© Author(s) 2019. CC Atribution 4.0 License GEOLOGIJA 62/2, 219-236, Ljubljana 2019 https://doi.org/10.5474/geologija.2019.010 Multielemental composition of some Slovenian coals determined with k0-INAA method and comparison with ICP-MS method Multielementna sestava nekaterih slovenskih premogov dolocena s k0-INAA metodo in primerjava z ICP-MS metodo Tjaša KANDUC1*, Timotej VERBOVŠEK2, Rok NOVAK1 & Radojko JACIMOVIC1 1Department of Environmental Sciences, Jožef Stefan Institute, Jamova cesta 39, SI-1000 Ljubljana, Slovenia; e-mail: tjasa.kanduc@ijs.si 2Department of Geology, Faculty of Natural Science and Engineering, University of Ljubljana, Aškerceva 12, 1000 Ljubljana, Slovenia Prejeto / Received 4. 10. 2019; Sprejeto / Accepted 4. 12. 2019; Objavljeno na spletu / Published online 24. 12. 2019 Key words: k0-instrumental neutron activation analysis (k0-INAA), multielemental composition, coal, PCA analysis, Slovenia Kljucne besede: k0-instrumentalna nevtronska aktivacijska anliza (k0-INAA), multielementna sestava, premog, PCA analiza, Slovenija Abstract In this multi-elemental study, 34 elements (Ag, As, Au, Ba, Br, Ca, Cd, Ce, Co, Cr, Cs, Eu, Fe, Ga, Hg, Hf, K, La, Mo, Na, Nd, Rb, Sb, Sc, Se, Sm, Sr, Ta, Tb, Th, U, Yb, Zn and Zr) were analysed in Slovenian coals from operative (Velenje) and non-operative (Kanižarica and Senovo) coal mines and an imported Indonesia coal using k0-Instrumental Neutron Activation Analysis (k0-INAA) and compared to inductively coupled plasma-mass spectroscopy (ICP-MS). Weaker regressions between both methods ICP-MS and k0-INAA are obtained for following elements: Cs, Co, Eu, Se, Sm and Tb with low concentration (below 1 mg/kg). The k0-INAA data are comparable to the ICP-MS data for the majority of elements. The levels of major elements measured with k0-INAA are as follows: Ca>Fe>K>Na>Sr>Ba. Minor and trace elements, as well as rare earth elements (REEs), are comparable with coal values worldwide. Data of trace elements in coal are important since they are related to air emissions. According to our data obtained with both methods (ICP-MS and k0-INAA) we can conclude that concentrations of trace elements, which impact to human health and are combusted (Indonesian and Velenje coal) in Slovenia are comparable to world averages coal. Izvlecek V tej raziskavi smo izmerili s k0-INAA (instrumentalno nevtronsko aktivacijsko analizo) metodo nekaj izbranih slovenskih premogov iz velenjskega premogovnika in ne operativnih premogovnikov: Kanižarica in Senovo. Prav tako smo s to metodo analizirali vzorec iz Indonezije (uvožen premog) in ga primerjali z že objavljenimi rezultati pridobljenimi z ICP – MS (masna spektrometrija z induktivno sklopljeno plazmo) metodo. S k0-INAA metodo smo dolocili naslednje elemente: Ag, As, Au, Ba, Br, Ca, Cd, Ce, Co, Cr, Cs, Eu, Fe, Ga, Hg, Hf, K, La, Mo, Na, Nd, Rb, Sb , Sc, Se, Sm, Sr, Ta, Tb, Th, U, Yb, Zn in Zr. Rezultati meritev pridobljeni s k0-INAA metodo so za vecino elementov, obravnavanih v tej raziskavi, primerljivi z rezultati meritev pridobljenih z ICP-MS metodo. Slabše regresije med metodami ICP-MS in k0-INAA dobimo le pri nekaterih elementih (Cs, Co, Eu, Se, Sm and Tb) za katere so znacilne nizke koncentracije (pod 1 mg/kg). Koncentracije glavnih elementov merjenih s k0-INAA metodo v premogu se znižujejo kot sledi: Ca> Fe> K> Na> Sr> Ba. Elementi z nizkimi koncentracijami in elementi redkih zemelj (REE) so primerljivi z vrednostmi premoga po vsem svetu. Podatki slednih elementov v premogu so pomembni, ker so povezani z emisijami v zraku. Glede na naše podatke pridobljene z obema metodama (ICP-MS, k0-INAA) lahko zakljucimo, da so koncentracije slednih elementov, ki vplivajo na clovekovo zdravje in jih sežigamo (premog iz Velenja in Indonezije) v Sloveniji primerljivi s povprecnimi vrednostmi svetovnih premogov. 220 Tjaša KANDUC, Timotej VERBOVŠEK, Rok NOVAK & Radojko JACIMOVIC Introduction The chemical analysis of coal includes, as well as, proximate (Khandelwal and Singh, 2010, Yi et al., 2017) (moisture, volatile compounds, ash content, fixed carbon) and ultimate analyses (car­bon, hydrogen, sulphur, oxygen, and nitrogen), the analysis of major, minor and trace elements. Usually, these elements are measured using in­ductively coupled plasma-mass spectrometry (ICP-MS) (Finkelman et al., 2018) and instrumen­tal neutron activation analysis (k0-INAA) (Wag­ner and Matiane, 2018, Lin et al., 2018) methods. Other methods for determining trace elements in­clude inductively coupled plasma optical emission spectrometry (ICP-OES) (Finkelman et al., 2018), hydride generation atomic absorption spectrome­try (HAAS) (Chen et al., 2011) and X-Ray Fluores­cence spectrometry (XRF) (Chen et al., 2011). It is widely known that these trace elements can occur in a wide variety of chemical forms or modes of occurrence, which determines the environmental, economic, technological impact, which in some cases can be significant (Finkelman, 1995, 2018). Twenty-five potential harmful trace elements (PHTEs) are typically present in coal in inorgan­ic and organic forms (Radenovic, 2006). Among them As, Be, Cd, Cr, Co, Hg, Mn, Ni, Pb, Se, Sb and U are all potential air pollutants (Gürdal, 2008). Ketris and Yudovich (2009) include rare earth elements, yttrium, and scandium (REY + Sc) in the table of coal Clarke values, which has been a highly useful tool for making geochemical comparisons of coals globally. Indonesian coals are generally low in ash and sulphur, but have high content of volatile matter. They are classified as low rank coals with low caloric value. The sulphur content varies from 0.1 to 1 % (Internet 1). Elemental composition (wt %, dry basis) of TOT S varied for Velenje samples from this study from 1.4 to 3.9 %, Kanižari­ca from 1.6 to 2.2 % and Senovo 1.9 % (Burnik Šturm et al., 2009). The geological composition of the Velenje ba­sin is described in detail in Brezigar et al. (1987). The origin of the Velenje basin is related to the transtention between Šoštanj and Smrekovec faults. In the pre-Pliocene basement of the ba­sin, Triassic carbonates and dolomites prevail on the northeastern side of the Velenje fault, while Oligocene to Miocene clastic strata, consisting predominantly of marls, sandstones and volca­noclastics are dominant on the south-western side of the fault. The alkaline, calcium-rich envi­ronment during formation of Velenje basin also caused a relatively high degree of gelification, which is significantly higher than the degree of gelification observed in other lignites (Markic & Sachsehofer, 1997; Šlejkovec & Kanduc, 2005; Markic & Sachsehofer, 2010) as well as coals in­vestigated in our study. A well known relation between alkalinity and gelification was clearly ascertained in the case of the Velenje lignite. Lignite samples with the highest calcium con­tents were also the samples with the strongest gelification (Markic & Sachsenhofer, 1997). The macroscopic description of the lignite samples, in term of lithotypes, was determined following the lithotype classification criteria for brown coals (lignites) provided by the International Committee for Coal Petrology (ICCP, 1993) and are described by Burnik Šturm et al. (2009). All of the samples from the Velenje excavation field -50/C in this study are classified as gelified de­trital lignite (Kanduc et al., 2018). The lithologi­cal columns for Senovo, Kanižarica and Trbovlje are also presented in Burnik Šturm et al. (2009) and references therein (Brezigar, 1987; Kušcer, 1967; Markic et al., 1991). The macroscopic de­scription of the lignite samples in terms of previ­ous petrological (Markic & Sachsenhofer, 1997), geochemical and isotopic studies of light ele­ments C, H, O, N, S (Bechtel et al., 2003; Kanduc et al, 2005; Burnik Šturm et al., 2009; Kanduc & Šlejkovec, 2005; Kanduc et al., 2012; Kanduc et al., 2018; 2019, Liu et al., 2019) were performed in the frame of various research projects. For example, three different lithotypes (xylitic, geli­fied and matrix) of Pliocene lignite for the Ve­lenje basin, Slovenia, were investigated to estab­lish the variations of biomarker compositions in solvent extracts and stable isotope composition of carbon and nitrogen in bulk material (Liu et al., 2019). All of these studies were focused on the Velenje basin since it is currently the only actively mined basin in Slovenia and is one of the biggest underground coal mines in Europe. All three of the Velenje lithotypes reflect the composition of the original plant material in the paleomire (Markic & Sachsenhofer, 1997). Arse­nic speciation studies and the different forms of calcite present in the coal suggest that bacterial activity was a significant factor during sedimen­tation of the basin (Kanduc & Šlejkovec, 2005; Kanduc et al., 2018; Kanduc et al., 2019a). The analysis of other geological matrixes such as coalbed gas (Kanduc & Pezdic, 2005; Kanduc et al., 2012, Sedlar et al., 2014) and groundwater (Kanduc et al., 2014; Kanduc et al., 2019b) reveal more evidence of bacterial activity during sedi­mentation of the basin. In the study of Kanduc et al. (2019a) organic and inorganic coal samples from -50/C excava­tion field of Velenje basin were measured using ICP-MS and revealed that the concentrations of the majority of the analysed elements were ei­ther equal to or below the global average for coal. Exceptions were Mo (7.76 ± 4.76 µg/g, 3.5 times higher) and U (5.24 ± 3.23 µg/g, 1.8 times higher) in organic-rich samples. It was found that higher than normal are concentrations of U (5-15 ppm – in comparison to 0.5-10 ppm concentrations in world coals), and of Mo (5-20 ppm – in comparison to 0.1-10 ppm in in world coals). Both elements are presumed to be organically bound (Markic & Sachsenhofer, 2010). This study aims to present results of major, minor and trace elements measured using k0-INAA method in coal samples collected from operative (Velenje) and non-operative (Kanižarica and Senovo) Slovenia coal mines. The study also analysed an Indonesia coal supplied by the thermal power plant Moste. Additionally, one of the objectives was to compare k0-INAA and ICP-MS methods used to analyse the same coal samples (Kanduc et al., 2019a, Supplementary material) from Velenje coal mine and perform a statistical analysis (PCA-Principal Component Analysis) of all data (Velenje, Senovo, Kanižarica, Indonesia coals) measured with k0-INAA method. Methods Sampling locations were taken from a local borehole database in the local coordinate system from the Velenje coal mine. Coordinates were then transformed to Gauss-Krüger D48 Sloveni­an national coordinate system and indicated on a hill-shaded relief map generated using the ESRI ArcGIS mapping software (Fig. 1). Figure 1A was produced using data from the Shuttle Radar To­pography Mission SRTM data at 90 m spatial reso­lution. A more detailed map (Fig. 1B), was created using the digital elevation model at a 1 × 1 m spa­tial resolution, using LiDAR data form the nation­al scanning campaign of the Slovenian territory (ARSO, 2014). Figure 1C includes the position of excavation field (-50/C) and cross-section of Velen­je basin with main geological and tectonic units. Samples of coal were collected from the fol­lowing mining areas in Slovenia (Fig. 1): Senovo (3 samples), Kanižarica (4 samples), Velenje ba­sin (7 delivery roadway samples, and 18 samples from excavation field -50/C), Indonesia (1 sample) in years 2004, 2005 and 2013. The Moste thermal power plant provided the sample of Indonesian coal. For k0-INAA analysis, samples (240-290 mg) were sealed in a pure polyethylene ampoule (SPRONK system, Lexmond, The Netherlands). For the determination of long-lived radionuclides, samples and standards (Al-0.1 %Au IRMM-530R disc of 7 mm in diameter and 0.1 mm thick) were stacked together and fixed in a polyethylene am­poule in sandwich form and irradiated for 12 hours in the carousel facility (CF) of a 250 kW TRIGA Mark II reactor (Jožef Stefan Institute, JSI) at a thermal neutron flux of 1.1×E+12 cm-2 s-1. Each sample was measured three times af­ter 2, 8-13 and 25-30 days cooling time on three absolutely calibrated HPGe detectors with 40 % and 45 % relative efficiency. Measurements were carried out at a distance such that the dead time remained below 10 % with negligible random coincidences. The detectors with 40 % relative efficiency were connected to a MULTIPORT II (Canberra) computerized multichannel analy­ser (MCA) in LT mode operating with GenieTM 2000 spectroscopy software, while the detector with 45 % relative efficiency was connected to a DSPEC PLUS (Ortec) multichannel analyser in ZDT mode operating with Maestro®-32 spectros­copy software. The HyperLab (2002) program was used for peak area evaluation, whereas for the determi­nation of f (thermal to epithermal flux ratio) and a (a parameter which represents the epithermal flux deviation from the ideal 1/E distribution) the “Cd-ratio” method for multi-monitor was applied (Jacimovic et al., 2003). The values obtained for f = 28.63 and a = -0.0011 were used to calculate the element concentrations. The elemental con­centrations and effective solid angle calculations were performed using the KayWin® (Kayzero for Windows, 2011) software package. Ranges of uncertainties with coverage factor k = 1 (%) for measured elements with k0-INAA method is as follows: i) uncertainty for elements: As, Br, Ca, Ce, Cs, Fe, Na, Sc, U, and Zn ranges from 3.5 to 7.3 % and ii) uncertainty for elements: Au, Ba, Co, Cr, Eu, Ga, Hf, Hg, K, La, Mo, Nd, Rb Sb, Se, Sm, Sr, Ta, Tb, Th, Yb, ranges from 3.5 to 28 %. Measured elements with higher concen­tration have lower uncertainties, while elements with lower concentration have higher uncertain­ties. Chemical analysis of Velenje coal samples (13-2123, 13-2125, 13-2130, 13-2134, 13-2138, 13-2141, 13-2145, 13-2157, 13-2162) were performed with ICP – MS method in ACME lab Canada (http://acmelab.com/services/). For the analysis of SiO2, Al2O3, Fe2O3, CaO, MgO, Na2O, K2O, MnO, TiO2, P2O5, Cr2O3, Ce, Co, Cu, and Zn samples were mixed with a LiBO2/Li2B4O7 flux. Crucibles were fused in a furnace. The cooled beads were dis­solved in ACS grade nitric acid and analyzed by ICP and or ICP-MS. Other elements (Ce, Co, Cs, Dy, Er, Eu, Gd, La, Ni, Nb, Nd, Pr, Rb, Sm, Sr, Tb, U, Th, V, Zr, Y) were measured with ICP-MS method. Total carbon (TOT C) and total sulphur (TOT S) were measured using LECO Carbon –Sulphur analyz­er. The mean limits of detection for both elements were 0.02 %. Loss on ignition (LOI) was deter­mined by igniting a sample split and then mea­suring the weight loss. For Ag, As, Au, Bi, Cd, Hg, Mo, Ni, Pb, Sb, Se, Tl, and Zn analysis, prepared samples were digested with modified Aqua Regia solution of equal parts of concentrated HCl, HNO3 and Mi­liQ H2O for 1 h in a heating block or in a hot water bath at 95°C. Samples were made up to volume witgh diluted HCl. Sample splits of 0.5 g were an­alyzed optional 15 g or 30 g digestion available for AQ200. Samples were analyzed using induc­tively coupled-mass spectrometry (ICP-MS). The following standards were used for quality assur­ance: STD-SO-18, STD-GGC-02, STD-GS311-1 and STD OREAS45EA. Statistical analysis was conducted using the R language (R Core Team, 2019), and the signifi­cance model was set at p<0.05. A Spearman’s cor­relation analysis was used to identify the rela­tionships between 27 elements (As, Ba, Br, Ca, Ce, Co, Cr, Cs, Eu, Fe, Hf, K, La, Mo, Na, Nd, Rb, Sb, Sc, Se, Sm, Sr, Tb, Th, U, Yb and Zn) with com­plete data sets. The crossed-out values indicate where p-values exceeded 0.05. Principal component analysis (PCA) was used to differentiate (same as for Spearman correla­tion analyses) between the coal from the differ­ent mines. Due to the broad range of elemental concentrations, the dataset was central log-ra­tio transformed. Studied mines were grouped as “Open” and “Closed”. The principle component plots were made using ggplot2 in R (Wickham, 2016). Results and discussion Tables 1 and 2 give the results of the k0-INAA of 34 elements (As, Ba, Br, Ca, Ce, Co, Cr, Cs, Eu, Fe, Hf, K, La, Mo, Na, Nd, Rb, Sb, Sc, Se, Sm, Sr, Tb, Th, U, Yb, Zn) for Velenje, Senovo, and Kanižarica mines and for an imported Indone­sian coal. In a previous study by Kanduc et al. (2019a), ten oxides (SiO2, Al2O3, Fe2O3, MgO, CaO, Na2O, K2O, TiO2, P2O5, MnO, Cr2O3), LOI (Loss on ignition), TOT C (Total carbon), TOT S (Total sul­phur) (Kanduc et al., 2019a), along the following toxicologically and environmentally relevant el­ements: As, Ba, Ce, Co, Cs, Cu, Dy, Er, Eu, Gd, Hf, La, Mo, Nb, Nd, Ni, Pb, Pr, Rb, Se, Sm, Sr, Tb, Th, U, V, Y, Zn, Zr were measured in the organic-rich component of the Velenje samples. In this study, nine samples from the Velenje coal mine (13-2123, 13-2125, 13-2130, 13-2134, 13-2138, 13-2141, 13-2145, 13-2157, 13-2162), were measured using ICP-MS and k0-INAA, while all other samples in this study were measured using only k0-INAA. For studying elemental composition of coal with k0-INAA method we choose only Velenje samples (from year 2013) that were organic rich, besides Kanižarica, Senovo and Indonesia coal samples that were sampled in years 2004 and 2005. Results of ICP – MS of major elements, LOI, TOT C, TOT S in coal samples (13-2123, 13-2125, 13-2130, 13-2134, 13-2138, 13-2141, 13-2145, 13-2157, 13-2162) collected from excavation field -50/C of Velenje basin are presented in Table 3a. Results of ICP – MS of trace elements in coal samples (13-2123, 13-2125, 13-2130, 13-2134, 13-2138, 13-2141, 13-2145, 13-2157, 13-2162) collect­ed from excavation field -50/C of Velenje basin are presented in Table 3b. Data (REEs) for coals from other locations (two power plants: Jungar (China), Tutuka (SA), Matla (SA), and the Wit­bank Coalfield (SA) are included for comparison purposes (Table 4). For quality assurance and quality control (QA/QC), in the study we used the certified refer­ence material BCR-180 Gas Coal (Fig. 2). The re­sults obtained by k0-INAA are in good agreement with the certified data for As, Hg, Se and Zn. It should be mentioned that expanded uncertainty (k=2) of mass fraction of Hg obtained via Hg-203 at 279.2 keV is relatively high in comparison with certified value due to correction from the mass fraction of Se via Se-75 at 279.5 keV, which was about 70 % (Fig. 2). Among the major elements, Ca prevails. Ma­jor oxides (CaO, Na2O, K2O, TiO2) and ultimate analysis (LOI, TOT C, TOT S) of the coal samples (13-2123, 13-2125, 13-2130, 13-2134, 13-2138, 13-2141, 13-2145, 13-2157, 13-2162) range as follows (Table 3 a): CaO from 1.91-5.21 %, Na2O rang­es from 0.04 to 0.13 %, K2O ranges from 0.007-0.08 %, TiO2 ranges from 0.07 to 0.08 %, TOT C ranges from 50.6 to 57.1 %, TOT S ranges from 1.17 to 2.46 %, and LOI (Loss on ignition) ranges from 86.7 to 97.1 % (Kanduc et al., 2019a). Fig­ure 3 represents the major oxides (MgO+CaO, Na2O+K2O, SiO2+Al2O3+Fe2O3) present in sam­ples of lignite. The data were obtained from the study by Kanduc et al. (2019a) and are present­ed in Table 3a. The major oxides in the Velen­je coal samples are CaO and MgO, suggesting that lignite was formed in a Ca-alkaline rich environment (Markic and Sachsenhofer, 1997). The most prevalent oxide is CaO (from 1.91 to 5.21 %). The concentration of oxides from the Velenje samples decrease in the following order: CaO>Fe2O3>Al2O3>SiO2>MgO>Na2O>K2O>TiO2. Only two Velenje coal samples (13-2134, 13-2145) have CaO + MgO concentrations less than 70 % (Fig. 3). Figure 4 A-C shows plots of the major (Ba, Ca, Fe, K, Sr), minor (As, Br, Ce, Co, Cr, La, Mo, Nd, Rb, Sc, U, Zn) and trace element levels (Cs, Eu, Hg, Sb, Se, Sm, Ta, Tb, Th, Yb) for each of the coal mine samples (Senovo, Kanižarica, Indone­sia, Velenje). From Figure 4A it can be observed that among major elements Ca prevailed for Ve­lenje coal mine samples, while in one sample of Senovo (Senovo 3) and Kanižarica (Kanižarica 15) Fe prevails. Some samples from Velenje mine (13-2166, 13-2167, 13-2181), from excavation field -50/C have high concentrations of Ca in the range from 163700 to 307100 mg/kg (Fig. 4 A), which is in compliance with thesis of Ca-rich envi­ronment during sedimentation of Velenje basin. Among minor elements there are huge differ­ences between coal samples between mines. The highest concentration of As, Br, Ce, Cr are ob­served in Kanižarica coal samples (Kanižarica 6, Kanižarica 15). Cr and Mo prevail in Velenje coal samples, while Br and Cr prevail in Senovo coal samples (Fig. 4 B). Kanižarica coal samples have also the highest concentration of rare elements (Cs, Eu, Hf, Ta, Th, Se, Sm) (Fig. 4 C). Sm and Th are enriched in three Velenje samples (B91, B105, 13-2149) (Fig. 4 C). Among all minor and trace el­ements Kanižarica coal samples have the highest concentrations (Figs. 4 B, C). Figure 5 presents box-plots of the k0-IN­AA data for all coal samples. From the box-plots it appears that the abundances of Ca>Fe>K>Na>S.r>Ba prevail among major ele­ments and Mo>Zn>U>Cr>As>Br in the case of mi­nor and trace elements. The elements: Ag, Au, Cd, Ga, Zr were excluded from plots since they were not measured in all of the coal samples, but are presented in Tables 1 and 2. Among the 16 REEs (Table 4) recorded in coals from other locations (Taylor and McLennan, 1985, Ketris and Yudovich, 2009, Dai et al., 2010, Akineyeni et al., 2012, Eze et al., 2013, Hart et al., 1982, Kanduc et al., 2019a) only eight elements (Ce, Eu, La, Nd, Sc, Sm, Tb and Yb were deter­mined using k0-INAA and compared with pub­lished REEs values (mg/kg) in coal Clarke val­ues, Jungar Power Plant (China), Tutuka Power Station (SA), Matla Power Station (SA) Witbank Coalfield (SA) (Wagner and Matiane, 2018) and the Velenje basin coal samples measured by ICP-MS. From a comparison of the data, all eight ele­ments from this study (Velenje, Senovo, Kanižar­ica and Indonesia) and the data for the coal from other locations fall in the same range. A comparison of the data for As, Ba, Ce, Co, Cs, Eu, La, Mo, Nd, Rb, Se, Sm, Sr, Tb, Th, U and Zn obtained using k0-INAA and ICP-MS (Tables 3a and 3b) in samples 2123, 2125, 2130, 2134, 2138, 2141, 2145, 2157 and 2162 revel a strong positive correlation (R2>0.8) in the case of As, Ba, Cs, Mo, Nd, Sr and U, (Fig. 6 A-D) and a good positive correlation (R2 from 0.6 to 0.8) was observed for Zn and Rb (Figs 6 B-C). Though less strong, cor­relations (R2 < 0.6) were found for Co, Eu, La, Se, Sm, Tb and Th (Figs. 6 C-D), which occur in low concentrations (< 1 mg/kg). Figure 7 shows the Spearman correlations (R2>90 %) for parameters (As, Ba, Br, Ca, Ce, Co, Cr, Cs, Eu, Fe, Hf, K, La, Mo, Na, Nd, Rb, Sb, Sc, Se, Sm, Sr, Tb, Th, U, Yb, Zn) measured with k0-INAA method from four different mining lo­cations (Kanižarica, Senovo, Indonesia, Velenje). Spearman’s correlation analysis revealed strong positive correlations (R2>0.95) between the fol­lowing elements: Ce-La, Cs-Rb, Cs-Sc, Hf-Sc, Eu-Tb, Cs-Tb, Sc-Tb, Cs-Yb, Hf-Yb, Sc-Yb, and Th-Yb. Principle component analysis (Fig. 8) reveals a strong gradient along the first PCA axis (49.5 %) and has the highest positive correlation with trace elements (e.g., Ce, Co, and Cs) and highest negative correlation with main elements (e.g. Ca, Na, B). These elements have the most discrimi­nant power separating coals from open (Velenje and Indonesia) and closed (Kanižarica and Se­novo) coal mines. The second axis explains an additional 16.1 % of the variance and correlates positively with Ba, Sr and negatively with U, Sb according to PCA multi-elemental grouping (Fig. 8). Conclusion Coal samples from Slovenia (Kanižarica, Ve­lenje, and Senovo) and Indonesia were sampled and analysed by k0-INAA in 2003, 2004 and 2013, while the Velenje coal mine samples (2013) was measured using both k0-INAA and ICP-MS to compare results obtained using both methods. Based on the comparison of both methods, it can be concluded that k0-INAA method is very accu­rate compared to ICP-MS method with no possi­bility of losses of material and contamination. A good correlation between both methods was ob­tained for Ba, Sr, Mo, Zn, U, As, Rb, Nd, while a weak correlation was observed for Th, Se, Cs, Eu, Sm and Tb. The major elements determined by k0-INAA in the Velenje lignite samples (n = 25) are Ca>Fe>Na>K>Sr>Ba while for minor and trace elements Zn>Zr>Mo>U>Br>Cr. In the coal samples from the Kanižarica mine (n=4), the levels of the main elements are Fe>Ca>K>Na>Sr>Ba, while for minor and trace elements Cr>Zr>U>Zn>Rb>Mo. In samples from Senovo mine (n = 3) the main elements are Fe>Ca>K>Na>Sr>Ba, and for trace elements Cr>Zn>As>Zr>Mo, whereas in the Indonesia coal had the following composition of main elements: Fe>Ca>K>Na>Ba>Sr and trace elements: Zn>Cr>Ce>Co. In all cases, Fe and Ca are the most abundant elements, while among trace elements; Zn and Cr are the most abundant. The levels of trace elements of samples from all investigated mines were also in the same range reported in the literature for other mining regions (SA and China). Principal component analysis based on 27 elements (As, Ba, Br, Ca, Ce, Co, Cr, Cs, Eu, Fe, Hg, K, La, Mo, Na, Nd, Rb, Sb, Sc, Se, Sm, Sr, Tb, Th, U, Yb, Zn) revealed good discrimination between coal from the closed (Senovo, Kanižarica) and open mines (Velenje, Indonesia). Further geochemical investigations of coal are required to investigate composition (proximate, ultimate analysis, major, minor and environmen­tally sensitive trace elements) of coal from active excavations in the Velenje coal mine in Slovenia, which is combusted in the Šoštanj thermal power plant and represents 30 % of energetic source in Slovenia. These analyses are essential to ensure the quality of combusted coal, which is related to atmospheric emissions. Acknowledgement The authors would like to thank L1-5451, P1-0143, P1-0195 and the Young researchers pro­gramme founded by Slovenian Research Agency. References Akinyemi, S.A., Gitari, W.M., Akinlua, A. & Petrik, I.F. 2012: Mineralogy and geoche­mistry of sub-bituminous coal and its combu­stion products from Mpumalanga Province, South Africa, Chapter 2. In: Krull, I.S. (ed.): Analytical Chemistry, inTechOpen. https://doi.org/10.5772/50692 ARSO (Agencija Republike Slovenije za oko­lje = Slovenian Environment Agency) 2014. http://gis.arso.gov.si/evode/profile.aspx?id=atlas_voda_Lidar.Arso Bechtel, A., Sachsenhofer, R.F., Markic, M., Gratzer, R., Lücke, A. & Püttman, W. 2003: Paleoenvironmental implications from bio­marker and stable isotope investigations on the Pliocene Velenje lignite seam (Slovenia). Organic Geochemistry, 34/9: 1277-2298. https://doi.org/10.1016/S0146-6380(03)00114-1 Brezigar, A. 1987: Premogovna plast Rudnika lignite Velenje = coal seam of the Velenje coal mine. Geologija, 28/29(1985/86): 319-336 (in Slovene with English summary). Burnik Šturm, M., Lojen, S., Markic, M. & Pezdic, J. 2009: Speciation and isotopic composition of sulphur in low-rank coals from Slovenian coal seams. Acta Chimica Slovenica, 36: 989-996. Chen J., Liu G., Jiang, M., Chan, C.L., Li, H., Wu, B., Zheng, L. & Jiang, D. 2011: Geochemistry of environmentally sensitive trace elements in Permian coals for the Huainan coalfield, Anhui, China. International Journal of coal geology, 88/1: 41-54. https://doi.org/10.1016/j.coal.2011.08.002 Connelly, N.G., Damhus, T., Hartshorn, R.M. & Hutton, A. T. 2005: Nomenclature of Inorganic Chemistry: IUPAC Recommendations 2005. The Red-Book. Royal Society of Chemistry, Cambridge, 340 p. Dai S., Zhao, I., Peng, S., Chou, C. I., Wang, X., Zhang, Y., Li, D. & Sun, Y. 2010: Abundances and distribution of minerals and elements in high-alumina coal fly ash from the Jungar Power Plant, Inner Mongolia, China. International Journal of Coal geology 81: 320-332. Eze, C.P., Fatabo, O., Madzivire, G., Ostrovnaya, T. M., Petrik, I.F., Frontasyeva, M.V. & Nechaev, A.N. 2013: Elemental composition of fly ash: a comparative study using nuclear and rela­ted analytical techiques. Chem. Didact. Ecol. Metrol. 18/1-2: 19-29. https://doi.org/10.2478/cdem-2013-0014 Finkelman, R.B. 1995: Modes of occurrence of environmentally sensitive trace elements of coal. In: Swaine, D.J. & Goodarzi, F. (eds.): Environmental aspects of trace elements of coal. Kluwer Academic Publishers, the Netherlands: 24-50. Finkelman, R.B., Palmer, C.A. & Wang, P. 2018: Quantification of the modes of occurren­ce of 42 elements in coal. Int. J. Coal Geol. 185/2: 138-160. https://doi.org/10.1016/j.coal.2017.09.005 Gürdal, G. 2008: Geochemistry of trace elements in Can coal (Miocene), Canakkale, Turkey. International Journal Coal Geology 74/1: 28-40. https://doi.org/10.1016/j.coal.2007.09.004 Hart, R.J., Leahy, R., & Falcon, R. M. 1982: Geochemical investigation of the Witbank Coalfield using Instrumental Neutron Activation Analysis. J. Radioanal. Chem. 71/1-2: 285-297. https://doi.org/10.1007/BF02516156 HyperLab 2002: System, Installation and quick start guide, HyperLabs Software, Budapest, Hungary. IRMM 2015: Joint Research Centre, Institute for reference materials and measurements, Certified Reference material BCR-180, Certificate of Analysis, 3 p. Jacimovic, R., Smodiš, B., Bucar, T. & Stegnar, P. 2003: k0-NAA Quality Assessment by Analysis of Different Certified Reference Materials Using the KAYZERO/SOLCOI Software. Journal of Radioanalytical and Nuclear Chemistry, 257/3: 659-663. https://doi.org/10.1023/A:1026116916580 Kanduc, T. & Pezdic, J. 2005: Origin and distri­bution of coalbed gases from the Velenje ba­sin, Slovenia. Geochem. J., 39: 397-409. Kanduc, T., Markic, M., Zavšek, S. & McIntosh, J. 2012: Carbon cycling in the Pliocene Velenje Coal Basin, Slovenia, inferred from stable carbon isotopes. International Journal of Coal Geology, 89/1: 70-83. https://doi.org/10.1016/j.coal.2011.08.008 Kanduc, T., Grassa, F., McIntosh, J., Stibilj, V., Ulrich-Supovec, M., Supovec, I. & Jamnikar, S. 2014: A geochemical and stable isotope investigation of groundwater/surface water interactions in the Velenje basin, Slovenia. Hydrogeol. J., 22/4: 971-984. https://doi.org/10.1007/s10040-014-1103-7 Kanduc, T., Vreca, P., Gregorin, Š., Vrabec, M., Vrabec, M. & Grassa, F. 2018. Authigenic mi­neralization in low rank coals from the Velenje basin, Slovenia. J. Sediment. Res., 88/2: 201-213. https://doi.org/10.2110/jsr.2018.7 Kanduc, T., Šlejkovec, Z., Mori, N., Vrabec, M., Verbovšek, T., Jamnikar, S. & Vrabec, Ma. 2019a: Multielemental compositi­on and arsenic speciation in low rank coal from the Velenje Basin, Slovenia. Journal of Geochemical Exploration, 200: 284-300. https://doi.org/10.1016/j.gexplo.2018.08.001 Kanduc, T., Šlejkovec, Z., Vreca, P., Samardžija, Z., Verbovšek, T., Božic, D., Jamnikar, S., Solomon, D. K. Fernandez, D. P., Eastoe, C., McIntosh, J., Mori, N. & Grassa, F. 2019b: The effect of geochemical processes on gro­undwater in the Velenje coal Bain, Slovenia: insights from mineralogy, trace elements and isotopes signatures, SN Applied Sciences, 1/11: 1518-1-1518-30. https://doi.org/10.1007/s42452-019-1661-6 Kayzero for Windows (KayWin®) User’s Manual, for reactor neutron activation analysis (NAA) using the k0-standardization method, Version 2.42 (2011). Ketris, M. & Yudovich, Y. 2009: Estimations of Clarkes for Carbonaceous Biolithes: world averages for trace element contents in black shales and coals. International Journal of Coal geology 78/2: 135-148. https://doi.org/10.1016/j.coal.2009.01.002 Khandelwal M. & Singh, T.N. 2010: Prediction of macerals contents of Indian coals from proximate and ultimate analyses using arti­ficial neural networks. Fuel, 89/5: 1101-1109. https://doi.org/10.1016/j.fuel.2009.11.028 Kušcer, D. 1967: Zagorski terciar = Tertiary Formations of Zagorje. Geologija, 10: 5-85. Lin R., Soong Y. & Granite E.J. 2018: Evaluation of trace elements in U.S. coals using the USGS COALQUAL database version 3.0. Part I: Rare earth elements and yttrium (REY). International Journal of Coal Geology, 192: 1-13. https://doi.org/10.1016/j.coal.2018.04.004 Liu, B., Vrabec, M. & Püttman, W. 2019: Reconstruction of paleobotanical and pa­leoenvironmental changes in the Pliocene Velenje Basin, Slovenia by molecular and sta­ble isotope analyses of lignites. International Journal of Coal Geology, 206: 31-45. https://doi.org/10.1016/j.coal.2019.03.006 Markic, M. 1995: Elaborat o kategorizaciji, klasi­fikaciji in izracunu zalog premoga na obmocju Rudnika rjavega premoga Kanižarica-stanje 31.12..1994. Elaborat, arhiv GeoZS, Ljubljana: 77 p. Markic, M. & Sachsehofer, R.F. 1997: Petrographic composition and depositional environments of the Pliocene Velenje lignite seam (Slovenia). International Journal of coal ge­ology, 33/3: 229-254. https://doi.org/10.1016/S0166-5162(96)00043-2 Markic, M. 2006: Anorgansko – geokemicna opre­delitev velenjskega lignite v reprezentativnem profile vrtine P-9k/92 = Inorganic geochemi­cal characterization of the Velenje lignite in the representative P-9k/92 borehole profile (Slovenia). Geologija, 49/2: 311-338. https://doi.org/10.5474/geologija.2006.023 Markic, M. & Sachsenhofer R.F. 2010: The Velenje lignite- its petrology and genesis. Geološki zavod Slovenije, Ljubljana: 218 p. R Core Team 2019: R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/. Radenovic, A. 2006: Inorganic constituents in coal. Kem Ind. 55/2, 65-71. Sedlar, J., Kanduc, T., Jamnikar, S., Grassa, F. & Zavšek, S. 2014: Distribution, composition and origin of coalbed gases in excavation fiel­ds from the Preloge and Pesaje mining areas, Velenje Basin, Slovenia. Int. J Coal Geol., 131: 363-377. Shuttle Radar Topography Mission SRTM data. http://srtm.csi.cgiar.org/ Taylor, S.R. & McLennan, S. M. 1985: The Continental crust: its composition and evolu­tion. Blackwell, Oxford: 312 p. USGS Fact Sheet, 2014. http://pubs.usgs.gov/fs/2014/3078/pdf/fs2014 3078.pdf. Yi, L., Feng, J., Qin, Y.H. & Li, W.Y. 2017: Prediction of elemental composition of coal using proximate analysis. Fuel, 193/1: 315-321. https://doi.org/10.1016/j.fuel.2016.12.044 Wagner, N. J. & Matiane, A. 2018: Rare ae­arth elements in select Main Karoo Basin (South Africa) coal and coal ash samples. International Journal of Coal Geology, 196: 82-92. https://doi.org/10.1016/j.coal.2018.06.020 Wickham, H. 201: ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York. Internet source: Internet 1: http://globalenergycertification.org/indonesian-coal-quality-coal-reserves/: Indonesian Coal quality and coal reserves (ci­ted 29.11.2019 221 Multielemental composition of some Slovenian coals determined with k0-INAA method and comparison with ICP-MS method 222 Tjaša KANDUC, Timotej VERBOVŠEK, Rok NOVAK & Radojko JACIMOVIC Fig. 1. General map of coals located in Slovenia showing the study area of sampled coals from Slovenia mines: Velenje Coal Basin (active coal mine, n = 25), Kanižarica (closed, n = 4), Senovo (closed, n = 3), and Indonesia (coal imported in Slovenia, n = 1). Velenje sampling locations from years 2004, n = 4 (B91, B105, B107, B113) and 2013, n = 18 (2113, 2116, 2119, 2123, 2125, 2126, 2130, 2134, 2138, 2141, 2143, 2145, 2149, 2157, 2162, 2166, 2167, 2181). B. Detailed map of Velenje sampling locations from years 2004 and 2013 are presented. C. Position of excavation field -50/C from where samples were taken and cross-section of the central part of the Velenje basin (modified from Brezigar, 1987) with main geological and tectonic units. 223 Multielemental composition of some Slovenian coals determined with k0-INAA method and comparison with ICP-MS method Fig. 1. 224 Tjaša KANDUC, Timotej VERBOVŠEK, Rok NOVAK & Radojko JACIMOVIC 225 Multielemental composition of some Slovenian coals determined with k0-INAA method and comparison with ICP-MS method Table 1. Elements (Ag, As, Au, Ba, Br, Ca, Cd, Ce, Co, Cr, Cs, Eu, Fe, Ga, Hf, Hg, K, La, Mo, Na, Nd, Rb, Sb, Sc, Se, Sm, Sr, Ta, Tb, Th, U, Yb, Zn, Zr) measured with k0-INAA method in following coal samples: Senovo (Sen., n = 3), Indonesia (Indon., n = 1), Kanižarica (Kan., n = 4), Velenje (Vel., n = 7), sampled in years 2004 and 2005. Code B91 Vel. B105 Vel. B106 Vel. B106 Vel. B113 Vel. B113 Vel. j.v. 3123 (1,8) Vel. Sen.1 Sen.2 Sen.3 Indon. Kan.6 Kan.9 Kan.19 Kan.15 Element mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg Ag 0.05). Fig. 8. PCA analysis of measured pa­rameters (As, Ba, Br, Ca, Ce, Co, Cr, Cs, Eu, Fe, Hf, K, La, Mo, Na, Nd, Rb, Sb, Sc, Se, Sm, Sr, Tb, Th, U, Yb, Zn) from different mining areas (Velenje-open, Kanižarica-closed, Senovo-closed, and Indonesia-open). 234 Tjaša KANDUC, Timotej VERBOVŠEK, Rok NOVAK & Radojko JACIMOVIC 235 Multielemental composition of some Slovenian coals determined with k0-INAA method and comparison with ICP-MS method 236 Tjaša KANDUC, Timotej VERBOVŠEK, Rok NOVAK & Radojko JACIMOVIC