J. Anal. Appl. Pyrolysis 76 (2006) 230–237 www.elsevier.com/locate/jaap
Fast pyrolysis of rice straw, sugarcane bagasse and coconut shell in an induction-heating reactor W.T. Tsai a,*, M.K. Lee b, Y.M. Chang a a
Department of Environmental Engineering and Science, Chia Nan University of Pharmacy and Science, Tainan 717, Taiwan b Department of Occupational Safety and Health, Chia Nan University of Pharmacy and Science, Tainan 717, Taiwan Received 11 April 2005; accepted 9 November 2005 Available online 24 January 2006
Abstract With the application of induction heating, a fast pyrolysis was used for producing valuable products from rice straw, sugarcane bagasse and coconut shell in an externally heated fixed-bed reactor. The effect of process parameters such as pyrolysis temperature, heating rate and holding time on the yields of pyrolysis products and their chemical compositions were investigated. The maximum yield of ca. 50% on the pyrolysis liquid product could be obtained at the proper process conditions. The chemical characterization by elemental (CHNO), calorific, Fourier transform infrared (FT-IR) spectroscopy and gas chromatography/mass spectrometry (GC–MS) showed that the pyrolysis liquid products contain large amounts of water (>65 wt.%), and fewer contents of oxygenated hydrocarbons composing of carbonyl groups, resulting in low pH and low heating values. The results were very similar to bio-oils obtained from other biomass materials. The residual solid (char or charcoal) was also characterized in the present study. # 2006 Published by Elsevier B.V. Keywords: Rice straw; Sugarcane bagasse; Coconut shell; Fast pyrolysis; Biomass and agricultural wastes
1. Introduction Since the energy crisis in the mid-1970s, the energy utilization from biomass resources (called biomass energy) has received considerable attention. The energy obtained from agricultural wastes or agricultural by-products is a form of renewable energy; in principle, utilizing this energy does not add carbon dioxide, which is a greenhouse gas, to the atmospheric environment, in contrast to fossil fuels [1]. Like other biomass wastes, agricultural wastes contain a high amount of organic constituents (i.e., cellulose, hemicellulose and lignin) and possess a high-energy content [2,3]. Therefore, it can be recognized as a potential source of renewable energy based on benefits of both energy recovery and environmental protection. Due to the lower contents of sulfur and nitrogen in biomass wastes, its energy utilization also creates less environmental pollution and health risk than fossil fuel
* Corresponding author. Tel.: +886 6 2660393; fax: +886 6 2669090. E-mail address:
[email protected] (W.T. Tsai). 0165-2370/$ – see front matter # 2006 Published by Elsevier B.V. doi:10.1016/j.jaap.2005.11.007
combustion. The biomass energy led to the developments of various thermo-chemical processes for producing bio-fuels from lignocellulosic biomass [4–7]. Pyrolysis, a promising route for biomass utilization, has been widely used for converting biomasses into fuel gases, liquids and solids among the thermo-chemical technologies. Generally, there are two approaches for the conversion technology. One approach, referred to as conventional or traditional pyrolysis, is to maximize the yield of fuel gas at the preferred conditions of high temperature, low heating rate and long gas resistance time, or to enhance the char production at the low temperature and low heating rate. Another approach, known as flash or fast pyrolysis, is to maximize the yield of liquid product at the processing conditions of (1) very highheating rate (>100 8C/min) and heat transfer rate, (2) finely ground biomass feed (<1 mm), (3) carefully controlled temperature (around 500 8C), and (4) rapid cooling of the pyrolysis vapors to give the bio-crude products. Fast pyrolysis of biomass solid-waste is at present considered as an emerging energy technology for liquid-tar and solid-char production [8– 11]. However, the pyrolysis oil from biomass was found to be
W.T. Tsai et al. / J. Anal. Appl. Pyrolysis 76 (2006) 230–237
highly oxygenated and complex, and chemically unstable. Thus, the liquid products still need to be upgraded by lowering the oxygen content and removing residues. In addition, induction heating has been applied in various technological and scientific fields such as preheating, thermal treatment, melting and welding processes because of its characteristics of rapid heating, direct heating, non-contact heating, and precision control. Therefore, induction heating has also been considered as an alternative to carry out the pyrolysis in comparison with other fast/flash pyrolysis systems due to its high-energy efficiency and low air pollution [11]. With respect to the production of bio-crude from the fast pyrolysis of agricultural wastes using induction heating, the information published is limited. The pyrolysis of coconut shells was studied in a concentric three tubes reactor using a combination of external and internal heating method [12]. It was found that the less charcoal and light oil, the more tar were produced at the higher temperature and longer time of pyrolysis. The pyrolysis of rice husk was investigated in a fluidized bed reactor with and without ZSM-5 catalyst at 400–600 8C [13], showing that the uncatalyzed pyrolysis bio-oil was found to be highly oxygenated and chemically complex. In the presence of the catalyst the yield of bio-oil was markedly reduced; however, the oxygen content of the bio-oils was significantly reduced due to the evolution of catalytic gases such as H2O, CO and CO2. The vacuum pyrolysis of sugarcane bagasse was performed at bench- and pilot-scale plants [14]. In comparison with the yield of charcoal at about 20 wt.%, the resulting bio-oil with the recovery of about 30 wt.% (bagasse anhydrous basis) could be found to be a potential liquid fuel. In this study, the fast pyrolysis of rice straw, sugarcane bagasse and coconut shell, which are the most primary agricultural wastes in Taiwan [15], was investigated in an externally heated fixed-bed furnace to assess the product yield. The influences of pyrolysis temperature, heating rate and holding time at specified temperature were studied so as to obtain useful data for the optimal design in the induction-heating process. In addition, the pyrolysis products obtained at the conditions of the maximum product yields were further analyzed, using elemental, Karl– Fischer, thermal and spectroscopic methods, to determine its possibility of being a potential source of renewable fuels.
231
Table 1 Main characteristics of biomass wastes fed Characteristics
Rice straw
Sugarcane bagasse
Coconut shell
Proximate analysisa (wt.%) Moisture Combustible matter Ash
13.61 76.85 9.54
16.07 79.59 4.34
11.26 85.36 3.38
Ultimate analysisb (wt.%) Carbon Hydrogen Nitrogen Sulfur Chlorine Oxygenc
50.93 6.04 0.83 0.23 0.36 41.61
58.14 6.05 0.69 0.19 0.36 34.57
63.45 6.73 0.43 0.17 0.95 28.27
Heating value analysis (MJ/kg)
16.35
18.61
22.83
a b c
As prepared (air-dried, sand and fines free and ground <0.5 mm). Elemental composition in the combustible matter, on a dry basis. Calculated by difference.
2.2. Pyrolysis experiments The fixed-bed fast pyrolysis experiments were performed in a horizontally and externally heated tubular reactor (3.67 cm i.d. and 60 cm long), constructed from 310 stainless steel and heated by high-frequency generator (i.e., induction heating), which is shown schematically in Fig. 1. For all experiments, the mass (10 g, in each set of experiments) of the sample was placed into a crucible made of stainless steel, and then housed at the center of the tubular reactor. During the experiments, temperature was measured above the bed, with a K-type thermocouple in the middle of the tubular reactor in order to control and monitor the reactor temperature. The sweep gas from a nitrogen cylinder was dried and purified by a molecular sieve tube. The constant nitrogen flow rate (ca. 1000 cm3/min) was precisely metered to the experimental system using a mass flow controller. The experimental conditions in the fast pyrolysis system were as follows: pyrolysis temperature of 400–800 8C, heating temperature of 100–500 8C/min, and holding time (at the specified
2. Experimental 2.1. Feed materials The biomass samples used as the feedstock for experimental runs were rice straw, sugarcane bagasse and coconut shell that were mainly obtained from a farm in southern Taiwan. These samples were first dried in the sunshine for the purpose of removing most of moistures, and then separated from physical impurities such as sand and fine fraction. Finally, the samples that were ground in a rotary cutting mill were further screened into fractions of particle diameter <0.50 mm. Each prepared sample was closely stored in glass bottles. Their results of proximate, ultimate and thermal analysis are given in Table 1.
Fig. 1. Schematic diagram of the fast pyrolysis system: (1) nitrogen gas cylinder; (2) regulator; (3) molecular sieve column; (4) mass flow controller; (5) flexible heating tape; (6) power system (high-frequency generator); (7) tubular reactor (including induction coil); (8) temperature controller; (9) K-type thermocouple; (10) temperature recorder; (11) cryogenic condensation (ethylene glycol/water system); (12) liquid product collectors.
232
W.T. Tsai et al. / J. Anal. Appl. Pyrolysis 76 (2006) 230–237
pyrolysis temperature) of 1–8 min. The resulting products after fast pyrolysis were cooled to room temperature using a cryogenic system ( 10 8C) with ethylene glycol solution, and then taken from holding crucible and condensable collectors in order to weigh the masses of residual solid and liquid tar, respectively. The yields of the resulting products were thus calculated based on the mass of the sample fed. 2.3. Characterization Elemental analyses of elements of C/H/N and O were carried out on the resulting products (i.e., liquid and residual solid) with a CHN-O-RAPID Element Analyzer (Heraeus Co., Germany). Prior to the measurement, the standard sample was first analyzed for checking the experimental error within 1%. The water content of the liquid product was measured using Karl– Fischer moisture titrator (Model: MKS-510; Kyoto Electronics Manufacturing Co., Japan) at least twice. The calorific values of the liquid product and residual solid were also determined by using a bomb calorimeter (Model: C2000 basic, IKA Co., Germany). Regarding the corrosion of the liquid product, the pH of the pyrolysis liquid product was measured using a pH meter (Model: SP-701, Suntex Co., Taiwan). In order to settle down char traces from liquid phase in the crude pyrolysis liquids, the pyrolysis liquid products were first decanted and then centrifuged for 15 min at about 2000 rpm prior to the chromatographic and spectrometric analyses. Functional group analysis of the pretreated pyrolysis liquid product was carried out using Fourier transform infrared (FTIR) spectrometry (Model: DA8.3, Bomen Co., Canada). The gas chromatography/mass spectroscopy (GC–MS) analysis of the pretreated pyrolysis liquid product was performed with a Hewlett-Packard HP 5890-Series II gas chromatograph equipped with a Hewlett-Packard HP 5972A mass selective detector (MS), using a 60 m 0.25 mm HP-1 capillary column (0.25 mm film thickness). The following temperature program was adopted: initial, intermediate and final temperatures were 35, 200 and 350 8C, respectively, times at initial, intermediate and final temperatures were 5, 5, and 2 min, respectively, and heating rates were 5 and 10 8C/min, respectively. The injector temperature and detector temperature were 250 and 280 8C, respectively. The following parameters were used for the integration of the GC–MS chromatograms: initial area rejection = 0, initial peak width = 0.058, and initial threshold (to determine the sensitivity of the integrator) = 14.9. The oil sample (ca. 0.2 ml) was injected with a Hamilton syringe. The MS operated in scan mode and its mass range was 45–500 amu. 3. Results and discussion 3.1. Product yield The yields of resulting products from fast prolysis of rice straw, sugarcane bagasse and coconut shell were discussed for examining the effect of pyrolysis temperature, heating rate and holding time. As shown in Fig. 2, the yield of residual solid had
Fig. 2. Dependence of water, bio-oil and char yields based on a wet basis from (a) rice straw, (b) bagasse, and (c) coconut shell on pyrolysis temperature at a heating rate of 200 8C/min, holding time of 1 min, particle size of <0.50 mm, nitrogen flow rate of 1000 cm3/min, and condensation temperature of 10 8C.
a declining trend as the final pyrolysis temperature increased from 400 to 800 8C because of the progressive pyrolysis conversion at the higher temperature. However, the rate of declination in the ranges of 500–800 8C is not so fast as that in the ranges of 400–500 8C (i.e., 74–38% versus 38–21% in rice straw; 65–33% versus 33–21% in sugarcane bagasse; 75–38% versus 38–25% in coconut shell). On the other hand, the total yield of liquid products (water and bio-oil) slightly increased in the rages of the pyrolysis temperature. As expected, the total yield of liquid products significantly increased as the pyrolysis temperature was raised from 400 to 500 8C (i.e., 15% versus 38% in rice straw; 23% versus 47% in sugarcane bagasse; 12% versus 36% in coconut shell). The results showed that most of volatile pyrolysis products from the feeding samples were evolved at temperatures lower than 500 8C, which was also observed at other biomass wastes in the literature [16–21]. Further, the pyrolysis liquid products contained large amounts of water (73–92 wt.%) obtained from the measurements by the Karl–Fischer moisture titrator, resulting in the low yield of biooil from these biomass wastes as shown in Fig. 2. The high water content in the pyrolysis liquid product may be due to the high moisture content in the feeding biomasses and the release of volatile organic products during the preparation of condensed liquid sample.
W.T. Tsai et al. / J. Anal. Appl. Pyrolysis 76 (2006) 230–237
233
To determine the effect of heating rate on the yields of the pyrolysis products, the experiments were conducted at a heating rate of either 100, 200, 300, 400 or 500 8C/min at the final pyrolysis temperature of 500 8C and its holding time of 1 min. Fig. 3 shows that the variations of yields in the liquid products and residual solid were observed in the cases of rice straw, sugarcane bagasse or coconut shell. It can be seen that the larger yield of solid product was generally obtained at the higher heating rate, which was also consistent with the literature [16–21]. The pyrolysis liquid products still contained large amounts of water (70–86 wt.%). The bio-oil liquid yields (lower than 14 wt.%) seemed to be extremely lower than those (60–80 wt.%) carried out by other fast pyrolysis just because the former is based on wet basis in comparison to the later based on wet basis. However, it was reported that the yield of bio-oil ranging 13–17 wt.% was obtained from the fast pyrolysys of rape seed at 100–800 8C/min [17]. As shown in Fig. 4, it seems that the holding time at the final prolysis temperature of 500 8C and heating rates of 200 8C/min plays a somewhat important role in the manufacture of pyrolysis products from rice straw, sugarcane bagasse, or coconut shell. Obviously, the lower yield of pyrolysis char product was observed at the longer holding time, which should be attributed to the progressive gasification and/or thermal cracking of the pyrolysis product [22]. On the other hand, the
Fig. 4. Dependence of water, bio-oil and char yields based on a wet basis from (a) rice straw, (b) bagasse, and (c) coconut shell on holding time at a heating rate of 200 8C/min, pyrolysis temperature of 500 8C, particle size of <0.50 mm, nitrogen flow rate of 1000 cm3/min, and condensation temperature of 10 8C.
pyrolysis liquid products again contained large amounts of water (65–87 wt.%), resulting in the low yield of bio-oil (below 16 wt.%). It was also found that the total yield of liquid product produced from sugarcane bagasse was higher than that from rice straw and coconut shell. 3.2. Chemical characterization
Fig. 3. Dependence of water, bio-oil and char yields based on a wet basis from (a) rice straw, (b) bagasse, and (c) coconut shell on heating rates at a pyrolysis temperature of 500 8C, holding time of 1 min, particle size of <0.50 mm, nitrogen flow rate of 1000 cm3/min, and condensation temperature of 10 8C.
The liquid products obtained in the fast pyrolysis of rice husk, sugarcane bagasse and coconut shell, which are usually termed crude bio-oils, are red-brown-colored products with irritable odor. The appearance and odor are common to all pyrolysis liquid products from biomass wastes [23]. The calorific values and pH of the liquid products obtained at typical pyrolysis conditions are presented in Table 2. The results listed in Table 2 also present the calorific values of the residual solids. It can be seen that the calorific values (i.e., 4.6–10.3 MJ/kg on the wet basis) of the pyrolysis liquid products are not as high as those of commercial heating oils (30–40 MJ/kg on the dry basis), showing that they should contain lots of nonhydrocarbons and condensed water as described above. However, the calorific values (i.e., 20–29 MJ/kg) of the residual solids are rather high, even higher than those of their
234
W.T. Tsai et al. / J. Anal. Appl. Pyrolysis 76 (2006) 230–237
Table 2 Calorific values and pH of the yrolysis productsa Biomass waste
Liquid productc
Pyrolysis conditions
Residual solid
d
b
Calorific value (MJ/kg)
pH
Calorific value (MJ/kg)
18.43 20.40 5.06 6.10 5.85
2.89 2.78 3.25 3.34 2.98
20.43 20.74 22.48 20.97 19.78
Rice straw
200 8C/min, 200 8C/min, 200 8C/min, 400 8C/min, 200 8C/min,
400 8C, 500 8C, 700 8C, 500 8C, 500 8C,
1 min 1 min 1 min 1 min 8 min
Sugarcane bagasse
200 8C/min, 200 8C/min, 200 8C/min, 400 8C/min, 200 8C/min,
400 8C, 500 8C, 700 8C, 500 8C, 500 8C,
1 min 1 min 1 min 1 min 8 min
3.79 4.43 3.72 4.88 5.12
2.10 2.32 2.04 2.28 2.24
24.57 24.80 25.24 26.47 27.00
Coconut shell
200 8C/min, 200 8C/min, 200 8C/min, 400 8C/min, 200 8C/min,
400 8C, 500 8C, 700 8C, 500 8C, 500 8C,
1 min 1 min 1 min 1 min 8 min
7.75 8.99 10.32 9.67 9.55
3.15 3.11 3.25 3.26 3.28
26.18 25.97 26.89 28.63 27.06
a
All pyrolysis products were derived from the same conditions: particle size of <0.50 mm, nitrogen flow rate of 1000 cm3/min, and condensation temperature of 10 8C. b 200 8C/min, 400 8C, 1 min means that the pyrolysis products were obtained at heating rate of 200 8C/min, pyrolysis temperature of 400 8C, and holding time of 1 min. c The liquid product is mainly composed of water (Figs. 2–4), and fewer amount of bio-oil. d On the wet basis.
feeding precursors listed in Table 1. It is obvious that considerable amounts of carbon organized in the combustible matters of feeding precursors are still present in the residual solids, not pyrolyzed to produce condensable components in the fast pyrolysis system. On the other hand, the water contents in the pyrolysis liquid products, which result from the original moisture in the feedstock and the product of the dehydration reactions occurring during pyrolysis, could be high up to 30% [10], thus lowering its heating value and bio-oil yield. Table 2 also shows that the pH values of the pyrolysis liquid products are in the ranges of 2.1–3.3. The finding is in agreement with the results in the literature; that is, pyrolysis liquid products generally contain substantial amounts of organic acids, mostly acetic acid and formic acid, which results in a pH of 2–3 [10,23]. The results described above were further demonstrated with the elemental analyses of pyrolysis liquid products. From the data in Table 3, the carbon, hydrogen and oxygen contents of the pyrolysis liquid products are 8–10%, 7–8% and 43–48%, respectively. Obviously, the pyrolysis liquid product contains considerable amounts of water and oxygenated component, resulting in higher H/C molar ratios than the feeding precursors (see Table 1) and
conventional fuels, and also resulting in the low energy content of the pyrolysis liquid product listed in Table 2. The FT-IR spectra of the pyrolysis liquid products produced from the fast pyrolysis of rice straw, sugarcane bagasse and coconut shell are given in Fig. 5. It can be seen that their peak patterns are very similar, showing that their corresponding functional groups should be also close. The presence of the significant peaks between 1640 and 1700 cm 1 could be ascribable to C–O (carbonyl) stretching vibration indicative of the ketones, phenols, carboxylic acids or aldehydes, and/or represent C C stretching vibrations indicative of alkenes and aromatics. The peak at around 1517 cm 1 could be the cause of C C stretching vibrations indicative of alkenes and aromatics. Below 1500 cm 1, all bands are complex and have their origin in a variety of vibrational modes. C–H stretching and bending vibrations between1380 and 1465 cm 1 indicate the presence of alkane groups in pyrolysis oils derived from biomass. Therefore, the band at 1395 cm 1 is ascribable to bending vibrations for CH3 groups. Absorptions possibly due to C–O vibrations from carbonyl components (i.e., alcohols, esters, carboxylic acids or ethers) occur between 1300 and 900 cm 1.
Table 3 Comparisons on major element analysis of the pyrolysis productsa Element
Carbon Hydrogen Nitrogen Oxygen
Liquid (wt.%)
Residual solid (wt.%)
Rice straw
Sugarcane bagasse
Coconut shell
Rice straw
Sugarcane bagasse
Coconut shell
8.39 7.21 0.08 49.85
10.36 7.68 0.18 46.19
10.29 7.01 0.27 43.22
49.29 2.83 1.32 15.15
71.41 3.32 1.77 15.46
69.33 3.26 0.94 15.29
a Obtained at heating rate of 200 8C/min, pyrolysis temperature of 500 8C, holding time of 8 min, particle size of <0.50 mm, nitrogen flow rate of 1000 cm3/min, and condensation temperature of 10 8C.
W.T. Tsai et al. / J. Anal. Appl. Pyrolysis 76 (2006) 230–237
235
Fig. 6. GC–MS chromatogram of the pyrolysis liquid products from (a) rice straw, (b) sugarcane bagasse, and (c) coconut shell.
Fig. 5. FT-IR spectra of the pyrolysis liquid products from (a) rice straw, (b) sugarcane bagasse, and (c) coconut shell.
Notably, the band at 1278 cm 1 may be connected with –C–O– C– (e.g., ethers). In the study of this spectral region described above, FT-IR analysis might be used as a fast screen technique to observe the extent of carbonyl group or oxygen content. Furthermore, some features in the elemental, calorific, pH and FT-IR analyses could be further demonstrated with the GC–MS analysis. GC–MS analysis was carried out in order to get an idea of the nature and type of organic compounds in the pyrolysis liquid products. Due to the rather similar peak patterns of chromatograms of the pyrolysis liquid products obtained at the other pyrolysis parameters, Fig. 6 shows, for example, the GC– MS chromatograms of the pyrolysis liquid products obtained at a heating rate of 200 8C/min, pyrolysis temperature of 500 8C, holding time of 1 min, particle size of <0.50 mm, and condensation temperature of 10 8C. Clearly, the pyrolysis
liquid products are such an unknown and complex mixture of organic compounds that no calibration of the MS detector was set, mainly due to the lack of an appropriate standard mixture for calibration [21,24]. Table 4 lists the tentative compounds of pyrolysis liquid products, which are the most probable compounds identified by the MS search file (HP MS ChemStation). The compounds have been numbered in Fig. 6 so that each compound has the same number in Table 4. The empirical formulae and their molecular weights are also included in Table 4. In view of the results presented in Table 4, it can be seen that, as expected, the chemical compositions of pyrolysis bio-oils are very similar to the inclusion of a lot of aromatics and oxygenated compounds such as carboxylic acids, phenols, ketones, etc. For a preliminary study, a corresponding GC–MS of pyrolysis gas sample was also performed in order to investigate the presence of condensable or non-condensable compounds in the vent gas. It was found that the pyrolysis gas still contains a comparatively small amount of condensable organic compounds such as aromatic hydrocarbons (e.g., benzene, 2-methyl-furan and toluene). This result explains why at the pyrolysis conditions total yields of water, bio-oil and char were about 75%, since at the pyrolysis conditions, all noncondensable and few condensable products were vented, resulting in the difficulty with
236
W.T. Tsai et al. / J. Anal. Appl. Pyrolysis 76 (2006) 230–237
Table 4 Tentative GC–MS characterization of pyrolysis liquid products from biomass wastesa Peak no.
tRb (min)
Component
% Area Rice straw
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
14.4 21.8 23.1 24.5 25.7 27.8 29.7 32.1 32.7 33.0 34.3 35.2 35.9 37.2 37.7 38.2 39.2 40.0 41.7 44.0 44.9 46.1 49.1 51.1
Acetic acid (C2H4O2) Furan, 2,5-dimethyl- (C6H8O) 2-Furanmethanol (C5H6O2) 2H-pyran, 3,4-dihydro (C5H8O) 2(3H)-furanone, 5-methyl- (C5H6O2) Phenol (C6H6O) 1,2-Cyclopentanedione-3-methyl- (C6H8O2) Phenol, 2-methoxy- (C7H8O2) Maltol (C6H6O3) 2-Cyclopentan-1-one, 3-ethyl-2-hydroxy- (C7H10O2) Phenol, 4-ethyl- (C8H10O) 1,2-Benzenediol (C6H6O2) Benzofuran, 2,3-dihydro- (C8H8O) 1,2-Benzenediol, 3-methyl- (C7H8O2) 1,2-Benzenediol, 3-methoxy- (C7H8O3) Phenol, 4-ethyl-2-methoxy- (C9H12O2) 4-Hydroxy-3-methylacetophenone (C9H10O2) Phenol, 2,6-dimethoxy (C8H10O3) Benzaldehyde, 3-hydroxy-4-methoxy (C8H8O3) Phenol, 2-methoxy-4-(1-propenyl)- (C10H12O2) Ethanone, 1-(4-hydroxy-3-methoxyphenyl)- (C9H10O3) 2-Propanone, 1-(4-hydroxy-3-methoxypheny)- (C10H12O3) Benzeneacetic acid, 4-hydroxy-3-methoxy- (C9H10O4) Phenol, 2,6-dimethoxy-4-(2-propenyl)- (C11H14O3)
Total
Sugarcane bagasse
Coconut shell
0.9 3.1 1.6 1.6 1.4 1.7 4.9 3.1 0.8 1.2 1.8 1.5 3.1 2.4 1.1 0.9 1.3 2.0 0.7 0.4 0.1 0.4 0.2 0.1
7.6 0.4 1.3 1.2 1.6 1.3 2.5 1.5 0.4 0.8 1.3 5.7 5.7 2.8 1.9 0.3 0.6 2.9 0.2 0.2 0.1 0.3 0.3 0.2
3.6 0.5 1.1 1.1 1.1 7.8 2.6 1.3 1.3 0.6 0.7 8.6 1.6 2.7 1.5 0.2 0.7 2.8 0.5 0.2 0.2 0.8 0.2 0.5
37.6
41.1
42.2
a
Obtained at heating rate of 400 8C/min, pyrolysis temperature of 500 8C, holding time of 1 min, particle size of <0.50 mm, nitrogen flow rate of 1000 cm3/min, and condensation temperature of 10 8C. b See Fig. 6.
the capture in the present cryogenic system. Further studies on the GC–MS analyses of the resulting gases would be helpful to elucidate the fast pyrolysis system. From the results of FT-IR and GC–MS analyses, it can be concluded that the pyrolysis liquid products are extremely complex and may be composed of hundreds of organic compounds. The main components of the pyrolysis liquid products include acids, phenols, ketones, aldehydes, ethers, and some species of aromatics, which are in accordance with those obtained from the fast pyrolysis of woods [25–27]. The presence of these aromatic and oxygenated compounds is attributable to its biopolymer textures such as cellulose and hemicellulose. These results are consistent with the elemental compositions and chemical characterization of the pyrolysis liquid products previously presented in Tables 2 and 3 and Figs. 2–4, that showed high water and oxygenated contents and low heating values. Obviously, the highly oxygenated organic components would need to be first separated from the water phase, and then upgraded using hydrotreating–hydrocracking process in order to further raise their heating values and reduce the corrosiveness when they are used as a potential fuel.
It was found that the yield of liquid tar product from sugarcane bagasse was higher than those of rice straw and coconut shell. Employing the higher pyrolysis temperature of >500 8C, faster heating rate of >200 8C/min, and longer holding time of >2 min, the pyrolysis tar yield reached a maximum of ca. 50%. All analytical results of the resulting pyrolysis products analyzed by elemental analyzer, pH meter and bomb calorimeter are in consistence with those analyses by FT-IR spectroscopy and GC–MS. However, it was noted that the pyrolysis liquid product contains a significant amount (>65%) of water, and fewer contents of complex compounds mostly composed of aromatic and carbonyl structures, resulting in low pH and low heating values. The results therefore indicate that a large quantity of water evolves within a short time due to the water content of biomass feedstock and dehydration reaction in the fast pyrolysis process. GC–MS analyses have shown that carboxylic acids, phenols, alcohols and branched oxygenated hydrocarbons are the main compounds of the bio-oil. It is necessary that the resultant liquid product would need further processing to remove the condensed water for chemical feedstock and/or biofuel production.
4. Conclusions
Acknowledgement
In this study, fast pyrolysis experiments of three biomass wastes including rice straw, sugarcane bagasse and coconut shell were conducted in a fixed-bed induction-heating system.
This research was partly supported by NSC (National Science Council), Taiwan, under contract no. NSC 92-2623-7041-001-ET.
W.T. Tsai et al. / J. Anal. Appl. Pyrolysis 76 (2006) 230–237
References [1] P. McKendry, Bioresour. Technol. 83 (2002) 37. [2] J.M. Ebeling, B.M. Jenkins, Trans. ASAE 28 (1958) 898. [3] B.M. Jenkins, L.L. Baxter, T.R. Miles Jr., T.R. Miles, Fuel Process Technol. 54 (1998) 17. [4] D.L. Klass, Biomass for Renewable Energy, Fuels, and Chemicals, Academic Press, San Diego, 1998, pp. 225–331. [5] A. Demirbas, Energy Convers. Manage. 42 (2001) 1357. [6] P. McKendry, Bioresour. Technol. 83 (2002) 47. [7] R. Gross, M. Leach, A. Bauen, Environ. Int. 29 (2003) 105. [8] A.V. Bridgwater, S.A. Bridge, in: A.V. Bridgwater, G. Grassi (Eds.), Biomass Pyrolysis Liquids Upgrading and Utilization, Elsevier Applied Science, London, 1991, pp. 11–92. [9] G. Maschio, C. Koufopanos, A. Lucchesi, Bioresour. Technol. 42 (1992) 219. [10] S. Czernik, A.V. Bridgwater, Energy Fuels 18 (2004) 228. [11] D. Meier, O. Faix, Bioresour. Technol. 68 (1999) 71. [12] S. Warnijati, I.B. Agra, Sudjono, Renew. Energy 9 (1996) 934. [13] P.T. Williams, N. Nugranad, Energy 25 (2000) 493. [14] M. Garcia-Perez, A. Chaala, C. Roy, J. Anal. Appl. Pyrol. 68 (2002) 111.
237
[15] W.T. Tsai, Y.H. Chou, Y.M. Yang, Renew. Sustain. Energy Rev. 8 (2004) 22. [16] A.E. Putun, A. Ozcan, E. Putun, J. Anal. Appl. Pyrol. 52 (1999) 33. [17] O. Onay, S.H. Beis, O.M. Kockar, J. Anal. Appl. Pyrol. 58/59 (2001) 995. [18] Y. Shinogi, Y. Kanri, Bioresour. Technol. 90 (2003) 241. [19] C. Acikgoz, O. Onay, O.M. Kockar, J. Anal. Appl. Pyrol. 71 (2004) 417. [20] F. Ates, E. Putun, A.E. Putun, J. Anal. Appl. Pyrol. 71 (2004) 779. [21] M.F. Laresgoiti, B.M. Caballero, I. de Marco, A. Torres, M.A. Cabrero, M.J. Chomon, J. Anal. Appl. Pyrol. 71 (2004) 917. [22] W.T. Tsai, C.Y. Chang, S.L. Lee, Carbon 35 (1997) 1198. [23] A.V. Bridgwater, D. Meier, D. Radlein, Org. Geochem. 30 (1999) 1479. [24] A. Torres, I. de Marco, B.M. Caballero, M.F. Laresgoiti, M.A. Cabrero, M.J. Chomon, J. Anal. Appl. Pyrol. 58/59 (2001) 189. [25] J. Piskorz, D.S. Scott, D. Radlein, in: E.J. Soltes, T.A. Milne (Eds.), Pyrolysis Oils from Biomass—Producing, Analyzing, and Upgrading, ACS Symposium Series 376, American Chemical Society, Washington, DC, (1988), pp. 168–178. [26] D.S. Scott, J. Piskorz, D. Radlein, in: R.M. Rowell, T.P. Schultz, R. Narayan (Eds.), Emerging Technologies for Materials and Chemicals from Biomass, ACS Symposium Series 476, American Chemical Society, Washington, DC, (1992), pp. 422–436. [27] X. Dai, C. Wu, H. Li, Y. Chen, Energy Fuels 14 (2000) 552.