The Kultuma Au–Cu–Fe-Skarn Deposit (Eastern Transbaikalia): Magmatism, Zircon Geochemistry, Mineralogy, Age, Formation Conditions and Isotope Geochemical Data

Abstract

The Kultuma deposit is among the largest and most representative Au–Cu–Fe–skarn deposits situated in Eastern Transbaikalia. However, its genetic classification is still a controversial issue. The deposit is confined to the similarly named massif of the Shakhtama complex, which is composed mainly of quartz monzodiorite-porphyry and second-phase monzodiorite-porphyry. The magmatic rocks are characterized by a low Fe₂O₃/FeO ratio, low magnetic susceptibility and belong to meta-aluminous, magnesian high-potassic calc-alkalic reduced granitoids of type I. The results of ⁴⁰Ar-³⁹Ar and laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) U-Pb dating showed that the formation of magmatic rocks proceeded during the Late Jurassic time: 161.5–156.8 Ma. Relatively low Ce/Ce*, Eu/Eu* and Dy/Yb ratios in the zircons indicate that the studied magmatic rocks were formed under relatively reduced conditions and initially contained a rather low amount of magmatic water. A mineralogical–geochemical investigation allowed us to outline five main stages (prograde skarn, retrograde skarn, potassic alteration, propylitic (hydrosilicate) alteration and late low-temperature alteration) of mineral formation, each of them being characterized by a definite paragenetic mineral association. The major iron, gold and copper ores were formed at the stage of retrograde skarn and potassic alteration, while the formation of polymetallic ores proceeded at the stage of propylitic alteration. The obtained timing of the formation of retrograde skarn (156.3 Ma) and magmatic rocks of the Shakhtama complex, along with the direct geological observations, suggest their spatial–temporal and genetic relationship. The data obtained on the age of magmatic rocks and ore mineralization are interpreted as indicating the formation of the Kultuma deposit that proceeded at the final stages of collision. Results of the investigation of the isotope composition of S in sulfide minerals point to their substantial enrichment with the heavy sulfur isotope (δ³⁴S from 6.6 to 16‰). The only exclusion with anomalous low δ³⁴S values (from 1.4 to 3.7‰) is pyrrhotite from retrograde skarns of the Ochunogda region. These differences are, first of all, due to the composition of the host rocks. Results of the studies of C and O isotope composition allow us to conclude that one of the main sources of carbon was the host rocks of the Bystrinskaya formation, while the changes in the isotope composition of oxygen are mainly connected with decarbonization processes and the interactions of magmatic fluids, host rocks and meteoric waters. The fluids that are responsible for the formation of the mineral associations of retrograde skarns and the zones of potassic alteration at the Kultuma deposit were reduced, moderately hot (~360–440 °C) and high-pressure (estimated pressure is up to 2.4 kbar). The distinguishing features of the fluids in the zones of potassic alteration at the Ochunogda region are a lower concentration and lower estimated pressure values (~1.7 kbar). The propylitic alteration took place with the participation of reduced lower-temperature (~280–320 °C) and lower-pressure (1–1.2 kbar) fluids saturated with carbon dioxide, which were later on diluted with meteoric waters to become more water-rich and low-temperature (~245–260 °C). The studies showed that the main factors that affected the distribution and specificity of mineralization are magmatic, lithological and structural–tectonic ones. Results of the studies allow us to classify the Kultuma deposit as a Au–Cu–Fe–skarn deposit related to reduced intrusion.

1. Introduction

The Kultuma deposit is situated in Eastern Transbaikalia, which is one of the oldest gold-mining regions of Russia. Many gold ore and gold-bearing complex deposits are known within this region. These deposits are paragenetically connected with Late Jurassic–Early Cretaceous magmatism. During recent years, a substantial increase in gold resources was achieved due to exploration and revaluation of gold–copper–iron–skarn deposits. These include the Bystrinskoye, Kultuma and Lugokan deposits. All the listed deposits possess a number of similar features; the major ones of them are deposit localization in the nodes of intersections of seep-seated faults, spatial link with the magmatic rocks of the Shakhtama complex, close formation ages and similar mineral composition. In spite of the fact that the deposits have been studied for a long time by many researchers, there are still different points of view concerning their genetic typing. Some researchers consider these deposits to be typical representatives of skarn formation, and others relate them to the gold–copper–molybdenum–porphyry type or to the combined skarn–porphyry type related with adakites [1,2,3,4,5,6]. Our previous investigation over the Lugokan gold–copper–iron–skarn deposit and the Antiinsky ore occurrence, which are also confined to the massives of the Shakhtama complex, showed that there are many prerequisites to distinguish new types of gold ore deposits in the region, such as “reduced” porphyry copper–gold deposit (RPCG) and reduced intrusion-related gold deposit (RIRGD) [7,8,9]. Among the numerous skarn deposits of Eastern Transbaikalia, the Kultuma deposit is one of the most representative but least studied ones. One of the main features of the deposit is its complex nature. Therefore, the studies of the genetic features of the Kultuma deposit are extremely interesting for understanding the genesis of large, complex skarn deposits, either in Eastern Transbaikalia or over the world.

Most of the Au–Cu–Fe–skarn and Au–Cu–porphyry deposits of the World relate with igneous rocks of the magnetite series. A distinctive feature of the Kultuma Au–Cu–Fe–skarn deposit is a spatial and genetic relationship with igneous rocks of the ilmenite series. The gold deposits located within Central Alaska and Yukon and composing the Tintina gold-bearing province are the most famous and well-studied gold deposits related with reduced granitoids. In the present paper, we report new data on the geological structure of the deposit, provide a detailed consideration of the features of ore mineral composition, magmatism (in particular, the chemical composition of zircon), the data on the age of magmatic rocks and ores, geochemistry of stable isotopes, as well as the results of the studies of fluid inclusions, which will finally allow us to specify a genetic type of the Kultuma deposit.

2. Geological Setting

2.1. The Major Features of Geodynamics in the Western Part of the Mongol–Okhotsk Orogenic Belt

The Kultuma deposit is situated in the Argun–Shilka interfluve, within the boundaries of the Aga–Borzya structural formation zone of the Mongol–Okhotsk orogenic belt (Figure 1).

Two main stages are clearly distinguished in the history of the belt: Paleozoic and Mesozoic. The Paleozoic stage embraced the formation of its basic structural elements: the fragments of Early Paleozoic oceanic blocks, Middle Paleozoic subduction island-arc volcanic belts and Late Paleozoic marginal-continental volcanic belts, the areas of orogenic granite magmatism [12]. Closure of the Mongol–Okhotsk ocean and collision of the marginal continental complexes of the Siberian and North Mongol–Chinese continents occurred between the Early and Middle Jurassic. The processes connected with the collision—in particular, thrusting, folding, metamorphism and magmatism—proceeded during the Middle and the major part of the Late Jurassic. The transition to stretching processes accompanied by the formation of metamorphic nuclei and sedimentary depressions relates to the end of Late Jurassic–Early Cretaceous [11,13]. The reasons for Early Cretaceous intracontinental rifting could be either a stretching of the continental crust after collision or a mantle convection, connected with the active hot spot, which is overlapped with the continental lithosphere [10,11]. The majority of large gold deposits and gold ore occurrences are confined to the zone of conjunction of two continental plates (Siberian and North Mongol–Chinese). Few deposits are located at a substantial distance from the suture and reveal a relation to regional faults. Large gold deposits and ore occurrences tend to mainly occur near the Mongol–Okhotsk suture and regional fault zones due to their increased permeability both for ore-producing melts and for gold-bearing fluids [13].

A specific feature of the Mesozoic magmatism of thee Mongol–Okhotsk orogenic belt is its vast area, while chemical features and isotope composition indicate that melt sources were located not only in the crust, but also in the mantle. Magmatic rocks that are in close genetic and paragenetic relations with gold ore mineralization were formed from latite, high-potassium calc-alkaline and less frequently calc-alkaline magmas [13].

An opinion exists that the Mesozoic magmatism in Transbaikalia started from the intrusion of shoshonite–latite magma, which initiated melting of the continental crust with the formation of calc-alkaline and high-potassium calc-alkaline magmas in intermediate chambers, which is the reason for temporal and spatial alternation of the derivatives of these three kinds of magma, as observed in the majority of ore magmatic systems. The upwelling asthenosphere that had been formed through compression during the collision of continents became the source of shoshonite–latite magmas (according to the model described by [14,15]). The action of its hot matter on the crust caused fusion of intermediate (in particular, felsic) magmas with higher alkalinity [16].

Another opinion was expressed by Kovalenko, Kuzmin and Yarmolyuk [17], who suggested that the reason for magmatic and ore activity there was in the manifestation of the Central Asian mantle plume [18].

2.2. Geological Structure of the Deposit

In terms of metallogenic position, the Kultuma deposit is part of the Budyumkan-Kultuma ore region (Au, Cu, Pb, Zn, Sn). The first discoveries of the deposits and ore occurrences at the territory were made as long ago as in the end of the 18th century. The ore occurrences and deposits are confined to a narrow band of Lower Cambrian terrigenous-carbonate rocks and are situated to the east and south-east from the Kultuma massif, at a distance of 200 m to 6 km (Figure 2). Below, we will present a generalized brief characterization of ore occurrences and deposits.

All of the deposits are commonly situated at the contacts of carbonate rocks of the Bystrinskaya formation and terrigenous-carbonate rocks of the Ernichenskaya formation. Ore bodies have diverse shapes, from layered accumulations of ore material to irregular impregnation spots, veins and tubular bodies. Embedding rocks are most frequently brecciated. Among ore minerals, the most widespread ones are galena, sphalerite, various Pb-Sb sulfosalts (for example, boulangerite), pyrite, arsenopyrite and less frequently chalcopyrite and fahlore. Sulfide mineralization of a similar kind widely occurs also within the boundaries of the Kultuma deposit, which will be considered below. Ore bodies were composed mainly of quartz and calcite (most frequently veins), while the presence of amphibole (tremolite, actinolite), biotite, sericite, rutile and chlorite was reported by researchers previously. More than 300 t of ore were mined as a total, but only Pb and Ag were later extracted from the ore. Sporadic detection of cerussite crystals decorated with native gold was reported [19].

The Kultuma deposit itself is confined to the massif of the same name, which is a part of the Shakhtama complex, intersecting the terrigenous-carbonate sediments of Late Proterozoic (according to drilling data) and Lower Cambrian (Figure 2). In earlier works at the territory under investigation, in addition to the Kultuma deposit, the Ochunogda ore occurrence situated at the southern margin of the Kultuma massif had also been distinguished. At present, the Ochunogda ore occurrence is not distinguished as a separate unit but is considered as a part of one of the regions of the Kultuma deposit. In addition, the Preobrazhenskiy and Inzhenerniy regions are also distinguished at the deposit, which will be considered herein as a whole and referred to as the Kultuma deposit.

2.3. Host Rocks

The most ancient objects are the sediments of the Beletuyskaya formation (Vbl), exposed by deep boreholes. The formation includes phyllites, gneiss, sandstone, gravelites, thin-layered siltstones, argillites, sublayers and lenses of limestone, dolomite, rare horizons of rhyoliths, their tuff and tuff siltstone. Carbonaceous and graphite-containing siltstones and shale rock were also detected within the formation. The rocks of the Beletuyskaya formation underwent regional metamorphism of the green schist facies everywhere. Upward along the section, the sediments of the Bystrinskaya formation are bedded conformably; they are most widespread at the Kultuma deposit. The Bystrinskaya formation (Є₁bs) is composed mainly of dolomites, dolomitized limestone and limestone, and less frequent are the horizons of terrigenous rocks (sandstone, gravelites, siltstones, argillites and carbonaceous-calc schist). Dark-colored varieties of carbonate rocks contain admixtures of carbonaceous and clayey substances and lean sulfide mineralization. Sublayers of phosphate-bearing schist and limestone are also established in the section of the formation. Upward along the section, the Ernichenskaya formation is conformably bedded. It is most widespread at the Ochunogda region of the Kultuma deposit. The Ernichenskaya formation (Є_1-2er) is composed of metamorphized siltstones, silt sandstone, phyllites, graphite-containing and other schist kinds, carbonate rocks, sandstone with gravelite sublayers. The lower horizon is characterized by homogeneity and composed of alternating carbonaceous, graphite-containing schist with sublayers and lenses of carbonate rocks. Upward along the section, they are gradually replaced by phyllites, siltstones with sandstone and silt sandstone sublayers. Carbonate rocks differ from the objects in the Bystrinskaya formation by a substantial amount of carbonaceous substance. The described sedimentary complex was formed under the conditions of shallow, isolated basin (a passive margin) at the margin of the Siberian Continent [21]. The major part of polymetal deposits and occurrences developed within the boundaries of Eastern Transbaikalia are localized in the Vendian (Late Proterozoic)–Cambrian sedimentary complex.

2.4. Structural–Tectonic Features and Morphology of Ore Bodies

The main features of the tectonic structure of the deposit are defined by a combination of widespread folded and fractured faults of different orders and ages. The deposit is situated in the core part of the Kultuma–Ushumun anticline, which is confined to the node of intersection of the north-eastern Levo–Gazimur and north-western Bogdat–Boshagochinsky deep faults. The major importance for the morphology of ore bodies is held by high-order anticlines complicating the eastern wing of the regional Kultuma–Ushumun anticlinal structure. In the cores of anticlines, carbonate-terrigenous sediments of the Beletuyskaya and Bystrinskaya formations are exposed. Sediments of the Ernichenskaya formation fill brachysynclines and wings of anticlinal folds. Fractures directed to the north-west appear as crush zones with a thickness from several meters to several ten meters. Tectonic faults of sub-latitudinal direction expressed as vast zones of intense fracturing and cleavage are also observed within the deposit. In addition to the listed major systems of faults, local tectonic faults are also intensively manifested.

More than 20 ore bodies are detected at the deposit. Ore bodies at the Kultuma deposit have a complicated shape and lie conformably to the anticlinal structures of the embedding terrigenous-carbonate sediments (Figure 3). They are traced in the meridional direction; ore bodies are often split into several branches from 6 to 25 m thick, usually in the hinges of anticlinal folds. The shapes of ore bodies are diverse; they are usually represented at the Kultuma deposit by paste-like bodies, while at the Ochunogda region they are most frequently vein-like and lens-like. The main ore bodies at the Kultuma deposit are confined to skarn formations and are located at different distances from the contact of magmatic rocks with embedding terrigenous-carbonate sediments. Skarns and ore bodies confined to them have been detected either directly at the contact between the sediments of the Bystrinskaya formation and the Kultuma massif or at a distance of several hundred meters to a kilometer from it. Sometimes skarns are spatially close to smaller bodies, dykes of the Shakhtama complex and apophyses of the Kultuma massif. At the Ochunogda region, skarns are confined to rare lenses of the carbonate rocks of the Ernichenskaya formation.

3. Analytical Methods

3.1. Whole-Rock Geochemistry

The analyses of igneous rocks were performed at the Analytical Center for multi-elemental and isotope research SB RAS in Novosibirsk, Russia. SiO₂, TiO₂, Al₂O₃, Fe₂O₃, FeO, MnO, MgO, CaO, Na₂O, K₂O, P₂O₅ and H₂O⁻ were determined by wet chemical digestion methods. A sample was fused with a mixture of NaHCO₃ and Na₂Ba₄O₇ at 940 °C, and the alloy was leached with hot water and dissolved in dilute hydrochloric acid (1:3). Further, various methods were used to determine the content of oxides. They included SiO₂, TiO₂, Al₂O₃, P₂O₅ and FeO (colorimetric method), MnO, MgO, CaO, Fe₂O₃ + FeO (atomic absorption spectrometry), Na₂O and K₂O (atomic emission spectrophotometry of flame) and H₂O⁻⁺ (gravimetry). The SO₃ content was defined by an X-ray fluorescence method.

3.2. Mineral Chemistry

Mineral compositions were analyzed at the Analytical Center for multi-elemental and isotope research SB RAS in Novosibirsk, Russia, using a JEOL JXA-8100 electron microprobe (Jeol, Tokyo, Japan) with five wavelength dispersive spectrometers and an energy dispersive spectrometer. A detailed study of the features of the composition of minerals was carried out by the electron micro probe analysis (EMPA) method on JEOL JXA-8100, CAMEBAX-Micro devices. Operating conditions were: 20 kV accelerating voltage, 30 nA beam current and 10 s counting time. Pyrope (Mg₃Al₂(SiO₄)₃), olivine ((Mg,Fe)₂SiO₄), diopside (CaMgSi₂O₆), albite (NaAlSi₃O₈) and synthetic fluorophlogopite (KMg₃AlSi₃O₁₀F₂) were used as an internal standard for silicates. Detection limits: Fe, Mn, Mg, K, Cr, Al, Ca, Ni, Ti, F–0.01 wt.%; Si, Na–0.04 wt.%. Synthetic alloy of gold and silver (60:40), chalcopyrite (CuFeS₂) and cinnabar (HgS) were used in the analysis of native gold. Detection limits: Au, Ag, Cu, Hg–0.01 wt.%. During EMPA studies of sulfide minerals, the following internal standards were used: synthetic alloys of Au–Ag (60:40), Fe–Ni–Co and In–Sb, arsenopyrite (FeAsS), pyrite (FeS₂), chalcopyrite (CuFeS₂), cinnabar (HgS), greenockite (CdS), bismuthinite (Bi₂S₃), stilleite (ZnSe), altaite (PbTe), galena (PbS), jacobsite (MnFe₂O₄), cassiterite (SnO₂), sphalerite (ZnS) and antimony (Sb₂S₃). Detection limits: Ag, Hg, As, In, Sb, Se, Sn, Te, Zn, Cu, Ni, Co, Fe, Mn, S–0.01 wt.%; Cd, Pb, Bi–0.04 wt.%. Some analyses were done on scanning electron microscopes JSM-6510 (Jeol, Tokyo, Japan) and LEO 1430VP (Carl Zeiss, Oberkochen, Germany), equipped with energy dispersive spectrometers (EDS). Operating conditions were: 20 kV accelerating voltage and amperage 1.9–2.6 nA.

3.3. U–Pb Dating and Trace Element Composition of Zircon

U–Pb dating and trace element composition of zircon grains was performed at the Center for MultiElemental and Isotope Research SB RAS in Novosibirsk, Russia, using a sector field inductively-coupled plasma mass spectrometer Element XR (Thermo Scientific, Waltham, MA, USA) coupled with a UV Nd:YAG New Wave Research UP 213 laser system (New Wave Research, Inc., Fremont, CA, USA). Instrument settings were optimized using NIST SRM 612 synthetic glass to achieve maximum intensity of the ²⁰⁸Pb isotope, while keeping low oxide production ²⁴⁸ThO⁺/²³²Th⁺ ratios (<2%). ²⁰²Hg, ²⁰⁴(Pb + Hg), ²⁰⁶Pb, ²⁰⁷Pb, ²⁰⁸Pb, ²³²Th, ²³⁵U and ²³⁸U isotopes were scanned using E-scan mode. ²³²Th and ²³⁸U were scanned in triple mode, while other isotopes were scanned in counting mode. The diameter of the laser beam was 30 μm. Pulse frequency was 5 Hz. Laser energy density was 3.0–3.5 J/cm³. LA-ICP-MS data were processed using “Glitter” software (GEMOC [22]). U–Pb ratios were normalized with the use of natural zircon standards GJ-1 [23] and Plesovice [24]. Errors of isotopic ratios and ages are presented at a 1σ level. Ages were calculated according to concordia diagrams using the IsoplotR software [25]. The measurements of ²⁰⁷Pb/²⁰⁶Pb, ²⁰⁶Pb/²³⁸U, ²⁰⁷Pb/²³⁵U and ²⁰⁸Pb/²³²Th ratios were used to calculate U–Pb ages on the ²⁰⁶Pb/²³⁸U–²⁰⁷Pb/²³⁵U concordia diagram.

Trace-element content was analyzed using the same mass spectrometer as for U–Pb dating. The NIST 610 standard glass was used as the external calibration standard.

Cathodoluminescence investigations were carried out at the Analytical Center for Multi-Elemental and Isotope Research Siberian Branch (RAS, Novosibirsk, Russia) with a MIRA 3LMU SEM (TESCAN Ltd., Redruth, UK) equipped with an INCA Energy 450 XMax 80 microanalysis system (Oxford Instruments Ltd., Abingdon, UK).

3.4. Isotope Analysis

The sulfur isotope analyses were made for six samples of pyrrhotite, nine samples of chalcopyrite, six samples of pyrite, two samples of arsenopyrite, one sample of sphalerite, one sample of galena, one sample of tetrahedrite and for two samples of anhydrite. Sulfides were handpicked under a binocular microscope, and then sulfur isotopic ratios were analyzed. The separation of SO₂ from sulfide minerals for sulfur isotopic analysis followed the method proposed by Han et al. (2002). The sulfur isotopic ratios were determined by using a mass spectrometer (Finnigan MAT (Thermo, Bremen, Germany) Delta dual inlet mode) at the Analytical Center for multi-elemental and isotope research SB RAS in Novosibirsk, Russia. The sulfur isotopic composition is expressed as δ³⁴S unit (‰) relative to Canyon Diablo Troilite standard, and its analytical precision is about ±0.2‰.

Carbon and oxygen isotope analyses were obtained using a Finnigan MAT-253 mass spectrometer at the Analytical Center for multi-elemental and isotope research SB RAS. The carbonates reacted with pure phosphoric acid to produce CO₂. The analytical precisions (2σ) are ±0.2‰ for δ¹³C value and ±0.5‰ for δ¹⁸O value. δ¹³C and δ¹⁸O values are reported relative to the Vienna Pee Dee Belemnite (V-PDB) and Vienna Standard Mean Ocean Water (V-SMOW), respectively.

3.5. Fluid Inclusions Study

Study of fluid inclusions was carried out in double-polished thin sections. Optical observation was conducting on a Zeiss microscope. Thermometric experiments were run on a microthermocamera Linkam THMSG-600 (Linkam, London, UK). It allows taking measurements in the temperature range from 196 to 600 °C and observing phase transitions in real time. Microthermometric measurements were calibrated using synthetic fluid inclusion standards for the freezing points of pure CO₂ (−56.6 °C) and pure H₂O (0 °C). Reproducibility of calibrations are ±0.6 °C for heating and ±0.2 °C for freezing. Gaseous and solid phase composition was defined by a Horiba Jobin Yvon LabRAM HR800 Raman microspectrometer (Horiba, Lille, France) equipped with a 532-nm Nd:YAG laser and an Olympus BX41 microscope. The RRUFF database and CrystalSleuth software [26] were used for mineral phase identification.

4. Results

4.1. Petrography of the Ore-Related Intrusion

The Kultuma massif forms a sill-like body, which is conformable with the layered and folded structure of the host rocks. It is composed mainly of quartz monzodiorites and monzodiorites of the second phase of the Shakhtama complex. The majority of magmatic rocks under study have porphyritic-like textures. They are represented by amphibole–biotite and hornblende–biotite varieties and have a porphyritic-like texture and hypidiomorphic (sometimes with the elements of micromonzonite and allotriomorphic), poikilitic texture of the main body. They contain phenocrysts of plagioclase, quartz, biotite, amphibole, hornblende and, less frequently, pyroxene. The major minerals of the main body are plagioclase, quartz, K–Na feldspar, biotite, amphibole and hornblende. Accessory minerals are apatite, titanite, zircon, while ilmenite occurs only rarely.

The dykes of quartz monzodiorite-porphyry and quartz diorite-porphyry of the Shakhtama complex are detected at the deposit and at a distance from it. The quartz monzonite-porphyry dykes are most frequently represented by biotite-hornblende varieties and have porphyritic-like structure and hypidiomorphic texture of the main body with the elements of micromonzonite and allotriomorphic texture. Phenocrysts are represented by plagioclase, quartz (idiomorphic crystals with dissolution figures), biotite, hornblende and pyroxene. The minerals detected in bulk included plagioclase, quartz, K–Na feldspar and hornblende. Accessory minerals are apatite, titanite, zircon and (only rarely) ilmenite. The dykes of quartz diorite-porphyry occur less frequently. They are represented by a biotite–pyroxene–amphibole variety and possess porphyritic texture and hypidiomorphic texture of the bulk material. They contain plagioclase, quartz, K–Na feldspar (idiomorphic crystals with dissolution figures), biotite and clinopyroxene. The bulk mass is composed of plagioclase, quartz, K–Na feldspar, biotite, hornblende and clinopyroxene. Accessory minerals are apatite, titanite and zircon, and less frequent are magnetite and ilmenite.

4.2. Bulk Compositions of the Ore-Related Intrusion

The bulk-rock major element composition is shown in the Table S1. The rocks of the massif are characterized by a broad range of SiO₂ content from 58.65 to 66.99 wt.% and fit within the field of intermediate rocks of the subalkaline series (Figure 4a). With respect to the alumina saturation index, the rocks are mainly metaluminious, less frequently peraluminious (ASI = 0.63–1.05) (Figure 4b) and, with respect to Fe-number (FeO_t/(FeO_t + MgO) = 0.42–0.54), they relate to magnesian granitoids (Figure 4c). K₂O content is 2.50 to 3.66 wt.% for monzodiorite-porphyry and 3.48 to 5.20 wt.% for quartz monzodiorite-porphyry. In the K₂O–SiO₂ diagram, quartz monzodiorite-porphyries fit in the fields of shoshonite and high-potassium calc-alkaline series, while monzodiorite-porphyries are only in the field of high-potassium calc-alkalinene series (Figure 4d).

The rocks of the dyke complex are characterized by SiO₂ content ranging from 60.54 to 63.51 wt.% and enter the fields of intermediate rocks of the normal and subalkaline series (Figure 4a). With respect to the alumina saturation index, the rocks relate to metaluminious (ASI = 0.77–0.87) (Figure 4b) and with respect to Fe-number (FeO_t/(FeO_t + MgO) = 0.47–0.55), to magnesian granitoids (Figure 4c). K₂O content is 3.23 wt.% for quartz diorite-porphyries and 3.64 wt.% for quartz monzodiorite-porphyries. At the K₂O–SiO₂ diagram, the rocks of the dyke complex are within the field of the high-potassium calc-alkaline series (Figure 4d).

Almost all magmatic rocks are characterized by the low Fe₂O₃/FeO ratio from 0.2 to 0.5 and are in the region of ilmenite (reduced) granitoids [27]. The only exclusions are the dykes of quartz diorite-porphyries for which the Fe₂O₃/FeO ratio is 0.59.

4.3. Magnetic Susceptibility

The data on magnetic susceptibility are a direct indicator of the average volume of magnetite in a rock. For instance, the rocks of ilmenite series formed under relatively reduced conditions should possess lower magnetic susceptibility in comparison with the rocks of the magnetite series formed under relatively oxidized conditions. Monzodiorite-porphyries and quartz monzodiorite-porphyries of the Kultuma massif are characterized by low magnetic susceptibility within the range (0.07–0.19) × 10⁻³ SI. The dykes of quartz monzodiorite-porphyries are also characterized by the low values: 0.14 × 10⁻³ SI. Unlike the latter, the dykes of quartz diorite-porphyries have higher magnetic susceptibility: 0.96 × 10⁻³ SI. It is known that the ilmenite series (reduced) includes the rocks with low magnetic susceptibility. For example, Ref. [33] assigned magmatic rocks with a magnetic susceptibility not higher than 0.5 × 10⁻³SI to the ilmenite (reduced) series, while [34] also considered magmatic rocks with higher magnetic susceptibility—≤3.0 × 10⁻³ SI as related to this series.

To ensure more correct assignment of magmatic rocks under investigation to the ilmenite or magnetite series, we plotted the diagram depicting Fe₂O₃/FeO at the ordinate axis and magnetic susceptibility values at abscissa axis. Other data on the magmatic rocks of the Shakhtama complex with known magnetic susceptibility values and Fe₂O₃/FeO are included in the diagram too. One can see in Figure 5 that monzodiorite-porphyries of the Kultuma massif, as well as the dykes of quartz monzodiorite-porphyries, fit (in two parameters) to the region of the ilmenite (reduced) series. Quite contrarily, the dykes of quartz diorite-porphyries fit to the region of the magnetite (oxidized) series on the basis of the values of Fe₂O₃/FeO ratios and magnetic susceptibility values (according to [33]), which is in agreement with petrographic data (the presence of magnetite among accessory minerals). With respect to two parameters, granodiorite-porphyries of the Lugokan massif (the Lugokan Au–Cu–Fe–skarn deposit) fit in the region of reduced granitoids. Monzodiorites of the Antiinsky massif fit the areas of both the reduced and oxidized granitoids.

4.4. Zircon Geochemistry

Integrated investigation of the indicator features of the elemental composition of a broad range of minerals is an essential element of genetic and forecast-survey models. Zircon is a common accessory mineral of many magmatic rocks. Investigation of the geochemical composition of zircon gives very important information on the chemical nature and the origin of magmas from which zircon was crystallized and on the role of magma in the formation of ore deposits. In addition, this mineral is a reliable indicator of the ore-bearing potential of magmatic rocks [35,36,37,38,39,40,41,42,43].

4.4.1. Zircon Cathodoluminescence (CL) Images

Cathodoluminescence (CL) images of representative zircon grains with laser spots are shown in Figure 6. The majority of zircons from the magmatic rocks of the Kultuma massif (quartz monzodiorite-porphyries and monzodiorite-porphyries) were 100 to 350 µm long, with the aspect ratios of 1:2, 1:3 and 1:4. Their textures were mainly bimodal, characterized by non-zoned or weakly zoned cores and clearly pronounced oscillatory zoning at the edges of grains (Figure 6a,b). Zircon grains were idiomorphic and prismatic, while oscillatory zoning was similar in shape with crystal shapes.

Zircons sampled from dykes (quartz monzodiorite-porphyries and quartz diorite-porphyries) were characterized by a smaller size (especially zircons from the dykes of quartz monzodiorite-porphyries, Figure 6c) and less clearly pronounced oscillatory zoning at the edges of grains. Zircons from quartz diorite-porphyries were 80 to 120 µm long, with the aspect ratios of 1:2 and 1:3. At the same time, zircons from the dykes of quartz diorite-porphyries (Figure 6d) were more similar in CL images with zircons from the rocks of the Kultuma massif. They were 120 to 250 µm long with an aspect ratio of 1:2. All zircon grains from the rocks of the dyke complex had more rounded shapes.

In summary, we may conclude that the parameters of all the studied zircons from the magmatic rocks of the Kultuma massif and the dyke complex are characteristic of the zircons of magmatic origin.

4.4.2. Zircon Trace Elements

The data on the content of trace and rare earth elements (REE) in zircon samples under investigation are listed in the Table S2. Zircons were isolated from 10 samples of magmatic rocks of the Kultuma deposit. The total number of zircon grains analyzed by means of laser ablation (LA-ICP-MS) was 110, but 29 among them were later excluded because they did not meet the requirements: La > 1 ppm, Ti > 50 ppm and Ba > 8 ppm [40]. The zircon separated from the dykes of quartz monzodiorite-porphyries, quartz diorite-porphyries and monzodiorite-porphyries. The dykes of quartz monzodiorite-porphyries, quartz diorite-porphyries and monzodiorite-porphyries occured.

All investigated zircon samples had similar chondrite-normalized REE patterns, characterized by heavy rare earth elements (HREE) enrichments and light rare earth elements (LREE) depletions with prominent positive Ce anomalies and relatively weak negative Eu anomalies. The REE content in the studied zircon samples was relatively low (ΣREE = 281–927 ppm). Chondrite-normalized spectra of REE distribution (C1 from [44]) shown in Figure 7 were characterized by a rather steep uprise from light REE to heavy ones, with an average (Yb/Sm)_N equal to 78 (the ratio (Yb/Sm)_N may be used to evaluate zircon enrichment with heavy REE [36]). The data obtained on the nature and degree of REE distribution in the studied zircon samples are typical for zircons of magmatic origin [45].

In the light-weight part of the REE distribution spectra, a clearly pronounced positive anomaly of Ce was observed (with the values ranging from 3 to 118 and the average value being 34). The Ce anomaly was determined from the ratio Ce/Ce*, where Ce is chonrdrite-normalized content of Ce and Ce* is a square root of the product of chondrite-normalized contents of La and Pr [46]. Zircons from the magmatic rocks of the Kultuma deposit were characterized by low Ce/Ce* values, which were within the range from 4.6 to 449.5 (97.6 on average) for quartz diorite-porphyries and 6.8 to 67.2 (27.5 on average) for monzodiorite-porphyries. The magmatic rocks of the dyke complex were also characterized by low Ce/Ce* values and were within the ranges: 5.6 to 114.1 (29.4 on average) for the dykes of quartz monzodiorite-porphyries and 5.6 to 181.4 (55.3 on average) for the dykes of quartz diorite-porphyries.

The Eu anomaly was calculated through the Eu/Eu* ratio, where Eu is chondrite-normalized Eu content, and Eu* is a square root of the product of chondrite-normalized Sm and Gd contents. All the studied zircon samples were characterized by a negative Eu anomaly varying within the range 0.15 to 0.62, with an average value of 0.28. Zircons from magmatic rocks of the Kultuma massif were characterized by low Eu/Eu* values, which were within the range 0.147 to 0.376 (0.264 on average) for quartz monzodiorite-porphyries and 0.209 to 0.321 (0.256 on average) for monzodiorite-porphyries. Unlike for the latter, the dykes of quartz monzodiorite-porphyries were characterized by higher Eu/Eu* values from 0.176 to 0.616 (0.511 on average). The values for the dykes of quartz diorite-porphyries were close to those for the magmatic rocks of the Kultuma massif, and the values of Eu/Eu* were 0.164 to 0.338 (0.259 on average).

Zircon crystallization temperatures were calculated with the help of the Ti-in-zircon thermometer according to Watson et al. [47]. The estimated temperature of zircon crystallization from the magmatic rocks of the Kultuma massif varied between 595 and 801 °C (638 °C on average) for quartz monzodiorite-porphyries and between 606 and 656 °C (638 °C on average) for monzodiorite-porphyries. Close values were also obtained for zircons from the magmatic rocks of the dyke complex: 648 to 719 °C (683 °C on average) for quartz monzodiorite-porphyries and 598 to 735 °C (639 °C on average) for quartz diorite-porphyries.

4.5. Mineral Composition and the Ore Formation Sequence

Several major types of ores with overprinted sulfide mineralization are distinguished at the Kultuma deposit: skarns, skarn-altered terrigenous carbonate sediments of the Bystrinskaya and Ernichenskaya formations and hydrothermally altered magmatic rocks of the Shakhtama complex and Ernichensky formation. The major ore minerals are magnetite, pyrite, chalcopyrite, pyrrhotite and arsenopyrite. The secondary minerals include sphalerite, galena and tetrahedrite; less widespread are marcasite, tennantite, bismuthinite, lollingite, alloclasite, siegenite, scheelite and cassiterite; more rarely occurring minerals are bournonite, boulangerite, cubanite, pekoite, bismuth sulfotellurides, ullmannite, molybdenite, aurostibite, native bismuth, native antimony, native gold, etc. Ore structures are represented by massive, disseminated, veinlet-disseminated, nest-disseminated and breccia-like varieties. Ore textures are fine-, medium- and coarse-grained. The total amount of ore minerals in the rocks varies substantially, from 5 to 20%, reaching almost 100% in some cases (in homogeneous massive magnetite ores).

The magnetite skarns are widespread at the deposit (Figure 8a); moreover, magnetite is often detected in phlogopite–magnetite (Figure 8b), amphibole–magnetite (Figure 8c,d), chondrodite–magnetite (Figure 8e) and other skarns of the retrograde stage. Olivine, pyroxene and garnet skarns of the progressive stage occur only sporadically. Serpentine (Figure 8f), serpentine–chlorite and other rocks related to the latest alterations of the skarns of prograde and retrograde stages also occur. Skarns are rare at the Ochunogda region; quartz–chlorite–calcite and quartz–chlorite–K–Na-feldspar (±epidote and apatite) metasomatites mainly occur. Sulfide ores are overprinted, regarding the host rocks. At the Kultuma deposit, widespread ores are magnetite (Figure 8g), magnetite–sulfide (Figure 8h) and sulfide. Their major mineral assemblage types are magnetite–chalcopyrite–pyrite, chalcopyrite–pyrrhotite–pyrite (Figure 8i) with the quantitative variations of the main minerals content (magnetite, chalcopyrite–pyrrhotite–magnetite etc.). They are characterized by massive, disseminated and veinlet-disseminated structures. Arsenopyrite–chalcopyrite–pyrite and polymetallic ores occur more rarely. Polymetallic ores are most frequently confined to skarn-altered dolomites (Figure 8j) and are located at a larger distance (in comparison with magnetite and magnetite–sulfide ores) from the direct contact between the magmatic rocks of the Kultuma massif and the host terrigenous carbonate rocks. Much more rarely, polymetallic mineralization occurs within quartz veinlets and veins among the altered magmatic rocks of the Kultuma massif. Molybdenum mineralization is also confined to these veinlets (Figure 8k). Unlike for the Kultuma deposit, arsenopyrite–chalcopyrite–pyrrhotite-pyrite ores dominate at the Ochunogda region (Figure 8l) with the quantitative variations of the content of main minerals: arsenopyrite (Figure 8m), pyrrhotite (Figure 8n), chalcopyrite–arsenopyrite (Figure 8o) and others. Sulfide minerals are frequently confined to quartz–chlorite–calcite veinlets and veins. The dominating structure is veinlet-disseminated, while the ores with massive structure occur rarely. Almost massive pyrrhotite and arsenopyrite–chalcopyrite ores with massive structure sometimes occur in the skarns and skarn-altered rocks.

4.5.1. Hydrothermal–Metasomatic Alteration of Host Rocks. Prograde Skarn

Prograde skarn is represented most frequently by the relicts of magnesian and (more rarely) calcareous (anhydrous) skarns, which are almost completely substituted by the skarns of the retrograde stage. We relate olivine, garnet and pyroxene skarns to the earliest prograde skarns. At present, it is difficult to establish the primary composition of magnesian and calcareous prograde skarns because of later processes, which strongly affected initial rocks. Among all the samples, the relicts of prograde skarns were established only in a few samples: the relicts of pyroxene (diopside and hedenbergite) and Fe-olivine, which are almost completely replaced by the skarns of the retrograde stage.

Olivine occurs extremely rarely in the form of grain relicts, which are almost completely replaced by later minerals. According to data previously obtained by researchers, olivine skarns only occur occasionally.

Garnet occurs very rarely as well, in the form of separate idiomorphic grains and their assemblies; it was more rarely represented by the grains of irregular and rounded shapes. Garnet skarns are not widespread at the deposit; garnet occurs most frequently within the relict pyroxene–garnet skarns.

Clinopyroxene was represented by diopside and hedenbergite; it occurs most frequently as the relicts of separate grains and their accumulations, which are almost completely replaced at the retrograde stage by tremolite (diopside) and actinolite (hedenbergite). Clinopyroxene relicts were detected in diopside (hedenbergite)–tremolite (actinolite)–phlogopite skarns.

4.5.2. Retrograde Skarn and Potassic Alteration

Retrograde skarns are the most widespread skarns at the deposit. They are represented mainly by chondrodite, amphibole and phlogopite skarns with variations of major mineral content (amphibole–chondrodite, amphibole–phlogopite, etc.). It is also necessary to note that prograde and retrograde skarns are mainly widespread at the Kultuma deposit, while skarns occur only rarely at the Ochunogda region.

Chondrodite (Table S3) consists either of small-grain aggregates or of separate rounded or irregular-shaped grains with clearly pronounced polysynthetic twinning. In addition to chondrodite norbergite (Table S4), another mineral of humite group was also detected in the skarns. Rounded norbergite aggregates are frequently accompanied by accumulations of sulfide minerals—chalcopyrite and pyrite (Figure 9a,b). According to EMPA data, an FeO admixture is permanently present in chondrodite at a level of 4.4 to 5.9 wt.%, while FeO content in norbergite is 1.2 to 2.5 wt.%. Humite group minerals are components of chondrodite, chondrodite–norbergite and tremolite–chondrodite skarns (Figure 9c) detected in the northern part of the deposit. Chondrodite and norbergite are associated mainly with fluorophlogopite (up to 5–6 wt.% F according to SEM data), tremolite, chlorite, serpentine, talc (Fe up to 5–6 wt.% according to SEM data), apatite (Figure 9d), fluorite (Figure 9b,d), magnesite and magnetite. Humite group minerals might have replaced early olivine at the retrograde stage. Then, chondrodite and norbergite were replaced by serpentine, chlorite and various carbonates. However, one more mechanism of the formation of humite group minerals is possible: the high activity of fluorine may stabilize many hydrous minerals at high temperatures at early stages of skarn formation [48].

Phlogopite is widespread at the deposit; sometimes it forms nearly monomineral skarns in which its content is up to 90–100% (Figure 9e). It also occurs in the major part of amphibole skarns in which its content is rarely above 10%. Phlogopite forms nests, rosette-shaped, laminated aggregates and small scales, which are distributed non-uniformly over the entire rock. The size of separate scales is usually no more than several centimeters. Phlogopite is often associated with tremolite, chlorite and apatite, rarely with anhydrite. Fluorophlogopite is characteristic of chondrodite (± norbergite) and tremolite–chondrodite skarns.

Amphibole is represented in the studied samples both by tremolite and by actinolite (Figure 9f). They form long prismatic, needle-shaped and, more rarely, radial and sheaf-shaped crystals not larger than several millimeters in size (Figure 9c). Tremolite contains insignificant admixtures (wt.%): FeO 1.64 to 2.37%, Na₂O up to 0.1% and Al₂O₃ up to 0.29% (according to EMPA data, Table S5). Some tremolite grains exhibit zonal structure, which is expressed as a decrease in FeO content from the center (3.23 wt.%) to the edges of crystal (2.68 → 1.56 wt.%) and an increase in MgO content (21.56 → 22.02 → 23.05 wt.%; according to SEM data). Actinolite also contains minor admixtures of Na₂O up to 0.3 wt.%, Al₂O₃ up to 2.1 wt.% and MnO up to 0.14 wt.% (according to EMPA data, Table S6). Tremolite and actinolite skarns are among the most widespread skarns at the deposit. Tremolite is mainly associated with phlogopite, serpentine and various carbonates and, more rarely, with chondrodite, norbergite and, much more rarely, with anhydrite. Actinolite forms both monomineral accumulations (retrograde skarn) or is observed together with chlorite, siderite and other carbonates (hydrosilicate (propylitic) alteration).

Potassic (calcic-potassic) alteration assemblage had minor development. It occured in intersecting veins and veinlets in the altered magmatic rocks of the Shakhtama complex. Quartz-potassic feldspar–carbonate veins with molybdenite were rare within the Kultuma massif. At the Ochunogda region, they were composed mainly of quartz, potassic feldspar, carbonate, biotite, chlorite and rarely of apatite. Sulfide minerals were represented most frequently by arsenopyrite, chalcopyrite and pyrrhotite. They were rarely overlaid on retrograde skarns, forming the intersecting quartz–biotite veinlets and veins with pyrite and chalcopyrite. Anhydrite was one more typical mineral of potassic hydrothermal alteration. It was detected only in one borehole. It occured as a matrix of breccias or filled cracks and veins.

Anhydrite occured as a granular mass and formed nests (breccia matrix) and veinlets in tremolite, tremolite–phlogopite and serpentine skarns. It was associated with tremolite, chlorite and phlogopite, as well as with various carbonates (Figure 9i).

K-feldspar was detected most frequently in quartz–carbonate (±chlorite, biotite, and apatite) veins and veinlets in hydrothermally altered magmatic rocks of the Shakhtama complex. It formed aggregates with irregular shapes.

Biotite was established in quartz–carbonate-(±K-feldspar, chlorite, apatite) veinlets and veins. It was composed of small scales, elongated grains and continuous dark-green scaly-granular aggregates.

4.5.3. Hydrosilicate (Propylitic) Alteration

We suppose that the latest alteration includes serpentization, chloritization and other processes that are generally related to hydrosilicate (propylitic) alteration of retrograde and prograde skarns. The typical minerals of this stage are serpentine, chlorite, talc, tourmaline, various carbonates, etc. It is often difficult to detect the minerals typical for retrograde skarns and the skarns of the propytilic stage in the samples because of their spatial overlap and combination of mineral associations. As a result, it is quite probable that some minerals characteristic of the retrograde stage (for example, actinolite, phlogopite and biotite) were also formed at the later stages. We suppose that propylitization processes at the Ochunogda region led to the formation of quartz–chlorite–carbonate (±epidote, rutile, apatite and significantly smaller amounts of K–Na-feldspar and biotite, in contrast to potassic alteration) metasomatites, which occur as veinlets and veins in the altered magmatic rocks of the Shakhtama complex and Ernichensky formation. The above-listed metasomatites contain various ore minerals—arsenopyrite, pyrrhotite, chalcopyrite, sphalerite and native gold.

Serpentine was not rare, sometimes forming dense, greenish-brown rocks. The mineral composition of these rocks is variable because, in addition to serpentine, they contained chlorite and various carbonates (most frequently magnesite) and relict minerals diopside, chondrodite, phlogopite, etc. There were monomineral serpentine rocks, as well as tremolite–serpentine, tremolite–chondrodite–serpentine, etc. Serpentine is composed of small-scale, thin-lamellar (Figure 9a,b,g) and cryptocrystalline aggregates and less frequently of tangled fibrous masses. An admixture of Fe up to 2 wt.% was permanently detected in it by SEM data.

Chlorite was widespread. It is composed of scaly, lamellar, elongated grains (Figure 9f), small-leafy and cryptocrystalline aggregates. Chlorite was detected most frequently in amphibole and phlogopite–amphibole skarns, as well as in quartz–chlorite–calcite and other metasomatites.

Talc occured infrequently. It was detected in chondrodite, chondrodite–norbergite skarns and in skarn-altered dolomites together with tourmaline, chlorite, magnesite and magnetite. It is composed of lamellar, scaly aggregates (Figure 9g), and sometimes it forms dense masses. It is remarkable that micro-veinlets of fluorite were observed in some cases at the cleavage planes (Figure 9g). An admixture of Fe up to 5–6 wt.% was permanently present in talc (according to SEM data).

Rutile was detected in quartz–chlorite–carbonate veinlets and veins at the Ochunogda region. It was present as aggregates of an irregular shape and sometimes formed elongated grains.

Tourmaline was discovered in skarn-altered dolomites. It most frequently appeared as schorl or dravite. It formed short prismatic (Figure 9h), extended and radial crystals. In skarn-altered dolomites, it formed veinlets and disseminations and associated with chlorite, talc, Fe-magnesite and magnetite.

Epidote occured rarely in the studied samples. It was detected as prismatic elongated crystals in tremolite skarns and quartz–chlorite–carbonate (± K-Na-feldspar) metasomatites.

Carbonates formed the bulk mass of skarns, as well as veinlets and veins. Carbonates were represented mainly by siderite, calcite, dolomite and magnesite. Siderite permanently contained (according to SEM data) impurities (wt.%): Mg up to 5% and Ca up to 4%. A characteristic feature of carbonates was the presence of manganese. For instance, Mn admixture (wt.%, according to SEM data) was permanently present in calcite (up to 2–3%), dolomite (up to 2%), magnesite (up to 0.7%) and siderite (up to 0.7%). The intermediate minerals of the magnesite–siderite series (for example, Fe-magnesite etc.) were frequently observed.

Fluorite formed both veinlets or separate disseminations and micro-veinlets in chondrodite, chondrodite–norbergite (Figure 9b) and tremolite–chondrodite skarns (Figure 9d). Fluorite was usually represented by grainy masses, elongated aggregates and sometimes separate grains. Quite rarely, fluorite formed micro-veinlets confined to the cleavage planes of separate talc scales and rims around chalcopyrite aggregates.

4.6. Distinctive Features of the Major Ore Minerals

Magnetite was one of the major ore minerals. It was represented by several generations. It was detected in skarns and skarn-altered dolomites. Magnetite (I) (early generation) occured as massive homogeneous aggregates composed of separate cataclastic corroded grains. Magnetite was also present in the form of relict grains confined to the central parts of idiomorphic pyrite (Figure 10a), which replaced magnetite (I). Under hypergenic conditions, magnetite was replaced by fine-grained hematite aggregates. The relicts of ilmenite (Figure 10b) and spinel were detected as the products of solid solution decomposition in magnetite (I). Magnetite (II) was formed due to the replacement of pyrrhotite and was represented most frequently by fine-grained aggregates and grains of irregular shapes. According to SEM data, Mg admixture up to 1 wt.% was detected in magnetite (I).

Scheelite occured rarely as separate grains of irregular shapes or their accumulations in the ground mass bulk of retrograde skarn.

Cassiterite was rather poorly spread in the ores. It formed idiomorphic crystals in the ground mass of skarns. It was also detected as inclusions in pyrite and in the intergrain space and cracks in magnetite (I), together with chalcopyrite (I) and sphalerite (I).

Lollingite was one of the earliest sulfide minerals; it was detected in the ores of the Ochunogda region. It formed micro-inclusions of prismatic, elongated, oval and irregular shapes in arsenopyrite (Figure 10c), which means that lollingite, as a rule, had crystallized prior to arsenopyrite, and then it underwent corrosion and was replaced by the latter. Admixtures that were determined in lollingite by micro X-ray diffraction analysis were (wt.%): S up to 2.7%, Co up to 0.52% and Ni up to 0.68%. According to the SEM data, the concentrations of admixtures in some cases reached (wt.%): Co up to 4.73% and Ni up to 0.95%. The lollingite composition is given in the Table S7.

Arsenopyrite was observed in the form of nests, disseminated formations, micro-veinlets in skarns, skarn-altered terrigenous carbonate sediments, as well as in silicified monzonite-porphyry of the Shakhtama complex. Arsenopyrite was most widespread at the Ochunogda region, where it was one of the most widespread sulfide minerals. It occured less frequently at other regions of the Kultuma deposit. Arsenopyrite either formed monomineral accumulations of idiomorphic crystals with a rhombic section or associated with pyrite (I), with the structural signs of nearly simultaneous crystallization. At the same time, the structures with the substitution of pyrite (I) by arsenopyrite were observed in some samples. Arsenopyrite was also associated with lollingite, ±pyrite, alloclasite. Arsenopyrite crystals were cataclastic and cemented with later sulfite minerals (chalcopyrite, pyrrhotite etc.) (Figure 10d).

Marcasite, pyrrhotite, chalcopyrite, lollingite, bismuthinite, native bismuth and native gold were detected as inclusions in arsenopyrite. Arsenopyrite was not homogeneous in its chemical composition (Figure 11a,b). There were two types of arsenopyrite in composition: one was the region of compositions with As/S > 1 (arsenous kind), and the other was As/S < 1 (sulfurous kind). Arsenous arsenopyrite was determined in association with lollingite as veinlets and disseminations in skarns; it was widespread mainly at the Ochunogda region. At other regions of the deposit, it occured much more rarely, in the form of disseminations in amphibole skarns in association with lollingite, alloclasite and pyrite (I). It was characterized by the high content of As from 33.4 to 37.5 at. % (36.0 on average) and by As/S > 1 (As/S from 0.99 to 1.25, 1.14 on average). The admixtures determined in it were (wt.%) up to Ni–0.16% and Co–0.38% (according to EMPA data, Table S8). Sulfurous arsenopyrite formed both separate veinlets and disseminations in the altered monzonite-porphyry (the Ochunogda region), as well as disseminations and nests in skarns and skarn-altered rocks in association with pyrite (I) (the Kultuma deposit). The sulfurous kind of arsonopyrite was also established at the Kultuma deposit as intersecting veinlets in the altered magmatic rocks of the Shakhtama complex. Arsenopyrite was associated with pyrite (I) and later minerals of the polymetallic association (galena, sphalerite (II), chalcopyrite (II), tetrahedrite, etc.). It was characterized by the lowest As content from 30.2 to 32.45 at. % (31.31 on average) and the low As/S ratio from 0.82 to 0.94 (0.87 on average). Arsenopyrite from skarn and skarn-altered rocks in association with pyrite (I) at the Kultuma deposit contained a higher As concentration—from 31.5 to 32.97 at. % (32.20 on average) and higher As/S ratio from 0.89 to 0.96 (0.92 on average). As far as other impurities are concerned, Ni up to 0.44 and Co up to 0.98 wt.% occured very rarely (sole grains). Unlike for the Kultuma deposit, sulfurous arsenopyrite from the Ochunogda region contained Co up to 2.73 and Ni up to 2.59 wt.%; As content was from 31.03 to 32.88 at. % (31.85 on average), and the As/S ratio was from 0.87 to 0.97 (0.92 on average).

Pyrite is a subjacent sulfide mineral. It was detected in the form of nests, disseminations and veinlets in skarn, skarn-altered terrigenous carbonate rocks and silicified magmatic rocks of the Shakhtama complex. Pyrite was represented by several generations. Pyrite (I) formed idiomorphic crystals of cubic shape, their accumulations and aggregates with irregular shapes. Pyrite crystals were often cataclastic and corroded. Some pyrite grains (mainly the central parts of grains) contained large amounts of ore and non-ore mineral inclusions: magnetite (Figure 10e), cassiterite, chalcopyrite, pyrrhotite, galena, fahlore, native gold, amphibole, chlorite, calcite, dolomite and siderite. In rare cases, pyrite bore the signs of magnetite replacement and, in turn, was replaced by marcasite. The deposition of early pyrite took place together with arsenopyrite. The later generation of pyrite (II) was represented by allotriomorphic aggregates (pyrite + marcasite mixture), which substituted earlier pyrrhotite. Admixtures determined in pyrite were (wt.%) up to: As—2.24%, Ni—0.11%, Co—0.14%, Sb—0.06% and Cu—0.14% (according to EMPA data, Table S9).

Alloclasite occured as rare disseminations in skarns. It formed idiomorphic cubic and octahedral grains. Alloclasite appeared either as intergrowth with pyrite (I) and arsenopyrite or as rims over the edges of pyrite grains (I), thus substituting it. Native gold was detected in alloclasite in the intergrain space and in cracks. Admixtures revealed in it included (wt.%) up to: Fe—6.06%, Ni—1.92% and Cu—0.53% (according to the data of EMPA, Table S10).

Siegenite was insignificantly spread in the ores. It occured as disseminations in amphibole and amphibole–phlogopite skarns. Siegenite formed intergrowth with pyrite (I) (Figure 10f) and appeared as isometric grains of cubic shape. Sole grains of native gold were detected as inclusions in siegenite. Similar to pyrite (I), it was one of the earliest sulfide minerals. The composition of siegenite was determined by SEM (wt.%): Co (18.3–20.5%), Ni (27.2–30%), S (41.6–42.2%), Fe (4.4–5%) and Cu (4.4–5.7%).

Marcasite formed disseminations and micro-veinlets; it was represented by fine-grained allotriomorphic and sheaf-shaped aggregates, and sometimes by elongated needle-like grains. The major mass of marcasite formed marcasite-pyrite aggregates, replacing earlier pyrrhotite (I). In some cases, marcasite replaced fahlore, chalcopyrite (I) along aggregate boundaries, pyrite (I) and magnetite (I).

Pyrrhotite formed aggregates of irregular shapes and often occured as inclusions in earlier sulfide minerals (pyrite (I) and arsenopyrite). Pyrrhotite was most widespread at the Ochunogda region, where it formed practically massive aggregates (Figure 10g). Pyrrhotite was represented in the ores by several generations. Pyrrhotite (I) formed intergrowth with chalcopyrite (I), and less frequently with sphalerite (I). The aggregates of pyrrhotite (I), chalcopyrite (I) and sphalerite (I) filled interstices and cracks between early sulfide minerals (pyrite (I), arsenopyrite, etc.), which is evidence that they were formed later. Inclusions in pyrrhotite (I) were established to be scheelite, bismuthinite, sphalerite, bismuth sulfotellurides and native gold. Pyrrhotite was substituted with pyrite (II) and marcasite, with the formation of a bird-eye type structure (Figure 10h). At the Kultuma deposit, pyrrhotite was often detected as the relicts of allotriomorphic aggregates, which were almost completely substituted by pyrite–marcasite aggregates. Among admixtures, Ni in the amount up to 0.89 wt.% was determined (Ochunogda region) in pyrrhotite, according to SEM data. Pyrrhotite (II) associated with galena (Figure 12a), sphalerite (II), fahlore, chalcopyrite (II), cubanite, boulangerite, ulmanite, native bismuth and other minerals of the latest mineral association. It either formed intergrowth with galena, fahlore and chalcopyrite (II), or was observed as inclusions in galena and ulmannite.

Cubanite was rare. It occured in the form of the aggregates of irregular shapes and associated with galena, sphalerite (II), tetrahedrite, chalcopyrite (II), pyrrhotite (II), etc. Cubanite was established in the form of inclusions in galena (bismuth-containing) (Figure 12b). Xenomorphic segregations of native bismuth either formed inclusions in cubanite aggregates or were arranged along their boundaries. Cubanite also formed intergrowth with chalcopyrite (II), galena and fahlore.

Chalcopyrite is one of the major sulfide minerals. It was detected almost in all types of ores. Several generations of chalcopyrite were established. The early generation of chalcopyrite (I) was characteristic both for skarn formations (Figure 10i) and for quartz–chlorite–carbonate and other veins and veinlets (potassic alteration), where it formed xenomorphic segregations, bulk mass and massive aggregates of irregular shapes. Chalcopyrite (I) often contained thin dust-like, dotting and stellar sphalerite particles as the products of the decomposition of solid solution; rounded and worm-like pyrrhotite inclusions occured rarely. The presence of stellar (paw-shaped) products of the decomposition of the solid solution of sphalerite and pyrrhotite in chalcopyrite are classic evidence that chalcopyrite was formed at a high temperature. Chalcopyrite inclusions were detected in pyrite (I) and arsenopyurite. Chalcopyrite (I) formed intergrowth with pyrrhotite (I) and sphalerite (I). Early chalcopyrite cemented earlier ore minerals (magnetite (I), pyrite (I), arsenopyrite, etc.), often filling the space between grains or cracks and veinlets. The formation of the early generation of chalcopyrite was associated with the formation of one of the native gold generations. Chalcopyrite and native gold often filled the same cracks and veinlets in earlier ore minerals, which is evidence of the close times of their formation. They were often observed as inclusions in magnetite (I), pyrite (I) and arsenopyrite. The late generation of chalcopyrite (II) was characteristic of the polymetallic mineral association, where it forms small grains of irregular shapes. Chalcopyrite (II) most frequently formed intergrowth with galena, sphalerite (II), bournonite and fahlore. Chalcopyrite of this generation did not contain the products of solid solution decomposition; quite contrarily, it often was itself the product of the decomposition of solid solution in sphalerite. The chemical composition of chalcopyrite was close to the theoretical one, independently of generations (according to EMPA data, Table S11).

Bismuthinite occured mainly in skarn in the form of small grains of irregular and rounded shapes. It often formed inclusions in pyrrhotite (I) and rarely in siderite and quartz. Bismuthinite was often observed in pyrrhotite (I) in intergrowth with chalcopyrite (I) and native gold. According to the SEM data, Pb admixture was permanently detected in bismuthinite from 4.1 to 6.2 wt.%, and less frequently, Se admixture was detected at a level up to 2.4 wt.%.

Pekoite occured mainly in skarn as fine round-shaped grains. Pekoite was usually detected as inclusions in pyrrhotite (I) and arsenopyrite, and sometimes in siderite and quartz. It often formed intergrowth with native gold. In rare cases, pekoite was substituted by matildite (Figure 12c). Impurities that were detected in pekoite according to SEM data were (up to, wt.%): Pb—8.69%, Se—2.46%, Te—0.33%, Ag—0.39% and Cu—0.98%.

Bismuth sulfotellurides occured rarely in the form of fine (several micrometers), rounded disseminations in pyrrhotite (I). Sometimes, they formed intergrowth with native gold. They were also detected as inclusions in pyrrhotite (I). Due to very small dimensions of separate grains, their composition could not be established reliably (the Bi:Te ratio was 2:1).

Sphalerite was detected mainly in skarns and skarn-altered dolomites, sometimes in silicified magmatic rocks of the Shakhtama complex; several generations were revealed. The earliest generation of sphalerite was represented by aggregates of irregular shape in intergrowth with chalcopyrite (I) and pyrrhotite (I). Sphalerite (I) was also detected in the form of the products of decomposition of the solid solution in chalcopyrite (I) (stellar, paw-like segregations) and as inclusions in pyrrhotite (I) and pyrite (I). The major part of sphalerite (II) (later generation) was crystallized during the polymetallic stage. Sphalerite (II) formed aggregates of irregular shapes and xenomorphous segregations, often saturated with chalcopyrite inclusions—the products of the decomposition of solid solution of chalcopyrite in sphalerite. Sphalerite (II) most frequently formed intergrowth with galena, chalcopyrite (II), bournonite and fahlore. There were substantial differences in chemical composition between the earlier and later generations of sphalerite (Figure 13) (according to EMPA data, Table S12). For instance, sphalerite (I) in association with chalcopyrite (I) and pyrrhotite (I) was characterized by high FeS content from 15.5 to 20 mol. %. However, there were some cases when sphalerite in association with chalcopyrite (I) contained anomalously low amounts of FeS—from 3.1 to 3.4 mol. %. Quite contrarily, sphalerite (II) in association with later sulfide minerals of the polymetallic stage was characterized by lower FeS, unlike for sphalerite (I). FeS content varied between 3.1 and 12.1 mol. %. However, sometimes even within one sample, a strong variation of Fe content in sphalerite was detected. For example, Fe content in sphalerite (II) in association with galena, chalcopyrite (II), fahlore, pyrrhotite (II) and cubanite varied between 3.49 and 6.82 wt.%. Other elements present in sphalerite as impurities were (wt.%): Cu up to 2.46% (most frequently detected in sphalerite (I)), Cd up to 0.73% (it was present almost permanently), Co up to 0.19% and Mn up to 0.19%.

Galena occured as micro-veinlets and disseminations mainly in skarn-altered dolomites, sometimes in silicified magmatic rocks of the Shakhtama complex. It appeared as aggregates of irregular shapes and xenomorphic segregations. Galena at the Kultuma deposit, detected in skarn-altered dolomites, was associated with fahlore (tennantite and tetrahedrite), sphalerite (II), chalcopyrite (II), and bournonite; their aggregates cemented cataclastic grains of earlier ore minerals (Figure 10j) and filled intersecting cracks and micro-veinlets. Native gold was detected as inclusions in galena. Impurities that were permanently present in galena included Te (up to 0.27 wt.%) and Cd (up to 0.28 wt.%), and much more rare admixtures were Ag (up to 0.25 wt.%, sole grains) and Se (up to 0.07 wt.%) (according to EMPA data, Table S13). More rarely, galena associated with sphalerite (II), chalcopyrite (II), fahlore (tetrahedrite), pyrrhotite (II), cubanite, aurostibite, native gold and native antimony, which filled micro-veinlets and veins in hydrothermally altered magmatic rocks of the Shakhtama complex. The composition of galena of this association was different from galena in association with fahlore, sphalerite (II), chalcopyrite (II) and bournonite. Admixtures that were permanently present in it were (wt.%): Ag from 0.18 to 0.43%, Bi from 0.21 to 0.99%, sometimes Te up to 0.15% and, very rarely, Se up to 0.15% (sole grains). Galena at the Ochunogda region was associated mainly with boulangerite, chalcopyrite (II), ullmannite, pyrrhotite (II) and native bismuth. Inclusions detected in galena were chalcopyrite (II), cubanite, jamesonite, native bismuth and sometimes nuffieldite; native gold occured only rarely. Galena substituted bismuth-containing boulangerite (Figure 12d). Admixtures permanently present in galena (from the Ochunogda region) were (wt.%): Ag from 0.1 to 0.47%, Bi from 0.48 to 1.19%, Se from 0.13 to 0.21% and, more rarely, Te up to 0.14% (according to EMPA data).

Fahlore, similar to galena, was one of the most widespread sulfide minerals of later stages, and sometimes it played a dominant part in the quantitative aspect. Though fahlore was not widespread in the ores, it was detected in small amounts in many samples under investigation. Fahlore formed allotriomorphic aggregates and occured as micro-veinlets and disseminations mainly in skarn-altered dolomites, while only rare sole dissemination or thin micro-veinlets in skarn and in silicified magmatic rocks of the Shakhtama complex. It was frequently associated with chalcopyrite (I-II), galena, sphalerite (II) and bournonite. Fahlore occured frequently at the edges of the aggregates of chalcopyrite (I), replacing it (Figure 12e). The aggregates of fahlore, sphalerite (II), galena, chalcopyrite (II) and bournonite (with even grain boundaries) filled interstices and cracks between early sulfide minerals (pyrite, arsenopyrite, chalcopyrite (I), etc.), substituting and corroding them. Clear boundaries between the indicated minerals, the absence of replacement in intergrowths and the joint filling of intergrain space and cracks allowed us to assume simultaneous crystallization. The presence of myrmekite-like intergrowths of galena, fahlore, sphalerite (II) and chalcopyrite (II) (Figure 10k) was additional evidence of the close age of their formation. Fahlore aggregates were sometimes substituted by marcasite. Less frequently, it associated with gersdorffite (Figure 12f), pyrrhotite (II), cubanite, aurostibite, native antimony and native gold. In the major part of the analyzed grains of fahlore (n = 55), the tetrahedrite component was prevailing (n = 44): the ratio Sb/(Sb + As) varied from 0.62 to 1 (Figure 14). Among 55 fahlore grains that were analyzed, 11 grains exhibited the Sb/(Sb + As) ratio below 0.5 (0.16–0.49); that is, their composition corresponding to tennantite was much rarer in the ores than tetrahedrite (Table S14). It formed intergrowth with galena, sphalerite (II), chalcopyrite (II) and native gold in skarn-altered dolomites with tourmaline. Tennantite was detected most frequently as inclusions and in intergrowth with tetrahedrite (Figure 12g). The structures of tennantite replacement with tetrahedrite were detected in the studied samples; tennantite was replaced with tetrahedrite along grain edges and cracks (Figure 12h). The presence of replacement structures points to the earlier formation of tennantite and a change of the physicochemical conditions of the formation. The Fe/(Fe + Zn) ratio varied from 0.37 to 0.75; that is, tennantite composition varied from zinc-bearing to iron-bearing kinds. For instance, the Zn-tennantite was detected mainly in the form of separate aggregates in association with galena, sphalerite (II), chalcopyrite (II) and native gold in skarn-altered dolomite with tourmaline, while the Fe-tennantite was detected as inclusions and intergrowth with tetrahedrite. Among all fahlore samples, only Zn-tennantite was detected to contain Cu²⁺ (from 0.1 to 0.37 apfu). Ag content up to 1.77 wt.% was also established in it. As mentioned above, tetrahedrite was more frequent in the ores and associated mainly with galena, sphalerite (II) and chalcopyrite (II), and less frequently with bournonite, pyrrhotite (II), cubanite, aurostibite, native antimony and native gold. The majority of the studied grains and aggregates of tetrahedrite were homogenous in their chemical composition; only several aggregates had a block-type structure (Figure 12i). This structure manifested itself in the changes of the composition of separate grains that formed tetrahedrite aggregates, which appeared uniform at first glance in the optical microscope. Tetrahedrite composition also varied from high-Zn to Fe-bearing kinds: the Fe/(Fe + Zn) ratio varied from 0.04 to 0.72. In general, Zn-tetrahredrite dominated in the ores. It was detected in intercrossing veinlets and veins in the altered magmatic rocks of the Shakhtama complex. In these samples, tetrahedrite associated with galena, sphalerite (II), chalcopyrite (II), pyrrhotite (II), cubanite, aurostibite, native antimony and native gold. Ag was permanently present in it at a level of 2.41 to 3.58 wt.% as admixture. Zn-tetrahedrite was also established in skarn-altered dolomites in association with galena, bournonite, sphalerite (II) and chalcopyrite (II). One of the characteristic admixtures was Ag from 1.62 to 2.16; less frequently, Bi was present in an amount of up to 0.22 wt.% (sole grains), Te up to 0.14 wt.% and Cd up to 0.17 wt.%. The highest Ag content up to 13.43 wt.% was established in Zn-tetrahedrite from quartz–biotite veins among amphibole skarns in association with gersdorffite and antimonite. As indicated above, Fe-tetrahedrite occured less frequently; it was detected as intersecting tetrahedrite–chalcopyrite veinlets in skarn-altered dolomites. Fe-tennantite was detected as inclusions in Fe-tetrahedrite. Among characteristic admixtures, Hg was always present at a level from 2.06 to 6.19 wt.%.

Bournonite occurd rarely in the form of disseminations and micro-veinlets in skarn-altered dolomites. It was represented by allotriomorphic aggregates of irregular shapes and formed intergrowth with fahlore, galena, sphalerite (II) and chalcopyrite (II) (Figure 12j).

Boulangerite (Bi-boulangerite) was detected only in the ores of the Ochunogda region in the form of micro-veinlets and impregnations in silicified magmatic rocks of the Shakhtama complex. It was represented by the aggregates of elongated and irregular shapes and associated with galena, ullmannite and chalcopyrite (I-II) (Figure 12k). Boulangerite was substituted by bismuth-containing galena. Admixtures that were permanently present in boulangerite were (wt.%): Bi from 5.4 to 7.2% and Cu from 0.68 to 0.92%; Ag up to 0.13% and Se up to 0.18% (according to EMPA data, Table S15) very rarely; and in some cases, admixtures reached (according to SEM data) rather high concentrations: Bi up to 13 wt.% and Cu up to 1 wt.%.

Ullmannite was detected mainly in the ores from the Ochunogda region. It formed impregnations and was represented by idiomorphic grains of cubic and octahedral shapes. Ullmannite frequently formed subgraphic intergrowth with galena (Figure 12l), which may be explained by the hypogene substitution of ullmannite by galena, though other mechanisms of the formation of these kinds of structures are also possible [49]. It was also associated with bismuth-containing boulangerite, with which it was formed closely simultaneously. Pyrrhotite (II) and chalcopyrite (II) were detected as inclusions in it. According to SEM data, Bi up to 2 wt.% was permanently present in ullmannite.

Molybdenite occured rather rarely in the ores; it was detected as separate disseminations and micro-veinlets in silicified magmatic rocks of the Kultuma massif. It formed flexed plates, thin-plate (Figure 10l) and scaly aggregates and was confined to the intersecting quartz–K-feldspar–carbonate micro-veinlets and cracks filling them (Figure 8k).

Aurostibite was detected in association with later sulfide minerals (galena, tetrahedrite, sphalerite (II), chalcopyrite (II), cubanite, pyrrhotite (II), bournonite, native gold and native antimony), forming intersecting quartz–carbonate veinlets in hydrothermally altered magmatic rocks of the Shakhtama complex. Aurostibite was represented by the aggregates of irregular shapes forming substitution rims over the boundaries of native gold grains (Figure 15a). This is a reactive mineral; it was formed under the action of Sb-containing hydrothermal solutions with low ƒS₂ on previously deposited native gold, which was indicated by the presence of native antimony in association with aurostibite and the substitution rim [49]. Aurostibite was oxidized at the edges of aggregates with the formation of AuSbO₃ (?) (Au—52.66 wt.%, Sb—37.31 wt.% and O—9.45 wt.% according to SEM data).

Native antimony was established in micro-cracks, caverns in galena and tetrahedrite, as well as in the form of later intercrossing micro-veinlets (Figure 15b). It associated with galena, tetrahedrite, sphalerite (II), pyrrhotite (II), chalcopyrite (II), cubanite, bournonite and native gold; it was formed later than the listed minerals, while the time of native antimony formation was close to that of aurostibite. Isolated inclusions of dyscrasite were detected in native antimony (Figure 15b). Native antimony was oxidized with the formation of senarmontite (Figure 15a).

Native bismuth was detected mainly in the ores from the Ochunogda region, and less frequently, it was detected in the ores of the Kultuma deposit. It was represented by xenomorphous segregations and grains of rounded shapes. Native bismuth filled micro-cracks in arsenopyrite, where it rarely formed intergrowth with native gold. In some cases, native bismuth occured along the edges of lollingite grains (lollingite inclusions in arsenopyrite). It was detected as inclusions in galena. Native bismuth often contained Sb up to 5 wt.% as impurity.

Native gold in the ores of the deposit was represented by several generations, distinguished, and taking into account the features of the composition (Table S16) and attribution to definite mineral associations. Broad variations of native gold composition were detected at the deposit: silver content varied from 2 to 44 wt.% (Figure 16a).

Three groups of native gold could be conventionally distinguished. Native gold with Ag admixture from 2 to 8 wt.% containing insignificant admixtures of Hg (up to 0.33 wt.%) and Cu (up to 0.23 wt.%) was assigned by us to the first group (Figure 16b,c). Native gold with Ag content from 10 to 22 wt.% was assigned to the second group. Other elements were detected as admixtures less frequently: Cu up to 1.55 wt.% and Hg up to 0.63 wt.% were present in significant amounts. Native gold with Ag admixture from 30 to 41 wt.% and a permanent admixture of Hg from 2.88 to 5.37 wt.% was assigned to the third group. Cu was only rarely present as admixture (up to 0.44 wt.%).

Native gold that we attributed to the earliest generation occurred in paragenesis with chalcopyrite (I), pyrrhotite (I) and sphalerite (I). It formed mainly intergrowth with chalcopyrite (I) and less frequently with sphalerite (I); together with the latter, it filled the same micro-cracks and veinlets in cataclastic grains and aggregates of earlier ore minerals: in magnetite (I), arsenopyrite and pyrite (I), etc. The fineness of native gold (I) varied between 800 and 980‰. Native gold with the highest fineness was detected in the samples from the surface of the deposit in magnetite skarns in intergrowth with chalcopyrite (I), where its fineness varied between 920 and 980‰ and rarely with Hg admixture up to 0.33 wt.%. At the same time, the samples from the core material in magnetite skarn were detected to contain native gold (I) with high fineness (920–950‰, with an insignificant Cu admixture up to 0.23 wt.%) and gold with lower fineness (800–810‰, with the permanent admixture of Hg from 0.39 to 0.63 wt.%). Fine native gold (I) together with chalcopyrite (I) was filling micro-cracks and caverns in magnetite (I) (Figure 17a). Native gold with lower fineness was detected in association with magnetite (I), pyrite (I), alloclasite, chalcopyrite (I) and sphalerite (I). It was localized in micro-cracks, caverns, at the contacts and in the intergrain space of earlier ore minerals (in magnetite (I), pyrite (I), alloclasite (Figure 17b)) in paragenesis with chalcopyrite (I) and sphalerite (I), the aggregates of which cemented the cataclastic grains of the earlier ore minerals.

It should be stressed that native gold with even lower fineness (650‰, with Hg admixture, 5.1–5.3 wt.%) was determined as sole findings in association with chalcopyrite (I) and sphalerite (I). For example, low-grade native gold was detected in magnetite skarns in the intergrain space and in micro-cracks in pyrite (I) (Figure 17c) together with chalcopyrite (I) and sphalerite (I). Close-in-time crystallization of the listed sulfide minerals and low-grade native gold cannot be stated unambiguously for several reasons. Mineragraphic studies reliably revealed that the finest native gold occurred in intergrowth with chalcopyrite (I), with which it filled the same micro-veinlets and inclusions in earlier ore minerals. This allows us to state that their formation is simultaneous. If Figure 17c is examined more thoroughly, one can notice that at tfirst glance, low-grade native gold seemed to form intergrowth with chalcopyrite (I), but at the same time, it filled intergrain space between pyrite (I) and sphalerite (I). With greater magnification, one may notice (Figure 17c, insert) that low-grade native gold filled micro-cracks in sphalerite (I) and grew onto it.

These features could be evidence in favor of the later formation of low-grade native gold with respect to chalcopyrite (I) and sphalerite (I). However, it is difficult to make unambiguous conclusions from these facts. Low-grade native gold was also established in actinolite–chlorite skarns, in the form of separate inclusion micro-cracks and caverns in pyrite (I) (Figure 17d). Another clear example of overlapping native gold of different compositions within the same mineral association is the Ochunogda region of the Kultuma deposit. It should be stressed that the early generation of native gold at the Ochunogda region was characterized by lower fineness ≈780–800‰ in comparison with the early generation of native gold established at other regions of the Kultuma deposit (800–980‰). For instance, native gold diverse in composition was detected in the intersecting quartz–chlorite–K-feldspar–carbonate veinlets with sulfide mineralization in the altered granodiorites of the Shakhtama complex. Native gold with the highest fineness (≈780–800‰, with Cu admixture up to 1.55 wt.% and Hg up to 0.38 wt.%), or native gold (I), were detected as inclusions in chalcopyrite (I), pyrrhotite (I), and in intergrowth with sphalerite (I) and chalcopyrite (I). Native gold (I), chalcopyrite (I) and sphalerite (I) filled interstices and micro-cracks in cataclastic arsenopyrite grains (Figure 17e), which is evidence of their later formation. Native gold with lower fineness (sole grains, fineness 610–620‰, with permanent Hg admixture from 2.8 to 3.1 wt.%) was also detected in the same sample. We refer this kind of native gold to the third group, with respect to chemical composition. Low-grade native gold also filled interstices and micro-cracks in cataclastic arsenopyrite grains (Figure 17f), but unlike for native gold (I), no intergrowth with chalcopyrite (I) and sphalerite (I) was formed. In addition to the above-listed sulfide minerals, less frequent ones in this sample were galena, boulangerite and ullmannite, which were formed later than chalcopyrite (I), sphalerite (I) and pyrrhotite (I). In our opinion, the presence of two generations of native gold in one sample, differing from each other both in Ag content and in Cu and Hg concentrations, could be considered as the spatial overlapping of earlier high-grade native gold (in paragenesis with chalcopyrite (I), sphalerite (I) and pyrrhotite (I)) with the latest low-grade native gold in association with galena, boulangerite and ullmannite. A similar pattern was also observed at the other region of the Kultuma deposit. Different compositions of native gold were established in skarn-altered dolomites with tourmaline. Here, native gold was detected in close intergrowth with galena (Figure 17g), in paragenesis with the later minerals of the polymetallic association (Zn-tennantite, sphalerite (II), chalcopyrite (II), bournonite). Our studies showed that the fineness of the dominating number of native gold grains varied within the range of 780–800‰, with insignificant permanent admixture of Hg up to 0.39 wt.%. At the same time, there were sole findings of gold with lower fineness (560‰, with Hg admixture up to 3 wt.%) in intergrowth with galena (Figure 17h). In our opinion, this may be explained not by the spatial overlapping of native gold of different compositions, but by a sharp change of the physicochemical conditions of its formation, which was also confirmed by the presence of the structures of tennantite replacement by tetrahedrite. In general, it may be stressed that low-grade native gold only rarely occured in the ores of the deposit. At the same time, native gold was frequently detected in paragenesis with later sulfide minerals of the polymetallic association. It was also detected in intercrossing veinlets and veins in the altered magmatic rocks of the Shakhtama complex. It associated in these samples with galena, Zn-tetrahedrite, sphalerite (II), chalcopyrite (II), pyrrhotite (II), cubanite, aurostibite and native antimony (Figure 17i). Native gold was characterized by somewhat higher fineness, in comparison with native gold in association with Zn-tennantite, galena (II), sphalerite (II), bournonite and chalcopyrite (II) in skarn-altered dolomites with tourmaline. The fineness of native gold varied from 820 to 890‰, and Hg and Cu admixtures were not detected. The higher fineness of native gold may be due to the major part of Ag, with simultaneous crystallization of Zn-tetrahedrite and native gold, binds in Zn-tetrahedrite, which was characterized by a higher Ag content (from 2.41 to 3.58 wt.%). In contrast, Zn-tennantite (Ag content from 1.61 to 1.77 wt.%) associated with native gold is characterized by lower fineness.

Native gold with bismuth minerals was another characteristic mineral paragenesis. Native gold was detected in intergrowth with bismuthine, pekoite and bismuth sulfotellurides. Native gold and bismuth minerals were frequently detected as inclusions in pyrrhotite (I) (Figure 17j), and they also formed integrowth with it and with chalcopyrite (I). These facts allow us to conclude that their crystallization proceeded contemporaneously. The fineness of native gold was 830–880‰, and no Hg and Cu admixtures were detected. Low-grade native gold (590‰, according to SEM data) was detected in a few samples of skarns at the Ochunogda region. Fine (5–6 µm in size) low-grade gold particles were detected in micro-cracks in arsenopyrite, along with native bismuth (Figure 17k).

Native gold in association with bismuth minerals was also observed in tremolite–phlogopite–magnetite skarns with anhydrite. Here, native gold was present in association with magnetite (I), pyrite (I), siegenite, chalcopyrite (I), low-Fe sphalerite (I?) and pekoite. Gold was detected in the form of inclusions (in caverns) and in micro-cracks in siegenite, as well as at the boundaries of its grains. Native gold was formed later than magnetite (I), pyrite (I) and siegenite. In the inclusions in siegenite, gold formed with a compound (or a mixture of minerals) with the composition: Ag—54.5 wt.%, Bi—14.2 wt.%, Te—4.3 wt.% and S—12.5 wt.%, according to SEM data (Figure 17l). Native gold was formed contemporarily with chalcopyrite (I), low-Fe sphalerite (I?) and pekoite. The fineness of gold varied from 830 to 880 ‰, with an admixture of Cu up to 1.26 wt.% and Hg up to 0.19 wt.%. In addition to the above-listed minerals, later sulfide minerals were detected: matildite, jalpaite (?) (Ag—64 wt.%, Cu—8.3 wt.%, Bi—5.9 wt.%, Sb—4.9 wt.%, Fe—2 wt.%, As—1 wt.% and S—14.9 wt.%, according to SEM data) and acanthite. In our opinion, the formation of these silver minerals relates to the latest hypergene processes, which is evidenced by the morphology of these minerals and their relationship with earlier ore minerals. For instance, acanthite formed very typical aggregates for hypergene minerals: bud-like and moss-like aggregates.

Summarizing the above-presented data on the chemical composition of native gold and on its relationship with various ore minerals, it appears most reasonable to distinguish the generations of native gold relying mainly (with some exceptions) on its relationship with definite mineral assemblages. We assigned native gold in assemblage with chalcopyrite (I), sphalerite (I) and pyrrhotite (I) to the earliest generation of native gold. It was characterized by fineness 800 to 980‰, with insignificant admixtures of Hg up to 0.63 wt.% and Cu up to 0.23 wt.%. As mentioned above, native gold at the Ochunogda region was characterized by several distinguishing features of chemical composition, but at the same time, similar to other regions of the Kultuma massif, native gold (I) was in association with chalcopyrite (I), sphalerite (I) and pyrrhotite (I). Its fineness varied from 780 to 800‰, with Cu admixture up to 1.55 wt.% and Hg up to 0.38 wt.%. Native gold in assemblage with bismuth minerals—in particular bismuthinite, pekoite and bismuth sulfotellurides, which were formed contemporaneously with chalcopyrite (I), sphalerite (I) and pyrrhotite (I) as our studies revealed—is referred by us to native gold (I). We refer the native gold in the paragenic series with galena, fahlore (tetrahedrite and tennantite), sphalerite (II), chalcopyrite (II), burnonite, pyrrhotite (II) and other minerals of the polymetallic association to the second generation. The fineness of this gold varied from 780 to 890‰, with Hg admixture up to 0.39 wt.%. Native gold with the lowest fineness (560–650‰), with consistently high Hg admixture from 2.8 to 5.3 wt.%, was related to the third generation. Sole grains of this gold were detected at all regions of the Kultuma deposit. Most frequently, it did not form any intergrowth with ore minerals. Unfortunately, it appears currently impossible to state any reliable paragenic relationship of native gold (III) with any ore minerals. It reliably formed close intergrowth with galena in only a single case.

4.7. Stable Isotope Geochemistry

4.7.1. Sulfur Isotope Data

Results of the investigation of sulfur isotope composition are listed in

Table 1. Sulfur isotopic composition of sulfide minerals from the Kultuma deposit.

Sample No.	Analytical Object	δ34SV-CDT (‰)
Km-1-313.7	anhydrite	24.4
Km-1-323.1	anhydrite	25.6
Od-1-61.2	arsenopyrite	9
Od-1-131.4	arsenopyrite	7
Km-1-313.7	chalcopyrite	12
Km-3-403.8	chalcopyrite	11.9
Km-3-405.4	chalcopyrite	14.4
Km-4-79.3	chalcopyrite	11.9
Km-11	chalcopyrite	6.8
Km-12	chalcopyrite	6.8
Km-17	chalcopyrite	8
Km-18	chalcopyrite	7.1
Od-1-61.2	chalcopyrite	6.6
Km-6-511.4	galena	9.2
Km-1-182.8	pyrite	7.2
Km-3-406.7	pyrite	11.8
Km-3-413.5	pyrite	7.4
Km-3-419.5	pyrite	8.2
Km-3-490.3	pyrite	16
Km-19-497.5	pyrite	7.9
Km-4-79.3	pyrrhotite	12.2
Od-5-99.4	pyrrhotite	1.9
Od-5-101.5	pyrrhotite	3.7
Od-5-141.5	pyrrhotite	3.4
Od-5-142.4	pyrrhotite	1.4
Od-5-144.5	pyrrhotite	2.5
Km-3-385.8	sphalerite	10.6
Km-3-481.5	tetrahedrite	12.3

.

The δ³⁴S values (in sulfide minerals) strongly varied from 1.4 to 16.0‰. For instance, at the Ochunodga region, anomalously low δ³⁴S values were detected (from 1.4 to 3.7‰) for pyrrhotite from retrograde skarns. Sulfide minerals with relative prevalence of the heavy isotope dominated at the Kultuma deposit, in general, and at the Ochunogda region. Early sulfide minerals from retrograde skarns contained relatively heavier sulfur isotopes: arsenopyrite—7.0 to 9.0‰ and pyrite from 7.2 to 11.8‰. The sample with the heaviest sulfur isotope composition (pyrite—16.0‰) was taken from the host dolomites of the Bystrinskaya formation. The isotope composition of sulfur in the host rocks of the Beletuyskaya formation was relatively lighter: 7.9‰. Sulfide minerals from the chalcopyrite–pyrrhotite–sphalerite paragenesis also contained relatively larger amounts of the heavy sulfur isotope: chalcopyrite from 6.6 to 14.4‰ and pyrrhotite 12.2‰ (from retrograde skarn). Similar δ³⁴S values—12.0‰—were also characteristic of chalcopyrite from the zone of propyllite alteration. Sulfide minerals of the polymetallic association from the zones of propyllite alteration also contained relatively more heavy sulfur isotope: galena—9.2‰, tetrahedrite—12.3‰ and sphalerite—10.6‰. The largest fraction of the heavy sulfur isotope was characteristic of anhydrite (24.4 and 25.6‰) from the potassium alteration zone.

4.7.2. Oxygen and Carbon Isotopes

Results of the investigation of C and O isotope composition are listed in

Table 2. Isotopic composition of carbon and oxygen from the Kultuma deposit.

Sample No.	Analytical Object	δ 13C (‰)	δ 18O (‰)
Km-3-481.5	dolomite	−3.1	17.1
Km-3-459.3	-//-	−3.5	23.9
Km-3-447.7	-//-	−3.1	22.2
Km-3-442.5	-//-	−2.7	16.6
Km-1-179.1	-//-	−2.4	18
Km-1-175.5	-//-	−2.1	14.2
Km-6-522.4	-//-	−2.7	15.5
Km-3-481.5	calcite	−2.1	12.6
Km-3-385.8	-//-	−4.5	4.4
Km-6-511.4	-//-	−2.5	14.3
Km-6-518.7	-//-	−3.8	12.8
Km-3-405.6	-//-	−3.4	8.7
Km-1-323.1	-//-	−2.5	2.7
Km-4-79.9	-//-	−5.3	3.3

. Host rocks, dolomites of the Bystrinskaya formation, had a rather uniform carbon isotope (δ¹³C) composition from −2.1 to −3.5‰. The obtained small negative δ¹³C values are characteristic of marine limestone. The isotope composition of oxygen varied in a broader range: δ¹⁸O varied from 14.2 to 23.9‰. The largest δ¹⁸O values (22.2 and 23.9‰) re characteristic of less altered dolomites of the Bystrinskaya formation and are typical for marine limestone. Lighter oxygen isotope composition (14.2, 15.5, 16.6, 17.1 and 18.0‰) for some samples of dolomites from the Bystrinskaya formation was most probably due to the interaction with the primary fluid, which had separated from the granite intrusion with the indicator values δ¹⁸O ≈ 6–10‰ [50].

Isotope labels of carbon in quartz–carbonate and carbonate veins (nests), with tourmaline, chlorite and sulfide minerals of the polymetallic association (in the zone of propylite alteration) had δ¹³C values from −2.1 to −4.5‰, which is close to the corresponding values for host dolomites of the Bystrinskaya formation. The isotope characteristics of oxygen exhibited a broader variation of the composition. For example, for calcite sampled from the propyllite alteration zone (with sulfide minerals of polymetallic association) of retrograde skarns, the δ¹⁸O value was 4.4‰, while δ¹⁸O values for calcite from the quartz–carbonate veins (with tourmaline, chlorite) were characterized by higher δ¹⁸O values from 12.6 to 14.3‰. Similar values were also obtained for calcite from retrograde skarns with magnetite and chalcopyrite: δ¹⁸O 12.8‰ and δ¹³C—3.8‰. The isotope characteristics of calcite from retrograde skarns were also close: δ¹³C from −2.5 to −5.3‰ and lower values of δ¹⁸O from 2.7 to 8.7‰, corresponding to the indicator values for granitoids.

4.8. Geochronological Studies

An essential problem of the genesis of many gold ore and gold-bearing deposits of Eastern Transbaikalia is the determination of the age of magmatic rocks, which is the basis for its correlation with the age of gold mineralization and analysis of gold metallogeny in Eastern Transbaikalia. To solve this problem, two methods were employed: LA-ICP-MS U–Pb and ⁴⁰Ar–³⁹Ar dating.

4.8.1. Zircon LA-ICP-MS U–Pb Geochronology

Representative CL images of zircons, laser spots (Figure 6) and other characteristics of zircon samples under investigation were considered above. For U–Pb dating (LA-ICP-MS), zircons were extracted from quartz monzodiorite-porphyries of the Kultuma massif. The total number of analyses of zircon samples from four magmatic rocks was 76 (results are presented in the Table S17). The weighted mean ²⁰⁶Pb/²³⁸U age of zircons from quartz monzodiorite-porphyries was 158.7 ± 0.56 Ma (MSWD = 0.78, n = 24, sample Km-8), 161.5 ± 1.0 Ma (MSWD = 1.2, n = 8, sample Km-9) and 156.8 ± 0.64 Ma (MSWD = 0.01, n = 18, sample Km-10-732.8). The weighted mean ²⁰⁶Pb/²³⁸U age of zircon samples from monzodiorite-porphyries (the Ochunogda region) was close to the above-listed data: 157.4 ± 0.53 Ma (MSWD = 0.057, n = 26, sample od-3-48.5) (Figure 18).

4.8.2. ⁴⁰Ar–³⁹Ar Geochronology

The obtained age data for biotite from the dykes of quartz monzodiorite-porphyries and phlogopite from retrograde phlogopite skarns are shown in Figure 19. One can see in Figure 20a that there was a clear plateau (corresponding to the criteria proposed by Fleck et al. (1977) with an age of 157.1 ± 1.6 Ma in the spectrum of biotite from quartz monzodiorite-porphyries, to which 96.8% of the ³⁹Ar released corresponds. The age of the retrograde phlogopite skarns is 156.3 ± 1.6 Ma (by phlogopite).

4.9. Fluid Inclusion Study

Petrographic and mineralogical examinations showed that retrograde skarn and potassic-propylitic alteration minerals are co-genetic and related paragenetically to sulfide mineralization. As the progressive stage skarns are quite rare and strongly altered at the deposit, we analyzed primary and pseudo-secondary fluid inclusions in quartz from zones of retrograde skarns, potassic and propylite alteration.

Based on the preliminary optical microscopy observations and Rödder’s [51] criteria, we distinguished primary fluid inclusions (FI) as isolated fluid inclusions and random groups, pseudo-secondary FI as fracture trails in intragranular quartz crystals and secondary FI as trails that cross boundaries of the quartz grains. The FI phase composition at room temperature was attributed to specify the following types: M—multiphase FI that contain gaseous, liquid and solid phases; VLC—FI with solution, liquid CO₂ and gas; VC—significantly carbonaceous FI (gaseous and liquid CO₂); V—vapor dominanted FI; VL—double-phase vapor-liquid FI. A summary of the results of the fluid inclusion study are given in

Table 3. Summary microthermometric and Raman spectroscopy data of fluid inclusions from quartz of the Kultuma deposit.

Sample	Od-5-101				Km-2-235			Od-2-50.5			Km-2-195		Km-3-414
Min. Ass.	Retrograde Skarn				Potassic Alteration, Kultuma			Potassic Alteration, Ochunogda			Propylitic Alteration
FI type	VL, pr	VC, pr	M, pr	V, ps	VC, pr	VL, ps	M, ps	VLC, pr	M, pr	VL, ps	VLC, pr	M, pr	VLC, ps	M, ps
Thom, °C	360–370	22–22.3	-	180–200	420–440	335–380	-	410–420	-	-	280–320	-	245–270	-
Pressure (bar)	-	-	-	-	2.0–2.4	-	-	1.65–1.7	-	-	1.0–1.2	-	1.06–1.2	-
Gaseous phase	CO2 + CH4 + N2	CO2	CH4 + N2	CO2	CO2	CO2	-	CO2	-	CO2 ± N2 ± CH4	CO2 ± N2 ± CH4 ± C2H6	CO2 + CH4	CO2 ± N2	CO2 ± N2
Solid phase	-	-	calcite	-	-	-	chloride	-	titanite	-	-	nahcolite	-	chloride

Note: FI—fluid inclusion; Th—homogenization temperatures; pr—primary; ps—pseudosecondary; M—multiphase FI; VLC—FI with solution, liquid CO₂ and gas; VC—significantly carbonaceous FI; V—vapor dominanted FI; VL—double-phase vapor-liquid FI; -—not detected.

.

4.10. Fluid Inclusions in Retrograde Skarn Minerals

Fluid inclusions were studied in quartz from retrograde skarns with pyrrhotite (I) (Ochunogda region of the Kultuma deposit). M type and VC type FI (Figure 20a,b) were located randomly in central parts of grains, suggesting a primary origin. The inclusions were round or isometric in shape with sizes ranging from 5 to 12 μm. Multiphase inclusions included a light-colored mineral phase, which was represented by carbonate (probably calcite). For VC type inclusions, the homogenization temperature into the liquid phase was set at 22–22.3 °C. Raman spectroscopy revealed an admixture of low-boiling gases in these inclusions—methane and nitrogen. The confirmed M and VC type FI propose ore fluid separation, probably as a result of pressure reduction.

VL type FI are defined as primary FI (Figure 20c). They had negative crystal shapes and a size of 4–9 μm. The gaseous phase was 20–30 vol.% of the inclusion and contained a mixture of CO₂ + CH₄ + N₂. The obtained homogenization (in liquid phase) temperatures fell within the range of 360–370 °C.

V type FI are described as small (3–8 μm), isometric, pseudo-secondary FI. CO₂ was detected in the composition of the gaseous phase. A thin rim of solution along the inclusion edges disappeared in the temperature range 180–200 °C, and the fluid became homogeneous.

4.11. Fluid Inclusions in Minerals from Potassic Alteration Assemblage

Quartz from quartz–feldspar–carbonate veins with molybdenum contained VC, VL and M types of FI. The VC type inclusions were primary (Figure 20e). They were isometric and 8–18 μm. Three phases were observed at room temperature: a narrow rim of solution, liquid carbon dioxide and gas. The dominant component of the gaseous phase was CO₂, and the minor one was N₂ (according to Raman spectroscopy and cryometric data). Homogenization of CO₂ into the liquid phase occured at temperatures of 31.3–31.6 °C and aqueous phase homogenization into the liquid at 420–440 °C. These values corresponded to the estimated pressure 2000–2400 bar, which corresponds with a depth of 8–9.5 km.

Pseudo-secondary VL type FI were isometric or slightly elongated in shape and up to 20 μm in size. The gaseous phase contained carbon dioxide. Homogenization of these inclusions occured in the liquid phase at temperatures of 335–380 °C.

Pseudo-secondary M type FI were characterized by irregular shapes and sizes of 13–25 μm. A light-colored isotropic cubic crystal, most likely chloride, represented the mineral phase.

Gas–liquid (VL) and multiphase (M) inclusions were located in the same cracks, which may indicate the boiling of the fluid during ore formation.

Formation of quartz–chlorite-feldspar–carbonate veins with sulfide mineralization (chalcopyrite (I), pyrrhotite (I) and sphalerite (I)) in the altered magmatic rocks of the Shakhtama complex (Ochunogda region of the Kultuma deposit) relates with potassium hydrothermal changes. Quartz from these rocks have been taken for fluid inclusion studies. VLC type FI were primary (Figure 20d,f) and had negative crystal shapes and sizes of 10–14 μm. Thermometric experiments showed that homogenization of carbon dioxide occured in liquid phase at temperatures of 24.8–25 °C. Temperatures of 410–420 °C matched with complete homogenization of inclusions. The calculated pressure, ranging from 1650 to 1700 bar, corresponds with a depth of 6.5–7 km.

A single M type FI was detected in the sample. According to Raman spectroscopy, the mineral phase was represented by titanite, which was highly likely xenogenic to the inclusion.

VL type inclusions were pseudo-secondary and had irregular shapes and sizes of 10–15 μm. The fraction of gas bubbles was 10–15 vol.%. In addition to CO₂, the gaseous phase contained N₂ and CH₄.

4.12. Fluid Inclusions in Minerals from Propylitic Alteration Assemblage

Fluid inclusions were studied in quartz from cutting quartz–chlorite–carbonate veins with sulfide minerals of polymetallic association (galena, Zn-tetrahedrite, sphalerite (II), chalcopyrite (II), pyrrhotite (II), cubanite, aurostibite, native gold (II), and native antimony). The primary VC type FI (Figure 20g) were isometric or a negative crystal in shape, with sizes varying from 5 to 18 microns (μm). The gaseous phase comprised CO₂, with traces of N₂, CH₄ and C₂H₆, which was confirmed by cryometric studies. VC inclusions displayed melting of the solid phase during thawing, observed at temperatures ranging from −57.5 to −57 °C. The temperatures were lower than the triple point for pure CO₂ (−56.6 °C), indicating the presence of tiny amounts of other gases. The CO₂ homogenization temperature (TCO₂h) varied between 26.2 and 29.6 °C (in liquid phase). The total homogenization proceeded at temperatures of 280–320 °C. This data corresponded with pressure matching 1000–1200 bar (corresponding depth 4–5 km).

The M type FI (Figure 20h) were observed in small clusters. They had negative crystal shapes and sizes ranging from 8–20 μm. Phase composition of these inclusions consisted of a solid phase, solution, liquid CO₂ and gaseous phase. The solid phase presented as light-colored anisotropic crystals, probably being nahcolite (NaHCO₃). The gaseous phase contents were CO₂ and CH₄.

Fluid inclusions were also studied in quartz from the zones of propylitic changes in retrograde skarns. The ore minerals were Zn-tetrahedrite, sphalerite (II), chalcopyrite (II) and marcasite. Pseudo-secondary FI in cracks extending from the centers of quartz grains to their boundaries with ore minerals were represented by VLC type and M type FI (Figure 20i,j). The VLC type FI had isometric or negative crystal shapes and sizes of 4–15 μm. The gaseous phase composition was CO₂ ± N₂. Homogenization of CO₂ occured in liquid at temperatures of 23.9–26.2 °C and complete homogenization of inclusions at 245–260 °C. Accordingly, the pressure was estimated as 1060-1200 bar. The M type inclusions contained a light-colored isotropic cubic mineral phase, probably chloride. The gaseous phase contained carbon dioxide and traces of nitrogen.

5. Discussion

The Au–Cu–Fe–skarn deposits related to the Shakhtama complex are situated mainly in the north-eastern and south-eastern parts of the Shilka–Argun interfluve. The largest deposits are the Bystrinsky, the Kultuma and the Lugokan deposits. According to the available data for open access, the total reserves of gold in these three deposits is estimated to be about 430 tons. A common feature of these deposits is the presence of many metals. In particular, only for skarn deposits, the industrially valuable resources of copper have been established. This fact should be stressed because copper is not a typical element and does not form industrial concentrations in the deposits of other genetic types related to the Shakhtama complex. In our opinion, this is due, first of all, to the geological structure of the deposits, and to the confinement of the skarn deposits (the Lugokan, Kultuma and Bystrinsky deposits) to the carbonate terrigenous rocks of the Lower Cambrian. The local geological conditions provide the specificity of mineralization pronounced within each deposit. The reasons of these specific features will be considered below to assess the genetic type of the Kultuma deposit, relying on the set of the data obtained in the studies.

5.1. Timing and Tectonic Setting of Magmatism and Mineralization

It is known that the majority of gold ore and gold-bearing deposits in Eastern Transbaikalia had been formed in the Middle and Late Jurassic under collision-related settings and in the Early Cretaceous under rift-related settings [10,11,52]. Our studies showed that the formation of the magmatic rocks of the Shakhtama complex proceeded at the modern territory of the Kultuma deposit during the Late Jurassic time: 161–156 Ma. A close age is also characteristic of the magmatic rocks of the Shakhtama complex, occurring at the Shakhtama deposit (161–155 Ma) and at the Lugokan deposit (154.7 ± 1.2 Ma). The age of these rocks corresponds to the final stage of the collision process [9,53]. A close age of formation (156.3 ± 1.6 Ma), determined by means of Ar–Ar dating, was also exhibited by phlogopite from retrograde phlogopite skarns, with which some magnetite ores are associated. The data obtained allow us to state that the formation of magmatic rocks of the Shakhtama complex and ore mineralization within the Kultuma deposit proceeded at the final stages of the collision of the marginal continental complexes of the Siberian and North Mongol–Chinese continents [7,8,9,10,11,13,52].

5.2. The Role of Lithological and Structural Tectonic Factors in Ore Mineralization Distribution and Specificity

Considering the major ore-controlling factors, it is necessary to pay attention first to the tectonic position of the Kultuma deposit in the structures of the Mongol–Okhotsk belt. The Kultuma deposit is confined to the node of intersection of the regional faults, namely, the north-eastern Levo–Gazimur and the north-western Bogdat–Boshagoch, which are parts of the Gazimur fault zone. It is well-known that the majority of gold ore and gold-bearing deposits in Eastern Transbaikalia either are directly confined to the Mongol–Okhotsk suture or occur at a substantial distance from it and exhibit the spatial relationship with rather extended fault zones. Fault zones such as the Gazimur were initially formed or renewed during collision events. Prevailing penchant of large gold deposits to the Mongol–Okhotsk suture and to large regional fault zones is connected with their increased permeability both for ore-producing magmatic melts and for gold-bearing fluids. Indeed, one may notice that all large complex skarn deposits related to the magmatic formations of the Shakhtama complex are also situated within the Gazimur fault zone. In particular, this caused a substantial impact both on the formation of the Kultuma massif as a whole and on many disjunctions occurring within the deposit. Due to the development of multiple relatively heterochronous systems of discontinuous faults, the territory of the Kultuma deposit was characterized by exclusively high permeability. It is this circumstance that provided the manifestation of diverse magmatic activity and hydrothermal processes within the boundaries of the deposit. It is impossible to provide a detailed consideration of each case in the present work; thus, in our opinion, it is necessary to reveal the most illustrative ones. One of the examples is polymetallic ores detected directly at the Kultuma deposit, as well as to the north and north-east of the Kultuma massif, where they form a number of deposits and ore occurrences (the Kultuma group). The major part of polymetallic ores at the Kultuma deposit are located (in comparison with magnetite and magnetite–sulfide ores) at a longer distance from the direct contact of the magmatic rocks of the Kultuma deposit and the host terrigene–carbonate rocks. A center of the development of polymetallic mineralization may be considered to be at the deposits and ore occurrences of the Kultuma group. Here, similarly to the case of the Kultuma deposit, the richest polymetallic ores are confined to the contact of two formations, which are in contrast to each other with respect to their composition and to the zones of fracturing and brecciation. The local fractures served as the routes of fluid transportation over long distances. A decrease in fluid temperature, as well as fluid mixing with meteoric waters, did not lead to large-scale formation of skarn. Skarnification is only local and tends to the faults crossing these rocks or the contacts of carbonate and aluminosilicate rocks, while propylitization processes and related sedimentation of the minerals of polymetallic association are pronounced most widely. These factors did not allow large-scale skarn formation, which was also depicted in the morphology of ore bodies. We suppose that polymetallic deposits and ore occurrences (of the Kultuma group) derived from the Kultuma ore-magmatic system. This is indicated by the similar mineral composition and by the hydrothermal–metasomatic alterations. These ore bodies are actually distal ones, remote from the direct contact of the Kultuma massif and the carbonate-terrigenous Lower Cambrian rocks.

Analysis of the spatial arrangement of various paragenetic mineral associations and the hydrothermal–metasomatic alternations of host rocks within the Kultuma deposit helps outline definite zoning, which is due not only to the magmatic factors, but also to the lithological and structural–tectonic factors. This zoning may be schematically represented as follows: the earliest and the most high-temperature ores with magnetite mineralization are confined mainly to the skarns of the regressive stage and form the margins of the contacts of the Kultuma massif. The early relatively high-temperature pyrite–arsenopyrite and gold–chalcopyrite–sphalerite–pyrrhotite associations are widespread in the endo- and exocontact parts of the Kultuma massif, while within the Kultuma deposit itself, potassium hydrothermal alteration is manifested only weakly; they were established both in the form of intersecting quartz–K-feldspar veins with molybdenite in the magmatic rocks of the Shakhtama complex and as superpositions over retrograde skarns. Potassic hydrothermal alterations are more widely expressed within the Ochunogda region, where veins and veinlets with arsenopyrite, chalcopyrite and pyrrhotite are formed. It should also be stressed that arsenopyrite at the Ochunogda region almost always dominates over pyrite, in contrast to the Kultuma deposit. The massive pyrrhotite ores were also determined there, confined to the regions with skarnified limestone of the Ernichenskaya formation. Relatively lower-temperature mineral parageneses and related hydrothermal–metasomatic alterations (for example, propylitic alteration) occur at a longer distance from the direct contact of the Kultuma massif with embedding rocks. The areal propylitization is characteristic of the Kultuma deposit, where it affects great volumes of rocks, while at the Ochunogda region, it often appears to be linear and is confined to the fault and fracture zones. This sequence is frequently complicated by ore telescoping.

The spatial arrangement of various paragenetic mineral associations over the territory of the Kultuma deposit provides evidence of the existence of horizontal zoning, which was formed around the Kultuma massif. Indeed, the richest magnetite ores are situated at the contact between the magmatic rocks of the Kultuma massif and the terrigenous carbonate sediments of the Bystrinskaya formation. In particular, this is also due to the lithology. The major part of retrograde skarns and the related large-scale magnetite mineralization are most widespread at the Kultuma deposit, where the carbonate terrigenous sediments of the Bystrinskaya formation occur most widely. At the Ochunogda region, the skarns occur locally and are confined to the lenses of carbonate rocks with the carbonaceous matter of the Ernichenskaya formation, which, in turn, has been depicted not only in the mineral composition, but also in the isotope characteristics to be considered below. This is why magnetite mineralization in particular is mainly concentrated at the Kultuma deposit, while it occurs only sporadically. It should be stressed that the specificity of mineralization is also strongly affected by the composition of host rocks. Carbonate masses taking part in hydrothermal–metasomatic processes within the Kultuma deposit are likely to have substantial differences in composition. The carbonate rocks of the Ernichenskaya formation often contain small amounts of the carbonaceous matter of biogenic origin. This carbonaceous matter, a strong reducing agent, interacts with fluids participating in the hydrothermal metasomatic processes and causes a reduction of their oxidizing ability (oxygen fugacity). In particular, for this reason, the massive pyrrhotite ores were determined at the Ochunogda region in retrograde skarns. On the other hand, the carbonate rocks of the Bystrinskaya formation most probably contained a definite amount of sulfates, which may also be evidenced by the isotope data that will be discussed below. These conditions could have occured in an isolated shallow-water basin during the formation of the carbonate rocks of the Bystrinskaya formation.

5.3. Magma Source and Oxidation State

The studies showed that the magmatic rocks that are widespread at the Kultuma deposit were represented by magnesian, mainly meta-aluminious (to a lower extent peraluminious) granitoids (type I) and related to the high-potassic, calc-alkalic and shoshonite series. The Late Jurassic granitoids of high-potassic, calc-alkalic and shoshonite series in Eastern Transbaikalia are associated with the overwhelming majority of endogenous deposits of nonferrous, rare and precious metals [13]. The Jurassic granitoid massifs are collision-related formations tracing the tectonic structures extending to the north-east [10,11]. In the opinion of Kovalev et al. [6], the formation of monzonitoids dominating in the Kultuma massif and the dyke complex proceeded through differentiation of the sub-alkalic basalt melt from the enriched mantle-related source.

The obtained set of new data allows us to consider the problem of the genetic type of the Kultuma deposit. The main factors affecting the arrangement and specificity of gold ore mineralization for gold ore deposits of the intrusion-related gold systems (IRGS) class are the chemical composition of magma and the extent of its oxidation. During the recent years, the magmatic rocks of the Shakhtama complex widespread over large complex (in particular, gold-bearing) deposits are considered by many researchers as analogues of adakites, which in turn are considered as indicators of highly productive copper–porphyry systems [4,5,53,54,55]. The question of whether the magmatic rocks of the Kultuma massif and the dykes belong to adakites or adakite-like rocks was considered separately by [6]. We agree with the opinion of these authors that the magmatic rocks of the Kultuma massif and the dykes do not have typical features of adakites, so other aspects will be considered below in more detail.

It is well-known that the gold–copper–porphyry deposits are spatially and genetically confirmed to the oxidized granitoids of the magnetite series. For this reason, the data on the degree of magma oxidation contain extremely important information for the genetic type of the deposits. To divide magmatic rocks into the magnetite (oxidized) and ilmenite (reduced) series, the data on the magnetic susceptibility of magmatic rocks, petrographic composition (the presence of accessory minerals—magnetite or ilmenite) and the ratio of Fe₂O₃/FeO are classically used. Our studies showed that almost all the studied rocks that occur at the Kultuma deposit were characterized by the low Fe₂O₃/FeO ratio (0.2 to 0.5) and low magnetic susceptibility (0.07 to 0.19 × 10⁻³ SI). Only the quartz diorite porphyry dykes were characterized by increased Fe₂O₃/FeO values (0.59), higher magnetic susceptibility (0.96 × 10⁻³ SI), and only in them, the presence of magnetite among accessory minerals was detected. One can see in Figure 5 that similar values were also characteristic of granodiorite–porphyries developed at the Lugokan Au–Cu–Fe–skarn deposit, while monzodiorites developed at the Antiinsky (Au) ore occurrence were characterized by slightly higher magnetic susceptibility, in contrast to magmatic rocks at the Au–Cu–Fe–skarn deposits. The highest magnetic susceptibility values and the high values of Fe₂O₃/FeO ratio were established for altered monzodiorites at the Antiinsky ore occurrence. These values were due to the degree of surface (exogenic) alteration of the rocks under study, which was described in detail in [8]. The data obtained allow us to conclude that the magmatic rocks of the Kultuma massif, as well as the quartz monzonite–diorite porphyry dykes relate to the reduced granitoids of the ilmenite series, while the dykes of quartz diorite-porphyries, based on all the three parameters, may be related to the oxidized granitoids of the magnetite series.

The data on the chemical composition of zircon are now widely used to determine the oxidation–reduction potential of the melt. In particular, it was established that the values of Ce and Eu anomalies changed depending on ƒO₂ in the melt [35,56,57,58,59]. For more correct assignment of the magmatic rocks under investigation to the reduced or oxidized granitoids, we also used the data on the chemical composition of zircons from the magmatic rocks of the Kultuma massif and from the dykes, as no data on the chemical composition of zircons from the magmatic rocks in paragenetic relationship with the deposits of various metals in Eastern Transbaikalia could be found in the literature. For comparison, we used the data on the concentrations of trace and rare earth elements in the zircons from different deposits of the world, which can be divided into two groups: porphyry and skarn. The data on the composition of zircons from the porphyry deposits situated in Northern China were also involved because these two deposits (Xing’a and Chalukou) are closest to the Kultuma deposit in a number of signs—in particular, the age of magmatic rocks, tectonic position and others. The results were plotted in the diagram as Ce/Ce* vs Eu/Eu*, along with the data on porphyry (Figure 21a) and skarn (Figure 21c) deposits. For instance, zircons that were crystallized from relatively oxidized magmas should be characterized by higher Ce/Ce* and Eu/Eu* values than zircons from more substantially reduced magmas.

One can see in Figure 21a that zircons from the magmatic rocks of the Kultuma massif and the dykes were characterized by low Ce/Ce* values than the porphyry deposits. At the same time, one may note that the ranges of values overlapped. A similar situation was also observed for skarn deposits, with some specific features. In particular, one can see in Figure 21c that zircons from the magmatic rocks of the Cu–Mo-skarn Tongshankou deposit had slightly higher Ce/Ce* values, though the ranges of values overlapped substantially, too.

The values of the Eu/Eu* ratio may also be used for the determination of the degree of melt oxidation. The low values of Eu/Eu* are characteristic for less oxidized conditions, while Eu/Eu* > 0.4 is characteristic of porphyry systems with the highest oxidation degree. One can see in Figure 21a that the majority of porphyry deposits were characterized by the larger values of Eu/Eu* ratios, unlike for the magmatic rocks of the Kultuma massif and the dykes of quartz diorite-porphyries. At the same time, zircons from the Mo-porphyry Xing’a deposit (northern China) were also characterized by low Eu/Eu* values. A similar picture was also observed for skarn deposits. For instance, zircons from the Cu–Mo-skarn Tongshankou deposit and the Cu–Au–Fe-skarn Tonglushan deposit, which are in the paragenetic relationship with stronger oxidized intrusions, had an Eu/Eu* more than 0.4. Unlike for these samples, the imaging points of the analyses of zircons from Fe-skarn deposits Lingxiang and Chengchao were generally characterized by lower Eu/Eu* values (<0.4), which is evidence of the higher reductive formation settings. The data obtained by us on the content of trace and rare earth elements in the zircons of the Kultuma massif, in particular the low values of Ce/Ce* and Eu/Eu* ratios, were in good agreement with the data on the magnetic susceptibility of magmatic rocks, petrographic composition (the presence of accessory minerals—magnetite or ilmenite) and Fe₂O₃/FeO ratios. In addition, a comparison of the results with the data for other porphyry and skarn deposits allows us to conclude that the studied magmatic rocks of the Kultuma massif were formed under relatively reductive conditions. Quite a different situation is observed for the magmatic rocks of the dyke complex. For instance, zircons from quartz diorite-porphyries, similar to the zircons from the Kultuma massif, were characterized by low Ce/Ce* and Eu/Eu* values, while to the contrary, with respect to the other three parameters, they fit in the region of oxidized granitoids. A similar picture is also observed with the dykes of quartz monzodiorite-porphyries, which have been related to the reduced granitoids of ilmenite series on the basis of three parameters as indicated above. However, unlike for the dykes of quartz diorite porphyries, they are characterized by the higher Eu/Eu* values and close values of Ce/Ce*.

In summary, it may be stressed that the magmatic rocks of the Kultuma massif were formed, most probably, under relatively reducing conditions. As far as the magmatic rocks of the dyke complex are concerned, further investigation is necessary to answer these questions.

5.4. Water Content in Magma

The high content of water in magma is a characteristic sign of the ore-bearing potential for many porphyry magmas. At present, it has been established by many researchers that the Eu/Eu* and Dy/Yb ratios in zircons may serve as an indirect indication of the content of magmatic water [40,59]. For comparison, we also took into consideration the data on porphyry [40,42,60,61] and skarn [59] deposits. One can see in Figure 21b that the major part of the image points of analysis data for zircons from different deposits exhibited higher values of both the Eu/Eu* and Dy/Yb ratios in comparison with the magmatic rocks of the Kultuma massif and the dykes of quartz diorite porphyries. A similar situation was also observed for the Tongshankou Cu–Mo-skarn deposit and the Tonglushan Cu–Au–-Fe-skarn deposit (Figure 21d), while the Fe-skarn deposits Lingxiang and Chengchao were characterized by the lower values of Eu/Eu* and Dy/Yb, close to the values for zircons from the magmatic rocks of the Kultuma deposit and the dykes of quartz diorite-porphyries. The only exception was the dykes of quartz monzodiorite porphyries, which were characterized by higher values of Eu/Eu* and Dy/Yb ratios. Comparing the data obtained by us with the data over other porphyry and skarn deposits listed above, it may be concluded that the melts from which the magmatic rocks of the Kultuma massif and the dykes of quartz diorite porphyries were crystallized originally contained relatively small amounts of magmatic water. Quite contrarily, the melts from which the dykes of quartz monzodiorite porphyries were crystallized contained a larger amount of magmatic water than the melts from which the magmatic rocks of the Kultuma massif and the dykes of quartz diorite porphyries were crystallized.

5.5. Ore Formation Sequence

The main criteria to distinguish the stages of mineral formation are first, the presence of definite paragenetic associations of minerals, their relationship with each other and the features of spatial arrangement of these mineral assemblages both within separate ore zones and over the deposit in general. An analysis of the whole data set described above allowed us to outline the stages of mineral formation, each of them being characterized by a definite paragenetic mineral association (Figure 22). Five main stages were distinguished: prograde skarn, retrograde skarn, potassic alteration, propylitic (hydrosilicate) alteration and late, low-temperature alteration. Hypergene minerals (hypergene alteration) and some rare minerals that do not disturb the general evolution of mineral formation process were not included into this scheme for simplicity.

The formation of the earliest and high-temperature mineral associations (olivinic, pyroxene and garnet skarns), which we relate to the stage of prograde (anhydrous) skarns, is connected with the intrusion of the magmatic rocks of the Shakhtama complex. It should be stressed that the prograde skarn occurs only sporadically at the deposit in the form of relicts, which are almost completely substituted by retrograde skarns and strongly affected by later hydrothermal alteration. The onset of the ore-forming process at the deposit, namely, the start of the deposition of magnetite mineralization, is related to this stage. The major amount of magnetite was formed during the retrograde skarn stage. As a result of prograde skarn alteration, in particular pyroxene–olivine and olivine skarns, the earliest mineral associations of this stage were formed: chondrodite, chondrodite–norbergite and tremolite–chondrodite skarns with magnetite mineralization. Then, against the general background of temperature drop, the formation of amphibole–magnetite and magnetite skarn with different concentrations of the main minerals most frequently occurring within the deposit took place. After the formation of the major amount of magnetite ores, a short pause in mineral formation processes occurred. During this pause, local tectonic processes accompanied, in particular, by the formation of fractured zones were pronounced at the deposit. Disjunctive tectonic processes paved the way for the new portions of ore-bearing solutions differing in their composition from early ore-bearing hydrotherms. In our opinion, they are in relation with the onset of the formation of the earliest sulfide minerals: pyrite, which often has the signs of substitution of magnetite, arsenopyrite and other minerals shown in Figure 22. Later on, new fracture systems were formed, which is evidenced by the extensive development of cataclasis. The grains and aggregates of early sulfide minerals are frequently crushed, cataclastic and healed with later sulfide minerals, mainly chalcopyrite and pyrrhotite. The formation of chalcopyrite (I), pyrrhotite (I) and sphalerite (I), bismuthinite, etc., along with the earliest generation of native gold, took place at a time close to the end of the retrograde stage. It should be noted that the major ores with chalcopyrite and pyrrhotite mineralization are confined mainly to the retrograde skarn.

In our opinion, the complex of calcic-potassic hydrothermal alterations proceeded close in time to the final steps of the retrograde stage. As mentioned above, this complex of hydrothermal alterations is most extensively manifested within the boundaries of the Ochunogda region of the deposit, while at other regions of the Kultuma deposit, they are weakly pronounced. Both for the retrograde stage skarn and for the zones of potassic hydrothermal alterations, a similar sequence of the formation of sulfide minerals is outlined. For instance, one of the earliest mineral associations within the Ochunogda region is the arsenopyrite–pyrite association with the prevalence of arsenopyrite, unlike for other regions of the Kultuma deposit, where arsenopyrite only holds a suborninate role. The grains of early sulfide minerals are often crushed and cataclastic; they are cemented with chalcopyrite (I), along with pyrrhotite (I) and sphalerite (I), which is certain evidence of their later formation. The early generation of native gold is also connected with the formation of the chalcopyrite–pyrrhotite–sphalerite mineral association. One can see that, independently of the regions of the deposit, the general sequence of the formation of sulfide minerals is conserved. In this connection, we did not distinguish more detailed additional generations of sulfide minerals, for example, one more generation for chalcopyrite, pyrrhotine and sphalerite characteristic of the zones of potassic alteration at the Ochunogda region. In our opinion, these processes (the terminal steps of the stage of retrograde skarn and potassic hydrothermal alteration) took place approximately at the same time but in different geological settings. We also relate the formation of rare thin quartz–K-feldspar–carbonate veins with molybdenite within the Kultuma massif to the potassic hydrothermal alteration. The formation of the major part of sulfide minerals (galena, fahlore, sphalerite (II), chalcopyrite (II), native gold (II), etc.) at the polymetallic stage proceeded at the stage of propylitic hydrothermal alteration, which we consider to also include chloritization, serpentinization and other alterations (for example, talc alteration). Propylitization affected great masses of rocks, both in the exo and endo contact zones. Here, typical minerals are chlorite, serpentine, talc, amphibole, tourmaline, quartz and various carbonates. At the same time, the major part of sulfide minerals of the polymetallic stage occur most frequently at a larger distance from the direct contact of the magmatic rocks of the Shakhtama complex with the embedding carbonate terrigenous sediments. However, locally, the sulfide minerals of the polymetallic stage also occur in the form of intercrossing thin veins and veinlets in the central parts of the Kultuma massif, as well as at its southern margin (the Ochunogda region). The veinlets and veins are composed mainly of quartz, various carbonates and chlorite and less frequently rutile, amphibole (actinolite) and epidote. Unlike for the zones of potassic hydrothermal alteration, such minerals as K-feldspar and biotite are of sharply subordinate significance. The formation of one of the latest mineral associations, namely, antimony one, proceeded at the final stages of propylitization. The formation of the latest generation of native gold is likely to be related to this stage. Antimony-containing hydrothermal solutions penetrated the same weakened zones and cracks and were overlapped on the previously deposited minerals of the polymetallic stage, thus forming reactive minerals—in particular, aurostibite. However, at the same time, the minerals of the late antimony stage occur extremely scarcely. We relate the latest low-temperature hydrothermal alterations to the terminal stages of hydrothermal activity and assign the formation of the later generation of pyrite and marcasite to these alterations. After that, the ores of the deposit underwent hypergene transformations, which was expressed in the formation of malachite, azurite, scorodite, iron hydroxides and other hypergene minerals.

5.6. Interpretation of the Data on Stable Isotopes

Results of the studies of the isotope composition of sulfur in sulfide minerals revealed substantial enrichment in the heavy sulfur isotope. According to modern opinion, enrichment of sulfide sulfur with the heavy isotope may be explained by a complete or partial intake from the sedimentary rocks enriched with sulfates from seawater or evaporates [62,63]. A comparison of the isotope composition of sulfur from the sulfide minerals of the Kultuma deposit with the data on other gold ore and gold-bearing deposits and occurrences in paragenetic relationship with the magmatic formations of the Shakhtama complex revealed both similarities and differences. The deposits in which the ores are localized in intrusive formations—either directly in the magmatic rocks of the Shakhtama complex or in the host rocks, which are most frequently the granitoids of the Undinsky complex—were characterized by the light isotope composition of sulfur. For example, for the W–Mo-porphyry gold-bearing Bugdainskoye deposit, the range of δ³⁴S in sulfide minerals is −0.9 to 4.6‰ [64]. Another example is the Antiinsky ore occurrence (vein type): here the range of δ³⁴S values in sulfide minerals varies from 1.2 to 4.4‰ [8]. One can see that this small increase in the content of heavy sulfur isotopes, with respect to the juvenile values, points to its magmatic source. The close values were also established for the sulfide minerals of the Lugokan Au–Cu–Fe–skarn deposit [9]. In this case, similarly to the Kultuma deposit, the host rocks were represented by the sediments of the Bystrinskaya formation. The δ³⁴S values for sulfide minerals were within the range of 2.5 to 6.6‰. A heavier sulfur isotope composition was detected for the sulfide minerals of the Bystrinskoye Au–Cu–Fe–skarn deposit, which varied from 6 to 13.4‰ (10.0‰ on average). In our opinion, these differences were due, first of all, to the composition of the host rocks. The enrichment of sulfides with a heavy sulfur isotope could also be explained by an increase of the ore-bearing fluids’ pH because of the interaction with host carbonates (especially with dolomites). Since an increasing pH affects the sulfur compounds in the fluid, the aqueous sulfide ions would tend to become enriched in the heavy isotope, as would a sulfide mineral precipitated in isotopic equilibrium with the fluid. According to Sakai [65] and Ohmoto [66], a simultaneous increase of both δ34S and pH is only possible at very low oxygen fugacity. In other words, the sulfides had to crystallize from the reduced fluids. This is consistent with our data that the magmatic rocks of the Kultuma massif belong to the reduced granitoids of the ilmenite series. Moreover, substantial enrichment of sulfur in the sulfide minerals of the Kultuma and Bystrinskoye deposits with the heavy isotope may have been caused by the interaction of a high-temperature fluid with the carbonate rocks of the Bystrinskaya formation. Most probably, the conditions of the formation of carbonate rocks in the Bystrinskaya formation were somewhat varyied depending on their regional location. In this case, it is quite natural to assume that some carbonate rocks of the Bystrinskaya formation could be enriched with sulfate from seawater. As a result, particularly at the Kultuma and Bystrinsky deposits, sulfide minerals are enriched with the heavy isotope of sulfur, unlike for the Lugokan deposit. One of the confirmations may be the measured isotope composition of anhydrite (24.4–25.6‰) from the zones of potassium hydrothermal alterations, which is comparable with the sulfur composition in the evaporates of the same age as that of the host carbonate rock mass of the Bystrinskaya formation [67]. In addition, the sample with the heaviest isotope composition of sulfur (pyrite—16.0‰) was taken from the host dolomites of the Bystrinksya formation. The only exception in which anomalously low δ³⁴S values (from 1.4 to 3.7‰) were detected was pyrrhotite from retrograde skarns of the Ochunogda region. Here, unlike for other regions of the Kultuma deposit, the dominating rocks are terrigenous carbonates of the Ernichenskaya formation, which often contain small amounts of carbonaceous matter of biogenic origin. Most probably, the anomalous low values were due to the contamination of the ore solution with sedimentary sulfides associated with the sublayers of carbonaceous substance in the host terrigenous carbonate sediments.

The results of the investigation of the isotope composition of C and O allowed us to trace several evolutionary trends (Figure 23). The first trend was expressed in a decrease in δ¹⁸O and a decrease in δ¹³C when passing from weakly altered dolomites of the Bystrinskaya formation to skarned dolomites, which may be explained by decarbonization and/or the interaction of carbonate rocks with the magmatic fluid. This trend was clearly observed for the Km-3 well. For instance, the least altered dolomites of the Bystrinskaya formation were characterized by δ¹⁸O 23.9‰ (Km-3/459.3) and 22.2‰ (Km-3/447.7), and δ¹³C −3.5‰ (Km-3/459.3) and −3.1 ‰ (Km-3/447.7), which is characteristic of marine carbonate rocks [50], while skarnified and/or weakly altered dolomites were characterized by lower δ¹⁸O values: 17.1‰ (Km-3/481.5) and 16.6‰ (Km-3/442.5), and higher δ¹³C values: −3.1‰ (Km-3/481.5) and −2.7‰ (Km-3/442.5). Quite a different situation was observed when passing from host rocks to retrograde skarns, with a stronger change in δ¹⁸O. One can see that calcite sampled from retrograde skarns (Km-3/405.6) was characterized by even lower δ¹⁸O values (8.7‰), while the change of δ¹³C was less clearly pronounced (−3.4‰). The value of δ¹⁸O (Km-3/405.6) corresponded to indicator values for granitoids and was connected with an increasing role of the magmatic fluid. Two trends were also outlined for calcite sampled from the zones of propylite alteration. For instance, in the Km-3/481.5 sample, the transition from weakly altered dolomite to calcite (with tourmaline and chlorite) veinlets with the minerals of the polymetallic association, δ¹³C increased from −3.1‰ (Km-3/481.5 dolomite) to −2.1‰ (Km-3/481.5 calcite veinlets), while δ¹⁸O decreased from 17.1‰ (Km-3/481.5 dolomite) to 12.6‰ (Km-3/481.5 calcite veinlets). This may also be explained by decarbonization and/or the interaction of carbonate rocks with the magmatic fluid. Another trend was expressed in a decrease in δ¹⁸O and δ¹³C. For example, for calcite sampled from the zone of propylite alteration (with the sulfide minerals of the polymetallic association) of retrograde skarn, δ¹⁸O values were equal to 4.4 ‰ and δ¹³C −4.5‰ (Km-3/385.8). This trend may be explained by the arrival of heated meteoric waters along the zones of intense fracturing. A decrease in δ¹³C may be connected with the arrival of light carbon isotopes from the organic sedimentary matter (the average value is δ¹³C = ~ −25‰ [50]) dissolved in meteoric waters and oxidized into carbonate. The trends considered with the example of well Km-3 are also characteristic of other regions of the deposit. To sum up the data on the isotope composition of carbon, it may be concluded that one of the major sources of carbon are the host rocks of the Bystrinskaya formation. The changes in the isotope composition of oxygen are connected mainly with decarbonization and interactions of magmatic fluids with host rocks. Some differences observed by us were also due to the local geological conditions; for example, the position of ore formations and host rocks with respect to the magmatic bodies of the Shakhtama complex, which are represented both by small bodies, dykes, apophyses of the Kultuma massif and by the Kultima massif itself. The possible effect of meteoric waters on the isotope composition of oxygen and carbon also should not be left unmentioned.

5.7. Hydrothermal Evolution

The data obtained in the studies of fluid inclusions showed that the formation of retrograde skarn was participated by the reduced (the presence of methane in the gas phase of fluid inclusions), moderately hot (~360–370 °C) fluid containing the high concentration of carbon dioxide. It may be suggested, relying on the positions of fluid inclusions within the grains, that mineral formation proceeded at first from the homogeneous fluid. However, later on, the fluid might have boiled, which is evidenced by the joint positions of VC and M type inclusions. Later hydrothermal processes imprinted in pseudo-secondary fluid inclusions took place with the participation of lower-temperature fluid (~200 °C) rich in carbon dioxide.

Higher-temperature (~420–440 °C), high-pressure (pressure during mineral formation up to 2.4 kbar) fluid enriched with carbon dioxide was responsible for the formation of mineral associations in the zones of potassic alteration. During mineral formation, the content of carbon dioxide in the fluid and fluid temperature decreased. In addition, the active deposition of ore minerals at this stage (at the Kultuma deposit) was associated with the boiling of ore fluid and its separation into significantly carbonaceous and significantly aqueous components. It should be stressed that the fluid inclusions in quartz from the Ochunogda region contained admixtures of methane and nitrogen and remained homogeneous. The estimated pressure for this region was lower (~1.7 kbar) than for the Kultuma deposit.

Later minerals from propylite alteration, including sulfites of the polymetallic association, were formed from a heterogeneous reduced low-temperature (~280–320 °C) fluid rich in carbon dioxide. Estimation of the pressure of mineral formation for it gave a minimal value: 1–1.2 kbar. It is worth noting that the major part of the studied fluid inclusions in quartz from the zones of propylitic alteration contained methane in the gas phase, in addition to carbon dioxide, and trace amounts of ethane, the second representative of the homologous series of alkanes, was detected in some inclusions. The dominant component of the fluid was generally carbon dioxide, though boiling of the solution was detected, with its separation into the gas phase containing mainly carbon dioxide and the aqueous phase rich in carbon dioxide (nahcolite crystals in fluid inclusions). Subsequent evolution was probably accompanied by a decrease in CO₂ concentration in the fluid, and mixing with meteoric waters occurred, which was expressed in a simpler composition of the gas phase, the formation of chloride as a daughter mineral phase and a decrease in the homogenization temperature of fluid inclusions (245–260 °C).

5.8. The Genetic Type of the Kultuma Deposit

The obtained set of new data allows us to consider the problem of the genetic typification of the Kultuma deposit. At present, many researchers refer Au–Cu–Fe–skarn deposits related to the Shakhtama complex to the Au–Cu–Mo-porphyry or combined skarn–porphyry type connected with adakites. These conclusions are based, in particular, on the assignments of magmatic rocks present within these deposits to the magnetite (oxidized) series [4,64]. The uncertainty of these ideas is demonstrated, first of all, by the absence of any data on the magnetic susceptibility of magmatic rocks developed at Au–Cu–Fe–skarn deposits, on their petrographic composition (the presence of accessory minerals—magnetite or ilmenite) and the Fe₂O₃/FeO ratio. The new data obtained by us on the magnetic susceptibility, petrographic composition and Fe₂O₃/FeO ratio for the magmatic rocks of the Kultuma and Lugokan massifs showed that they belong to the ilmenite (reduced) series, as indicated by all the parameters. As far as the assignment of these rocks to adakites or adakite-like rocks is concerned, we agree with the conclusions made by Kovalev et al. [6] that the magmatic rocks of the Kultuma massif and the dykes do not exhibit typical features of adakites. In addition, the first data obtained on the chemical composition of zircons from the magmatic rocks of the Kultuma massif and the dyke complex also provide evidence in favor of their assignment to the ilmenite series. In particular, comparing the values of Ce/Ce* and Eu/Eu* anomalies with those of porphyry and skarn deposits related to the magmatic rocks of the magnetite series, one may note that the magmatic rocks of the Kultuma massif are characterized by the lower values of Ce/Ce* and Eu/Eu* anomalies, which is evidence of their formation under relatively reductive conditions. These conclusions are also confirmed by the ratios of Eu/Eu* and Dy/Yb in zircons, which may serve as indirect signs of the presence of magmatic water. As far as the role of the magmatic rocks of the dyke complex is concerned, their part could hardly be essential because the degree of their occurrence is low, and they are most widespread outside the boundaries of the deposit. It is most probable that the magmatic rocks of the dyke complex only have a local effect on the oxidation–reduction setting of ore formation.

It is known from the published studies that the reductive or oxidative environment is conserved for the skarn and post-skarn hydrothermal–metasomatic processes, which is particularly considered to be related with economic gold mineralization [68,69]. This is how the data on the composition of rock-forming minerals and ores may be used to evaluate the oxidation–reduction characteristics of the environment. The reduced prograde skarn is characterized by the prevalence of pyroxene over garnet, and pyroxene itself is most frequently represented by hedenbergite (or > Hd₅₀). In the samples studied by us, the minerals of prograde skarn occurred most frequently as relicts, so a reliable quantitative evaluation seems impossible. We may only rely on indirect signs. For instance, garnet occurred only rarely in the samples under investigation, while pyroxene was more frequent. This may suggest that pyroxene dominated over garnet in prograde skarn. Pyroxene was represented both by diopside and by hedenbergite (according to the SEM data), which may in turn be evidence of the change of an oxidation–reduction potential as early as at the progressive stage of skarn formation. Analyzing the data on the mineral composition, we may state the high role of Fe²⁺, both at the stages of the formation of prograde and retrograde skarns, and later at hydrothermal–metasomatic alterations. The presence of divalent iron in amphiboles, some carbonate and sulfide minerals, as well as the very poor occurrence of minerals with trivalent iron (for example, epidote) at later stages points to the reductive environment. At the same time, for the data on the mineral composition of separate rocks, we may conclude that the process of mineral formation included several changes of the oxidation–reduction potential. These conclusions are also confirmed by the results obtained in the investigation of fluid inclusions, which demonstrated that the formation of both retrograde skarn and the zones of potassic and propylite alteration were participated by reduced fluids rich in carbon dioxide. Such a high carbon dioxide content is explained by the interaction of magmatic fluids with the host carbonate masses. Carbon dioxide is the key component in the formation and evolution of metal-bearing fluids in ore systems. According to the available data, CO₂ may act either as a direct agent of copper and gold transfer or play a secondary role, accelerating the separation of the ore-bearing fluid phase [70]. It may be assumed that the saturation of the melt with carbon dioxide (CO₂) during the formation of skarn at the Kultuma deposit promoted the removal of copper from it [71]. If we take it for granted that only the degree of magma oxidation is responsible for the oxidation–reduction environment of mineral formation, some discrepancies may be noticed. This is one of the reasons why we suppose that the conditions of the formation of mineral parageneses were substantially affected by the composition of host rocks and by the local geological factors, which may also include structural–tectonic factors. These conclusions may also be confirmed by the data on stable isotopes. As stated above, the carbonate rocks of the Bystrinskaya formation might have been enriched with sulfates. The interaction of reduced magmatic fluids with the host rocks of the Bystrinskaya formation has most probably led to the oxidation of ore-bearing fluids and to the formation of magnetite ores. A mechanism of this type was described in the works by Wen et al. [72], Duan et al. [73] and is likely to be responsible for the formation of magnetite ores at the Kultuma deposit.

In our opinion, the data presented above provide evidence in favor of the assignment of the Kultuma deposit to Au–Cu–Fe–skarn deposits related to reduced intrusions.

6. Conclusions

• The Kultuma Au–Cu–Fe–skarn deposit was related with a meta-aluminous, magnesian, high-potassic calc-alkalic reduced (ilmenite-series) I type granitoids.
• Zircon Ce/Ce*, Eu/Eu* and Dy/Yb ratios can be used as a proxy for magmatic oxidation state and magmatic hydrous states. Low Ce/Ce*, Eu/Eu* and Dy/Yb in zircon of Kultuma massif indicated reduced conditions and a low amount of magmatic water, unlike Cu–Mo–Au-porphyry deposits. These results are in good agreement with the data on the magnetic susceptibility of magmatic rocks, petrographic composition and the Fe2O3/FeO ratio.
• For the magmatic rocks of the Kultuma massif, the calculations yielded zircon crystallization temperatures of 595–801 °C (averaging 638 °C) (indicated by Ti-in-zircon thermometer) for magmatic rocks of the dyke complex, 648 to 719 °C (averaging 683 °C) for quartz monzodiorite-porphyries and 598 to 735 °C (averaging 639 °C) for quartz diorite-porphyries.
• Formation of magmatic rocks of the Shakhtama complex (161–156 Ma) and ore mineralization (156 Ma) within the Kultuma deposit proceeded during the late Jurassic at the final stages of the marginal continental complex of the Siberian and North-Mongol–Chinese continent collision.
• The Kultuma deposit represents a complex Au–Cu–Fe mineralized system that comprises early prograde skarn, retrograde skarn, potassic alteration, later propylitic (hydrosilicate) alteration and the latest low temperature alteration. The major gold, copper and iron ores related with retrograde skarn and potassic alteration, while polymetallic ores related with later propylitic alteration. The detailed data on the chemical composition of various minerals provided evidence of the multiple changes of physicochemical conditions during the formation of the ores at the deposit.
• The changes in isotope composition oxygen, carbon and sulfur were related mainly with decarbonization and interactions of magmatic fluids with host rocks and effect meteoric waters.
• The reduced, moderately hot (~360–440 °C), high-pressure (up to 2.4 kbar) fluids participated in the formation of retrograde skarn and the zones of potassic alteration mineralization. The fluids in the potassic alteration zones at the Ochunogda region were characterized by the absence of daughter mineral phases in fluid inclusions and lower estimated pressure values (~1.7 kbar). As for propylitic alteration, the reduced lower-temperature (~280–320 °C) and lower-pressure (1–1.2 kbar) fluids saturated with carbon dioxide were ore-forming ones. They became more water-rich and low-temperature (~245–260 °C) because of mixing with meteoric waters. It is necessary to stress the general saturation of the fluids with carbon dioxide, which might be one of the major factors establishing the features of the propagation of ore mineralization, in particular copper.
• The deposit could be part of a broader regional-scale magmatic–hydrothermal system comprising almost coeval reduced intrusion-related Au–Cu–Fe–skarn deposits and vein/stockwork Au deposits.