Timing and Evolution of Gold Mineralization in the Maljavr Occurrence (NW Russia), NE Part of the Fennoscandian Shield

Abstract

Gold occurrence Maljavr is the first Archean conglomerate-hosted gold mineralization found in the Fennoscandian Shield. Gold-mineralized metasomatic rocks form a set of lenses within a 10 m thick linear zone, conformable to the bedding of host conglomerates. The lenses are up to 10 m long and up to 1 m thick and they clearly exhibit three alteration envelopes: the rock in the central part consists of garnet and quartz or garnet-only; biotite, garnet, and quartz make the intermediate biotite–garnet envelope; hornblende, hedenbergite, and quartz are the principal rock-forming minerals in the outer zone of the lenses. All metasomatic rocks contain sulfide mineralization up to 15–20 vol.% and up to 0.6 g/t Au. The main ore mineral is pyrrhotite, and the minor minerals are arsenopyrite, chalcopyrite, pentlandite, löllingite, and troilite. The age of zircon from biotite gneiss in the zone of alteration is 2664 ± 18 Ma, this is considered as the time of formation of lenses of metasomatic rocks. Biotite gneiss-conglomerate and metasomatic rocks were later intruded by tourmaline granite pegmatite 2508 ± 7 Ma. The injection of pegmatite caused re-crystallization of sulfides (mainly arsenopyrite and löllingite) and redistribution of gold. Visible gold in association with Bi minerals native bismuth, ehrigite, maldonite, bismuthinite, joseite-B, and hedleyite was found in inclusions in recrystallized arsenopyrite and löllingite. Au content in the rocks with recrystallized arsenopyrite and löllingite is >1 g/t, up to 30 g/t in hand samples. The 2508 Ma pegmatite is interpreted as synchronous with formation of gold mineralization in its present form. The linkage of gold mineralization with pegmatite and geochemical association Au-As-Se-Te-Bi in the mineralized rocks agree with characteristics of intrusion-related gold deposits worldwide. Biotite gneiss–metaconglomerate, hosting the mineralized altered rocks, was the probable primary source of arsenic and gold for mineralization.

1. Introduction

Paleoplacers in Precambrian conglomerates are the most important genetic type of gold deposits: half of gold ever mined on the Earth was mined from Witwatersrand Mesoarchean conglomerate basin in South Africa [1]. Conglomerates with gold are known in other Precambrian shields in the world as well—for example, Tarkwa in Ghana (Western Africa) [2], Jacobina in Brazil (South America) [3], Pardo in Canada (North America) [4,5], Purdy’s Reward in Pilbara Craton (Australia) [6,7], Gorumahisani-Badampahar conglomerates in Craton Singhbhum, India (Asia) [8]. High gold grades are mainly connected with alluvial placers along river paleovalleys, but sea beach paleoplacers are also known (for example, in Australia—[6]). In the Fennoscandian shield, conglomerates with gold were found in the Lehtinskaya and Onezhskaya structures in Russian Karelia and in the Central Lapland greenstone belt in Finland, where these gold occurrences are considered as Paleoproterozoic gold paleoplacers in fluvio-deltaic fans [9,10,11].

The Precambrian gold-bearing conglomerates are either monomictic with quartz pebbles or polymictic with pebbles of different composition, up to ultramafic pebbles [4,5,6,7]. Gold concentrates mainly in the conglomerate matrix, not in the pebbles [1,2,3]. The matrix may be clayey, sandy, or up to coarse-grained sands.

Paleoplacers (and gold as a mineral) were more or less modified during late regional metamorphic and hydrothermal-metasomatic events. The paleoplacers, mentioned in the previous paragraph, are mainly metamorphosed greenschist [1,2,3,4,5,6,7,8,9,10,11]. If these modifying events are more intensive (amphibolite facies), then gold deposits and occurrences will display features of orogenic and/or intrusion-related deposits: structural-control of mineralized rocks, mineral associations (amphibole, pyroxene, biotite, garnet, etc.) indicating high PT-conditions of metasomatic processes in the altered rocks, geochemical association of minor elements introduced by fluids of metamorphic or magmatic origin. In this case paleoplacers play a role as interim gold concentrators.

The Maljavr gold occurrence, located in amphibolite metamorphosed rocks in the Uragubsky greenstone belt (north-eastern part of the Fennoscandian Shield) (Figure 1, inset), is an occurrence of this kind. It is the first conglomerate-hosted gold occurrence of Archean age to be found in the Fennoscandian Shield.

2. Geological Overview of the Study Area

The Maljavr gold occurrence is located in the north-western part of Russian Arctic within the Uragubsky greenstone belt (Figure 1, inset). This belt, together with the larger and more extensive Kolmozero-Voron’ya belt in the south-east, makes a system of greenstone belts, which traces the boundary between two big blocks of the Fennoscandian Shield—the Murmansk Craton and the Kola-Norwegian Province (or simply Kola Province) (Figure 1, inset).

The Uragubsky belt has an approximately 80 km strike length and a width of up to 4 km. The history of the formation of the Uragubsky and Kolmozero-Voron’ya belts, as described in [13], includes the following events: formation of an island arc at 2.87–2.83 Ga; a break of 50 million years and subsequent accretion of the island arc to the Murmansk continent; formation of an accretionary orogen (2.78–2.76 Ga); probable collapse of the orogen with formation of post-orogenic (or possibly anorogenic) granite intrusions.

The geological section of the Uragubsky belt includes the lower terrigenous sequence with basal conglomerates, the middle komatiite–basalt volcanic sequence, and the upper sedimentary-volcanic sequence [14]. The lower terrigenous sequence is composed mainly of fine and medium grained biotite and two-mica gneisses and cataclastic pebble and boulder–pebble rocks. The middle sequence is composed of mafic–ultramafic volcanics; it also contains tuff horizons, fine-grained biotite and biotite–amphibole gneiss horizons and lenses, layers of BIF and magnetite–amphibole rocks (Figure 1). The upper sedimentary-volcanic sequence consists predominantly of biotite, biotite–garnet, biotite–staurolite–sillimanite, biotite–sillimanite–garnet gneisses with subordinate thin foliated amphibolite bodies. Chemical composition of the gneisses corresponds to dacite and rhyolite; paragneisses are subordinate. The U-Pb SHRIMP II age of zircon from gneiss–metadacite is 2838 ± 23 Ma, this age is considered as the age of volcanism [15,16].

The volcanic–sedimentary rocks were metamorphosed amphibolite at a temperature of 600–620 °C, under a pressure of 4 kbar [17]. The U-Pb age of metamorphism was reported as 2786 ± 31, 2774 ± 12 Ma (SHRIMP-II, zircons of metamorphic genesis) [15], 2763 ± 8 Ma [14].

The volcanic–sedimentary sequences are cut by intrusions of plagioclase/microcline granite (2696 ± 9 Ma), aplite veins (2697 ± 10 Ma) [18], and tourmaline granite pegmatite veins.

Pebble conglomerates were mapped at the southern contact of the belt with migmatized gneisses and plagiogranites of the Kola province in the lower part of the basal terrigenous sequence [14,19,20]. Thickness of the conglomerate layer is 2–5 m, sometimes up to 10 m. The conglomerates are matrix-supported, pebble dimensions vary predominantly from 3 × 5 to 5 × 10 cm. Fine grained greywacke (biotite and muscovite–biotite gneiss) makes the conglomerate matrix.

Layers and lenses of pebble conglomerate and gritstone were also observed throughout the entire biotite gneiss sequence. Every conglomerate layer is an alternation of conglomerate, gritstone, and sandstone (Figure 2A,B) with thickness of strata 10–20 cm. The rocks preserved obvious signatures of their terrigenous origin such as graded rhythmical bedding (Figure 2A,B) and psephytic–psammitic structures. The clasts consist predominantly of plagiogranite (up to diorite) and quartz; pebbles and gravel are cemented by biotite or biotite–muscovite gneiss–metagraywacke. Plagiogranite in the pebbles contains partially sericitizated oligoclase, quartz, and minor biotite and tourmaline. Gneiss in the matrix is similar to the biotite paragneiss in the lower sequence of the volcanic–sedimentary section as it consists of quartz, plagioclase (labradorite), biotite, ±muscovite; accessory minerals are apatite, zircon, chlorite, tourmaline, allanite.

3. Materials and Methods

Investigations of wall rock alteration, metasomatic zoning, and determination of pre-ore, ore-related, and post-ore mineral assemblages in altered rocks were based on the study of rocks in the outcrops, on examination of mineral relations in thin and polished sections, as well as on the results of assaying the primary and altered rocks. More than 50 specimens and samples (1.5–10 kg), supplemented by chips for thin and polished sections, were collected by the authors in the outcrops and trenches in 2022–2023. The samples were assayed for 12 major (rock-forming) elements in the chemical laboratory of the Geological Institute, Kola Science Centre of Russian Academy of Sciences, Apatity, Russia, with flame atomic absorption spectrometry (FAAS). Data on 40 trace elements, which determined the geochemical characteristics of the rocks, were obtained by ICP-MS with nitric acid digestion in the Institute of Geology and Geochemistry of the Ural Branch of the Russian Academy of Sciences, Ekaterinburg,

Mineral composition of the ores was studied in thin and polished sections of the rocks and in epoxy-polished sections of heavy mineral concentrates, separated with the Sochnev magnet and bromoform from the crushed hand samples. Thin and polished sections were studied with the reflected light microscope Axioplan 2 Imaging (Carl Zeiss, Jena, Germany) and with the electron microscope Zeiss Evo-25 (Carl Zeiss, Jena, Germany), equipped with the energy-dispersive system (EDS) AZtec UltimMax-100 (Oxford Instruments, Oxford, UK) in the Geological Institute of the Kola Science Center. Sulfides, tellurides, and native metals were identified and photographed in the regime of back-scattered electron image (BSE). Chemical composition of mineral phases was defined with EDS AZtec UltimMax-100 (Oxford Instruments), operating at accelerating voltage 20 kV, beam current 2 nA, spectrum accumulation time 100 s. The following analytical lines and standards were used for silicates and oxides: FKα (fluorapatite), SiKα and CaKα (wollastonite), ClKα (atacamite), FeKα (hematite), LaLα (LaSe), CeLα (CeS), NdLα (LiNd(MoO₄)₂), TiKα and NaKα (lorenzenite), MnKα (MnCO₃), SKα, and SrLα, MgKα, AlKα, KKα, NiKα (metal); analytical lines (standards) for sulfides and metals were PbMα and TeLα (PbTe, in some cases HgTe for TeLα), BiMα and SeLα (Bi₂Se₃), AsKα (InAs), AgLα (metal), AuMα (metal), FeKα и SKα (pyrite) (in some cases metal for FeKα), NiKα (metal), CoKα, (metal), ZnKα (metal or sphalerite), BrKα (KBr), CdLα (metal or CdSe).

Microprobe analysis (MS-46, CAMECA, Gennevilliers Cedex, France), accelerating voltage 22 kV, beam current 30–40 nA, standards (analytical lines; detection limit, wt%): Fe₁₀S₁₁ (FeKα; 0.01 and SKα; 0.05), Bi₂Se₃ (BiMα; 0.1 and SeKα; 0.05), LiNd(MoO₄)₂ (MoLα; 0.1), PbS (PbMα; 0.1), pure metals Co (CoKα; 0.01), Ni (NiKα; 0.01), Cu (CuKα; 0.01), Pd (PdLα; 0.05), Ag (AgLα; 0.05), Te (TeLα; 0.05), Re (ReLα; 0.05), Au (AuLα; 0.05), was performed for grains larger than 20 μm (analyst Ye. Savchenko). Beam size was 2–10 nm, depending on stability of the matter under the beam current. Measurement time was 10 s for major elements, 20 s for minor elements, and 10 s for background measurements. The results of 4–5 measurements for each element were averaged.

Zircon U-Pb dating was undertaken at the Geological Institute of the Kola Science Center (Apatity, Murmansk region, Russia). Isotope dilution-thermal ionization mass spectrometry (ID-TIMS) was used. Prior to analysis, zircons were extracted using magnetic separation with the Sochnev magnet and heavy liquid separation in bromoform, with surface contamination removed using alcohol, acetone, and 1 M HNO₃.

The zircon dissolution and chemical recovery of Pb and U were performed using the technique described in [21], with U and Pb concentrations determined by isotope dilution employing a Finnigan MAT-262 (RPQ) mass spectrometer and a mixed ²⁰⁸Pb + ²³⁵U tracer, with silica gel used as an ion emitter. Blank levels had maximum values of 100 ng Pb and 10–50 ng U, and all isotope ratios were corrected for mass fractionation by analysis of the SRM-981 and SRM-982 standards (0.12 ± 0.44%). The uncertainties on the resulting U-Pb ratios are 0.5%. The raw experimental data were processed using PbDAT and ISOPLOT [22,23], with age values calculated using conventional U decay constant values [24] and common Pb corrections following [25]. All uncertainties are reported at a 2σ confidence level.

4. Results

4.1. The Maljavr Occurrence: Geological Structure, Alteration, Ore Minerals

The Maljavr gold occurrence is located at the SW flank of the belt, where biotite gneiss of the lower sequence forms an anticline fold with the upper bend dipping SE (Figure 1). Biotite gneiss (metamorphosed sandstone, gritstone, and conglomerate) is cut by a tourmaline granite pegmatite N-striking vein, about 40 m thick (Figure 3). Rock forming minerals in pegmatite are quartz, plagioclase (albite–oligoclase), microcline, muscovite, and tourmaline, and minor minerals are biotite, apatite, zircon.

The western contact of the vein is sharp, but the eastern contact is transitional, through a 10 m-wide zone of foliation (Figure 4A) and migmatization (Figure 4B). In the zone of foliation, pebbles are deformed together with the matrix (Figure 4A). The biotite gneiss is intensely tourmalinizated at both sides of the pegmatite vein (Figure 3). Deformation and tourmalinization intensity in the biotite gneiss decreases within ~20 m distance from the pegmatite contact.

Smaller lenses of pegmatite were found up to 50 m away from the main pegmatite vein and zone of migmatization. The biggest lenses are up to 20 m long, smaller lenses are less than 20–30 cm (Figure 4C,D). The pegmatite lenses cut biotite gneiss, including the pebbles in metaconglomerate (Figure 4C).

A series of lenses of metasomatic rocks is controlled by a ~10 m thick zone, striking NNE 10–15° and dipping vertically, conformably to schistosity in the biotite gneiss (Figure 3). This zone was traced for 50 m in outcrop and farther on is covered by soil. The biggest lenses of metasomatic rocks reach the length of 10 m and 1 m thickness (Figure 3), small lenses have less than 1 m strike length (Figure 5).

The lenses of metasomatic rocks clearly exhibit three alteration envelopes (Figure 5). The central parts (the cores) of the lenses consist of garnet and quartz (Figure 6D) or garnet-only (garnetite) (Figure 5), the cores are 10–40 cm thick. Biotite, garnet, and quartz make the intermediate biotite–garnet envelope (Figure 6C). Thickness of this envelope is up to 20 cm in the biggest lenses. Hornblende, hedenbergite, and quartz are the principal rock-forming minerals in the outer zone, where minor minerals are epidote, quartz, and garnet (Figure 6B). Thickness of the outer hornblende–hedenbergite envelope is variable, not more than 20 cm; in some lenses, this zone is reduced or absent.

Metasomatic rocks are massive, medium- to coarse-grained: garnet grains reach 1 cm, other minerals are 1–3 mm in size. Biotite flakes and needle crystals of amphibole and pyroxene are differently oriented (Figure 6B,C). Quartz veinlets penetrate all zones of metasomatic rocks, the veinlets are oriented along the strike of the lenses (Figure 5).

Biotite gneiss in between the lenses is generally similar to the gneiss throughout the entire biotite gneiss sequence, but garnet appears (1–2 vol.%) and sulfide content increases from <1 to 1–2 vol.% close to the contact with metasomatic lenses.

The spidergrams (Figure 7) illustrate change in chemical composition of the rocks during alteration processes. The diagrams exhibit general loss of SiO₂ and Na₂O and gain of FeO in all metasomatic rocks: Fe content systematically increases from the outer envelope to the core of the lenses. Al₂O₃, CaO, MgO, K₂O were re-deposited during alteration and formed metasomatic zoning. Hornblende–hedenbergite rock has high CaO, low Al₂O₃, Na₂O, and K₂O. Biotite–garnet rock contains high Al₂O₃, MgO, and K₂O, but low CaO. Garnet–quartz rock in the central part of the lenses has high FeO, and lower content of other general elements (Figure 7, Table S1).

Change of mineral composition of the rocks during alteration processes is chiefly limited to severe alteration of plagioclase in the biotite gneiss, recrystallization of biotite and quartz, and formation of garnet, amphiboles, and pyroxene. Sodium, released during plagioclase alteration, was removed from the altered rocks. Calcium was re-deposited in the outer hornblende–hedenbergite envelope, where it formed hornblende, hedenbergite, and high-Ca almandine. The residual alumina from the altered plagioclase together with iron, which was added to the system, formed garnet (almandine) in biotite–garnet and garnet–quartz rocks. Potassium, concentrated in the primary biotite in gneiss, was re-deposited mainly in the garnet–biotite rock.

GER (general element ratio) diagrams [27] (Figure 8) illustrate change of both mineral and chemical composition during alteration and formation of metasomatic zoning in terms of element ratios: loss of sodium due to severe alteration of plagioclase (Figure 8A,B), redistribution of potassium, connected with recrystallization of biotite (Figure 8A,B), addition of Fe and redistribution of Ca and Al, which lead to formation of Ca-rich almandine, hornblende, and hedenbergite (Figure 8C,D).

All Fe-Mg silicates in the altered rocks are Fe-rich (Table S2): pyroxene is hedenbergite, amphiboles are ferro–hornblende and grunerite, chlorite is chamosite, and mica is Fe-biotite. Garnet is almandine in the garnet–quartz and garnet–biotite rocks, and Ca-rich almandine with mole proportion of grossular 14–28% in the hornblende–hedenbergite rock (Table S2).

Altered rocks, if compared with the unaltered biotite gneiss, display loss of Rb, Cs, Sr, Ba, Zr, Nb, LREE, U, Pb, and gain of As, Ag, Te, Se, Bi, Cu (Table S3, Figure 9). It is important to note high As in the unaltered biotite gneiss; arsenic is 3–15 times higher than the average for the upper continental crust [28]. Concentration of As, Te, Se, and Bi in the gold-mineralized rocks is by 1–2 orders of magnitude more than in non-mineralized metasomatic rocks (Table S3, Figure 9).

Gold content in the unaltered biotite gneiss is 0.004 g/t (Table S1) and increases in the lenses of metasomatic rocks. Distribution of gold is very uneven: sulfide-mineralized rocks in all metasomatic envelopes contain 0.03–0.5 g/t Au, but where coarse-grained arsenopyrite appears, gold content increases to >1 g/t, and reaches 30 g/t in some hand samples [12]. Channel sampling across one of the lenses of altered rocks gave 1.78 g/t Au for 0.8 m [12].

Unaltered biotite gneiss (metaconglomerate, metagritstone, metasandstone) contains disseminated sulfide mineralization < 1 vol.%; sulfide minerals are pyrrhotite (prevails), arsenopyrite, rarely chalcopyrite, and late pyrite; oxide minerals are ilmenite and rare magnetite. No sulfide mineralization was found in the migmatizated rocks at the contact of biotite gneiss with pegmatite.

Sulfide content increases significantly to 1 to 25 vol.% in the lenses of metasomatic rocks, sulfide-rich rocks (>10 vol.%) were found in different alteration envelopes of the lenses. Texture of mineralization is disseminated, veinlet-disseminated, or nested. Mineral composition is not complicated: the main minerals are pyrrhotite and arsenopyrite, minor sulfides are chalcopyrite, löllingite, troilite, pentlandite, molybdenite. Oxide minerals are ilmenite and magnetite (the latter is more rare, but up to 10% in some samples).

Two generations of arsenopyrite were detected in the metasomatic rocks, the generations differ in grain form and size (Figure 10 and Figure 11), and in chemical composition (

Table 1. Microprobe data for arsenopyrite (Apy) and löllingite (Lo) of the Maljavr gold occurrence, wt%.

Sample No.	TУ-25-1	TУ-25-1	TУ-25-2	TУ-25-2	TУ-31	TУ-31	TУ-41	TУ-41	TУ-45	TУ-45	TY-41-1	TY-41-1	TY-41-2	TY-41-2	TY-29	TY-42	TY-42	TY-41	TY-41	TY-29	TY-42	TY-42
Mineral	Apy-1-C	Apy-1-R	Apy-1-C	Apy-1-R	Apy-1-C	Apy-1-R	Apy-1-C	Apy-1-R	Apy-1-C	Apy-1-R	Apy-2-C	Apy-2-R	Apy-2-C	Apy-2-R	Apy-2	Apy-2	Apy-2	Lo	Lo	Lo	Lo	Lo
S	14.94	18.32	16.08	18.94	15.49	18.36	18.16	20.03	17.93	19.28	18.66	19.30	17.77	19.28	17.37	18.06	18.21	1.40	1.98	1.84	1.83	2.04
Fe	26.46	31.4	27.92	30.56	27.73	31.49	33.15	34.61	31.89	33.65	33.25	34.54	33.43	34.36	33.78	33.62	33.63	27.66	28.23	28.03	25.11	26.31
Co	3.76	2.32	3.12	2.44	4.11	2.23	0.47	0.13	1.28	0.48	0.34	0.12	0.39	0.09	0.07	0.24	0.17	0.08	0.13	0.10	0.49	0.42
Ni	2.89	0.64	2.25	0.81	1.62	0.38	0.24	bdl	0.29	0.02	0.49	bdl	0.38	bdl	0.17	0.26	0.11	0.09	0.10	0.13	2.07	0.68
As	51.83	47.69	50.57	47.45	50.93	46.8	47.86	45.47	48.52	46.38	47.96	46.72	48.59	45.80	48.80	47.87	48.09	70.27	69.35	69.94	70.26	70.04
Total	99.88	100.37	99.93	100.2	99.88	99.26	99.88	100.25	99.91	99.81	100.70	100.67	100.55	99.53	100.19	100.05	100.21	99.51	99.79	100.04	99.76	99.49
Atoms per formula unit
S	0.805	0.946	0.852	0.965	0.831	0.957	0.94	1.014	0.927	0.986	0.952	0.982	0.921	0.992	0.908	0.937	0.939	0.089	0.125	0.116	0.114	0.127
Fe	0.819	0.931	0.850	0.894	0.854	0.942	0.985	1.006	0.946	0.987	0.975	1.009	0.996	1.015	1.014	1.001	0.996	1.009	1.024	1.013	0.904	0.944
Co	0.110	0.065	0.090	0.068	0.120	0.063	0.013	0.004	0.036	0.013	0.009	0.003	0.011	0.002	0.002	0.007	0.005	0.003	0.004	0.003	0.017	0.014
Ni	0.085	0.018	0.065	0.023	0.048	0.011	0.007	0.000	0.008	0.001	0.014	0.000	0.011	0.000	0.005	0.007	0.003	0.003	0.003	0.004	0.071	0.023
As	1.195	1.054	1.148	1.035	1.169	1.043	1.06	0.986	1.073	1.014	1.048	1.018	1.079	1.008	1.092	1.063	1.061	1.911	1.875	1.884	1.886	1.873
at.% As	39.7	35.0	38.2	34.7	38.7	34.6	35.3	32.7	35.9	33.8	34.9	33.8	35.7	33.4	36.1	35.2	35.3	63.4	61.8	62.4	63.0	62.8

Note: bdl = below the detection limit. Atoms per formula unit are calculated for As + S = 2. Apy-1 and Apy-2—arsenopyrite of the first and the second generations, correspondingly. C—data for the central part of the grain, R—data for the rim.

). The early arsenopyrite-1 occurs as small grains up to 0.2 mm in disseminated sulfides. It contains high Co and Ni, and some grains are clearly zoned (Figure 11A) with As-, Co- and Ni-rich core (As 38.2–39.7 at.%, Co 3.12–4.11 wt%, Ni 1.62–2.89 wt%) and lower As, Co and Ni in the rim (As 34.6–35.0 at.%, Co 2.23–2.44, and Ni 0.38–0.81 wt%) (Table 1). Late re-crystallized coarse-grained (0.5–3 mm in size) arsenopyrite-2 is disseminated in the altered rocks, or it forms chains of euhedral grains (Figure 11B and Figure 12–the detailed fragment). Zoning in arsenopyrite-2 is complicated—there are zones with lower (32.7–33.8 at.% As) and higher (34.6–35.9 at.% As) arsenic alternate in the grains (Figure 11C,D). Co and Ni impurities are below 0.5 wt%.

Löllingite occurs as inclusions in arsenopyrite-2 (Figure 13), rarer as separate euhedral grains up to 0.5 mm in size. Löllingite contains impurities of Ni (<2 wt%), Co (<0.5 wt%), and S (<2 wt%) (Table 1).

Gold was found in numerous inclusions in arsenopyrite-2 and löllingite (Figure 13), except the only gold grain in quartz, reported in [12]. The inclusions are monomineral or polymineral, in the polymineral inclusions gold associates with native bismuth and bismuth telluride ehrigite Bi₈Te₃ (the second finding in the world after Good Hope gold mine [29]) (

Table 2. Microprobe data for ehrigite (Ehg), hedleyite (Hdl), maldonite (Mdo), and native bismuth (Bi) of the Maljavr gold occurrence, wt%.

Mineral	Ehg	Ehg	Ehg	Ehg	Ehg	Ehg	Ehg	Hdl	Hdl	Mdo	Mdo	Bi	Bi	Bi
S	bdl	bdl	bdl	bdl	bdl	bdl	bdl	bdl	bdl	bdl	bdl	0.06	bdl	bdl
Fe	0.27	bdl	bdl	0.45	0.27	0.96	1.33	bdl	bdl	bdl	2.89	0.27	0.27	1.37
As	0.08	bdl	bdl	0.00	0.03		0.10	bdl	bdl	bdl	0.12	0.13	0.05	0.14
Te	18.61	18.68	18.34	16.38	18.29	18.36	18.25	23.45	24.93	bdl	bdl	0.00	bdl	bdl
Au	bdl	bdl	bdl	bdl	bdl	bdl	bdl	bdl	bdl	63.78	62.82	bdl	bdl	bdl
Pb	bdl	0.78	bdl	bdl	bdl	0.94	bdl	4.08	bdl	bdl	bdl	bdl	bdl	bdl
Bi	81.49	80.54	80.23	82.49	80.36	79.64	79.05	72.47	75.07	36.22	33.85	99.02	99.88	98.43
Total	100.45	100.00	98.56	99.39	98.95	99.89	98.73	100.00	100.00	100.00	99.68	99.49	100.20	99.94
Atoms per formula unit
S				0.047								0.004
Fe	0.099			0.168	0.099	0.347	0.479				0.291	0.010	0.010	0.049
As	0.022				0.008		0.028				0.009	0.004	0.001	0.004
Te	2.962	3.006	2.998	2.646	2.956	2.896	2.879	3.340	3.523			0.000
Au										1.954	1.791
Pb		0.077				0.091		0.358
Bi	7.917	7.915	8.003	8.140	7.934	7.667	7.615	6.302	6.477	1.046	0.910	0.982	0.989	0.947

Notes: bdl = below the detection limit. Atoms per formula unit are calculated for 11 atoms in ehrigite, 10 atoms in hedleyite, and 3 atoms in maldonite.

, Figure 14 and Figure 15). Rarer mineral phases, found in the inclusions, are maldonite, bismuthinite, joseite-B, hedleyite, hessite, bismoclite, and demicheleite-Br. The inclusions are of irregular angular form, up to 0.2 mm sized (Figure 13 and Figure 14). High concentration of bismuth–ehrigite–gold inclusions is reached at the contact arsenopyrite–löllingite, or these minerals form veinlets along arsenopyrite–löllingite boundary (Figure 13B).

Gold fineness is high or medium, it contains from 53 to 99 wt% Au with two peaks at 72–82 and 88–98 wt%. The main impurities are Ag (4–30 wt%), Fe and As (<0.2 wt%) (

Table 3. Microprobe data for native gold of the Maljavr gold occurrence, wt%.

Fe	1.94	0.48	1.48	1.12	1.71	0.77	0.24	1.11	0.11	0.24	1.35	1.67
As	0.04	0.01	0.04	bdl	0.03	0.07	0.10	0.11	bdl	bdl	bdl	bdl
Ag	7.34	21.22	35.73	45.78	3.78	4.78	29.71	8.78	18.17	18.68	6.23	bdl
Au	88.97	77.08	61.42	53.10	94.19	92.83	69.32	89.45	77.69	78.62	91.16	99.22
Total	98.29	98.80	98.67	100.00	99.71	98.46	99.37	99.45	95.98	97.53	98.75	100.89
Atoms per formula unit
Fe	0.063	0.014	0.040	0.028	0.056	0.026	0.007	0.036	0.003	0.007	0.044	0.056
As	0.001	0.000	0.001		0.001	0.002	0.002	0.003
Ag	0.123	0.330	0.494	0.594	0.064	0.084	0.435	0.146	0.298	0.300	0.106
Au	0.814	0.656	0.465	0.378	0.879	0.889	0.556	0.816	0.698	0.692	0.850	0.944

). One more gold mineral in the deposit is maldonite Au₂Bi (Table 2), which occurs in the inclusions in arsenopyrite and löllingite together with gold, native bismuth, and ehrigite (Figure 15).

4.2. Zircons in the Rocks of the Maljavr Occurrence and Age of Zircons

Zircon from sample SHM-12 of biotite gneiss–metaconglomerate forms euhedral transparent pale yellow fractured crystals, with plain facets and smoothed edges. Prism {110} and di-pyramid {111} facets are predominant. The crystal size is 50–150 μm, the elongation coefficient varies from 2.0 to 3.0 (Figure 16). Some grains contain inclusions of monazite and biotite (Figure 16). Coarse zoning was noted in the outer part of some zircon crystals.

Ten samples of the least altered zircon crystals, selected from different size fractions, were assayed for U and Pb isotopes (

Table 4. Results of U-Pb geochronological studies of zircons from biotite gneiss–metaconglomerate and tourmaline granite–pegmatite.

Sample No./ Fraction No.	Weight, mg/ Size, μm	Pb, ppm	U, ppm	Isotope Ratios					Rho	Age, Ma
				206Pb/204Pb *	207Pb/206Pb *	208Pb/206Pb *	206Pb/238U	207Pb/235U		206Pb/238U	207Pb/235U	207Pb/206Pb
Sample SHM-12—biotite gneiss
SHM-12/1	0.2/<100	165	358	533	0.1999 ± 0.0001	0.1281 ± 0.0002	0.3958 ± 0.0015	9.669 ± 0.038	0.91	2150 ± 9	2404 ± 10	2627 ± 3
SHM-12/2	0.5/>100	130	341	702	0.1907 ± 0.0001	0.1246 ± 0.0002	0.3322 ± 0.0049	7.9400 ± 0.012	0.99	1849 ± 27	2224 ± 33	2590 ± 2
SHM-12/3	0.5/>100	141	337	822	0.1891 ± 0.0001	0.1291 ± 0.0002	0.3639 ± 0.0011	8.7448 ± 0.026	0.96	2000 ± 6	2312 ± 7	2599 ± 2
SHM-12/4	0.2/>50	164	363	485	0.1972 ± 0.0001	0.3042 ± 0.0001	0.3418 ± 0.0014	8.1055 ± 0.040	0.91	1896 ± 8	2243 ± 11	2577 ± 4
SHM-12/5	0.5/<100	361	803	1178	0.1845 ± 0.0001	0.2709 ± 0.0001	0.3551 ± 0.0007	8.5313 ± 0.017	0.94	1959 ± 4	2289 ± 5	2599 ± 1
SHM-12/6	0.3/<50	158	371	539	0.1909 ± 0.0001	0.2773 ± 0.0001	0.3226 ± 0.0016	7.6289 ± 0.038	0.96	1834 ± 9	2188 ± 11	2539 ± 3
SHM-12/7	0.4/<50	186	432	818	0.1829 ± 0.0001	0.2064 ± 0.0001	0.3457 ± 0.0010	8.2387 ± 0.025	0.88	1962 ± 6	2258 ± 7	2538 ± 2
SHM-12/8	0.4/<100	175	406	1242	0.1829 ± 0.0001	0.2062 ± 0.0001	0.3565 ± 0.0011	8.5121 ± 0.025	0.92	1966 ± 6	2287 ± 7	2588 ± 2
SHM-12/9	0.3/<50	255	615	574	0.1942 ± 0.0001	0.1505 ± 0.0001	0.3502 ± 0.0017	8.3507 ± 0.0011	0.96	1935 ± 9	2270 ± 11	2586 ± 3
SHM-12/10	0.5/<50	142	305	550	0.1968 ± 0.0001	0.2235 ± 0.0001	0.3723 ± 0.0016	8.9664 ± 0.0011	0.94	2040 ± 9	2335 ± 10	2603 ± 3
Sample SHM-11—tourmaline granite pegmatite
SHM-11/1	0.3/<200	1218	2587	1426	0.1724 ± 0.0001	0.03001 ± 0.0001	0.4633 ± 0.0005	10.509 ± 0.011	0.90	2463 ± 2	2481 ± 2	2495 ± 1
SHM-11/2	0.9/<150	729	1534	1233	0.1741 ± 0.0001	0.05311 ± 0.0002	0.4449 ± 0.0007	10.068 ± 0.015	0.95	2373 ± 4	2441 ± 4	2499 ± 1
SHM-11/3	0.6/>100	793	1619	1723	0.1719 ± 0.0001	0.02737 ± 0.0001	0.4715 ± 0.0011	10.714 ± 0.026	0.99	2490 ± 6	2499 ± 6	2506 ± 1
SHM-11/4	1.7/>250	1109	2370	2347	0.1695 ± 0.0001	0.02410 ± 0.0001	0.4529 ± 0.0009	10.265 ± 0.021	0.96	2408 ± 5	2459 ± 5	2501 ± 1
SHM-11/5	1.2/>200	1140	2643	2091	0.1686 ± 0.0001	0.02238 ± 0.0001	0.4182 ± 0.0009	9.3869 ± 0.017	0.94	2252 ± 5	2377 ± 5	2485 ± 1

* Isotope ratios are corrected for the blank and common lead; Rho—the correlation coefficient of the ²⁰⁷Pb/²³⁵U-²⁰⁶Pb/²³⁸U ratios.

). Analytical points of isotopic composition of ten zircon fractions lie in discordia with the age 2664 ± 18 Ma at the upper intersection with the concordia, MSWD = 9.5 (Figure 17). The lower intersection with the age 378 ± 77 Ma reflects the Paleozoic tectonic activity and formation of the Khibiny and Lovozero alkaline intrusions [30].

Zircon from sample SHM-11 of tourmaline granite pegmatite forms euhedral long-prysmatic transparent fractured crystals of brown color. Prism {110} and di-pyramid {111} facets are predominant, the crystal faces are rough, and the edges are sharp. The inner structure is nonhomogeneous, with dark and light areas of irregular form, some grains contain small inclusions of uraninite (Figure 16). The crystal size is 100–300 μm, the elongation coefficient varies from 3.0 to 4.0. Five samples of the least altered and free of inclusions zircon crystals were selected from different size fractions and assayed for U and Pb isotopes (Table 4). Analytical points of isotopic composition of five zircon fractions lie in discordia with the age 2508 ± 7 Ma at the upper intersection with the concordia, MSWD = 1.4 (Figure 18). The lower intersection with the concordia occurs at 358 ± 57 Ma.

5. Discussion

The sequence of volcanic–sedimentary rocks, including the conglomerate hosting the Maljavr gold occurrence, formed during the period from 2.95 to 2.83 Ga. The lower boundary of the time interval is defined by the age of detrital zircons from the basal conglomerates 2952–2939 Ma: the zircons are primarily magmatic, the probable parental rock was TTG gneiss in the Kola province [14]. The upper time boundary of the interval is specified by the age gneiss–metadacite 2838 ± 23 Ma (zircon, U-Pb, SHRIMP II) in the upper sequence of the Uragubsky belt [16].

The volcanic sedimentary sequence was metamorphosed amphibolite at a temperature of 600–620 °C, under a pressure of 4 kbar [17] during the period from 2.87 to 2.66 Ga. It was the time of several accretion events, which lead to formation of the continent Kenorland [13]. An ancient island arc (the Kolmozero-Voron’ya and Uragubsky greenstone belts, at present) and the Kola province were successively accreted to the Murmansk block 2.78–2.76 Ga ago [13]. The peak of the Neoarchean metamorphic event in the Uragubsky greenstone belts was reached at ~2.77 Ga: the U-Pb age of metamorphism is 2786 ± 31, 2774 ± 12 Ma (SHRIMP-II, zircons of metamorphic genesis) [16,17], 2763 ± 8 Ma [15].

Age of zircon from the biotite gneiss in the zone of alteration in the Maljavr occurrence is 2664 ± 18 Ma (Figure 17, Table 4), i.e., it is younger than the age of volcanic sedimentary rocks and younger than regional metamorphism in the Uragubsky belt. Textural and structural characteristics of the metasomatic rocks in the Maljavr gold occurrence, such as massive fabrics with different orientation of biotite flakes and amphibole crystals (Figure 6B,C), indicate alteration processes after the peak of regional metamorphism under conditions of undirected pressure. Biotite–garnet [31,32] and amphibole–garnet [31] geothermometers, applied to the composition of coexisting biotite–almandine and hornblende–almandine, correspondingly (Table S2), indicate temperature of metasomatic rocks formation 600–630 °C, it is close to the temperature of regional metamorphism in the Uragubsky belt [17].

This zircon age of biotite gneiss in the zone of alteration corresponds well to the age of granulite metamorphism in the Kola-Norwegian province, adjacent to the Uragubsky belt from SW (Figure 1, inset). The rocks of the Kola-Norwegian province were metamorphosed granulite during the period 2.71–2.50 Ga [33,34]. This metamorphic event was dated with the outer zones of zircon crystals, which defined the interval 2707–2656 Ma [33], and with zircon from biotite–garnet–sillimanite gneiss 2648 ± 18 Ma [34].

The tectonic events 2.71–2.50 Ga in the Kola-Norwegian province are reflected in the neighboring Uragubsky and Kolmozero-Voron’ya greenstone belts. The volcanic–sedimentary sequences of the belts were intruded with plagioclase/microcline granites and granite aplite dykes at 2696 ± 9 Ma [18]. These tectonic events probably caused increasing intensity of hydrothermal processes in the greenstone belts: in the Kolmozero-Voron’ya belt metasomatic rocks were dated 2629 ± 64 Ma [35], and we consider the age of zircon from the Maljavr occurrence 2664 ± 18 Ma as the time of alteration of the biotite gneiss, when the lenses of metasomatic rocks with sulfide (pyrrhotite-arsenopyrite) mineralization formed. Morphological and isotope characteristics of zircons from the biotite gneiss agree with the zircons of metamorphic genesis [36,37]: zircon grains are isometric or ellipsoid, crystal zoning is weakly displayed or absent (Figure 16A). The discordancy of the analytical points (Figure 17) appears probably due to zircon fracturing, which promoted partial loss of radiogenic lead.

The more recent recognized stage of formation of the Maljavr occurrence was the injection of tourmaline pegmatite with age of 2508 ± 7 Ma. Morphological and isotope characteristics of zircons from the tourmaline granite pegmatite agree with the magmatic zircons (Figure 16B, Table 4). Thin euhedral zoning can be seen in the outer crystal zones, and zoning is blurred in the central parts of the zircon grains. Inhomogeneity of the inner structure (Figure 16B) may be explained by later fluid alteration, which did not significantly modify the crystal structure.

The pegmatite affected the biotite gneiss and metasomatic rocks and caused tourmalinization and recrystallization of sulfides (arsenopyrite and löllingite): a chain of arsenopyrite grains, recrystallized due to pegmatite injection, can be seen in a thin veinlet of biotite gneiss between two lenses of pegmatite (Figure 12). Recrystallization of sulfides resulted in gold–bismuth mineralization in a form of numerous inclusions in arsenopyrite-2 and löllingite grains.

The temperature of formation of arsenopyrite-2 (and gold–bismuth mineralization) was determined with the arsenopyrite geothermometer [38]. The arsenopyrite geothermometer is believed to be the most reliable for sulfide systems, because arsenopyrite composition is resistant to change of PT-parameters in the surrounding medium [39]. The thermometer can be applied only to arsenopyrites with Co and Ni impurities less than 0.5 wt%: arsenopyrite-1 does not fit this requirement, but arsenopyrite-2 is acceptable (Table 1). Composition of arsenopyrite-2 and its association with pyrrhotite indicate two temperature intervals: 500–600 °C for arsenopyrite with As 34.6–35.9 at.% and 400–510 °C for arsenopyrite with As 32.7–33.8 at.%. The first interval can be related to the temperature of metasomatic rock’s formation. The second temperature interval 400–510 °C corresponds well to the conditions of pegmatite crystallization: T = 415–460 °C was estimated for pegmatite in the area of the Maljavr occurrence with different mineral thermometers and study of fluid inclusions in beryl [40].

Two stages of gold mineralization evolution in the Maljavr gold occurrence are supposed: 1—the stage of formation of lenses of metasomatic rocks with sulfide pyrrhotite-arsenopyrite mineralization (0.03–0.5 g/t Au) at ~2.66–2.67 Ga; 2—gold concentration in re-crystallized arsenopyrite and löllingite at the time of pegmatite injection ~2.51 Ga. Therefore, the pegmatite age of 2508 ± 7 Ma reflects the time of formation of gold–bismuth mineralization in its present form.

As it was shown in the example of the Suurikuusikko and Iso-Kuotko deposits in the Central Lapland belt in Northern Finland [41], gold-mineralized rocks form with a series of impulses of hydrothermal activity. In the Central Lapland belt, these impulses correlate with magmatic events—the formation of different generations of minor granite intrusions and dykes. Multi-stage formation was also supposed for gold mineralization in the Rompas-Rajapalot deposits in the Perapohja belt (Northern Finland) [42].

The geological position of the Maljavr occurrence, the geochemical association Au-As-Se-Te-Bi in the mineralized rocks, and the linkage of gold mineralization with pegmatite—all these characteristics agree with the reduced intrusion related type of gold deposits [43,44,45], but paucity of relevant data (fluid inclusions and isotope study) does not allow constraint of the genetic type with certainty. Biotite gneiss–metaconglomerate, hosting the mineralized metasomatic rocks, could be the primary source of arsenic and, probably, gold for the mineralization, which formed during the later metasomatic event.

In Northern Fennoscandia, two stages of gold deposit formation are identified—the Neoarchean (2.7–2.6 Ga) and Paleoproterozoic (1.92–1.74 Ga) [46]. Gold occurrences of Neoarchean age are located in the north-eastern part of the region, within the Kola province and north of it, in the Kolmozero-Voron’ya belt [46]. The deposits and occurrences, located south of the Kola province, are of Paleoproterozoic age, even those hosted by the Archean greenstone belts [46].

The Neoarchean Maljavr gold occurrence (2.67–2.51 Ga) in the Uragubsky belt (north of the Kola province) fully fits this model. Close age 2.64–2.68 Ga (U-Pb, SHRIMP, outer zones of zircon) was obtained for gold-mineralized in skarnoids in the Olenegorsk BIF deposit [47] in the Kola province. Gold deposits in the Kolmozero-Voron’ya belt are older: ~2.82 Ga (Oleninskoe) and ~2.78 Ga (Nyalm-1) [46].

In the northern part of the Fennoscandian shield, two conglomerate-hosted gold occurrences (Kaarestunturi and Outapää) were found in the 1970′s in the Central Lapland greenstone belt in Finland [10,11]. The occurrences are hosted in quartzite and oligomictic conglomerate of the Kumpu Group (<1850 Ma), deposited discordantly on graywackes and carbonate metasedimentary rocks; the rocks are metamorphosed greenschist. In the Kaarestunturi occurrence, Au-bearing lenses, containing up to 5.0 ppm Au, form a set of up to 30 m long and up to 1 m (typically 30 cm) thick lenses within a 30–60 m thick conglomerate sequence. Free gold (about 5% Ag), with grain size of 0.03–0.4 mm, occur in the matrix of the conglomerate. Other ore minerals are chalcopyrite, minor galena, pyrite, magnetite, ilmenite, and hematite at Kaarestunturi, and rare magnetite and hematite at Outapää. The gold occurrences are considered as Paleoproterozoic gold paleoplacers in fluvio-deltaic fans [10,11]. Probable sources for gold are the “mesothermal” lode-gold deposits in the greenstones stratigraphically below the metasedimentary host rocks [10,11].

The Maljavr occurrence is the first gold occurrence in the Fennoscandian Shield, hosted in Archean conglomerates. It differs from the occurrences in Paleoproterozoic conglomerates in high grade metamorphism, intensive alteration of the rocks, modification by pegmatite injection, and in more diverse composition of mineralization.

6. Conclusions

The following story of formation of the Maljavr gold occurrence is reconstructed:

• Deposition of sedimentary rocks, including the conglomerates hosting the Maljavr gold occurrence, in the interval 2.95–2.83 Ga.
• Regional upper amphibolite metamorphism of the volcanic–sedimentary sequence in the interval 2.77–2.79 Ga.
• Alteration of the biotite gneiss–metaconglomerate and formation of lenses of metasomatic rocks with sulfide pyrrhotite–arsenopyrite mineralization; this event took place in a local zone of NNE strike at ~2.66–2.67 Ga. Geochemical association of the metasomatic rocks includes Au, As, Ag, Se, Te, Bi, and Cu.
• Injection of tourmaline pegmatite veins at ~2.51 Ga caused recrystallization of sulfides and formation of gold–bismuth mineralization in its present form. Gold and bismuth minerals (native gold, maldonite, native bismuth, ehrigite, and other tellurides) occur in inclusions in recrystallized arsenopyrite (arsenopyrite-2) and löllingite.

Biotite gneiss–metaconglomerate, hosting the mineralized metasomatic rocks, was the probable primary source of arsenic and gold for mineralization, which formed later. The geological position of the Maljavr occurrence, the linkage of gold mineralization with pegmatite, the geochemical association Au-As-Se-Te-Bi in the mineralized rocks all correspond to the type of gold deposits, which are associated with intrusions. The Maljavr occurrence is the first gold occurrence in the Fennoscandian Shield, to be reported from Archean conglomerates.