Carbonaceous Shale Deposits as Potential Unconventional Sources for Rare Earth Elements at the Witbank Coalfield, Permian Vryheid Formation, South Africa

Abstract

Carbonaceous shale has garnered significant interest as a viable alternative source of rare earth elements (REEs) besides conventional REE-bearing ores. This study characterized rare earth element + Yttrium+ Scandium (REYs) enrichment in the 11 core samples of carbonaceous shale (7) and coal (4) collected from Arnot Mine. Major elements of the studied carbonaceous shale (CS) and coal showed high amounts of SiO₂, Al₂O₃, and Fe₂O₃, indicating a high content of aluminosilicate and iron-rich minerals. The plots Na₂O + K₂O against SiO₂ suggested alkali granite, granite, and granodiorite provenance sources for the studied shale and coal. The samples showed enrichment in low and heavy rare elements crystallized from a low potassium tholeiitic and medium calc-alkaline magma based on the plots of La_N/Yb_N and K₂O vs. SiO₂. The mineralogical and maceral analysis revealed the dominant presence of kaolinite (15%–45%), and it was suggested as the cation exchange site resulting from the isomorphous substitution of Al³⁺ for Si⁴⁺. Additionally, siderite was suggested as one of the REY hosts due to the Fe³⁺ site forming a complex with the REE³⁺ ions. Furthermore, the samples were classified as lignite to sub-bituminous coal category with dominant minerals including kaolinite, quartz, and siderite. The outlook coefficient (Coutl) of REY in CS revealed a promising area for economically viable, having two enrichment types, including low (La, Ce, Pr, Nd, and Sm) and heavy (Ho, Er, Tm, Yb, and Lu). The Eu_N/Eu_N* and Ce_N/Ce_N* ratio for the current studied samples exhibited a weak negative to no anomaly, and most of the studied samples were characterized by distinctive positive Gd anomalies derived from sediment source regions weathered from alkali granite, granite, and granodiorite provenance formed from a low potassium tholeiitic and medium calc-alkaline magma.

1. Introduction

Rare earth elements are transition metals significant for clean energy, high-technology applications, and national security. Rare earth elements (REEs) are a set of 15 lanthanide group elements plus scandium (Sc) and yttrium (Y) that share similar chemical characteristics, denoted as REY. REY are rarely found in concentrated economic ore deposits [1], are difficult to extract [2], and are often found in low concentrations [3] but are relatively more abundant in the upper continental crust than most minerals, such as galena or hematite deposits [4]. The need for REY in the modern economy has been tremendous, addressing climatic change and global warming challenges and providing materials for green energy utilization, including hybrid cars, electric vehicle batteries, and wind turbines. In recent years, there has been a growing demand for energy-efficient gadgets that are lighter, faster, smaller, and more efficient, driven by the unique chemical, electronic, conductive, magnetic, and luminescent properties of REEs [5].

Geologically, carbonatite formations that contain bastnäsite minerals are the most common, though they are not the only types of REY minerals. The minerals commonly occur in deep-seated plutonic and magmatic-hydrothermal settings formed by various geological processes. Bastnäsite, the carbonate fluoride mineral, tends to be enriched in light REY composed of cerium, lanthanum, neodymium, and europium in ores that are high grade from >4 wt.% Rare Earth Oxide (REO) [6,7]. A prolific REY occurrence associated with carbonatite has been reported in the Bayan Obo deposit in China [8]. It has high-grade ores of >6 wt.% REO, and lower grade resources of <4.1 wt.% REO [8]. Still, mining of high grade of REE > 12% REO in Mountain Pass is reported to have reopened following the increasing demand for modern energy and technology devices [1]. Other world’s significant REE-bearing deposits are typically associated with peralkaline rocks [9,10], hydrothermal vein [11], iron-oxide apatite [12], ion-adsorption clay [13], sediments-hosted hydrothermal [14], and placer deposits [15]. In South Africa, the ores of the Steenkampskraal monazite have been reported [16] to host high-grade (45 wt.% REO) REE mineralization. However, the Steenkampskraal deposit is composed of monazite–apatite bearing high radioactive thorium concentration [17].

Coal-based resources have been identified as secondary sources of rare earth elements [18,19]. The aforementioned studies showed that REE enrichment in coal ash can range from parts per billion (ppb) to thousands of ppm. It further demonstrated the possibility of recovering critical minerals and 2 wt.% purity REEs from low-grade source materials (~300 ppm). Other recent studies on REE associated with coal and coal byproducts [20,21,22] from Waterberg and Witbank Coalfields in South Africa have suggested enrichment of LREE.

With an increasing demand for REY, carbonaceous shale intercalated with coal is supposedly considered a significant unconventional source because REEs have high ionic charges, forming stronger complexes with hard ligands of carbonates [19]. Considering the composite strata of carbonaceous shales interbedded with coal seams in the Witbank Coalfield [18,19,20,21,22], the successful characterization of the deposits could account for significant technically recoverable REY in South Africa. As such, this study seeks to characterize REY occurrence in carbonaceous shales interbedded with coal to understand its potential economic viability.

2. Geological Setting

The study area, Arnot Colliery, is situated within 29°51′19.38″ N and 26°01′50.44″ at the eastern end of the Witbank Coalfield (Figure 1). The sedimentary sequence underlying the study area comprises sandstone, siltstone, carbonaceous shale, and coal [23]. It belongs to the Vryheid Formation of the Main Karoo Basin (MKB), and paleo paleo-depositional environment of the carbonaceous shale is fluvial-deltaic [24]. The proximal fluvial-deltaic carbonaceous shale of the formation emanated from northern MKB and thinned southwards into the Pietermaritzburg and Volksrust Formations [24]. The stratigraphy of the coalfield is delineated into five coal seams from seam 5 to seam 1 [24].

3. Materials and Methods

A total of 11 core samples representing the lithologies of carbonaceous shale and coal were retrieved from boreholes drilled at the Permian Vryheid Formation of the Ecca Group at the Arnot Mine in the Witbank Coalfield, Mpumalanga, South Africa, as shown in Figure 2. The carbonaceous shale varies from 1–3 m in thickness, and the coal thickness between 2–20 m.

Samples were pulverized and prepared for X-ray diffractometry (XRD) using a backloading method. The analysis was conducted using a PANalytical X’Pert Pro-powder diffractometer equipped with an X’Celerator detector coupled with receiving slits, variable divergence, and Fe-filtered Cu-Kα radiation from the Energy Centre CSIR Pretoria. The procedure for the experiment has been described in Hutton and Mandile [23].

Additionally, the macerals and mineralogical contents of the coal and carbonaceous shale were examined using petrographic microscopy. The polished blocks were assessed petrographically using a Zeiss AxioImager M2M petrographic microscope at a magnification of x500 under oil immersion at the Department of Geology at the University of Johannesburg. The samples were crushed to a minimum of 1 mm with minimal particles for the petrographic analysis. The blocks were then ground using a Struers Tegraforce polisher to provide a polished surface in accordance with the South African National Standard (SANS) that is based on the International Organization for Standardization (ISO) standard SANS ISO 7407-2 (2015) ISO 7404 after the particles had been cured in epoxy resin under vacuum for 24 h. The analysis was carried out following the procedure of Wagner et al. [24].

The proximate analysis of the carbonaceous shale and coal was carried out to ascertain the analyzed moisture, volatile matter, fixed carbon, and ash yield at the Bureau Veritas Testing and Inspecting in South Africa. A detailed description of the procedures is provided by Donahue and Rais [25].

The whole rock analysis was carried out using X-ray Fluorescence (XRF) spectrometry. The samples were analyzed using PANalytical XRF at Stellenbosch University in Western Cape, South Africa) [26]. The procedure of the whole rock geochemistry analysis has been described in detail by Akintola et al. [26]

Additionally, the sequential chemical extraction procedures (SCEP) and Laser Ablation inductively Coupled Plasma Mass Spectroscopy (LA-ICP-MS) of the samples were conducted at Stellenbosch University in Western Cape, South Africa). The SCEP procedure followed the procedures described by Dai et al. [19] Dai et al. (2004), as shown in

Table 1. Steps of sequential chemical extraction procedure.

	Reaction Phases	Experimental Method
1	Water soluble	8 g Carbonaceous shale and Coal samples + 60 mL water, 25 °C, 24 h
2	Ion exchangeable	Residue of Phase 1 + 60 mL NH4Ac, 25 °C, 24 h
3	Organic bonded	Residue of Phase 2 + 1.47 g/cm3 the floating dried at 40 °C, ashed at 650 °C, +3 mL HClO4, 200 °C 60 h
4	Carbonate	The sinking of Residue of Phase 2, washed by alcohol and dried at 40° C, +20 mL 0.5% HCl
5	Silicate	Residue of Phase 3 + 2.89 g/cm3, CHBr3, the floating dried at 40° C, ashed at 650° C, +3 mL HNO3 and 3 mL HF, 200 °C, 60 h
6	Sulfide	The sinking of Residue of Phase 3, washed by water and dried at 40 °C, +HNO3, 5 h

. The REE concentration was determined from steps using the ICP-MS [27]. The details of the operating conditions for both the laser and the ICP-MS are described in Thomas et al. [28].

4. Results

The studied carbonaceous shale (CS) and coal showed the presence of major elements, including SiO₂, TiO_2, Al₂O₃, CaO, Cr₂O₃, Fe₂O₃, K₂O, MgO, MnO, Na₂O, and P₂O₅,

Table 2. Major elements of studied samples.

Samples Name	Samples Type	Al2O3 (%)	CaO (%)	Cr2O3 (%)	Fe2O3 (%)	K2O (%)	MgO (%)	MnO (%)	Na2O (%)	P2O5 (%)	SiO2 (%)	TiO2 (%)	Al2O3/TiO2 (%)	L.O.I.	CIA
BS-1	CS	22.05	0.27	0.01	3.54	2.18	0.92	0.05	0.44	0.06	54.81	0.92	23.97	14.70	88.41
BS-2		22.06	0.35	0.01	6.14	2.50	1.10	0.11	0.67	0.08	50.96	0.80	27.58	15.29	86.24
BS-3		17.72	0.95	0.01	8.45	2.79	1.60	0.10	0.46	0.11	53.04	0.83	21.35	14.00	80.84
BS2-1		21.54	0.37	0.01	4.79	2.19	1.19	0.15	0.10	0.10	47.57	0.86	25.05	20.82	89.01
BS2-4		13.91	1.33	0.07	14.37	1.19	4.69	0.19	0.28	0.06	44.18	0.33	42.15	19.02	83.24
BS2-2		10.22	0.93	0.02	1.67	0.99	0.57	0.02	0.07	0.01	44.12	0.81	12.62	39.87	83.70
BS2-3		18.99	0.68	0.01	6.37	1.43	0.95	0.18	0.08	0.06	42.04	0.87	21.83	28.29	89.66
BSC-1	Coal	20.01	0.59	0.02	3.86	1.89	0.71	0.08	0.30	0.12	48.98	0.79	25.33	22.21	87.80
BSC-2		5.18	1.00	0.01	3.18	0.22	0.31	0.01	0.06	0.01	22.30	0.29	17.86	66.59	80.19
BSC-3		7.55	1.38	bdl	0.79	0.21	0.48	0.02	0.10	0.02	20.15	0.46	16.41	67.85	81.71
BSC2-1		17.00	0.20	0.01	1.74	1.37	0.57	0.02	0.08	0.04	42.76	0.69	24.64	35.50	91.15
Ave		11.79	0.84	0.01	2.75	0.84	0.58	0.05	0.08	0.03	34.27	0.62	19.02	47.62	85.68

CIA = [Al₂O₃/(Al₂O₃ + Na₂O + K₂O + CaO⁎)] × 100.

). The amounts of SiO₂ (42.04%–54.8%), Al₂O₃ (10.22%–22.06%), and Fe₂O₃ (1.67%–8.45%) were relatively high compared to other elements in CS. A similar trend was found in the coal sample showing a relatively high amount of SiO₂ (20.15%–48.98%), Al₂O₃ (5.18%–20.01%), and Fe₂O₃ (0.79%–3.86%). The Chemical Index of Alteration (CIA) values [29], calculated from the chemical analysis of carbonaceous shale and coal using the molecular proportions of Al₂O₃, Na₂O, K₂O, and CaO (CIA = [Al₂O₃/(Al₂O₃ + Na₂O + K₂O + CaO⁎)] × 100, where CaO⁎ is in silicates), increased as the alteration process proceeds.

The XRD semi-quantitatively revealed the presence of quartz (18%–36%), kaolinite (15%–45%), pyrite (0%–0.6%), muscovite (4%–9%), anatase (0.3%–0.7%), rutile (0%–0.8%), microcline (5%–10%), siderite (1.2%–23%), plagioclase (0.1%–3%) and organic carbon (11%–20%) in the studied carbonaceous shale (

Table 3. Mineralogical characteristics of the samples studied.

Samples	S/Type	Qtz	Kaol	Pyr	Mus	Anat	Rut	Micro	Sid	Plag	Organic C
BS-1	CS	26.1	44.2	0	7.7	0.6	0.4	7.4	1.2	0.7	11.7
BS-2		18.5	45.1	0	9	0.7	0.6	7.2	2.8	3	13.2
BS2-1		21.8	42.3	0	8.5	0.6	0.5	5	3.4	0.1	17.7
BS2-4		30.6	15.7	0.6	6.5	0.3	0	5.3	23.5	0.6	17
BS-3		36.3	26.8	0.4	4.9	0.4	0.1	10.2	6.9	0.8	13.2
BS2-2		26.3	27.1	0	3.5	0.1	0.7	4.7	0	0	37.6
BS2-3		16	40.1	0	5.8	0.4	0.8	5.9	4.9	0.1	26
BSC-1	Coal	22.2	41.6	0	5.2	0.4	0.8	7.5	1.3	0.8	20.3
BSC-2		16.2	13.3	1.4	1.3	0	0.1	3.3	0	0	64.3
BSC2-1		17.8	36.8	0	6.1	0.2	0.4	5.5	0.1	0.1	33
BSC-3		12	19.6	0	1.3	0.1	0.1	1.5	0	0	65.4

Qtz—Quartz, Kaol—Kaolinite, Pyr—Pyrite, Mus—Muscovite, Anat—Anatase, Rut—Rutile, Micro—Microcline, Sid—Siderite, Plag Plagioclase, Organic carbon.

and Figure 3). At the same time, the studied coal samples showed similar composition and amount except for a higher content of organic carbon (26%–65%) and a lower content of quartz (12%–26%) and kaolinite (13%–40%). The kaolinite minerals predominated the CS samples with an average value of 35.95%, followed by quartz (22.2%) and organic carbon (15.52%). On the other hand, kaolinite was the second dominant mineral in the coal samples, with an average amount of 27.38%, but it was exceeded by organic carbon, with an average amount of 45.26%.

Kaolinite, a mineral assemblage of the carbonaceous shale and coal, has been identified as a REY-bearing mineral because its enriched cation valencies provide an ion-exchangeable site for the adsorption of free ion REE³⁺ [30]. The mechanism of the secondary enrichment of REE Kaolinite has been interpreted by cation exchange ensued from isomorphous substitution of Al³⁺ for Si⁴⁺. Some studies [31,32,33] have documented that exchangeable kaolinite cations are found on the mineral’s edges and basal surface. Additionally, the geochemical analysis of the diagenetic change of kaolinite’s rare earth element mobilization and fractionation [34] revealed that Cretaceous kaolin has greater light REE mobility than recent kaolin, which exhibits very low REE fractionation. The reason was attributed to the differences in sedimentary kaolin physical properties and the presence of organic acids in groundwater [34].

Furthermore, siderites are commonly formed in organic-rich sediment and are one of the REE-bearing minerals due to the Fe³⁺ site forming complexes with the REE³⁺ ions [35]. Siderite occurs as a fine-grained ellipsoidal shape, indicating an authigenic setting, in carbonaceous shale and coal seams [36]. Carbonaceous sediments bearing siderites are said to be characterized by low sulfur concentrations and little or absence of pyrite deposited in nonmarine environments [37]. The findings by Shen et al. [35] slightly attributed alkaline conditions to the co-existence of authigenic kaolinite and epigenetic siderite in the same sample.

Petrographically, the studied carbonaceous shale and coal samples under a non-polarized reflected white light depicted the maceral group and observable mineral compositions (Figure 3). The macerals mainly comprised vitrinite, inertinite groups, and mineral compositions (Figure 4a,d). Most of the examined samples were dominated by the vitrinite group, suggesting a prevalence of organic input from higher plant vascular tissues, such as xylem and phloem tissues from the terrestrial environment.

Figure 4a,b showed an elongated shape of vitrinite and expanded bands of inertinite bands, indicating an intense diagenetic alteration. This diagenetic alteration mechanism might have been the explanation for the mobilization and fractionation of rare earth elements of kaolinite’s diagenetic alteration by cation exchange ensued from the isomorphous substitution of Al³⁺ for Si⁴⁺ in the studied samples [34]. Furthermore, pyrite framboids appear spherical, infilling cracks in the inertinite and vitrinite. Pyrite is the predominant sulfide in CS and coal [38]. The study CS and coal samples showed low sulfur content (<1% S), suggesting that they are derived primarily from the parent plant material in fluvial depositional environments [39]. Quartz appeared as angular grains, found mainly with inertinite and clay minerals, suggesting a detrital origin [40]. Minerals such as quartz, carbonates, and clay fill were also detected.

The presence of inertinite, or vitrinite, is evidence of authigenesis, likely resulting from remobilization of the elements or precipitation from syngenetic or epigenetic solutions [37].

The proximate analysis shows the assay content of the moisture, volatile matter, ash content, and fixed carbon of the studied carbonaceous shale and coal samples (

Table 4. Proximate analysis of the studied samples.

Samples	CS BS1	CS BS2	CS BS3	CS BS2-1	CS BS2-4	CS BS2-2	CS BS2-3	Coal BSC-1	Coal BSC-2	Coal BSC-3	Coal BSC-4
% inherent moisture content	1.3	1.4	0.6	2.0	4.0	1.7	1.6	1.3	2.3	2.5	2.0
%Ash content	86.0	85.5	86.5	81.2	83.3	62.3	73.2	78.5	33.1	32.1	66.4
%volatile Matter	9.1	9.9	10.7	10.1	13.4	14.1	13.7	11.0	18.6	18.9	11.5
%Fixed carbon (by Cal)	3.6	3.2	2.2	6.7	0.7	11.9	11.5	29.2	46	46.5	20.1
%Total carbon	5.41	5.37	5.23	8.68	2.76	26.14	14.84	11.42	50.28	51.06	22.61
%Organic carbon	4.69	4.6	4.9	6.95	2.77	24.04	14.36	11.32	48.69	49.32	21.11
%Elemental Carbon	0.72	0.77	0.39	1.78	<0.05	2.06	0.55	0.18	1.58	1.59	1.49

). The moisture contents were generally low, ranging from 0.6%–4.0% in CS and 1.3%–2.5% in coal samples. The frictional heat generated by grinding the samples in the pulverizing machine may have resulted in moisture loss. The moisture content represents water that may be physically or chemically bound in the CS and coal, except mineral hydrates that decompose above 110 °C [25].

The relatively low volatile matter in CS (9.1%–14.1%) compared to coal 11.0%–18.9%) may be attributed to nearly double sulfide (0.28%) in the coal than in CS (0.17%). The volatile compounds, such as sulfur, carbon dioxide, and water vapor, that developed between 110 and 900 °C under N₂ due to thermal breakdown are equivalent to the volatile matter content of coal and CS [41]. In this study, the volatile matter content varies with the coal rank. The lowest volatile matter content (2%–12%) is typically found in anthracite, followed by bituminous (15%–45%), sub-bituminous (28%–45%), and lignite (24%–32%) [25,42].

The fixed carbon content of coal samples ranged from 20.1% to 46.5%, while the CS showed very low fixed carbon content from 0.7% to 11.9%. Similarly, the total carbon and organic carbon content values are very high relative to CS, probably due to the high vitrinite and inertinite macerals in the coal samples. Bituminous (50%–70%), sub-bituminous (30%–57%), and lignite (25%–30%) have the next highest fixed carbon content of coal, after anthracite (75%–85%) [25,42]. The studied coal samples were categorized into lignite to sub-bituminous coal based on the fixed carbon and volatile matter content.

The ash yield shows the highest values in CS, ranging from 62.3–86.5 and lower in coal samples (32.1%–78.5%). It is what is left over after the coal sample has lost moisture and volatile materials and the fixed carbon has burned in the air at 900 °C. The ash concentration does not necessarily match the mineral content of the coal sample since the minerals may have experienced thermal decomposition, such as the oxidation of pyrite and FeS₂ to iron oxide or the loss of CO₂ in carbonates.

5. Discussion

5.1. Evaluation of Rare Earth Elements in Carbonaceous Shale

The outlook coefficient (Coutl) of REY is a critical parameter that indicates the economic viability of REY in host resources. The higher the Coutl, the more promising the REY resource with respect to potential industrial value [43]. The ratio of the relative amount of critical REY metals in the REY sum to the relative amount of excessive REY was proposed for the primary estimation of ore quality, as calculated using Equation (1).

Table 5. Concentrations of REY (ppm) together with REY values estimated for the upper continental crust (UCC).

Sample	Sc	La	Ce	Pr	Nd	Sm	Eu	Gd	Tb	Dy	Ho	Er	Tm	Yb	Lu	Y	ΣREE	ΣREY	ΣREY + Sc	Coutl
BS-1	19.11	67.51	134.84	14.7	55.26	10.14	1.99	8.65	1.33	8.6	1.67	4.59	0.63	4.13	0.68	44.79	314.72	359.51	378.62	0.86
BS-2	17.74	67.29	149.33	15.51	58.87	11.67	1.97	8.67	1.25	7.7	1.54	4.32	0.61	3.8	0.58	38.84	333.11	371.95	389.69	0.76
BS2-1	21.16	82.51	179.1	19.02	68	12.75	2.26	11.69	1.66	9.49	1.9	5.3	0.74	5.04	0.71	50.52	400.17	450.69	471.85	0.77
BS2-4	27.87	9.86	19.21	2.37	9.15	1.72	0.65	1.9	0.35	2.28	0.53	1.51	0.23	1.73	0.25	15.46	51.74	67.2	95.07	1.38
BS-3	16.44	60.55	113.89	14.14	50.68	9.74	1.62	8.14	1.3	6.79	1.33	3.75	0.62	3.42	0.58	38.56	276.55	315.11	331.55	0.90
BS2-2	28.15	94.24	152.17	16.32	52.98	15.2	1.51	10.65	0.25	10.26	1.5	4.09	0.48	6.93	0.62	60.75	367.2	427.95	456.1	0.86
BS2-3	39.52	115.42	173.77	27.41	63.73	26.2	2.02	21.54	1.14	9.15	0.41	5.18	0.72	5.82	0.54	71	453.05	524.05	563.57	0.94
BSC-1	18.33	67.26	143.84	15.23	56.91	11.22	1.87	9.37	1.38	8.26	1.66	4.29	0.63	4.34	0.62	42.67	326.88	369.55	387.88	0.80
BSC-2	40.25	90.46	185.65	37.45	65.73	34.35	1.74	20.48	1.96	9.67	1.12	5.61	1.25	5.28	0.41	80.98	461.16	542.14	582.39	0.94
BSC2-1	16.63	56.17	138.11	14.4	47.76	10.56	0.85	7.46	0.14	6.51	0.42	3.21	0.33	2.6	0.53	27.73	289.05	316.78	333.41	0.65
BSC-3	39.12	78.37	154.65	26.14	65.73	23.14	2.56	10.28	2.14	10.17	2.54	6.18	0.54	6.45	0.51	51.75	389.4	441.15	480.27	0.88

shows the concentrations of REY together with Rey values estimated for the upper continental crust (UCC). Seredin and Dai [42] proposed the outlook coefficient (Coutl) as a criterion for assessing REY as economic raw materials. Coutl values that range from 0.7 to 1.9 are considered promising for REY recovery, exhibiting REO ≥ 1000 ppm. Also, values < 0.7 and >2.4 are regarded as unpromising and highly promising, respectively.

$$Coutl={\displaystyle \frac{(\mathrm{N}\mathrm{d}+\mathrm{E}\mathrm{u}+\mathrm{D}\mathrm{y}+\mathrm{E}\mathrm{r}+\mathrm{Y}/\mathrm{\Sigma }\mathrm{R}\mathrm{E}\mathrm{E}}{(\mathrm{C}\mathrm{e}+\mathrm{H}\mathrm{o}+\mathrm{T}\mathrm{m}+\mathrm{T}\mathrm{b}+\mathrm{L}\mathrm{u}/\mathrm{\Sigma }\mathrm{R}\mathrm{E}\mathrm{E})}}$$

(1)

The total REY concentration indicates that the studied carbonaceous shale from the Witbank Coalfield is a promising area and is economically viable for recovery (Figure 5). The Coutl value for all the studied carbonaceous shale ranged from 0.76 to 1.38, except the BS2-4 sample, indicating a promising area for REY recovery. The BS2-4 shale sample shows an unpromising source, probably due to the relatively low content of ion-adsorption kaolinite clay [7]. Furthermore, all the studied coal samples at the Witbank Coalfield associated with the carbonaceous shale indicate a promising source, consistent with a previous study [44] which evaluated the Coutl values of selected South African coal ash samples from power stations. The study [44] showed that the Coutl ranged from 0.793 to 0.815, falling under cluster II, a promising source.

5.2. Enrichment Types and Anomalies of REY in Carbonaceous Shale

Rare earth elements have been recently classified into three types based on geochemical and industrial perspectives [43]. The three types of REY enrichment are light (La, Ce, Pr, Nd, and Sm), medium (Eu, Gd, Tb, Dy, and Y), and heavy (Ho, Er, Tm, Yb, and Lu) REY corresponding to La_N/Lu_N > 1, La_N/Sm_N > 1 and Gd_N/Lu_N < 1, and La_N/Lu_N < 1 respectively [42]. The studied samples identified two REY enrichment types, L and H. Most samples exhibited L-REY enrichment type, except for one carbonaceous shale sample, BS2-4, which showed H-type enrichment (

Table 6. Enrichment Types and anomalies of REY in studied Carbonaceous Shale and Coal Samples.

Samples	LaN/LuN	LaN/SmN	GdN/LuN	EuN/EuN*	CeN/CeN*	GdN/GdN*
BS-1	1.06	1.00	1.07	1.01	0.98	1.07
BS-2	1.24	0.86	1.26	0.92	1.05	1.05
BS2-1	1.24	0.97	1.39	0.91	1.03	1.15
BS2-4	0.42	0.86	0.64	1.65	0.91	1.02
BS-3	1.11	0.93	1.18	0.85	0.89	1.03
BS2-2	1.62	0.93	1.45	0.70	0.87	2.04
BS2-3	2.28	0.66	3.36	0.50	0.70	1.82
BSC-1	1.16	0.90	1.27	0.87	1.02	1.09
BSC-2	2.35	0.40	4.21	0.32	0.70	1.18
BSC2-1	1.13	0.80	1.19	0.57	1.11	2.13
BSC-3	1.64	0.51	1.70	0.63	0.77	0.69

and Figure 6). The anomalous distribution of REY has been determined using Equations (2)–(4) [19], decoupling individual Ce, Eu, and Gd concentrations from the other REY in the UCC-normalized distribution patterns. All the studied samples are characterized by distinctive positive Gd anomalies with values > 1, ranging from 0.69 to 2.13, except one sample, BSC-3, with a value of 0.69. Furthermore, Eu and Ce positive anomalies have characterized the REY concentration in some carbonaceous shale samples BS-1 and BS2-4, while other samples exhibited negatively weak anomalies. The Eu_N/Eu_N* and Ce_N/Ce_N* ratio ranges from 0.32 to 1.65 and from 0.70–1.11 to 1.04, respectively, for the current studied samples (Table 6), exhibiting a weak negative to no anomaly.

$${\displaystyle \frac{CeN}{CeN}}\ast ={\displaystyle \frac{CeN}{0.5LaN+0.5PrN}}$$

(2)

$${\displaystyle \frac{EuN}{EuN}}\ast ={\displaystyle \frac{EuN}{\left[\left(SmN\ast 0.67\right)+\left(TbN\ast 0.33\right)\right]}}$$

(3)

$${\displaystyle \frac{GdN}{GdN}}\ast ={\displaystyle \frac{GdN}{\left[\left(SmN\ast 0.33\right)+\left(TbN\ast 0.67\right)\right]}}$$

(4)

5.3. Sediment Provenance Region for REY

Carbonaceous shale deposits with significant amounts of REY minerals may be derived from source regions, including granitic rocks, high-grade metamorphic rocks, or re-worked sedimentary rocks. The ratio of immobile elements such as La, Th, Co, Sc, Cr, Ni, Al, and Ti has been a veritable provenance indicator tool for typical clastic sedimentary deposits and coal seams [45]. Typical ratios of Al₂O₃ and TiO₂ for sedimentary rocks derived from mafic, intermediate, and felsic-dominated sediment source regions have expressed values ranging from 3–8, 8–21, and 21–70, respectively [46]. The studied carbonaceous shale samples (12.6–42.15) and associated coal (16.41–25.33) suggest that they are derived from intermediate and felsic volcanic rocks. Furthermore, Figure 7a,b show the geochemical plots Na₂O + K₂O against SiO₂, suggesting alkali granite, granite, and granodiorite provenance sources for the studied shale and coal. The samples showed enrichment in low and heavy rare elements (Figure 7c,d) crystallized from a low potassium tholeiitic and medium calc-alkaline magma (Figure 7e,f). The findings show that CaO and Na₂O were eliminated from the system, most likely as a result of plagioclase loss, and that K₂O was depleted in the more thoroughly examined material, which was mostly related to the loss of microcline and biotite [34]. The presence of Cr₂O₃ suggests that throughout the diagenesis and coalification processes of black shale, heavier Cr isotopes preferentially leach into fluids, leaving residues concentrated in lighter isotopes [47]

Additionally, Ce anomalies are one controlling factor of the geochemical composition of the sediment source region [19]. The Ce_N/Ce_N* ratio ranges from 0.70 to 1.05 and from 0.70 to 1.11 for the carbonaceous shale and coal samples, respectively, exhibiting a weak negative to no anomaly. The Ce anomalies in the studied samples indicate that the materials were derived from different sediment source regions. The Witbank Coalfield, belonging to the Early Permian sedimentary sequence of the Vryheid Formation of the Karoo Super Group sediments, comprises a wide range of compositions of felsic, intermediate, and mafic [48].

6. Conclusions

The studied carbonaceous shale and the associated coal seams revealed a predominance of kaolinite minerals acting as an ion-adsorption clay for REY enrichment. Siderite was identified as one of the REY hosts due to the Fe³⁺ site forming complexes with the REE³⁺ ions. The petrographic analysis revealed that siderite is intimately mixed with organic carbon, a common cleat-infilling material in coal seams and carbonaceous shale. The associated coal samples fall into the lignite to sub-bituminous coal category based on the fixed carbon and volatile matter content, while the carbonaceous shale has the highest ash content.

Based on the outlook coefficient (Coutl) of REY, the prospect of rare earth elements deposit in studied carbonaceous shale is promising for economic recovery. The two REY enrichment types that were identified include light (La, Ce, Pr, Nd, and Sm) and heavy (Ho, Er, Tm, Yb, and Lu). The Eu_N/Eu_N* and Ce_N/Ce_N* ratio for the current studied samples exhibited a weak negative to no anomaly, and most of the studied samples are characterized by distinctive positive Gd anomalies derived from sediment source regions weathered from alkali granite, granite, and granodiorite provenance formed from a low potassium tholeiitic and medium calc-alkaline magma.