Three-D Mineralogical Mapping of the Kovdor Phoscorite-Carbonatite Complex, NW Russia: II. Sulfides

Abstract

The world largest phoscorite-carbonatite complexes of the Kovdor (Russia) and Palabora (South Africa) alkaline-ultrabasic massifs have comparable composition, structure and metallogenic specialization, and can be considered close relatives. Distribution of rock-forming sulfides within the Kovdor phoscorite-carbonatite complex reflects gradual concentric zonation of the pipe: pyrrhotite with exsolution inclusions of pentlandite in marginal (apatite)-forsterite phoscorite, pyrrhotite with exsolution inclusions of cobaltpentlandite in intermediate low-carbonate magnetite-rich phoscorite and chalcopyrite (±pyrrhotite with exsolution inclusions of cobaltpentlandite) in axial carbonate-rich phoscorite and phoscorite-related carbonatite. Chalcopyrite (with relicts of earlier bornite and exsolution inclusions of cubanite and mackinawite) predominates in the axial carbonate-bearing phoscorite and carbonatite, where it crystallizes around grains of pyrrhotite (with inclusions of pentlandite-cobaltpentlandite and pyrite), and both of these minerals contain exsolution inclusions of sphalerite. In natural sequence of the Kovdor rocks, iron content in pyrrhotite gradually increases from Fe₇S₈ (pyrrhotite-4C, Imm2) to Fe₉S₁₀ (pyrrhotite-5C, C2 and P2₁) and Fe₁₁S₁₂ (pyrrhotite-6C) due to gradual decrease of crystallization temperature and oxygen fugacity. Low-temperature pyrrhotite 2C (troilite) occurs as lens-like exsolition inclusions in grains of pyrrhotite-4C (in marginal phoscorite) and pyrrhotite-5C (in axial phoscorite-related carbonatite). Within the phoscorite-carbonatite complex, Co content in pyrrhotite gradually increases from host silicate rocks and marginal forsterite-dominant phoscorite to axial carbonate-rich phoscorite and carbonatite at the expense of Ni and Fe. Probably, this dependence reflects a gradually decreasing temperature of the primary monosulfide solid solutions crystallization from the pipe margin toward its axis. The Kovdor and Loolekop phoscorite-carbonatite pipes in the Palabora massif have similar sequences of sulfide formation, and the copper specialization of the Palabora massif can be caused by higher water content in its initial melt allowing it to dissolve much larger amounts of sulfur and, correspondingly, chalcophile metals.

1. Introduction

Sulfur, alongside H₂O, CO₂ and halogens, is an important volatile constituent in magmas [1,2,3]. Unlike all other major volatile elements, sulfur changes its oxidation state depending on the oxygen fugacity (fO₂) regime. The two most abundant forms of S in silicate melts are sulfide (S²⁻) and sulfate (S⁶⁺) [4,5,6]. The transition between these two oxidation states is very sharp and occurs over the range of oxygen fugacity between the FMQ equilibrium and 2logfO₂ units above (FMQ + 2). However, sulfur shows radically different behavior in these two oxidation states, and, consequently, the existing sulfur species may provide important insights into the evolution of magmas and mineralizing fluids.

Reduced sulfur actively reacts with iron forming FeS(melt) species; therefore, iron content determines “sulfide capacity” of silicate melts [4,7,8]. Besides, iron influences the sulfur oxidation state in silicate melts at given redox conditions [4,5]. Various sulfur-bearing species (H₂S, SO₂, HS^–, S_n⁻, S_nS²⁻, HSO₃⁻, HSO₄⁻, S₂O₃²⁻, SO₃²⁻, SO₄²⁻) serve as ligands for transportation of chalcophile and highly siderophile elements (Cu, Ni, Co, Zn, Mo, Ag, Au and PGE) in silicate melts and magmatic volatiles [9,10], and also control precipitation of the ore-forming elements in magmatic-hydrothermal environment.

Sulfide mineralization is a typical component of the (phoscorite)-carbonatite complexes in alkaline-ultrabasic massifs. The Loolekop phoscorite-carbonatite pipe in the Palabora (Phalaborwa) massif, South Africa (RSA) is a typical example of such complexes, where sulfide mineralization reached the economic level. Sulfides are abundant in alkaline-ultrabasic massifs of the Kola Alkaline Province (NW Russia), and the Kovdor massif is one of the richest. Like the Palabora massif, the Kovdor massif predominantly concentrates sulfides within the Kovdor phoscorite-carbonatite complex, while host silicate rocks (peridotite and foidolite) contain much less sulfides [11,12]. Different aspects of sulfide mineralization in the Kovdor phoscorite-carbonatite complex were discussed in [11,12,13,14,15,16,17]. However, almost all of these works presented results of the studies implemented on few random samples with comparatively rare sulfides, while rock-forming sulfides remained almost unstudied.

The general geology of the Kovdor alkaline-ultrabasic massif and its phoscorite-carbonatitre complex has been described by [11,13,18,19,20,21], and there is a short geological digest of these works in the first article of this series [22]. In recent years, we have obtained new 3D data on petrography [23,24,25], composition and properties of economic [22,26,27,28,29] and rare minerals [30,31,32,33,34,35,36] of the Kovdor phoscorite-carbonatite complex and host ultrabasic and alkaline rocks. This data enabled us to establish formation sequences of the Kovdor massif and its phoscorite-carbonatite complex [21]: peridotite → foidolite and melilitolite → metasomatic rocks (fenite, diopsidite, phlogopitite, scarn-like rocks) → phoscorite and phoscorite-related carbonatite → vein calcite carbonatite → vein dolomite carbonatite and dolomite-magnetite-serpentine rock. The 3D data on distribution of content, grain size and composition of sulfides presented here will help us to understand the behavior of both sulfur and sulfur-related metals during crystallization and subsolidus evolution of the Kovdor massif.

2. Materials and Methods

About 550 core samples of phoscorite, carbonatites and host rocks were taken from 108 boreholes drilled within the Kovdor phoscorite-carbonatite complex. We analyzed thin polished sections of these core samples with a petrographic microscope to estimate textural characteristics, mineral relations and grain size of sulfides. Quantitative relations between magnetic and non-magnetic pyrrhotites were determined in polished sections with a nematic liquid crystal MBBA [37]. Sulfide grain sizes were estimated with the Image Tool 3.0 program [38] as a mean equivalent circular diameter.

Electron-microscope analyses were carried out using a LEO-1450 scanning electron microscope with a Röntek energy-dispersive spectrometer. Chemical compositions of sulfides were determined with the Cameca MS-46 electron microprobe (Geological Institute of the Kola Science Center, Russian Academy of Sciences) operating in a wavelength-dispersive mode at 20 kV and 20–30 nA. The electron beam diameter used was 1–10 μm. The applied standards and detection limits are listed in

Table 1. Parameters of EPMA analyses.

Element	Detection Limit, wt %	Standards for EPMA Analyses	Element	Detection Limit, wt %	Standards for EPMA Analyses
Mg	0.1	Pyrope	Se	0.08	Synthetic PbSe
Al	0.05	Pyrope	Mo	0.1	Metallic molybdenum
Si	0.05	Diopside	Pd	0.05	Metallic palladium
S	0.05	Synthetic Fe10S11	Ag	0.05	Metallic silver
K	0.03	Wadeite	Cd	0.05	Synthetic CdS
Ca	0.03	Diopside	Sn	0.05	Metallic tin
Mn	0.01	Synthetic MnCO3	Sb	0.05	Antimony
Fe	0.01	Synthetic Fe10S11	Te	0.05	Synthetic PbTe
Co	0.01–0.03	Metallic cobalt	Pt	0.05	Metallic platinum
Ni	0.01	Synthetic NiAs	Au	0.05	Metallic gold
Cu	0.01	Metallic copper	Pb	0.05	Synthetic PbSe
Zn	0.01	Synthetic ZnO	Bi	0.06	Bismuth
As	0.05	Synthetic NiAs

. Abbreviations used include Acn (acanthite), Ap (hydroxylapatite), Apn (argentopentlandite), Bdy (baddeleyite), Bn (bornite), Brt (barite), Cal (calcite), Cbn (cubanite), Ccp (chalcopyrite), Clc (clinochlore), Cls (clausthalite), Djf (djerfisherite), Dol (dolomite), Fo (forsterite), Ght (goethite), Gn (galena), Hss (hessite), Hwl (hawleyite), Mag (magnetite), Mch (moncheite), Mck (mackinawite), Mrc (marcasite), Nph (nepheline), Pcl (pyrochlore), Phl (phlogopite), Pn (pentlandite), Po (pyrrhotite), Py (pyrite), Sgn (siegenite), Sp (sphalerite), Spl (spinel), Srp (serpentine), Tro (troilite), Ttn (titanite), Val (valleriite) and Vlt (violarite).

Bulk-rock samples were analyzed by the Tananaev Institute of Chemistry of KSC RAS (Apatity) by means of inductively coupled plasma-mass spectrometry (ICP-MS) performed with an ELAN 9000 DRC-e mass spectrometer (Perkin Elmer, Waltham, MA, USA). For the analyses, the samples were dissolved in a mixture of concentrated hydrofluoric and nitric acids with distillation of silicon and further addition of hydrogen peroxide to a cooled solution to suppress hydrolysis of polyvalent metals [39].

Cation contents were calculated with the MINAL program of D. Dolivo-Dobrovolsky [40]. Statistical analyses were carried out with the STATISTICA 8.0 [41] and TableCurve 2.0 [42] programs. For the statistics, resulting values of the analyses below the detection limit (see Table 1) were considered to be 10 times lower than the limit. Geostatistical studies and 3D modelling were conducted with the MICROMINE 16 program [43]. Interpolation was performed with ordinary kriging. Automatic 3D geological mapping (Figure 1a) was performed by means of conversion of the rocks chemical composition to mineral composition by logical computation [24].

Crystal structures of pyrrhotite samples (00-01, 36/33, 00-10-41, 01-11-91, 00-51) were studied by means of Agilent Technologies Xcalibur Eos diffractometer operated at 50 kV and 40 mA. More than a hemisphere of three-dimensional data was collected at 100 K temperature by monochromatic MoKα X-radiation with frame widths of 1° and 5–15 s count for each frame. The crystal structures were refined in standard and non-standard settings of different space groups: P-6m2, P312, P32, P-3m, P-3, P3m, Cmce, Cca, Cc, C222₁, C2, P2₁ for the 5C-polytype and P-6m2, Aem2, Amm2, C2/m, C/m, C2, F2/d, Imm2 for the 4C-polytype by means of the SHELX program [44] incorporated in the OLEX2 program package [45]. The final models were chosen according to the following criteria: absence of violating reflections, lower means of R-factors and GOOF, absence or low number of atoms with physically unrealistic anisotropic displacement parameters (without restraints). Crystal structures were refined with a pseudomerohedral twin model with [−1 0 0 0 −1 0 0 0 1 2] twining matrix for the C2 and P2₁ models. Empirical absorption correction was applied in the CrysAlisPro program [46] using spherical harmonics implemented in the SCALE3 ABSPACK scaling algorithm. The crystal structures were visualized with the Diamond 3.2f program [47].

3. Results

3.1. Sulfur Mineralization

In primary silicate rocks of the Kovdor massif, average sulfur content is comparatively low (

Table 2. Median contents of sulfide-forming elements in rocks of the Kovdor massif.

Rock	Host Silicate Rock	(Ap)-Fo Phoscorite	Low-Cb Mag-Rich Phoscorite	Cal-Rich Phoscorite and Related Carbonatite	Vein Calcite Carbonatite
n	44	17	90	50	16
S, wt %	0.22	0.20	0.15	0.29	1.53
Fe	6.49	7.88	29.79	17.48	3.85
Cu, ppm	56	44	37	78	55
Zn	89	100	198	117	28
Co	40	65	92	70	29
Ni	111	49	39	17	17
Ag	2	1	8	6	1
Pb	1	1	1	2	3

), but it becomes higher in apo-peridotite metasomatic diopsidite and phlogopitite. Within the phoscorite-carbonatite pipe, sulfur content increases from marginal low-carbonate phoscorite to axial calcite-rich phoscorite and calcite carbonatite (Figure 1b). Besides sulfur, the Kovdor sulfuric compounds include Fe, Co, Ni, Cu, Zn, Ag, Pb and Ba (see Table 1). Many of these metals initially concentrated in rock-forming forsterite (Ni), apatite (Ba), magnetite (Fe, Co, Zn) and pyrochlore (Pb), and demonstrate unclear or even negative correlation with sulfur. Nevertheless, sulfuric compounds inherited initial distribution of these metals in earlier minerals resulting in concentric distribution corresponding to the secondary sulfides and sulfates within the phoscorite-carbonatite complex.

In the Kovdor rocks, major concentrators of sulfur are sulfides, mainly, pyrrhotite and chalcopyrite (Figure 1c), while sulfates, mainly, barite, are the products of low-temperature alteration of sulfides. Pyrrhotite is a common accessory to a rock-forming mineral of the most phoscorite and carbonatites varieties, apart from vein dolomite carbonatite and the related dolomite-magnetite-serpentine metasomatic rock. Chalcopyrite is closely associated with pyrrhotite, being predominantly concentrated in сarbonate-rich phoscorite, phoscorite-related carbonatite and vein calcite carbonatite of the ore-pipe apical part (up to 20 modal %). In the apical part of the pipe axial zone, there is calcite-rich phoscorite with rock-forming chalcopyrite (up to 30 modal %).

All other sulfides and sulfates resulted from subsolidus exsolution and low-temperature hydrothermal alteration of pyrrhotite and chalcopyrite. Pyrrhotite contains exsolution inclusions of pentlandite–cobaltpentlandite (almost always) and troilite (rarely), and chalcopyrite carries exsolution inclusions of sphalerite (common) and cubanite (rare). Pyrite and valleriite are spread products of pyrrhotite and chalcopyrite alteration, while primary bornite sometimes occurs as relicts within secondary chalcopyrite. Rare secondary sulfides include djerfisherite, marcasite (after both pyrrhotite and chalcopyrite), chalcocite and covellite (after chalcopyrite only). In addition, galena, violarite, mackinawite, moncheite, petzite, hessite and clausthalite form separate inclusions in pyrrhotite, while acanthite, argentopentlandite, altaite, galena, hawleyite, siegenite, tsumoite, volynskite, wittichenite occur as inclusions in bornite and chalcopyrite.

3.2. Pyrrhotite and Products of Its Alteration

In the Kovdor massif, pyrrhotite-4C, Fe₇S₈, pyrrhotite-5C, Fe₉S₁₀, and pyrrhotite-2C (troilite), FeS were found and structurally verified, while presence of pyrrhotite-6C, Fe₁₀S₁₁ was established only according to its composition. Few pyrrhotite modifications often occur within one sample, and even within one grain of pyrrhotite (Figure 2a); however, non-magnetic monoclinic modification 5C always predominates (in average, 88 vol. %). Content of ferrimagnetic orthorhombic pyrrhotite-4C reaches 50 vol. % in host silicate rocks and marginal (apatite)-forsterite phoscorite, 30 vol. % in intermediate low-carbonate magnetite-rich phoscorite, 20 vol. % in axial calcite-rich phoscorite and related carbonatite, and 15 vol. % in vein calcite and dolomite carbonatite [14]. Pyrrhotite-6C is the latest modification that substitutes magnetite and chalcopyrite (mainly, in calcite carbonatite), and forms pyrrhotite-pyrite intergrowth resulting from exsolution of earlier pyrrhotite-5C or -4C. Pyrrhotite-4C forms separate grains, marginal zones of non-magnetic pyrrhotite crystals and thin (up to 50 µm) lens-like inclusions in pyrrhotite-5C (see Figure 2a). Exsolution hexagonal pyrrhotite-2C (troilite) is found in about 5% of the samples (mainly, in marginal apatite-forsterite phoscorite and axial phoscorite-related carbonatite) as lens-like inclusions in grains of pyrrhotite-4C and pyrrhotite-5C (Figure 2b).

In peridotite, rounded or irregularly shaped pyrrhotite grains (up to 120 μm in diameter) with exsolution inclusions of pentlandite fill interstices between rock-forming olivine and diopside. In foidolite, there are sporadic pyrrhotite grains in cancrinitized nepheline grains (Figure 2c) and late calcite veinlets. Apo-peridotite metasomatites, especially carbonatized and apatitized, contain up to 3 modal % of pyrrhotite that fills interstices between grains of rock-forming minerals in close association with chalcopyrite, magnetite and ilmenite (Figure 2d).

As mentioned above, pyrrhotite is a predominant accessory to rock-forming sulfide of phoscorite and carbonatite. In marginal forsteritite and apatite-forsterite phoscorite, pyrrhotite (up to 0.2 modal %) occurs within and between forsterite grains as pseudohexagonal plate crystals (up to 500 μm in diameter) with exsolution lamellae of pentlandite inside (see Figure 2a), as well as irregularly shaped grains in close intergrowth with magnetite (Figure 2e). Content of pyrrhotite in phoscorite increases with growth of magnetite amount and reaches 8 modal % in intermediate low-carbonate magnetite-rich phoscorite. Herein, pyrrhotite usually forms close intergrowths with magnetite, and, together with valleriite, replaces earlier magnetite and fills fractures in its grains (Figure 2f). Such secondary pyrrhotite often inherits cubic inclusions of spinel from exsolved Mg-Al-rich magnetite.

Gradual transition of low-carbonate magnetite-rich phoscorite into axial calcite-rich phoscorite and phoscorite-related carbonatite is accompanied by growth of pyrrhotite content and disappearance of its connection with magnetite. In this type of rock, there are sulfide-rich areas (up to 8 m in diameter), where pyrrhotite (up to 50 modal %) and chalcopyrite cement separate grains of all other minerals (Figure 2g). Besides, pyrrhotite forms here porous rims around chalcopyrite grains (Figure 2h) and tabular to short prismatic crystals (up to 2 cm in diameter) with hexagonal prismatic {10-10} and pinacoidal {0001} faces. In selvages of vein calcite and dolomite carbonatites, pyrrhotite occurs as bronze-yellow platy crystals (up to 15 cm in diameter and 1.5 cm thick), partially replaced with pyrite and goethite (Figure 2i).

Mean equivalent circular diameter of pyrrhotite grains (

Table 3. Grain size and chemical composition of the pyrrhotite group minerals (mean ± SD/min–max).

Rock	Host Rock		Phoscorite and Related Carbonatite			Vein Carbonatite
	Foidolite	Diopsidite	(Ap)-Fo	Low-Cb Mag-rich	Cal-rich	Cal	Dol
n	9	27	13	60	94	26	25
D, µm	100 ± 80 15–300	140 ± 90 10–300	120 ± 90 50–400	200 ± 80 50–450	220 ± 90 10–600	230 ± 90 100–600	200 ± 100 50–600
S, wt %	39.2 ± 0.5 38.54–40.38	39.1 ± 0.5 37.67–39.98	39 ± 1 36.72–39.82	39.1 ± 0.6 36.31–40.52	38.9 ± 0.7 35.81–40.36	39.0 ± 0.4 38.27–39.77	39.0 ± 0.5 37.74–39.77
Fe	60.5 ± 0.5 59.83–61.11	60.6 ± 0.9 59.31–63.01	61 ± 2 59.77–64.13	60.8 ± 0.8 58.18–63.51	60.9 ± 0.8 58.93–63.37	60.8 ± 0.5 59.77–61.87	60.7 ± 0.6 59.37–61.92
Co	0.1 ± 0.1 <0.01–0.43	0.06 ± 0.06 <0.01–0.19	0.1 ± 0.1 <0.01–0.39	0.2 ± 0.1 <0.01–0.39	0.2 ± 0.1 <0.01–0.54	0.14 ± 0.09 0.02–0.36	0.13 ± 0.08 <0.01–0.33
Ni	0.3 ± 0.1 0.03–0.44	0.3 ± 0.2 <0.01–0.82	0.2 ± 0.1 <0.01–0.38	0.06 ± 0.09 <0.01–0.36	0.04 ± 0.09 <0.01–0.77	0.01 ± 0.02 <0.01–0.08	0.1 ± 0.1 <0.01–0.38
Fe, at. %	46.8 ± 0.4 45.85–47.35	47.0 ± 0.7 46.02–48.98	48 ± 1 46.34–50.02	47.1 ± 0.7 45.26–50.10	47.2 ± 0.8 45.80–50.22	47.2 ± 0.4 46.20–47.87	47.2 ± 0.5 46.21–48.08
Ni	0.2 ± 0.1 0.00–0.32	0.2 ± 0.2 0.00–0.59	0.1 ± 0.1 0.00–0.26	0.04 ± 0.07 0.00–0.26	0.03 ± 0.07 0.00–0.58	0.00 ± 0.01 0.00–0.05	0.06 ± 0.08 0.00–0.31
Co	0.1 ± 0.1 0.00–0.32	0.04 ± 0.05 0.00–0.16	0.05 ± 0.07 0.00–0.27	0.11 ± 0.08 0.00–0.27	0.14 ± 0.09 0.00–0.38	0.10 ± 0.06 0.00–0.27	0.10 ± 0.07 0.00–0.27
S	52.9 ± 0.4 52.55–53.88	52.8 ± 0.6 51.02–53.56	52 ± 1 49.98–53.39	52.7 ± 0.7 49.90–54.47	52.6 ± 0.7 49.73–53.82	52.7 ± 0.4 52.03–53.53	52.7 ± 0.5 51.76–53.36

) increases insignificantly from 80 µm in marginal (apatite)-forsterite phoscorite to 160 μm in intermediate low-carbonate magnetite-rich phoscorite, and then to 200 µm in axial calcite-rich phoscorite and carbonatite. This trend is similar to that of co-existing rock-forming and accessory minerals [21,26,27].

Chemical composition of pyrrhotite (Table 3, Figure 3a) varies from Me₇S₈ (pyrrhotite-4C) to MeS (pyrrhotite-2C), with intensive maximum at Me₉S₁₀ (pyrrhotite-5C). Approximation of total metal content histogram by the corresponding Gaussians with fixed positions at Me = 46.66 for Fe₇S₈, 47.36 for Fe₉S₁₀, 47.82 for Fe₁₁S₁₂ and 50.00 for FeS has confirmed a sharp predominance of pyrrhotite-5C with subordinate roles of 6C, 4C and, especially, 2C modifications. Contents of iron and sulfur are distributed in accordance with normal law (Figure 3b), while cobalt and nickel are characterized by lognormal and exponential distributions correspondingly (Figure 3c,d).

Iron content in pyrrhotite (exclusive of exsolution troilite) gradually increases from earlier host silicate rocks to the latest vein carbonatites (Figure 4a), which is fully in accordance with gradual increase of pyrrhotite-5C fraction due to presence of pyrrhotite-4C in this sequence. Taking into account pyrrhotite structural data according to [48], where Fe₇S₈ = Fe²⁺₅Fe³⁺₂☐S²⁻₈, Fe₉S₁₀ = Fe²⁺₇Fe³⁺₂☐S²⁻₁₀ and Fe₁₁S₁₂ = Fe²⁺₉Fe³⁺₂☐S²⁻₁₂, one can see a gradual decrease of Fe³⁺/Me²⁺ ratio in natural rock sequence of the Kovdor massif (Figure 4b), and from marginal (apatite)-forsterite phoscorite with pyrrhotite-4C to intermediate low-carbonate magnetite-rich phoscorite with pyrrhotite-5C and, finally, calcite-rich phoscorite and calcite carbonaite with pyrrhotite-5C and -6C (Figure 5a).

Experimental details and crystallographic parameters of structurally different pyrrhotites are given in

Table 4. Structural data for studied pyrrhotites.

Sample	1 (00-01)	2 (36-33)	3 (00-10-41)	4 (01-11-91)	5 (00-51)
Modification	4C	5C	5C	5C	5C
Refined formula	Fe6.78S8	Fe8.91Ni0.25S10	Fe8.99S10	Fe8.84S10	Fe8.88S10
Formula weight	634.86	833.04	822.41	814.45	816.41
Temperature/K	100 (2)	100 (2)	100 (2)	100 (2)	100 (2)
Crystal system	orthorhombic	monoclinic	monoclinic	monoclinic	monoclinic
Space group	Imm2	C2	C2	C2	P21
a (Å)	22.678 (4)	11.8624 (9)	11.8875 (11)	11.8706 (7)	6.8477 (5)
b (Å)	3.4131 (5)	6.8613 (5)	6.8667 (6)	6.8589 (5)	28.584 (4)
c (Å)	5.9083 (13)	28.593 (2)	28.661 (2)	28.5953 (16)	6.8518 (5)
α (°)	90	90	90	90	90
β (°)	90	89.897 (8)	90.023 (8)	89.982 (5)	119.972 (11)
γ (°)	90	90	90	90	90
Volume/Å3	457.31 (15)	2327.3 (3)	2339.6 (4)	2328.2 (3)	1161.8 (2)
Z	2	8	8	8	4
ρcalc (g/cm3)	4.611	4.755	4.670	4.647	4.668
μ (mm−1)	12.202	12.914	12.542	12.432	12.499
F (000)	608.0	3190.0	3149.0	3119.0	1563.0
Crystal size (mm3)	0.15 × 0.15 × 0.15	0.18 × 0.18 × 0.10	0.13 × 0.13 × 0.13	0.13 × 0.13 × 0.05	0.12 × 0.12 × 0.12
Radiation	MoKα (λ = 0.71073)
2Θ range for data collection (°)	7.126–54.92	5.7–61.33	5.69–62.04	5.70–50.00	5.7–61.932
Index ranges	−29 ≤ h ≤ 27, −3 ≤ k ≤ 4, −4 ≤ l ≤ 7	−15 ≤ h ≤ 16, −8 ≤ k ≤ 9, −40 ≤ l ≤ 23	−16 ≤ h ≤ 16, −6 ≤ k ≤ 9, −33 ≤ l ≤ 40	−14 ≤ h ≤ 14, −8 ≤ k ≤ 7, −34 ≤ l ≤ 33	−9 ≤ h ≤ 7, −38 ≤ k ≤ 32, −9 ≤ l ≤ 9
Reflections collected	1018	5920	10,932	7368	6471
Independent reflections	475	4494	4797	3597	4708
Rint, Rsigma	0.0171, 0.0189	0.0354, 0.0397	0.0396, 0.0545	0.0607, 0.0571	0.0278, 0.0271
Data/restraints/parameters	475/31/53	4494/127/255	4797/1/136	3597/1/135	4708/211/216
Goodness-of-fit on F2	1.202	1.074	1.145	1.608	1.125
Final R indexes [I>= 2σ(I)]	R1 = 0.0780 wR2 = 0.1173	R1 = 0.0896, wR2 = 0.2586	R1 = 0.0884, wR2 = 0.2361	R1 = 0.1170, wR2 = 0.3522	R1 = 0.0881, wR2 = 0.1802
Final R indexes [all data]	R1 = 0.0817 wR2 = 0.1197	R1 = 0.1009, wR2 = 0.2728	R1 = 0.1311, wR2 = 0.2941	R1 = 0.1286, wR2 = 0.3608	R1 = 0.0990, wR2 = 0.1874
Largest diff. peak/hole/e Å−3	1.61/−2.06	4.02/−3.57	5.13/−5.73	4.01/−4.36	4.93/−4.66
Flack parameter	0.49 (12)	0.53 (19)	0.5 (2)	0.26 (15)	0.09 (6)

. The site occupancy factors (s.o.f.) were refined using the scattering curves for neutral atoms given in the International Tables for Crystallography [49]. Due to the large scope of structural information (atom coordinates, bond-lengths, displacement parameters for samples 1–5), these data can be obtained from the CIFs, which are available as Supplementary Materials.

The pyrrhotite crystal structure was firstly described by Nils Alsen [50] in the space group P6₃/mmc with unit cell parameters a = 3.43, c = 5.68 Å. The description was based upon hexagonal-close packing stacked perpendicular to the c axis, where sulfur atoms approximately occupy the nodes, and iron atoms are regularly arranged in octahedral interstices of sulfur atoms [51]. The vacancies between iron sites are in ordered arrangement, and confined to alternate layers of iron atoms and normal to the c axis [52,53]. Formation of different pyrrhotite polytypes is caused by ordered arrangement of vacancies and/or Fe²⁺/Fe³⁺ atoms that results in formation of superstructures [53]. Random arrangement of vacancies leads to the formation of hexagonal 1C polytype, their partial ordering causes orthorhombic 4C modification, and fully ordered vacancy distribution produces monoclinic 5C, 6C and modulated NC structures. Up today, 4C, 5C, 6C pyrrhotite polytypes and modulated NC structures with N = 4.88, 5.5 etc. were reported [52,54,55]. The N is defined as a number of supercell reflections along c* direction (Figure 6) situating between bright reflections of a hexagonal cell.

The crystal structure of ferrimagnetic pyrrhotite-4C, Fe_6.78S₈, was refined in the Imm2 space group. It differs from the previous refinements of crystal structure of stoichiometric Fe₇S₈ in C2/c and C2 space groups [51,56] by distribution of vacancies in Fe sites (Figure 7) based on the observed differences in their chemical compositions. The Kovdor pyrrhotite-4C contains four independent iron sites with refined occupancies of 0.89, 0.75, 1 and 0.75 for Fe1, Fe2, Fe3 and Fe4 respectively.

In the crystal structure of the pyrrhotite-5C with the P2₁ symmetry, anomalously short distance 2.470 Å has been observed between Fe2 and Fe4 sites (Figure 8). There are two possible explanations for this fact: (1) one of these sites is vacant, while the second one is populated (their refined occupancies are 0.24 for the Fe2 site and 0.75 for the Fe4 site); (2) the Fe2-Fe4 interaction has at least partially bonding character. The same structural effect has been observed in [57].

Most samples of the Kovdor pyrrhotite-5C are crystallized in C2 space group. Their chemical composition widely varies, Fe_8.84–8.99S₁₀, which causes differences in occupancy of Fe-sites (Figure 9). From 6 to 9 iron sites (from 22 ones in C2 model) are observed. They have a different vacancy proportion ranging from 1% to 100%. Such variability of 5C polytype can explain its domination in the Kovdor massif (see Figure 3). In samples 1–5, the mean <Fe-S> distance ranges from 2.426 to 2.454 Å, and the expected elongation of bonds can be compensated by partial incorporation of Fe³⁺ in low-occupied sites [48,57].

Similar to nickel content in the rocks (see Table 2), the amount of Ni in pyrrhotite gradually decreases from host foidolite and diopsidite–glimmerite to marginal (apatite)-forsterite phoscorite, intermediate low-carbonate magnetite-rich phoscorite, calcite-rich phoscorite, phoscorite-related and vein calcite carbonatite, and then insignificantly increases in vein dolomite carbonatite and the related dolomite-magnetite-serpentine rock. In contrast, content of Co both in rock and pyrrhotite increases from host silicate rocks and marginal calcite-poor phoscorite to calcite-bearing phoscorite and phoscorite-related carbonatite and then slightly decreases in the vein carbonatites (Figure 4c). Correspondingly, Ni-dominant pyrrhotite is concentrated in the marginal zone of the phoscorite-carbonatite complex, Ni-Co-bearing pyrrhotite occurs in the intermediate zone, and Co-dominant pyrrhotite occupies the axial zone of the pipe (Figure 5b).

Almost all pyrrhotite grains contain exsolution inclusions of pentlandite, (Ni_4.5Fe_4.5)S₈, or cobaltpentlandite, Co₉S₈, and, much rarely, troilite, FeS:

Fe₁₁S₁₂ → Fe₉S₁₀ + 2FeS and Fe₉S₁₀ → Fe₇S₈ + 2FeS;

9(Fe_8.5Ni_0.5)S₁₀ → 9Fe₇S₈ + Fe₉S₁₀ + (Fe_4.5Ni_4.5)S₈;

9(Fe₈Co)S₁₀ → 9Fe₇S₈ + Fe₉S₁₀ + Co₉S₈.

Troilite occurs only as lens-like lamellae along (001) planes of host pyrrhotite from marginal (apatite)-forsterite phoscorite (see Figure 5a) and, rarely, phoscorite-related carbonatite (Figure 10a). The host pyrrhotite is predominantly represented by its 4C modification in marginal (apatite)-forsterite phoscorite, and by 5C modification in axial phoscorite-related carbonatite. In host silicate rocks and vein calcite and dolomite carbonatites, troilite has not been found.

Exsolution inclusions of pentlandite in pyrrhotite occur mainly in host silicate rocks and (apatite)-forsterite phoscorite of the phoscorite-carbonatite complex marginal zone, and independent pentlandite grains (up to 120 μm in diameter) occur rarely in (аpatite)-forsterite phoscorite and dolomite-magnetite-serpentine rock. Pyrrhotite with inclusions of cobaltpentlandite dominates in the pipe axial carbonate-rich zone, while intermediate low-carbonate magnetite-rich phoscorite usually contains pyrrhotite with inclusions of both pentlandite and cobaltpentlandite with approximately equal proportions of Co and Ni-Fe. Exsolution pentlandite–cobaltpentlandite forms distinctive flame-like (up to 200 µm long, Figure 2b) round or hexangular lamellar inclusions (up to 100 µm in diameter and 10 µm thick, Figure 2a and Figure 10b) that predominantly grow along {0001} planes of host pyrrhotite from boundaries of its grains and fissures. Sometimes, the oriented inclusions of pentlandite–cobaltpentlandite are found in pseudomorphs of pyrite, valleriite, carbonates (see Figure 10b) and secondary magnetite after pyrrhotite.

Chemical composition of pentlandite–cobaltpentlandite varies in all possible ranges (

Table 5. Chemical composition of pentlandite–cobaltpentlandite (mean ± SD/min–max).

Rock	Host Rock		Phoscorite and Related Carbonatite			Vein Carbonatite
	Foidolite	Diopsidite	(Ap)-Fo	Low-Cb Mag-rich	Cal-rich	Cal	Dol
n	6	24	13	51	67	17	8
S, wt %	33 ± 1 32.34–34.80	33 ± 1 27.51–34.73	33.0 ± 0.5 32.06–33.74	32.8 ± 0.7 31.84–35.24	32.7 ± 0.8 31.61–36.12	32.6 ± 0.4 31.80–33.57	32.4 ± 0.9 30.64–33.66
Fe	26 ± 9 11.64–30.64	28 ± 3 15.52–32.87	23 ± 7 9.25–30.94	14 ± 7 4.81–28.00	13 ± 8 3.52–35.02	9 ± 3 4.24–15.42	17 ± 9 4.58–28.42
Co	10 ± 20 1.72–47.16	10 ± 10 0.77–54.33	20 ± 20 4.50–52.27	40 ± 10 10.11–61.91	50 ± 20 1.21–61.92	57 ± 4 47.63–62.36	30 ± 20 8.16–63.16
Ni	30 ± 10 8.86–34.53	30 ± 8 2.65–37.12	20 ± 10 5.11–34.47	10 ± 9 0.7–30.14	10 ± 10 0.62–32.96	2 ± 3 0.25–10.26	20 ± 10 3.94–32.21
Ni, apfu	3 ± 2 1.20–4.64	4 ± 1 0.42–5.01	3 ± 1 0.68–4.64	1 ± 1 0.09–3.93	1 ± 1 0.08–4.31	0.3 ± 0.3 0.03–1.36	2 ± 2 0.53–4.24
Fe	4 ± 1 1.65–4.23	3.9 ± 0.4 2.59–4.53	3.2 ± 0.9 1.28–4.38	1.9 ± 0.9 0.68–3.87	2 ± 1 0.50–4.51	1.2 ± 0.4 0.60–2.17	0.65 ± 4.07 3.87–23.32
Co	2 ± 3 0.22–6.35	1 ± 2 0.10–8.60	3 ± 2 0.60–6.88	6 ± 2 1.31–8.38	6 ± 2 0.15–8.26	7.6 ± 0.6 6.30–8.45	4 ± 3 1.07–8.52
S	8	8	8	8	8	8	8

). Increase of Co/Ni-ratio in pyrrhotite causes logarithmical growth of this ratio in exsolution pentlandite–cobaltpentlandite (Figure 11). In the sequence of rock formation, there is a gradual increase of Co content in pentlandite–cobaltpentlandite due to Ni and Fe (see Figure 4c), which causes corresponding spatial zonation of the phoscorite-carbonatite complex (see Figure 5c).

On the M₉S₈ plane of the Fe-Ni-Co-S tetrahedron (Figure 12), chemical compositions of pentlandite are concentrated within the field of the solid-solution stability at ≥200 °C [58]. Chemical compositions of cobaltpentlandite correspond to the Fe_4.5Ni_4.5S–Co₉S₈ trend, whereas most samples from host silicate rocks, marginal forsterite-dominant and intermediate low-carbonate magnetite-rich phoscorite are within the stability field at 300–400 °C, and most samples from axial carbonate rich phoscorite and carbonatites are within the stability field at 200–400 °C.

Comparatively rare exsolution sphalerite occurs as rounded, irregularly shaped, cross- and butterfly-like inclusions (up to 100 µm, Figure 10c) in pyrrhotite from any types of rocks. Sphalerite together with chalcopyrite also form rims (up to 120 µm thick) around pyrrhotite grains in dolomite-magnetite-serpentine rock. Since chemical composition of exsolution sphalerite (

Table 6. Chemical composition of sphalerite inclusions in pyrrhotite (mean ± SD/min–max) and pressure of their formation estimated with a sphalerite-pyrite-pyrrhotite geobarometer [62].

Rock	Host Rock		Phoscorite and Related Carbonatite			Vein Carbonatite
	Foidolite	Diopsidite	(Ap)-Fo	Low-Cb Mag-rich	Cal-rich	Cal	Dol
n	3	13	3	14	24	9	17
S, wt %	31.6 ± 0.6	32.0 ± 0.6	32 ± 1	31.7 ± 0.9	31.7 ± 0.9	31.5 ± 0.7	32 ± 1
	31.02–32.22	30.96–32.83	30.43–32.88	29.60–33.22	29.78–33.86	30.70–32.59	27.60–33.44
Mn	0.5 ± 0.5	0.3 ± 0.5	0.8 ± 0.2	0.6 ± 0.9	1 ± 1	1 ± 1	0.5 ± 0.5
	<0.01–0.95	<0.01–1.44	0.57–0.99	<0.01–3.51	<0.01–6.17	<0.01–2.89	<0.01–1.52
Fe	10 ± 1	9 ± 1	8.8 ± 0.9	10 ± 2	10 ± 2	9 ± 2	10 ± 2
	9.51–11.51	6.18–11.33	8.13–9.79	6.67–15.06	5.51–15.82	5.69–13.23	5.86–12.77
Co	<0.01	<0.01	0.1 ± 0.2	0.2 ± 0.3	0.1 ± 0.2	0.2 ± 0.4	0.2 ± 0.3
			<0.01–0.40	<0.01–0.99	<0.01–0.63	<0.01–1.15	<0.01–1.04
Ni	<0.01	<0.01	<0.01	<0.01	0.01 ± 0.03	<0.01	<0.01
					<0.01–0.13
Zn	56 ± 1	58 ± 2	58 ± 2	57 ± 3	54 ± 5	57 ± 5	57 ± 3
	54.52–57.01	54.04–61.50	56.78–60.27	51.71–62.99	42.41–63.11	49.17–62.79	52.99–62.12
Cd	2 ± 1	1 ± 1	<0.01	0.6 ± 0.8	3 ± 3	2 ± 2	0.1 ± 0.3
	0.73–3.37	<0.01–3.57		<0.01–2.98	<0.01–10.01	<0.01–5.96	<0.01–1.03
Zn, apfu	0.86 ± 0.01	0.89 ± 0.04	0.90 ± 0.06	0.89 ± 0.07	0.83 ± 0.09	0.88 ± 0.09	0.88 ± 0.07
	0.85–0.87	0.81–0.95	0.86–0.97	0.77–1.04	0.63–0.99	0.75–1.00	0–0
Fe	0.19 ± 0.02	0.16 ± 0.03	0.16 ± 0.02	0.17 ± 0.04	0.19 ± 0.04	0.17 ± 0.05	0.19 ± 0.04
	0.18–0.21	0.11–0.20	0.14–0.17	0.12–0.28	0.10–0.27	0.11–0.25	0.11–0.27
Mn	0.01 ± 0.01	0.01 ± 0.01	0.01	0.01 ± 0.02	0.03 ± 0.02	0.01 ± 0.02	0.01 ± 0.01
	0.00–0.02	0.00–0.03	0.01–0.02	0.00–0.06	0.00–0.11	0.00–0.05	0.00–0.03
Cd	0.02 ± 0.01	0.01 ± 0.01	–	0.01 ± 0.01	0.03 ± 0.03	0.02 ± 0.02	0.00
	0.01–0.03	0.00–0.03		0.00–0.03	0.00–0.09	0.00–0.05	0.00–0.01
Co	–	–	0.00	0.00 ± 0.01	0.00	0.00 ± 0.01	0.00 ± 0.01
			0.00–0.01	0.00–0.02	0.00–0.01	0.00–0.02	0.00–0.02
S	1	1	1	1	1	1	1
P, kbar	2.2 ± 0.1	4 ± 2	4 ± 1	4 ± 2	3 ± 2	5 ± 3	3 ± 2
	2.1–2.3	0.5–8.9	2.6–5.4	1.1–7.6	0.4–9.9	1.8–9.4	0.2–9.5

) at temperature below 500 °C depends on pressure only [59,60,61], we calculate the pressure using the following equation [62]:

FeS_Sp = 20.53 − 1.313P + 0.0271P²,

where FeS_Sp is in mol %, and P is in kbar. The results (see Table 6) shows gradual pressure decrease from 4 kbar in marginal (apatite)-forsterite phoscorite to 2 kbar in axial carbonate-rich phoscorite and phoscorite-related carbonatite, and then again growth to 4 kbar in vein calcite carbonatite, and decrease to 2 kbar in vein dolomite carbonatite (i.e., decrease by 2 kbar from earlier to the latest rock within both pipe and vein series). It is necessary to note similar behavior of oxygen fugacity [26] estimated with the Fe²⁺Ti–Fe³⁺₂ magnetite-ilmenite exchange geothermometer/oxometer of Ghiorso and Evans [63].

In intermediate low-carbonate magnetite-rich and, especially, axial calcite-rich phoscorite and phoscorite-related carbonatite, there are grains of pyrrhotite-6C with rounded inclusions of pyrite (up to 100 µm in diameter, Figure 10d) that can be regarded as products of exsolution of sulfur-rich pyrrhotite-4C and -5C [62,64,65]:

4Fe₇S₈ ↔ 3Fe₉S₁₀ + FeS₂,

5Fe₇S₈ ↔ 4Fe₁₁S₁₂ + FeS₂, or

5Fe₉S₁₀ ↔ Fe₁₁S₁₂ + FeS₂.

In this case, distribution of impurities between the coexisting pyrrhotite and pyrite depends on temperature: with temperature decrease, the impurities undergo redistribution from pyrite into pyrrhotite, and vice versa [66]. Constant presence of cobalt impurity in the Kovdor pyrite-pyrrhotite pairs (

Table 7. Chemical composition of pyrite inclusions in pyrrhotite (mean ± SD/min–max), and temperature of their formation estimated with a Co-exchange pyrite-pyrrhotite geothermometer [66].

Rock	Low-Cb Mag-Rich Phoscorite	Cal-Rich Phoscorite/Carbonatite	Dol-Carbonatite
n	2	8	2
S, wt %	52.4 ± 0.6/51.94–52.80	53.0 ± 0.2/52.76–53.25	53.1 ± 0.2/52.97–53.30
Fe	48 ± 1/47.86–48.65	46.8 ± 0.9/45.52–48.51	46.1 ± 0.2/45.93–46.27
Co	0.4 ± 0.2/0.26–0.49	0.1 ± 0.2/<0.01–0.01	0.8 ± 0.6/0.35–1.21
Ni	<0.01	0.03 ± 0.05/<0.01–0.13	0.02 ± 0.01/0.02–0.03
Fe, apfu	1.05 ± 0.04/1.02–1.08	1.02 ± 0.02/0.98–1.05	1.00 ± 0.01/0.99–1.00
Co	0.01	0.00/0.00–0.01	0.02 ± 0.01/0.01–0.03
S	2	2	2
T, °C	170 ± 20/159–188	300 ± 100/235–444	170 ± 90/106–228

) allows us to estimate temperature of their formation using the Co-Fe exchange pyrite-pyrrhotite geothermometer:

$$T=\frac{1000}{1.907-0.538{\log }_{10}{K}_D^{\mathrm{Co}}}$$

where T is temperature in K, K_D^Co = (Co_at.%/Fe_at.%)_Po : (Co_at.%/Fe_at.%)_Py [66]. Our calculations showed (see Table 7) that temperature of Co equilibration between pyrite and pyrrhotite increases from 170 °C in intermediate low-carbonate magnetite-rich phoscorite to 300 °C in axial carbonate-rich phoscorite, and then decrease to 230 °C in vein dolomite carbonatite.

It is necessarily to note that pyrrhotite-4C may become stable at about 140 °C in nature [67], while 6C and 5C superstructures can be formed at temperatures below approx. 60 °C [62]. Therefore, in natural rock sequence, the ratio of pyrrhotite-5C + 6C to pyrrhotite-4C (see Figure 4a,b) gradually increases due to the gradual decrease of the pyrrhotite superstructure formation temperature. These hypotheses are in good agreement with the results of calcite-dolomite and ilmenite-magnetite geothermometry for this rock sequence [26], where the average temperature of the mineral equilibration gradually decreases from 500 °C to 400 °C for carbonates and from 500 °C to 300 °C for oxides.

Besides exsolution inclusions of troilite, pentlandite–cobaltpentlandite, sphalerite and pyrite, there are unit inclusions of galena, mackinawite, clausthalite, violarite, hessite, moncheite and petzite (

Table 8. Chemical composition of characteristic inclusions in pyrrhotite grains (wt %/apfu).

Mineral	Mck	Vlt	Cls
Sample	K-02-124	74/67.2	73/205.8
Rock	CM-phoscorite	Mag-Dol-Srp rock	Vein Cal-carbonatite
S	35.61/8.00	43.31/4.00	bd
Mn	bd	bd	bd
Fe	58.27/7.52	16.99/0.90	bd
Co	0.94/0.12	13.33/0.67	bd
Ni	5.56/0.68	26.40/1.33	bd
Zn	bd	bd	bd
Se	bd	bd	26.82/1.00
Cd	bd	bd	bd
Pb	bd	bd	73.98/1.05
Total	100.38/16.32	100.03/6.90	100.80/2.05
Mineral	Hss	Mch	Ptz
Sample	931/341.2
Rock	Diopsidite
Ag	56.57/1.97	bd	37.07/2.84
Te	34.04/1.00	44.54/1.54	30.84/2.00
Pd	bd	8.38/0.35	bd
Pt	bd	25.40/0.58	bd
Au	bd	bd	17.93/0.75
Pb	8.91/0.16	bd	14.40/0.58
Bi	bd	21.68/0.46	bd
Total	99.52/3.13	100.00/2.93	100.24/6.17

) in pyrrhotite grains. Mackinawite is a typical mineral for axial calcite-rich phoscorite and phoscorite-related carbonatite, where it forms equant or wedge-shaped inclusions (up to 100 µm long) in pyrrhotite and, especially, chalcopyrite. Violarite is found together with pentlandite in dolomite-magnetite-serpentine rock as lens-like inclusions (up to 30 µm long and 2 µm thick) oriented along the {0001} planes of host pyrrhotite. Both these minerals are not independent from pyrrhotite. Galena forms rounded inclusions (up to 20 µm in diameter) in peripheral parts of pyrrhotite grains and pyrrhotite-chalcopyrite boundaries.

In nepheline-bearing diopsidite, pyrrhotite contains tabular inclusions of moncheite (up to 15 µm in diameter and 4 µm thick), and irregularly shaped polyphase inclusions of hessite, petzite, galena and clausthalite (up to 6 µm in diameter). All these inclusions may result from pyrrhotite self-cleaning from the corresponding impurities during rock cooling.

Under the influence of low-temperature hydrothermal solutions, pyrrhotite-4C (often) and pyrrhotite-5C (rarely) can be replaced by pyrite, marcasite, valleriite, djerfisherite, pyrite and goethite (

Table 9. Chemical composition of characteristic products of pyrrhotite alteration (wt %/apfu).

Mineral	Py	Mrc	Djf	Val	Brt
Sample	1011/79/6	K-96-19-1	1004/656.5	K-051	941/41.5
Rock	AF-phoscorite	Dol-carbonatite	Cal-carbonatite	Dol-carbonatite	MAF-foscorite
Mg	bd	bd	bd	11.08/2.53	bd
Al	bd	bd	bd	0.67/0.14	bd
S	53.05/2.00	53.34/2.00	32.86/26.00	23.16/4.00	13.65/1.00
Cl	bd	bd	1.43/1.02	bd	bd
K	bd	bd	9.33/6.05	bd	bd
Ca	bd	bd	bd	bd	0.37/0.02
Fe	46.19/1.00	46.79/1.01	35.21/15.99	31.81/3.15	bd
Co	0.39/0.01	bd	0.03/0.01	bd	bd
Ni	0.05/0.00	bd	0.05/0.02	2.93/0.28	bd
Cu	bd	bd	21.52/8.59	9.23/0.80	bd
Sr	bd	bd	bd	bd	0.87/0.02
Ba	bd	bd	bd	bd	57.02/0.96
Total	99.68/3.01	100.13/3.01	100.43/57.70	78.88/10.90	71.91/2.00

):

4Po + 10H₂O + 15O₂ = 24Py + 20Ght,

Po + 9Mgs + 9H₂O + Fe²⁺ = 3Val + 9CO₂,

13Po + 6NaCl + 36KCl + 7Fe²⁺ = 6Djf,

Po + 24O₂ + 12Ba²⁺ = 12Brt + 11Fe²⁺, etc.

where Brt is barite, BaSO₄; Djf—djerfisherite, K₆NaFe₂₅S₂₆Cl; Gth—goethite, FeO(OH); Mgs—magnesite, MgCO₃; Po—pyrrhotite, Fe₁₁S₁₂; Py—pyrite, FeS₂ and Val—valleriite, Mg₃Fe₄S₄(OH)₆.

Pyrite and other secondary minerals form partial to complete pseudomorphs after pyrrhotite (Figure 13), which are mostly spread within ring-like stockwork of vein calcite carbonatite and linear zone of vein dolomite carbonatite (see Figure 1), and also in well-shaped crystals in surrounding voids and fissures. In particular, pyrite forms cubic to pentagonal-dodecahedral crystals (up to 20 cm in diameter); marcasite composes sheaf-like aggregates of flattened prismatic crystals (up to 3 mm long); valleriite and tochilinite, 6Fe0.9S·5Mg(OH)₂, occur as twisted platy crystals (up to 4 mm in diameter); and barite forms well shaped prismatic crystals (up to 4 cm long).

In addition, Fe²⁺ and Co²⁺ cations resulted from low-temperature hydrothermal alteration of pyrrhotite react with (PO₄)³⁻ and (CO₃)²⁻ anions, and form late hydrothermal phosphates and carbonates: mitridatite, (Ca_1.72Fe²⁺_0.02)_Σ1.74Fe³⁺_3.10[P_3.04O₁₂]O₂·3H₂O, strengite, Fe³⁺_0.99[P_1.01O₄]·2H₂O, gladiusite, (Fe²⁺_2.02 Mg_1.61Mn_0.06)_Σ3.69Fe³⁺_2.17[P_1.02O₄](OH)₁₁·H₂O, baricite, (Mg_1.75Fe_1.18)_Σ2.93[P_2.03O₈]·7H₂O, vivianite, (Fe_2.88Mg_0.02)_Σ2.90[P_2.04O₈]·8H₂O, pakhomovskyite, (Co_2.38Mg_0.38Mn_0.17Ni_0.04Fe²⁺_0.03)_Σ2.99[P_2.01O₈]·8H₂O, siderite, (Fe_0.97Ca_0.02Mg_0.01)_Σ1.00[CO₃], and pyroaurite (Mg_5.59 Fe²⁺_0.27)_Σ5.86(Fe³⁺_1.92Al_0.08)_2.00(OH)₁₆(CO₃)_0.86·4H₂O [11,33,35,68,69,70]. In particular, pakhomovskyite and gladiusite together with pyrite precipitated on walls of leached fissures in dolomite carbonatite vein (see Figure 5b,c) as a result of the following approximate reaction [35]:

3Po + 32H₂O + 4(PO₄)³⁻ = 18Py + Pkh + 2Gld + 11H₂,

where Po is Co-bearing pyrrhotite, (Fe₁₀Co)S₁₂, Py—pyrite, Pkh—pakhomovskyite, Gld—gladiusite. Both phosphates form small (up to 0.5 mm in diameter) radiated aggregates of elongated plate crystals coloured dark brown (gladiusite) and bright pink (pakhomovskyite).

3.3. Chalcopyrite and Products of Its Alteration

In natural sequence of the Kovdor rocks, content of copper gradually increases from host silicate rocks to earlier (apatite)-forsterite phoscorite, intermediate low-carbonate magnetite-rich phoscorite, and then to the latest calcite-rich phoscorite and carbonatites (see Table 2). On this reason, chalcopyrite content and grain size increase with growth of carbonate amount in the rock (up to 15 modal % and 1 cm in diameter, Figure 14a). In fact, this mineral is absent in peridotite and, rarer, in non-altered foidolite. In hydrothermally altered ijolite-urtite, it occurs as inclusions (up to 40 µm in diameter) in newly formed cancrinite (see Figure 2c) and calcite. Apo-peridotite diopsidite and phlogopitite contain irregularly shaped chalcopyrite grains (up to 1 mm in diameter) associated with pyrrhotite, magnetite and ilmenite in interstices of rock-forming silicates and apatite (see Figure 2d).

Marginal (apatite)-forsterite phoscorite is usually free from chalcopyrite. In intermediate low-carbonate phoscorite, chalcopyrite, together with pyrrhotite, fills interstices between grains of earlier forsterite, phlogopite, hydroxylapatite and magnetite (Figure 14b). Chalcopyrite usually occurs in marginal parts of pyrrhotite-chalcopyrite segregations or forms irregularly shaped gulf-like inclusions growing inside pyrrhotite grains from their margins. It also fills cleavage fractures in phlogopite, impregnates serpentine pseudomorphs after forsterite, accompanies late dolomite veinlets and segregations. In voids of vein dolomite carbonatite, late chalcopyrite forms druses of well-shaped tetrahedral crystals (up to 1 mm in diameter) in association with pyrite, anatase and titanite [11].

In calcite-rich phoscorite and carbonatite, chalcopyrite forms irregularly shaped chalcopyrite grains (up to 1 cm in diameter, see Figure 14a) partially replaced and rimmed by fine-grained pyrrhotite (see Figure 2h), as well as inclusions in magnetite, calcite and dolomite. Сhalcopyrite grains sometimes contain irregularly shaped relicts (up to 15 µm in diameter, Figure 14c) of high bornite (cubic, a = 5.47 Å) accompanied by newly formed covellite [11].

Chemical composition of chalcopyrite varies insignificantly (

Table 10. Chemical composition of chalcopyrite (mean ± SD/min–max).

Rock	Phoscorite and Related Carbonatite			Vein Carbonatite
	(Ap)-Fo	Low-Cb Mag-Rich	Cal-Rich	Cal	Dol
n	2	2	2	2	2
S, wt %	34.1 ± 0.8 33.53–34.70	34.7 ± 0.1 34.60–34.79	35.0 ± 0.9 34.40–35.68	34.9 ± 0.1 34.86–35.00	35.1 ± 0.3 34.83–35.29
Fe	30.6 ± 0.2 30.40–30.73	29.9 ± 0.2 29.76–30.02	30.4 ± 0.3 30.22–30.65	30.1 ± 0.5 29.80–30.50	30.8 ± 0.1 30.74–30.92
Co	<0.01	0.01 ± 0.01 <0.01–0.02	0.03 ± 0.1 0.02–0.04	<0.01	<0.01
Ni	<0.01	0.02 ± 0.02 <0.01–0.04	0.01 ± 0.01 <0.01–0.02	0.01 ± 0.02 <0.01–0.02	0.02 ± 0.02 <0.01–0.03
Cu	35.0 ± 0.3 34.80–35.29	34.29 ± 0.05 34.26–34.32	34.3 ± 0.5 33.93–34.58	34.0 ± 0.3 33.80–34.24	34.5 ± 0.1 34.39–34.53
Zn	<0.01	<0.01	<0.01	0.3 ± 0.3 0.00–0.50	<0.01
Cu, apfu	1.04 ± 0.04 1.01–1.06	1.00	0.99 ± 0.04 0.96–1.01	0.99 ± 0.01 0.98–0.99	0.99 ± 0.01 0.99–1.00
Fe	1.03 ± 0.03 1.01–1.05	0.99	1.00 ± 0.02 0.99–1.01	0.99 ± 0.02 0.98–1.00	1.01 ± 0.01 1.00–1.01
S	2	2	2	2	2

), and its averaged formula corresponds to a theoretical one. Nevertheless, like pyrrhotite, chalcopyrite contains different exsolution inclusions (

Table 11. Chemical composition of characteristic inclusions in chalcopyrite (wt %/apfu).

Mineral	Bn	Cbn	Mck	Sgn	Sp
Sample	KZh-25b	K-97-3	K-97-3	2002/Bn2	K-97-3
Rock	CM-phoscorite	CMF-phoscorite	CMF-phoscorite	CMAF-phoscorite	CMF-phoscorite
S	25.34/4.00	35.57/3.00	35.87/8.00	41.90/4.00	33.94/1.00
Mn	bd	bd	bd	bd	1.59/0.03
Fe	11.82/1.07	41.09/1.99	51.59/6.61	0.57/0.03	13.38/0.23
Co	bd	bd	12.50/1.52	19.72/1.02	bd
Ni	bd	bd	0.40/0.05	35.98/1.88	bd
Cu	62.67/4.99	23.70/1.01	bd	2.12/0.10	0.71/0.01
Zn	bd	bd	bd	bd	46.17/0.67
Cd	bd	bd	bd	bd	4.80/0.04
Total	99.83/10.06	100.36/6.00	100.36/16.18	100.29/7.03	100.59/1.98
Mineral	Hwl	Apn	Acn	Vol	Tsu
Sample	1009/121.4	910/348.2	999/76.6	913/57.1	913/57.1
Rock	MF-phoscorite	Diopsidite	CMAF-phoscorite	CM-phoscorite	CM-phoscorite
S	23.76/1.00	30.78/8.00	12.45/1.00	bd	bd
Fe	3.12/0.08	32.43/4.84	bd	bd	3.58/0.11
Co	bd	0.34/0.05	bd	bd	bd
Ni	bd	23.08/3.28	bd	bd	bd
Cu	6.50/0.14	bd	bd	bd	bd
Zn	8.29/0.17	bd	bd	bd	bd
Ag	bd	13.35/1.03	87.03/2.08	19.82/1.08	bd
Cd	58.11/0.70	bd	bd	bd	bd
Te	bd	bd	bd	43.52/2.00	38.07/1.00
Bi	bd	bd	bd	35.79/1.00	58.42/0.94
Total	99.78/2.09	99.98/17.20	99.48/3.08	99.13/4.08	100.07/2.05

); therefore, we can assume that its initial composition is more complex. In particular, chalcopyrite grains usually carry sharply bounded cross- or star-like exsolution inclusions of sphalerite (up to 100 µm in diameter), as well as its gradually bounded rims around chalcopyrite grains and veinlets within them (up to 20 µm thick, Figure 14d). In axial carbonate-rich phoscorite and phoscorite-related carbonatite, there are exsolution inclusions of cubanite in mackinawite-bearing chalcopyrite (Figure 14e). Cubanite forms brass-yellow, uniformly oriented, long prismatic inclusions (up to 3 mm × 1.5 mm) with split ends, as well as micro-granular aggregates, up to 1 mm in diameter [11,14]. Mackinawite occurs as wedge-shaped grains, skeletal crystals and stellate intergrowths (up to 100 µm in diameter, see Figure 14e), which are abundant in chalcopyrite, but relict or even absent in co-existing cubanite [11].

In addition to the minerals described above, <50 µm sized inclusions of galena, hawleyite, argentopentlandite, siegenite and acanthite were found in chalcopyrite grains [11]. Galena and hawleyite form rounded grains at the contacts between chalcopyrite and pyrrhotite, bornite, covellite, carbonates. Argentopentlandite (Figure 14f) occurs in ijolite, diopsidite and phlogopitite as irregularly shaped inclusions (up to 30 µm) in marginal parts of chalcopyrite grains. Siegenite together with galena, hawleyite, wittichenite and covellite form small (10–70 μm) inclusions in bornite and chalcopyrite pseudomorphs after bornite (see Figure 14c). Irregularly shaped grains of acanthite (up to 20 µm) were found at the contact between pyritized chalcopyrite and calcite.

Besides, chalcopyrite may contain fine impregnation of volynskite, tsumoite and native silver. Volynskite and tsumoite were found in calcite-magnetite phoscorite as thin intergrowth (up to 5 µm in diameter) within chalcopyrite inclusions in magnetite. Native silver forms thin (<1 µm thick) veinlets and xenomorphic inclusions (up to 3 µm in diameter) in chalcopyrite grains, especially, at the contacts between chalcopyrite grains and pyrrhotite, sphalerite.

Typical products of chalcopyrite alteration include valleriite, djerfisherite, chalcocite, covellite, pyrite and goethite (

Table 12. Chemical composition of characteristic products of chalcopyrite alteration (wt %/apfu).

Mineral	Val	Djf	Cct	Cv	Py
Sample	KZh-25b	K-97-3	K-01-1110	K-02-122	K-0042
Rock	CM-phoscorite	CMF-phoscorite	CMAF-phoscorite	CM-phoscorite	Dol-carbonatite
Mg	11.12/2.54	bd	bd	bd	bd
Al	3.57/0.73	bd	bd	bd	bd
S	23.14/4.00	32.91/26.00	19.64/1.00	31.68/1.00	53.20/2.00
Cl	bd	0.63/0.45	bd	bd	bd
K	bd	9.06/5.87	bd	bd	bd
Fe	25.66/2.55	35.55/16.13	bd	1.54/0.03	46.76/1.01
Co	bd	0.06/0.03	bd	bd	0.36/0.01
Ni	0.12/0.01	1.40/0.60	bd	bd	0.02/0.00
Cu	12.06/1.05	16.51/6.58	79.12/2.03	66.30/1.06	bd
Ag	bd	3.34/0.78	bd	bd	bd
Total	75.67/10.88	99.46/56.44	98.76/3.03	99.52/2.09	100.34/3.02

), which replace chalcopyrite grains from margins and fractures (see Figure 14c), up to formation of complete pseudomorphs after chalcopyrite:

16Ccp + 2H₂O + 3O₂ = 12Py + 8Cct + 4Gth,

2Ccp + 3Mgs + 3H₂O = Val + CO₂,

25Ccp + 12KCl + 2NaCl + 2S + Cl₂ = 2Djf, etc.

where Ccp is chalcopyrite, FeCuS₂; Cct—chalcocite, Cu₂S; Djf—djerfisherite, K₆NaFe_12.5Cu_12.5S₂₆Cl; Gth—goethite, FeO(OH); Mgs—magnesite, MgCO₃; Py—pyrite, FeS₂ and Val—valleriite, Mg₃Fe₂Cu₂S₄(OH)₆.

Different low-temperature copper minerals appear also in neighboring voids and fractures. In particular, several funnels of staffelite breccia in apical (now excavated) part of the phoscorite-carbonatite pipe axial zone are formed by fragments of altered chalcopyrite-rich phoscorite and carbonatite cemented by bluish green colloform fluorapatite with inclusions of brochantite, Cu₄[SO₄](OH)₆, malachite, (Cu_1.98Fe_0.02)_2.00[CO₃](OH)₂, pseudomalachite, (Cu_4.85Ca_0.07)_4.92[P_2.03O₈](OH)₄, chrisocolla, (Cu_1.85Ca_0.03)_1.88H_2.00 Si_2.06O₅(OH)₄·nH₂O, tenorite, Cu_1.00O, and goethite (Fe_0.86Si_0.11)_0.97O(OH) [11].

4. Discussion

The Kovdor phoscorite-carbonatite complex has gradual concentric zonation in terms of the rocks modal and chemical composition, as well as grain size, chemical composition, crystallochemical features and properties of rock-forming and accessory minerals [11,21,22,24,26,27,28,29]. This zonation mainly results from gradual change in chemical composition of the residual phoscorite melt from silicate-rich to carbonate-rich during crystallization from the pipe margins to its axial zone [22].

Geochemically, nickel behaves like Mg and Fe²⁺, and readily substitutes them in most crystalline phases including olivine [71]. Besides, there is a strong negative dependence of Ni activity on FeO_tot content in silicate melt [72,73]. However, even in a high-magnesian system, like the Kovdor peridotite and forsteritite, nickel can remain in the residual silicate melt (SM) because its partition coefficient D^Ol/SM depends significantly on sulfur content [74]. In fact, at fixed P–T conditions, the partition coefficients of chalcophile elements M between monosulfide solid solution (MSS), (Fe,Ni,Co)_1-xS, and silicate melt are controlled by the ratio fS₂/fO₂, which, in turn, depends on FeO_tot content in sulfide-saturated silicate melt [75,76,77]:

(Fe,M)O_(SM) + 0.5S₂ = (Fe,M)S_(MSS) + 0.5O₂,

MO_(SM) + FeS_(MMS) = MS_(MSS) + FeO_(SM).

As a result, the MSS–SM partition coefficients for Ni, Cu and Co substantially increase with FeO_tot decrease in silicate melt (Figure 15a), and we can conclude that crystallization of Mg-rich (apatite)-forsterite phoscorite can produce Fe-Ni-Co-Cu-rich sulfide melt.

In earlier low-carbonate phoscorite, pyrrhotite is formed at the final stage of magnetite crystallization, because comparatively low sulfur content (see Table 2) still requires melt saturation in FeS complexes. For this reason, ferrimagnetic pyrrhotite-4C (or the corresponding MSS) first closely associates with magnetite forming gulf-like ingrowths inside its grains. Then, with transition to carbonate-rich phoscorite and carbonatite, iron-rich pyrrhotite-5-6C (or the corresponding MSS) crystallizes independently of magnetite. Increase of mean equivalent circular diameter of pyrrhotite grains reflects a decrease of the rock crystallization rate from the pipe wall to its axis. This trend is similar to that of co-existing rock-forming and accessory minerals, in particular, forsterite, magnetite and baddeleyite [22,26,27].

Our data on crystal structure of the Kovdor pyrrhotites are consistent with numerous previous studies of the structure [48,51,54,55,56,57,78], which demonstrate different space groups for the most common 4C, 5C and 6C polytypes. Such structural diversity is caused by different P-T conditions of crystallization, amount of incorporated Fe³⁺ or different mechanisms of vacancies distribution in the structure. According to the structural complexity classification [79,80], 4C polytype is “simple”, while 5C and 6C modifications are “intermediate” (Figure 16). Relatively low-temperature formation of pyrrhotite-5C in comparison with pyrrhotite-4C is in good agreement with a common tendency of structural complexity growth with crystallization temperature decrease. From this point of view, use of the term ‘pyrrhotite-nC’ for 5C and 6C polytypes is justified by the absence of significant differences between their complexity.

Decrease of monosulfide crystallization temperature causes growth of D^MSS/SM for Cu and Co (Figure 15b), while Ni partition coefficient remains the same or even decreases [77,81], which enables fractionation of Ni and Cu-Co between low- and high-temperature sulfides respectively. Therefore, we believe that concentrically zoned distribution of pyrrhotite with different Fe, Ni and Co contents is a response to a gradual decrease of crystallization temperature from the ore-pipe marginal forsterite-dominant zone to its axial carbonate-rich zone. Exsolution of MSS or the corresponding higher-temperature pyrrhotite modifications produced lamella or flames of pentlandite in the ore-pipe marginal zone, and cobaltpentlandite in the pipe axial zone.

For the same reason, copper is concentrated in the pipe axial carbonate-rich zone. Carbonate melt appears to contain a significant amount of water [82] and copper as stable hydrosulfide complexes [83]. Their destruction by bornite and chalcopyrite crystallization after carbonates caused intensive hydrothermal alteration of associated minerals, and formation of secondary pyrite, valleriite and other phases.

It is interesting to compare sulfide mineralization of the phoscorite-carbonatite complexes of the Kovdor (Russia) and Palabora (South Africa) alkaline-ultrabasic massifs that have comparable composition, structure and metallogenic specialization [12]. However, the Loolekop phoscorite-carbonatite pipe in the Palabora massif is the largest carbonatite-hosted copper deposit in the world 850 Mt @ 0.5% Cu [84], and the Kovdor phoscorite-carbonatite complex has comparatively low copper content. What caused such a significant enrichment of the Loolekop phoscorite and carbonatites in copper?

The Loolekop phoscorite-carbonatite pipe is a close relative to the Kovdor one. The Loolekop pipe (1.4 km × 0.8 km) is situated in the eastern part of the Proterozoic central-type Palabora complex of apo-peridotite serpentinite (central stock), shonkinite (outer ring intrusion) and the related diopsidite and phlogopitite between them. It has a clear concentric zonation with marginal low-carbonate phoscorite, intermediate “banded” zone of interlayered carbonate-rich phoscorite and phoscorite-related calcite carbonatite and axial stockwork of “transgressive” calcite and dolomite carbonatite [85,86,87].

All these rocks are rich in pyrrhotite with exsolution inclusions of pentlandite–cobaltpentlandite [87], and contain also sufficient amount of copper sulfides: dominant bornite and chalcocite in phoscorite and banded carbonatite and dominant chalcopyrite with relicts of bornite and exsolution inclusions of cubanite in transgressive carbonatite [85,86,88,89]. Minor copper minerals include covellite, valleriite, tetrahedrite, and native copper [12,85,90]. As a result, copper content increases from <0.3 wt % in marginal phoscorite to 0.3–0.9 wt % in intermediate banded carbonatite and about 1 wt % in axial transgressive carbonatite [85,86].

In phoscorite and phoscorite-related “banded” carbonatite, sulfides fill interstices and thin fractures in magnetite-apatite-forsterite-calcite aggregate, and form irregularly shaped segregations with pyrrhotite in core and copper-bearing sulfides in marginal zone, while in the late “transgressive” carbonatite, they form subparallel vertical lenses and veinlets (usually, up to 3 cm thick) within a vertical ore-zone with cross-section about 200 m × 600 m [85,87]. In both rocks, pyrrhotite and pentlandite–cobaltpentlandite are the earliest sulfides, high bornite is the next one, and chalcopyrite and other minerals follow them.

Thus, the Kovdor and Loolekop phoscorite-carbonatite pipes have similar sequences of sulfide formation (Figure 17a), from earlier pyrrhotite to intermediate copper sulfides and late pyrite and valleriite. There are only differences in positions of argentopentlandite and mackinawite that are common products of chalcopyrite exsolution in the Kovdor complex, and substitute pentlandite in the Loolekop pipe [12,87,89]. It is necessary to note that secondary djerfisherite is widely spread in the Kovdor complex due to higher alkalinity of this massif.

The sequences of sulfide formation in the phoscorite-carbonatite pipes are in a good accordance with the upper limits of their stability [62] that linearly decrease from 1100 °C for high bornite to 103 °C for chalcocite (Figure 17b). For the Kovdor complex, these relations can be supplemented with temperature of pyrite-pyrrhotite equilibration ranged from 170 °C in intermediate low-carbonate magnetite-rich phoscorite and dolomite carbonatite to 300 °C in axial carbonate-rich phoscorite and phoscorite-related carbonatite, and temperatures of the pyrrhotite superstructure formation from 140 °C for pyrrhotite-4C to 60 °C for pyrrhotite-5C [62].

One of the most likely reasons for sulfide specialization in the Loolekop phoscorite-carbonatite pipe in comparison with the Kovdor one is insignificant differences in oxygen fugacity of initial phoscorite melts (due to higher water content), because the larger oxygen fugacity causes more intensive copper extraction by melt from surrounding rocks [5,91,92,93]. In fact, even a slight increase in oxidation state (to ≥FMQ + 1) can shift the balance between sulfide and sulfate components dissolved in melt towards more soluble sulfates, allowing magmas to dissolve much larger amounts of sulfur than reduced melts [94]. Higher water content in the Palabora’s magma is confirmed, in particular, by much deeper alteration of peridotite and phoscorite into serpentinite, much abundant pegmatites, much richer valleriite mineralization, etc. [85,90,95].

5. Conclusions

• Primary silicate rocks of the Kovdor massif (peridotite, foidolite–melilitolite and forsteritite) are free of sulfides due to sulfur fractionation in fluid phase. For the same reason, hydrothermally altered parts of these rocks, including diopsidite, phlogopitite and skarn-like rocks, carry rich sulfide mineralization associated with secondary minerals (cancrinite, natrolite, phlogopite, clinochlore, serpentine, vesuvianite, etc.).
• Distribution of rock-forming sulfides within the Kovdor phoscorite-carbonatite complex reflects gradual concentric zonation of the pipe: pyrrhotite with exsolution inclusions of pentlandite in marginal (apatite)-forsterite phoscorite, pyrrhotite with exsolution inclusions of cobaltpentlandite in intermediate low-carbonate magnetite-rich phoscorite, and chalcopyrite (±pentlandite with exsolution inclusions of cobaltpentlandite) in axial carbonate-rich phoscorite and phoscorite-related carbonatite;
• Both pyrrhotite and chalcopyrite fill interstices between the main rock-forming minerals, including carbonates, and form late veinlets and irregularly shaped segregations with inclusions of surrounding minerals. Usually, chalcopyrite (with relicts of earlier bornite and exsolution inclusions of cubanite and mackinawite) crystallizes around grains of pyrrhotite (with inclusions of pentlandite-cobaltpentlandite and pyrite), and both these minerals contain common exsolution inclusions of sphalerite. Temperature of pyrite-pyrrhotite equilibration ranges from 100 °C to 400 °C, and pressure of pyrrhotite-sphalerite equilibration reaches 10 kbar;
• For the most part, pyrrhotite corresponds to its non-magnetic 5C polytype. Ferrimagnetic pyrrhotite-4C forms individual grains, marginal zones of non-magnetic pyrrhotite crystals and thin lens-like inclusions in pyrrhotite-5C. Low-temperature pyrrhotite 2C (troilite) occurs as lens-like exsolution inclusions in grains of pyrrhotite-4C and -5C. In natural sequence of the Kovdor rocks, iron content in pyrrhotite gradually increases from Fe₇S₈ to Fe₉S₁₀ and Fe₁₁S₁₂ in accordance with gradual decrease of crystallization temperature and oxygen fugacity.
• High complexity of crystal structure of pyrrhotite-5C (I_G,total = 429.75 bits per unit cell), which is the most common within the Kovdor massif, confirms the lowest temperature of its formation in comparison with other polytypes. Wide dissemination of this modification is associated with structural stability (variations in composition Fe_8.84–8.99S₁₀) due to different vacancies ordering mechanisms resulting in two structural modifications: P2₁ and C2.
• Within the phoscorite-carbonatite complex, content of Co in pyrrhotite gradually increases from host silicate rocks and marginal forsterite-dominant phoscorite to axial carbonate-rich phoscorite and carbonatite due to Ni and Fe. This dependence probably reflects a gradual decrease of crystallyzation temperature in primary monosulfide solid solutions from the pipe margin toward its axis.
• Low-temperature hydrothermal alteration of pyrrhotite and chalcopyrite first produces partial or complete pseudomorphs of djerfisherite, pyrite, chalcocite, valleriite and goethite after the rock-forming sulfides, and then numerous phosphates and carbonates of Fe, Co and Cu (baricite, gladiusite, malachite, mitridatite, pakhomovskyite, pseudomalachite, pyroaurite, siderite, strengite, vivianite) in surrounding voids and fissures.
• The Kovdor and Loolekop phoscorite-carbonatite pipes have similar sequences of sulfide formation: pyrrhotite–pentlandite–cobaltpentlandite–bornite–chalcopyrite–covellite–cubanite–sphalerite–pyrite-chalcocite–galena–vallereite. This sequence corresponds to the sulfide stability limits ranged from 1100 °C (high bornite) to 103 °C (chalcocite) and even lower (for valleriite). Thus, copper specialization of the Loolekop pipe is not caused by any specific process but reflects stochastic specifics of the phoscorite-carbonatite genesis.
• Sulfide specialization of the Palabora massif can be caused by higher water content in its initial melt allowing it to dissolve much larger amounts of sulfur and, consequently, concentrate more chalcophile metals.