Work on a number of PGE-relevant alloy-sulfide phase systems has been completed or is close to completion. Experiments were carried out by reacting pure elements in evacuated silica glass tubes at desired temperatures for extended reaction times with one homogenization.
Solubility of PGE in Fe1-xS (resp. mss) and in the Fe-S based sulfide melt is very important for the geochemistry of PGE. Detailed experiments show that 1.1 at.%Pt and 5.8 at.% Ir dissolve in S-rich Fe1-xS and 8.5 at.%Pt in the S-rich sulfide melt at 1100°C. The experiments in the system Fe-Pt-Ir-S allowed the first exact delineation of the composition ranges of Fe-Pt and Fe-Ir-Pt alloys. Pt3Fe partakes in all important phase assemblages justifying its abundance in PGE deposits but at the same time being of little use in analysing the mineral associations in them.
Several instances of liquid immiscibility were found in the PGE-based systems: (1) the Rh-rich and Rh-poor sulfide liquids in the system Cu-Fe-Rh-S at 1000° and 1100°C; (2) Pd-rich and Pd-poor sulfide liquids in the Cu-rich portions of the system Cu-Fe-Pd-S at 1000°C; (3) the Cu-rich and the Pd-enriched liquid in the system Cu-Pd-S at 725°C. Interesting are also the As-rich and As-poor liquid in the system Ni-Au-As at 950°C (with F. Gervilla). Some unexpected PGE partitioning values can result from such phenomena.
The phase systems Fe-Os-S and Fe-Re-S display at 900°-1200°C a range of alloys (determined exactly for the first time), disulfides of Os and Re and Fe1-xS resp. the Fe-S melt with low solubility of Os and Re. Interestingly, the iron-rich sulfide melts do not dissolve appreciable amounts of Os, Re, Ir and Pt and these elements partition strongly into the respective alloys. Does this influence the core/mantle boundary or is the situation entirely altered by the high pressures?
The solid solubility of PGE and Au in the Fe, Cu sulfides is a sensitive function of sulfur fugacity. In bornite, up to 1.8 at.%Au and 1.2 at.%Pd dissolve at elevated S fugacities at 1000°C. The solubility of these elements is close to nil for the metal-excess bornite compositions. The solubility trends for PGE in Fe1-xS and changes of partition coefficients Fe1-xS/melt for changing sulfur concentrations have been refined for several PGE-containing systems at several important temperatures.
All these experiments may not offer explanation for the origin and preconcentration of PGE in the deposits, but they help to understand the processes and the conditions of initial PGE fixation and their further mobility in the deposits. It is important to get data for a broad range of temperatures comprising both the magmatic and the postmagmatic processes. Surprises are still possible, e.g. the appearance of a very extensive high-T solid solution (Ni, Fe)3±xS2 (Fedorova & Sinyakova 1993, Makovicky & Karup-Møller 1995) which is a PGE collector of primary importance (2 at.%Pd at 725°C) before its pentlandite-yielding decomposition at about 610°C.
I have reinvestigated the system Fe-Ni-Cu-S to determine the solubility of PGE in magmatic sulfide liquids as a function of S fugacity. The principal aim was to find out if the sulfide fraction in the Merensky reef was oversaturated with respect to PGE at magmatic temperatures. Micro-inclusions of PGE phases in natural sulfides do suggest that this may have been the case (Ballhaus and Sylvester 1998). In the course of these investigations I found an extensive, previously unreported miscibility gap the Fe-Ni-Cu-S system at superliquidus conditions that may have profound effects on the differentiation paths of magmatic sulfide liquids.
Experiments were carried out in evacuated SiO2-glass capsules. Starting mixtures are from stoichiometric FeS, Cu2S, and NiS. S fugacity, i.e. the metal/S ratio of the bulk charge, was varied by adding appropriate amounts of elemental S. Any oxygen or water present in the charge was captured by adding a trace of metallic Si. Run temperatures were from 1000 to 1050°C, monitored by placing a cromel-alumel thermocouple alongside the charge.
At 1000°C the run product contains crystalline mss coexisting with a Cu-Ni rich melt. At about 1020°C the product consists of two liquids, one of which is dominated by Cu-Ni (M/S ~ 1.04) and the other by Fe±Ni±Cu (M/S ~ 1.2), dispersed as perfectly rounded droplets in each other. The width of the solvus appears to depend critically on the metal/S ratio (S fugacity) of the bulk charge as well as on temperature. At bulk metal/S~1 the miscibility gap disappears at all temperatures and the product is a homogeneous quenched monosulfide whose composition resembles exactly that of the starting mix. At 1050°C the persistence of the solvus is difficult to prove on textural grounds because at this temperature the Fe-S rich melt is well above the liquidus of mss and quench modification is extensive. Ideal conditions are (1) quench rates as large as possible, and (2) run temperatures just above the solidus of mss to keep quench modification of at least one melt fraction to a minimum.
Conjugate average melt compositions for one experiment at 1020°C are given below. I anticipate that the miscibility gap widens the more Cu is added, since Cu is the most siderophile element in the charge. It seems feasible that highly reduced, differentiating sulfide melts may be able to circumvent fractionation of iss by physical separation of two liquids, hence evolve to Cu/Fe and Cu/Ni ratios unusual for most magmatic sulfide deposits.
Fe 60.3 47.5, Ni 1.6 10.1, Cu 1.6 8.7, S 34.7 30.8
Sum 98.2 97.2, M/S 1.05 1.21
Ballhaus C, Sylvester P, Inst. Mining Metall. Symp. Series, S18, 25-28, (1998).
The conception of genesis of PGE-bearing sulphide deposits as a part of large-scale process of fractional crystallization of melts in layered mafic-ultramafic massifs is a very popular. In particular, it was included into discussion about genesis of both intrusions and deposits of Norilsk ore region (Distler et al., 1979), Sudbury (Naldrett, 1981, Naldrett and Barnes, 1986) etc. Ioko-Dovyren massif, located in North Baikal region, demonstrates typical features of layered plutons, described previously by a number of authors and recently by Kislov (1998). There are two types of Cu-Ni suphide mineralization, including massive and disseminated ores as well as low-sulphide PGE-containing stratified layers (reefs) in the upper part of the section (Orsoev et al., 1995 etc). Two various sulphide paragenesises are developed in both ore types. The first is enriched in Fe-suplhides (troilite, hexagonal pyrrhotite, pentlandite with low Ni cocnent) but with rather limited amount of Cu-sulphides (chalcopyrite and others). The second association differs by the high chalcopyrite percentage and by the presence of cubanite, talnakhite, bornite, Ni-pentlandite and other Ni-sulphides. Other authors for more PGE-bearing sulphide deposits have distinguished similar types of sulphide ores too.
The experimental data, carried out by Kolonin et al. (1993, 1994, 1998), Sinyakova et al. (1996, 1998), Peregoedova et al. (1996) give a chance to estimate trends of sulfur fugacity during formation of both ore types and to estimate possible mineral forms of the PGE mineralization there. We believe that base metal sulphides (BSM) play a role of mineral buffer assemblages to sustain sulphur fugacity. As a principal parameter of the system they predetermine both conditions of precipitation and stable forms of PGE mineralization. In particular, just after pentlandite appearance at 600-570°C, the presence of the mentioned BSM minerals in paragenesis shows the following expected values of -logfS2 and probable PGE forms: Fe-pentlandite Ni-pentlandite Cu-pentlandite Chalcopyrite+other Cu sulphides 11.5-8 up to 6.5 up to 6 >6 tetraferroplatinum Pd and Rh in pentlandite cooperite, vysotskite, braggite, tulameenite khaeralakhite, malaniteThe minerals observed at Ioko-Dovyren are underlined. As to rustenburgite found there (isoferroplatinum is absent), its presence could be explained by a rather high sulfur fugacity in the Sn-containing ore-forming system, as Fe-S chemical affinity is higher in comparison with Sn-S (Mills, 1974), i.e. by shift to right the equilibrium Pt3Fe + SnS = Pt3Sn + FeS.
This research was supported by the Russian-French project through the grant of RFBR-CNRS N 98-05-22020 and by the program "Universities of Russia", the project "Crystalgenesis", grant N 3H-325-98.
Pentlandite (Fe,Ni)9S8 is the dominant PGE-bearing base metal sulphide in platinum-copper-nickel deposits. The experimental study of phase relations in the plane Me9S8 of the system Cu-Fe-Ni-S at temperatures in the range 760-400°C was undertaken to determine the influence of copper on the composition and conditions of formation of pentlandite (Pn). Minor amounts of Pd and Rh (< 1 at.%) were then added in order to investigate the variation of PGE-content of pentlandite as a function of bulk composition and base metal sulphide associations. Sulphur fugacities were systematically estimated using pyrrhotite as an indicator.
Two kinds of pentlandite can be distinguished: (1) pentlandite formed at temperatures in the range 613-574°C by a reaction between a monosulphide solid solution (Mss) and a quaternary one extended from heazlewoodite solid solution (Hzss) to intermediate solid solution (Iss) from the ternary Fe-Ni-S and Cu-Fe-S systems; (2) pentlandite arising from low-temperature subsolidus transformations of sulphide solid solutions (Hzss, Ni-Iss or Mss).
The former generally contain less than 1.3 at.% of Pd and Rh and show a strong positive correlation between PGE-content and Ni-content. Both increase with increasing (Ni+Cu)/Fe ratio and sulphur fugacity. Secondary pentlandite is found to be richer in light PGE. At 550°C maximum Pd and Rh contents are around 4 at.% in Ni-rich pentlandite exsolved on quenching from PGE-containing heazlewoodite solid solution stable under the conditions of higher sulphur fugacity (lgfS2=-7) in the Cu-rich samples. Pentlandite is associated with vysotskite (Pd,Ni)S and intermediate solid solution having a Cu/Fe ratio corresponding to the composition of chalcopyrite.
A similar dependence of PGE content of pentlandite on the mineral assemblage has been observed in some disseminated PGE ores. In the Munni-Munni layered complex, Australia, pentlandite extremely rich in palladium (4 - 5 at.%) has been found in association with vysotskite in the enriched copper ores, while Pd-poor pentlandite is present in Fe-rich assemblages containing sperrylite (PtAs2).
This study was supported by joint Russian-French grant RFBR-CNRS 98-05-22020.
The most spread in nature Platinum-Group Minerals (PGM) belong to the system Pt-Fe-S: (Pt,Fe) natural alloys, isoferroplatinum (Pt3Fe), tetraferroplatinum (PtFe), cooperite (PtS). They are found in different geological occurrences and can be formed under different temperature and physico-chemical conditions. The series of mineral transformation when the oxidation degree increases corresponds to "PtFe3" -> PtFe -> Pt3Fe -> (Pt,Fe)/PtS. The different reactions result in same phase. For example, Pt3Fe can be formed during simple oxidation of PtFe: 9PtFe +4O2 = 3Pt3Fe + 2Fe3O4. (1) Another possibility is the reaction with Fe-sulphide under reducing conditions: 3PtFe + 7FeS = Pt3Fe + Fe6S7 + 3Fe. (2) Cooperite can be produced at the higher initial ratio FeS:PtFe: PtFe + 8FeS = PtS + Fe6S7 + 3Fe. (3) When the oxygen is includes in the reaction sulfidisation the final association contains magnetite: 3PtFe + 7FeS + 2O2 = Pt3Fe + Fe6S7 + Fe3O4 (4) and PtFe + 8FeS + 2O2 = PtS + Fe6S7 + Fe3O4. (5) Note, that the formation of isoferroplatinum and "sulfidisation" of PtFe (with the formation of PtS takes place at the same proportion of PtFe and FeS with and without O2 (reactions 2,3 and 4,5). The association PtS + Pt3Fe is "intermediate" result of not completely accomplished reaction: 4PtFe + 9FeS + 4O2 = Pt3Fe + 2PtS + Fe6S7 + 2Fe3O4 (6) or 5PtFe + 2FeS + 4O2 = Pt3Fe + 2PtS + 2Fe3O4 (6a) Based on these reactions natural assemblages from Noril'sk copper-nickel sulphide ores and some other occurrences are discussed using experimental data and calculated thermodynamic parameters. According to the analysis of Pt-Fe-S- phase assemblages in nature and experiments it is possible to propose the composition of Fe-S phases associated with different PGM for estimate the sulphur fugacity for determine the position of PGM in the field of "pyrrhotite" stability. This field includes the most important natural sulphide equilibria. Using such approach it is possible to compare the conditions of PGM formation in some natural occurrences.
Typical Fe-Ni-Cu sulfide deposits of the Sudbury Structure (1.85 Ga) formed by coalescence of sulfide-magma droplets in the Contact Sublayer and Footwall Breccia that underlie noritic rocks of the Sudbury Igneous Complex (SIC), and in "offset" quartz-diorite dikes that extend into the Archean and Proterozoic footwall rocks. Cu-Ni-PGE-enriched deposits occur as veins, disseminations, and massive ore in the footwall to the SIC and near its contact with sublayer and footwall breccias. The veins are rarely physically connected to typical Sudbury Ni ores. Cu-rich ore might result from fractionation of sulfide magma; however, it commonly is enclosed by hydrothermal alteration assemblages, and both silicate and sulfide minerals contain extremely saline fluid inclusions. The PGE, Au, Ag-enriched footwall ores are Cu, Ni-rich, Fe-poor sulfides that are intergrown with, and enveloped by calcic mineral assemblages involving epidote, actinolite, garnet, titanite, biotite, stilpnomelane, pyrosmalite, chlorite, quartz and various carbonates. PGM are commonly found in chalcopyrite or enclosed by hydrous silicates. Fluid-inclusion studies of quartz from footwall deposits reveal multiple trapping events at about 400-100°C; some fluids precipitated polyphase Ca-Na-K-Pb-Ba-Fe-Mn chloride daughter-mineral assemblages. These fluid inclusions also occur in sulfide minerals. The range of fluid compositions is consistent, but there are deposit-specific chemical signatures, reflected in deposit-specific assemblages of the precious-metal minerals. Stable-isotope (O,H) data from alteration minerals are compatible with fluid derivation by mixing of formational brines and fluids that had equilibrated at high temperature with igneous and metamorphic rocks. These saline fluids, similar to present-day formational brines residing in the Archean basement rocks of the Sudbury Structure, have compositions appropriate for dissolution of significant quantities of PGE, especially under conditions of low pH at high Eh and T. During emplacement of igneous breccias, and crystallization of immiscible sulfide and silicate magma at Sudbury, a hydrothermal system was established under the SIC. Formational brines entered permeable units (impact and igneous breccias) and interacted with igneous rocks and their Fe-Ni-Cu sulfide deposits. Base and precious metals, remobilized by chloride complexes, migrated along and near the sublayer contact in permeable rocks. The brines precipitated chalcopyrite, pentlandite, millerite, bornite, pyrite, and other base- and precious-metal minerals as a result of fluid-mixing, decrease in T and Eh, and increase in pH. This model has application around the whole Sudbury structure, as well as in the footwall rocks of other magmatic sulfide occurrences.
The presence of two types of PGE-containing sulphide mineralization at the Ioko-Dovyren dunite-troctolite-massif allows to consider it as model object for research of noble metals behaviour in ore-magmatic process. It is PGE-bearing intrusion with the precisely expressed tendency of magmatite sulphides accumulation at the Lower border zone. PGE-Ni-Cu sulphide mineralization is presented by syngenetic disseminated and net-textured, epigenetic massive (vein) types. The low-sulphide PGE mineralization is located at the anorthosites of Gabbroic zone.
The analysis of general trends of PGE, gold and silver distribution between various parts of cross-section and types of mineralization was made. The Ni-Cu sulphide mineralization are characterised by smaller contents of noble metals on the 100% sulphide basis (Pt - 0.11-4.8; Pd - 0.48-3.40; Rh - 0.012-0.61; Ru - 0.002-0.170; Ir - 0.002-0.122; Os - 0.003-0.122; Au - 0.37-94.75; Ag - 2.8-13.4 ppm). The low sulphide PGE mineralization differs: Pt - 3.94-458.15; Pd - 8.02-1109.72; Rh - 0.35 33.66 ; Ru - 0.05-12.38; Ir - 0.11-7.87; Os - 0.07-2.50; Au - 3.31-72.05; Ag - 10.8-60.0 ppm.
The Ni-Cu sulphide mineralization is similar with sulphide ores, related to the comatiite. The higher concentration of PGE, gold and silver in sulphide phase of disseminated mineralization in comparison with massive that are caused by early sulphides fractionation from not depleted in these components melt and change of physical-chemical conditions to the increasing of fugacity of oxygen at the advanced stages of magma crystallization and sulphide concentration.
The increase of Pt, Au and Ag ratio at the general balance of noble metals in the massif cross-section is explained by their primary redistribution into residual melt, enriched by fluid phase, and the carry from the Lower border zone of the chamber. The Pd accumulated here mainly at the sulphide melt during sulphide-silicate differentiation, because of more chalcophilic properties, and siderophilic that at the reducing conditions, than Pt, Au and Ag. The increased level of noble metals concentration at the all types of ores testifies about PGE-bearing of initial melt, and, hence, Ioko-Dovyren complex as a whole.
This work was supported by Russian Foundation of the Basic Researches 98-05-65309 and joint RFBR-CNRS program 98-05-22020.
Workable Ni-Cu-Fe sulfide deposits containing PGE-Au-Ag mineralization associate with ultramafic pipe-like bodies crosscutting the Main Gabbro unit and the roof amphibolitic metasediments of the Ivrea-Verbano Complex. Single zircon grain geochronology places intrusion of the pipes at an average age of 288± 4 M.a.. Compared with N-MORB the parental magma to the pipes has picrite-type major element characteristics, although alkaline affinity is supported by depletion in Si and HREE, high K/Na and sensible anomalies in Rb, LREE, Sr, Zr, Ba, U, Th, Pb, Cu, Zn, Ag and Mo, along with high concentrations of volatiles (H2O, CO2, F, Cl, P2O5). Consistently the main mineral assemblage (olivine, pyroxenes, spinel, plagioclase) is accompanied by abundant amphibole, phlogopite, ilmenite, rutile, hematite, magnetite, and accessory F-Cl rich apatite, zircon, baddeleyite, titanite, primary carbonates and barite.
The sulfur isotopic composition of the pipes (34S from 0.00 to +2.00) is indicative of mantle source. Sulfides segregated as immiscible liquid from the volatile-rich, ultramafic magma, giving rise to the assemblage pyrrhotite, pentlandite, chalcopyrite, with minor pyrite, mackinawite, cubanite, and molybdenite. Accessory cobaltite, gersdorffite, melonite, hessite, pilsenite, irarsite and electrum are the major carriers of noble metals. Bulk sulfide displays sensible Cu enrichment (Cu/(Cu+Ni) = 0.18-0.54). Except for Pt and Pd, PGE concentrations are below chondritic. Normalized PGE patterns are fairly unfractionated having (Pt+Pd)/(Ru+Os+Ir) of 3.7-8.67 intermediate between sulfide ores in komatiites and layered intrusions, but exhibit extreme Au-enrichment up to more than fifty times CC1.
Petrology of the pipes displays contrasting aspects that cannot be explained by partial melting of primitive mantle. The ultramafic nature versus high concentrations of incompatibles, volatiles and MARID-IRPS type accessory minerals, and the almost unfractionated PGE-patterns versus the Cu-Au enrichment might indicate origin of the pipe magma by second-stage melting of a depleted mantle source, metasomatically re-enriched by influx of volatile-rich alkalic fluids. Geochronology and <epsilon>Sr-<epsilon>Nd290 isotope data suggest possible genetic relationships with the metasomatic event in the residual mantle peridotite of Finero. Accordingly, fluxed harzburgites and chromitites from Finero have metallogenic signature similar to the pipes, characterized by unusual Au-enrichment of the PGE-patterns, while PGM inclusions in chromitites contain sensible amounts of Ag and Pb, and associate with of Ni-Cu-Mo sulfides.
Intrusion into the deep crust of ultramafic alkaline magmas carrying Ni-Cu sulfide and PGE-Au-Ag mineralization is possibly related with a large-scale metasomatic episode that affected the deep mantle below the Ivrea Zone in Upper-Carboniferous ages.
The largest Pt-Cu-Ni ore deposits of Russia are linked with zones of the midcontinental riftogenesis. Among them are large and unique the early Mesozoic (0.23-0.22 Ga) deposits of the Noril'sk ore region, which are controlled by complicate triple-juncted rift system. The late Palaeoproterozoic (2.1.-1.8 Ga) intrusions of the Pechenga and East-Pechenga ore fields, are linked with the Pechenga-Varzuga rift belt, and the early Palaeoproterozoic (c.2.5 Ga) Monchegorsk, Fedorovopansky and Imandra layered complexes, also linked with Pechenga-Varzuga Belt. The latter type of the layered intrusions with PGE-Cu-Ni mineralization was also linked with the early Palaeoproterozoic rifts within the Karelian craton (Burakovsky, Lukkulaisvaara, Penikat, etc. massifs). All these intrusions form the huge Kola-Karelian platineferous province owes to its origin to midcontinental riftogenic processes. The main feature of the ore-forming systems of these regions is their prolonged (more then 300 Ma) multistage development as it follows from analysis of geological-petrological data. There are reasons to think that geological structures which control of ore fields and ore deposits of these regions were regenerated over and over again along faults, inherited from the Paleoproterozic and the late Riphean (in case of the Noril'sk region) riftogenic belts. As this takes place in the each regions, the ore deposits contain massive, dissiminated, essential cuprous and low-sulfur mineralization. There are ground to think that the main reasons for formation of large and unique ore deposits of the Noril'sk and the Pechenga riftogenic strictures was a deep-seated liquation of an ore-silicate material under condition of a high concentration of fluide components, which played essential role during formation of the different types of the platinum-copper-nickel ores. Also very important is a specialization of ultrabasic-basic melts for Pt, Pd, Cr, Ni, and K. So, in the areas of the midcontinental riftogenesis, which owe to their origin of ascending of mantle diapirs both in the early Precambrian and in the Phanerozoic, occurred large ore-forming systems in which concentration of typical for the upper mantle ore components (Ni, Cu, PGE and Cr) took place.
Numerical modelling of freezing magma chambers along with experience in other disciplines (e.g., industry, limnology, etc) dealing with suspended loads in fluids suggests that the growth of suspended crystal load in the melt (which increases the effective density) serves to split the magma chamber up into a number of stratifications as it evolves. The thicknesses and composition of these stratifications are amenable to quantitative estimates and indicate the following broad trend: the lowermost layers are closest to the original composition of the magma but enhanced in refractories. The middle layers are of a more evolved composition and significantly contaminated as the chamber by this time has been emplaced long enough to bring country rock to temperature which enhances assimilation. The uppermost stratifications are the most evolved but by the time they are formed, convection has died down, hence possessa contamination level between that of the lower and intermediate layers. These processes apparently track trends seen in the field such as Sr isotope ratios and place the majority of the sulfides near the bottom. Double diffusive effects and convective scavenging also attend to provide the fine detail of large layered intrusives. Convective scavenging serves to place the PGM's in the boundary layers of the convective flow, yielding the double peak PGE distribution that attends, say, the chromitite layers which host them in the Bushveld Complex. Ore grades are expected to be higher where the geometry of the chamber enhances heat transfer, leading to thinner convective boundary layers. One aspect of this scenario is that it provides a rationale for the observation that only those layered intrusions of significant initial thickness possess PGE ore bodies that are of economic interest.
The lower layered horizon (50 - 100 m in thickness) is distinguished among the gabbronorite of the Early Proterozoic Western Pana intrusion due to a distinct alternation of complementary leucocratic (anorthosite, leucogabbro) and melanocratic (gabbronorite, norite, pyroxenite) rocks. Unlike the underlying gabbronorite, rocks of the LLH contain little or no cumulus clinopyroxene; the anortite content in the LLH plagioclase is much higher (An=70-80%) than in the underlying gabbronorite (An=60-70%), the ferrugeneity (f#=Fe*/(Fe*+Mg)) of orthopyroxene in the LLH is lower (f#=23-26%) than in the gabbronorite (f#=25-30%), and the chromium content increases 3-4 fold (from 30-40 ppm, to 150-200 ppm). Oxide-sulfide mineralization with high PGE contents is associated with the layers of leucogabbro and anorthosite, which commonly show intrusive relationships with the surrounding mafic rocks of the LLH. Poikilitic intercumulus olivine, which is absent in the surrounding melanocratic rocks, commonly closely associates with the platinum-rich oxide-sulfide mineralization. Petrographic features and the abundances of some incompatible elements (Zr, Rb) in the leucogabbro and anorthosite suggest that these rocks contain a considerable amount of crystallization products of the trapped intercumulus liquid. The anorthositic cumulates are estimated to contain about 20-40 volume percent of the primarily trapped intercumulus liquid. On the whole, the mutual intercorrelations of plagioclase and pyroxene composition in the gabbronorite section of the LLH indicate that the crystallization zone was located close to the temporary floor of the magmatic chamber. The LLH rocks are characterized by a well-manifested effect on mafic silicate compositions of solidification of trapped intercumulus liquid, and by a distinct dependence of incompatible elements (Zr, Rb) contents on the petrochemical parameters of differentiation (F#=Fe*/(Fe*+Mg)). These features suggest that there was no large-scale postcumulus migration of the trapped intercumulus liquid or aqueous fluid phase exsolved from it in the intrusion. The diverse petrographic composition and complicated relationships between the rocks, the changes in the order of crystallization, an abrupt shift of mineral composition towards high-temperature members of their solid solutions, and the presence of PGE-rich sulfide mineralization permit us to consider the LLH a critical zone of the West-Pana intrusion. The zone is supposed to have been formed by mixing of the intrachamber melt with fresh influxes of a magma that was similar in composition to the parental melt. Origin of PGE-rich sulfide mineralization is attributed to a separation of immisible sulfide liquid from trapped intercumulus melt of anorthositic cumulates.
A detailed textural and mineralogical study was undertaken in the Merensky Reef and adjacent footwall and hanging wall leuconorites in the Rustenburg type section, Western Bushveld, to determine relationships between PGE mineralization with host rocks. Four distinct silicate assemblages were recognized: (1) early cumulus plagioclase (An75-73) and orthopyroxene which dominate both the leuconorites and Merensky pyroxenites; (2) mesostasis assemblages around opx which comprise HFSE- and LILE-bearing oxides (loveringite, davidite, zirconolite, etc), as well as apatite, amphibole, biotite, clinopyroxene, sodic plag (<An60), K-feldspar and quartz. The nature of these assemblages suggests derivation from a fluid-rich magma (OH, F, Cl) with a high proportion of LIL- and HFS- and alkali elements; (3) poikilitic plag (An73-62) restricted to the pyroxenite layer; (4) polyphase inclusions in plag of pyroxenites and leuconorites which are compositionally similar to the mesostasis assemblage. The bulk composition of the inclusions varies, however, from gabbro (An60 + cpx) to granite (quartz, K-feldspar, biotite). The cumulus plag and opx, as well as the poikilitic plag, have homogeneous compositions. This suggests that they crystallized either from magma resident in the chamber, or from a new magma injected just before formation of the Merensky Reef. The fluid-rich magma parental to both the mesostasis assemblage and the polyphase inclusions was emplaced in the Merensky protoreef layer after crystallization of opx and simultaneously with plag crystallization. Local mixing resulted in hybrid poikilitic plag in pyroxenites. The fluid-rich magma in interstices between opx or in plag inclusions may have resulted from extreme fractionation of the resident magma parental to the Lower Ultramafic Zone, or it may be of distinct origin. The emplacement of this magma is associated with deposition of PGE minerals, as suggested by the location and textural relationships of the phase assemblages. An understanding of the ore-forming process requires a better knowledge of the nature and origin of this magma.
The Pansky Tundra layered intrusion belongs to a group of the Early Proterozoic layered intrusions of about 2.45-2.50 Ga located in the north-eastern Fennoscandian Shield, Kola Peninsula (Russia). The intrusions were emplaced during just that epoch of intracontinental rifting. This intrusion consists of norite-gabbronorite-gabbroic megacycles and contain rhythmically layered horizon (up to 200-300 m thick). This last includes thin "layers" of norite, olivine-gabbronorite, melanonorite, troctolite and sill-like pyroxene anorthosite. Namely the pyroxene anorthosite lenses host low sulphide Ni-Cu and PGE mineralization and strongly metasomatic altered. The U-Pb ages of gabbronorite by zircon is 2470±9 Ma (Balashov et al., 1993) and anorthosites by baddeleyite is 2449±12 Ma (Bayanova and Mitrofanov, 1995). Ore-bearing pyroxene anorthosite lenses have 10-100 meters long and 10-30 cm thick. Nonaltered rocks consist of plagioclase (75An) adcumulate with intercumulus pyroxene (up to 5%) and quartz. With ore-bearing anorthosites associate «mixed» rocks which consist of plagioclase adcumulate and dark irregular plagioclase-bronsite cumulate and have similar features to the "raisin pudding" anorthosite in the J-M Reef (Campbell et al., 1983) or PGE reef in the Penikat intrusion, Finland (Huhtelin et al., 1990). Ore-bearing altered pyroxene anorthosites are composed by two successively-formed metasomatic association: 1) replacement of pyroxene by tremolite and magnetite; 2) epidote-zoisite-chlorite-albite-sericite-quartz association in the paragenesis with the sulphide-PGE mineralization. The PGE (tellurides, bismutides and sulphides of Pd and Pt) minerals closely associate with chalcopyrite and pentlandite, and are rarely hosted within pyrrhotite and bornite grains. Comparing with host gabbronorites, the REE patterns of ore-bearing anorthosites are characterised by positive Eu-anomaly and lower contents of HREE. Mineralised rocks are clearly enriched by Te, Au, Se, Ir, Cu, Ag, Ni and especially of Br, Cl (neutron-activation analysis, Interfaculty Reactor Institute, Delft, The Netherlands) in the relation to an average gabbronorite of the Pansky Tundra intrusion. Simultaneously for these rocks are typical a sharp depletion in Cr content. Mass-spectrometric thermion-emission study of fluid phases in altered anorthosites and gabbronorites shows that most part of H2O, N2, SO2 and CO, CH4 gases were extracted within temperature interval 800-1200°C as well as its higher content comparably with host gabbronorites. The occurrence PGE minerals and sulphides in the paragenesis with hydroxyl-bearing silicates in metasomatic altered anorthosites can indicate high temperature hydrothermal origin of the PGE mineralization in the Pansky Tundra layered intrusion. Such correlations are typical also for large layered plutons, such as Sudbery (Farrow and Watkinson, 1992) or even for Alpine-type ultramafic intrusions (Ohnenstetter, 1992). The data do not support a simple «magmatic model», but rather argue in favour to late magmatic high temperature hydrothermal origin of PGE ores and sulphide mineralization close to crystallisation of most later vapour-saturated anorthosite within rhythmically layered horizons.
Balashov YA, Bayanova TB & Mitrofanov FP, Precambrian Research, 64, 197-205, (1993).
Bayanova TB, Mitrofanov FP, IConfAbs Fennoscandian Geol. Correlation, 1, 113-114, (1995).
Campbell IH, Naldret AJ & Barnes SJ, Journal of Petrology, 24, 133-165, (1983).
Farrow CEG & Watkinson DH, Mineralogy and Petrology, 46, 67-83, (1992).
Huhtelin TA, Alapieti TT & Lahtinen JJ, Mineralogy and Petrology, 42, 52-70, (1990).
Ohnenstetter M, Mineralogy and Petrology, 46, 85-107, (1992).
Average contents of PGE, Au, Ag in Pilgujarvi layered intrusion (Pechenga Region, Kola Peninsula) aged 1980±40 Ma are calculated from PGE, Au, Ag contents in the gabbro (65 vol.%), piroxenites (2.5%), peridotites (32%) and ore (0.5%) zones. The AAS method with preliminary concentration of elements in sulfide melts was used to determine them in the above units. The difference between the PGE concentrations in the ores (Or) and all the intrusion (In) is found, respectively: Os 17± 2.7 ppb, 0.33±0.06 ppb, ratio Or/In 51.7; Ir 17±2.5, 1.24±0.23, 13.7; Ru 27±2.7, 1.32±0,22, 20.5; Rh 7±0.1, 44±7.9, 0.16; Pt 54±5, 13.2±1.8, 4.1; Pd 100±7, 44±3.5, 2.3; Au 49±6.8, 9.2±1.3, 5.3; Ag 5210±781.5, 405±60.7, 12.9. It follows from this that the Os, Ir and Ru enrichment takes place in the ores if compared with their contents in the initial melt. The study was supported by the Russian Foundf of Fundamental Investigations ( grants #97-05-64863, 98-05-64276).
Gold in the late Archean Witwatersrand Basin in South Africa, the world's largest known depository of this precious metal, occurs concentrated in quartz pebble conglomerates (reefs), where it is often associated with hydrocarbons, occurring in stratiform carbon seams and bitumen nodules. The insoluble organic matter (kerogen, pyrobitumen), the soluble organic matter (extract), and the individual n-alkanes and acyclic isoprenoids (pristane and phytane) extracted from this carbonaceous material were subjected to carbon isotope analysis in order to gain insight into the possible sources of hydrocarbons associated with Witwatersrand gold. To this effect we chose samples from the reef which shows particularly extensive development of carbon seams, the Vaal Reef, and from the reef which had undergone the most intense hydrothermal alteration, the Ventersdorp Contact Reef (VCR). In order to examine local variability, all these samples were taken from the same mine (Vaal Reefs) in the Klerksdorp goldfield. For comparison, one sample from the Basal Reef, which is believed to be at a stratigraphic level similar to that of the Vaal Reef, from the Welkom goldfield was also analyzed. To address the question whether the hydrocarbons in the Witwatersrand reefs could be sourced in the Transvaal Supergroup dolostones, samples from there were also included. The extracts from the samples are depleted or enriched in 13C (up to ±2.4‰) compared to the associated kerogens. This suggests that the auriferous rocks were stained by mobile hydrocarbons produced by thermal and oxidative alteration of indigenous organic carbon. The variability in the isotopic omposition of the indigenous kerogen (-26.6 to 39.2‰) indicates that a variety of biomass and reducing conditions existed in the depositional sedimentary environment. The isotopic signature of this organic source was changed during secondary, post-depositional, thermal and chemical alteration processes during diagenesis and multiple infiltration of metamorphic/hydrothermal fluids. The small variability in the 13C values of the individual n-alkanes (-31.1±1.7‰) and isoprenoids (-30.7±1.1‰) indicates fractionation during local migration/remobilization of the mobile hydrocarbons. Carbonaceous matter in dolostones from the lower Transvaal Supergroup shows distinctly different isotopic and molecular characteristics and thus cannot have been the source of the hydrocarbons in the Witwatersrand Basin. The new isotopic organic geochemical results are in good agreement with the model of a hydrothermally altered metamorphosed placer deposit, whereby most phases, including hydrocarbons, sulfides, uraninite and gold, which appear late in the paragenetic sequence are authochthonous products of basin-internal, and often reef-internal mobilization.
Platinum is not a traditional ore in the north of the Urals and Timans. However, appreciable amounts of platinum and PGM were discovered in some gold placers. Gold-to-PGM ratio varies from 100:1 to 100:2.5. We have investigated the PGM content in the alluvium of the Pechora river basin. Twenty six minerals and their varieties have been found. The most widespread among them are isoferroplatinum, iridium, osmium and ruthenium, observed in the majority of streams. The age of PGM from a Timanian placer Kyvvoge varies rather widely. So isoferroplatinum age is evaluated to be from 1090 to 400 million years with 187Os/188Os ratio from 0.1199 up to 0.1246, ruthenium and osmium from 1360 to 330 million years with the isotopic 187Os/188Os ratio from 0.1180 up to 0.1251. The obtained dates agree well with the age of Timanian formations and probably reflect the multistage formation of platinum mineralization in massifs (primary crystallization, disintegration of solid solution, recrystallization, opening of a system and escape of volatile components) on the one hand, and combination in the placer of PGM from several massifs of different age or even belts, on the other. Re content ranges from traces up to 0.35% in the investigated PGM. Given below for comparison are some age estimates for PGM from a placer of the river Kojim (Circumpolar Urals): isoferroplatinum from 690 to 550 mln. yr., ruthenium from 1040 to 520 mln. yr. It is not excluded that the older age of ruthenium in comparison with isoferro-platinum is stipulated by the different age of their sources: platinum-bearing concentriczonal ultrabasic massifs of the Uralian-Alaskan type and chrome-bearing alpinotype massifs. The youngest age (490-370 mln. yr.) is attributed to the PGM (osmium and ruthenium) from the alluvium of the Yelyets and Longot-Yugan rivers (Polar Urals) which take their source from alpinotype Riyiz and Syum-Kew massifs. It is clear that the Polar Urals alpi-notype belt cannot be a source of PGM for the Circumpolar Uralian and Timanian placers.
The main scheme of pentlandite formation in the Fe-Ni-S system on base of experimental (Kullerud et al., 1969; Naldrett, 1967; Fedorova, Sinyakova, 1993) and natural data can be represented as: 870°C 620°C melt-mss + heazlewoodite ss - pentlandite I. According to recent data (Sugaki et al., 1998) the existence of high temperature cubic (pc) pentlandite phase between crystallization of melt and appearance of normal (cubic Fm3 m) pentlanditeI (pnI) is possible as well. Some quantitative estimations of Pt and light PGE partition have been elucidated on the base of our experimental data (Sinyakova, 1996, 1998 a,b; Kolonin et al., 1997) and discussed within the wide Fe:Ni ratio. For example, the Pt, Ir and light PGE partition coefficients between mss and sulfide melt at 900°C and at S-content in the system of 51 at.% change in the following manner: Ni/Ni+Fe lg fS2 Pt Pd Rh Ir Ru 0.4 2.4 0.07 0.02 2.1 2.9 8.2 0.6 2.0 0.1 0.03 6.4 7.2 >10. They demonstrate different partition of PGE at early crystallization of mss from sulfide melt. It can be seen that platinum and palladium are concentrated in sulfide melt, whereas rhodium, iridium and ruthenium are scattered in mss.
The light PGE partition between mss and pentlanditeI, forming at lower temperature (600°C), is more important for ore-geologist. It is shown that the stability conditions of pentllandite I gradually shift towards higher values of -lgfS2 from 11.5 - 8 for Fe-pentlandite to 8 - 6.6 for Ni-pentlandite. It is assigned that even at anomalously high content of platinum in the initial melt (0.5 wt.%) it is absent in pentlandites of any composition, as electron microprobe analysis shows the light platinoids and Ir in the following quantities (in at.%):PdRhRuIrFe-pentlanditeup to 0.4 up to 0.3 up to 0.4 up to 0.2 Ni-pentlandite up to 3.1 up to 4.4 up to 5.4 up to 1.1 (so-called <pi>-phase) Because Rh, Ir and Ru are scattered in mss too, their parition coefficients between mss and pentlandite I are close to 1. The obtained experimental data for S-rich conditions demonstrate the growth of chalcophility, and consequently the degree of their scattering in sulfides towards: Pt < Pd < Rh < Ir < Ru.
This work was supported by the Russian-French project through grant of RFBR-CNRS N 98-05-22020 and through the grant of RFBR N 98-05-65314
The concentration and composition of volatile components (H2O, CO2, CO, CH4, H2) at the plagioclase samples were studied by gas chromatography procedure. The plagioclase samples were selected from anorthosites of PGE-bearing horizon and taxitic rocks (olivine leucogabbroes and gabbronorites, gabbro pegmatites) from the Layered zone of Ioko-Dovyren massif. The investigated samples are characterised by the higher contents of H2O (1.84-10.32 ml/g), CO2 (0.0635-1.234), H2 (0.08-0.27), CO (0.0199-0.1687), CH4 (0.0087-0.0472) in comparison with plagioclase from normal rocks of the massif Layered zone (Balykin et al., 1983). The H2 and CO2 are leading components at the gas phase of all plagioclase samples. The CO and CH4 have the subordinated role. At the same time the plagioclases from rocks of PGE-bearing horizon differ by a higher proportion of H2 and CO2.
This composition of volatile components could not be in equilibrium with melt under magmatic temperatures (Neruchev, Prasolov, 1995). Presumably it is result of influx of fluids from intermediate magmatic chamber. On the other hand, the increased role of CO2 can be caused by not only primary solubility of gas in equilibrium with melt, but also by partial oxidation of fluids during massif cooling and the crustal waters circulation. The heavy isotope composition of sulphur at the Ni-Cu and PGE mineralization (Glotov et al., 1998) and that of hydrogen and oxygen in lower margin rocks (Kislov et al., 1991) are evidences about this phenomena. This process was displayed more heavily in rocks of PGE-bearing horizon than in Mafic zone rocks by a ratio of the CO2 and CO contents in a cross-section of the massif.
Thus, the evolution of volatile components during formation of the PGE-bearing horizon take place from carbonaceous composition mainly, that is connected directly with initial melt, through hydrogen-methane composition, playing an important role in PGE industrial concentrations formation, to water mainly composition as a result of hydrothermal system occurrence on the late magmatic and postmagmatic stages.
This work was supported by Russian Foundation of the Basic Researches 98-05-65309 and joint RFBR-CNRS program 98-05-22020.
Balykin AP, Yurkovskiy SA, Proskuryakov, Sov. Geol. Geoph, 3, 36-42, (1983).
Glotov AI, Kislov EV, Orsoev DA, Podlipskiy MY Pertseva AP, Zyuzin VI, Rus. Geol. Geoph, 39, 228-233, (1998).
Kislov EV, Vetshtein VE, Konnikov EG, Sov. Geol. Geoph, 5, 88-92, (1991).
Neruchev SS, Prasolov EM, Platinum of Russia, II, 94-101, (1995).
The Gremyakha-Vyrmes igneous complex of about 100 km2 in area is confined to the Central Kola accretion terrane (68038'N; 32028'E). The complex consists of three phases, each being represented by spatially distinct group of rocks. The earliest phase comprises verlite, gabbro, anorthosite and ferrodiorite which make up the southern part of the complex. A later phase consists of a series of melteigites, ijolites, urtites and nepheline syenites which are also located in the southern part of the complex. Igneous activity is completed by the intrusion of peralkaline, but more siliceous, magma which has formed a large body of granite in the nothern part of the complex. According to radiometric data (K-Ar; U-Pb, and Pb-Pb methods) all of the above rock series were formed during the time interval of 1800-2000 Ma (Pushkarev 1990).
The most important chemical feature of rocks and mafic minerals of a layered mafic-ultrmafsic series is their high iron-magnesium ratio. With these characteristics, the gabbroic rocks from the Gremyakha-Vyrmes complex have much in common with platiniferous gabbro from the Coldwell complex, Nothern Ontario (Good & Crocket 1994), which also includes nepheline syenite. Gremyakha-Vyrmes complex contains ilmenite-titanomagnetite ores with apatite concentration of up to 8% form layers of 100-400 m thickness. The estimation of the PGE and Au distribution in mafic-ultrmafsic rocks has been made on the basis of 56 atomic-absorbtion analyses(accuracy 5%) of rocks and minerals of the layered gabbroic series. The preliminary results fron analyses show that concentrations of noble metalls increase from the less evolved members (verlite) to more evolved (gabbro-anorthosite with magnetite-ilmenite ores) from 10-12 ppb Pd+Rh and 5 ppb Au to the Pd - 63 ppb, Rh - 14 ppb and Au - 13 ppb. The estimation of the PGE and Au distribution in Fe-Ti oxides and sulphides (pyrrhotite, pentlandit and chalcopyrite) shows that the highest concentrations of noble metalls are in titanomagnetite (Au - 610 ppb, Pd - 110 ppb) and ilmenite (Au - 420 ppb, Pd - 58 ppb) from gabbro containing magnetite-ilmenite ores. Sulphides of these rocks contain Au - 38 ppm and Pd - 75 ppb. According to A.N.Korobeinicov et. al. (1997) the relatively high concentration of Pt and Pd (up to 220 and 240 ppb, respectively) has been determined also in samples contacting with younger alkaline rocks. This gabbro has obvious signs of metasomatic changes caused by the nepheline syenite intrusion.
The Copper Cliff Offset is a dike of quartz diorite which extends from a funnel-shaped embayment of Sublayer south into the footwall of the Sudbury Igneous Complex and then rapidly thins to become a steeply dipping, 40 m-thick dike cutting metamorphosed sedimentary and volcanic rocks of the Huronian Supergroup. Ni-Cu-PGE deposits such as the 120 orebody of the Copper Cliff North Mine are located in inclusion-rich zones near the centre of the dike. The sulfides occur as disseminated blebs of pyrrhotite, pentlandite, and chalcopyrite in quartz diorite; the sulfide content increases both toward the centre of the dike and down-dip forming massive ore. Mineralogical and petrological studies of ore from the 1200 level underground, as well as from drillcore on the Lady Violet property, revealed that the predominant sulfide assemblage comprises pyrrhotite, pentlandite, chalcopyrite, ilmenite. Silicates are feldspar, quartz, amphibole, biotite, chlorite, epidote, titanite.
Detailed mapping of the surface expression of two ore bodies (100 and 880) revealed zonation of the dyke. Inclusion-rich quartz diorite with abundant disseminated sulfide blebs (3-5%) is located in the 10-20 m wide centre of the dike. Inclusion- and sulfide-proportions as well as grain size decrease toward the dike's contact with the enclosing granitic and metavolcanic rocks. Narrow felsic dikes without sulfides occur at the centre or at the margin of the dike. The disseminated sulfides consist of pyrrhotite with minor pentlandite. Chalcopyrite content is variable and it partially replaces pyrrhotite. Oriented ilmenite lamellae reflect oxy-exsolution from partially replaced titanomagnetite. Idiomorphic 10-30 µ m-sized grains of cobaltite-gersdorffite occur mainly in pyrrhotite and chalcopyrite, but also in epidote. Cobaltite-gersdorfitte from both massive and disseminated ore contains 2-5 µ m-sized sperrylite, and 1-3 µ m-sized hollingworthite (RhRuAsS) and irarsite (IrAsS). Pd content of cobaltite may exceed 1 wt.%. Hollingworthite contains varying amounts of Os (0.9-1.57wt.%), Pt (1.46-3.93wt.%), Pd (0.48-1.42wt.%), Ru (0.65-5.75wt.%). Bismuth tellurides are mainly tetradymite, tsumoite, and tellurobismuthite. Massive ores and blebs are intergrown with OH-bearing silicates such as epidote, amphibole, biotite, chlorite, and with plagioclase and calcite. These minerals also occur as inclusions in massive ore composed of pyrrhotite-pentlandite-chalcopyrite-ilmenite and in blebs of pyrrhotite-chalcopyrite.
The original magmatic textures of immiscible sulfide droplets in quartz diorite have been partially modified revealed by marginal intergrowths with hydrous minerals. This feature is probably not only the result of regional metamorphism, but related to fluids of metamorphic or hydrothermal origin.
The Monchegorsk and Monchetundra layered intrusions (~2504 Ma) are placed in south inner corner of the Paleoproterozoic rift-transform system between the Pechenga and Imandra rift zones. Magmatic suits are predominantly ultramafic in the Monchegorsk Pluton and gabbroic in the Monchetundra massif; possibly, they con-stitute a single evolutionarily complex. The Monchegorsk Pluton contains economic or sub-economic PGE-Cu-Ni deposits, recently discovered PGE-bearing chromite horizons, and firstly reporting PGE-bearing disseminated sulfide mineralization.
PGE-Cu-Ni deposits are placed within layered series of the Pluton and presented by veinlet and disseminated sulfide ores with bismutotellurides Pd and Pt as prevailing PGM. The PGE-bearing chromitite horizon (up to 15 m in thickness) was found re-cently in dunites of the Sopcha massif. Rock-forming minerals of this horizon are olivine, minor pyroxene, and high chromium chromite; the latter contains the tiny in-clusions of laurite.
The newly discovered PGE-bearing disseminated sulfide mineralization (2-7 ppm PGE+Au) was revealed in gabbronorites of the marginal zone of the Monchegorsk Pluton (Mt. Vurechuaivench). The PGE reef (1 - 4 m thick) has been traced over 2 km through strike. Major base metal sulfides are chalckopyrite, millerite and pen-tlandite with minor pyrrhotite and pyrite. PGM (1 to 150 mkm in size) occur in sul-fide grains, along contacts between sulfides and silicates, and also in silicate matrix. Mineral association includes bismutho-tellurides of Pd and Pt, sperrylite, holling-worthite, platarsite, irarsite, Pd-arsenides, Pt-gersdorffite. PGE are incoming as iso-morphic admixture into Ni-Co-Fe sulfarsenides and pentlandite. Principal features of the Vurechuaivench PGE mineralization are high whole rock contents of Pd and Au, and close association of PGM with post-magmatic sulfides. Low-temperature envi-ronment of ore formation is evidenced also by admixture of Hg, Ag, Se and Au in PGM, as well as by such typical post-magmatic minerals as electrum, hessite, nau-mannite, galenite and clausthalite.
Prospective mineralized occurrences enriched in PGE are encountered within transi-tion zone between the Monchegorsk and Monchetundra massifs. Preliminary studies show that two types of mineralization exist here. Densely disseminated Cu-Ni, mainly pyrrhotite ores in pyroxenites and norites contain PGE up to 1-3 ppm. Cuperite, merenskyite and hessite were found here. PGE-enriched occurrences (up to 10 ppm) in the Pentlandite Valley are characterized by very erratic distribution and located in pegmatitic bodies and layers of plagiopyroxenites and norites within Mon-chetundra gabbronorites. PGM assemblage is dominated by cooperite-braggite-vysotskite isomorphic group, with minor sperrylite, kotulskite, moncheite, meren-skyite, electrum. The work is supported by RFFI, grant 98-05-64671
The Dinamitnoe, Zelenoe, Ozerscoe deposits are connected with a tectonic contact of carbonate-sedimentary rock, wich belong to ophiolitas belt, with Archean granite-gneiss of Gargan block. The original of such type gold deposits remains controversial one yet. Wide complex of investigations was carried out for elucidation this problem, the role quartz during formation of quartz-gold-tennantit deposits and its peculiarity was considered separately. The four different types of quartz were determined by microscopic styding.of veins ores. First type is a milk-white gangue, middle- and small-grained quartz, which being exposed of collecting recrystallization. This is early generation. Second type is transparent, blue-grey middlegrain ore quartz. Ore minerals ( galena, tetraedrite, pyrite, tennantite, sphalerite, native gold) are characterized accumulation in intergranular place 'ores' quartz. Third type is a small-grained, almost transparent more late quartz being developed over early types with formation of rims and replacement of cracks. Last type is a fine-grained quartz, which together with carbonates and chlorites forming fine veins. Such separation of quartz types is in accordance with thermobarometry data, which were received during investigation of fluid inclusions in quartz by homogenization of gas-liquid inclusions, Raman spectroscopy and cryometry methods. So only quartz-1 combine two- and three-phase inclusions with complex gas phase composition (CO2, N2, CNs) and presence of alkanes in liquid. Also it has two-phase with CO2 both gas-liquid and two-liquid inclusions. The temperature of homogenization fluctuate from 300 to 350 C. The quartz-2, "ores" quartz, has less inclusions comparison with Q-1. Its inclusions represent two-liquid (70) and gas-liquid (30)phases with CO2. Average homogenization temperature is 240-260 C, what is correspond to temperature of galena-tennantite stage. The Q-3 generation contains the round one phase inclusions only. For determination influence of different quartz types in processes of gold-sorption and deposition the methods of radioisotope indicator (195 Au) combined with autoradiography were employed too. Polished section of different parts quartz veins were used. They were placed in a plastic vessel. Weak hydrochloric acid solution were prepared by adding stable 195 Au were induced as radioactive labels. The Au content of solution were 5-10 microgramm/l at pH=5. Experiments were carried out at room temperature and atmospheric pressure. After 72 hours, the sections were removed from solution, dried and exposed to film of FO-5 types. The experimental results have confirming of the different sorption ability of quartz. Maximum of blacking density corresponds to the quartz, which low temperature generation and lead to forming late viens of substitution.
The Chineisky massif is a large layered lopolith which is chiefly composed by gabbro varieties. Sulfide mineralization tends to the lower contact of the lopolith with surrounding sandstones and occurs either as impregnation in gabbroids, or as ore nests, lenses and irregularly shaped nodules in hornfels sandstones. Correspondingly, two types of mineralization can be considered: endocontact and exocontact.
Endocontact ores are gabbroids with sulfide impregnation dominated by pyrrhotite and chalcopyrite. Pentlandite, siegenite, violarite, millerite, gersdorffite and cobaltite, safflorite, as well as pyrite, sphalerite, and galena are present in minor amounts. PGE minerals are defined by PtAs2, PdTe, PdBi and Pd2As). Isomorphic series from sobolevskite to kotulskite is typical of the ores Besides, native silver and Au-Ag and Au-Ag-Hg alloys are found.
Exocontact ores are sulfide nodules in hornfels sandstones occurring 10-15 m from the lower contact with gabbroids. Sulfide mineralization is accompanied by metasomatic alteration of hornfels resulting in the formation of complex quartz-feldspar-amphibole-biotite metasomatic rocks. Ore minerals are dominated by chalcopyrite and bornite with small amounts of pentlandite, millerite, siegenite, linnaeite, violarite, gersdorffite maucherite and nickeline. Huebnerite was found in one ore body. Palladium (up to 2.2 mass%) was found only in maucherite (Ni11As8). Gold-silver solid solutions from the exocontact ores are characterized by the increased amount of Ag (36-77 atomic%). PGE minerals are represented by PtAs2 and numerous Pd-bearing compounds with Bi, Sb, As, and Sn with dominating PdBi2, PdBi, Pd5Sb2 and PdSb. Less abundant are Pd2Sn, Pd11Sb2As2, Pd8Sb3, PdBiTe, PdNiAs and some other unnamed Pd-compounds. Rh mineral RhAsS - has been found. Pd minerals from exocontact ores the Chineisky massif are specified by the presence of Pd2Sn-Pd2 (Sb,Sn) and PdBi-PdSb isomorphic series. This feature differentiates the Chineisky paragenesis of Pd minerals from that of PGE minerals from other Cu-Ni deposits.
The endocontact ores are enriched in pyrrhotite and contain 16 mass% of chalcopyrite end member and about 1% of Ni-component. In the triangle CuS-NiS-FeS exocontact ore compositions fall in CuFeS2-CuS interval: this is explained by the presence of bornite. Copper-rich ores are found in many Cu-Ni deposits, their formation being related to the evolution of sulfide melts. PGE distribution between Cu-Ni and Cu-rich ores from different deposits correlates well with experimental data (Fleet et al., 1993). Pt, Pd and Au coefficients of distribution between Fe-rich and Cu-rich ores are over 1 when magmatically formed Cu-rich ores. Some other values of PGE and Au distribution coefficients were estimated for the Chineisky massif ores, thus, uggesting another, non-magmatic, origin of exocontact ores. The available values of D exo/endo seem to be related to the distribution of PGE, Au and Ag between sulfide melt and high-temperature fluid which was responsible for the transportation of native and other metals to the hornfels sandstones. We acknowledge the financial support from RFFR grant 97-05-65215.
Fleet ME, Chryssoulis SL, Stone WE & Weisner CG, Contrib Mineral Petrol, 15, 36-44, (1993).
Precise determination of PGE abundances in silicate and sulfide rocks is a prerequisite for the study of a variety of geological processes, such as core-mantle interaction, mantle evolution, magmatic differentiation, and hydrothermal alteration. However, common methods used for PGE measurements do not adequately control the chemical yield during chemical separation procedures. This among other factors limits the attainable total analytical uncertainty (TAU) to 5-10% at best.
Isotope dilution ICP-MS circumvents these problems. We chemically separate the PGE applying a modified procedure originally developed by Rehkaemper and Halliday (1997), with the addition of the elements Re and Os and Os isotopic analysis by N-TIMS. After adding the spike-solution the samples are dissolved using a high pressure asher at 320°C and 130 bar with concentrate acids of HNO3 + HCl for about 24 h. We first collect Os by solvent extraction with Br2. The other elements are then separated using the anion-exchange separation procedure of Rehkaemper and Halliday (1997). This procedure gives high yields for all PGE (>70%).
Os concentrations and Os-isotopic compositions are analyzed by N-TIMS. The remaining PGE concentrations are measured by isotope-dilution using a MC-ICPMS (Isoprobe). This allows measurements by static multiple collection and has the advantages that fractionation corrections are done by internal normalization or by monitoring an admixed element of similar mass. Interfering element corrections can also be employed. Isotope dilution offers the best analytical precision (TAU <1%), but monoisotopic elements (e.g., Rh and Au) cannot be measured. In this study, we obtained PGE concentration data on silicate and sulfide samples such as volcanic rocks (e.g., Kerguelen Plateau), mantle peridotites (e.g., Troodos Complex in Cyprus and Ivrea Zone, Italy), and magmatic or hydrothermal sulfides (e.g., TAG and Sullivan, British Columbia, Canada). The results reveal different distributions of PGE in various types of rocks and sulfides, and their origin is further discussed on the poster.
Imandra complex about 2.44 Ga in age is represented by number of layered bodies up to 3 km thick and from a few to several tens of kilometres in length within the Palaeoproterozoic riftogenic Imandra-Varzuga Belt in the central Kola Peninisula. The bodies have similar structures, rock composition and ore component distribution, which suggests that they may form large (1500 km2) intrusion. However, they could be individual intrusions emplaced within the Sumian volcano-sedimentary complex and Archaean basement in the vicinity. The complex is a part of the early Palaeoproterozoic platineferous Kola province, formed by such layered intrusions as Fedorovopansky, Monchegorsky, Mt. Generalskaya, etc., derived from the specific siliceous high-Mg (boninite-like) series melts.Border zones of the bodies from 5-7 till 90 m thick represented by taxitic mesocratic gabbronorites with schlieren and veins of pegmatoid rocks. Layered series occurs above it. It begins from alternation of bronzitite and bronzite-plagioclase cumulates with thickness about 45-60 m, sometimes enriched in Crt; thickness of mineralised layers varied from 10-30 cm to 1.5-2.1 m, their number till 4. The main zones of the bodies (till 2000 m) formed by norites and gabbronorites with gabbro-anorthosites in the upper part with thickness from 7-20 till 70 m. The near-roof zone (150-500 m) is composed by mesocratic quartz gabbro and gabbrodiorites with disseminated V-bearing titanomagnetite mineralization; there are two the most enriched in Ti-Mt layers; in the base of the zone and another about 100 m further the section. Cr and PGE mineralization occur in the Imandra complex intrusions. Economic Cr mineralization is found in the lower part of the Layered series and represent mainly by disseminated Crt (up to 50-70 vol.%), which form ore beds or veins among plagiopyroxenites and melanocratic norites, and schlieren in the mesocratic gabbroids of the Border group. The Cr2O3 content in the ore beds varies from 17 to 30 wt.%; sulfides are practically absent. Chromspinelides belong to the subferrialumochromite-ferrialumochromite series and characterised by Cr2O3 content till 42-48 wt.%, total iron 31-33 wt.% and Al2O3 6-8 wt.%. The PGE contents in chromitites is 0.3-1.8 ppm and Au till 1 ppm. Minerals of laurite-erlichmanite series are predominating; sperrilite is less abundant, and platarsite, hollingworthite, malanite, cooperite and isoferroplatinum are rare.
We have studied the Pd mineralization from oxidation zone of the Chineisky deposit. Chineisky massiff occurs in the western part of the Aldan Shield, Russia. The oxidation zone is weak and is exhibited in some areas. The ores, oxidized to a varying degree, contains both primary relict minerals (chalcopyrite, arsenopyrite, maucherite, pyrrhotite, michenerite, Te- and Sb-bearing sobolevskite, Sb-rich paolovite, and sperrylite) and typical minerals of oxidation zones (goethite and covellite, formed after sulfides, and scorodite - a product of arsenopyrite replacement). The fractions of both weakly and intensely oxidized ores were investigated.
A sample of weakly oxidized ore, washed from brown iron, consists mainly of chalcopyrite. The PGE are dominated by sperrylite, whose euhedral crystals contain small inclusions of maucherite, arsenopyrite, and chalcopyrite. In case sperrylite intergrows with chalcopyrite and arsenopyrite, the latter to a varying degree undergo oxidation to goethite and scorodite. Goethite contains 0.28 - 6.72 wt.% Pd. It is very important for elucidation of mode of occurence of PGE in oxidation zone. Out of three Pd minerals, which are present in the weakly oxidized ore, sobolevskite and paolovite do not decompose and remain stable, and only michenerite undergoes partial alterations. Scarce relics of unaltered michenerite are found among the replaced mass of colloform and banded-colloform structure. The concentrate of intensevly oxidized ores consists from brown iron and goethite with a minor amount of unaltered chalcopyrite which is often surrounded by chalcocite rims. All michenerite grains are oxidized to form of banded-colloform structures. Low- and high-refractivity zones are alternated and observed the optical zoning. The ratio of Pd,Bi,Te elements varies according to oxidation degree. Occasionally a component of altered michenerite is Fe which concentrations range from 0.09 to 13.99 wt.%. The optical zoning do not depend from Fe concentrations. The composition of oxidized michenerite versus to unaltered mineral observes the differenses in lower concentration of Te and Bi, which are, probably, evacuated first. The deficit of total oxidized compound is about 10 - 25 wt.% and may be related to (1) presence of oxygen, which was not determined by microprobe, as well as (2) to the resorbed surface of partially dissolved mineral. Sobolevskite also observe the alterations, but in a lesser degree than intergrown michenerite. The surface of Pd(Bi,Sb) has a colloform structure as a result of partial resorption. Analysis of oxidized sobolevskite has a 11%-deficit of the total, due to the removing of Te and Bi from the decomposed mineral. Paolovite, which seems to be unaltered, also observed in paragenesis of michenerite and sobolevskite, and has the homogeneous surface, and its analysis corresponds to the Sb- and Te-bearing variety.
The PGE minerals in the oxidation zone of sulfide ores decompose. Their degree of stability increases in the following order: PdBiTe ¾Pd (Bi,Te,Sb)¾Pd2 (Sn,Sb). The most stable Pd mineral is paolovite which, the same as the sperrylite, does not decompose under hypergene conditions but it accumulated as a part of placers. Palladium is removed from decomposed minerals. It is sorbed by brown iron and goethite. That is a secondary enrichment of Pd in oxidation zone. We acknowledge the financial support from RFFR grant 97-05-65215.
The distribution of the platinum group elements (PGE), Au and Ag has been studied in the sulphide ores of Cu-massive deposits, the South Urals, Russia. The presence Pt and Pd has been established in 100 of investigated speciments of ores, ore-concentrates and sulphides; the results of 50 chemical analyses have shown the presence of Au and Ag. The contents of PGE are different and depend on the type of ore-deposit. The contents of Pt and Pd vary from 0.002 ppm to 0.025 ppm and from 0.002 to 0.12 ppm respectively in the Sibay and Alexandrinskoe deposits occurring in association with the volcanites of the basalt - rhyolite series of the primary eugeosynclines. The Bakr-Tau, Tash-Tau and Balta-Tau deposits occurring in association with the K-Na volcanites of the andesite-dacite series of ancient island arcs are characterized by higher contents of Pt and Pd varying within the range: Pt=0.002 - 1.2 ppm and Pd=0.004 - 0.9 ppm. The Rh content is different but does not exceed - 0.007 ppm in any type of the deposit. The Cu and Cu-Zn concentrates show the highest contents of Pt + Pd. The first results of the detailed studies allow to say that PGE occur in dispersed form in sulphides. It is also possible that PGE may occur in form of individual minerals in the Cu-Zn- rich ores. The native Au has been found in the Bakr-Tau and Alexandrinskoe deposits. Au and Ag also occur in dispersed form. The contents of Au in ores vary from 0.1 to 8.4 ppm, Ag - from 1 to 290 ppm. The ores enriched by Zn also show highest contents of Au and Ag. Ag concentrates mainly in fahlore, bornite, chalcocite and sphalerite. The Bakr-Tau deposit is characterized by the higher contents of Au and Ag. The relatively high contents of Pt and Pd in the metasomatically altered sulphide-poor host rocks and in the magnetite-ores of the Sibay deposit as well as in the sphalerite-chalcopyrite-ores of the Bakr-Tau and Tash-Tau deposits are related to early stages of ore formation. The pyrite-bornite-sphalerite-fahlore mineralization appeared during a late stage of the ore formation, when Au and Ag were also accumulated. The distribution of noble metals in Cu-massive ore deposits of the South Urals corresponds to the model of the hydrothermal deposits formation.
Index of EUG 10 Volume
Further EUG 10 Information
Index of the Journal of Conference Abstracts
Cambridge Publications Home Page
Last Updated on Wednesday, March 17, 1999.
© 1997 Cambridge Publications