Influence of Nitric Acid Treatment in Different Media on X-rayStructural Parameters of Coal

Sudip Maity* and Ashim Choudhury

Central Institute of Mining and Fuel Research, FRI, Dhanbad 828 108, India

ReceiVed June 3, 2008. ReVised Manuscript ReceiVed August 14, 2008

The treatment of coal with nitric acid in aqueous and non-aqueous media introduces changes in the chemicaland spatial structure of the organic mass. Four coals of different rank have been treated with nitric acid inaqueous and glacial acetic acid media for assessing the changes in the structural parameters by the X-raydiffraction (XRD) technique. Slow-scan XRD has been performed for the raw and treated coals, and X-raystructural parameters (d002, Lc, and Nc) and aromaticity (fa) have been determined by profile-fitting software.Considerable variation of the structural parameters has been observed with respect to the raw coals. The d002

values have decreased in aqueous medium but increased in acetic acid medium; however, Lc, Nc, and fa valueshave increased in aqueous medium but decreased in acetic acid medium. It is also observed that considerableoxidation takes place during nitric acid treatment in aqueous medium, but nitration is the predominantphenomenon in acetic acid medium. Disordering of the coal structure increases in acetic acid medium, but areverse trend is observed in the aqueous medium. As a result, structurally modified coals (SMCs) are derivedas new coal-derived substances.

Introduction

The heterogeneous nature of coal has led many coal research-ers to study this material using different kinds of sophisticatedinstrumental techniques. Many structural models have beenproposed for coal considering the presence of aliphatic andaromatic species, amount of oxygen containing functionalgroups, and several other factors. Cartz et al. first used the X-raydiffraction (XRD) technique to decipher the carbon stackingstructure.1 Interlayer spacing (d002), average height of thecrystallite (Lc), and average number of layers (Nc) in crystalliteshave been established as the structural parameters for evaluatingthe stacking structure of highly crystalline carbon materials, suchas graphite.2,3 One of the difficulties of using the XRD techniquefor characterization of carbon materials, such as coal with lowcrystallinity, is the broadness of the corresponding XRD peaks.Many researchers have developed standard techniques for XRDmeasurement to quantify the stacking structure of aromaticlayers for such carbon materials.4-6 The existence of crystallitesin the coal structure has been proven by the appearances of thepeaks corresponding to the 002, 100, and 110 reflections ofgraphite.7 Slow-scan XRD analysis has been used to give higherresolution of the diffractograms, classifying the carbon relatedpeaks around 20-26° basically into two categories: one derivedfrom aromatic ring stacking around 26° (so called π band) and

the other is named γ band around 20°, which is believed to bederived from aliphatic chains.8,9

Moreover, coal has been used for the synthesis of severalcarbon derived materials, such as activated carbon and humicacid, and for these purposes, different kinds of structuralmodification have been carried out for coal. Structural modifica-tions have been carried by several methods, such as nitric acidtreatment in acetic anhydride medium or in aqueous medium10,11

(∼100 °C). In the present study, four demineralized Indian coalsof different carbon content (% Cdmf ∼ 80.0-90.0) have beentreated with nitric acid in aqueous and glacial acetic acidmedia.12 X-ray structural parameters (d002, Lc, and Nc) and fa ofall of the coals have been determined by the slow-scan XRDtechnique. Considerable variations are noticed for the differentX-ray structural parameters, including changes in aliphatic/aromatic ratios for the treated and untreated coals.

Experimental Section

Four coal samples with carbon content varying from 80-90%from different coalfields, namely, Shatabdih, Belpahar, SonepurBazari, and Sirna, of India were demineralized and are coded asSADM, BEDM, SODM, and SIDM, respectively. Elementalconstituent and volatile matter (VM) content were estimated forthe coals on dry mineral-free (dmf) basis (Table 1).

All coal samples were chemically demineralized using conven-tional HF + HCl acid treatment to avoid the effect of mineral matterin the XRD pattern.9 Weight loss during the demineralizationprocess was checked to see that demineralization had occurred toat least less than 1% effective mineral matter.

* To whom correspondence should be addressed. Telephone: +913262388363.Fax: 0326-2381113. E-mail: sudip_maity@yahoo.com.

(1) Cartz, L.; Diamond, R.; Hirsch, P. B. Nature 1956, 177, 500.(2) Houska, C. R.; Warren, B. E. J. Appl. Phys. 1954, 25 (12), 1503–

1509.(3) Iwashita, N.; Inagaki, M. Carbon 1993, 31 (7), 1107–1113.(4) Wertz, D. L.; Bissel, M. Energy Fuels 1994, 8, 613–617.(5) Lu, L.; Sahajwalla, V.; Kong, C.; Harris, D. Carbon 2001, 39, 1821–

1833.(6) Takagi, H.; Maruyama, K.; Yoshizawa, N.; Yamada, Y.; Sato, Y.

Fuel 2004, 83, 2427–2433.(7) Coal; Van Krevelen, D. W., Ed.; Elsevier: Amsterdam, The

Netherlands, 1993; pp 225-227.

(8) Watanabe, I.; Sakanishi, K.; Mochida, I. Energy Fuels 2002, 16,18–22.

(9) Maity, S.; Mukherjee, P. Curr. Sci. 2006, 91 (3), 337–340.(10) Alvarez, R.; Clemente, C.; Gomez-Limon, D. Fuel 2003, 82, 2007–

2015.(11) Tamarkina, Y. V.; Shendrik, T. G.; Krzton, A.; Kucherenko, V. A.

Fuel Proc. Technol. 2002, 77-78, 9–15.(12) am Ende, D. J.; Clifford, P. J.; Northrup, D. L. Thermochim. Acta

1996, 289, 143–154.

Energy & Fuels 2008, 22, 4087–4091 4087

10.1021/ef800424w CCC: $40.75 2008 American Chemical SocietyPublished on Web 11/19/2008

Chemical Treatments of Coal. Two kinds of chemical treat-ments have been performed on coal: (i) 30 wt % HNO3 in aqueousmedium at 105 °C in a refluxing condenser and (ii) a mixture ofglacial acetic acid and concentrated HNO3 in a 3:1 volume ratio.About 2 g of demineralized coal was treated with 60 mL of 30%nitric acid for a period of 8-10 h or until the end of any gasevolution from the coal-acid mixture. The residue was washedwith warm distilled water (∼60°C) for removal of any trace of nitricacid, and neutralization was confirmed by the pH test of the washedout water. In the second treatment, 10 mL of glacial acetic acidwas taken in a conical flask, which was connected to fused calciumchloride tower for keeping the mixture of coal, acetic acid, andnitric acid free from any environmental moisture. A total of 2 g ofdemineralized dry coal was transferred to the conical flask, and 10mL of glacial acetic acid was added and mixed thoroughly, keepingthe flask in a water bath of temperature 15-20 °C, while 3 mL of68% nitric acid was added to the mixture drop by drop for at least30 min duration and was kept overnight for 24 h. The treatedmixture was diluted with 250 mL/g distilled water, left for 24 h,and washed with distilled water until neutralization (pH ∼ 7.0).After each kind of treatment, coal was dried in an air oven anddifference weight (∆ wt %) was measured (Table 2). The percentagechange in weight has been calculated using the following formula:

∆ wt %)Wf -Wi

Wi× 100 (1)

where Wf is the weight of coal after treatment and Wi is the initialweight of coal. Before undertaking the treatment of the coals, theprocedures were standardized and the repeat experiment carriedon one coal showed that the variation in weight change is within3%.

The nitric-acid-treated coals in aqueous medium are coded witha suffix “H” viz. SAH (i.e., demineralized Shatabdih coal wastreated with nitric acid in aqueous medium at 105 °C), and acetic-and nitric-acid-treated coals are coded with a suffix “A” viz. SAA(i.e., demineralized Shatabdhi coal was treated with nitric acid inglacial acetic acid medium).

XRD Analysis and Profile Fitting. A D-8 ADVANCE (BrukerAXS, Germany) X-ray diffractometer was used to collect the X-rayintensities of the demineralized and treated coals in the 2θ rangeof 10-60° with Bragg-Brentano geometry, using parallel beam CuKR (40 kV, 40 mA) radiation. An X-ray amorphous sample holderwas used for coal sample loading, and the scan was made in lockedcouple, step scan mode (0.02°/step), with 4 s at each step.

A Windows NT-based Diffrac Plus profile fitting software (M/sBruker AXS, Germany) was used for determination of d002, Lc, Nc,and fa of the coals in the 2θ region of 15-32° of the diffractograms.This software was used to obtain line positions, intensities, widths,and shapes from both resolved and unresolved XRD spectra usings least-square refinement technique. To remove the systematic errorrelated to the least-square fitting for the peak positions andintensities, alignment of the X-ray diffractometer was checked

before every scan of the coals by running a standard quartzspecimen. Repeat XRD runs on few samples have been performed,and the deviation in the d002 values is less than 1%. The 2θ scanwas refined with KR1 and KR2 doublet and the KR2/KR1 relativeintensity. The least-square algorithm refines the background intensi-ties and peak intensities, positions, and widths.

The split-Pearson VII function combined with least-squareoptimization algorithm gives excellent fits to asymmetric X-raydiffraction lines obtained under a variety of conditions.13 Theapproach is to split each diffraction profile into a low- and high-angle part by dividing at the profile maximum and fit a separatePearson VII to each side. The split-Pearson VII lineshape functionwith variable width (2θ) and shape parameters was used on theleft and right side for the peak, and accordingly, the averageinterlayer distance (d002) and crystallite size (Lc) were determinedfrom the peak position and half-width of the broad hump in the 2θrange of 15-32° of the diffractograms for each coal sample.

The value of Lc is estimated using the empirical Scherrer formula(eq 2)

Lc )Kλ

B cos θ(2)

where B is the width of the corresponding band at half-maximumintensity, K is a constant (0.89), λ is the wavelength of incidentX-ray (Cu KR1 ) 1.5409 Å), and θ is the peak position of d002.The average number of aromatic layers in the stacking structure,Nc, was estimated by means of the following equation:14

Nc )Lc

d002(3)

The asymmetrical nature of the 002 band indicates the contributionfrom both the π and γ bands to the X-ray intensity in the defined2θ region. To estimate the relative contribution of the saidcomponents, two symmetrical Gaussian peaks were fitted in thesame region, corresponding to the π and γ bands. Initial guessvalues for the peak positions and half-widths were chosen, andprofile fitting was performed on diffused humps for 170 intensitypoints. Goodness-of-fit was measured by minimizing weightedreliability (WR) by several trials. This is expressed by eq 4:

WR) 100 %∑√w|lo - lc|

∑√wlo

(4)

where lo and lc are observed and calculated intensities and w is theweighting factor. A good profile fit is achieved when error valuesare less than 5% for crystalline materials; however, in the presentstudy, WR values could be minimized up to 2.3%, which may beconsidered as a quite good fitting result in the diffused humpregions. Theoretically, the areas under the π and γ peaks indicatethe relative proportion of aromatic atoms (Car) and saturated carbonatoms (Cal), respectively.15

(13) Brown, A.; Edmonds, J. W. AdV. X-ray Anal. 1980, 23, 361.(14) Sharma, A.; Kyotani, T.; Tomita, A. Carbon 2000, 38 (14), 1977–

1984.(15) Yen, T. F.; Erdman, J. G.; Pollack, S. S. Anal. Chem 1961, 33,

1587–1594.

Table 1. Elemental and Maceral Analyses of Coals

elemental analysis (wt %, dmf basis) maceral analysis (vol %) (mineral-matter-free basis)

C H S N Odiff VM vitrinite liptinite inertinite

Satabdih 90.93 5.10 0.64 1.82 1.51 21.05 37.3 2.9 59.8Sonpur Bazari 85.21 4.69 0.53 2.19 7.38 40.06 64.6 7.7 27.7Belpahar 82.78 5.38 0.67 1.65 9.52 41.61 9.8 23.3 66.9Sirna 79.64 4.69 1.81 2.53 11.33 37.15 36.1 16.3 47.6

Table 2. Percentage of Weight Change of Coal by (H2O +HNO3) and (AcCOOH + HNO3) Treatments

∆ wt %(H2O + HNO3)

∆ wt %(AcCOOH + HNO3)

Satabdih (SA) +14.9 +15.4Sonepur Bazari (SO) -33.3 +5.6Belpahar (BE) -26.0 +10.4Sirna (SI) -37.1 -1.9

4088 Energy & Fuels, Vol. 22, No. 6, 2008 Maity and Choudhury

Therefore, the aromaticity (fa) of coal can be estimated from eq 5

fa )Car

Car +Cal)

Aπ

Aπ +Aγ(5)

where A denotes the integrated area under the corresponding peakand Cal and Car are the number of aliphatic and aromatic carbonatoms per structure unit, respectively. Using this relation, the fa

values for both the demineralized and treated coals have beencalculated. A graphical representation of the π peaks after profilefitting for SA coal is depicted in Figure 1.

Using the above-mentioned procedure and equations (2, 3, and5), d002, Lc, Nc, and fa values for the demineralized and treated coalshave been evaluated and the results are summarized in Table 3.

Results

Raw Coals. X-ray diffractograms of the untreated deminer-alized coals is shown in Figure 2. All of the coals exhibit ahump in the 2θ range of 20-30°, corresponding to the 002 band.The high background intensity shows the presence of amorphouscarbon in the coal samples. The highest mature SA coal depictsa relatively sharper 002 band compared to the others, and it ismore symmetric on the low-angle side of the band. The presenceof the (10) two-dimensional band (2θ range of 42-48°) is moreconspicuous for the high-rank SA coal (Figures 2 and 3).Deviation from the expected trend of decreasing interlayerspacing (d002) with an increase in carbon content is observed.

However, the Lc, Nc, and fa of SA are higher than that of lowerrank coals. Surprisingly, the structural parameters of SO coal,whose carbon content is higher than SI and BE, reflect relativelyless ordered structure than the later, along with a loweraromaticity (Table 3). These variations could arise because ofthe lower inertinite content and hence less aromatic carbon ofSO coals (Table 1) compared to the two less mature coals.

Treatment in Aqueous Medium. Nitric acid treatment inaqueous medium has resulted in increased mass of the low-volatile bituminous coal SA, with significant loss in weight ofthe other three less mature coals (Table 2). The increase inweight of highly matured SA coal is due to the nitration of thecoal matrix with a minor loss of aliphatic side chains. In thecase of the other three high volatile coals, the loss of weight isdue to a significant loss of aliphatic side chains by oxidationduring this treatment. Analyses of the X-ray data show that, incase of the three coals of relatively lower carbon values, thedegree of ordering of organic framework increased comparedto the untreated ones, as reflected by the lower d002 and higherLc and Nc values. In contrast, lower values of Lc and Nc areobserved in the case of treated SA coal. A significant increase,however, is observed in the aromaticity (fa) of all four coals.This increase in aromaticity is primarily due to the loss of thealiphatic carbons, which is reflected in the reduced integratedarea of the γ band during profile fitting. The variation of thestructural parameters with the carbon content is shown in Figures4-6.

Treatment in Acetic Acid Medium. With the exception ofthe SI with the lowest carbon content, all three coals showedan increase in weight after treatment, which could be ascribedto the phenomenon of nitration of the organic matrix. Therewas minor weight loss in the lowest rank SI coal. The X-rayresults suggest a lowering of structural ordering of all four coalsas evident from the increase in d002 and decrease in the Lc andNc values. In all four coals, the aromaticity (fa) decreased fromthat of the parent coals.

Discussion

The change in the XRD pattern caused by the differenttreatment is shown in Figure 3 for SA coal. The nature of the

Figure 1. Fitting of two Gaussian peaks for the Shatabdih demineralized coal in 2θ ∼ 15-32° (upper graph is the net difference).

Table 3. Variation of d002, fa, Lc, and Nc, as Determined byProfile Fitting of the X-ray Diffractogram

sample d002 (Å) Lc fa Nc

SAA 3.54 13.8 0.74 3.9SADM 3.50 17.2 0.79 4.9SAH 3.50 16.4 0.89 4.7SOA 3.65 8.7 0.53 2.4SODM 3.55 9.0 0.61 2.5SOH 3.47 14.7 0.81 4.2BEA 3.50 12.2 0.72 3.5BEDM 3.45 15.9 0.74 4.6BEH 3.43 17.1 0.82 5.0SIA 3.50 10.3 0.75 2.9SIDM 3.46 10.8 0.76 3.1SIH 3.46 16.1 0.80 4.6

Nitric Acid Treatment in Different Media on Coal Energy & Fuels, Vol. 22, No. 6, 2008 4089

002 band of nitric-acid-treated coal in aqueous medium is moresymmetric compared to the raw coal and coal treated by nitricacid in acidic acid medium. This variation is observed in all ofthe other three coals and their treated samples (Figure 6). Itmay be noted that the variations is most significant in the caseof SO coals, which may be due to the presence of high vitrinite+ liptinite of the coal.

Two types of phenomena occur during nitric acid treatmentin different media, which are (i) removal of the aliphatic sidechains through oxidation and (ii) nitration in certain parts ofcoal molecules, primarily in aromatic zones. Both phenomenonare likely to occur simultaneously, the extent of which maydepend upon the type of coal and the medium of treatment. Itis generally observed that the oxidation effect is predominantin aqueous medium compared to that in acetic acid medium,leading to organic matter solubilization. It is observed that in

the case of coals that have undergone nitration, as evident froman increase in weight during treatment in either of the media,some structural disordering has taken place, which has beenreflected in the X-ray structural parameters. The inclusion ofthe additional heteroatoms, such as nitro groups and oxygen,associated with the nitration in the coal matrix has contributedto the increase of the average interlayer spacing and decreaseof the crystallite size (Figures 4 and 5). With regard to thechanges in the aromaticity of the residues, it has been observedthat, in the case where weight loss has taken place, particularlyin aqueous medium, the aromaticity has increased possiblybecause of the preferential removal of the aliphatic side chainsby oxidation. In the case where weight uptake was observed,in acetic acid medium, the aromaticity decreased. The highlymatured SA coal in aqueous medium, despite weight gain,reflects an increase in the aromaticity. The likely reason is the

Figure 2. XRD profiles of demineralized raw coals (A, SADM; B, SODM; C, SIDM; D, BEDM).

Figure 3. XRD profiles of treated and raw Shatabdih coal.

4090 Energy & Fuels, Vol. 22, No. 6, 2008 Maity and Choudhury

predominance of nitration phenomenon with simultaneous oxy-destruction of the aliphatic side chains, which led to the increaseof aromaticity.

Conclusion

X-ray investigations on four coals of different rank and theirstructurally modified samples through nitric acid treatment intwo different media show that the changes in the structuralparameters and the aromaticity of the SMCs are dependent uponcoal type and the medium of treatment. In general, weightincrease was observed in acetic acid medium with significantnitration of the coal matrix, making it more disordered alongwith the a decrease in aromaticity. In aqueous medium, nitrationhas occurred simultaneously with oxidation of the aliphatic sidechains. The significant weight loss of the lower rank coals is

due to the predominance of the oxidation phenomenon inaqueous medium. However, the present study shows thatnitration is the only predominant phenomena during nitric acidtreatment in acetic acid medium.

Acknowledgment. One of the authors (S.M.) is thankful to DST,Government of India, for providing funds under the Fast TrackYoung Scientist’s Project (DST SR/FTP/ES-015/2003, Sept 30,2003), to work on the micro-structure of Indian coals. The authorsthankfully acknowledge the RQA Division and Smt. N. Choudhuryof this Institute for providing the elemental and maceral data ofcoal, respectively. Authors are also thankful to Director, CIMFR,for permitting them to publish this work.

EF800424W

Figure 4. Variation of d002 with respect to Cdmf (%) for coals: (i) [, demineralized raw; (ii) 9, water + HNO3 treated; and (iii) 2, AcCOOH +HNO3 treated.

Figure 5. Variation of Lc with respect to Cdmf (%) for coals: (i) [, demineralized raw; (ii) 9, water + HNO3 treated; and (iii) 2, AcCOOH + HNO3

treated.

Figure 6. Variation of fa with respect to Cdmf (%) for coals: (i) [, demineralized raw; (ii) 9, water + HNO3 treated; and (iii) 2, AcCOOH + HNO3

treated.

Nitric Acid Treatment in Different Media on Coal Energy & Fuels, Vol. 22, No. 6, 2008 4091