Introduction
Bentonite is an industrial clay-based rock, predominantly composed of minerals from the smectite group (usually montmorillonite). Bentonites retain remarkable properties, including but not limited to excellent sorption capacity, ability to swell, plasticity, cohesiveness, large specific surface area, and a substantial capacity for exchangeable cations. These attributes facilitate their diverse applications, ranging from pet litter to foundry work, pelletizing, building industry, geotechnical barriers, and even in pharmaceutical and cosmetics industries (Grim and Güven, Reference Grim and Güven1978; Eisenhour et al., Reference Eisenhour, Reisch, Kogel, Trivedi, Barker and Krukowski2006; Murray, Reference Murray2007; Carrado and Komadel, Reference Carrado and Komadel2009; Eisenhour and Brown, Reference Eisenhour and Brown2009; Christidis, Reference Christidis2011; Harvey and Lagaly, Reference Harvey and Lagaly2013; Dehn and McNutt, Reference Dehn and McNutt2015; Alkrad et al., Reference Alkrad, Al-Sammarraie, Dahmash, Qinna and Naser2022; Kaufhold et al., Reference Kaufhold, Dohrmann, Wallis and Weber2023).
Bentonite forms primarily through the transformation of volcanic glass, which is mainly found in fine pyroclastic materials (tuff or volcanic ash), and occasionally in perlites or perlitic breccias (Kraus et al., Reference Kraus, Čičel and Machajdik1982; Kraus et al., Reference Kraus, Šamajová, Šucha, Lexa and Hroncová1994; Yildiz and Kuscu, Reference Yildiz and Kuscu2004). Economic deposits of bentonite are formed mainly by diagenesis and hydrothermal alterations of these volcanic glass-bearing materials (e.g. Güven, Reference Güven and Bailey1988; Christidis and Dunham, Reference Christidis and Dunham1997; Belousov and Karelina, Reference Belousov and Karelina2022; Gilg et al., Reference Gilg, Kaufhold, Ufer and Kaufhold2024). The quality and grade of bentonite during its formation are influenced by many factors such as the proportion of volcanic glass, the chemistry and permeability of the parent volcanic rock, and the presence and chemistry of fluids, as well as their circulation (e.g. Grim and Güven, Reference Grim and Güven1978; Christidis, Reference Christidis2008a; Christidis and Huff, Reference Christidis and Huff2009; Kadir et al., Reference Kadir, Külah, Erkoyun, Christidis and Arslanyan2019; Ünal Ercan et al., Reference Ünal Ercan, Schroeder, Gençoğlu Korkmaz, Ece and Çiftçi2025). A common source rock of the most industrial bentonites is rhyolitic glassy material; nevertheless, the SiO2 content in rhyolite is significantly greater than that in bentonite. For the formation of high-quality bentonites rich in montmorillonite, an open system and/or relatively high-temperature conditions are essential, enabling the leaching and washing away of excess Si under favorable fluid circulation. In contrast, a closed or semi-closed system, and/or lower temperature conditions lead to the precipitation of excess SiO2 in the form of (opal-C/CT) opal-C and/or opal-CT (Kraus et al., Reference Kraus, Šamajová, Šucha, Lexa and Hroncová1994; Ddani et al., Reference Ddani, Meunier, Zahraoui, Beaufort, El Wartiti, Fontaine, Boukili and El Mahi2005; Christidis and Huff, Reference Christidis and Huff2009).
Bentonite global mining has experienced significant growth, increasing from 9.6 million tons (mt) in 1990 to an estimated 21 mt in 2024 (Virta, Reference Virta1994; USGS, 2025). The leading countries in bentonite mining are USA, India, Turkey, China, and Greece, with Slovakia also playing a crucial role as one of the top six bentonite producers in Europe (Reichl and Schatz, Reference Reichl and Schatz2025; USGS, 2025). More than 90% of bentonite mining in Slovakia takes place within the Central Slovak Volcanic Field (CSVF: Andrejkovičová et al., Reference Andrejkovičová, Madejová, Czimerova, Galko, Dohrmann and Komadel2006; Uhlík et al., Reference Uhlík, Jánošík, Kraus, Pentrák and Čaplovičová2012; Górniak et al., Reference Górniak, Szydłak, Gaweł, Klimek, Tomczyk, Sulikowski, Olejniczak, Motyka, Serwicka and Bahranowski2016; Osacký et al., Reference Osacký, Binčík, Paľo, Uhlík, Madejová and Czímerová2019; Šoltés et al., Reference Šoltés, Mižák, Kúšik, Kyrc and Kubač2024; Hlavný banský úrad, 2025). The oldest known bentonite deposits in the Slovak Republic are found in the eastern part of the country (e.g. Fintice, Kuzmice, and Lastovce deposits), with research on these deposits commencing in the 1950s and 1960s (Gregor and Číčel, Reference Gregor and Číčel1969). These bentonites were primarily recognized as effective plasticizers in foundry molding sands, and when assessing reserves for the first qualitative category of bentonite raw material, critical criteria included green compression strength (minimum 55 KPa) and cation exchange capacity (CEC; minimum 35 meq per 100 g) were determined (Harcek, Reference Harcek1963; Slávik et al., Reference Slávik, Hruskovic, Ocenas and Zuberec1971; Horský, Reference Horský1994). Within the East Slovak Neogene Basin (ESNB), the Michaľany-Lastovce deposit stands out as the largest bentonite deposit, with estimated reserves of 8.8 mt (Harcek, Reference Harcek1963; Horský, Reference Horský1994). This deposit is currently being mined by LB MINERALS Ltd, with peak production reaching 42.4 kt in 2018. Following a substantial reduction in mining activity in 2020, production is on the rise again, projected to be 31.2 kt in 2024 (Fig. 1). This production represents 10% of Slovakia’s total bentonite production (Šoltés et al., Reference Šoltés, Mižák, Kúšik, Kyrc and Kubač2024).
Records of bentonite production (2000–2024) in Slovakia, and in the Michaľany-Lastovce deposit (Šoltés et al., Reference Šoltés, Mižák, Kúšik, Kyrc and Kubač2024).
Figure 1. Long description
A combination line and bar graph. The horizontal X axis represents the Year from 2000 to 2024, labeled as 00 through 24. The primary vertical Y axis on the left measures production for Michalany-Lastovce in k t, ranging from 0 to 45. The secondary vertical Y axis on the right measures production for Slovakia in k t, ranging from 0 to 350.
* Slovakia data (light blue bars): Shows a gradual increase from approximately 50 k t in 2000 to 100 k t by 2009. A sharp jump occurs in 2010 to nearly 200 k t, followed by a fluctuating upward trend that peaks at over 300 k t in 2024.
* Michalany-Lastovce data (red line with white circular markers): Remains very low, near 0 to 2 k t, from 2000 until 2009. In 2010, it begins a significant upward climb, reaching a major peak of approximately 42 k t in 2018. After a sharp drop to roughly 13 k t in 2020, it rises again to approximately 31 k t by 2024.
The legend in the bottom right quadrant identifies the light blue bars as Slovakia and the red line with markers as Michalany-Lastovce.
Although bentonite from the Michaľany-Lastovce area is among the oldest known bentonite deposits in Slovakia, only a limited number of studies have been conducted on it since the 1960s. The first and only detailed report of these bentonites was published by Harcek (Reference Harcek1963), followed by a brief reserve estimation report by Horský (Reference Horský1994). Moreover, in 2005, as part of a project aimed at verifying the properties of Slovak bentonites as sealing barriers for the deep disposal of radioactive waste, mineralogical, chemical, and geotechnical properties were assessed on a single representative sample from this deposit; among various Slovak bentonite samples, the Lastovce sample was found to consist of 41 wt.% of montmorillonite and 27 wt.% opal-C/CT, and had a relatively low CEC value (37 meq per 100 g) which was consistent with the mineralogical findings in the present report (Šucha et al., Reference Šucha, Adamcová, Bujdák, Haasová, Honty, Komadel, Kufčáková, Madejová, Rajec, Stríček, Uhlík and Valúchová2005; Adamcová et al., Reference Adamcová, Frankovská and Durmeková2009; Galamboš et al., Reference Galamboš, Kufčáková and Rajec2009; Osacký et al., Reference Osacký, Honty, Madejová, Bakas and Sucha2009). There has been less focus on the bentonite from the Michaľany-Lastovce deposit than on others due to its lower montmorillonite content.
As such, this work represents the first comprehensive study on the Michaľany-Lastovce deposit in over six decades. In the current study, thorough mineralogical, petrological, and geochemical characterizations of new borehole samples, obtained by REGOS Ltd during their exploration campaign of the Michaľany-Lastovce deposit in 2022, were done. The main objective was to obtain new data on these bentonites using contemporary methods for ongoing prospecting, to compare these findings with previously mined bentonites in Slovakia, and to gain a better understanding of the bentonite deposit in question, and to some extent, its genesis.
Geological setting
The ESNB is a volcano-sedimentary basin located at the junction of the Western and Eastern Carpathians, in the northwestern part of the Trans-Carpathian Depression. This basin spans ~60 km in length (north to south) and 80 km in width west to east (Fig. 2a,b; see also Fig. S1 in the Supplementary material). The tectonic dynamics of the basin have been influenced by the collision between the Carpathian orogen and the North European platform. The Neogene basin fill, ranging from the Eggenburgian to the Pannonian period (21.5–4.2 Ma), consists of terrigenous, marine (including evaporitic) deposits, and volcanic complexes. Significant hydrocarbon reservoirs are also found within the basin. The main accumulation of the volcano-sedimentary materials, reaching a total thickness of 7–9 km, is found in the NW part of the depression (Vass et al., Reference Vass, Kováč, Konečný and Lexa1988; Rudinec, Reference Rudinec1989; Rudinec, Reference Rudinec1990; Kováč et al., Reference Kováč, Kováč, Marko, Karoli and Janočko1995; Pécskay et al., Reference Pécskay, Lexa and Kovacs1995; Galamay and Karoli, Reference Galamay and Karoli1997; Karoli et al., Reference Karoli, Janočko, Kotulák and Verdon1997; Vass et al., Reference Vass, Elečko, Janočko, Karoli, Pereszlényi, Slávik and Kaličiak2000; Honty et al., Reference Honty, Uhlík, Šucha, Čaplovičová, Franců, Clauer and Biroň2004; Subová et al., Reference Subová, Rybár, Šarinová, Hudáčková, Jamrich, Sliva, Šály and Hlavatý2022).
Simplified geological map of the Michaľany-Lastovce study area and Tokaj Mountain surroundings, modified after Lexa et al. (Reference Lexa, Bezák, Elečko, Mello, Polák, Potfaj, Vozár, Eliáš, Konečný and Less2000), including the locations of the drilled boreholes (VLa-1 to VLa-7).
Figure 2. Long description
A multi-panel geological map set.
Panel A, top-right: A small-scale map of Slovakia and surrounding countries (Czechia, Poland, Ukraine, Hungary, Austria). A dashed box in Eastern Slovakia indicates the study area near Lastovce.
Panel B, left: A regional geological map of the Michaľany-Lastovce area. The map features a North arrow in the top-left and a scale bar of 0 to 15 kilometers at the bottom. Major cities labeled include Košice to the West, Michalovce to the Northeast, and Trebišov in the center. The geology is dominated by green neovolcanic areas in the West and light blue Quaternary sediments in the East. A dashed black line indicates major faults, thicker dashed black line indicates boundary between Slovakia and Hungary.
Panel C, right: A detailed inset map of the study area. Seven boreholes, labeled V L a 1 through V L a 7, are marked with red circles. These are concentrated in a pink-shaded region identified as Neogene and Quaternary sediments with intercalations of tuffite (from late Badenian).Bentonitized rhyolite tuffs and bentonites from the Middle Sarmatian are in the middle with pinkish shaded areas. To the West are Quaternary sediments (Holocene and Middle Pleistocene). To the East are Neogene and Quaternary sediments with intercalations of tuffite. Blue dashed line outlines the deposit area.
Legend:
Section A lists regional belts including Flysh Belt (brown), Neovolcanic areas (green), and Pieniny Klippen Belt (blue). Pre-Neogene of inner orogenic zones (light brown). Tertiary mollase zone (grey/yellow). Inner Carpatian Paleogene (violet).
Section B details Quaternary fluvial gravel and sands (light blue); Neogene andesites (orange/green) and porphyry diorite (red); Rhyolites to rhyodacites (orange). Clastic sediments (greenish yellow). Inner Carpathian Palaeogene clastic sediments (brown); and Mesozoic limestone (white).
Section C identifies symbols for faults (solid and dashed lines), drill holes (red circles), and cross sections (blue lines with perpendicular ticks).
The Upper Badenian to Lower Sarmatian syn-rift phase (15–12.7 Ma) resulted from active subduction in the Eastern Carpathian region, associated with a rollback effect of the subducted slab and subsequent uplift of the hinterland mantle, which was characterized by extensive calc-alkaline volcanism (Lexa et al., Reference Lexa, Konečný, Kaličiak, Hojstričová, Rakús and Vozár1993). During this period, the ESNB evolved into a back-arc basin, where sediments displaying a shallowing-upward trend, intercalated with tuffs and tuffites, were deposited. The final phase of rifting within the ESNB occurred during the Upper Sarmatian (~12 Ma), which was followed by a peak in volcanic activity (Lexa et al., Reference Lexa, Konečný, Kaličiak, Hojstričová, Rakús and Vozár1993; Kováč et al., Reference Kováč, Kováč, Marko, Karoli and Janočko1995; Lukács et al., Reference Lukács, Guillong, Szepesi, Szymanowski, Portnyagin, Józsa, Bachmann, Petrelli, Müller, Schiller, Fodor, Chelle-Michou and Harangi2024).
The Michaľany-Lastovce deposit consists of Sarmatian volcano-sedimentary rock formations with a thickness of up to 30 m. The rock composition includes a variety of intensively argillitized rhyolite tuff and tuffites, tuffitic clays, and bentonites, along with scattered limnosilicite and diatomite (reaching up to 30 cm; Fig. S1). Beneath the deposit, sub-horizontally deposited marly calcareous clays from the Upper Badenian to Lower Sarmatian periods are present in the bedrock. Overlying these deposits is a quaternary eluvial-diluvial clay loam, which was historically used as a raw material for brick-making. The average thickness of the overburden is ~10 m (Harcek, Reference Harcek1963; Horský, Reference Horský1994).
Materials and methods
The samples utilized in the present study were provided by the LB MINERALS SK Ltd Mining Company (samples from quarry and landfill heaps) and the REGOS Ltd Mining Company during its exploration campaign (2022–2024) across a currently mined area in eastern Slovakia (Michaľany-Lastovce deposit). Twenty-two samples were collected from both the exposed surfaces of the deposit and drilled boreholes at various depths (Table 1). Specifically, 17 samples were obtained from six boreholes (VLa-1 to VLa-3, and VLa-5 to VLa-7; Fig. 2c), while three samples were collected from the exposed surface in an open quarry (LA1 to LA3; Fig. 3), and two samples were taken from a landfill heap (LA4 to LA5; Fig. S2 in the Supplementary material).
Codes for the samples collected, the intervals from which they were collected, and macroscopic field descriptions
Table 1. Long description
The table contains 19 rows of data under the headers Code, Interval (m), and Description.
* L A 1, L A 2, L A 3: Surface samples. Collected from an open quarry. L A 3 is top, fine-grained black soil; L A 2 is brownish.
* L A 4, L A 5: Surface samples. Mined bentonite in heaps. L A 4 is gray high quality; L A 5 is gray with ochre fragments.
* V L a dash 1 A: 8 to 9 meters. Light gray with pink to brown tint, fine grained clay, smooth touch, small amount of sandy grains and brown iron oxides.
* V L a dash 1 B: 11 to 12 meters. Mix of altered rocks and clay, gray to yellowish with white spots and brown iron oxides.
* V L a dash 1 C: 16.5 to 17 meters. Mainly altered rocks, crystals altered to gray clay plus brown iron oxides and black grains.
* V L a dash 1 D: 20 to 21 meters. Light gray, fine grained clay with small amount of fine sand.
* V L a dash 1 E: 24.5 to 25 meters. Altered rock plus gray clay, coarse grains altered to white or light gray clay.
* V L a dash 1 F: 26 to 27 meters. Homogeneous, dark gray, fine-grained clay with black spots of organic matter.
* V L a dash 2 A: 3 to 4 meters. Homogeneous, dark gray, fine-grained clay with brownish-yellowish to gray clay.
* V L a dash 2 B: 8 to 9 meters. Lamellae of white clay with brownish-yellowish to gray fine-grained clay plus black minerals.
* V L a dash 2 C: 17 to 18 meters. Dark gray with white spots, fine-grained clay with small organic grains.
* V L a dash 3 A: 9 to 10 meters. Orange-brown, fine-grained sand with gray clay spots.
* V L a dash 3 B: 18 to 19 meters. Same as V L a dash 3 A but with more clay.
* V L a dash 5 B: 16.3 to 17.4 meters. Brown-gray sandy, slightly plastic clay with limnosilicite or kaolin.
* V L a dash 5 E: 22.5 to 24 meters. Light gray-blue bentonite with black rounded grains.
* V L a dash 6 D: 13.8 to 15.5 meters. Light gray, ochreous bentonite clays with variable sand and F e streaks.
* V L a dash 6 H: 26 to 27 meters. Dark-gray to black, weakly plastic to dense clays with organic matter and sand.
* V L a dash 7 A: 3 to 5 meters. Gray-brown dense plastic clays with F e smudges and limnosilicite fragments.
* V L a dash 7 E: 26 to 28 meters. Bluish to light-gray bentonite clay with sand and altered rhyolite tuffs.
Samples collected from the surface (open quarry), and from boreholes.
The raw samples were dried naturally at room temperature for several days, and representative samples were obtained through quartering. Around 25 g of each bulk sample was pulverized using a Fritsch Pulverisette agate mill and prepared for the mineral and chemical composition analysis after sieving to a particle size of <250 μm.
Approximately 100 g of each dried bulk material was dispersed in distilled water by stirring for 24 h at room temperature. Several size fractions were then isolated through wet sieving, specifically >2 mm, 2–0.5 mm, 0.5–0.25 mm, 0.25–0.063 mm, and <0.063 mm. The size fractions in the ranges of 63–2 μm and <2 μm were collected using a sedimentation column in accordance with Stokes’s law. The <0.2 μm fraction was obtained by centrifugation in distilled water using an MPW-340 centrifuge (from Mechanika Precyzyjna) at 3500 rpm (1028 RCF) for 20 min. All separated fractions were weighed after drying at 60°C, the clay fraction samples (<2 and <0.2 μm) were passed through a 250 μm sieve after gently grinding in an agate mortar and subsequently prepared for the various analyses.
In addition, ~50 mL of the solution containing the isolated <0.063 mm fraction was analyzed for particle-size distribution (PSD). This analysis was conducted in water as a dispersant and was repeated three times with a reaction time of 5 s, utilizing a laser particle sizer (Malvern Mastersizer 2000).
X-ray diffraction (XRD)
Analysis by XRD of both oriented and randomly oriented preparations was conducted using a Phillips PW1710 diffractometer equipped with CuKα radiation and a graphite monochromator, operated at 20 mA and 35 kV. The oriented samples were prepared by dispersing ~70 mg of the clay fraction samples (<2 and <0.2 μm) in 1.5 mL of distilled water using an ultrasonic bath and pipetting the suspension onto (2.4 cm2) glass slides and drying overnight at room temperature. The XRD patterns of oriented preparations were recorded (from 4 to 60°2θ with a step size of 0.02°2θ and a counting time of 0.80 s per step) in air-dried (AD) and ethylene glycol (EG) solvated form. The EG solvation was achieved by keeping the samples inside a desiccator containing ethylene glycol overnight at 60°C. Additionally, in order to determine the particle thickness of crystallites and their distributions, the 001 XRD basal reflection of selected oriented EG-solvated samples (<2 μm, Na-saturated) was analyzed using a Bruker D8 Advance diffractometer equipped with CuKα radiation over a range of 2–8°2θ, with a step size of 0.02°2θ and a 5 s per step count time. Subsequently, the mean crystallite thicknesses and their distributions for smectite particles were calculated using the Bertaut-Warren-Averbach (BWA) method (Drits et al., Reference Drits, Eberl and Środoń1998) through the MudMaster program (Eberl et al., Reference Eberl, Drits, Środoń and Nuesch1996).
Quantitative X-ray diffraction (QXRD) was performed on randomly oriented preparations, and the mineral quantities were retrieved using RockJock software (Eberl, Reference Eberl2003). Samples were prepared according to the method reported by Środoń et al. (Reference Środoń, Drits, McCarty, Hsieh and Eberl2001) and modified by Omotoso and Eberl (Reference Omotoso and Eberl2009). The sample was passed through a 250 μm sieve, then 1 g of sample was mixed with 0.250 g of pure corundum (as internal standard) in yttrium-stabilized zirconium oxide grinding cylinders and micronized with 4 mL of denatured ethanol (96%) in a McCrone micronizing mill for 5 min. The mixture was then dried overnight at 60°C. The mixture (sample + corundum) was shaken for 10 min in a plastic vial (25 mL) with three plastic balls (9 mm diameter) using a Retsch MM 200 mill, n-Hexane was added to the mixture (to minimize the preferred orientation) in the ratio of maximum 0.5 mL hexane to 1 g of pure clay, and the vial was shaken for an additional 10 min. The powder was then side-loaded into an XRD holder after passing through a 250 μm sieve and X-rayed from 4 to 65°2θ with a step size of 0.02°2θ and a counting time of 2 s per step.
Furthermore, five samples were selected for the identification test of opaline silica polymorphs based on the morphology and intensity of the (101) XRD reflection. These samples were heated at 1050°C for 24 h (Elzea et al., Reference Elzea, Odom and Miles1994) and subsequently subjected to X-ray analysis using a Bruker D8 Advance diffractometer equipped with CuKa radiation (4 to 65°2θ, step size of 0.02°2θ, and 2 s per step).
Fourier transform infrared spectroscopy (FTIR)
The mid-infrared (MIR) spectra for the clay fraction samples (<0.2 μm) were obtained using a Nicolet iS50 FTIR spectrometer (ThermoFisher Scientific), equipped with a KBr beamsplitter and a DTGS detector. Each sample underwent 64 scans at a resolution of 1 cm–1. The KBr pressed-disk technique was employed for the measurements, and the pellets were heated overnight at 130°C to minimize adsorbed water.
Cation exchange capacity (CEC)
The CEC was determined by the copper(II) triethylenetetramine [Cu(Trien)]2+ method according to Meier and Kahr (Reference Meier and Kahr1999). One liter of the 0.01 M solution of [Cu(Trien)]2+ was prepared by mixing 1.504 mL of triethylenetetramine solution with 1.596 g of CuSO4. Approximately 150 mg of sample dried overnight at 105°C was mixed with 50 mL of distilled water and dispersed ultrasonically for ~3 min. The suspension was then shaken for 1 h after adding 10 mL of [Cu(Trien)]2+ and then centrifuged at 4500 rpm (1698 RCF) for 20 min using a MPW-340 centrifuge (from Mechanika Precyzyjna). The concentration of Cu2+ ions in the supernatant was determined by ultraviolet-visible (UV-vis) spectrophotometry (Evolution 220, ThermoScientific) at 578 nm after passing the supernatant through a syringe filter with 0.45 μm pores (cellulose acetate, from AZ Chrom). The CEC values were calculated following Pentrák et al. (Reference Pentrák, Czímerová, Madejová and Komadel2012). Moreover, the CEC values of four samples (VLa-1A, 1E, VLa-2B, and 2C) were determined at Salamanca University, Spain, using the ammonium acetate solutions method (Vicente et al., Reference Vicente, Suárez, López-González and Bañares-Muñoz1996).
Thermal analyses (TA)
Thermogravimetric (TG), derivative thermogravimetric (DTG), and differential thermal analysis (DTA) of the clay fraction samples (<0.2 μm) were performed using a Netzsch STA 449 F3 Jupiter analyzer in the temperature range 20–1000°C by analyzing 40 mg of sample at a heating rate of 10 K min–1 and at a N2 flow rate of 50 mL min–1. The samples were stored at ambient laboratory conditions and measured without pre-treatment.
Specific surface area (SSA)
The textural properties of the samples (bulk and clay fractions (<0.2 μm)) were determined by physical adsorption of nitrogen at the temperature of liquid nitrogen (–196°C) using an ASAP 2400 instrument (Micromeritics). Prior to measurements, samples were de-gassed overnight at 110°C and in a vacuum of 2 Pa. The specific surface area (S BET) was calculated by treatment of adsorption data by BET-isotherm in a standard range of P/P 0 = 0.05–0.30. The specific volume of micropores (V micro) and the specific surface areas of mesopores + external surface (S t) were calculated by the t-plot method using Harkins-Jura standard isotherm. Total pore volume (V a) was estimated from data of the adsorption isotherm at relative pressure of P/P 0 = 0.99. The pore-size distribution of mesopores (D p) was determined from the desorption part of isotherm by the standard Barrett-Joyner-Halenda (BJH) method (Barrett et al, Reference Barrett, Joyner and Halenda1951).
Chemical analysis
The pulverized natural bulk samples (n = 19) were analyzed at Bureau Veritas Commodities Canada Ltd (Vancouver, Canada). Inductively coupled plasma emission spectroscopy (ICP-ES) was used to analyze the major and minor elements, while ICP-MS (ICP-mass spectrometry) was used for the trace and rare earth elements (REE) analyses (using the LF200 (lithium borate fusion, major and minor elements by ICP-ES, trace elements by ICP-MS) and AQ200 (modified aqua regia digestion (1:1:1 HNO3:HCl:H2O), ICP-ES/MS) code methods). The detection limits ranged from 0.002 to 0.04 wt.% (0.01–20 ppm, and 0.01–0.3 ppm) for major elements, trace elements, and REE, respectively.
Scanning electron microscopy (SEM) and electron probe microanalysis (EPMA)
The natural bulk samples (coated with gold film) were examined using a JEOL JSM-6390LV scanning electron microscope (SEM), operated at 15 kV with an energy-dispersive X-ray (EDX/EDS) detector (Oxford Instruments INCA x-act) for elemental microanalysis (at the Earth Science Institute of the Slovak Academy of Sciences - Banská Bystrica, Slovakia). Siliceous sinter fragments were examined using a JEOL JSM-IT210 scanning electron microscope at low vacuum (at Faculty of Natural Sciences - Comenius University in Bratislava).
The point chemical compositions of the natural bentonite polished section samples were obtained by EPMA using a JEOL Super-probe JXA-8530F instrument equipped with wavelength-dispersive X-ray spectroscopy (WDXS/WDS) (at the Earth Science Institute of the Slovak Academy of Sciences - Banská Bystrica, Slovakia), with detection limits within 0.01–0.05 wt.% of oxide. The voltage acceleration was 15 kV, with a beam current and focus of 20 nA and 3–5 μm, respectively. The analysis results of the selected points were utilized to determine the structural formula of smectites, in accordance with the method outlined by Číčel et al. (Reference Číčel, Novák and Horváth1981) and Kelley (Reference Kelley1955). The general composition M(Si4-y Al y)(Al3+, Fe3+, Mg2+)2O10(OH)2 was adopted for montmorillonite.
X-ray micro-computed tomography (μCT)
The natural bulk samples were trimmed roughly into cuboid or cylindrical shapes (~1.5 cm) and analyzed using a high-resolution μCT v|tome|x L 240 scanner (at the Earth Science Institute, Slovak Academy of Sciences - Banská Bystrica, Slovakia). The scanner was equipped with a 240 kV open unipolar X-ray tube, a 180 kV high-power nano focus tube, and maximal resolution of <1 μm. The resulting data were processed and visualized using VGStudio MAX 2.2 software. The visualized sections were used to determine particle roundness, shape, and size (according to Powers, Reference Powers1953).
Results
The samples from the Michaľany-Lastovce deposit were plastic materials which exhibit a range of colors, including shades of gray (grayish-white, grayish-yellow, and off-white). The deeper bentonitic layers in the formation tend to have a grayish-blue appearance. These materials were intercalated with volcano-sedimentary formations. The samples collected from six boreholes exhibited vertical variations in texture, lithology, mineral composition, and slightly in smectite content.
Particle-size distribution (PSD) and texture
The results of the particle-size distribution analysis (wet sieving-based) indicated that the 2–63 μm and <2 μm size fractions accounted for 32.5–78.2 wt.% and 11.8–54.2 wt.% of the total weight of the samples, respectively (Fig. 4). These findings were consistent with those obtained from X-ray micro-computed tomography (μCT) where the samples mainly seem to be dominated by fine-size grains (Fig. 5). The minor presence of the sand-fraction particles, however, was used for their characterization. Most of particles analyzed were phenocrysts with medium sand fractions <0.4 mm. Lithic fragments were found very rarely. Low-sphericity, elongate particles were prevalent. The particle roundness of the analyzed samples was angular or subangular (Fig. 5; Table S3, Supplementary material). The significant angularity of particles suggested that they did not encounter substantial transport. Despite the intense alteration of the original material, several features such as relicts of volcanic shards, spherulites, and flow textures have been identified through microscopic studies that confirm the volcanic origin of the parent rock.
Particle-size distribution achieved by wet sieving analysis (wt.%) of the bulk samples.
Figure 4. Long description
A grouped bar chart with a logarithmic Y-axis labeled percent fraction weight ranging from 0.0 to 100.0. The X-axis is labeled Size fraction and is divided into six vertical panels representing different size ranges.
From left to right, the size fractions are:
1. Greater than 2 m m: Most samples are below 10 percent, with several falling below the 1.0 percent baseline.
2. 2 to 0.5 m m: Distribution is varied, with V L a-1 C and V L a-1 E showing peaks around 10 percent, while V L a-3 A and V L a-3 B drop significantly below 0.1 percent.
3. 0.5 to 0.25 m m: V L a-1 B shows a prominent peak above 10 percent. Other samples fluctuate between 1.0 and 10 percent.
4. 0.25 to 0.063 m m: A general upward trend where most samples cluster between 10 and 40 percent. V L a-1 D is the highest in this category.
5. 2 to 63 micrometers: This category contains the highest concentrations for nearly all samples, with most bars reaching between 40 and 80 percent. L A 1, L A 2, and V L a-1 B are among the highest.
6. Less than 2 micrometers: Samples remain high but generally lower than the previous category, ranging between 10 and 50 percent.
A legend in the bottom right identifies 16 samples: L A 1 through L A 5, V L a-1 A through V L a-1 F, V L a-2 A through V L a-2 C, and V L a-3 A through V L a-3 B.
Variations in texture across different samples: three types of bentonites (LA1, VLa-2B, VLa-1F), and one clay sample (VLa-3A) studied by μCT imaging. The dark color indicates the presence of voids, grey denotes the fine bentonite/clay matrix, bright grey highlights the grains of feldspar and quartz crystals, and the brighter colors are attributed to minerals containing heavier elements (e.g. Fe-biotite, pyrite).
Figure 5. Long description
A four-panel mu C T scan display. Each panel shows a cross-section of a geological sample against a black background.
* Top-left panel: Sample L A 1. A large, irregular dark gray mass with a 2.5 millimeter scale bar. It features a dense distribution of small black voids and scattered bright white mineral grains. A faint diagonal vein of lighter gray material runs from the top-right toward the center.
* Top-right panel: Sample V L a dash 1 F. A vertical rectangular section with a 2 millimeter scale bar. It shows a distinct layered or foliated texture with horizontal bands of gray and dark gray. A prominent dark crack cuts diagonally from the bottom-right edge toward the center.
* Bottom-left panel: Sample V L a dash 2 B. An irregular block with a 2 millimeter scale bar. This sample is highly fractured with numerous dark, jagged cracks throughout. It contains a high concentration of large, bright white and light gray mineral inclusions (feldspar, quartz, or pyrite) embedded in the darker matrix.
* Bottom-right panel: Sample V L a dash 3 A. A vertical section with a 2 millimeter scale bar. This sample exhibits the most uniform texture, appearing as a consistent medium-grey fine-grained matrix with very few visible voids or bright mineral inclusions.
Technical metadata in the corners of each panel includes scene coordinate system values such as 1.82 millimeters, minus 0.23 millimeters, minus 4.02 millimeters, and minus 1.98 millimeters, along with zoom percentages ranging from 70 percent to 263 percent.
The PSD analyzed using a laser-based diffraction technique (Mastersizer) for the <63 μm and <2 μm fractions (suspended in distilled water after wet sieving and without pre-treatment) showed that these size fractions consisted predominantly of particle populations of ~10 μm and 0.2 μm, respectively (Fig. 6).
Particle-size distribution of selected samples: (a) <63 μm fraction; (b) <2 μm fraction, showing the particle populations of the Michaľany-Lastovce samples and of JP3 (from Jelšový potok).
Figure 6. Long description
Two line graphs, labeled a and b, plot Volume Density in percent on the Y-axis against Particle size in mu m on a logarithmic X-axis ranging from 0.01 to 1000.
Panel a, labeled less than 63 micrometers, contains 11 solid lines and one dotted line for J P 3. Most samples show a primary peak between 4 and 10 micrometers, reaching volume densities between 5 and 8 percent. A secondary, smaller peak is visible around 0.8 micrometers. The dotted line for J P 3 is shifted to the right, with a broad peak centered around 60 micrometers.
Panel b, labeled less than 2 micrometers, contains 10 solid lines. The distribution is shifted toward smaller particles compared to panel a. The primary peaks are concentrated between 0.1 and 0.3 micrometers, with volume densities ranging from 5 to 8 percent. Most curves show a gradual decline with smaller secondary humps between 1 and 10 micrometers before dropping to zero.
Both panels include a legend in the top-left or top-right corner listing samples: L A 4, V L a-1 A, V L a-1 B, V L a-1 D, V L a-2 A, V L a-2 B, V L a-5 E, V L a-6 D, V L a-6 H, and V L a-7 A.
Mineral composition
The samples analyzed exhibited slightly varying amounts of dioctahedral smectite, typically accompanied by a variety of mineral admixtures, including opal-C and/or opal-CT (opal-C/CT), feldspars (both K-feldspar and plagioclase), quartz, micas, kaolinite, clinoptilolite, and calcite ± dolomite. Some feldspars and biotite underwent local partial alteration into smectite through dissolution–precipitation processes (Fig. S4, Supplementary material). Four samples (LA3, VLa-2A, 5B, and 7A) exhibit larger quartz contents (9.4–30.2 wt.%) compared with the remaining studied samples (Table 2). With the exception of sample LA3, all samples contained thin fragments (up to 5 mm thick) of siliceous sinter (limnosilicite; Table 1, Fig. S5, Supplementary material). Quartz was confirmed as the dominant mineral phase by XRD analysis, whereas opal-C/CT and smectite were identified as minor phases. SEM revealed the presence of very fine-grained siliceous material containing organic relics within these fragments (Fig. S5). In addition, some samples contained iron oxides, pyrite, and organic matter (Table 2). The presence of smectite and bladed lepispheres of opal-C/CT of the same size is distinctly observable in the electron microscopy images (Fig. 7). The samples obtained from borehole VLa-3 (VLa-3A and 3B) consist predominantly of fine-grained sandstone, chiefly composed of quartz and carbonate minerals. Therefore, these samples will not be described further or discussed.
Mineralogical compositions (wt.%) of the bentonites analyzed, determined using RockJock
Table 2. Long description
The table lists mineralogical data for samples categorized by borehole location: Surface, V L a-1, V L a -2, V L a-5, V L a-6, V L a-7, and V L a-3.
Columns include: Borehole, Sample, Q z (quartz), K f s (K-feldspar), P l (plagioclase), C a l (calcite), G t h (goethite), P y (pyrite), C p t (clinoptilolite), K l n (kaolinite), S m e (smectite), Opal-C/CT, B t (biotite), M s (muscovite), and O M (organic matter) or Dolomite.
Key data points:
* Surface samples (L A 1 to L A 5) show high S m e (32.0 to 40.7) and Opal-C/CT (1.0 to 46.3).
* V L a-1 samples (A to F) show S m e ranging from 23.3 to 32.6 and Opal-C/CT from 25.7 to 45.6.
* V L a-2 samples show high Q z in sample A (28.7) and high Opal-C/CT in sample B (44.0).
* V L a-3 samples (A and B) are distinct for containing Dolomite (8.9 and 8.4) and higher Q z (36.8 and 48.5) compared to other boreholes.
* S m e (smectite) is a major component across most samples, typically exceeding 24 percent except in V L a- 3.
* Trace minerals like G t h and P y are frequently less than 1 percent or absent.
Qz = quartz, Kfs = K-feldspar, Pl = plagioclase, Cal = calcite, Gth = goethite, Py = pyrite, Cpt = clinoptilolite, Kln = kaolinite, Sme = smectite, Bt = biotite, Ms = muscovite, OM = amorphous organic matter. Minerals abbreviations according to Warr (Reference Warr2021).
Smectite and opal-C/CT (bladed lepispheres) of the samples studied by SEM.
Figure 7. Long description
A collection of six S E M micrographs arranged in two columns.
Top-left panel L A 2 at 3,000 times magnification with a 5 micrometer scale bar. Red labels Sme point to flaky, honeycomb-like smectite textures. A red arrow points to Opal-C or C T bladed lepispheres.
Top-right panel L A 3 at 8,000 times magnification with a 2 micrometer scale bar. A red label Sme identifies a dense, crinkled sheet-like smectite surface.
Middle-left panel L A 4 at 4,000 times magnification with a 5 micrometer scale bar. The label Opal-C or C T identifies a cluster of bladed, spherical lepispheres resting on a bed of flaky smectite labeled Sme.
Middle-right panel L A 5 at 1,200 times magnification with a 10 micrometer scale bar. Red arrows point to rows of Opal-C or C T lepispheres lining a cavity wall, with Sme labeled on the surrounding matrix.
Bottom-left panel V L a-2 B at 4,300 times magnification with a 5 micrometer scale bar. Red labels identify Opal-C or C T lepispheres embedded within a dense, wavy smectite matrix labeled Sme.
Bottom-right panel V L a-2 B at 4,300 times magnification with a 5 micrometer scale bar. Red arrows point to individual bladed lepispheres of Opal-C or C T scattered across a highly textured smectite surface labeled Sme.
Certain samples (VLa-1E, VLa-1F) contained trace amounts of pyrite; however, this mineral phase was identified by electron microscopy through its recognizable framboidal morphology (Fig. 8). Consequently, a sufficient amount of these framboidal-like crystals was isolated for sulfur isotope analysis (δ34S); the analysis methodology is provided in the Supplementary material. The δ34S values exhibited variability, ranging from –20.22 to 2.35‰.
BSE images of framboidal pyrite identified in samples VLa-1E (a) and VLa-1F (b).
Figure 8. Long description
A two-panel set of B S E micrographs.
Panel a, located on the left, shows a cluster of bright white spherical framboids and granular aggregates labeled plus P y. These are situated within a dark gray matrix containing large angular mineral grains labeled plus K f s on the left side. A prominent dark fracture runs diagonally through the center. A scale bar at the bottom right indicates 10 micrometers at 330 times magnification.
Panel b, located on the right, displays a higher magnification view of a large, irregular cluster of bright white micro-crystals labeled plus P y. The cluster is surrounded by a mottled grey mineral matrix. An angular grain labeled plus K f s is visible in the top right corner. A scale bar at the bottom right indicates 10 micrometers at 650 times magnification.
Both panels include technical metadata at the bottom including 15.0 k V C O M P O and specific sample identifiers V L a dash 1 E and V L a dash 1 F.
Dioctahedral smectite (montmorillonite) was identified as the predominant clay mineral phase in all samples, with the exception of those from the borehole (VLa-3), as confirmed by XRD (Table 2; Fig. 9a). The (060) reflection was observed at 1.49 Å in randomly oriented preparations, while the basal reflection (001) for oriented clay fraction samples was observed at 14.7 and 15.05 Å for air-dried (AD) samples, shifting to 16.9 and 17.02 Å after EG solvation for the <2 and <0.2 μm fraction samples, respectively (Fig. 9b,c). Note that montmorillonite was the only swelling clay mineral detected in all analyzed samples, as demonstrated by the comparison of AD and EG oriented XRD patterns. Furthermore, the reflection observed around 4 Å in the XRD oriented pattern (Fig. 9b,c) indicates that it was not possible to remove the opaline silica phases (opal-C/CT) from smectite in the bentonite samples from this deposit, even within the <0.2 μm fraction (a similar observation was reported by Gread et al. (Reference Gread, Kureková, Santamaría-López, Osacký and Uhlík2025) in samples from the Lutila-I bentonite deposit in the CSVF area). The XRD results for the (101) reflection at 4.04–4.1 Å indicated that the opaline silica polymorphs in all samples were opal-CT (cristobalite-like structure with varying degrees of stacking disorder leading to the appearance of tridymite reflections (Jones and Segnit, Reference Jones and Segnit1971; Elzea and Rice, Reference Elzea and Rice1996)), with the exception of sample VLa-1C (Fig. 10), which contained opal-C (almost identical to that of α-cristobalite and more ordered than opal-CT (Jones and Segnit, Reference Jones and Segnit1971; Elzea and Rice, Reference Elzea and Rice1996)) which was identified by a relatively intense (101) peak at 4.04 Å. These XRD reflections appeared broad with low intensity in the natural samples; following heating to 1050°C, they became sharper and more intense, a change attributed to the transformation into well-ordered cristobalite (Fig. 10; Elzea et al., Reference Elzea, Odom and Miles1994; Uhlík et al., Reference Uhlík, Jánošík, Kraus, Pentrák and Čaplovičová2012). The characteristic bladed lepishere of the opal-CT, with diameters of a few micrometers, was identified by SEM (Fig. 7; Lynne et al., Reference Lynne, Campbell, James, Browne and Moore2007; Fröhlich, Reference Fröhlich2020). The large silica content caused the lepispheres to coalesce into larger aggregates without specific morphology (Fig. 7).
XRD, randomly oriented (a); and selected oriented <2 μm samples (b); and <0.2 μm (c), offset for clarity. Sme = smectite; Qz = quartz; Cpt = clinoptilolite; Kln = kaolinite; Kfs = K-feldspars; Pl = plagioclase; Cal = calcite; Dol = dolomite; Py = pyrite; Std = internal standard (corundum). Mineral abbreviations according to Warr (Reference Warr2021).
Figure 9. Long description
A multi-panel figure containing three X-ray diffraction (X R D) plots labeled a, b, and c.
Panel a: A large plot with the X-axis showing Diffraction Angle from 5 to 65 degrees 2-theta and the Y-axis showing Intensity in counts. It contains 22 stacked spectra labeled a through v. A legend at the top right identifies sample codes from L A 1 to V L a-7 E. Numerous vertical peaks are labeled with mineral abbreviations and Angstrom values. The most prominent peak is Q z at 3.36 Angstroms near 26.6 degrees. Other labeled peaks include Sme (001) at 15.05 Angstroms, mica/illite at 10.04 Angstroms, C p t at 9.03 and 7.91 Angstroms, K l n at 7.149 Angstroms, P l at 5.60 Angstroms, Sme at 4.49 Angstroms, K f s at 4.28 Angstroms, Opal-C T at 4.06 Angstroms, and several S t d (corundum) peaks.
Panel b: A smaller plot for oriented samples less than 2 micrometers. The X-axis ranges from 2.0 to 36.0 degrees. It shows four groups of paired spectra (A D and E G treatments) labeled a, b, c, and d. Key peaks include a large Sme (001) peak shifting from 14.7 Angstroms (A D) to 16.9 Angstroms (E G), Mica at 10 Angstroms, K l n at 7.22 Angstroms, and Opal-C T at 4.09 Angstroms.
Panel c: A plot for oriented samples less than 0.2 micrometers, similar in scale to panel b. It shows the same four sample groups. The Sme (001) peaks are more pronounced, shifting from 15.05 Angstroms to 17.02 Angstroms upon E G treatment. Other labeled features include Sme, Opal-C T at 4.07 Angstroms, and Q z.
XRD pattern of the samples prior to heating (black continuous line) and after heating at 1050°C (red dotted line) demonstrates the increase in the 4 Å peak intensity following heating.
Figure 10. Long description
The graph consists of five vertically stacked panels sharing a common X-axis representing Diffraction angle in degrees 2-theta, ranging from 15 to 25. The Y-axis represents Intensity in counts. Each panel contains two data lines: a black continuous line for the sample prior to heating and a red dotted line for the sample after heating at 1050 degrees Celsius.
* Top Panel (L A 5): The black line is relatively flat with minor fluctuations. The red dotted line shows a sharp, high-intensity peak at approximately 21.8 degrees 2-theta, labeled cristobalite and 4.09 Angstroms.
* Second Panel (V L a dash 1 B): Similar to the first, the black line shows low-intensity broad peaks, while the red dotted line exhibits a dominant sharp peak at 21.8 degrees 2-theta.
* Third Panel (V L a dash 1 C): The black line features a distinct sharp peak at approximately 22 degrees 2-theta labeled 4.04 Angstroms. The red dotted line shows a much taller peak shifted slightly left to 21.8 degrees 2-theta, labeled 4.09 Angstroms.
* Fourth Panel (V L a dash 1 F): The black line shows a very broad, low-intensity hump around 21.8 degrees. The red dotted line displays a sharp, high-intensity peak at the same position.
* Bottom Panel (V L a dash 2 B): The black line is nearly baseline. The red dotted line shows a sharp, high-intensity peak at 21.8 degrees 2-theta.
Across all panels, heating consistently results in the emergence or significant intensification of a peak at 21.8 degrees 2-theta, corresponding to the 4.09 Angstrom d-spacing of cristobalite.
Crystallite mean thickness and distribution
The mean crystallite thickness (T mean) and crystallite thickness distributions (CTDs) of the Na-saturated and natural clay-fraction samples (<2 μm) indicated that the calculated T mean values for the bentonites analyzed from this deposit were heterogeneous (Table 3; Fig. 11). The samples analyzed can be categorized into two groups based on T mean values. The first group comprises values of <4 nm, while the second encompasses values >7 nm. The calculated mean thickness of the natural samples yielded nearly identical results to those of the Na-saturated counterparts (Table 3). The CTDs of representative samples from the Michaľany-Lastovce deposit (both natural and Na-saturated forms), along with JP3 (from Jelšový potok, Slovakia) and SWy-3 (Wyoming, USA) indicated that the distribution profile of smectites exhibits predominantly a lognormal shape across most samples (Fig. 11).
Cation exchange capacity (CEC), specific surface area (S BET), mean crystallite thickness (T mean), and sample classification based on DTG analysis (according to Wolters and Emmerich, Reference Wolters and Emmerich2007), of the bulk and clay fractions (<2 and ± <0.2 μm) of bentonites
Table 3. Long description
The table is organized into columns for Sample name, C E C in meq per 100 g, S sub B E T in m eters squared per gram, T sub mean in nm, and Type.
Header Structure:
- C E C is divided into Bulk, <2 micrometers, <0.2 micrometers, and Bulk*.
- S sub B E T is divided into Bulk and <0.2 micrometers.
- T sub mean is divided into <2 micrometers natural, <2 micrometers Na-saturated, and <0.2 micrometers.
Data Rows:
- L A 1: C E C Bulk 54.9, <2 micrometers 74.7.
- L A 2: C E C Bulk 54.3, <2 micrometers 68.3.
- L A 3: C E C Bulk 51.8, <2 micrometers 61.4; T sub mean <2 micrometers natural 3.7, Na-saturated 3.7; Type tv/cv.
- L A 5: C E C Bulk 56.4, <2 micrometers 77.2; T sub mean Na-saturated 7.1.
- V L a-1 A: C E C Bulk 46.1, <2 micrometers 58.5, Bulk* 53.9; S sub B E T Bulk 73.0, <0.2 micrometers 101.0; T sub mean Na-saturated 3.2; Type tv.
- V L a-1 B: C E C Bulk 53.9, <2 micrometers 65.6, <0.2 micrometers 77.0; S sub B E T Bulk 66.3, <0.2 micrometers 73.6; T sub mean <2 micrometers natural 7.0, Na-saturated 7.3; Type cv/tv.
- V L a-1 C: C E C Bulk 44.3, <2 micrometers 59.2; T sub mean <2 micrometers natural 6.9; Type cv/tv.
- V L a-1 D: C E C Bulk 39.3, <2 micrometers 57.4, <0.2 micrometers 70.7; S sub B E T Bulk 65.3, <0.2 micrometers 82.2; T sub mean <2 micrometers natural 7.1, Na-saturated 7.3; Type cv/tv.
- V L a-1 E: C E C Bulk 35.3, <2 micrometers 52.7, Bulk* 61.9; S sub B E T Bulk 50.4, <0.2 micrometers 92.5; T sub mean <2 micrometers natural 7.4; Type cv/tv.
- V L a-1 F: C E C Bulk 35.3, <2 micrometers 45.5; Type tv/cv.
- V L a-2 A: C E C Bulk 46.7, <2 micrometers 64.9, <0.2 micrometers 80.7; S sub B E T Bulk 63.1, <0.2 micrometers 69.0; T sub mean <2 micrometers natural 3.7, Na-saturated 3.7; Type tv/cv.
- V L a-2 B: C E C Bulk 52.4, <2 micrometers 64.3, <0.2 micrometers 75.4, Bulk* 59.0; T sub mean <2 micrometers natural 8.2, Na-saturated 8.7; Type cv/tv.
- V L a-2 C: C E C Bulk 48.2, <2 micrometers 53.5, Bulk* 69.3; T sub mean <2 micrometers natural 7.6; Type cv.
- V L a-5 B: C E C Bulk 44.9, <2 micrometers 75.1.
- V L a-5 E: C E C Bulk 47.5, <2 micrometers 63.5.
- V L a-6 B: C E C Bulk 54.9, <2 micrometers 71.5.
- V L a-6 H: C E C Bulk 58.4, <2 micrometers 70.6.
- V L a-7 E: C E C Bulk 36.9, <2 micrometers 68.3; T sub mean <2 micrometers natural 8.7; Type cv.
- J P 3: C E C Bulk 83.4; S sub B E T Bulk 29.1, <0.2 micrometers 54.1; T sub mean Na-saturated 9.5.
- S W y-3: C E C Bulk 82.4; T sub mean Na-saturated 8.3.
Footnote: Bulk* equals C E C values measured by the ammonium acetate method.
Bulk* = CEC values measured by the ammonium acetate method.
Crystallite thickness distributions (CTDs) of the <2 μm fractions: (a) Na-saturated; and (b) natural.
Figure 11. Long description
A two-panel line graph. Both panels share the same axes. The x-axis represents Thickness in nanometers ranging from 0 to 40. The y-axis represents Frequency ranging from 0 to 0.25.
Panel a, labeled Na-saturated, shows several data series with distinct peaks. Three dashed lines representing V L a-2 A, V L a-1 A, and L A 3 show sharp, narrow peaks between 2 and 5 nanometers, reaching frequencies between 0.18 and 0.23 before dropping rapidly to near zero by 10 n m. Solid and dotted lines representing S W y 3, J P 3, L A 5, V L a-1 B, V L a-1 D, and V L a-2 B show broader distributions. These series peak between 5 and 7 n m with frequencies around 0.15 to 0.22 and exhibit a long tail that gradually approaches zero as thickness increases toward 40 nanometers.
Panel b, labeled natural, shows a more uniform distribution across all samples. Most series, including L A 3, V L a-1 A, V L a-2 A, V L a-1 B, V L a-1 C, V L a-1 D, V L a-1 E, V L a-2 B, and V L a-2 C, exhibit a sharp increase to a peak frequency between 0.15 and 0.21 at approximately 3 to 5 nanometers. Following the peak, all curves show a similar asymptotic decay, converging toward zero frequency as thickness reaches 30 to 40 nanometers.
FTIR spectroscopy
The FTIR spectra of the clay fraction samples analyzed indicate clearly that smectite is the predominant mineral present in all samples (Fig. S6, Supplementary material). The broad vibrational band near 3630 cm-1 corresponds to the stretching vibrations of structural OH groups (νOH) in montmorillonite. The bending vibration bands of structural OH groups in smectites are located in the 940 to 830 cm–1 range and are influenced significantly by the occupancy of the octahedral sheets by various atoms within the smectites (Madejová et al., Reference Madejová, Gates, Petit, Gates, Kloprogge, Madejová and Bergaya2017).
Two bands observed near 920 cm–1 (δ Al-Al-OH) and 880 cm–1 (δ Al-Fe-OH) indicate partial substitution of octahedral Al by Fe in all samples. Additionally, partial substitution of Al by Mg was evident, but this was noted only in the LA3 sample, where a weak band corresponding to (δ Al-Mg-OH) was detected at 830 cm–1. The bands around 3430 cm–1 and 1640 cm–1 could be attributed to the vibrational modes of water molecules that remained bound in the KBr pellet upon heating (Farmer, Reference Farmer and Farmer1974). The most prominent band near 1040 cm–1 is mainly due to the stretching vibrations of Si-O groups in montmorillonite (Madejová et al., Reference Madejová, Komadel and Číčel1992). Some samples, such as VLa-1D, VLa-1C, and VLa-1A, also displayed a distinctive shoulder near 1100 cm–1, which can be attributed to Si-O stretching vibrations of kaolinite, amorphous silica, but probably by opal-C/CT.
The presence of opal-C/CT could also be confirmed by the vibration band at 625 cm-1, accompanied by a band at 796 cm–1 (Wilson, Reference Wilson1994), characteristic of cristobalite. In the VLa-1A and VLa-1F samples, only a negligible band was detected, despite the significant band near 700 cm–1, which suggests that this band arises primarily from amorphous silica or tridymite, rather than cristobalite, in these samples. Additionally, a shoulder at 605 cm-1 was noted for the VLa-1C sample, which may indicate the presence of ferrous oxides, known to exhibit stretching vibration bands in this region. The strong bands observed at ~525 and 470 cm–1 correspond to bending vibrations of Al-O-Si and Si-O-Si, respectively.
In the VLa-1C, VLa-2A, and LA3 samples, a shoulder near 3700 cm–1 was also observed, which correlates with the OH-stretching vibrations of structural OH groups found in kaolinite (Madejová et al., Reference Madejová, Kečkéš, Pálková and Komadel2002). The Si-O stretching vibrations further corroborated the presence of amorphous silica observed at 796 cm–1 in all samples, except for VLa-2A and LA3, where only a weak band was observed. The subtle peak, along with a broad band near 700 cm–1, could be assigned to the Si-O vibrations characteristic of kaolinite.
The spectral bands between 3000 and 2700 cm–1 correspond to the vibrational stretching modes of aliphatic CH groups. The low-intensity absorption band at 2854 cm–1 could be attributed to the stretching vibration of aliphatic CH3 groups, while the band at 2925 cm–1 could be associated with the stretching vibrations of CH2 groups (Iglesias et al., Reference Iglesias, de la Puente, Fuente and Pis1998; Osacký et al., Reference Osacký, Geramian, Ivey, Liu and Etsell2013). These results were consistent with the detection of organic matter by XRD and supported by geochemical analyses as total carbon (Table 4).
Chemical compositions of natural bulk samples (major elements (wt.%), ∑REE (ppm) = sum (La:Lu+Y), ∑LREE = sum (La:Sm), ∑HREE = sum (Eu:Lu), Eu/Eu* = EuN/√(SmN*GdN); N refers to chondrite-normalized values (Anders and Grevesse, Reference Anders and Grevesse1989)
Table 4. Long description
The table details the chemical composition of samples LA1, LA3, LA4, LA5, and various VLa series samples (1A through 7E).
Major oxides (wt.%) include:
* S i O sub 2: Ranges from 57.09 (VLa-1C) to 68.85 (VLa-1F).
* Al sub 2 O sub 3: Ranges from 12.49 (LA5) to 16.39 (VLa-6H).
* Fe sub 2 O sub 3: Ranges from 1.85 (VLa-1D) to 10.50 (VLa-1C).
* M g O: Generally between 0.75 and 1.66.
* C a O: Ranges from 0.95 to 2.68.
* Na sub 2 O: Ranges from 0.33 to 1.54.
* K sub 2 O: Ranges from 0.60 to 2.74.
* Ti O sub 2: Ranges from 0.23 to 0.80.
* P sub 2 O sub 5: Low concentrations from 0.02 to 0.15.
* M n O: Ranges from 0.00 to 0.13.
* Cr sub 2 O sub 3: Trace amounts from 0.00 to 0.02.
Additional measurements include:
* L O I (Loss on Ignition): Ranges from 9.5 to 14.5.
* T O T / C and T O T / S: Total carbon and sulfur values, mostly below 1.0.
* S i O sub 2 / Al sub 2 O sub 3 ratio: Ranges from 3.71 to 5.42.
Rare Earth Element (R E E) data (ppm):
* Sum L R E E: Ranges from 82.8 to 161.4.
* Sum H R E E: Ranges from 12.4 to 23.4.
* Sum H R E E plus Y: Ranges from 29.3 to 56.6.
* Sum R E E: Total values range from 115.6 to 215.6.
* E u / E u star: Chondrite-normalized europium anomaly values ranging from 0.26 to 0.65.
Chemical composition and crystal chemistry
The SiO2 concentration varied between 57.09 and 68.85 wt.%, while Al2O3 ranged from 12.49 to 16.39 wt.%; the average Fe2O3 content was 3.22 wt.% (excluding sample VLa-1C). The average values of MgO, CaO, Na2O, and K2O were 1.11, 1.71, 0.74, and 1.28 wt.%, respectively (Table 4). Loss on ignition (LOI) ranges from 9.50 to 14.50 wt.%. A concentration of total rare earth elements (∑REE), ranging from 115.56 to 215.61 ppm, was observed among the samples analyzed. It is evident that these samples are enriched in total light rare earth elements (∑LREE (La–Sm) = 82.75–161.35 ppm) in comparison with heavy rare earth elements (∑HREE (Eu–Lu) + Y) = 29.30–56.59 ppm), accompanied by negative europium anomalies (Eu/Eu* = 0.26–0.65). Among the samples, VLa-2A, VLa-6H, and VLa-7A displayed the highest concentrations of both ∑REE and ∑LREE, whereas the LA1 and VLa-1F samples exhibited the lowest values (Table 4). The samples analyzed were distinguished by their high concentrations of Ba (166.00–845.00 ppm) and Zr (98.60–318.80 ppm) compared with other trace elements (Table S7, Supplementary material).
Four samples (VLa-1B, VLa-2B, VLa-1E, and VLa-1F) were chosen for analysis to determine the overall structural formula of the smectite from this deposit, utilizing spot analyses through EMPA-WDS of polished natural specimens. Although this method may yield challenging and potentially questionable results due to issues with sample preparation, particle thickness, and the likelihood of obtaining chemical data from beneath the clay particles (Christidis, Reference Christidis2008b), a total of 87 spots were analyzed (Table S8, Supplementary material); of these, only 16 analyses were selected for the calculation of structural formula, and the selection was based on the coefficient value for Si atoms (Si <8). The chemical composition of the selected analyses was relatively homogeneous except for three analyses that have a significantly larger Fe2O3 content (13.7–17%) and total oxides (91–96.8%) than others (Table 5). The assumption was made that the remainder of the total amount of analyzed oxides belongs to water. Although these 16 spots may not represent precisely the overall composition of the entire deposit, the results can be insightful for understanding the anticipated crystal chemistry of these smectite minerals. Crystallochemical formulae derived from EMPA-WDS point analyses were also calculated for smaller subsets of analytical points representing individual samples (VLa-1B and VLa-2B) as well as for more compositionally homogeneous domains within sample VLa-1B, specifically Fe-rich areas and areas characterized by lower Fe contents. The most pronounced compositional differences were observed in the Fe-rich domains of sample VLa-1B; nevertheless, the calculated formula remained consistent with montmorillonite. In addition, the formulae obtained for samples VLa-1B and VLa-2B were found to be very similar (Table 5).
Chemical composition of the major elements (wt.%) of 16 spot analyses, and the calculated structural formula from these points (ΣIV = total tetrahedral, ΣVI = total octahedral, ξT = total charge, ξIV = tetrahedral charge, ξVI = octahedral charge, ξI = interlayer charge, %ξIV = tetrahedral charge percentage of total charge, %ξVI = octahedral charge percentage of total, charge calculated per formula unit (f.u.). Values of Na2O (close to zero) were excluded because the analysis error was (239.4%)
Table 5. Long description
The first table lists 16 spot analyses across samples V L a-1 B and V L a-2 B. Columns include S i O sub 2, A l sub 2 O sub 3, F e sub 2 O sub 3, M g O, C a O, and K sub 2 O. Average weight percentages are S i O sub 2 54.8, A l sub 2 O sub 3 18.4, F e sub 2 O sub 3 6.9, M g O 3.0, C a O 1.4, and K sub 2 O 0.3.
The second table details the structural formula calculations.
* Tetrahedral site IV: S i super 4 plus 3.86, A l super 3 plus 0.14, with a total tetrahedral sum Sigma super I V of 4.00 and tetrahedral charge xi sub I V of minus 0.14.
* Octahedral site V I: A l super 3 plus 1.40, F e super 3 plus 0.36, M g super 2 plus 0.24, with a total octahedral sum Sigma super V I of 2.00 and octahedral charge xi sub V I of minus 0.26.
* Interlayer space: C a super 2 plus 0.10, M g super 2 plus 0.07, K super plus 0.03, with an interlayer charge xi sub I of 0.38 and total charge xi sub T of minus 0.40.
* Charge percentages: Tetrahedral charge percentage percent xi sub I V is 36.07 and octahedral charge percentage percent xi sub V I is 63.93.
The bottom section provides average formula units for different sample groupings, such as the overall average (n = 16): C a sub 0.10 M g sub 0.07 K sub 0.03 (S i sub 3.86 A l sub 0.14) (A l sub 1.40 F e sub 0.36 M g sub 0.24) O sub 10 (O H) sub 2.
Cation exchange capacity
The CEC values for both bulk samples and their corresponding clay subfractions (<2 and <0.2 μm) were measured by the [Cu(Trien)]2+ method (Table 3). The CEC values varied as follows: bulk samples ranged from 35 to 58 meq per 100 g, the <2 μm fractions ranged from 45 to 77 meq per 100 g, and the <0.2 μm subfractions ranged from 70 to 80 meq per 100 g, with an average 48, 64, and 76 meq per 100 g, respectively (Fig. S9, Supplementary material). Compared with the Michaľany-Lastovce samples, the natural bulk samples SWy-3 (Wyoming, USA) and JP3 (Jelšový potok, Slovakia) exhibited considerably larger cation exchange capacities (82.4 and 83.4 meq per 100 g, respectively), attributable to the elevated smectite content (~80 wt.%) in these bentonites. Furthermore, the CEC values determined using the ammonium acetate method were similar for two of the four natural bulk samples, while the remaining two samples exhibited significantly higher values compared with those measured by the [Cu(Trien)]2+ method (Table 3).
Specific surface area (S BET) and pore-size distributions
The specific surface area (S BET) results (Table 3) were measured for selected samples, including both bulk materials and their corresponding clay fractions (<0.2 μm). The S BET values for bulk bentonites ranged from 50.4 to 73 m2 g–1, while the corresponding clay fractions showed values between 69 and 101 m2 g–1. From these data, it is evident that the S BET values increase in relation to the amount of smectite present in the samples analyzed, which aligns with expectations. However, the Jelšový potok (JP3) sample, despite having a much larger smectite content, showed relatively low S BET values for both bulk and clay fractions (29.1–54.1 m2 g–1). This discrepancy may be due to the smaller particle size of the Michaľany-Lastovce smectite (Suárez et al., Reference Suárez, Lorenzo, García-Vicente, Morales, García-Rivas and García-Romero2022).
The N2 adsorption-desorption isotherms of the analyzed bulk and clay fraction (<0.2 μm) samples are shown in Fig. 12. All studied samples could be described by the presence of hysteresis loops which are generally associated with capillary condensation, hysteresis loop type H3 (according to IUPAC classification; Thommes et al., Reference Thommes, Kaneko, Neimark, Olivier, Rodriguez-Reinoso, Rouquerol and Sing2015) typically characterizes the results of the bulk samples, where the adsorption branch resembles the type II nitrogen adsorption isotherm, with the curvature gradually increasing with rising pressure, except for samples VLa-2A and JP3, which was more representative of the type H4 hysteresis loop. Conversely, the clay fraction samples display a predominantly type H2b hysteresis loop, except for the samples VLa-2A and JP3, which resemble more closely the type H4 hysteresis loop.
Adsorption (filled squares)−desorption (open circles) isotherms of bulk and <0.2 μm samples.
Figure 12. Long description
A multi-panel figure with six line graphs. All graphs share an x-axis labeled P divided by P sub 0 ranging from 0 to 1, and a y-axis labeled V sub a in cubic centimeters S T P per gram. Each graph contains two sets of data: bulk samples represented by squares and less than 0.2 mu m samples represented by circles. Filled markers indicate adsorption and open markers indicate desorption. Arrows indicate the direction of the hysteresis loops.
* Top-left panel (V L a dash 1 A): The bulk sample (blue) shows a steady increase, while the less than 0.2 micrometers sample (orange) exhibits a significantly wider hysteresis loop starting at P divided by P sub 0 of 0.4.
* Top-right panel (V L a dash 1 B): Similar to 1 A, the less than 0.2 micrometers sample (orange) shows a sharp desorption drop at 0.8 P divided by P sub 0, while the bulk (gray) has a narrower loop.
* Middle-left panel (V L a dash 1 D): The bulk (dark green) and less than 0.2 micrometers (brown) curves are closely aligned until 0.8 P divided by P sub 0, where the less than 0.2 micrometers sample shows a larger desorption volume.
* Middle-right panel (V L a dash 2 A): The bulk (light green) and less than 0.2 micrometers (orange) samples show a distinct step-like desorption plateau between 0.5 and 0.9 P divided by P sub 0.
* Bottom-left panel (V L a dash 2 B): The bulk (black) and less than 0.2 micrometers (orange) curves cross at 0.4 P divided by P sub 0, with the bulk sample maintaining higher adsorption at high pressures.
* Bottom-right panel (J P 3): The bulk (purple) shows a very steep rise near 1.0 P divided P sub 0, whereas the less than 0.2 micrometer sample (orange) plateaus much earlier with a prominent hysteresis step at 0.5 P divided by P sub 0.
In general, at very low relative pressures (P/P 0 <0.01), the amount of adsorbed nitrogen (N2) was marginally high, indicating the presence of a small amount of micropores (diameter <2 nm) within the samples. Conversely, at high relative pressures (P/P 0 >0.98), these samples were capable of adsorbing a larger volume of N2, and the hysteresis loops observed were broader at the relative pressure (P/P 0 = 0.5–0.95). This, along with the hysteresis loops suggests a predominance of mesopores (2−50 nm pore size). The isotherms also displayed a steep slope rather than a plateau nearing 1.0 relative pressure (P/P 0), which could indicate the presence of macropores (with sizes >200 nm) and a greater overall porosity (refer to Fig. 12). These findings were consistent with the measured values for total pore volume (V a) and micropore volume (V micro).
The BJH pore-size distribution (Fig. 13) was nearly consistent across all samples. Although all three categories of pores were present, the most prominent characteristic was the predominance of mesoporosity (pore sizes ranging from 2 to 50 nm) in all bulk and clay fraction samples. Notably, the pore diameter was mostly <10 nm in all samples. Furthermore, the sharp spikes observed around 4 nm in the pore-size distributions for both the bulk and clay fractions of the four samples (VLa-1A, VLa-2A, VLa-2B, and JP3) are probably artifacts, as suggested by Groen et al. (Reference Groen, Peffer and Pérez-Ramı́rez2003), and do not accurately represent the true porous properties of the samples analyzed.
Pore-size distribution of the bulk (a), and <0.2 μm (b) samples obtained by treatment of desorption branches of the isotherms.
Figure 13. Long description
A two-panel line graph displaying pore volume in cubic centimeters per gram on the y-axis against pore diameter in nanometers on a logarithmic x-axis ranging from 1 to 300.
Panel a, labeled bulk, shows several samples with a primary sharp peak between 3 and 5 nanometers. Sample J P 3 bulk reaches the highest peak at approximately 0.85 cubic centimeters per gram. Most other samples show a secondary, broader distribution between 10 and 100 nanometers with lower intensities around 0.1 to 0.2 cubic centimeters per gram. The legend includes V L a 1 A bulk, V L a 1 B bulk, V L a 1 D bulk, V L a 1 E bulk, V L a 2 A bulk, V L a 2 B bulk, and J P 3 bulk.
Panel b, labeled less than 0.2 micrometers, shows a significant shift in intensity. The y-axis scale is higher, reaching 1.8. Most samples exhibit two distinct sharp peaks. The first peak is between 3 and 5 nanometers, and a second, more prominent peak occurs between 8 and 10 nanometers. Sample V L a 2 A less than 0.2 micrometers shows the highest peak at 1.7 cubic centimeters per gram near 4 nanometers. The legend lists the same sample names as panel a but with the less than 0.2 micrometers suffix.
Thermal analysis
The mass loss observed in the TG curves can be categorized into four distinct regions (Fig. 14): a dehydration (<300°C), two dehydroxylation events (500–550°C and 650–700°C), and a decomposition/recrystallization phase (>800°C). The first dehydroxylation event is linked to trans-vacant (tv) types of smectite, while the second is associated with cis-vacant (cv) varieties of smectite (Drits et al., Reference Drits, Besson and Muller1995; Wolters and Emmerich, Reference Wolters and Emmerich2007).
Thermal analysis (TGA, DTA, and DTG) results of the clay fraction (<0.2 μm) samples.
Figure 14. Long description
A multi-panel figure containing 10 line graphs arranged in two columns of five. Each graph shares a common X-axis representing Temperature in degrees Celsius from 0 to 1000. The Y-axes include D T A in microvolts per milligram on the left, Mass percentage in red on the right, and D T G in percentage per minute on the far right.
Each panel represents a different sample: V L a-1 A, V L a-1 F, V L a-1 B, V L a-2 A, V L a-1 C, V L a-2 B, V L a-1 D, V L a-2 C, V L a-1 E, and L A 3, all for the less than 0.2 micrometer fraction.
In every graph, three curves are present:
1. A black D T A curve showing endothermic troughs, most notably a sharp deep trough between 100 and 200 degrees and a second significant trough around 500 to 700 degrees.
2. A red Mass percentage curve showing a stepped decline. Vertical dashed arrows indicate mass change percentages. The first major drop occurs before 200 degrees (ranging from -5.66 percent to -12.56 percent), followed by a second drop between 400 and 700 degrees (ranging from -3.44 percent to -5.74 percent).
3. A blue D T G curve mirroring the rate of mass loss with downward peaks corresponding to the steepest slopes of the red mass curve.
An ‘ex’ arrow in the top right of each panel points upward, indicating the direction of exothermic reactions for the D T A signal.
At temperatures below 225°C, the TG curves indicated significant mass loss (see Fig. 14), attributed primarily to the release of surface-absorbed water and the dehydration of exchangeable cations in the interlayer space of swellable clay minerals (Wolters and Emmerich, Reference Wolters and Emmerich2007; Derkowski and Kuligiewicz, Reference Derkowski and Kuligiewicz2022). The first step of mass loss was between 5.66 and 10% for all samples; however, this might not reflect the actual hydration ability of the samples because the samples were in the natural form and were not undergoing any special treatment.
The exothermic peak in the DTA curve with maxima near 350°C for the samples LA3 and VLa-1E (these samples are distinguished by a darker color) is probably related to the presence of organic matter in these samples (Wolters and Emmerich, Reference Wolters and Emmerich2007), which was confirmed by FTIR and XRD and was recognizable in the chemical data as total carbon (TOT/C) for the bulk samples (Table 2; Table 4; Fig. S6). The DTA curves of the samples examined exhibited a prominent endothermic event at ~280°C for sample VLa-1C, while a less intense event was noted for samples VLa-1A, VLa-1B, VLa-1F, and VLa-2C (see Fig. 14). This endothermic event is probably related to the oxidation of Fe oxides, which was identified through XRD and FTIR analyses (VLa-1C contains ~5 wt.% Fe oxides; Table 2; Fig. S6). Most of the samples analyzed, except for sample LA3, exhibited a pronounced bi-modal peak in the DTG and DTA curves, signifying the presence of both trans-vacant octahedral sheet (tv, 500–550°C) and cis-vacant (cv, 600–700°C) types of dioctahedral smectites (Drits et al., Reference Drits, Besson and Muller1995; Wolters and Emmerich, Reference Wolters and Emmerich2007; Derkowski and Kuligiewicz, Reference Derkowski and Kuligiewicz2022).
Furthermore, the presence of both endothermic and exothermic events at temperatures exceeding 880–900°C in all samples could be attributed to the formation of new amorphous phases following the breakdown of the anhydrous/dehydroxylated smectite structure and the subsequent crystallization of new phases (Bradley and Grim, Reference Bradley and Grim1951; Wolters and Emmerich, Reference Wolters and Emmerich2007).
Discussion
Characterization of the Michaľany-Lastovce bentonite
Smectite and opal-C/CT were recognized as the primary components in most of the analyzed samples, averaging 32 wt.% for both, with ranges from 23 to 40 wt.% for smectite and 25 to 50 wt.% for opal-C/CT, reflecting an almost 1:1 ratio between these two phases, with the exception of three samples (LA3, VLa-2A, and VLa-7A), which contained significantly less opal-C/CT (<5.5 wt.%) and more quartz (>25 wt.%) than the majority of the studied samples (Table 2). Based on the heating tests of selected bentonites in this study, the dominance of opal-CT has been confirmed as a low-temperature opaline polymorph (Fig. 10). Cristobalite, classified as a high-temperature opaline phase, typically linked to volcanic rocks, was not observed. The LA3 sample was collected from the uppermost section of the open quarry (Fig. 3); this sample displayed a rich black color, a fine-grained texture, and contained the smallest amount of opal-C/CT, along with 10 wt.% of amorphous organic phases. Consequently, it is believed to represent the secondary reformed bentonite that was formed as a soil layer by the weathering of the primary bentonite in the surrounding area. Therefore, sample LA3 was excluded from further discussion within this specific subset of samples, whereas sample VLa-5B was included due to its comparatively elevated quartz content, despite exhibiting an opal-CT content exceeding the average observed across the studied samples.
An initial explanation proposed for the increased quartz content observed in a limited number of samples involved the contribution of fine-grained sandy to dusty epiclastic material derived from the surrounding lithologies. Under such conditions, an increase in the SiO2/Al2O3 ratio would be expected in the respective samples. However, the SiO2/Al2O3 ratios of these three samples remain within the range defined by the remaining samples (Table 4). Consequently, a more reasonable explanation for the elevated quartz content is the occurrence of thin layers of siliceous sinters (limnosilicites; Table 1, Fig. S5), with SiO2 presumed to originate from bentonitized volcanic glass, i.e. from internal sources within a closed system.
The occurrence of siliceous sinter raises the question of the potential presence of hot springs, which are characteristic of geothermally active environments (e.g. Lynne et al., Reference Lynne, Campbell, James, Browne and Moore2007; Jones, Reference Jones2021). Nevertheless, a purely diagenetic transformation of biogenic siliceous material (e.g. diatoms and sponges) also represents a viable alternative explanation. This process involves a phase transformation from X-ray amorphous opal-A through opal-CT to microcrystalline quartz. The extent and rate of this transformation are controlled not only by temperature and time, but also by the composition of the siliceous sediment and the overall SiO2 content. In sediments containing large proportions of biogenic SiO2 (>85%) and subjected to sufficiently long diagenetic evolution, transformation to microcrystalline quartz may occur even at relatively low temperatures below 50°C (Mizutani, Reference Mizutani1970; Behl and Kassa, Reference Behl and Kassa2026). Given the presently available evidence, both above-mentioned possibilities remain plausible.
According to the calculated formulae (Table 5), the smectite was composed predominantly of Si and Al in the tetrahedral position, and Al, Fe, and Mg in the octahedral position, with Mg, Ca, as major, and K as minor, interlayer cations. Three analyses that have significantly larger amounts of Fe2O3 (Table 5) could suggest some andesitic admixture to the main rhyolitic source for bentonite formation, but admixture mineral phases of Fe oxides in the EMPA cannot be ruled out either. The layer charge arises primarily from the octahedral position where Mg substitutes for Al. This supports the identification of the smectite as primarily montmorillonite (Brindley, Reference Brindley, Brindley and Brown1980; Brindley and Brown, Reference Brindley and Brown1980) and it could be classified as a medium-charged beidellitic montmorillonite as the calculated layer charge was 0.38 per formula unit (f.u.), and 63% of the charge was located in the octahedral position (Emmerich et al., Reference Emmerich, Wolters, Kahr and Lagaly2009).
By comparing the smectite compositions outlined above with those from other Slovak bentonite deposits such as Stará Kremnička-Jelšový potok, and Stará Kremnička-III (Číčel et al., Reference Číčel, Novák and Horváth1974; Osacký et al., Reference Osacký, Binčík, Paľo, Uhlík, Madejová and Czímerová2019), it is evident that both bentonites exhibit nearly comparable medium layer charges (0.38 per f.u. for the current deposit and 0.42 per f.u. for the Stará Kremnička-III deposit). The primary distinction arises from the considerably larger amount of Fe substituting for Al in the octahedral positions of the smectites from Michaľany-Lastovce (0.36), compared with the smaller Fe content found in the Stará Kremnička-III smectites (0.11).
The crystallite thickness of smectites is influenced by several factors, including the type of interlayer cations, size fraction, and sample pretreatments (Mystkowski et al., Reference Mystkowski, Środoń and Elsass2000). A positive correlation between the c-dimension (crystallite thickness) and the lateral dimensions (a–b dimensions) of smectite was discussed by Mystkowski et al. (Reference Mystkowski, Środoń and Elsass2000), who proposed that larger crystallites may indicate larger particles. The JP3 and SWy-3 samples exhibit larger T mean values compared with the Michaľany-Lastovce samples (except for the VLa-2B sample), while both sets maintain a similar lognormal distribution. The smallest T mean values were calculated from samples collected at shallower depths (VLa-1A, VLa-2A, and LA3), whereas the largest T mean values were derived from deeper samples (e.g. VLa-2B, VLa-7E) for both natural and Na-saturated forms (Table 3). The sample LA4 with the larger value was taken from a landfill heap (Fig. S2); therefore, the exact depth of emplacement of this sample is unknown. However, the deepest part of the bentonite quarry is up to ~11 m (Fig. 3). The reduced crystallite thickness of smectite observed in LA3 (which we suggest to have formed by weathering as a secondary bentonite) aligns with the observations of Šucha et al. (Reference Šucha, Środoń, Clauer, Elsass, Eberl, Kraus and Madejová2001), and Stríček et al. (Reference Stríček, Šucha, Uhlík and Madejová2006) who noted that weathered smectites exhibit lower crystallite thickness compared with those not exposed to weathering. Although the depths of samples VLa-1A and VLa-2A are relatively shallow (8–9 m and 3–4 m, respectively), considering that the overlying soil horizon and quaternary sediments extend only up to 2 m, attributing the significantly lower crystal thickness of these two samples to weathering remains questionable. Comparison of mean thickness results with other bentonites from the CSVF, as studied by Osacký et al. (Reference Osacký, Binčík, Paľo, Uhlík, Madejová and Czímerová2019), indicates that most smectites from the CSVF area exhibit larger T mean values than those from the ESNB examined in the present study. This difference may reflect distinct formation processes; specifically, the CSVF bentonites may have formed under hydrothermal conditions (Osacký et al., Reference Osacký, Binčík, Paľo, Uhlík, Madejová and Czímerová2019), which could promote the growth of larger smectite crystallites compared with those formed through diagenetic processes, as in the Michaľany-Lastovce deposit in the ESNB.
A positive correlation was observed between CEC values and smectite content highlighting that the CEC is influenced by smectite content, which is not unexpected. While the clay fraction samples (<0.2 μm) exhibited the largest CEC values (Table 3; Fig. S9), these remain considerably lower than those of the bulk samples from central Slovak bentonites (reported also by Osacký et al., Reference Osacký, Binčík, Paľo, Uhlík, Madejová and Czímerová2019). This discrepancy was attributed to the presence of the non-clay admixtures specifically opal-C/CT in all fractions of samples from the Michaľany-Lastovce deposit, which reduces the smectite content.
For comparison, the CEC values determined by the ammonium acetate method for four bulk samples (Table 3) were nearly identical for VLa-1A and VLa-2B, but substantially greater for VLa-1E and VLa-2C than those determined by [Cu(Trien)]2+. The elevated CEC values in VLa-1E and VLa-2C are probably due to the presence of clinoptilolite (VLa-1E = 24.1 wt.% and VLa-2C = 7.4 wt.%; Table 2), the structural cavities of which are more accessible to small ammonium ions than to the larger Cu2+ complex, as reported by Meier and Kahr (Reference Meier and Kahr1999). The proportions of cations determined after ammonium acetate saturation were compared with the proportions of exchangeable cations determined in the structural formula by EPMA-WDS (Table 5; Table S10, Supplementary material). The observed enrichment of Ca following ammonium acetate treatment may be explained by calcite dissolution (Dohrmann, Reference Dohrmann2006). In contrast, the depletion of Na in the structural formula may be attributed to the EPMA-WDS method’s limitations in measuring lighter elements such as Na (the average error percentage for Na was 239.4% of the selected samples).
The varying intensities of the DTA curves in the analyzed samples probably indicate the relative proportions of tv and cv vacancies (Table 3; Wolters and Emmerich, Reference Wolters and Emmerich2007). The thermal analysis results suggest that the tv varieties of smectite were found predominantly in samples collected from shallower depths, while cv varieties tend to increase with depth, except for sample VLa-1F. At the same time, samples characterized by a higher proportion of the tv varieties exhibit lower T mean values, whereas samples dominated by the cv varieties show greater T mean values of smectite particles (Table 3). Although there is currently no definitive explanation for this trend, it certainly merits further discussion and exploration in future studies.
The Michaľany-Lastovce deposit continues to serve consumers, in spite of the smaller smectite content (~40 wt.%). The positive properties of the deposit include fine-grained raw material, light color, reasonable CEC, and larger SSA values than high-quality bentonite from the Jelšový potok deposit, CSVF (Table 3). The significant difference in SSA is not just the result of thinner smectite crystallites of bentonite with smaller smectite contents. It seems that, in the case of the Michaľany-Lastovce deposit, the presence of a significant amount of opal-CT also plays positive role. Generally, opaline silica is undesirable in bentonite because it decreases swelling and the CEC, and although it has a positive influence on the color (white or light), it increases abrasiveness, thus rendering them unsuitable for certain industries, such as pharmaceutical and cosmetics (Christidis and Scott, Reference Christidis and Scott1996; Christidis and Scott, Reference Christidis and Dunham1997). Individual opal-CT bladed lepispheres in the bentonite studied have diameters of several micrometers (Fig. 7). The blades typically possess a thickness of ~200 nm, with free space between them and a preferred orientation along three crystallographic axes (Lynne et al., Reference Lynne, Campbell, James, Browne and Moore2007; Fröhlich, Reference Fröhlich2020). This morphology could contribute to large SSA values. Kwon et al. (Reference Kwon, Kim and Roh2021) described an opaline mudstone, composed of nearly 90% opal-CT (with minor quartz and feldspar) that exhibited a negative zeta potential of −6.66 mV. The opaline mudstone also had a large specific surface area (S BET) of 67.58 m2/g, similar to that of the Michaľany-Lastovce samples (Table 3). These properties render the opaline mudstone suitable for Cs adsorption (Kwon et al., Reference Kwon, Kim and Roh2021). On the other hand, Milos bentonite containing up to 30 wt.% opaline silica has been described as a suitable material that meets all required bentonite-cement slurry properties. This suitability was attributed to the presence of soda-soluble silica, which dissolves upon contact with alkaline cement pore water and may subsequently react with the cement (Ohrdorf et al., Reference Ohrdorf, Kaufhold, Rüßmann, Ufer and Flachberger2011). Therefore, the Michaľany-Lastovce deposit may also be considered as promising, particularly given the increasing demand for supplementary cementitious materials (Snellings et al., Reference Snellings, Suraneni and Skibsted2023). Its potential as a supplementary cementitious material will be investigated in future studies.
Although several definitions do not specify an exact threshold for the smectite content required to classify a material as bentonite (e.g. Grim and Güven, Reference Grim and Güven1978; Gilg et al., Reference Gilg, Kaufhold, Ufer and Kaufhold2024), we chose to use the terms bentonite-like deposit or bentonitic deposit to emphasize the specific character of the studied deposit. This designation primarily reflects the relatively low montmorillonite content, which is accompanied by an approximately comparable proportion of opal-CT.
Parent materials and genesis of the bentonite
The Michaľany-Lastovce bentonite-like deposit is situated within the Carpathian-Pannonian Region (CPR) of eastern-central Europe. The CPR comprises an inter-arc extensional basin system surrounded by orogenic belts, where Miocene volcanism in the region resulted from the convergent collision between the Carpathian orogeny and the North European platform (e.g. Lexa et al., Reference Lexa, Konečný, Kaličiak, Hojstričová, Rakús and Vozár1993; Handy et al., Reference Handy, Ustaszewski and Kissling2015; Lukács et al., Reference Lukács, Harangi, Guillong, Bachmann, Fodor, Buret, Dunkl, Sliwinski, von Quadt, Peytcheva and Zimmerer2018, Reference Lukács, Guillong, Szepesi, Szymanowski, Portnyagin, Józsa, Bachmann, Petrelli, Müller, Schiller, Fodor, Chelle-Michou and Harangi2024). The Pannonian Basin experienced the largest silicic explosive eruptions in Europe over the past 20 million years, producing >4000 km3 of tephra during the syn-rift phase between 18.1 Ma and 14.4 Ma (Brlek et al., Reference Brlek, Richard Tapster, Schindlbeck-Belo, Gaynor, Kutterolf, Hauff, Georgiev, Trinajstić, Šuica, Brčić, Wang, Lee, Beier, Abersteiner, Mišur, Peytcheva, Kukoč, Németh, Trajanova, Balen, Guillong, Szymanowski and Lukács2023; Lukács et al., Reference Lukács, Guillong, Szepesi, Szymanowski, Portnyagin, Józsa, Bachmann, Petrelli, Müller, Schiller, Fodor, Chelle-Michou and Harangi2024). Subsequently, silicic volcanism progressed eastward, characterized by bi-modal andesitic-rhyolitic activity in the Tokaj and Slánske vrchy Mountains, which also contributed to the infilling of the ESNB (Baňacký et al., Reference Baňacký, Elečko, Kaličiak, Straka, Škvarka, Šucha, Vass, Vozárová and Vozár1989; Kaličiak et al., Reference Kaličiak, Baňacký, Bodnár, Dubéciová, Jacko, Janočko, Jetel, Karoli, Petro, Spišák, Syčev, Zlinská and Žec1996; Subová et al., Reference Subová, Rybár, Šarinová, Hudáčková, Jamrich, Sliva, Šály and Hlavatý2022; Lukács et al., Reference Lukács, Guillong, Szepesi, Szymanowski, Portnyagin, Józsa, Bachmann, Petrelli, Müller, Schiller, Fodor, Chelle-Michou and Harangi2024; Šarinová et al., Reference Šarinová, Rybár, Brčeková, Szymanowski, Fred, Marcel, Kotulová, Ľubomír, Hudáčková and Lukács2026).
Bentonite is formed primarily through the alteration of volcanic glass. One objective of the present study was to determine the parent rock of the bentonitic rocks analyzed by examining its physicochemical properties. Several lines of evidence suggest that the original material was fine juvenile pyroclastic ash. First, the combined silt and clay fractions range from 62 to 85 wt.% (Fig. 4, Fig. 6). Second, there was a predominance of angular and subangular grains (Fig. 5; Table S3), and no non-volcanic rock fragments were identified in the studied samples. Basaltic and andesitic fragments were also absent. Only trace amounts of calcite (<1.3 wt.%) and generally <5% of muscovite or illite were present, which suggests minimal secondary re-deposition or mixing with surrounding rocks (Table 2; Fig. 5). Third, volcanic shards were occasionally but distinctly present (Fig. S11, Supplementary material). Fourth, lithic felsitic fragments and partially altered feldspar and biotite crystals further support a volcanic origin, as they were embedded in the altered groundmass (Fig. S4, Fig. S12, Supplementary material). Fifth, in some samples (Fig. 15), the groundmass retains relicts of rheomorphic flow despite significant alteration. This observation suggests an ignimbrite origin, which refers to a deposit from a pyroclastic flow, either welded or unwelded (Fisher and Schmincke, Reference Fisher and Schmincke1984). Spherulitic aggregates (spherulites) also form during the cooling of silica-rich glassy volcanic bodies (Breitkreuz, Reference Breitkreuz2013; Lexa et al., Reference Lexa, Varga, Uhlik, Koděra, Biroň and Rajnoha2021).
Photomicrograph (PPL images, a–d) and BSE images (e–f) show the preserved rheomorphic flow structure (a–c, e–f) and spherulitic aggregates (d) in the bentonite samples.
Figure 15. Long description
A six-panel grid labeled a through f.
Panel a: P P L photomicrograph of sample V L a-1 B showing a brown, fibrous flow structure. A red rectangular box highlights a central area of dense alignment. Scale bar indicates 2000 micrometers.
Panel b: P P L photomicrograph of sample V L a-1 B with a similar flow structure. A red box on the right side highlights a region of swirling textures. Scale bar indicates 2000 micrometers.
Panel c: P P L photomicrograph of sample V L a-1 B showing a distinct vertical band of dark brown flow structure on the left, highlighted by a red box, adjacent to a lighter area with circular bubbles. Scale bar indicates 2000 micrometers.
Panel d: P P L photomicrograph of sample V L a-1 E showing a spherulitic structure. The center contains dark, radiating circular mineral aggregates. Scale bar indicates 500 micrometers.
Panel e: B S E image of sample V L a-1 B at 160 times magnification. It shows a grayscale view of the flow structure with red numerical markers and mineral labels like plus q z. Scale bar indicates 100 micrometers.
Panel f: B S E image of sample V L a-1 B at 400 times magnification. It displays highly detailed, elongated, and folded flow bands in shades of gray. Red markers and mineral labels are scattered across the texture. Scale bar indicates 10 micrometers.
No non-bentonized volcanic ash has been identified near the studied area, and the precise volcanic source of the Sarmatian ash remains undetermined. Consequently, mass-balance calculations comparing volcanic ash composition before and after bentonitization have not been conducted. Potential sources within a 15–20 km radius include rhyolitic pyroclasts with felsic, spherulitic, and vitric rhyolite extrusions near the villages Viničky and Borsuk in the Slovak Tokaj area, SSE of the Michaľany-Lastovce deposit. The Viničky obsidians have been dated to 11.04–13.52 Ma (Bačo et al., Reference Bačo, Kaminská, Lexa, Pecskay, Bačová and Konečný2017; Kohút et al., Reference Kohút, C. Sherlock and M. Halton2021a; Kohút et al., Reference Kohút, Westgate, Pearce and Bačo2021b), whereas the Hegyköz Unit, a silicic explosive volcaniclastic unit in the northern Hungarian Tokaj Mountains, northeast of the deposit, has been dated to 12.4±0.8 to 11.7±0.5 Ma (Pécskay et al., Reference Pécskay, Balogh, Székyné and Gyarmati1987; Lukács et al., Reference Lukács, Guillong, Szepesi, Szymanowski, Portnyagin, Józsa, Bachmann, Petrelli, Müller, Schiller, Fodor, Chelle-Michou and Harangi2024). Furthermore, three andesitic stratovolcanoes (Bogota, Hradisko, Bradlo) in the Slánske vrchy Mountains exhibited rhyolitic activity during the Sarmatian (Kaličiak et al., Reference Kaličiak, Baňacký, Bodnár, Dubéciová, Jacko, Janočko, Jetel, Karoli, Petro, Spišák, Syčev, Zlinská and Žec1996).
Geochemical analyses of the samples in this study and volcanic rock samples from the Tokaj area (Lukács et al., Reference Lukács, Guillong, Szepesi, Szymanowski, Portnyagin, Józsa, Bachmann, Petrelli, Müller, Schiller, Fodor, Chelle-Michou and Harangi2024), as plotted on the Nb/Y versus Zr/TiO2 discrimination diagram (Fig. 16; Winchester and Floyd, Reference Winchester and Floyd1977), indicate that bentonites possess similar Nb/Y values. Most bentonitic samples exhibit rhyodacitic to dacitic affinities; however, certain samples, such as VLa-1C with elevated iron oxide content (Table 4; Fig. S6), display more andesitic characteristics in contrast to the predominantly rhyolitic Tokaj volcanic rocks (Fig. 16). Chondrite-normalized rare earth element (REE) patterns (Fig. 17; Anders and Grevesse, Reference Anders and Grevesse1989) demonstrate enrichment in light rare earth elements (∑LREE) relative to heavy rare earth elements (∑HREE), accompanied by negative europium anomalies (Eu/Eu* = 0.26–0.65). The total REE content (∑REE) in the Michaľany-Lastovce deposit falls between published values for rhyolitic and average crustal rocks (Table S13, Supplementary material) and remains below the threshold for economic potential (Van Gosen et al., Reference Van Gosen, Verplanck, Seal, Long and Gambogi2017; Berti et al., Reference Berti, Slowey, Yancey and Deng2022; Berti et al., Reference Berti, Slowey, Deng, Yancey and Velazquez2023). The presence of at least some ∑REE could be attributed to trace quantities of REE-bearing phases (undetected by XRD), such as perrierite-(La), allanite-(Ce), and monazite-(Ce), as identified by SEM-EDS analysis (Fig. 18). Comparison of normalized REE patterns with those of Tokaj volcanic rocks (Lukács et al., Reference Lukács, Guillong, Szepesi, Szymanowski, Portnyagin, Józsa, Bachmann, Petrelli, Müller, Schiller, Fodor, Chelle-Michou and Harangi2024; Fig. 17) reveals substantial similarities, although not complete agreement. These findings suggest that Eu and HREE may have been slightly depleted during the alteration of parent rocks to bentonitic rocks, or alternatively, the pyroclasts from the Tokaj Mountains are not the source of the Michaľany-Lastovce deposit. In addition to the geochemical significance of REE in determining the origin of geological materials, there is also an economic factor of their importance driven by increasing demand for these trace elements. Therefore, recent research has been focusing on monitoring the economic potential for REE accumulation from unconventional sources. The economic viability of the Michaľany-Lastovce deposit as a source of REE has so far proven to be unpromising.
Immobile elements classification diagram of Winchester and Floyd (Reference Winchester and Floyd1977), comparing the bulk bentonite samples from the present study with the volcanic rock samples from the Tokaj area (data from Lukács et al., Reference Lukács, Guillong, Szepesi, Szymanowski, Portnyagin, Józsa, Bachmann, Petrelli, Müller, Schiller, Fodor, Chelle-Michou and Harangi2024).
Figure 16. Long description
A log-log line graph with the X-axis representing the ratio N b / Y ranging from 0.01 to 10.00 and the Y-axis representing the ratio Z r / T i O sub 2 ranging from 0.001 to 5.000. The plot is divided into several rock classification fields by solid and dashed lines.
From bottom to top and left to right, the fields are:
* SubAlkaline Basalt at the bottom left.
* Andesite/Basalt and Andesite in the middle left.
* Rhyodacite/Dacite and Rhyolite in the upper middle.
* Alk-Bas (Alkali Basalt), Trachy-andesite, and Trachyte in the middle right.
* Basanite Nephelinite at the far right.
* Comendite Pantellerite and Phonolite at the top right.
Data points are plotted as various black symbols for bentonite samples and colored open circles for volcanic samples. The bentonite samples (labeled L A 1 through V L a-7 E) are tightly clustered in the Rhyodacite/Dacite field, centered around an N b / Y ratio of 0.3 to 0.5 and a Z r / T i O sub 2 ratio of 0.03 to 0.05. The volcanic samples are grouped by unit: Szerencs unit (light blue), Vizsoly Unit (red), Hegyköz unit (green), and Sátoraljaújhely unit (dark blue). These volcanic samples are primarily located in the Rhyolite field, showing higher Z r / T i O sub 2 ratios than the bentonite samples.
Chondrite-normalized rare earth element (REE) patterns (after Anders and Grevesse, Reference Anders and Grevesse1989) of the bentonite samples from this study (black dotted lines) and volcanic rock samples from the Tokaj area (data from Lukács et al., Reference Lukács, Guillong, Szepesi, Szymanowski, Portnyagin, Józsa, Bachmann, Petrelli, Müller, Schiller, Fodor, Chelle-Michou and Harangi2024).
Figure 17. Long description
A line graph on a logarithmic scale. The Y-axis is labeled Sample or R E E chondrite with values ranging from 1 to 1000. The X-axis lists rare earth elements from left to right: L a, C e, P r, N d, S m, E u, G d, T b, D y, H o, E r, T m, Y b, and L u.
Data series include:
* Szerencs unit (purple lines)
* Vizsoly Unit (red lines)
* Hegyköz unit (green lines)
* Sátoraljaújhely unit (gold lines)
* Lastovce (black dotted lines)
All samples follow a similar trend. They begin at the left with high values between 80 and 200 for L a. The lines trend downward toward S m. At E u, every sample exhibits a sharp, V-shaped negative anomaly, with the Vizsoly and Szerencs units dropping most significantly to values between 2 and 4, while other samples drop to between 6 and 20. Following E u, the patterns rise to a plateau between 10 and 40 for the remaining heavy rare earth elements from G d to L u, maintaining a relatively horizontal or slightly descending profile.
Graphs and images from SEM-EDS of selected REE-rich crystals (separated from the sand fraction).
Figure 18. Long description
The figure contains two side-by-side panels.
Left Panel (V L a 2 A, Spectrum 52):
* Micrograph: A grayscale S E M image of a faceted, angular crystal approximately 300 micrometers wide. A red dot labeled Spot 52 marks an analysis point on a smooth crystal face.
* Spectrum: The x-axis represents energy in k e V from 0 to 10, and the y-axis represents intensity in counts. Major peaks are identified at 0.5 k e V (O), 2.1 k e V (P and A u), and a broad cluster between 4.5 and 6.0 k e V containing multiple peaks for L a, C e, and E u. Minor peaks for T h are visible between 2.5 and 4.0 k e V.
Right Panel (V L a 2 C, Spectrum 102):
* Micrograph: A grayscale S E M image of a more irregular, blocky crystal. A red dot labeled Spot 102 marks an analysis point near the center.
* Spectrum: The x-axis and y-axis scales match the left panel. Dominant peaks occur at 0.5 k e V (O), 1.5 k e V (A l), and 1.8 k e V (S i and A u). A distinct peak for C a is located at 3.7 k e V. A broad series of lower-intensity peaks for L a, C e, and F e are distributed between 4.5 and 7.0 k e V.
Sarmatian pyroclasts were probably deposited in brackish or freshwater environments associated with coal accumulation (Baňacký et al., Reference Baňacký, Elečko, Kaličiak, Straka, Škvarka, Šucha, Vass, Vozárová and Vozár1989; Kováč et al., Reference Kováč, Kováč, Marko, Karoli and Janočko1995). A water-rich environment is recognized as a key factor in bentonite formation (e.g. Christidis, Reference Christidis1998; Caballero et al., Reference Caballero, De Cisneros, Huertas, Huertas, Pozzuoli and Linares2005). No indicators of hydrothermal alteration, such as zonation, fractures, or veins, were observed in the studied area. The present findings, together with previous studies, indicate a diagenetic origin for the Michaľany-Lastovce deposit. Moreover, the absence of minerals such as chlorite and epidote, which are typical secondary products of biotite alteration under hydrothermal conditions, further supports a diagenetic origin (Schwartz, Reference Schwartz1958; Eggleton and Banfield, Reference Eggleton and Banfield1985; Dong et al., Reference Dong, Peacor and Murphy1998; Kadir et al., Reference Kadir, Külah, Erkoyun, Christidis and Arslanyan2019). Low sulfur isotope values (δ34S = –20.22 to 2.35‰) confirm a biological origin for sulfur of the analyzed pyrite (Rickard, Reference Rickard2012; Hammerli et al., Reference Hammerli, Greber, Martin, Bouvier, Kemp, Fiorentini, Spangenberg, Ueno and Schaltegger2021). If smectite and pyrite were assumed to be formed simultaneously, this would suggest that smectites developed from the alteration of volcanic glass and dissolution of pyroclastic materials under anaerobic, diagenetic conditions. Ambient conditions and low temperatures, probably <50°C, were probably sufficient for the alteration of volcanic ash to bentonitic rock.
Opal-CT precipitated simultaneously from residual, dissolved silica after its release during the alteration of volcanic glass to montmorillonite or shortly thereafter (Grim, Reference Grim1962). This assumption is supported by the presence of comparable quantities of both phases (Table 2; Fig. 7), as well as the botryoidal texture (Fig. 15) that results from the filling of the pore environment with supersaturated fluids (Dong et al., Reference Dong, Morrison and Jaireth1995). The presence of approximately equal proportions of montmorillonite and opal-CT in the studied samples is explained by a closed system with insufficient fluid circulation. The fluid circulation usually depends on the convective flow supported by thermal gradient (Christidis and Huff, Reference Christidis and Huff2009) or paleorelief (Koděra et al., Reference Koděra, Lexa, Fallick, Wälle and Biroň2014).
The term closed system refers to an environment in which fluid migration was absent or significantly restricted. Under such conditions, the SiO2 released during the alteration of volcanic glass to bentonite remained within the system and subsequently served as a source for the formation of opal-CT. In addition to the abundance of opal-CT (30–40 wt.%), the SiO2/Al2O3 ratio provides further evidence supporting the interpretation of a closed-system environment. Unaltered acidic volcanic rocks (e.g. perlite and pyroclastic rocks) typically exhibit SiO2/Al2O3 ratios ranging from 4.2 to 5.6, with an average value of 5.4 based on 58 analyzed samples (Ddani et al., Reference Ddani, Meunier, Zahraoui, Beaufort, El Wartiti, Fontaine, Boukili and El Mahi2005; Kadir et al., Reference Kadir, Külah, Erkoyun, Christidis and Arslanyan2019; Lexa et al., Reference Lexa, Varga, Uhlik, Koděra, Biroň and Rajnoha2021; Lukács et al., Reference Lukács, Guillong, Szepesi, Szymanowski, Portnyagin, Józsa, Bachmann, Petrelli, Müller, Schiller, Fodor, Chelle-Michou and Harangi2024). In contrast, bentonites lacking a significant opal-CT component exhibit SiO2/Al2O3 ratios ranging from 2.7 to 4.2, with an average value of 3.3 based on 32 analyzed samples (Christidis et al., Reference Christidis, Scott and Marcopoulos1995; Ddani et al., Reference Ddani, Meunier, Zahraoui, Beaufort, El Wartiti, Fontaine, Boukili and El Mahi2005; Kadir et al., Reference Kadir, Külah, Erkoyun, Christidis and Arslanyan2019; Osacký et al., Reference Osacký, Binčík, Paľo, Uhlík, Madejová and Czímerová2019; and unpublished data from Central Slovakia). This decrease reflects substantial SiO2 depletion resulting from its removal by migrating fluids. Conversely, bentonites enriched in opal-CT exhibit elevated SiO2/Al2O3 ratios, ranging from 5.0 to 6.0 (average 5.8; four samples from unpublished data from Central Slovakia) and from 3.7 to 5.4 (average 4.7; 19 samples from this study (Table 4), and Osacký et al. (Reference Osacký, Binčík, Paľo, Uhlík, Madejová and Czímerová2019)).
In contrast, significant volumes of clinoptilolite-rich tuff formed in the volcano-sedimentary marine environment of the ESNB, located 50 km south of the studied deposit. Rhyodacite pyroclastic material was deposited with high-density turbidites in a middle shelf to deep basin environment during the lower Badenian (Subová et al., Reference Subová, Rybár, Šarinová, Hudáčková, Jamrich, Sliva, Šály and Hlavatý2022; Šarinová et al., Reference Šarinová, Rybár, Brčeková, Szymanowski, Fred, Marcel, Kotulová, Ľubomír, Hudáčková and Lukács2026). Zeolitization in this area occurred under higher diagenetic temperatures, ranging from 44 to 84°C (Šamajová, Reference Šamajová, Kirov, Filizova and Petrov1997; Tschegg et al., Reference Tschegg, Rice, Grasemann, Matiasek, Kobulej, Dzivák and Berger2019), compared with the bentonitic rock studied. Further research on the genesis of the Michaľany-Lastovce deposit will be conducted, including analyses of stable isotopes of hydrogen and oxygen, and possibly boron, as well as high-precision zircon U-Pb dating.
Conclusions
This study provides the first comprehensive characterization of the Michaľany-Lastovce deposit in the East Slovak Neogene Basin (ESNB) in more than six decades. The bentonitic samples studied contain a small amount of Ca-montmorillonite (~40 wt.%), alongside an almost equal proportion of opaline silica phases, which remained inseparable from the smectite, even within the <0.2 μm clay fraction. The CEC of the samples analyzed, however, was just 35–58 for bulk samples but 70–80 meq per 100 g for the <0.2 μm clay fractions.
The opaline silica phases were identified as opal-CT, as confirmed by XRD analysis following heating of the samples, apart from one sample which contains opal-C. Cristobalite has not been identified. Typical bladed lepisheres of the opal-CT, with diameter of a few micrometers have been identified by SEM. The specific surface area (S BET) of Michaľany-Lastovce bentonitic rock, ranges between 50 to 73 and 69 to 101 m2 g–1 for bulk samples and the corresponding (<0.2 μm) clay fractions, respectively. These values were considerably higher than those determined for high-grade bentonite from the Jelšový potok deposit in the Central Slovak Volcanic Field. The difference has been explained by thinner montmorillonite crystallites and an abundance of opal-CT, with its bladed morphology and nanosized blades.
Additionally, Michaľany-Lastovce smectites demonstrate medium charge (0.38 per f.u.), attributed mainly to the substitution of the trivalent cations (Al) by divalent ones (Mg) within the octahedral position. They can also be categorized as beidellitic montmorillonite based on the charge distribution, with 63% in the octahedral position, and were classified as cv/tv and tv/cv types according to the ratios of dehydroxylation peaks obtained from DTG analyses. Generally, the shallower samples from the Michaľany-Lastovce deposit display a greater variety of tv smectite, while samples collected from deeper depths reveal a predominance of cv smectite varieties. This statement, however, needs confirmation by a larger, statistically reliable series of measurements.
The study presented here confirmed the volcano-sedimentary origin of Michaľany-Lastovce bentonitic rock with new findings such as the presence of preserved relicts of volcanic glass shards, rheomorphic flow, spherulitic aggregates and botryoidal texture. The assumption that the fine pyroclastic material was a parent rock for bentonite formation has been confirmed also by the dominant presence of angular and subangular grains determined by X-ray micro-computed tomography, which was demonstrated to be a useful method for studying bentonite texture mainly when it is difficult or impossible to prepare thin sections. The major and trace elements, including REE, the structural formula of montmorillonite and the absence of mafic minerals (amphibole and pyroxene), indicate the rhyodacitic character of the source rocks. Episodic increases in the Fe content in the whole-rock chemical analyses and in the montmorillonite structure, however, do not exclude a partial influence of andesite rocks from the surroundings. Missing indicators of hydrothermal alteration have been used as an indirect marker of diagenetic genesis of the bentonitic rocks studied. This interpretation was further supported by the low sulfur isotope (δ34S) values. The identification of framboidal pyrite indicates anaerobic conditions.
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1017/cmn.2026.10049.
Data availability statement
Most of data supporting the findings of this study are included within the article and the supplementary material. The rest data will be available from corresponding author.Most of data supporting the findings of this study are included within the article and the supplementary material. The rest data will be available from corresponding author.
Acknowledgements
The authors thank Ondrej Nemec for help with SEM operation, REGOS Ltd and LB MINERALS Ltd for providing the samples, and the University of Salamanca for measurements.
Author contributions
Study conception: F.A.M.G., P.U.; sample preparation: F.A.M.G.; methodology, analyses and their interpretation: all authors; preparation of illustrative materials: F.A.M.G., M.O., J.B., J.Š., P.U.; writing and funding acquisition: F.A.M.G., M.O., P.U.Study conception: F.A.M.G., P.U.; sample preparation: F.A.M.G.; methodology, analyses and their interpretation: all authors; preparation of illustrative materials: F.A.M.G., M.O., J.B., J.Š., P.U.; writing and funding acquisition: F.A.M.G., M.O., P.U.
Financial support
This work was supported by the Slovak Research and Development Agency (APVV-20–0175), UK/3062/2024 (funded by the EU NextGenerationEU through the Recovery and Resilience Plan for Slovakia under project no. 09I03-03-V05-00012), UK/1025/2025, and by the Scientific Grant Agency of the Ministry of Education, Science and Culture of the Slovak Republic under contract VEGA 1/0336/25 and 1/0306/23. This work was also supported by the Research and Innovation Authority (VAIA): NextGenerationEU – Recovery and Resilience Plan of the Slovak Republic under the contract 09103-03-V06-00076.
Competing interests
The authors declare none.
