Studying Natural Radioactivity of Coals and Ash and Slag Waste as Potential Raw Materials for Quality Assessment and Extraction of Rare Earth Elements

Abstract

A significant portion of coal mined in Kazakhstan is mainly used for fuel energy and metallurgy. Approximately 60% of electricity is generated by coal-fired power engineering. About 19 million tons of ash and slag waste (ASW) are annually sent to dumps. After coal combustion, in ASW not only are natural radioactive nuclides NRN (U²³⁸, Th²³², K⁴⁰) concentrated, but also rare and rare earth elements (REE). In this regard, ASW that essentially turns into quasi-technogenic deposits of NRN and REE, requires systemic measures for their utilization. The possibilities of extracting REE from coal power-industry waste are estimated based on the analysis of the concentration of REE (Ce, La, Nd, Sm, etc.), NRN (U²³⁸, Th²³² and their decay products, K⁴⁰) and the established significant correlations between rare earth and radioactive elements. The purpose of this paper is to study the natural radioactivity of coals and ash and slag waste as potential raw materials for assessing the quality and extracting rare earth metals. The stated purpose involves solving the following problems: studying the features of the NRN and REE distribution in coals and ash and slag waste; assessing the possibility of using ash and slag waste as a promising source of REE extraction based on nuclear radiometric studies; and studying the spectrometry of natural gamma radiation for assessing the quality of coals.

1. Introduction

Kazakhstan coals are widely used in the energy sector, but their value as a source of rare earth and radioactive elements has not yet been sufficiently studied. Major coal deposits in Kazakhstan include Ekibastuz coal basin, Karaganda basin, Karazhyra and Shubarkol coal deposits (Figure 1). Modern industrialization and the development of high-tech industries have led to increasing the demand for rare earth elements (REE) that are key components in the production of electronics, batteries, permanent magnets and renewable energy [1]. Due to the limited availability of traditional REE sources, increasing attention is being paid to alternative sources, among which coals and their combustion products—ash and slag waste—occupy a special place. The study of rare earth elements and natural radioactivity in coals and ash waste in Kazakhstan is a relatively poorly researched area. Some publications deal with the radioecological aspects of the natural radioactivity of coals and ash in the context of their impact on the environment [2,3].

Coals are complex geochemical systems capable of accumulating a wide range of impurity elements, including REE, U²³⁸, Th²³² and K⁴⁰ [4,5,6,7]. Coal ash has attracted the greatest interest in the last decade as a more concentrated and technologically accessible object for extracting REE. This is due to the fact that, during the combustion process, elements in the ash phase are redistributed and enriched, especially in fine fractions [8]. Ash and slag waste generated as a result of coal combustion at thermal power plants is not only an environmental problem but also a valuable source for extracting a number of valuable metals. In addition to rare earth elements, ash and slag waste can contain rare, noble and other metals, which makes it a potentially valuable raw material for processing.

One of the key aspects that complicates the use of ash waste in Kazakhstan is the absence of effective technologies for their disposal or recycling. Large accumulations of ash generated during coal combustion at thermal power plants remain in ash dumps, occupying huge areas and having a long-term impact on the environment. However, most ash dumps remain inactively used, and the issues of waste recycling and land restoration do not receive due attention.

Currently, there are many works that deal with studying trace elements in coal and ash waste, which consider industrial concentrations of various metals. This topic is especially relevant for countries with highly developed coal mining industries and large coal resources, such as China, Russia, the USA, Australia, Poland and others [1,9,10]. In Poland, ~2 million tons of ash waste are generated annually [11]. It contains various trace elements, including REE. The discovery of coal deposits with high REE content (up to 1%) has drawn attention to the possibility of extracting REE as a by-product from coal deposits [12]. In China, such critical elements such as lithium (Li), gallium (Ga) and germanium are actively extracted from coal deposits, making the country a major supplier of these valuable materials. In Russia, comprehensive studies are being conducted on the metal content of coal and ash and slag waste to extract rare and rare earth elements. The potential of ash waste in the USA and Australia is being actively studied to assess the possibility of extracting rare earth elements, which is important for meeting the growing global demand for these resources [13,14,15,16].

The prospects for extracting REE from ASW should be assessed by studying the content of these trace elements in ASW and the individual composition of REE. The average REE content in coal ash from the world deposits is 0.04% [17], which is approximately three times higher than that in the upper continental crust (0.017%). The role of REE will significantly increase in many areas of nanotechnology over the next decade. According to [18], coal ash contains an average of 445 mg/kg REE. There are data on the concentration of REE (0.1–1%) in coal deposits in the Russian Far East [4,12,15], Western Siberia of Russia [19], in coal deposits in Mongolia [20,21], China [22], and the USA [23,24]. A review of coal deposits as a potential source of REE is given in [4].

In Kazakhstan, approximately 60% of electricity is generated by coal power engineering. Annually, about 19 million tons of ASW end up in ash and slag dumps [25]. According to the most conservative estimates, they contain about 1300 tons of REE. In this regard, the possibility of extracting REE from ASW of coal power engineering is a promising area of research.

The authors of work [26] studied coal and ash from the Shubarkol deposit located in Central Kazakhstan. It was shown that the coals differed in significant variations in the contents of the bulk of the elements studied, including rare earth elements.

Studying the regularities of the chemical elements’ occurrence, including REE in coals, is of great importance for determining the conditions of coal formation, modeling the behavior of chemical elements during coal combustion at thermal power plants (TPPs), and developing comprehensive technology for processing coal and ash [27,28]. The methods of finding uranium in coal have been studied for more than half a century. At an early stage in the 1940s–1950s, coals containing uranium were of interest as a possible source of uranium for the nuclear industry [29,30]. Later, attention to the forms of uranium in coal was attracted by the potential radioecological hazard associated with the use of coal in the power industry. It was found that the forms of uranium occurrence, along with the combustion technology, determine the behavior of uranium in the processes of coal utilization at TPPs [16]. Despite the relatively long period of study of the radiogeochemical features of the distribution of NRE in coals, there are still no clear ideas about the forms of occurrence (uranium, thorium) in coals. This is caused by the extreme complexity of this problem, associated with many factors (types of coal deposits, age, coal grade, degree of coalification, coal formation conditions, ash content, etc.).

The results of studying NRN and REE concentrations in the coal power industry waste in Poland are presented in [31]. Ash and slag waste excluded from social and economic use can have a negative impact on the environment [32]; systemic measures are needed for waste storage and disposal, assessment of their quality and ecological and radiological hazard.

Specific radioactivity of coal power plant ASW depends on many factors: the coal ash content and its variation range, the radionuclide concentration and variation range, their forms, the combustion temperature conditions, etc. The multifactorial nature and variability of natural radioactivity of coals and combustion products have become the subject of studies to assess radioactive emissions from coal and its products [33].

Establishing a relationship between the REE content and NRN deserves special attention. Many REE-bearing minerals, such as monazite, xenotime, and bastnaesite, simultaneously include U and Th radionuclides in their structure [34]. Consequently, enrichment of coals and ASW with REE is often accompanied by increased radioactivity.

The possibilities of testing coals for gamma radiation emitted by natural radioactive elements have been poorly studied in the world practice [35,36]. This possibility was first described in [37]. It was found that the mineral (ash-forming) part of coal, in contrast to its organic matter, has increased natural radioactivity. Natural gamma-activity of coals is mainly caused by U²³⁸ and its decay products, Th²³² and its decay products, and K⁴⁰.

Depending on the type of coal deposit, radiogeochemical characteristics of coals and their component composition and the contribution of the specific activity of these radionuclides can vary. In Donbass coals (Ukraine), the content of radioactive elements varies within the range of (1–8.5) × 10⁻⁴% uranium; (8.7–49) × 10⁻⁴% thorium. In the Pechora Basin coals (Russia), the content of NRN varies within the range of (0.4–10) × 10⁻⁴% radium. The main contribution is made by K⁴⁰ that is common in clay compounds [35].

Thus, the objective of the article is to study natural radioactivity of coals and ash and slag waste as potential raw materials for assessing the quality and extracting rare earth metals. The stated objective involves solving the following research problems:

-

Studying radiogeochemical and petrophysical features of the NRN and REE distribution in coals and ash and slag waste;

-

Analytical assessing of the possibility of using ash and slag waste of coal power plants as a potentially promising source of REE extraction based on nuclear radiometric studies;

-

Studying the spectrometry of natural gamma radiation of natural radionuclides for assessing the quality of coals.

Figure 1. Geological and tectonic map of coal deposits in Central and Eastern Kazakhstan. Note—Compiled from source [38].

Figure 1. Geological and tectonic map of coal deposits in Central and Eastern Kazakhstan. Note—Compiled from source [38].

2. Materials and Methods

In this study, coals from the Ekibastuz and Karaganda coal basins were examined. The Ekibastuz coal basin is located in Central Kazakhstan and is one of the largest coal basins in the country. Coal mined in this basin is characterized by a high moisture content, a medium calorific value and a high ash content of 30–45%. This coal is mainly used for electricity generation. The Ekibastuz coal basin is one of the largest in Kazakhstan and is of considerable interest both for energy and for assessing the potential for extracting rare earth elements. The Karaganda coal basin is located in Central Kazakhstan. It produces hard coal that possesses high thermal properties and is used in energy and metallurgy. The ash content is on average 20–35%.

The authors tested and analyzed a total of 25 coal samples and 17 samples of ash and slag waste of the Ekibastuz basin taken from the State District Power Plant, Temirtau (taking into account the areas of technological impact), and 42 samples of Karaganda coal taken from the Saranskaya mine. The samples served as test materials for determining the ash content of coal, the concentration of NRE (U, Th, K⁴⁰) and rare earth elements (La, Ce, Sm, Eu, Nd, etc.) in coal and ash and slag waste. The overall workflow of the study is shown schematically in Figure 2.

Gamma-spectrometric analysis of coal samples and ash and slag waste weighing ~100 g was used to determine the concentrations of U²³⁸, Ra²²⁶, Th²³², K⁴⁰ at the Institute of Nuclear Physics, Almaty. A CANBERRA spectrometer with Genie-2000 software (Mirion Technologies, Atlanta, GA, USA) was used. The duration of the analyses (12–15 h) was selected in terms of achieving the minimum statistical measurement error. To calculate the specific activity of radionuclides, gamma lines were used: U²³⁸ was determined by the intensity of γ-radiation of its decay product Th²³⁴ (63 keV), Pb²¹⁴ (242, 295, 352 keV), Bi²¹⁴ (609 keV), Ra²²⁶ (186.2 keV); the activity of U²³⁵ by (185.7 keV), Th²²⁷ (236 keV); the activity of Th²³² was determined by Tl²⁰⁸ (583 keV), Pb²¹² (238 keV), Ac²²⁸ (911.2 keV); and the activity of K⁴⁰ by (1460.8 keV).

Partial spectrometric analysis of coal and ash samples (weighing ~1 kg) was performed using an MKC-01A MULTIRAD scintillation gamma spectrometer with the PROGRESS software implementing an advanced technique for selecting analytical gamma lines and optimizing energy characteristics, which allows for increased sensitivity of the analysis. The device was manufactured by the SPP Gamma LLC, Almaty, Kazakhstan. The instrument was calibrated using standard reference radiation sources of Cs-137 (with a gamma-quantum energy of 662 keV) and K-40 (1460 keV), ensuring the precise adjustment of the energy scale. The measurement energy range was from 60 to 2000 keV that allowed recording of the main photopeaks of natural radionuclides such as Ra-226, Th-232, and K-40. Background radiation was determined in the same measurement geometry as the analyzed samples, using an empty cuvette (containing no material). Background measurements were conducted for at least 2000 s to ensure sufficient statistical reliability. The resulting background spectrum was subtracted from the sample spectra using the built-in algorithm of the PROGRESS software. Spectrum processing and calculation of the specific activity of radionuclides were performed automatically according to the methodology approved for use in radiation monitoring.

The instrumental spectra of natural gamma radiation from coal and ash (Figure 3 and Figure 4) recorded by a gamma spectrometer with a scintillation detector clearly show the main gamma lines of U²³⁸: Th²³⁴ (92 keV), Ra²²⁶ (186.2 keV); Pb²¹⁴ (241, 295 keV, 352 keV); Bi²¹⁴ (609 keV); gamma lines Th²³²: Pb²¹² (238 keV), Tl²⁰⁸ (511 keV) and monoline K⁴⁰ (1460 keV).

At the same time, for quantitative analysis of rare earth, noble and the other elements, modern highly sensitive neutron activation analysis (HSNAA) with thermal neutron irradiation was used at the IRT-T research nuclear reactor of the Institute of Nuclear Physics in the laboratory of nuclear geochemical research methods of Tomsk Polytechnic University (TPU, Tomsk, Russia). The thermal neutron flux density in the irradiation channel was 2 × 1013 neutrons/(cm² × s). Sample irradiation duration ranged from 10 to 20 h. After irradiation, the samples were held for a period of time, and after the induced activity had subsided, they were sent to a gamma spectrometer to measure the emission intensity of radioactive isotopes. Measurements were performed on a Canberra multichannel pulse analyzer with a GX3518 semiconductor Ge detector. The content of the elements being determined is calculated by comparing the gamma line intensities of the corresponding radionuclides with those of standard samples or control samples. A total of 11 samples of Ekibastuz coal and 10 samples of ASW were analyzed. In the laboratory of nuclear physical methods of analysis (INP, Almaty), quantitative determination of REE and NRE in six samples of ASW of Ekibastuz coals was carried out using the neutron activation analysis. Reproducibility errors of gamma spectrometric and neutron activation analyses performed in two different laboratories were estimated (21–27%). The relative average discrepancy between the data of gamma spectrometric and neutron activation analysis for natural radionuclides was 34%.

The petrophysical characteristics of coal and ash waste samples were determined to identify relationships between the physicochemical properties of the material and the content of radionuclides and rare earth elements. The ash content was determined using a standard laboratory method in accordance with GOST 11022–95 [39]. The total content of light rare earth elements (∑LREE) was calculated as the sum of the concentrations of La, Ce, Nd, and Sm. Additionally, concentration coefficients and correlation coefficients between the contents of ∑LREE, Th, and U were calculated, along with the statistical significance level (p). These calculations allowed for an analysis of the possible co-migration and mineralogical association of elements, as well as the identification of radiogenic sources of REE in coal matter.

Based on the generalization of estimated data [40], the relationship between the gamma activity (Nγ) of coal in percent equivalent uranium from its ash content (Ad) is expressed as follows:

Ad = 2.156 + 0.403 × 10⁵ Nγ

(1)

It is quite obvious that this dependence is of an ambiguous nature, caused by the inconsistency of the NRN content in coals of different component composition and quality. The error in determining the ash content of coal by the value of natural gamma activity varies within 5–7% abs. The comparatively high error in assessing the quality of coal is due to the fact that natural gamma activity of coals is in a complex relationship with their quality. The reasons are the heterogeneity of the component composition, the variety of forms of NRN in organic and mineral components, different degrees of coalification, the polyenergetic composition of natural gamma radiation emitted by radionuclides, etc. Based on the measurement of natural gamma radiation of coals of the Karaganda and Ekibastuz basins, the ambiguity of the integral intensity of gamma radiation depending on the ash content was established. Relative sensitivity to the ash content that means the relative percentage increase in the measured gamma radiation intensity with a single (1% abs.) change in ash content varies within 1.57–2.62%/% abs. These practical results that are in satisfactory agreement with earlier studies [35] indicate that for coals with near-clarke NRN contents, the statistical error and variability of a specific activity of radionuclides play a significant role in the metrology of natural gamma radiation. These are the most significant factors reducing the sensitivity and accuracy of assessing the quality of coals of variable composition.

To assess the redistribution of elements during the ash formation process, the coefficient of the relative accumulation of the element in ash compared to coal was also calculated using the formula

$$\mathrm{K}={\displaystyle \frac{C_{ash}}{C_{coal}}},$$

(2)

where C_ash is the element content in the ash and C_coal is the element content in the coal.

To assess the reliability of the obtained analytical data and the degree of variation in the concentrations of elements, the mean values (Mean), standard deviation (SD) and coefficient of variation (CV, %) were calculated.

$$\mathrm{SD}=\sqrt{{\displaystyle \frac{{\sum }_{i=1}^n(x_i-\overline{\mathit{x}}{)}^2}{n-1}}},$$

(3)

where x_i is an individual value of the element concentration, $\overline{x}$ is the mean value, and n is the number of samples.

The variance coefficient is determined as follows:

$$\mathrm{CV} = {\displaystyle \frac{\mathrm{SD}}{\overline{x}}}\times 100\%$$

(4)

The calculations were performed in Python 3.11 (Anaconda, Jupyter Notebook) with NumPy and Pandas libraries and verified in MS Excel.

Statistical analysis was performed separately for coal and ash waste. The principal component analysis (PCA) was used to analyze the joint variability of rare earth elements (La, Ce, Nd, Sm) and natural radionuclides (U, Th). Before calculations, the data were normalized using z-standardization, after which the eigenvalues and vectors of the covariance matrix were calculated. PCA was performed in Python using the scikit-learn library.

3. Results

The results of the instrumental analysis of coal and ash and slag samples for the content of REE and radioactive nuclides are given, and their concentration coefficients during combustion were estimated.

Table 1. Coal and ASW analysis results.

Element	Coal	ASW	Concentration Factor
	${\displaystyle \frac{\mathbf{C}\mathbf{o}\mathbf{n}\mathbf{c}\mathbf{e}\mathbf{n}\mathbf{t}\mathbf{r}\mathbf{a}\mathbf{t}\mathbf{i}\mathbf{o}\mathbf{n} \mathbf{R}\mathbf{a}\mathbf{n}\mathbf{g}\mathbf{e}}{\mathbf{A}\mathbf{v}\mathbf{e}\mathbf{r}\mathbf{a}\mathbf{g}\mathbf{e} \mathbf{C}\mathbf{o}\mathbf{n}\mathbf{c}\mathbf{e}\mathbf{n}\mathbf{t}\mathbf{r}\mathbf{a}\mathbf{t}\mathbf{i}\mathbf{o}\mathbf{n},(\mathbf{m}\mathbf{g}/\mathbf{k}\mathbf{g})}}$	${\displaystyle \frac{\mathbf{C}\mathbf{o}\mathbf{n}\mathbf{c}\mathbf{e}\mathbf{n}\mathbf{t}\mathbf{r}\mathbf{a}\mathbf{t}\mathbf{i}\mathbf{o}\mathbf{n} \mathbf{R}\mathbf{a}\mathbf{n}\mathbf{g}\mathbf{e}}{\mathbf{A}\mathbf{v}\mathbf{e}\mathbf{r}\mathbf{a}\mathbf{g}\mathbf{e} \mathbf{C}\mathbf{o}\mathbf{n}\mathbf{c}\mathbf{e}\mathbf{n}\mathbf{t}\mathbf{r}\mathbf{a}\mathbf{t}\mathbf{i}\mathbf{o}\mathbf{n},(\mathbf{m}\mathbf{g}/\mathbf{k}\mathbf{g})}}$
Sm	${\displaystyle \frac{2.5-3.4}{2.8}}$	${\displaystyle \frac{2.9-9.0}{6.2}}$	${\displaystyle \frac{1.16-2.6}{2.2}}$
Ce	${\displaystyle \frac{20.5-28.9}{24.4}}$	${\displaystyle \frac{52.0-57.9}{54.8}}$	${\displaystyle \frac{2.5-2.0}{2.2}}$
Lu	${\displaystyle \frac{0.2-0.3}{0.3}}$	${\displaystyle \frac{0.7-0.9}{0.8}}$	${\displaystyle \frac{3.5-3.0}{2.7}}$
U	${\displaystyle \frac{0.5-1.5}{1.0}}$	${\displaystyle \frac{1.05-3.62}{2.7}}$	${\displaystyle \frac{2.1-2.4}{2.7}}$
Th	${\displaystyle \frac{2.1-3.1}{2.5}}$	${\displaystyle \frac{6.3-7.1}{6.5}}$	${\displaystyle \frac{3.0-2.3}{2.6}}$
Yb	${\displaystyle \frac{1.5-2.1}{1.8}}$	${\displaystyle \frac{4.8-5.6}{5.1}}$	${\displaystyle \frac{3.2-2.7}{2.8}}$
Nd	${\displaystyle \frac{2.9-15.8}{11.4}}$	${\displaystyle \frac{13.8-46.9}{28.9}}$	${\displaystyle \frac{4.7-3.0}{2.5}}$
Tb	${\displaystyle \frac{0.3-0.6}{0.5}}$	${\displaystyle \frac{0.7-1.7}{1.2}}$	${\displaystyle \frac{2.3-2.8}{2.4}}$
Eu	${\displaystyle \frac{0.5-0.8}{0.7}}$	${\displaystyle \frac{1.7-2.1}{1.9}}$	${\displaystyle \frac{3.4-2.6}{2.7}}$
La	${\displaystyle \frac{9.0-13.1}{11.0}}$	${\displaystyle \frac{25.9-28.5}{27.4}}$	${\displaystyle \frac{2.9-2.2}{2.5}}$
K40	${\displaystyle \frac{9.4-87.6}{59.9}}$	${\displaystyle \frac{59.5-115.8}{80.0}}$	${\displaystyle \frac{6.3-1.3}{1.3}}$

presents the range of concentrations and the average content of elements in coal and ash and slag waste according to neutron activation analysis data (11 coal samples and 9 ash and slag waste samples).

The results of neutron activation analysis (Table 1) show that the REE content in the studied samples varies widely. In the studied samples, the highest content was recorded for light lanthanides Sm, Ce, La, Nd, which correspondy to the global trend for the prevalence of light rare earth elements (LREE) [4,10,12,15].

A summary assessment of the average content of rare earth elements (Ce, La, Nd, Sm, Eu, Tb, Yb, Lu) and natural radionuclide radioactive elements (U, Th) in coal and ash and slag waste samples in comparison with their clarke values given in [41] allows estimating the degree of geochemical enrichment of the original fuel, as well as identifying the possible technogenic concentration of these elements in combustion products. It was found that the content of most REE and radionuclides in coal exceeded or was comparable to clarke levels; this is especially pronounced for Ce, Yb, Sm and Lu. In ash and slag waste, the concentrations of all the studied elements increase. For comparison, the authors used the geochemical classification of REEs: light (La, Ce, Nd, Sm), heavy (Tb, Yb, Lu), and Eu. These observations highlight the need for further study of the behavior of these elements during incineration, as well as their forms in waste, from an environmental and resource-assessment perspective. Concentration factors were calculated using Formula (2).

An assessment was made of the relationship between the total content of light rare earth elements (∑LREE) and the concentration of radionuclides (U and Th), as well as their sum in Ekibastuz coal samples. In coal samples, a moderate positive correlation is observed between ∑LREE and the uranium content (r = 0.68 p = 0.022), as well as between ∑LREE and thorium (r = 0.63, p = 0.038). To confirm the statistical significance of the correlation coefficients between the total content of light rare earth elements and the content of natural radionuclides, p-value (p) calculations were performed. The correlation is higher between ∑LREE and U + Th (r = 0.75, p = 0.008) (Figure 5). In addition, descriptive statistics were performed: standard deviations for elements are within the range of ±0.03–±3.61 mg/kg, variation coefficients (CV) = 8.87–31.74% (low–moderate variability for La, Ce, Sm, Th and increased variability for Nd and especially U). According to the results of the principal component analysis (PCA), the first component PC1 explains about 60–65% of the total variance, the second PC2—about 20–25%. High positive loadings are observed on PC1 for La, Ce, Nd, Sm, and Th (and, on average, U), indicating their consistent variation. The contribution of U is also partially manifested in PC2, which is consistent with its higher intersample variability. Taken together, the results of the correlation analysis, SD/CV and PCA confirm the joint trend of ∑LREE and Th + U variation in coals while maintaining the increased heterogeneity of uranium distribution.

Table 2. ASW analysis results.

Sample	La	Ce	Nd	Eu	Sm	Tb	Yb	Lu	Th	U
3–8	27.6	66.3	31.4	1.82	7.03	1.27	4.7	0.69	7.13	2.72
3–10	28.8	67	34.8	1.95	7.58	1.28	5.15	0.74	7.49	2.85
3–12	27.9	66.2	31.4	1.88	7.4	1.3	4.93	0.72	7.39	3.04
3–14	29.8	71.1	33.9	1.97	7.87	1.44	5.11	0.76	8.06	3.08
3–2	29	67.3	31.7	1.85	7.62	1.26	4.93	0.73	7.51	2.97
3–6	28.2	66.1	34.9	1.85	7.27	1.22	4.73	0.69	7.32	2.8

presents the results of analyzing Ekibastuz coal ash and waste for the quantitative content of REE and uranium and thorium radionuclides using the neutron activation analysis method in the laboratory of the Institute of Nuclear Physics in Almaty. The SD values ranged from ±0.03 to ±1.74 mg/kg, and the CV ranged from 2.6 to 6% that indicated low variability in the chemical composition of the samples.

The correlation coefficient between (∑LREE) and Th is r = 0.89, p = 0.01, which is a very high value. The correlation coefficient between (∑LREE) and U is (r = 0.45), ∑LREE and U + Th (r = 0.80, p = 0.05), (Figure 6). Light rare earth elements (LREE) in ash and slag waste samples demonstrate a positive correlation with thorium and the sum of U + Th. Moreover, Th alone shows a higher correlation with ∑LREE (r = 0.89), confirming local enrichment in rare earth elements in areas with elevated thorium content. According to PCA results, the first principal component (PC1) explains approximately 62% of the total variance, and the second (PC2) explains approximately 24%. High loadings of La, Ce, Nd, Sm, and Th on PC1 indicate a possible common source of their accumulation. The element U is characterized by a lower loading on PC1, reflecting a partial divergence in the behavior of radionuclides and rare earth elements during ash formation. Low CV values (≈2.6–6%) for the major elements are consistent with the PCA conclusions and indicate a relative compositional homogeneity of the studied ash samples while maintaining the general geochemical interrelationship of rare earth elements and Th.

Similar results were recorded in a number of international studies, in particular, in [4,34,42], it was noted that a high positive correlation between ΣREE and U, Th can indicate a joint origin of these elements associated with phosphate, organic or aluminosilicate minerals. Minerals—carriers of REE and radionuclides confirming joint enrichment were determined [31]. The data obtained are in good agreement with these conclusions, indicating the possibility of using ERN as indicators of REE in coals and ash and slag waste.

The correlation analysis showed the presence of stable relationships between the REE content and the concentration of NRN in the ash and slag waste. This indicates a possible commonality of sources and sorption mechanisms that ensure the joint accumulation of these elements. This nature of the relationship can indicate the geochemical conjugacy of these elements, which allows considering REE as potential indicators of the presence of natural radionuclides in coal-bearing strata. This is of practical importance for geological exploration aimed at identifying areas of increased radioactivity, as well as for radioecological monitoring of coal deposits and thermal power plants, where both REE and radionuclides can accumulate in ash.

The results of analyzing coal samples from the Karaganda basin taken from the Saranskaya mine are given in

Table 3. Karaganda coal analysis results.

Element	${\displaystyle \frac{\mathbf{C}\mathbf{o}\mathbf{n}\mathbf{c}\mathbf{e}\mathbf{n}\mathbf{t}\mathbf{r}\mathbf{a}\mathbf{t}\mathbf{i}\mathbf{o}\mathbf{n} \mathbf{C}\mathbf{h}\mathbf{a}\mathbf{n}\mathbf{g}\mathbf{i}\mathbf{n}\mathbf{g} \mathbf{R}\mathbf{a}\mathbf{n}\mathbf{g}\mathbf{e}}{\mathbf{A}\mathbf{v}\mathbf{e}\mathbf{r}\mathbf{a}\mathbf{g}\mathbf{e} \mathbf{C}\mathbf{o}\mathbf{n}\mathbf{c}\mathbf{e}\mathbf{n}\mathbf{t}\mathbf{r}\mathbf{a}\mathbf{t}\mathbf{i}\mathbf{o}\mathbf{n} (\mathbf{m}\mathbf{g}/\mathbf{k}\mathbf{g})}}$	Clarke in Coal (mg/kg)	Concentration Factor
La	${\displaystyle \frac{10.7-18.8}{13.98}}$	11.00	1.27
Ce	${\displaystyle \frac{23.4-45.8}{31.95}}$	23.00	1.38
Nd	${\displaystyle \frac{11.5-26.1}{17.61}}$	12.00	1.47
Eu	${\displaystyle \frac{0.65-1.21}{0.86}}$	0.43	2.0
Sm	${\displaystyle \frac{2.87-6.08}{3.84}}$	2.20	1.75
Tb	${\displaystyle \frac{0.42-0.86}{0.59}}$	0.31	1.9
Yb	${\displaystyle \frac{1.83-3.0}{2.31}}$	1.00	2.31
Lu	${\displaystyle \frac{0.23-0.41}{0.31}}$	0.20	1.55
Th	${\displaystyle \frac{2.62-5.2}{3.84}}$	3.20	1.2
U	${\displaystyle \frac{1.11-6.85}{2.03}}$	1.90	1.1

. The data are compared with clarke values [41]. Standard deviations are in the range of ±0.03–±4.47 μg/g, and the variation coefficients CV = 9.39–42.56%. La, Ce, Sm, Th, and HREE (Lu, Yb) are characterized by low to moderate variability (CV ≈ 9–16%), while U (CV ≈ 42.6%) and partially Nd (≈15–16%) exhibit increased scatter, reflecting local anomalies or differences in the phase distribution of uranium.

Comparison of the results of studying rare earth and radioactive elements of Ekibastuz and Karaganda coals shows that the content in Karaganda coals is higher than that in Ekibastuz.

Karaganda coals also show a predominance of light rare earth elements (Ce, La, Nd, Sm), which corresponds to the geochemical characteristics of coals and the global trend in LREE accumulation. The average contents of Ce (31.95 mg/g) and La (13.98 mg/g) are at the level characteristic of Chinese lignites and some US coals [42,43]. A significant positive correlation was established between ∑LREE and the Th content (r = 0.63, p = 0.001), which can indicate joint migration or coaccumulation of these elements in the organo-mineral phase of coal (Figure 7). The principal component analysis (PCA) confirmed the general distribution structure: PC1 = 69%, PC2 = 11% (together accounting for ~81% of the variance). High positive correlations are observed for La, Ce, Nd, Sm, and Th in PC1, reflecting their combined variability and a common geochemical factor. U partially projects onto PC2, consistent with its greater intersample variability. Taken together, correlation analysis and PCA confirm a stable relationship between light REE and Th in Karaganda coals.

Figure 8 shows the average content of radionuclides (Ra-226, Th-232, K-40) determined by the gamma spectrometric method for 25 coal samples and 17 samples of ash and slag waste. It has been shown that radionuclides are concentrated in ash and slag waste. This confirms the known effect of radionuclide accumulation in ash during combustion, which was previously noted in studies [2,4,11,44].

The specific activity of radium in coal is 6.64 Bq/kg, while in ash it is 12.38 Bq/kg; the specific activity of thorium-232 increases from 10.32 to 15.24 Bq/kg, and potassium-40 increases from 138.80 Bq/kg to 222.10 Bq/kg.

The comparatively low values of the radionuclide concentration factors in the studied coal and ash and slag waste samples can be caused by specific features of the material selection. Ash samples were taken from ash and slag dumps formed as a result of long-term operation, where the ash was stored for a long time. Over the years of storage, the composition of the ash could have been affected by various physicochemical processes, such as weathering, leaching, migration of elements and contact with atmospheric precipitation. These processes probably contributed to the redistribution and partial removal of both rare earth and radioactive elements, which reduced their concentrations in the tested samples.

The ash content can be an indirect indicator of the potential radioactivity of coals. These results are consistent with the data of earlier studies noting the accumulation of radionuclides in the ash component of coal [45,46,47].

The authors have proposed innovative approaches to instrumental testing based on measuring the spectrometry of natural gamma radiation of coals, which allow increasing the sensitivity of coal ash content determination under conditions of their variability. The energy distribution of natural gamma radiation is a complex polyenergetic spectrum, including a set of many gamma lines of U²³⁸, Th²³² and their decay products, and K⁴⁰ in the range of 63–2620 keV. The photopeaks of gamma lines are located against the background of a continuous Compton distribution of higher-energy gamma radiation (Figure 1 and Figure 2). Hence, the integral intensity of natural gamma radiation is a complex function depending on the ash content of coal, its component composition, specific activity of radionuclides and their decay products, as well as the spectrometer hardware parameters—in particular, the energy discrimination threshold [48]. The optimal energy discrimination threshold (Ei) is selected from the point of view of the minimum relative statistical measurement error (ƒi) and maximum relative sensitivity to ash content (S). It is easy to imagine that the ratio value (ƒi/S) characterizes the minimum statistical error expressed in fractions of coal ash content. The approach to choosing the optimal energy threshold that we propose allows us to minimize statistical errors, reduce the contribution of background gamma radiation, and increase sensitivity to ash content by 23% rel., which is especially important for the radiometric testing of coals with a low radionuclide content.

Exploratory studies have been started on instrumental testing of coals in large masses based on NRN gamma-radiation spectrometry with selective consideration of the contribution of each radionuclide, optimization of instrumental and energy parameters and development of an individual interpretational and algorithmic assessment of coal quality taking into account the measured instrumental signals from analytical gamma lines as benchmarks, taking into account the diversity of radiogeochemical and petrophysical characteristics of the sampled coal in order to increase the information content, sensitivity and accuracy of coal testing in large masses.

The main instrumental and methodological signals from analytical gamma lines of uranium, thorium and potassium-40 as benchmarks taking into account the features of the sampled coal have been identified. One such benchmark is the value of the Th/U ratio significantly related to the coal ash content (Figure 9). Additional research is needed to improve the accuracy and sensitivity of radiometric testing using the spectrometry of natural gamma radiation of coals under conditions of variability in their quality and component composition.

4. Conclusions

• The instrumental nuclear radiometric analysis of test samples of Ekibastuz and Karaganda coals have shown that the contents of REE (Ce, La, Nd, Sm, etc.) and NRN (uranium, thorium and their decay products) are comparable to the clarke values. In coal power plant ash and slag waste, the concentrations of these elements increase significantly.
• Correlations were found between the concentrations of individual REE and the concentrations of uranium and thorium. The most stable (significant) relationships were found between the sum of REE (Ce, La, Nd, Sm) and the thorium content, as well as the total content of thorium and uranium. This indicates a probable geochemical relationship between REE and NRN.
• Comparatively high concentrations of REE in coal power plant ash and slag waste and the close correlations found between the concentrations of REE and NRN allow for the prompt assessment of the REE content using gamma-radiometric measurements.
• In the context of the global economy, growing demand for REE and the search for alternative sources of their production, ash and slag waste seem to be the most accessible raw material for their extraction.
• The studies have established that the integral intensity of natural gamma radiation of coals is a complex function depending on the ash content of coal, the specific activity of radionuclides, the energy composition of natural gamma radiation and the spectrometer hardware parameters. It is proposed to select the optimal energy discrimination threshold from the point of view of minimal statistical error and maximum sensitivity to the ash content. Spectrometric signals from analytical gamma lines of NRN as benchmarks taking into account variability of the sampled coals have been found, which will increase the sensitivity and accuracy of the coal ash content assessment.