Solid fossil fuels thermal decomposition features in air and argon

Fuel 199 (2017) 145–156

Contents lists available at ScienceDirect

Fuel journal homepage: www.elsevier.com/locate/fuel

Full Length Article

Solid fossil fuels thermal decomposition features in air and argon Svetlana A. Epshtein, Elena L. Kossovich ⇑, Vladimir A. Kaminskii, Nikolay M. Durov, Nadezhda N. Dobryakova National University of Science and Technology ‘‘MISiS”, 4, Leninsky prospect, Moscow 119049, Russian Federation

h i g h l i g h t s
Coals oxidation is described by TGA data processing.
TGA experiments were held in inert and oxidizing environments.
Sample set contains coals of different nature and ranks.
Two groups of coals are found in context of their interaction with oxygen.
New kinetic model is proposed to mathematically process TGA data.

a r t i c l e

i n f o

Article history: Received 21 February 2016 Received in revised form 20 January 2017 Accepted 23 February 2017

Keywords: Solid fossil fuels Coal rank Low-temperature oxidation Thermogravimetric analysis Activation energy Decomposition rate

a b s t r a c t Investigation and characterization of coals oxidation is one of the most current problems due to the fact that it may lead to quality loss and spontaneous combustion hazards at mining, storage and utilization. A detailed analysis was performed for thermogravimetric analysis of selection of coals with a wide rank range, different structural characteristics, petrographic and elements composition and origins during experiments in inert and oxidizing environments. It was found that for both the considered environments, the coals thermogravimetric curves retain their space disposition order. Values of the temperatures corresponding to maximal decomposition rates correlate with coal rank as for inert, as for oxidizing environments. For bituminous coals and anthracite, temperatures corresponding to maximal decomposition rate drifted to the zone of higher temperatures in case of experiments in air in comparison with tests in inert. For lignites, they moved to lower temperature intervals. Values of the maximal thermal decomposition rate for bituminous coals well correlate with aromaticity degree. For lignites, the maximal decomposition rates in air and argon have comparable values, whereas for high-rank coals and anthracite they grow in oxidizing environment in comparison with inert one. Kinetic parameters were evaluated for the most common stages of coals mass change during thermogravimetric experiments in air and argon. Kinetic parameters of coals pyrolysis and combustion correlate not only with rank but also with aromaticity degree. For bituminous coals at high temperature intervals at experiments in inert environment, there was found a decrease of activation energy values with rank and aromaticity degree growth. As for corresponding values for bituminous coals tested in oxidizing environment, there was an increase of activation energy with rank and aromaticity degree. Two groups were allocated for coals in context of their interaction with oxygen. The first group included lignites and peat and is characterized by enhanced volatiles release at temperatures over 100 °C. The second one contains coals that are prone to oxygen adsorption at low-temperature intervals (up to 300–400 °C) (bituminous coals and anthracite). New parameter was proposed for preliminary describing coals group affiliation. This parameter denotes mass gain or decrease for coals tested in oxidizing environment. The complex of kinetic parameters may serve as an additional but informative tool to view at coals oxidation mechanisms. This, presumably, could be performed at simultaneous characterization of thermogravimetric data with heat flow and gas analysis. Ó 2017 Elsevier Ltd. All rights reserved.

⇑ Corresponding author. E-mail addresses: [email protected] (S.A. Epshtein), [email protected] (E.L. Kossovich), [email protected] (V.A. Kaminskii), [email protected] (N.M. Durov), [email protected] (N.N. Dobryakova). http://dx.doi.org/10.1016/j.fuel.2017.02.084 0016-2361/Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction Characterization of different solid fossil fuels behavioral features is becoming increasingly popular due to the related issues

146

S.A. Epshtein et al. / Fuel 199 (2017) 145–156

of mining, storage and utilization. Recent researches are concentrated on variation of quality, characteristics and combustion properties of fossil fuels with rank [1–4], oxidation degree [5] and structure [6]. One of the most relevant problem nowadays is studying of solid fossil fuels oxidation that may lead to quality loss and spontaneous combustion hazards at mining, storage and utilization [7]. In order to study the effects of structure and properties of coals and other solid fossil fuels at their pyrolysis and combustion behavior, thermogravimetric analysis (TGA) is widely applied [8,9]. Coals pyrolysis stages detailed experimental and analytical investigation was shown in [10]. It was shown that vitrinite reflectance index of coals is in a good correlation with thermal decomposition kinetic parameters obtained from TGA studies [11,12]. Propensity of coals of different rank to oxidation and spontaneous combustion was studied in [13,14], where authors introduced new parameters derived from TGA tests and found correlation between them with rank. As part of analysis of coals thermochemical behavior during oxidation, authors of [4] discussed that thermogravimetric (TG) curves shapes variate with coals rank. They tested low- and high-rank coals in oxygencontaining gas flow. And observed that for the high-rank coal there exists a pronounced stage of oxygen adsorption (mass increase at temperatures 150–300 °C), whereas for low-rank coals such phenomena were not found at similar thermal intervals. TGA of bituminous coals in environments containing different values of oxygen concentration were performed by [14–18] and the phenomenon of oxygen adsorption at low-temperature intervals (up to 400 °C) also was highlighted. In [13] it was shown that for bituminous coals such stage remains for all the considered heating rates, and, along with the heating rate growth, the oxygen adsorption becomes less pronounced, whereas for lignites no mass growth was found for all the heating rates. In [19] authors revealed that for the low-rank coal there exist no such stage for tests in oxidizing environment, and increase of heating rate reduces mass loss rate. In [15], while investigating of a bituminous coal, authors revealed that growth of oxygen concentration in atmosphere of TG experiment enhances its adsorption into coal matter at lowtemperature intervals, and increase of heating rate induces growth of the maximal weight loss rate and the corresponding temperatures. As for lignites, increase of oxygen concentration in experiment environment lead to pronounced shifts of maximal weight loss rate temperatures to the lower values [20]. Some lowtemperature coal oxidation features were observed and discussed in [17,18,21], such as thermal decomposition kinetic parameters comparison at low and high temperature intervals. Solely TGA of coals do not provide unambiguous information on oxidation mechanisms because non-stationary heating regime leads to simultaneous proceeding of various processes, e.g. physical and chemical sorption, oxidation, decomposition, etc. Therefore, the majority of the coal thermogravimetric analysis researches nowadays are dedicated to combining of TGA and differential scanning calorimetry (DSC) or differential thermogravimetric methods (DTA), along with the recently developed approach of simultaneous TGA/Fourier transform infrared spectroscopy (FTIR) for studying the thermal decomposition features of coals. Such combinations of research methods are especially important in context of studying of coals propensity to oxidation, in order to obtain more information on the mechanisms of coals interaction with oxygen. For example, TG-FTIR was used for the quantitative characterizing of coals oxidation degree by analyzing the volatiles gas composition [22–24]. On the other hand, TGA of coals could serve as a supplementary but comprehensive tool for characterization of the prevailing mechanisms of oxidation and spontaneous combustion. To this end, one may use comparative analysis of solid fossil fuels behavior at low- and hightemperature intervals in inert and oxidizing environments during

the TGA experiments, as shown in [25]. Unfortunately, only a few papers were found with comparative analysis (see, e.g., [15,18,25,26]), and there is lack of unified information of such type for coals of different rank and structure. In this work we performed thermogravimetric analysis of coals in inert (argon) and oxidizing (air) gas flow under coherent conditions of experiment. A sample set of 15 coals with wide rank range, different structure, of various basins and origins was used. Mass loss patterns obtained in air and argon were compared at lowand high-temperature intervals for each sample. Kinetic parameters were evaluated by model [27] and compared for low- and high-temperature intervals for all the samples at experiments in air and argon. Also, a new parameter was introduced for indication of dominating processes of oxygen sorption or enhanced matter decomposition of coals at low-temperature intervals prior to combustion stage. Complex of kinetic parameters and the introduced one could serve as an additional informative tool (along with heat flow measurements, gas analysis, etc.) for clarification of coals oxidation mechanisms. This data may be useful for prognosis of coals propensity to spontaneous combustion as well as characterization of tendency to oxidation in context of quality loss at mining and storage, coals gasification, and many other related issues.

2. Materials and methods 2.1. Coals samples characterization and preparation In the work the studies were performed at coals of different rank and origins from different deposits of the Russian Federation. The sample set of coals counts 14 items. Characteristics of samples are shown in Table 1. On using samples chemical composition data, the atomic ratio H/C (at.) was determined. This atomic ratio well correlates with aromaticity degree fa, as it was shown previously by [28–30]. Therefore, further in the text we will use this parameter, H/C, as substituent of coals aromaticity degree. Data shown in Table 1 reveals that the selected samples set contains coals with different metamorphism degree, petrographic and chemical compositions. All lignites are characterized by very close values of vitrinite reflectance index (Ror, %) (range 0.30–0.38%) but by relatively wide range of H/C. For example, lignite 5 has the largest value of H/C ratio and low index Ror. At the same value of Ror., lignites 3 and 5 differ by H/C. There also exist variations in petrographic composition of coals: lignites 5 and 2 are mostly consist of vitrinite (Vt), whereas ##3 and 4 have much lower Vt percentage. A major share of samples is occupied by bituminous coals of different ranks. Their vitrinite reflectance indexes vary from 0.5 to 1.8%, carbon 76–88%, hydrogen 3.90–5.70%, atomic ratio H/C is 0.53–0.91 at. Also, the bituminous coals samples have different petrographic composition with vitrinite share varying from 31 to 94%. An oxidized bituminous coal also was included (#6) characterized by higher moisture, volatile matter, lower carbon and hydrogen contents along with higher H/C ratio in comparison with coals with the same rank. Also, within the chosen set of solid fossil fuels, there exists anthracite with high Sulphur contents, lowest H/C ratio (i.e. highest aromaticity degree) and with Ror = 3.58%. And, a sample of peat was included. This sample is characterized by high hydrogen content and low carbon one. H/C for peat has the largest value among all the samples set. As it was expected, there is a good correlation between vitrinite reflectance index and atomic ratio H/C of the studied samples (Fig. 1). With Ror growth, the atomic ratio values of H/C decrease, but for lignites there exist high level of dispersion. At the moment, this samples set was used partially in work [31] aimed at evaluation of coals propensity to oxidation and spontaneous combustion.

2.85 1.57 1.26 2.25 0.96 2.24 2.23 2.25 1.58 2.24 2.51 2.23 2.17 2.29 1.15 6.72 5.37 4.95 4.93 5.36 4.97 5.15 5.16 5.68 4.56 4.67 3.90 4.04 4.08 1.61 59.05 73.59 66.09 73.94 64.25 66.45 76.39 80.85 74.66 79.84 85.39 88.31 85.48 85.61 92.34 0,68 0.61 0.35 1.79 0.26 0.42 0.37 0.41 1.1 0.51 0.33 0.51 0.16 0.65 1.6 Not determined 48.8 59.3 41.91 47.5 41.6 38.8 35.7 45.9 37.6 23.7 13.4 21.9 18.7 3.5

H Total Sulphur Sdt (on dry basis)

9.7 17.9 11.5 13.35 4.3 14.8 15.3 13.1 23.1 17.5 18.9 19.6 11.3 26.4 4.3 9.4 5.9 7.4 14.93 10.8 6.8 3.1 2.3 4.8 2.1 1.0 1.2 1.2 1.8 1.2 Not determined 0.38 0.30 0.33 0.30 0.43 0.50 0.66 0.51 0.64 0.96 1.80 1.04 1.52 3.58 Not determined 86 75 79 94 80 54 35 94 39 34 64 31 55 85 Peat Lignite Lignite Lignite Lignite Bituminous oxidized Bituminous Bituminous Bituminous Bituminous Bituminous Bituminous Bituminous Bituminous Anthracite 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Murmansk region Irkutsk basin Hankaiski basin Irkutsk basin Kansko-Achinsky basin Kuzbass basin Kuzbass basin Kuzbass basin Irkutsk basin Kuzbass basin Kuzbass basin Kuzbass basin Kuzbass basin Kuzbass basin Donetsk basin

Type

Moisture Wa

Ash Ad (on dry basis)

Volatile matter Vdaf (on dry, ash free basis)

C

N

1.3656 0.8763 0.8994 0.7945 1.0019 0.8974 0.8091 0.7651 0.9137 0.6852 0.6566 0.5304 0.5674 0.5715 0.2098

147

Fig. 1. Correlation between vitrinite reflectance index of samples and their atomic ratio H/C.

Due to the large number of samples, further in work we will present graphical results mainly for typical ones or some exceptions. Tables will contain all the data obtained from the studied fuels. 2.2. Installation and experimental procedure

Sample #

Table 1 Characteristics of samples.

Origin

Vitrinite content Vt,% (on mineral matter free basis)

Vitrinite reflectance index Ror, %

Proximate analysis,% mass

Ultimate analysis,% mass (on dry, ash free basis)

Atomic H/C ratio, at

S.A. Epshtein et al. / Fuel 199 (2017) 145–156

All samples were conditioned to air-dry state, grinded to the particles size of 200 lm. Afterwards, the samples were stored in containers under conditions excluding their oxidation prior to the main experimental procedures. Thermogravimetric studies were performed at Netzsch STA 449C Jupiter Thermo-microbalance with vacuum-tight camera construction providing experiments in pure gas atmospheres. Preliminary, an empty crucible was tested in environments and conditions identical to the main experiments for further accounting of its own thermal effects. Experiments were performed in two different environments: inert (argon) and oxidizing (air). Calibration of the installment was performed both for temperature and mass signals. Due to features of the TG-analyzer, the materials are only applicable characterized by first-order phase transitions. The main experiments conditions were used for calibration procedure: alundium crucibles reasoned utilization of the pure metals as a reference material (indium, bismuth, stannum, aluminum, zinc and aurum); heating rate was 20 °C/min, with two environments: argon and air 25 ml/min flow. The statistical errors were evaluated using 3 melting (heating) and 2 crystallization (cooling) cycles, i.e. the program had five steps. Each test sample was heated on average by 10–50 °C higher temperature values than its melting temperature to achieve full melt. All this was done to reach accuracy not only of TG, but also DTA signals. The calibration curve design files were constructed for each environment regime to avoid experimental errors. Finally, the accuracy of the quantitative measurements was found to be ±0.1% wt. for mass alterations and ±0.5 °C for temperature. During the main experimental procedure, each sample was heated in the temperature interval 30–900 °C with rate of 20 °C/ min with argon and air flow of 25 ml/min. All the tests were held in the same alundium crucible to exclude its effect on the results. After each experiment, the crucible was mechanically cleaned and washed in acetone. Also, its mass was controlled and, in case of its variation, the corrections were introduced. Prepared 50 mg mass sample was inserted into an open crucible installed into the furnace at the thermobalance. The sample was stored in the gas flow at 30 °C during 15 min for reaching its thermostability. Heating was switched on afterwards. At the end of experiment, the crucible with sample was cooled in the inert gas flow (argon). The experi-

148

S.A. Epshtein et al. / Fuel 199 (2017) 145–156

mental data were processed by Proteus software (Netzsch) by extracting the numerical data of temperature growth, mass decrease and differential TG (DTG) quantities. This allowed experimental data processing by mathematical methods. Additionally, to evaluate precision of TG mass signals related to the samples under study, we used the following procedure. Experiments were held in two environments twice at each sample. The results (TG curves and mass alteration numerical data) were compared at each corresponding temperature point, and deviation values (i.e. precision at each point in percent quantities) were found as a ratio of absolute difference to an average mass decrease. Then, a maximum was found for each set of measurements. Such maxima were considered as a measurement precision for each sample under the repeatability conditions of experiments. It was found that precision of TGA at parallel tests on the same samples have larger value for inert environment (not exceeding 2.9% (of average) for peat and lignites and 2.6% (of average) for bituminous coals of different rank, 1% for anthracite) as compared with oxidizing one (2.0%, 1.4% and 1.0%, respectively), although its behavior is consistent.

      Ai T Ei Ei Mi ðTÞ ¼ M i0 Ai exp  exp   exp  a T0 RT RT 0     Ei Ei Ei Ei ; Ei 1;  Ei 1;  þ  RT 0 RT RT 0 RT 0 where Eiðn; xÞ ¼

R1 1

ext tn

ð3Þ

dt,



    E i Ai T Ei Ai Ei W i ðTÞ ¼ M i0 Ai exp exp   þ exp  RT a T 0 RT a RT 0      Ai Ei Ei Ei : ð4Þ Ei 1;   Ei 1;  þ a RT 0 RT RT 0 Then the sample mass evolution throughout the experiment could be written as

MðTÞ ¼

N1 X M i ðTÞ þ M R ;

ð5Þ

i¼0

where MR is a remainder of the sample not involved in fractions decomposition. Analogously, an expression for rate W(T) could be written as (6) N1 N1 X 1 X dM i ðTÞ : W i ðTÞ ¼  T dT a 0 i¼0 i¼0

ð6Þ

2.3. Thermal decomposition kinetic characteristics evaluation

WðTÞ ¼

Kinetic parameters, such as activation energies and preexponential factors, were found using TG analysis data with help of modification of methodology presented in paper [27]. The modification was based on choosing the non-intersecting fractions corresponding to different stages of thermal decomposition. Studying of thermal decomposition of coals and other heterogeneous systems is performed by division by fractions (each with initial mass M0i) describing dominating process leading to mass alteration of the sample at the current temperature interval. Note that at any fraction, despite the prevailing processes, primarily reasoning mass change, there exist other ones with non-dominating contribution. Such contribution is also included to the results of the model calculations, but is considered negligible in comparison with the prevailing processes. Usually, there could be found a finite number of such fractions (let us call it N). For any fraction (i.e. prevailing process of thermal decomposition, oxidation, oxygen sorption, etc., along with accompanying ones), its decomposition kinetics is determined by simple first-order reactions (1)

Kinetic parameters adjustment was performed numerically by the least squares method based on explicit expressions for dependencies Mi(T) (3), (5) and digitalized TG curves as a twodimensional matrix containing temperature growth data with step of 1 °C and corresponding values of sample mass at each point. Control of correctness was performed by comparison of curves W (T) from (4) and (6) with DTG graphs. The least squares method was based on solving of the following problem at each fraction i

dM i ¼ ki M i ðtÞ; dt

ð1Þ

Ei

where ki ¼ Ai eRT is rate constant of fraction thermal decomposition, T = T0(1 + at) – dependence of temperature growth with time with rate a, Ei is activation energy, Ai is pre-exponential factor, T0 is the initial temperature, R is the gas constant, t is time, Mi – mass of the coal substance involved in the current fraction with number i. After introduction of fraction decomposition rate by equality (2)

Wi ¼ 

Ei dM i  ¼ Ai e RT 0 ð1þatÞ M i ðtÞ; dt

ð2Þ

i and under the condition of dW ¼ 0 at T = Tim, where Tim is temperadT ture of maximal thermal decomposition rate of fraction i, it is possible to find connection of Arrhenius parameters Ei, Ai with values
 0 Mi m ¼ Mi ðt m Þ t im ¼ T imaTT , Tim and Wim (maximal rate). This 0

assumption is valid for non-intersecting fraction decomposition thermal intervals. Using these correlations and above equalities (1) and (2), it is possible to write an explicit expression for the kinetic equation solution, as it was presented in [27] (see Eqs. (3) and (4))

1 0T Nf X 2 ðMðTÞ  Mexp ðTÞÞ C B C BT¼T C B 0 F i ðki ; Ai ; Ei Þ ¼ minfraction B C; 2 C B M 0i A @ where M(T) are the model prediction mass functions from (3) and (5), Mexp(T) are the experimentally derived values of mass evolution, taken at the same temperature points as M(T), T0 is the temperature of the fraction interval beginning, TNf – at the end, M0i is the current fraction initial mass. Note that among the sought-for kinetic parameters ki, Ai, Ei there are only two independent ones, whereas the third parameter could be found from Arrhenius law equalities. Therefore, it is possible to use function Fi(ki, Ai, Ei) for fitting only two parameters, for example ki, Ai, and, on their basis, to evaluate activation energy. Examples for steps of experimental TG curves fitting by fractions are shown in Fig. 2a–d. The latter demonstrates DTG curve matched by rate W(T). Accuracy of parameters evaluation by curves fitting was found to be ±0.5%. Precision for evaluation of activation energies Ei and rate constants ki was found as a ratio (in percent) of the absolute difference between values for parallel TGA measurements divided by the averaged one. For all the samples, precision does not exceed 9% for activation energies and 5% for rate constants. 3. Results and discussion 3.1. Analysis of characteristics of solid fossil fuels thermogravimetric curves obtained in air and argon During the experiments, a series of TG curves were obtained for samples tested in different environments. In Fig. 3 only typical

S.A. Epshtein et al. / Fuel 199 (2017) 145–156

149

Fig. 2. Fitting procedure of experimental TG curves by fractions.

ones are shown due to the large total number of curves. It could be noticed from comparison of Fig. 3a and b that for both the considered environments, the coals’ TG curves retain their space disposition order. Therefore, it can be assumed that the general pattern of mass change during TG analysis primarily depend on coal rank, petrographic composition, element composition and its aromaticity degree for the experiments held both in inert [32] and oxidizing environments [13]. Analysis of the maximal values of thermal decomposition rates (Wmax) was performed at high-temperature intervals (pyrolysis or combustion). Such values found as altitudes of DTG peaks located within temperature intervals corresponding to intensive decompo-

sition stage. Also, the temperature values related to peaks (TWmax) were fixed. Results are shown in Table 2. It could be seen that, for experiments in inert gas flow, Wmax values tend to descend with coals rank growth. The previous results (at a collection of bituminous coals of Donetsk basin) revealed a good correlation between the maximal decomposition rates in inert environment and the metamorphism degree (vitrinite reflectance index) [12]. As for Wmax found for experiments in oxidizing environment, there is no obvious dependence of such kind. In experiments in argon gas flow, temperatures TWmax significantly increase with rank. The latter is retained for the experiments held in oxidizing environment: TWmax also grow with rank. It should be

150

S.A. Epshtein et al. / Fuel 199 (2017) 145–156

Fig. 3. Typical TG curves of solid fossil fuels studied in a) inert environment and b) air (oxidizing) environment.

also noted that Wmax for peat tested in oxidizing environment has an extremely high value, and this could be related to matter ablation at combustion stage. For bituminous coals and anthracite, values TWmax shifted to the zone of higher temperatures in case of experiments in air in comparison with tests in inert (see, e.g. [15]). As for lignites and peat, one may observe a converse situation, namely, shift to the zone of lower temperatures. This was previously discussed by authors of [20]. For lignites, in most cases, maximal decomposition rates in air and argon have comparable values, whereas for high-rank

coals and anthracite they grow in oxidizing environment in comparison with inert one. An observation on dependence of TWmax on coal rank is graphically demonstrated in Fig. 4 for both environments of experiment. Such a pronounced correlation was found for bituminous coals. At the same time, the range of vitrinite reflectance index for lignites is too short to find a good correlation with TWmax values. As previously was observed, there was no univalent correlation found for values of maximal decomposition rate and coals rank at experiments in oxidizing environment. But a relation

S.A. Epshtein et al. / Fuel 199 (2017) 145–156 Table 2 Comparison of maximal values of thermal decomposition rate of solid fossil fuels at TG analysis in different environments. Sample #

Values of maximal decomposition rate Wmax,%/min

Peat 1 Lignite 2 Lignite 3 Lignite 4 Lignite 5 Bituminous Oxidized 6 Bituminous 7 Bituminous 8 Bituminous 9 Bituminous 10 Bituminous 11 Bituminous 12 Bituminous 13 Bituminous 14 Anthracite 15

Corresponding temperature TWmax, °C

Argon

Air

Argon

Air

0.090 0.040 0.055 0.070 0.050 0.035 0.083 0.065 0.093 0.066 0.032 0.013 0.015 0.010 0.011

0.210 0.045 0.058 0.055 0.069 0.055 0.050 0.068 0.095 0.080 0.055 0.053 0.061 0.050 0.065

315 440 420 480 455 455 460 485 460 480 500 550 515 535 735

275 310 300 430 370 430 510 565 510 550 600 605 610 595 750

was found between Wmax and aromaticity degree H/C, as shown in Fig. 5 for bituminous coals. On the other hand, due to the number of lignites in the selected samples set, there was no univalent correlation found between Wmax and H/C. In Fig. 5 in could be noticed that for bituminous coals values of Wmax in oxidizing environment are relatively larger in comparison with corresponding ones in inert, but along with H/C ratio growth such difference diminishes. According to the above observations, the following may be highlighted: - Values of the temperatures corresponding to maximal decomposition rates correlate with the coal rank for both inert and oxidizing environments. In oxidizing environment in comparison with inert one, TWmax values for lignites shift to the lower temperatures, whereas for bituminous coals – to higher.

151

- Values of the maximal thermal decomposition rate for bituminous coals, as in oxidizing, as in inert environments, well correlate with H/C parameter. Along with increase of H/C atomic ratio, Wmax grow. In general, Wmax are larger for experiments in oxidizing environment with respect to inert ones. The largest difference is observed for coals with low H/C values. With increase of H/C, such differences gradually diminish.

3.2. Kinetics of solid fossil fuels decomposition in different environments Relation between TG curve shapes and coal rank was previously observed by [4] and other researchers. Moreover, according to [4,10], for bituminous coals there are at least three most common stages characterized by pronounced prevailing processes: moisture removal, oxygen adsorption or chemisorption (in case of experiments in oxidizing environment) or constant mass (in case of inert environment) and combustion or pyrolysis (stage of intensive thermal decomposition). As for low-rank coals including lignites and peat, the stage of constant mass (in inert environment) or oxygen adsorption (in oxidizing environment) is not pronounced, moreover, it is considered to be involved in the stage of moisture removal [20,33]. In order to obtain kinetics characteristics of the complex processes that reason coals behavior during TG analysis, the following stages and corresponding fractions were allocated. Fraction 0 corresponds to a low-temperature interval of prevailing process of moisture removal [18] accompanied by thermal decomposition (below 150–200 °C). Fraction 2 corresponds to the hightemperature interval of prevailing pyrolysis or combustion processes. Fraction 1 is characteristic only for bituminous coals 8–14 and anthracite 15 and corresponds to the pre-intensive decomposition stage in case of experiment in inert environment [10] and mass gain interval for the experiments in oxidizing environment [4,10,16]. Results of kinetic parameters evaluation are shown in Tables 3 (for inert environment) and 4 (for the oxidizing one). It could be noted that, for all the samples, temperature intervals corresponding to fraction 2 (pyrolysis or combustion stage) shifted to the area of lower temperatures at the experiments in oxidizing

Fig. 4. Correlation between TWmax and rank for bituminous coals for both inert and oxidizing environments.

Fig. 5. Change of maximal decomposition rate of bituminous coals with aromaticity degree in inert and oxidizing environments.

152

S.A. Epshtein et al. / Fuel 199 (2017) 145–156

Table 3 Kinetics of thermal decomposition of solid fossil fuels in inert environment. Sample #

Peat 1 Lignite 2 Lignite 3 Lignite 4 Lignite 5 Bituminous Oxidized 6 Bituminous 7 Bituminous 8 Bituminous 9 Bituminous 10 Bituminous 11 Bituminous 12 Bituminous 13 Bituminous 14 Anthracite 15

Fraction 0

Fraction 1

Fraction 2

Interval T0, °C

M00,%

k0, 1/ min

A0, 1/ min

E0, kJ/mol

Interval T1, °C

M01, %

k1, 1/ min

A1, 1/ min

E1, kJ/mol

Interval T2, °C

M02, %

k2, 1/ min

A2, 1/min

E2, kJ/mol

30–210 30–190 30–203 30–250 30–250 30–250

9.97 6.81 16.61 14.48 9.41 8.55

12 13.5 14.25 15.5 12 14

110 500 530 1200 80 200

32.7 36.3 36.7 39.0 33.1 35.3

– – – – – –

– – – – – –

– – – – – –

– – – – – –

– – – – – –

210–550 190–600 203–550 250–750 250–700 250–700

43.23 25.74 28.70 29.76 32.44 23.86

17 19 21 31 15 18

32 25 50 900 1.8 3

46.4 51.0 54.1 78.1 41.4 45.3

30–203 30–180 20–187 30–200 30–197 30–250 30–220 30–217 30–200

3.12 1.74 2.84 2.02 1.57 0.98 1.13 1.55 2.18

19 14 17 11 10 7 9 9 11

0.6 * 106 1100 11500 80 42 1.7 18.6 8.7 70

50.9 38.4 44.1 29.6 26.85 19.1 24.3 23.6 27.7

– 180–285 187–266 200–320 197–370 250–400 220–380 217–370 200–450

– 0.36 0.18 0.39 0.59 0.32 0.29 0.85 0.04

– 25 35 16.5 20 10 11 21 20

– 0.1 * 106 30 * 106 200 300 0.7 0.1 500 400

– 68.6 90.8 44.4 53.7 27.35 29.7 55.0 50.4

203–650 285–650 266–800 320–700 370–700 400–700 380–700 370–700 450–900

22.44 22.74 29.18 18.23 13.67 6.34 10.35 7.59 11.57

47 39 40 47 32.5 31 26 26 25

8 * 106 0.13 * 106 0.8 * 106 6 * 105 1.5 * 103 600 79 65 4

125.8 107.0 103.8 126.6 87.3 84.8 70.2 68.1 63.0

peat it is less expressed. Values of activation energy at fraction 2 obtained by the proposed model are in a good agreement with previously found for bituminous coals (25–120 kJ/mol (e.g., see [34– 36])), lignites (45–200 kJ/mol (e.g. see [37,38])), and for peats (25 – 70 kJ/mol (e.g. see [35,39,40])). As for lignites, as for bituminous coals tested in oxidizing environment (Table 4), the growth of activation energy and rate constant values with temperature (at corresponding fractions) is persistent in almost all considered cases. Such increase is especially seen for bituminous coals, whereas kinetic parameters for fractions 0 and 2 of lignites had congruent values. As for the significant growth of activation energy and rate constants for peat, it might be explained by possible substance ablation during combustion. At fraction 2, for bituminous coals tested in oxidizing environment, activation energy and rate constant are lower in comparison

environment in comparison with the inert one. Moreover, for bituminous coals, it became wider, whereas for lignites and peat – shorter. But, according to previous observations, despite these shifts, the temperature corresponding to maximal decomposition rate (TWmax) for bituminous coals increases in oxidizing environment in comparison with inert one, whereas for lignites and peat it grows smaller. As for fraction 0 intervals, one may see a nonpronounced shift to the lower temperatures. Temperature intervals corresponding to fraction 1 in inert environment varied within close values (200–450 °C). As for experiments in air, such intervals became relatively shorter (200–300 °C). Analysis of data shown in Table 3 (for inert environment) reveals that for all the considered samples we obtain an expected increase of activation energy and rate constant values with temperature (at corresponding fractions). This amplification is especially pronounced for bituminous coals, whereas for lignites and

Table 4 Kinetics of thermal decomposition of solid fossil fuels in oxidizing environment. Sample #

Fraction 0

Fraction 1

Fraction 2

Interval T0, °C

M00,%

k0, 1/ min

A0, 1/ min

E0, kJ/mol

Interval T1, °C

M01, %

k1, 1/ min

A1, 1/ min

E1. kJ/mol

Interval T2, °C

M02. %

k2, 1/ min

A2, 1/ min

E2, kJ/mol

Peat 1 Lignite 2 Lignite 3 Lignite 4 Lignite 5 Bituminous Oxidized 6 Bituminous 7 Bituminous 8

30–203 30–190 30–203 30–220 30–206 30–218

12.24 28.52 20.98 14.49 10.80 14.12

12 14.9 15.5 14 12.5 13

65 700 1150 650 120 150

30.3 37.5 39.2 36.4 32.1 32.6

– – – – – –

– – – – – –

– – – – – –

– – – – – –

– – – – – –

203–400 190–600 203–550 220–750 206–650 218–700

46.43 46.83 47.52 56.52 60.84 51.20

37 10 13.95 12.5 12 14.8

2.9 * 106 0.2 1.9 0.3 0.45 0.75

93.5 25.2 35.5 32.5 30.8 37.2

30–230 30–180

3.33 1.33

15 15

1000 3000

37.7 41.03

– 180–285

– 38

– 5 * 108

– 103.9

230–800 285–800

58.40 65.18

14 14.5

0.29 0.55

35.2 39.7

Bituminous 9

30–187

2.97

16

3250

40.7

187–266

– 0.77 0.45 0.69 0.93 1.28 1.33 0.45 0.45

35

25 * 106

Bituminous 10

30–188

1.38

13

600

36.0

188–296

Bituminous 11

30–197

1.73

9

24

24.2

197–319

Bituminous 12

30–200

0.60

10

34

26.5

200–363

Bituminous 13

30–180

1.23

13

500

32.85

180–328

Bituminous 14

30–217

2.08

10.1

25

25.6

217–331

Anthracite 15

30–227

1.44

12

290

32.4

227–420

89.0

266–800

58.27

23

21

58.5

30

6

2.2 * 10

83.3

296–700

50.95

22

30

61.1

29

0.3 * 106

77.9

319–700

38.4

21

9

56.4

33

8.8 * 106

87.5

363–700

32.03

27.6

83

73.2

6

84.3

328–700

40.05

25.7

26

64.95

33.35

1.2 * 10

11

57

3 * 10

144.5

331–800

47.89

20.5

2.7

52.0

25

5800

67.5

420–900

52.60

25

8

67.5

S.A. Epshtein et al. / Fuel 199 (2017) 145–156

with the corresponding parameters obtained in inert. At the same fraction, for lignites there exist a less pronounced decrease. At fraction 0, for both the lignites and bituminous coals, the kinetic parameters have congruent values at experiments in air and argon. It seems that at fraction 0 there exist no visible correlation between kinetic parameters (for inert and oxidizing environments) and rank. At fraction 2, in experiments in argon, activation energy and rate constants values in general decrease with rank, whereas for experiments in air, there exist slight growth with rank. For bituminous coals tested in inert environment, at fraction 2, decrease of activation energy values with rank growth is clearly seen (Fig. 6a). For experiments in air, we observe a converse situation: along with rank growth, the activation energy values tend to increase. Unfortunately, due to narrow range of vitrinite reflectance index for the studied lignites, for them it is not possible to find such a pronounced relation. Also, relation between activation energy values at fraction 2 for bituminous coals and H/C atomic ratio was found, as shown in Fig. 6b. It could be seen that along with H/C ratio growth, values of activation energy increase for experiments in inert environments, whereas for tests in oxidizing gas flow they are being reduced. As for the bituminous coals with high rank or low H/C (such as ##12–14 and anthracite #15), their activation energies almost coincide for both the environments. As for fraction 1, it should be noted that in the case of experiments in air it denotes a complex of various processes occurring in coals matter: thermal decomposition, oxygen adsorption, etc. Therefore, the kinetic parameters evaluated at such stage (in oxidizing environment) should be considered only as informational ones due to lack of TGA data for separating the effects of such processes. But, in work [17] authors evaluated kinetic parameters for oxygen absorbance stage (corresponding, in our case, to fraction 1), and their results are in good agreements with our data. In the view of above, some additional parameters should be introduced in order to characterize coals interaction with oxygen at low-temperature intervals. To this end, it was convenient to compare TG curves of coals tested in two considered environments. Examples of TG curves of coals tested in inert and oxidizing environments are shown in Fig. 7. Comparative analysis (for each

153

of the considered samples) revealed two groups of coals in context of their interaction with oxygen. The first group included peat, lignites and bituminous oxidized coal #6 (see Fig. 7 a, b and e). It could be characterized by the following: at temperatures over 100 °C oxygen presence enhances volatiles release. This was previously observed by many researchers, including [41,42]. The second group included bituminous coals and anthracite. For this group, at relatively low temperatures (less than 300–400 °C), mass increase is seen during the experiment in air environments in comparison with inert ones (see Fig. 7c, d, f and g). This phenomenon of mass gain is connected presumably with oxygen adsorption or chemisorption, as it was noticed by authors of [14–18,43] and others. Similar difference between TG curves obtained for bituminous coals tested in nitrogen and air environments was shown and discussed in detail in [25,26]. DTG data for the bituminous coals samples also revealed characteristic zones of mass gain, as it was discussed in some works (see, e.g. [16,18,25,26]). Similar observations (mass gain or its absence) were highlighted by researchers in [13]. They connected this phenomenon with fuels reactivity to oxygen and concluded that coals with the largest reactivity do not show any mass gain during TG experiments in air. In order to quantitatively describe these two groups of coals, the following parameter was proposed. We calculated a difference between TG curves in air and argon at the temperature corresponding to beginning of intensive thermal decomposition stage (initial temperature of fraction 2 interval). This parameter, dM (% mas) denotes mass difference that occur between TG curves in inert and oxidizing environments at low-temperature interval prior to intensive thermal decomposition initiation. It could be regarded as a parameter characterizing mass gain or loss at lowtemperature interval. The results are shown in Table 5. Undoubtedly, in order to fully characterize the observed behavioral features of coals interaction with oxygen and to reveal the mechanisms of their oxidation, it is necessary to use simultaneous TGA and DSC (for evaluation of heat release and characterization of sorption processes by exothermic peaks [44,45]) and/or gas analysis (e.g. FTIR [18,24]). But, to our opinion, the values of dM allow to view at the prevailing reaction direction for coals interaction with

Fig. 6. Correlations between fraction 2 activation energy values and a) coals rank; b) aromaticity degree.

154

S.A. Epshtein et al. / Fuel 199 (2017) 145–156

Fig. 7. Comparison of TG curves of coals tested in different environments.

S.A. Epshtein et al. / Fuel 199 (2017) 145–156 Table 5 Values of parameter dM for considered coals. Sample #

dM, mas %

Peat 1 Lignite 2 Lignite 3 Lignite 4 Lignite 5 Bituminous oxidized 6 Bituminous 7 Bituminous 8 Bituminous 9 Bituminous 10 Bituminous 11 Bituminous 12 Bituminous 13 Bituminous 14 Anthracite 15

2.40 21.41 4.25 1.09 2.66 6.06 2.85 1.49 0.53 1.71 1.00 2.28 1.16 0.42 1.01

155

Two groups were allocated for coals in context of their interaction with oxygen. The first group (peat and lignites) is characterized by enhanced volatiles release at temperatures over 100 °C. The second one (bituminous coals and anthracite) contains coals that are prone to oxygen adsorption at low-temperature intervals (up to 300–400 °C). A new parameter was proposed denoting mass gain or loss at low-temperature interval. Acknowledgements The work was supported by the Federal Target Program ‘‘Research and development on priority directions of Russia scientific-technological complex for 2014-2020”, event 1.2. Unique identifier of the project RFMEFI57514X0062. References

oxygen, e.g. oxidative degradation or sorption. For example, for lignites and peat, oxidation processes enhance decomposition (i.e. dM has negative values), whereas bituminous coals tend to oxygen attachment or sorption into their structure (and dM values are positive). Parameter dM could be used as an additional one along with the kinetic parameters evaluated at low- and high-temperature intervals (corresponding to fractions 0 and 2, respectively). Unfortunately, our attempts to find correlations between the proposed parameter dM and rank or aromaticity degree of coals did not give any reliable dependencies. But it could be assumed that parameter dM shall be informative in context of its correlation with heat flow, gas analyses, etc. This requires additional investigations.

4. Conclusions In this work qualitative and quantitative analysis was presented for thermogravimetric data obtained at experiments in inert (argon) and oxidizing (air) environments for a selection of coals with a wide rank range, different structural characteristics, petrographic and elements composition and of different basins and deposits. It was found that for both the considered environments, the coals thermogravimetric curves retain their space disposition order. It was shown that the general pattern of mass change during TG analysis mostly depend on coal rank and aromaticity degree for the experiments in argon and air gas flow. Values of the temperatures corresponding to maximal decomposition rates correlate with coal rank for both environments. For bituminous coals and anthracite, temperatures corresponding to maximal decomposition rate shifted to the zone of higher temperatures in case of experiments in air in comparison with tests in inert. Whereas for lignites they shifted to the zone of lower temperatures. Values of the maximal thermal decomposition rate well correlate with aromaticity degree. For lignites, in most cases, maximal decomposition rates in air and argon have comparable values, whereas for high-rank coals and anthracite they grow in oxidizing environment in comparison with inert one. Kinetic parameters were evaluated for the most common stages of coals mass change during thermogravimetric experiments in air and argon. Kinetic parameters of coals pyrolysis and combustion correlate not only with rank but also with aromaticity degree. At the pyrolysis stage, for bituminous coals, decrease of activation energy values with rank growth and aromaticity degree was found. As for bituminous coals combustion stage (tests in oxidizing environment), there was a slow increase of activation energy values with rank and aromaticity degree.

[1] Khatami R, Stivers C, Levendis YA. Ignition characteristics of single coal particles from three different ranks in O2/N2 and O2/CO2 atmospheres. Combust Flame 2012;159:3554–68. http://dx.doi.org/10.1016/ j.combustflame.2012.06.019. [2] Moron´ W, Rybak W. Ignition behaviour and flame stability of different ranks coals in oxy fuel atmosphere. Fuel 2015;161:174–81. http://dx.doi.org/ 10.1016/j.fuel.2015.08.065. [3] Radovic LR, editor. Coal, energy and fuels in society: analysis of bills and media reports. Boston: McGraw-Hill Primis Custom Publishing; 1997. p. 113–36 [Chapter 7]. [4] Moon C, Sung Y, Ahn S, Kim T, Choi G, Kim D. Thermochemical and combustion behaviors of coals of different ranks and their blends for pulverized-coal combustion. Appl Therm Eng 2013;54:111–9. http://dx.doi.org/10.1016/j. applthermaleng.2013.01.009. [5] Qi G, Wang D, Zheng K, Xu J, Qi X, Zhong X. Kinetics characteristics of coal lowtemperature oxidation in oxygen-depleted air. J Loss Prev Process Ind 2015;35:224–31. http://dx.doi.org/10.1016/j.jlp.2015.05.011. [6] Li W, Zhu Y, Chen S, Zhou Y. Research on the structural characteristics of vitrinite in different coal ranks. Fuel 2013;107:647–52. http://dx.doi.org/ 10.1016/j.fuel.2012.10.050. [7] de la Puente G, Iglesias MJ, Fuente E, Pis JJ. Changes in the structure of coals of different rank due to oxidation—effects on pyrolysis behaviour. J Anal Appl Pyrol 1998;47:33–42. http://dx.doi.org/10.1016/S0165-2370(98)00087-4. [8] Kok MV. Fossil fuels – application of thermal analysis techniques. Handbook of thermal analysis and calorimetry, vol. 2, 2003. p. 371–95. http://dx.doi.org/ 10.1016/S1573-4374(03)80013-4 [Chapter 9]. [9] van Heek KH, Hodek W. Structure and pyrolysis behaviour of different coals and relevant model substances. Fuel 1994;73:886–96. http://dx.doi.org/ 10.1016/0016-2361(94)90283-6. [10] Strezov V, Lucas JA, Strezov L. Experimental and modelling of the thermal regions of activity during pyrolysis of bituminous coals. J Anal Appl Pyrol 2004;71:375–92. http://dx.doi.org/10.1016/S0165-2370(03)00123-2. [11] Vijayakumar CT, Vinayagamoorthi S, Fink JK, Sivasamy P. Characterization of low rank alpine coals: thermogravimetric studies. J Anal Appl Pyrol 2006;76:191–7. http://dx.doi.org/10.1016/j.jaap.2005.11.001. [12] Epshtein SA, Barabanova OV, Minaev VI, Weber J, Shirochin DL. Effect of dimethylformamide treatment of coals on their thermal degradation and elastic-plastic properties. Solid Fuel Chem 2007;41:210–5. http://dx.doi.org/ 10.3103/S0361521907040052. [13] Avila C, Wu T, Lester E. Estimating the spontaneous combustion potential of coals using thermogravimetric analysis. Energy Fuels 2014;28:1765–73. http://dx.doi.org/10.1021/ef402119f. [14] Yi B, Zhang L, Huang F, Xia Z, Mao Z, Ding J, et al. Investigating the combustion characteristic temperature of 28 kinds of Chinese coal in oxy-fuel conditions. Energy Convers Manage 2015;103:439–47. http://dx.doi.org/10.1016/j. enconman.2015.06.053. [15] Li Q, Zhao C, Chen X, Wu W, Li Y. Comparison of pulverized coal combustion in air and in O2/CO2 mixtures by thermo-gravimetric analysis. J Anal Appl Pyrol 2009;85:521–8. http://dx.doi.org/10.1016/j.jaap.2008.10.018. [16] Li B, Chen G, Zhang H, Sheng C. Development of non-isothermal TGA–DSC for kinetics analysis of low temperature coal oxidation prior to ignition. Fuel 2014;118:385–91. http://dx.doi.org/10.1016/j.fuel.2013.11.011. [17] Chen G, Ma X, Lin M, Lin Y, Yu Z. Study on thermochemical kinetic characteristics and interaction during low temperature oxidation of blended coals. J Energy Inst 2015;88:221–8. http://dx.doi.org/10.1016/j. joei.2014.09.007. [18] Zhan J, Wang H, Zhu F, Song S. Analysis on the Governing Reactions in Coal Oxidation at Temperatures up to 400 C. Int J Clean Coal Energy 2014;3:19–28. http://dx.doi.org/10.4236/ijcce.2014.32003. [19] Idris SS, Rahman NA, Ismail K. Combustion characteristics of Malaysian oil palm biomass, sub-bituminous coal and their respective blends via thermogravimetric analysis (TGA). Bioresour Technol 2012;123:581–91. http://dx.doi.org/10.1016/j.biortech.2012.07.065.

156

S.A. Epshtein et al. / Fuel 199 (2017) 145–156

[20] Wang M, Zhao R, Qing S, Liu Y, Zhang A. Study on combustion characteristics of young lignite in mixed O2/CO2 atmosphere. Appl Therm Eng 2017;110:1240–6. http://dx.doi.org/10.1016/j.applthermaleng.2016.09.026. [21] Ma B, Qin A, Li X, Zhao X, He C. Bioinspired design and chitin whisker reinforced chitosan membrane. Mater Lett 2014;120:82–5. http://dx.doi.org/ 10.1016/j.matlet.2014.01.015. [22] Solomon PR, Serio MA, Carangelo RM, Bassilakis R, Yu ZZ, Charpenay S, et al. Analysis of coal by thermogravimetry—fourier transform infrared spectroscopy and pyrolysis modeling. J Anal Appl Pyrol 1991;19:1–14. http://dx.doi.org/10.1016/0165-2370(91)80031-3. [23] MacPhee JA, Giroux L, Charland J-P, Gransden JF, Price JT. Detection of natural oxidation of coking coal by TG-FTIR—mechanistic implications. Fuel 2004;83:1855–60. http://dx.doi.org/10.1016/j.fuel.2004.02.017. [24] Giroux L, Charland J-P, MacPhee JA. Application of Thermogravimetric Fourier Transform Infrared Spectroscopy (TGFTIR) to the Analysis of Oxygen Functional Groups in Coal. Energy Fuels 2006;20:1988–96. http://dx.doi.org/ 10.1021/ef0600917. [25] Zhang Y, Wu J, Chang L, Wang J, Xue S, Li Z. Kinetic and thermodynamic studies on the mechanism of low-temperature oxidation of coal: a case study of Shendong coal (China). Int J Coal Geol 2013;120:41–9. http://dx.doi.org/ 10.1016/j.coal.2013.09.005. [26] Zhang Y, Wang J, Xue S, Wu Y, Li Z, Chang L. Evaluation of the susceptibility of coal to spontaneous combustion by a TG profile subtraction method. Korean J Chem Eng 2016;33:862–72. http://dx.doi.org/10.1007/s11814-015-0230-8. [27] Kaminskii VA, Epshtein SA, Shirochin DL, Timashev SF. The determination of kinetic parameters for the decomposition of complex substances from thermogravimetry data. Russ J Phys Chem A 2011;85:561–7. http://dx.doi. org/10.1134/S0036024411030137. [28] Ito O, Seki H, Iino M. Diffuse reflectance spectra in near-i.r. region of coals; a new index for degrees of coalification and carbonization. Fuel 1988;67:573–8. http://dx.doi.org/10.1016/0016-2361(88)90357-2. [29] Maroto-Valer MM, Love GD, Snape CE. Relationship between carbon aromaticities and HC ratios for bituminous coals. Fuel 1994;73:1926–8. http://dx.doi.org/10.1016/0016-2361(94)90224-0. [30] Maroto-Valer MM. Verification of the linear relationship between carbon aromaticities and H/C ratios for bituminous coals. Fuel 1998;77:783–5. http:// dx.doi.org/10.1016/S0016-2361(97)00227-5. [31] Korovushkin VV, Epshtein SA, Durov NM, Dobryakova NN. Mineral and valent forms of iron and their effects on coals oxidation and self-ignition. Gornyi Zhurnal 2015:70–4. http://dx.doi.org/10.17580/gzh.2015.11.14. [32] Arenillas A, Rubiera F, Pis JJ. Simultaneous thermogravimetric–mass spectrometric study on the pyrolysis behaviour of different rank coals. J Anal Appl Pyrol 1999;50:31–46. http://dx.doi.org/10.1016/S0165-2370(99)000248.

[33] Abbasi-Atibeh E, Yozgatligil A. A study on the effects of catalysts on pyrolysis and combustion characteristics of Turkish lignite in oxy-fuel conditions. Fuel 2014;115:841–9. http://dx.doi.org/10.1016/j.fuel.2013.01.073. [34] Folgueras MB, Díaz RM, Xiberta J. Pyrolysis of blends of different types of sewage sludge with one bituminous coal. Energy 2005;30:1079–91. http://dx. doi.org/10.1016/j.energy.2004.08.001. [35] Sutcu H. Pyrolysis by thermogravimetric analysis of blends of peat with coals of different characteristics and biomass. J Chin Inst Chem Eng, 2007;38:245–9. http://dx.doi.org/10.1016/j.jcice.2007.03.002. [36] Ferrara F, Orsini A, Plaisant A, Pettinau A. Pyrolysis of coal, biomass and their blends: performance assessment by thermogravimetric analysis. Bioresour Technol 2014;171:433–41. http://dx.doi.org/10.1016/j.biortech.2014.08.104. [37] Haykiri-Acma H, Yaman S. Synergy in devolatilization characteristics of lignite and hazelnut shell during co-pyrolysis. Fuel 2007;86:373–80. http://dx.doi. org/10.1016/j.fuel.2006.07.005. [38] Zhou Z, Hu X, You Z, Wang Z, Zhou J, Cen K. Oxy-fuel combustion characteristics and kinetic parameters of lignite coal from thermogravimetric data. Thermochim Acta 2013;553:54–9. http://dx.doi.org/ 10.1016/j.tca.2012.11.030. [39] Kim JK, Lee HD, Kim HS, Park HY, Kim SC. Combustion possibility of low rank Russian peat as a blended fuel of pulverized coal fired power plant. J Ind Eng Chem 2014;20:1752–60. http://dx.doi.org/10.1016/j.jiec.2013.08.027. [40] Huang X, Rein G. Smouldering combustion of peat in wildfires: inverse modelling of the drying and the thermal and oxidative decomposition kinetics. Combust Flame 2014;161:1633–44. http://dx.doi.org/10.1016/ j.combustflame.2013.12.013. [41] Korotkikh AG, Slyusarskiy KV, Larionov KB, Osipov VI. Comparison of coal reactivity during conversion into different oxidizing medium. J Phys: Conf Ser 2016;754:52005. http://dx.doi.org/10.1088/1742-6596/754/5/052005. [42] Jayaraman K, Gokalp I. Thermogravimetric and evolved gas analyses of high ash Indian and Turkish coal pyrolysis and gasification. J Therm Anal Calorim 2015;121:919–27. http://dx.doi.org/10.1007/s10973-015-4500-9. [43] van der Plaats G, Soons H, Chermin HAG. Low-temperature oxidation of coal. Thermochim Acta 1984;82:131–6. http://dx.doi.org/10.1016/0040-6031(84) 87281-0. [44] Rotaru A. Thermal analysis and kinetic study of Petrosßani bituminous coal from Romania in comparison with a sample of Ural bituminous coal. J Therm Anal Calorim 2012;110:1283–91. http://dx.doi.org/10.1007/s10973-0112022-7. [45] Slovák V, Taraba B. Effect of experimental conditions on parameters derived from TG-DSC measurements of low-temperature oxidation of coal. J Therm Anal Calorim 2010;101:641–6. http://dx.doi.org/10.1007/s10973-010-0878-6.