Organic components in thermally altered coal waste: Preliminary petrographic and geochemical investigations

International Journal of Coal Geology 71 (2007) 405 – 424 www.elsevier.com/locate/ijcoalgeo

Organic components in thermally altered coal waste: Preliminary petrographic and geochemical investigations M. Misz ⁎, M. Fabiańska, S. Ćmiel University of Silesia, Faculty of Earth Sciences, ul. Będzińska 60, 41 - 200 Sosnowiec, Poland Received 28 February 2006; received in revised form 30 August 2006; accepted 30 August 2006 Available online 24 October 2006

Abstract The petrographic and geochemical composition of coal wastes exposed to fire in the minestone dump of Piekary Śląskie town (Upper Silesia, Poland) was investigated using samples collected at various distances from a recent fire site. The question as to whether geochemical biomarker maturity parameters could be applied to assess thermal changes in organic matter caused by waste dump fires, was examined using the data obtained. Geochemical parameters were correlated with observed petrographic changes in the organic matter caused by oxidation and heating. Petrographic analyses included the determination of maceral group contents (vitrinite, liptinite and inertinite), mineral matter and coke contents, and reflectance measurements on organic matter. All results were supported by proximate and ultimate analyses. Geochemical analysis included ultrasonic solvent extraction of bitumen followed by gas chromatography–mass spectrometry (GC–MS) of the extracts. In petrographic terms, the influence of heating was seen in reflectance variations and as oxidation rims, cracks, pores and coke development. Some zoned oxidation rims may be interpreted as re-heating episodes. In terms of chemical fingerprints, less thermally-stable compounds such as lighter n-alkanes, cyclic isoprenoids, methyl- and dimethylnaphthalenes, methyphenanthrenes and five-ring polycyclic aromatic hydrocarbons were destroyed or evaporated in the most fire-affected material. The presence/absence of particular compound groups was used to assess heating temperatures. Biomarker parameters of thermal maturity were used to assess alterations in organic matter around the waste dump fires, especially those indices and ratios with higher maturity ranges, e.g. (3-methylbiphenyl + 4-methylbiphenyl)/dibenzofurane and Σdimethylbiphenyls/Σmethyldibenzofuranes. © 2006 Elsevier B.V. All rights reserved. Keywords: Coal waste; Macerals; Organic biomarkers; Thermal alteration; Oxidation; Gas chromatography–mass spectrometry

1. Introduction Coal mining is the most important industry in Upper Silesia (Poland). Some 40 coal mines are currently operating. These mines produce annually 95.62 Mt generating about 40 Mt of coal waste (Jabłońska, 2004). In Poland, the exploitation of 1 t of coal generates 0.6– 0.7 Mt of waste (Skarżyńska, 1995). The mining waste is ⁎ Corresponding author. Tel.: +48 32 3689546; fax: +48 32 2915865. E-mail address: [email protected] (M. Misz). 0166-5162/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.coal.2006.08.009

collected in 136 centralized dumps and on sites adjacent to the collieries (Skarżyńska, 1995; Sikorska-Maykowska, 2001). The waste derives from pre-mining and actual mining activities and from coal-separation processes such as those generated by washing and flotation procedures (Skarżyńska, 1995). The grain size of the waste material is variable and ranges from less than 1 mm to decimeters; the typical shape is angular (Skarżyńska, 1995). Lithologically, the waste is composed mainly of sandstones, shales, mudstones and, occasionally, conglomerates. Organic matter,

406

M. Misz et al. / International Journal of Coal Geology 71 (2007) 405–424

in quantities ranging from 3–30 wt.% (Skarżyńska, 1995), occurs as dispersed particles of various sizes, in lenses and as laminae. Air access to individual organic particles in the coal waste dumps varies. Some particles are easily oxidised, while others with limited air access are not oxidised. Several coal waste dumps in the Upper Silesia Region, where organic matter is oxidised, are also prone to selfheating. During this process, the oxygen content in the organic matter increases (van Krevelen, 1993), while the carbon and the hydrogen contents decrease (Tekely et al., 1987). Plasticity, calorific values, and coking ability also decrease (Banerjee et al., 2000; Machnikowska et al., 2003). Organic matter oxidation also leads to an increase in carboxylic groups and to changes in the proportions of different bitumen fractions (Martinez and Escobar, 1995). During oxidation, vitrinite particles develop brighter ‘oxidation rims’ (Chandra, 1962; Stach et al., 1982; Calemma et al., 1995; Ndaji and Thomas, 1995). However, this is not always the case as Ingram and Rimstidt (1984) have reported darker rims. During vitrinite oxidation, irregular cracks have also been shown to develop (Stach et al., 1982; Calemma et al., 1995; Ndaji and Thomas, 1995). These changes occur mainly in the vitrinite macerals; inertinite particles tend to develop microfractures and pores (Ingram and Rimstidt, 1984). The occurrence of fires in coal waste dumps is only possible if three conditions are simultaneously satisfied, namely easy access to air, the presence of organic matter and special conditions that make heat accumulation possible (Urbański, 1983). Two stages in the process of waste self-ignition have been defined by Sawicki (2004): an incubation stage during which the temperature does not increase, and a self-ignition stage during which the temperature rises progressively. The temperature of self-ignition depends on the rank of the organic matter and, for bituminous coals, ranges from 250– 350 °C (Cygankiewicz, 1996). Waste burning temperatures can reach between 1200–1300 °C (Sawicki, 2004). The process of self-heating- and self-ignition — can stop at any time if the heat generated falls below a critical threshold. The self-heating process is also dependent on a number of factors. The maceral composition and the rank of the organic matter are important. The physico-chemical properties of the organic matter, the pore structure and the oxidation degree are also influential (Rosiek and Urbański, 1990; Sawicki, 2004). The aim of this paper is to: (a) determine the rank and the petrographic composition of the organic matter in the coal waste after

oxidation and heating and the nature of their alteration due to these processes, (b) assess the occurrence and distribution of organic compounds in the waste in relation to the petrographic composition and the distance from the heat source, and (c) establish whether ratios of geochemical biomarker contents, normally used as maturity parameters, can be applied in the present instance, to assess the thermal changes caused by the waste dump fire. Since the phenomenon is accompanied by organic matter pyrolysis close to the fire zone, an attempt to estimate temperature ranges is made. 2. Sample collection and analytical methods Pieces of coal waste were collected from the coal waste dump in Piekary Śląskie town (Upper Silesia, Poland). The waste deposited in the Piekary Śląskie dump originates mainly from the closed Rozalia Mine where, in the last years of its operation, the Poręba, Anticlinal and Ruda Beds were exploited. These beds contained coals with a reflectance Rr between 0.5 and 1.0%. Twelve samples were collected at various distances (0.05 m, 0.2 m, 0.5 m, 0.6 m, 0.75 m, 1 m and N4 m) from the heat source. At each sampling point, a sample of coal waste approximately 0.5 kg in weight was collected. Care was taken, especially at locations close to the heat source, to avoid mixing samples exhibiting different levels of thermal transformations. Ultimate and proximate analyses were carried out on all samples in the laboratory of the Main Mining Institute in Katowice (Poland) according to Polish Standards and their ISO equivalents. Samples for petrographic analyses were prepared according to the procedure described in ISO 7404-2 (1985). Petrographic analyses carried out on particulate pellets included the content determination of all three maceral groups (vitrinite, liptinite, inertinite and their forms of alteration), mineral matter and coke; they were carried out in the Faculty of Earth Sciences, University of Silesia in Sosnowiec. In each sample, the petrographic composition was determined at 500 points using an Axioplan II optical microscope at a magnification of 500X according to the procedure described in ISO 7404-3 (1994). Reflectance measurements were carried out at 100–500 points on each sample at a magnification of 500X according to the procedure described in ISO 7404-5 (1994). The random and the maximum reflectance values are the mean statistical values of the measurements taken. The type of measurement depended on the rank of the organic matter and its level of transformation. In most samples (1, 3, 6–

M. Misz et al. / International Journal of Coal Geology 71 (2007) 405–424

12), random reflectance was measured. On samples 1, 3, 6–10 the reflectance analyses were carried out on vitrinite (collotelinite), while on samples 11 and 12, devoid of vitrinite, they were carried out on coke. On samples 1 and 3, in addition to Rr measured on vitrinite particles, maximum reflectance was also measured. In the highly altered samples (2, 4, 5), devoid of vitrinite, maximum reflectance was measured on coke particles. Geochemical analysis involved ultrasonic solvent extraction of powdered samples and the analysis of the total extracts by gas chromatography–mass spectrometry (GC–MS). About 20 g of each sample, powdered to a grain size of b0.2 mm, were extracted in dichloromethane for about 20 min in an ultrasonic bath. In each case, this procedure was repeated 3–4 times in order to obtain an extract. All extracts were collected, evaporated and weighed. The extracts were not separated into compound groups due to the very low extractability of some of the samples. An Agilent gas chromatograph 6890 with a DB — 35 column (60 m × 0.25 mm i.d.), coated by a 0.25 μm stationary phase film coupled with an Agilent Technology mass spectrometer 5973 was used. The experimental conditions were as follows: carrier gas — He; temperature — 50 °C (isothermal for 2 min); heating rate — up to 175 °C at 10 °C/min, to 225 °C at 6 °C/min and, finally, to 300 °C at 4 °C/min. The final temperature (300 °C) was held for 20 min. The mass spectrometer was operated in the electron impact ionisation mode at 70 eV and scanned from 50–650 da. Data were acquired in a full scan mode and processed with the Hewlett Packard Chemstation software. All compounds were identified by using their mass spectra, comparison of peak retention times with those of standard compounds, interpretation

407

of MS fragmentation patterns and literature data (Philp, 1985; Wiley/NBS Registry of Mass Spectral, 2000). Geochemical parameters were calculated using peak areas acquired in the manual integration mode. 3. Results 3.1. Macroscopic characterization Most of the samples contain visible layers and lenses of black organic matter displaying a strong lustre. The layers and lenses range from a few cm to about 10 cm in length and up to a few mm in thickness. In many cases, the organic matter is cracked and shows signs of heating, e.g., cracks parallel to bedding (samples 1–3, 5, 9), cracks on the surface of organic matter (samples 1–3, 5) and coking. Other samples are red, pinkish or dark grey due to intense thermal transformation or even show signs of slagging (samples 11, 12). These samples are hard and solid and do not show any visible organic matter. Uniquely, sample 10 is characterized by a strong hydrocarbon smell and leaves a yellowish stain on blotting paper. None of the other samples exhibited these characteristics. 3.2. Proximate and ultimate analyses The results of ultimate and proximate analyses, supported by the petrographic composition and reflectance data, show that samples 1–3, 5, 6, 9 are humic coals, sample 10 is a sapropelic shale and that the remainder are coaly waste material (Tables 1, 2a and 2b). The moisture of the samples varies from 0.09– 4.27 wt.%, the ash yields from 42.55–99.45 wt.%, and

Table 1 Proximate and ulimate analyses of coal waste samples Sample no.

Type of sample

Wa (wt.%)

Aa (wt.%)

VMdaf (wt.%)

Qdaf s (MJ/kg)

dr (g/cm3)

Cdaf (wt.%)

Hdaf (wt.%)

Ndaf (wt.%)

Odaf (wt.%)

Sat (wt.%)

Distance from the heat source [m]

1 2 3 4 5 6 7 8 9 10 11 12

Bituminous coal Bituminous coal Bituminous coal Waste coaly rock Bituminous coal Bituminous coal Waste coaly rock Waste coaly rock Bituminous coal Sapropelic shale Waste coaly rock Waste coaly rock

4.27 4.13 2.87 0.23 3.63 1.94 2.08 2.48 3.47 0.59 0.09 0.20

48.25 43.96 49.73 98.38 45.91 44.22 53.20 54.35 42.55 69.71 99.19 99.45

3.45 2.18 8.21 52.50 5.07 41.30 35.08 40.34 43.45 57.64 68.06 8.57

23.95 33.01 28.66 32.68 24.42 29.28 33.18 25.17 25.35 29.46 32.99 37.27

2.17 1.78 2.03 2.16 1.95 1.82 2.19 1.98 1.81 2.08 2.34 2.47

93.15 95.93 91.95 57.55 94.60 75.59 72.16 69.25 67.97 72.12 53.82 67.57

0.21 0.64 0.97 6.18 0.67 3.69 5.46 4.42 3.50 7.91 7.28 8.57

1.24 0.00 1.46 1.86 1.07 1.35 2.08 1.67 1.65 1.62 2.33 1.20

3.24 1.50 3.99 31.53 2.50 19.57 19.52 23.44 25.58 17.37 31.02 14.09

1.06 1.00 0.85 0.04 0.76 0.34 0.35 0.57 0.71 0.32 0.04 0.03

0.05 0.20 0.50 0.60 0.75 1.00 2.00 N4.00 N4.00 N4.00 N4.00 N4.00

W — moisture, A— ash, VM — volatile matter, Q — calorific value, dr — density, C — carbon, H — hydrogen, N — nitrogen, O — oxygen, S — sulphur, a — air dry basis, daf — dry ash free basis, t— total.

408

M. Misz et al. / International Journal of Coal Geology 71 (2007) 405–424

Table 2a Petrographic properties of coal waste samples Sample number

Unaltered vitrinite (vol.%) Vitrinite with cracks (vol. %) Oxidised vitrinite (vol. %) Oxidised vitrinite with cracks (vol. %) Vitrinite with oxidation rims (vol. %) Plasticised vitrinite (vol. %) Plasticised vitrinite with oxidation rims (vol. %) Vitrinite surrounding other particles (vol. %) Sum of vitrinite (vol. %) Unaltered liptinite (vol. %) Altered liptinite (vol. %) Sum of liptinite (vol. %) Unaltered inertinite (vol. %) Altered inertinite (vol. %) Sum of inertinite (vol. %) Porous coke (vol. %) Massive coke (vol. %) Sum of coke (vol. %) Unaltered mineral matter (vol. %) Altered mineral matter (vol. %) Sum of mineral matter (vol. %) Rr vitrinite (%) Rr coke (%) Rmax coke (%) Distance from the heat source (m)

1

2

3

4

5

6

7

8

9

10

11

12

0.0 0.0 2.2 0.0 0.0 4.0 0.0 0.0 6.2 0.0 0.0 0.0 0.2 1.2 1.4 64.6 25.4 90.0 0.0 2.4 2.4 2.36 – 6.57 0.05

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.8 1.8 77.8 20.0 97.8 0.0 0.4 0.4 – – 6.47 0.20

31.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 31.4 6.4 0.0 6.4 1.8 0.0 1.8 16.0 38.2 54.2 6.2 0.0 6.2 0.76 – 6.32 0.50

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.0 0.2 0.0 1.4 1.4 0.0 98.4 98.4 – – 5.57 0.60

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 5.2 2.0 7.2 25.2 32.6 57.8 31.0 4.0 35.0 – – 6.27 0.75

32.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 32.2 17.2 0.0 17.2 16.0 0.0 16.0 0.0 0.0 0.0 34.6 0.0 34.6 0.58 – – 1.00

12.6 0.4 0.0 0.0 3.2 1.6 3.8 1.6 23.2 5.8 3.6 9.4 12.8 0.0 12.8 0.0 0.0 0.0 54.0 0.6 54.6 0.69 – – 2.00

12.8 0.8 7.2 1.2 2.6 6.0 1.0 2.2 33.8 4.8 2.4 7.2 25.4 0.0 25.4 0.0 0.0 0.0 32.8 0.8 33.6 0.90 – – N4.00

23.2 0.4 32.4 0.0 6.8 0.0 0.0 0.0 62.8 7.6 1.4 9.0 3.6 0.0 3.6 0.0 0.0 0.0 24.2 0.4 24.6 0.87 – – N4.00

10.8 0.6 0.0 0.0 0.0 0.0 0.0 0.0 11.4 8.0 0.8 8.8 3.6 0.0 3.6 0.0 0.0 0.0 76.2 0.0 76.2 0.62 – – N4.00

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.2 0.4 1.6 0.0 0.4 0.4 0.0 98.0 98.0 – 2.43 – N4.00

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.4 0.0 0.4 0.0 0.6 0.6 0.0 99.0 99.0 – 2.60 – N4.00

– not determined. Table 2b Petrographic properties of coal waste samples calculated on a mineral matter free basis Sample number

Unaltered vitrinite (vol. %) Vitrinite with cracks (vol. %) Oxidised vitrinite (vol. %) Oxidised vitrinite with cracks (vol. %) Vitrinite with oxidation rims (vol. %) Plasticised vitrinite (vol. %) Plasticised vitrinite with oxidation rims (vol. %) Vitrinite surrounding other particles (vol. %) Sum of vitrinite (vol. %) Unaltered liptinite (vol. %) Altered liptinite (vol. %) Sum of liptinite (vol. %) Unaltered inertinite (vol. %) Altered inertinite (vol. %) Sum of inertinite (vol. %) Porous coke (vol. %) Massive coke (vol. %) Sum of coke (vol. %) Rr vitrinite (%) Rr coke (%) Rmax coke (%) Distance from the heat source (m) – not determined.

1

2

3

4

5

6

7

8

9

10

11

12

0.0 0.0 2.3 0.0 0.0 4.1 0.0 0.0 6.4 0.0 0.0 0.0 0.2 18.9 1.4 66.2 26.0 92.2 2.36 – 6.57 0.05

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.8 1.8 78.1 20.1 98.2 – – 6.47 0.20

33.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 33.5 6.8 0.0 6.8 1.9 0.0 1.9 17.1 40.7 57.8 0.76 – 6.32 0.50

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 12.5 0.0 12.5 0.0 87.5 87.5 – – 5.57 0.60

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 8.0 3.1 11.1 38.8 50.2 88.9 – – 6.27 0.75

49.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 49.2 26.3 0.0 26.3 24.5 0.0 24.5 0.0 0.0 0.0 0.58 – – 1.00

27.8 0.9 0.0 0.0 7.0 3.5 8.4 3.5 51.1 12.8 7.9 20.7 28.2 0.0 28.2 0.0 0.0 0.0 0.69 – – 2.00

19.3 1.2 10.8 1.8 3.9 9.0 1.5 3.3 50.9 7.2 3.6 10.8 38.3 0.0 38.3 0.0 0.0 0.0 0.90 – – N4.00

30.8 0.5 43.0 0.0 9.0 0.0 0.0 0.0 83.3 10.1 1.9 11.9 4.8 0.0 4.8 0.0 0.0 0.0 0.87 – – N4.00

45.4 2.5 0.0 0.0 0.0 0.0 0.0 0.0 47.9 33.6 3.4 37.0 15.1 0.0 15.1 0.0 0.0 0.0 0.62 – – N4.00

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 60.0 20.0 80.0 0.0 20.0 20.0 – 2.43 – N4.00

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 40.0 0.0 40.0 0.0 60.0 60.0 – 2.60 – N4.00

M. Misz et al. / International Journal of Coal Geology 71 (2007) 405–424

409

Fig. 1. Relationship between carbon content and various chemical parameters. A — volatile matter, B — hydrogen, C — oxygen, D — nitrogen.

the calorific values from 23.95–37.27 MJ/kg (Table 1). Carbon contents range from 53.82–95.93 wt.%, hydrogen from 0.21% to 8.57 wt.%, nitrogen from 0.00– 2.33 wt.%, oxygen from 1.50–31.53 wt.% and total sulphur varies from 0.03–1.06 wt.%. Sample densities range between 1.78–2.47 g/cm3. The volatile matter content fluctuates between 2.18– 68.06 wt.%. Sample 11 contains the highest volatile matter and the lowest carbon content. In contrast, sample 2 has the lowest volatile matter and the highest carbon content. The samples show good negative correlations between carbon and volatile matter content (correlation coefficient r = − 0.84), as well as between the contents of carbon and hydrogen (r = − 0.86), oxygen (r = − 0.96) and nitrogen (r = −0.75) (Fig. 1). The heating influence is seen in the contents of carbon, hydrogen, oxygen and ash. Samples collected within 1 m of the heat source reveal higher carbon contents and lower hydrogen and oxygen contents (Table 1). The correlation coefficients (r) between the distance from the heat source and the carbon, hydrogen and oxygen contents are − 0.67, 0.73 and 0.58, respectively (Fig. 2). Samples obtained from distances greater than 1 m from the heat source typically have higher hydrogen and oxygen contents and lower carbon contents. This pattern of gain and loss directly reflects increasing carbonization and indicates that, at distances smaller than 1 m from the heat source, the heat was the dominant factor of transformations.

Fig. 2. Relationship between distance from the heat source and selected chemical parameters. A — carbon, B — hydrogen, C — oxygen.

410

M. Misz et al. / International Journal of Coal Geology 71 (2007) 405–424

3.3. Petrographic analyses All three maceral groups (vitrinite, liptinite and inertinite), mineral matter and coke were determined (Tables 2a and 2b). In vitrinite, the following types of changes were recognized: vitrinite with cracks, oxidised vitrinite, oxidised vitrinite with cracks, vitrinite with oxidation rims, plasticised vitrinite with and without oxidation rims and vitrinite surrounding other particles. Unaltered and altered particles of liptinite and inertinite macerals were determined as were particles of mineral matter and of porous and massive coke particles (Tables 2a and 2b). Samples 1–3, and 5 containing the highest amounts of coke, also reveal relatively low mineral matter contents. From all these, liptinite was observed only in sample 3, which is also characterized by a relatively low reflectance (0.76%) indicating that only part of the sample had been strongly affected by heat. The vitrinite content in this sample is 31.4 vol.%, while the inertinite content is low (∼2 vol.%). The remaining samples (1, 2, 5) are characterized by high reflectance values (2.36–

6.47%). In addition to the dominating coke, the four samples also contain up to 7 vol.% inertinite. Vitrinite (6.2 vol.%) occurs only in sample 1 in form of plasticised vitrinite and oxidised vitrinite. In samples with the highest coke contents (samples 1 and 2), porous coke prevails over massive coke. In samples 3 and 5, massive coke prevails over porous coke (Tables 2a and 2b). Exclusively massive coke also occurs in small amounts (b 1.4 vol.%) in samples 4, 11 and 12. These samples reveal the highest mineral matter (98–99 vol.%) and up to 1.6 vol.% inertinite contents. The absence of vitrinite and liptinite indicates severe transformation. In general, the coke contents decrease with increasing distance from the heat source (Fig. 3A). The remaining samples (6–10) are characterized by random reflectances in the range between 0.58–0.90% (Tables 2a and 2b). They contain all three maceral groups but not coke. Vitrinite (b62.8 vol.%) is the dominating maceral group (Tables 2a and 2b) followed by inertinite (b 25.4 vol.%) and liptinite (b 17.2 vol.%). In samples 6–8 and 10, unaltered vitrinite (10.8– 32.2 vol.%) is the dominating form, whereas in sample

Fig. 3. Relationship between coke content and various chemical parameters. A — distance from the heat source, B — volatile matter, C — carbon, D — hydrogen, E — oxygen, F — total sulphur.

M. Misz et al. / International Journal of Coal Geology 71 (2007) 405–424

9, oxidised vitrinite (32.4 vol.%) prevails over other vitrinite forms. Oxidised vitrinite is also important in sample 8. In samples 7–9, vitrinite with oxidation rims is a component (2.6–6.8 vol.%) as are both plasticised

411

vitrinite and plasticised vitrinite with oxidation rims in sample 7 (1.6 and 3.8 vol.%, respectively) and in sample 8 (6.0 and 1.0 vol.%, respectively). Vitrinite enclosing other particles is also significant in samples 7

Fig. 4. Vitrinite reflectograms. A — sample 3, B — sample 6, C — sample 10, D — sample 7, E — sample 9, F — sample 1.

412

M. Misz et al. / International Journal of Coal Geology 71 (2007) 405–424

tain any liptinite. Coke contents (0.4 to 97.8 vol.%) are inversely related to mineral matter contents. Samples rich in mineral matter (4, 11 and 12) contain less than 2 vol.% coke whereas samples (1 and 2) with less than 3 vol.% mineral matter contain over 90 vol.% coke. The contents of the microscopic components were compared with the results of the proximate and ultimate analyses. Many of the chemical parameters correlated well with the coke content (Fig. 3B–F). The correlation coefficients for the contents of coke, volatile matter, carbon, hydrogen, oxygen and total sulphur are, respectively,− 0.72, 0.92,− 0.93,− 0.85 and 0.98. Reflectance values ranges from 0.58–6.57%. In general, samples collected within a distance of 1 m from the heat source show the highest reflectance values (Tables 2a and 2b). The differing shapes of the reflectograms for individual samples provide an indication of the changes that the samples have undergone. Some reflectograms (samples 3, 6 and 10) are narrow

(1.6 vol.%) and 8 (2.2 vol.%). Altered inertinite particles occur only in samples with the highest coke contents (samples 1, 2 and 5). Liptinite particles are found in samples 3, 6–10; in samples 3 and 6, these are invariably unaltered, whereas in samples 7–10, unaltered liptinite occurs in higher quantity than altered liptinite. Mineral matter ranges from about 24.6–76.2 vol.% in samples 6–10 with the greatest amount characterising the sapropelic shale (sample 10). This shale is also characterized by the highest liptinite content (37.0 vol.%, mmf ) and by a significant presence (47.9 vol.%, mmf) of vitrinite (Table 2b). In samples 3, 6 and 10, mineral matter is present exclusively in an unaltered form and, in samples 1, 2, 4, 11 and 12, exclusively in an altered form. The remaining samples (5, 7–9) contain both mineral forms though unaltered forms are the more prominent. Samples with mineral matter present only in unaltered form are characterized by relatively low reflectances and, those with exclusively altered mineral matter, by relatively high reflectances. The latter also do not con-

Table 3 Aliphatic and aromatic biomarker parameters showing the thermal influence of waste dump fires on coal organic matter Sample Extract yields CPI (wt.%) 1) 1 2 3 4 5 6 7 8 9 10 11 12

0.2 0.2 0.3 0.5 0.6 6.1 11.0 0.7 5.2 3.5 0.4 0.2

Σ2/Σ1 Pr/Ph Pr/n-C17 C31αβ22S/ Ts/(Ts + MNR MPI-3 P/A (22S + 22R) Tm)

2-MP/2-MA (3MB + Σ DMB/ 4 MB)/DBF Σ MDBF

2)

1.06 0.49 1.21 1.18 0.92 0.91 1.19 1.58 1.09 4.44 1.39 0.56 1.14 4.70 1.09 17.50 1.07 6.64 1.17 0.43 1.12 0.08 1.16 1.49

3)

4)

5)

6)

7)

8)

9)

10)

11)

12)

1.41 0.65 2.33 1.40 1.01 5.89 1.40 3.60 1.81 2.03 4.29 1.29

0.42 0.96 0.39 0.41 1.02 0.58 0.18 0.45 0.52 0.64 0.23 1.43

– – – –13) – 0.56 0.53 0.55 0.58 0.50 0.60 0.63

– – – –13) – 0.89 0.85 0.84 0.87 0.80 0.90 0.86

– – – – – 1.00 1.22 1.06 0.69 1.28 1.23 –

– – – – 1.05 1.04 1.51 0.84 0.87 0.85 1.68 1.10

14.67 3.37 6.17 3.75 14.43 12.50 5.15 6.00 4.86 5.10 9.8 32.7

– – – – 8.33 2.32 2.69 1.63 2.06 3.73 10.29 –

1.40 – 0.41 2.06 0.62 0.56 0.20 0.34 0.89 0.68 0.98 0.34

– – 0.71 1.82 1.06 0.99 0.45 0.78 0.84 – 1.12 –

1) = 0,5{[(n-C25 + n-C27 + n-C29 + n-C31 + n-C33)/(n-C24 + n-C26 + n-C28 + n-C30 + n-C32)] + [(n-C25 + n-C27 + n-C29 + n-C31 + n-C33)/(n-C26 + n-C28 + nC30 + n-C32 + n-C34)]; Carbon Preference Index; m/z = 71; thermal maturity parameter (Bray and Evans, 1963). 2) Σ2/Σ1 = [Σ (from n-C13 to n-C22)]/[Σ (from n-C23 to n-C35) ]; m/z = 71, source indicator. 3) Pr/Ph = pristane/phytane; parameter of environment oxicity (with exception of coals); m/z = 71 (Didyk et al., 1978). 4) Pr/n-C17 = pristane/n-heptadecane; m/z = 71 (Didyk et al., 1978). 5) C31αβ22S/(22S + 22R) = 17α(H),21β(H)-29-homohopane 22S/(17α(H),21β(H)-29-homohopane 22S + 17α(H),21β(H)-29-homohopane 22R); m/z = 191; thermal maturity parameter (Peters et al., 2005). 6) Ts/(Ts + Tm) = 18α(H)-22,29,30-trisnorneohopane/(18α(H)-22,29,30-trisnorneohopane + 17α(H)-22,29,30-trisnorhopane); m/z = 191; thermal maturity parameter (Peters et al., 2005). 7) MNR = 2-methylnaphthalene/1-methylnaphthalene; m/z = 142; thermal maturity parameter (Radke et al., 1994). 8) MPI-3 = (2-methylphenanthrene + 3-methylphenathrene)/(1-methylphenathrene + 9-methylphenathrene); m/z = 192; thermal maturity parameter (Radke and Welte, 1983). 9) P/A = phenanthrene/anthracene; m/z = 178. 10) 2-MP/2-MA = 2-methylphenanthrene/2-methylanthracene; m/z = 192. 11) (3MB + 4MB)/DBF = (3-methylbiphenyl + 4-methylbiphenyl)/dibenzofurane; m/z = 168; thermal maturity parameter (Radke, 1987). 12) Σ DMB/ Σ MDBF = a sum of dimethylbiphenyls/a sum of methyldibenzofuranes; m/z = 182; thermal maturity parameter. 13) “–” compounds present, concentrations too low to calculate a parameter value.

M. Misz et al. / International Journal of Coal Geology 71 (2007) 405–424

(b3.5 v-classes) indicating limited organic matter change (Fig. 4); the random reflectance (Rr) of these samples varies between 0.58–0.76% (Tables 2a and 2b). In these cases, the reflectance was measured on vitrinite only. Reflectograms for samples 1, 7–9 are wider, suggesting greater heterogeneity. For samples 2, 4 and 5, mean maximum reflectance values are given on Tables 2a and 2b and, for samples 11 and 12, Rr measured on coke particles are likewise given. In addition to Rr measurements only on vitrinite particles in sample 1 and 3, the maximum reflectance were measured on coke particles present in those samples. The mean maximum reflectance is 6.57% and 6.32%, respectively (Tables 2a and 2b). 3.4. Gas chromatography–mass spectrometry analysis Sample extractabilities vary depending on the thermal influence (0.2–11.0 wt.%). Samples 6 (6.1 wt.%), 7

413

(11.0 wt.%), 9 (5.2 wt.%) and 10 (3.5 wt.%) yielded the highest extract amounts. For the remainder, extract yields were very low at about 0.2–0.5 wt.% (Table 3). Several groups of compounds were found in the waste-material extracts. The aliphatic hydrocarbon group is dominated by n-alkanes (m/z = 71) together with n-alkenes (m/z = 69) present in some of the extracts. Generally, three types of n-alkane distributions can be distinguished (Fig. 5): 1) Bimodal with almost equal contents of short-chain (n-C13–n-C22) and long-chain (n-C23–n-C36) compounds (samples 1, 4, 6), 2) Monomodal dominated by short-chain n-alkanes (nC15–n-C16), similar to low-temperature coal tars (samples 10 and 11; see also Fabiańska and Matuszewska, 1998), 3) Monomodal, long-chain n-alkanes (n-C23–n-C36) dominated (samples 2, 3, 5, 7, 8, 9 and 12).

Fig. 5. Main types of n-alkane distribution in the extracts of samples from the Piekary Śląskie coal waste dump.

414

M. Misz et al. / International Journal of Coal Geology 71 (2007) 405–424

Differences in n-alkane distributions are expressed as a ratio of short- to long-chain n-alkanes (Table 3). The highest long-chain n-alkane dominance is shown in samples 8 and 9. A predominance of odd-over-even carbon number n-alkanes, commonly found in subbituminous and bituminous coals or in dispersed organic matter with terrestrial input, is not evident in most of the investigated extracts (see Carbon Preference Index values, Table 3). n-Alkenes (m/z = 69) were found in samples 1 and 11. Their distributions roughly follow the n-alkane envelope with a slight predominance of even carbon-number compounds. Their presence testifies to pyrolysis (Solomon et al., 1992; Love et al., 1995; Fabiańska and Matuszewska, 1998). Other compounds identified include pentacyclic triterpanes (hopanes and moretanes), methyl- and dimethylbiphenyls, dibenzofurane and methyldibenzofuranes. The hopane distribution in the range of C29–C33 compounds is typical for organic matter that reached the mid-catagenesis stage at least. The hopane distribution shows a high concentration of 17α(H),21β(H)-hopanes (geodiastereomers), an absence of 17β(H),21β(H)hopanes (biodiastereomers) and a concentration of 18α(H)-22,29,30-trisnorneohopane (Ts) over 17α(H)22,29,30-trisnorhopane (Tm) (Peters et al., 2005); the Ts/(Ts + Tm) ratio has reached values higher than 80%. The values of C31αβ22S/(22S + 22R) ratio has attained its equilibrium (∼ 56%). Moretanes (17β(H),21α(H)hopane diastereomers) are also present in relatively high amounts (Fig. 6). A wide range of polycyclic aromatic hydrocarbons with 2–6 rings was identified in the extracts together with their aliphatic derivatives. Sample 6 is the

richest in these compounds with naphthalene, phenanthrene, anthracene, 2-phenylnaphthalene, 1-phenylnaphthalene, retene, fluoranthene, pyrene, benzo(c) phenanthrene, benzo(a)anthracene, chrysene, benzo(b) fluoranthene, benzo(k)fluoranthene, benzo(a)fluoranthene, benzo(e)pyrene, benzo(a)pyrene, perylene and benzo(ghi)perylene present (Fig. 7). Samples 1–4 and 10, especially those with higher vitrinite reflectance values, contain much lower amounts of these compounds. Relative percentage contents of selected polycyclic aromatic hydrocarbons were calculated to reveal changes caused by heat (Fig. 8). Generally, it is phenanthrene that occurs in the highest concentration in all samples, whereas naphthalene and five- and six-ring PAHs show the lowest contents in all samples. To assess the extent of thermal change in organic matter caused by fire, values of several thermal maturity biomarker parameters were calculated based on both aliphatic and aromatic hydrocarbons (see Bray and Evans, 1963; Radke and Welte, 1983; Radke et al., 1986; Radke, 1987; Radke et al., 1994; Peters et al., 2005; see Table 3). Only those parameters were selected for which appropriate compounds were present in most of the extracts. 4. Discussion 4.1. Petrography Petrographic changes in organic matter, particularly in the vitrinite group, reflect oxidation and/or heating (Stach et al., 1982). Though some samples (e.g., 2, 4, 5,

Fig. 6. Distribution of hopanes and moretanes in sample 7; m/z = 191.

M. Misz et al. / International Journal of Coal Geology 71 (2007) 405–424

415

Fig. 7. Distribution of polycyclic hydrocarbons (PAHs) in sample 6; the combined ion chromatograms m/z = 128 + 178 + 202 + 204 + 228 + 252 + 276.

11, 12) are composed entirely of transformed organic matter, and others (sample 6) of completely unchanged organic matter, altered and unaltered particles may also occur together as in samples 1, 3, and 7–10 (Ćmiel and Misz, 2005). Non-altered particles are grey in colour and do not show cracks, pores or oxidation rims. These typically comprise a single maceral, e.g., vitrinite, inertinite, or various macerals and/or inorganic matter, e.g., trimacerites (Fig. 9A) and carbominerites. Some vitrinite particles show cracks (Fig. 9B) and oxidation rims (Fig. 9C). These are formed at temperatures of 200 °C (Calemma et al., 1995; Ndaji and Thomas, 1995) or caused by oxygen diffusion (Calemma et al., 1995). In a number of cases, the oxidation rims are zoned (Fig. 9D). The zone nearest the unaltered particle centre has the lowest reflectance and the darkest colour of all, while the outermost zone is characterized by the highest reflectance and the lightest colour. This zonal

distribution could be an indication of several heating periods. Altered organic matter may be defined by rims surrounding particles of trimacerite, carbominerite or mineral matter. In some cases, these grey–white rims are smooth and of constant width (Fig. 9E), in others the outer edge is wavy (Fig. 9F). These rims are probably made of vitrinite that has passed through a plastic stage (Ćmiel and Misz, 2005). Alteration involving the development of pores may also reflect particle plasticity that takes place at a temperature range between 350–500 °C (Speight, 1994; Taylor et al., 1998). Pores form when volatiles are released in vitrinite particles exposed to short-term heating (Tsai and Scaroni, 1987; Bailey et al., 1990; Rosenberg et al., 1996; Taylor et al., 1998). Commonly, pores are also surrounded by pale rims (Fig. 9G). Similar pores also occur in particles composed of other macerals and/or mineral matter (Fig. 9H).

416

M. Misz et al. / International Journal of Coal Geology 71 (2007) 405–424

Fig. 8. Relative concentrations of selected polycyclic aromatic hydrocarbons in the coals affected by fire in the Piekary Śląskie coal waste dump.

Coke is another form of transformed organic matter typically containing pores of various sizes and shapes (Goodarzi, 1987; Karayigit and Whateley, 1997; Taylor et al., 1998; Thrope et al., 1998; Kwiecińska and Petersen, 2004). Some particles are highly porous (Fig. 9I), others are dense (Fig. 9J). Layers of porous and massive coke (Fig. 9K), and coked matter occurring as thin lenses of varying length and high reflectance dispersed within mineral matter (e.g. sample 5), probably reflect primary rock layering. Coke formation is connected with temperatures above 500 °C (Speight, 1994; Taylor et al., 1998). Spherical hydrocarbon droplets (1–3 μm) occurring in clusters in the sapropelic shale (sample 10) also reflect heating. The hydrocarbons were probably generated from liptinite (Fig. 9L). 4.2. Geochemistry The chemical composition of the organic extracts shows evidence of thermal alteration. Most commonly found compounds such as hopanes, methylnaphthalenes, dimethylnaphthalenes and methylphenanthrenes are absent in samples more thermally affected. Generally, these compounds commonly occur in bituminous coals even of higher rank (Willsch and Radke, 1995). However, it has been shown that pyrolysis over a long time can cause their destruction (Lu and Kaplan, 1992).

Presumably, the aforementioned compounds were destroyed as a result of high thermal stress or they evaporated. The distribution of other compounds such as n-alkanes, has been altered as a result of pyrolysis. The geochemical parameters based on biomarkers show variations ranging between values indicating the beginning of catagenesis – equivalent to coals mined in the region – in relatively unaltered material (sample 6) up to values characterising samples thermally altered to a high degree (sample 2). In the case of some parameters such as CPI, Pr/Ph or Pr/n-C17, it is difficult to discern any significant trend due to the high degree of thermal alteration of the organic matter. Values of Σ2/Σ1 (Table 3) reflect two processes resulted from the thermal influence of fire — the expulsion of aliphatic compounds from organic-matter macromolecules and the thermal destruction (or evaporation) of n-alkanes starting with lighter short-chain compounds. The beginning of the process is best seen in sample 10 extract with its wide band of n-C13–n-C22 in high concentration (Σ2/Σ1 = 0.43). Subsequently, with increasing temperature, these compounds were removed by evaporation leading to n-alkane distribution of monomodal type with long-chain n-alkanes (n-C23–nC36) such as in samples 3, 7–9 (Table 3). This trend is similar to that determined for coal pyrolysates from Upper Silesia Coal Basin, Poland (Fabiańska and Matuszewska, 1998). In these samples the higher the

M. Misz et al. / International Journal of Coal Geology 71 (2007) 405–424

417

Fig. 9. Microphotographs of altered organic matter from the Piekary Śląskie coal waste dump. A — non-altered trimacerite particle, B — vitrinite particle with cracks linked to oxidation, C — oxidation rim in vitrinite particle, D — zoned oxidation rim, E — rim of constant thickness surrounding minerogenic particle, F — wavy vitrinite rim surrounding minerogenic particle, G — vitrinite particle with oxidation rim, H — plastically-deformed trimacerite particle, I — porous coke particle, J — massive coke particle, K — layers of porous and massive coke, L — hydrocarbon droplets exhibiting fluorescence in sapropelic shale.

vitrinite reflectance, the greater the predominance of long-chain n-alkanes. It may be assumed that cracking of long-chain n-alkanes did not occur in these samples or that it played only a minor role as n-alkenes, the major cracking products, are absent in their extracts.

Samples 1 and 2 with high Rr values of 2.36% and 6.47% respectively, probably contain some bitumen substances expelled from nearby material of lower maturity. Extracts from sample 5 at a distance of 0.75 m from the fire site, and from samples 11 and 12 at

418

M. Misz et al. / International Journal of Coal Geology 71 (2007) 405–424

Fig. 9 (continued ).

N4.00 m, probably contain significant quantities of substances evaporated from organic matter near the fire site that would tend to precipitate on colder more distal material. The petrographic compositions of these last three samples indicate that they could not have been the source of these evaporated/precipitated extract compounds; these samples comprise mostly coke and small

amounts of inertinite. Moreover, the extracts of these samples contain n-alkenes formed as a result of organic matter cracking and/or thermal dehydrogenation of nalkanols (Huizinga et al., 1987; Solomon et al., 1992; Oros and Simoneit, 2000; Simoneit, 2002). In these samples, n-alkene distribution roughly follows n-alkane distribution as is commonly reported for pyrolysates of

M. Misz et al. / International Journal of Coal Geology 71 (2007) 405–424

organic matter obtained in laboratory experiments (Love et al., 1995; Han and Kruge, 1999; Leif and Simoneit, 2000). Sample 11 extract is dominated by lighter compounds (for example, Σ2/Σ1 = 0.08) evaporated at relatively low temperatures; compounds with higher boiling points are absent. The sample 12 bitumen was exposed to higher temperatures because it is relatively enriched in much heavier compounds (especially longchain n-alkanes) and lacks less thermally resistant compounds such as hopanes and alkylnaphthalenes. Since the coal waste contained organic matter of relatively high maturity (catagenic stage at least) even prior to the fire event, Carbon Preference Index (CPI) values are all near 1.00. However, sample 6 (CPI = 1.39) shows a slight odd-over-even carbon number predominance not found in samples more intensively heated (Fig. 5). This feature vanished as a result of heating and macromolecule cracking which expelled n-alkanes showing the smooth distribution envelope (Peters et al., 2005). Pristane and phytane concentrations follow the changes in n-alkane distribution being reflected by Pr/n-C17 and Pr/Ph ratios (Table 3). The uniform absence of steranes and diterpanes in all extracts may indicate that, despite large differences in chemical composition and physico-chemical properties, most of the primary organic (coal) material dumped in the waste had features in common, especially those related to their biogenic origin and depositional environment. Diterpenes are considered indicators of conifer input into sedimentary organic matter (Noble et al., 1985; Isaksen, 1993; Otto and Simoineit, 2001). They have been found in coals of various ages, including Pennsylvanian coals where they are considered to have been synthesized by early gymnosperms (Schultze and Michaelis, 1990; Fabiańska et al., 2003). Pentacyclic triterpane distributions are very similar in all of the extracts in which these compounds occur; the highest peaks are for 17α(H),21β(H)-hopane (C30) and a series of 22R and 22S distereomers from 17α(H),21β(H)29-homohopane (C31) to 17α(H),21β(H)-29-trishomohopane (C33). Despite the heating, moretanes are present in relatively high quantities. A comparable pattern of hopane distribution has been noted in pyrolysates of humic coals, even after heating for 1000 h (Lu and Kaplan, 1992). It may be concluded that, in the case of the investigated coal waste, a rise in temperature led to a similar simultaneous destruction of all pentacyclic triterpanes as none is present in samples with Rr N 2.36%. In contrast to this trend during natural maturation in a deposit moretanes are transformed to αβ-hopanes and, as a result they disappear from coal extracts or oils (Peters et al., 2005).

419

Except for sample 2 (Rr = 6.47%), ethylbiphenyls occur in all extracts. Dimetylbiphenyls are generally found in the samples of lower rank (Rr b 2.36%) only. PAHs distribution varies with heat influence. Five- and four-ring hydrocarbons only occur in samples 6–9 which have been less affected by fire. The ratio of phenanthrene to anthracene is lowest for samples 2, 4, 9 and 10. Since anthracene is formed only during incomplete combustion of organic matter or its pyrolysis (Simoneit, 1998; Oros and Simoneit, 2000; Simoneit, 2002), it may be concluded that this high concentration (ratio) reflects thermal alteration due to fire. The contamination of samples 1, 5, 6 and 12 by material less thermally changed is confirmed (as phenanthrene/anthracene values are too high). The ratio 2-methylphenanthrene/2-methylanthracene may likewise reflect alteration caused by fire. However, these compounds are less thermally stable and were not found in most of the highly altered coals (samples 1–4, 12). Both anthracene and 2-methylanthracene are also formed during natural fires (Simoneit, 1998) and are often found in coals rich in fusinite and semifusinite (Fabiańska and Kruszewska, 2003). Distributions of alkylnaphthalenes and alkylphenanthrenes are also affected by fire. As shown in Fig. 8, alkylnaphthalenes were all removed from the samples relatively early in the heating process; alkylnaphthalenes are absent in samples with Rr N 2.36%. As boiling points for methylnaphthalenes and dimethylnaphthalenes are 245 °C and 262 °C, respectively, these samples must have been heated to temperatures far exceeding 270 °C. In the case of sample 8 (Rr = 0.90%), distributions of alkylnaphthalenes are, except for the methylnaphthalenes, unsuitable for geochemical interpretation (Fig. 10). In this case, the temperature range was b200 °C. Geochemical parameters based on alkyl derivatives of aromatic hydrocarbons such as MNR, MPI-3, (3MB + 4MB)/DBF and Σ DMB/Σ MDBF seem to reflect the thermal influence of fire better than parameters based on aliphatic biomarkers. Other processes apart from fire may influence organic material in waste dumps. Weathering of organic matter and water washing may affect distributions of alkylnaphthalenes by removing lighter and more water soluble compounds (Palmer, 1993; Martinez and Escobar, 1995; Ahmed et al., 1999). Alkylphenanthrenes, with higher boiling points (data from ChemIdPlus), show greater resistance to heating than alkylnaphthalenes. In those samples not affected by fire (sample 6 — though collected close to the fire site) or samples only slightly affected, alkylphenanthrenes occur in much lower concentrations than phenanthrene (Fig. 11). At Rr = 0.90% (sample 8), alkylphenanthrenes were probably expelled from the macromolecule in

420

M. Misz et al. / International Journal of Coal Geology 71 (2007) 405–424

Fig. 10. Distribution of alkylnaphthalenes in organic material affected by waste dump fire (sample 8 and 6); the combined ion chromatograms m/z = 128 + 142 + 156 + 170.

significant amounts; their concentrations in the extract are seen to increase. However, with increasing thermal alteration (samples 1–3, 5, 11) they were destroyed. While they occur in samples 1 and 11 (2.36% and 2.43% Rr , respectively), their concentrations are too low to calculate the Methylphenanthrene Index (Radke and Welte, 1983). For samples 3, 7, 8 and 9, which probably contain only organic matter heated in situ and not evaporated, temperature ranges may be roughly estimated using the presence/absence of compounds found in their extracts. Sample 7 (Rr = 0.69%; 2.00 m from the fire site) contains only small quantities of methyl-, dimethylnaphthalenes and n-alkanes lighter than n-C16 (b.p. = 287 °C). However, the presence of phenanthrene (b.p. = 340 °C), and the lack of any obvious loss of alkylphenanthrenes, suggest that the temperature range was about 280– 330 °C. In the case of samples 8 and 9 (N 4.00 m from the fire zone), the temperature was about 20 °C lower as n-

C15 (b.p. = 268 °C) is present. Sample 3 probably reached a temperature in the range 290–350 °C, assuming no input of vaporised pyrolysates from nearby. For samples 1, 2 and 4 obtained closest to the fire zone, the estimation of the temperature range is more difficult; their extracts include compounds from both organic matter heated in situ and introduced re-precipitated pyrolysates. The presence of PAH markers may indicate that the upper temperature limit of pyrolysis did not exceed 450 °C for samples 1 and 4 (chrysene present, b.p. = 448 °C) and about 400 °C for sample 2 (pyrene present, b.p. = 404 °C, while benzo(a)anthracene absent, b.p. = 438 °C). Possible correlations between geochemical parameters of thermal alteration and petrographic and technological properties were examined to see if these parameters might be applied in the assessment and evaluation of alterations in coal organic matter caused by waste dump fires. Combining such parameters should be valid over relatively wide ranges of rank and should not

M. Misz et al. / International Journal of Coal Geology 71 (2007) 405–424

421

Fig. 11. Distribution of alkylphenanthrenes in organic material affected by waste dump fire (samples shown in the order of diminishing thermal influence from the most affected sample 11, intermediate sample 8 and unchanged sample 6); the combined ion chromatograms m/z = 178 + 192 + 206 + 220.

be sensitive to the influence of organic matter type as described, for instance by Radke et al. (1986). First of all, the study has shown that the Carbon Preference Index is too near to the end of its value range (1.00) to be successfully applied, as is the case for most of the aliphatic biomarker parameters. Potentially, the best geochemical parameters are those used for organic matter of advanced thermal alteration such as the ratio of 3-methylbiphenyl plus 4-methylbiphenyl to dibenzofurane and/or the sum of dimethylbiphenyls to a sum of methyldibenzofuranes (Fig. 12C). However, the plot of their values against reflectance (Fig. 12) indicates that samples 5, 11 and 12, with extracts composed entirely or

almost entirely of evaporated and precipitated pyrolysates, do not fit the correlation; the values of the geochemical parameters are too high for the measured Rr values. The C31αβ22S/(22S + 22R) and Ts/(Ts + Tm) values show the same trend, although these biomarkers occur only in a limited number of samples. The Methylnaphthalene Ratio (MNR) shows a significant correlation with the vitrinite content (r = 0.90). Several geochemical parameters such as C31αβ22S/ (22S + 22R), MNR and MPI-3, show high correlation coefficients with vitrinite or inertinite contents (r in the range 0.70–0.90). In particular, it is the relative concentrations of certain polycyclic aromatic

422

M. Misz et al. / International Journal of Coal Geology 71 (2007) 405–424

Fig. 12. Relationship between reflectance (Rr) and A — Ts/(Ts + Tm), B — C31αβ22S/(22S + 22R), C — Σ DMB/Σ MDBF, D — (3MB + 4MB)/ DBF, E — Perylene (%).

hydrocarbons which show negative or positive correlations with vitrinite (r = − 0.71 and −0.77 for naphthalene and fluoranthene concentrations, respectively, and r = 0.86 and 0.82 for perylene and benzo(e)pyrene). Four- and five-ring PAHs such as pyrene and benzo(a) anthracene, are well correlated with the coke content (r = 0.71 and 0.99, respectively). However, it should be noted that these parameters are of limited validity due to the small number of petrographically unaltered samples with a wide range of PAHs (only five samples 6–9 and 12; see Figs. 7 and 8).

5. Conclusions • In the waste material of the Piekary Śląskie minestone dumps, all three maceral groups are affected by oxidation and heating to various extents. Unaltered or altered particles may co-exist in some samples. • Samples collected in a distance N 1 m from the heat source contain all three maceral groups and do not tend to contain coke. • The influence of heating is seen in oxidation rims, cracks, pores and coke formation. Samples collected

M. Misz et al. / International Journal of Coal Geology 71 (2007) 405–424

•

• •

•

•

•

nearest to the heat source show the greatest alteration expressed in high reflectance and the coke content. Zonation of oxidation rims could be an indication of successive heating events. The reflectance of the investigated material ranges from 0.58–6.47%. Reflectogram shapes vary with the level of material transformation — being narrow for relatively unaltered material, wider for altered. Bitumen occurring in extracts originates from organic matter heated in situ and is a result of pyrolysate evaporation and precipitation elsewhere. The chemical composition of the waste material clearly reflects the thermal influence of fire. Less thermally stable compounds, such as lighter nalkanes, cyclic isoprenoids, methyl and dimethylnaphthalenes, methyphenanthrenes and five-ring polycyclic aromatic hydrocarbons, are destroyed or evaporated in the most affected material. Distributions of several compound groups are altered as a result of heating. Examples are n-alkanes and nalkenes formed by macromolecule cracking, acyclic isoprenoids, methylbiphenyls and dimethylbiphenyls. While biomarker parameters based on alkanes (CPI) are of limited use, others may be successfully applied to assess alterations in coal organic matter around waste dump fires, especially those with validity ranges corresponding to higher mature organic matter, e.g. (3methylbiphenyl + 4-methylbiphenyl)/dibenzofurane and Σdimethylbiphenyls/Σmethyldibenzofuranes. On the basis of compounds found in extracts, the alteration temperature ranges can be roughly estimated for samples containing organic matter heated in situ and not containing pyrolysates evaporated from organic matter elsewhere.

Acknowledgements The authors would like to acknowledge the two anonymous reviewers for their constructive remarks that helped to greatly improve the paper. They would also like to thank Dr. Pádhraig S. Kennan, University College Dublin, Ireland, for his help with the English. References Ahmed, M., Smith, J.W., George, S.C., 1999. Effects of biodegradation on Australian Permian coals. Organic Geochemistry 30, 1311–1322. Bailey, J.G., Tate, A., Diessel, C.F.K., Wall, T.F., 1990. A char morphology system with application to coal combustion. Fuel 69, 225–239. Banerjee, D., Hirani, M., Sanyal, S.K., 2000. Coal — quality deterioration in a coal stack of a power station. Applied Energy 66, 267–275.

423

Bray, E.E., Evans, E.D., 1963. Distribution of n-paraffins as a clue to recognition of source beds. Geochimica et Cosmochimica Acta 22, 2–15. Calemma, V., Del Piero, G., Rausa, R., Girardi, E., 1995. Changes in optical properties of coals during air oxidation at moderate temperature. Fuel 74, 383–388. Chandra, D., 1962. Reflectance and microstructure of weathered coals. Fuel 41, 185–193. ChemIdPlus (www.chem.sis.nlm.nih.gov/chemidplus/). Ćmiel, S., Misz, M., 2005. Petrographical changes caused by the fires of coal waste dumps. Proceedings of LXXVI Meeting of Polish Geological Institute, 43−50 (In Polish). Cygankiewicz, J., 1996. Estimation of the development of self-ignition centers on the basis of precise analysis of coal mine air samples. Scientific Papers of Main Mining Institute No. 14, pp. 505–530 (in Polish). Didyk, B.M., Simoneit, B.R.T., Brassell, S.C., Eglinton, G., 1978. Organic geochemical indicators of palaeoenvironmental conditions of sedimentation. Nature 272, 216–222. Fabiańska, M., Kruszewska, K.J., 2003. Relationship between petrographic and geochemical characterisation of selected South African coals. International Journal of Coal Geology 54, 95–114. Fabiańska, M., Matuszewska, A., 1998. Preliminary geochemical analysis of coal extracts and low-temperature tars of the UpperVistula Coal Region (Poland). Polish Journal of Applied Chemistry 3–4, 259–274. Fabiańska, M., Bzowska, G., Matuszewska, A., Racka, M., Skręt, U., 2003. Gas Chromatography–mass spectrometry in appraisal of organic matter of the Grodziec Beds (Upper Carboniferous): Upper Silesian Coal Basin, Poland. Chemie der Erde (Geochemistry) 63, 63–91. Goodarzi, F., 1987. Reflectance and petrology of a burning bituminous coal seam. Fuel 66, 1073–1078. Han, Z., Kruge, M.A., 1999. Chemistry of maceral and groundmass density fractions of torbanite and cannel coal. Organic Geochemistry 30, 1381–1401. Huizinga, B.J., Tannenbaum, E., Kaplan, I.R., 1987. The role of minerals in the thermal alteration of organic matter — IV. Generation of n-alkanes, acyclic isoprenoids, and alkenes in laboratory experiments. Geochimica et Cosmochimica Acta 51, 1083–1097. Ingram, G.R., Rimstidt, J.D., 1984. Natural weathering of coal. Fuel 63, 292–296. Isaksen, G.H., 1993. Organic geochemistry of paleodepositional environments with predominance of terrigenous higher-plant organic matter. In: Huc, A.-Y. (Eds.), Paleography. Paleoclimate, and Source Rocks, AAPG Studies in Geology, No 40. The American Association of Petroleum Geologists, Tulsa, Oklahoma, USA, 81−104. ISO 7404-2, 1985. Methods for the Petrographic Analysis of Bituminous Coal and Anthracite — Part 2: Methods for Preparing Coal Samples. International Organization for Standardization, Geneva, Switzerland. 8 pp. ISO 7404-3, 1994. Methods for the Petrographic Analysis of Bituminous Coal and Anthracite — Part 3: Methods Maceral Group. International Organization for Standardization, Geneva, Switzerland. 4 pp. ISO 7404-5, 1994. Methods for the Petrographic Analysis of Bituminous Coal and Anthracite — Part 5: Methods Determining Microscopically the Reflectance of Vitrinite. International Organization for Standardization, Geneva, Switzerland. 11 pp. Jabłońska, M., 2004. Thousands of tons of waste. Biuletyn Górniczy 11–12, 113–114 (in Polish).

424

M. Misz et al. / International Journal of Coal Geology 71 (2007) 405–424

Karayigit, A.I., Whateley, M.K.G., 1997. Properties of a lacustrine subbituminous (k1) seam, with special reference to contact metamorphism, Soma—Turkey. International Journal of Coal Geology 34, 131–155. Kwiecińska, B., Petersen, H.I., 2004. Graphite, sami-graphite, natura coke, and natural char classification — ICCP system. International Journal of Coal Geology 57, 99–116. Leif, R.N., Simoneit, B.R.T., 2000. The role of alkenes produced during hydrous pyrolysis of a shale. Organic Geochemistry 31, 1189–1208. Love, G.D., Snape, C.E., Carr, A.D., Houghton, R.C., 1995. Release of covalently-bound alkane biomarkers in high yields from kerogen via catalytic hydropyrolysis. Organic Geochemistry 10, 981–986. Lu, S.-T., Kaplan, I.R., 1992. Diterpanes, triterpanes steranes and aromatic hydrocarbons in natural bitumens and pyrolysates from different humic coals. Geochimica et Cosmochimica Acta 56, 2761–2788. Machnikowska, H., Łuczak, A., Kubacki, A., 2003. Effect of oxidation method on properties and structure of coals and lithotypes. Karbo 3, 157–164 (in Polish). Martinez, M., Escobar, M., 1995. Effect of coal weathering on some geochemical parameters. Organic Geochemistry 23, 253–261. Ndaji, F.E., Thomas, K.M., 1995. The effect of oxidation on the macromolecular structure of coals. Fuel 74, 932–937. Noble, R.A., Alexander, R., Kagi, R.I., Knox, J., 1985. Tetracyclic diterpenoid hydrocarbons in some Australian coals, sediments and crude oils. Geochimica et Cosmochimica Acta 49, 2141–2147. Oros, D.R., Simoneit, B.R.T., 2000. Identification and emission rates of molecular tracers in coal smoke particulate matter. Fuel 79, 515–536. Otto, A., Simoineit, B.R.T., 2001. Chemosystematics and diagenesis of terpenoids in fossil conifer species and sediment from the Eocene Zeitz formation, Saxony, Germany. Geochimica at Cosmochimica Acta 65, 3505–3527. Palmer, S.E., 1993. Effect of biodegradation and water washing on crude oil composition. In: Engel, M.H., Macko, S.A. (Eds.), Organic Geochemistry. Principles and Applications, pp. 511–534. Peters, K.E., Walters, C.C., Moldowan, J.M., 2005. The Biomarker Guide. Biomarkers and Isotopes in Petroleum Exploration and Earth History, 2nd ed. Cambridge University Press. 1155 pp. Philp, R.P., 1985. Fossil Fuel Biomarkers. Application and Spectra. Elsevier, Amsterdam. 294 pp. Radke, M., 1987. Organic geochemistry of aromatic hydrocarbons. Advances in Petroleum Geochemistry, vol. 2. Academic Press, London, pp. 141–203. Radke, M., Welte, D.H., 1983. The methylphenanthrene index (MPI): a maturity parameter based on aromatic hydrocarbons. In: Bjoroy, M., et al. (Eds.), Advances in Organic Geochemistry 1981. Wiley, Chichester, 504−512. Radke, M., Welte, D., Willsch, H., 1986. Maturity parameters based on aromatic hydrocarbons: influence of the organic matter type. Organic Geochemistry 10, 51–63. Radke, M., Rullkotter, J., Vriend, S.P., 1994. Distribution of naphthalenes in crude oils from the Java Sea: source and maturation effects. Geochimica et Cosmochimica Acta 58, 3675–3689. Rosenberg, P., Petersen, H.I., Thomsen, E., 1996. Combustion char morphology related to combustion temperature and coal petrography. Fuel 75, 1071–1082.

Rosiek, F., Urbański, J., 1990. Influence of some physical properties of coal on their selfignition. Scientific Publications of Mining Institute of Wrocław Technical University 59, 27–32 (in Polish). Sawicki, T., 2004. Spontaneous combustion in stock piles as a reason of a fire. Karbo 1, 56–59 (in Polish). Schultze, T., Michaelis, W., 1990. Structure and origin of terpenoid hydrocarbons in some German coals. Organic Geochemistry 16, 1051–1058. Sikorska-Maykowska, M. (Eds.), 2001. Valorization of the Natural Environment and Identification of Environmental Threats in Silesia Voivodship. Polish Geological Institute and Marshal Office of the Silesian Voivodeship, Warszawa. 23 pp. Simoneit, B.R.T., 1998. Biomarker PAHs in the environment. In: Neilson, A.H. (Eds.), The Handbook of Environmental Chemistry, vol. 3, Part I, PAHs and Related Compounds. Springer-Verlag, Berlin, Heidelberg, pp. 176–221. Chapt. 5. Simoneit, B.R.T., 2002. Biomass burning — a review of organic tracers for smoke from incomplete combustion. Applied Geochemistry 17, 129–162. Skarżyńska, K.M., 1995. Reuse of coal mining wastes in civil engineering — Part 1: properties of minestone. Waste Management 13, 3–42. Solomon, P.R., Serio, M.A., Suuberg, E.M., 1992. Coal pyrolysis experiments, kinetic rates and mechanisms. Progress in Energy and Combustion Science 18, 133–220. Speight, J.C., 1994. The Chemistry and Technology of Coal. Marcel Dekker, Inc., New York. 642 pp. Stach, E., Mackowsky, M.-Th., Teichmüller, M., Taylor, G.H., Chandra, D., Teichmüller, R., 1982. Stach's Textbook of Coal Petrology. Gebrüder Borntraeger, Berlin. 535 pp. Taylor, G.H., Teichmüller, M., Davis, A., Diessel, C.F.K., Littke, R., Robert, R., 1998. Organic Petrology. Gebrüder Borntraeger, Berlin. 704 pp. Tekely, P., Nicole, D., Delpuech, J., Totino, E., Muller, J., 1987. Chemical structure changes in coals after low-temperature oxidation and demineralization by acid treatments as revealed by high resolution solid state 13C NMR. Fuel Processing Technology 15, 225–231. The Wiley/NBS Registry of Mass Spectral Data. Wiley, New York. Thrope, A.N., Senftle, F.E., Finkelman, R.B., Dulong, F.T., Bostick, N.H., 1998. Change in the magnetic properties of bituminous coal intruded by an igneous dike, Dutch Creek Mine, Pitkin County, Colorado. International Journal of Coal Geology 36, 243–258. Tsai, C.-Y., Scaroni, A.W., 1987. The structural changes of bituminous coal particles during the initial stages of pulverised coal combustion. Fuel 66, 200–206. Urbański, J., 1983. Technical Recultivation of Mine Waste Dumps with Particular Consideration of Fire Protection. Training Materials. The Association of Mining Engineers and Technics, Katowice. (in Polish), 61 pp. van Krevelen, D.W., 1993. Coal: Typology–Physics–ChemistryConstitution. Elsevier. 979 pp. Willsch, H., Radke, M., 1995. Distribution of polycyclic aromatic compounds in coals of high rank. Polycyclic Aromatic Compounds 7, 231–251.