ap-3-12.dvi Acta Polytechnica Vol. 52 No. 3/2012 A Thermogravimetric Study of the Behaviour of Biomass Blends During Combustion Ivo Jǐŕıček1, Pavla Rudasová2, Tereza Žemlová1 1 Power Engineering Department, Institute of Chemical Technology, Prague, Technická 5, Prague 6, Czech Republic 2 Škoda Power, s. r. o., Tylova 1/57, 301 28 Pilsen, Czech Republic Correspondence to: jiriceki@vscht.cz Abstract The ignition and combustion behavior of biomass and biomass blends under typical heating conditions were investigated. Thermogravimetric analyses were performed on stalk and woody biomass, alone and blended with various additive weight ratios. The combustion process was enhanced by adding oxygen to the primary air. This led to shorter devolatiliza- tion/pyrolysis and char burnout stages, which both took place at lower temperatures than in air alone. The results of the ignition study of stalk biomass show a decrease in ignition temperature as the particle size decreases. This indicates homogeneous ignition, where the volatiles burn in the gas phase, preventing oxygen from reaching the particle surface. The behavior of biomass fuels in the burning process was analyzed, and the effects of heat production and additive type were investigated. Mixing with additives is a method for modifying biofuel and obtaining a more continuous heat release process. Differential scanning calorimetric-thermogravimetric (DSC-TGA) analysis revealed that when the additive is added to biomass, the volatilization rate is modified, the heat release is affected, and the combustion residue is reduced at the same final combustion temperature. Keywords: straw, lignin, peat, charcoal, combustion behavior, biomass blends, thermogravimetry. 1 Introduction Biomass, such as straw, grasses and wood, is used in various forms for energy production. Many tech- nologies for biomass utilization have been studied in the last two decades, e.g. (co)-combustion, pyroly- sis, gasification and liquefaction. These technologies are in various stages of development, whereby com- bustion is most developed and most frequently ap- plied. Biofuel products are sometimes mixed with other biomass, semi-fossil peat, fossil coal and cat- alyst to achieve better control of the burning pro- cess [1]. Until recently, there have been few studies on the co-firing of biomass blends for energy gen- eration [2]. It is anticipated that blending low-grade biomass with higher-quality biomas will reduce flame stability problems, and will also minimize corrosion effects due to the deposited ash containing low melt- ing point salts. The present work was undertaken to determine whether blending different biomass fuels influences the combustion performance, and whether the addition of a specific char additive can modify the burning velocity in the burnout stage and unify the thermal properties. Non-isothermal thermogravime- try was applied to determine the combustion char- acteristic of six samples, namely wheat straw, rape straw, flax straw (leftover after scutching), pulp-mill lignin, garden peat, and hardwood charcoal. 2 Materials and experiments All the samples originated from the Eastern EU coun- tries. The initial samples were milled, sieved and seven different particle fractions between 0.08 mm < D < 2 mm in diameter were used for the ignition study. For the combustion study, the sample parti- cle size was in the range of 0.15–0.25 mm. This size enables uniform packing on the sample pan. Prox- imate and ultimate analyses of these biomass sam- ples were made using standard procedures, i.e. the ASTM D 3172-89 method and the TGA method [3]. The gross calorific value (GCV) was determined as per ASTM D2015-66. The results are given in Ta- ble 1. Thermogravimetric tests were performed in the TA instruments Q600 simultaneous differential scanning calorimetry-thermogravometry(DSC-TGA) apparatus. The weight precision of the instrument is 0.1 μg. Small samples weighing about 5 mg were placed in an open platinum sample pan, and uniform packing of the samples was ascertained. The samples were heated to 1 000 ◦C at a constant heating rate of 10 ◦C/min, under a constant air-oxygen flow rate of 120 ml/min (air flow rate of 100 ml/min and high purity oxygen flow rate of 20 ml/min) through the sample chamber. The thermograms were analyzed to determine the relevant combustion parameters. (dw/dt)max indi- 39 Acta Polytechnica Vol. 52 No. 3/2012 cates the maximum reactivity attained in terms of rate of weight loss (%/min) at DTG (1st derivative of the TG curve) peak temperatures, (dw/dt)mean is the average rate of weight loss. To determine the ignition temperature, two points on the TG curve were first identified. One is the point at which a vertical line from the sharp DTG peak (dw/dt)max crosses the TG curve. The other is the point at which devolatilization begins. Two tangents drawn to the TG curve at these points intersect at the ignition temperature (Ti). The burnout temperature (TBO) was obtained in a similar way. 3 Results and Discussion The effect of particle size on ignition temperature was investigated on wheat straw and flax straw sam- ples. Other samples are being tested, and the results will be reported later. The ignition temperature (Ti) for seven particle fractions was found to follow linear behavior between 0.08 mm < D < 2 mm. Linear regressions gave the following relationships: Wheat straw Ti ( ◦C) = 241.9 + 2.731 · D, Flax straw Ti ( ◦C) = 265.3 + 2.427 · D. These values imply that wheat straw ignites at lower temperatures than flax straw. Although the proximate analysis results differ considerably, the ig- nition temperatures of the biomass samples changed only in a narrow range, see Table 2. Generally, volatile matter, flammability of vo- latiles, and transport from the particle determine whether ignition of an isolated particle occurs hetero- geneously or homogeneously (gas ignition). Accord- ing to thermal explosion theory, heterogeneous igni- tion characterizes the decrease in ignition tempera- ture as the particle size increases [1]. Our data shows the decrease in ignition temperature as the particle size decreases, which indicates homogeneous ignition. Volatiles seem to evolve early in the combustion se- quence. After ignition, they burn in the gas phase, preventing oxygen from reaching the particle surface. When a volatile hydrocarbon is burned, a large num- ber of different oxygen radicals are involved in various radical chain reactions. The combustion study showed that biomass degradation takes place in two steps: between 180 ◦C and 370 ◦C, volatiles are released and burned (de- volatilization/pyrolysis step), and at 370–490 ◦C char combustion takes place (oxidation step). Biomass can be divided into three categories according to the heat release process. Most biomass falls into the first category, which is characterized by major heat re- lease in the devolatilization/pyrolysis step. Wheat straw and rape straw are typical representatives of this category, where the predominant form of com- bustion is gas-phase oxidation of the volatile species, see Figure 1. Peat and pulp-mill lignin fall into the second category, characterized by the highest heat re- lease in the char oxidation step. The third category is reserved for biomass chars. Table 1: Analyses of the samples in wt. % dry base and low heat value(LHV) Fuel Volatile matter Ash C K Cl S LHV (wt. %) (wt. %) (wt. %) (wt. %) (wt. %) S(wt. %) (MJ/kg) Wheat straw 81.0 6.6 41.6 2.36 0.11 0.10 15.5 Rape straw 79.5 3.5 40.0 2.50 0.20 0.20 14.8 Flax straw 78.6 3.6 45.2 n.a. 0.06 0.04 19.1 Pulp-mill lignin 63.6 5.3 55.7 0.01 0.074 0.73 21.0 Peat 65.0 2.8 60.0 2.86 0.17 0.20 23.0 Charcoal 20.8 0.6 86.0 0.43 0.001 0.01 33.0 Table 2: Ignition temperature Ti, burn-out temperature TBO and combustion characteristic factor CCF for biomass fuels of particle size 0.15–0.25 mm Fuel Ti, ◦C TBO, ◦C CCF · 10−7 Fuel Ti, ◦C TBO, ◦C CCF · 10−7 Wheat straw 243 447 1.97 Lignin 257 462 1.19 Rape straw 247 448 1.98 Peat 251 445 1.40 Flax straw 266 440 2.22 Charcoal 425 513 2.95 40 Acta Polytechnica Vol. 52 No. 3/2012 Figure 1: Heat release process in biomass fuels Figure 2: Heat release process in straw-lignin blends Blending seems advantageous, if samples from dif- ferent categories are chosen. For example 20 % and 40 % addition of pulp-mill lignin to wheat straw mod- ifies the volatilization rate and affects the oxidation. The heat release is more balanced and more contin- uous, see Figure 2. The released heat can be calculated from the DSC curves as: Q = β ∫ ∞ ignition (T − Ti) dt = β · S = β · Δw · h, (1) where β is the heat transfer constant from the sample to the metal wall; S is the area under the DSC curve; Δw is the average width of the heat release peak; h is the exothermic peak height. Neglecting the difference in the β constant between these biomass species, the area under the DSC curve reflects the heat releasing state. However, the value of this heat is lower than the gross calorific value, because of the loss from in- completely burned volatiles. The effect of the heat release process on charcoal and its blends with the additive is shown in Figure 3. The additive contains a catalyst that enables oxy- gen transport. The charcoal used in the study seems to have some semi-char content which upon heating degrades up to a temperature of about 500 ◦C. Af- ter this, the final char oxidizing step takes place. If mixed with a suitable ratio of additive, for example below 20 %, the heat release can be more continu- ous, and the burnout temperature decreases; thus, the combustion efficiency is increased. Wood char- coal combustion is a solid phase reaction. A hetero- geneous reaction involving carbon is usually slower than a gas-phase reaction, and this seems to be the reason for the higher burnout temperatures (TBO) that are observed. The difference in burnout temper- ature between charcoal and other biomass samples was in the range of 51–73 ◦C, see Tabular 2. The additive ignites at a temperature of 440 ◦C and is burned out at a temperature of 508 ◦C, which is 5 ◦C lower than the burnout temperature of charcoal. The additive blends investigated here were able to lower the burnout temperature of charcoal by up to 4 ◦C. Figure 3: Heat release process in charcoal-additive blends A parameter called the combustion characteristic factor CCF [4] can be used as a criterion for fuel combustion performance, defined as: CCF = ( dw dt ) max · ( dw dt ) mean T 2i · TBO , (2) where (dw/dt)max and (dw/dt)mean are the maxi- mum and average burning velocity (%/min); Ti and TBO are the ignition temperature and the burnout temperature (K). This factor, which includes the ease of ignition, the firing velocity and the burnout temperature, is a comprehensive parameter, used here to compare the combustion performance of biomass fuels. CCF values were calculated for several biomass fuels, and are listed in Table 2. With the exception of pulp- mill lignin and peat, the values are near to or greater than 2 for all other biomass fuels, indicating their good general burning performance. The CCF values 41 Acta Polytechnica Vol. 52 No. 3/2012 for the blends were found to fall between the val- ues of the original fuels. The additive CCF value was found to be the highest. Fuels with the highest CCF values e.g. charcoal (CCF = 2.95 · 10−7) and the additive (CCF = 3.23 · 10−7) suggest that they may be advantageously used for blending with other biomass. 4 Conclusion In this study, thermogravimetric experiments were performed on a number of biomass species intended for use as fuels. Some quantitative characteristics during ignition, devolatilization/pyrolysis, char burn- ing and the burnout stages are listed and compared. Combustion of wheat straw showed a longer tran- sition stage between volatilization and char burning, so mixing wheat straw with a lignin additive is a method for modifying the biofuel and obtaining a more continuous heat release process. When an additive is added to charcoal, the heat release is affected and the burnout temperature de- creases; thus, the combustion efficiency increases. The comprehensive parameter CCF for straw biomass fuels and charcoal in this project is near to or greater than 2, indicating good combustion per- formance. In further work, we intend to study additives for intensifying heterogeneous charcoal oxidation, which is the slowest step in the overall wood combustion process. Acknowledgement Financial support for specific university research in MSMT project no. 21/2010 is gratefully acknowl- edged. References [1] Sami, M., Annamalai, K., Wooldridge, M.: Co- firing of coal and biomass fuel blends, Progress in Energy and Combustion Science, 2001, Vol. 27, p. 171–214. [2] Van Loo, S., Koppejan, J.: The handbook of biomass combustion and co-firing. Earthscan, 2008, p. 22–38. ISBN 978-1-84407-249-1. [3] Jǐŕıček, I., Žemlová, T., Macák, J., Janda, V., Viana, M.: Paliva, 2009, Vol. 1, p. 19–22. [4] Nie, Q. H., Sun, S. Z., Li, Z. Q.: Thermogravi- metric analysis of the combustion characteristic of the brown coal blends, Combustion Science and Technology, 2001, Vol. 7, p. 71–76. 42