Combustion behavior and thermal analysis of agricultural and woody biomass blends.
Global wood pellet demand is expanding rapidly and consumers, businesses and regulators are looking for alternatives to fossil fuels . Currently, pellet industries rely heavily on forestry biomass, which could face threat from biomass power station in the near future. Wood pellet demand is projected to grow from 23 million tons from year 2014 to 50 million tons in year 2024 . This means new feedstock resources have to be exploited in order to support the future demand. Therefore, pellet industry is now looking for alternate biomass resources to increase the pellet production. Canada as one of the main producers of pellets has increased the pellets fuel export by 25 percent every year . Wood and agricultural residues account about 42% of the total Canadian biomass . About 25 Mtoe (Megatonne of oil equivalent) of agricultural residues is generated annually in Canada .
Furthermore, agricultural biomass has been highly recommended by industries as a feedstock for making pellet [5, 6 and 7]. This would help to increase the production of pellets and obviously can fill the demand.
Generally, pellets made from agricultural biomass show poor combustion behavior due to the chemical components specifically the inorganic components as compared to woody biomass. Other than the combustion behavior, the inorganic mineral contents also affect the thermal characteristics and emissions. For instance, Werther et al.  reported that some agricultural residues have high alkali oxides and salts, which may lead to various problems during combustion due to their low melting points. Few others have also reported the ash related problem due to inorganic matters such as potassium, chlorine and sodium in the biomass during combustion [9 and 10]. According to Demirbas :
* The high moisture and ash contents in biomass fuels can cause ignition and combustion problems.
* The melting point of the ash is usually low and it causes fouling and slagging problems.
* And due to the above problems it was anticipated that blending biomass with higher quality coal will reduce flame stability problems, as well as minimize corrosion effects.
Further, agricultural biomass has also been used as fuel for combustion but its use can be limited due to low heating value as compared to woody biomass. Therefore, it was anticipated that blending agricultural residues with woody biomass may improve the combustion behavior. Furthermore, the studies related to blends are limited to biomass with coal, biomass with plastic, and other materials. However, information on blending of agricultural residues with woody biomass is very limited in the literature. Therefore, none has investigated the combustion behavior of agricultural and woody biomass blends using TGA and DSC techniques. Except, TGADSC analysis of mischantus and poplar [12, 13] was found.
The present study focuses on the combustion behavior of the selected agricultural (Reed Canary grass, timothy hay and switchgrass) residues and their blends with woody biomass (pine and spruce). In this work, agricultural residues from reed canary grass, timothy hat and swicthgrass were chosen as biomass feedstock since they are available in large quantity in New Brunswick. A particular attention was also given to the mineral content, TGA and DSC analysis, and heat release.
Agricultural biomass such as reed canary grass, timothy hay and switchgrass was selected and acquired from agricultural farm in New Brunswick (NB). The woody biomass (spruce and pine) were obtained from a wood mill situated near Fredericton, NB. The raw biomass materials were stored in a closed container at ambient conditions. Prior to analysis, the samples were dried in the oven at 105[degrees]C for 24 h, and then grinded using Wiley mill to less than 1 mm particle size. The grinded materials were sieved according to ASTM standards (D 2013-72) to collect the particle size in the range of 150-300 pm. The grinded biomass and their blends were stored in airtight container for further analysis. It should be noted, that heating value, proximate, and ultimate analysis was conducted for individual and blended biomass as presented in Table 1.
Proximate and ultimate analysis:
The ultimate analysis was performed using LECO CHNS 932 elemental analyzer as per the ASTM D5291 method and was repeated twice. Element oxygen was calculated by difference. Proximate analysis was determined twice using TG analyzer (Q500 TA Instrument Inc.) as per ASTM E1641-04 method, while the ash content was investigated as per NREL/TP-510-42622 method with 6 replications. The higher heating value (HHV) of a biomass was measured using a bomb calorimeter (Isoperibol Oxygen Bomb Calorimeter) according to the standard method, ASTM D-5865.
The mineral concentration was determined according to U.S. EPA methods and procedures. The samples were prepared using microwave-assisted digestion (MarsXpress, CEM Corp., Matthews, N.C.) in nitric acid according to EPA Method 3051. The resulting solutions were analyzed for trace elements by ICP-MS (Thermo X Series II, Thermo Fischer Scientific, Inc., Waltham, Mass.) according to EPA Method 200.8. Phosphorus was determined by ICP-ES (Varian Vista AX, Agilent Technologies, Inc., Santa Clara, Cal.) according to EPA Method 200.7. Silica was determined by ICP-ES on sodium peroxide fusions of samples ash. Chlorine was determined calorimetrically on aqueous leaches of the samples using a photometric analyzer (AquaKem 250, Thermo Fischer Scientific, Inc., Waltham Mass.) according to APHA Standard Method 4500-CL-E.
To evaluate the effect of blending and the type of biomass (agricultural and forestry) on the combustion behavior, the samples were subjected to TG (TA Instruments, USA) and DSC analyzer. The samples were heated from room temperature to 1000[degrees]C with heating rate of 20[degrees]C/min. This heating rate was selected based on the highest volatiles and char yield that it produced from spruce and pine , and from miscanthus, poplar and rice husk . Air was used as purge gas with a flow rate of 100 ml/min.
RESULTS AND DISCUSSION
Chemical and Mineral content of individual and blended biomass:
The heating value of blended biomass (18-19 MJ/kg) was between agricultural residues (17-18 MJ/kg) and forestry biomass (19-20 MJ/kg), as presented in Table 2. There was a slight improvement in the heating value of agricultural residues after blending them with forestry residues. For instance, the heating value (15.64 MJ/kg) of the blends (50% fir, 30% beech and 20% wheat straw) was higher than wheat straw (14.7 MJ/kg) and beech (15.00 MJ/kg) . Barmina et al.  also found that the heating value of blend biomass (19.1 MJ/kg) was slightly higher than the reed canary grass biomass (18.6 MJ/kg).
In comparison with spruce and pine, the selected agricultural biomass contained higher proportion of oxygen, but less hydrogen and carbon (Table 2). These could be the reason for lower heating value of agricultural biomass since the energy contained in carbon-oxygen bonds is usually lower  than carbon-hydrogen bond. However, after blending, the carbon content was increased by around 2% for RCG, 2.5% for hay, and 4.5% for switchgrass, whereas hydrogen content increased by around 2% (averaged) for all blend biomass. Higher oxygen content indicates that the biomass will have higher thermal reactivity .
The moisture content of the blended biomass increased slightly, which might have contributed from woody biomass. However, the moisture content of blend biomass was comparable to that of bituminous coal . Volatile matters in agricultural biomass ranged from 65 to 78%. After blending, its values were close to woody biomass (69-73%). The ash content in blend biomass reduced by about 62% (averaged) as compared to individual agricultural residues and this could be due to considerable reduction in char content of blend biomass.
A particular attention was also given to analyze the individual and blend biomass for mineral content. Table 3 shows that all blend biomass had a lower sulfur content (< 0.05%) that can comply with the European pellet standard . These characteristics are favorable for combustion applications , as compared to coal.
All other chemical and mineral elements such as nitrogen, silver, cadmium, arsenic, lead, chromium, nickel and zinc, were observed lower than the standard limits by European Standards for wood pellets, EN 14961 -2 (as shown in Table 3) in all biomass including blend. The chlorine content was higher in individual agricultural residues and some blend biomass than the standard limit of 0.03%, which is not favorable for thermal system since it can cause agglomeration on the heating wall. Overall, switchgrass was found to have higher nickel, copper, zinc content compared to other biomass including blend.
The DSC and DTA curves revealed two different exothermic reaction regions as shown in Figure 1. The curve obtained for each biomass was analyzed to determine the relevant combustion parameters, such as peak temperatures, ignition temperature for combustion, and burnout temperature. First region was attributed due to the combustion of light volatile matters and the second region might be due to the combustion of heavy volatile and fixed carbon . Similar reaction regions were observed for rice husk using TGA and DSC . In this study, it is observed that the temperature range of each region was different for each biomass species and blends (Figure 1).
The high volatile content in spruce and pine as presented in Table 2 was also confirmed by the DTA curve (first region) in Figure 1 and DTG curve in Figure 2. Karampinis et al.  characterized miscanthus and poplar using DSC and it revealed that these biomass are reactive fuels because of high volatile matter. As presented in Figure 1, the combustion behavior of blended biomass (RCG+S and RCG+P) was found to follow the combined effect of chemical composition present in agricultural residues (RCG) and woody biomass (S and P). Since spruce and pine showed highest area under the curves of heat flow profiles, they can be considered as highly exothermic or reactive biomass fuel. On the other hand, the heat released by the blended biomass (RCG+S and RCG+P) was higher than the individual agricultural (RCG) biomass.
Similarly, two main reaction regions during the combustion of RCG, spruce, pine and their blends were observed from DTG profiles. The first region refers to devolatization of biomass that took place at temperature around 200[degrees]C due to combustion of light volatiles present in the biomass. As observed in Figure 2, two "shoulder peaks" occur in the first region of the DTG curve at around 220[degrees]C and 315[degrees]C, respectively, for reed canary grass. The same behavior was also found for other agricultural biomass (timothy hay and switchgrass) . This peak could be attributed due to the decomposition of hemicellulose and cellulose, respectively. On the contrary, only one peak was observed in spruce and pine. This could possibly be due to their lower hemicellulose content than cellulose, which caused to merge the reaction mechanism . Another reason could be due to delay in the thermal decomposition of the hemicellulose . The DTG result in Figure 2 shows only one reaction peak for reed canary grass and woody biomass blend. This could be due to greater contribution of woody biomass characteristics in the blend. Apparently, in first reaction region the main reaction peak for woody biomass appeared earlier (335[degrees]C) than RCG (315[degrees]C) and their blends (328[degrees]C).
From Figure 2, the initial ignition temperature (180[degrees]C) was similar in case of RCG and it blends (RCG+S and RCG+P). For spruce and pine it was slightly higher (220[degrees]C) than woody biomass. The reactivity in combustion regions is proportional with the height of DTG peak (Kok and Ozgur, 2013). The DTG peaks as shown in Figure 2 for reed canary grass was lower (0.7%/[degrees]C) than reed canary grass blends (1.1%/[degrees]C) and woody biomass (1.5%/[degrees]C). However, the degradation trend of reed canary grass blends followed the degradation of woody biomass. Therefore, blending process is believed to improve the reactivity as compared to individual agricultural biomass.
For the second region, combustion of more complex and thermally stable structure and formation of char took place. In this phase, DSC curves (Figure 1) corresponds to decomposition of lignin. The curves indicated that the decomposition of woody biomass (spruce and pine) was the highest as compared to blends, and individual agricultural biomass. For other biomass samples, the change in ignition and burnout temperature of biomass samples is presented in Table 3, and for the blended biomass is presented in Fig. 3. The combustion of woody biomass and switchgrass started at 220[degrees]C temperature (Table 4). However, combustion of reed canary grass and timothy hay begin earlier (180 and 160[degrees]C, respectively). High ignition temperature indicates higher heat capacity of a biomass . Combustion of heavy volatiles in all agricultural biomass completed within 150 [degrees]C temperature difference. Compare to woody biomass, combustion of heavy volatiles in agricultural biomass was within 80[degrees]C temperature difference. This could be due to higher ash and mineral content in agricultural biomass as presented in Table 2 and 3, respectively. Similar observation was reported by Kok and Ozgur .
From Figure 3, the combustion temperature range for all blended biomass was almost similar, which was around 180-500[degrees]C. Irrespective of parent biomass as presented in Table 3, the combustion characteristics of all blended biomass were uniform. The ignition temperature, when the light volatiles starts to eliminate rapidly, was found almost similar at 180[degrees]C, for most blended biomass. However the peak was highest (1.64%/[degrees]C and 1.43%/[degrees]C) for swicthgrass blends (SW+S and SW+P), and the lowest peak (1.1%/[degrees]C) was observed for reed canary grass blends (RCG+S and RCG+P) and timothy hay blends (H+S and H+P). This could be due to the difference in the hemicellulose and cellulose fractions  of blended biomass samples.
Heat released during the combustion of biomass samples at heating rate of 20[degrees]C/min was calculated based on the area under DSC curve as presented in Figure 1 and Figure 3b. The low heat release (4.5-6.7 kJ/kg) in agricultural biomass might be due to breaking of low volatile chemical bound. Similar results were found by by Kok and Ozgur  for rice husk and miscanthus biomass. In woody biomass the chemical bonds are not easily broken and this might result in the release of higher amount of heat (10-12 kJ/kg) than agricultural biomass. The heat released due to combustion of individual agricultural and woody biomass as well as their blends is shown in Figure 5.
Blending of agricultural with woody biomass improved the release of energy. Several reason could be behind such result. For instance, increase in carbon and hydrogen content, and contribution of lignin from woody biomass in the blend. As presented in Figure 5, the energy released from the blend biomass was in the range of 7.0 to 10 kJ/kg.
Combustion behavior and thermal analysis of selected agricultural and woody biomass and their blend was investigated in this study. The differences in thermal behavior of agricultural and woody biomass were analyzed and the following conclusions were derived:
* The reactivity, combustion behavior and degradation rate peak of the agricultural biomass changed after blending them with woody biomass.
* Typically, agricultural biomass contains higher amount of inorganic minerals which can result in higher emissions and problem during combustion process. Blending agricultural residues with forestry biomass helped to reduce the sulfur, copper, chlorine and ash content that can possibly meet the standard limit of biomass fuel.
* There was some improvement in heat released (6.94 to 9.26 kJ/kg) during combustion for blended biomass as compared to individual agricultural biomass (reed canary grass-5.4 kJ/kg, timothy hay- 6.78 kJ/kg, and switchgrass-4.59 kJ/kg).
Overall, blending of agricultural biomass with woody biomass can be considered as one of the alternative options to increase the biomass fuel feedstock such as pellet production to meet the future demand. Future works will be proposed to study the biomass availability in Malaysia for industrial application and meeting pellets demand of neighbourhood country.
Received 6 June 2015
Accepted 19 July 2015
Available online 1 August 2015
The authors appreciate the financial assistance from New Brunswick Department of Agriculture, Aquaculture and Fishries; New Brunswick Soil and Crop Improvement Association; New Brunswick Agricultural Council; Agriculture and Agri-Food Canada; New Brunswick Innovation Foundation; and Universiti Teknologi PETRONAS, Malaysia.
 Walker, S., 2015. Wood Bioenergy Magazine. Available at: www.woodbioenergy.com. Accessed.
 Exportation of wood pellets from North America to Europe reached a new record, [Online]. Available at: http://www.forestbioenergyreview.com/pellets/item/127-exportation-of-wood-pellets-from-north-america-toeurope-reached- a-new-record. Accessed 15 April 2013.
 Wood, S.M., and D.B. Layzell, 2003. A Canadian biomass inventory: feedstocks for a bio-based economy. BIOCAP Canada Foundation Final Report Industry Canada Contract #: 5006125.
 Helwig, T., R. Jannasch, R. Samson, A. DeMaio and D. Caumartin, 2002. Agricultural biomass residue inventories and conversion systems for energy production in Eastern Canada Final Report Prepared for Natural Resources-Canada.
 Samson, R., S. Bailey, and C. Ho Lem, 2007. The Emerging Agro-Pellet Industry in Canada Resource Final Report. Efficient Agricultural Production (REAP)-Canada
 Than, D. and C. Hennig, 2011. Global Wood Pellet Industry Market and Trade Study. In IEA Bioenergy Task 40: Sustainable Bioenergy Trade., pp: 190.
 BioDimension, 2012. West Kentucky Agricultural Biomass Pellet Report. Report for West Kentucky Center for Emerging Technology, Murray State University, and Memphis Bioworks Foundation., pp: 55.
 Werther, J., M. Saenger, E.U. Hartge, T. Ogada and Z. Siagi, 2000. Combustion of agricultural residues. Progress in Engineering and Combustion Science, 26: 1-27.
 Wang, K.S., K.Y. Chiang, C.C. Tsai, C.J. Sun, C.C. Tsai and K.L. Lin, 2001. The effects of Fe[Cl.sub.3] on the distribution of the heavy metals Cd, Cu, Cr, and Zn in a simulated multimetal incineration system. Environment International., 26(4): 257-63.
 Khan, A.A., W. de Jong, P.J. Jansens and H. Spliethoff, 2009. Biomass combustion in fluidized bed boilers: potential problems and remedies. Fuel Processing Technology, 90(1): 21-50.
 Demirbas, B., 2009. The global climate challenge: Recent trends in CO2 emissions from fuel combustion. Energy Education Science Technology, 22: 179-193.
 Karampinis, E., D. Vamvuka, S. Sfakiotakis, P. Grammelis, G. Itskos and E. Kakaras, 2012. Comparative study of combustion properties of five energy crops and Greek lignite, Energy Fuels, 26(2): 869-878.
 Kok, M.V. and E. Ozgur, 2013. Thermal analysis and kinetic of biomass samples. Fuel Processing Technology, 106: 739-743.
 Harun, N.Y. and M.T. Afzal, 2010. Thermal Decomposition Kinetics of Forest Residue. Journal Applied Science, 10(12): 1122-1127
 Prvulovic, S., Z. Gluvakov, J. Tolmac, D. Tolmac, M. Matic and M. Brkic, 2014. Methods for determination of biomass energy pellet quality. Energy fuels, 28: 2013-2018.
 Barmina, I., A. Lickrastina, R. Valdmanis, M. Zake, A. Arshanitsa, V. Solodovnik and G. Telysheva, 2013. Effects of Biomass Composition Variations on Gasification and Combustion Characteristics. Engineering for Rural Development, pp: 282-287.
 Munir, S., S.S. Daood, W. Nimmo, A.M. Cunliffe and B.M. Gibbs, 2009. Thermal analysis and devolatilization kinetics of cotton stalk, sugar cane bagasse and shea meal under nitrogen and air atmospheres. Bioresource Technology, 100: 1413-1418.
 Haykiri-Acma, H. and S. Yaman, 2008. Effect of co-combustion on the burnout of lignite/biomass blends: A Turkish case study. Waste Management, 28: 2077-2084.
 Vamvuka, D., E. Kakaras, E. Kastanaki and P. Grammelis, 2003a. Pyrolysis characteristics and kinetics of biomass residuals mixtures with lignite. Fuel, 82: 1949-1960.
 Vamvuka, D., N. Pasadakis, E. Kastanaki, P. Grammelis and E. Kakaras, 2003b. Kinetic modeling of coal/agricultural by-product blends. Energy Fuel, 17: 549-558.
 Biagini, E., F. Lippi, L. Petarca and L. Tognotti, 2002. Devolatilization rate of biomasses and coal-biomass blends: an experimental investigation. Fuel, 81: 1041-1050.
 Biagini, E., A. Fantei and L. Tognotti, 2008. Effect of the heating rate on the devolatilization of biomass residues. Thermochim Acta. 472: 55-63. DOI: 10.1016/j.tca.2008.03.015.
 Harun, N.Y. and M.T. Afzal, 2015. Chemical and Mechanical Properties of Pellets Made from Agricultural and Woody Biomass Blends. Transactions of the ASABE, 58(4): American Society of Agricultural and Biological Engineers ISSN 2151-0032 DOI 10.13031/trans.58.11027
 Shen, D.K, S. Gu, K.H. Luo, A.V. Bridgwater and M.X. Fang, 2009. Kinetic study on thermal decomposition of woods in oxidative environment. Fuel, 88: 1024-1030.
 Garcia-Maraver, A., L.C. Terron, M. Zamorano and A.F. Ramos-Ridao, 2013. Thermal events during the combustion of agricultural and forestry lopping residues. In Materials and process for energy: communicating current research and technological developments (A. Mendez-Vilas. Ed.)
(1) Noorfidza Y Harun and (2) Muhammad T Afzal
(1) Universiti Teknologi PETRONAS, Department of Chemical Engineering, 32610 Bandar Seri Iskandar, Perak, Malaysia.
(2) University of New Brunswick, Department of Mechanical Engineering, 15 Dineen Drive, E3B 5A3 Fredericton, New Brunswick, Canada.
Corresponding Author: Noorfidza Y Harun, Universiti Teknologi PETRONAS, Department of Chemical Engineering, 32610 Bandar Seri Iskandar, Perak, Malaysia.
Table 1: Biomass and blending ratio. Agricultural Forestry RCG H SW S P Proportion 100 100 100 100 100 (wt. %) Blends S+RCG P+RCG S+H P+H S+SW P+SW Proportion 50:50 50:50 50:50 50:50 50:50 50:50 (wt. %) Notes: RCG=Reed Canary grass, H=Timothy hay, SW=Switchgrass, S=Spruce and P=Pine Table 2: Heating value, ultimate and proximate analysis of individual and blended biomass. RCG H SW S P RCG+S HH MJ/kg V 18.61 17.58 18.20 19.15 19.75 18.56 C (wt. %, 45.45 45.46 45.03 47.14 46.80 46.64 H dry 5.89 5.95 5.98 6.06 6.14 6.06 O basis) 48.44 48.37 48.84 46.76 47.02 47.13 Ash (wt. %, 5.34 4.06 3.61 traces traces 2.07 MC dry 6.4 6.9 7.0 8.0 9.3 7.2 VM basis) 65 66 78 70 70 69 FC 17.84 18.51 11.70 21.33 20.70 21.57 RCG+P H+S H+P SW+S SW+P HH MJ/kg V 19.00 18.07 18.68 18.46 18.44 C (wt. %, 45.98 46.57 46.72 47.12 47.21 H dry 6.02 6.01 6.08 6.08 6.12 O basis) 46.97 47.36 47.14 46.75 46.64 Ash (wt. %, 1.54 1.51 1.62 1.53 1.55 MC dry 7.5 7.3 7.6 7.7 7.5 VM basis) 70 70 69 73 71 FC 20.94 21.03 21.33 17.77 19.96 Notes:RCG=reed canary grass, H=timothy hay, SW=switchgrass, S=spruce and P=pine Table 3: Chemical analysis of biomass and blended biomass. RL RCG H SW S P RCG+S Chlorine 0.03 0.13 0.27 0.03 0.01 0.26 0.06 Sulfur 0.05 0.08 0.09 0.04 0.01 0.01 0.04 Nitrogen 0.3 0.17 0.18 0.12 0.04 0.04 0.04 Silver 0.1 0.02 0.02 0.02 0.02 0.02 0.02 Cadmium 0.5 0.02 0.02 0.08 0.22 0.11 0.12 Arsenic 1 < 0.2 < 0.2 < 0.2 < 0.2 < 0.2 < 0.2 Lead 10 0.14 0.11 0.22 0.03 0.09 0.09 Chromium 10 0.2 0.3 0.4 0.4 < 0.2 0.3 Nickel 10 4.2 5.5 7.3 0.6 0.5 2.4 Copper 10 6.2 9.8 12.8 1.8 2.6 5.1 Zinc 100 12.2 9.3 13.9 8.5 10.1 10.4 RCG+P H+S H+P SW+S SW+P Chlorine 0.01 0.07 0.06 0.01 0.08 Sulfur 0.04 0.03 0.03 0.01 0.02 Nitrogen 0.04 0.04 0.04 0.03 0.03 Silver 0.02 0.02 0.02 0.02 0.02 Cadmium 0.06 0.17 0.08 0.19 0.09 Arsenic < 0.2 < 0.2 < 0.2 < 0.2 < 0.2 Lead 0.11 0.06 0.12 0.18 0.26 Chromium < 0.2 0.4 < 0.2 0.4 < 0.2 Nickel 2.5 2 1.8 2.3 2.2 Copper 5 3.9 6.7 9.7 9.1 Zinc 11.4 8.7 8.5 10 10.9 RL=Relative Limits based on European Standards for Energy Pellet (Prvulovic et al., 2014) RCG=reed canary grass, H=timothy hay, SW=switchgrass, S=spruce and P=pine Table 4: Reaction regions of parent biomass Combustion of Combustion of light volatiles heavy volatiles Spruce (S) 220-400 400-470 Pine (P) 220-400 400-480 Reed canary grass (RCG) 180-370 370-520 Timothy hay (H) 160-370 370-510 Switchgrass (SW) 220-375 375-520
|Printer friendly Cite/link Email Feedback|
|Author:||Harun, Noorfidza Y.; Afzal, Muhammad T.|
|Publication:||Advances in Environmental Biology|
|Date:||Aug 1, 2015|
|Previous Article:||Occurrence of 17a-ethinylestradiol (EE2) in Inanam River, Sabah, Malaysia.|
|Next Article:||Development of automated model for standardized estimation of the Biotope Area Ratio(BAR) based on GIS.|