The permanent physico-chemical changes attributed to coals of different ‘rank’ reflect the degree of maturation of the original plant matter. Variation in the ranks of this metamorphosed plant matter, ranging from peat, lignite to anthracite through bituminous coal, therefore illustrates a fascinating story of transformation in the molecular level (Burnham 2017). The concept of coal rank has been established firmly through the works of several decades, considering thermal maturity of the organic matter as the key. Defining the rank of a coal usually and most commonly uses rank parameters like vitrinite reflectance (VR_o), thermal alteration index (TAI) and the maximum temperature of metamorphism. However, no single parameter is sufficient to ascribe the rank of a coal over the whole range and rank parameter like vitrinite reflectance can be tricky when it comes to characterizing low rank coals such as lignite (O’Keefe et al. 2013). Other than inertinite and liptinite macerals, lignite comprises of huminite macerals which, with increasing temperature and prolonged depth of burial, matures into vitrinite—dominating the ranks of sub-bituminous through anthracite. The study of the extractable organic fractions in coal have gained significant scientific insights through the use of different analytical methods over the past decades (Mille et al. 1988; Katoh et al. 1980; Benkhedda et al. 1992; Doskočil et al. 2016; Ryder 2004; Von der Dick and Kalkreuth 1986; Enev et al. 2014; Doskočil et al. 2018; Fabianska et al. 2013).

With the advancement of technology, it is now possible to use different instrumental techniques to qualitatively as well as quantitatively determine and demarcate the constituent moieties in coal and organic matter in terms of their chemical compositions and structures. Identification of these polycyclic aromatic hydrocarbon (PAH) components can be feasibly carried out with the help of spectroscopic techniques. Although the introduction of spectroscopy in the domain of organic geochemistry is not new and it has been considered as a powerful and accurate tool to define the polycyclic polyaromatic hydrocarbon moieties present in coal as well as crude oil and other organic-rich rocks (Mille et al. 1988; Katoh et al. 1980; Benkhedda et al. 1992; Doskočil et al. 2016; Ryder 2004; Von der Dick and Kalkreuth 1986; Enev et al. 2014; Doskočil et al. 2018; Fabianska et al. 2013). Molecular fingerprints of different PAHs have been established by many workers (Katoh et al. 1980; Benkhedda et al. 1992; Doskočil et al. 2016, 2018; Ryder 2004; Enev et al. 2014; Fabiańska and Kurkiewicz 2013) for various polycyclic-polyaromatic hydrocarbon families and some literatures (Mille et al. 1988; Kershaw 1978) have also reported the possibility of distinguishing different coal ranks using UV-assisted fluorescence spectrometry. Von der Dick and Kalkreuth (1986) and Mille et al. (1988) had illustrated systematic trends and supported the potential of fluorescence spectrometry as a powerful tool to identify “fingerprints” in solvent-extracts from different ranks of coal, although some researches (Doskočil et al. 2016) had criticized the idea due to the scarcity of available experimental data. Fourier transform infrared spectroscopy and UV–visible absorption spectroscopy techniques, along with total synchronous and total luminescence fluorescence spectroscopy have proven to be reliable and convenient analytical methods for characterizing and delineating the structures of organic substances. These techniques require low quantity of analyte and provide quick, comprehensive and accurate results (Doskočil et al. 2016; Kister et al. 1996; Matuszewska et al. 2002). Fluorescence spectroscopy is a highly useful tool to analyse organic matter as it reflects only those compounds in which moieties with conjugated bonds are present (Birdwell and Engel 2010; Lakowicz 2006). The present study is the first attempt to implement spectroscopic techniques to understand the macromolecules present in solvent-extracts of the coals from different parts of India. Our work demonstrates the integrated insights from spectroscopic and microscopic observations on different ranks of coals from various coalfields of India, reflecting their different origins and compositions.

Indian coals are divided broadly into two age groups viz. upper Palaeozoic Gondwana coals which are mostly comprised of the ranks subbituminous to bituminous, and the Cenozoic lignites. The Gondwana coalfields span across the Indian states of Jharkhand, West Bengal, Madhya Pradesh, Maharashtra, etc. and are geologically restricted to intra-cratonic basins (Ghosh et al. 1985) whereas the Cenozoic lignites in India are found in the states of Rajasthan, Gujarat, Jammu & Kashmir and the North-eastern provinces of India, in the pericratonic rift basins (Biswas 1992; Dutta et al. 2011). The rank variation within these coals arises from their particular depositional environments, along with the duration of burial which has control over the maturation of the organic matter. A comprehensive representation of the stratigraphic and geological distribution of the samples with their corresponding ages are provided in Table 1 which can be correlated with Fig. 1 for a better understanding.

Stratigraphic successions of the studied formations

2.1 Gondwana coals of Jharkhand, Odisha and Sikkim

The Permian sections of the Jharia Basin in Jharkhand are likely to have suffered a major Permian–Triassic thermal event. Coals of the Lower Permian Barakar Formation in Jharia Basin are late mature to overmature in nature (Mishra et al. 1992), and coals of the Ib Valley Basin (part of Mahanadi master basin) hold exceptionally high potential for oil generation (Varma et al. 2015).

Multiple fold-thrust systems are present in the Himalayas which extend from south to north and demarcate the boundaries between different lithological sequences viz. the Lesser-, Greater- and sub-Himalayan rocks. In the state of Sikkim, Teesta half-window exposes these relatively low-grade lesser Himalayan rocks with highly metamorphosed Greater Himalayan crystallites. The Rangit Duplex lying within the Teesta half-window, is entirely bound by Ramgarh Thrust (Bhattacharyya and Mitra 2009; Bhattacharyya et al. 2006), and exposes the Rangit Duplex. The Rangit Duplex system consists of rocks of the lesser Himalayan sequence and the main constituent lithologies are the Gondwana, Buxa and upper Daling litho-units. Upper Palaeozoic Gondwana-equivalent rocks are exposed in the Rangit window at the core of the Sikkim domal structure and includes the coal-bearing lithology of Lower Gondwana exposed as tectonic strips within the Himalayan thrust belts (Ghosh et al. 2018; Ghosh 1997). In the present study, samples were acquired from the Lower Permian Barakar Formation exposed in different parts of the aforementioned Gondwana basins (see Table 1).

2.2 Cenozoic lignites of Gujarat

One of the best established and undisturbed Mesozoic–Cenozoic sequences within India is contained in the pericratonic rift basin of Kachchh, Gujarat, in western India. The formation of the basin occurred as a result of the East Gondwana Rift in the Upper Triassic (Biswas 1992). In the western part of the basin, Cenozoic sediments are present and extend to the continental shelf in the offshore region (Sarkar et al. 1996). These several hundred of meters thick sediments lie unconformably over the Deccan traps and Mesozoic rocks and contains lignite-bearing horizons such as the Naredi Formation (Dutta et al. 2011). In the Cambay Basin of Gujarat, a lignite deposit is mined at the Amod Lignite block in Rajpardi where the major litho-units comprise of the Babaguru Formation underlain by Tadkeshwar Formation which, in turn, is underlain by Nummulitic Formation. In Tadkeshwar Formation, the grey clay-bed at the bottom is continuous with varying thickness and is overlain by carbonaceous clays. Lenses of sandstone, carbonaceous clay and lignite occur embedded within this grey clay. The lignite occurring as lenses have variable thickness and lateral extent. The carbonaceous clay bed is conformably overlain by a five meter thick lignite seam, which acts as a marker bed. The lignite seam is soft, friable and brown containing yellow resin and pockets of clay. Upper carbonaceous clay lies conformably over the lignite seam (Singh et al. 2017).

Six coal samples were collected from different coalfields across India and their ranks were established using vitrinite reflectance (VR_o). The two lignite samples (LR and LM) were collected from Rajpardi and Matanomadh lignite mines of Gujarat state of western India, whereas the subbituminous sample (SB) was collected from the Hingula Opencast Project of Talcher coalfields in the state of Odisha, eastern India. Two bituminous samples (BKC and BNCA) were collected from the Jharia and Raniganj coalfields situated in the state of Jharkhand. One anthracite sample (AS) was acquired from the road-cut section of Jorethang area in Sikkim state. Geological background of the study areas is provided in the previous section.

Samples were prepared on as received basis for reflectance measurements and petrographic studies, using blocks mounted on araldite. The reflectance measurements were carried out in room temperature in a Leica DM RXP microscope. Yttrium Aluminium Garnet or YAG (RI value 0.917), Gadolinium Gallium Garnet or 3G (RI value 1.726) and Leucosapphire (RI value 0.59) were used as standards for calibration before measurement, as per as the IS 912 (Part 5)—1986 norm. For micropetrography, a Leica DM2700 microscope with fluorescence, blue and white light illumination modes were used.

Samples were ground to fine powder of less than 212-micron size and 5 gm of each sample was placed in a beaker with Dichloromethane (CH₂Cl₂) and Methanol mixed in 7:3 w/w ratio. Samples were weighed carefully and stirred before being placed in ultrasonic bath. Here, they were subjected to repeated ultrasonication following filtration in four continuous cycles. The extracted fractions were collected in individual glass vials and were placed under a fume hood to evaporate the remaining solvent present in them. The whole procedure was carried out in room temperature (~ 30 °C). During the spectroscopic measurements, 1:3 sample to solvent ratio in µL was maintained for all the extracted samples, using the same solvent base (Dichloromethane and Methanol in 7:3 w/w).

FT-IR spectral measurements were performed on the dried extracts using Attenuated Total Reflection (ATR) mode in a Perkin Elmer Spectrum Two spectrometer. Individual spectra were recorded in the wavenumber range of 400 to 4000 cm⁻¹ with a scanning resolution of 4 cm⁻¹ and a spectrum of air on the clean dry diamond crystal was selected as background. Auto-correction of the samples were done prior to the measurements.

UV–visible spectroscopy was carried out in solvent-diluted extracted fractions using a Perkin Elmer Lambda 35 double beam spectrophotometer with baseline correction. The extracted samples were diluted in a medium of dichloromethane and methanol mixed in the same w/w ratio of 7:3. The absorption spectra were recorded in the wavelength range of 200 nm to 800 nm, in a 10 mm quartz cuvette with a path length of 1 mm.

For fluorescence spectroscopy, the extracts were dissolved in the 7:3 w/w DCM and methanol mixture in a concentration of 72 mg/L. Spectra were recorded in a Perkin Elmer LS 55 spectrofluorometer. Scanning speed was maintained at 500 nm/min with a slit width of 10 mm. The excitation-emission matrices were obtained by simultaneously scanning excitation and emission wavelengths over 250 nm to 600 nm range with an interval of 20 nm. The experiment was carried out in room temperature. For both fluorescence and UV–visible spectral measurements, baseline correction was performed by subtracting the solvent base from the data.

4.1 Micropetrography and bulk composition study of original coal samples

The original coal samples were characterised as per the ICCP classifications of macerals (ICCP 1998, 2001; Pickel et al. 2017; Sýkorová et al. 2005) (see Fig. 2 for photomicrographs). Samples of the cenozoic lignites LM and LR from western India as well as the subbituminous Gondwana coal from Talcher coalfield (SB) exhibit a higher liptinite content. The liptinite macerals in the LM and LR samples are mostly constituted of sporinite, alginite, cutinite and resinite with sparse contribution from phlobaphinite (Figs. 2 a–f). Plant resins deposited inside the cell walls were identified. Sporadic presence of liptodetrinite occurring in lenses are also observed. The SB sample contains sporinite in unusually large proportions. In sample SB, the amount of liptinite present suppresses the reflectance of vitrinite. AS shows intense signatures of structural deformation which can be attributed to the stress conditions of the Himalayan fold-thrust belt (Figs. 2 g–h). The intensity of the deformation has essentially removed majority of liptinite from the coal and has also increased the reflectance of vitrinite, imparting the anthracite rank to it (Ghosh et al. 2018). The micropetrography data of the original coal samples are presented in Table 2.

Photomicrographs of the original coal samples: a Suberinite with phlobaphinite, sporinite and alginite in lignite LR; b Fluorinite and liptodetrinite dispersed in lignite LM; c Alginite in lignite LM; d Sporinite in subbituminous SB; e Funginite and corpogelinite in subbituminous SB; f Resinite infilling inside the cell wall structures in lignite LM; g and h Semifusinite, detrital vitrinite and intertodetrinite (inertinite) in anthracite AS; i Semifusinite and collotelinite in bituminous coal BNCA. Slides a, b, c, d and f are observations under fluorescence mode while e, g, h and i are observations carried out under white light

For the bulk composition study, industrial analysis of the original samples was carried out. The data is presented in Table 3. The BKC, BNCA and AS samples are having higher fixed carbon content compared to the samples of lower rank, while the lignite samples LR and LM are having high amount of volatile matter compared to the rest. For sample LM, the calculated Mineral Matter is 70.32 vol%, and the corresponding ash percentage of the same sample was found to be 24.28% on as received (AR) basis. Our inference as per the data and observation of the same sample and the area’s geology is that the original organic matter was deposited possibly in topogenous mire with fluctuating groundwater levels. This inference is drawn based on the available literature on similar type of samples reported from the same geological age and nearby area (Kumar et al. 2021). The fluctuation in the groundwater level can cause sudden increase in the mineral matter input during the deposition. According to literature (Taylor et al. 1998), an ash yield between 5 wt% and 40 wt% generally indicates the deposition of organic matter in a topogenous mire. Although the supply of mineral matter would have been lessened with a decline in the groundwater table level, an increase in the groundwater table level might increase the amount of minerals added to the mire (Jasper et al. 2010).

4.2 UV–visible spectroscopy

The normalized UV–Visible absorbance spectra of the extracts are shown in Fig. 3. The wavelength region in the spectra can be broadly divided into two segments viz. the UV region and the visible light region. Electronic absorption occurs in the UV part of the spectra while the visible part reflects the scattering due to Rayleigh components. Except those for the extracts from LM and LR lignite, the spectra are largely featureless and characterized by lesser absorption in the UV region; absorbance decreases with increasing acquisition wavelength. These spectra contain broadly recognizable shoulders. In the spectrum of AS, a sharp decrease in absorption is observed near the end of the UV region. Earlier works have associated higher wavelength transitions and lower UV absorption features with PAHs containing higher number of fused aromatic rings and a higher ratio of isolated double-bond carbon/sextet (Berlman 2012; Gargiulo et al. 2015). Broad shoulders observed in the mid UV range in the case of BNCA, BKC and SB samples might typically indicate the presence of aliphatic structures or smaller PAHs, as smaller PAHs tend to participate in fluorescence more actively compared to their larger counterparts (Gargiulo et al. 2015). Therefore, on a comparative basis, it can be assumed that the extracts from lignites LM and LR contain aliphatics and smaller PAHs with lesser degrees of conjugation compared to the extracts from bituminous-subbituminous coals i.e., BNCA, BKC and SB. The distribution of PAHs in case of the medium rank coal-extracts i.e., SB, BNCA and BKC are characteristically exhibiting two different absorption ranges in the UV region—one nearing 270 nm and the other at 300 nm wavelength. This feature can be attributed to the occurrence of different fluorophore groups present in the extracts and is evident from the EEM data as well (discussed in Sect. 4.4).

UV–visible absorption spectra of solvent-extracts

4.3 FT-IR spectroscopy

The solvent-extracts used in the present study were extracted from coals of different ranks collected from different basins with variable depositional settings and these differences have set the FT-IR spectra of the samples apart from one another (see Fig. 4). If the samples are to be categorized according to their corresponding ranks, then a similarity among the spectra is found for each rank (see Table 4 for FT-IR band assignments). For LM and LR which are both lignite, the spectra are characterized by bands of aliphatic groups. Bands at 2923 and 2921 cm⁻¹ were ascribed to asymmetric stretching in methylene groups for both the lignite while the band at 2849 cm⁻¹ corresponding to the symmetric stretching of methylene is more prominent in case of LR and occurs as a shoulder in case of LM. Rest of the samples show both the signature of symmetric and asymmetric stretching of methyl and methylene groups. Extracts from bituminous and subbituminous coals show signatures of in-plane C=C stretching in aromatic rings and hydrogen-bonded carbonyl and aromatic hydrocarbon (C=C) with –O– substitution (Wu et al. 2019). Bending vibrations of methylene and methyl groups are also present in spectra from all the samples except for AS near the 1463 cm⁻¹ and 1377 cm⁻¹ range.

FT-IR spectra of solvent-extracts of studied coal samples

Out-of-plane aromatic CH bending modes were observed in the 900–700 cm⁻¹ region for samples BNCA and BKC only and these were assigned to aromatic structures with isolated aromatic hydrogens (877 and 867 cm⁻¹), two adjacent hydrogens per ring (812 cm⁻¹) and four adjacent hydrogens (747 and 754 cm⁻¹). These OPLA modes are sensitive to ring substitution and give rise to different components according to the number of adjacent hydrogen atoms on the ring (Centrone et al. 2005). The bituminous samples BNCA and BKC show the highest intensities of absorption bands in the 1300–500 cm⁻¹ region. Of all the spectra, the highest intensity for the band at 2918 cm⁻¹ was found in solvent-extracts isolated from the subbituminous coal of Talcher, reflecting the aliphatic C–H stretching in methyl and methylene groups. The appearance of bands near 1038 cm⁻¹ indicates the presence of strongly hydrogen-bonded OH groups such as alcohols and aliphatic ethers.

Complications and overlapping of the functional groups were observed mostly in case of extracts from the bituminous samples viz. BNCA and BKC and lignite LM. This might indicate that the samples, even though differ in rank and depositional conditions, have followed a similar path for the evolution of important functional groups. Evolution of some important functional groups such as carbonyls, ethers and amines have been correlated with thermal maturation in previous literatures (Burnham 2017; Sarkar et al. 2023). When correlated with the maceral composition of the sources, these observations can provide a good insight on the contribution of the maceral groups in the formation of mainly lipids and has been discussed in the conclusion part.

4.4 Steady-state fluorescence emission spectroscopy

The fluorescence spectroscopic measurements were carried out in the steady-state fluorescence emission (SSFE) scan mode and the numerical data obtained was used to delineate the Excitation-Emission Matrix (EEM). The EEM for each extract were displayed as colour-coded contour plots where the PAH concentration has been categorised in different peaks. First and second order Rayleigh scattering peaks were not corrected for the plots and can be easily identified as diagonal bands occurring at the same and double of the wavelength as the excitation light, respectively (Birdwell and Engel 2010; Doskočil 2016).

Maxima of the extracts were devised and classified in the fashion mentioned in previously reported literatures. Enev et al. (2014) and Rodriguez et al. (2014) had designated Peak A for humic substances, whereas Birdwell and Engel (2010) reported the occurrence of peak A at excitation/emission wavelength pair of 240–260/400–460 nm. A modified peak A’ have been proposed by Doskočil et al. (2016) for fluorophores with maxima in the ultraviolet region [260–275/370–440 nm], highlighting the possibility of contribution of lipids in the total fluoresce of humic materials. It is plausible that these lipids participate in the fluorescence of humic acids at excitation wavelength longer than 400 nm (Enev et al. 2014). Peak C corresponds to 300–400 nm excitation wavelength, as reported by Coble et al. (1996), whereas Birdwell and Engel (2010) had designated the 320–360/420–460 nm region as peak C. Peak M at 290–310/370–410 nm, peak B at 270–280/300–315 nm and peak T at 270–280/345–360 nm has also been reported (Birdwell and Engel 2010; Doskočil et al. 2016). Maxima with excitation wavelength greater than 400 nm are also reported from previous literatures (Enev et al. 2014; Tamamura et al. 2015; Doskočil et al. 2016). Doskočil et al. (2016) had proposed a peak V for values greater than 400/470 nm. The longer emission wavelength (> 470 nm) values can result from intramolecular charge transfer states, instead of from individual fluorophores.

Shorter emission wavelength (350–400 nm) with high fluorescence intensity have been associated with lighter components having lower aromatic content and prevalent electron-donor groups like hydroxyls and alkoxyls. Higher emission wavelength with relatively low fluorescence intensity, however, are characteristic of heavier components with condensed aromatic rings and electron-receptor groups like carboxyls and carbonyls (Peuravuori et al. 2002).

In the present study, the devised peaks based on excitation/emission wavelength were identified and correlated on the basis of the ranks of their sources (Fig. 5). Characterisation of the extracts have distinguished between the occurrence of different types of fluorophores reflecting their origins. LM and LR samples which are the derivatives from lignite, show the presence of a single peak corresponding to the UVA-excited region that reflects the terrestrial origin of the humic organic matter (Birdwell and Engel 2010). This observation also indicates the allochthonous nature of the carbon present in the OM content. The bituminous-extracts viz. BNCA and BKC are dominated by humic fluorophores and corresponds to the region of UVC-excitation. In previous literatures (Birdwell and Engel 2010; Coble 1996), the shorter emission wavelength region of this feature has been linked to autochthonous carbon production in the DOM. One interesting observation from the EEM is that for SB, the distribution of PAHs tends to spread towards the protein-like (tryptophan) fluorophore end. The development of peak in the UVA-excited region is taking place in case of SB, which is not well developed in case of BKC and BNCA, reflecting the allochthonous nature of the original organic matter. On the other hand, sample AS indicates a bimodal distribution evident from the presence of smaller fluorophores reflected from the low excitation and low emission value region and larger fluorophores of terrestrial humic origin from the high emission value region. Further correlation of the spectroscopic and petrographic data is presented in the summary section. 3D representations of the distribution of PAHs as per the EEM is provided in Fig. 6 for better visualization purpose. Table 5 contains the peak positions of excitation–emission wavelength pairs for fluorescence regions identified from the EEM.

Excitation-emission matrix (EEM) plots of studied solvent-extracts

3D topographic representation of fluorescence EEM plots for the studied solvent-extracts. Horizontal axes represent the excitation wavelength and emission wavelength intervals at 20 nm; vertical axis represent relative fluorescence intensity. Spectra were acquired in 3D scan mode in a quartz cuvette of 1 cm path length

The extracts have a predominantly aliphatic character, highest for the extracts from LR lignite and least for anthracite. The extracts include oxygen-containing functional groups such as carboxyls, hydroxyls (including phenols), esters, ethers, and conjugated carbonyls. Extracts from LR and LM lignite comprise a greater number of shorter carbon chains compared to the other samples. These extracts contain lipids with fluorophores that participate in the total fluorescence of the extracts. The EEM spectra of lignites were dominated by concentration of fluorophores in the humic-acid region which possibly signifies the origin of the precursor of the lipids in the coals. The maxima are at the same position or very similar excitation/emission positions, which implies that the fluorophores are of the same form, or belong to very similar fluorophore families. However, the presence of multiple peaks in samples from higher ranks of coals viz. subbituminous and bituminous suggest the occurrence of different polycyclic aromatic hydrocarbon ring structures present within the same solvent-extract. It was observed that the anthracite contains very less amount of these optically active fluorophores hence there was negligible response from the AS sample when fluorescence spectroscopy was performed. Our study integrated micropetrography of the original coal samples with spectroscopic techniques, in order to provide insights on the relation of macerals with the solvent-extracted fractions.

The EEM spectra of solvent-extracts can be used as fingerprints to distinguish between solvent-extracted fractions of different ranks of coals and among coals of the same rank as well. With the help of existing literature, it was possible to classify the EEM into different regions which reflect the characters of the extracts based on their constituent moieties. The peak regions presented in Table 5 indicate the fluorescing regions of the humic materials in the extracts according to Birdwell and Engel (2010), Chen et al. (2003), and Coble (1996), and could be correlated with the regions mentioned by Doskočil et al. (2016). In case of our samples, it has been observed that the solvent-extracted fraction contains such fluorophores which reflect their particularly humic nature. For the low rank coals, the solvent-extracts show a very strong humic influence while for medium rank coals, the solvent-extracts tend to show a more bimodal distribution of PAHs. The peak regions are well developed in case of a particular solvent-extract from the subbituminous sample SB which contains a higher amount of liptinite and perhydrous vitrinite. The signature of simple aromatic proteins might come from the influence of the higher hydrogen content exerted by these maceral groups, which is not seen in case of any other samples. For high rank coal, the loss in hydrogen content is reflected through the low intensity of fluorescence and overall, a smaller peak region.

In the present study, characterisation of solvent-extracts from different ranks of coals, ranging from lignite to anthracite, was performed and the analyses demonstrate the following conclusions. The original anthracite sample (AS) was almost completely devoid of liptinite macerals and was highly metamorphosed as a result of intense deformation in the Himalayan fold-thrust belt. From the FT-IR spectroscopy data it is evident that coal with no liptinite content is still capable of producing aliphatic CH signatures in the solvent-extracted fraction, possibly as a contribution of the vitrinite macerals present in the precursor coal. Therefore, it can be inferred that both the liptinite and vitrinite maceral groups are capable of contributing towards aliphatic CH₂ and CH₃ present in the solvent-extracts. It has been observed from the present study that the extracted fraction from SB—a subbituminous coal with unusually higher liptinite content for its rank, has the highest degree of electronic absorption among the samples of similar rank. The nature of the spectrum follows the pattern of solvent-extracts from the two other bituminous samples but with a higher intensity.

Many of the earlier workers have suggested that liptinites, the main contributor to coal lipids, are dominated by the presence of aliphatic components. The use of EEM in conjunction with micropetrography and FT-IR spectroscopy in this study reveals that vitrinite present in coal samples is also capable of generating aliphatic and polyaromatic signatures in the solvent-extracted fraction. This indicates the contribution of the hydrogen present in vitrinite that imparts its reactive character.

Extracts from lignites (LM and LR) containing lesser aromatics and more aliphatics tend to exhibit fluorescence in the 400–450 nm region and the fluorophore responses are restricted to a specific type only. Bimodal distribution of the peaks has been observed in the extracts from the bituminous (BNCA and BKC) and subbituminous (SB) samples, containing both aliphatic and polycondensed aromatic structures. Smaller PAHs are more fluorescence sensitive and quick to respond to EEM but when variation is introduced to the structure, multimodal distribution of these fluorophores have been noted. Solvent-extracts of the lignite samples are dominated by a small number of individual fluorophores or the same fluorophore family, while the solvent-extracts from the more mature coals are broadly influenced by the combined effects of different fluorophores.

The use of EEM with other geochemical proxies can be a very useful and potent way of obtaining new perspectives on coal characterisation on a molecular level, and highlights the necessity for refinement of classical analytical techniques such as micropetrography. It is worth mentioning that micropetrography is definitely useful for establishing the rank and maceral compositions of coals and can provide ready insight into the character of coals, but the procedure is time-consuming and tedious. With the help of different spectroscopic techniques, finer details with more clarity can be attained, but it requires experiments with a larger number of coal samples. The future scope of this work includes the association and incorporation of coal samples from all over India so that a clearer correlation can be drawn. Utilization of coal as a source for cleaner fuels in terms of liquefaction, requires more in-depth knowledge of the moieties which is not possible to obtain solely through the help of tools such as petrography. Therefore, the future scope of this study includes the possibility to provide more detailed knowledge on coal composition which can further enhance the utilization potentiality of coals.