Scholarly article on topic 'Experimental Analysis of Minimum Ignition Temperature of Coal Dust Layers in Oxy-fuel Combustion Atmospheres'

Experimental Analysis of Minimum Ignition Temperature of Coal Dust Layers in Oxy-fuel Combustion Atmospheres Academic research paper on "Chemical engineering"

CC BY-NC-ND
0
0
Share paper
Academic journal
Procedia Engineering
OECD Field of science
Keywords
{"oxy-fuel combustion" / "minimum ignition temperature" / "dust layer" / "hot surface" / "thermal and kinetic parameters"}

Abstract of research paper on Chemical engineering, author of scientific article — Dejian Wu, Frederik Norman, Filip Verplaetsen, Jan Berghmans, Eric Van Den Bulck

Abstract An experimental investigation into the hot surface ignition of coal dust layers has been undertaken in oxy-fuel combustion atmospheres, with the oxygen mole fractions in the range of 21-50 vol. % (21%, 30%, 40% and 50%). Three coals were used to determine the minimum layer ignition temperature (MLIT) with a hot surface ignition apparatus according to the European EN50281-2-1 norm. In addition, thermal and kinetic parameters were also determined. Layer thicknesses of 5, 12.5, 15, 20 and 30mm were investigated. Firstly, 5 and 12.5mm thick coal dust layers were used to investigate the influence of the oxygen mole fraction on the MLIT, and the steady state dust layer temperature profile of a 30mm thick coal dust layer was used to estimate the thermal conductivity (k) at a low hot surface temperature. Secondly, the South African coal was used in this study to determine the kinetic parameters on the basis of the Frank-Kamenetskii theory. Results show that oxygen concentration has a significant influence on the MILT, (i.e., MLIT decreases with increasing oxygen concentration). It also shows that replacing N2 in the combustion air with CO2 a small increase of the MLIT will occur. The estimated k of the coal dust samples is respectively 0.1, 0.11 and 0.1 W·m-1·K-1 for South African coal, Sebuku coal and Pittsburgh coal. The region of the kinetic parameters of South African coal changes when there is a shift in the test from air to oxy-fuel combustion atmospheres,which means that the reaction mechanism of self-heating or ignition changed somewhat.

Academic research paper on topic "Experimental Analysis of Minimum Ignition Temperature of Coal Dust Layers in Oxy-fuel Combustion Atmospheres"

CrossMark

Available online at www.sciencedirect.com

ScienceDirect

Procedía Engineering 84 (2014) 330 - 339

Procedía Engineering

www.elsevier.com/locate/procedia

"2014ISSST", 2014 International Symposium on Safety Science and Technology

Experimental analysis of minimum ignition temperature of coal dust layers in oxy-fuel combustion atmospheres

Dejian WUa, Frederik NORMANb, Filip VERPLAETSENb, Jan BERGHMANSa,

Eric Van DEN BULCKa*

aKatholieke Universiteit Leuven, Department of Mechanical Engineering, Celestijnenlaan 300A, B3001 Leuven, Belgium bAdinex NV, Brouwerijstraat 5/3, B 2200 Noorderwijk, Belgium

Abstract

An experimental investigation into the hot surface ignition of coal dust layers has been undertaken in oxy-fuel combustion atmospheres, with the oxygen mole fractions in the range of 21-50 vol. % (21%, 30%, 40% and 50%). Three coals were used to determine the minimum layer ignition temperature (MLIT) with a hot surface ignition apparatus according to the European EN50281-2-1 norm. In addition, thermal and kinetic parameters were also determined. Layer thicknesses of 5, 12.5, 15, 20 and 30mm were investigated. Firstly, 5 and 12.5mm thick coal dust layers were used to investigate the influence of the oxygen mole fraction on the MLIT, and the steady state dust layer temperature profile of a 30mm thick coal dust layer was used to estimate the thermal conductivity (k) at a low hot surface temperature. Secondly, the South African coal was used in this study to determine the kinetic parameters on the basis of the Frank-Kamenetskii theory. Results show that oxygen concentration has a significant influence on the MILT, (i.e., MLIT decreases with increasing oxygen concentration). It also shows that replacing N2 in the combustion air with CO2 a small increase of the MLIT will occur. The estimated k of the coal dust samples is respectively 0.1,

0.11 and 0.1 W • m-1 • K-1 for South African coal, Sebuku coal and Pittsburgh coal. The region of the kinetic parameters of South African coal changes when there is a shift in the test from air to oxy-fuel combustion atmospheres,which means that the reaction mechanism of self-heating or ignition changed somewhat.

© 2014 PublishedbyElsevier Ltd.Thisisan open access article under the CC BY-NC-ND license (http://creativecommons.Org/licenses/by-nc-nd/3.0/).

Peer-review under responsibility of scientific committee of Beijing Institute of Technology

Keywords: oxy-fuel combustion; minimum ignition temperature; dust layer; hot surface; thermal and kinetic parameters

* Corresponding author. Tel.: +0032 16 322 509; Fax: +0032 16 322 985. E-mail address: eric.vandenbulck@kuleuven.be

1877-7058 © 2014 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

Peer-review under responsibility of scientific committee of Beijing Institute of Technology doi: 10.1016/j.proeng.2014.10.441

1. Introduction

Self-heating of coal is the process by which coal achieves temperatures higher than ambient caused by exothermic reactions when coal is exposed to surrounding atmosphere without an external ignition source [1-3]. Self-ignition or spontaneous combustion, also defined as supercritical self-heating or thermal runaway, is quite hazardous in many powder industrial processes and in applications such as storage of porous bulk materials [4-7] and dust accumulations on a hot surface [8-14]. Self-ignition may cause fires and even dust explosions in the worst case [15]. Self-ignition of a dust layer depends on many factors, which can be divided into two main types: properties of coal (internal factors) and the environmental conditions (external factors) [16]. Previous researchers have investigated the effect of layer shrinkage [8], inerts [9], heat flux [10], air flow [11] and the estimation of thermo-kinetic parameters [12], as well as confinement geometry [13] and ignition position [14]. Despite the vast studies on self-ignition of coal, only a few works have investigated the effect of oxygen concentration on self-ignition [4, 6].

Oxy-fuel combustion, as one of the most promising technologies for CO2 emissions, has been advancing in recent years [17-19]. In the oxy-fuel combustion system, dust may be processed or deposited in oxygen-enriched environments where little or nothing is known about the spontaneous ignition risk. The minimum ignition temperature of coal dust layers is studied with various oxygen concentrations by means of a heated surface. In addition, one test series was carried out to determine the thermal and kinetic parameters that are necessary in mathematical modeling.

2. Experiments

Hot plate ignition tests of 3 coal dust layers from different locations were conducted at various gas environments: South African coal, Indonesian Sebuku coal, and Pittsburgh No.8 coal. Hereafter these coals are referred to as "SA coal", "IS coal", and "P8 coal" respectively. 5 stainless rings of different height (5, 12.5, 15, 20 and 30mm) were employed and corresponding thermocouples were placed at the specified heights for each ring as shown in Table 1. Specifically, the results of the coal samples from the 5 and 12.5mm layer were used to estimate the influence of the oxygen concentration on the minimum ignition temperature of dust layer. The steady state temperature distribution in the 30mm layer was used to determine the thermal conductivity. The Kinetic parameters of SA coal were determined with the results of the MLIT experiments with 5 different thicknesses.

Table l.The experimental outline and objectives.

Ring with thermocouples Gas environment

Ring height L (mm) Thermocouple height (mm) air 21 vol.% 30 vol.% O2 in CO2 O2 in CO2 40 vol.% 50 vol.% O2 in CO2 O2 in CO2

5 2 3 samples

12.5 4, 7 and 10 3 samples -

15 3, 6, 9 and 12 SA SA SA SA -

20 5, 10 and 15 SA SA SA SA -

30 6, 12, 18 and 24 3 samples (steady state) SA - - SA -

2.1. Coal dust samples and gas composition

The coal samples were first milled or sieved (SA coal and IS coal were milled and sieved < 63 ^m, and Pittsburgh No.8 coal was milled < 200 ^m). Next these coals were dried in a vacuum oven at 80 °C and 0.1 bar until the moisture content was maximum 3%. Proximate and ultimate analysis data are given in Table 2. According to their compositions, IS coal can be classified as a subbituminous coal, while SA coal and P8 coal are bituminous coals.The gas mixtures consisted of air and various oxygen mixtures with carbon dioxide. Beginning with 21 vol. %

O2 and 79 vol. % CO2, the oxygen mole fraction was increased in increments of 10% from 30 vol. % up to 50 vol. % O2 in CO2 with an accuracy of ± 1%.

Table 2. Properties of the coal dust samples.

Properties

Coal Samples

Proximate analysis (wt. %)

Fixed carbon 47 55.6 56.6

Volatile material 38.2 31 27.1

Moisture content 3 3 2.1

Ash 11.8 10.4 14.2

Ultimate analysis (%)

Carbon 65.7 73.8 67.5

Hydrogen 5.22 4.82 4.26

Nitrogen 1.58 1.29 1.76

Oxygen (by difference) 27.5 20.09 26.48

Gross calorific value ( MJ • kg f1) 27.65 29.78 27.37

-3 Packing density (kg • m ) 645±5 620±5 600±5

2.2. Experimental apparatus and procedure

The fundamental apparatus and procedure of the dust layer test on a constant hot surface are similar to the European EN50281-2-1 norm [20]. A dust layer is placed on a hot plate, its temperature controlled by an automated control system. In order to carry out the test at oxy-fuel combustion atmospheres, an isolation glass box (about 400L) with two gloves is added as shown in Fig. 1. The experimental setup consisted of a circular stainless steel plate 25.4mm in thickness and 203mm in diameter, and stainless steel rings with an internal diameter (D ) of 100mm. The rings were placed at the center of the temperature-controlled plate with a thermal stability of 1°C. The thermocouples were set in the specified rings (see in Table 1) in order to measure the temperature profile of the dust layer. Once the hot plate temperature is stable at a predetermined temperature, the ring was gently filled with dust sample and was leveled with a spatula without compressing. Any extraneous dust on the hot plate was removed. The thermocouples were connected to the data collection system. It should be noted that all coal samples were put in the corner of the glass box that was already flushed by the premixed O2/CO2 gas. All the operations were conducted in the isolation glass box with vacuum chamber gloves.

Fig. 1. The schematic of the hot plate test.

When the test starts, the mixture gas was blown into the oven at a rate of 270 L/h measured by a rotameter. Ventilation was provided to exhaust any gases released from the dust sample. The concentration of oxygen in the box was adjusted by the gas flow controller. An iterative procedure was used to determine the ignition temperature within 5°C. The temperatures were recorded simultaneously every 10 seconds. Ignition is considered to have occurred if visible glowing or flaming is observed, a temperature of 450°C is reached, or a temperature rise of 250°C above the hot surface temperature is measured. If none of these conditions is satisfied after a delay (e.g. 30 min for 5 mm layer), ignition is considered to have not occurred.

3. Experimental results and discussion

3.1. Dust Layer temperature evolutions: 5mm

Typical temperature profiles for self-heating and ignition of dust layers are shown in Fig.2 (a) for 5mm layers. The critical ignition temperature (the MLIT or Tc) is supposed to be at the region between the maximum subcritical

and the minimum supercritical, e.g. 235-240°C for the IS coal, and 240-245°C for the SA coal and the P8 coal in the case of Figure 2 (a). It is clear that the critical ignition temperature of the IS coal is a bit lower than that of the others, which may result from IS coal's higher volatile contents (see in Table 2). The observation that IS coal was more likely to produce smoke can also support this result. Even so, the induction time to ignition of the IS coal is much longer than that of the others. One of the possible reasons is that the ignition reaction of the IS coal occurs more slowly than the other coals due to the higher activation energy at the same conditions at which it was determined in a previous study [21].

Fig. 2 (b) illustrates the influence of oxygen concentration on the thermal behavior of the 5mm thick SA coal. The temperatures in the sample rise rapidly following ignition and large quantities of smoke are present, especially at the higher concentrations of oxygen. Similarly, more visible glowing was found in 40 vol. % O2 in CO2. Moreover, the results show that both the induction time to ignition and minimum ignition temperature decrease with increasing oxygen concentration, while the maximum temperature can be reached after ignition increases with oxygen concentration.

Fig. 2. Thermal behaviour of the 5mm coal dust layer: (a) the subcritical and supercritical of three coal dust layers at 30 vol. % O2 in CO2; (b) the SA coal at various oxygen concentrations.

3.2. Dust Layer temperature evolutions: 12.5mm

Temperature evolutions at different heights in the 12.5mm dust layer are shown in Fig. 3. The subcritical and supercritical conditions were also determined at various oxygen concentrations. It shows that the maximum

temperature can be reached after ignition ( Tmax ) elevates with increasing ambient oxygen concentration. It is

worthwhile to mention that no crossing point can be found between the temperature figures of the different ring

heights for all the subcritical conditions (the dash lines in Figure 3). However, crossing points can be found for all the supercritical conditions (the solid lines in Figure 3), e.g., the temperature at the height of 10mm overtakes that of the 4mm and 7mm height in 12.5mm SA layer. Therefore, the crossing point might be also regarded as a criterion to judge if ignition will occur or not [11].

Fig. 3.Temperature profiles of SA coal of thickness of 12.5mm: (a) air; (b) 21 vol. % O2 in CO2; (c) 30 vol. % O2 in CO2; (d) 40 vol. % O2 in CO2.

3.3. Influence of the combustion environment on the MLIT

The MLIT of three different coal samples with a height of 5mm and 12.5mm were determined at various oxygen concentrations. The results are plotted in Fig. 4 (a) and (b), respectively. As already illustrated by the temperature profiles, it is clear that oxygen concentration plays an important role in the self-heating or self-ignition process. The MLIT of the three coal samples decreases with increasing oxygen concentration, which agrees with the previous observation of the MCIT (minimum ignition temperature of dust cloud) at oxy-fuel atmospheres [22]. Specifically, the MLIT decreases dramatically when the oxygen concentration increases from 21 vol. % to 30 vol. % and then the MLIT goes down gradually with further increasing oxygen concentration as shown in Fig. 4 (a). In contrast, the influence of oxygen concentration is more moderate and the MLIT shows an approximate linear correlation with oxygen concentration from 21 vol. % to 40 vol. % as shown in Fig. 4 (b).

Table 3 lists the MLIT of South African coal at different thicknesses. This table shows that MLIT decreases dramatically from 5mm to 12.5mm thick rings, and then this trend continues on gradually. With an increased thickness in the dust layer, oxygen diffusion to the bottom of the layer becomes difficult, which is the reason why the maximum temperature of self-ignition occurred at the top surface where oxygen is easily available. At first, the thermal decomposition or oxidation reactions occur at the bottom of the layer and generate gas products such as carbon oxides and volatiles, which can slow down the oxygen diffusion from the top surface layer.

°„ 250

— ■ — Sebuku coal

— • — Pittsburgh coal

—♦— South African coal

20 25 30 35 40 O2 concentration, vol. %

20 25 30 35

O2 concentration, vol. %

Fig. 4. The influence of oxygen concentration on the MLIT for 5mm (a) and 12.5mm (b) dust layers. Table 3. The MLIT of South African coal at different thicknesses.

Oxygen concentration

the MLIT (°C)

12.5mm

air 21 vol.% 30 vol.% 40 vol.%

255-260 210-215

260-265 215-220

240-245 205-210

230-235 198-200

205-210 208-213 200-205 190-195

196-200 180-185

197-201 -190-195 -185-188 170-173

4. Estimation of thermodynamic and kinetic parameters

4.1. Steady state energy conservation

With the assumption of negligible reactant depletion, a one dimensional steady state heat conduction equation for the dust layer can be written as [1, 12]:

d2T E k —- + pbAHA exp(--) = 0

&2 b RT

where pb is the packing density ( kg • m-3), k is the thermal conductivity of a dust layer ( W • m-1 • K-1), and AH is the heat of reaction (J • kg-1) which is assumed to equal the gross calorific value in this paper.

According to the experimental procedure, the dust layer equals the ambient room temperature at time t=0:

t = 0, T = Ta (2)

Boundary conditions of constant temperature at the bottom surface and convective and radiative cooling on the top surface are

x = 0, T = T

x = r , "k~T= h(Ts - Ta) dx

where h = h + h is the total heat transfer coefficient (W • m-2 • K-1) accounting for the convective and radiative heat

transfer coefficient, Ts is the top surface temperature, Tp is the hot plate temperature and Ta is the ambient

temperature.

4.2. Estimation of thermal conductivity

Effective thermal conductivity plays an important role in the heat transfer in the dust layer. Therefore, some tests were carried out to estimate thermal conductivity using the method described by Park et al [12] and based on the assumption that heat generation can be neglected at low temperature. Then Equation (1) can be rewritten as

= 0 (5)

which implies a the linear temperature profile in the dust layer ( — =constant). Rearranging equation (4) yields

* = h (Ta - Ts)— = irhTzT) (6)

d T (Ts " Tp)

where dx is the thickness of the dust layer (30mm). Ts can be obtained from the steady state experiment. h can be calculated by the following equations:

0.54R 0 25k

hc =-^^ (7)

hr =«T(TS2 + Ta2)(Ts + Ta) (8)

where l is characteristic length equal to the side of a square having the same area as the dust layer surface ( l =0.09m in this case), Ra = gfi(Ts -Ta )l3 / (vDa), g is the gravitational constant, p = 2/(Ts + Ta), v is the kinematic viscosity

of air, Da is thermal diffusivity of air, ka is the thermal conductivity of air, e is the coal emissivity=0.9, and cr is the Stefan-Boltzmann constant (5.67x 10-8W• m-2 • K-4).

In order to evaluate the value of Ts, 4 thermocouples were used to measure temperatures at heights of 6, 12, 18

and 24mm in the 30mm dust layer exposed at a hot surface at 50°C. The temperature profiles of the locations are shown in Fig. 5 (a). After 5000s, the average temperatures at 0, 6, 12, 18 and 24mm in the South African coal dust layer were 49.6, 44.3, 39.8, 34.7 and 30.9°C, respectively. A linear extrapolation from the 5 temperatures yields the temperature of the top surface (25.9°C) of the dust layer as shown in Fig. 5 (b). Then the effective thermal

conductivity of the dust layer (0.1 W • m 1 • K 1) can be obtained. Similarly, the thermal conductivity of IS coal and

P8 coal are 0.11 and 0.1 W • m-1 • K-1, respectively.

4.3. Estimation of kinetic parameters

Previous work [21] found that to determine the kinetic parameters, the Frank-Kamenetzkii (F-K) method is more reliable when compared to other Arrhenius equation variations. Hence, the F-K method was used in this paper. The

F-K parameter ^cr, which is defined as [1]:

„ S^pA exp[~Ea/( RTC)]

o =--(9)

cr kRT2 1 J

where £cr depends on the geometry of the dust sample (e.g. £cr =0.88 for an infinite slab), ro is the radius or

characteristic length of the dust sample (L/2, see Fig. 1).

I 30 H

- Experiment Linear approximation

10 20 Distance, mm

Fig. 5. 30mm SA dust layer temperatures at different heights when exposed to a hot surface of 50°C: (a) temperature profiles; (b) equilibrium temperatures distribution and its linear trend line.

Rearranging equation (9) gives

Mcr c —2"

pQE E ) = ln(—— A)- — k R RT

Equation (10) also gives a linear correlation between the terms ln(5rTc2 /ro2)and 1/Tc with E /R as the slope, which offers the possibility to determine the pre-exponential factor from the intercept of the linear correlation [1].

Fig. 6 shows the correlation between the characteristic lengths of the coal samples and the critical ignition temperature of both the upper and the lower limits using the F-K method and based on the results of the MLIT of Table 3. From Fig. 6, combined with Equation (10), the kinetic parameters of South African coal can be derived. The results are listed in Table 4. Clearly, the activation energy in air is similar with that in 21 vol. % O2 in CO2, but smaller than those in 30 and 40 vol. % O2 in CO2 and the magnitude of the pre-exponential factor increases by a factor of 2. Strangely both values of the kinetic parameters are the same on the whole between the 30 and 40 vol. % O2 in CO2.

1,8 1,9 2,0 2,1 2,2

1000/7, Tc in K

Fig. 6. Interceptions for the linear correlation (Frank-Kamenetskii theory).

Table 4. The kinetic parameters of South African coal.

Oxygen concentration E ( kJ • mol 1) A (s"1)

lower upper mean lower upper mean

Air 100.2 108.7 104.4 1.4e5 1e6 3.9e5

21 vol.% O2 in CO2 97.5 105.1 102.3 7.3e4 3.4e5 1.6e5

30 vol.% O2 in CO2 115.8 122 118.9 9.8e6 3.3e7 1.8e7

40 vol.% O2 in CO2 114.6 124 119.3 9.1e6 1.2e8 3.4e7

5. Conclusions

The experimental investigation of the minimum ignition temperature of dust layers has been performed by means of aheated plate. The influence of oxygen concentration and dilution of air with CO2was studied. Several results of the MLIT under various conditions arepresented to provide data and experimental basis for the assessment and prevention of coal dust fires, and explosion risks related to hot surfaces. The thermal and kinetic parameters have been estimated which can be used for further modelling work. The conclusions as follows:

(1) As the thickness of the dust layer increases, the critical temperature for ignition decreases significantly both in air and oxy-fuel combustion atmospheres.

(2) A remarkable increase in the spontaneous ignition risk of the coal dust layer has been found when increasing the ambient oxygen concentration. However, the effect of ambient oxygen concentration gradually lowers with increased thickness of the dust layer.

(3) The obvious variation in the values of the apparent activation energy and pre-exponential factor is observed when the combustion environment shifts from air to oxy-fuel atmospheres, which means that the reaction mechanism of self-heating or ignition more or less changed. Further investigation is necessary to elucidate this phenomenon.

Acknowledgements

The authors gratefully acknowledge financial support from the European FP7 project RELCOM (Reliable and Efficient Combustion of Oxygen/Coal/Recycled Flue Gas Mixtures).

References

[1] P.C. Bowes, Self-Heating: Evaluating and Controlling the Hazards, Elsevier, London, 1984.

[2] J.N. Carras, & B.C. Young, Self-heating of coal and related materials: models, application and test methods,Prog.Energy Combust. Sci. 20(1994)1-15.

[3] H. Wang, B. Z. Dlugogorski, E. M. Kennedy. Coal oxidation at low temperatures oxygen consumption, oxidation products, reaction mechanism and kinetic modelling,Prog.Energy Combust. Sci.29 (2003) 487-513.

[4] P.C. Bowes, & P.H. Thomas, Ignition and extinction phenomena accompanying oxygen-depend self-heating of porous bodies. Combust. Flame 10 (3)(1996) 221-230.

[5] Y.S. Nugrono, A.C. Mcintosh, M. Gibbs. Low-temperature oxidation of single and blend coals. Fuel 79 (2000) 1951-1961.

[6] M. Schmidt, C. Lohrer, U. Krause. Self-ignition of dust at reduced volume fractions of ambient oxygen. J. Loss Prev. Process Ind. 16 (2003) 141 -147.

[7] J. García-Torrent, Á. Ramírez-Gómez, E. Querol-Aragón, C. Grima-Olmedo, L. Medic-Pejic. Determination of the risk of self-ignition of coals and biomass materials,! Hazard. Mater. 213-214 (2012) 230-235.

[8] H.M. Kim, C.C. Hwang, Heating and ignition of combustible dust layers on a hot surface: Influence of layer shrinkage, Combust. Flame 105 (1996) 471-485.

[9] P.D. Reddy, P.R. Amyotte, M.J. Pegg, Effect of inerts on layer ignition temperatures of coal dust, Combust. Flame 114 (1998) 41-53.

[10] K. Lebecki, Z. Dyduch, A. Fibich, J. ' Sli'z, Ignition of a dust layer by a constant heat flux, J. Loss Prev. Process Ind. 16 (2003) 243-248.

[11] H.J. Park, R. Zalosh, Air flow and oxygen concentration effects in dust layer hot surface ignition temperature tests, in: 5th International Seminar on Fire and Explosion Hazards, University of Edinburgh, Edinburgh, UK, 2007.

[12] H. Park, A.S. Rangwala, N.A. Dembsey, A means to estimate thermal and kinetic parameters of coal dust layer from hot surface ignition tests, J. Hazard. Mater. 168 (2009) 145-155.

[13] K.A. Joshi, V. Raghavan, A.S. Rangwala, An experimental study of coal dust ignition in wedge shaped hot plate configurations, Combust. Flame 159 (2012) 376-384.

[14] C.Yuan, D.Huang, C.Li, G.Li,Ignition behavior of magnesium powder layers on a plate heated at constant temperature,J. Hazard. Mater.246- 247 (2013) 283- 290.

[15] R.K. Eckhoff, Dust explosions in the process industries. Gulf Professional Publishing, Amsterdam, 3rd edition, 2003.

[16] V. Babrauskas, Ignition Handbook, Fire Science Publishers, Washington, 2003.

[17] B.J.P. Buhre, L.K. Elliott, C.D. Sheng, R.P. Gupta, T.F. Wall. Oxy-fuel combustion technology for coal-fired power generation,Prog.Energy Combust. Sci. 31 (2005) 283-307.

[18] M.B. Toftegaard, J. Brix, P.A. Jensen, P. Glarborg, A.D. Jensen. Oxy-fuel combustion of solid fuels, Prog. Energy Combust. Sci. 36 (2010) 581-625.

[19] L. Chen, S.Z. Yong, A.F. Ghoniem. Oxy-fuel combustion of pulverized coal: Characterization, fundamentals stabilization and CFD modeling. Prog.Energy Combust. Sci. 38 (2012) 156-214.

[20] EN 50281-2-1, Electrical apparatus for use in the presence of combustible dust-Part 2-1: test methods— Methods of determining minimum ignition temperatures, 1999.

[21] D.Wu, F.Norman, F.Verplaetsen, J. Berghmans,E.Van den Bulck, Auto-ignition Parameters of Bituminous Coal Dusts. The 38th International Technical Conference on Clean Coal & Fuel Systems, 2-6 June, 2013, Clearwater, Florida, USA.

[22] D. Wu, F. Norman, F. Verplaetsen, J. Berghmans, E.Van den Bulck,Experimental analysis of minimum ignition temperature of coal dust clouds in oxy-fuel atmospheres, The 6th Clean Coal Technologies Conference, 12-16 May, 2013, Thessaloniki, Greece.