ARTICLE pubs.acs.org/EF
Formation of Aromatic Structures during the Pyrolysis of Bio-oil Yi Wang, Xiang Li, Daniel Mourant, Richard Gunawan, Shu Zhang, and Chun-Zhu Li* Fuels and Energy Technology Institute, Curtin University of Technology, GPO Box U1987, Perth, Western Australia 6845, Australia ABSTRACT: The pyrolysis of biomass to produce bio-oil is a very effective way of biomass use. Bio-oil undergoes drastic structural changes as it is upgraded into biofuels or used as a fuel for gasification/combustion. The evolution of aromatic ring systems in bio-oil is a key consideration in bio-oil use. A bio-oil sample produced from the fast pyrolysis of mallee wood at 500 °C, its lignin-derived oligomers, and pure cellulose have been pyrolyzed in a novel two-stage fluidized-bed/fixed-bed reactor at temperatures between 350 and 850 °C. The product tars were characterized with ultraviolet (UV) fluorescence spectroscopy. Our results indicate that significant portions of aromatic ring systems in the bio-oil could turn/polymerize into solids not soluble in CHCl3 + CH3OH during the pyrolysis at relatively low temperatures, e.g., 350 400 °C. This process can be enhanced by the presence of cellulose-/ hemicellulose-derived species in the bio-oil, which are reactive and produce radicals to enhance the polymerization reactions. The pyrolysis of cellulose-derived species in the bio-oil tended to form additional very large aromatic ring systems at temperatures higher than 700 °C.
1. INTRODUCTION The pyrolysis of biomass is a very effective means of energy densification. With the biochar returned to the field as a soil conditioner and for carbon biosequestration, bio-oil is a complex mixture of chemical compounds that can be used in many ways, including being upgraded into liquid transport biofuels or being used as a feedstock for gasifiers or conventional boilers (e.g., cofired with coal).1 5 However, bio-oil is exceedingly reactive and would undergo drastic changes when it is further heated, which could impact its use.6,7 The formation of various large/complex aromatic ring systems is especially important. For example, coke formation during the upgrading of bio-oil can deactivate catalysts and is a major operating problem. Tar formation in a gasifier is a major problem for the use of the gasification product gas.8,9 Insufficient knowledge exists about the formation of various aromatic ring systems during the pyrolysis of biomass and the subsequent use of bio-oil. Lignin, cellulose, and hemicellulose are the most important cellular components of woody biomass in the production of bio-oil. While lignin contains abundant monoaromatic structures, the woody biomass generally contains little polycyclic aromatic ring structures.10 Even the monoaromatic ring structures in lignin could lead to the formation of aromatic ring systems containing more than one benzene ring during pyrolysis. As part of our ongoing efforts to understand the formation of aromatic ring systems during the pyrolysis of biomass and the further use of bio-oil, this study aims to investigate the evolution of aromatic ring structures during the thermal decomposition (pyrolysis) of bio-oil between 350 and 850 °C. To gain further insight into the reactions involved, cellulose- and lignin-derived oligomers separated from bio-oil were also pyrolyzed under similar conditions. While relatively simple aromatic compounds were quantified with gas chromatography mass spectrometry (GC MS), ultraviolet (UV) fluorescence spectroscopy was used to trace the overall development of aromatic ring systems during pyrolysis. r 2011 American Chemical Society
2. EXPERIMENTAL SECTION 2.1. Preparation of Bio-oil and Bio-oil Fractions. The bio-oil used in this study was produced from the pyrolysis of mallee eucalypt (Eucalyptus loxopheba ssp. lissophloia) wood at fast heating rates in a fluidized-bed reactor (nominally 1 kg/h) at 500 °C. A detailed description of the reactor system and the experimental procedure can be found elsewhere.11 13 The bio-oil was stored in a freezer (about 10 °C) until required in further experiments described herein. The bio-oil was separated into water-soluble and water-insoluble fractions through the cold water precipitation.14,15 The water-insoluble fraction (filter cake) was further washed with CH2Cl2 to produce CH2Cl2-soluble (∼18.0 wt % bio-oil) and CH2Cl2-insoluble (∼3.6 wt % bio-oil) fractions. This process was repeated to prepare a significant amount of water-insoluble but CH2Cl2-soluble fraction for the pyrolysis experiments described below. The water-insoluble/CH2Cl2-soluble fraction is mainly the lignin-derived oligomers in the bio-oil.16,17 Again, the sample was dried and stored in a freezer until required. The pure cellulose powder (α-cellulose, Sigma-Aldrich) was dried in the oven at 105 °C for 24 h prior to use in the experiment. 2.2. Pyrolysis Experiment. The pyrolysis of bio-oil, cellulose, and lignin-derived oligomers was carried out in a novel two-stage fluidizedbed/fixed-bed quartz reactor (Figure 1).18,19 A total of 60 g of silica sand ranging from 0.21 to 0.30 mm was used as a bed material in the bottom stage of the reactor, which was fluidized with argon fed into the reactor from the bottom. The bio-oil was fed into the reactor via an air-cooled injection probe, with the help of argon carrier gas (0.21 L/min), at a rate of 100 mg/min using a syringe pump (KD Scientific 410CE) equipped with a 20 mL stainless-steel syringe. The weights of the syringe and the injection probe were recorded before and after an experiment to calculate the amount of bio-oil fed into the reactor. The solid feedstock (water-insoluble/CH2Cl2-soluble fraction and cellulose) particles were Special Issue: 2011 Sino-Australian Symposium on Advanced Coal and Biomass Utilisation Technologies Received: August 1, 2011 Revised: September 14, 2011 Published: September 14, 2011 241
dx.doi.org/10.1021/ef201155e | Energy Fuels 2012, 26, 241–247
Energy & Fuels
ARTICLE
Figure 2. Constant energy ( 2800 cm 1) synchronous spectra of different fractions produced from the precipitation of bio-oil in cold water and the subsequent washing of the filter cake with CH2Cl2. The fluorescence intensity is expressed on the basis of per gram of bio-oil.
2.3. Characterization of Tar. 2.3.1. UV Fluorescence Spectroscopy. UV fluorescence spectroscopy has been widely used to characterize the structural features of “liquids” (such as bio-oil or tar) derived from coal and biomass, giving information about the relative size and concentration of aromatic ring systems in the sample.16,23 25 In this study, the UV fluorescence spectra of tars were recorded using a PerkinElmer LS50B spectrometer. The bio-oil/tar solution was diluted with methanol (Uvasol for spectroscopy; purity (GC), g99.9%) to 4 ppm (wt). The synchronous spectra were recorded with a constant energy difference of 2800 cm 1. The slit widths were 2.5 nm, and the scan speed was 200 nm/min. The “wavelength” shown for each spectrum refers to that of the excitation monochromator. The wavelength is a brief indication of the aromatic ring sizes (e.g., 370 nm; Figure 10B) aromatic ring systems. The GC MS analysis indeed detected some new aromatic compounds (such as fluorene and pyrene) after pyrolysis at high temperatures. Some large aromatic ring systems may be present in large tar molecules and have thus not been detected by GC MS. Because the bio-oil contained a significant amount (∼19 wt %)11,13,16 of water and the thermal cracking reactions would also have produced more water, it then becomes clear that the self-reforming of bio-oil with steam under our experimental conditions was insufficient to suppress the formation of aromatic ring systems. This is in agreement with our earlier studies on the reforming of biomass volatiles.18,32 3.3. Comparison between the Pyrolysis of Bio-oil and That of Lignin-Derived Oligomers. The data in Figure 2 indicate that the fluorescence properties of the raw bio-oil are dominated largely by its lignin-derived oligomers, in agreement with our previous study.16 However, the pyrolysis behavior of the ligninderived oligomers (Figures 7 and 8) and that of the raw bio-oil (Figures 9 and 10) are not completely similar, although they do have great similarities. In particular, more aromatic ring structures during the pyrolysis of bio-oil at 350 and 400 °C (Figure 10) have turned into solids, i.e., not detected as tar, than during the pyrolysis of lignin-derived oligomers (Figure 8), which had been separated from the same bio-oil. The amount of aromatic structures turning into solids decreased drastically with an increasing pyrolysis temperature (Figure 10). The exact reasons for the differences in pyrolysis behavior between the bio-oil and lignin-derived oligomers remain unclear. However, it is believed that the differences originate from the differences in their composition. Bio-oil is an exceedingly complex mixture. In particular, bio-oil contains a lot of reactive, often O-containing, species, such as acids, aldehydes, and sugars, formed from the breakdown of cellulose/hemicellulose.12,16,32 These reactive species would dissociate to form radicals at relatively low temperatures, e.g., 350 °C or lower. In comparison, the lignin-derived species, particularly larger aromatic ring systems (likely containing various substitutional groups), are relatively stable. When bio-oil is heated to 350 °C or higher (Figure 10), the radicals generated from the decomposition of these reactive species would attack and activate the structures containing larger aromatic ring systems, which in turn would combine/polymerize to become insoluble in the CHCl3 + CH3OH solvent mixture. This would include the combination not only among lignin-derived species but also among celluloseand lignin-derived species. At low temperatures around 350 °C, the radical combination reactions (including polymerization) dominate over the decomposition reactions. This appears to
provide a plausible explanation for the transformation of a large portion of larger aromatic ring systems in the bio-oil into solids (coke) at 350 and 400 °C (Figure 10) that is insoluble in CHCl3 + CH3OH. With an increasing temperature, the decomposition reactions became significant. Even if some solids have been formed involving larger aromatic ring systems because of the enhanced polymerization by the cellulose-/hemicellulose-derived species, the solids would appear to be unstable and, thus, would decompose. This would seem to explain why the “yields” of larger aromatic ring systems increased with an increasing temperature (Figure 10) during the pyrolysis of bio-oil. In the case of the pyrolysis of lignin-derived oligomers, in the absence of cellulose-/hemicellulose-derived reactive species, any radicals required for the combination (polymerization) involving larger aromatic ring systems would have to come from the bond breakage of the lignin-derived oligomers, which is relatively stable at low temperatures. This explains the relatively high “yields” of larger aromatic ring systems in the lignin-derived oligomers remaining as soluble in CHCl3 + CH3OH after pyrolysis at low temperatures. Another distinct difference in the pyrolysis behavior between bio-oil and its lignin-derived oligomers at high temperatures (>700 °C) is the presence of very large (>375 nm) aromatic ring systems in the tar of the bio-oil (Figure 10) but not so much in the tar of the lignin-derived oligomers (Figure 8). These very large aromatic ring systems appear to have originated in the cellulose-derived products. The pyrolysis of cellulose did generate these very large aromatic systems (Figure 5). The bio-oil also contains some small amounts (∼3.6 wt %) of water-insoluble/CH2Cl2-insoluble materials, which are mainly the high-molecular-mass materials. The pyrolysis behavior of this fraction remains unknown because the preparation of a significant amount of this fraction for the pyrolysis experiments was difficult because of its low concentration in the bio-oil.
4. CONCLUSION The bio-oil produced from the pyrolysis of mallee wood at 500 °C at fast heating rates in a 1 kg/h fluidized-bed reactor has been pyrolyzed again in a bench-scale novel two-stage fluidizedbed/fixed-bed reactor at temperatures between 350 and 850 °C. The majority of larger (g2 rings) aromatic ring systems as detected by UV fluorescence spectroscopy are in the ligninderived oligomers. To gain insights into the reactions responsible for the formation and evolution of aromatic ring systems during the repyrolysis of bio-oil, the lignin-derived oligomers were separated from the bio-oil and pyrolyzed under the same conditions. The pyrolysis of cellulose (especially the volatiles formed in situ in the bottom reactor) has given further information about the formation of aromatic systems from the cellulose-derived species during the pyrolysis of bio-oil. Our results indicate that significant portions of aromatic ring systems in the bio-oil could turn/polymerize into solids not soluble in CHCl3 + CH3OH during the pyrolysis at relatively low temperatures, e.g., 350 400 °C. This process can be enhanced by the presence of cellulose-/hemicellulose-derived species in the bio-oil, which are reactive and produce radicals to enhance the polymerization reactions. The pyrolysis of cellulose-derived species in the bio-oil at high temperatures (e.g., >700 °C) tended to form additional very large aromatic ring systems. 246
dx.doi.org/10.1021/ef201155e |Energy Fuels 2012, 26, 241–247
Energy & Fuels
ARTICLE
’ AUTHOR INFORMATION
(26) Li, C.-Z.; Sathe, C.; Kershaw, J. R.; Pang, Y. Fuel 2000, 79 (3 4), 427–438. (27) Mullen, C. A.; Boateng, A. A. J. Anal. Appl. Pyrolysis 2010, 90 (2), 197–203. (28) Scholze, B.; Hanser, C.; Meier, D. J. Anal. Appl. Pyrolysis 2001, 58 59, 387–400. (29) Bayerbach, R.; Nguyen, V. D.; Schurr, U.; Meier, D. J. Anal. Appl. Pyrolysis 2006, 77 (2), 95–101. (30) Bayerbach, R.; Meier, D. J. Anal. Appl. Pyrolysis 2009, 85 (1 2), 98–107. (31) Richter, H.; Howard, J. B. Prog. Energy Combust. Sci. 2000, 26 (4 6), 565–608. (32) Lievens, C.; Mourant, D.; He, M.; Gunawan, R.; Li, X.; Li, C.-Z. Fuel 2011, 90 (11), 3417–3423.
Corresponding Author
*Telephone: +61-8-92661131. Fax: +61-8-92661138. E-mail:
[email protected].
’ ACKNOWLEDGMENT Australian Government funding through the Second Generation Biofuels Research and Development Grant Program and the International Science Linkages Program supports this project. The study also received support from the Government of Western Australia via the Centre for Research into Energy for Sustainable Transport (CREST). The authors also thank Hongwei Wu for helpful discussions. ’ REFERENCES (1) Chiaramonti, D.; Oasmaa, A.; Solantausta, Y. Renewable Sustainable Energy Rev. 2007, 11 (6), 1056–1086. (2) Demirbas, A. Energy Convers. Manage. 2009, 50 (1), 14–34. (3) Iborra, S.; Huber, G. Chem. Rev. 2006, 106 (9), 4044–4098. (4) Elliott, D. C.; Oasmaa, A. Energy Fuels 1991, 5 (1), 102–109. (5) Li, X.; Gunawan, R.; Lievens, C.; Wang, Y.; Mourant, D.; Wang, S.; Wu, H.; Garcia-Perez, M.; Li, C.-Z. Fuel 2011, 90 (7), 2530–2537. (6) Czernik, S.; Bridgwater, A. V. Energy Fuels 2004, 18 (2), 590– 598. (7) Bridgwater, A. V. Fuel 1995, 74 (5), 631–653. (8) Gayubo, A. G.; Valle, B.; Aguayo, A. T.; Olazar, M.; Bilbao, J. J. Chem. Technol. Biotechnol. 2010, 85 (1), 132–144. (9) Milne, T. A.; Abatzoglou, N.; Evans, R. J. Biomass Gasifier “Tars”: Their Nature, Formation, and Conversion; National Renewable Energy Laboratory (NREL): Golden, CO, 1998; NREL/TP-570-25357. (10) Zakzeski, J.; Bruijnincx, P. C. A.; Jongerius, A. L.; Weckhuysen, B. M. Chem. Rev. 2010, 110 (6), 3552–3599. (11) Garcia-Perez, M.; Wang, X. S.; Shen, J.; Rhodes, M. J.; Tian, F.; Lee, W.-J.; Wu, H.; Li, C.-Z. Ind. Eng. Chem. Res. 2008, 47 (6), 1846– 1854. (12) Shen, J.; Wang, X.-S.; Garcia-Perez, M.; Mourant, D.; Rhodes, M. J.; Li, C.-Z. Fuel 2009, 88 (10), 1810–1817. (13) Mourant, D.; Wang, Z.; He, M.; Wang, X. S.; Garcia-Perez, M.; Ling, K.; Li, C.-Z. Fuel 2011, 90 (9), 2915–2922. (14) Oasmaa, A.; Lepp€am€aki, E.; Koponen, P.; Levander, J.; Tapola, E. Application of Standard Fuel Oil Analyses; VTT Energy: Espoo, Finland, 1997; VTT Publications 3 06. (15) Garcia-Perez, M.; Chaala, A.; Pakdel, H.; Kretschmer, D.; Roy, C. Biomass Bioenergy 2007, 31 (4), 222–242. (16) Garcia-Perez, M.; Wang, S.; Shen, J.; Rhodes, M.; Lee, W. J.; Li, C.-Z. Energy Fuels 2008, 22 (3), 2022–2032. (17) Scholze, B.; Meier, D. J. Anal. Appl. Pyrolysis 2001, 60 (1), 41–54. (18) Min, Z.; Asadullah, M.; Yimsiri, P.; Zhang, S.; Wu, H.; Li, C.-Z. Fuel 2011, 90 (5), 1847–1854. (19) Min, Z.; Yimsiri, P.; Asadullah, M.; Zhang, S.; Li, C.-Z. Fuel 2011, 90 (7), 2545–2552. (20) Li, C.-Z.; Bartle, K. D.; Kandiyoti, R. Fuel 1993, 72 (11), 1459–1468. (21) Li, C.-Z.; Bartle, K. D.; Kandiyoti, R. Fuel 1993, 72 (1), 3–11. (22) Sathe, C.; Pang, Y.; Li, C.-Z. Energy Fuels 1999, 13 (3), 748–755. (23) Li, C.-Z.; Wu, F.; Cai, H.-Y.; Kandiyoti, R. Energy Fuels 1994, 8 (5), 1039–1048. (24) Zeng, C.; Favas, G.; Wu, H.; Chaffee, A. L.; Hayashi, J.-i.; Li, C.-Z. Energy Fuels 2005, 20 (1), 281–286. (25) Peacocke, G. V. C.; Madrali, E. S.; Li, C. Z.; Guell, A. J.; Wu, F.; Kandiyoti, R.; Bridgwater, A. V. Biomass Bioenergy 1994, 7 (1 6), 155–167. 247
dx.doi.org/10.1021/ef201155e |Energy Fuels 2012, 26, 241–247
