Thermogravimetric analysis of microalgae combustion under different oxygen supply concentrations
Chunxiang Chen,Xiaoqian Ma,Kai Liu
DOI: https://doi.org/10.1016/j.apenergy.2011.03.003
IF: 11.2
2011-01-01
Applied Energy
Abstract:Highlights ► The combustion processes of Chlorella vulgaris could be divided into three stages. ► The weight loss rate of C. vulgaris attained its maximum at the first peaks. ► The major weight loss of C. vulgaris occurred at the second stage. ► The apparent activation energy of C. vulgaris was 134.03–242.33 kJ mol −1 . ► The optimal oxygen concentration for C. vulgaris combustion was 25–35 vol.%. Abstract Recently, studies of microalgae in China have increased a lot because of their obvious advantages over other biological fuels. In this paper, the combustion behavior of Chlorella vulgaris (a genus of unicellular green microalgae) was investigated in a thermogravimetric analyzer (TGA) from room temperature to 800 °C in O 2 /N 2 atmospheres. The effects of different oxygen concentrations (20, 50, 60, 80 vol.%) and different heating rates (10, 20 and 40 °C min −1 ) on the combustion processes of C. vulgaris had been studied. The results indicated that the combustion processes of C. vulgaris could be divided into three stages. The oxygen concentrations and heating rates had important effects on the main combustion processes of C. vulgaris . The iso-conversional method involving Flynn–Wall–Ozawa (FWO) and the Kissinger–Akahira–Sunose (KAS) methods were used for the kinetic analysis of the main combustion process. The results indicated that, when the oxygen concentration varied from 20 to 80 vol.%, the value of activation energy increased respectively from 134.03 to 241.04 kJ mol −1 by using FWO method and from 134.53 to 242.33 kJ mol −1 by KAS method. Moreover, the optimal oxygen concentration for C. vulgaris combustion was 25–35 vol.% . Keywords Chlorella vulgaris Oxygen concentration Combustion Thermogravimetric analysis (TGA) Kinetics 1 Introduction It is generally accepted that fossil fuels will be completely depleted in the near future: At current production rates, global proven reserves of crude oil and natural gas are estimated to last for 40.5 and 66.7 years, respectively [1] ; and for coal, about 164 years. Moreover, the excessive combustion of fossil fuels has led to serious environmental problems and even global climate change. In order to alter this trend and protect the planet, much effort is needed to develop new fuel technologies and to find alternative energy sources [2] . Lately photosynthetic capable biomass has been considered as an alternative energy source because it is a renewable resource and fixes CO 2 in the atmosphere through photosynthesis [3,4] , hence, it was also considered to be CO 2 -neutral fuels when being combusted [5] . Moreover, biomass can be converted into useful biofuels and biochemicals via biomass upgrading and biorefinery technologies [6] . In recent years, the biofuel production has increased [7] . However, for the first generation biofuels, primarily produced from food crops and mostly oil seeds, their ability were limited to achieve targets for biofuel production, climate change mitigation and economic growth. Therefore, more attention was given to developing the second generation biofuels produced from non-food feedstocks such as microalgae, which potentially offer the big opportunities in the long term because of their many advantages over existing energy crops [8–13] . If fuel is recovered efficiently from microalgae, the microalgae can be used as second generation biofuels instead of using fossil fuel [3] . At present, research and development activities include fundamentals of microalgae strain improvement, reactor design and process integration, with auxiliary power requirements and specific reactor cost being addressed as most critical issues [14,15] . It is worthy to note that the pyrolysis process of microalgae biomass has attracted a great deal of attention in recent years. Scott et al. [10] and Miao et al. [16] have studied thermal characteristics of microalgae under slow and fast pyrolysis conditions, respectively. Peng et al. [17] have reported two kinds of autotrophic microalgae, spirulina platensis (SP) and Chlorella protothecoides (CP) were pyrolyzed at the different heating rates in a thermogravimetric analyzer to investigate their pyrolytic characteristics. Zou et al. [3,18] have reported a species of unicellular green marine microalgae, dunaliella tertiolecta , was pyrolyzed in a thermogravimetric analyzer in a highly purified N 2 atmosphere at the different heating rates. However, the combustion behavior of microalgae under different oxygen concentrations in a thermogravimetric analyzer (TGA) has not been explored yet. Nowadays combustion is the most simple and direct technology available for biomass utilization, which is responsible for over 97% of the world’s bio-energy production [19] . However, in a real combustion chambers, the surface temperature of solid fuel increases rapidly and volatiles are released quickly due to high thermal radiation. There is not sufficient air to diffuse into the solids, and ambient oxygen concentration varies from time to time, which will affect combustion reactions of solids. Therefore, the kinetics of combustion of solid fuel under different oxygen concentrations would be of interest in fuel research [20] . In addition, with the development of low cost of oxygen making technology in present, such as permeable membranes and pressure swing adsorption, oxygen-enriched combustion technology is gradually developing with its application range continuously expanded [21] . Many studies have been done on the behavior of combustion of biomass in oxygen-enriched atmosphere by using thermogravimetry. The oxygen-enriched combustion of biomass micro fuel (BMF) was carried out respectively in the thermogravimetric analyzer and cyclone furnace, in order to investigate the effects of oxygen concentration on combustion performance [21] . Besides, the influences of different catalysts on the ignition and combustion of rice and wheat straw in air and oxygen-enriched atmospheres have been investigated through thermogravimetric analysis (TGA) and kinetic study [22,23] . In addition, pyrolysis and combustion of wood under different oxygen concentrations were studied by using TG-FTIR analysis and effects of oxygen concentration on TGA of wood and evolved gas analysis were thoroughly analyzed [20] . In this paper, thermogravimetric analysis was employed to investigate the combustion characteristics of Chlorella vulgaris under different oxygen supply concentrations. The objective was the kinetic analysis. Different heating rates were used for the kinetic analysis to obtain the apparent activation energy via Flynn–Wall–Ozawa (FWO) and the Kissinger–Akahira–Sunose (KAS) methods. Finally, the optimal oxygen concentration for oxygen-enriched combustion of Chlorella vulgaris was determined. 2 Experimental 2.1 Material The powder of microalgae, Chlorella vulgaris , was provided by the Jiangmen Yue Jian Biotechnologies Co. Ltd. (Guangdong Province, China). The proximate and ultimate analyses results and the calorific value were shown in Table 1 . The ultimate analysis and proximate analysis were based on ASTM D5373 criterion and GB212-91/GB212-84 criterion, respectively. ASTM D240-92, ASTM D4809-95, ISO 1928 and BSI 1016 criterions provided foundation for detecting heating values. The C. vulgaris sample was dried in oven at 105 °C for 24 h, then milled and sieved to obtain pure fractions of uniform particle with a size less than 200 μm. After above treatment, the sample was stored in a desiccator for test. 2.2 Experimental facility and procedure Thermogravimetry (TG) tests were carried out on NETZSCH STA 409 PC simultaneous analyzer. The TG baselines were corrected by subtraction of predetermined baselines determined under identical conditions except for the absence of a sample. To eliminate the effects caused by the mass and heat transfer limitations, small samples (10 mg) were loaded into an Al 2 O 3 ceramic crucible for each run under non-isothermal conditions. The small sample in the reactor was heated up from ambient temperature to 105 °C and held at this temperature for approximately 10 min to ensure that free-water was completely removed. Then, the sample was further heated to 800 °C with the heating rates of 10, 20 and 40 °C min −1 . In addition, the combustion experiments were performed in 20% O 2 /80% N 2 , 50% O 2 /50% N 2 , 60% O 2 /40% N 2 and 80% O 2 /20% N 2 with the flow rate of 100 ml/min in the pressure range of 0.05–0.1 MP. For each atmosphere and heating rate, three repetitive TG curves were obtained in order to assure reproducibility of the results. 2.3 Kinetic method In the non-isothermal experiments carried out with a thermo balance, the sample mass was measured as a function of temperature. The rate of degradation or conversion, dα / dt , is the linear function of a temperature-dependent rate constant, k ( T ), and the temperature-independent function of conversion, f ( α ), (1) d α dt = k ( T ) f ( α ) where α is the conversion degree, t (s) is time, T (K) is the absolute temperature, k ( T ) usually described by the Arrhenius equation: (2) k = A exp - E RT where A (s −1 ) is pre-exponential or frequency factor, E (kJ mol −1 ) is the activation energy, R (kJ mol −1 K −1 ) is the universal gas constant. The function f ( α ) is expressed as: (3) f ( α ) = ( 1 - α ) n where n is the reaction order. The degree of conversion of the reduction process is expressed as [24,25] : (4) α = m i - m t m i - m ∞ where m i is the initial mass of the sample, m t is the mass of the sample at time t , m ∞ the final mass of the sample in the reaction. Inserting Eqs. (2) and (3) in Eq. (1) , we obtain the kinetic equation in the following form: (5) d α dt = A exp - E RT ( 1 - α ) n (6) as β = dT dt where β is the heating rate.Eq. (5) could be transferred to (7) d α ( 1 - α ) n = A β exp - E RT dT An integration function of Eq. (7) is shown as (8) g ( α ) = ∫ 0 α d α ( 1 - α ) n = A β ∫ T 0 T exp - E RT dT (9) T = T 0 + β t where T 0 (K) is the starting temperature [26] . Isoconversional methods involve carrying out a series of experiments at different heating rates [27] . In this work, activation energy from dynamic data was obtained from iso-conversional method by Flynn–Wall–Ozawa (FWO) and Kissinger–Akahira–Sunose (KAS) methods. The Flynn–Wall–Ozawa (FWO) method [3] based on the following equation: (10) ln β = ln 0.0048 AE α Rg ( α ) - 1.0516 E α RT For α = const., ln β vs. 1/T obtained at several heating rates yields a straight line whose slope allows evaluation of the apparent activation energy. The Kissinger–Akahira–Sunose (KAS) method based on the following equation [3,28] : (11) ln β T α 2 = ln AR E α g ( α ) - E α RT α The E α for different conversion values can be calculated from a plotting ln β T α 2 vs. 1/T. 3 Results and discussion 3.1 Thermogravimetric study 3.1.1 TG and DTG curves analysis The weight loss (TG) and the rate of weight loss (DTG) curves during the combustion of Chlorella vulgaris in different atmospheres at a heating rate of 20 °C min −1 were showed in Figs. 1 and 2 , respectively. It can be seen that three individual stages could be distinguished during the heating process of C. vulgaris sample. The first stage was from room temperature to about 170 °C (the temperature of initial devolatilization), corresponded to a loss of moisture and the very light volatile materials. The second stage was from 170 °C to 612.5–657.3 °C (depending on the oxygen concentration) and was characterized by a major weight loss, corresponded to the main release and combustion of organic components. The third stage was from the end of the second stage to 800 °C, and the weight loss was very slight in this stage. As the combustion of microalgae was a very complex process, these three stages were assigned by using a broad classification. Table 2 showed the characteristic parameters of the combustion of C. vulgaris in different atmospheres at a heating rate of 20 °C min −1 . The analysis of the rate of weight loss curves showed that the weight loss rate of the second stage was notably higher than that of other two stages. Three peaks, occurred at 274.90–279.30 °C, 459.91–474.70 °C and 543.90–569.10 °C (depending on the oxygen concentration), were observed from the DTG curves in Fig. 2 . At the first peaks of DTG curves, the rate of weight loss attained its maximum ( Table 2 ). The weight loss of the second stage was mainly due to the release and combustion of volatile matter and combustion of fixed carbon. As shown in Table 2 , the main weight loss of the second stage occurred at the temperature ranges of 251.4–657.3 °C, 250.6–629.8 °C, 246.6–617.8 °C and 244–612.5 °C at the concentration of 20, 50, 60 and 80 vol.% respectively. 3.1.2 Effect of mineral matter loss on the weight loss of C. vulgaris Metals and other inorganic components of C. vulgaris were determined by digestion, and carried out by Accuris ICPES (induced coupled plasma emission spectrometer), as shown in Table 3 . It can be seen from Table 3 that sulfur, potassium, phosphorus, chlorine, magnesium and calcium are the major mineral elements present in C. vulgaris ; sodium, iron, zinc and selenium are present in smaller amounts. These mineral elements occur as oxides, silicates, carbonates, sulfates, chlorides, and phosphates in biomass [29] . During combustion or gasification of biomass, some of chlorine and potassium are released into the gas phase, such as HCl(g), KCl(g), (KCl) 2 (g), K 2 SO 4 (g), KOH(g), and K(g), etc. And the others may be retained in ash, forming potassium silicate, aluminosilicate or sulfate. Because of the relatively higher content of sulfur than chlorine in biomass, potassium sulfate and aluminosilicate becomes important in ash [30] . Moreover, it can be roughly estimated from Table 3 that the total volatile mineral matter content is small in C. vulgaris . Besides, the burnout temperature of volatile matters and fixed carbon of C. vulgaris is less than about 650 °C. At this low temperature range (before about 650 °C), only part of volatile mineral matter (potassium and chlorine, etc.) is loss. For example, when the heating temperature reaches about 500 °C, organic potassium is mainly released into the gas phase, while the inorganic potassium is released at higher temperature [31] . Therefore, the mineral matter loss has a small influence on the weight loss of Chlorella vulgaris at low temperatures. Further study aiming at the catalytic effect of metals such as sodium, potassium, calcium and magnesium on the volatile release and combustion is still needed. 3.1.3 Effects of oxygen concentration on the combustion process of C. vulgaris Fig. 1 showed that all the TG curves of samples under different oxygen concentrations were similar before 170 °C, and so were the DTG curves ( Fig. 2 ). This phenomenon can be explained as follows: the mass loss is attributed to drying and thermal decomposition of sample, in the kinetic control zone (before 170 °C), mainly affected by temperature, and the effect of oxygen concentration is almost negligible [21] . When the oxygen concentration increased, the mass loss rate increased in a temperature range from 170 to 500 °C, and the shorter time was consumed to burn out. Thus, thermal decomposition was accelerated by the increase of oxygen concentration [20,25] . The ignition temperature ( T I ) and final combustion temperature ( T F ) are usually used to describe the combustion characteristic of fuels. The lower the ignition temperature is, the easier the fuel is ignited. The lower the burnout temperature is, the less the time for fuels to burn out. It could be seen from Table 2 that both the ignition temperature and final combustion temperature decreased gradually with the increase of oxygen concentration. The effect of oxygen concentration on the final combustion temperature was much more remarkable than that of the ignition temperature. In addition, an increase in oxygen concentration caused a shift of the main weight loss curves to the lower temperature range without significant change in its shape. This result indicates that increasing oxygen concentration can intensify the combustion reaction of samples. The maximum reaction rates were 7.92% min −1 , 8.26% min −1 , 8.67% min −1 and 9.26% min −1 at oxygen concentration of 20, 50, 60 and 80 vol.% respectively. It could be observed that the maximum mass loss rate of sample increased with the increase of oxygen concentration. Thus, when the oxygen concentration was enhanced, the volatile materials would be released quickly, causing the maximum mass loss rate of samples to be increased [32] . In addition, with the increase of oxygen concentration, the residue mass of the samples tended to decrease, as shown in Table 2 . This result indicates that oxygen enhances the decomposition of the material at low temperatures, and when the temperature reaches high enough, oxygen is likely to promote the combustion of the char residue generated in the early stages of the solid decomposition [33] . As shown in Table 2 , when the oxygen concentration increased, the weight loss rate of three DTG peaks tended to increase. The reasons caused this phenomenon are that when the oxygen concentration increases, the amount of inert nitrogen contained in the flue gases is lowered, thereby reducing the amount of heat loss through the furnace exhaust and obtaining higher efficiency. Also the heat transfer process is favored because the higher temperature reached by the oxygen-enriched combustion increases the driving force of the heat exchange [21] . The average reaction rates of C. vulgaris in the second stage were calculated. Fig. 3a illustrated the fitting curve of average reaction rate of the second stage varied following the oxygen concentration, and the slope ( R D ) of this fitting curve was shown in Fig. 3b . As can be seen from Fig. 3a , the average reaction rates gradually raised with the increase of oxygen concentration. At 20–25% O 2 , the effect of oxygen concentration on the average reaction rates was not obvious, while at 25–48%, R D increased significantly and up to a maximum value at oxygen concentration 48 vol.% (a critical value). But above 48 vol.%, R D decreased sharply. Therefore, the optimal oxygen concentration was approximately 25–48 vol.% when combined with the cost of oxygen production. 3.2 Analysis of kinetic parameters According to Eqs. (10) and (11) , a plot of ln β , ln ( β/T 2 ) against 1/T should be a straight line. From the slope of line, we can estimate apparent activation energy ( E α ) of the dynamic combustion at various conversions α . In the present study, three different heating rates (10, 20 and 40 °C·min −1 ) were used to evaluate the relationship between E α and α by the iso-conversional method. The characteristic parameters of combustion of C. vulgaris at these three heating rates under a representative of oxygen concentration (50 vol.%) were shown in Table 4 . It could be noted that the heating rate not only influenced the release and burning temperatures of the volatile, but also influenced the weight loss rate and the residual mass of C. vulgaris combustion. The plots for determination of activation energy for the main stage of weight loss (i.e., second stage) at different α by FWO and KAS methods corresponding to the combustion of C. vulgaris in different atmospheres with the heating rates of 10, 20 and 40 °C min −1 were depicted in Figs. 4–7 . The activation energies in 0.2 ⩽ α ⩽ 0.8 calculated by the KAS and FWO methods were listed in Table 5 . As shown in Table 5 , the activation energy had the high linear correlation coefficients ( R 2 , related coefficient) in the range 0.93352–0.99999, and the average values of activation energy worked out through the KAS and FWO methods varied slightly, so the results were credible. Because the average values of activation energy worked out by the KAS and FWO methods were very close to each other, we took the average values of two methods as the values of activation energy of C. vulgaris , and they were 134.28, 164.62, 210.58 and 241.69 kJ mol −1 for the oxygen concentration of 20, 50, 60, 80 vol.% respectively. It should be noted that the activation energy increased with the increase of oxygen concentration. This result is consistent with what obtained by Munir et al. [34] and Liu et al. [35] . The activation energy is affected by decrease of activated molecule concentration, diffusion limitation and organic impurities during the process of combustion of samples [20] . As oxygen concentration increases, heat release from semi-coke oxidization increases and thus surface temperature of semi-coke increases. Besides, semi-coke structure expands the size of grain and increases ash content with the increase of final temperature. Therefore, the activation energy increases with increased oxygen concentration. To further show the mathematical relationship between the activation energy and oxygen concentration, the fitting curve of activation energy varied following the oxygen concentration was plotted in Fig. 8a . The fitted result agreed well with the calculated values and the fitting curve nearly liked an “ S ”. In addition, the slope ( R E ) of the fitting curve was shown in Fig. 8b . R E was 0.03748 and 0.09368 at oxygen concentration 20 and 35 vol.%, respectively. When the oxygen concentration was above 35 vol.%, R E increased sharply, up to 4.88564 at oxygen concentration 55 vol.% (a critical value). However, when oxygen concentration was above 55 vol.%, R E decreased sharply. Combined with the cost of oxygen production and the reaction rates of the combustion of C. vulgaris which were discussed in Section 3.1.3 , the optimal oxygen concentration was 25–35 vol.% for oxygen-enriched combustion of C. vulgaris , which was slightly different from that of biomass micro fuel (30–40%) [21] . 4 Conclusions The combustion characteristic and kinetic behavior of the C. vulgaris were studied at different oxygen concentration (20, 50, 60, 80 vol.%) and different heating rates (10, 20 and 40 °C min −1 ). The following conclusions could be drawn from this work: (1) During combustion processes of C. vulgaris sample at different oxygen concentration with a heating rate of 20 °C min −1 , three stages could be distinguished. The second stage was from 170 °C to 657.3 °C, which corresponded to the main de-volatilization and combustion processes. The DTG curves of C. vulgaris had three strong peaks, and the rate of weight loss attained its maximum at the first peaks. (2) The main elemental constituents of C. vulgaris minerals are sulfur, potassium, phosphorus, chlorine, magnesium and calcium, together with trace amounts of sodium, iron, zinc and selenium. The influence of mineral matter loss on the weight loss of C. vulgaris is small. (3) As the oxygen concentration increased, the ignition temperature, the final combustion temperature and the residue mass of C. vulgaris tended to decrease, while the maximum reaction rates increased. (4) With the increase of the heating rate, the ignition temperature, the weight loss rate of the peaks and the residual mass of C. vulgaris combustion tended to increase. (5) The average values of activation energy at the oxygen concentrations of 20, 50, 60 and 80 vol.% were 134.28, 164.62, 210.58 and 241.69 kJ mol −1 respectively. It was obvious that the activation energy was increased with the increase of oxygen concentration. (6) The optimal oxygen concentration was 25–35 vol.% for oxygen-enriched combustion of C. vulgaris when combined with the cost of oxygen production and the reaction rates of C. vulgaris . Acknowledgements This work was supported by the National Natural Science Foundation of China (No. 50906025 ); the Fundamental Research Funds for the Central Universities, SCUT (No. x2dlD2105280 ); Guangdong Key Laboratory of Clean Energy Technology ( 2008A060301002 ) and the National Basic Research Program of China (973 Program) : 2011CB201500 . References [1] A.K. Hossain O. Badr Prospects of renewable energy utilisation for electricity generation in Bangladesh Renew Sustain Energy Rev 11 8 2007 1617 1649 [2] H. Haykiri-Acma S. Yaman S. Kucukbayrak Effect of heating rate on the pyrolysis yields of rapeseed Renew Energy 31 6 2006 803 810 [3] S.P. Zou Y.L. Wu M.D. Yang C. Li J.M. Tong Pyrolysis characteristics and kinetics of the marine microalgae Dunaliella tertiolecta using thermogravimetric analyzer Bioresour Technol 101 1 2010 359 365 [4] Z.A. Sun J.Z. Shen B.S. Jin L.Y. Wei Combustion characteristics of cotton stalk in FBC Biomass Bioenergy 34 5 2010 761 770 [5] S. Munir W. Nimmo B.M. Gibbs Shea meal and cotton stalk as potential fuels for co-combustion with coal Bioresour Technol 101 19 2010 7614 7623 [6] K. Kita S. Okada H. Sekino K. Imou S. Yokoyama T. Amano Thermal pre-treatment of wet microalgae harvest for efficient hydrocarbon recovery Appl Energy 87 7 2010 2420 2423 [7] A. Demirbas Competitive liquid biofuels from biomass Appl Energy 88 1 2011 17 28 [8] T.M. Mata A.A. Martins N.S. Caetano Microalgae for biodiesel production and other applications: a review Renew Sustain Energy Rev 14 1 2010 217 232 [9] X. Meng J.M. Yang X. Xu L. Zhang Q.J. Nie M. Xian Biodiesel production from oleaginous microorganisms Renew Energy 34 1 2009 1 5 [10] S. Grierson V. Strezov G. Ellem R. Mcgregor J. Herbertson Thermal characterisation of microalgae under slow pyrolysis conditions J Anal Appl Pyrolysis 85 1–2 2009 118 123 [11] J.Y. Lee C. Yoo S.Y. Jun C.Y. Ahn H.M. Oh Comparison of several methods for effective lipid extraction from microalgae Bioresour Technol 101 1 2010 S75 S77 [12] L. Brennan P. Owende Biofuels from microalgae – a review of technologies for production, processing, and extractions of biofuels and co-products Renew Sustain Energy Rev 14 2 2010 557 577 [13] G.H. Huang F. Chen D. Wei X.W. Zhang G. Chen Biodiesel production by microalgal biotechnology Appl Energy 87 1 2010 38 46 [14] C. Postena G. Schaub Microalgae and terrestrial biomass as source for fuels—a process view J Biotechnol 142 1 2009 64 69 [15] C.H. Hsieh W.T. Wu Cultivation of microalgae for oil production with a cultivation strategy of urea limitation Bioresour Technol 100 17 2009 3921 3926 [16] X.L. Miao Q.Y. Wu C.Y. Yang Fast pyrolysis of microalgae to produce renewable fuels J Anal Appl Pyrolysis 71 2 2004 855 863 [17] W.M. Peng Q.Y. Wu P.G. Tu N. Zhao Pyrolytic characteristics of microalgae as renewable energy source determined by thermogravimetric analysis Bioresour Technol 80 1 2001 1 7 [18] S.P. Zou Y.L. Wu M.D. Yang C. Li J.M. Tong Characteristics and dynamics of pyrolysis process microalgae J Combust Sci Technol 13 4 2007 330 334 [in Chinese] [19] M.F. Demirbas Biorefineries for biofuel upgrading: a critical review Appl Energy 86 Supplement 1 2009 S151 S161 [20] M.X. Fang D.K. Shen Y.X. Li C.J. Yu Z.Y. Luo K.F. Cen Kinetic study on pyrolysis and combustion of wood under different oxygen concentrations by using TG-FTIR analysis J Anal Appl Pyrolysis 77 1 2006 22 27 [21] S.Y. Luo B. Xiao Z.Q. Hu S.M. Liu Y.W. Guan Experimental study on oxygen-enriched combustion of biomass micro fuel Energy 34 11 2009 1880 1884 [22] Z.S. Yu X.Q. Ma A. Liu Thermogravimetric analysis of rice and wheat straw catalytic combustion in air- and oxygen-enriched atmospheres Energy Convers Manage 50 3 2009 561 566 [23] Z.S. Yu X.Q. Ma A. Liu Kinetic studies on catalytic combustion of rice and wheat straw under air- and oxygen-enriched atmospheres, by using thermogravimetric analysis Biomass Bioenergy 32 11 2008 1046 1055 [24] A.A. Zuru S.M. Dangoggo U.A. Birnin-Yauri A.D. Tambuwal Adoption of thermogravimetric kinetic models for kinetic analysis of biogas production Renew Energy 29 1 2004 97 107 [25] H.M. Xiao X.Q. Ma Z.Y. Lai Isoconversional kinetic analysis of co-combustion of sewage sludge with straw and coal Appl Energy 86 9 2009 1741 1745 [26] J.M. Cai R.H. Liu Kinetic analysis of solid-state reactions: precision of the activation energy obtained from one type of integral methods without neglecting the low temperature end of the temperature integral Solid State Sci 10 5 2008 659 663 [27] M.E. Sanchez M. Otero X. Gomez A. Moran Thermogravimetric kinetic analysis of the combustion of biowastes Renew Energy 34 6 2009 1622 1627 [28] B. Boonchom S. Puttawong Thermodynamics and kinetics of the dehydration reaction of FePO 4 ·2H 2 O Phys B 405 9 2010 2350 2355 [29] K. Raveendran Anuradda Ganesh Kartic C. Khilart Influence of mineral matter on biomass pyrolysis characteristics Fuel 74 12 1995 1812 1822 [30] Xiaolin Wei, Uwe Schnell, Hein Klaus RG. Behaviour of gaseous chlorine and alkali metals during biomass thermal utilization. Fuel 2005;84(7–8):841–8. [31] Yu Chun-jiang, Luo Zhong-yang, Zhang Wen-nan, Fang Meng-xiang, Zhou Jin-song, Cen Ke-fa. Inorganic material emission during biomass pyrolysis. J Fuel Chem Technol 2000;28(5):420–5 [in Chinese]. [32] Y.F. Liao X.Q. Ma Thermogravimetric analysis of the co-combustion of coal and paper mill sludge Appl Energy 87 11 2010 3526 3532 [33] N.A. Liu W.C. Fan R. Dobashi L.S. Huang Kinetic modeling of thermal decomposition of natural cellulosic materials in air atmosphere J Anal Appl Pyrolysis 63 2 2002 303 325 [34] S. Munir S.S. Daood W. Nimmo A.M. Cunliffe B.M. Gibbs Thermal analysis and devolatilization kinetics of cotton stalk, sugar cane bagasse and shea meal under nitrogen and air atmospheres Bioresour Technol 100 3 2009 1413 1418 [35] G.H. Liu X.Q. Ma Z.S. Yu Experimental and kinetic modeling of oxygen-enriched air combustion of municipal solid waste Waste Manage 29 2 2009 792 796