Mineralogical and Chemical Properties and REE Content of Bauxites in the Seydişehir (Konya, Türkiye) Region

Abstract

The most important bauxite deposits in Türkiye are located in the Seydişehir (Konya) and Akseki (Antalya) regions, situated along the western Taurus Mountain, with a total reserve of approximately 44 million tons. Some of the bauxite deposits have been exploited for alumina since the 1970s. In this study, bauxite samples, collected from six different deposits were examined to determine their mineralogical and chemical composition, as well as their REE content, with the aim of identifying which bauxite types are enriched in REEs and assessing their economic potential. The samples included massive, oolitic, and brecciated bauxite types, which were analyzed using optical microscopy, X-ray diffraction (XRD), X-ray fluorescence (XRF) and inductive coupled plasma-mass spectrometry (ICP-MS), field emission scanning electron microscopy (FESEM-EDX), and electron probe micro-analysis (EPMA). Massive bauxites were found to be more homogeneous in both mineralogical and chemical composition, predominantly composed of diaspore, boehmite, and rare gibbsite. Hematite is the most abundant iron oxide mineral in all bauxites, while goethite, rutile, and anatase occur in smaller quantities. Quartz, feldspar, kaolinite, dolomite, and pyrite were specifically determined in brecciated bauxites. Average oxide contents were determined as 52.94% Al₂O₃, 18.21% Fe₂O₃, 7.04% TiO₂, and 2.69% SiO₂. Na₂O, K₂O, and MgO values are typically below 0.5%, while CaO averages 3.54%. The total REE content of the bauxites ranged from 161 to 4072 ppm, with an average of 723 ppm. Oolitic-massive bauxites exhibit the highest REE enrichment. Cerium (Ce) was the most abundant REE, ranging from 87 to 453 ppm (avg. 218 ppm), followed by lanthanum (La), which reached up to 2561 ppm in some of the massive bauxite samples. LREEs such as La, Ce, Pr, and Nd were notably enriched compared to HREEs. The lack of a positive correlation between REEs and major element oxides, as well as with their occurrences in distinct association with Al- and Fe-oxides-hydroxides based on FESEM-EDS and EPMA analyses, suggests that the REEs are present as discrete mineral phases. Furthermore, these findings indicate that the REEs are not incorporated into the crystal structures of other minerals through isomorphic substitution or adsorption.

1. Introduction

Bauxite deposits, beyond their primary role as the main source of aluminum, have gained increasing attention due to their potential as alternative sources of critical and strategic elements such as rare earth elements (REE), gallium (Ga), germanium (Ge), scandium (Sc), and vanadium (V). Understanding the enrichment, mineralogical associations, and occurrence modes of these elements is of great importance, particularly in the context of resource security and sustainable raw material supply. REEs are used in almost every field, and their areas of use are increasing day by day. The demand for these elements has significantly increased in recent years, especially due to their usage in many high-tech applications such as high-power permanent magnets, lasers, lenses, phosphors used in electronic displays, various renewable energy technologies, hybrid vehicles, and the production of new metal alloys [1,2,3,4]. Furthermore, since each of these elements has different areas of use, the substitution of one for another is quite limited. In this regard, the Seydişehir–Akseki region in southern Türkiye hosts significant bauxite deposits that remain underexplored in terms of their critical element content.

The textural characteristics, types, origin, and mineralogical-chemical characteristics of karstic bauxites in the Mediterranean region and various parts of the world have been investigated by many researchers [5,6,7,8,9,10,11,12,13,14,15,16,17]. Regardless of whether the underlying carbonate rock is karstified or not, bauxites developed on these rocks are generally classified as karstic bauxites, whereas those formed on aluminosilicate rocks are referred to as lateritic bauxites [5,6,17,18,19,20,21]. In addition to the exploitation of bauxite deposits for aluminum production, interest in REEs has increased due to the significant concentration in some bauxite deposits. Türkiye’s economic bauxite reserves are estimated at 87 million tons, a substantial portion of which is located in the Seydişehir–Akseki region in the west of the Taurus Mountains [18]. Numerous karstic bauxite deposits with varying reserves are found in this area, extending from Konya–Seydişehir to Antalya–Akseki in the Western Taurides (Figure 1). According to Karadağ et al. [18] the total reserve in Seydişehir region is approximately 32 million tons, while the combined reserves of around 90 large and small bauxite deposits in the broader Seydisehir–Akseki region are estimated at about 85 million tons. The most significant bauxite deposits in the region include Mortaş, Doğankuzu, Değirmenlik, and Morçukur. Mining operations have been completed in the Kaklıktaştepe and Çatmakaya deposits. Investigations are ongoing to assess the reserves and characteristics of the Değirmenlik and Morçukur deposits.

Since the 1970s, mining activities have been concentrated around the Mortaş and Doğankuzu deposits. The physical properties, mineralogical composition, and major-element content of the bauxite deposits in the region are generally similar. The bauxites are typically boehmitic, with Al₂O₃ contents ranging from 55% to 67% [18]. Previous studies have suggested that some of these deposits may also possess economic potential in terms of REE content [18,22,23,24]. It has been reported that REE and certain trace elements (Ga, Zr, Ti, Sc, etc.) are more enriched in karstic bauxite deposits compared to lateritic bauxites [25], and that both light REE (LREE) and heavy REE (HREE) exhibit considerable mobility during the formation of these bauxites [26,27,28]. REE enrichment in karstic bauxites is primarily attributed to two mechanisms: (1) the absorption of REEs onto the surfaces of diaspore, gibbsite, and clay minerals and (2) isomorphic substation, whereby REEs replace chemically similar ions within the crystal structures of these minerals [12,28].

The trace-element contents of the bauxite deposits—particularly REE, Ga, and several critical/strategic elements—have not been comprehensively investigated. Previous studies have reported that LREEs are more abundant than HREEs in the karstic-type bauxite deposit. The karstic bauxite deposits in the Seydişehir (Konya) and Akseki (Antalya) regions are notable for their high average REE contents, exceeding 1000 ppm [18,29]. Karadağ et al. [18] reported that the REE concentrations in the Seydişehir bauxites show a wide range—from 17 to 2494 ppm—with LREEs comprising a major portion of this content. They further emphasized that the REE levels in these deposits may hold economic significance based on chemical analysis results. However, the study did not specify which bauxite types or specific ores exhibit REE enrichment. To date, there has been limited research on which bauxite deposits and types contain high REE concentrations in terms of their economic potential. Furthermore, image analyses such as EPMA and FESEM-EDS have not previously been conducted to investigate REE-bearing phases in the bauxite deposits of the region. The objectives of this study are to determine the following: (1) the concentration of critical elements, particularly REE and certain trace elements (Ga, Ge, Sc, V, etc.), in the bauxite deposits of the Seydişehir–Akseki region; (2) which types of bauxite are enriched in REEs; and (3) the REE-bearing minerals present in high-REE bauxite, or the form in which REEs occur.

2. Geology

The study area is situated within the Seydişehir–Akseki region, characterized by complex stratigraphy dominated by autochthones and allocthones tectonic units [18]. The Çaltepe Formation, representing the oldest lithostratigraphic unit in the region and dated to the Lower–Middle Cambrian, is composed of dolomite, dolomitic limestone, limestone, and phyllite-schist lithologies [19,30]. The Seydişehir Formation, which dates to the Upper Cambrian–Lower Ordovician, comprises low-grade metamorphosed schist, phyllite, quartzite, sandstone, and nodular limestone. This formation is widely observed across the region and has also been proposed as a potential source rock for bauxite and found is observed at the base of the study area [18]. The Middle–Late Cambrian Seydişehir Formation exhibits a transitional contact with the Çaltepe Formation and unconformably overlies the Bozkır ophiolitic mélange, which consists predominantly of gabbro/diabase, amphibolite, serpentinite, radiolarite, limestone, and tuff lithologies of Jurassic–Cretaceous age. The emplacement of this mélange occurred through tectonic contact during the Late Eocene–Early Miocene period. The Dogger-aged unit is conformably overlain by the Katrangediği Formation, which is of Cenomanian–Maastrichtian age and entirely composed of carbonate rocks including different sizes with dissolution cavities that serve as primary sites for bauxite mineralization. This unit is unconformably overlain by the Doğankuzu Formation, also of Cenomanian–Maastrichtian age (Figure 2). At the base of the unit, brecciated limestones weakly cemented by bauxite and earthy bauxite containing pyrite crystals of several millimeters in size are observed. In the upper levels, oolitic–pisolitic bauxite and partially oolitic massive bauxite are present. These two types of bauxite generally exhibit lateral gradation into one another. In addition, dark reddish-brown bauxites are present at these levels. Above the massive bauxite, weakly consolidated earthy bauxite with a yellowish to light reddish-brown color and a composition distinct from that of the basal earthy bauxite is observed. Bauxite deposits are located either on or within the Katrangediği Formation, primarily within large, lens-shaped karstic depressions that are several hundred meters in length and width, as well as in smaller sinkholes and fractures. The fact that the Katrangediği Formation is composed of micritic limestone suggests deposition in a low-energy, shallow marine (lagoonal) environment.

3. Materials and Methods

The bauxite samples were collected from various levels of six different bauxite deposits located in the Seydişehir–Akseki region, particularly from the dissolution cavities within the Cenomanian–Maastrichtian aged Katrangediği Formation. The sampled bauxite deposits and their corresponding codes are as follows: Çatmakaya (B1), Değirmenlik (B2), Doğankuzu (B3), Kaklıktaştepe (B4), Mortaş (B5), and Morçukur (B6) (Table 1, Figure 1).

In general, samples were taken from the lower, middle, and upper parts of each deposit. However, due to operational and geological limitations, sampling strategies were adopted accordingly. At the Değirmenlik and Kaklıktaştepe deposits, mining operations had already been completed, leaving only very limited in situ bauxite exposures. Similarly, in the Çatmakaya and Morçukur bauxites, where exploration drilling is still ongoing and surface outcrops are restricted, sampling was conducted in a way that best represents the observable ore. At the Mortaş deposit, where mining began in the 1970s and reserves have been largely exhausted, only a limited number of samples could be collected.

Given the difficulty in capturing the full horizontal and vertical variability of all deposits, the sampling strategy focused on collecting specimens from all macroscopically distinguishable bauxite types in order to determine their mineralogical and chemical characteristics. More detailed sampling was carried out at the Doğankuzu mine, which is actively operated and considered as the most significant deposit in the study area due to its relatively high REE content and economic potential. Mineralogical and textural characteristics of the bauxites were examined in thin and polished sections using transmitted and reflected light petrographic microscopy.

The petrographic analyses were conducted with a Nikon Eclipse Ci Pol model microscope in the Department of Geological Engineering at the Konya Technical University. Prior to analysis, bauxite samples were ground in an agate ring mill (Fritsch Vibrating Cup Mill Pulverisette 9, Idar-Oberstein, Germany) at 600 rpm for 1 min to prepare them for whole rock mineralogical and geochemical investigations. The mineralogical compositions of bulk rocks and clay fractions were determined using X-ray diffraction (XRD) with a fully automated Rigaku D/MAX 2200 PC diffractometer (Tokyo, Japan) at Hacettepe University (Ankara). Analyses were conducted on both randomly oriented and oriented-treated samples using a CuKα radiation (λ = 1.54082 Å), operating at 40 kV and 40 mA, over a 2 to 70°2θ angular range with a scanning speed of 2°/minute. Clay minerals were identified from the oriented XRD patterns of clay-sized fractions, which were air-dried, heated (350 and 550 °C for 4 h), and ethylene-glycol treated using a Phillips PW 3710/1830 model diffractometer (Almelo, Netherlands) at the General Directorate of Mineral Research and Exploration (MTA) Ankara.

The morphological properties and elemental compositions of high-REE-content bauxite samples were investigated using a field emission scanning electron microscope (FESEM; ZEISS Gemini SEM 500 equipped with Ultim Extreme EDX and BSD detectors; Oberkochen, Germany, and High Wycombe, UK, respectively) at an accelerated voltage of 2.0 kV. The presence and distribution of REE were analyzed qualitatively using energy-dispersive X-ray spectroscopy (EDX) through both linear element spectra and spatial element mapping. The samples were coated with iridium and mounted on metal holders prior to analysis. Based on the chemical analysis results, the B3-4 sample from the Doğankuzu bauxite deposit—which exhibited particularly high REE content—was subjected to oxalic acid leaching to remove hematite, following the method described by Zhang et al. [31]. Post-treatment analysis using FESEM-EDX and microprobe techniques revealed that hematite was largely eliminated and allowed a clearer assessment of REE distribution. Quantitative chemical compositions of high-REE-bearing bauxites—including minerals, grains, and matrix (mineral, oolite, grain, and matrix)—were determined using an electron probe X-ray microanalyzer (EPMA-1600; Kyto, Japan). Microprobe analyses were conducted at the YEBİM laboratory, Ankara University, using the JEOL JXA 8230 (Tokyo, Japan) instrument under the following operating conditions: accelerating voltage 20 kV, a beam current of 10 nm, and beam diameter of 2 µm. Polished thin sections for microprobe analyses were carbon-coated using Quorum Q150T ES device (Lewes, UK). The detection limits for Na, Mg, Al, Si, Fe, Mn, K, Ca, F, and Ti oxides were below 0.04 wt%. Natural oxide and mineral standards were used for calibration, and matrix correction were applied using JEOL’s ZAF/PB-FP software (version 3.7) integrated with the JXA-8230 EPMA system (Tokyo, Japan). The chemical analysis of the bauxite samples —including major, trace, and REEs—was conducted at the ITU-JAL (İstanbul) and the MTA laboratory (Ankara).

All reagents used in this study were of analytical grade and certified for impurity levels. Ultrapure water with a maximum resistivity of 18.3 MΩ·cm was obtained using a Zeneer UP 900 water purification system (Seoul, Republic of Korea). The following chemicals were supplied by Merck: Hoechst Wax C micropowder (Darmstadt, Germany), orthophosphoric acid (85% H₃PO₄), sulfuric acid (95%–97% H₂SO₄), hydrochloric acid (37% HCl), nitric acid (65% HNO₃), hydrofluoric acid (38%–40% HF), and boric acid (>90% H₃BO₃). Mixed standard solutions containing 10 µg·mL⁻¹ of Ga, Rb, Cd, In, Cs, Tl, Y, REEs, Th, and U, as well as an internal standard solution containing 1000 µg·mL⁻¹ of Re, were obtained from PerkinElmer. The certified reference material (CRM), Jamaican Bauxite (NIST 698), was supplied by the National Institute of Standards and Technology (NIST).

The collected bauxite samples were initially crushed using a jaw crusher (Haan, Germany). The crushed materials were then oven-dried at 105 °C for 24 h. Approximately 20–30 g of dried sample was milled to a particle size of <45 µm using a RETSCH RS-200 milling system (Haan, Germany) equipped with a tungsten carbide grinding set. Milling was performed for 5 min at 1250 rpm. The powdered samples were subsequently stored in a desiccator to prevent moisture absorption.

Major oxides and trace elements were semiquantitatively determined using a BRUKER S8 TIGER wavelength-dispersive X-ray fluorescence spectrometer (WDXRF) (Karlsruhe, Germany). For pellet preparation, the milled sample was mixed with wax at a ratio of 1:5 (w/w) and pressed into pellets using a HERZOG pellet press (Osnabrück, Germany). Loss on ignition (LOI) was determined by combusting approximately 2–3 g of the ground sample in a muffle furnace at 1000 °C for 1.5 h. The CRM was also pelletized and analyzed to validate the XRF measurements. The accuracy of the XRF analysis was evaluated using a certified reference material (CRM), NIST SRM 698–Jamaican Bauxite (Gaithersburg, MD, USA). The CRM was subjected to the same sample preparation and analytical conditions as the bauxite samples. Ten replicate measurements were performed, and the mean values were compared with the certified and/or reference values. The results demonstrated good agreement with the certified data, confirming the reliability and accuracy of the analytical method (Table S1a). In the XRF analysis, the detection limits for major element oxides, the lowest, and the highest measurable concentrations ranged between 0.01% and 99.9%, respectively. For trace elements, the detection limits were element-specific, with ranges as follows: barium (Ba), 5–4000 ppm; niobium (Nb), 0.6–700 ppm; scandium (Sc), 1.6–60 ppm; strontium (Sr), 0.9–1650 ppm; vanadium (V), 1.4–340 ppm; and zirconium (Zr), 4.9–550 ppm.

Prior to ICP-MS analysis, the milled samples were fully decomposed using a BERGHOF Speedwave™ MWS-3+ (Eningen, Germany) microwave digestion system. The digestion procedure was conducted in two sequential steps. In the first step, approximately 100–200 mg of powdered sample was placed into high-pressure Teflon digestion vessels, and an acid mixture of H₃PO₄:H₂SO₄ (6.5:3.5, v/v) was added. In the second step, an additional acid mixture of HCl:HNO₃:HF (1:1:1, v/v) was introduced into the same vessels.

Following microwave digestion, the resulting solutions were transferred to 50 mL volumetric flasks after rinsing with ultrapure water, resulting in clear sample solutions suitable for ICP-MS analysis. A blank solution was prepared using the same digestion procedure for background correction. Additionally, a certified reference material (CRM) was digested under identical conditions to validate both the microwave digestion and the ICP-MS analytical method. PerkinElmer ELAN 6000 DRC-e model ICP-MS system (Massachusetts, Germany) was employed to determine the concentrations of Cd, Cs, Ga, In, Rb, Tl, REEs, Th, and U. Calibration standards were prepared from stock solutions at concentration levels of 5, 10, 50, and 100 µg L⁻¹. The calibration curves were linear, with correlation coefficients (R²) of at least 0.999 for all analyzed elements. To minimize matrix effects and analytical deviations, a 50 ppb rhenium (Re) internal standard was added to each sample solution. Both CRM and sample solutions were diluted and analyzed in triplicate. Relative standard deviation (RSD%) values were maintained below 5% throughout the analysis. After every three sample analyses, a CRM solution was reanalyzed to monitor the accuracy and stability of the instrument. The validation of the ICP-MS analysis was carried out using the certified reference material (CRM) NIST SRM 698–Jamaican Bauxite. The accuracy of the analytical method was assessed by analyzing the CRM under identical sample preparation and measurement conditions. Fifteen replicate analyses were performed, and the mean values were compared with the certified and/or reference values, demonstrating good agreement. The limits of detection (LOD) and limits of quantification (LOQ) for each element were determined according to the IUPAC definition, calculated as 3σ and 10σ of the blank signal, respectively. Here, σ represents the standard deviation obtained from five independent blank solutions, each analyzed in triplicate (n = 15). Calibration slopes derived from a four-point external calibration curve (5, 10, 50, and 100 µg L⁻¹) were used for each element to calculate LOD and LOQ values (Table S1b). Europium (Eu) and cerium (Ce) anomalies were calculated using REE concentrations normalized to Post-Archaean Australian Shale (PAAS) values [32].

Table 1. Location and mineralogical composition of the analyzed bauxite types.

Sample Number	Location/Bauxite Mine	Mineralogical Composition	Bauxite Type
B1-1	Çatmakaya	Bhm + Hem + Dsp + Ant	Massive
B1-2		Bhm + Hem + Dsp + Go	Massive
B2-1	Değirmenlik	Bhm + Hem + Kln	Massive
B2-2		Bhm + Hem + Gbs + Kln	Massive
B3-1	Doğankuzu	Bhm + Hem + Kln + Gbs + Ant + Py	Oolitic-pisolitic
B3-2		Dsp + Hem + Bhm + Kln + Ant	Massive
B3-3		Dsp + Hem + Bhm + Kln + Ant	Massive-oolitic
B3-4		Bhm + Hem + Dsp + Kln + Gbs + Ant + Py	Massive-oolitic
B3-5		Bhm + Hem + Kln + Ant	Massive
B3-6		Bhm + Hem + Kln + Ant + Kln	Massive
B3-7		Bhm + Hem + Kln + Ant + Kln	Earthy
B4-1	Kaklıktaştepe	Bhm + Cal + Hem + Go	Brecciated
B4-2		Bhm + Hem + Ant	Massive-oolitic
B4-3		Bhm + Cal + Hem + Ant	Brecciated
B4-4		Bhm + Hem + Ant + Kln	Massive-oolitic
B5-1	Morçukur	Bhm + Hem + Kln + Gbs	Oolitic
B5-2		Bhm + Hem + Kln	Oolitic-pisolitic
B6-1	Mortaş	Bhm + Hem + Kln + Go	Brecciated
B6-2		Bhm + Hem + Kln	Massive

Notes: Abbreviations of mineral names: Ant: anatase, Bhm: boehmite, Cal: calcite, Dsp: diaspore, Gbs: gibbsite, Go: goethite, Hem: hematite, Kln: kaolinite, Py: pyrite, Qz: quartz [33].

Table 1. Location and mineralogical composition of the analyzed bauxite types.

Sample Number	Location/Bauxite Mine	Mineralogical Composition	Bauxite Type
B1-1	Çatmakaya	Bhm + Hem + Dsp + Ant	Massive
B1-2		Bhm + Hem + Dsp + Go	Massive
B2-1	Değirmenlik	Bhm + Hem + Kln	Massive
B2-2		Bhm + Hem + Gbs + Kln	Massive
B3-1	Doğankuzu	Bhm + Hem + Kln + Gbs + Ant + Py	Oolitic-pisolitic
B3-2		Dsp + Hem + Bhm + Kln + Ant	Massive
B3-3		Dsp + Hem + Bhm + Kln + Ant	Massive-oolitic
B3-4		Bhm + Hem + Dsp + Kln + Gbs + Ant + Py	Massive-oolitic
B3-5		Bhm + Hem + Kln + Ant	Massive
B3-6		Bhm + Hem + Kln + Ant + Kln	Massive
B3-7		Bhm + Hem + Kln + Ant + Kln	Earthy
B4-1	Kaklıktaştepe	Bhm + Cal + Hem + Go	Brecciated
B4-2		Bhm + Hem + Ant	Massive-oolitic
B4-3		Bhm + Cal + Hem + Ant	Brecciated
B4-4		Bhm + Hem + Ant + Kln	Massive-oolitic
B5-1	Morçukur	Bhm + Hem + Kln + Gbs	Oolitic
B5-2		Bhm + Hem + Kln	Oolitic-pisolitic
B6-1	Mortaş	Bhm + Hem + Kln + Go	Brecciated
B6-2		Bhm + Hem + Kln	Massive

Notes: Abbreviations of mineral names: Ant: anatase, Bhm: boehmite, Cal: calcite, Dsp: diaspore, Gbs: gibbsite, Go: goethite, Hem: hematite, Kln: kaolinite, Py: pyrite, Qz: quartz [33].

4. Results

4.1. Macroscopic Properties

The bauxite deposits in the region exhibit a range of colors, including light brick red, brownish red, dirty white-gray, light yellow and dark red, and burgundy (Figure 2). The bauxite shows predominantly fractured (diaclastic), brecciated, nodular, and cavenous textures. It commonly displays massive, colloform, and oolitic/pisolitic structure and texture types (Figure 3). Additionally, massive, granular, earthy, and occasionally plate-like or banded appearances are also observed. In the upper zones of the bauxite deposits, the bauxite is typically yellowish to light red in color, earthy in texture, and weakly consolidated.

The massive bauxites are typically purple-dark brown in color and locally display banded or oolitic structures. In some instances, dark brown to black oolites, with diameters ranging from 0.1 to 0.5 cm, are present within the massive bauxite. These bauxites are commonly found in the middle to upper levels of bauxite ores and are generally harder than other bauxite types. The bauxites containing pisolites exhibit a light burgundy color, with pisolite diameters ranging from 2 to 3 cm. These bauxites are typically found beneath the massive-oolitic bauxite layers and commonly show both horizontal and vertical transitions. Brecciate bauxites are typically observed at lower level of karstic cavities. The bauxites contain limestone and schist fragments of 0.1–1.0 cm in size, as well as native sulfur and several millimeter-sized pyrite crystals with a size of a few mm. Brecciated bauxites, on the other hand, are typically observed in the lower sections of karstic cavities. These brecciated bauxites contain limestone and schist fragments ranging in size from 0.1 to 1.0 cm, as well as native sulfur within cavities, along with several millimeter-sized pyrite crystals.

4.2. Mineralogical Composition and Properties

Microscopic examination of thin and polished sections clearly reveals the microtextural relationships between minerals, showing that iron oxides and aluminum hydroxide minerals are the primary constituents of the bauxites (Figure 4). Boehmite and diaspore are observed in microcrystalline form, commonly associated with hematite and clay minerals, indicating that these phases acted as matrix-forming minerals and major components during the diagenetic process. Goethite, on the other hand, is generally observed as irregularly shaped grains, mostly occurring within microcracks or in association with hematite, suggesting that it formed during the later stages of supergene oxidation processes. Iron oxihydroxide minerals (hematite, goethite) and sulfur-bearing minerals (e.g., pyrite) are commonly present throughout the samples (Figure 4H). In addition, a small amount of pyrite (Py) in framboidal form was found, with sizes ranging between 5 and 10 µm in some samples (Figure 4E). Among Fe-oxide/hydroxide minerals, hematite is observed in all bauxite types, whereas goethite is detected to a lesser extent (Figure 4D–H). The ooides exhibiting a core of Al-hyroxide (boehmite or diaspor) mineral surrounded by hematite, and generally range in size from 50 to 500 µm, with considerable variation in the number and thickness of concentric rings (Figure 4F,G). Opaque minerals are mostly dispersed within the matrix, along fractures, and partially around the oolites; they are rarely observed inside the oolites. Furthermore, some internal features within oolitic structures are interpreted as possibly representing kaolinite.

XRD analyses revealed that massive and predominantly oolitic bauxites typically contain diaspore, boehmite, hematite, rutile/anatase, and kaolinite along with minor amounts of gibbsite, ilmenite, quartz, and feldspar (Figure 5). Among the Al-oxide/hydroxide minerals, boehmite is more commonly observed, while diaspore appears as the dominant Al-hydroxide mineral in some samples (Table 1). Gibbsite is less abundant, particularly in massive bauxites compared to other bauxite types (Figure 5). The brecciated bauxites are primarily composed of boehmite, gibbsite (Al(OH)₃), and, less frequently, diaspore (AlO(OH)), along with rutile/anatase, illite, calcite, dolomite, pyrite, quartz, and feldspar. Although the characteristic peaks of diaspore are not distinctly visible, the diffraction peaks of boehmite and hematite are well-defined. The earth bauxites consist mainly of boehmite, hematite/lepidocrocite, kaolinite, gibbsite, and calcite. The characteristic reflections of boehmite are observed at °2θ values of 6.11, 3.16, 2,36, 1.96, and 1.83 Å values, while diaspore displays typical reflections at °2θ values of at 3.99, 2.32, 2.13, Å, and 2.08 Å (Figure 5).

4.3. Chemical Composition

While the contents of major element in bauxite samples are generally similar, significant differences are observed in trace-element contents (Table S2). Among the major oxides, Al₂O₃ exhibits the highest standard deviation, ranging from 0.02 to 9.96 wt%. The average contents of SiO₂, Al₂O₃, Fe₂O₃, and TiO₂ in bauxite samples are 7.04%, 52.94%, 18.21%, and 2.69%, respectively. The contents of alkali and alkaline earth element of bauxites are relatively low in all samples.

The bauxites are significantly enriched in Ga, with concentrations ranging from 20.5 to 84.1 ppm (average 64.3 ppm), and the standard deviation is 19.7. The ∑REE contents of the bauxite samples varies widely from 161 to 4072 ppm. The LREE content ranges from 146 to 3931 ppm (average 685 ppm) and HREE from 8 to 141 ppm (average 51 ppm), while the ∑LREE/∑HREE ratios very between 9 and 40. Among the REE group, Ce is the most abundant element in all bauxite samples except in B3-1 and B3-2, with concentrations ranging from 87 to 453 ppm (average: 218 ppm; standard deviation: 94.5). The La content is notably high in sample B3-2, reaching 2561 ppm; it ranges from 16.6 to 2561 ppm (average: 260 ppm; standard deviation: 626 ppm) in the remaining samples, its significant variability in the bauxite samples (Table S1). The Y content also displays a wide range, from 19 to 10,378 ppm (average: 785 ppm; standard deviation: 2567 ppm), suggesting considerable enrichment in certain samples. To calculate the Ce anomaly, the normalized Ce value ([Ce]ₙ) was compared with the normalized values of neighboring REEs using the formula (Ce/Ce*)ₙ = Ceₙ/((0.5Laₙ) + (0.5 Ndₙ))n, as proposed by Bau et al. [34].

There is no correlation between Fe₂O₃-TiO₂ and Al₂O₃-TiO₂ (Table S3). Overall, no significant correlations were observed between the major oxides and the trace elements, except for SiO₂, Fe₂O₃, and TiO₂, which show moderate positive correlations with certain trace elements (Table S3). Additionally, the major element oxides do not exhibit any meaningful correlation with the REEs, except Ce (Table S3). All the REEs, except Ce, exhibit strong positive correlations with each other, no significant correlation with other REEs, but display a moderate positive correlation with Fe₂O₃, suggesting that it may be partially associated with Fe-bearing minerals (Table S3). Cluster analysis of the major element oxides and trace elements in the bauxite samples reveals three distinct groups in the dendrogram (Figure 6). The dendrogram, which illustrates the geochemical relationships among the samples, shows that the rare earth elements (REEs) form a strongly correlated cluster, while the major oxides and other trace elements are grouped into separate clusters—supporting the correlation results presented in Table S3. The first group in the dendrogram consists of all REEs except Ce, along with K₂O, which is closely associated with them. The second group comprises Ba, Rb, V, TiO₂, Nb, Zr, Ga, Th, and U—elements that also exhibit positive correlations in the correlation analysis—with Fe₂O₃. Al₂O₃, Ce, and Sc, which are subsequently incorporated into this group. Lastly, SiO_2, and Rb are added to this cluster.

The evaluation of the cluster and factor analysis of bauxites suggests that the increase in the contents of Al₂O₃, Ce, and Sc is related to the decomposition of the source rocks (e.g., schist, psammites, partially limestones), and it has been interpreted that Sc generally behaves as less mobile during bauxite formation, as it is commonly found associated with Fe and Ti oxides. The later addition of Al₂O₃ to the group may be related to its geochemical properties, which is different from those of other elements. The separation of Ce from the other REE group, when they should be present together, demonstrates the differences in the oxidation states of these elements, and the subsequent addition of Ce to the group reveals its different behavior and sensitivity to redox conditions. The separation of Ce from other REE through factor analysis explains why the content of this element in the environment can be relatively lower than that of La. The formation of a third group between CaO, Y, and Sr may indicate that they are likely controlled by similar geochemical processes.

Factor analysis of the bauxite samples indicates that the first six factors with initial eigenvalues greater than 1 account for approximately 90% of the total variance (

Table 2. Result of factor analysis of elements in the bauxite samples.

Component	Initial Eigen Values			Extraction Sums of Squared Loadings
	Total	% of Variance	Cumulative %	Total	% of Variance	Cumulative %
1	12,588	40,605	40,605	12,588	40,605	40,605
2	8291	26,744	67,349	8291	26,744	67,349
3	2634	8496	75,846	2634	8496	75,846
4	1806	5827	81,672	1806	5827	81,672
5	1411	4550	86,222	1411	4550	86,222
6	1098	3541	89,763	1098	3541	89,763

). The first factor, explaining about 41% of the total variance, is characterized by a strong negative loading of CaO, along with negative loadings of trace elements such as Sc, Y, V, and Sr (Figure 7a). In contrast, other major oxides exhibit positive factor loadings, suggesting that CaO and REEs follow different geochemical pathways. The negative loading of CaO may reflect its distinct geochemical behavior relative to the environments enriched in REEs. The second factor, which explains roughly 27% of the variation, is defined by positive loadings of Al₂O₃, TiO₂, Fe₂O₃, Zr, Nb, Th, U, V, Ga, Ba, and SiO₂, and strong negative loadings of CaO and Y (Figure 7b). This factor may reflect the persistence of relatively resistant minerals hosting these elements. The third factor, accounting for 8.5% of the variance, is represented by strong negative loadings of Al₂O₃, Fe₂O₃, and TiO₂, along with Ce, whereas SiO₂ and K₂O exhibit positive loadings (Figure 7c). This factor may correspond to a source material undergoing partial weathering, characterized by the presence of silica, K-feldspar, and quartz. The fourth factor, contributing approximately 6% of the variance, includes high positive loadings of some alkaline earth and trace elements such as Sc, Th, and U (Figure 7d). The fifth and sixth factor, explaining about 4.5% and 3.5% of the variance, respectively, show a clear geochemical separation between LREEs and HREEs, with Ce distinctly separating from other REEs—likely reflecting redox-controlled differentiation processes (Figure 7e).

While the PAAS-normalized REE patterns of all bauxite ores and rocks from different lithologies in the region are nearly flat, the Doğankuzu (B3) bauxites exhibit notably higher REE enrichment compared to other bauxite deposits and associated rocks. Additionally, B1, B3, B4, B5, and carbonate rocks show negative Ce anomalies, whereas B2 and B6 display clearly positive Ce anomalies. Ophiolite and Seydişehir Formation samples also show slightly positive Ce anomalies (Figure 8).

4.4. Microprobe (EPMA and FESEM-EDS Analysis)

EPMA analyses were performed on samples with high REE content, excluding the bauxite deposits labeled B1 and B2, and specifically on the oxalic acid-treated B3-2 sample. In the microscopic examination of these samples, oolitic grains were commonly observed (Figure 9A,B,D). As a result of the analysis, it was determined that the REE content of the bright whitish and mostly prismatic grains around/next to the oolites is much higher compared to dark gray-black grains, matrix, or oolites (Figure 9,

Table 3. REE oxide contents (wt%) of crystals/grains analyzed by electron microprobe in sample B3-2. The analysis points correspond to those shown in Figure 9. Bolded values indicate points with relatively high REO contents.

No	La2O3	Ce2O3	Pr2O3	Sm2O3	Eu2O3	Gd2O3	Ho2O3	Er2O3	Nd2O3	Total
A-1	21.87	0.94	7.67	0.59	0.43	0.81	0.36	0.00	3.01	35.72
A-2	21.86	1.17	7.80	0.61	0.49	1.30	0.47	0.02	3.20	36.93
A-3	0.06	0.01	0.01	0.02	0.01	0.04	0.02	0.00	0.04	0.23
B-1	24.92	0.97	7.73	0.66	0.46	0.93	0.39	0.00	3.39	39.53
B-2	22.79	1.10	6.73	0.65	0.47	0.99	0.35	0.06	3.39	36.53
B-3	28.38	0.75	8.91	0.75	0.62	0.65	0.31	0.03	4.00	44.40
B-4	0.01	0.00	0.01	0.01	0.04	0.00	0.00	0.00	0.02	0.08
B-5	0.00	0.00	0.04	0.00	0.00	0.00	0.03	0.00	0.01	0.09
B-6	0.03	0.01	0.01	0.02	0.03	0.01	0.00	0.04	0.02	0.20
B-7	0.00	0.00	0.04	0.00	0.01	0.02	0.02	0.00	0.00	0.10
C-1	26.79	1.38	8.90	0.79	0.62	1.45	0.53	0.04	4.10	44.68
C-2	25.66	1.15	8.28	0.75	0.53	1.36	0.51	0.06	3.94	42.25
C-3	0.14	0.03	0.07	0.00	0.00	0.00	0.00	0.04	0.03	0.31
D-1	26.41	1.36	8.13	0.80	0.59	1.48	0.54	0.03	4.04	43.39
D-2	23.02	1.43	7.11	0.83	0.49	1.49	0.54	0.04	3.95	38.92
D-3	21.57	1.06	6.64	0.60	0.45	1.21	0.41	0.04	3.39	35.37
D-4	27.06	1.19	8.42	0.71	0.53	0.99	0.39	0.06	3.81	43.26
D-5	0.00	0.00	0.03	0.00	0.00	0.01	0.00	0.01	0.02	0.09

). No significant REE content was determined in any area other than the short prismatic bright whitish crystal/grains (Figure 9). The rings from the center to outside of the oolitic grains were also analyzed, but no significant enrichment in terms of REE content was determined (Figure 9B).

In particular, bright white prismatic crystals/grains in the B3-2 and B3-4 samples exhibited significantly high concentrations of REEs, ranging from 35.4% to 44.7%, in contrast to the surrounding matrix and other mineral phases (Table 3, Figure 9). Among the trace elements, fluorine (F) concentrations ranged between 2.0% and 4.5%. The concentric rings of the oolitic structures were found to be significantly depleted in REEs. In sample B3-2, the La₂O₃ content reached as high as 28.38%, indicating an exceptionally high concentration. Pr₂O₃, Nd₂O₃, Gd₂O₃, and Ce₂O₃ all exhibited concentrations exceeding 1%. In sample B3-4, the REE content of the white grains was also high (40.74%). Although the F content was lower than that of B3-4OX, it ranged from 2.0% to 3.86% at several points. The CaO and ThO₂ contents were relatively low, at 1.48% and 0.13%, respectively. As in sample B3-2, La₂O₃ was the dominant REE (25.47%), and additional REEs with concentrations above 1% were identified in many analyses.

In the element maps obtained from FESEM, LREE was found to be concentrated in the specific areas, and the elemental compositions identified by EDS were consistent with the region showing high LREE intensities on the maps. In sample B3-2, the LREEs, along with Al, Fe, Ca, and F, were mapped within the oolite and the surrounding white prismatic crystals (Figure 10). It was observed that La, Ce, Nd, and Pr—among the REEs—were concentrated in the white grains, whereas other REESs were not detected in these LREE-enriched areas. Ca and F were also observed in the regions with high LREE content, while Al and Fe were distributed in different rings of the oolite. Al is also more concentrated in distinct concentric rings within the oolite, with Al being particularly concentrated in the dark-gray ring. The co-occurrence of LREEs, Ca, and F in the same areas suggests that the REE-bearing mineral is La-parisite, given its composition (containing Ca and F) and the predominance of La over Ce.

In samples with high REE content, the compositional variations in the oolites and grains surrounding the oolites were examined in terms of major elements and REE concentrations using linear spectrum analysis (Figure 11). The spectrum revealed that the gray-black oolite rings are enriched in Al (30.6%), while the thin, light-colored rings exhibit elevated Fe content (20.6%) (Figure 11b,c). The concentrations of Si, Ti, and La were all below 2%. The compositional variations in oxygen levels from the core of the oolite to the outer ring generally followed a trend parallel to that of Al, and to a lesser extent, Fe. In the linear spectrum, the inverse relationship between Fe and Al-bearing mineral precipitated during different stages under oxidizing conditions (Figure 11c). Furthermore, no clear correlation was observed between the variation in La content and changes in Al and Fe concentrations.

In the areal spectrum analysis (Figure 12), it was determined that the concentration of Al and La are similar; while LREEs (La, Ce, Nd, and Pr) and Y are co-located with Ca, Al, Fe, and Si are concentrated in separate regions (Figure 12). The spatial distribution of the elements identified by FESEM-EDS is quite similar to the EPMA results (Figure 10). In all REE-bearing crystals/minerals analyzed via FESEM-EDS, La consistently exhibited a significantly higher concentration than Ce, with La content ranging between 30 and 78%. Other LREEs with concentrations exceeding 1.0% include Pr, Nd, Gd, and Ce, in descending order. These findings align well across methods; however, while EPMA results show Y₂O₃ content below 0.5%, FESEM-EDS recorded maximum values of 5.3% in the B3-2 sample and 4.7% in the B3-4OX sample. Additionally, these values correspond closely with the Y concentrations obtained from ICP-MS result of the B3-2 sample (Table S2).

5. Discussion

Although these morphological bauxite types have slightly different mineralogical compositions, boehmite is the most common alumina mineral found in almost all bauxite deposits. Diaspore and gibbsite are found in small amounts along with boehmite, especially in dark-brown, massive, and earthy bauxites (e.g., Doğankuzu, Kaklıktaştepe), while hematite is found as the second most abundant mineral aside from alumina minerals (Table 3, Figure 5). While the mineralogical composition of brecciated bauxites resembles other bauxites, they differ in that they contain calcite and small amounts of dolomite and pyrite, whereas earthy bauxites contain calcite and illite. Anatase/rutile is the main titanium mineral and has been identified in small amounts in most bauxites. In the clay size XRD analysis of bauxites, except for earthy bauxites containing illite, no other clay minerals have been identified apart from kaolinite. This may be attributed to the depletion of alkaline elements such as Na and K due to their leaching and mobilization during the chemical weathering, along with the concurrent enrichment of relatively immobile elements such as Al, Ti, and Fe in their in oxidized states.

During the bauxitization of bedrock and/or source rock, some elements remain in the environment while others move away according to their chemical properties. Therefore, during the bauxitization process, alkali and alkaline earth elements, in particular, become impoverished in the forming secondary deposits, while elements such as LREE, Ga, Hf, Zr, Nb, and Ti become enriched [37]. During the Jurassic–Cretaceous period, precipitation of limestones in the region, and the weathering process of bauxites, as well as the decomposition of organic matter, bacterial sulfate reduction, and especially the alteration of mica (biotite) in the source material, may have resulted in the formation of iron minerals and pyrite [38]. It has been suggested that various Fe-bearing minerals (goethite, lepidocrocite, and biotite) play a significant role in the formation of sedimentary pyrite, and that iron sulfide precipitates as framboidal pyrite in environments where conditions are anoxic waters [39]. Pyrite precipitation may have occurred during the bauxite stage, particularly when the groundwater level rose within kartic or dissolution cavities, through both biotic and abiotic reactions [18]. The REE content of bauxites varies significantly, ranging from 161 to 4072 ppm, both within individual bauxite deposits and among different mines. This wide variation may be related to processes involved in bauxite formation. Analyses indicate that the REE (particularly LREE) content in massive bauxite is generally higher than in other bauxite types. Additionally, massive bauxite contains significantly lower—or entirely absent—amounts of detrital components such as quartz, feldspar, calcite, and dolomite. These high-grade bauxites are classified as first quality by the processing company and are used directly for alumina production. Their formation may be attributed to the final stages of bauxitization or to more stable geochemical conditions.

In addition to the changes in the major-element content, the changes in the trace-element contents are related to the mineralogical-chemical composition of the host rock [40], the physicochemical conditions during bauxite formation, the solubility and composition of elements in the parent rock, water mobility, tectonic features of the environment, etc. It has been reported that trace elements in bauxite deposits such as Zr, V, Nb, Cr, Y, Ga, and Th, as well as LREE, can be significantly enriched [41]. It has also been reported that the REEs in solutions are affected by many factors such as the properties of parent rock, adsorption, desorption, pH, Eh, complexation, and the chemical properties of groundwater [26,41,42,43]. The concentration of Fe in leach profiles, the degree of mineral leaching, and the geochemistry of elements play a role in the enrichment of trace and REEs. In all the bauxite deposits, LREE contents are significantly more enriched than HREE contents; LREE contents range from 146 to 3931 ppm (average 950 ppm) and HREE range from 8 to 141 ppm (average 51 ppm) (Table S2). Typically, HREEs are preferentially extracted under alkaline and weak alkaline conditions, while LREEs are extracted under acidic conditions [42,43,44]. Karadağ et al. [18] reported the average REE contents of earthy (n = 8), brecciated (n = 13), oolitic (n = 16), and massive (n = 17) bauxites in the Mortaş deposit as 282, 453, 530, and 986 ppm, respectively (total of 54 samples). Although the average REE concentrations of these bauxite types are lower than those in the Doğankuzu bauxite mine, the massive bauxites are enriched in REEs by approximately two to three times compared to the other bauxite types.

Generally, trace elements such as gallium, cobalt, nickel, zirconium, and vanadium are found in all types of bauxite. Secondary minerals like phosphates (e.g., apatite) [45], clay minerals [46], and oxides and hydroxides of Fe and Mn [18,47,48,49] are important traps for REE. The degree of leaching in bauxites can affect the concentration of the primary minerals containing REE. The strong negative Ce anomaly of the Doğankuzu bauxite (B3) indicates strong oxidation conditions, while the weak negative anomalies of Kaklıktaştepe (B4) and Morçukur (B5) bauxites indicate weak oxidation conditions. Ce anomaly is also influenced by factors such as pH and water depth [50] (Table S1). Circulating waters play an important role in the formation process of karstic bauxite deposits, and these waters typically have a neutral to slightly alkaline pH, being in equilibrium with calcite [51].

In addition, Ce³⁺ is more stable in low pH solutions and Ce shows a negative anomaly as the pH increases [52]. REE–carbonate complexes are more common at these pH levels [53,54,55]. Liu et al. [56] stated that the waters were alkaline and reductive during the formation process of Permian-aged bauxite ore (Western Guangxi), and these conditions favor the widespread formation of minerals such as bastnäsite and parisite (REE–fluorocarbonates) in various bauxite deposits. It was explained that in the matrix of the bauxites and in the inner regions, centers, and rings of the ooids, the REE content is high and cerianite forms. It has been suggested that REE is transported to the karst environment, primarily adsorbed on the surfaces of clay minerals or in the form REE minerals such as cerianite in the upper part of the weathering profile [6,12].

The different behavior of Ce in the upper levels of bauxites compared to other REEs, provides information about the paleo-redox evolution of the environment during chemical weathering processes [57]. The fractionation of Ce is related to the supergene oxidation of Ce³⁺ to Ce⁴⁺ at pH = 5–6 [29,57,58,59,60,61], and some researchers [57,58,59,60,61] have also mentioned that the formation of cerianite is associated with the oxidation of Ce³⁺→Ce⁴⁺. Fluctuations in the groundwater level during bauxite formation, increases in pH, and the adsorption of Ce onto the surface of goethite or hematite may have played a role in the formation of the positive Ce anomaly (1.95, 18). The value of the Eu anomaly is mostly similar, close to unity, and ranging from 0.70 to 1.57 (mean 1.03, STD = 0.19) in all bauxite samples (Table S1). The lack of correlation of REEs with Al₂O₃, Fe₂O₃, and SiO₂ may indicate that aluminum, iron hydroxides, and clay minerals (mostly kaolinite and illite) in the studied bauxites are not the primary host phases for REE adsorption and/or absorption. This interpretation is supported by the FESEM-EDS and EPMA analyses, which reveal that REEs, Al, and Fe are located in distinct mineral phases. Additionally, the lack of correlation between REEs and Al₂O₃ or Fe₂O₃, along with the observation that aluminum- and iron-bearing minerals do not contain REEs, further indicates that REEs are not structurally incorporated into alumina or iron hydroxide minerals. In some instances (B3-1-4 and B4-2, 4, Table S2), the REE content, being much higher than that of other samples, may be related to the formation conditions of the bauxites or variations in the source material. Generally, intense chemical weathering can promote the mobilization and fractionations of LREE and HREE—as well as Ce and/or Eu [18,62,63]—with pH playing a key role in controlling the mobility. Under acidic conditions REEs are easily leached from weathering products [64]. Therefore, neutral to alkaline conditions may have played a more prominent role in the formation of bauxites with exceptionally high REE contents.

In the EPMA conducted from the centers of the oolites to the outer rim, the REE content is less than 1.0%, while the REE content of all the white short prismatic grains surrounding the oolites exceeds 20%. This may indicate that REE-containing minerals have deposited around the oolitic grains in a partially energetic, turbulent environment. In the areal and linear analyses of the oolites using FESEM-EDS and EPMA, the Al-rich dark-brown rings and the Fe-rich light-colored rings, along with the absence of REE in the rings, indicate that REEs are not present in the structure of the aluminum and/or iron minerals of the bauxites.

In the analyses conducted, La was found to be the most abundant REE, with concentrations ranging from 20% to 28%, exceeding the typical La content reported for La-bastnäsite and parisite in the literature [64,65]. According to Leleyter et al. [66], unbalanced REE distribution may be primarily associated with carbonates, Fe-oxides/hydroxides, or organic matter. However, the absence of a correlation between REEs and CaO or Fe₂O₃, as well as the spatial separation of these elements in elemental mapping, suggests that REEs in the studied samples are not bound to these phases.

Bastnäsite is recognized as the most common REE-bearing mineral in lateritic and karst-type bauxite deposits, as well as in carbonate environments [5]. It is a rare carbonate mineral containing REEs and forms a solid solution series with hydroxyl-bastnäsite, [(Ce,La)CO₃(OH,F)], and [(Ce,La,Y)CO₃F] [60]. Although bastnäsite is generally rich in Ce, there are also species reported in the literature that have high La, Y content [67]. Srinivasan et al. [66] described bastnäsite as a fluorocarbonate mineral composed of alternating MF²⁺ (where M = La³⁺, Ce³⁺, Y³⁺) and CO₃²⁻ layers, noting that Ce-bastnäsite and La-bastnäsite exhibit similar isomorphic structures. Additionally, bastnäsite and parisite are closely related minerals and exhibit similar isomorphic structures [64,65,66,67]. In the chemical composition of La-parisite, the oxide contents of La, Ce, Nd, and Pr are reported as approximately 25%, 13%, 15%, and 8.5%, respectively [62]. These values closely resemble the compositions of the small prismatic crystals observed in the analyzed bauxites (Table 2; Figure 9 and Figure 11). Therefore, based on the combined results from EPMA, EDS, and, to some extent, ICP-MS analyses, it was concluded that the REE-bearing mineral phase identified in the bauxites is La-parisite, due to its compositional similarity to the La-parisite described in the literature.

It has been stated that the important factors affecting REE fractionation and mobility influence the distribution of REEs. CE concentrations consistently remained around 1.0% in areal EPMA analyses across all sample groups, even in areas where La content exceeded 20%. These EPMA findings are consistent with ICP-MS and EDS results, yet deviate from the compositions typically reported for La-bastnäsite in comparable rock types [68]. In addition to the EPMA, EDS, and chemical analysis data, the lack of correlation between major oxides—particularly CaO—and LREEs, the distinct geochemical behavior of Ce compared to other LREEs, as well as its separate clustering in both cluster and factor analyses, indicate that the REE-bearing mineral phase is not associated with carbonate minerals. Furthermore, the relatively low Ce content compared to other LREEs suggests that Ce is not significantly concentrated in the studied environment. It is well established that REE mobility and fractionation are strongly influenced by environmental pH and redox potential (Eh). It has been indicated that the Ce content in the fluids responsible for transporting REEs during the formation of La-parisite is relatively low. This is attributed to the mildly oxidizing conditions under which Ce is partially oxidized to Ce⁴⁺, resulting in its immobilization and subsequent depletion from the solution [69]. In particular, low pH and reducing conditions enhance the leaching and mobility of REEs. The mobility of La, Gd, and Y is primarily controlled by pH, while Ce behavior is additionally dependent on the redox potential [12,27,70]. The pronounced enrichment of La, Nd, Pr, and Y compared to Ce, in the examined bauxite samples may therefore reflect the prevailing geochemical conditions, particularly variations in pH and Eh.

Notably, the behavior of Eu and Ce under redox conditions is of particular geochemical significance. These two elements often exhibit marked fractionation due to their variable oxidation states. Under reducing conditions (Eh < −350 mV), Eu³⁺ may be reduced to Eu²⁺, whereas Ce³⁺ is readily oxidized to Ce⁴⁺ and can precipitate under moderately high pH or Eh conditions [12,27,60,69,71,72]. These redox-driven transformations may explain the relative depletion of Ce and the distinctive REE patterns observed in the studied bauxites.

In the context of the European economic scenario, bauxite deposits have been reported to contain an average of approximately 0.18% TREO in Europe and 0.2% in Sweden, highlighting their potential as future REE resources [68]. The massive-oolitic bauxites of the Doğankuzu deposit, with an average total rare earth oxide (TREO) content of 0.57%, suggest a significantly greater resource potential compared to the average TREO concentrations reported for the European bauxite deposits.

Given its relatively high REE concentration, Doğankuzu bauxites can be considered an economically viable candidate for REE recovery. As of 2018, Türkiye’s probable bauxite reserves were estimated at 63 million tons, with an annual production of approximately 1 million tons [73,74]. The Mortaş and Doğankuzu bauxite deposits in the study area are still actively mined. Although the REE content in the Değirmenlik and Morçukur bauxite deposits is relatively low, their similar geological origin suggests a high potential for REE-rich bauxite in these areas as well.

This elevated REE content, particularly in LREEs, positions the Doğankuzu deposit as a potentially strategic resource, not only for aluminum production but also for the extraction of critical raw materials essential to advanced technologies. Considering the increasing global demand for REEs—particularly Nd, Pr, and Y—which are essential in many high-tech applications, the development of economically and environmentally sustainable technologies for the extraction of REEs from bauxite deposits is becoming increasingly important.

The elements in question are not traded on international commodity exchanges, making it difficult to determine a definitive price or assess market demand with precision. Although REE prices vary depending on their specific applications, they are generally offered to the market at significantly different and often fluctuating rates. Given the presence of 10 million tons of bauxite in the study area—home to Türkiye’s most important bauxite reserves—the total LREO content is estimated to be around 34,000 tons. Based on 2025 market data, the approximate prices of selected rare earth elements (REEs) per kilogram are as follows: La—USD 6.93, Ce—USD 3.31, Pr—USD 119, Nd—USD 148, Y—USD 38, and Sm—USD 4.89. These values reflect the projected price ranges reported by Garside [75] for the 2020–2030 period. Although La and Ce are more abundant and therefore lower in price, Pr, Nd, and Y are economically significant due to their essential roles in strategic technologies. Accordingly, using the projected 2025 prices [75], the total estimated value of these elements is approximately USD 154 million.

It was determined that the massive-oolitic bauxites in the Doğankuzu mine contain TREO at levels of potential economic significance. While numerous analytical studies have focused on the recovery of REEs from red mud—the residue generated after alumina extraction from bauxite—the number of investigations directly addressing REE recovery from primary bauxite ore remains limited. Identifying the REE content of bauxites and developing economically and environmentally sustainable technologies for their extraction could enable the direct recovery of LREEs from bauxite deposits. Such advances would provide a valuable alternative to red mud-based recovery, particularly for deposits with REE-rich. Moreover, the integration of REE recovery during the alumina production process, without generating red mud or requiring separate extraction stages, could further enhance the economic and environmental sustainability of the operation. Our results highlight the economic viability of recovering LREEs during the alumina production process from high-TREO bauxites—not through post-processing of red mud, but via integrated methods applied simultaneously with alumina extraction. This approach underscores the potential of bauxite deposits as strategic sources of critical REEs.

6. Conclusions

Seydişehir bauxites, characterized as karstic-type bauxite ores, contain between 54 and 63 wt% Al₂O₃ and present a promising potential source of REEs. The findings from this study can contribute significantly to REE research, and particularly to the recovery and valorization of REEs from bauxite resources. The key results obtained from the ore analyses are summarized below:

(1)

The primary minerals identified in bauxites are boehmite, diaspore, hematite, and kaolinite. Minor phases include gibbsite, goethite, pyrite, anatase/rutile, illite, calcite, quartz, and feldspar minerals.

(2)

The REE content of the bauxite ores varies widely. The B3 ore from the Doğankuzu mine shows the highest REE concentration, ranging from 161 to 4072 ppm. Bauxites with high REE content (e.g., B3, B4) exhibit weak to moderately negative Ce anomalies, whereas those with lower REE contents display positive Ce anomalies. All samples exhibit positive Eu anomalies.

(3)

The absence of correlation between LREEs and major elements such as Al, Fe, and Si and the fact that these elements are spatially co-located EPMA and FESEM element maps suggests that REEs are neither adsorbed onto nor structurally incorporated into alumina phases (diaspore, boehmite), goethite, or clay minerals.

(4)

The co-localization of LREEs with Ca and F on the element maps, along with the low Ce content compared to La, suggests that the REE-bearing minerals include La-rich parisite and/or La-bastnäsite from the bastnäsite group.

(5)

The Doğankuzu bauxite, from Seydişehir deposits, stands out as particularly rich in critical LREEs. These findings indicate that the deposit has strong potential for future exploitation to meet the growing demand particularly for LREEs in advanced technological applications.