- Email: [email protected]
Experimental investigation of wood pellet swelling and shrinking during pyrolysis
Fuel 142 (2015) 145–151
Contents lists available at ScienceDirect
Fuel journal homepage: www.elsevier.com/locate/fuel
Experimental investigation of wood pellet swelling and shrinking during pyrolysis gas Rolandas Paulauskas ⇑, Algis Dzˇiugys, Nerijus Striu Laboratory of Combustion Processes, Lithuanian Energy Institute, Breslaujos Str. 3, LT-44403 Kaunas, Lithuania
h i g h l i g h t s The wood pellet changes in the radial direction during pyrolysis was captured with the digital camera. The swelling phenomenon of wood pellet was observed during pyrolysis. Heating temperature of pyrolysis influenced the radial swelling and shrinking of wood pellet. The estimated peaks of the heating rates indicate the chemical processes and their intensity inside the wood pellet. A discussion was given for the mechanism of the wood pellet swelling phenomenon during pyrolysis.
a r t i c l e
i n f o
Article history: Received 9 July 2014 Received in revised form 30 September 2014 Accepted 5 November 2014 Available online 17 November 2014 Keywords: Wood pellets Pyrolysis The center temperature Swelling Shrinking
a b s t r a c t Gasification of wood pellets in downdraft gasifier is confronted with the problem of granulated fuel agglomeration. It occurs when the wood pellets are moving from the pyrolysis zone to the oxidation zone and pellets stick together in lumps, which disrupts the movement of fuel and stops further process. In order to investigate the cause and regularities of fuel agglomeration, experimental studies of wood pellets thermal deformations during pyrolysis were carried out. Experimental studies were performed in an electrically heated horizontal tubular furnace from 300 °C to 1000 °C temperature in an inert atmosphere capturing wood pellet thermal deformations by a digital camera. During investigation the center temperature of the pellet was also measured. Analyzed results showed that wood pellet final diameter decreased when the heating temperature increased. It was also established that the pellet expanded from 400 °C to 900 °C heating temperature at the beginning of pyrolysis process. The maximum swelling effect was reached at 550 °C temperature and after it swelling intensity was decreasing till 900 °C temperature. Over 900 °C heating temperature, the expansion phenomenon was no longer visible. The tendency of wood pellet size alteration (swelling and shrinking) depends on heating temperature. The obtained results explain the reason for adhesion of wood pellets in the gasifier and reveal regularity of wood pellet size changes with increasing heating temperature. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Renewable fuel resources are widely used to produce electricity and heat. The growing demand for these stocks and their price results in searching for ways how to use low-quality biofuels. One of the ways is gasification [1]. During this process, the solid fuel is gasified and turned into higher quality gas, which is used to generate heat or electricity. However, use of granulated biofuel for the gasification results in fuel agglomeration that stops the entire process. More often agglomeration problem occurs in hoppers [2] when the particulate solid fuel forms a ⇑ Corresponding author. Tel.: +370 37 401875; fax: +370 37 351271. E-mail address: [email protected] (R. Paulauskas). http://dx.doi.org/10.1016/j.fuel.2014.11.023 0016-2361/Ó 2014 Elsevier Ltd. All rights reserved.
stable structure interfering the movement of the fuel. The blockage of fuel movement is caused by different size and shape of particles, moisture content and incorrect designed transport system. The problem is avoided by improving the fuel transport system in several ways [3]: use of nitrogen flow to flush the fuel or brake the fuel blockage, or the fuel vibration powered by an eccentric motor. Many publications have also reported that the agglomeration phenomenon occurs in the oxidation zone due to the use of fuel of high ash and moisture content [4,5]. When the temperature is too high, the fused ashes form slag, which grows and thus stops the gasification process. In order to avoid slag formation, the temperature of oxidation zone is decreased by a large quantity of steam or an additional system of ash removal is installed to the gasifier [6–8].
146
R. Paulauskas et al. / Fuel 142 (2015) 145–151
However, the wood pellet gasification tests showed that wood pellets stick together moving from the pyrolysis zone to the oxidation zone due to the initial expansion of pellets and stop the further gasification. This fuel agglomeration problem is commonly encountered in the gasification of pellets made from recycled waste [9]. However, much more studies of wood or straw pellets size changes due to high temperature are published. Several researchers investigated thermal deformations of wood particles during pyrolysis at 700 °C temperature [10]. The cylindrical particles with a diameter of 25 mm, length of 300 mm, were prepared from 150 mm diameter birch trunk. The prepared sample was inserted in an electrically heated vertical furnace with an inner diameter of 150 mm and a height of 1100 mm, where was created an inert environment. The temperature was measured by five thermocouples installed at the sample surface and by one seated in the center of the sample. The results showed that wood particle heated up to 700 °C during 800 s and shrank till 20 mm in the radial direction. Longitudinal shrinkage was not observed. Davidsson and Pettersson [11] performed a more detailed study of the shrinkage of cubic wood particle during pyrolysis from 350 to 900 °C temperature. The wood particle (5 mm) was inserted into the stationary temperature vertical pyrolysis reactor on alumina plates, which was connected to the scales. Particle size changes were captured through the window by the digital camera installed in the reactor. According to the obtained data, the authors [11] found that the wood particle lost from 45% to 70% of its initial volume, and established the shrinkage of the particle in three directions: longitudinal shrinkage varies from 5% to 25%, tangential – from 25% to 40% and radial – from 15% to 40%. The most intensive shrinkage of particle in the radial direction was measured in 500–700 °C temperature range. Authors Kumar et al. [12] researched changes of wood particle size during combustion process at 650 °C, 750 °C and 850 °C temperature. The wood particles of shapes like cylinder (l/d 1), disk (l/d = 0.2–0.67) and rod (l/d = 2–10) were used for investigation. The particle was weighed and its size was measured before the experimental test. The particle dimensions varied from 5 to 100 mm. The prepared sample was inserted into a special basket made from 1 mm wire in the combustion reactor where the required temperature and air content were maintained. The view of particle ignition moment and combustion process was monitored from a mirror system. After extinction of the volatiles flame, the basket with the particle was removed from the reactor and the residual size of the particle was measured. The experiment results showed that the particle shrinkage in longitudinal direction depended on the particle length: the longitudinal shrinkage varied from 17% to 11% due to the variation of length from 8 mm to 20 mm. The radial shrinkage varied from 14% to 28.6% with the increase of the particle’s length and diameter ratio. Wood particle lost from 38% to 58% of its initial volume irrespective of its length and diameter ratio during the combustion process. Scientist group from Sweden [13] presented the recovered solid waste pellets with 8 mm diameter swelling and shrinking investigation during pyrolysis in 500–900 °C temperature range. The sample was placed on a holder in the horizontal pyrolysis reactor, where the constant temperature and nitrogen flow were maintained. The holder was connected to the digital scales for pellet mass change recording and the pellet surface and center temperature was measured by K-type thermocouples during the pyrolysis process. The change in the size of the pellet was captured by the CCD camera and the collected data analyzed using the ImageJ software. Obtained results showed that the recovered solid waste pellets expanded to 54% of the initial volume per 70 s at 550 °C heating temperature and after that the particle shrank to the initial volume value per 80 s. The swelling of the pellet reached 58% of the initial volume per 40 s at higher heating temperature (660 °C). For comparison purposes, the authors repeated the experimental investigation
with the straw pellets. The tendency of the straw pellet swelling and shrinking was different: the pellet expanded to 18% of the initial volume at 660 °C heating temperature and after that began to shrink till 44% of the initial volume [13]. The reviewed works mostly analyze the changes in size of wood particles [10–12], straw and recycled solid waste pellets [13] at high temperatures. In order to determine the origin of the adhesion of wood pellets due the expansion and find possible solutions to decrease the risk of the bridging, it is necessary to investigate size changes of wood pellets at high temperatures. According to this, experimental investigations were performed in the electrically heated horizontal furnace from 300 °C to 1000 °C temperature at an inert ambient. The wood pellet with 6 mm diameter was captured by the digital camera during the pyrolysis process and the obtained results are presented in this paper. 2. Experimental methodology 2.1. Sample characterization The tested material for research was pellets made from softwood sawdust. The diameter of samples varied from 6.1 to 6.4 mm, length from 15 to 17 mm. The material analyzes were performed using an IKA C5000 calorimeter, a Flash 2000 CHNS analyzer and a NETZSCH STA 449 F3 Jupiter thermogravimeter (TG, DTG,) coupled Fourier-transform infrared (FTIR) spectrometer Bruker Tensor 27 with external TGA-IR module (Gran Schmidt) in accordance with: LST EN 14774-1 (moisture content), LST EN 14918 (HHV), LST EN 14775 (ash content), LST EN 15148 (volatile content) and LST EN 15104 (CHNS content) and obtained characteristics are presented in Table 1. 2.2. Research methodology Investigation of pelletized wood fuel swelling and shrinking was carried out using the electrically heated horizontal tubular furnace Nabertherm RS 80/500/13 with a temperature control unit (measurement accuracy ±3 °C). A scheme of this experimental rig is shown in Fig. 1. A working sillimantin tube (2) with outer diameter of 80 mm and length of 850 mm was mounted inside the furnace (1) with heating zone length of 500 mm; the tube was heated on both sides. One end of the working tube was supplied by a nitrogen flow, whose temperature with 2.2 °C accuracy was measured by an installed K-type thermocouple. The other end of the tube was left open for placement of a special pad with wood pellet and for capturing the sample size changes. The nitrogen flow of 7 l/min
Table 1 Characteristics of wood pellets in experimental investigation. Parameter
Softwood wood pellets
Proximate analysis wt.% Moisture Volatiles Fixed carbon Ash HHV (MJ/kg)
5.2 79.2 15.2 0.4 19.0
Ultimate analysis wt.% (dry basis) Carbon Hydrogen Oxygen (diff.) Nitrogen Sulphur
49.20 6.20 44.06 0.08 0.06
Ash softening Temperature (°C)
1315
R. Paulauskas et al. / Fuel 142 (2015) 145–151
147
Fig. 1. Scheme of the experimental rig: 1 – electrical furnace RS 80/500/13; 2 – working sillimantin tube; 3 – nitrogen flow meter; 4 – sample holder with thermocouple; 5 – wood pellet; 6 – K-type thermocouple; 7 – computer; 8 – digital camera Canon SX30 IS.
controlled by the flow meter (3) was fed into the furnace when the temperature reached the desired level. The special pad (4) with the sample (5) was inserted in the middle of the working tube through the open end. One additional thermocouple (6) was installed in the pad and used to measure the center temperature of the wood pellet during the pyrolysis process. The measured center temperature values were collected by data logger PICO and sent to the computer (7). The wood pellet was captured with the digital camera Canon S 30 IS (8) through the open end of the working tube throughout the pyrolysis process until the pellet stopped shrinking. The digital camera has integrated wide-angle (24–840 mm) lens, and the distance between the camera and the sample was 700 mm. The initial diameter of the sample was measured with a varnier caliper, which provided precision to 0.05 mm before the sample load. Once the pyrolysis process had ended, capturing of the wood pellet was stopped and the pad with the wood pellet was removed. The measurements were performed at constant heating temperature from 300 °C to 1000 °C by 50 °C step. The high-resolution (1280 720 pixels) recorded videos of wood pellet changes were analyzed using GIMP software. Each 145th recorded video frame was converted into a photo and the wood pellet diameter was measured with a digital ruler in pixels with 0.5 pixel accuracy. The wood pellet diameter of 6 mm matched to 58 ± 0.5 pixels at the initial time of the pyrolysis process. The measured sample diameter in pixels was expressed as the relative units by the following equation:
D ¼ Dt =D0
ð1Þ
where Dt is measured wood pellet diameter during the time t, D0 – initial diameter of wood pellet. In order to reduce the occurrence of errors due to accuracy of measuring devices and variability of the pellets, the tests of wood pellet changes were repeated 5 times at each heating temperature point. 3. Results 3.1. Wood pellet center temperature profiles The center temperature of the wood pellet was measured during the pyrolysis process at different heating temperatures. The pellet center temperature versus time is plotted in Fig. 2. It is shown in Fig. 2 that the center temperature profile changes depending on the heating temperature and as the temperature increases the profile changes become more intensive. The temperature profile variations proceed due to chemical processes inside the wood pellet. According to the curves of TG, Gram Schmidt and DTG (see Fig. 3) and literature review [14], these chemical processes
Fig. 2. Wood pellet center temperature profiles at different heating temperatures.
can be divided into several main phases: the first is the removal of water from the pellets (from 20 to 130 °C temperature), the second – the lignin, hemicellulose and cellulose decomposition (from 160 to 400 °C), and the third – the residual lignin decomposition and char formation (from 420 to 600 °C). In order to evaluate the regularities of these phases more closely, heating rates of wood pellets were calculated by the first order derivative (DTc/Dt) at 300 °C, 400 °C, 550 °C and 900 °C heating temperature from the obtained data (see Fig. 4). Evaporation of water in 300–400 °C ambient temperature range starts at 30 °C temperature and lasts for about 30 s until the center temperature of the pellet reaches 120 °C value. At higher heating temperature, heating rate increases due to more intensive heat transfer and final dehydration process shifts towards to higher temperatures, respectively to 140 °C center temperature at 550 °C and to 175 °C center temperature at 900 °C. The heat absorbed during the dehydration process is used to preheat the water inside the particle from 30 °C to 100 °C and evaporate it. Since the specific heat of water (cp = 420 kJ/kg) is lower than the specific heat of evaporation of water (cp = 2250 kJ/kg), evaporation of water from the pellet requires more heat than to rise water temperature up to 100 °C inside the pellet. Due to this cause, the heating rate decreases approximately 3.5 times in the particle center (see Fig. 4). After the evaporation of water, hemicellulose, cellulose and lignin decomposition begins, which leads to the release of volatiles from the wood pellet. In the literature [15–17], hemicellulose decomposition occurs at 150–300 °C temperature with activation energy of 132.9 kJ/mol, cellulose decomposes at 200–400 °C with activation energy of 175.6 kJ/mol and lignin decomposes at 150–600 °C with activation energy of 101 kJ/mol. The heating rate of the wood pellet center in Fig. 4(A) indicates that only negligible decomposition of wood composition occurs at 300 °C heating temperature and it does not cause any significant temperature changes in the pellet center. At higher heating temperature (400 °C), partial hemicellulose, lignin (at 160 °C) and cellulose (at 200 °C) decomposition takes place. Increasing heating temperature causes more intensive wood decomposition which is defined by variations of the center temperature in the wood pellet due to various thermal effects (see Fig. 2). Lignin, hemicellulose and cellulose decomposes when the sample center temperature grows from 140 to 371 °C at 550 °C heating temperature. At this point heating rate increases from 1.6 °C/s to 3.2 °C/s and linearly decreases to 0.45 °C/s (see Fig. 4(C)). According to the temperature range of lignin, hemicellulose and cellulose decomposition [15,16] and activation energies [17], it can be stated that the hemicellulose and part of cellulose are decomposed over 125 s at 140–371 °C center temperature range. After the aforementioned decomposition
148
R. Paulauskas et al. / Fuel 142 (2015) 145–151
Fig. 3. Pyrolysis curves of wood pellet in TGA.
Fig. 4. Profiles of wood pellet center temperature and heating rate.
ends, the residual lignin starts to decompose and an unbounded energy is only used to heat residual particle (heating rate increases from 0.45 °C/s to 1.2 °C/s). During lignin decomposition char formation occurs and the heating rate decreases till 0.07 °C/s (see Fig. 4(C)). This significant decrease of the heating rate appears due to the
increase of the thermal conductivity caused by the formatted char. At higher temperature (900 °C), the trend of chemical processes in the wood pellet remains the same as at 550 °C temperature (see Fig. 4(C) and (D)), however, the decomposition is 1.65 time faster due to the received higher amount of heat. This wood decomposition
R. Paulauskas et al. / Fuel 142 (2015) 145–151
proceeding and duration depend on the heat transfer, which is determined by the heating temperature (see Fig. 4). 3.2. Wood pellet swelling and shrinking The wood pellet changes in size calculated by (1) equation and center temperature curves at 300 °C, 400 °C, 550 °C and 900 °C pyrolysis temperature are presented in Fig. 5. At low heating temperature (300 °C), the wood pellet slowly overheats due to the small amount of heat from ambient causing barely noticeable wood pellet shrinkage in Fig. 5(A). The shrinkage of the pellet occurs after 45 s, and the wood pellet loses 2% of the initial diameter. At the end of the pyrolysis process, the pellet shrinks to 96% of the initial diameter. Further increase of the ambient temperature to 400 °C, the wood pellet diameter changes became more intensive. When the center of the pellet heats up to 61 °C, the swelling phenomenon occurs causing wood pellet expansion by 3.8% of the original diameter and lasts 60 s till the pellet center temperature reaches 163 °C value (see Fig. 5(B)). During the swelling process, the heating rate of the pellet center decreases from 4.2 °C/s to 0.7 °C/s due to water evaporation (see Figs. 3 and 4(B)). The expanded wood pellet starts to shrink after 5 s and the pellet diameter loses 7% of the initial diameter at the end of the pyrolysis process. During the contraction process, the center temperature increases from 160 °C to 310 °C which indicates a partial decomposition of wood (see Fig. 3). The most intensive swelling
149
effect is observed at 550 °C heating temperature after 20 s from the beginning of the pyrolysis process, lasts 55 s and then stabilizes (see Fig. 5(C)). Wood pellet expands in the radial direction approximately 12% of the initial diameter. During the pellet expansion, the center temperature grows from 120 °C to 250 °C. After the expansion, the wood pellet shrinks for 170 s and loses 14% of the initial diameter. When the wood pellet center reaches 420 °C temperature, no significant changes in the diameter is observed. At higher temperature (900 °C) only the shrinkage of wood pellet is detected. The pellet shrinks in radial direction by 21% of the initial diameter at 608 °C center temperature. Although the pellet center temperature increases until it reaches the heating temperature, the pellet diameter changes are not detected and in accordance with TGA analyses presented in Fig. 3, only fixed carbon is left. In order to estimate the regularities of wood pellet diameter changes due to the heating temperature, the maximum swelling and shrinking values of wood pellet at different heating temperatures are plotted in Fig. 6. It is shown in Fig. 6 that the increase of heating temperature results in more intensive swelling process up to 550 °C temperature. When this temperature is exceeded, the swelling process begins to decline till the disappearance at 900 °C heating temperature. Wood pellets shrinkage increases with temperature in the range 300–1000 °C and the wood pellet shrinks approximately 2% of the initial diameter at every 50 °C step of the heating temperature. In the literature [11] it is reported that the radial shrinkage
Fig. 5. Wood pellet diameter changes and center temperature profile at different heating temperature.
150
R. Paulauskas et al. / Fuel 142 (2015) 145–151
Fig. 6. Wood pellet swelling and shrinking at different temperatures.
maximum of a wood particle is between 500 and 700 °C temperature and the particle shrinks from 15% to 40% of the initial diameter. Comparing the wood pellet and the wood particle changes in the radial direction, it is observed that attribution of wood pellet shrinking is different. The wood pellet loses only 13–18% of the initial diameter in the heating temperature range 500–700 °C and the shrinkage maximum of 23% of the initial diameter is reached at 1000 °C heating temperature. From the obtained results, the wood pellet diameter changes depend on the pyrolysis temperature and heat transfer in the particle. The increase of heating temperature causes more intensive
wood decomposition which influences faster volatile release from the pellets (see Figs. 4 and 7). It is considered that the expansion of the wood pellets is due to a small amount of heat which indicates slow overheat of the wood pellet and causes irregular volatile release from the pellet. At low temperatures (up to 400 °C), the particle heats evenly (heating rate 1.2 °C/s) (see Fig. 4(A)) causing water evaporation and negligible hemicellulose, cellulose and lignin decomposition (see Figs. 3 and 7). During this process the pressure of water vapour and volatile compounds do not reach a critical value inside the particles and the pellet’s surface is not destructed by water and volatile diffusion from it. At higher 400 °C heating temperature, the heating rate increases only by 0.3 °C/s, but release process of the volatiles is more intensive due to a higher amount of heat (see Fig. 4(B)). Formed water vapour and volatile compounds fall to evaporate from the pellet center due to the chemical processes ongoing on the surface which influence the pellet porosity decrease. In this way, water and volatile substances accumulate inside the pellet and internal pressure is growing near the pellet’s surface, which destroys the surface structure of the pellet by expanding it to 4% of the initial diameter and freeing accumulated water vapour and volatiles (see Fig. 7). Shrinkage of wood pellet begins after volatile evaporation. The swelling of pellets starts to decrease when the temperature is higher than 550 °C (see Fig. 2), possibly due to faster particle overheat and more intensive wood decomposition (see Figs. 4(C) and 7). The swelling phenomenon is no longer detected when the heating temperature is over 900 °C (see Fig. 7). The wood particle overheats so quickly that the compounds emitted from the surface layer are decomposed by high temperatures and the evaporation of volatiles becomes possible from the deeper layers (see Fig. 4(D)). According to the obtained results it can be stated that the fuel agglomeration in
Fig. 7. Photo of wood pellet diameter changes at different temperature.
R. Paulauskas et al. / Fuel 142 (2015) 145–151
the gasifier using wood pellets occurs due to the pellet expansion in pyrolysis zone between 400 and 900 °C temperatures. In order to avoid the bridging problem it is necessary to maintain 900 °C temperature at the end of the pyrolysis zone of gasifier. 4. Conclusions Experimental investigation of wood pellet shrinkage and swelling during pyrolysis process can be summarized in few statements. The observed changes of the wood pellet diameter and the center temperature variation reveal that when increasing the heating temperature from 300 °C to 1000 °C, the wood pellet loses from 5% to 23% of the initial diameter respectively. At low temperatures (up to 400 °C), the center of the wood pellet heats evenly up to 230 °C temperature and only water evaporation and negligible wood decomposition occur; the latter process influences the pellet shrinking only by 5% of the initial diameter. At higher 400 °C heating temperature, the wood decomposition is more intensive due to the higher amount of heat what causes the wood pellet expansion approximately by 3.8% of the initial diameter. The swelling phenomenon intensifies with the heating temperature increase up to 550 °C causing maximum expansion of the wood pellet by 12% of the initial diameter at 250 °C center temperature. When the heating temperature is higher than 550 °C, the wood pellet expansion starts decreasing and disappears when the temperature is higher than 900 °C due to the intensive wood decomposition which causes full release of volatiles and char formation. The heating rate increases about 2 times from 4 °C/s to 8 °C/s in the heating temperature range 550–900 °C. From the results obtained from the heating rate in the wood pellet center, it can be concluded that the wood pellet diameter changes are determined by the heating temperature, and not by the heating rate in the pellet center. The estimated peaks of the heating rates indicate the chemical processes and their intensity inside the wood pellet. Acknowledgment This research was funded by grant (No. ATE-02/2012) from the Research Council of Lithuania.
151
References [1] Mayerhofer M, Fendt S, Spliethoff H, Gaderer M. Fluidized bed gasification of biomass – in bed investigation of gas and tar formation. Fuel 2014;117:1248–55. [2] Hamel S, Hasselbach H, Weil S, Krumm W. Autothermal two stage gasification of low density waste derived fuels. Energy 2007;32:95–107. [3] Paterson N, Zhou Y, Dugwell DR, Kandiyoti R. Investigation of ammonia formation during gasification in an air blown spouted bed: reactor design and initial tests. Energy Fuels 2002;16:127–35. [4] Brown MD, Baker EG, Mudge LK. Environmental design considerations for thermochemical biomass energy. Biomass 1986;11:255–70. [5] Gezer I, Dogru M, Akay G. Gasification of appicot pit shells in a downdraft gasifier. Int J Green Energy 2009;6:218–27. [6] Sheth a, Pratik N, Babu BV. Experimental studies on producer gas generation from wood waste in a downdraft biomass gasifier. Bioresour Technol 2009;100:3127–33. [7] Corman JC. Coal to electricity: integrated gasification combined cycle. Appl Energy 1982;10:243–59. [8] Chopra S, Jain A. A review of fixed bed gasification systems for biomass. Agr Eng Int CIGR EJ 2007:9. [9] Dogru M, Midilli A, Howarth CR. Gasification of sewage sludge using a throated downdraft gasifier and uncertainty analysis. Fuel Process Technol 2002;75:55–82. [10] Larfeldt J, Leckner B, Malaaen MC. Modelling and measurements of the pyrolysis of large wood particles. Fuel 2000;79:1637–43. [11] Davidsson KO, Pettersson JBC. Birch wood particle shrinkage during rapid pyrolysis. Fuel 2002;81:263–70. [12] Kumar RR, Kolar AK, Leckner B. Shrinkage characteristics of Casuarina wood during devolatilization in a fluidized bed combustor. Biomass Bioenergy 2006;30:153–65. [13] Zhou C, Zhang Q, Arnold L, Yang W, Blasiak W. A study of the pyrolysis behaviors of pelletized recovered municipal solid waste fuels. Appl Energy 2013;107:173–82. [14] De Wild PJ, Reith H, Heeres E. Biomass pyrolysis for chemicals. Biofuels 2011;2:185–208. [15] Koukios EG. Progress in thermochemical, solid-state refining of biomass – from research to commercialization. Adv Thermochem Biomass Convers 1993;2:1678–92. [16] Park WC, Atreya A, Baum HR. Experimental and theoretical investigation of heat and mass transfer processes during wood pyrolysis. Combust Flame 2010;157:481–94. [17] Gašparovicˇ L, Labovsky´ J, Markoš J, Jelemensky´ L. Calculation of kinetic parameters of the thermal decomposition of wood by distributed activation energy model (DAEM). Chem Biochem Eng Q 2012;26:45–53.
Contents lists available at ScienceDirect
Fuel journal homepage: www.elsevier.com/locate/fuel
Experimental investigation of wood pellet swelling and shrinking during pyrolysis gas Rolandas Paulauskas ⇑, Algis Dzˇiugys, Nerijus Striu Laboratory of Combustion Processes, Lithuanian Energy Institute, Breslaujos Str. 3, LT-44403 Kaunas, Lithuania
h i g h l i g h t s The wood pellet changes in the radial direction during pyrolysis was captured with the digital camera. The swelling phenomenon of wood pellet was observed during pyrolysis. Heating temperature of pyrolysis influenced the radial swelling and shrinking of wood pellet. The estimated peaks of the heating rates indicate the chemical processes and their intensity inside the wood pellet. A discussion was given for the mechanism of the wood pellet swelling phenomenon during pyrolysis.
a r t i c l e
i n f o
Article history: Received 9 July 2014 Received in revised form 30 September 2014 Accepted 5 November 2014 Available online 17 November 2014 Keywords: Wood pellets Pyrolysis The center temperature Swelling Shrinking
a b s t r a c t Gasification of wood pellets in downdraft gasifier is confronted with the problem of granulated fuel agglomeration. It occurs when the wood pellets are moving from the pyrolysis zone to the oxidation zone and pellets stick together in lumps, which disrupts the movement of fuel and stops further process. In order to investigate the cause and regularities of fuel agglomeration, experimental studies of wood pellets thermal deformations during pyrolysis were carried out. Experimental studies were performed in an electrically heated horizontal tubular furnace from 300 °C to 1000 °C temperature in an inert atmosphere capturing wood pellet thermal deformations by a digital camera. During investigation the center temperature of the pellet was also measured. Analyzed results showed that wood pellet final diameter decreased when the heating temperature increased. It was also established that the pellet expanded from 400 °C to 900 °C heating temperature at the beginning of pyrolysis process. The maximum swelling effect was reached at 550 °C temperature and after it swelling intensity was decreasing till 900 °C temperature. Over 900 °C heating temperature, the expansion phenomenon was no longer visible. The tendency of wood pellet size alteration (swelling and shrinking) depends on heating temperature. The obtained results explain the reason for adhesion of wood pellets in the gasifier and reveal regularity of wood pellet size changes with increasing heating temperature. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Renewable fuel resources are widely used to produce electricity and heat. The growing demand for these stocks and their price results in searching for ways how to use low-quality biofuels. One of the ways is gasification [1]. During this process, the solid fuel is gasified and turned into higher quality gas, which is used to generate heat or electricity. However, use of granulated biofuel for the gasification results in fuel agglomeration that stops the entire process. More often agglomeration problem occurs in hoppers [2] when the particulate solid fuel forms a ⇑ Corresponding author. Tel.: +370 37 401875; fax: +370 37 351271. E-mail address: [email protected] (R. Paulauskas). http://dx.doi.org/10.1016/j.fuel.2014.11.023 0016-2361/Ó 2014 Elsevier Ltd. All rights reserved.
stable structure interfering the movement of the fuel. The blockage of fuel movement is caused by different size and shape of particles, moisture content and incorrect designed transport system. The problem is avoided by improving the fuel transport system in several ways [3]: use of nitrogen flow to flush the fuel or brake the fuel blockage, or the fuel vibration powered by an eccentric motor. Many publications have also reported that the agglomeration phenomenon occurs in the oxidation zone due to the use of fuel of high ash and moisture content [4,5]. When the temperature is too high, the fused ashes form slag, which grows and thus stops the gasification process. In order to avoid slag formation, the temperature of oxidation zone is decreased by a large quantity of steam or an additional system of ash removal is installed to the gasifier [6–8].
146
R. Paulauskas et al. / Fuel 142 (2015) 145–151
However, the wood pellet gasification tests showed that wood pellets stick together moving from the pyrolysis zone to the oxidation zone due to the initial expansion of pellets and stop the further gasification. This fuel agglomeration problem is commonly encountered in the gasification of pellets made from recycled waste [9]. However, much more studies of wood or straw pellets size changes due to high temperature are published. Several researchers investigated thermal deformations of wood particles during pyrolysis at 700 °C temperature [10]. The cylindrical particles with a diameter of 25 mm, length of 300 mm, were prepared from 150 mm diameter birch trunk. The prepared sample was inserted in an electrically heated vertical furnace with an inner diameter of 150 mm and a height of 1100 mm, where was created an inert environment. The temperature was measured by five thermocouples installed at the sample surface and by one seated in the center of the sample. The results showed that wood particle heated up to 700 °C during 800 s and shrank till 20 mm in the radial direction. Longitudinal shrinkage was not observed. Davidsson and Pettersson [11] performed a more detailed study of the shrinkage of cubic wood particle during pyrolysis from 350 to 900 °C temperature. The wood particle (5 mm) was inserted into the stationary temperature vertical pyrolysis reactor on alumina plates, which was connected to the scales. Particle size changes were captured through the window by the digital camera installed in the reactor. According to the obtained data, the authors [11] found that the wood particle lost from 45% to 70% of its initial volume, and established the shrinkage of the particle in three directions: longitudinal shrinkage varies from 5% to 25%, tangential – from 25% to 40% and radial – from 15% to 40%. The most intensive shrinkage of particle in the radial direction was measured in 500–700 °C temperature range. Authors Kumar et al. [12] researched changes of wood particle size during combustion process at 650 °C, 750 °C and 850 °C temperature. The wood particles of shapes like cylinder (l/d 1), disk (l/d = 0.2–0.67) and rod (l/d = 2–10) were used for investigation. The particle was weighed and its size was measured before the experimental test. The particle dimensions varied from 5 to 100 mm. The prepared sample was inserted into a special basket made from 1 mm wire in the combustion reactor where the required temperature and air content were maintained. The view of particle ignition moment and combustion process was monitored from a mirror system. After extinction of the volatiles flame, the basket with the particle was removed from the reactor and the residual size of the particle was measured. The experiment results showed that the particle shrinkage in longitudinal direction depended on the particle length: the longitudinal shrinkage varied from 17% to 11% due to the variation of length from 8 mm to 20 mm. The radial shrinkage varied from 14% to 28.6% with the increase of the particle’s length and diameter ratio. Wood particle lost from 38% to 58% of its initial volume irrespective of its length and diameter ratio during the combustion process. Scientist group from Sweden [13] presented the recovered solid waste pellets with 8 mm diameter swelling and shrinking investigation during pyrolysis in 500–900 °C temperature range. The sample was placed on a holder in the horizontal pyrolysis reactor, where the constant temperature and nitrogen flow were maintained. The holder was connected to the digital scales for pellet mass change recording and the pellet surface and center temperature was measured by K-type thermocouples during the pyrolysis process. The change in the size of the pellet was captured by the CCD camera and the collected data analyzed using the ImageJ software. Obtained results showed that the recovered solid waste pellets expanded to 54% of the initial volume per 70 s at 550 °C heating temperature and after that the particle shrank to the initial volume value per 80 s. The swelling of the pellet reached 58% of the initial volume per 40 s at higher heating temperature (660 °C). For comparison purposes, the authors repeated the experimental investigation
with the straw pellets. The tendency of the straw pellet swelling and shrinking was different: the pellet expanded to 18% of the initial volume at 660 °C heating temperature and after that began to shrink till 44% of the initial volume [13]. The reviewed works mostly analyze the changes in size of wood particles [10–12], straw and recycled solid waste pellets [13] at high temperatures. In order to determine the origin of the adhesion of wood pellets due the expansion and find possible solutions to decrease the risk of the bridging, it is necessary to investigate size changes of wood pellets at high temperatures. According to this, experimental investigations were performed in the electrically heated horizontal furnace from 300 °C to 1000 °C temperature at an inert ambient. The wood pellet with 6 mm diameter was captured by the digital camera during the pyrolysis process and the obtained results are presented in this paper. 2. Experimental methodology 2.1. Sample characterization The tested material for research was pellets made from softwood sawdust. The diameter of samples varied from 6.1 to 6.4 mm, length from 15 to 17 mm. The material analyzes were performed using an IKA C5000 calorimeter, a Flash 2000 CHNS analyzer and a NETZSCH STA 449 F3 Jupiter thermogravimeter (TG, DTG,) coupled Fourier-transform infrared (FTIR) spectrometer Bruker Tensor 27 with external TGA-IR module (Gran Schmidt) in accordance with: LST EN 14774-1 (moisture content), LST EN 14918 (HHV), LST EN 14775 (ash content), LST EN 15148 (volatile content) and LST EN 15104 (CHNS content) and obtained characteristics are presented in Table 1. 2.2. Research methodology Investigation of pelletized wood fuel swelling and shrinking was carried out using the electrically heated horizontal tubular furnace Nabertherm RS 80/500/13 with a temperature control unit (measurement accuracy ±3 °C). A scheme of this experimental rig is shown in Fig. 1. A working sillimantin tube (2) with outer diameter of 80 mm and length of 850 mm was mounted inside the furnace (1) with heating zone length of 500 mm; the tube was heated on both sides. One end of the working tube was supplied by a nitrogen flow, whose temperature with 2.2 °C accuracy was measured by an installed K-type thermocouple. The other end of the tube was left open for placement of a special pad with wood pellet and for capturing the sample size changes. The nitrogen flow of 7 l/min
Table 1 Characteristics of wood pellets in experimental investigation. Parameter
Softwood wood pellets
Proximate analysis wt.% Moisture Volatiles Fixed carbon Ash HHV (MJ/kg)
5.2 79.2 15.2 0.4 19.0
Ultimate analysis wt.% (dry basis) Carbon Hydrogen Oxygen (diff.) Nitrogen Sulphur
49.20 6.20 44.06 0.08 0.06
Ash softening Temperature (°C)
1315
R. Paulauskas et al. / Fuel 142 (2015) 145–151
147
Fig. 1. Scheme of the experimental rig: 1 – electrical furnace RS 80/500/13; 2 – working sillimantin tube; 3 – nitrogen flow meter; 4 – sample holder with thermocouple; 5 – wood pellet; 6 – K-type thermocouple; 7 – computer; 8 – digital camera Canon SX30 IS.
controlled by the flow meter (3) was fed into the furnace when the temperature reached the desired level. The special pad (4) with the sample (5) was inserted in the middle of the working tube through the open end. One additional thermocouple (6) was installed in the pad and used to measure the center temperature of the wood pellet during the pyrolysis process. The measured center temperature values were collected by data logger PICO and sent to the computer (7). The wood pellet was captured with the digital camera Canon S 30 IS (8) through the open end of the working tube throughout the pyrolysis process until the pellet stopped shrinking. The digital camera has integrated wide-angle (24–840 mm) lens, and the distance between the camera and the sample was 700 mm. The initial diameter of the sample was measured with a varnier caliper, which provided precision to 0.05 mm before the sample load. Once the pyrolysis process had ended, capturing of the wood pellet was stopped and the pad with the wood pellet was removed. The measurements were performed at constant heating temperature from 300 °C to 1000 °C by 50 °C step. The high-resolution (1280 720 pixels) recorded videos of wood pellet changes were analyzed using GIMP software. Each 145th recorded video frame was converted into a photo and the wood pellet diameter was measured with a digital ruler in pixels with 0.5 pixel accuracy. The wood pellet diameter of 6 mm matched to 58 ± 0.5 pixels at the initial time of the pyrolysis process. The measured sample diameter in pixels was expressed as the relative units by the following equation:
D ¼ Dt =D0
ð1Þ
where Dt is measured wood pellet diameter during the time t, D0 – initial diameter of wood pellet. In order to reduce the occurrence of errors due to accuracy of measuring devices and variability of the pellets, the tests of wood pellet changes were repeated 5 times at each heating temperature point. 3. Results 3.1. Wood pellet center temperature profiles The center temperature of the wood pellet was measured during the pyrolysis process at different heating temperatures. The pellet center temperature versus time is plotted in Fig. 2. It is shown in Fig. 2 that the center temperature profile changes depending on the heating temperature and as the temperature increases the profile changes become more intensive. The temperature profile variations proceed due to chemical processes inside the wood pellet. According to the curves of TG, Gram Schmidt and DTG (see Fig. 3) and literature review [14], these chemical processes
Fig. 2. Wood pellet center temperature profiles at different heating temperatures.
can be divided into several main phases: the first is the removal of water from the pellets (from 20 to 130 °C temperature), the second – the lignin, hemicellulose and cellulose decomposition (from 160 to 400 °C), and the third – the residual lignin decomposition and char formation (from 420 to 600 °C). In order to evaluate the regularities of these phases more closely, heating rates of wood pellets were calculated by the first order derivative (DTc/Dt) at 300 °C, 400 °C, 550 °C and 900 °C heating temperature from the obtained data (see Fig. 4). Evaporation of water in 300–400 °C ambient temperature range starts at 30 °C temperature and lasts for about 30 s until the center temperature of the pellet reaches 120 °C value. At higher heating temperature, heating rate increases due to more intensive heat transfer and final dehydration process shifts towards to higher temperatures, respectively to 140 °C center temperature at 550 °C and to 175 °C center temperature at 900 °C. The heat absorbed during the dehydration process is used to preheat the water inside the particle from 30 °C to 100 °C and evaporate it. Since the specific heat of water (cp = 420 kJ/kg) is lower than the specific heat of evaporation of water (cp = 2250 kJ/kg), evaporation of water from the pellet requires more heat than to rise water temperature up to 100 °C inside the pellet. Due to this cause, the heating rate decreases approximately 3.5 times in the particle center (see Fig. 4). After the evaporation of water, hemicellulose, cellulose and lignin decomposition begins, which leads to the release of volatiles from the wood pellet. In the literature [15–17], hemicellulose decomposition occurs at 150–300 °C temperature with activation energy of 132.9 kJ/mol, cellulose decomposes at 200–400 °C with activation energy of 175.6 kJ/mol and lignin decomposes at 150–600 °C with activation energy of 101 kJ/mol. The heating rate of the wood pellet center in Fig. 4(A) indicates that only negligible decomposition of wood composition occurs at 300 °C heating temperature and it does not cause any significant temperature changes in the pellet center. At higher heating temperature (400 °C), partial hemicellulose, lignin (at 160 °C) and cellulose (at 200 °C) decomposition takes place. Increasing heating temperature causes more intensive wood decomposition which is defined by variations of the center temperature in the wood pellet due to various thermal effects (see Fig. 2). Lignin, hemicellulose and cellulose decomposes when the sample center temperature grows from 140 to 371 °C at 550 °C heating temperature. At this point heating rate increases from 1.6 °C/s to 3.2 °C/s and linearly decreases to 0.45 °C/s (see Fig. 4(C)). According to the temperature range of lignin, hemicellulose and cellulose decomposition [15,16] and activation energies [17], it can be stated that the hemicellulose and part of cellulose are decomposed over 125 s at 140–371 °C center temperature range. After the aforementioned decomposition
148
R. Paulauskas et al. / Fuel 142 (2015) 145–151
Fig. 3. Pyrolysis curves of wood pellet in TGA.
Fig. 4. Profiles of wood pellet center temperature and heating rate.
ends, the residual lignin starts to decompose and an unbounded energy is only used to heat residual particle (heating rate increases from 0.45 °C/s to 1.2 °C/s). During lignin decomposition char formation occurs and the heating rate decreases till 0.07 °C/s (see Fig. 4(C)). This significant decrease of the heating rate appears due to the
increase of the thermal conductivity caused by the formatted char. At higher temperature (900 °C), the trend of chemical processes in the wood pellet remains the same as at 550 °C temperature (see Fig. 4(C) and (D)), however, the decomposition is 1.65 time faster due to the received higher amount of heat. This wood decomposition
R. Paulauskas et al. / Fuel 142 (2015) 145–151
proceeding and duration depend on the heat transfer, which is determined by the heating temperature (see Fig. 4). 3.2. Wood pellet swelling and shrinking The wood pellet changes in size calculated by (1) equation and center temperature curves at 300 °C, 400 °C, 550 °C and 900 °C pyrolysis temperature are presented in Fig. 5. At low heating temperature (300 °C), the wood pellet slowly overheats due to the small amount of heat from ambient causing barely noticeable wood pellet shrinkage in Fig. 5(A). The shrinkage of the pellet occurs after 45 s, and the wood pellet loses 2% of the initial diameter. At the end of the pyrolysis process, the pellet shrinks to 96% of the initial diameter. Further increase of the ambient temperature to 400 °C, the wood pellet diameter changes became more intensive. When the center of the pellet heats up to 61 °C, the swelling phenomenon occurs causing wood pellet expansion by 3.8% of the original diameter and lasts 60 s till the pellet center temperature reaches 163 °C value (see Fig. 5(B)). During the swelling process, the heating rate of the pellet center decreases from 4.2 °C/s to 0.7 °C/s due to water evaporation (see Figs. 3 and 4(B)). The expanded wood pellet starts to shrink after 5 s and the pellet diameter loses 7% of the initial diameter at the end of the pyrolysis process. During the contraction process, the center temperature increases from 160 °C to 310 °C which indicates a partial decomposition of wood (see Fig. 3). The most intensive swelling
149
effect is observed at 550 °C heating temperature after 20 s from the beginning of the pyrolysis process, lasts 55 s and then stabilizes (see Fig. 5(C)). Wood pellet expands in the radial direction approximately 12% of the initial diameter. During the pellet expansion, the center temperature grows from 120 °C to 250 °C. After the expansion, the wood pellet shrinks for 170 s and loses 14% of the initial diameter. When the wood pellet center reaches 420 °C temperature, no significant changes in the diameter is observed. At higher temperature (900 °C) only the shrinkage of wood pellet is detected. The pellet shrinks in radial direction by 21% of the initial diameter at 608 °C center temperature. Although the pellet center temperature increases until it reaches the heating temperature, the pellet diameter changes are not detected and in accordance with TGA analyses presented in Fig. 3, only fixed carbon is left. In order to estimate the regularities of wood pellet diameter changes due to the heating temperature, the maximum swelling and shrinking values of wood pellet at different heating temperatures are plotted in Fig. 6. It is shown in Fig. 6 that the increase of heating temperature results in more intensive swelling process up to 550 °C temperature. When this temperature is exceeded, the swelling process begins to decline till the disappearance at 900 °C heating temperature. Wood pellets shrinkage increases with temperature in the range 300–1000 °C and the wood pellet shrinks approximately 2% of the initial diameter at every 50 °C step of the heating temperature. In the literature [11] it is reported that the radial shrinkage
Fig. 5. Wood pellet diameter changes and center temperature profile at different heating temperature.
150
R. Paulauskas et al. / Fuel 142 (2015) 145–151
Fig. 6. Wood pellet swelling and shrinking at different temperatures.
maximum of a wood particle is between 500 and 700 °C temperature and the particle shrinks from 15% to 40% of the initial diameter. Comparing the wood pellet and the wood particle changes in the radial direction, it is observed that attribution of wood pellet shrinking is different. The wood pellet loses only 13–18% of the initial diameter in the heating temperature range 500–700 °C and the shrinkage maximum of 23% of the initial diameter is reached at 1000 °C heating temperature. From the obtained results, the wood pellet diameter changes depend on the pyrolysis temperature and heat transfer in the particle. The increase of heating temperature causes more intensive
wood decomposition which influences faster volatile release from the pellets (see Figs. 4 and 7). It is considered that the expansion of the wood pellets is due to a small amount of heat which indicates slow overheat of the wood pellet and causes irregular volatile release from the pellet. At low temperatures (up to 400 °C), the particle heats evenly (heating rate 1.2 °C/s) (see Fig. 4(A)) causing water evaporation and negligible hemicellulose, cellulose and lignin decomposition (see Figs. 3 and 7). During this process the pressure of water vapour and volatile compounds do not reach a critical value inside the particles and the pellet’s surface is not destructed by water and volatile diffusion from it. At higher 400 °C heating temperature, the heating rate increases only by 0.3 °C/s, but release process of the volatiles is more intensive due to a higher amount of heat (see Fig. 4(B)). Formed water vapour and volatile compounds fall to evaporate from the pellet center due to the chemical processes ongoing on the surface which influence the pellet porosity decrease. In this way, water and volatile substances accumulate inside the pellet and internal pressure is growing near the pellet’s surface, which destroys the surface structure of the pellet by expanding it to 4% of the initial diameter and freeing accumulated water vapour and volatiles (see Fig. 7). Shrinkage of wood pellet begins after volatile evaporation. The swelling of pellets starts to decrease when the temperature is higher than 550 °C (see Fig. 2), possibly due to faster particle overheat and more intensive wood decomposition (see Figs. 4(C) and 7). The swelling phenomenon is no longer detected when the heating temperature is over 900 °C (see Fig. 7). The wood particle overheats so quickly that the compounds emitted from the surface layer are decomposed by high temperatures and the evaporation of volatiles becomes possible from the deeper layers (see Fig. 4(D)). According to the obtained results it can be stated that the fuel agglomeration in
Fig. 7. Photo of wood pellet diameter changes at different temperature.
R. Paulauskas et al. / Fuel 142 (2015) 145–151
the gasifier using wood pellets occurs due to the pellet expansion in pyrolysis zone between 400 and 900 °C temperatures. In order to avoid the bridging problem it is necessary to maintain 900 °C temperature at the end of the pyrolysis zone of gasifier. 4. Conclusions Experimental investigation of wood pellet shrinkage and swelling during pyrolysis process can be summarized in few statements. The observed changes of the wood pellet diameter and the center temperature variation reveal that when increasing the heating temperature from 300 °C to 1000 °C, the wood pellet loses from 5% to 23% of the initial diameter respectively. At low temperatures (up to 400 °C), the center of the wood pellet heats evenly up to 230 °C temperature and only water evaporation and negligible wood decomposition occur; the latter process influences the pellet shrinking only by 5% of the initial diameter. At higher 400 °C heating temperature, the wood decomposition is more intensive due to the higher amount of heat what causes the wood pellet expansion approximately by 3.8% of the initial diameter. The swelling phenomenon intensifies with the heating temperature increase up to 550 °C causing maximum expansion of the wood pellet by 12% of the initial diameter at 250 °C center temperature. When the heating temperature is higher than 550 °C, the wood pellet expansion starts decreasing and disappears when the temperature is higher than 900 °C due to the intensive wood decomposition which causes full release of volatiles and char formation. The heating rate increases about 2 times from 4 °C/s to 8 °C/s in the heating temperature range 550–900 °C. From the results obtained from the heating rate in the wood pellet center, it can be concluded that the wood pellet diameter changes are determined by the heating temperature, and not by the heating rate in the pellet center. The estimated peaks of the heating rates indicate the chemical processes and their intensity inside the wood pellet. Acknowledgment This research was funded by grant (No. ATE-02/2012) from the Research Council of Lithuania.
151
References [1] Mayerhofer M, Fendt S, Spliethoff H, Gaderer M. Fluidized bed gasification of biomass – in bed investigation of gas and tar formation. Fuel 2014;117:1248–55. [2] Hamel S, Hasselbach H, Weil S, Krumm W. Autothermal two stage gasification of low density waste derived fuels. Energy 2007;32:95–107. [3] Paterson N, Zhou Y, Dugwell DR, Kandiyoti R. Investigation of ammonia formation during gasification in an air blown spouted bed: reactor design and initial tests. Energy Fuels 2002;16:127–35. [4] Brown MD, Baker EG, Mudge LK. Environmental design considerations for thermochemical biomass energy. Biomass 1986;11:255–70. [5] Gezer I, Dogru M, Akay G. Gasification of appicot pit shells in a downdraft gasifier. Int J Green Energy 2009;6:218–27. [6] Sheth a, Pratik N, Babu BV. Experimental studies on producer gas generation from wood waste in a downdraft biomass gasifier. Bioresour Technol 2009;100:3127–33. [7] Corman JC. Coal to electricity: integrated gasification combined cycle. Appl Energy 1982;10:243–59. [8] Chopra S, Jain A. A review of fixed bed gasification systems for biomass. Agr Eng Int CIGR EJ 2007:9. [9] Dogru M, Midilli A, Howarth CR. Gasification of sewage sludge using a throated downdraft gasifier and uncertainty analysis. Fuel Process Technol 2002;75:55–82. [10] Larfeldt J, Leckner B, Malaaen MC. Modelling and measurements of the pyrolysis of large wood particles. Fuel 2000;79:1637–43. [11] Davidsson KO, Pettersson JBC. Birch wood particle shrinkage during rapid pyrolysis. Fuel 2002;81:263–70. [12] Kumar RR, Kolar AK, Leckner B. Shrinkage characteristics of Casuarina wood during devolatilization in a fluidized bed combustor. Biomass Bioenergy 2006;30:153–65. [13] Zhou C, Zhang Q, Arnold L, Yang W, Blasiak W. A study of the pyrolysis behaviors of pelletized recovered municipal solid waste fuels. Appl Energy 2013;107:173–82. [14] De Wild PJ, Reith H, Heeres E. Biomass pyrolysis for chemicals. Biofuels 2011;2:185–208. [15] Koukios EG. Progress in thermochemical, solid-state refining of biomass – from research to commercialization. Adv Thermochem Biomass Convers 1993;2:1678–92. [16] Park WC, Atreya A, Baum HR. Experimental and theoretical investigation of heat and mass transfer processes during wood pyrolysis. Combust Flame 2010;157:481–94. [17] Gašparovicˇ L, Labovsky´ J, Markoš J, Jelemensky´ L. Calculation of kinetic parameters of the thermal decomposition of wood by distributed activation energy model (DAEM). Chem Biochem Eng Q 2012;26:45–53.