Bioresource Technology 99 (2008) 83948399

Contents lists available at ScienceDirect

Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech

Thermal conversion of elephant grass (Pennisetum Purpureum Schum) to bio-gas, bio-oil and charcoal
Vladimir Strezov a,*, Tim J. Evans b, Chris Hayman c
a

Graduate School of the Environment, Macquarie University, Sydney, NSW 2109, Australia Hamersley Iron, GPO Box A42, Perth, WA 6837, Australia c Rio Tinto Brazil Ltda., Torre Rio Sul Rua Lauro Mller, 116 Sala 3505, Botafogo Cep 22.290-160, Rio de Janeiro, Brazil
b

a r t i c l e

i n f o

a b s t r a c t
Elephant grass is an abundant, fast growing plant with signicant potential as a renewable energy source and for conversion to higher caloric value fuels. This work investigates thermal conversion of elephant grass to bio-gas, bio-oil and charcoal under two heating rates of 10 and 50 C/min. The energy required to pyrolyse elephant grass was evaluated using computer aided thermal analysis technique, while composition of the resultant bio-gas and bio-oil products were monitored with gas chromatographic and mass spectroscopic techniques. At 500 C, the bio-gas compounds consisted primarily of CO2 and CO with small amounts of methane and higher hydrocarbon compounds. The heat of combustion of the bio-gas compounds was estimated to be 3.77.4 times higher than the heat required to pyrolyse elephant grass under both heating rates, which conrms that the pyrolysis process can be self-maintained. Faster heating rate was found to increase the amount of liquid products by 10%, while charcoal yields remained almost the same at 30%. The bio-oil mainly consisted of organic acids, phthalate esters, benzene compounds and amides. The amount of organic acids and benzene compounds were signicantly reduced at 50 C/min, while the yields of phthalate esters and naphthalene compounds increased. The difference in bio-oil composition with increased heating rate is believed to be associated with the reduction of the secondary reactions of pyrolysis, which are more pronounced under lower heating rate. 2008 Elsevier Ltd. All rights reserved.

Article history: Received 19 December 2007 Received in revised form 24 February 2008 Accepted 27 February 2008 Available online 11 April 2008 Keywords: Energy crop Pyrolysis Charcoal Bio-oil Renewable energy

1. Introduction The importance for economically viable and environmentally sustainable energy supply is becoming the primary factor of the current international developments in the energy transformation technology. Biomass as an energy source is now at the forefront of the renewable energy developments due to its renewable character and a net carbon neutral energy conversion. Biomass sources are abundant and can be supplied from various industries which produce organic plant materials as wastes, such as agricultural, sugar, wood processing, food industry, etc., or alternatively, biomass plants can be purposely grown for energy supply. Sustainability of purposely cultivating plants for energy use is often challenged with the main argument focused on the anticipated threat from competition between energy and food crops placing further pressure on available agricultural land. Lemus and Lal (2005), however, claimed that worldwide over 2 Gha of available, non-arable soils are suitable for energy crop cultivation. These soils are marginal and degraded with little agricultural use. They currently require
* Corresponding author. Tel.: +61 2 9850 6959; fax: +61 2 9850 7972. E-mail address: vstrezov@gse.mq.edu.au (V. Strezov). 0960-8524/$ - see front matter 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2008.02.039

protection from soil erosion and, moreover, energy crop cultivation can signicantly improve soil health and quality. One of the promising methods for biomass conversion is the thermal pyrolytic conversion. Under pyrolytic conditions biomass is heated in absence of air to the maximum temperatures of 450500 C. During heating biomass decomposes liberating noncondensable (bio-gas) and condensable (bio-oil) volatiles, while the remaining solid residue is carbon rich charcoal. Generally, slow heating conditions, also termed as carbonisation, favour charcoal production, while high heating rates, known as ash pyrolysis, enhance the bio-oil yields. Pyrolysis is one of the oldest technologies where charcoal was originally the main product of interest, primarily for metallurgy (Antal and Grnli, 2003), but more recently pyrolysis is considered as a technology suitable for agricultural waste to energy conversion (Maiti et al., 2006; Strezov et al., 2007). Schell et al. (2008) have noted that the development of a viable biorenewable industry has to integrate all of the co-products of pyrolysis in the process of value adding. Bio-gas and bio-oils are attractive co-products of pyrolysis as they are possible energy carriers and feedstock for bio-diesel and petrochemicals. The industrial scale biomass pyrolysing facilities for charcoal production are currently based on the principle of either industrial

V. Strezov et al. / Bioresource Technology 99 (2008) 83948399

8395

kilns or circulating retort systems (FAO, 1987). Heating conditions applied in these systems vary signicantly between each other. The charcoal making process in industrial kilns is batch based and requires between 15 and 30 days to complete the cycle, most of which is required for cooling of the charge. The more efcient circulating retort system operates under continuous or batch mode with a range of 812 h of carbonisation time. Faster carbonisation rates of 30 min have recently being proposed (Antal et al., 2003) as an alternative to the conventional carbonisation processes. The primary interest of the current study was to evaluate the potential for production of charcoal from the energy crop Pennisetum Purpureum Schum, known as elephant grass. Elephant grass is a plant species native to Africa with growth rates as high as 40 tons of dry biomass per hectare per annum (Woodard and Prine, 1993). Over a century ago the plant was also introduced to South America and Australia as forage for livestock. Elephant grass requires very little supplementary nutrients for growth and can be harvested up to four times a year, which makes this plant one of the most prospective crops for energy use (Osava, 2007). The work presented here explores the opportunity for production of charcoal and bio-oils from elephant grass. The application of this work is directed for the rural parts of Brazil, but the same energy crop can have similar applications in other countries where elephant grass is a native plant. The thermal conversion of this crop has potential to assist local communities and industry by supplying raw materials and energy. While parts of rural Brazil are rich in iron ore, the sources of solid carbon to drive the reduction are very limited. Charcoal produced from energy crops can be applied as an energy and reductant source for ironmaking of the locally supplied iron ore, while the bio-oils can be used for various liquid fuel applications, including conversion to bio-diesel. Cultivation of energy crops may prove benecial to serve this purpose. One of the critical points to assess feasibility of elephant grass pyrolysis is to quantify if the pyrolysis process can be self-sustained by supplying pyrolysis heat through combustion of the produced bio-gas products. The aim of this research was to evaluate the potential of elephant grass to produce charcoal, bio-oils and suitable bio-gas that can provide combustion heat to drive the pyrolytic heating. The type of oils and effect of heating conditions on the nal products were also investigated. 2. Experimental 2.1. Materials An energy crop Pennisetum Purpureum Schum, known as elephant grass or napier grass, was supplied from Brazil and was used to carry out the pyrolysis experiments for this study. The sample was crushed to size fractions of 4 mm and 212 lm. The size fraction used for analysis was 4 mm, except for thermogravimetric analysis (TGA) where samples with size fraction of 212 lm were applied. Prior to each analysis the samples were dried in an oven at 80 C for 2 h.
Table 1 Analysis of elephant grass Analysis basis Total moisture (%) Proximate analysis Moisture (%) Fixed carbon (%) Volatile matter (%) Ash (%) Ultimate analysis Carbon (%) Hydrogen (%) Nitrogen (%) As received 12.4 7.9 18.8 70.3 3.0 43.8 5.08 0.45 Air dried

Table 2 Mineralogical analysis of the ash produced from elephant grass Ash analysis Silicon Aluminium Iron Calcium Magnesium Sodium Potassium Titanium Manganese Phosphorus Sulphur Strontium Barium Zinc Vanadium SiO2 Al2O3 Fe2O3 CaO MgO Na2O K2O TiO2 Mn3O4 P2O5 SO3 SrO BaO ZnO V2O5 (% db) 43 <0.1 1.4 1.9 9.9 <0.01 30.5 0.03 0.17 7.2 5.7 0.03 0.08 0.08 <0.01

The proximate and ultimate analysis of the sample is shown in Table 1. On a dry basis, the xed carbon content of the sample was 20.4%, while volatile matter content was 76.3%. Table 2 shows mineralogical data of the ash samples analysed by X-ray uorescence (XRF) spectrometry based on the Australian Standard Method AS1038.14.3. Ashing of the biomass was performed by heating under air at 815 C for 30 min. The ash sample was then cooled and subjected to XRF spectroscopy. The sample was found to contain high silicon and potassium content, followed by magnesium, phosphorus and sulphur. 2.2. Thermal analysis Specic heat of the elephant grass sample was determined using a novel Computer Aided Thermal Analysis technique. Detailed description of the experimental procedure can be found elsewhere (Strezov et al., 2003a). The thermal analysis apparatus consisted of an infrared furnace and arrangement of internals for heating of a packed bed of sample. The analysed sample with 0.8 g mass was packed in a silica glass tube to the density of 310 kg/m3. The sample was insulated on the sides with ceramics and heated under argon atmosphere with a graphite heating element positioned inside the furnace. In this work, the heating rate of the furnace was xed at 10 and 50 C/min and the heating was carried out until graphite temperature reached 900 C. Temperatures of the graphite, surface and centre of the packed sample were acquired at a frequency of 1 Hz using chromelalumel thermocouples. The specic heat was estimated simultaneously by applying an inverse numerical technique to the measured temperatures. For the purpose of calculations, the sample was divided into a grid with assumed 530 nodes across the radius. The heat balance for each node was calculated based on the heat conduction principle where heat accumulated by the node equals the difference of the input and output heats from the node. The estimated specic heat had apparent values, hence heats evolved during decomposition of the heated sample were included in the specic heat. Therefore, during an endothermic heat effect, the specic heat showed increasing values, while during an exothermic reaction the specic heat values decreased. Reported calculations for the specic heat and energy balance in the current work are generally based on the initial mass of the sample. 2.3. Gas chromatographic analysis of volatiles Volatiles evolved during pyrolysis of elephant grass were analysed separately with gas chromatograph. M200 Micro gas chromatograph from MTI Analytical Instruments was connected to

17.8 66.9 2.9 41.6 4.83 0.43

8396

V. Strezov et al. / Bioresource Technology 99 (2008) 83948399

the gas outlet of the glass sample tube. A molecular sieve 5A column at 90 C was used to separate H2 and CO while analysis of CO2, CH4, C2H4, and C2H6 was performed on a Poraplot U column at 55 C. Chromatograms were obtained every 85 s using a gas thermal conductivity detector. Helium gas at a rate of 50 mL/min was passed through 50 mg of biomass while maintaining continuous heating rate up to the maximum heating temperature of 900 C. Two samples were analysed, one at a heating rate of 10 C/min and the second at 50 C/min. Two reproducible experiments were arranged for each heating rate to conrm the analysis. 2.4. Analysis of bio-oils The bio-oils produced from pyrolysis of elephant grass were condensed at room temperature for further analysis. The bio-oils were produced by heating 0.5 g of elephant grass at a heating rate of 10 and 50 C/min up to the maximum temperature of 500 C. The bio-oils were then dissolved using a 4:1 volume mixture of chloroform and methanol, and were subjected to analysis using matrix assisted laser desorption ionization mass spectroscopy (MALDI). The MALDI mass spectroscopy was performed on a Micromass/Waters TOFSPEC 2E time of ight mass spectrometer system. The system used nitrogen laser of 337 nm with a 4 ns pulse. All spectra were in the positive-ion mode. Bio-oils were also analysed by ash pyrolysis GC/MS and results used to identify individual species present in the bio-oils. 2.5. Thermogravimetric analysis Approximately 20 mg of elephant grass was analysed with the thermogravimetric analyser from TA Instruments 2050 TGA using nitrogen as carrier gas. Two heating rates of 10 and 50 C/min were applied for the measurement. The TGA analysis was performed up to the maximum temperature of 900 C. 3. Results and discussion The volumetric specic heat of elephant grass determined with the computer aided thermal analysis technique at a heating rate of 10 C/min and the corresponding rate of volatile evolution during pyrolysis are shown in Fig. 1. During heating elephant grass exhibited a low temperature endothermic reaction, followed by primarily exothermic heat event. In the temperature range between 80 C and 200 C the heat of reaction was found to be substantially endothermic. Since the sample was dried prior to analysis, this endothermic reaction is believed to correspond to decomposition of the strongly bonded hydrated compounds releasing water upon

heating (Liou, 2004; Yaman, 2004). At temperatures above 250 C the reaction was exothermic and evolved at temperatures of up to 400 C. The high temperature exotherm is most likely due to decomposition of the lignin fraction, considering lignin decomposition is overall exothermic and occurs at a similar temperature range as the observed exothermic reaction in this case (Strezov et al., 2003b). Once temperatures exceed 400 C there is no evidence of any further signicant reactions. Evolution of volatiles was continuously monitored in this work using micro gas chromatograph. The oxides of carbon were found to evolve at lower temperature range compared to hydrocarbons and hydrogen. While CO and CO2 are products of primary devolatilisation of biomass, hydrocarbons and hydrogen are most likely released after secondary thermal cracking of the evolved oils and reformation of the high temperature vapour phase. The results from gas analysis and separate thermogravimetric measurements were applied to calculate the evolution of pyrolytic liquids with temperature. Fig. 2 shows a comparison of the total evolution rate of volatiles, liquids and the rate of weight loss superimposed to the volumetric specic heat. The rate of weight loss and rates of evolution of both liquid and gas products exhibited peaks at approximately 320 C. This temperature coincides with the overall exothermic reaction of elephant grass pyrolysis. It should be noted that the liquid fraction shown in Fig. 2 comprises of a mixture of aqueous and oil compounds. Generally, under industrial pyrolytic conditions for either charcoal or bio-oil production, the pyrolysis process occurs at temperatures of maximum 500 C. Most of the liquids, CO and CO2 in this analysis evolved prior to reaching 500 C, while the hydrocarbons evolved primarily at temperatures above 500 C. In this work CO2 was found to be the largest volatile compound of elephant grass pyrolysis reaching 18.5% of the dry sample weight at 500 C and 20% at 900 C. Large amount of released CO2 is generally undesirable because it is non-combustible gas and decreases the potential for energy recovery from combustion of the volatile products. Regardless of the large amounts of released CO2, the net caloric value that could theoretically be recovered through stoichiometric combustion of the evolved pyrolytic volatiles at 500 C is estimated at 10 MJ/kg of dry elephant grass. The estimate is calculated accounting for the amounts of measured combustible volatiles (CO, CH4, C2H4 and C2H6) at 500 C and using the corresponding heat of combustion for each volatile. These measurements revealed that at temperatures of up to 500 C hydrogen was still not released. In comparison to the heat of combustion of resultant volatiles, the energy required to heat dry elephant grass to temperature of 500 C is estimated at only 1.35 MJ/kg. This estimate is provided by integration of the apparent specic heat shown in Fig. 1 and

1.6

2.5 CO2 2 CO Cp 1.5

0.08

2.5

Evolution Reate (wt%/min)

Evolution Reate (wt%/min)

1.4 1.2 1 0.8 0.6 0.4 0.2 0 0 200 400 600

Specific Heat (MJ/m K)

0.06 0.05 0.04 0.03 0.02 0.01 0 0 200 400 600

1.5

0.5

0.5

0 800

0 800

Temperature (C)

Temperature (C)

Fig. 1. Rate of evolution of volatile compounds from pyrolysis of elephant grass superimposed to the apparent specic heat for the heating rate of 10 C.

Specific Heat (MJ/m K)

0.07

CH4 C2H4 C2H6 H2 Cp

V. Strezov et al. / Bioresource Technology 99 (2008) 83948399

8397

7
dwt%/dt

2.5

Evolution Rate (wt%/min)

Total Gases

Specific Heat (MJ/m3K)

6 5 4 3 2

Total Liquids Cp
1.5

Table 3 Individual compounds detected in elephant grass oils obtained at 500 C and heating rate of 10 C/min Compound name Phthalate esters Benzene compounds Phenols Octadecanoic acid Hexadecanoic acid Pyrans Amides Propenoic acid 1,2-Benzenedicarboxylic acid Octadecene Furans Acetate Phenyls Bromomethylbutene Benzaldehyde Acetic acid Docosanoic acid 1,3-Dioxolane-2,2diethanol Amphetamines Oleic Acid 1-Hexacosene Ethanone Oleanolic acid Cholest-5-en-3-ol (3b)Dimethyl 2-chloro-5triuorome Cyclopentenones 6,7-Epoxypregn wt% Compound name wt% 0.69 0.68 0.66 0.65 0.63 0.63 0.62 0.61 0.50 0.49 0.48 0.47 0.47 0.45 0.42 0.42 0.42 0.38 0.35 0.35 0.34 0.33 0.29 0.29 0.28 0.26 0.09

0.5

1 0 0 100 200 300

0

Temperature (C)

400

500

600

700

800

900

11.90 3,12-Oleandione 11.20 2,2,4-Trimethyl-3-(3,8,12,16)-tetramethylheptadeca-3,7 8.61 Docosahexaenoic acid, 1,2,3-pro 7.22 Piperazinyl 5.50 Naphthalenes 5.08 2-Propanone, 1-(4-hydroxy-3methoxyphenyl) 4.23 Propanoic acid 3.48 Adamantane 2.95 Hydroquinone 2.58 2.49 2.34 2.28 2.02 1.70 1.60 1.53 1.46 1.28 1.23 1.14 1.10 0.79 0.75 0.72 0.72 0.70 Acetoxyhydroxy Pregnanoic acid Ergostatrienol Pyrazoles 3,5-Diacetoxy-cyclopent-1-ene 4-Ethylcatechol 1,4:3,6-Dianhydro-a-d-glucopyranose Rhodopin Trimethoprim 30 ,8,80 -Trimethoxy-3-piperidyl Tricosanoic acid, methyl ester Dodecanoic acid 5a-Androstane-3,17-dione Cyclooctanone Flopropione 7-Dehydrodiosgenin Tetradecanoic acid Phenethylamine

Fig. 2. Rate of weight loss and evolution rate of volatiles and liquids superimposed to the volumetric specic heat.

using the initial packing density of the sample. The initial estimate shows that potential energy recovery from combustion of the released pyrolytic volatile products should be more than sufcient to provide energy to self-maintain the pyrolysis of this energy crop. However, when an industrial pyrolysis process is modelled, it is critical that the heat losses of the system, losses from non-stoichiometric combustion conditions, as well as excess energy required to remove free moisture should also be accounted and carefully considered. The pyrolytic bio-oils are one of the desirable products of biomass pyrolysis. They are usually mixed with condensed water forming liquid fraction of pyrolysis. Their composition is important to evaluate the potential for direct combustion of the bio-oils or for their further processing and utilisation as bio-diesel or chemicals. For this purpose, the pyrolysis liquids were collected when elephant grass was heated to 500 C and subjected to detailed evaluation for molecular weight distribution and compound identication. Fig. 3 shows the MALDI spectra for higher molecular weight hydrocarbon liquid compounds. The result shows a range of compounds, for which identication was possible when ash pyrolysis GC/MS technique was applied on the collected liquid sample. Table 3 summarises the individual bio-oil compounds that represent 99% of the total bio-oil weight. Results in Table 3 are shown as a weight percent of the total amount of bio-oil applied for analysis. The primary products of the collected bio-oil were found to consist of various groups of organic acids representing a total of 27.3% of the bio-oil weight (or 12.2% of dry biomass weight). The phthalate esters comprised a total of 11.9% of bio-oil weight (5.3% of dry biomass weight), which also appeared as a peak in Fig. 3 with molecular weight of 458 amu. The bio-oils also contained larger fraction of various benzene chemicals, phenols and

100 90 80 70 60 50 40 30 20 10 0 250 300 350 400 450 500 550

Mass (m/z)
Fig. 3. Molecular weight distribution of liquids produced by pyrolysis of elephant grass to 500 C at 10 C/min determined using MALDI mass spectroscopy.

pyrans. Bio-oils are generally high in acids with their pH ranging between 2 and 3.7 which is considered as a signicant disadvantage, as the oils can become corrosive to common construction materials (Elliott, 2007; Oasmaa and Czernik, 1999; Yaman, 2004). Hydrothermal processing of bio-oils using various catalysts is recommended to reduce the bio-oil acidity as well as to remove the high oxygen content of the bio-oils. During this process bio-oils are upgraded under steam reforming conditions resulting in a product that allows production of bio-diesel, or the processed bio-oils can be used as base material to produce marketable chemicals (Sanders et al., 2007). The effect of heating rate on pyrolysis of elephant grass was investigated in this work by increasing the heating rate to 50 C/ min and monitoring the thermophysical properties of the heated sample. Fig. 4 compares the apparent specic heats for the two applied heating rates of 10 and 50 C/min. The result shows differences in the heats of reactions for the two heating rates, which were estimated by subtracting the integral of the measured specic heat curve in the temperature range of the observed reactions and the integral of an assumed straight line baseline which connects the start and end temperatures of each reaction. The heat of the rst endothermic reaction in the range of 80 and 200 C was estimated to be 30 MJ/m3 when the sample was heated at 10 C/min, while for the higher heating rate of 50 C/min, the reaction was 57.5 MJ/m3. The heat of the second exothermic reaction was estimated at 18.5 MJ/m3 for the lower heating rate, and 47.5 MJ/m3 for the higher heating rate. Higher heating rate was found to not only increase the overall heat of reactions, but also shift the reactions to higher temperature range. Table 4 summarises product yields of elephant grass pyrolysis for the two examined heating rates at the maximum temperature of 500 C. Higher heating rate was found to decrease the amount of volatiles with the oxides of carbon being the most affected.

% Intensity

8398

V. Strezov et al. / Bioresource Technology 99 (2008) 83948399

Volumetric Specific Heat (MJ/m3K)

4 3.5 3 2.5 2 1.5 1 0.5 0 0 100 200 300 400 500 600 700

50C/min

10C/min

Temperature (C)
Fig. 4. Volumetric specic heat of elephant grass sample at 10 and 50 C/min.

Table 4 Comparison of the yields of volatiles, liquid products and charcoal expressed in% of dry biomass for heating rates of 10 and 50 C/min and for the maximum temperature of 500 C Yields (%) at 500 C 10 C/min CO2 CO CH4 C2H4 C2H6 H2 Liquids Charcoal 18.6 6.6 0.6 0.10 0.17 0 44.7 29.3 50 C/min 11.0 3.1 0.39 0.07 0.12 0 54.37 31

by-products. Considering increase in the heating rate also reduces evolution rate of combustible volatiles and, consequently, the theoretical caloric recoverable value from combustion of the resultant volatiles, a threshold may exist in how fast the carbonisation process can be carried, or otherwise the liquid products would have to be compromised to drive the pyrolysis. Elephant grass has vast energy potential, even if cultivated only in agriculturally non-arable and degraded soils. Brazil currently has an estimated 100 Mha of land facing desertication (Malheiros, 2004), which could be potentially used for cultivation of elephant grass. Using the average growth rates of 40 tons of dry biomass per hectare per annum (Woodard and Prine, 1993), the non-arable soils in Brazil could theoretically cultivate 4 Gt of dry biomass per year, which, when pyrolysed, could produce 1.2 Gt of charcoal and between 1.8 and 2.2 Gt of bio-oils. Based on the statistics derived from the Global Market Information Database and using the data obtained from the current work, Brazils theoretical potential for charcoal production from elephant grass is 4.7 times larger than worlds total charcoal production for year 2007, while its theoretical bio-oil production yield from elephant grass pyrolysis is in the range of 4555% from the worlds total crude oil production for 2007. The above initial estimates indicate that elephant grass has potential to play one of the critical roles in addressing the challenges that arise from the worlds growing energy demands. 4. Conclusion Elephant grass is fast growing and low nutrient demand plant that has potential for use in charcoal making and bio-oil production. The work shown here demonstrates that pyrolysis of elephant grass can produce combustible bio-gas compounds with sufcient caloric value to provide the internal heat of pyrolysis. Investigation of the effect of heating rate on composition of the pyrolysis products revealed that higher heating rate of carbonisation (50 C/min), not only reduces the carbonisation time, but also produces bio-oils with smaller acidic and benzene fractions than the case of lower heating rate carbonisation (10 C/min). Initial estimates on the potential of elephant grass for charcoal and bio-oil production suggest that Brazilian non-arable soils alone have theoretical substantial potential to produce 1.2 Gt of charcoal and an average of 2 Gt of bio-oils per year. References
Antal, M.J., Grnli, M., 2003. The art, science, and technology of charcoal production. Industrial and Engineering Chemistry Research 42, 16191640. Antal, M.J., Mochidzuki, K., Paredes, L.S., 2003. Flash carbonization of biomass. Industrial and Engineering Chemistry Research 42, 36903699. Branca, C., Giudicianni, P., Di Blasi, C., 2003. GC/MS characterization of liquids generated from low-temperature pyrolysis of wood. Industrial and Engineering Chemistry Research 42, 31903202. Elliott, D.C., 2007. Historical developments in hydroprocessing bio-oils. Energy and Fuels 21, 17921815. Food and Agriculture Organization of the United Nations (FAO), 1987. Simple technologies for charcoal making. FAO Forestry Paper 41, Rome. Lemus, R., Lal, R., 2005. Bioenergy crops and carbon sequestration. Critical Reviews in Plant Science 24, 121. Liou, T.-H., 2004. Evolution of chemistry and morphology during the carbonization and combustion of rice husk. Carbon 42, 785794. Maiti, S., Dey, S., Purakayastha, S., Ghosh, B., 2006. Physical and thermochemical characterization of rice husk char as a potential biomass energy source. Bioresource Technology 97, 20652070. Malheiros, J.O., 2004. 17 puntos que ajudam a explicar o que la deserticaco, a convenco da ONU e o processo de construco do PAN-LCD Brasileiro. RIOD BRASIL, 11 pp. Oasmaa, A., Czernik, S., 1999. Fuel oil quality of biomass pyrolysis oils state of the art for the end users. Energy and Fuels 13, 914921. Osava, M., 2007. Elephant Grass for Biomass. Inter Press Service News Agency, October 10. Sanders, J., Scott, E., Weusthuis, R., Mooibroek, H., 2007. Bio-renery as the bioinspired process to bulk chemicals. Macromolecular Bioscience 7, 105177.

The yield of liquid products was found to increase with heating rate, while the estimated charcoal yields were little affected. The main reason for the change in pyrolytic product composition is believed to be the reduction of secondary reactions under higher heating rate. Consequently, the theoretical caloric recoverable value from combustion of the resultant volatiles for the heating rate of 50 C/min is almost halved to 6 MJ/kg compared to 10 C/min, while the energy required to heat dry elephant grass to 500 C increased to 1.6 MJ/kg. The composition of bio-oils is also affected by heating rate. Increased heating rate was found to reduce oil acidity producing 11% of organic acids by bio-oil weight (or 6% dry biomass weight) comparing to 27.3% (12.2% of biomass weight) at 10 C/min. This amount is in a similar range to acid yields obtained previously by Branca et al. (2003) from pyrolysis of wood (5.26.1% of dry wood weight), although the composition differed. Wood pyrolysis was reported to produce primarily acetic acid and small amounts of formic acid, while the primary acidic compounds of elephant grass pyrolysis were octadecanoic and hexadecenoic acids. Higher heating rate was also found to reduce the amounts of benzene from 11.2% at 10 C/min to only 2% of bio-oil weight at 50 C/min. The yields of phthalate esters were signicantly increased from 11.9% to 25.4% at 50 C/min, while phenols remained at a similar level of 8%. Naphthalene compounds were found to represent 24.3% of the bio-oils produced at 50 C/min compared to only 0.63% at 10 C/min. It is very likely that under lower heating rates naphthalene products as well as phthalates undergo the most severe secondary reactions producing organic acids, oxides of carbon and hydrocarbon volatiles. Current work suggests that elephant grass carbonisation under faster heating rates will not only signicantly reduce the required carbonisation time but it will also improve properties of the liquid

V. Strezov et al. / Bioresource Technology 99 (2008) 83948399 Schell, C., Riley, C., Peterson, G.R., 2008. Pathways for development of a biorenewable industry. Bioresource Technology, 99, 51605164. Strezov, V., Lucas, J.A., Strezov, L., 2003a. Computer aided thermal analysis. Journal of Thermal Analysis and Calorimetry 72, 907918. Strezov, V., Moghtaderi, B., Lucas, J.A., 2003b. Thermal study of decomposition of selected biomass samples. Journal of Thermal Analysis and Calorimetry 72, 10411048.

8399

Strezov, V., Patterson, M., Zymla, V., Fisher, K., Evans, T.J., Nelson, P.F., 2007. Fundamental aspects of biomass carbonisation. Journal of Analytical and Applied Pyrolysis 79, 91100. Woodard, K.R., Prine, G.M., 1993. Dry matter accumulation of elephantgrass, energycane and elephantmillet in a subtropical climate. Crop Science 33, 818824. Yaman, S., 2004. Pyrolysis of biomass to produce fuels and chemical feedstocks. Energy Conversion and Management 45, 651671.