Origin of Zn-Pb Mineralization of the Vein Bt23C, Bytíz Deposit, Příbram Uranium and Base-Metal Ore District, Czech Republic: Constraints from Occurrence of Immiscible Aqueous–Carbonic Fluids

Abstract

The mineralogical, fluid inclusion, and stable isotope (C, O) study was conducted on a Late Variscan Zn-Pb vein Bt23C, Příbram uranium and base-metal district, Bohemian Massif, Czech Republic. The vein is hosted by folded Proterozoic clastic sediments in exo-contact of a Devonian-to-Lower-Carboniferous granitic pluton. Siderite, dolomite-ankerite, calcite, quartz, baryte, galena, sphalerite, V-rich mica (roscoelite to an unnamed V-analogue of illite), and chlorite (chamosite) form the studied vein samples. The banded texture of the vein was modified by the episodic dissolution of earlier carbonates and/or sphalerite. Petrographic, microthermometric, and Raman studies of fluid inclusions proved a complicated fluid evolution, related to the activity of aqueous fluids and to an episode involving an aqueous–carbonic fluid mixture. Homogenization temperatures of aqueous inclusions decreased from ~210 to ~50 °C during the evolution of the vein, and salinity varied significantly from pure water up to 27 wt.% NaCl eq. The aqueous–carbonic fluid inclusions hosted by late quartz show highly variable phase compositions caused by the entrapment of accidental mixtures of a carbonic and an aqueous phase. Carbonic fluid is dominated by CO₂ with minor CH₄ and N₂, and the associated aqueous solution has a medium salinity (6–14 wt.% NaCl eq.). The low calculated fluid δ¹⁸O values (−4.7 to +3.6‰ V-SMOW) suggest a predominance of surface waters during the crystallization of dolomite-ankerite and calcite, combined with a well-mixed source of carbon with δ¹³C values ranging between −8.2 and −10.5‰ V-PDB. The participation of three fluid endmembers is probable: (i) early high-temperature high-salinity Na>Ca-Cl fluids from an unspecified “deep” source; (ii) late low-salinity low-temperature waters, likely infiltrating from overlying Permian freshwater partly evaporated piedmont basins; (iii) late high-salinity chloridic solutions with both high and low Ca/Na ratios, which can represent externally derived marine brines, and/or local shield brines. The source of volatiles can be (i) in deep crust, (ii) from interactions of fluids with sedimentary wall rocks and/or (iii) in overlying Permian piedmont basins containing, in places, coal seams. The event dealing with heterogeneous CO₂-bearing fluids yielded constraints on pressure conditions of ore formation (100–270 bar) as well as on the clarification of some additional genetic aspects of the Příbram’s ores, including the reasons for the widespread dissolution of older vein fill, the possible re-cycling of some ore-forming components, pH changes, and occasionally observed carbon isotope shift due to CO₂ degassing.

1. Introduction

The Příbram ore area consists of two principal districts comprising the Late Variscan uranium and base-metal vein mineralizations: the westward lying Březové Hory ore district, where base-metal mineralization prevails over uranium ores, and the eastward Příbram uranium and base-metal district, where uranium ores predominate (Figure 1). The Příbram uranium and base-metal district is the most important hydrothermal vein-type uranium deposit in the Czech Republic and one of the largest vein-type uranium deposits worldwide. About 50,000 tons of U were mined out there during 40+ years of post-WWII exploitation (1948–1991), and the parallel extraction of base metals and silver yielded more than 6000 t Pb, 2400 t Zn and 28 t Ag [1]. The Příbram uranium and base-metal district consists of 9 individual ore deposits containing 20 vein clusters with more than 2500 hydrothermal veins within an area of 40 × 20 km.

Despite its economic significance, only a very limited amount of modern genetic data is available for this ore district, including the fluid inclusion studies. To date, only three works deal with investigation of fluid inclusions in the Příbram ore area. The most comprehensive study by [2] was based on the investigation of 30 specimens, originated mostly from the base-metal deposits. Only six samples (formed exclusively by calcite) of their dataset were sourced from uranium deposits. Two additional samples from base-metal deposits were studied in the bachelor thesis by [3]. Both works reported exclusively aqueous fluid inclusions without any volatile gases in concentrations detectable by microthermometric measurements; Raman analysis of gas bubbles of L + V fluid inclusions was not applied. The most recent paper by [4] was focused on the base-metal vein H32A, Háje deposit, Příbram uranium and base-metal district. That study confirmed the predominance of gases-absent aqueous inclusions, but traces of CO₂, CH₄, and N₂ were identified by Raman analysis of vapour bubbles in a small part of aqueous inclusions. These gases occurred in various mutual proportions across the whole paragenesis of the vein H32A except of late calcite. The total amount of gases was always very low, as in none of the studied inclusions was the formation of clathrate or the liquid-vapour phase equilibria in gas bubbles observed during microthermometry.

In this contribution, we present new evidence on the occurrence of gases in the fluids giving rise to Příbram‘s base-metal mineralization. In the studied Zn-Pb mineralization of the vein Bt23C, we found aqueous fluid inclusions containing volatile gases in high concentrations allowing for the formation of clathrate or the heterogenization of a non-aqueous phase during cooling as well as fluid inclusions exclusively formed by a non-aqueous phase (i.e., carbonic inclusions). Such fluid inclusions were not previously reported from any other vein in the whole Příbram ore area. The detailed study of this fluid event allowed us to precisely characterize P-T conditions of vein formation, which were poorly constrained up to date. The source of some fluid components was checked by the carbon and oxygen isotope study of carbonates and quartz. Moreover, we have discussed some additional genetic consequences resulting from the recognition of activity of such gas-enriched fluids in the Příbram ore district.

2. Background

2.1. Geological Setting

The Příbram area is located in the core part of the Bohemian Massif (Figure 1), which belongs to the Central European Variscides consolidated during the Variscan Orogeny at the end of Palaeozoic. The ore deposits are situated in metasedimentary rocks of the Teplá-Barrandian Unit in exo-contact of the Central Bohemian Plutonic Complex (CBPC). The Teplá-Barrandian Unit comprises several mutually tectonically separated crustal blocks formed by Neo-Proterozoic rocks with different pre-Variscan tectono-metamorphic evolution, magmatic history and provenance. The folded Proterozoic rocks are covered by volcano-sedimentary sequences of the Barrandian Palaeozoic, which originated during the Cambrian to Devonian [5,6,7,8,9,10]. The crust-scale [11] tectonic contact between the Teplá-Barrandian and the Moldanubian units was intruded by the CBPC in the Uppermost Devonian-to-Lower Carboniferous (~354–336 Ma; [12,13,14,15]). The CBPC is dominated by various types of granitic rocks with subordinate bodies of quartz diorites, durbachites, and gabbros. The end of Variscan orogenetic processes was followed by a collapse of the orogen associated with fault segmentation of the basement, its uplift, and rapid erosion [11]. The early post-orogenic evolution also comprised the moulding of piedmont basins, which were filled up by predominantly clastic freshwater molasse of the Uppermost Carboniferous-to-Lower Permian age [16,17,18]. However, most of these Upper Palaeozoic sediments were eroded during the Mesozoic and Tertiary, with negligible preserved relics being preserved only today (Blanice Graben in inset of Figure 1).

The studied area is situated in contactly metamorphosed Upper-Proterozoic marine siliciclastic sediments of the Dobříš series forming a large anticline structure. There are five sequences of sedimentary rocks within the Dobříš series differing in the content of siltstones, mudstones, conglomerates, and sandstones [19]. The so-called conglomerate-sandstone P2 succession hosts the base-metal ore bodies of the vein Bt23C. The facially very variable character of the rocks with prevailing sandstone and siltstone layers, elongated lenses of conglomerates and thin tuffite interbeds are typical for this series. The overlying volcano-sedimentary sequence of the Barrandian Palaeozoic is represented by Cambrian marine clastic sediments (mainly graywackes and siltstones) in the Příbram ore area (Figure 1). The whole sedimentary sequence is intruded by two populations of dykes of magmatic rocks. The first group is represented by “diabase” (=strongly hydrothermally altered palaeobasalt) dykes, which are older than Variscan granites of the CBPC. The second group comprises porphyres, porphyrites, lamprophyres, aplites, and pegmatites, which are younger than granites and occur commonly either in granitic rocks, or in the close exo-contact of the CBPC [19].

2.2. Ore Mineralization of the Příbram Area

The earliest ore mineralization recognized in the Příbram ore area comprises rare Au-bearing coulisse-arranged quartz veins. Quartz gangue with minor carbonates, chlorite, and illite-muscovite hosts high-fineness gold associated with arsenopyrite and Bi-tellurides [20]. The Au contents reach up to 100 g/t. Origin of this early mineralization was bound to progressive cooling (300–400 °C at the beginning, ca. 180 °C at the end) of low-salinity (1.2 to 7.0 wt.% NaCl eq.) aqueous solutions, which were likely generated during thermal alteration of host Proterozoic sediments caused by intrusion of the CBPC [20]. A significant mineralogical reworking of early gold mineralization can occur in places where younger base-metal veins crosscut the Au-bearing veins, mainly due to input of base metals, Ag, and Sb [20].

The superimposed steep N–S to NW–SE striking ore veins are mainly hosted by the Upper Proterozoic or Cambrian sediments and partly follow the dykes of hydrothermally strongly altered “diabases”. The paragenetic situation is largely similar in both uranium and base-metal districts of the Příbram ore area, although significant differences can occur locally. Previous studies have invariably emphasized the complicated multi-stage evolution of vein ore deposits in the Příbram area, typically with 7 to 15 respective hypogene mineralization (sub)stages [4,21,22,23]. These features are associated with the position of the ore area near the crust-scale tectonic structure [11]. Four major mineralization stages were recognized in the Příbram area: the earliest siderite-sulphidic, calcite, calcite-uraninite and the latest calcite-sulphidic. In the longitudinal direction from SW to NE, the extent of the younger mineralization stages grows at the expense of the older ones [19]. The mineralization is very rich, with more than 300 mineral species described from the Příbram area [24]. The Fe-poor sphalerite, galena, pyrite, minerals of the tetrahedrite group, and locally also Ag-Pb-Sb, Pb-Sb, and/or Ag-Cu-Sb sulphosalts are the typical ore minerals of the base-metal mineralization. The ore minerals are usually hosted by carbonate (siderite, dolomite-ankerite, and most commonly calcite), less frequently quartz, and/or baryte gangue. The uranium mineralization consists especially of uraninite, less frequently coffinite and/or U-bearing anthraxolite, which are hosted by a multi-phase calcite vein.

The poly-phase nature of the Příbram uranium and base-metal mineralizations is also confirmed by available geochronological data, which suggest long-lasting hydrothermal activity from the Lower Carboniferous to the Triassic [25,26,27]. The biotite Ar-Ar cooling ages of CBPC magmatites and associated dyke rocks constrained the uppermost possible age of the early base-metal mineralization at 336–338 Ma [25]. The Rb-Sr dating of siderite and sphalerite from the earliest siderite-sulphidic stage gave age of 330 ± 4 Ma [26]. The U-Pb dating of uraninite from the superimposed calcite-uraninite stage yielded Permian dates 275 ± 4 and 278 ± 4 Ma [27], which well agree with results obtained by K-Ar dating of associated micas (274 ± 7 and 268 ± 7 Ma) [26]. Finally, the Rb-Sr dating of sphalerite and associated calcite and dolomite from the late post-uranium sulphidic mineralization gave ages 276 ± 3 Ma and 224 ± 3 Ma, respectively [26].

The previous fluid inclusion and stable isotope studies showed that the early Pb-Zn mineralization of the siderite-sulphidic stage from the Březové Hory district has formed from exclusively aqueous fluids featured by high temperatures (250–300 °C), high salinities (14–23 wt.% NaCl eq.), and highly positive δ¹⁸O values (6‰–10‰ V-SMOW). The source of these fluids is interpreted in a deep circulation in the hot crustal rocks [2]. The origin of reduced sulphur with calculated δ³⁴S values −5 to −8‰ V-CDT is sought in the host rocks. A decrease in fluid temperature (to ~60 °C), fluid δ¹⁸O values (2‰–6‰ V-SMOW), and fluid salinities (3–10 wt.% NaCl eq.) characterize late stages of early Pb-Zn mineralization with dolomite, calcite, and baryte gangue. The superimposed uranium mineralization formed from low-salinity (<5 wt.% NaCl eq.), low-temperature (<150 °C), near-zero δ¹⁸O (−3 to 3‰ V-SMOW) aqueous fluids, which were dominated by surface waters experiencing only shallowly seated circulation [2]. During the late stages of hydrothermal activity, the participation of low-temperature high-salinity Ca–Na–Cl solutions was also found, representing either brines of external marine provenance or locally generated “shield brines” [4].

2.3. Vein Bt23C

The vein Bt23C belongs to the Bytíz deposit, which was the most important uranium deposit in the Příbram area. The Bytíz deposit, situated in the central part of the Příbram uranium and base-metal district, yielded more than 50% of the Příbram’s uranium ores, which were extracted from a total of 782 veins.

The vein Bt23C is located in the vein cluster Bt17B–Bt22 in the arch part of the Příbram anticline. This vein cluster was explored from the surface to the depth of 1300 m. The steep (dip 65–90°) hydrothermal veins and their branches strike NW–SE and N–S, exceptionally NE–SW. Some parts of this vein cluster can be characterized as stockworks. The main NW–SE striking veins and part of their veinlets represent the prolongation of the NW part of the Zduchovice tectonic zone continuing from the CBPC into Proterozoic sediments. The base-metal ores occurred in all depth levels in many NW–SE striking veins containing the siderite-sulphidic mineralization [19]. There is a predominance of sphalerite over galena in the vein cluster Bt17B–Bt22 and sphalerite content increases with increasing depth [28].

The vein Bt23C is one of the important veins of the NW–SE direction (310–330°) with the dip 65–80° to the SW. The width of the vein Bt23C is 20 cm on average, and 75 cm in maximum. The vein was exposed on the 15th, 17th, 20th, 21st, 22nd, and 23rd levels of the shaft No. 19.

The shape of the ore bodies is elongated in most cases. They occur in the proximity of the junctions with faults and other veins, especially in sites where (i) strike of the vein changes to NNW–SSE and the dip is around 70°, (ii) a wider thickness of the vein occurs or (iii) additional fracturing manifested by the presence of veinlets is observed. The vertical extent of the ore-mineralized part of the vein is limited by a fault dipping ca. 45° to the SE above the 20th level. In depth, the Bt23C vein is terminated by a fault structure on the 23rd level [28,29].

The mineralization of the vein Bt23C consists of simple Zn>Pb base-metal association with banded or drusy texture, in development typical for the Příbram ore area. Sphalerite and galena are reported as the main ore minerals of the vein Bt23C, which are accompanied by calcite, siderite, quartz, and minor baryte, whereas pyrite, pyrrhotite, bournonite, and cubanite are accessories [19].

3. Samples and Analytical Methods

The well-documented archive samples used for this study were collected at the 20th level (in depth of 900–950 m) at place, where the vein Bt23C crosscuts an Au-bearing quartz-sulphide vein, which was studied in detail by [20].

The standard doubly polished plates (~100-μm-thick) were prepared from selected representative samples that are deposited in the Mineralogical collection of the National Museum in Prague. Mineral composition and textures of samples were studied by a polarizing microscope Nikon Eclipse ME600 (Nikon Co., Tokyo, Japan) and an electron microprobe Cameca SX-100 (AMETEK, Inc., Berwyn, PA, USA). The chemical composition of selected minerals was measured by the latter apparatus in the wave-dispersive mode using acceleration voltage of 15 kV, beam current of 10 nA (micas, chlorites) or 5 nA (carbonates) and beam diameter of 2 µm (micas, chlorites) or 4 µm (carbonates). The following elements were determined in micas and chlorites: Al, Ba, Ca, Co, Cs, Cu, F, Fe, K, Mg, Mn, N, Na, Ni, P, Pb, Rb, Sb, Si, Sr, and Zn. The following elements were determinedin carbonates: Al, Ba, Ca, Co, Cu, Fe, Mg, Mn, Na, Ni, P, Pb, Si, Sr, and Zn. The following analytical lines and standards were used for determination of individual elements: Kα lines: albite (Na), fluorapatite (P), BN (N), celestine (S), Cr₂O₃ (Cr), diopside (Mg), halite (Cl), hematite (Fe), chalcopyrite (Cu), LiF (F), Ni (Ni), rhodonite (Mn), sanidine (K, Al), TiO₂ (Ti), V (V), wollastonite (Ca, Si); Lα lines: baryte (Ba), Cs-glass (Cs), Rb-Ge glass (Rb); Lβ line: celestine (Sr); Mα line: vanadinite (Pb). The acquisition time on the peak was typically between 10 and 30 s (150 s for N), whereas acquisition time on each background was half of the peak time. Content of oxygen was calculated from stoichiometry. The raw counts were converted to wt.% using the standard PAP procedure [30]. The contents of elements, which are not included in tables of mineral compositions, were always below the detection limits. The detection limits varied between 0.05 and 0.1 wt.% for most elements except for F and N, which were around 0.2 wt.%. The data were corrected for overlaps Ti vs. V, Mn vs. Cr, and Ca vs. P.

After removing the carbon coating by re-polishing, the fluid inclusion petrography was carried out following the descriptive and genetic criteria given by [31] and fluid inclusion assemblages (FIAs) concept [32]. Microthermometric measurements were obtained from 561 fluid inclusions using a Linkam THMSG 600 heating-freezing stage (Linkam Scientific Instruments, Surrey, UK) operating in a temperature range of −196 to 600 °C and attached to an Olympus BX-51 polarizing microscope (Olympus Co., Tokyo, Japan) with 20× and 50× long-working-distance objectives. The stage was calibrated using natural fluid inclusions with known temperatures of phase transitions between −56.6 and 374.1 °C. The precision of measurements is ±0.1 °C in the temperature span between 0 and 50 °C and ±0.5 °C at 374.1 °C. In aqueous inclusions, the following parameters were measured: freezing temperature (T_f), temperature of initial melting (T_i), homogenization temperature (Th), melting temperature of last crystal of hydrohalite (Tm_hh), ice (Tm_ice) and clathrate (Tm_cla). In aqueous–carbonic and carbonic inclusions, melting temperature of solid CO₂ (Tm_CO2) and temperature of partial homogenization of carbonic phase (Th_car) were also obtained. In inclusions, where Th_car ˂ Tm_cla, the Th_car was measured before freezing of the inclusion. The degree of fill (F) was estimated as the L/(V + L) ratio at room temperature. Salinities of aqueous inclusions were calculated as wt.% NaCl eq. using the equation by [33]. For aqueous–carbonic inclusions, salinities were calculated using equations by [34] and [35] for three- and four-phase clathrate-melting associations, respectively. Microthermometric data on carbonic inclusions were further processed (calculation of bulk fluid composition, densities, and isochores) using the FLUIDS software packages [36] with calibrations by [37,38,39].

The chemical composition of carbonic phase of fluid inclusions was determined by a DXR dispersive Raman Spectrometer (Thermo Fisher Scientific, Inc., Waltham, MA, USA) attached to a confocal Olympus microscope. The Raman signal was obtained using an unpolarised 532 nm solid-state, diode-pumped laser and detected by a CCD detector. The parameters of measurements were as follows: 100× objective, 5 s exposure time, 200 exposures (in order to increase the signal/noise ratio), 10 mW laser power, 50 μm pinhole spectrograph aperture, and spectral range 1000–4200 cm⁻¹. The quantification of gaseous components was based on quantitative measurement of peak areas [40] and calibration using natural fluid inclusions with known proportions of CH₄, N₂, and CO₂.

For stable isotope analyses, ca. 0.1 g of pure mineral phase was handpicked under binocular microscope and ground in an agate mortar. The carbonates were converted to CO₂ by reaction with 100% orthophosphoric acid [41] in vacuum at 25 °C. The measurement of isotopic compositions of carbon and oxygen of the resultant CO₂ gas was realized using Delta V Advantage isotope-ratio mass spectrometer (Thermo Fisher Scientific, Inc., Waltham, MA, USA) operating using dual inlet methodology in the Laboratories of the Czech Geological Survey in Prague. Powdered samples of quartz were washed with hydrochloric acid at 25 °C to remove possible impurities (especially carbonates). Oxygen was extracted by fluorine, cleaned on 5Å molecular sieve, and its δ¹⁸O value was determined on a Finnigan MAT 253 mass spectrometer (Finnigan MAT GmbH, Bremen, Germany) at the University of Lausanne (T. Vennemann analyst). The results of C and O isotope analyses are conventionally expressed in delta (δ) notation as per mil (‰) deviation from commonly used standards V-PDB and V-SMOW, respectively. Uncertainty is better than ±0.05, ±0.1, and ±0.2‰ for carbonate δ¹³C, carbonate δ¹⁸O, and quartz δ¹⁸O, respectively. The dolomite δ¹⁸O value was corrected for fractionation during chemical preparation of the sample by a value of −0.84 [42]. The isotopic composition of the parent fluid was calculated using the fractionation factors published by [43,44,45].

4. Results

4.1. Vein Mineralogy

Mineralogical investigations of studied samples revealed three mineralizing events (Figure 2). Banded, veinlet, and drusy textures of the vein fill are common and often complicated by the replacement of older mineral phases by younger minerals. The presence of siderite, dolomite-ankerite, quartz, calcite, baryte, sphalerite, galena, chlorite, and mica was confirmed in the vein fill.

4.1.1. Stage 1

Stage 1 comprises siderite, sphalerite, and galena (Figure 2). The coarse-grained Mn-rich siderite (Sid_56–71Rdc_14–28Mag_11–18Cal_1–7) is the oldest mineral of the vein (Figure 3a,b), which occurs in relatively small amounts. An indistinct oscillatory zoning is sometimes observed in a BSE image. Siderite is always strongly corroded and replaced by younger phases, especially calcite, quartz, sphalerite, and galena.

The coarse-grained brown sphalerite grows on siderite in bands up to 3 cm thick. In polished sections, no anisotropy, or inclusions of other ore phases (i.e., “chalcopyrite disease”) are observed. In thin sections, a very detailed oscillatory zoning of sphalerite is observed with yellow, orange, red, brownish, and dark-brown zones (Figure 3c). Sphalerite is strongly cracked with fissures filled mainly by calcite and less by dolomite-ankerite.

Galena is a subordinate ore mineral, forming a 5 mm thick outer “growth zone” on sphalerite, lined by own crystal faces in apical part. In reflected light, no marks of deformation are observed on galena grains.

4.1.2. Stage 2

Stage 2 involved crystallization of carbonate of the dolomite-ankerite series (Figure 2), which are the least frequent carbonates in the vein Bt23C. They form fillings of some cracks in ore bands (Figure 3c) as well as replacements at the expense of siderite, but they predate calcite and baryte. They display strong oscillatory compositional zoning (Figure 3b) with compositions evolving from dolomite to Mg-rich ankerite (Dol_31–65Ank_34–50Ktn_1–19).

4.1.3. Stage 3

Stage 3 is characterized by crystallization of mica, chlorite, quartz, and calcite (Figure 2), which followed the most intense fracturing of the early vein (Figure 3e,f). Rare small nests of fine-grained mica are enclosed in quartz-calcite aggregates. Mica has an uncommon composition, as vanadium (1.06–1.37 apfu V, based on 11 atoms of oxygen) prevails over ^viAl (0.21–0.33 apfu) in all analyses, and the contents of P (≤0.15 apfu), Fe (≤0.26 apfu), Mg (≤0.27 apfu), Ca (≤0.13 apfu), Zn (≤0.05 apfu), and Pb (≤0.03 apfu) are elevated (

Table 1. Representative compositions of mica from the vein Bt23C. Oxides are in wt.%, the apfu values are calculated on the basis of 11 atoms of oxygen.

An. No	1	2	3	4	5	6
Sample	Pb-666	Pb-666	Pb-666	Pb-666	Pb-666	Pb-666
P2O5	1.58	1.26	1.58	2.45	2.26	2.32
SiO2	44.02	44.62	43.66	43.84	44.44	42.71
Al2O3	12.91	12.23	12.52	13.35	12.73	12.43
V2O3	21.96	23.86	22.16	18.41	21.36	21.18
Sb2O3	b.d.	b.d.	b.d.	0.26	0.18	0.21
MgO	1.95	1.75	1.94	2.55	1.98	1.93
MnO	0.09	0.07	0.08	0.14	0.12	0.11
FeO	2.90	2.70	3.00	4.37	3.14	3.09
ZnO	0.62	0.70	0.78	0.77	1.04	0.29
CaO	1.20	0.98	1.29	1.64	1.61	1.64
PbO	0.89	0.87	1.09	1.40	1.48	1.74
Na2O	b.d.	0.09	b.d.	b.d.	b.d.	b.d.
K2O	6.23	6.74	6.31	5.85	6.12	6.24
F	0.11	0.26	0.28	0.33	0.11	0.32
Total	94.46	96.13	94.69	95.36	96.57	94.21
P5+	0.097	0.077	0.097	0.149	0.136	0.144
Si4+	3.186	3.201	3.176	3.156	3.165	3.134
ivAl3+	0.717	0.723	0.726	0.694	0.699	0.721
T sum	4.000	4.000	4.000	4.000	4.000	4.000
viAl3+	0.384	0.311	0.347	0.439	0.370	0.354
V3+	1.274	1.372	1.293	1.063	1.220	1.246
Sb3+	b.d.	b.d.	b.d.	0.008	0.005	0.006
Mg2+	0.210	0.187	0.210	0.274	0.210	0.211
Mn2+	0.006	0.004	0.005	0.009	0.007	0.007
Fe2+	0.176	0.162	0.183	0.263	0.187	0.190
Zn2+	0.033	0.037	0.042	0.041	0.055	0.016
O sum	2.083	2.074	2.080	2.096	2.054	2.030
Ca2+	0.093	0.075	0.101	0.127	0.123	0.129
Pb2+	0.017	0.017	0.021	0.027	0.028	0.034
Na+	b.d.	0.013	b.d.	b.d.	b.d.	b.d.
K+	0.575	0.617	0.586	0.537	0.556	0.584
I sum	0.796	0.814	0.829	0.845	0.859	0.911
Catsum	6.769	6.796	6.787	6.787	6.762	6.777
F-	0.025	0.059	0.064	0.075	0.025	0.074

b.d.—below detection limit; I sum = K + Na + 2Ca + 2Pb.

). The sum of interlayer cations varies at 0.80–0.91, suggesting compositions between roscoelite and an unnamed ^viV-dominated analogue of illite.

Trioctahedral chlorite is commonly associated with calcite, quartz, and mica and usually forms thin layers on crystals of sulphides or on the walls of cracks in them (Figure 3d,e). It belongs to chamosite in classification scheme by [46] with Si = 2.81–2.97 apfu (based on 14 atoms of oxygen), Fe²⁺/(Fe²⁺ + Mg) = 0.58–0.73, low Mn (≤0.04 apfu), low Ca (≤0.09 apfu), often elevated V (≤0.51 apfu), and sometimes elevated Zn (≤0.14 apfu;

Table 2. Representative compositions of chlorite from the vein Bt23C. Oxides are in wt.%, the apfu values are calculated on the basis of 14 atoms of oxygen. Where possible, total Fe was recalculated to Fe²⁺ and Fe³⁺ according to [47].

An. No.	1	2	3	4	5	6	7	8	9	10	11	12	13
Sample	Pb-670	Pb-670	Pb-670	Pb-670	Pb-670	Pb-667	Pb-667	Pb-667	Pb-666	Pb-666	Pb-666	Pb-666	Pb-666
P2O5	b.d.	b.d.	b.d.	b.d.	b.d.	0.13	0.14	b.d.	b.d.	b.d.	b.d.	b.d.	0.53
SiO2	25.83	25.53	26.00	26.15	25.08	27.47	27.83	26.70	27.33	27.33	26.53	26.82	25.62
Al2O3	18.06	17.72	18.56	17.58	19.04	18.15	19.37	19.12	18.48	18.04	16.95	17.30	15.41
V2O3	b.d.	b.d.	b.d.	b.d.	b.d.	0.46	0.49	0.57	1.11	2.78	3.73	4.24	5.56
Fe2O3	1.96	1.93	2.11	1.98	2.38	2.14	1.90	2.25	3.45	3.50	n.a.	n.a.	n.a.
MgO	7.11	7.21	6.98	7.39	7.12	10.88	10.39	9.07	10.93	8.92	9.81	9.53	7.90
CaO	0.56	0.56	0.67	0.73	0.68	0.22	0.12	0.10	b.d.	0.13	0.34	b.d.	0.41
MnO	0.22	0.21	0.21	0.24	0.28	0.19	0.17	0.32	0.14	0.24	0.23	0.14	0.17
FeO	33.77	32.94	33.31	33.00	32.92	28.14	27.68	29.97	26.90	27.42	30.33	29.29	27.75
ZnO	b.d.	b.d.	b.d.	b.d.	b.d.	0.17	0.25	b.d.	1.75	0.81	1.33	b.d.	0.23
PbO	0.20	0.23	b.d.	b.d.	b.d.	b.d.	b.d.	b.d.	b.d.	b.d.	0.17	0.65	0.51
Na2O	b.d.	b.d.	b.d.	b.d.	0.15	b.d.	b.d.	b.d.	b.d.	0.22	0.21	b.d.	b.d.
K2O	b.d.	b.d.	b.d.	b.d.	b.d.	b.d.	0.18	b.d.	b.d.	b.d.	b.d.	b.d.	0.28
Total	87.71	86.33	87.84	87.07	87.65	87.95	88.52	88.10	90.09	89.39	89.75	87.97	84.37
P5+	b.d.	b.d.	b.d.	b.d.	b.d.	0.012	0.013	b.d.	b.d.	b.d.	b.d.	b.d.	0.051
Si4+	2.882	2.889	2.881	2.924	2.792	2.946	2.948	2.889	2.882	2.916	2.873	2.926	2.932
Al3+	2.375	2.364	2.424	2.317	2.498	2.294	2.419	2.438	2.297	2.268	2.163	2.225	2.078
V3+	b.d.	b.d.	b.d.	b.d.	b.d.	0.040	0.042	0.049	0.094	0.238	0.324	0.371	0.510
Fe3+	0.165	0.164	0.176	0.167	0.199	0.173	0.151	0.183	0.274	0.281	n.a.	n.a.	n.a.
Mg2+	1.183	1.216	1.153	1.232	1.182	1.739	1.641	1.463	1.718	1.419	1.583	1.550	1.348
Ca2+	0.067	0.068	0.080	0.087	0.081	0.025	0.014	0.012	b.d.	0.015	0.039	b.d.	0.050
Mn2+	0.021	0.020	0.020	0.023	0.026	0.017	0.015	0.029	0.013	0.022	0.021	0.013	0.016
Fe2+	3.151	3.118	3.087	3.086	3.065	2.524	2.453	2.712	2.372	2.446	2.747	2.673	2.656
Zn2+	b.d.	b.d.	b.d.	b.d.	b.d.	0.013	0.020	b.d.	0.136	0.064	0.106	b.d.	0.019
Pb2+	0.006	0.007	b.d.	b.d.	b.d.	b.d.	b.d.	b.d.	b.d.	b.d.	0.005	0.019	0.016
Na+	b.d.	b.d.	b.d.	b.d.	0.032	b.d.	b.d.	b.d.	b.d.	0.045	0.044	b.d.	b.d.
K+	b.d.	b.d.	b.d.	b.d.	b.d.	b.d.	0.024	b.d.	b.d.	b.d.	b.d.	b.d.	0.041
Catsum	9.848	9.847	9.819	9.835	9.876	9.783	9.739	9.775	9.786	9.714	9.906	9.776	9.718
F/FM	0.73	0.72	0.73	0.71	0.72	0.59	0.60	0.65	0.58	0.63	0.63	0.63	0.66
T	138	135	122	118	153	98	75	108	88	51	n.a.	n.a.	n.a.
log fO2	−59.4	−59.7	−61.3	−62.2	−56.4	−63.6	−68.4	−62.4	−63.1	−71.1	n.a.	n.a.	n.a.

b.d.—below detection limit; F/FM = Fe²⁺/(Fe²⁺ + Mg²⁺); T—temperature (°C) calculated using chlorite compositional thermometry by [47]; log fO₂—oxygen fugacity calculated according to [47]; n.a.—not applicable.

).

Quartz occurs in two forms. Frequent are idiomorphic crystals (up to 2 cm in size) with milky white or light gray color, which either overgrow (and partly also along cracks replace) sulphidic aggregates (Figure 3f,g and Figure 4a,b) or accompany chlorite (Figure 3d). Less prominent are microscopic xenomorphic quartz grains, which replace, often together with chlorite and calcite, the older siderite (less frequently dolomite-ankerite) gangue, especially along their boundaries with other minerals (Figure 3a,b). In thin section, a distinct growth zoning of quartz megacrysts is visible, where clear parts alternate with bleary zones rich in fluid inclusions (Figure 4a,b).

White calcite grows on or replaces older phases or fills younger cavities or cracks (Figure 3a,d,e). Its chemical composition is characterized by a variable amount of rhodochrosite component (Cal_88–98Rdc_2–12Sid_0–2).

4.1.4. Baryte

Baryte was only found in one hand specimen, where it postdates sulphides and predates calcite (Figure 3f); however, its relationship to quartz, dolomite-ankerite, and phyllosilicates remains unclear due to the absence of crosscut evidence. Baryte forms milky white coarse-tabular crystals up to several cm in size in the central band of the vein with common ancient drusy cavities fulfilled by calcite (Figure 3f).

4.2. Fluid Inclusions

4.2.1. Petrography and Typology of Fluid Inclusions

The fluid inclusions were studied in six doubly polished plates prepared from three hand specimens covering the whole paragenetic evolution of the vein Bt23C. Fluid inclusions suitable for conventional microthermometric analysis were found in quartz, sphalerite, dolomite, calcite, and baryte. Three main types of fluid inclusions were distinguished based on their phase composition at room temperature: aqueous–carbonic (AC), carbonic (C), and aqueous (A).

The AC inclusions were found exclusively in the quartz host, where they usually occur within variably long trails (suggesting their pseudosecondary and secondary origin). In these trails AC inclusions almost always coexist with aqueous inclusions, whereas AC inclusions predominate, and A inclusions are relatively rare (Figure 4c–f). Sporadically, the AC inclusions were found also as solitary inclusions. In one section (Pb-666), which was cut parallel with crystallographic axis z of a quartz crystal, AC inclusions (as well as C inclusions) were found only in the central part of the quartz crystal in a limited area between two well-defined growth zones (Figure 4a,b). The AC inclusions have irregular, oval or negative-crystal shape with sizes from 6 to 55 µm (Figure 4c–f). At room temperature, they are either three-phase (aqueous solution + liquid carbonic phase + vapour carbonic phase) or two-phase (aqueous solution + liquid carbonic phase). Based on their filling ratio, they can be divided into two subtypes, which coexist on the same trails. In subtype AC₁, the aqueous phase occupies 50–90 vol.%, whereas in subtype AC₂, the aqueous phase takes less than 50 vol.%. Wide variations in mutual proportions of aqueous and carbonic phase are typical for inclusions present in individual trails (Figure 4c–f) as well as for solitary inclusions.

The C inclusions coexist with AC and A inclusions on trails in quartz host (Figure 4c). In addition, one pseudosecondary or secondary inclusion belonging to this type was also found in sphalerite. At room temperature, C inclusions appear to be monophase (liquid carbonic phase) or two-phase (liquid carbonic phase + vapour carbonic phase). Their shape is usually rounded or slightly irregular, with sizes between 6 and 32 µm. Interestingly, within some trails the C and AC inclusions richest in the carbonic phase clearly tend to concentrate in a narrow zone, which follows the places where change of the curvature of the ancient microfracture appeared (Figure 5).

The A inclusions are widespread in all studied minerals. Primary (P), pseudosecondary (PS) and secondary (S) inclusions were identified. In some inclusions in quartz and sphalerite, solids were also rarely present, which were identified by Raman spectroscopy as siderite and in one case as chlorite. Due to their nature and rare occurrence, it is evident that these solids are accidentally trapped phases, which have no influence on the behaviour of fluid inclusions during microthermometry, and thus their presence is neglected in the following text. According to their phase composition and association, the aqueous inclusions can be divided into three subtypes. The first subtype (A_c) represents inclusions in quartz coexisting on trails (PS or S character) with AC and C inclusions. These aqueous inclusions are mostly two-phase with up to 20 vol.% of vapour phase or liquid monophase. Their shapes are irregular, rounded or elongated with sizes up to 30 µm (Figure 4c–f).

The second subtype (A_L+V) represents two-phase aqueous inclusions, in which the vapour phase occupies 5–20 vol.% (mostly around 10 vol.%). They are abundant in quartz, where they occur on distinct growth zones (Figure 6a) or on trails. They are the most frequent type in sphalerite, where they occur mostly on trails (Figure 6c), rarely as solitary inclusions or sometimes on growth zones (Figure 6b), where they have a very small size (up to 6 µm) and have a very dark (sometimes even black) appearance making them very difficult objects for microthermometry. In calcite and dolomite, they occur in three-dimensional clusters or on trails. Their shape varies from irregular, rounded to negative-crystal shape (mostly carbonates, sometimes sphalerite) with sizes from 1 to 100 µm (mostly around 15 µm).

The third subtype (A_L) represents monophase liquid inclusions, that are the only type of inclusions in baryte, where they occur only on trails. They are the prevailing type in calcite, where they appear in three-dimensional clusters as well as on trails, often together with A_L+V inclusions. They are also abundant as secondary inclusions in sphalerite, dolomite, and quartz, where they also occur as primary inclusions on the outer growth zones coexisting with A_L+V inclusions. These inclusions have irregular or oval shapes and vary from 1 to 55 µm in size.

4.2.2. Microthermometry

Only four bulk homogenization temperatures were measured for AC inclusions ranging between 274 and 310 °C to both liquid (AC₁) and vapour (AC₂;

Table 3. Microthermometric data from A_c, AC, and C inclusions. Temperature parameters in °C, salinity in wt.% NaCl eq. n—number of measurements, F—degree of fill.

Sample	Mineral	Trail	Gen.	Type	n	F	Thtot	Thcar	TmCO2	Tmice	Tmcla	Salinity
Pb-666	Quartz	1	PS	AC1	2	0.80–0.85	287–295 (L)	14.1–27.7 (L)	−57.5		5.8–7.8	4.3–7.8
Pb-666	Quartz	1	PS	AC2	3	0.15–0.35	310 (V)	28.0–29.2 (V)	−57.5		3.5–8.2	3.6–11.3
Pb-666	Quartz	1	PS	C	2			27.7–29.0 (V)	−57.5
Pb-666	Quartz	1	PS	Ac	11	0.90–0.95	107–141 (L)			−10.1/−7.8		11.5–14.1
Pb-666	Quartz	2	PS	AC1	3	0.10–0.40		25.1–28.1 (V)	−57.5		3.6–4.0	10.6–11.2
Pb-666	Quartz	2	PS	C	2			26.2–28.1 (V)	−57.5
Pb-666	Quartz	2	PS	Ac	8	0.90–0.95	109–170 (L)			−8.5/−7.9		11.6–12.3
Pb-666	Quartz	3	PS	AC1	1	0.65		29 (V)	−57.5		3.5	11.3
Pb-666	Quartz	3	PS	C	2			28.5–29.1 (V)	−57.5
Pb-666	Quartz	3	PS	Ac	6	0.95	105–236 (L)			−8.7/−8.4		12.2–12.6
Pb-668	Quartz	4	PS	AC2	1	0.30		29.8 (V)	−57.2		8.9	2.2
Pb-668	Quartz	4	PS	Ac	1	0.80	268 (L)			−4.1		6.6
Pb-668	Quartz	5	PS	C	1	0.00		29.9 (L)	−57.4
Pb-668	Quartz	5	PS	Ac	1	0.80	210 (L)			−4.3		6.7
Pb-668	Quartz	6	PS	AC1	1	0.65		15 (L)	−57.5		9.1	1.8
Pb-668	Quartz	7	PS	AC1	1	0.55		29.9 (L)	−57.3		8.9	2.2
Pb-668	Quartz	8	PS	C	1			28.9 (V)
Pb-669	Quartz	9	PS	AC2	1	0.20		28.9 (V)
Pb-666b	Quartz	10	PS/S	AC1	3	0.7–0.8		−5.5/−2.1 (L)	−57.8		8.8–8.9	4.0–4.5
Pb-666b	Quartz	10	PS/S	C	1			30.0 (V)	−57.7
Pb-666b	Quartz	11	PS/S	AC1	1	0.75		−1.9 (L)	−57.8		8.8	4.1
Pb-666b	Quartz	12	PS/S	AC1	6	0.50–0.85		−9.3/−1.8 (L)	−57.9/−57.8		9.0–11.7	3.0–4.8
Pb-666b	Quartz	12	PS/S	AC2	5	0.10–0.45		−2.2/−1.7 (L)	−57.9/−57.8		9.0–9.4	3.0–3.8
Pb-666b	Quartz	12	PS/S	C	3			−2.3/−1.8 (L)	−57.9/−57.9
Pb-666b	Quartz	13	PS/S	AC1	3	0.50–0.60		−4.3/0.6 (L)	−57.3		8.9–9.4	3.3–4.3
Pb-666b	Quartz	13	PS/S	Ac	1	0.90	187 (L)
Pb-666b	Quartz	14	PS	C	2			28.6–30.2 (V)	−57.7
Pb-666b	Quartz	14	PS	Ac	1	0.95	140 (L)			−7.9		11.6
Pb-666b	Quartz	15	PS/S	C	3			27.7–30.3 (V)	−57.7
Pb-666b	Quartz	15	PS/S	Ac	2	0.95	93–117 (L)			−10.1/−7.9		11.2–14.1
Pb-666b	Quartz	16	PS/S	AC1	5	0.50–0.80		−11.5/−7.6 (L)	−57.8		8.8–10.8	4.1–5.1
Pb-666b	Quartz	17	PS/S	AC2	2	0.20–0.30		30.3–30.7 (V)	−57.4		7.5–7.8	4.3–4.9
Pb-666b	Quartz	17	PS/S	C	5			27.8–30.6 (V), 30.9 (L)	−57.4/−57.5
Pb-666b	Quartz	17	PS/S	Ac	1	0.95	193 (L)
Pb-666b	Quartz	18	PS/S	AC1	7	0.5–0.9		1.3–30.5 (L)	−57.5/−57.2		7.9–9.5	1.9–4.1
Pb-666b	Quartz	18	PS/S	AC2	11	0.1–0.4		14.0–29.5 (L), 29.9–30.8 (V)	−57.3/−57.2		5.5–8.4	3.2–8.3
Pb-666b	Quartz	18	PS/S	C	2			−0.6 (L), 29.8 (V)	−57.9/−57.3
Pb-666b	Quartz	18	PS/S	Ac	7	0.8–0.95	69–90 (L)			−7.2/−5.2		8.1–10.7
Pb-666b	Quartz	19	PS/S	AC2	1	0.10		30.1 (V)	−57.4		4.6	9.7
Pb-666b	Quartz	19	PS/S	C	2			29.5–29.8 (V)	−57.4
Pb-666b	Quartz	20	PS/S	AC1	1	0.50		26.3 (L)	−57.4		7.8	9.4
Pb-666b	Quartz	20	PS/S	C	2		274 (V)	29.6–30.8 (V)	−57.4
Pb-666b	Quartz	20	PS/S	Ac	3	0.85–0.95	55–185 (L)			−7.8/−7.2		10.7–11.5
Pb-666c	Quartz	21	P?	AC1	1	0.70		0.5 (L)			10.9
Pb-666b	Sphalerite	22	PS/S	C	1			28.8 (V)	−57.6

). Other inclusions would homogenize at even higher temperatures, but they decrepitated prior reaching the homogeneous state. The carbonic phase of AC and C inclusions froze around −100 °C. Melting of solid CO₂ was observed in a close range of temperatures between −57.9 and −57.2 °C, which suggest minor amounts of other low-condensable gases in addition to CO₂. Clathrate melting temperatures in AC₁ and AC₂ inclusions range between 3.5–11.7 °C and 3.5–9.4 °C, respectively; in C inclusions, either clathrate formation or melting was not observed. It is interesting that inclusions, where the carbonic phase homogenizes to vapour, yielded a lower and wider interval of Tm_cla (3.5–8.9 °C) than inclusions displaying Th_car to liquid (7.8–11.7 °C), no matter whether they were rich in the carbonic phase or aqueous solution. Partial homogenization of the carbonic phase occurs either to liquid (L) in wide temperature intervals ranging from −11.5 to 30.5 °C, −2.2 to 29.5 °C and −0.6 to 30.9 °C or to vapour (V) at temperatures 25.1–29.2 °C, 28.0–30.7 °C and 27.7–30.9 °C for AC₁, AC₂ and C inclusions, respectively. Within single trails (containing AC₁, AC₂ and C inclusions together), Th_car occurred in three modes: (i) only to liquid in a relatively narrow range (within 7.6 °C at maximum; four trails); (ii) only to vapour always in narrow temperature intervals (within up to 3 °C; six trails); (iii) in both homogenization modes in a wide interval (five trails, for example trail #1: 14.1–27.7 °C (L) and 27.7–29.2 °C (V); Table 3).

The A_c inclusions homogenized always to liquid in a wide interval of temperatures (55–193 °C). They froze out between −57 and −41 °C and remained colorless. Only one temperature of initial melting was observed at −37.0 °C. The last phase to melt was always ice, which disappeared at temperatures between −10.1 and −4.0 °C (with maximum variability in a single trail within 2.3 °C; Table 3).

All A_L+V inclusions homogenized to liquid. Primary and pseudosecondary inclusions homogenized at temperatures 60–220 °C, 162–205 °C, 175–195 °C, and 64–136 °C in quartz, sphalerite, dolomite, and calcite, respectively (

Table 4. Microthermometric data from aqueous A_L+V and A_L inclusions. Temperature parameters in °C, salinity in wt.% NaCl eq. n—number of measurements.

Sample	Mineral	Gen.	Type	n	Thtot (L)	Ti	Tmhh	Tmice	Tmcla	Salinity
Pb-666	Quartz	P	AL+V, AL	42	60–205	−49		−24.9/−0.1	−5.1/−4.8	0.4–25.7
Pb-666	Sphalerite	PS/S	AL+V	42	116–198	−50/−47	−34.0/−24.0	−25.9/−9.2	−9.6/−0.9	13.1–26.4
Pb-666	Calcite	P	AL+V, AL	16	120–136			−21.9/−0.2		0.4–24.8
Pb-667	Calcite	P	AL+V, AL	24	64			−10.2/−0.2		0.4–14.2
Pb-668	Quartz	P	AL+V	13	84–213			−10.4/−3.1		5.1–14.4
Pb-668	Quartz	PS	AL+V	11	137–220			−9.2/−0.9		1.6–13.1
Pb-668	Quartz	PS/S	AL+V	5	131–204			−11.2/−3.9		6.3–15.2
Pb-668	Sphalerite	P	AL+V	7	173–203	−50/−49	−24.7/−24.3	−14.0/−12.6	−7.8/−6.5	16.6–17.8
Pb-668	Sphalerite	PS	AL+V	41	173–196	−55/−49	−26.0/−24.0	−15.6/−11.0	−5.5/−1.7	15.0–19.2
Pb-668	Dolomite	P	AL+V	22	175–195			−12.1/−3.1		5.1–16.1
Pb-668	Dolomite	PS	AL+V	4	177–180			−11.4/−2.5		4.2–15.4
Pb-668	Dolomite	PS/S	AL+V	24	108–187			−8.6/−7.6	−4.8/1.3	11.2–12.4
Pb-669	Baryte	PS/S	AL	51			−26.1/−24.7	−19.6/−0.2		0.4–22.2
Pb-670	Calcite	P	AL+V, AL	15	82–109			−3.3/−0.1		0.2–5.4
Pb-666b	Quartz	P	AL+V	35	65–198	−49	−31.1	−27.2/−2.5	−17.2/−3.2	3.9–27.2
Pb-666b	Quartz	PS/S	AL+V, AL	31	93–219	−49	−24.5/−22.6	−14.5/−1.8	−9.1/+3.4	3.1–18.3
Pb-666b	Sphalerite	P	AL+V	1	205		−23.8	−12.5		16.5
Pb-666b	Sphalerite	PS	AL+V	3	190–197	−50	−26.2	−19.6/−19.1	−7.7/−7.4	21.9–22.0
Pb-666b	Sphalerite	PS/S	AL+V	11	105–188	−50	−24.7	−12.9/−12.7	−1.1/−1.0	16.6–16.8
Pb-666b	Sphalerite	S	AL+V, AL	7	105			−0.9/−0.2		0.4–1.6
Pb-666c	Quartz	P	AL+V	19	120–188			−14.3/−12.8	−4.0	16.7–18.1
Pb-666c	Sphalerite	PS	AL+V	7	162–188		−26.3/−25.8	−14.6	−4.2/−2.0	18.3

). The pseudosecondary or secondary (PS/S) inclusions homogenized between 93 and 219 °C. The A_L+V inclusions froze at temperatures from −88 to −30 °C and often became darker or brownish. Temperatures of initial melting for all genetic types of fluid inclusions ranged from −49 to −37 °C and −55 to −49 °C in quartz and sphalerite, respectively. Melting of hydrohalite was recorded for primary and pseudosecondary inclusions at −31.1 °C and from −26.3 to −23.8 °C, for pseudosecondary or secondary inclusions from −24.5 to −22.6 °C and from −33.2 to −24.0 °C in quartz and sphalerite, respectively (Table 4). In most inclusions, ice was the last phase to melt, which occurred at temperatures from −27.2 to −0.1 °C, from −19.6 to −11.0 °C, from −12.1 to −2.5 °C, and from −21.9 to −0.1 °C for primary and pseudosecondary inclusions in quartz, sphalerite, dolomite, and calcite, respectively. In pseudosecondary or secondary inclusions, ice melts from −25.9 to −7.6 °C. In ca. one third of A_L+V inclusions hosted by quartz, sphalerite and dolomite, clathrate was the last melting phase (Figure 6d–g); however, in some small, dark and/or low-salinity inclusions (especially those hosted by carbonates or sphalerite) the clathrate observation was very difficult, and it is possible that it was overlooked in some inclusions. Last crystal of clathrate melts at temperatures from −17.2 to −3.2 °C and from −7.8 to −1.8 °C for P and PS inclusions in quartz and sphalerite, respectively, and from −9.1 to 3.4 °C for pseudosecondary or secondary inclusions.

The A_L inclusions were artificially stretched by overheating to allow for the measurement of cryometric data in the presence of vapour phase. The freezing out of A_L inclusions occurred between −73 and −33 °C. Initial melting was observed only in pseudosecondary or secondary inclusions in baryte at −54.0 and −53.0 °C. Ice as the last solid phase melted at temperatures from −2.6 to −0.2 °C, from −12.8 to −0.2 °C, and from −11.8 to −0.8 °C for P and PS inclusions in quartz, calcite, and baryte, respectively. In pseudosecondary or secondary inclusions in baryte and in late secondary inclusions hosted by sphalerite, ice melted from −19.6 to −0.2 °C and from −0.9 to −0.2 °C, respectively (Table 4).

4.2.3. Raman Spectroscopy

The Raman analysis was used to determine the composition of the non-aqueous phase in C, AC, A_c, and A_L+V inclusions. The CO₂ is the predominating volatile in the carbonic phase of quartz-hosted AC and C inclusions (Figure 7). In most of them, the contents of CO₂ are 95–100 mol.%; only two inclusions of analyzed 43 contained a lower amount of CO₂ (87 and 88 mol.%). The other identified gases were N₂ (0–5 mol.%) and CH₄ (0–9 mol.%). Gaseous phase of A_c inclusions hosted by quartz has a similar composition (CO₂ = 86–98, N₂ = 0–3, CH₄ = 1–11 mol.%). The vapour bubbles in A_L+V inclusions in trails yielded a similar composition (CO₂ = 96–98, N₂ = 1–2, CH₄ = 1–2 mol.%), as well as that in a primary A_L+V inclusion in dolomite (CO₂ = 99, N₂ = 0, CH₄ = 1 mol.%). In contrast, A_L+V inclusions from the quartz-hosted growth zones seem to have very different compositions: while one inclusion from the central part of the crystal has predominant N₂ (68 mol.%) with 21 mol.% of CH₄ and 11 mol.% of CO₂, an inclusion from the edge of the same crystal was dominated by CH₄ (58 mol.%) with 31 mol.% of CO₂ and 11 mol.% of N₂. Gaseous phase of A_L+V inclusions in sphalerite has variable compositions, but CO₂ is always the prevailing component (CO₂ = 59–99, N₂ = 0–20, CH₄ = 1–36 mol.%). In many cases the acquisition of Raman spectra from water-dominated inclusions was hampered by the movement of usually small gaseous bubbles.

4.3. Stable Isotopes

The isotopic composition of C and/or O was determined in three carbonate samples and two quartz separates from the same samples that were used for the fluid inclusion study. The carbonate (calcite and dolomite-ankerite) samples gave very limited ranges of both δ¹³C (between −7.1 and −7.8‰ V-PDB) and δ¹⁸O (between 16.2 and 18.5‰ V-SMOW) values (

Table 5. The determined δ¹³C and δ¹⁸O values for carbonates and quartz from the vein Bt23C and fluid δ¹³C and δ¹⁸O values calculated for the ranges of homogenization temperatures of primary A_L+V fluid inclusions. Note that one Th outlier was neglected for quartz-r Pb-666.

Sample	Mineral	Paragenetic Stage	δ13Cmineral (‰ V-PDB)	δ18Omineral (‰ V-SMOW)	Temperature (°C)	δ13Cfluid (‰ V-PDB)	δ18Ofluid (‰ V-SMOW)
Pb-668	Dolomite	2	−7.8	16.2	175–195	−8.2/−8.9	+2.3/+3.6
Pb-667	Calcite	3	−7.1	18.5	50–64	−9.9/−10.5	−4.7/−2.6
Pb-670	Calcite	3	−7.8	16.3	82–109	−9.8/−10.1	−2.4/+0.6
Pb-666	Quartz-c	3	-	19.0	60–136	-	−8.8/+2.2
Pb-666	Quartz-r	3	-	19.4	68–156	-	−6.8/+4.5

c—core; r—rim of the crystal.

). Quartz also yielded mutually very similar δ¹⁸O values: a turbid central part and a colorless margin of a crystal gave 19.0 and 19.4‰ V-SMOW, respectively (Table 5).

5. Interpretation and Discussion

5.1. Composition, Trapping Mode, and P–T Conditions of Fluid Inclusions

5.1.1. AC, C, and A_c Inclusions

The chemical composition of the non-aqueous phase in AC, C and A_c inclusions is, except for three methane-richer outliers, CO₂-dominated and not very variable (0–3 mol.% CH₄ and 0–3 mol.% N₂). Moreover, salinities of A_c inclusions lie in quite close interval (6.6–14.1 wt.% NaCl eq., mostly around 11 wt.% NaCl eq.) compared to A_L+V and A_L inclusions from quartz (see below) and they produce a horizontal trend in a Th-Tm_ice plot (Figure 8b). The AC inclusions yield salinities 1.8 to 11.3 wt.% NaCl eq., which are mostly in a comparably wide interval as A_c inclusions. One measured temperature of initial melting (T_i = −37.5 °C) indicates the presence of Mg- and/or Fe-chlorides beside Na-chloride [48].

The huge variability in phase composition of trail-hosted co-existing fluid inclusions belonging to types C, AC and A_c in quartz host suggests that these inclusions might have been trapped from a heterogeneous fluid mixture. The occurrence of such fluids is welcome for geologic interpretation because it allows one to determine the trapping P-T conditions from microthermometric data only. However, wide variations of phase composition of a population of fluid inclusions can also be the result of some superimposed processes. Therefore, it is highly important to have solid evidence for heterogenous trapping. According to [49], the following four main criteria should be complied: 1. simultaneous trapping of all the inclusions of the population of interest; 2. very scattered degree of filling, homogenization temperatures, and compositions; 3. Th frequency distribution diagrams non-symmetrical and flattened, particularly towards high temperatures, but more or less similar for the liquid and the vapour; 4. serious evidence indicating that no leakage and/or necking-down can be suspected.

The inclusions with contrasting phase compositions occur on the same trails and thus seem to be formed contemporaneously (e.g., [32]). Moreover, the specific arrangement of fluid inclusions further supports an idea of heterogeneous trapping. This applies for some trails containing “linear” groups of inclusions richest in the carbonic phase, which concentrate in places where a change of curvature of ancient microfracture appeared (Figure 5). Such behaviour is consistent with the coeval trapping of a heterogeneous fluid mixture in an open fracture (future vein), where “bubbles” of carbonic fluid experienced some gravitational separation from aqueous solution due to their lower density. Consequently, bubbles of carbonic fluid can be trapped only in upward blinded “gravity traps” occurring within suitably shaped microfractures (case A in Figure 9). The criterion 3 is difficult to assess because not many bulk homogenization temperatures were measured due to the common limitations. There are usually only a few water-rich inclusions within each trail, and sometimes they are completely missing; homogenization data of water-rich inclusions are thus rare. Similarly, only a very limited amount of bulk homogenization data was obtained for aqueous–carbonic inclusions, most of which decrepitated prior to reaching the homogeneous state. Although very limited, the obtained homogenization data seem to follow the basic condition of wide variability. The involvement of post-entrapment alteration due to refilling, stretching, leakage or necking-down can be difficult to identify. The petrographic evidence for necking-down is missing in our case, as inclusions with very complex morphology or pairs of fluid inclusions still connected through a thin bridge were not observed. However, such petrographic evidence is not definitely unequivocal. Nevertheless, the occurrence of solitary inclusions with a comparably variable phase composition, such as show trail-hosted fluid inclusions, strongly supports the opinion that necking-down is not the key reason explaining the wide phase variability of the studied fluid inclusions. The absence of necking-down is also evidenced from differences in the salinity of aqueous solutions in neighbouring fluid inclusions, which was observed in a few trails (i.e., #1, #18; Figure 5). Partial leakage and stretching, if any, did not affect the studied inclusions significantly. Both these processes would alter the density of enclosed fluid, which is sensitively mirrored by the Th_car values. Fluid inclusions in two thirds of investigated trails show the minimum variability of this parameter (Table 3, Figure 8a), which is in line with a complete absence of secondary halos of micro-inclusions containing ejected fluid (e.g., [50,51]). In contrast, fluid inclusions in trails #1 and #18 show bimodal or wide variability in multiple parameters: in salinities of A_c inclusions, Th_car of AC and C inclusions, and/or Tm_CO2 of AC and C inclusions (Table 3, Figure 5 and Figure 8a). This is likely consistent with a longer healing period for these microfractures rather than with a multi-stage refilling of part of the present inclusions (no crosscuts by another trails of inclusions were observed in these cases). Although these trails do not follow the definition of FIAs, they are also useful because they illustrate highly dynamic changes of chemical composition and/or density of fluids in the studied hydrothermal system over a period somewhat longer than those represented by FIAs. In summary, we suggest that most of the fluid inclusions showing highly variable phase composition were not significantly disturbed due to post-entrapment alterations and thus represent a geologically meaningful record.

In a population of fluid inclusions that enclosed a heterogeneous fluid mixture, the minimum homogenization temperature of inclusions, which trapped an uncontaminated fluid end member (pure aqueous solution or carbonic phase), represents the true trapping temperature, and the fluid pressure can be located for this temperature at the isochore of non-aqueous fluid [52,53,54]. For P-T modeling, the lowest Th_tot of A_c inclusions (55–140 °C) of each trail were considered as true trapping temperatures; these values correspond very well with temperatures calculated from the semi-empirical chlorite thermometry, according to [47], applied to associated chlorite (51–153 °C; Table 2). Trapping pressures were obtained from isochores constructed for carbonic inclusions, which likely represent the non-aqueous fluid endmember. In such carbonic-rich inclusions, thanks to low trapping temperatures and very low solubility of water in the carbonic phase in low-P, low-T conditions (e.g., [55]), only a thin film of the aqueous phase can be expected, which is invisible under a polarizing microscope due to optical reasons [56], and therefore no bulk homogenization temperatures (Th) can be measured. Hence, the amount of water in this invisible (but essential for hermetization of the inclusions) aqueous rim was quantified according to experimental data about the solubility of water in CO₂ presented by [55] and included in the calculation of bulk compositions (

Table 6. Bulk chemical composition, bulk density and some microthermometric parameters for carbonic inclusions used for estimation of P-T conditions. Composition of non-aqueous phase is based on Raman analysis. Temperature parameters are in °C. Water contents in carbonic fluid were corrected according to experimental data by [55].

FI No.	Trail	Thcar	TmCO2	vol.% H2O	X (H2O)	X (CO2)	X (CH4)	X (N2)	D (g/cc)
2146	1	29.0 (V)	−57.5	0.2	0.0180	0.9325	0.0281	0.0214	0.249
2148	1	27.7 (V)	−57.5	0.2	0.0183	0.9513	0.0216	0.0088	0.247
2154	2	26.2 (V)	−57.5	0.2	0.0194	0.9466	0.0226	0.0113	0.231
2155	2	28.1 (V)	−57.5	0.2	0.0186	0.9295	0.0234	0.0286	0.240
2149	3	28.8 (V)	−57.5	0.2	0.0176	0.9465	0.0269	0.0089	0.255
2150	3	28.5 (V)	−57.5	0.2	0.0177	0.9510	0.0243	0.0070	0.255
3363	12	−2.3 (L)	−57.9	0.7	0.0198	0.9574	0.0117	0.0112	0.801
3380	14	30.2 (V)	−57.7	0.6	0.0457	0.9417	0.0065	0.0060	0.291
3381	14	28.6 (V)	−57.7	0.6	0.0495	0.9375	0.0053	0.0077	0.268
3383	15	30.3 (V)	−57.7	0.2	0.0154	0.9757	0.0040	0.0048	0.296
3384	15	27.7 (V)	−57.7	0.2	0.0163	0.9749	0.0040	0.0048	0.280
3409	18	−0.6 (L)	−57.9	0.4	0.0114	0.9697	0.0132	0.0055	0.791
3413	18	29.8 (V)	−57.3	0.1	0.0092	0.9878	0.0031	0.0000	0.302
3432	20	30.8 (V)	−57.4	0.1	0.0079	0.9784	0.0025	0.0112	0.292
3433	20	29.6 (V)	−57.4	0.1	0.0084	0.9779	0.0025	0.0112	0.276

), bulk densities (Table 6), and isochores of carbonic fluids (Figure 10). The obtained P–T conditions of the trapping of these fluid inclusions are visualized in Figure 10. The lowest pressure conditions were located at 100 bar and 55 °C (trail #20). Most of the trails had a pressure condition located between 122 and 170 bar with a narrow range for each trail (max. 18 bars), except for trail #18, where the pressure ranges from 115 to 268 bar and the temperature lies around 69 °C. Even higher-pressure conditions could be suggested for trail #12; however, no A_c inclusions were present in this trail, so only isochore of carbonic phase limits the uppermost possible pressure conditions (Figure 10).

The present-day position of the studied samples, which were taken in the mine from depths of around 900–950 m, requires minimum pressures of 90–95 bars under hydrostatic conditions. The higher pressures than these minimum possible values reflect the missing overburden formed by the rocks of the Teplá-Barrandian unit and its sedimentary cover, which were eroded since the time of mineralization (i.e., over time span from ~Permian to Quaternary). Whereas the thickness of eroded basement rocks is difficult to quantify, the original thickness of the Permian sedimentary cover in the adjacent Blanice Graben (Figure 1) is estimated at ca. 700 [18] or 2500 m [57]. Therefore, the calculated pressure variations can reflect either the changing thickness of this sedimentary cover due to sedimentation or subsequent erosion (in case if the duration of the mineralizing process took place over tens of million years), or fluctuations between lithostatic and hydrostatic fluid regime (in case if the duration of the mineralizing process was shorter than the velocity of sedimentation/erosion). In any case, the participation of the latter process can be potentially suggested from the highly variable densities of the carbonic phase of fluid inclusions recorded in trails #1, #17, and #18. Considering the unchanged fluid temperature, the pressures calculated for the minimum and maximum densities of carbonic fluid hosted by fluid inclusions of trail #18 yielded a maximum value (268 bar) ca. 2.5 times higher than the minimum one (115 bar), which is typical for fluctuations between hydrostatic and lithostatic load and participation of the crack-seal mechanism of vein development (e.g., [53,58,59]).

5.1.2. A_L+V and A_L Inclusions

Contrary to AC and A_c inclusions, the aqueous (A_L+V and A_L) inclusions are mostly considered to contain a homogenously trapped fluid phase due to limited ranges of homogenization temperatures. The only exception represents the primary A_L+V inclusions in quartz, which yielded most variable data scattered in both Th and salinities (Th = 60–220 °C, 0.4–27.2 wt.% NaCl eq.; Figure 8b). Wide differences in fluid salinity are usually attributed to different growth zones (Figure 4a), pointing to oscillatory changes in fluid source(s); a rare episode of mutual mixing of both low-salinity and high-salinity end-members is also illustrated in a growth zone containing inclusions with variable salinities within 13.5 wt.% NaCl eq. However, the broad variability is often observed in homogenization temperatures of aqueous fluid inclusions hosted within single well-defined growth zone (with Th ranges up to 145 °C), even in zones displaying essentially constant salinity. Because some of these inclusions contain Raman-detectable gases and/or display formation of clathrate during cryometry, we suggest that they likely trapped water-dominated members of a heterogeneous fluid mixture, similar to the above discussed A_c and AC inclusions. The difference is in a complete absence of inclusions rich in the carbonic phase. This feature, however, can be again potentially explained due to the gravity separation of both fluid endmembers differing in density and thus the formation of local microenvironments around growing crystal essentially devoid of larger “bubbles” of the carbonic phase (e.g., growth zone E in Figure 9), whose presence is a necessary pre-requisite for the formation of carbonic inclusions. Finally, it must be emphasized that fluid events giving rise to a majority of aqueous inclusions in quartz growth zones were not identical to those in which trail-hosted coexisting A_c, AC and C inclusions formed, as clearly indicated by differences in both the salinity of aqueous solutions and the composition of non-aqueous phase.

The temperatures of the initial melting of primary inclusions in quartz (from −49 to −37 °C) and sphalerite (from −55 to −49 °C) and for PS/S inclusions from baryte (from −54.0 to −53.0 °C) mostly suggest the presence of CaCl₂ [48], which is also likely present in dolomite and in some inclusions in calcite because of their low freezing temperatures. Projecting the Tm_hh and Tm_ice data into the phase diagram of the system NaCl–CaCl₂–H₂O (Figure 11), we can recognize at least three types of fluids. The first one is a high-salinity CaCl₂-rich fluid with bulk salinity between 22 and 24 wt.% NaCl eq. and with a NaCl/(NaCl + CaCl₂) ratio between 0.25 and 0.43, showing a faint subhorizontal trend in Figure 11 that corresponds to one primary inclusion in quartz and in one trail of PS/S inclusions in sphalerite. The second fluid is a high-salinity (16–21 wt.% NaCl eq.) NaCl-rich fluid with a NaCl/(NaCl + CaCl₂) ratio between 0.63 and 0.88 occurring in PS, PS/S inclusions in sphalerite and PS/S inclusions in quartz and baryte. The third one is a medium-salinity (10–11 wt.% NaCl eq.) NaCl-rich fluid occurring only in late secondary inclusions in quartz and baryte with NaCl/(NaCl + CaCl₂) ratio between 0.63 and 0.82.

The vapour bubbles of primary aqueous inclusions in quartz show completely different proportions of gases (i.e., N₂- or CH₄-dominated; Figure 7) than those of trail-hosted AC, C and A_c inclusions (CO₂-dominated). Such differences could suggest different sources of volatiles at different stages of mineralization, significant fractionation of gases during fluid migration and evolution, such as the consumption of some gas compound(s) during fluid evolution due to their reactions with wall rocks, earlier vein fill or other compounds present in the fluid. The latter case applies especially to CO₂, which can be reacted in the presence of aqueous solutions with limestones or other carbonates, forming bicarbonates well soluble in aqueous fluids [63]:

CaCO₃ + CO₂ + H₂O → Ca(HCO₃)₂

(1)

Similarly, methane is often mentioned as a reductant in hydrothermal systems [64,65,66], for example, for sulphate dissolved in fluids, giving rise to hydrogen sulphide necessary for precipitation of sulphides:

CH₄ + SO₄²⁻ → H₂S + CO₃²⁻ + H₂O

(2)

To further complicate the overall picture, the trail-bound aqueous inclusions hosted by quartz have a very similar chemical composition of non-aqueous phase, when compared with the AC, C and A_c inclusions, but a markedly wider range of salinity. This observation implies that occurrence of gases was not strictly coeval with only a single portion of aqueous fluid. The chemical composition of the non-aqueous phase of PS/S inclusions in sphalerite is quite variable and, in some inclusions, it is similar to inclusions in quartz and, in another ones, it is different, which can also manifest the re-filling phenomena, in addition to above-mentioned possibilities.

5.2. Isotopic Composition of the Fluids

In Table 5, there are listed the fluid δ¹⁸O and δ¹³C values calculated for the ranges of homogenization temperatures of primary fluid inclusions in the appropriate samples. Both calcites yield low fluid δ¹⁸O values between −4.7 and +0.6‰ V-SMOW, which suggests the predominance of surface waters (e.g., meteoric waters with negative δ¹⁸O values, seawater with near-zero δ¹⁸O values) that did not undergo significant isotope exchange with rocks at elevated temperatures [67]. Similar results, but with much greater ranges, also yielded both quartz samples (−8.8 to +4.5‰ V-SMOW). The early dolomite fluids have slightly positive δ¹⁸O values between +2.3 and +3.6‰ V-SMOW, which may be compatible either with a more significant isotope exchange of the above-mentioned surface fluids with rocks, or with some mixing of surface fluids with waters rich in ¹⁸O (e.g., magmatic waters or metamorphic waters; [67]). The fluid salinities decrease with the fluid δ¹⁸O values in carbonate samples (Figure 12a), which favors the mixing hypothesis involving two fluids with contrasting salinities and δ¹⁸O values. The quartz data yielded an opposite trend (Figure 12a); however, broad overlaps of fluid δ¹⁸O values of both quartz samples are obvious, which leads to serious uncertainty. Nevertheless, quartz median values also fall to near-zero δ¹⁸O area, but, in contrast to carbonates, there is a lack of correlation with fluid salinity. On the other hand, the fluid temperature shows good positive correlation when plotted against fluid δ¹⁸O for all studied samples (Figure 12c).

The fluid δ¹³C values calculated for all carbonates are much more consistent and range between −8.2 and −10.5‰ V-PDB (Table 5). Such values suggest a well-mixed source of carbon, originating potentially from limestones/marbles (δ¹³C close to 0‰ V-PDB), igneous or lower crust carbon (δ¹³C values between −5 and −8‰ V-PDB), and carbon derived from oxidation of organic matter (δ¹³C values between −20 and −35‰ V-PDB or even lower; [68]). Alternatively, it can be assumed that a significant intra-vein re-cycling of carbon took place during processes of dissolution of early vein carbonates, which were followed by precipitation of later carbonates from the same fluid. When plotting fluid δ¹³C values against fluid salinity or temperature, similar negatively correlated trends are observed in both cases (Figure 12b,d), which can only be explained by the mixing of fluids with contrasting characteristics.

5.3. Possible Sources of the Fluids

Most important fluid events recorded in the Bt23C vein were also reported from other parts of the Příbram ore area, implying for the same fluid source and evolution. Previous studies [2,4] have demonstrated that three basic types of aqueous fluids participated during the formation of ore deposits in the Příbram area.

The earliest reported fluid is a high-temperature (170–300 °C), high-salinity (15–24 wt.% NaCl eq.), high-δ¹⁸O (7‰–10‰ V-SMOW) Na>>Ca–Cl brine, which participated during the crystallization of the early portion of the siderite-sulphidic stage [2,4]. The ultimate origin of these fluids is not known, because although these fluids operated relatively soon after the ending of the Variscan metamorphic and magmatic activity, their composition is very different from those of metamorphic and/or magmatic fluids in the given area [29,62,69,70] (cf. composition of late-magmatic fluids from a porphyry-style Au deposit Petráčkova hora; Figure 11). The authors of [2,4] interpreted these fluids as deeply circulated waters, whose original signature was strongly modified due to a pronounced interaction with hot crustal rocks. In the Bt23C vein, these fluids were likely associated with the crystallization of early sphalerite, whose primary and pseudosecondary inclusions occupy an essentially identical field in the Th-Tm plot (Figure 8b). The end of this early stage could be associated with a decrease in salinity, homogenization temperature and/or fluid δ¹⁸O value [2,4], as is also manifested by Bt23C dolomite data (Figure 8b and Figure 12a,c).

The late portions of the vein paragenesis in the Příbram ore area crystallized from fluids, which were essentially sourced from surface waters. The most frequent are low-temperature (Th < 150 °C), low-salinity (<7 wt.% NaCl eq.), near-zero δ¹⁸O (−3 to 3‰ V-SMOW) fluids, whose origin is mainly associated with partly evaporated Lower Permian freshwater piedmont basins, developed in the area at the time of mineralization [2,4]. At the vein Bt23C, the content of numerous low-salinity, low-temperature fluid inclusions hosted by quartz and calcite, characterized by near-zero fluid δ¹⁸O values, can be related to this fluid source. The feasibility of the hypothesis dealing with the source of fluids in Permian basins is also supported by the V-rich compositions of phyllosilicates associated with quartz (Table 1 and Table 2), since vanadium is a characteristic element of Permian basins and their mineralizations in the wider area [71,72,73,74].

The third type of fluids reported by [4] is hosted by some secondary fluid inclusions in calcite. Fluid inclusions are characterized by low homogenization temperatures (<50–62 °C), high salinity (ca. 30 wt.% NaCl eq.) and Ca>Na–Cl composition. The authors of [4] interpreted these fluids as external basinal brines of marine provenance (e.g., [75]) or local upward migrating shield brines originating from the long-lasting interaction of meteoric waters and relatively cold crustal rocks [76]. This fluid endmember can be present in part of high-salinity inclusions hosted by quartz and calcite, namely, those with predominance of CaCl₂ over NaCl (Figure 11).

In addition, at least two other types of fluids participated during the formation of the vein Bt23C, up to date unknown from other sites in the Příbram ore area. The first one represents the low-temperature NaCl-dominated Na–Ca–Cl brines hosted by part of primary fluid inclusions in paragenetically late quartz and possibly also calcite (the Na/Ca ratios were not determined in calcite fluids). Although these fluids seem to have essentially identical chemical compositions to those of early NaCl-dominated Na–Ca–Cl brines hosted by sphalerite (Figure 11), we suggest that they must be predominantly sourced from a surface source because of their much lower δ¹⁸O values. These late fluids can be derived from marine sources, as was already above suggested for CaCl₂-rich Na–Ca–Cl brines. The Late Variscan to post-Variscan low-temperature high-salinity chloridic fluids with highly variable Ca/Na ratios and near-zero δ¹⁸O values are frequently reported from various types of vein mineralizations of the Bohemian Massif including the uranium deposits (e.g., [77,78,79]).

Another specific fluid is the quartz-hosted heterogeneous mixture of a carbonic phase and a medium-salinity aqueous solution. Such fluid was recognized for the first time in the Příbram ore area. The aqueous solution can represent a mixture of both the above-mentioned types of surface waters (i.e., chloride-rich brines and low-salinity basinal waters). The origin of associated volatiles can be explained in several ways. Firstly, gaseous components could be sourced from the deeper litosphere and migrated upward along deep fault structures. This may be valid especially for CO₂, because a significant portion of carbon bound in carbonates has a “deep” isotopic signature [2,4] and this work and CO₂ is usually a dominating compound among volatiles derived from lower parts of the Earth’s crust [31]. Secondly, some gases (including CO₂, CH₄, and N₂) could have been produced during the hydrothermal alteration of organic matter contained in sedimentary wall rocks and by radiolysis of hydrocarbons that were migrated together with aqueous fluids and gave rise to anthraxolite-bearing uranium mineralization, whose source is supposed in organic-matter rich horizons of the host Proterozoic sequences [80,81]. Thirdly, considering the participation of fluids from overlying Permian piedmont basins, we can assume that the source of volatiles (including CO₂, CH₄, and N₂) was due to the maturation of organic matter from coal seams, which occur at the base of sedimentary sequences of the neighbouring Blanice Graben [16,57]; sedimentary basins are in general the fertile sources of fluids with various gaseous compounds [53,82,83]. To date, we do not have enough information to allow for the unequivocal interpretation of the origin of volatiles, which is why further research is necessary to distinguish among the mentioned possibilities.

5.4. Constraints on Ore Genesis in the Příbram Area

The recognition of a fluid event characterized by activity of a CO₂-rich carbonic phase together with an aqueous solution during formation of a typical Zn-Pb ore vein has some implications on interpretations of the genesis of the base-metal mineralization in the Příbram ore area.

A widespread feature of ore veins in the Příbram uranium and base-metal district is the partial corrosion of older vein filling followed by the cementation of free space by younger hydrothermal minerals. The dissolution experienced especially by carbonates (siderite, dolomite-ankerite) and sulphides (galena, sphalerite, Pb-Sb sulphosalts) was evidently a process, which occurred several times during the paragenetic evolution of the vein (e.g., [4,20,84,85,86]). Across the district, the most prominent signs of dissolution were recorded around the drusy cavities, fractures, grain boundaries, growth zones and cleavage planes of a mineral (Figure 13). The textural evidence indicates that an intense dissolution was experienced also by siderite, dolomite-ankerite and sphalerite in the vein Bt23C (Figure 3a,b). The best developed corrosion cavities were filled up by mineral phases coeval with or postdating the activity of CO₂-dominated carbonic fluid. This suggests that the dissolution was caused by CO₂-induced increased acidity of aqueous fluids. In this case, the mechanism of dissolution of carbonates can be characterized by reactions such as those illustrated by Equation (1).

The corrosion of earlier mineral phases was often also associated with the fracturing of the vein. Hence, the invasion of corroding fluids would be connected with tectonic movements, which likely re-opened the migration paths from the source of carbonic fluids to the hydrothermal system of the originating vein. As such, the flow of carbonic fluid would be a relatively short episode, which is consistent with a relatively rare occurrence of fluid inclusions containing this fluid system.

The interaction of CO₂-rich carbonic fluid with earlier vein fill rich in sulphides and carbonates would lead to a progressive decrease in the acidity of the original fluid due to the consumption of CO₂ (Equation (1)) down to levels at which new portions of carbonates and/or sulphides begin to crystallize. In such a scenario, the substantial re-cycling of mineral-forming components is likely, which can be suggested from the existing chemical and isotopic data, although some alternative explanations could also be possible. The close δ¹³C values of carbonates (siderite, dolomite-ankerite, and calcite) from various mineralization stages were confirmed by this study and [4]. Furthermore, elevated contents of Mn are reported in all these carbonates, although they evidently crystallized from distinctly sourced fluids ([2,4], as well as this work). Similarly, all generations of sphalerite in the vein H32A are anomalously enriched in In and Sn [4] pointing to possible re-cycling of these elements by later fluids. The uniform Pb isotopic composition of galena from various paragenetic stages was interpreted in terms of the re-cycling of ore lead deposited in the earliest stage [2]. The very local re-cycling of vein components is also nicely mirrored in the chemical composition of chlorite: if chlorite crystallized in the vein portion containing Pb-Sb sulphosalts, it becomes enriched in Sb (H32A vein—[4]), whereas if it grew in another part of the same vein without Sb-minerals, it is Sb-absent (Bt23C vein—this work; H32A vein—[4]).

The slight increase in δ¹³C values of parent fluids, from which the paragenetically late calcite crystallized in the vein H32A (Figure 12b,d) but was unrecognized in the vein Bt23C (Figure 12b,d), can be related to isotope fractionation caused by the local CO₂ degassing of the fluids, as hypothesized by [4]. The pH changes of hydrothermal fluids caused by fluid-rock/gangue reactions involving CO₂ can be the reason for sulphur isotopic disequilibrium between co-existing galena and sphalerite, as suggested by [4] in case of the H32A vein.

The widely scattered Th and salinity data similar to those of the studied quartz also characterize aqueous inclusions hosted by quartz II from the base-metal vein H32A occurring in essentially identical paragenetic position (i.e., postdating dolomite-ankerite, contemporaneous with chlorite, and preceding calcite [4]). The H32A sample, however, contained only a very small amount of CO₂ in the fluid inclusions, detectable in vapour bubbles by Raman analysis, but not by corresponding phase transitions during cooling. Nevertheless, it seems plausible that the quartz-hosted CO₂-bearing fluid inclusions may represent a district-scale fluid episode. The traceability of this mineralizing event across the wider area is also highlighted by very low oxygen fugacities of parent fluids (log fO₂ ranging between −71 and −56 bar) calculated from the chemical composition of associated chlorite (Table 2, Figure 14), which are essentially identical for the veins Bt23C and H32A (Figure 14).

6. Conclusions

The Bt23C vein, hosted by folded Proterozoic clastic sediments in exo-contact of a large Variscan granite pluton, represents a typical development of Late-Variscan base-metal Zn>Pb mineralization of the Příbram ore area. Sphalerite and galena ores are hosted by gangue formed especially by Mn-enriched carbonates including early siderite, younger dolomite-ankerite, and late calcite. Quartz, baryte, chlorite, and V-rich mica (roscoelite to an unnamed V-analogue of illite) are subordinate components of the gangue. Banded and drusy textures of the vein were significantly modified by intra-vein dissolution and replacement by later hydrothermal mineral phases.

The fluid inclusion study conducted on sphalerite, dolomite-ankerite, calcite, baryte, and quartz revealed the overall decrease in fluid temperature (from ~210 to ~50 °C) throughout the paragenetic sequence, whereas salinity varied significantly from pure water up to 27 wt.% NaCl eq. A short episode characterized by the activity of a mixture of a medium-salinity aqueous solution and a CO₂-dominated carbonic phase was identified during the crystallization of quartz.

The C and O stable isotope study conducted on carbonates and quartz yielded low calculated fluid δ¹⁸O values (−4.7 to +3.6‰ V-SMOW), suggesting the predominance of surface waters, while fluid δ¹³C values (−8.2 to −10.5‰ V-PDB) are likely linked to well-mixed “deep” and organic sources of carbon.

Three sources of aqueous fluids are suggested based on available data: (i) early high-temperature high-salinity Na>Ca-Cl fluids from an unspecified “deep” source; (ii) late low-salinity low-temperature waters, likely infiltrating from overlying Permian partly evaporated freshwater piedmont basins; (iii) late high-salinity chloridic solutions with both high and low Ca/Na ratios, which can represent either externally derived marine brines, or local shield brines. The source of CO₂-rich volatiles can be seen in (i) deep crust, (ii) sedimentary wall rocks, and (iii) overlying Permian piedmont basins containing, in places, coal seams.

The newly recognized fluid episode dealing with an immiscible aqueous–carbonic fluid mixture allowed for the precise calculation of pressure conditions of ore formation (100–270 bar), which were poorly constrained up to now. Moreover, the occurrence of a CO₂-rich fluid episode gave possible reasons for some previously reported genetic aspects of the Příbram ores, including the widespread corrosion of the early vein fill, re-cycling of some vein components, S isotope disequilibrium among co-existing sulphides, and δ¹³C shifts due to CO₂ degassing.