J. Anal. Appl. Pyrolysis 86 (2009) 161–167
Contents lists available at ScienceDirect
Journal of Analytical and Applied Pyrolysis
journal homepage: www.elsevier.com/
Microwave-assisted pyrolysis of biomass: Catalysts to improve product selectivity
Yiqin Wan a,b, Paul Chen b,*, Bo Zhang b, Changyang Yang b, Yuhuan Liu a,b, Xiangyang Lin a,b, Roger Ruan a,b
Biomass Energy Center and State Key Laboratory of Food Science, Nanchang University, Nanchang, China Center for Biore?ning and Department of Bioproducts and Biosystems Engineering, University of Minnesota, 1390 Eckles Ave., St. Paul, MN 55108, United States
A R T I C L E I N F O
A B S T R A C T
Article history: Received 17 November 2008 Accepted 13 May 2009 Available online 29 May 2009 Keywords: Pyrolysis Microwave Bio-oils Catalysis Catalysts Corn stover Aspen Biofuels
This study was intended to evaluate the effects of catalysts on product selectivity of microwave-assisted pyrolysis of corn stover and aspen wood. Metal oxides, salts, and acids including K2Cr2O7, Al2O3, KAc, H3BO3, Na2HPO4, MgCl2, AlCl3, CoCl2, and ZnCl2 were pre-mixed with corn stover or aspen wood pellets prior to pyrolysis using microwave heating. The thermal process produced three product fractions, namely bio-oil, gas, and charcoal. The effects of the catalysts on the fractional yields were studied. KAc, Al2O3, MgCl2, H3BO3, and Na2HPO4 were found to increase the bio-oil yield by either suppressing charcoal yield or gas yield or both. These catalysts may function as a microwave absorbent to speed up heating or participate in so-called ‘‘in situ upgrading’’ of pyrolytic vapors during the microwave-assisted pyrolysis of biomass. GC–MS analysis of the bio-oils found that chloride salts promoted a few reactions while suppressing most of the other reactions observed for the control samples. At 8 g MgCl2/100 biomass level, the GC–MS total ion chromatograms of the bio-oils from the treated corn stover or aspen show only one major furfural peak accounting for about 80% of the area under the spectrum. We conclude that some catalysts improve bio-oil yields, and chloride salts in particular simplify the chemical compositions of the resultant bio-oils and therefore improve the product selectivity of the pyrolysis process. ? 2009 Elsevier B.V. All rights reserved.
1. Introduction Declining supplies of fossil energy resources and adverse impacts of fossil energy uses on the global environment have prompted strong interests in renewable energy. A wide array of renewable energy technologies is being researched. Among them are biomass-based energy technologies such as processes converting corn to ethanol, soybean to biodiesel, lignocellulosics to bioethanol or bio-oils. Rising food prices have caused serious concerns over corn to bioethanol and soybean to biodiesel approaches. Increasing attention is drawn to conversion of lignocellulosics to biofuels such as ethanol, bio-oils, and gas. Pyrolysis, a thermochemical conversion process, is an attractive way to produce liquid fuels from solid biomass feedstock. Pyrolysis is one of the thermochemical conversion processes widely studied. Pyrolysis is normally carried out in the absence of oxygen at temperature above 400–500 8C. During pyrolysis, large molecules in biomass decompose or depolymerize at high temperature to gaseous phase leaving some solid charcoal behind . The gaseous phase consists of condensable and incondensable compounds. The condensable compounds can be cooled down to yield bio-oils and
* Corresponding author. Tel.: +1 612 625 7721; fax: +1 612 624 3005. E-mail address: firstname.lastname@example.org (P. Chen). 0165-2370/$ – see front matter ? 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jaap.2009.05.006
chemicals. Like all other renewable technologies, biomass pyrolysis is facing many technical challenges. For example, the liquid fuels produced from pyrolysis of biomass are complex in chemical composition and highly unstable in terms of physical consistency, chemical properties, and combustion characteristics, and often highly acidic, seriously limiting the practical use of these liquid fuels. Researchers have been seeking solutions to these problems. Post-pyrolysis upgrading is a common approach to stabilizing biooils. Upgrading processes may involve physical or chemical methods. Solvent blending and ?ltration are physical upgrading methods [2–6], but they are not very effective. We recently developed a two-step fractionation procedure to separate bio-oil to light oil, heavy oil, and chemicals . The light oil fraction, with no combustion residue and a much lower boiling point, is a good alternative fuel for engines. However, this procedure produces only about 25% light oil fraction. Chemical upgrading employs catalytic cracking and reforming such as hydrotreatment and thermal cracking to de-oxygenate the bio-oil. Recent research in this area has made incremental advances in improving the bio-oil quality. Nevertheless, use of pyrolytic oils remains unpractical due to the aforementioned problems. Pyrolysis of biomass occurs at high temperature and is not well understood. Some suggest a ?ve-step process : (1) biomass is heated, (2) volatiles evolve from the organics, and carbonization occurs, (3) out?ow of hot volatiles and cooler unpyrolyed fuel
Y. Wan et al. / J. Anal. Appl. Pyrolysis 86 (2009) 161–167 Table 1 Properties of corn stover and aspen pellets. Properties Diameter (mm) Length Bulk density at 20 8C (kg/m3) Moisture Elemental composition (wt.%) Carbon Hydrogen Nitrogen Sulfur Oxygen (by difference) Gross heating value (MJ/kg) Chemical composition (wt.%) Lignin Cellulose Hemicellulose Others Corn stover 6.2 10–20 340 6.4 Aspen wood 4.8 4–8 240 5.9
volatiles, (4) volatiles condense to liquid (tar) with incondensable gas, and (5) autocatalytic secondary reactions (decomposition or repolymerization) take place . Some interventions in step 2 may induce and/or alter certain chemical reactions, resulting in different chemical pro?les of the volatiles. Recently, researchers became interested in in situ upgrading of the biomass pyrolysis vapors, i.e., evolved volatiles from thermal decomposition of organics react directly and immediately on catalysts pre-mixed with the biomass feedstock [10–13]. Such in situ catalytic upgrading of the biomass pyrolysis vapors is also known as catalytic pyrolysis. This approach could eventually eliminate the costly condensation and re-evaporation procedures used in upgrading traditional pyrolytic oil. Studies on a range of catalysts have been reported in the literature. Some catalysts produced more hydrocarbons than others and some not at all. Catalytic pyrolysis usually produces additional water and coke-solid residue and thus reduces the yield of the organic phase of the bio-oil. We found that chlorides favor production of light oil and especially water solubles and metaloxides favor heavy oil, and thus total oil yield, while nitrates favor gas production . Bio-oils from current catalytic pyrolysis processes are still not up to the industry standards, primarily due to the still complex chemical composition. However, a positive effect on the quality of the organic phase was noticed, and research has been directed towards the design of selective catalysts for either increasing the production of speci?c compounds (e.g. phenols) or minimizing the formation of undesirable compounds (e.g. acids, carbonyls). Zinc chloride was found to catalyze decomposition of cellulose in solution at high temperature, and chloride salts were found to promote conversion of sugars to furfural [15–17]. Some studies employed alkali and alkaline earth metal in pyrolysis and gasi?cation of different biomass [18–21]. They found that metal salts, especially alkaline earth metal chlorides such as MgCl2 and ZnCl2, increased the yield of char at low temperature but decreased the yield of char at high temperature. Our earlier studies show that chloride salts favor liquid production in microwave-assisted pyrolysis of corn stover and aspen . Therefore, it is necessary to study how catalysts change the chemical pro?les of bio-oils under microwave-assisted pyrolysis conditions. Microwave-assisted pyrolysis (MAP) is a relatively new pyrolysis technique which provides many advantages over conventional processes . The objective of the present study was to evaluate the product selectivity of microwave-assisted pyrolysis (MAP) of cellulosic materials as affected by using metal oxides and chloride salts as catalysts. The selected catalysts were mixed with the biomass materials at different ratios prior to the microwave pyrolysis. The condensable volatiles collected from the biomass pyrolysis runs were analyzed using GC–MS. 2. Experimental process 2.1. Materials Dry corn stover and aspen pellets were provided by Lone Tree Manufacturing (Bagley, MN). The major physical and chemical properties of corn stover and aspen are given in Table 1. Analytical grade chemicals MgCl2?6H2O, AlCl3?2H2O, LiCl, and CuSO4?6H2O were purchased from Mallinckrodt Baker Inc., ZnCl2 from Fisher Scienti?c Inc., MgSO4 and MgO from J.T. Baker Inc., Al2O3 from Matheson Coleman & Bell, and Ni(NO3)2, Mg(NO3)2 from Alfa Acsar. 2.2. Preparation of mixture of biomass and catalysts
40.4 5.3 1.1 0.1 53.1 24.5
45.4 4.8 0.5 0.0 49.3 17.9
18 37 27 18
19 53 27 1
calcium, zinc, and copper, and some acids prior to the pyrolysis. After early analysis, we found that MgCl2 had strong impact on the MAP products. Therefore different dosages of MgCl2 were further tested. 2.3. Microwave-assisted pyrolysis (MAP) The pyrolysis of biomass was carried out in a Panasonic NNSD787S microwave cavity oven by placing 100 g prepared samples in a quartz ?ask, which in turn was placed inside the microwave cavity. The time for microwave treatment was around 20 min. A constant power input of 875 W at the microwave frequency of 2450 MHz was used for each batch. Since monitoring temperature during microwave heating was dif?cult, the temperatures of the biomass mixtures were measured immediately after heating was terminated. The observed temperatures ranged from 450 to 550 8C. The volatile pyrolyzates were condensed with ?ve condensers. The temperature of the cooling water was about 0–5 8C. The fraction collected from the bottles connected on the bottom of the condensers is a pyrolytic liquid (bio-oil). The solid char residue was allowed to cool to room temperature before it was weighed. The condensates adhering to the interior wall of the quartz ?ask were washed with ethanol into the pyrolytic liquid collection bottle. All liquids collected were concentrated at 40 8C using a vacuum rotovap (Buchi R-141, Flawil, Switzerland) to a near constant weight, and the weight was recorded. The weight of gas product was calculated using following equation: weight of gas ? initial biomass ? pyrolytic liquid mass ? char residue mass 2.4. GC–MS analysis of bio-oils Chemical compositions of the liquid products were analyzed using an Agilent 7890-5975C gas chromatography/mass spectrometer (Santa Clara, CA) with a DB-5-MS capillary column. The GC was programmed at 45 8C for 0.5 min and then increased at 15 8C/ min to 290 8C, and ?nally held with an isothermal for 5 min. The injector temperature was 250 8C, and the injection size was 1 ml. The ?ow rate of the carrier gas (helium) was 1.2 ml/min. The ion source temperature was 230 8C for the mass selective detector. The compounds were identi?ed by comparison with the NIST Mass Spectral Database. 2.5. Data analysis
To determine the in?uence of different chemicals on the microwave-assisted pyrolysis, 100 g dry biomass was mixed with 8 g powders/crystals of oxides of sodium, potassium, magnesium,
All experiments were performed in quintuplicate. The data presented in this paper are averaged values. Statistical analysis of
Y. Wan et al. / J. Anal. Appl. Pyrolysis 86 (2009) 161–167
the data was conducted using MS Excel?. The standard deviations are presented along with the fractional yield data. 3. Results and discussion 3.1. Product fractional yields The microwave-assisted pyrolysis of corn stover and aspen wastes produced three fractions of products, i.e., bio-oil, gas, and solid charcoal. The fractional yields are shown in Fig. 1. The result shows that the catalysts used in our study affected the fractional yields to different degrees. In general, KAc, Al2O3, MgCl2, H3BO3, and Na2HPO4 increased the bio-oil yield. For KAc, the increase in bio-oil fraction is accompanied by a reduction in charcoal without a change in gas production. The situation for Al2O3 is just opposite of that for KAc, i.e., the increase in bio-oil fraction is associated mainly with a decrease in gas. For MgCl2, H3BO3, and Na2HPO4, the increase in bio-oil fraction is accompanied by reduction in both charcoal and gas but by different degrees. For MgCl2 and H3BO3, the increase in bio-oil is associated mainly with gas reduction while for Na2HPO4, the high yield of bio-oil is coincident more signi?cantly with charcoal reduction than with gas reduction. These observed differences indicate that catalysts affected bio-oil production differently. Al2O3 did not improve the thermal decomposition of biomass ef?ciency (no signi?cant change in charcoal fraction) but promoted the generation of condensable volatiles (more gaseous compounds can be condensed to liquid). KAc helped convert more solid biomass to volatiles among which the amount of noncondensable (gas) was similar to that in the control experiment with a net gain in the condensables. MgCl2, H3BO3, and NaHPO4 are
Fig. 1. Product fractional yields of microwave-assisted pyrolysis of corn stover.
the mixed cases of Al2O3 and KAc where solid biomass conversion and the ratio of condensables to non-condensables increased to different degrees. How the catalysts affected the fractional yields are unknown. We believe that both physical and chemical mechanisms are involved. The added chemical compounds may have acted as
Fig. 2. Total ion chromatograms from GC–MS analysis of pyrolytic oils from corn stover when different catalysts were used (8 g/100 g biomass). FF: furfural.
Y. Wan et al. / J. Anal. Appl. Pyrolysis 86 (2009) 161–167 Table 2 Bio-oil and charcoal fractional yields of microwave-assisted pyrolysis of corn stover and aspen pellets. Biomass Corn Stover MgCl2?6H2O (g/100 g sample) 0 2 4 8 0 2 4 8 Bio-oil (%) 37.03 ? 1.96 39.44 ? 2.25 42.70 ? 1.11 42.61 ? 1.89 35.87 ? 0.06 35.46 ? 2.58 38.97 ? 0.36 41.12 ? 1.41 Charcoal (%) 28.63 ? 0.58 28.33 ? 0.63 29.26 ? 0.36 28.01 ? 0.97 25.24 ? 0.16 30.69 ? 2.07 28.75 ? 0.25 28.04 ? 0.71
microwave absorbers and/or chemical catalysts. Earlier studies involving mixing salts or oxides with feedstock were unable to separate the microwave absorbent effect from chemical catalysis effect. For example, Monsef-Mirzai et al.  used CuO and Fe3O4 as microwave receptors for microwave-assisted pyrolysis of coal. They attributed the observed differences in the fractional yields and gas and tar compositions between CuO and Fe3O4 treated samples to heating rate. However, the chemical catalysis induced ? by CuO and Fe3O4 should not be ruled out. Dom?nguez et al.  found that the use of graphite and char as microwave absorbers did change the heating rate and fractional yields however the chemical pro?les of the liquid products remained the same regardless of the type of microwave absorber used. Our previous studies showed that varying microwave input power resulted in noticeable difference in fractional yields and gas composition but not liquid composition. In the present study, the increased bio-oil yields are probably related to the so-called ‘‘in situ upgrading’’ of the pyrolytic vapors emitted from thermal decomposition of biomass. Different catalysts may affect the upgrading reactions differently. The reduced charcoal fraction yield may be related to two factors. The ?rst factor is that some catalyst may actually promote thermal decomposition of biomass. The second factor is that some catalysts may act as a microwave absorption promoter, which may increase the heating rate that is often responsible for increased conversion ef?ciency as well as liquid yield. The ?rst factor may also be responsible for the signi?cant variations in chemical composition of the bio-oils as described later on.
We conducted more experiments on MgCl2. Table 2 shows the bio-oil and charcoal yields as a function of MgCl2 percentage when MgCl2 was used as catalyst in microwave-assisted pyrolysis of corn stover and aspen. For corn stover, the charcoal yield did not change much with MgCl2 addition, but the bio-oil yield increased with increasing MgCl2 addition. For aspen, MgCl2 tends to increase the charcoal yield as well as the bio-oil yield. The results shown in Table 2 suggest that MgCl2 is effective in ‘‘in situ upgrading’’ which changes the ratio of condensables to non-condensables. This is consistent with the MgCl2 results shown in Fig. 1. The results in Table 2 also indicate that corn stover and aspen differ in fractional yields. It is unclear how this difference is related to the chemical compositions of the two materials (Table 1).
Fig. 3. Total ion chromatograms from GC–MS analysis of pyrolytic oils from corn stover when MgCl2 was used at different levels. X: unidenti?ed compound; FF: furfural; MFE: beta-methoxy-(S)-2-furanethanol.
Y. Wan et al. / J. Anal. Appl. Pyrolysis 86 (2009) 161–167
Fig. 4. Total ion chromatograms from GC–MS analysis of pyrolytic oils from aspen when MgCl2 was used at different levels. X: unidenti?ed compound; FF: furfural; MFE: betamethoxy-(S)-2-furanethanol.
3.2. GC–MS analysis of the bio-oils The pyrolytic bio-oils were characterized using GC–MS, and the results are shown in Figs. 2–4 and Tables 3 and 4. Fig. 2 shows the
total ion chromatograms (TIC) of GC–MS analysis of bio-oil samples from microwave-assisted pyrolysis of corn stover when different chloride salts were used as catalysts. There were about 53 compounds detected by GC–MS in the bio-oil from non-catalyst
Table 3 Compounds detected with GC–MS in the bio-oils from microwave-assisted pyrolysis of corn stover. Retention time Peak name Area% Without catalyst MgCl2 (g/100 biomass) 2 2.654 2.912 3.226 3.360 3.426 3.679 3.840 3.967 4.063 4.163 4.475 4.713 5.159 5.625 5.906 5.966 6.560 6.892 6.968 7.068 8.030 8.356 9.583 12.951 14.114 1-Hydroxy-2-butanone Unidenti?ed compound X Furfural 2-Furanmethanol Furan, tetrahydro-2,5-dimethoxyFuran, tetrahydro-2,5-dimethoxy2-Cyclopenten-1-one, 2-methylButyrolactone 2-Cyclopenten-1-one, 2-hydroxy2-Furanethanol, beta-methoxy-(S) Phenol 2-Cyclopenten-1-one, 2-hydroxy-3-methylPhenol, 4-methylPhenol, 2-methoxy1,3-Propanediamine, N-methylPhenol, 4-ethyl1,2-Benzenediol Butanoic acid, pentyl ester Benzofuran, 2,3-dihydro2-Methoxy-4-vinylphenol Phenol, 2,6-dimethoxyLevoglucosan n-Hexadecanoic acid Oleic Acid 2.63 – 4.31 – 4.24 – – – 2.68 1.96 – 6.74 3.93 3.51 4.33 5.56 3.51 2.16 – 9.63 3.33 5.08 – 2.34 2.88 1.92 – 6.80 4.17 1.78 0.71 0.69 0.91 2.21 3.43 9.76 6.28 3.08 2.84 3.49 – 2.11 10.31 3.13 1.87 3.98 4.18 – – 4 – 6.65 13.40 – – 1.36 1.24 0.63 – 1.68 1.72 3.56 2.27 1.36 2.58 – – 3.78 2.44 2.66 7.70 – – 8 2.00 – 79.58 – – 4.33 2.95 1.82 – – – 5.68 – – 3.65 – – – – – – – – – –
Note: About 36 kinds of components of bio-oil from corn stover without any catalyst whose area % less than 1.82% were not displayed.
Y. Wan et al. / J. Anal. Appl. Pyrolysis 86 (2009) 161–167
Table 4 Compounds detected with GC–MS in the bio-oils from microwave-assisted pyrolysis of aspen. Retention time Peak name Area % Without catalyst MgCl2 (g/100 biomass) 2 2.658 2.912 3.226 3.426 3.679 3.840 4.163 4.498 4.713 5.159 5.625 5.906 6.401 6.588 6.867 6.892 8.030 8.356 9.196 9.237 9.706 10.179 11.229 1-Propanol, 2-methylUnidenti?ed compound X Furfural 2-Furanmethanol Furan, tetrahydro-2,5-dimethoxyFuran, tetrahydro-2,5-dimethoxy2-Cyclopenten-1-one, 2-hydroxy2-Furanethanol, beta-methoxy-(S) Phenol 2-Cyclopenten-1-one, 2-hydroxy-3-methylPhenol, 4-methylPhenol, 2-methoxyPhenol, 3,5-dimethylPhenol, 2-methoxy-4-methyl1,2-Benzenediol 2-Methoxy-4-vinylphenol Phenol, 2,6-dimethoxy3-Hydroxy-4-methoxybenzoic acid Phenol, 2-methoxy-4-(1-propenyl)1,6-Anhydro-beta-D-glucopyranose (E)-Stilbene Phenol, 2,6-dimethoxy-4-(2-propenyl)2.22 – 3.58 3.17 – – 2.13 – 4.47 2.58 2.69 3.20 2.60 2.78 2.72 2.12 4.51 2.63 2.10 4.45 1.92 2.41 – 10.52 3.34 – 4 – 14.09 9.94 – 8 – – 83.00 – 7.34 4.94 – – – – – 4.72 – – – – – – – – – –
1.36 8.37 3.24 1.45 1.43 2.43 6.92 2.65 2.52 3.08 1.98 1.04 1.62 – 1.09
0.36 11.26 3.09 1.83 0.85 1.75 3.97 2.04 2.57 1.97 0.96 – 4.73 – –
Note: About 81 kinds of components of bio-oil from aspen without any catalyst whose area % less than 1.82% were not displayed.
treated samples (control). We found that all chloride salts promoted furfural production, but ZnCl2 and MgCl2 also effectively suppressed the production of other chemicals. MgCl2 gave a very clean spectrum with only one major peak which is furfural. This is evidenced by data shown in Table 3. The area for furfural increased from 4.31% in the control sample to 79.58% in the sample treated with 8 g MgCl2. Most of other chemicals decreased or simply disappeared when MgCl2 was used. The rest of the area (around 20%) is made up of 1-hydroxy-2-butanone, tetrahydro-2,5dimethoxy-furan, 2-methyl-2-cyclopenten-1-one, phenol, and 2methoxy-phenol. We conducted more experiments on corn stover and aspen using MgCl2 at different levels (Figs. 3 and 4). We noticed a gradual
increase in furfural and a decrease in other chemicals with increasing MgCl2. At lower levels of MgCl2, we also noticed a sudden increase and/or appearance of furfural (FF), beta-methoxy(S)-2-furanethanol (MFE) peaks and an unidenti?ed peak X. Fig. 5 shows the electron ionization (EI) spectrum for this unidenti?ed compound X. The MFE and X peaks disappear when 8 g of MgCl2 was used. Our results show that 8 g MgCl2 per 100 g of dry biomass is necessary to produce spectra with a single major furfural peak with minimal appearance of other chemicals. However, these results also indicate that the chemical pro?les can be altered through adjusting the dosage of catalysts, suggesting that it is possible to target certain desirable products by controlling the dosage. Further study is necessary to explore other catalysts which
Fig. 5. Electron ionization spectrum of unidenti?ed compound X.
Y. Wan et al. / J. Anal. Appl. Pyrolysis 86 (2009) 161–167
could suppress most undesirable chemical reactions while maximizing reactions leading to production of desirable products (other than furfural). Table 4 shows the chemical composition of bio-oils from microwave-assisted pyrolysis of aspen. There were about 99 chemicals in the bio-oil from the control sample but only 3 major chemicals in the bio-oil from aspen samples treated with MgCl2. Furfural accounts for approximately 83%, and tetrahydro-2,5dimethoxy-furan and, 2-methoxy-phenol for 17%. The difference in furfural yield between corn stover and aspen may be attributed to their large difference in cellulose content instead of hemicellulose since they have the same hemicellulose content. Commercial production of furfural is achieved through converting the hemicellulosic fraction of the lignocellulosic to monosaccharides (pentoses) by acid hydrolysis, which are further dehydrated to produce furfural. Additional chemical pathways for conversion of cellulose to furanic compounds in different reaction systems including pyrolysis have been proposed by researchers [25–27]. We tested other catalysts (results not shown). Among them, Mg(ClO4)2 behaved similarly with MgCl2, while most of other catalysts had little effect on the composition of the bio-oils. In addition, we tested MAP of aspen and corn stover using a mixture of MgCl2 and some metal oxides including MgO, MgSO4, H3BO3. These oxides did not affect or improve the function of MgCl2 (results not shown). 4. Conclusions Medal oxides, salts, and acids including K2Cr2O7, Al2O3, KAc, H3BO3, Na2HPO4, MgCl2, AlCl3, CoCl2, and ZnCl2 when pre-mixed with corn stover or aspen wood pellets prior to pyrolysis were found to affect the fractional yields of bio-oil, gas and char and the chemical pro?les of the bio-oils. KAc, Al2O3, MgCl2, H3BO3, and Na2HPO4 were found to increase the bio-oil yield. The increase in bio-oil yield was accompanied by change in charcoal or gas yields or both, suggesting that the catalysts used may affect the fractional yields through different mechanisms. The catalysts may function as a microwave absorbent to speed up heating or they may participate in the so-called ‘‘in situ upgrading’’ of pyrolytic vapors during the microwave-assisted pyrolysis of biomass. Catalysts may play an important role in solid-to-vapor conversion and reforming of vapors during the microwaveassisted pyrolysis process. The GC–MS analysis indicated that use of the catalysts signi?cantly reduce the number of compounds in the bio-oils. At lower levels of MgCl2, we observed two to three major peaks on the spectra. At 8 g MgCl2 per 100 biomass level, furfural peak accounting for about 80% of the area under the spectrum. These results indicate that MgCl2 is effective in improving the product selectivity of the microwave-assisted
pyrolysis and that it is possible to control the product pro?les by varying the catalysts and the dosages used. We are conducting similar experiments using a much larger continuous pilot scale microwave-assisted pyrolysis system. Further study is needed to study additional catalysts and investigate the feasibility of controlling microwave-assisted pyrolysis for production of target chemicals. Acknowledgements This work was supported by the Open Foundation of the State Key Laboratory of Food Science, Nanchang University, China Ministry of Education PCSIRT Program (IRT0540), and by US Department of Energy and US Department of Agriculture and University of Minnesota Initiative for Renewable Energy and the Environment grants. References
 D. Klass, Biomass for Renewable Energy, Fuels, and Chemicals, Academic Press, San Diego, 1998, p. 651.  S. Zhang, Y. Yan, T. Li, Z. Ren, Bioresource Technol. 96 (5) (2005) 545.  R.K. Sharma, N.N. Bakhshi, Can. J. Chem. Eng. 71 (3) (1993) 383–391.  R.K. Sharma, N.N. Bakhshi, Energy Fuels 7 (2) (1993) 306.  E. Putun, F. Ates, A.E. Putun, Fuel 87 (6) (2008) 815–824.  B.V. Babu, A.S. Chaurasia, Model. Energy Convers. Manage. 44 (2003) 2135.  C. Yang, B. Zhang, J. Moen, K. Hennessy, Y. Liu, X. Lin, Y. Wan, P. Chen, R. Ruan, 30th Symposium on Biotechnology for Fuels and Chemicals, New Orleans, LA, in press.  M. Ringer, V. Putsche, J. Scahill, Large-scale Pyrolysis Oil Production, National Renewable Energy Laboratory, Golden, 2006.  Q. Zhang, J. Chang, T. Wang, Y. Xu, Energy Convers. Manage. 48 (1) (2007) 87.  P.T. Williams, N. Nugranad, Energy 25 (6) (2000) 493.  E. Antonakou, A. Lappas, M.H. Nilsen, A. Bouzga, M. Stocker, Fuel 85 (14–15) (2006) 2202.  P.T. Williams, P.A. Horne, J. Anal. Appl. Pyrol. 31 (1995) 15.  J. Adam, E. Antonakou, A. Lappas, M. Stocker, M.H. Nilsen, A. Bouzga, J.E. Hustad, G. Oye, Micropor. Mesopor. Mater. 96 (1–3) (2006) 93.  J. Moen, C. Yang, A.L. Olson, B. Zhang, H. Lei, K. Hennessy, Y. Wan, Z. Le, Y. Liu, P. Chen, R. Ruan, 30th Symposium on Biotechnology for Fuels and Chemicals, New Orleans, LA, in press.  R.J. Ferrier, W.B. Severn, R.H. Furneaux, I.J. Miller, Carbohydr. Res. 237 (1992) 79.  H. Zhao, J.E. Holladay, H. Brown, Z.C. Zhang, Science 316 (2007) 1597.  N. Cao, Q. Xu, C. Chen, C.S. Gong, L.F. Chen, Appl. Biochem. Blotechnol. 45/46 (1994) 521.  M.R. Gray, W.H. Corcoran, G.R. Gavalas, Ind. Eng. Chem. Process. Design 24 (1985) 646.  K. Raveendran, A. Ganesh, K.C. Khilart, Fuel 74/12 (1995) 1812–1822.  A. Gani, I. Naruse, Renewable Energy 32 (2007) 649.  H. Kawamoto, D. Yamamoto, S. Saka, J. Wood Sci. 54 (2008) 242.  F. Yu, S. Deng, P. Chen, Y. Liu, Y. Wan, A. Olson, D. Kittelson, R. Ruan, Appl. Biochem. Biotechnol. 137–140 (1–12) (2007) 957.  P. Monsef-Mirzaia, M. Ravindrana, W.R. McWhinniea, P. Burchillb, Fuel 74 (1) (1995) 20. ? ?  A. Dom?nguez, J.A. Menendez, M. Inguanzo, J.J. Pis, Fuel Process. Technol. 86 (9) (2005) 1007.  J. Scheirs, G. Camino, M. Avidano, J. Appl. Polym. Sci. 69 (13) (1998) 2541.  H. Kawamoto, S. Saito, W. Hatanaka, S. Saka, J. Wood Sci. 53 (2) (2007) 127.  E.B. Sanders, A.I. Goldsmith, J.I. Seeman, J. Anal. Appl. Pyrol. 66 (1) (2003) 29.