Behavior and Kinetics of Non-isothermal Pyrolysis of Coal at Different Heating Rates

Fan Junfeng; Tian Bin,; An Xiaoxi; Yin Mengmeng; Qiao Yingyun; Tian Yuanyu,

(1. Key Laboratory of Low Carbon Energy and Chemical Engineering, Shandong University of Science and Technology, Qingdao 266590;2. State Key Laboratory of Heavy Oil Processing, China University of Petroleum, Qingdao 266580)

Behavior and Kinetics of Non-isothermal Pyrolysis of Coal at Different Heating Rates

Fan Junfeng1; Tian Bin1,2; An Xiaoxi1; Yin Mengmeng1; Qiao Yingyun2; Tian Yuanyu1,2

(1. Key Laboratory of Low Carbon Energy and Chemical Engineering, Shandong University of Science and Technology, Qingdao 266590;2. State Key Laboratory of Heavy Oil Processing, China University of Petroleum, Qingdao 266580)

Investigations on the pyrolysis and kinetic behaviors during pyrolysis offossil fuel samples, such as coal, are fundamental for developing the related technology and practical application. In this work, pyrolysis behavior and kinetics in the pyrolysis process of a coal sample were investigated by the thermogravimetric analyzer at a heating rate of 10, 40, 70,100, 200, and 500 °C/min, respectively. The pyrolysis process could be divided into four stages according to the TG/DTG curves. The first stage was mainly attributed to the removal of volatiles, and the second peak was the polycondensation reaction between the volatile components. It was found that more volatiles could be released from coal during pyrolysis at higher heating rate and the higher pyrolysis temperature was necessary for a rapid pyrolysis process. Kinetic analysis revealed that both the model-free (KAS and FWO) and the Coats-Redfern methods were suitable to describing the pyrolysis process, and the variation of activation energy (E) with the two types of kinetic models was consistent during the whole pyrolysis process. Furthermore, the heating rate did not alter the reaction sequence of the whole pyrolysis process, but higher heating rate could make the E value of the initial pyrolysis stage smaller.

pyrolysis behavior; kinetic analysis; TG/DTG curves; activation energy

1 Introduction

China is rich in coal reserves, but short in oil reserves along with less gas resources. Therefore, coal plays an important role in China’s energy structure, and it is expected that the coal-based energy structure in China will not change for a long period of time. Currently,coal is mainly used for combustion to produce power and serious environmental problems are caused by huge combustion products emissions. In this context,the development of clean and efficient coal utilization technology is extremely crucial in changing the energy structure[1].

Pyrolysis is a simple and an effective technology to utilize coal under mild conditions[2], and many fine chemical reagents can be obtained from the pyrolysis reaction. The pyrolysis tar is generally used in the preparation offuel oil through hydrogenation[3].

Coal pyrolysis refers to the complex process of treating coal in a series of physical and chemical reactions at different temperatures in an isolated air or inert atmosphere. Many high-quality activated semi-coke,low-temperature or medium-temperature coal tar and high calorific value gas can be produced by the pyrolysis process, from which a lot of raw materials and fuel can be applied by the industry. For example, the high-quality bituminous coals are usually used as raw materials to produce the syngas, hydrogen and industrial or civil gas[4]. However, the gas or liquid products formed thereby are generally derived from the coal volatiles, which are closely related to the types of coal and its chemical structures. In addition, the decomposition reaction can provide a basic information concerning the structural features of bituminous coal[5]. As some researchers have reported, the pyrolysis temperature and heating rate are dominant factors in the pyrolysis process as comparedwith the pyrolysis pressure, gas and coal particle size.Tian, et al. reported that the performance for release of gaseous products and devolatilization profiles are significantly influenced by the heating rate[6]. K?k used the thermogravimetric method to investigate the effect of particle size on coal pyrolysis and the thermal resistance of the coal particles, and it was found out that the peak temperature at the maximum weight loss rate gradually increased with an increasing particle size[7]. Therefore, a deep understanding of reaction mechanism and chemical kinetics of coal pyrolysis can help to improve the conversion efficiency and provide fundamental data for designing the scale-up practice of industrial application.Thermogravimetric (TG) analysis has been widely used to investigate the characteristic parameters of the decomposition, as well as the pyrolysis kinetics under either the inert gas or the air atmosphere conditions[8].The kinetic analysis is a basic process in coal pyrolysis technology development. Investigation on the factors influencing the coal pyrolysis can help researchers completely understand the pyrolysis process. In addition,the design and simulation of the pyrolysis process also require the accurate and reliable kinetic parameters.Du, et al. predicted the influence of heating rate on the reaction rate constant of pulverized coal by the widely adopted amended Arrhenius equation. The calculated results presented a better predictive ability for the pyrolysis kinetics and better extrapolation reliability in a wide range of the study[9]. Zhang, et al. studied the kinetic parameters and used the solid heat carrier to investigate the characteristics for release of typical volatile compounds in the pyrolysis process[10]. Moreover, the basic investigations on the fast coal pyrolysis process can better delve into the actual coal utilization technology.Although there are many research results in coal pyrolysis, the investigation on the coal pyrolysis behavior and the kinetic characteristics, especially under the fast heating conditions are still insufficient.

In this work, TG experiments were conducted on a bituminous coal at different heating rates to investigate the process and kinetics of pyrolysis. The influence of heating rate (at 10 °C/min, 40 °C/min, 70 °C/min, 100°C/min, 200 °C/min, and 500 °C/min, respectively,) on the variation in weight loss and pyrolysis characteristic parameters was investigated in detail during the nonisothermal pyrolysis process. After that, the kinetic analysis was also investigated by the Coats-Redfern method, the Kissinger-Akahira-Sunose method and the Flynn-Wall-Ozawa method. All the above investigation work can provide the basic data and theoretical reference for the heat transfer of coal pyrolysis process, which was beneficial to better understanding of the pyrolysis behavior and kinetics for the rapid coal pyrolysis at high temperatures.

2 Experimental

2.1 Material and apparatus

The coal sample studied in this work was collected from the Yichun coal mine in Heilongjiang Province. The coal sample was crushed and sieved to a particle size of smaller than 74 μm and then dried at 110 °C for 5 h prior to use. A TG analyzer (STA449F3, Netzsch-Ger?tebau GmbH, Germany) was applied to determine the proximate analysis and carry out the pyrolysis experiments.Meanwhile, an elemental analyzer (Vario Macro cube,Elementar Corp., Germany) was used to conduct the ultimate analysis. The properties of the coal sample are listed in Table 1.

Table 1 The proximate and ultimate analyses of Yichun coal w, %

2.2 Experimental method

The high-purity nitrogen (99.999%) was introduced into the TG system at a flow rate of 50 mL/min, in an attempt to keep an inert gas environment. In the TG analyzer,the coal sample (5 mg) was added into the analyzer and then heated from the room temperature to 1 000 °C at a heating rate of 10 °C/min, 40 °C/min, 70 °C/min, 100 °C/min, 200 °C/min, and 500 °C/min, respectively. The mass loss of the sample was recorded during pyrolysis of the coal sample.

2.3 Pyrolysis kinetic methods

Pyrolysis kinetics is an important method to analyze a series of physical and chemical reactions taking place during the coal pyrolysis process. The thermodynamic parameters, activation energy and frequency factor of the pyrolysis process can be obtained according to the nonisothermal thermogravimetric curve and the reaction kinetic equation. Meanwhile, the activation energy can reflect the difficulty of the pyrolysis reaction. The Coats-Redfern method and the Horowitz-Metzger method are adopted to analyze the kinetics of coal pyrolysis at different heating rates, which is simple and accurate.Thus, the above two methods are adopted to describe the pyrolysis process of coal sample in this work. The Coats-Redfern method equation is presented as follows[11]：

where the mass fractional conversion at one point is expressed as：

There is a linear relationship between pyrolysis temperature T and time t during the non-isothermal TG experiment conducted at a constant heating rate, and the pyrolysis temperature is expressed as T=T0+βt (in which β is the heating rate, °C /min). By substituting this relationship into Equation (1), the rearranged new results are presented as：

After integration of both sides in Eq. (3), the integral expressions can be expressed as：

After taking the integration and logarithm of both sides in Eq. (4), the result can be expressed as：

The 2RT/E term can be approximated to zero, since the E values in Eq. (5) and (6) are very large. If the reaction order is made right, the left side in Eq. (4) or (5) has a linear relation with 1/T. Finally, the activation energy E and the pre-exponential factor A can be obtained,respectively, according to the slope and intercept of the plotted straight line.

Secondly, the Kissinger-Akahira-Sunose method is presented as follows[12]：

Thirdly, the Flynn-Wall-Ozawa method is presented as follows[13]：

3 Results and Discussion

3.1 Pyrolysis behavior

The TG pro files and the corresponding DTG (derivative thermogravimetry) of coal sample tested at different heating rates as a function of temperature are shown in Figure 1. During the pyrolysis process, nearly 36.5%of mass loss were observed for a heating rate of 500°C/min at an end pyrolysis temperature of 1 000 °C.Theoretically, thermal decomposition of coal can take place when the temperature reaches the initial pyrolysis temperature. However, the pyrolysis process is exceedingly complicated and many competing reactions can contribute to the mass loss curves, while the coal samples from different sources demonstrate different pyrolysis behaviors. Therefore, the temperature at the maximum decomposition rate (Tmax) and mass loss in two main pyrolysis regions during pyrolysis at different heating rates are determined and shown in Table 2.

Figure 1 TG/DTG curves of coal sample at different heating rates

As shown in Figure 1 and Table 2, the DTG curves for the coal sample at different heating rates displayed similar mass loss behavior, which presented two main pyrolysis regions. For the first region in the pyrolysis process, the DTG peak was identified at 453 °C for the heating rate of 10 °C/min, at 487 °C for the heating rate of 40 °C/min, at 500 °C for the heating rate of 70 °C /min, at 508 °C for the heating rate of 100 °C /min, at 518 °C for the heating rate of 200 °C /min, and at 554 °C for the heating rate of 500 °C /min, respectively. The first decomposition peak was mainly attributed to the removal of volatiles from coal. Moreover, when the heating rate increased from 10 °C/min to 500 °C/min, the increased mass loss of coal at the highest temperature was equal to 12.49%, 12.33%,12.75%, 13.28%, 13.30%, and 14.64%, respectively. In other words, more volatiles could be released from coal during pyrolysis at the high heating rate region. This phenomenon can be used in the reactor design of practical coal pyrolysis process, during which the rapid heating of fresh coal sample should be considered. The second peak was the polycondensation reaction between the volatile components of the coal sample. Furthermore, the mass loss at this stage was also attributed to the decomposition of mineral species, such as calcite[14]. However, since the Yichun coal contained less mineral species (7.36% of ash content), the decomposition of minerals gave limited contribution to the mass loss. In the secondary pyrolysis region, the DTG peak was identified at 738 °C for the heating rate of 10 °C/min, at 789 °C for the heating rate of 40 °C/min, at 819 °C for the heating rate of 70 °C/min,at 836 °C for the heating rate of 100 °C/min, at 866 °C for the heating rate of 200 °C/min, and at 893 °C for the heating rate of 500 °C/min, respectively. The above experimental results indicated that the decomposition temperature changed to the high temperature region with the increase in heating rate. Therefore, the pyrolysis mechanism had changed to a certain extent when the slow pyrolysis process was gradually transformed into a fast pyrolysis process. This could occur because the mass and heat transfers were almost constant and independent on the heating rate. This is another important fundamental phenomenon that can be applied in the practical industrial process. In a word, the industrial pyrolysis reactor should operate at a rapid heating rate to obtain more volatiles, but it needs higher pyrolysis temperature as compared with the case of slow pyrolysis process. Moreover, when the heating rate increased from 10 °C/min to 500 °C/min, the decreased mass loss of coal at the highest temperature was equal to 30.37%, 29.90%, 30.75%, 31.86%, 32.52%, and 33.20%,respectively, meaning that the second pyrolysis region could release more volatiles than the first pyrolysis region. Meanwhile, a larger maximum derivative of mass loss could be observed when the heating rate increased.This phenomenon was closely related to the poor thermal conductivity of coal[15].

Table 2 Tmax value and mass loss in two main pyrolysis regions obtained at different heating rates

3.2 Kinetics of pyrolysis process

3.2.1 Coats-Redfern method

In this work, a popular three-dimensional diffusion(spherical symmetry) was adopted to determine the pyrolysis mechanism of coal sample. It can be described as follows：

During the slow pyrolysis process (at the heating rate of 10 °C/min), the calculated E value of four stages was 55.07 kJ/mol, 91.49 kJ/mol, 16.19 kJ/mol, and 64.94 kJ/mol, respectively. Different activation energy may represent different reaction mechanisms and the reaction difficulties. The lower the E is, the easier the reaction would be. The cleavage of weak bonds and the release of the trapped small molecular phase in coal can take place before 416 °C, and these reactions can readily proceed[3].Therefore, the E value at the first pyrolysis stage is relatively low. The main pyrolysis stage of coal is often in the temperature range of 416—508 °C, and a large number of bonds, including the aliphatic C—C, C—H,and C—O bonds, will break down, leading to the formation of volatiles and gases[16]. Since the pyrolysis reaction in this stage is very drastic and violent, the E is also the largest among the four pyrolysis stages. However,the pyrolysis reactions become slow and the number of bond scission in the same time duration decreases at temperatures exceeding 500 °C so that the E becomes small again. At the final pyrolysis stage in the temperature range of 697—770 °C, carbonates decomposition and aromatic ring condensation in coal become dominative.These reactions are commonly associated with the relatively high bond energy molecules, in which the E value is greater than the former pyrolysis stage.

Interestingly, the trend for variation of E in the four pyrolysis stages was consistent under every heating rate condition, revealing that the heating rate did not alter the reaction sequence of the whole pyrolysis process.Furthermore, only E in the first pyrolysis stage decreased obviously with an increasing heating rate and E in other pyrolysis stages was not dependent on the heating rate.It is well known that coal pyrolysis reactions are mainly related to the bond cleavage and also some free radical reactions.Rapid heating can provide sufficient heat on coal sample in a very short time, the chemical bonds with weak bond energies will be cracked down simultaneously that do not follow the order of bond energies. Therefore, E at the first pyrolysis stage becomes smaller with an increasing heating rate. However, once the pyrolysis reaction starts,the radical reactions become more intensive and these reactions are not affected by the heating rate, and the E at other pyrolysis stages is not dependent upon the heating rate.

Figure 2 Analysis of the kinetic curves at different heating rates using the Coats-Redfern method

Table 3 Kinetic parameters at different heating rates determined by the Coats-Redfern method

3.2.2 KAS and FWO methods

The plots of the Kissinger-Akahira-Sunose (KAS) and the Flynn-Wall-Ozawa (FWO) methods are shown in Figures 3 and 4, respectively. In addition, the distribution of the activation energy values calculated from the experimental results at different conversion rates are shown in Figure 5.For the two methods (KAS and FWO), the shapes of the activation energy are nearly the same. It can be divided into four parts clearly, the activation energy in the first part, α ＜ 0.2, increased rapidly. Then, in the second part,0.2 ＜ α ＜ 0.5, the activation energy increased slowly. In the third part, 0.5 ＜ α ＜ 0.8, the activation energy increased rapidly again and reached a highest value. After that,in the fourth part, 0.8 ＜ α ＜ 0.9, the activation energy decreased rapidly. These results were also consistent with the TG/DTG curves. The advantage of the KAS and FWO methods is that it can re flect the activation energy distribution to more objectively re flect the real process.

Figure 3 The plots of the Kissinger-Akahira-Sunose method

Figure 4 The plots of the Flynn-Wall-Ozawa method

The iso-conversional or mode-free methods can obtain the kinetic parameters (activation energy) using TGA data without the explicit model for concentration dependence.Since the KAS and FWO kinetic models are all similar methods of this type, obtaining the activation energy needs the experimental data from at least three different heating rates. In this work, four heating rates ranging from 10 °C/min to 200 °C/min were incorporated. The variations of E from the model-free methods and the Coats-Redfern method were consistent, implying that these kinetic models were suitable for describing the coal pyrolysis process, especially at higher heating rate regions. The E values of different pyrolysis stages calculated from model-free methods were greater than those of the C-R method. Similar results were also observed by Jain, et al.[17], when using a high-ash coal as the pyrolysis feedstock. It is recommended that TGA data should first be processed using various isoconversional methods to acquire E vs. α dependence, and then the model-fitting kinetic analysis, such as the C-R method,could be applied[18]. The C-R method is derived from simplification of the mathematic models and cannot entirely represent the real reaction mechanism, since the coal is an extremely complex material[19]. Thus, the C-R and the model-free methods give inconsistent E values at different pyrolysis stages. However, the C-R method can at least provide the most probable reaction mechanism and obtain E and A simultaneously for the pyrolysis process that cannot be estimated by the model-free methods.

Furthermore, E calculated from the KAS and the FWO methods using high heating rate data in this work were greater than that using the low heating rate data with the same kinetic method. Values of E varied from 150 kJ/mol to 280 kJ/mol when the pyrolysis conversion increased from 0.15 to 0.90 at the present study. However, an E range of 90 kJ/mol—278 kJ/mol was observed for the Shenmu long flame coal when using the FWO method with a heating rate of 5 °C/min, 10 °C/min, and 15 °C/min,respectively[20]. In another example[21], the E value was determined in a range of 50 kJ/mol—200 kJ/mol for a Ningdong coking coal during pyrolysis at a heating rate of 5 °C/min, 10 °C/min, and 15 °C/min, respectively. Higher E value observed in this work might be ascribed to more amounts of covalent bonds that were broken down in a shorter period of time during the rapid heating process,leading to the needs for more energies to overcome the energy barrier.

Figure 5 Activation energy calculated by the Kissinger-Akahira-Sunose and the Flynn-Wall-Ozawa methods

4 Conclusions

The pyrolysis behavior and kinetics in the pyrolysis process of coal were investigated by the thermogravimetric analyzer at a heating rate of 10 °C/min, 40 °C/min,70 °C/min, 100 °C/min, 200 °C/min, and 500 °C/min,respectively, in this work. The pyrolysis process of coal sample was divided into four parts. According to the TG/DTG curves, the first peak was mainly attributed to the removal of volatiles, and the second peak was ascribed to the polycondensation reaction between the volatile components. It is found that more volatiles could be released from coal during pyrolysis in the high heating rate region and a higher pyrolysis temperature was necessary to achieve a rapid pyrolysis process. Kinetic analysis reveals that both the model-free methods (KAS and FWO) and the Coats-Redfern method were suitable for describing the pyrolysis process, and the variation in E value between the two types of kinetic models was consistent during the whole pyrolysis process. Furthermore, the heating rate did not alter the reaction sequence of the whole pyrolysis process, but higher heating rate could make the E value of the initial pyrolysis stage smaller.

Acknowledgements: The authors gratefully acknowledge the financial support from the National Natural Science Foundation of China (Grant Nos. 21576293 and 21576294).

[1] Xue J, Wu Y, Cui Q, et al. Feasibility analysis and parameter optimization of organic steam to drive a MED system dealing with high salinity waste water[J].Environmental Science： Water Research & Technology,2017, 3： 546-552

[2] Baruah BP, Khare P. Pyrolysis of high sulfur Indian coals[J]. Energy Fuels, 2007, 21(6)： 3346-3352

[3] Tian B, Qiao Y, Bai L, et al. Pyrolysis behavior and kinetics of the trapped small molecular phase in a lignite[J]. Energy Convers Manage, 2017, 140： 109-120

[4] Cui G, Lin M. Gasification characteristics and kinetics of briquette using humic acid binder[J]. Journal of fuel Chemistry and Technology, 2012, 40(11)： 1289-1294

[5] Wiktorsson LP, Wanzl W. Kinetic parameters for coal pyrolysis at low and high heating rates—a comparison of data from different laboratory equipment[J]. Fuel, 2000,79(6)： 701-716

[6] Tian B, Qiao YY, Tian YY, et al. Investigation on the effect of particle size and heating rate on pyrolysis characteristics of a bituminous coal by TG-FTIR[J]. J Anal Appl Pyrolysis, 2016, 121： 376-386

[7] Versan K?k M, ?zbas E, Karacan O, et al. Effect of particle size on coal pyrolysis[J]. J Anal Appl Pyrolysis,1998, 45(2)： 103-110

[8] Ren Q, Zhao C, Wu X, et al. TG-FTIR study on copyrolysis of municipal solid waste with biomass[J].Bioresour Technol, 2009, 100(17)： 4054-4057

[9] Du R L, Wu K, Xu D A, et al. A modified Arrhenius equation to predict the reaction rate constant of Anyuan pulverized-coal pyrolysis at different heating rates[J]. Fuel Processing Technology, 2016, 148： 295-301

[10] Zhang Y Q, Zhu J L, Wang X H, et al. Simulation of large coal particles pyrolysis by circulating ash heat carrier toward the axial dimension of the moving bed[J]. Fuel Process Technol, 2016, 154： 227-234

[11] Zhang S, Zhu F, Bai C, et al. Thermal behavior and kinetics of the pyrolysis of the coal used in the COREX process[J]. Journal of Analytical and Applied Pyrolysis,2013, 104(112)： 660-666

[12] Starink MJ. The determination of activation energy from linear heating rate experiments： A comparison of the accuracy of isoconversion methods[J]. Thermochim Acta,2003, 404： 163-176

[13] Sharara MA, Holeman N, Sadaka SS, et al. Pyrolysis kinetics of algal consortia grown using swine manure wastewater[J]. Bioresour Technol, 2014, 169： 658-666

[14] Zhao H L, Bai Z Q, Yan J C, et al. Transformations of pyrite in different associations during pyrolysis of coal[J].Fuel Process Technol, 2015, 131： 304-310

[15] Liang P, Zhang Y, Jiang W, et al. Simulation study of Shenmu coal pyrolysis by gas heat carrier based on a moving bed[J]. Energy Fuels, 2015, 29(11)： 7727-7733

[16] Shi L, Liu Q, Guo X, et al. Pyrolysis behavior and bonding information of coal—a TGA study[J]. Fuel Process Technol, 2013, 108： 125-132

[17] Jain A A, Mehra A, Ranade V V. Processing of TGA data：Analysis of isoconversional and model fitting methods[J].Fuel, 2016, 165： 490-498

[18] Vyazovkin S, Burnham A K, Criado J M, et al. ICTAC Kinetics Committee recommendations for performing kinetic computations on thermal analysis data[J].Thermochim Acta, 2011, 520(112)： 1-19

[19] White J E, Catallo W J, Legendre B L. Biomass pyrolysis kinetics： A comparative critical review with relevant agricultural residue case studies[J]. J Anal Appl Pyrolysis,2011, 91(1)： 1-33

[20] Lin Y, Li Q, Li X, et al. Pyrolysates distribution and kinetics of Shenmu long flame coal[J]. Energy Convers Manage, 2014, 86： 428-434

[21] Feng S, Li P, Liu Z, et al. Experimental study on pyrolysis characteristic of coking coal from Ningdong coal field[J].Journal of the Energy Institute, in press

date: 2017-05-22; Accepted date: 2017-07-09.

Professor Tian Yuanyu, Telephone：+86-532-86057766; E-mail： tianyy1008@126.com.