Evaluation of elemental sulphur in biodesulphurized low rank coals

Fuel 90 (2011) 2923–2930

Contents lists available at ScienceDirect

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

Evaluation of elemental sulphur in biodesulphurized low rank coals L. Gonsalvesh a, S.P. Marinov a, M. Stefanova a, R. Carleer b, J. Yperman b,⇑ a b

Institute of Organic Chemistry, Bulgarian Academy of Sciences, Sofia 1113, Bulgaria Hasselt University, CMK, Research Group of Applied and Analytical Chemistry, Agoralaan – Gebouw D, B-3590 Diepenbeek, Belgium

a r t i c l e

i n f o

Article history: Received 9 February 2011 Received in revised form 5 April 2011 Accepted 28 April 2011 Available online 10 May 2011 Keywords: Coal Elemental sulphur HPLC Biodesulphurization

a b s t r a c t A new procedure for elemental sulphur (Sel) determination in coal and its fractions is offered. It includes exhaustive CHCl3 extraction and subsequent quantitative analysis of the extracts by HPLC using C18 reversed phase column. Its application gives ground to achieve better sulphur balance and to specify the changes in the organic and elemental sulphur as a result of biotreatments. Two Bulgarian high sulphur containing coal samples, i.e. subbituminious (Pirin) and lignite (Maritza East), and one Turkish lignite (Cayirhan-Beypazari) are used. Prior to biotreatments, the samples are demineralized and depyritized. In the biodesulphurization processes, the applied microorganisms are: the white rot fungi ‘‘Phanerochaeta Chrysosporium’’ – ME446 and the thermophilic and acidophilic archae ‘‘Sulfolobus Solfataricus’’ – ATCC 35091. In the preliminary demineralized and depyritized coals, the highest presence of Sel is registered, which is explained by their natural weathering. As a result of the implemented biotreatments, the amount of Sel could be reduced in the range of 16.1–53.8%. The content of Sel is also assessed as part of the total sulphur and organic sulphur. The following range of Sel content is measured: 0.01–0.16 wt.% or 0.3–4.6% of total sulphur and 0.3–5.1% of organic sulphur. In this way, more precise information is obtained concerning the content of organic sulphur presence. Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction Recently coal biodesulphurization became to the forefront of coal technology research as a method with potential toward sulphur removal. The main advantage is mild experimental conditions with no harmful reaction products where the technological value of coal is slightly affected. Development of new tactics for sulphur compounds removal from coal depends, in part, upon knowledge of their chemical constitution. A fundamental requirement of any research into desulphurization is an accurate method for assessment the various forms of sulphur in coal [1]. For the evaluation of the effects of desulphurization methods and for the selection of suitable coals for sulphur species-specific desulphurization process, the determination of the different sulphur types in coal is required. Sulphur in coal is recognized as existing in three forms: (1) inorganic sulphates (Ss), mainly as ferrous sulphate and gypsum; (2) inorganic sulphides (Sp), generally as ferrous sulphide (pyrite), although sulphides of zinc and lead may occur in some coals; and (3) as organic sulphur compounds (So) [2,3]. A forth sulphur form should be mentioned as well – elemental sulphur is detected in weathered coal (Sel). ⇑ Corresponding author. E-mail address: [email protected] (J. Yperman). 0016-2361/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.fuel.2011.04.041

There are different theories for the appearance of Sel in coal. Some authors emphasize the importance of Sel in the formation of pyritic and organic sulphur compounds and suggested that Sel found in coal today is a primary substance formed during coal formation [3]. However, this view is still under discussion. Duran et al. [4] used extraction and GC analysis to determine Sel in a suite of US coals. They found that Sel (0.03–0.17%) is present in coal that has been exposed to the atmosphere, but it is absent in pristine samples that have been processes and sealed under a nitrogen atmosphere. According to the same study, Sel is not a natural constituent of coal but rather a product of atmospheric oxidation of pyrite. Beyer et al. [5] proved that, microbial desulphurization in acidic environment results in Sp decreasing and Sel increasing with time, whereas the organic sulphur remains unchanged. They suggested that microbial oxidation of pyrite produces ferric sulphate and that the simultaneous inorganic reaction of ferric iron with pyrite produces Sel and ferrous iron, as follows:

2Fe3þ þ FeS2 ðpyriteÞ ! 3Fe2þ þ 2S There are some consequences when the coal under consideration contains Sel [6]. This is related to the fact that there are direct standard analytical procedures for total sulphur (St), Sp and Ss determination. In the absence of standard analytical procedure for direct measurement of the So, the latter is calculated indirectly, i.e. by subtracting the sum of inorganic sulphur (Ss + Sp) from St.

2924

L. Gonsalvesh et al. / Fuel 90 (2011) 2923–2930

Thus, cumulative errors in determining St, Ss and Sp are reflected in the values of So. Moreover if Sel is present in coal, it will be included into the so called organic sulphur and an overestimation of the organic sulphur magnitude will take place. Although Sel actively participates in many biogeochemical processes, the studies related to this species are rather poor, mainly due to the limitations of the few available analytical methods for its determination [7]. There were several attempts for quantification of Sel in coal and related material. X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS) and Raman spectroscopy were among the methods used to detect Sel at the mineral surfaces [8,9] and coal [5], but accurate quantification of Sel with these techniques is difficult. Chromatographic methods to solve the problem have been described as well. Generally the procedures for the estimation of Sel in coal and environmental samples encompass the extraction of Sel and subsequent quantification by spectral or chromatographic methods. Respectively, appropriate extraction method should be developed. For that purpose Gryglewicz and Gryglewicz [6] tested the effect of different solvents and time on the efficiency of Sel extraction. A protocol for Sel quantification in coal, based on the extraction with cyclohexane for 6 h with it subsequent quantitative analysis in the extract by GC/MS, is proposed. Duran et al. [4] used GC coupled with a Hall electrolytic conductivity detector, with a high selectivity to sulphur. Another technique, based on perchloroethylene extraction and subsequent analysis by high-performance liquid chromatography (HPLC) on a C18 reversephase column with UV detection, was used by Buchanan et al. [10]. According to Steudel [11], the so-called C18 stationary phase turned out to be the best in the case of sulphur homocycles analysis. The author claims that the best detection for Sel is UV absorption at/or near 254 nm, since all compounds containing S-S bonds strongly absorb in this wavelength region. In a previous study, in the process of effective biodesulphurization of coals from different sources, the targets were: appreciation of the changes in sulphur, its different species and organic forms in soluble and insoluble products [12]. The interest was directed towards specification of organic and elemental sulphur variations as a result of biotreatments. Therefore, the aim of the present study is to apply appropriate analytical procedure for Sel determination and to study the effect of biotreatments on the contents of Sel in coals. Development of a precise procedure for Sel measurement will give us ground to attain a better sulphur balances for initial and biotreated samples. 2. Experimental 2.1. Coal Two Bulgarian coal samples – Humovitrain Maritza East (M), Rr = 0.20% [13] and Pirin subbituminous coal (P), Rr = 0.46% [14], and the Turkish lignite – Cayirhan-Beypazari (B), Rr = 0.38% [15] are used. The samples are selected in view of their high sulphur contents. In order to concentrate our efforts on organic sulphur study and to improve the information obtained by the Atmospheric Pressure–Temperature Programmed Reduction (AP-TPR) technique concerning organic sulphur functionalities, the coal samples are preliminary demineralized [16,17] and subsequently treated by diluted nitric acid for depyritization [18]. Demineralized and depyritized samples are assigned as APF (Ash and Pyrite Free). 2.2. Biodesulphurization Microorganisms effective towards organosulphur compounds are selected and used according to the previous studies [12,19– 21]. Demineralized and depyritized coal samples (<0.063 mm)

are added to the selected microorganism media, with a ratio of 3 g coal per 100 ml microorganism in the recommended by the suppliers’ media. In the biodesulphurization processes, the microorganisms fungus Phanerochaeta Chrysosporium – ME446 and thermophilic bacteria Sulfolobus Solfataricus – ATCC No. 35091 are used under following conditions: P. Chrysosporium (ME464) – (PC), white rot fungi, pH = 4.7 (at ambient temperature), temperature 30 °C, 125 rpm shaking rate, 6 days duration; S. Solfataricus (ATCC No. 35091) – (SS), thermophilic and acidopfilic archae, pH = 4 (at ambient temperature), temperature 70 °C, 40 rpm shaking rate, 14 days duration. Biotreated coal samples are separated by filtration from the media and then washed with 5% HCl (aq.) solution, and subsequently with hot distiled water. This procedure is performed in order to avoid any biomass contamination and to purge eventually generated SO2 4 byproducts. After that, the coal samples are dried at 105 °C and analyzed for total sulphur content by Eshka method. Other sulphur forms, i.e. pyrite (Sp) and sulphate sulphur (Ss) are determined as well [22]. In Table 1 are included: – total sulphur content in investigated samples together with other sulphur forms calculated on a dry base; – proximate analysis determined by TGA method [23]; – higher heating value (HHV) calculated by the formula of Channiwala [24]. In this table, organic sulphur is calculated by the difference from St and the sum of Sp and Ss. Thus, the calculated So values comprises So and Sel. 2.3. Extraction procedure The following extracts are under consideration: (i) cyclohexane (c-He) extracts; (ii) soluble in chloroform part (bitumen) and its fraction of neutral oils; (iii) c-He extracts of the insoluble in chloroform part. The extraction sequence is illustrated in Fig. 1. Separation and fractionations are included in it as well. The extractions are performed at the following experimental conditions: (i) Extraction with c-He A 500 mg coal sample (particle size < 0.063 mm) is extracted in E-flask. The extraction is performed with 50 ml c-He at ambient temperature and stirring with magnetic stirrer for 6 h. Perchlorethylene (PCE), 1000 lL, is added to the dry extract and Sel presence is determined by HPLC. One subsequent extraction with fresh portion of c-He is performed as well. The dry extract is anylized after disolving in 600 lL PCE. (ii) Extraction with chloroform and assessment of the soluble parts (bitumen). The extraction is carried out for 6 h on a 4 g of coal sample with 50 ml of chloroform reflux at 70 °C and stirring. The obtained bitumen is divided into asphaltenes and maltenes (neutral oils) by asphalthene precipitation (4:1, v/v n-hexane (n-He):benzene (Bz)). The soluble parts (neutral oils) are fractionated by SiO2 chromatography into aliphatic (n-He eluent), aromatic (Bz eluent) and polar fractions (Acetone (Ac) eluent). A half of the aliphatic fraction is dissolved in 600 lL PCE and subjected to Sel determination by HPLC. Serial following dillutions with PCE are performed to set the concentration of the sample within the linear range of standard solutions. (iii) Subsequent extraction with c-He of the insoluble part, obtained after chloroform extraction.

2925

L. Gonsalvesh et al. / Fuel 90 (2011) 2923–2930 Table 1 Characteristics of coal samples. Sample

S content (%)db

Proximate analysis (%) ad

daf

ad

Fix C

daf

Caloric valuea (MJ kg1)

Ash

VM

St

Sp

Ss

So

P-in P-APF P-APF-PC P-APF-SS

10.36 0.01 0.53 1.69

37.23 42.27 42.86 41.01

6.45 6.16 6.60 5.34

62.77 57.73 57.12 59.00

4.88 4.10 3.09 3.47

0.51 0.23 0.16 0.20

0.48 0 0 0

3.89 3.87 2.93 3.27

22.68 23.65 23.08 26.30

M-in M-APF M-APF-PC M-APF-SS

8.47 1.20 0.89 1.43

48.83 50.75 50.51 49.85

8.06 6.25 7.37 6.99

51.17 49.23 49.49 50.15

6.41 3.96 3.01 3.81

1.49 0.37 0.20 0.37

1.40 0 0 0

3.52 3.59 2.81 3.44

19.29 19.93 20.15 24.24

B-in B-APF B-APF-PC B-APF-SS

29.10 0.64 0.96 1.14

40.22 43.67 44.02 42.39

12.34 6.49 6.19 6.37

59.78 56.33 55.98 57.61

5.09 3.47 3.10 2.86

1.12 0.38 0.11 0.36

0.82 0 0 0

3.15 3.09 2.99 2.50

16.51 22.03 22.58 23.00

W

P – Pirin subbituminous coal; M – Humovitrain Maritza East; B – Cayirhan-Beypazari lignite; in – initial coal; APF – Ash and Pyrite Free coal; PC – treated with Phanerochaeta Chrysosporium coal; SS – treated with Solfolobus Solfataricus coal; W – moisture; VM – volatile matter; Fix C – fixed carbon; adair dried; dbdry basis; dafdry, ash free. a Calculated on dry bases by the formula of Channiwala [21].

Initial coal Demineralization, Depyritization

Ash Pyrite Free coal (APF) Extraction with c-He (i)

Biotreatments by Phanerochaete Chrysosporium (PC) or Solfolobus Solfataricus (SS)

Sel determination by HPLC

Biodesulphurized APF coal (APF-PC and APF-SS)

Extraction with CHCl3 (ii) Insoluble part

Soluble part (bitumen)

n-He:Bz (4:1, v/v)

Asphaltenes (insoluble)

Neutral Oils (soluble)

Extraction with c-He (iii)

SiO2 chromatography

Insoluble part

Soluble part Aliphatic fraction

AP-TPR analysis

Sel determination by HPLC

Aromatic fraction

Polar fraction

Sel determination by HPLC Fig. 1. Experimental strategy.

The extraction is performed in a similar way as described in paragraph (i). 2.4. HPLC analysis HPLC procedure for Sel determination has been applied. All analyses are performed on an Agilent HPLC system with 20 lL injection volumes and a diode array detector operating at 254 nm with an 8 nm bandwidth. A Varian Chromosphere 5 l C18 reversed phase column (4.6 mm  250 mm) is used with an eluent 95:5 methanol (HPLC grade):water at a flow rate of 1 ml/min. 2.5. Elemental sulphur standards for HPLC The external standard is prepared using commercially available Sel (Aldrich, 99.998%). Namely, 24 mg Sel is dissolved in a beaker

containing 20–30 ml of PCE (Acros, spectrophotometric grade). The solution is stirred (1 h) and quantitatively transferred to a 100 ml volumetric flask and diluted with PCE to the mark. A final concentration for the stock solution of 240 mg/l is obtained. Subsequent dilutions of the stock solution produced additional standards, varying in the concentration range of 0.1–240 mg/l. 2.6. Analysis of extracted solid residue AP-TPR technique is used for identification of the volatile species released during pyrolysis [12,17,18,25]. A mixture of 40 mg sample and equivalent amount of fumed silica is placed in the quartz reactor and subjected to a linear heating of 5 °C/min from 25 up to 1025 °C. Helium gas is passed into the sample with a flow of 100 cm3/min. The AP-TPR reactor is coupled ‘‘off-line’’ to the TDGC/MS apparatus. In this case, the AP-TPR reactor is connected to

2926

L. Gonsalvesh et al. / Fuel 90 (2011) 2923–2930

ice-cooled metal adsorption tubes, with SilcoSteel Coating, filled with Tenax/Carbopack B/Carbosieve SIII (Marks) as adsorbents. The adsorption tubes are desorbed systematically and analyzed by a TD-GC/MS. TD-GC/MS apparatus is used with He, as carrier gas at 85 kPa, at the following conditions:

3

6x10

3

5x10

3

Peak area

4x10

(a) Unity thermal desorber (Marks): primary desorption 20 min at 320 °C; cold trap at 8 °C, heated at maximum heating rate to 320 °C, hold time 15 min; flow path temperature 200 °C; (b) Trace GC Ultra-Gaschromatography (Thermo Instruments): capillary column 30 m ZB 5-MS 0.25 mm  0.25 lm film thickness (Phenomenex); temperature program – 3 min at 30 °C, heated 8 °C/min to 100 °C, heated 12 °C/min to 310 °C, hold time 5 min; (c) DSQ-Mass spectrometer (Thermo Instruments): EI spectra; ionization energy – 70 eV; scan range – m/z 33–480 in 0.4 s. Each tube is spiked with 3 lg d4-thiophene to quantify results for the target sulphur species. NIST library spectra are used for peak identification with special interest to the different sulphur species, liberated or in situ formed during the AP-TPR pyrolysis.

3

3x10

3

2x10

3

1x10

0 0

50

100

3.2. Analysis of the extracts

The quantity of Sel extracted from coal is determined using a calibration curve prepared from standard solutions of Sel in PCE. Fig. 2 presents the overlaid chromatograms for a series of standard solutions of Sel in PCE. The peak approximately at 11.4–12.1 min is assigned to Sel. Its area increases with increasing concentration of Sel standards. The broad peak registered at 4–6 min, as well as the small peak around 3.6 min, are assigned to PCE. The negligible peak appearing around 3.1 min could be attributed to acetone, a rest from glassware washing. Other peaks are not observed in the studied range of elution region up to 15 min. Fig. 3 shows the calibration curve for a series of Sel standards. The points represent the area of the Sel peak for three independent

At ambient temperature in non-polar liquids Sel dissolves without decomposition [26]. As a rule, the solubility of the sulphur allotropes in organic solvents decreases with increasing molecular size [27]. Carbon disulphide is by far the best solvent, followed by toluene and dichloromethane, while cycloalkanes are suitable for the smaller ring molecules only as far as ambient temperatures are concerned. It should be mentioned that the data cited here refer to non-associated free Sel. In the case of Sel extraction from environmental samples, presented data for Sel extractability of various organic solvents [6,8] are different from those listed for the

0.1mg/L 1 mg/L 7.5 mg/L 15 mg/L 30 mg/L 60 mg/L 120 mg/L 240 mg/L

400

Absorbance (mAU)

Absorbance (mAU)

3000

2000

250

series of standard solutions (I–III), with concentration in the range of 0.1–240 mg/l. Linear regression analysis of the calibration points for I–III series result in correlation coefficients of 0.99999, 0.99980 and 0.99985, respectively. The small standard deviation in peak area, found for each standard concentration, is an indication for the high robustness and reliability of HPLC with UV–Vis detection.

3.1. Analysis of standards

Solvent

200

Fig. 3. Calibration curve for three series of Sel standard solutions.

3. Results and discussion

2500

150

Concentration (mg/L)

300

200

100

1500 0 11,4

1000

11,6

11,8

12,0

Retention time (min)

Elemental sulphur

500 0 2

4

6

8

10

12

14

Retention time (min) Fig. 2. Overlaid HPLC chromatograms for a series of standard solutions of Sel in PCE with concentration range of 0.1–240 mg/l. The inset presents the chromatograms in the retention time window of 11.40–12.05 min.

L. Gonsalvesh et al. / Fuel 90 (2011) 2923–2930

solubility of Sel in organic solvents [26]. Moreover, it is proved that Sel extractability with organic solvents depends on Sel surroundings [8]. Therefore to apply chromatography procedures for Sel analysis in any environmental samples, an appropriate extraction method should be developed. In the literature several reliable organic solvents are proposed. According to Buchanan et al. [10,28] PCE is an excellent solvent for extracting Sel from bituminous coal. Temperature and contact time are important variables for the yield of Sel by PCE extraction, because of reaction between sulphur and coal. For three different rank Polish coals (lignite to medium volatile bituminous), Gryglewicz and Gryglewicz [6] tested a number of solvents at different extraction times, in order to choose the solvent with the highest Sel extractability. It was demonstrated that c-He and PCE had the highest effectiveness in the extraction of Sel for 6 h. A prolongation of extraction procedure to 10 h results in a decrease of Sel amount in PCE extract, whereas for c-He, the amount remains the same. According to the same authors, this indicates that for Sel extraction, c-He is more suitable than PCE, and a 6 h extraction time is sufficient. Therefore in the present study initially exactly c-He is selected as a proper solvent for the extraction of Sel. The changes that occur in the soluble coal part (bitumen) and its neutral oils are also under consideration. It is well known that Sel eluted with aliphatic fraction of the neutral oil. As a result of aforementioned reasons, the investigation strategy given in the outline on Fig. 1 is applied. 3.2.1. Extraction with c-He (i) The Sel content in the studied sample extracted for 6 h with cHe, as well as the Sel content extracted with subsequent 4 h c-He extraction, is presented in Table 2. As can be seen in it, the content of determined Sel by HPLC UV–Vis varied in the range of 0.001– 0.032 wt.%, calculated on dry, ash free base (daf). The highest amount of Sel is detected in the initial samples. This is probably related to the progress of oxidation processes (weathering), supported by the increased sulphate content (Table 1) in the initial samples. According to some authors, a sulphate sulphur content exceeding 0.3% is an indication for the progress of oxidation processes in which part of pyrite sulphur transfers into sulphate sulphur [29]. It is known that Sel can be considered as a product of atmospheric oxidation of pyrite by water and oxygen, and in situ bacteriological action or both together. There are two basic reactions proposed [30]: The first one takes place during microbial oxidation of pyrite;

2Fe3þ þ FeS2 ! 3Fe2þ þ 2Sel

Table 2 Content of Sel in c-He extracts of untreated by organic solvents coals. Sample

6 h c-He extraction

lg S/g

daf

Subsequent 4 h c-He extraction lg S/gdaf

S (wt.%)

R

lg S/g

daf

P-in P-APF P-APF-PC P-APF-SS

27.9 8.7 7.3 7.7

8.6 2.7 2.4 2.3

36.5 11.4 9.7 10.0

0.004 0.001 0.001 0.001

M-in M-APF M-APF-PC M-APF-SS

316.5 26.5 15.4 8.5

6.2 2.9 2.3 0.2

322.7 29.4 17.7 8.7

0.032 0.003 0.002 0.001

B-in B-APF B-APF-PC B-APF-SS

182.3 72.8 31.6 48.8

18.2 48.3 14.5 25.7

200.5 121.0 46.2 74.5

0.020 0.012 0.005 0.007

2927

The second one is the Sel producing reaction from pyrite;

4FeS2 þ 6H2 O þ 3O2 ! 4FeðOHÞ3 þ 8Sel In both cases, the Sel is an intermediate product in a series of reactions that ultimately produce sulphate (SO2 4 ). Detectable amounts of Sel can also be formed at the pyrite surfaces in a suitable condition. In our studies, it appears that the amount of detected Sel, as well as the content of Ss, correlates with the content of Sp (Fig. 4). The highest content of Sel is detected in M-initial sample, characterized by the highest content of Sp and Ss, followed by B-initial sample and the lowest content is found back in P-initial sample. In consequence of demineralization and depyritization treatment with mineral acid, the content of Sel is reduced in the range of 40–91%, related to the oxidation potential of the used mineral acids. Nitric acid, characterized by a strong oxidation potential towards pyrite, is widely used for the extraction of pyritic sulphur from coal. The reaction between HNO3 and pyrite is strongly dependent on the temperature and concentration of the acid. Reaction equations cited in the literature for the HNO3 – pyrite system are the following [31]:

FeS2 þ 4HNO3 ! FeðNO3 Þ3 þ 2S þ NO þ 2H2 O

ð1Þ

2HNO3 þ S ! H2 SO4 þ 2NO

ð2Þ

6FeS2 þ 30HNO3 ! 3Fe2 ðSO4 Þ3 þ 3H2 SO4 þ 30NO þ 12H2 O

ð3Þ

It is obvious that pyrite oxidation by the HNO3 produces simultaneously Sel and sulphate by separate oxidation processes (Reactions (1) and (3)). Produced Sel is an intermediate product, which is subsequently oxidized to sulphate (Reaction (2)). ‘‘Naturally’’ occurring Sel in initial coal is also oxidized by HNO3 through Reaction (2). The results found for Sel in the c-He extracts of the studied samples gave us ground to conclude that, due to biotreatments, the amount of Sel decreases in all PC and SS biotreated APF samples. 3.2.2. Extraction with chloroform and assessment of the soluble parts (bitumen), (ii) The results for separation of bitumens, neutral oils and their fractionation in aliphatic, aromatic and polar fractions are included in Table 3. Determined content of Sel, as a part of the aliphatic fraction, is shown as well. It varies in the range of 15–76 wt.%, calculated on dry ash free bases. HPLC determined Sel in the aliphatic neutral oil fraction is higher than the content of Sel extracted with c-He. In some cases, the amount of Sel extracted with chloroform is even 50 times higher than the one in c-He extracts. From Table 4, it can be stated that, similar to the extraction with c-He (i), the highest Sel amount in demineralized and depyritized samples is detected for B-APF coal followed by M-APF. In all biotreated APF samples, the content of extracted Sel decreases. 3.2.3. Extraction with c-hexane of already extracted with chloroform samples (iii) c-He is a reliable organic solvent for the extraction of Sel, as it was already noted in previous studies [6]. The initial strategy of our study was: (1) to examine changes that occur with the soluble part in organic solvents (in this case chloroform) bitumen, as a result of the applied biotreatments (ii); (2) to determine Sel changes by applying subsequent c-He extraction on the insoluble in chloroform residue (iii). The obtained results for the quantity of Sel in (iii), (see Table 4) were extremely low. This fact provoked the question, whether part of the Sel is already completely extracted by chloroform. Our previous GC/MS studies of neutral oils demonstrated that, if there was a presence of Sel in the bitumen, it would be co-eluted with aliphatic fraction of the neutral oils. It was therefore

2928

L. Gonsalvesh et al. / Fuel 90 (2011) 2923–2930

1.6

Sp vs Ss linear fit; R = 0.96

1.4

0.035 M

0.025 B

0.020

1.0

0.015

Sel (wt %)

Sp (wt %)

1.2

0.030

0.8 0.010 0.6

Sel vs Ss linear fit; R = 0.97

P

0.4

0.005 0.000

0.4

0.6

0.8

1.0

1.2

1.4

Ss (wt %) Fig. 4. Correlations between Sp/Ss and Sel/Ss for initial coals (P – Pirin subbituminous coal, B – Cayirhan-Beypazari lignite, M – Humovitrain Maritza East).

Table 3 Yields of bitumens, neutral oils and their fractions for APF and biotreated by PC and SS APF coal samples, in rel. %. Sample

a b c d

Bitumena

Neutral oilsb

Neutral oil fractionsc

Sel (%)d

Aliphatic

Aromatic

Polar

P-APF P-APF-PC P-APF-SS

1.9 2.0 1.3

64.7 71.1 65.1

16.2 14.3 12.4

30.1 22.8 12.4

43.6 48.1 58.1

18.1 15.1 15.2

M-APF M-APF-PC M-APF-SS

1.5 0.9 0.6

64.2 52.8 54.6

23.2 32.1 24.6

46.0 36.6 18.7

30.0 34.6 58.0

46.0 46.8 61.5

B-APF B-APF-PC B-APF-SS

2.0 2.3 1.2

66.9 74.0 59.7

20.7 17.1 23.8

24.0 27.1 13.0

40.5 43.5 49.7

63.8 41.4 76.0

Rel. % of coal. Rel. % of bitumen. Rel. % of neutral oil. Sel as a part of aliphatic fraction.

Table 4 Content of Sel in CHCl3 extracts (aliphatic fractions of the neutral oils) and c-He extracts of already chloroform extracted samples. Sample

CHCl3 extraction lg S/gdaf

Subsequent c-He extraction lg S/gdaf

R

lg S/ gdaf

S wt.%

P-APF P-APF-PC P-APF-SS

309.5 256.7 138.1

1.0 1.6 2.5

310.5 258.4 140.6

0.03 0.03 0.01

M-APF M-APFPC M-APFSS

882.1 616.4

2.1 0.5

884.3 616.9

0.09 0.06

416.1

0.7

416.9

0.04

1539.2 1016.0 1112.5

17.8 5.1 3.2

1557.0 1021.1 1115.6

0.16 0.10 0.11

B-APF B-APF-PC B-APF-SS

necessary to quantify Sel amount in the aliphatic fraction of the neutral oils (ii) and in c-He extracts of none extracted with chloroform samples (i). The obtained results demonstrate that in our experimental setup and coal set, chloroform is a better solvent for the extraction of

Sel compared to c-He (Tables 2 and 4). This fact could be attributed to differences in the polarity of both solvents or eventually, to Sel surroundings. It appears that accessibility of Sel for various organic solvents is of utmost importance. 3.3. Analysis of the solid extraction residue It is possible that a small amount of Sel is retained in already chloroform and c-He extracted samples. AP-TPR technique with its extension ‘‘off-line’’ TD-GC/MS is applied to receive more detailed information for the presence of Sel in already extracted products. The pyrolysis experiments were performed in inert (He) atmosphere. GC/MS spectra were quantitatively interpreted. Typical products of coal pyrolysis, i.e. alkylbenzenes, alkylnaphthalenes, phenols, etc., are accompanied by their sulphur containing analogues and are clearly detected in the chromatograms of the samples under study. The last ones are dominant present, due to special sulphur compounds selective sorbents. Sel is also clearly detected. The results for Sel are shown in Table 5. Detected low quantities of Sel in the extracted solid residues are a proof of the effectiveness of the applied extraction procedure with chloroform. Reactions of sulphur with various organic compounds are also possible [26]. At high temperature (above 180 °C), liquid sulphur becomes extremely reactive and can react with aromatic and aliphatic hydrocarbons and their derivatives. This reaction is due to the small sulphur species formed at corresponding temperature and it results in the formation of organic sulphur compounds similar to those believed naturally occurring in some coals, i.e. sulphides, polysulfides and thiophenes. In the studied samples, the content of sulphides, polysulfides and thiophenes is also traced. In the present study we cannot prove unequivocally: (1) how many of these compounds are formed, due to reactions of Sel with the organic part of the coal during pyrolysis, and (2) how many are naturally present. In order to estimate the part of Sel, which in the process of pyrolysis is liberated as volatile organic sulphur compounds, further deeper work is necessary. 3.4. Elemental sulphur balance Formula for calculation of total Sel distribution in the studied samples based on the results for Sel content in (i), (ii), and (iii) extracts as well as in the extracted solid residue is proposed. The total

2929

L. Gonsalvesh et al. / Fuel 90 (2011) 2923–2930 Table 5 Content of Sel in solid residue after CHCl3 and c-He extraction determined by AP-TPR ‘‘offline’’ TD-GC/MS. Sample

lg S/gdaf

P-APF P-APF-PC P-APF-SS M-APF M-APF-PC M-APF-SS B-APF B-APF-PC B-APF-SS

2.3 3.9 4.0 4.4 3.0 1.0 – 3.5 7.0

elemental sulphur distribution can be calculated taking in consideration: – Sel content in chloroform extract (ii); – Sel content in c-He extract of already chloroform extracted samples; – Sel content in solid residue after (ii) and (iii) determined by APTPR ‘‘off-line’’ TD-GC/MS in He atmosphere. Total Sel balance can thus be calculated as follow:

Table 6 Elemental sulphur balance. Sample

Sel(tot)daf

lg S/g coal

Correcteda St

So

wt.%

%

%

So (wt.%)db

P-APF P-APF-PC P-APF-SS

312.8 262.3 144.6

0.03 0.03 0.01

0.73 0.96 0.28

0.78 1.02 0.30

3.84 2.90 3.26

M-APF M-APF-PC M-APF-SS

888.7 619.9 417.9

0.09 0.06 0.04

2.24 1.97 1.03

2.47 2.11 1.15

3.51 2.75 3.40

B-APF B-APF-PC B-APF-SS

1557.0 1024.6 1122.6

0.16 0.10 0.11

4.58 3.19 3.79

5.14 3.31 4.35

2.93 2.89 2.39

%St – Sel(tot) as a part (%) from St (on dry, ash free basis) in the samples. %So – Sel(tot) as a part (%) from So (on dry, ash free basis) in the samples. a So corrected by Sel.

the range of 16–54%. S. Solfataricus appears to be a better biodesulphurizing agent toward Sel, than the applied white rot fungi P. Chrysosporium. The maximum reduction is achieved for the most mature coal sample P-APF-SS, with 54%. More precise information for content of organic sulphur is obtained. The following range of Sel is measured: 0,01 – 0,16 wt.% or 0,3%  4,6% of total sulfur and 0.3 – 5.1% of organic sulphur.

SelðtotÞ ¼ SelðiiÞ þ SelðiiiÞ þ SelðresÞ Acknowledgements where Sel(tot) is the total Sel distribution; Sel(ii) is Sel in (ii) extract; Sel(iii) is Sel in (iii) extract; Sel(res) is Sel in solid residue after (ii) and (iii) extractions. The obtained values are summarized in Table 6, together with Sel content (%) as a part of St and So. Corrected organic sulphur values by Sel(tot) results are included in Table 6 as well. The highest presence of Sel is registered for preliminary demineralized and depyritized coal. This can be explained by the weathered nature of initial coals. The content of Sel in APF samples increases in the following sequence: Pirin < Maritza East < Beypazari. This increase is expected for the first two members, bearing in mind their Sp contents. On the other hand, Sp content cannot explain the expressed difference in total Sel distribution for M-APF and B-APF samples. Most likely, this difference could be related to the characteristics of the coal matrix, differences in the crystallographic characteristics of the present pyrite, different impact of weathering effects or in situ different impact of bacteriological action. Through the implemented biotreatments, the amount of Sel is reduced in all samples. More promising biodesulphurization effect toward Sel appears in the treatment with S. Solfataricus, except for B-APF-SS. The maximum decrease in Sel is 54% and it is achieved for P-APF-SS coal sample. In the treatment with white rot fungi P. Chrysosporium, the biodesulphurization effect varied in the range of 16–34%, with maximum for B-APF-PC sample. 4. Conclusions A new procedure for Sel determination in coal and its fractions is offered. It includes exhaustive CHCl3 extraction and subsequent quantitative analysis of the extracts by HPLC using C18 reversed phase column. The highest presence of Sel is registered in the demineralized and depyritized coals, explained by their natural weathering. The differences in the Sel amount for all APF samples could be related to special characteristics of the coal matrix. Other explorations might be the Sp content of the initial samples as well as the crystallographic characteristics of the present pyrite. As a result of the implemented biotreatments, the amount of Sel could be reduced in

The study was supported within the framework of Cooperation agreement for joint supervision and award of a doctorate between Hasselt University, Belgium and Bulgarian Academy of Sciences (BAS). The authors are grateful to Prof. Y. Yurum, Dr. G. Dinler-Doganay and Dr. A. Dumanli for the biotreatment assistances, done in the frame of Project collaboration between IOCH-BAS and Sabanci University.

References [1] Newton RJ, Bottrell SH, Louie PKK, Hawthorne SB, Timpe RC. Coal-specific systematic errors in estimation of pyritic and organic forms of sulphur by ASTM and BS methods. Fuel 1996;75:1251–4. [2] Chakrabaritty SK, Iacchelli A. The dilemma of estimating forms of sulfur in lowsulphur low rank coals. Can J Chem 1986;64:861–4. [3] Chou CL. Geochemistry of sulfur in coal. In: Wilson LO, Curt MW, editors. Geochemistry of sulfur in fossil fuels. Washington (DC): ACS Symposium Series No 429; 1990. p. 30–52. [4] Duran JE, Mahasay SR, Stock LM. The occurrence of elemental sulphur in coals. Fuel 1986;65:1167–8. [5] Beyer M, Ebner HG, Assenmacher H, Frigge J. Elemental sulphur in microbiologically desulphurized coals. Fuel 1987;66:551–5. [6] Gryglewicz G, Gryglewicz St. Determination of elemental sulphur in coal by gas chromatography–mass spectrometry. Fresen J Anal Chem 2001;370:60–3. [7] Chen YW, Joly HA, Belzile N. Determination of elemental sulphur in environmental samples by gas chromatography–mass spectrometry. Chem Geol 1997;137:195–200. [8] Toniazzo V, Mustin C, Portal JM, Humbert B, Benoit R, Erre R. Elemental sulphur at the pyrite surfaces: speciation and quantification. Appl Surf Sci 1999;143:229–37. [9] McGuire MM, Hamers RJ. Extraction and quantitative analysis of elemental sulphur from sulphide mineral surfaces by high-performance liquid chromatography. Environ Sci Technol 2000;34:4651–5. [10] Buchanan DH, Coombs KJ, Murphy PM, Chaven C. A convenient method for the quantitative determination of elemental sulfur in coal by HPLC analysis of perchloroethylene extracts. Energy Fuels 1993;7(2):219–21. [11] Steudel R. Liquid sulfur. In: Steudel S, editor. Elemental sulfur and sulfur-rich compounds I, vol. 230. Top Curr Chem; 2003. p. 97–102. [12] Gonsalvesh L, Marinov SP, Stefanova M, Yürüm Y, Dumanli AG, DinlerDoganay G, et al. Biodesulphurized subbituminous coal by different fungi and bacteria studied by reductive pyrolysis. Part 1: initial coal. Fuel 2008;87:2533–43. [13] Šiškov GD. Bulgarian low rank coals: geology and petrology. In: Gayer R, Pešek J, editors. European coal geology and technology, vol. 125. Geol Soc Spec Publ; 1997. p. 140–8.

2930

L. Gonsalvesh et al. / Fuel 90 (2011) 2923–2930

[14] Kortenski J, Zdravkov A. Petrology of the subbituminous coal from the pirin basin, Bulgaria. Ann Univ Min Geol ‘‘St Ivan Rilski’’ Part: Geol Geophys 2005;48:56–66. [15] Whately MKG, Tuncali E. Quality variations in the high sulphur lignite of the Neogene Beypazari Basin, Central Anatolia, Turkey. Int J Coal Geol 1995;27:131–51. [16] Radmacher W, Mohrhauer P. Brennst Chem 1956;37:353. [17] Jorjani E, Yperman J, Carleer R, Rezai B. Reductive pyrolysis study of sulfur compounds in different Tabas coal samples (Iran). Fuel 2006;85:114–20. [18] Marinov SP, Stefanova M, Stamenova V, Carleer R, Yperman J. Sulphur functionality study of steam pyrolyzed ‘‘Mequinenza’’ lignite using reductive pyrolysis technique coupled with MS and GC/MS detection systems. Fuel Process Technol 2005;86:523–34. [19] Aytar P, Sam M, Çabuk A. Microbial desulphurization of Turkish lignites by white rot fungi. Energy Fuels 2008;22:1196–9. [20] Marinov SP, Gonsalvesh L, Stefanova M, Yperman J, Carleer R, Reggers G, et al. Combustion behaviour of some biodesulphurized coals assessed by TGA/DTA. Thermochim Acta 2010;497:46–51. [21] Durusoy T, Özbas T, Tanyolaç A, Yürüm Y. Biodesulfurization of some Turkish lignites by Sulfolobus Solfataricus. Energy Fuels 1992;6:804–8. [22] ASTM. D2492-02 standard test method for forms of sulfur in coal; 2002.

[23] Hassel RL. Du Pont company, No. Ta-54, instr products, scient and process div Wilmington, DE 19898 Proximate analysis of coal and coke. [24] Channiwala SA, Parikh PP. A unified correlation for estimating HHV of solid, liquid and gaseous fuels. Fuel 2002;81:1051–63. [25] Yperman J, Maes II, Van den Rul H, Mullens S, Van Aelst J, Franco DV, et al. Sulphur group analysis in solid matrices by atmospheric pressure– temperature programmed reduction. J Anal Chim Acta 1999;395(1–2):143–55. [26] Meyer B. Elemental sulfur. Chem Rev 1976;76(3):367–88. [27] Steudel R, Eckert B. Solid sulfur allotropes. In: Steudel S, editor. Elemental sulfur and sulfur-rich compounds I, vol. 230. Top Curr Chem; 2003. p. 55–6. [28] Buchanan DH, Warfel LC. Perchloroethylene extraction of sulphur and sulphur compounds from coal. Prepr Pap – Am Chem Soc, Div Fuel Chem 1990;35(2):516–22. [29] Ruschev D, Bekiarova E. Chemistry and technology of solid fuels. Sofia: Tehnika; 1979. p. 120. [30] Kawatra SK, Eisele TC. Froth flotation. In: Coal desulphurization: highefficiency preparation methods. 1st ed. New York: Taylor & Francis INC; 2001. p. 133. [31] Kadioglu Y, Karaca S, Bayrakceken S. Kinetics of pyrite oxidation in aqueous suspension by nitric acid. Fuel Process Technol 1995;41:273–87.