Paleoproterozoic Layered Intrusions of the Monchegorsk Ore District: Geochemistry and U–Pb, Sm–Nd, Re–Os Isotope Analysis

Abstract

The paper concerns the geochemical analysis of rocks from the ore-bearing layered intrusions that belong to two age groups of the Monchepluton and the Imandra–Umbarechka Complex (2.50 and 2.44 Ga) and the largest gabbro-anorthosite of the Main Ridge Complex (2.51–2.45 Ga). The intrusion of these complexes happened at different depths when the endogenous and geodynamic settings changed at the beginning of the Paleoproterozoic Era. Five megacycles are distinguished in a generalized cross-section of the two-chamber Monchepluton. The megacycles differ in rock composition, rock geochemical features, and mineralization types, i.e., the chromite, sulfide Cu–Ni–PGE and low-sulfide PGE types. The abrupt changes in isotope indicators (εNd, ⁸⁷Sr/⁸⁶Sr) mark their boundaries. At a depth of 2037–2383 m, the M-1 borehole intersects a standalone intrusive body that is essentially a magma feeder channel. The intrusive body’s geochemical characteristics and U–Pb isotope age correlate to the Monchepluton rocks. The gabbro-anorthosite massifs united in the Main Ridge Complex were intruded in the following order: the Monchetundra, Chunatundra, Volchetundra, and Losevo–Medvezhye tundras. The largest Monchetundra massif was formed as a result of multiple intrusions of mafic magmatic melt from the deep reservoirs. The melts intruded in two stages, i.e., 2.51–2.49 Ga and 2.48–2.47 Ga, and their composition changed gradually. The gabbro-pegmatites and coeval harrisite dykes are more recent ones (2.46–2.45 Ga). The summarized results of the U–Pb, Sm–Nd, and Re–Os systems research allowed us to establish genetic relations between the studied geological objects. We proposed a model where there was an uplift of a mantle plume to the lower crust area at the age of 2.5 Ga, the deep mantle reservoirs were formed, and a large-scale interaction happened between the parental magma and granulite–eclogite complex rocks. Local contamination and assimilation processes took place during the uplifting of magmas in areas where the magmatic feeding system contacted the host amphibolite–gneiss Archean complexes.

1. Introduction

Being the most ancient part of the Fennoscandian Shield, the Kola–Lapland–Karelian Province contains widespread layered intrusions of the Paleoproterozoic age that comprise a series of dunite–peridotite–orthopyroxenite–gabbronorite–anorthosite rocks [1,2,3,4]. They have spatial and genetic relations with occurrences and deposits of chromite, sulfide Cu–Ni–PGE sulfide, low-sulfide PGE, and Ti–V ores. The intrusions serve as an important indicator of the endogenous and geodynamic setting change that happened between the Neoarchean and Paleoproterozoic, so their importance can hardly be overestimated.

The intrusions belong to two heterochronous groups judging by their geological interrelation with the Paleoproterozoic volcanogenic–sedimentary complexes and by their U–Pb isotopic age. The more ancient Kola group includes the following intrusions, i.e., Mt. Generalskaya, Ulitoozerskaya, Monchepluton, Fedorovo–Pansky Complex, and Pados–Tundra. Their formation at the age of 2.5 Ga and glacial erosion preceded the formation of the large Paleoproterozoic Pechenga–Varzuga Paleorift Belt, which extends at a length of 600 km from Northern Norway to the White Sea. Later, the Lapland–Karelian group (2.45 Ga) formed in the already existing riftogenic conditions, i.e., the Burakovsky Pluton, Akanvaara, Koitelainen, Penikat, Kemi, Kivakka, Tsipringa, Lukkulaisvaara, and the complexes of Portimo, Koillismaa, and Imandra–Umbarechka. The dunite–orthopyroxenite massifs of the Notozero Complex (the Pados–Tundra, Chapes–Varaka, and others) are spatially separated and located to the south of the Lapland Granulite Belt.

This paper concerns the intrusions of the Monchegorsk ore district. The Monchegorsk ore district, located in the central part of the Kola Region, is a unique test site to study layered intrusions (Figure 1). The point is that there are layered mafic–ultramafic intrusions and comagmatic dykes of two age groups (2.5 and 2.45 Ga); coeval gabbro-anorthosite massifs; and various well-exposed and accessible deposits and ore occurrences of sulfideCu–Ni–PGE sulfide, low-sulfide PGE, chromite, and titano-magnetite ores on its territory. The work provides a detailed history of their study and their geological location, structure, and petrography [1,5,6].

A large amount of isotope data on the U–Pb (ID-TIMS, SIMS), Sm–Nd, and Re–Os analyses obtained since the 1990s to the present [6,7,8,9,10,11,12,13,14,15,16,17,18] have allowed us to determine the age, period, and model of the formation of these layered intrusions. However, the question of their genetic relations remains open. One of the methods to resolve the question is a geochemical analysis of rocks composing layered intrusions. In our studies, the main objective is to find similarities and differences between them and, as a result, determine their genetic relations based on the geochemical analysis of the rocks of Paleoproterozoic intrusions and the comagmatic dykes of the Monchegorsk ore district in addition to the results of isotope analyses.

2. Geologic Setting and Rock Characteristics

For our analysis, we chose layered intrusions of both age groups, i.e., the gabbro-anorthosite of the Main Ridge Complex and coeval dykes. The first and more ancient group of layered intrusions is represented by the Monchepluton, and the second group is represented by the massifs of the Imandra–Umbarechka Complex.

According to the petrographical studies, the most common rocks of these intrusions are dunite (90%–96% Ol, 2%–8% OPx), plagiodunite (the presence of interstitial Pl, 6%–8%), harzburgite (55%–65% Ol, 25%–30% OPx, 3%–8% Cpx), plagiogarzburgite (65%–85% Ol, 10%–25% OPx, 3%–5% Cpx, 6%–10% Pl), lherzolite (55%–60% Ol, 22%–30% OPx, 15%–20% Cpx), olivine orthopyroxenite (10%–15% Ol, 80%–85% OPx), orthopyroxenite (90%–95% OPx, <5% Ol, <1% CPx), olivine norite (8%–10% Ol, 40%–60% OPx, 30%–40% Pl, 1%–3% Cpx), melanonorite (40%–70% OPx, 25%–40% Pl, 5%–10% Cpx), norite (30%–45% OPx, 40%–50% Pl, 5%–15% Cpx), pyjonite gabbronorite (45%–70% Pl, 15%–25% Cpx + Pi, 10%–20% Opx), gabbronorite (20%–60% Pl, 20%–70% Opx, 15%–25%Cpx), and anorthosite (85%–90% Pl, 10%–15% Opx + Cpx). The most variable mineral composition was revealed for gabbronorites. Harrisite dykes are represented by melanocratic pyroxene troctolites (60%–70% Ol-fayalite, 15%–25% Pl, 3%–10% Opx, 1%–8% Cpx) with harrisite structures. The veins comprise mostly coarse-grained gabbronorites (hereinafter referred to as gabbro-pegmatites). Cr-spinel is typical accessory mineral (1%–3%) in olivine-bearing rocks, while accessory titanomagnetite, ilmenite, and apatite (1%–4%) occur in plagioclase-bearing rocks. The secondary minerals comprise serpentine, chlorite, and amphibole. Within the Monchetundra fault zone, the rocks of the intrusions and Archean basement are transformed into amphibole–garnet blastocataclasites and blastomylonites. The main mineral assemblage of the Archean plagiogneisses is 60%–78% Pl, 15%–35% Qu, and 5%–10% Amph + Bt. The hypersthenic diorites are composed of 40%–60% Pl, 20%–30% Opx, and 2%–25% Qu.

The Monchepluton is arc-shaped and consists of two magmatic chambers on the section plan (Figure 2). The Northern Chamber is 7 km long, oriented to the north-east, and presented by the Nittis, Kumuzhya, and Travyanaya mountains (hereinafter referred to as the NKT). The Southern Chamber is 9 km long and stretches out to the west from the top of Mt. Sopcha to the Nyud and Poaz mountains (hereinafter referred to as the SNP) and then to the south-east to the Mt. Vurechuaivench foothills.

Each magmatic chamber is symmetrically through-shaped, while the wings fall at angles of 30–40° (NKT) and flatten out to the axial parts (SNP) from 40–45° to 20–25° [19]. Both chambers are inclined to the south-west. The internal constitution of the Monchepluton was deformed by the tectonic faults during the post-intrusion period, when the Imandra–Varzuga Zone was established. The Dunite Block rocks and the Sopcheozero chromite deposit were moved downward in relation to the northern and southern chambers. In the Vurechuaivench foothill, the intrusive rocks are overlaid by polymictic conglomerate and metavolcanics of the Seidorechka formation, which compose the bottom part of the Imandra–Varzuga Zone section.

From bottom to top, the Monchepluton is composed of quartz norites and gabbronorites of the Marginal Zone and rocks of five sequentially formed megacycles. Each of them starts with rocks enriched in MgO (Figure 3). Within the northern chamber, there are the following megacycles: I—harzburgite–orthopyroxenite; II—dunite–harzburgite–orthopyroxenite. Within the southern chamber, there are the following megacycles: III—dunite–orthopyroxenite–norite; IV—harzburgite–norite–gabbronorite; and V—gabbronorite–anorthosite. The megacycles were formed due to the multiple pulse injections into the magma chambers. The new magma pulse was accompanied by the partial dissolution of the olivine phase of the previous megacycle and followed by the crystallization of chromite, olivine, and orthopyroxene in nonequilibrium conditions [5,6].

The I megacycle has a typical rhythmic layering caused by the alternation of harzburgite and orthopyroxenite layers; the III megacycle has a more complex microrhythmic layering, where the alternation passes from dunite–harzburgite into harzburgite–olivine orthopyroxenite.

The high-alumina hornfelses with cordierite and their melting products occur on the boundary between the III and IV megacycles within the “Terrasa” deposit area of Mt. Nyud. They represent a fragment of the primary roof of a magma chamber that existed before the IV megacycle magma pulse.

The dunite and plagiodunite of the II megacycle underwent severe plastic deformations under relatively high temperatures (over 400 °C) and total pressure (ca. 5 kbar) [5]. Rocks of the Vurechuaivench foothill are metamorphosed under an epidote–amphibolite facies, while most of the rocks composing the Monchepluton are not.

The Monchepluton includes various deposits and occurrences of Cu–Ni, Cr, and PGE ores. Sulfide-disseminated ores are deposited within the Marginal Zone, i.e., the I megacycle contains veined sulfide ores along with the ore–silicate pegmatites; the II megacycle contains chromite layers of the Sopcheozero deposit; the III megacycle contains Ore Horizon 330” of disseminated Cu–Ni ores; the IV megacycle contains sulfide nest-disseminated ores of the “Terrasa” and “Nyud-II” deposits; and the V megacycle contains low-sulfide PGE mineralization of the Vurechuaivench reef-type deposit. A thin layer with unprofitable titano-magnetite mineralization has occurred within the metagabbro of the “10th anomaly”.

The age of the Monchepluton rocks and dykes was determined using the U–Pb (ID-TIMS) analysis of zircon and baddeleyite [6,7,10], accompanied by SIMS SHRIMP-II analysis [11,20,21] (Figure 3). Some samples have been analyzed by two methods. Taking into account the study of the obtained results, the multiple pulse emplacements into the magma chambers, their crystallization, and the conditions of the U–Pb isotope system’s closure in baddeleyite and zircon, the Monchepluton formation took place in a relatively short period of time, 2507–2498 (average 2502 ± 5) Ma. The age of dykes intersecting the II megacycle dunites and chromite ores is similar, being 2506–2496 Ma.

The Imandra–Umbarechka Complex occurs as closely located massifs divided by tectonic faults. They formed later in riftogenic conditions alongside the intensive basalt and medium acid volcanism [18]. They occur on the eastern (the Prikhibinsky, Umbarechensky, and Bolshaya Varaka) and western (the Severny, Mayavr–Devichya, and Yagelny) coasts of Lake Imandra (Figure 1). The western massifs intruded the Archean basement rocks, while the eastern massifs intruded under the felsic volcanics of the Seidorechka formation during the period of 2442–2437 Ma [6,22]. The massifs’ total area is 225 km², and their section thickness varies from 0.5 to 1.5 km. The chromite ores occur as sheet-like bodies within the Bolshaya Varaka and Mayavr–Devichya massifs. The Imandra–Umbarechka Complex is the closest to the Akanvaara and Koitelainen intrusions of Northern Finland in regard to the internal structure, rock composition, and ore mineralization [23].

There is a Lower Border Zone of microgabbronorites (5–15 m); a Lower Layered Zone of norites with chromite layers (up to 120 m); a Main Layered Zone of gabbronorites (up to 2000 m) with signs of layering and thin chromite layers; an Upper Layered Zone of gabbro and gabbronorites (up to 300 m); an Upper Border Zone of gabbro and gabbro-anorthosites (300–500 m) with a disseminated titano-magnetite mineralization; and a Granophyre Zone in the roof (up to 600 m) on the contact with dacitic volcanics.

Intrusive rocks of the Imandra–Umbarechka Complex have undergone uneven metamorphism under conditions varying from greenschist facies in the north-east to epidote–amphibolite facies in the south-west. Maximum metamorphism temperatures (610–670 °C) have been obtained for the Yagel’ny massif rocks.

Ore-free, lens-shaped massifs of the Luvtuaivench, Ospe, and Ekostrovsky are located near the Ekostrovsky strait of Lake Imandra. They comprise mainly orthoamphibolites derived from gabbro-anorthosites and ferrogabbro, and actinolite-chlorite–serpentine schists also developed after ultramafites. They are comparable to the Imandra–Umbarechka Complex massifs [24] regarding their composition, geochemical features, and U–Pb isotope age (2456–2445 Ma); thus, they are included into the analysis.

The ore-free Ostrovsky massif of the same age is located 10 km westward of the Imandra–Umbarechka Complex. It has typical rocks, that contain both clinopyroxene and orthopyroxene, i.e., lherzolites and websterites. There is a Lower Border Zone of plagiopyroxenites and gabbronorites (as relics); a Lower Zone of lherzolites, harzburgites, and dunites (over 750 m); a Middle Zone of websterites and lherzolites (100–150 m); and a severely eroded Upper Zone of gabbronorites and norites (50–150 m) in its section. The age of the gabbronorite pegmatite is 2445 ± 11 Ma [6].

The Main Ridge Complex of the gabbro-anorthosites stretches out by 80 km to the near north-south at a width of 1–2 to 15–20 km and separates the Central Kola Block from the Belomorian Block (Belt). It is composed of several massifs, i.e., the Monchetundra and Chuna–Volche–Losevye–Medvezhye tundras (Figure 4).

The massifs were formed within the same period: the Monchetundra—2505–2470 Ma; Chunatundra—2467 ± 7 Ma; and Volchetundra—2473–2463 Ma [6,12,14,25]. The Monchetundra massif was separated from the Chunatundra massif and the Monchepluton by a submeridional steep fault system. The tectonic movements occurred in the Paleoproterozoic (2.0–1.9 Ga), synchronously with the Svecofennian Orogeny in the central part of the region [26].

The major part of the Monchetundra massif section, the tectonic area, and underlying rocks of the Archean basement were revealed by the 2472 m deep structure borehole M-1 (Figure 5), as well as by smaller 753, 734, 765, and 742 boreholes. The Lower and Middle Zone rocks were partially also revealed in the Loipishnyun area by MT-3, MT-25, and MT-69 boreholes and the MT-79 borehole of a 251–355 m depth [14,27]. The following zones are distinguished within the central part of the Monchetundra massif section: the Lower Zone (M-1 borehole, interval of 750–1020 m), the Middle Zone (M-1 borehole, interval of 0–750 m), and the Upper Zone (from the M-1 borehole head to the tops of Mt. Monchetundra and Mt. Hipiknjunchorr, over 500 m thick) (Figure 5).

The Lower Zone is the most heterogeneous regarding the composition. Its bottom part (908–1020 m) comprises mostly interlayered gabbronorites and norites, and its top part comprises orthopyroxenites, norites, and gabbronorites, along with plagiodunites and plagioharzburgites. The latter were revealed at depths of 782 m and 810 m; they transfer to melanocratic norites and orthopyroxenites on the contacts.

The Middle Zone comprises trachytoid, mesocratic, and, less frequently, leucocratic gabbronorites. The Upper Zone consists of meso- and leucocratic gabbronorites and anorthosites. The younger narrow bodies of the coarse-grained gabbronorites and gabbroic pegmatites occur within all zones. At a depth of 2037–2383 m, the M-1 borehole intersected a standalone intrusive body that is essentially a magma feeder channel (Figure 5). It comprises mostly plagioharzburgite, and the chill margins remained on the contacts.

The Border Norite Zone (20–400 m) and Main Zone are located in a section of the Volchetundra massif [12]. The total volume of the Main Zone is composed of coarse-grained leucogabbro and gabbronorites, and the central part is composed of coarse-grained anorthosites with bodies of harrisites (melanocratic troctolites) and pegmatoid gabbronorites. Analogous rocks are exposed in the Chunatundra massif. The gabbroids were metamorphosed under the amphibolites facies. The rocks are severely altered near the faults. In addition to the intrusions, dykes, and harrisite sheet bodies breaking through the rocks of the Monchetundra, a gabbronorite dyke in the Olenegorsk district and dykes of gabbro, norite, ferrogabbro, and microgranite intersecting the rocks and ores of the Sopcheozero deposit were studied. A composition of rocks from the magma feeder body of the Monchepluton revealed by M-1 borehole was analyzed, too.

3. Analytical Methods

A modern database was established in order to perform the studies. This database contains the analyses of major elements (wt.% oxides) obtained from the ICP-AES method and analyses of trace and rare-earth elements (REE) obtained from the ICP-MS method. In addition to that, we used published data [12,24,25,27,28,29,30,31].

Major and Trace Elements. Whole-rock major and trace element concentrations were determined at the Centre de Récherche Pétrographiques et Géochimique (SRPG-CNRS, Nancy, France). Analyses of major elements were performed by inductively coupled plasma-atomic emission spectrometers (ICP-AES) (Jobin Yvon, JY 70) after alkaline melting with lithium borate and nitric acid dissolution. The analytical error is 1%–10% for the concentrations exceeding 1 wt.% and 2%–20% for element concentrations below 1 wt.%. Concentrations of trace elements were analyzed by inductively coupled plasma mass-spectrometry (ICP-MS) (Perkin Elmer Elan 5000) and showed variable precision and accuracy detection which strongly depend on the elemental concentration in the samples and related limits of determination. At a concentration level above 10 ppm (i.e., for the most common concentrations between 1 and 50 ppm in the studied samples), the trace elements were analyzed with an analytical error better than 15%. The REE concentrations (within the 0.1–15 ppm range in the studied samples) were analyzed with an analytical error better than 15% (except for the Dy, which had an analytical error of 20%). Analyses for the international standards used as well as detection limits and analytical uncertainties in the CRPG laboratory at Nancy can be viewed at http://dx.doi.org/10.17632/432y98sbvx.1.

The data processing scheme included the construction of diagrams for the main elements, such as SiO₂, TiO₂, Al₂O₃, Fe₂O_3tot, CaO, Na₂O + K₂O, P₂O₅, with respect to MgO; for ore elements, Ni–Cu, Ni–Cr, Ni/Cu–MgO, Ni/Co–Fe# rock, Ni/Cu–S; and REE spectra normalized to chondrite [32] and spider-giagrams of rare element contents normalized to the depleted mantle (DM) [33]. The diagrams were made with respect to the specific features of the formation and structure of the intrusions: five megacycles were determined for the Monchepluton due to the phase magma intrusion; three main areas were detected in the section structure of the Monchetundra and the Imandra–Uvbarechka Complex; rocks of the Chunatunda and Volchetundra occur separately.

Representative electron microprobe data of the main mineral phases (plagioclase, olivine, pyroxenes, and accessory minerals) are openly available online at http://dx.doi.org/10.17632/432y98sbvx.1. Mineral analyses were performed on a Camebax SX 50 Microprobe at the SRPG-CNRS (Nancy, France). The operating conditions were an accelerating voltage of 15 kV, a beam current of 10 nA, and a count time per element of 10 s for major elements and 20 s for minor elements. For accessory minerals, a bivoltage program was used with 15 kV for major elements and 20 kV for trace elements. Standards used were a combination of natural and synthetic minerals. Mineral analyses were also performed on a JXA—8200 LEOL (IGEM RAS, Moscow, Russia)—and MS-46 CAMECA (GI KSC RAS, Apatity, Russia).

U–Pb and U–Th–Pb Isotope Analysis. Data regarding the age and sequence of the intrusions and dykes in the Monchegorsk ore district are based on the results of the two analytical procedures. The first one is a classical U–Pb analysis of zircon and baddeleyite using the ID-TIMS method [6,10]. The second one is an analysis of single grains of zircon and baddeleyite from duplicate samples using the ²⁰⁵Pb–²³⁵U tracer with/without ion-exchange chromatography [9,10]. All U–Pb studies were carried out with a seven-channel thermionic mass-spectrometer Finnigan MAT-262 in dynamic mode using a multiplier or a quadrupole RPQ accessory in an ion-counting mode. All the measured isotope ratios were adjusted for mass-discrimination obtained from the study of parallel analyses of SRM-981 and SRM-982, the mass-discrimination value being 0.12% ± 0.04%. Measurements were performed at the Geological Institute of the Kola Science Centre, Russian Academy of Sciences (GI KSC RAS). The method of the U–Pb isotope analysis is published in the work [10].

To study the reproducibility issue regarding the isotope study results, we performed a local U–Th–Pb analysis of zircon and baddeleyite applying the secondary-ion mass-spectrometry method (SIMS) with the SHRIMP-II device [20,21]. Monofractions were extracted from repeatedly selected samples. The isotope studies were carried out at the Centre of Isotopic Research of the A.P. Karpinsky Russian Geological Research Institute (VSEGEI).

Mass-spectrometric measurements were performed in the course of a single uninterrupted session in order to avoid the cross-correlation issues regarding the results of analytical sessions and standards, which was not a trivial task concerning the baddeleyite measurements. The baddeleyite grains were randomly oriented in relation to the analytical disc surface, thus minimizing the possible “matrix inhomogeneity” effect. Then, the grains were mounted in a single compound together with the reference materials, i.e., zircon 91,500 and the “Phalaborwa” baddeleyite (aged 2060 Ma). Analytical error was calculated by measuring the reference materials from the prepared compound. The error corresponded to the value of −1.2% (2σ), which is generally comparable to the errors obtained from the zircon measurements [34].

Sm–Nd Isotope Analysis. The analysis was performed using sample collections of intrusive rocks from the Monchepluton and the Monchetundra massifs alongside the dykes. Measurements of the Nd isotope composition and Sm–Nd concentration were carried out applying the isotope dissolution method with the Finnigan MAT-262 seven-channel solid-phase mass-spectrometer in a static double-band mode and collectors with Re + Re and Ta-Re filaments. Measurements were carried out at GI KSC RAS in accordance with procedures described in [10]. Some of the Sm–Nd analyses were performed at VSEGEI.

Re–Os Isotope Analysis. Re and Os concentrations and the Os isotopic composition were analyzed at Laboratoire de Géochimie et Cosmochimie, IPG de Paris, France. The analysis procedure was published in [35,36].

Samples were prepared for the Re–Os analysis using a low-temperature acid digestion technique. Approximately 02–05 g of rock sample powders was spiked with a mixed ¹⁹⁰Os–¹⁸⁵Re spike and then dissolved with HBr and HF in a Teflon bomb at 145 °C. After evaporation, Os was oxidized to OsO₄ in a nitric acid solution containing chromium trioxide to ensure spike–sample equilibration. Finally, Os was extracted in liquid bromine and purified by a microdistillation technique. The supernatant was reduced by ethanol, and Re was extracted and purified by liquid–liquid extraction with isoamylic alcohol and 2N HNO₃. The purified Os and Re fractions were loaded on Pt and Ni filaments, respectively, and measured using negative thermal ionization mass spectrometry (N-TIMS) on a Finnigan MAT-262. The IPGP Merck Os standard yielded ¹⁸⁷Os/¹⁸⁸Os = 0.1746 ± 8 (n = 4, 2σ standard deviation) during the period of the measurements, and the total procedural blank for Os was 0.040 ± 0.015 pg (n = 7). The ¹⁸⁷Os/¹⁸⁸Os ratios for the blanks ranged between 0.138 and 0.475 with a mean value of 0.279. The Re blanks ranged between 8 and 10 pg. Because the total blanks for both Re and Os were run as part of each batch of dissolutions, the appropriate blank correction was applied to each batch of samples.

4. Geochemistry

4.1. Monchepluton

The initial review of variations in the major petrogenic components and ore elements (Figure 6) shows their asymmetrical distribution at the Monchepluton generalized cross-section, which is determined by the increase in the MgO content and, to a lesser extent, Fe₂O_3tot content in dunite–harzburgite–orthopyroxenites of the I, II, and III megacycles, as well as the Al₂O₃, CaO, Na₂O, and K₂O content in norite–gabbronorite–anorthosites of the IV and V megacycles.

The MgO distribution at the generalized cross-section vividly indicates a process of consecutive accumulation of olivine and orthopyroxene accumulating rocks, whereas the distribution of other components is associated with the crystallization and accumulation of plagioclase. The content of anorthite (An%) in plagioclase from the orthopyroxenes of Mt. Nittis varies from 67.4 to 51.5. The crystallization differentiation of magmatic melt was accompanied by the change in composition of accumulated minerals, which can be demonstrated through the example of the olivine chemical composition. The content of forsterite (Fo%) in olivine varies from 95 in chromitites and 89 in dunites to 87–85 in harzburgites and 84–81 in orthopyroxenites [6,31]. The asymmetry caused by the changes in the mineral assemblage and the cumulated rock composition is broken at the boundaries of megacycles and within the zones of rhythmical layering.

The increased Cr contents (>1% Cr₂O₃) are associated with the dunites of the II megacycle, while the high Cr contents (up to 37.8% Cr₂O₃) are associated with the ore layers of the Sopcheozero deposit. The Ni distribution is more complicated. Its content in harzburgites and orthopyroxenites of the I megacycle that are intersected by subvertical ore veins is lower than in dunites of the II megacycle. The dunites contain Ni (0.4%–1.1% NiO) in the form of a silicate in olivine. The highest contents of the sulfide Ni were discovered in the “Ore Horizon 330” of the III megacycle. Within the Marginal Zone, we observe an abrupt decrease in the content of the mafic components, i.e., MgO, Fe₂O_3tot, and Cr₂O₃, yet the content of salic components (Al₂O₃, CaO, Na₂O + K₂O, and TiO₂) increases.

Variations in the major component contents in rocks of all distinguished megacycles are depicted in petrochemical diagrams (Figure 7).

The analytical points form two distinct trends on the SiO₂–MgO diagram. The first trend includes a dunite–harzburgite–orthopyroxenite series of the I, II, and III megacycles, which have typically increased SiO₂ content, while the MgO content decreases. The second trend is a norite–gabbronorite–anorthosite series of the IV and V megacycles with significantly increased Al₂O₃ and CaO contents, where SiO₂ is relatively stable with decreasing MgO.

All the rocks form a common trend from dunites to anorthosites, as shown by the diagrams showing the ratio of Al₂O₃, CaO, and Na₂O + K₂O to MgO. The Marginal Zone gabbronorites take a medium position, except for the low P₂O₅ content. We observe a clear decrease in Fe₂O_3tot content alongside with the decrease in MgO for the III and IV megacycles, whereas the variations in Fe₂O_3tot content are significantly higher in rocks of different megacycles. The metagabbro of the “10th anomaly” is close to the rocks of the V megacycle with regard to the major elemental content, featuring high TiO₂ content and low NiO content. The plagioharzburgites of the intrusive body intersected by the M-1 borehole match the rocks of the III megacycle in regard to the SiO₂/MgO ratio, but feature higher contents of Al₂O₃, CaO, and Na₂O + K₂O. These higher contents result from the addition of plagioclase (43%–53% An). The increased volatile and P₂O₅ contents are observed in pegmatoid rocks that occur in the upper part of the “Ore Horizon 330”, the III megacycle.

Rocks of all megacycles are characterized by the uniform REE spectra that are normalized to chondrite, the light rare earth elements (LREE) slightly exceeding the heavy rare earth elements (HREE) (Figure 8) and the differentiation degree being various. The highest differentiation degree is observed in the rocks of the I and III megacycles, while the lowest one is observed in the rocks of the V megacycle. The rocks of the IV and, to a lesser extent, V megacycles have a slightly positive Eu anomaly. The spider-diagrams depicting the content of coherent elements normalized to the depleted magma—DM (Figure 9)—clearly show the enrichment of the Monchepluton and the magma feeder channel rocks with the low-field-strength, large-ion lithophyle elements (LILE: Rb, Ba) and high-field-strength elements (HFSE: Th, U), as well as with light rare earth elements (LREE: La–Nd). Rocks of all megacycles feature a negative Nb and Ta anomaly and a positive Sr anomaly. The Marginal Zone gabbronorites take a medium position on both diagrams (Figure 8 and Figure 9). The metagabbro of the “10th anomaly” bears the most resemblance to the rocks of the V megacycle.

The process of the crystallization differentiation is shown in Figure 10, where the two trends are vividly expressed. The first trend is determined by the decrease in the Ni/Co ratio while the Fe# increases from the III megacycle dunites to the metagabbro of the “10th anomaly”. This trend indicates a transition from a sulfur-free system (silicate Ni only) to a sulfuric one (sulfide Ni). The second trend features an abrupt and anomalous increase in the Ni/Co ratio within the single III megacycle (Mt. Sopcha) with a slight variation in the Fe#.

4.2. Main Ridge Complex

The duration of the Main Ridge Complex formation (over 35 Ma) is considerably longer than that of the Monchepluton (8–10 Ma). It should be noted that the Main Ridge Complex rocks are metamorphosed under the conditions of amphibolite facies which have an irregular character. The rocks are mostly metamorphosed within the tectonic zones, where the garnet–amphibole paragenesis appears.

Let us review the geochemical features of rocks of the mentioned massifs regarding their inner structure. We shall carry out separate analyses for each area, i.e., the three areas of the Monchetundra massif, the Chunatundra, and the Volchetundra massif, with harrisites as an addition and the feeder body for comparison.

The rocks of the three Monchetundra massif areas are grouped in the SiO₂–MgO diagram (Figure 11) in the form of disconnected or partially overlaid fields that do not form definite trends by contrast with the Monchepluton. The Lower Zone rocks differ considerably from that of the Middle and Upper Zones in regard to higher MgO and Cr content and lower Al₂O₃ and CaO content. Compositions of rocks of the Middle and Upper Zones are largely overlaid. The Chunatundra massif rocks are maximally enriched with Al₂O₃ and Na₂O + K₂O and depleted of MgO and Fe₂O_3tot.

All rocks of the Monchetundra massif feature a uniform flat partition of REE-normalized spectra, slightly expressed an excess of LREE over HREE, and clearly expressed positive Eu anomalies (Figure 12) determined by the plagioclase accumulation. All rocks of the Chunatundra and Volchetundra are enriched with LREE and have strongly expressed positive Eu anomalies (Figure 12). Judging by the spider-diagrams (Figure 13), the negative Nb and Ta anomalies and the positive Sr anomaly are detected for all rocks of the Main Ridge Complex, as well as for the Monchepluton rocks. Two clusters can be distinguished in the Monchetundra massif cross-section. The first cluster corresponds to the Lower Zone, and the second one unites rocks of the Middle and Upper Zones. All rocks of the Chunatundra and Volchetundra are the two most evolved differentiates with predominant anorthosites.

The intrusive body plagioharzburgites form a separate field on the petrochemical diagrams (Figure 11) that differs from the Monchetundra massif rocks. Plagioharzburgites contain lenses of melanocratic olivine norites occurred in the Lower Zone of the Monchetundra massif. Plagioharzburgites also differ from the mafic rocks of the Monchetundra massif by a considerable excess of LREE over HREE and the absence of Eu anomalies (Figure 12 and Figure 13).

Harrisites breaking through the Upper Zone gabbroids of the Monchetundra massif have no parallels among the magmatic formations of the district. They have specific features, i.e., high Fe₂O_3tot content (18–21 wt.%) and low contents of SiO₂ and P₂O₅, yet they match the Monchetundra massif rocks (Figure 11) concerning the normalized REE spectra and spider-diagrams (Figure 13).

4.3. Imandra–Umbarechka Complex and Ostrovsky Massif

The Imandra–Umbarechka Complex rocks do not form one trend and are divided into two clusters. The first cluster comprises rocks of the Lower Zone (the Bolshaya Varaka and Umbarechensky massifs) with increased MgO, Ni, and Cr contents; the second cluster comprises rocks of the Main and Upper Zones (the Prikhibinsky massif) with increased Al₂O₃ and CaO contents (Figure 14).

The Upper Border Zone rocks with high contents of TiO₂ and Fe₂O_3tot are considered specific, as well as granophyres with anomalously high contents of SiO₂, K₂O, and P₂O₅ that appeared as the result of partial melting of the overlaying dacitic volcanics. These data confirm that the massifs are independent and do not form a single lopolith. The massifs should be united into two units, i.e., the Umbarechensky and the Prikhibinsky ones. Most of their volume was formed in the period of 2442–2437 Ma, and the additional phases were formed in the period of 2396–2395 Ma [5].

Rocks of the Ostrovsky massif are the most similar to rocks of the second cluster regarding the normalized REE spectra and differ in their high Ni and low Cr contents. The increased total REE content and relative enrichment with LREE in comparison to the Monchepluton rocks, negative Nb and Ta, positive Sr anomalies, as well as the absence of the Eu anomaly were determined for the rocks of the Imandra–Umbarechka Complex and the Ostrovsky massif (Figure 15).

The orthoamphibolites prevail in the Luvtuaivench, Ospe, and Ekostrovsky massifs. They bear quite a resemblance to the rocks of the Main Zone and partial resemblance to the Upper Border Zone of the Imandra–Umbarechka Complex (Figure 14 and Figure 15) in regard to their petrochemical features, the normalized REE spectra, and spider-diagrams, including the negative Nb and the positive Sr anomalies.

4.4. Dykes

Dykes of different composition and age are widely present within the Monchegorsk ore district. The dykes are divided into local and regional complexes [5]. The first ones are presented only in the district territory; the second ones are widely presented outside of it. The analysis of dykes was carried out in relation to their age, as we chose dykes that are coeval with the Monchepluton (2.5 Ga). The U–Pb age of the studied dykes varies within the 2506–2495 Ma range (

Table 1. The content of oxides and ore elements in the dyke rocks according to atomic absorption analysis and the U–Pb age [6].

No. Sample	M-12	M-14	M-61	M-52	2066
U–Pb age (Ma)	2506 ± 10	2496 ± 14	2491 ± 9	2495 ± 13	2455 ± 10
SiO2	52.44	54.85	52.81	64.53	40.89
TiO2	0.15	0.15	0.52	0.42	0.23
Al2O3	16.36	5.33	11.72	13.51	8.21
Fe2O3	2.38	1.24	1.01	1.07	20.63
FeO	4.77	8.31	7.68	4.56	-
MnO	0.13	0.18	0.13	0.08	0.26
MgO	7.48	22.13	14.25	4.40	22.94
CaO	10.83	4.52	6.74	3.66	4.49
Na2O	2.39	0.75	1.90	2.97	0.74
K2O	0.3	0.21	0.92	2.78	0.10
H2O-	0.56	0.56	0.20	0.11	-
H2O+	1.59	1.33	1.17	1.79	1.29
P2O5	0.01	0.01	0.12	0.16	0.06
CO2	0.15	0.1	0.43	0.44	0.43
S	0.02	0.01	0.20	0.07	0.20
F	0.009	0.009	0.027	0.065	0.027
Cl	0.020	0.035	0.004	0.006	0.004
Total (wt.%)	99.59	99.72	99.83	100.62	99.84
Cu, ppm	120	52	60	99	60
Ni, ppm	180	590	600	280	632
Co, ppm	39	69	50	29	164
Cr, ppm	26	2100	2300	770	36
V, ppm	140	120	130	110	67

Note: M-12—coarse-grained gabbronorite, Dunite block, borehole C-1518/70 m; M-14—coarse-grained melanonorite, Dunite block, borehole C-1586/63 m; M-61—olivine gabbronorite, Olenegorsk quarry; M-52—quartz diorite, Olenegorsk quarry; 2066—harrisite, north-east slope of Mt. Monchetundra.

). Most dykes were revealed by the boreholes. They intersect the Monchepluton rocks and occur within the Loipishnyun area, located in the Pentlandite Canyon between the Monchepluton and Monchetundra massifs.

The dykes linked to the stage of the Imandra–Varguga Zone volcanic strata formation are still out of focus. These dykes are composed of felsic and mafic rocks. The felsic dykes are represented by microgranite and microgranophyre forming the marginal quenching zone of composite dykes intersecting the II megacycle dunites. One of such dykes was discovered under moraine in the eastern side of the Sopcheozero pit. Quartz diorites of similar composition are found in a fragment of the Olenegorsk dyke exposed by the quarry. These rocks feature the highest contents of SiO₂ (62.9–75.5 wt.%), Al₂O₃ (12.1–13.5 wt.%), and Na₂O + K₂O (5.6–8.5 wt.%) and the lowest contents of TiO₂ and P₂O₅ (Figure 16). They are enriched with LREE, Rb, Ba, Th, U, and have clear Nb, Sr, and Eu negative anomalies (Figure 17). The negative Sr and Eu anomalies allow us to clearly differentiate these rocks from the mafic dyke rocks.

The mafic dykes may be divided into four groups (Figure 16) according to the increasing MgO, Cr, and Ni content with respect to the other components. The I group includes dykes of microgabbro and coarse-grained gabbroids intersecting the II megacycle dunites. They show quite low MgO content (3.59–4.74 wt.%) and increased TiO₂ content (1.38–1.89 wt.%). The II group dykes are revealed by the boreholes at the II megacycle dunites and are represented by microgabbronorite and gabbronorites with an MgO content of 6.91–8.24 wt.%. These rocks show a moderate content of TiO₂ (0.15–1.10 wt.%) and increased content of Al₂O₃ (14.53–16.56 wt.%). The III group dykes intersect rocks of the I and II megacycles and the Neoarchean ferruginous quartzites of the Olenegorsk iron ore pit. They contain mostly homogenous medium-grained olivine gabbronorites with an MgO content of 13.91–14.64 wt.%, Al₂O₃ content of 10.47–11.72 wt.%, and Na₂O + K₂O content of 1.35–2.82 wt.%. The IV group dykes also intersect the I and II megacycle rocks and are represented by the olivine micronorites and melanonorites with the highest MgO (19.51–25.42 wt.%), medium Al₂O₃, low Na₂O + K₂O, and increased PGE content. The contents of Cr and Ni increase spasmodically in dyke rocks from the first group to the fourth group: Cr increased from 36 to 3841 ppm, and Ni increased from 66 to 1130 ppm. The composition of V decreases in the same direction—from 645 to 152 ppm.

Melanocratic gabbronorite and norite dykes with the U–Pb age of 2487 ± 12 Ma were discovered by the boreholes (C-1717, C-1720) in the Loipishnyun region. They are very close to fourth-group dykes but differ in their lower contents of MgO (17.51–18.09 wt.%) and Cr (700 ppm).

The first group of mafic dyke rocks are also enriched with Rb, Ba, Th, U, and REE, which is comparable with acidic rocks. They are characterized by considerable excess of LREE over HREE and clearly expressed negative anomalies of Nb and Ta (Figure 17). The high REE content can be determined by the increased content of apatite and other fluid-bearing mineral phases. The dyke rocks of the II–IV groups have a lower REE content and close to normal REE spectra and spider-diagrams (Figure 17). They are also characterized by the domination of LREE over HREE, negative Nb and Ta anomalies, and positive Eu anomaly. The additional factors for the second group of dyke rocks include a poorly expressed Eu positive anomaly and a negative Zr anomaly.

Harrisites with the U–Pb age of 2455 ± 10 Ma that break through the rocks of the Monchetundra massif Upper zone are the closest to the fourth-group dykes but differ in their anomalous Fe₂O₃ (20.63 wt.%) content and lowered SiO₂ (40.89 wt.%).

According to the petrogeochemical diagrams, normalized REE spectra, and spider-diagrams, the mafic dyke rocks are very similar to the IV and V megacycle rocks of the Monchepluton, including the metagabbro of the “10th anomaly”, but contain less Na₂O + K₂O. These results combined with the mafic dyke age data confirm that the rocks are comagmatic with the Monchepluton. This allows us to attribute the dykes to melts that filled contractional cracks. The dyke integration into the dunite block rocks could take place when the magma filled the Southern chamber or in the process of its crystallization with the residual melts enriched with fluids and PGE.

According to the normalized REE spectra and spider-diagrams, the komatiitic basalts from the Vetreny Belt with the U–Pb age of 2405 ± 5 Ma [37] bear a significant resemblance to the rocks of all (I–IV) groups (Figure 17). Yet, they are distinguished from each of these groups by a higher degree of differentiation of rock-forming components in relation to MgO. The volcanics, as well as the intrusive rocks, show negative Nb and Ta anomalies and a positive Sr anomaly. The positive Eu anomaly is absent. Similar komatiitic basalts occur within the Polisar formation of the Imandra–Varzuga Zone.

5. The U–Pb, U–Th–Pb, Sm–Nd, and Re–Os Isotope Systems

5.1. Reproducibility of the U–Pb and U–Th–Pb Isotope Data

To estimate the reproducibility of the isotopic age of zircon and baddeleyite, we took four reference objects (

Table 2. Data on isotope analysis of zircon and baddeleyite from the rocks of the Monchepluton, Monchetundra massif, and dykes.

Methods	ID-TIMS * [7]	ID-TIMS * [6,10]	ID-TIMS ** [9]	SIMS-SHRIMP II [20,21]
Gabbronorite–pegmatite, “Critical Horizon”, Mt. Nyud, IV Megacycle	2504.4 ± 1.5 (Zr)	M-2: 2505 ± 5 (Zr)	M-2: 2503.5 ± 4.6 (Zr)	M-64: 2500 ± 11 (Zr)
Metagabbronorite, Pt-reef, Mt. Vurechuaivench, V Megacycle		M-42: 2497 ± 21 (Zr + Bd)	M-42: 2498 ± 6.7 (Bd)
Gabbronorite, Monchetundra massif, the bottom of the Upper Zone		M-6, M-54: 2501 ± 8 (Zr) M-55: 2505 ± 6 (Zr)	M-55: 2504 ± 7.4 (Zr)	M-55: 2494.6 ± 7.3 (Zr) 1841 ± 25 (Zr1)
Gabbronorite, dyke, the Olenegorsk quarry		M-52: 2495 ± 13 (Bd)		M-61: 2490.5 ± 8.9 (Bd)

Note: (*) U–Pb method with ²⁰⁸Pb–²³⁵U tracer, (**) ²⁰⁵Pb–²³⁵U tracer; M-2–M-61 are the sample numbers; Zr—zircon; Bd—baddeleyite; Zr¹—metamorphic zircon.

). The first one is the “Critical Horizon” of Mt. Nyud (the IV megacycle of the Monchepluton). Earlier, a zircon from pegmatoidal gabbronorite showed a concordant value U–Pb age of 2504.4 ± 1.5 Ma [7].

Later, the same rocks were used to obtain zircon and baddeleyite (M-2 sample). According to the U–Pb analysis data (ID-TIMS), their age is 2500 ± 5 Ma [6]. According to the U–Pb analysis of single grains of zircon, their age is 2503.5 ± 4.6 Ma [10]. The zircon from the M-64 sample (duplicate of M-2 sample) was analyzed using the in situ U–Th–Pb age analysis method (SIMS SHRIMP II). Its concordant age is 2500 ± 11 Ma [20,21].

The second object is used to characterize the metagabbronorite of the ore Pt-reef of the Mt. Vurechuaivench foothill (the V megacycle of the Monchepluton). The U–Pb age of zircon and baddeleyite (M-42 sample) is 2497 ± 21 Ma [6]. The U–Pb analysis of single grains of baddeleyite from the same sample allowed us to adjust the age and reduce the error of estimate to 2498 ± 6.7 Ma. According to the in situ U–Th–Pb age analysis of zircon from metaanorthosite (sample B-1) and metagabbronorite (sample B-2) from the Pt-reef, its age values are 2507.9 ± 6.6 and 2504.2 ± 8.4 Ma, respectively [11]. These values are slightly greater than those obtained from the U–Pb analysis, but they are close to the age of zircon from the Marginal Zone (MZ), which is 2507 ± 9 Ma (Figure 3).

The third object is used to characterize a horizon of unchanged trachytoid gabbronorites from the Monchetundra massif (Figure 5). M-6 and M-54 samples were taken near the head of the M-1 structure borehole. An M-55 sample was collected at a distance of 1.5 km north-eastward along the strike. The U–Pb zircon age values (ID-TIMS) from this horizon are 2501 ± 8 Ma (M-6 and M-54 samples) and 2505 ± 6 Ma (M-55 sample) [6]. A close U–Pb age value of 2504 ± 7.4 Ma was obtained from the single zircon grains of the M-55 sample [9]. Zircon from M-65 sample (duplicate of M-55 sample) was analyzed using the in situ U–Th–Pb age analysis method (SIMS-SHRIMP II) [20,21]. The magmatic zircon with undisturbed REE spectrum shows the concordant age value of 2494 ± 7.3 Ma, and the metamorphic zircon shows the concordant age value of 1841 ± 25 Ma. The metamorphic zircon age complies with the age of tectonic movements at the Monchetundra fault, which was obtained in the result of the Sm–Nd and Rb–Sr analyses of garnet–amphibole parageneses from blastocataclasites after gabbroids [26]. According to the mineralogical thermobarometry, the metamorphism parameters are the following: p = 6.9–7.6 kbar and T = 620–640 °C, which is equivalent to the conditions of amphibolite facies.

The obtained age ranges of 2498–2506 Ma for the Monchepluton and 2501–2505 Ma for the Monchetundra massif do not exceed the range of analytical errors and confirm a proper reproducibility of given data. This allows us to estimate the duration of the Monchepluton formation at 8–10 Ma with great probability.

The fourth object is essentially a gabbronorite dyke exposed at the Olenegorsk iron ore pit. The baddeleyite prismatic crystals (M-52 sample) were separated out from the quartz diorite that forms the central part of the dyke. Its isochron U–Pb age (ID-TIMS) is 2495 ± 13 Ma [6]. The M-61 duplicate sample characterized a homogenous part of the dyke revealed at the pit slope. The results of the U–Th–Pb analysis of baddeleyite (SIMS SHRIMP II) show comparable age value of 2491 ± 8.9 Ma [20,21]. The obtained results indicate the intrusion of gabbronorite dykes into the Archean basement rocks at the later stages of the Monchepluton formation, not the earlier ones.

5.2. Sm–Nd Isotope System

The model age of the mantle substratum (depleted mantle (DM)) and homogenous reservoir (CHUR) was determined on the basis of previously published results of the Sm–Nd isotope analysis of intrusive rock bulk samples taken from the Monchepluton and Main Ridge massifs [4,6]. The model age for the Monchepluton protolith varies within the range of 3.09–2.9 Ga, and for the Main Ridge Complex rocks, this range is 3.08–2.91 Ga. This shows that their protoliths were formed in close succession. The CHUR model age also varies slightly within the range of 2.75–2.68 Ga.

Figure 18 shows variations in Nd content and primary ratios of εNd and ⁸⁷Sr/⁸⁶Sr at the generalized cross-section of the Monchepluton regarding its megacycle inner structure. Amidst the general tendency of low Nd content, we observe its sharp increase into the Marginal Zone rocks, then decrease on the boundaries of the II–III and III–IV megacycles, and sharp increase in gabbro-anorthosites of the V megacycle. With rare exception, the εNd ratio is characterized by the negative values, and the primary ⁸⁷Sr/⁸⁶Sr ratio is characterized by values bigger than 0.7040, which indicates the anomalous type of mantle source enriched with lithophylic elements and different from the depleted mantle. At the same time, the variations in primary εNd and ⁸⁷Sr/⁸⁶Sr ratios at the generalized cross-section of the pluton are distributed heterogeneously, i.e., clear εNd changes are detected on the boundaries of the II–III and the III–IV megacycles, while the ⁸⁷Sr/⁸⁶Sr changes are detected on the boundaries of the II–III megacycles.

5.3. Re–Os Isotope System

This system is a unique tool to reconstruct the processes of mantle magma generation because Os stays in mantle restite, unlike the other isotope systems, and Re transfers into the melt completely as it is an incompatible element. This leads to fractionation of the Re/Os ratio during the partial mantle melting. The mantle restites feature very low Re/Os ratios, and the melts feature very high ones.

The analysis of Re and Os in intrusive rocks and minerals (chromite, sulfides) of the Monchepluton and the Monchetundra massif gives the following results: Os content varies (ppb) within the range of 0.1–1.7 in gabbroids, orthopyroxenites, and harzburgites; 4.6–5.6 in the Monchepluton dunites; and 3.2–17.2 in sulfides. In chromite ores it increases to 44.7–161.1 [3]. The Re/Os ratio is low in the Monchepluton chromites (0.04–0.08) and dunites (0.08–0.10) and the feeder harzburgites (0.23). It has a medium value in orthopyroxenites and gabbroids (0.44–12.45) and the maximum value in sulfides (10.86–15.09).

The primary γOs ratio in the Monchepluton chromites varies from +1.6 to +2.7, which is comparable to the data obtained from the komatiitic basalts of the Vetreny Belt (+1.7) [38].

6. Discussion

6.1. The Monchepluton Megacyclicity and Trends of Magma Differentiation

Relatively simple differentiation trends are defined for volcanic flows of the komatiitic basalts from the Vetreny Belt. Their mafic composition is close to that of parental magma for the studied layered intrusions. They are characterized by a clearly expressed inverse relation between MgO and SiO₂, Al₂O₃, CaO, and Na₂O + K₂O and by direct relation with Fe₂O_3tot [5,39,40]. Such trends are determined by the crystallization and accumulation of olivine accumulate rock. Both these processes resulted from the decreasing magmatic melt temperature during the cooling of the melt in a closed system.

Unlike the volcanics, a combination of two consecutive trends on a MgO–SiO₂ diagram is typical for the Monchepluton. The first trend shows sequential SiO₂ increases, while the second one shows relatively stable SiO₂ content (Figure 7). Both trends are determined by the change from paragenesis CrSp + Ol + Opx to paragenesis Opx + Pl. The Monchepluton provides no direct relation between MgO and Fe₂O_3tot due to considerable spread of Fe₂O_3tot content within the megacycle limits.

One of the reasons for this is the system opening and the injection of a new portion of magma enriched with iron (and nickel). Earlier, direct signs of a new injection within the limits of the III megacycle (“Ore Horizon 330”) were established—erosion of the cumulative olivine layer, its partial dissolution and crystallization of new phases of Cr-spinel, olivine and orthopyroxene under conditions of imbalance [5]. Against this background, the variation in dark-colored mineral ferruginosity in each rhythm due to multiple interchanges of decreasing and increasing of the phase crystallization temperatures. The decrease in temperature is determined by the general magma cooling, while the increase in temperature is determined by the relief of concealed crystallization temperature [41,42].

The analysis of Nd content variations and isotopic labels at the Monchepluton generalized cross-section (Figure 18) shows that there were repeated violations of rock formation sequences, the latter being determined by the crystallization differentiation. Within the theoretical framework of the megacyclicity model, we can explain these violations by a pulsating magma flow. Taking into consideration the different values of primary εNd and ⁸⁷Sr/⁸⁶Sr ratios, this model assumes that new magma batches would come from the deep-laid reservoir and intermittent vent and would be further differentiated. A sharp increasing (2–3 times) in the Nd content in the Marginal Zone rocks is determined by the interaction with the host rocks of the Archean age.

6.2. Issues of Formation Period and Composition of the Main Ridge Complex

For quite a long time, the Main Ridge Complex was considered to be a gabbro-anorthosite structure of Archean age [19]. A complete rethink of this model happened at the beginning of the XXI century, when we obtained the first valid results of the U–Pb analysis of zircon from the trachytoid gabbronorites occurring on the boundary between the Medium and Upper Zones of the Monchetundra massif. This result is 2501 ± 8 Ma [6], and it indicates that the Monchepluton and the Monchetundra massif are close in age. Later, a considerable time span of 2521–2445 Ma was determined for the Monchetundra massif rocks, which is far longer than the duration of the Monchepluton formation (8–10 million years).

By now a lot of U–Pb and Sm–Nd isotope analyses have been performed for the Monchetundra massif, and single analyses have been performed for the other massifs of the Main Ridge Complex [6,10,12,14,17,25,43,44].

We can distinguish four stages in accordance with the results of the U–Pb isotope analysis of zircon and baddeleyite from the Monchetundra massif. The first one is the Early Stage of 2521–2516 Ma (this stage is conventional due to the uncertainty of the geological position and metamorphic genesis of the sampled rocks). The second one is the Main Stage of 2507–2496 Ma. The third one is the Late Stage of 2476–2471 Ma. Finally, the fourth one is the late intrusive Pegmatoid Stage of 2456–2445 Ma. The average age of the II Main Stage is 2502 ± 5 Ma (six analyses). The average age of the III Late Stage is 2473 ± 8 Ma (three analyses). The average age of the IV Pegmatoid Stage is 2451 ± 4 Ma (three analyses). The Chunatundra and Volchetundra massif rocks are close in age to the III Stage rocks, being of 2467 ± 7 Ma and 2473–2463 Ma, correspondingly. The harrisite dyke (2455 ± 10 Ma) and metadolerite dykes (2450 ± 10 Ma) are close in age to the IV Pegmatoid Stage. The Imandra–Umbarechka Complex rocks (2442–2437 Ma) are the closest in age to the IV Stage [6].

The rocks of the Main Stage lie on the border of the Middle and Upper Zones and within the Lower zone. The Late Stage rocks are found within the limits of the Middle and the Upper Zones. The Pegmatoid Stage rocks are found mainly within the limits of the Upper Zone and also in the Lower Zone. As revealed by the abovementioned age data, the Monchetundra massif was formed in the result of multiple invasions of mafic magmatic melts from the deep-laid reservoirs. The composition of melts gradually changed. More ancient melts are represented mainly by norites and trachytoid mesocratic gabbronorites and their olivine-bearing forms. The younger melts are represented by massive meso- and leuco-gabbronorites and gabbro-anorthosites. The formation of harrisites and gabbro-pegmatites and the Main Stage events are separated in time. Narrow bodies of olivine orthopyroxenites and harzburgites occur within the limits of the Lower Zone. They are close to the intrusion body rocks in regard to the geochemical features (Figure 11, Figure 12 and Figure 13).

Chashchin V.V. [14] reconstructed the P-T formation conditions for the rocks of the Monchetundra massif and the Monchepluton based on various previously published geothermometers and geobarometers. It was defined that the crystallization of rocks in the Lower Zone of the Monchetundra massif happened at a temperature of 1190–1000 °C and pressure from 5.3 to 6.4 kbar (taking into account similar results from different geobarometres). The Monchepluton rocks crystallized at a temperature of 1300–1200 °C (average value is 1230 ± 35 °C) and pressure of 3.0 kbar. Thus, the formation of the abovementioned intrusions happened at different depths, i.e., the Monchetundra massif was formed at a depth of ca. 20 km, and the Monchepluton was formed at a depth of ca. 10 km. They contacted along the deep fault zone (intersected by M-1 borehole) at a period of 2.04–1.90 Ga [26]. These data do not confirm that both massifs belong to a single intrusive complex, which is in agreement with the above-mentioned results of geochemical analysis of rocks comprising the massifs (Figure 7, Figure 8, Figure 9, Figure 10, Figure 11, Figure 12 and Figure 13).

6.3. Issues of the Magma Feeder Channel

The 350 m thick intrusive ultramafic body that is intersected by M-1 borehole occurs at a depth of 2037–2337 m. It is separated from the overlaying Monchetundra massif by a metamorphic rock complex of Archean age and over 1 km thickness (Figure 5). It comprises mostly plagioharzburgite and dunite, its lower chill margin consists of orthopyroxenite, and the upper chill margin consists of melanocratic norite. Their grain coarseness decreases towards the contacts. Narrow lens-shaped bodies of olivine melanocratic gabbronorites occur at depths of 2065–2073 and 2083–2108 m.

The composition of the main phase (olivine) features a high content of Fo (86.5%–90.8%) and increased NiO (0.14%–0.64%). In its composition, it is close to olivine from the dunites of the Dunite Block (II megacycle). Olivine crystallized under the conditions close to the equilibrium with high-magnesium orthopyroxene and Cr-spinel. Orthopyroxene is represented by a high-magnesium variety (84.8%–89.9% En, 9.5%–13.4% Fs). The Fs amount in orthopyroxene increases towards the contacts. The subordinate amount of clinopyroxene (diopside) is registered. It contains 0.47%–0.62% TiO₂. Interstitial plagioclase contains 43.6%–55.9% An. A reaction kelyphite rim composed of amphibole is observed on the boundary between olivine and plagioclase. The accessory Cr-spinel contains (wt.%) 37.11–46.14 of Cr₂O₃, 12.81–18.36 of Al₂O₃, 4.01–8.49 of MgO, 28.08–41.63 of Fe₂O_3tot, as well as an admixture of TiO₂ (0.25–1.42; 2.84; 4.73).

A symmetrical distribution of MgO, Cr, and Ni content with a trend of decreasing towards the both contact zones is vividly expressed at the body vertical section (Figure 19). This agrees quite well with the distribution of crystallization temperature (small values on the contacts and stable values within the major part of the body) and also with the absence of olivine and Cr-spinel accumulation. The Fe₂O_3tot is distributed in different manner, i.e., it increases gradually from the bottom to the top of the section and breaks within the chill margin limits. The first tendency is connected with the gradual decrease in the crystallization temperature, and the second one is connected with the relatively quick cooling of melt on the contacts (the more intense cooling on the upper contact). An equal Nd distribution in the majority of the section (3.23–3.99 ppm) and its abrupt increasing on the lower contact (6.55 ppm) indicate the interaction with the host rocks, as it was described for the Monchepluton (Figure 18).

In order to define the body intrusion age, we performed the U–Pb isotope analysis of zircon using the ID-TIMS method. Zircon was separated from the medium-grained melanocratic olivine gabbronorite (2100–2102 m). The age of magmatic zircon is 2510 ± 9 Ma, the MSWD = 1.4, and the age of the more ancient xenogenous metamorphosed zircon is 2780 ± 10 Ma [17]. Both kinds are clearly distinguished by the U content values, which are 169–524 and 35 ppm, respectively.

An equilibrium pair of olivine–Cr-spinel was used to reconstruct the crystallization temperature of mineral phases. Samples were taken from different depths (2059.5, 2072, 2132, 2138 m). The T–fO₂ conditions were calculated using the modified O’Neil–Wall–Ballhaus–Berry–Green (O’NWBBG) equations [45,46], which made it possible to take into account the effect of Ti on the partition of Fe and Mg between the mineral phases. The calculated values of the crystallization temperatures of olivine–Cr-spinel pairs range from 1103 to 1222 °C with a calculation accuracy equal to 50 °C (

Table 3. Olivine–Cr-spinel oxythermobarometry of ultramafic rocks of the Magma Feeder Channel.

N Analysis, Olivine *	N Analysis, Cr-Spinel *	T, °C	dlogfO2 (QFM) **
M-1/2059,5-2	M-1/2059,5-2	1103	+2.6
M-1/2072,0-1-5	M-1/2072,0-1-12	1119	+1.9
M-1/2132-1	M-1/2132-1c	1222	+2.2
M-1/2132-2	M-1/2132-1c	1217	+2.3
M-1/2338-1	M-1/2338-1	1221	+1.8
M-1/2338-2	M-1/2338-1	1197	+1.9

Note: * Chemical composition of minerals are available at http://dx.doi.org/10.17632/432y98sbvx.1. ** dlogfO₂ (QFM)—logfO₂ relative to the QFM buffer.

). There is a tendency of temperature to increase with depth. The average temperature for the six coexisting olivine–Cr-spinel pairs is about 1180 °C.

There are broad variations in the redox state of the inspected samples, with the oxygen fugacity dlogfO₂ (QFM) varying from +1.8 to +2.6. Calculated values of crystallization temperatures and the oxygen fugacity values in the ultramafic rocks of the magma feeder channel are practically identical to those of the host dunites of the Sopcheozero chromite deposit (1163–1165 °C, dlogfO₂ (QFM) = +2) [5].

The above data confirm that the plagioharzburgite body revealed at a depth of 2037–2337 m is essentially a magma feeder channel. The intrusion of magma and its crystallization happened almost synchronously with the formation of the Monchepluton and the beginning of formation of the Monchetundra massif.

6.4. Sources of Magmas for Layered Intrusions

The very first results of studying the Sm–Nd and Rb–Sr systems in the Monchepluton mafic rocks indicated the anomalous values of primary isotopic ratios. There were negative εNd values and increased ⁸⁷Sr/⁸⁶Sr values [47]. Later, there were studies on Sm–Nd, Rb–Sr, and Re–Os systems in rock-forming minerals, rocks, and ores of layered intrusions, performed independently in several laboratories by different researchers [3,4,6,8,10,44]. The results of that studies confirmed the anomalous character of mantle source enriched with lithophylic elements. In this case, the primary εNd ratio in intrusive rocks is less than –1, although, according to the De Paolo–Wasserburg model, the rocks of mantle genesis (aged 2.5–2.4 Ga) should have featured the primary ratio value εNd = +3.5 in the case of depleted magma (DM).

Let us analyze the diagram εNd–T(Ma) (Figure 20). It is constructed on the basis of the Sm–Nd isotope analysis results (authorial and published) of the following objects: the layered intrusions of the Monchegorsk ore district in Karelia and Finland; the gabbro-anorthosite massifs of the Main Ridge and the Pyrshin; the “Drusite Complex” massifs of the Belomorian Block; and the gabbronorite dykes and komatiitic basalts of the Vetreny Belt. The age data were input with an adjustment to the U–Pb analysis of zircon; therefore, we mostly took into account samples with a defined U–Pb isotope age.

These magmatic rocks are distributed practically over the whole Kola–Lapland–Karelian Province, which is the most ancient part of the Fennoscandian Shield. Most of analytical data, including the volcanic data, lie within the range from −1 to −3 εNd. The maximum range is defined for gabbro-anorthosites, while the minimum range is defined for the “Drusite Complex” rocks. The plagiogneisses of the Archean complexes that host the Monchetundra massif and the Koitelainen intrusion feature the lowest possible values of the εNd(T) ratio from −3.5 to −18.1. These data allow us to assume active interaction between the parental magmas for layered intrusions and gabbro-anorthosite massifs and the crustal matter enriched with lithophylic elements and fluids.

Values of the primary εNd ratio that are anomalous for all analyzed magmatites indicate a single isotopic source. The distribution of magmatites on a large area indicates a common geological position of this source. The analysis of the deep-laid mantle xenolithes situated in the Paleozoic diatremes and explosive dykes within the Belomorian Block [50] provides basis for discussion concerning the composition of reservoirs existed 2.5 Ga. Values of the primary εNd ratio vary from −0.8 to −2.5 for the xenolithes of spinel peridotites and pyroxenites which U-Pb age is 2.47–2.41 Ga. The obtained results are comparable to the data on layered intrusions

A group of analytical data with positive values of the primary εNd ratio is also depicted on the diagram (Figure 20). The group comprises dunites–harzburgites–orthopyroxenites of the Pados massif (+2), dunites (+2.5), and chromitites (+2.9) of the Sopcheozero deposit and the plagioharzburgites of the magma feeder channel (+1). They represent a relict matter that is close to the depleted mantle and was not taken into consideration before. Similar rocks occur in the Lower Zone of the Monchetundra massif, in the Lopishnyun area [14,27].

Rocks from layered intrusions that belong to two age groups (2.50 and 2.44 Ga) and contain chromite ore deposits were analyzed using the Re–Os isotope method. The results are depicted on the γOs–Age (T) diagram (Figure 21). Data on the Neoarchean Kostomuksha komatiites, the Paleoproterozoic komatiitic basalts from the Vetreny Belt, and the Pechenga ferropicrites are given for reference. As the subcontinental lithospheric mantle (SCLM), we took ophiolites of the Jormua Complex (Finland).

The diagram analysis shows that the layered intrusions feature a considerable spread of the γOs ratio. Judging by the negative γOs values, we should assume that the Monchepluton interacted with the SCLM, and the Akanvaara and Koitelainen massifs interacted with the continental crust. Granophyre zone (200–260 m thick), microgabbro, and magnetite gabbro that occur in the Upper Zones of the Akanvaara and Koitelainen massifs confirm the active interaction between magma and continental crust. Similar rocks are observed in the Upper Zone of the Imandra–Umbarechka Complex. Quite frequently found relics of ancient zircon (aged 2.7 Ga) also indicate the presence of the crustal component. The komatiitic basalts are considered to have a minimum interaction with the SCLM.

Taking into consideration the above-mentioned geochemical and isotopic data, we may conclude that at a time of 2.5 Ga, an uplift of the mantle plume to the limits of lower crust, the formation of the deep-laid mantle reservoirs, and a large-scale interaction of parental magma with the granulite–eclogite complex rocks occurred, preserving the depleted mantle relics.

The magma uplift was accompanied by the locally developed contamination and assimilation processes. These processes happened on the contacts of the intermittent vents and magmatic chambers with the host amphibolite–gneissic Archean complexes. Later, the plume moved to the south judging by the isotopic age of layered intrusions.

The mantle magmatism was accomplished by the komatiitic basalts from the Vetreny Belt (2.41 Ga). This indicates the decreasing depth of intermittent magmatic vents feeding the Paleoproterozoic volcanoes.

7. Conclusions

The following conclusions were made based on the geochemical analysis of intrusive and dyke rocks, the chromite ores of the Monchegorsk ore district, and the research of the U–Pb, Sm–Nd, and Re–Os isotope systems.

(1)

The mafic–ultramafic Monchepluton and the gabbro-anorthosite Monchetundra massif are close in age. Initially, they were formed at different depths (ca. 10 and 20 km), and they differ in rock set and differentiation trends. Therefore, they cannot be considered a single complex. They were spatially close in the Svecofennian Orogeny (2.04–1.90 Ga) due to movement along the deep-laid fault.

(2)

The Monchepluton is a two-chamber layered intrusion that was formed 2502 ± 5 Ma. Five megacycles are distinguished in its generalized cross-section. The megacycles differ in rock composition, volatile and ore element contents, stratification and differentiation character, and mineralization types (chromite, Cu–Ni–PGE sulfide, and low-sulfide PGE reef types). They were formed as the result of multiple injections of high-magnesium magma, the washing-out of previously settled accumulative rocks, the breaking of equilibrium crystallization conditions, and an abrupt change in isotopic indicators (εNd, ⁸⁷Sr/⁸⁶Sr). At the late magmatic stage, a filling of contractional cracks by the residual melts enriched with the volatile components occurred. The formation of mafic dykes of four groups with the age of 2506–2495 Ma also occurred.

(3)

At a depth of 2.2 km, the M-1 structural borehole intersected the ultramafic rocks that composed a bed-lens-shaped body (350 m thick) with the remained chill margins. Judging by the rock composition, their geochemical and isotopic features, this body is comparable to the Monchepluton rocks and is essentially a magma feeder channel. The intrusion time defined with the U–Pb isotope analysis of zircon (ID-TIMS) is 2510 ± 9 Ma, which is comparable to the Monchepluton formation period.

(4)

The gabbro-anorthosite massifs comprise a single complex of the Main Ridge that was consequently formed by the Monchetundra, Chunatundra, Volchetundra, and Losevo–Medvezhye tundras. The positive Eu anomaly increases in the same sequence determined by the increase in a basic plagioclase accumulation and fractionation degree. The largest Monchetundra massif was formed as the result of multiple injections of mafic magmatic melts from the deep-laid reservoirs. The melt composition changed gradually. The major volume of the massif was formed in two stages, i.e., 2507–2496 Ma and 2476–2471 Ma. The intrusion of the Chunatundra, Volchetundra, and Losevo–Medvezhye tundra massifs happened at the second stage. The gabbroic pegmatites (2456–2445 Ma) and coeval harrisite dykes that are anomalously enriched with iron are the most recent ones.

(5)

The Imandra–Umbarechka Complex massifs and the nearby massifs of the Ostrovsky, Ospe, and others did not form a single intrusion. Essentially, they are near-surface structures formed as the result of the injection of a more ferruginous magma. The studies revealed active interaction between these structures and overlaying dacitic volcanics with further formation of granophyre zones.

(6)

The composition of parental magmas of layered intrusions is comparable to the composition of the komatiitic basalts concluding the high-magnesium magmatism. However, they are depleted in regard to the ore elements. The geochemical analysis of rocks and the study of the Sm–Nd and Re–Os systems showed that the parental mantle magmas had an active interaction with the granulite–eclogite and amphibolite–gneiss complexes, including the metasedimentary rocks of the Archean greenstone belts enriched with sulfur. Deep-laid reservoirs and intermittent vents of these magmas were situated within the limits of the lower and middle crust. The Monchepluton and Pados massif preserved a relict matter close to the depleted mantle with positive εNd values (2.0–2.9) in the form of dunite–chromitite complexes.

(7)

Based on the analysis of the results of geological, geochemical, and isotope studies of intrusions of the Monchegorsk ore district, a model of their formation is proposed. According to the model, at the turn of the Neoarchean and Paleoproterozoic (2.5 Ga), there was a rise in superplume, generation, and introduction of high-magnesian mantle magmas into the lower part of the Earth’s crust; the formation of a system of deep and intermediate magmatic chambers; a preliminary differentiation of magmas within them; intensive contamination by magmas of the substance of the granulite–eclogite and amphibolite–gneiss complex; the formation of magmatic chambers as a result of multiple injections; and the sequential formation of ores—chromite, Cu–Ni–PGE sulfide, and low-sulfide PGE ores.