline, acidic and neutral components of the bio-oil samples, the acidic and neutral components had strong absorptions at 1700 cm1 and 1250 cm1. This suggests that the acidic and neutral fractions mainly contained carbonyl and ethers compounds. The alkaline components had strong absorption at 2950 cm1 and 2850 cm1, implying that this fraction mainly contained hydrocarbon materials. Comparison among the different components eluted with polar solvents suggests that the large polar components had strong absorptions at 3500 cm1, 2950 cm1, 2850 cm1, 1070 cm1 and 700e900 cm1. This demonstrates that this fraction contained vinyl, ethers and benzene ring compounds, and most likely contained acidic compounds. The middle-polar components had strong absorptions at 3500 cm1, 3000 cm1 and 1750 cm1, suggesting that this fraction contained carbonyl compounds. The weak polar components had strong absorption at 2950 cm1, 2850 cm1 and weak absorption at 3500 cm1, showing that this fraction mainly contained alkane or cycloalkane derivatives.

Ex (nm)

600 500

350

400 300

300

3.4.

GCeMS results

The identification of all the fractions in different samples was performed with GC/MS (Fig. 7). In the GC spectrum, more

200 100

250 250

0

300

350

400

450

500

550

Wave number (cm1)

450 900

c

800

400

700

Ex (nm)

600 500

350

400 300

300

200 100

250 250

Table 1 e FTIR peaks of the function groups of the bio-oil fractions.

0

300

350

400

450

500

550

Em (nm) Fig. 5 e EEM fluorescence spectra of the different bio-oil fractions washed with: (a) methanol; (b) ethyl acetate; and (c) cyclohexane.

3600e3400 3040e3000 2980e2930 2870e2850 1770e1700 1625e1590 1600,1580,1450 1450,1380 1465 1275e1070 1350e1260 900e700 740e720 730e675

Infrared absorption Stretching OeH Stretching aromatic ring CeH vas CeH vs CeH Stretching C]O Stretching C]C Stretching aromatic ring C]C d0 CeH d00 CeH Stretching CeOeC Bending OeH Substituents in aromatic ring Bending out of the plane CeH Bending out of the plane C]H

vas is the symmetrical CeH stretching vibration of aliphatic CH3 and CH2, vs is the symmetrical CeH stretching vibration of aliphatic CH3 and CH2, d0 is the symmetrical CeH bending vibration of aliphatic CH3, d00 is the scissoring bending vibration of aliphatic CH3 and CH2.

676

b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 6 7 1 e6 7 8

Fig. 6 e FTIR spectra of the different bio-oil fractions.

Fig. 7 e Total ion chromatogram of the different bio-oil fractions.

677

b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 6 7 1 e6 7 8

Table 2 e GC/MS analysis of the compounds in the bio-oils dissolved in methanol (RT is used for “retention time (min)”). RT 2.22 2.73 4.09 5.71 6.83 7.54 7.82 8.48 8.61 9.63 9.80 11.17 11.43 11.64 11.97 12.59 12.96 13.88 14.72 15.28 16.00

Compounds

RT

Compounds

Formic acid Acetic acid Pyridine Furfural Tetrahydro-2,5-dimethoxyfuran 2-Methyl-2-cyclopenten-1-one 2(5H) Furanone 5-Methyl-2(5H)furanone Dihydro-3-methylene-2,5-furandione 3-Methyl-2(5H)furanone Phenol 2-Hydroxy-3-methyl-2-cyclopenten-1-one 2,3-Dimethyl-2-cyclopenten-1-one 4-Methyl-5H-furan-2-one 2-Methylphenol 3-Methylphenol 2-Methoxyphenol 3-Ethyl-2-hydroxy-2-chclopenten-1-one 2,4-Dimethyphenol 4-Ethylphenol 2-Methoxy-4-methylphenol

16.46 18.08 18.44 18.95 19.40 20.41 20.56 20.81 21.40 21.73 21.88 22.94 23.92 25.40 27.73 28.00 28.85 29.63 34.06 37.37

1,2-Benzenediol 3-Methyl-1,2-benzenediol 4-Ethyl-2-methoxyphenol 4-Ethyl-1,2-benzenediol 2-Methoxy-4-vinylphenol 2,6-Dimethoxyphenol 2-Methoxy-3-(2-propenyl)phenol 2-Methoxy-4-propylphenol 4-Ethylcatechol Vanillin 2-Methoxy-4-(1-propenyl)phenol 2-Methoxy-4-(1-propenyl)phenol 1-(4-Hydroxy-3-methoxyphenyl)-ethanone D-Allose 4-(Ethoxymethyl)-2-methoxyphenol 4-Hydroxy-3,5-dimethoxy-benzaldehyde 2,6-Dimethoxy-4-(2-propenyl)-phenol 1-(4-Hydroxy-3,5-dimethoxyphenyl)-ethanone n-Hexadecanoic acid (Z )-6-Octadecenoic acid

than 30 peaks were observed. MS spectra of these peaks were identified using the MS database. The compositions of the fractions dissolved in methanol analyzed by GC/MS are listed in Table 2. The main compounds in the bio-oil were furan derivatives, phenolic derivatives, alcohol derivatives and acid derivatives. Song et al. [21] and Wang et al. [22] had used GC/MS to investigate the compositions of the bio-oil from fast pyrolysis of rice husk. They also found that benzene derivatives, phenol derivatives, alkanes, cycloalkanes and aromatic hydrocarbons were the main compounds in the bio-oils. Compared to Sample 1, Sample 2 contained few ketones and phenols only, while Sample 3 had few acids and alcohols. This is because different solvents contained different fractions. The main compounds identified by GC/MS were alcohols, ketones and aldehydes in Sample 4, while they were organic acids and phenols in Sample 5 and were phenols and some alkanes in Sample 6. Compared with Sample 9, Sample 7 did not contain alkane (i.e., cyclopentane, cyclohexane, norbornane, undecane), ketones and phenols. Sample 8 had no alkane, but contained ketones and phenols. The methanoleluted fraction had no ketones or phenols, compared to the ethyl acetate-L-eluted one. Both methanol- and ethyl acetateeluted fractions had no alkane derivatives. The bio-oil dissolved in methanol contained the greatest number of compounds, where the bio-oil dissolved in ethyl acetate contained the least. This might be the reason why Sample 1 had the strongest intensity in the UVeVis spectra among the various bio-oil fractions. Comparison among Samples 1e9 shows that the locations of the fluorescence peaks were different. The shift was also attributed to the different compounds in them. Different fractions mainly contained different compounds. The alkaline fractions contained many alcohols but no fluorescence, whereas the acidic fractions had some phenols, which have strong fluorescence compared to alcohols.

4.

Conclusions

Chemical compositions of bio-oil from fast pyrolysis of rice husk were fractioned and analyzed with the spectrometric and GC/MS methods. The following conclusions could be drawn:  The bio-oil fractions dissolved in various solvents had similar components with a slight difference in contents. The FTIR and EEM fluorescence spectroscopy results show that their main components were strongly polar and could be identified.  There are good consistency between the spectral characteristics and the GC/MS results of the bio-oils. The spectroscopic methods could be used as a powerful tool to analyze the bio-oil samples in-situ.  Molecular spectroscopy is an efficient technique for bio-oil characterization. The establishment of the relationship between the bio-oil properties and its spectral characteristics would be useful for the on-line monitoring for bio-oil production process.

Acknowledgments The authors wish to thank the National Hi-Technology Development 863 Program of China (2007AA061502 and 2007AA06A407) for the partial support of this study.

references

[1] Bridgwater AV. Production of high grade fuels and chemicals from catalytic pyrolysis of biomass. Catal Today 1996;29: 285e95.

678

b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 6 7 1 e6 7 8

[2] Garcia L, Salvador ML, Arauzo J, Bilbao R. Catalytic pyrolysis of biomass: influence of the catalyst pretreatment on gas yields. J Anal Appl Pyrolysis 2001;58:491e501. [3] Aznar MP, Caballero MA, Corella J, Molina G, Toledo JM. Hydrogen production by biomass gasification with steamO-2 mixtures followed by a catalytic steam reformer and a CO-shift system. Energy Fuels 2006;20:1305e9. [4] Radmanesh R, Chaouki J, Guy C. Biomass gasification in a bubbling fluidized bed reactor: experiments and modeling. AIChE J 2006;52:4258e72. [5] McKendry P. Energy production from biomass: conversion technologies. Bioresour Technol 2002;83:47e54. [6] Bridgwater AV, Meier D, Radlein D. An overview of fast pyrolysis of biomass. Org Geochem 1999;30:1479e93. [7] Czernik S, Bridgwater AV. Overview of applications of biomass fast pyrolysis oil. Energy Fuels 2004;18:590e8. [8] Bridgwater AV. Renewable fuels and chemicals by thermal processing of biomass. Chem Eng J 2003;91:87e102. [9] Mohan D, Pittman CU, Steele PH. Pyrolysis of wood/ biomass for bio-oil: a critical review. Energy Fuels 2006;20: 848e89. [10] Wu LB, Guo SP, Wang C, Yang ZY. Direct deoxy-liquefaction of poplar leaves to biopetroleum with two kinds of catalysts. Ind Eng Chem Res 2008;47:9248e55. [11] Boateng AA, Hicks KB, Vogel KP. Pyrolysis of switchgrass (Panicum virgatum) harvested at several stages of maturity. J Anal Appl Pyrolysis 2006;75:55e64. [12] Sipila K, Kuoppala E, Fagernas L, Oasmaa A. Characterization of biomass-based flash pyrolysis oils. Biomass Bioenerg 1998; 14:103e13.

[13] Ba TY, Chaala A, Garcia-Perez M, Roy C. Colloidal properties of bio-oils obtained by vacuum pyrolysis of softwood bark. Storage stability. Energy Fuels 2004;18:188e201. [14] Murwanashyaka JN, Pakdel H, Roy C. Separation of syringol from birch wood-derived vacuum pyrolysis oil. Sep Purif Technol 2001;24:155e65. [15] Demirbas A. Production of gasoline and diesel fuels from biomaterials. Energy Sources Part A-Recovery Util Environ Eff 2007;29:753e60. [16] Uzun BB, Putun AE, Putun E. Rapid pyrolysis of olive residue. 1. Effect of heat and mass transfer limitations on product yields and bio-oil compositions. Energy Fuels 2007;21:1768e76. [17] Sheng GP, Yu HQ. Characterization of extracellular polymeric substances of aerobic and anaerobic sludge using 3dimensional fluorescence spectroscopy. Water Res 2006;40: 1233e9. [18] Zheng JL, Zhu XF, Guo QX, Zhu QS. Thermal conversion of rice husks and sawdust to liquid fuel. Waste Manage 2006;26: 1430e5. [19] Chan JKS, Duff SJB. Methods for mitigation of bio-oil extract toxicity. Bioresour Technol 2010;101:3755e9. [20] Torri C, Fabbri D. Application of off-line pyrolysis with dynamic solid-phase microextraction to the GCeMS analysis of biomass pyrolysis products. Microchem J 2009;93:133e9. [21] Song QH, Nie JQ, Ren MG, Guo QX. Effective phase separation of biomass pyrolysis oils by adding aqueous salt solutions. Energy Fuels 2009;23:3397e412. [22] Wang C, Pan JX, Li JH, Yang ZY. Comparative studies of products produced from four different biomass samples via deoxy-liquefaction. Bioresour Technol 2008;99:2778e86.