Pyrite transformation and sulfur dioxide release during calcination of coal gangue

Abstract

Calcination is a typical process associated with the utilization of coal gangue. A concern associated with this procedure is the emission of sulfur dioxide (SO₂). In this work, the behavior of SO₂ release during coal gangue calcination under an air atmosphere were systematically investigated and compared to the characteristics of SO₂ evolution from pure pyrite calcination. Results show that although sulfur in coal gangue mainly exists in the form of pyrite, it represents different transformation behaviors from that in pure mineral pyrite. At 500 °C, the release rate of SO₂ is significantly higher in coal gangue than in mineral pyrite due to the fact that coal gangue combustion can occur at low temperatures, which favors the SO₂ release, while at 600 and 700 °C they become almost the same. The shrinking core model cannot describe the SO₂ emission profiles in coal gangue and, instead, a hybrid 3D diffusion, i.e. the Jander model, is successfully developed in this study.

---

1. Introduction

With the increasing environmental and economic burden of waste disposal, utilization of waste materials as an alternative for declining fossil fuel and mineral resources has gained considerable interest worldwide. However, attention should be paid to the potential secondary pollution generated during this utilization, as it may constrain the development of the utilization and pose new challenges to the environment protection.¹ Coal gangue is a problematic solid waste discharged from coal mining and beneficiation.^2–4 Substantial quantities of coal gangue were dumped in every coal-producing country, causing a series of environmental problems, such as acid drainage, heavy metals leaching as well as atmospheric pollution.^5–7 In China, there is still about 659 million tons of coal gangue produced each year.⁸ As coal gangue has high contents of silica and alumina, it finds application as a substitute for clay in the construction industry.^9,10 The amount of coal gangue used in the construction industry was up to 50 million tons in 2011 in China.⁸

Coal gangue utilized in the construction industry is generally subjected to a calcination process. As coal gangue contains sulfur derived from both coal components and minerals, the emission of SO₂ during calcination is one of the major environmental concerns. With increasing environmental awareness, recent studies have been carried out on the emission of trace elements of coal gangue during combustion and brick making.^3,11,12 However, much less is known about the SO₂ emission during calcination of coal gangue. Sulfur in coal gangue can exist in both organic and inorganic forms. The forms of organic sulfur are rarely reported and their amounts are likely small. For inorganic sulfur compounds, pyrite is a frequently observed and abundant inclusion. The SO₂ emission of coal gangue during calcination is expected to be highly related to the transformation behavior of embedded pyrite. As coal gangue contains both combustible matter and large fractions of mineral species, including kaolinite, quartz and calcite, pyrite may interact with these substances during calcination. Thus, the embedded pyrite in coal gangue may exhibit a distinct transformation behavior and SO₂ emission profile compared with individual coal or pyrite. With increasingly stringent regulations on pollution emissions, investigations regarding the SO₂ formation mechanism and release behavior during coal gangue calcination are imperative and desirable for the sustainable development of coal gangue utilization.

In this work, the SO₂ release behaviors and pyrite transformation of coal gangue were investigated in comparison with pure pyrite. Compared with the widely used thermal analysis method, a tube furnace was employed to achieve the high flow rate and obtain effective gas–solid contact, which are rather close to the practical conditions. The effects of reaction temperatures, the reaction mechanism and the kinetics of pyrite transformation in coal gangue during calcination were also discussed.

2. Materials and methods

2.1. Materials

The two coal gangue samples used in the present investigation, denoted as CG1 and CG2, were collected from two centralized coal waste dumps in Shanxi province, China. The samples were crushed and sieved to a particle size less than 150 μm before use. Proximate analysis was carried out by a thermogravimetric analyzer (TGA-701, LECO) and ultimate analysis was performed by an elemental analyzer (vario Macro CHNS, Elementar). Analyses of the total sulfur and sulfur forms were conducted according to the Chinese standards GB/T 214 ¹³ and GB/T 215,¹⁴ respectively. The chemical compositions were determined by an X-ray fluorescence spectrometer (S4-Explorer, Bruker). A standard mineral pyrite sample (GBW07267, Fe 48.08 ± 0.29 wt%, S 52.72 ± 0.21 wt%) was purchased from Aikong biological technology (Beijing, China), Ltd. The air (purity > 99%) used was provided by Qianxi gas company (Beijing, China), Ltd.

2.2. Calcination and emission experiments

The calcination experiments of coal gangue samples and mineral pyrite were carried out using a vertical tube furnace and the release of SO₂ was recorded simultaneously by a flue gas analyzer (Testo pro350, Testo) at regular intervals of 2 s. The experimental setup is illustrated in Fig. 1. Note that both non-isothermal and isothermal experiments were conducted. During the non-isothermal experiments, the samples were placed into the furnace at ambient temperature and heated up to 1200 °C at a heating rate of 5 °C min⁻¹. During the isothermal experiments, the furnace was first heated up to a constant temperature (500 °C, 600 °C and 700 °C), and then the samples were quickly loaded into the furnace. Coal gangue samples of 250.0 mg and pure pyrite of 10.0 mg, which has an equivalent sulfur content to that of CG2, were used. The non-isothermal and isothermal measurements were both performed under an air atmosphere at a flow rate of 1.2 L min⁻¹. To investigate the mineral phase transformation, 200.0 mg pyrite was calcined under the same conditions. A corundum crucible (Tangshan industrial ceramic plant, China) with a depth of 1 cm and a diameter of 6 cm was used; therefore, all samples can be well distributed within a thin layer to improve the gas–solid contact.

Fig. 1 The schematic diagram of the vertical tube furnace experimental system. 1 – gas cylinder; 2 – flow meter; 3 – tube furnace; 4 – thermocouple; 5 – ceramic crucible; 6 – gas analyzer; 7 – computer.

2.3. Thermal analysis

The thermogravimetric (TGA), derivative thermogravimetric (DTG) and differential scanning calorimetric (DSC) curves were measured using a thermal analyzer (Q600SDT, Thermal analysis) with 10 mg sample inside the alumina crucible. The analyses were performed with an air flow rate of 100 ml min⁻¹ and a linear heating rate of 10 K min⁻¹ from room temperature to 1200 °C.

2.4. X-ray diffraction (XRD)

The XRD analyses of coal gangue samples, mineral pyrite and their residues after combustion were run on a Rigaku D/max 2500 PC X-ray diffractometer with Cu Kα radiation (λ = 0.15406 nm) and operating at 40 kV and 100 mA. The scanning range of 2θ was 10°–60° with an increment of 0.02° under the scanning speed of 2° min⁻¹.

3. Results and discussion

3.1. Characterization of coal gangue

The chemical compositions of CG1 and CG2 are listed in Table 1. It shows that sulfur in coal gangue is predominately in pyritic forms, accounting for 78.8% and 89.0% of the total sulfur mass in CG1 and CG2, respectively. The mass fractions of organic sulfur to total sulfur in coal gangue are relatively low, only 15.2% in CG1 and 7.7% in CG2. The distributions of different types of sulfur in coal gangue are clearly different from those reported in coal. For instance, Gornostayev et al.¹⁵ reported that the fraction of pyritic sulfur varied considerably from 3% to 63% in coals, and organic sulfur could occupy a significant fraction of the total sulfur. This difference might be ascribed to the compositional features of coal gangue, which contains low level of organics but a high content of ash. The dominance of pyritic sulfur in coal gangue confirms that pyritic sulfur or pyrite plays a major role in the subsequent SO₂ formation during calcination of coal gangue.

Table 1 Proximate and ultimate analysis of coal gangue

	CG1	CG2
a ad, air dried.b d, dried.
Proximate analysis (wt%)
Moisture, ad^a	1.27	0.87
Ash, d^b	64.96	74.01
Volatile matter, d	16.36	11.12
Fixed carbon, d	18.68	14.88
Ultimate analysis (wt%)
C, ad	20.37	15.78
H, ad	2.14	1.41
N, ad	0.60	0.55
Sulfur (d, wt%)
Total	1.84	0.91
Organic	0.28	0.07
Pyritic	1.45	0.81
Sulfate	0.11	0.03
Major composition (wt%)
SiO2	31.8	41.2
Al2O3	28.1	26.3
Fe2O3	1.35	1.28
CaO	0.13	2.40
MgO	<0.1	0.27
K2O	0.12	0.86
Na2O	<0.1	0.13
TiO2	0.96	0.85

---

The major mineral phases found in the two coal gangues are kaolinite and quartz, with minor contributions from pyrite, illite and calcite (Fig. 2a). The morphology of pyrite in coal gangue (CG2) is shown in Fig. 2b. It can be seen that pyrite was closely surrounded by aluminosilicate mineral species. In contrast, O'Brien et al.¹⁶ observed that pyrite in coal was mostly dotted in organic material or next to silicates. The different distribution of pyrite in coal gangue, as well as its mineral components, may lead to different SO₂ release behavior compared with coal.

Fig. 2 (a) The XRD patterns of coal gangue. K – kaolinite (Al₂Si₂O₅(OH)₄), Q – quartz (SiO₂), C – calcite (CaCO₃), Py – pyrite (FeS₂); (b) the scanning electron microscopy images of pyrite in coal gangue (CG2).

The general thermal properties of coal gangue are shown by the TGA, DTG and DSC curves in Fig. 3. The thermal behavior of pure pyrite is also illustrated as a comparison. Sharp exothermal peaks are observed from 400 °C to 600 °C on the DSC curves of coal gangue, accompanied by the dramatic weight loss shown on the TGA curves in the same temperature interval, indicating that the carbon content in coal gangue undergoes combustion in this temperature range. It is known that kaolinite also received dehydroxylation from 400 °C to 600 °C.^11,17 Therefore, it appears that the exothermal effect caused by coal gangue combustion is dominant, and outweighs the endothermic effect due to kaolinite dehydroxylation. The pure pyrite exhibits a similar behavior with a dramatic weight loss and significant exothermic effect around 500 °C. Overall, from 400 °C to 600 °C, coal gangue may release low molecular weight volatile chemicals due to uncompleted combustion, H₂O (g) due to kaolinite dehydroxylation and SO₂ due to oxidation of pyrite. These gaseous species together may complicate the release behavior of SO₂ in coal gangue.

Fig. 3 The thermogravimetric (TGA), derivative thermogravimetric (DTG) and differential scanning calorimetric (DSC) curves of coal gangue and pyrite. The upward peaks on DSC curves represent an exothermic reaction.

3.2. SO₂ release during calcination of coal gangue

The following equation can be used to describe the conversion of sulfur to SO₂:^18,19where x_s is the conversion rate of the sulfur content in sample to SO₂, t (min) is a certain time during the experiment, C_s(t) (μg L⁻¹) is the concentration of SO₂ in flue gas corresponding to t, V(t) (L min⁻¹) is the flow rate of the flue gas, which is kept constant at 1.2 L min⁻¹ in this study, and m_s (μg) is the mass of total sulfur in the sample.

The SO₂ release and conversion profiles of two coal gangue samples (CG1 and CG2) up to 1200 °C are presented in Fig. 4, along with the results of pure pyrite. It can be seen that both the two coal gangue samples release SO₂ at lower temperature than that of pure pyrite. CG1 and CG2 start to release SO₂ at 400 °C and 450 °C, respectively, while a measurable SO₂ release is only observed above 500 °C for pure pyrite. These results are in accordance with the thermal behaviors observed in the DSC curves of Fig. 3. In Fig. 3, both the starting and peak temperatures of the exothermic peak of pure pyrite are higher than that of coal gangue. The starting temperature for SO₂ release from coal gangue is lower than that for pure pyrite, which is consistent with the observed results in a high sulfide containing shale by Hansen et al.¹⁸ in which the shale–pyrite mixture released SO₂ at a lower temperature than pure pyrite as well. Considering that both shale¹⁸ and coal gangue experience combustion and release heat at lower temperatures than that of pure pyrite, the advance release of SO₂ may result from the elevated local temperature of embedded pyrite in coal gangue due to the heat released by the combustion. It can be seen from the DSC curves of coal gangue (shown in Fig. 3) that coal gangue does indeed have an exothermal effect at 400 °C. Moreover, the conversion profiles in Fig. 4 show that the sulfur in CG1 was almost completely converted to SO₂, while CG2 and pyrite failed to achieve 100% conversion due to the formation of sulfate in residue ash. Both the relatively high content of alkali metal oxides in CG2 and the exposure to abundant oxygen for pyrite can contribute to the formation of sulfate.

Fig. 4 The SO₂ release and conversion profiles of coal gangue and pure pyrite.

The effects of temperature on the release profiles and conversion rates of sulfur in coal gangue are illustrated in Fig. 5a and b. It can be seen that the concentration profiles of SO₂ become narrower and higher as the temperature rises and their peaks appear earlier. The two samples show overall similar behaviors. In both samples, the increase of the slope of the conversion curve is relatively significant as the temperature rises from 500 °C to 600 °C, and becomes small when the temperature further rises to 700 °C. The slope of conversion curve of CG1 at 500 °C is higher than that of CG2 at 500 °C. In addition, the sulfur in CG1 reaches complete conversion to SO₂ in 30 minutes at 500 °C, while the release of SO₂ in CG2 is still in progress. Given that CG1 actually contains more sulfur than CG2, this result reveals a much higher reaction rate of sulfur transformation in CG1 than that in CG2. At 600 °C, the sulfur in coal gangue was entirely converted to SO₂ in both CG1 and CG2 while at 700 °C, they both failed to reach 100% conversion. It may be because of the formation of sulfate, which retained part of the sulfur in residue ash. CG2 has a significant content of calcite. The calcite decomposes to lime and carbon dioxide at 700 °C (shown in Fig. 3), which favors the formation of calcium sulfate in CG2.

Fig. 5 Effect of temperature on SO₂ release rate and conversion of coal gangue as well as pure pyrite under isothermal conditions.

By comparison, the results for pure pyrite under the same experimental conditions are shown in Fig. 5c. It should be noted that the SO₂ release behavior of pyrite is largely different from that of coal gangue at 500 °C. In comparison with coal gangue, the release rate of SO₂ from pyrite is much slower at 500 °C. The lag becomes smaller at 600 °C, while at 700 °C, it becomes almost the same as that of coal gangue. Conversion of pyrite into SO₂ at 500 °C proceeds slowly and reaches only 58%, even after a long reaction time of 170 min. Note that a 100% conversion can only be achieved when the temperatures are increased to 600 °C and higher.

In addition, it is interesting that we observed a thin white layer adhered to the wall of the cooling zone of the tube after many runs of coal gangue experiments. After dissolving it into 1 mL deionized water, the pH value of the solution was measured at ∼3.0. This result indicates that some acidic species can evaporate during calcination of coal gangue. These acidic materials may be formed by the interactions of SO₂ with other organic substances or volatile elements, such as Na. Moreover, since kaolinite can dehydroxylate at 500–700 °C, the H₂O released may help with dissolution and can condense on the tube wall when cooled down. Furthermore, the XRD pattern of the collected powder, shown in Fig. SI1,† confirmed the generation of sulfates and hydrates.

3.3. Transformation of pyrite

As previously mentioned, the formation of SO₂ during the calcination of coal gangue is mainly governed by the transformation of pyrite. In an oxygen-containing atmosphere, the transformation process of pyrite is complicated,^20,21 involving at least four reactions as listed below.²⁰ The standard Gibbs free energies (Δ_rG^θ) of the reactions are listed in Table 2. It can be seen that the Δ_rG^θ of all reactions except the decomposition of pyrite (eqn (2)) give large negative values, indicating the relative difficulty of the pyrite decomposition and strong possibilities for all the other reactions to occur.

FeS₂(s) → FeS_x(s) + (1 − 0.5x)S₂(g) (2)

S₂(g) + 2O₂(g) → 2SO₂(g) (3)

2FeS_x(s) + (1.5 + 2x)O₂(g) → Fe₂O₃(s) + 2xSO₂(g) (4)

2FeS₂(s) + 5.5O₂(g) → Fe₂O₃(s) + 4SO₂(g) (5)

Table 2 The standard Gibbs free energies (Δ_rG^θ) of the reactions

Reactions	ΔrG^θ (kJ mol^−1)
	500 °C	600 °C	700 °C
FeS2(s) → FeSx(s) + (1 − 0.5x)S2(g)	34.259 (x = 1)	20.311 (x = 1)	6.383 (x = 1)
S2(g) + 2O2(g) → 2SO2(g)	−610.785	−596.205	−581.660
2FeSx(s) + (1.5 + 2x)O2(g) → Fe2O3(s) + 2xSO2(g)	−1004.577 (x = 1)	−976.188 (x = 1)	−948.230 (x = 1)
2FeS2(s) + 5.5O2(g) → Fe2O3(s) + 4SO2(g)	−1546.844	−1531.771	−1517.125
2FeS2(s) + 7O2(g) → Fe2(SO4)3(s) + SO2(g)	−1776.886	−1680.175	−1584.124
FeS2(s) + 3O2(g) → FeSO4(s) + SO2(g)	−823.684	−794.688	−765.978

---

Ferrous sulfate and ferric sulfate may also form according to the following reactions:

2FeS₂(s) + 7O₂(g) → Fe₂(SO₄)₃(s) + SO₂(g) (6)

FeS₂(s) + 3O₂(g) → FeSO₄(s) + SO₂(g) (7)

To further elucidate the transformation of pyrite during calcination of coal gangue, the mineral phases in the coal gangue residues as well as pure pyrite after calcination were analyzed. The XRD patterns of residual ashes of CG1, CG2 and pyrite at 500–700 °C are shown in Fig. 6. It can be seen that the major mineral phase is hematite (Fe₂O₃) for both CG1 and CG2, which is in good agreement with the phase equilibrium diagram of iron oxides and oxygen presented by Darken and Gurry.²² There is no indication that any FeS_x is formed while small amounts of ferric sulfate (Fe₂(SO₄)₃) or ferrous sulfate (FeSO₄) exist. The XRD pattern of pure pyrite calcined at 500 °C for 20 min (Fig. 6c) also demonstrates that no FeS_x is formed while both Fe₂(SO₄)₃ and FeSO₄ can be observed. A similar result was also obtained by Schorr et al.²³ In addition, the pyrite calcined at 600 °C and 700 °C also proved the generation of hematite.

Fig. 6 The XRD patterns of residual ash after calcination at different temperatures. Py – pyrite (FeS₂), H – hematite (Fe₂O₃), m – mikasaite (Fe₂(SO₄)₃), S – iron sulfate (FeSO₄), z – szomolnokite (FeSO₄·H₂O).

Based on these observations, the reaction mechanism is proposed to be the direct oxidation of pyrite rather than a two-step process (i.e. first decomposition and successive oxidation), i.e., the pyrite is directly oxidized to form hematite according to reaction (4) via a surface reaction. As the experiment was conducted under an air atmosphere, the supply of oxygen is sufficient. At 500, 600 and 700 °C, the decomposition rates of pyrite may be lower than the oxidation rates, and therefore, pyrite undergoes direct oxidation. This mechanism is also favored by other investigators in the experiments of pyrite transformation under similar conditions.^18,24

3.4. Kinetics analysis

The reaction rate r of SO₂ generation can be expressed as follows:²⁵where x is the fractional conversion, k(T) is the rate constant at the temperature T, and f(x) is the mechanism function. Alternatively, in terms of integral form, we obtain

By plotting the function F(x) against the time of reaction, the correlation coefficient can be calculated. An appropriate mechanism function could be estimated by comparing the linear relationships between different F(x) and t.

The results show that there is no single mechanism function that can conform to the experimental data across the whole range. However, some functions are in good agreement with different segments of the curves. The selected most probable mechanism functions under different conditions are shown in Table 3. As shown in Fig. 7, the experimental data of CG1 and CG2 at 500 °C, 600 °C and 700 °C were reasonably reproduced by a 3-D diffusion (Jander, n = 1/2) model²⁶ at the initial stage of the reaction (Fig. 7a) and a Jander model²⁶ at the following stage (Fig. 7b). For samples CG1 and CG2, the probable mechanisms are the same at different temperatures.

Table 3 The kinetic mechanism functions and parameters of sulfur dioxide emissions of coal gangue under different temperatures

Reaction model	M1					M2
	3-D diffusion (Jander)					Jander
F(x)	[1 − (1 − x)^1/3]^1/2					[1 − (1 − x)^1/3]^2
Sample	T/°C	t/min	R^2	k/min^−1	Ea/kJ mol^−1	t/min	R^2	k/min^−1	Ea/kJ mol^−1
CG1	500	0–6	0.9458	0.03166	31	6–26	0.9963	0.03684	33
	600	0–5	0.9219	0.05755		5–12	0.9918	0.07436
	700	0–3	0.9091	0.08506		3–8	0.9912	0.10588
CG2	500	0–5	0.9674	0.03082	40	5–45	0.9807	0.00806	77
	600	0–3	0.9520	0.06069		3–16	0.9895	0.04424
	700	0–2	0.9547	0.11126		2–9	0.9971	0.0921

---

Fig. 7 Diagram of different mechanism functions versus time.

Many previous studies employed the shrinking core model to describe the direct oxidation of mineral pyrite,^27–29 but the model is unable to reproduce the experimental results on coal gangue in this study. This difference may be attributed to the surrounding minerals of pyrite in coal gangue. As shown in Fig. 3, pyrite in coal gangue is embedded in the aluminosilicate, i.e., kaolinite and quartz, in separate or connected spherical aggregations. The general size of the aggregations is 30 to 50 μm. As the coal gangue samples employed in the present study were sieved to 150 μm, pyrite in the coal gangue should be coated by kaolinite or quartz. The gas diffusion rate, i.e., the outward diffusion of formed SO₂, through the mineral layer is limited, which deviates from the shrinking core model.

According to eqn (9), the intrinsic surface reaction rate k(T) can be obtained by the slope of the curves of F(x) versus t in Fig. 7. By plotting ln k(T) against 1/T (shown in Fig. SI2†), the activation energy E can thus be derived according to the following equation:²⁵

The results of the intrinsic surface reaction rate k(T) and activation energy E are also listed in Table 3. The activation energy only slightly increased for CG1 at the later stage compared to the initial stage of the reaction, while it increased significantly for CG2, indicating the increase of diffusion resistance in CG2.

4. Conclusions

The SO₂ evolution and pyrite transformation during calcination of coal gangue was investigated. It can be found that sulfur in coal gangue was mainly in the form of pyrite, whereas the content of organic sulfur was low, which is different from the sulfur content of coal. The release of SO₂ in coal gangue was accelerated by the combustion of the carbon content in coal gangue, which resulted in a lower start temperature and significantly higher release rate at 500 °C compared with pure pyrite. Pyrite was observed to undergo direct oxidation under the studied experimental conditions and the final product was majorly hematite. The evolution of SO₂ initially followed the 3D diffusion models and correlated with the Jander model in subsequent conversions, which showed no difference between two coal gangues. In addition, we also find that since coal gangue usually undergoes incomplete combustion, and carbon monoxide and other volatile matters generated during calcination may react with SO₂ to produce low-volatility acidic species in the presence of H₂O due to the dehydroxylation of clay minerals. These species may result in the corrosion of devices and various other operational problems, and the result of the present study is significant to the further abatement of SO₂ emissions during coal gangue calcination.

Acknowledgements

This study was sponsored by the Common Development Fund of Beijing and the National Natural Science Foundation of China (51172003 and 51074009), the National High Technology Research and Development Program of China (863 Program, 2012AA06A114), and the China National Key Technology R&D Program (2011BAB03B02 and 2013BAC14B07).

References

1. Z. F. Bian, X. X. Miao, S. G. Lei, S. E. Chen, W. F. Wang and S. Struthers, Science, 2012, 337, 702–703.
2. M. Frias, R. V. de la Villa, M. I. S. de Rojas, C. Medina and A. J. Valdes, J. Am. Ceram. Soc., 2012, 95, 386–391.
3. C. C. Zhou, G. J. Liu, S. C. Wu and P. K. S. Lam, Chemosphere, 2014, 95, 274–280.
4. X. Wang, C. Zhou, G. Liu and Z. Dong, Anal. Lett., 2013, 46, 1962–1977.
5. Y. C. Zhao, J. Y. Zhang, C. L. Chou, Y. Li, Z. H. Wang, Y. T. Ge and C. G. Zheng, Int. J. Coal Geol., 2008, 73, 52–62.
6. D. Nichol and N. P. Tovey, Eng. Geol., 1998, 50, 309–318.
7. F. D. Ardejani, B. J. Shokri, M. Bagheri and E. Soleimani, Environ. Earth Sci., 2010, 61, 1547–1560.
8. http://hzs.ndrc.gov.cn/newgzdt/201304/t20130412_536759.html, accessed 13 June 2014.
9. F. R. Meng, J. L. Yu, A. Tahmasebi and Y. N. Han, Energy Fuels, 2013, 27, 2923–2932.
10. D. X. Li, X. Y. Song, C. C. Gong and Z. H. Pan, Cem. Concr. Res., 2006, 36, 1752–1759.
11. C. C. Zhou, G. J. Liu, Z. C. Yan, T. Fang and R. W. Wang, Fuel, 2012, 97, 644–650.
12. C. Zhou, G. Liu, S. Cheng, T. Fang and P. K. S. Lam, Bioresour. Technol., 2014, 166, 243–251.
13. Determination of Total Sulfur in Coal (GB/T 214), China, 2007.
14. Determination of Forms of Sulfur in Coal (GB/T 215), China, 2003.
15. S. Gornostayev, J. Harkki and O. Kerkkonen, Fuel, 2009, 88, 2032–2036.
16. G. O'Brien, Y. Gu, B. J. I. Adair and B. Firth, Miner. Eng., 2011, 24, 1299–1304.
17. A. Á. B. Maia, R. S. Angélica, R. de Freitas Neves, H. Pöllmann, C. Straub and K. Saalwächter, Appl. Clay Sci., 2014, 87, 189–196.
18. J. P. Hansen, L. S. Jensen, S. Wedel and K. Dam-Johansen, Ind. Eng. Chem. Res., 2003, 42, 4290–4295.
19. S. L. Niu, K. H. Han and C. M. Lu, Chem. Eng. J., 2011, 168, 255–261.
20. F. Paulik, J. Paulik and M. Arnold, J. Therm. Anal., 1982, 25, 313–325.
21. G. L. Hu, K. Dam-Johansen, W. Stig and J. P. Hansen, Prog. Energy Combust. Sci., 2006, 32, 295–314.
22. L. Darken and R. Gurry, J. Am. Chem. Soc., 1946, 68, 798–816.
23. J. Schorr and J. Everhart, J. Am. Ceram. Soc., 1969, 52, 351–354.
24. G. M. Schwab and J. Philinis, J. Am. Chem. Soc., 1947, 69, 2588–2596.
25. S. Vyazovkin, A. K. Burnham, J. M. Criado, L. A. Perez-Maqueda, C. Popescu and N. Sbirrazzuoli, Thermochim. Acta, 2011, 520, 1–19.
26. P. Li, Q. B. Yu, H. Q. Xie, Q. Qin and K. Wang, Energy Fuels, 2013, 27, 4810–4817.
27. N. Boyabat, A. K. Ozer, S. Bayrakceken and M. S. Gulaboglu, Fuel Process. Technol., 2004, 85, 179–188.
28. Y. Kajiwara, A. Sasaki and O. Matsubaya, Geochem. J., 1981, 15, 193–197.
29. I. C. Hoare, H. J. Hurst, W. I. Stuart and T. J. White, J. Chem. Soc., Faraday Trans. 1, 1988, 84, 3071–3077.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra06954d

---