Sarah-Jane Barnes, sjbarnes@uqac.ca, Tél. 418-545-5011 poste 5274, Fax. 418-545-5012

    Abstract

The Bushveld Complex of South Africa contains ~ 80 % of the world’s Pt and almost half of its Pd resources in the form of three large ore deposits, the Merensky Reef, the UG-2 reef and the Platreef. Two questions arise a) were the magmas that formed the Bushveld Complex pariticularily Pt- and Pdrich? and b) what is the origin of these magmas? In order to consider these questions we have estimated the compositon of the Bushveld magmas based on 40 new whole rock analyses of quench textured rocks from the margins of the intrusion. Broadly speaking, there are two types of magmas present, a Mg-rich basaltic andesite and a tholeiitc basalt. Both of these magmas are enriched in large ion lithophile elemnts, light rare earth elements and in Pb. Both magmas have negative Ta, Nb, P and Ti anonmalies. The PGE contents of both magmas are similar to primary basalts and they do not appear to be enriched in PGE, except for Pt. The Pt/Pd ratio and Pt/Ti ratios are 1.5 to 2 times that of most basalts. It is possible to model the lithophile element composition of the two magma types by up to ~ 50 % crustal contamination of komatiitic basalt or picrite. This conclusion is supported by Sr, Nd and oxgen isotopic work. However, the Pt contents of the model magmas and the Pt/Pd and Pt/Ti ratios are much lower than those observed. An alternative is to suggest that the Bushveld magmas formed by parital melting of the metasomatised lithosphere. But, modeling using Kaapvaal mantle xenoliths compositions and MELTS shows that this magma would be too Al2O3 rich and SiO2 poor. Two possible solutions to this are: a) the melts formed by zone refining melting of the lithosphere (to attain usually high Pt concentrations) followed by contamination with crustal melts; OR b) the current estimate of primitive upper mantle with a Pt/Pd of 1 is incorrect and it should be closer to 2, in which case mixing of a plume derived magma with continental crust melts approximates the composition of the Bushveld magmas.

PGE reference material heterogeneity – Estimating minimum analytical mass

  Bédard, L.P., Kim, H.E.,  Barnes, S.-J., 2011 

 Abstract

    Very small analytical masses (mg to ng) are dictated by modern analytical instruments, either because of their high sensitivity, analytical protocols or because of the small sampling volume of micro-beam techniques. But small sample masses can create problems when trace elements are major constituent in some minerals (e.g. PGE as PGM) and ar  irregularly distributed in samples or reference materials. Accidental inclusion or exclusion could change the spatial concentration realised simply due to spatial heterogeneity issues [1,2]. Such effects influence mass balance calculation and element budgets. Recommended minimum mass for reference materials is generally determined by analysing smaller sample mass until variance become unacceptable. Is there a way to determine minimum sample mass directly related to RM heterogeneity? And is it identical for all analytes, or unique for each analyte, in each sample? In order to systematically investigate these questions, pressed-pellets of PGE-bearing reference materials (CHR-Bkg, CHR-Pt+, MASS-1, MASS-3, WMS-1, WMS-1a) were analysed with an EDAX EAGLE III micro-XRF. Approximately 10 000 contiguous measurements were made with a beam of 50 μm covering a total area of about 25 mm2. Up to 25 analytes were measured including precious metals. Reproducibility was determined by analyzing the same location 1 000 times. In order to express the empirical heterogeneity, Angle Measure Technique (AMT) was employed [2]. The AMT transform describe the elemental map complexity as a function of geometrical scale from local to global. AMT provides a quantitative measure of the empirical heterogeneity for each element, RM, or analytical sample. From such results a minimum analytical mass for each analyte is proposed to ensure that analytical results are representative all the way down to the scale of RM fields-of-view. The proposed technique offers the advantage of defining the effective minimum mass for each analyte with better elemental sensitivity by allowing for a more meaningful estimation of the global minimum analytical mass.

[1] Savard, Barnes & Meisel (2010) Geostandards and Geoanalytical Research 34, 281-291. [2] Huang & Esbensen (2000) Chemometrics and Intelligent Laboratory Systems 54, 1-19.

The origin of platinum-group elements in accessory pyrite from magmatic sulfides from a Sudbury Ni-Cu-PGE deposit

 Dare, S.A.S.and Barnes, S-J., 2011

 Abstract
    Magmatic platinum-group element (PGE) and Ni-Cu-PGE sulfide deposits consist of pyrrhotite, pentlandite, chalcopyrite ± accessory pyrite.  Laser ablation (LA) ICP-MS of these base-metal sulfide (BMS) phases is a useful tool to constrain the mineralogical sites of the PGE in these deposits which is important for the petrogenetic models of the ores and for the efficient extraction of the PGE.  Previous work has shown that  1) pyrrhotite and pentlandite typically host much of the IPGE (Os, Ir, Ru, Rh),  2) pentlandite also hosts much of the Pd and Co,  3) chalcopyrite does not host PGE,  4) Pt is absent from the BMS and forms Pt minerals spatially associated with the sulfides and  5) the role of pyrite has not been closely investigated.
We have determined the trace element content of BMS from 5 massive sulfide samples from the McCreedy East deposit of Sudbury, using a 213nm Nd:YAG UV laser and a Thermo X7 ICP-MS at UQAC.  The samples are pyrrhotite-rich, with minor pentlandite, chalcopyrite ± accessory pyrite (< 2 wt. %).  Pyrite forms small, euhedral grains which are associated with pyrrhotite and pentlandite.  A surprising result is that the PGE are hosted not only in pyrrhotite and pentlandite but also in pyrite.  Relative to the co-existing pyrrhotite and pentlandite, pyrite is significantly enriched in IPGE (e.g. < 130 ppm Rh), As (< 30 ppm) ± Pt (< 0.15 ppm).  Furthermore, the concentrations of these elements in pyrite, together with Co and Se, are oscillatory zoned and the PGE-As concentrations decrease from core to rim as the Co-Se concentrations increase.
The pyrrhotite-rich assemblage of McCreedy East represents a cumulate of Fe-rich monosulfide solid solution (MSS) which crystallized early from the sulfide melt (~1000°C).  Experiments show that IPGE and Co preferentially partition into MSS whereas Pt, Pd and As remain in the fractionated liquid.  Upon cooling of the MSS cumulate (<600°C), pyrrhotite, pentlandite ± accessory pyrite exsolved and inherited the IPGE and Co.  The enrichment of IPGE in pyrite indicates that 1) pyrite exsolved early from MSS and 2) the IPGE, and the small amount of Pt and As in MSS, both showed a preference for the pyrite structure over pyrrhotite or pentlandite.  The oscillatory zonation of the PGE, As, Co and Se in pyrite resulted from a boundary layer effect during its complex growth from MSS.

Chalcophile and platinum-group element (PGE) concentrations in the sulfide minerals from theMcCreedy East deposit, Sudbury, Canada, and the origin of PGE in pyrite

 Dare, S.A.S., Barnes, S-J., Hazel M. Prichard & Peter C. Fisher., 2011

 Abstract

    Magmatic sulfide deposits consist of pyrrhotite, pentlandite, chalcopyrite (± pyrite), and platinum-group minerals (PGM). Understanding the distribution of the chalcophile and platinum-group element (PGE) concentrations among the base metal sulfide phases and PGM is important both for the petrogenetic models of the ores and for the efficient extraction of the PGE. Typically, pyrrhotite and pentlandite host much of the PGE, except Pt which forms Pt minerals. Chalcopyrite does not host PGE and the role of pyrite has not been closely investigated. The Ni–Cu–PGE ores from the South Range of Sudbury are unusual in that sulfarsenide PGM, rather than pyrrhotite and pentlandite, are the main carrier of PGE, probably as the result of arsenic contribution to the sulfide liquid by the As-bearing metasedimentary footwall rocks. In comparison, the North Range deposits of Sudbury, such as the McCreedy East deposit, have As-poor granites in the footwall, and the ores commonly contain pyrite. Our results show that in the pyrrhotite-rich ores of the McCreedy East deposit Os, Ir, Ru, Rh (IPGE), and Re are concentrated in pyrrhotite, pentlandite, and surprisingly in pyrite. This indicates that sulfarsenides, which are not present in the ores, were not important in concentrating PGE in the North Range of Sudbury. Palladium is present in pentlandite and, together with Pt, form PGM such as (PtPd)(TeBi)2.
Platinum is also found in pyrite. Two generations of pyrite are present. One pyrite is primary and locally exsolved from monosulfide solid solution (MSS) in small amounts (<2 wt.%) together with pyrrhotite and pentlandite. This pyrite is unexpectedly enriched in IPGE, As (±Pt) and the concentrations of these elements are oscillatory zoned. The other pyrite is secondary and formed by alteration of the MSS cumulates by late magmatic/hydrothermal fluids. This pyrite is unzoned and has inherited the low concentrations of IPGE and Re from the pyrrhotite and pentlandite that it has replaced.

Keywords: Platinum-group elements. Sulfides. Pyrite.Oscillatory zoning. Laser ablation ICP-MS. Platinum-group minerals. Sudbury. Canada
 

Formation of Platinum-Group Minerals from an evolving sulfide liquid at Sudbury, Canada

Dare, S.A.S.,  Barnes, S.-J. , Prichard H.M. and Fisher P.C., 2011

 Abstract

    Chalcophile and platinum-group elements (PGE) are collected by a magmatic sulfide liquid and form PGE (± Ni-Cu) deposits. Early-crystallizing monosulfide solid solution (MSS) concentrates Os, Ir, Ru and Rh (IPGE) and the residual liquid concentrates Cu, Pt, Pd, Ag, As, Bi, Te, and Sn. It is important to determine the host phases of the PGE, which are sulfide minerals and/or platinum-group minerals (PGM), in order to understand the petrogenesis of the deposit and to improve PGE extraction. Previous work shows that PGM form by: 1) exsolution from base metal sulfides, 2) crystallization from the residual Cu-rich liquid and/or a late-stage immiscible melt and 3) remobilization during metamorphism or by hydrothermal fluids. We have investigated the origin of PGM from Sudbury Ni-Cu-PGE deposits by combining a PGM study with wholerock data and laser ablation-ICP-MS analysis of the sulfides. We found that a large proportion of the PGE are hosted in As-, Bi-, Te- and Sn-rich PGM, which formed over a wide range of temperatures during the evolution of the sulfide deposit. The amount of As in the initial sulfide melt, which varied according to the As content of the assimilated country rocks, was critical in determining whether As-PGM (IrAsS, RhAsS ± PtAs2) crystallized early (1200-900°C) from the sulfide liquid togther with MSS. An As-rich sulfide melt crystallizes sulfarsenides so that the co-existing MSS is depleted in Ir and Rh whereas an As-poor sulfide melt does not crystallize sulfarsenides and the IPGE remain in MSS. The amount of Bi, Te and Sn increases during sulfide fractionation so that the following Pt-Pd-minerals crystallized from the residual liquid. Solitary grains of Pd2Sn and PtSn crystallized early (1300-800°C) from the Cu-rich sulfide liquid whereas an unnamed Pt(Sn,Bi)Te phase together with numerous Pt-Pd-Bi-Te-Ag minerals crystallized as composite grains from microdroplets of an immiscible late-stage melt (1000-600°C). The small amount of Pt, Pd, Bi and Te, which partitioned into MSS, later exsolved (< 600°C) as laths of (PtPd)(BiTe)2 and PdBiTe and were remobilized during greenschist metamorphism.

Petrology in three dimensions:  Insights from high-resolution X-ray computed tomography combined with in situ mineral chemistry and mineralogy

  Godel, B. Barnes, S.J., 2011

 Abstract

    High-Resolution X-Ray Computed Tomography (HRXCT) is a non-destructive technique allowing the 3D visualization at the sample scale (few centimetres to less than 1 millimetre in size) of rock texture, crystal size, shape and mineral intergrowth inaccessible by any other methods.  Over the past few years, HRXCT has elucidated some fundamental problems in geosciences and opened new doors toward a better understanding of igneous processes and ore genesis at the micron scale.  The combination of HRXCT with conventional 2D image analysis (e.g. backscattered electron images), and in situ chemical analysis using state of the art techniques including laser-ablation ICP-MS and synchrotron based X-ray fluorescence mapping has a wide range of applications in petrology with potential benefit in mineral exploration and extraction.  Examples are: (i) New insights into the spatial distribution of platinum-group element minerals (mainly consisting of Pt-Pd sulphides, Pt-Fe alloy and Pt-Pd bismuthotellurides) in chromitites from the Merensky Reef of the Bushveld Complex (South Africa) with results illustrating the consistent relationship of PGM with the edges of magmatic sulphide blebs and chromite or silicate grain boundaries which strongly support an orthomagmatic origin for these PGM; (ii) Imaging of complex magmatic morphologies of chromite, magnetite and sulphide blebs and the determination of mineral proportions in komatiitic dunites from the Yilgarn Craton (Australia) and their relationships with whole-rock major and in situ mineral trace element (including Se, As, Te, Bi, Sb, Au, Ag, Pt, Pd, Ir, Os, Ru, Rh) chemistry providing insights on element mobility during alteration and metamorphism of komatiite-hosted nickel sulphide deposits.

 Concentration of chalcophile and siderophile elements in MORB sulphide droplets: New sulphide meltsilicate melt partition coefficients

Patten, C.,  Barnes, S.-J and  Mathez, E.A., 2011

    Abstract

We have determined the concentrations of chalcophile and siderophile elements by LA-ICP-MS from sulphide droplets and fresh glass in contact with them from MORB pillow rims. MORBs play an important role in the understanding of mantle petrogenesis, providing information on chemical fractionation of elements in the mantle. However, chalcophile element behaviour is not completely understood, partly due to the lack of data for partition coefficients between sulphide and silicate melts. Some droplets present homogenous textures and others portions rich in monosulphide solid solution (Mss) and intermediate solid solution (Iss) indicating that they have undergone crystal fractionation. For homogenous droplets, concentrations of Ni and Cu are 10 to 1%; Co and Zn 1000 to 100 ppm; Se, Te, Ag and Pb 100 to 10 ppm; Cd, Sn, Pd, Bi and Pt 10 to 1 ppm; Au, Ru and Re 1 to 0.1 ppm. For some of these elements it was also possible to obtain data in fresh glass allowing the calculation of partition coefficients. These were calculated for Ni (745± 252), Cu (1219 ± 381), Co (42 ± 5.5), Zn (3.4 ± 0.9), Sn (10.4 ± 1.8) and Pb (55.6 ± 9.3). Values for Ni, Cu and Co are in agreement with literature [1], suggesting that values for Zn, Sn and Pb are realistic.

[1] Peach et al. (1990) Sulfide melt-silicate melt distribution coefficients for noble metals & other chalcophile elements as deduced from MORB, Implications for partial melting, Geochimica et Cosmochimica Acta 54, 3379–3389.

Microstructural control on pentlandite exsolutions from monosulfide solid solution in komatiite hosted Ni sulfides from the Yilgarn Craton, Western Australia

  Vukmanovic, Z., Barnes, S.J., Reddy, S.M. and Fiorentini, M.L., 2011

    Abstract

Initial crystallization of monosulfide solid solution (MSS) from magmatic sulfide liquid occurs at about 1150˚C.  For typical Ni-rich sulfides as found in komatiites, pentlandite ((Fe, Ni)₉S₈) first starts to exsolve from MSS at around 600˚C, and continues to exsolve down to below 250˚C, changing in morphology from intergranular blebs to fine lamellae with falling temperature. Sulfide assemblage in komatiite-hosted deposits consist of intergrowths of pyrrhotite (Fe_1-xS), pentlandite, pyrite (FeS₂) and chalcopyrite (FeCuS).  To understand the relationship between these phases we used optical microscopy, electron backscatter diffraction (EBSD) and microprobe analyses on samples from high and low grade metamorphic terranes in Western Australia.  The sulfide assemblages at low metamorphic grade preserve pockets of annealed “foam” textures inherited from original post-magmatic cooling, whereas sulfides that have been exposed to higher metamorphic grade and high strain retain high-temperature deformation textures.  Pyrrhotite from both types shows deformation features (kink bands, subgrains and new grains) closely related to the presence of pyrite and/or pentlandite. EBSD is used as a tool to understand crystal-plastic deformation in both isotropic (pyrite and pentlandite) and anisotropic (pyrrhotite) minerals.  We succeeded in producing for the first time EBSD maps on pentlandite grains.  EBSD maps showed that the majority of the deformation is accommodated by pyrrhotite, followed by pentlandite and pyrite respectively.  EBSD maps of pentlandite grain show the presence of internal subgrains and enable determination of active slip systems in all the sulfide phases. Combined EBSD analysis and microprobe elemental mapping shows that some of the pentlandite exsolutions from pyrrhotite are closely related to low angle subgrain boundaries in pyrrhotite.  Ongoing research is addressing the question of whether exsolution is governed by the deformation microstructures and if so, at what stage in the cooling and deformation history does exsolution takes place.  Results are significant for the interpretation of tenor variations within nickel sulfide ores.

The location of the chalcophile and siderophile elements in platinum-group element ore deposits (a textural, microbeam and whole rock geochemical study): Implications for the formation of the deposits

Barnes, S.-J., Pritchard, H.M., Cox, R.A., Fisher, P.C., Godel, B., 2008

Abstract

    Recent analytical developments now make it possible to determine chalcophile and siderophile elements in situ in base metal sulfide minerals (BMS). Three points can be considered using these analyses: a) are the different elements preferentially concentrated in any particular BMS; b) what percentages of the siderophile and chalcophile elements are present in the BMS; c) what processes affect the distribution of the siderophile and chalcophile elements among the BMS. We have compared siderophile and chalcophile element distributions in pentlandite, chalcopyrite and pyrrhotite from platinum-rich ore deposits that have undergone different cooling rates and degrees of metamorphism to address these questions. We found that Re, Os, Ir, Ru and Rh are concentrated in both pentlandite and pyrrhotite. In addition to these elements pentlandite concentrates Ni, Co and Pd. Copper, Zn, Cd and Ag are concentrated in chalcopyrite or cubanite. Gold and Pt do not preferentially concentrate in any particular BMS, with very little of these elements located in BMS. The BMS from sulfide droplets of the Noril'sk (Russia) host almost all of the siderophile elements (except Pt and Au) and much of the Co and Ag. Platinum occurs as Pt-bearing mineral inclusions within the BMS. The droplets occur in unmetamorphosed subvolcanic sills, which would have cooled relatively quickly. The high percentage of PGE in the BMS and the close association of the Pt-minerals with the BMS suggest that the model whereby a base metal sulfide liquid collected the siderophile and chalcophile elements to form the deposit is correct. We suggest that the Pt partitioned into the sulfide liquid and could have partitioned into the BMS at high temperatures, but that a lower temperatures the BMS structure would not accommodate the Pt and Pt-minerals exsolved during cooling. Alternatively, if Pt could not partition into the BMS then Pt would have concentrated in the fractionated sulfide liquid and crystallized as Pt-minerals from the final liquid. In the platinum-group element (PGE) reefs of unmetamorphosed layered intrusions (Busveld Complex, South Africa and Great Dyke, Zimbabwe) 30 to 60% of the siderophile elements (except Pt and Au) are present in pentlandite, pyrrhotite and chalcopyrite. The balance is found in platinum-group minerals (PGM), which occur associated with the BMS. We suggest that the reason that a larger percentage of PGE are in the form of PGM is the result of the slower cooling of the BMS in the layered intrusion, which would allow more time for exsolution of the PGE than in the case of the BMS from subvolcanic sills. In the PGE-reefs from the metamorphosed layered intrusion (Penikat, Finland) the percentage of siderophile elements present in BMS covers a larger range, of 8 to 70%. There are many more PGM present and there has been extensive recrystallization of the BMS. Most of the PGM are associated with BMS, but in many cases the Pd-bismuthotellurides and in a few cases the Pt-sulfarsenides and -arsenides are not found associated with BMS. Three processes could possibly have led to this. The BMS, which originally contained the PGM as exsolutions, dissolved during metamorphism, leaving behind insoluble PGM. Alternatively the bismuthotellurides, arsenides and sulfarsenides could have been locally remobilized from the BMS into the surrounding silicates during metamorphism. The third possibility is that Pd and to a lesser extent Pt could have been introduced to the PGE reef by fluids and precipitated as PGM from the fluid.


Total sulfur concentrations in geological reference materials by elemental infrared analyser

Bedard, LP, Savard, D. and Barnes, S-J., 2008

Abstract

    Total sulfur is an analyte for which there are few determinations published, despite the fact that it is a very important element (e.g., a major element in most ores, an important gas constituent in global warming, an active participant in acid drainage). Most geological reference materials have very poor quality sulfur results, that is with relative standard deviations (RSD) in the range of 30-50%, even for concentrations over 100 µg g-1 S, which compromises their use as calibrators. In order to provide modern results with low RSD, sulfur was determined in twenty-nine geological reference materials with a state-of-the-art elemental S/C analyser using metal chips (certified reference materials with a traceability link) and analytical grade sulfur for high concentration samples. Analytical parameters (sample mass, crucible degassing, calibration strategy, etc.) were optimised by testing. Our results agreed with reference material values provided by issuing bodies. Results for CCRMP SY-2 (129 ± 13 µg g-1 S), which has been proposed as a sulfur reference material, were in agreement with the proposed modern value of 122 ± 3.7 µg g-1 S.

Résumé

    Le soufre total est un analyte pour lequel il existe très peu de déterminations publiées, bien que cet élément soit très important (par exemple, c'est un élément majeur de la plupart des minerais, un gaz très important dans le problème de réchauffement global et un participant actif au problème des eaux acides). La plupart des matériaux de référence certifiés ont des données sur le soufre de très mauvaise qualité, avec des écart-types relatifs (RSD) de l'ordre de 30 à 50%, même pour des concentrations supérieures à 100 µg g-1, ce qui compromet leur utilisation comme calibrant. Afin de fournir des résultats assortis de très faibles RSD, le soufre a été mesuré dans vingt neuf matériaux de référence certifiés avec un analyseur S/C élémentaire de dernière génération, en utilisant des fragments de métal (matériaux de référence certifiés ayant un lien de traçabilité) et du soufre de qualité analytique pour les échantillons les plus concentrés. Les paramètres analytiques (poids de l'échantillon, dégazage du creuset, stratégie de calibration, etc.) ont été optimisés via une série de tests. Nos résultats sont en bon accord avec les valeurs données pour les matériaux de référence par leurs organismes de certification respectifs. La concentration mesurée (129 ± 13 µg g-1 S) pour CCRMP SY-2 qui a été proposé comme matériau de référence pour le soufre, est en accord avec la valeur actuelle de 122 ± 3.7 µg g-1 S.


Image analysis and composition of platinum-group minerals in the J-M reef, Stillwater Complex (U.S.A.)

Godel, B. and Barnes, S-J., 2008

Abstract
Detailed imaging, bulk geochemical analyses, and laser ablation ICP-MS analyses were conducted on mineralized samples from the platinum-group element (PGE)–bearing J-M reef, Stillwater Complex (Montana). The aims of the study were to determine which phases are the main hosts of the PGE, to examine the textural relationships of the PGE-bearing minerals, and the associated base metal sulfides, silicates, and secondary magnetites. The platinum-group minerals (PGM) observed in the studied samples consist of Pd ± Pt sulfides, Pt-Fe alloy (isoferroplatinum), Pd ± Pt tellurides, and Pd-Cu alloy (skaergaardite) with minor native Pd, [Ru(Ir, Os)S2] (laurite), and Au-Pd-Ag alloy (palladian electrum). These minerals account for all the Pt, half the Pd, Ru, and Ir, and a smaller proportion of the Os in whole rocks (the balance being found in the base metal sulfide). Most of these PGM are closely associated with base metal sulfides, either included in the base metal sulfide or located at the contact between the base metal sulfide and the silicate or oxide minerals. The textures of the PGM within and at the margins of the base metal sulfide suggest that they exsolved from the sulfides. We suggest that the PGE, Ni, Cu, and Te partitioned into a magmatic sulfide liquid that crystallized into base metal sulfide. The Pt-Fe alloys and some of the Pd sulfides exsolved from the base metal sulfide during desulfurization at fairly high temperatures. In contrast, the skaergaardite and some Pd sulfides are found in association with secondary magnetite and probably formed at relatively low temperatures (~250°–465°C) during a second fluid migration event. The Pd/Pt ratio of the reef (~3.3) is higher than most mafic magmas and the Pd/Pt ratio of the footwall is lower (0.3–0.7). It is possible that some of the Pd in the reef was leached from the underlying rocks and deposited in the reef as PGM associated with secondary magnetite.



Platinum-group elements in sulfide minerals and the whole rock of the J-M Reef (Stillwater complex): Implication for the formation of the reef

Godel, B. and Barnes, S-J., 2008

Abstract
The concentration of platinum-group elements (PGE), Co, Ni, Cu, Re, Au and Ag was determined in base metal sulfides (BMS) minerals and whole rocks of the J-M Reef (footwall, reef and hanging wall) of the Stillwater Complex (U.S.A.). The aims of the study were to establish: (i) whether the BMS minerals (pyrrhotite, pentlandite, and chalcopyrite) are the principal host of these elements; (ii) whether these elements preferentially partition into a specific BMS mineral. The results of this study allowed us to consider and evaluate: the possible parental magma composition, the role of sulfide liquid and the role of alteration in the formation of the J-M Reef.

Only a minor quantity of PGE ( 10 to 140 ppb) were found in the footwall and hanging wall samples of the J-M Reef, which contain only small amounts of sulfur ( 140 to 400 ppm). In contrast, the reef samples enriched in sulfides contain higher values of PGE (49 to 419 ppm). The S, Ni, Cu, and PGE contents follow the same trends indicating that BMS minerals are the principal phases controlling these elements. Palladium and Pt are the more abundant PGE (up to 244 ppm for Pd and 166 ppm for Pt) with an average Pd/Pt ratio of 3.3. Similarly, the Pd/Ir ratio and the mantle normalized PGE patterns indicate that Pd is more strongly enriched in the reef relative to surrounding rocks than the other PGE.

Pentlandite is the BMS mineral that contains the largest weight fraction of PGE. Palladium represents 95% of the PGE found in solution in BMS minerals and is mainly partitioned in pentlandite. In contrast, Pt is almost exclusively found as platinum-group minerals and do not partition in any BMS minerals. The other PGE are largely found in the BMS minerals, mainly pentlandite.

The variation of S/Se ratios and the presence of secondary magnetite in some samples indicate that these samples may have lost 20 to 50% of their original S. The highest Pd (and by analogy PGE contents) are found in samples containing secondary magnetite, but these are not necessarily those which have experienced the highest S removal. This indicates that the processes which caused the removal of S and the alteration to magnetite are probably distinct.

Our modeling results suggest that PGE (and to a greater extent Pd) enrichment may be summarized into two different steps. First, an immiscible sulfide liquid interacted with a large volume of magma with PGE composition close to high-Mg basalt and collected the PGE. The sulfide liquid percolated down through the crystal mush to collect at a level where the porosity did not permit any further migration. During cooling, magma chamber instabilities triggered the partial desulfurization of the sulfides. Finally, in some parts of reef, a fluid deposited Pd (possibly removed from the footwall) and altered the BMS minerals to magnetite. During this step, Pd was precipitated as an alloy in the most Pd-enriched samples and Pd possibly diffused into pentlandite forming a high-Pd bearing pentlandite. The different contributions of each of the above processes may explain the variability of the Pd grade observed along the J-M Reef.



Platinum-group elements in Merensky and JM reefs: A review of recent studied

Godel, B., Maier, W.D. and Barnes, S-J., 2008

Abstract
The Merensky Reef of the Bushveld Complex contains one of the world’s largest concentrations of platinum-group elements (PGE). We have investigated ‘normal’ reef, its footwall and its hanging wall at Impala Platinum Mines. The Reef is 46 cm thick and consists from bottom to top of leuconorite, anorthosite, chromitite and a very coarse-grained melanorite. The footwall is leuconorite and the hanging wall is melanorite. The only hydrous mineral present is biotite, which amounts to 1%, or less, of the rock. All of the rocks contain 0•1–5% interstitial sulphides (pyrrhotite, pentlandite and chalcopyrite), with the Reef rocks containing the most sulphides (1–5%). Lithophile inter-element ratios suggest that the magma from which the rocks formed was a mixture of the two parental magmas of the Bushveld Complex (a high-Mg basaltic andesite and a tholeiitic basalt). The Reef rocks have low incompatible element contents indicating that they contain 10% or less melt fraction. Nickel, Cu, Se, Ag, Au and the PGE show good correlations with S in the silicate rocks, suggesting control of the abundance of these metals by sulphides. The concentration of the chalcophile elements and PGE in the silicate rocks may be modelled by assuming that the rocks contain sulphide liquid formed in equilibrium with the evolving silicate magma. It is, however, difficult to model the Os, Ir, Ru, Rh and Pt concentrations in the chromitites by sulphide liquid collection alone, as the rocks contain 3–4 times more Os, Ir, Ru, Rh and Pt than the sulphide-collection model would predict. Two possible solutions to this are: (1) platinum-group minerals (PGM) crystallize from the sulphide liquid in the chromitites; (2) PGM crystallize directly from the silicate magma. To model the concentrations of Os, Ir, Ru, Rh and Pt in the chromitites it is necessary to postulate that in addition to the 1% sulphides in the chromitites there is a small quantity (0•005%) of cumulus PGM (laurite, cooperite and malanite) present. Sulphide liquids do crystallize PGM at low fS2. Possibly the sulphide liquid that was trapped between the chromite grains lost some Fe and S by reaction with the chromite and this provoked the crystallization of PGM from the sulphide liquid. Alternatively, the PGM could have crystallized directly from the silicate magma when it became saturated in chromite. A weakness of this model is that at present the exact mechanism of how and why the magma becomes saturated in PGM and chromite synchronously is not understood. A third model for the concentration of PGE in the Reef is that the PGE are collected from the underlying cumulus pile by Cl-rich hydrous fluids and concentrated in the Reef at a reaction front. Although there is ample evidence of compaction and intercumulus melt migration in the Impala rocks, we do not think that the PGE were introduced into the Reef from below, because the rocks underlying the Reef are not depleted in PGE, whereas those overlying the Reef are depleted. This distribution pattern is inconsistent with a model that requires introduction of PGE by intercumulus fluid percolation from below.



The metal rich portions of the phase diagram Cu-Fe-Pd-S at 1000 ºC, 900 ºC and 725º C: Implications for mineralization in the Skaergaard intrusion

Karup-Moller, S., Makovicky, E. and Barnes, S-J., 2008

Abstract
The sulphur-poor portions of the dry condensed Cu-Fe-Pd-S system were studied at 1000ºC, 900ºC and 725ºC by synthesis in evacuated silicate glass tubes, along with textural observations and electron microprobe analyses of equilibrated reaction products. Sulphide melt coexists with Cu-Fe-Pd alloys, bornite, Fe1–xS and iss (intermediate solid solution, Cabri, 1973) and Pd4S. Compositional data were obtained for the associations bornite-alloy-melt, pyrrhotite-alloy-melt and for immiscible Cu-rich sulphide melts. Partition coefficients for all three metals were derived for the association alloy-melt. Formation of the two new Cu-Pd alloy minerals, skaergaardite and nielsenite, is discussed in terms of the present findings.



The composition of magmatic Ni–Cu–(PGE) sulfide deposits in the Tati and Selebi-Phikwe belts of eastern Botswana

Maier, W.D., Barnes, S-J.,Chinyepi, G, Barton, J. M., Eglington, B and I. Setshedi, 2008

Abstract
We studied a number of magmatic Ni Cu (PGE) sulfide deposits in two distinct belts in eastern Botswana. The Tati belt contains several relatively small deposits (up to 4.5 Mt of ore at 2.05% Ni and 0.85% Cu) at Phoenix, Selkirk and Tekwane. The deposits are hosted by ca 2.7 Ga, low- to medium-grade metamorphosed gabbroic troctolitic intrusions situated within or at the periphery of a greenstone belt. The deposits of the Selebi-Phikwe belt are larger in size (up to 31 Mt of ore grade). They are hosted by high-grade metamorphosed gabbronorites, pyroxenites and peridotites believed to be older than ca 2.0 Ga that intruded gneisses of the Central Zone of the Limpopo metamorphic belt. The composition of the sulfide mineralisation in the two belts shows systematic variation. Most of the mineralisation in the Tati belt contains 2 9% Ni and 0.05 4% Cu (Cu/Cu + Ni = 0.4 0.7), whereas most of the mineralisation in the Selebi-Phikwe belt contains 1 3% Ni and 0.1 4% Cu (Cu/Cu + Ni = 0.4 0.9). The Cu Ni tenors of the ores in both belts are consistent with crystallization from a basaltic magma. The Tati ores contain mostly >3 ppm Pt + Pd (Pt/Pd 0.1 1), with Pd/Ir = 100 1,000, indicative of a differentiated basaltic magma that remained S-undersaturated before emplacement. Most of the Selebi-Phikwe ores have <0.5 ppm Pt + Pd (Pt/Pd < 0.1 1), with Pd/Ir = 10 500. This suggests a relatively less differentiated magma that reached S saturation before emplacement. The Tati rocks show flat mantle-normalised incompatible trace element patterns (average Th/YbN = 1.57), except for strong enrichments in large ion lithophile elements (Cs, Rb, Ba, U, K). Such patterns are characteristic of relatively uncontaminated oceanic arc magmas and suggest that the Tati intrusions were emplaced in a destructive plate margin setting. Most of the Selebi-Phikwe rocks (notably Dikoloti) have more fractionated trace element signatures (average Th/YbN = 4.22), possibly indicating digestion of upper crustal material during magma emplacement. However, as there are also samples that have oceanic arc-like signatures, an alternative possibility is that the composition of most Selebi-Phikwe rocks reflects tectonic mingling of the intrusive rocks with the country rocks. The implication is that orogenic belts may have a higher prospectivity for magmatic Ni Cu ores than presently recognised. The trigger mechanism for sulfide saturation and segregation in all intrusions remains unclear. Whereas the host rocks to the intrusions appear to be relatively sulfur poor, addition of crustal S to the magmas is suggested by low Se/S ratios in some of the ores (notably at Selebi-Phikwe). External S sources may thus remain unidentified due to poor exposure and/or S mobility in response to metamorphism.



Early Kibaran rift-related mafic–ultramafic magmatism in western Tanzania and Burundi: Petrogenesis and ore potential of the Kapalagulu and Musongati layered intrusions

Maier, W.D., Barnes, S-J., Bandyayera, D. Livesey, T, Li, C. and Ripley, E., 2008

Abstract
The Kapalagulu and Musongati intrusions are differentiated mafic–ultramafic intrusions, more than 1 km in stratigraphic thickness and several 10 s of km2 in size. They form part of the Kabanga–Musongati belt of intrusions in western Tanzania and Burundi. The intrusions of the Kabanga–Musongati belt were emplaced at ca 1.4 Ga into pelitic sediments of the Burundi and Karagwe–Ankolean Supergroups that accumulated during an early rifting phase of the Kibaran orogeny. The parental magmas to the intrusions were of picritic composition (ca 15% MgO) that assimilated variable amounts of sulfidic sedimentary rocks during emplacement. Modeling suggests that the Musongati magma assimilated ca. 5% of sedimentary material, whereas the Kapalagulu magma assimilated ca. 15% of sediment. Contamination caused enrichment of the magma and the cumulates in incompatible trace elements, the development of negative Nb–Ta–Ti anomalies, and crustal sulfur isotopic signatures (d34S = + 4.5 to + 20). At Kapalagulu, contamination of the parent magma led to the formation of basal olivine melanorite cumulates. In the less contaminated Musongati intrusion dunites and harzburgites formed at the base. Both intrusions are prospective for magmatic Ni and PGE deposits. This is indicated by empirical observations, notably the presence of important Ni sulfide ores at Kabanga and reef-type PGE concentrations at Musongati and Kapalagulu. It is also supported by theoretical considerations, namely the high-magnesian composition of the parental magmas and the abundance of sulfides in the host sedimentary rocks. Weathering of the ultramafic rocks resulted in a thick lateritic crust that contains up to > 4 ppm PGE and, at Musongati, hosts one the world's largest Ni-laterite deposits.



Petrogenesis of contact-style PGE mineralization in the northern lobe of the Bushveld Complex: comparison of data from the farms Rooipoort, Townlands, Drenthe and Nonnenwerth

Maier, W.D., de Klerk, L., Blaine, J., Manyeruke, M., Barnes, S-J., Stevens, M.V.A. and Mavrogenes, J. A., 2008

Abstract
In the present study, we document the nature of contact-style platinum-group element (PGE) mineralization along >100 km of strike in the northern lobe of the Bushveld Complex. New data from the farm Rooipoort are compared to existing data from the farms Townlands, Drenthe, and Nonnenwerth. The data indicate that the nature of the contact-style mineralization shows considerable variation along strike. In the southernmost portion of the northern Bushveld, on Rooipoort and adjoining farms, the mineralized sequence reaches a thickness of 700 m. Varied-textured gabbronorites are the most common rock type. Anorthosites and pyroxenites are less common. Chromitite stringers and xenoliths of calcsilicate and shale are largely confined to the lower part of the sequence. Layering is locally prominent and shows considerable lateral continuity. Disseminated sulfides may reach ca. 3 modal % and tend to be concentrated in chromitites and melanorites. Geochemistry indicates that the rocks can be correlated with the Upper Critical Zone. This model is supported by the fact that, in a down-dip direction, the mineralized rocks transform into the UG2-Merensky Reef interval. Between Townlands and Drenthe, the contact-mineralized sequence is thinner (up to ca. 400 m) than in the South. Chromitite stringers occur only sporadically, but ultramafic rocks (pyroxenites, serpentinites, and peridotites) are common. Xenoliths of calcsilicate, shale, and iron formation are abundant indicating significant assimilation of the floor rocks. Sulfides may locally form decimeter- to meter-sized massive lenses. PGE grades tend to be higher than elsewhere in the northern Bushveld. The compositions of the rocks show both Upper Critical Zone and Main Zone characteristics. At Nonnenwerth, the mineralized interval is up to ca. 400 m thick. It consists largely of varied-textured gabbronorites, with minor amounts of igneous ultramafic rocks and locally abundant and large xenoliths of calcsilicate. Layering is mostly weakly defined and discontinuous. Disseminated sulfides (



Trace element concentrations in apatites from the Sept-Îles Intrusive Suite, Canada — Implications for the genesis of nelsonites

Tollari, N., Barnes, S.-J, Cox, R.A. and Nabil H., 2008

Abstract
Apatite-oxide-rich rocks (e.g. nelsonite) occur in a wide variety of rock types. Nelsonites occur in felsic and calco-alkaline granitic rocks but are more common in mafic systems, layered intrusions and massif type anorthosites. Models for their formation are currently highly debated. They could form by crystallization of apatite and oxide from a fractionated magma, and accumulation on the crystal pile. Alternatively, an Fe-Ti-P-rich liquid may segregate from the fractionated magma and the nelsonite could crystallize from this liquid. In order to provide further constraints on the formation processes of nelsonites in layered intrusions, we have studied the petrography, the major and trace element contents of apatite and whole rock in 14 apatite-oxide-rich rocks from the Sept-Îles Intrusive Suite (Canada). The mineralogical association and mineral compositions (REE, Cl content in apatite and MgO content in ilmenite) are consistent with the crystal accumulation model. Combining the in situ LA-ICP-MS analyses of the apatite and the major element content of the Fe-Ti oxides, the trace element modeled concentrations in the liquid from which the nelsonites could have formed are similar to the dykes at the margins of the Sept-Îles Intrusive Suite. The PELE software was used to estimate the composition of the magma derived from the dykes at the time of apatite crystallization. Our results support the crystal accumulation hypothesis for the formation of the nelsonites and associated rocks.



Experimental effects of pressure and fluorine on apatite saturation in mafic magmas, with reference to layered intrusions and massif anorthosites

Tollari, N, Baker, D. and Barnes, S-J., 2008

Abstract
Apatite is a cumulate phase in the upper parts of some mafic layered intrusions and anorthositic complexes. We investigated the effect of pressure and fluorine on apatite saturation in mafic magmas to better understand under which conditions this mineral crystallizes. Apatite saturation gives information about the formation of silicate rocks, and is of interest in explaining the formation of apatite oxide-rich rocks (e.g. nelsonites comprising approximately, one-third apatite and two-third Fe Ti oxide). Two models of formation are proposed for this rock type: crystal fractionation followed by accumulation of apatite and Fe Ti oxides and liquid immiscibility. New experiments carried out with mafic compositions at 500 MPa confirm that the most important variables on phosphate saturation are SiO2 and CaO. Fluorine addition leads to apatite saturation at lower SiO2 and higher CaO concentrations. Comparison of our results with those of previous experimental studies on liquid liquid immiscibility at upper-to-mid-crustal conditions allows us to investigate the relative importance of apatite saturation versus liquid liquid immiscibility in the petrogenesis of nelsonites and similar rocks. The liquid line of descent of three natural examples studied (the Sept-Îles intrusive suite, the anorthositic Complex of the Lac-St-Jean and the Skaergaard layered intrusion) do not cross the liquid liquid immiscibility field before they reach apatite saturation. Thus, the apatite oxide-rich rock associated with these three intrusive suites are best explained by crystal fractionation followed by accumulation of apatite and Fe Ti oxides.



Platinum-group elements in sulphide minerals, platinum-group minerals, and the whole rocks of the Merensky Reef (Bushveld Complex, South Africa): Implication for the Formation of the Reef

Godel, B. Barnes, S.-J. and Maier, W. D., 2007

Abstract
The concentrations of platinum-group elements (PGE), Co, Re, Au and Ag have been determined in the base-metal sulphide (BMS) of a section of the Merensky Reef. In addition we performed detailed image analysis of the platinum-group minerals (PGM). The aims of the study were to establish: (1) whether the BMS are the principal host of these elements; (2) whether individual elements preferentially partition into a specific BMS; (3) whether the concentration of the elements varies with stratigraphy or lithology; (4) what is the proportion of PGE hosted by PGM; (5) whether the PGM and the PGE found in BMS could account for the complete PGE budget of the whole-rocks. In all lithologies, most of the PGE ( 65 up to 85%) are hosted by PGM (essentially Pt–Fe alloy, Pt–Pd sulphide, Pt–Pd bismuthotelluride). Lesser amounts of PGE occur in solid solution within the BMS. In most cases, the PGM occur at the contact between the BMS and silicates or oxides, or are included within the BMS. Pentlandite is the principal BMS host of all of the PGE, except Pt, and contains up to 600 ppm combined PGE. It is preferentially enriched in Pd, Rh and Co. Pyrrhotite contains, Rh, Os, Ir and Ru, but excludes both Pt and Pd. Chalcopyrite contains very little of the PGE, but does concentrate Ag and Cd. Platinum and Au do not partition into any of the BMS. Instead, they occur in the form of PGM and electrum. In the chromitite layers the whole-rock concentrations of all the PGE except Pd are enriched by a factor of five relative to S, Ni, Cu and Au. This enrichment could be attributed to BMS in these layers being richer in PGE than the BMS in the silicate layers. However, the PGE content in the BMS varies only slightly as a function of the stratigraphy. The BMS in the chromitites contain twice as much PGE as the BMS in the silicate rocks, but this is not sufficient to explain the strong enrichment of PGE in the chromitites. In the light of our results, we propose that the collection of the PGE occurred in two steps in the chromitites: some PGM formed before sulphide saturation during chromitite layer formation. The remaining PGE were collected by an immiscible sulphide liquid that percolated downward until it encountered the chromitite layers. In the silicate rocks, PGE were collected by only the sulphide liquid.



Petrogenesis of contact-style PGE mineralization in the northern lobe of the Bushveld Complex: comparison of data from the farms Rooipoort, Townlands, Drenthe and Nonnenwerth

Maier, W.D., de Klerk, L, Blaine, J. Manyeruke, T. Barnes, S.-J., . Stevens, M. V. A and Mavrogenes, J. A., 2007

Abstract
In the present study, we document the nature of contact-style platinum-group element (PGE) mineralization along >100 km of strike in the northern lobe of the Bushveld Complex. New data from the farm Rooipoort are compared to existing data from the farms Townlands, Drenthe, and Nonnenwerth. The data indicate that the nature of the contact-style mineralization shows considerable variation along strike. In the southernmost portion of the northern Bushveld, on Rooipoort and adjoining farms, the mineralized sequence reaches a thickness of 700 m. Varied-textured gabbronorites are the most common rock type. Anorthosites and pyroxenites are less common. Chromitite stringers and xenoliths of calcsilicate and shale are largely confined to the lower part of the sequence. Layering is locally prominent and shows considerable lateral continuity. Disseminated sulfides may reach ca. 3 modal % and tend to be concentrated in chromitites and melanorites. Geochemistry indicates that the rocks can be correlated with the Upper Critical Zone. This model is supported by the fact that, in a down-dip direction, the mineralized rocks transform into the UG2-Merensky Reef interval. Between Townlands and Drenthe, the contact-mineralized sequence is thinner (up to ca. 400 m) than in the South. Chromitite stringers occur only sporadically, but ultramafic rocks (pyroxenites, serpentinites, and peridotites) are common. Xenoliths of calcsilicate, shale, and iron formation are abundant indicating significant assimilation of the floor rocks. Sulfides may locally form decimeter- to meter-sized massive lenses. PGE grades tend to be higher than elsewhere in the northern Bushveld. The compositions of the rocks show both Upper Critical Zone and Main Zone characteristics. At Nonnenwerth, the mineralized interval is up to ca. 400 m thick. It consists largely of varied-textured gabbronorites, with minor amounts of igneous ultramafic rocks and locally abundant and large xenoliths of calcsilicate. Layering is mostly weakly defined and discontinuous. Disseminated sulfides ( < ca. 3 modal %) occur throughout much of the sequence. Geochemistry indicates that the rocks crystallized mainly from tholeiitic magma and thus have a Main Zone signature. The implication of our findings is that contact-style PGE mineralization in the northern lobe of the Bushveld Complex cannot be correlated with specific stratigraphic units or magma types, but that it formed in response to several different processes. At all localities, the magmas were contaminated with the floor rocks. Contamination with shale led to the addition of external sulfur to the magma, whereas contamination with dolomite may have oxidized the magma and lowered its sulfur solubility. In addition to contamination, some of the magmas, notably those of Upper Critical Zone lineage present at the south-central localities, contained entrained sulfides, which precipitated during cooling and crystallization.



Platinum-group element, Gold, Silver and Base Metal distribution in compositionally zoned sulfide droplets from the Medvezky Creek Mine, Norilsk, Russia

Barnes, S-J, Cox, R. A., Zientek, M. L., 2006

Abstract
Concentrations of Ag, Au, Cd, Co, Re, Zn and Platinum-group elements (PGE) have been determined in sulfide minerals from zoned sulfide droplets of the Noril’sk 1 Medvezky Creek Mine. The aims of the study were; to establish whether these elements are located in the major sulfide minerals (pentlandite, pyrrhotite, chalcopyrite and cubanite), to establish whether the elements show a preference for a particular sulfide mineral and to investigate the model, which suggests that the zonation in the droplets is caused by the crystal fractionation of monosulfide solid solution (mss). Nickel, Cu, Ag, Re, Os, Ir, Ru, Rh and Pd, were found to be largely located in the major sulfide minerals. In contrast, less than 25% of the Au, Cd, Pt and Zn in the rock was found to be present in these sulfides. Osmium, Ir, Ru, Rh and Re were found to be concentrated in pyrrhotite and pentlandite. Palladium and Co was found to be concentrated in pentlandite. Silver, Cd and Zn concentrations are highest in chalcopyrite and cubanite. Gold and platinum showed no preference for any of the major sulfide minerals. The enrichment of Os, Ir, Ru, Rh and Re in pyrrhotite and pentlandite (exsolution products of mss) and the low levels of these elements in the cubanite and chalcopyrite (exsolution products of intermediate solid solution, iss) support the mss crystal fractionation model, because Os, Ir, Ru, Rh and Re are compatible with mss. The enrichment of Ag, Cd and Zn in chalcopyrite and cubanite also supports the mss fractionation model these minerals are derived from the fractionated liquid and these elements are incompatible with mss and thus should be enriched in the fractionated liquid. Gold and Pt do not partition into either iss or mss and become sufficiently enriched in the final fractionated liquid to crystallize among the iss and mss grains as tellurides, bismithides and alloys. During pentlandite exsolution Pd appears to have diffused from the Cu-rich portion of the droplet into pentlandite.



3-D Distribution of Sulphide Minerals in the Merensky Reef (Bushveld Complex, South Africa) and the J-M Reef (Stillwater Complex, USA) and their Relationship to Microstructures Using X-Ray Computed Tomography

Godel, B., Barnes, S-J. and Maier, W.D., 2006

Abstract
Large mafic–ultramafic layered intrusions may contain layers enriched in platinum-group elements (PGE). In many cases, the PGE are hosted by disseminated sulphides. We have investigated the distribution of the sulphides in three dimensions in two oriented samples of the Merensky Reef and the J-M Reef. The aim of the study was to test the hypothesis that the sulphides crystallized from a base metal sulphide liquid that percolated through the cumulate pile during compaction. The distribution of sulphides was quantified using: (1) X-ray computed tomography; (2) microstructural analysis of polished thin sections oriented parallel to the paleovertical; (3) measurement of dihedral angles between sulphides and silicates or oxides. In the Merensky Reef and the J-M Reef, sulphides are connected in three dimensions and fill paleovertical dilatancies formed during compaction, which facilitated the downward migration of sulphide liquid in the cumulate. In the melanorite of the Merensky Reef, the sulphide content increases from top to bottom, reaching a maximum value above the underlying chromitite layer. In the chromitite layers sulphide melt connectivity is negligible. Thus, the chromitite may have acted as a filter, preventing extensive migration of sulphide melt downwards into the footwall. This could partially explain the enrichment in PGE of the chromitite layer and the observed paucity of sulphide in the footwall.



An experimental study of mass transfer of platinum-group elements, gold, nickel and copper in sulfur-dominated vapor at magmatic temperatures

Peregoedova, A., Barnes, S-J., and Baker, D.R., 2006

Abstract
We report results of an experimental study on platinum-group elements (PGE) and Au mass transfer by an S-vapor in the Fe-Ni-Cu sulfide system at magmatic temperatures. Using the tube-in-tube technique, we have examined the quantity of PGE, Au and base-metals (BM) transferred via the vapor from the PGE donor [S-rich (Fe,Ni,Cu)1-x S doped with about 2000 ppm of each PGE and Au] to a S-poor PGE-free pyrrhotite (Po) used as the PGE receiver. At the end of the experiments, the receiver Po contained significant quantities ofNi, Cu, Au, Pt and Pd, but little Ir, Ru and Rh. The most important factors influencing the vapor mobility of Ni, Cu, Au and PGE are the phase assemblage present in the donor, which in turn is controlled by temperature, S, Ni and Cu content of the system, and the sulfur fugacity (fS2). In experiments containing only Au-alloy and monosulfide solid-solution (Mss) in the donor, Cu is transferred more than Ni. Gold is transferred 10 times more efficiently than Pd and Pt, and Pd and Pt are transferred 10 times more efficiently than Rh, Ru and Ir. At the lowest S contents, when Pt-alloys form in the donor system, Pt is transferred less than Pd. In experiments containing Mss and sulfide liquid, the amount of all PGE transferred is higher, although the order remains the same Pd-Pt >Rh-Ru-Ir. The amount ofNi transferred is higher than in the Mss-alloy system. In contrast, the amount of Au transferred is lower where the sulfide liquid is present. The amount of S in the receiver Po may be used as a proxy for fS2. Within each phase assemblage the amount of Ni and PGE transferred increases with the amount of S in the receiver Po suggesting that these elements are transferred as sulfide complexes. In contrast, Cu and Au show no correlation with fS2 suggesting that these elements are transferred as metals. These results are not directly applicable to natural systems because the simplicity of the experiments. Nevertheless three geological systems where these results could be relevant are be considered. Firstly, in the case of PGE deposits some are enriched in Pd, Cu and Au, while some are rich in Pt. According to our experiments this could occur if a magmatic sulfide underwent S devolatilization to form S-poor Mss plus Pt-alloy (to form the Pt-rich deposits) and the vapor transported Cu, Au and Pd to a lower pressure and temperature site and deposited these metals there to form the Pd-rich deposits. Secondly, in the case of magmatic Ni-Cu sulfide deposits, which in many cases are enriched in Cu and Pd; S, Cu and Pd could be transferred from the country rock as a vapor leaving Mss and Pt-alloy in the country rock. Finally, during S devolatilization of mantle nodules, Pd, Cu and Au could be removed by the vapor.



TCF selenium preconcentration in geological materials for determination at sub-microg g-1 with INAA (Se/TCF-INAA)

Savard, D., Bedard, P., and Barnes, S-J., 2006

Abstract
In geological samples, Se concentration ranges from 1 × 10-9 g g-1 up to 1 × 10-3 g g-1. The analytical difficulty at low concentration (<1 µg g-1), is one of the main reasons why the geological cycle of Se is poorly known. The analytical method that consisted of preconcentration of Se with thiol cotton fiber (TCF) followed by graphite furnace atomic absorption spectrometry (GFAAS) has been modified by finishing with instrumental neutron activation analysis (INAA). The modified technique involves sample dissolution (HF-HNO3-H2O2) and evaporation to dryness at low temperature (55–60 °C) to avoid selenium volatilization. SeVI is converted to SeIV by adding 6 M HCl to the dry residuum and the solution is then heated in a covered boiling bath (95–100 °C). The solution is diluted to obtain 0.6 M HCl and then collected on TCF. The TCF is placed in a polyethylene vial for irradiation in the SLOWPOKE II reactor (Montréal) for 30 s at a neutron flux of 1015 m-2 s-1. The 162 keV peak of 77mSe (half-life 17.36 s) is read for 20 s after a decay of 7 s. The amount of sample to be dissolved is controlled by two competing effects. To obtain low detection limits, a larger amount of sample should be dissolved. On the other hand, the TCF could become saturated with chalcophile elements when large sample is used. Sulfur is a good indicator of the amount of Se and chalcophile elements present. In S poor sample (<100 µg g-1) 3.0 g of sample was used and the LD was 2 ng g-1. In S high samples (>1.5% S) 0.05 g of sample was used and the LD was 120 ng g-1. The present work also includes suggested Se concentration for eight international geological reference materials (IGRM) that compare favorably with literature values.



Predicting phosphate saturation in silicate magmas: An experimental study of the effects of melt composition and temperature.

Tollari, N., Toplis, M.J., and Barnes, S.-J., 2006

Abstract
A series of 1 atm experiments has been performed to test the influence of iron content and oxidation state on the saturation of phosphate minerals in magmatic systems. Four bulk compositions of different iron content have been studied. The experiments cover a range of temperature from 1030 to 1070 °C and oxygen fugacity from 1.5 log units below to 1.5 log units above the Fayalite–Magnetite–Quartz buffer. The results demonstrate that neither iron content of the liquid nor oxidation state play a significant role on phosphate saturation. On the other hand, SiO2 and CaO contents of the liquid strongly influence the appearance of a crystalline phosphate. Our results are combined with data from the literature to define an equation which predicts the P2O5 content of silicate liquids saturated in either whitlockite or fluorapatite: ,

where M represents the molar percentage of the relevant oxides and T is temperature in K. This equation is valid over extremely wide ranges of liquid composition and temperature (e.g., M SiO2 from 10% to 80%), including peraluminous liquids. The equation is used to illustrate the relative effects of melt chemistry and temperature on phosphate saturation, both in general terms and in particular for the case of ferrobasaltic differentiation relevant to the late stage differentiation of mafic layered intrusions. It is concluded that magmatic liquids may reach high concentrations in both iron and phosphorus, not through direct association of P5+ and Fe3+, but rather as a consequence of the variations of CaO and SiO2 content of the liquid. These results may help explain the petrogenesis of certain enigmatic rock types dominated by association of apatite and iron–titanium oxides, such as nelsonites.



Formation of magmatic nickel-sulfide ore deposits and processses affecting their copper and platinum-group element contents.

Barnes, S-J. and Lightfoot, P.C., 2005

Abstract
Nickel-copper sulfide ore deposits are found at the base of mafic and ultramafic bodies. All their host rocks, except the Sudbury Igneous Complex, are thought to be mantle-derived melts. The Sudbury Igneous Complex is thought to be the product of complete melting of continental crust.

In the case of mantle-derived magmas, a high degree of partial melting of the mantle serves to enrich the silicate magma in Ni and platinum group elements (PGE). This magma must then be transported to the crust by an efficient process in order to reduce the possibility that Ni is removed from the magma by crystallization of olivine. Once the magma is emplaced into the crust, S from some source must be added to bring about saturation of the base metal sulfide liquid. An ideal site for all of these processes is where a mantle plume intersects a continental rift. The plume provides a large volume of magma, produced by a high degree of partial melting. The normal faults of the rift provide easy access to the crust so that the magma is transported efficiently. In many cases rifts contain sedimentary rocks rich in S, thus providing an ideal source of S for sulphide saturation. The heat from the plume can lead to melting of a large volume of rift sediments and release of S from the sediments to the Ni-PGE–rich primary magma. In the case of the Sudbury Igneous Complex, a very large volume of superheated magma formed by flash melting of the crust. This melting event was the result of the impact of shock waves from the explosion of a large meteor in the atmosphere.

The sulfide droplets collected at the base of intrusions and lava flows because they are denser than the silicate magma. The largest concentrations are typically found in locations where there are changes in the geometry of the contacts between intrusions or flows and the country rock. In some cases the accumulated sulphide liquid fractionated to form an Fe-rich monosulfide solid-solution (mss) cumulate and a Cu-rich sulfide liquid which later crystallized as an intermediate solid solution (iss). As a result of crystal fractionation of mss many Ni sulfide orebodies show a strong zonation with respect to Cu and PGE. During mss fractionation Os, Ir, Ru, and Rh concentrated in the mss cumulate and Cu, Pt, Pd, and Au concentrated in the Cu-rich sulfide liquid. The partition coefficient for Ni into mss is close to 1; thus, mss fractionation would not have caused large variations in Ni concentrations. The silicate magma solidified at or above 1,000°C whereas the Cu-rich sulfide liquid solidified at ~900°C. Thus, at many localities the Cu-rich sulfide liquid appears to have migrated into dilatent spaces in the footwall or the hanging wall to form veins that extend into the country rock for up to 2 km. At subsolidus temperatures a number of processes modify the orebodies. Both the mss and iss are not stable below 600°C. As the sulfides cooled mss exsolved to form pyrrhotite and pentlandite (±pyrite), and iss exsolved to form chalcopyrite and pyrrhotite (±cubanite, ±pyrite). Most of the PGE and chalcophile elements that originally partitioned into mss or iss are not readily accommodated in the structure of pyrrhotite, pentlandite, and chalcopyrite; therefore, they exsolve from the mss and iss at low temperature and form a wide variety of platinum group minerals (PGM).

During deformation stress may focus in the structurally incompetent massive sulfide units, which are generally located at the lower contact of the mafic or ultramafic host rock. In this situation the massive sulfides may then be displaced relative to the host rocks. Finally, during greenschist to amphibolite metamorphism, olivine is unstable and Ni released from the olivine will partition into disseminated sulfides, thereby upgrading the sulfides.



Major and trace element geochemistry of the Platreef on the farm Townlands, northern Bushveld Complex.

Manyeruke, T.D., Maier, W.D. and Barnes, S-J., 2005

Abstract
The Platreef is a platinum group elements and base metal enriched mafic/ultramafic layer situated along the base of the northern limb of the Bushveld Complex. The present study contains a detailed petrographic and geochemical investigation of a borehole core intersection through the Platreef on the farm Townlands. At this locality, the Platreef rests on metasedimentary rocks of the Silverton Formation of the Transvaal Supergroup, and is comprised of three medium-grained units of gabbronorite/feldspathic pyroxenite that are separated by hornfels interlayers. We refer to the three platiniferous layers as the Lower, Middle and Upper Platreef. The Middle Platreef is the main mineralised layer, with total PGE contents up to 4 ppm. The Lower and Upper Platreefs are less well mineralised (up to 1.5 ppm). Trace element and S-isotope data show compositional breaks between the different platiniferous layers suggesting that they represent distinct sill-like intrusions of pyroxene and sulphide enriched crystal mushes. The study also reveals a reversed differentiation trend of more primitive rocks towards the top of the succession. For example, orthopyroxene shows an increase in Cr2O3 from 0.07 to 0.37 weight % with height and the whole rock concentration of incompatible trace elements such as Y and Zr decreases. This pattern is interpreted to reflect enhanced crustal contamination of the lower Platreef layers. All three Platreef layers are enriched in heavy S ( d34S of 2.6 to 9.1 ‰ ) indicating addition of crustal sulphur, and they have elevated K, Ca, Zr and Y contents and high Zr/Y ratio relative to Critical Zone rocks from elsewhere in the Bushveld Complex, suggesting a model of crustal contamination in ore formation. Well defined correlations between the concentrations of the individual PGE, and between the PGE and S suggest that the concentration of the PGE was controlled by segregating sulphide melt. Alteration of the rocks, possibly due to infiltration by fluids derived from the floor rocks, caused localized redistribution of Cu, S and, to a lesser degree, the PGE. However, alteration, sulphur and metal mobility was apparently much less pronounced at Townlands than at other Platreef localities further to the north, notably at Sandsloot mine where the PGE are largely hosted by PGM (Armitage et al., 2002). We suggest that this is due to more pronounced devolatisation of the dolomites relative to the shales, implying that the nature of the floor rocks plays an important role in ore formation.



The distribution of base metals and platinum-group elements in magnetitite and its host rocks in the Rio Jacaré Intrusion, Northeastern Brazil Sá, J.H.S., Barnes, S.-J., Prichard, H.M., and Fisher, P.C., 2005

Abstract
Anomalously high Pt and Pd values have been found in three magnetite bodies in the Rio Jacaré intrusion of northeastern Brazil. The intrusion hosting these magnetite bodies consist predominantly of pyroxenite and gabbro. One magnetite body occurs in the Lower zone and two in the Upper zone of the intrusion. These bodies contain approximately 0.04 percent Ni, 0.1 percent Cu, 0.18 percent S, 1 ppb Ir, 3 ppb Rh, 160 ppb Pt, 120 ppb Pd, and 37 ppb Au. They are much richer in platinum-group element (PGE) than the surrounding silicate rock, and there are significant correlations among all of the PGE and between PGE and Ni. However, the correlations between PGE and Au, Cu, and S are much weaker than the correlation between Au, Cu, and S.

In the magnetite bodies palladium-rich minerals, especially bismuthides and antimonides, are the most abundant platinum-group minerals (PGM). In most case these occur with interstitial silicates or within silicate inclusions in magnetite and ilmenite grains and are associated with Co-bearing pentlandite and in few case with Co-Ni sulfarsenides and arsenides. Sperrylite (PtAs2) is the most abundant Pt mineral and is associated with silicates interstitial to magnetite and ilmenite grains and sometimes with Co-Ni arsenides. At sites where the igneous mafic minerals have been altered to amphiboles, sperrylite may be altered to Pt-Fe alloys. Other alloys present include Pd-Sn-Cu, Pt-Ni, and Pt-Au.

It is suggested that Ni and PGE were concentrated in the magnetite bodies by the coprecipitation of a small quantity of sulphide with magnetite. These PGE-bearing base metal sulfides subsequently exsolved PGM. The association of Pd minerals with base metal sulfides and the small variation in the Pt/Pd ratio (ca. 1.4) suggests that the PGE have not been extensively remobilized in the magnetitite. In contrast, the strong correlation between S, Cu and Au suggests that, in addition to the redistribution of S, it is likely that Cu and Au were remobilized. It is not possible to say whether the redistribution of sulphur was due to late magmatic fluids dissolving S or the later metamorphic events.

The association of PGE enrichment with magnetite layers in the Rio Jacaré intrusion contrats with that of the Bushveld, Stillwater, Great Dyke, and Munni Munni Complexes. In these complexes PGE-enriched layers or reefs are found in the lower third of the complexes and the oxide associated with the reefs is chromite. Magnetite-bearing layers, wich form from an evolved magma in the upper parts of the intrusions, are generally barren of PGE because, at the time of magnetite crystallisation, the PGE had already precipitated either in sulfides or PGM. However in a number of intrusions (e.g., Rincon del Tigre, Skaergaard, Stella, and Rio Jacaré) the upper magnetite-bearing portion of the intrusion shows PGE enrichment. This enrichment is rarely associated with visible sulfides but suggests a possible new target for PGE exploration.



Platinum-group element distribution in the Main Zone and Upper Zone of the Bushveld Complex, South Africa

Sarah-Jane Barnes, W. D. Maier and L. D. Ashwal

Abstract
The platinum-group element (PGE) contents of the upper portions of the Bushveld Complex were investigated with three questions in mind: (a) In natural systems does magnetite concentrate Os, Ir, Ru and Rh (IPGE), as has been observed in experimental systems? (b) Is there a Au–Pd-enriched layer present such as observed in the upper parts of the Skaergaard intrusion? (c) Can changes in metal ratios be used for prospecting for PGE deposits?

In the sulfide-poor Main Magnetite Layer of the eastern Bushveld, Ir and Rh are enriched relative to Pt, Pd and Au and this could be because magnetite preferentially concentrated Ir and Rh over Pt, Pd and Au. However, in most other magnetite layers no enrichment was observed. This could be because most magnetite layers contain approximately 1% sulfides and the PGE budget is dominated by the sulfides. These sulfides obscured the effects of magnetite collecting IPGE, because sulfides collect all the PGE and the partition coefficients for the PGE into a sulfide liquid are much greater than the partition coefficient for IPGE into magnetite.

The weighted average of the platinum-group elements (PGE) and Au over the 2000 m of sampled stratigraphy is Au 2.1 ppb, Pd 1.7 ppb, Pt 1.7 ppb, Rh 0.18 ppb, Ru<0.5, Ir 0.16 ppb, Os<0.5 ppb. Compared to the marginal rocks (presumed initial liquids) of the Bushveld Complex the PGE and Au are severely depleted. Only one sample (a leuconorite in the first cyclic unit) contained Pt and Pd at economic grade (Pt 2 ppm, Pd 2 ppm). The overall depletion of the PGE in the Upper Zone (despite the presence of 1% to 3% sulfides) could be the result of the PGE having been stripped from the magma by early sulfide liquid which had already settled out of the magma to form the world famous platinum reefs lower in the magma chamber. In addition to the overall depletion of PGE, PGE/S ratios decrease up section indicating that the sulfide fraction is poorer in PGE up section. This is interpreted to be the result of continued depletion of the silicate liquid as sulfides constantly settle out of the silicate liquid. There appears to be little prospect of a Pd-reef type deposit in the Upper Zone.

Comparison of the composition of the marginal rocks of the Bushveld with a weighted average for the complete 6 km of the Bushveld cumulates shows that the cumulate pile is much richer in compatible elements (Ir, Rh, Cr) and poorer in incompatible elements (Sm and Hf) than the marginal rocks. Two possible solutions to this are: (a) The magma emplaced into the chamber was a crystal mush and thus more mafic than marginal rocks to the intrusion or (b) fractionated magma has been removed from the cumulate sequence and either irrupted or intruded the country rocks as granites and granophyres.

Author Keywords: Platinum-group elements; Gold; Copper; Nickel; Vanadium; Titanium; Magnetite; Upper Zone; Bushveld



Pt/Pd and Pd/Ir ratios in mantle-derived magmas: A possible role for mantle metasomatism

W.D. Maier, and Barnes, S.-J., 2004

Abstract
We compare concentrations of Pt, Pd, and Ir in mantle-derived magmas, ranging from tholeiitic basalts to komatiitic basalts, komatiites, and various alkaline magmas, and in oceanic andcontinental settings. The alkaline magmas tend to have higher Pt/Pd ratios, but lower Pd/Ir ratios than most of the other magmas. We suggest this is attributable to different melting conditions in the mantle. Under relatively "dry" melting conditions applicable to tholeiites and komatiites, Pt-alloys and Os-Ir-Ru-Rh-enriched monosulfide solid solution (Mss) behave in a refractory manner, resulting in sub-chondritic Pt/Pd and super-chondritic Pd/Ir. Under fluid-rich melting regimes in metasomatised lithospheric mantle sources that may be applicable to the generation of many alkaline magmas, the alloys/mss are more fusible, resulting in PGE ratios closer to chondrite. Bushveld magmas and some continental flood basalts also have relatively high Pt/Pd ratios and may thus contain a component of the metasomatised sub-continental lithospheric mantle. Komatiites have relatively low Pt/Pd suggesting that they are derived from a dry mantle source.



The formation of Pt–Ir alloys and Cu–Pd-rich sulfide melts by partial desulfurization of Fe–Ni–Cu sulfides: results of experiments and implications for natural systems

Anna Peregoedova, Sarah-Jane Barnes and Don R. Baker

Abstract
We propose a model of high-temperature formation of platinum-group element (PGE) alloys from base–metal sulfides subjected to decreasing sulfur fugacity (fS2). For this purpose we experimentally investigated the effect of partial desulfurization of monosulfide solid-solution (Mss) with variable Ni and Cu contents on the distribution behaviour of Pt, Ir and Pd at 1000 °C. We partially removed sulfur from the PGE-bearing Mss using the "tube-in-tube" technique and pyrrhotite as an fS2 buffer. We found that Mss undergoing S loss can produce (1) Pt and Ir exsolution from the Mss matrix in the form of PGE-bearing alloys, (2) partial melting of the Cu–Ni-bearing Mss to form Fe-rich Mss, Cu–Ni–Pd-rich sulfide liquid and Fe–Ir–Pt alloy. Both of these processes are geologically important. The partial melting of the S-depleted sulfides could explain the presence of two types of sulfides found in the mantle, Mss-dominated sulfides and S-poor sulfides consisting mainly of pentlandite and chalcopyrite. We also suggest that the Cu–Ni–Pd-rich liquid formed by partial melting of the sulfides could migrate away from the Mss and Fe–Ir–Pt alloys thus spatially decoupling Ir–Pt and Cu–Pd as observed in reefs and mantle nodules. In addition, the formation of PGE alloys in response to the desulfurization processes potentially could occur (1) after pressure falls during transport of the basalt magma with entrained sulfide droplets; (2) after sulfur removal by S-undersaturated melts or fluids from PGE-enriched sulfide proto-ore; (3) after degassing of the sulfide droplets occurring in a sub-volcanic chamber and (4) as a result of sulfide interaction with chromite.

Author Keywords: Platinum; Iridium and palladium alloys; Sulfur fugacity; System Fe–Ni–Cu–S; Experiments



Platinum group elements in the Uitkomst Complex, South Africa

Maier, W.D., Gomwe, T., et Barnes, S.-J., 2004

Abstract
The mafic-ultramafic Uitkomst Complex of South Africa is a tubular intrusion that is believed to be coeval and cogenetic with the 2054 Ma Bushveld Complex. It consists of a thin (3.5-m) basal gabbronoritic phase that is overlain by ca. 450 m of ultramafic rocks and 250 m of gabbroic and dioritic rocks. The basal 300 m of the intrusion hosts disseminated Ni-Cu-platinum group element (PGE) sulfides (100 Mt at ca. 0.55% Ni, 0.17% Cu). In addition, several lenses of massive sulfides (2.9 Mt at 2.04% Ni, 1.13% Cu, and 6 ppm total PGE) are situated in the immediate floor of the intrusion. We determined the PGE concentrations in 86 samples from the Uitkomst Complex. The rocks containing disseminated sulfides have up to 3 ppm total PGE and the massive sulfides have up to 7 ppm total PGE. Sulfide segregation was triggered by assimilation of external sulfur from the country rocks adjacent to the intrusion. The most important contaminant was probably dolomite of the Malmani Subgroup that includes sulfidic shale interlayers. The composition of the magmatic sulfides can be modeled by applying an R factor (mass ratio of silicate melt to sulfide melt) of between 800 and 1,000. The sulfides in the ultramafic rocks show little tendency of metal depletion with height and thus appear to have segregated from successive surges of fertile magma. This suggests that the basal portion of the Uitkomst Complex crystallized in an open magmatic system, e.g., a magma conduit. The gabbroic and dioritic rocks in the upper portion of the complex are relatively PGE depleted (Cu/Pd mostly higher than in primitive mantle), suggesting that they crystallized from the residual magmas to the ultramafic rocks in a closed magmatic system. Some of the ultramafic rocks in the central portion of the complex are S poor (mostly <700 ppb S) but contain relatively high amounts (30-250 ppb) of total PGE. These rocks have unusually low Pd/Ir (0.6-ca. 10), Cu/Zr (typically <0.5), and high Pt/Pd (up to 19), suggesting that either Pd, Cu, and S were remobilized or that the magma was undersaturated in sulfide liquid at the time the rocks formed and that platinum group minerals (PGM) crystallized. The present study suggests that sill-like bodies of Bushveld affinity that intruded dolomitic rocks may generally have an enhanced potential to host massive magmatic Ni-Cu-PGE sulfides.



A Ni-Cu-Co-PGE massive sulfide prospect in a gabbronorite dike at Lac Volant, eastern Grenville Province, Quebec

Nabil, H., Clark, T., and Barnes, S.-J., 2004

Abstract
The Lac Volant Ni-Cu-Co-PGE prospect, located 75 km northeast of Sept-Iles in the Grenville Province of Quebec, is an example of magmatic sulfide mineralization associated with mafic magmas in a high-grade metamorphic terrain. Disseminated and massive sulfides occur within a 20- to 25-m-thick zone in a gabbronorite dike intruding the gabbronorite of the host Matamec Complex. The average composition (n = 29) of the sulfides is estimated to be 2.0% Cu, 1.5% Ni, 0.12% Co, 67 ppb Pt, and 256 ppb Pd. The age of the dike is 1351+ or -6 Ma (U-Pb zircon age), which is identical to the age of the Riviere-Pentecote anorthosite (1354+ or -3 Ma) located 130 km to the southwest. The dike originated by means of multiple injections of sulfide-bearing, tholeiitic magmas derived from a depleted, N-type mid-ocean ridge basalt (N-MORB)-type source. The dike is chemically similar to the gabbronoritic host, suggesting similar parental magmas. Geochemical variation within the dike was caused by fractional crystallization of silicates (orthopyroxene, clinopyroxene, and plagioclase) and by ingestion of crustal rocks. Assimilation is suggested by the presence of xenoliths of granite and metasediment within the dike or narrow branching dikes; enrichment of the dike in Rb, Th, Ba, and light rare earth elements; and a negative Ta anomaly with respect to Th. The observed composition of the dike can be explained by 15% crustal contamination, which would have occurred at depth before the magma began to crystallize mafic minerals. Massive, matrix, and disseminated sulfides, interpreted to be of magmatic origin, are composed of pyrrhotite (the principal species, 75%), pentlandite (altered to bravoite and violarite), chalcopyrite, and pyrite. Magmatic breccia structures are common in gabbronorite containing disseminated sulfides. The presence of metasedimentary xenoliths and a high S/Se ratio (9,000-16,000) suggest that sulfide saturation was caused by contamination of the magma. The sulfide liquid interacted with a small volume of magma (R = silicate liquid/sulfide liquid = 200). Impoverishment of the sulfides in platinum-group elements (PGE) relative to Ni and Cu may have been caused by the loss of PGE during an earlier and deeper episode of sulfide separation. Limited sulfide fractionation is suggested by the chemical similarity of the three types of sulfide. Metamorphic recrystallization produced xenomorphic pyrrhotite and idiomorphic pyrite crystals up to 25 cm and 4 cm in size, respectively. Alteration by meteoric water caused the transformation of pentlandite into bravoite and violarite. The Lac Volant prospect is similar in terms of composition, style of emplacement, magma type, and age to certain Norwegian deposits (the Ertelian and Flat deposits in the Sveconorwegian Province) and to the Voisey's Bay deposit in Labrador.



Variations in the nature of the platinum-group minerals in a cross-section through the Merensky Reef at Impala Platinum: implications for the mode of formation of the reef [pdf file, kb] Hazel M. Prichard, Sarah-Jane Barnes, Wolfgang D. Maier, and Peter C. Fisher, 2004



Abstract
A study of the abundance, size, distribution and composition of platinum-group minerals in samples from a section of the Merensky Reef at Impala Platinum, on the farm Reinkoyalskraal, in the western Bushveld Complex, South Africa, has shown that melanorite, leuconorite and anorthosite contain a PGM assemblage that consists almost exclusively of Pt and Pd bismuthotellurides, predominantly moncheite and merenskyite. In the chromite-rich lithologies, this assemblage of Pt-Pd-Bi telluride PGM is joined by a Pt-Pd-Rh sulfide PGM assemblage of cooperite, braggite and an unnamed Cu-Pt-Rh sulfide, with laurite and rare Sn-bearing PGM. This additional assemblage tends to be Pd-poor. The PGM are rarely enclosed by chromite. All the PGM are predominantly associated with base-metal sulfides, either as euhedral PGM or laths forming an exsolution texture within the base-metal sulfides. Rhodium, present as the unnamed Cu-Pt-Rh sulfide, is associated with pyrrhotite and pentlandite. Throughout this section of the reef, the PGM are commonly located at the edge of base-metal sulfides adjacent to serpentine, chlorite and amphibole that form on the edges of silicate grains. In the chromite-poor samples, Pt-Pd-Bi tellurides and their associated base-metal sulfides are located commonly within silicates, including plagioclase and quartz. The chromite-bearing rocks in this section of the Merensky Reef are enriched in Os, Ir, Ru, Rh and Pt. We test three models for the formation of the PGM: coprecipitation of PGM and chromite, crystallization of PGM from a sulfide liquid, and redistribution of PGE and base metals by hydrous intercumulus fluid. The strong association of PGM with base-metal sulfides suggests that the PGE were collected by an immiscible base-metal sulfide liquid. This liquid crystallized as Mss, with Rh being concentrated in the Mss, and then as Iss. These exsolved to pyrrhotite, pentlandite and chalcopyite and PGM. In the chromite-rich layer, we note a lack of minerals containing Pd in the PGM assemblage. No one model satisfactorily explains the PGM distribution. Rather, the PGM observed are likely to result from late, low-temperature processes superimposed on the magmatic ones.

Key words: platinum-group minerals, Merensky Reef, Impala Platinum mine, chromitite, sulfide, telluride, Bushveld Complex, South Africa.

Résumé
Notre étude de l'abondance, la taille, la distribution et la composition des minéraux du groupe du platine (MGP) dans les échantillons prélevés d'une section du banc de Merensky à la mine Impala Platinum, sur la ferme Reinkoyalskraal, dans la partie occidentale du complexe de Bushveld, en Afrique du Sud, démontre que la mélanorite, la leuconorite et l'anorthosite contiennent un assemblage de MGP fait presqu'exclusivement de bismuthotellurures de Pt et Pd, surtout monchéite et merenskyite. Dans les roches riches en chromite, un assemblage de sulfures de Pt-Pd-Rh, soit cooperite, braggite et un sulfure à Cu-Pt-Rh sans nom, avec laurite et de rares MGP stannifères, viennent s'ajouter aux bismuthotellurures. Ce deuxième assemblage tend à contenir très peu de Pd. Les MGP sont rarement inclus dans la chromite. Tous les MGP sont surtout associés aux sulfures des métaux de base, soit sous forme de cristaux idiomorphes ou bien de lamelles d'exsolution dans ces sulfures. Le rhodium, par sa présence dans le sulfure à Cu-Pt-Rh sans nom, est associé à la pyrrhotite et la pentlandite. De part et d'autre de cette section du banc, les MGP sont généralement situés en bordure des sulfures des métaux de base, avoisinant la serpentine, la chlorite et l'amphibole qui forment les parties externes des grains de silicate. Dans les échantillons à faible teneur en chromite, les tellurures de Pt-Pd-Bi et les sulfures des métaux de base associés sont inclus dans les silicates, y inclus le plagioclase et le quartz. Les roches de cette section du banc de Merensky contenant la chromite sont enrichies en Os, Ir, Ru, Rh et Pt. Nous évaluons trois modèles de formation des MGP: coprécipitation des MGP avec la chromite, cristallisation des MGP à partir d'un liquide sulfuré, et redistribution des MGP et des métaux de base par l'intermédiaire d'une phase fluide intercumulus. L'association marquée des MGP avec les sulfures des métaux de base nous fait penser que les éléments du groupe du platine ont d'abord été concentrés dans un liquide sulfuré immiscible. Ce liquide a cristallisé sous forme de Mss, avec le Rh concentré dans la phase Mss, et ensuite sous forme de Iss. Ces minéraux ont par la suite exsolvé la pyrrhotite, la pentlandite, la chalcopyrite et les MGP. Dans le niveau enrichi en chromite, nous soulignons l'absence de minéraux contenant le Pd, parmi les MGP. Aucun des trois modèles explique avec satisfaction la distribution des MGP. A notre avis, les MGP résulteraient plutôt de processus tardifs, surimposés à basse température aux produits des processus magmatiques.

Mots clés : minéraux du groupe du platine, banc de Merensky, mine Impala Platinum, chromitite, sulfure, tellurure, complexe de Bushveld, Afrique du Sud.



Platinum-group elements in the Boulder Bed, western Bushveld Complex, South Africa

Wolfgang D. Maier and Sarah-Jane Barnes, 2003

Abstract
Concentrations of platinum-group elements in samples from the Boulder Bed at five localities in the western Bushveld Complex range between 50 ppb and 70 ppm. Boulders thus have much more variable, and sometimes highly enriched, PGE contents relative to the other lithologies in the immediate foot-wall sequence of the Merensky Reef. The PGE enrichment can largely be modelled as a result of primary magmatic processes including collection of PGE by segregating sulphide melt and fractionation of mss. Other features of the Boulder Bed, such as the selvages of pure anorthosite and the chromitite stringers surrounding some of the boulders, bear evidence of recrystallisation. A model is proposed by which the Boulder Bed formed as a result of a combination of early and late magmatic processes. The PGEs were collected by magmatic sulphide melt which accumulated in a pyroxenite layer. The host rock to the pyroxenite was a thick package of norites which recrystallised in response to upward-migrating magmatic fluids. The fluids caused partial hydration melting of the norites adjacent to the pyroxenite, producing anorthosite. The boulders represent the broken-up remnants of the pyroxenite layer. The selvages of chromite and pure anorthosite around some of the boulders remain poorly understood, but may represent the latest recrystallisation event, in response to localised late-magmatic fluid overpressure upon cooling.

Keywords: Platinum-group elements, Boulder Bed, Merensky Reef, Bushveld Complex, South Africa



Pt-Pd reefs in magnetitites of the Stella layered intrusion, South Africa: A world of new exploration opportunities for platinum group elements

W.D. Maier, S.-J. Barnes, V. Gartz, and G. Andrews, 2003

Abstract
The 3033 Ma Stella layered intrusion of South Africa consists largely of magnetite gabbros and gabbros that are hosted by greenstones of the Kraaipan belt. The intrusion contains a 100-m-thick, platinum group element (PGE)–enriched interval that includes a number of laterally continuous PGE reefs constituting the oldest mineralization of this type known on Earth. The richest of the reefs is hosted by magnetitite and contains 10–15 ppm Pt + Pd over 1 m, representing by far the highest PGE grades known up to this time in magnetitite-hosted Pt-Pd reefs. The PGEs are interpreted to have been concentrated by sulfide melt, after S saturation had been reached in the advanced stages of magmatic differentiation, in response to magnetite crystallization. Reaction between sulfide melt and oxides led to late magmatic S loss, causing a paucity of sulfides in most of the PGE mineralized interval. As a result, the reefs cannot be distinguished macroscopically from their unmineralized host rocks, and we suggest that similar mineralization may have been overlooked in the upper parts of other tholeiitic intrusions elsewhere.

Keywords: Stella intrusion, platinum group elements, South Africa, greenstone belts, magnetite.



The concentrations of the noble metals in Southern African flood-type basalts and MORB: implications for petrogenesis and magmatic sulphide exploration

Wolfgang D. Maier, Sarah-Jane Barnes and Julian S. Marsh, 2003

Abstract
Concentrations of the platinum-group elements have been determined in several suites of southern African flood-type basalts and mid-ocean ridge basalt (MORB), covering some 3 Ga of geologic evolution and including the Etendeka, Karoo, Soutpansberg, Machadodorp, Hekpoort, Ventersdorp and Dominion magmas. The magmas cover a compositional range from 3.7 to 18.7% MgO, 26–720 ppm Ni, 16–250 ppm Cu, and <1–255 ppb total platinum-group elements (PGE). The younger basalts (Etendeka, Karoo) tend to be depleted in PGE relative to Cu, while most of the older basalts (Hekpoort, Machadodorp, Ventersdorp, Dominion) show no PGE depletion relative to Cu. Further, the younger basalts tend to have lower average Pt/Pd ratios than the older basalts, and the MORBs have lower average Pt/Pd than the continental basalts within the broad groupings of "old" and "young" basalts. This may reflect (1) a decreasing degree of mantle melting through geologic time, and (2) source heterogeneity, in that the MORBs are derived from predominantly asthenospheric mantle, whereas the continental basalts also contain a lithospheric mantle component enriched in Pt. In addition to these factors, some PGE fractionation also occurred during differentiation of the magmas, with Pd showing incompatible behaviour and the other PGE variably compatible behaviour. The examined southern African flood-type basalts and MORB appear to offer limited prospects for magmatic sulfide ores, largely because they show little evidence for significant chalcophile metal depletion that could be the result of sulphide extraction during ascent and crystallization.



The Concentration of the Platinum-Group Elements in South African Komatiites: Implications for Mantle Sources, Melting Regime and PGE Fractionation during Crystallization

Wolfgang D. Maier, Frederick Roelofse, and Sarah-Jane Barnes, 2003

Abstract
We have analysed 18 samples of komatiite from five consecutive lava flows of the Komati Formation at Spinifex Creek, Barberton Mountain Land. Our samples include massive komatiite, various types of spinifex-textured komatiite, and flow-top breccias. The rocks have low platinum-group element (PGE) contents and Pd/Ir ratios relative to komatiites from elsewhere, at 0·45–2 ppb Os, 1–1·4 ppb Ir, <1–5 ppb Ru, 0·33–0·79 ppb Rh, 1·7–6 ppb Pt, 1·6–6·1 ppb Pd, and Pd/Ir 3·3. Pt/Pd ratios are c. 1·1. Platinum-group elements are depleted relative to Cu (Cu/Pd = 15 300). They display a tendency to increase in the less magnesian samples, suggesting that the magmas were S-undersaturated upon eruption and that all PGE were incompatible with respect to crystallizing olivine. Komatiites from the Westonaria Formation of the Ventersdorp Supergroup and the Roodekrans Complex near Johannesburg have broadly similar PGE patterns and concentrations to the Komati rocks, suggesting that the PGE contents of South African ultrabasic magmas are controlled by similar processes during partial mantle melting and low-P magmatic crystallization. Most workers believe that the Barberton komatiites formed by relatively moderate-degree batch melting of the mantle at high pressure. Based on the concentration of Zr in the Komati samples, we estimate that the degree of partial melting was between 26 and 33%. We suggest that the low PGE contents and Pd/Ir ratios of all analysed South African komatiites are the result of sulphides having been retained in the mantle source during partial melting. The difference in Pd/Ir between our samples and Al-undepletedkomatiites from elsewhere further suggests that the PGE are fractionated during progressive partial melting of the mantle. Thus, our data are in agreement with other recent studies showing that the PGE are hosted by different phases in the mantle, with Pd being concentrated by interstitial Cu-rich sulphide, and the IPGE (Os, Ir, Ru) and Rh resting in monosulphide solid solution included within silicates. Pt is possibly controlled by a discrete refractory phase, as Pt/Pd ratios of most komatiites worldwide are sub-chondritic.

Key words: platinum-group elements; komatiites; Barberton; mantle melting; South Africa



The use of mantle normalization and metal ratios in the identification of the sources of platinum-group elements in various metal-rich black shales

Jan Paava, Sarah-Jane Barnes and Anna Vymazalová, 2003

Abstract
It has been shown that Ni/Cu versus Pd/Ir and Cu/Ir versus Ni/Pd ratios, as well as mantle-normalized metal patterns can be successfully used to evaluate the effects of partial melting, crystal fractionation and sulfide saturation in mafic and ultramafic rocks. Platinum-group element (PGE) enrichments occur in Zn-, Cu- and Ni-rich black shales in a number of geological settings. These facies are generally associated with the development of continental rift structures, both without significant volcanic activity (examples in Canada, China, Finland and Poland) and with volcanic activity (Czech Republic and Namibia). Using the same element ratios as were used in mafic igneous rocks, it is evident that metal-rich black shales associated with volcanic rocks reflect the fractionation of PGEs in volcanogenic facies. It is further concluded that, in metal-rich black shales in which distributions of Ni and Cu have been altered by various processes, the Cu/Ir versus Ni/Pd and Ni/Cu versus Pd/Ir ratios plots are applicable for determination of the source of PGEs only after careful evaluation of Ni and Cu enrichments.

Keywords: Metal-rich black shale - Mantle normalization - Metal ratios - Source of PGEs



Platinum-group Elements and Microstructures of Normal Merensky Reef from Impala Platinum Mines, Bushveld Complex

Sarah-Jane Barnes and Wolfgang D. Maier, 2002

Abstract
The Merensky Reef of the Bushveld Complex contains one of the world’s largest concentrations of platinum-group elements(PGE). We have investigated ‘normal’ reef, its footwall and its hanging wall at Impala Platinum Mines. The Reef is 46 cm thick and consists from bottom to top of leuconorite, anorthosite, chromitite and a very coarse-grained melanorite. The footwall is leuconorite and the hanging wall is melanorite. The only hydrous mineral present is biotite, which amounts to 1%, or less, of the rock. All of the rocks contain 0·1–5% interstitial sulphides (pyrrhotite, pentlandite and chalcopyrite), with the Reef rocks containing the most sulphides (1–5%). Lithophile inter-element ratios suggest that the magma fromwhich the rocks formed was a mixture of the two parental magmas of the Bushveld Complex (a high-Mg basaltic andesite and a tholeiitic basalt). The Reef rocks have low incompatible element contents indicating that they contain 10% or less melt fraction. Nickel, Cu, Se, Ag, Au and the PGE show good correlations with S in the silicate rocks, suggesting control of the abundance of these metals by sulphides. The concentration of the chalcophile elements and PGE in the silicate rocks may be modelled by assuming that the rocks contain sulphide liquid formed in equilibrium with the evolving silicate magma. It is, however, difficult to model the Os, Ir, Ru, Rh and Pt concentrations in the chromitites by sulphide liquid collection alone, as the rocks contain 3–4 times more Os, Ir, Ru, Rh and Pt than the sulphide-collection model would predict. Two possible solutions to this are: (1) platinum-group minerals (PGM) crystallize from the sulphide liquid in the chromitites; (2) PGM crystallize directly fromthe silicate magma. To model the concentrations of Os, Ir, Ru, Rh and Pt in the chromitites it is necessary to postulate that in addition to the 1% sulphides in the chromitites there is a small quantity (0·005%) of cumulus PGM (laurite, cooperite and malanite) present. Sulphide liquids do crystallize PGM at low fS2. Possibly the sulphide liquid that was trapped between the chromite grains lost some Fe and S by reaction with the chromite and this provoked the crystallization of PGM from the sulphide liquid. Alternatively, the PGM could have crystallized directly from the silicate magma when it became saturated in chromite. A weakness of this model is that at present the exact mechanism of how and why the magma becomes saturated in PGM and chromite synchronously is not understood. A third model for the concentration of PGE in the Reef is that the PGE are collected from the underlying cumulus pile by Cl-rich hydrous fluids and concentrated in the Reef at a reaction front. Although there is ample evidence of compaction and intercumulus meltmigration in the Impala rocks, we do not think that the PGE were introduced into the Reef from below, because the rocks underlying the Reef are not depleted in PGE, whereas those overlying the Reef are depleted. This distribution pattern is inconsistent with a model that requires introduction of PGE by intercumulus fluid percolation from below.

Key words : Merensky Reef; platinum-group elements; chalcophile elements; microstructures



A comparison of N-type semi-planar and coaxial INAA detectors for 33 geochemical reference samples

L. P. Bédard and S.-J. Barnes, 2002

Abstract
While INAA is becoming a less popular analytical technique and it is a mature tool, there are still many improvement happening in the field. The effect of the new semi-planar detector is evaluated as compared to geological reference material and as its performance to the classical coaxial detector. The semi-planar detector offers improved accuracy (about 5%) for many analytes (As, Ba, Ce, Co, Cr, Eu, Hf, Lu, Nd, Rb, Sm, Th, U, Yb and Zn) while the coaxial gives an accuracy in the range of 10-15%.



A comparison of the capacity of FA-ICP-MS and FA-INAA

L. P. Bédard and S.-J. Barnes, 2002

Abstract
The platinum-group elements (PGEs) are commonly determined by INAA and ICP-MS after a NiS fire assay preconcentration. The results of the initial round robin for the PGEs and gold were examined for geological Canadian reference materials (WGB-1, TDB-1, UMT-1, WPR-1, WMG-1, and WMS-1). The Au accuracy is generally within 15% for both methods. For Ir, Os, Pd, Pt and Rh the accuracy for most samples is better than 10% for FA-ICP-MS and FA-INAA (true only for sulfide-bearing samples in the case of FA-INAA). Ru is not very accurate by either methods. Ru and Au have problems with precision which is interpreted to be related to the loss of gold in the dissolution step and for Ru, the source of the problem is not yet understood. Kurtosis show that FA-INAA has higher clustering than FA-ICP-MS for most analytes. It suggests a slightly better precision for FA-INAA. This is explained by the robustness of INAA after the NiS preconcentration despite its lower instrumental precision versus the complex dissolution steps involved in ICP-MS. For samples richer in PGEs (sulfide- and/or oxide-bearing rocks) both methods perform adequately but for low PGEs concentration samples (crustal rocks) ICP-MS shows an advantage.



The distribution of platinum group elements in the Insizwa lobe, Mount Ayliff Complex, South Africa: implications for Ni-Cu-PGE sulphide exploration in the Karoo igneous province

Maier, W.D., Marsh, J.S., Barnes, S.-J., and Dodd, D.C., 2002

Abstract
The Mount Ayliff Complex of the Eastern Cape province of South Africa is a layered intrusion of some 800 km2 surface area and up to 1200 m thickness. On the basis of compositional similarities between the Mount Ayliff Complex and the staging chambers and feeder conduits to flood basalts that host magmatic sulphide ores elsewhere in the world suggest that the Mount Ayliff Complex may have an enhanced potential for Noril’sk-Talnakh-type massive Ni-Cu sulphide ores, an idea that is supported by the well-know sulphide occurrence at Waterfall Gorge. Here, we present major-, trace-, and noble-element data for 30 samples of cumulate rocks from a continuous 1200 m drill core through the Insizwa lobe of the complex, as well as six samples of a footwall sill, the Taylor’s Koppie dike, that was considered a possible feeder zone of the lobe, and through a massive sulphide lens in the footwall of the lobe at Ndzongiseni. We show that most of the Insizwa cumulate rock contain small amounts of cumulus sulfides. The sulfides, including those at Waterfall Gorge, can be explained by a model in wich the sulfides were segregated from magma having chilled-margin composition. On the basis of the composition of the chilled margins, the Insziwa magma contained no entrained sulfides and was depleted in platinum group elements (PGE), although not in Cu and Ni, before emplacement. These compositional characteristics are also observed in lavas of the central Karoo igneous province and suggest only limited sulphide segregation from these magmas at depth. Our findings thus indicate relatively little potential for economic magmatic Ni-Cu-PGE sulphide deposits in the Insizwa lobe.



Fluid transport of sulfur and metals between sulfide melt and basaltic melt [pdf file, kb]

Don R. Baker, Sarah-Jane Barnes, Gail Simon and Frederic Bernier, 2001

Abstract
Experiments were performed at 1.0 GPa and temperatures between 1225 and 1450°C to demonstrate the transport of sulfur and metals Cu, Ni and Pt into a basaltic melt via a sulfurous fluid phase. Experiments were performed inside graphite-lined Pt capsules by converting a mixture of FeS2 + NiS2 + CuS into sulfide melt + a sulfurous fluid, which was kept separate from a basaltic melt inside the same Pt capsule by a gas-permeable graphite membrane. The sulfurous fluid is rapidly transported, and the basaltic melt achieves S concentrations similar to saturation values (within 2s) in times less than 1 hour, even though the two melts are not in physical contact. The sulfurous fluid is shown to transport significant quantities of Cu, Ni and Pt in 1 hour. Gold is transported out of the melt. A process similar to the one experimentally modeled could occur in nature as magmatic systems heat country rocks to temperatures at which pyrite converts to pyrrhotite + a sulfurous fluid, potentially containing significant quantities of ore metals. The volume change of this reaction could crack the country rocks and allow the fluid to enter the magma chamber. These fluids could then saturate the melt, causing the formation of immiscible sulfide droplets enriched in base and precious metals, eventually forming either the progenitor of an ore deposit or the ore deposit itself.

Key words: ore deposits, sulfur, metal transport, copper, nickel, platinum, basalt, experimental petrology.

Résumé
Nous avons étudié le transfert du soufre et des métaux Cu, Ni et Pt, incorporés dans un bain fondu basaltique par le biais d'une phase volatile sulfureuse à une pression de 1.0 GPa et des températures entre 1225 et 1450°C. Ces expériences, effectuées dans un récipient de platine avec gaine de graphite, ont donné un liquide sulfuré et une phase volatile sulfureuse à partir d'un mélange de FeS2 + NiS2 + CuS, maintenu séparé du liquide basaltique à l'intérieur de la même capsule en platine au moyen d'une membrane de graphite perméable au gaz. Le fluide sulfuré est rapidement transporté, et le liquide basaltique atteint une saturation en soufre, les teneurs étant semblables à celles prévues dans un cas de saturation (à 2s près) en moins d'une heure, malgré le fait que les deux liquides n'étaient pas en contact. Le fluide sulfuré peut transporter des quantités importantes de Cu, Ni et Pt en une heure. En revanche, l'or a été extrait du liquide. Un processus semblable à celui que nous avons simulé par expériences pourrait être important dans les cas où un système magmatique réchauffe les roches encaissantes suffisamment pour déstabiliser la pyrite et produite la pyrrhotite + une phase fluide sulfurée qui pourrait contenir des quantités importantes de métaux d'intérêt économique. Le changement de volume associé à cette réaction de dégazage pourrait craquer les roches encaissantes et ainsi donner à la phase fluide libre accès à la chambre magmatique. Le magma pourrait par la suite devenir saturé en soufre, ce qui mènerait à la formation de goutelettes de liquide sulfuré immiscibles enrichies en métaux de base et métaux précieux. Ce processus pourrait former un précurseur à un gîte minéral, ou bien le gisement lui-même.

Mots clés : gîte minéral, soufre, transport de métaux, cuivre, nickel, platine, basalte, pétrologie expérimentale.



Proton microprobe results for the partitioning of platinum-group elements between monosulphide solid solution and sulphide liquid

Sarah-Jane Barnes, Esme van Achterbergh, Emil Makovicky, and Chusi Li, 2001

Abstract
Partition coefficients (D) for Ni, Cu, and platinum-group elements (PGE) between monosulphide solid solution (mss) and Fe-sulphide liquid (liq) have been determined experimentally using an electron microprobe (EMP) to analyze experimental run products. The EMP detection limit is approximately 0.05 weight per cent for the PGE, consequently few results were obtained for Pt and Ir and the precision for Pd and Rh at low concentrations was poor. These run products have been reanalyzed using a proton microprobe(PMP), which has a detection limit between 10 and 50ppm for these elements.

It is now clear that Dmss/liquid for all the elements show a strong dependence on the S content of the run in S-undersaturated and S-saturated runs. However, in S-oversaturated runs the S content of the run does not appear to influence Dmss/liq. The greater precision of the PMP data establishes that in S-oversaturated runs Dmss/liq at 1000°C are consistently higher than those at 1100°C. In contrast, Dmss/liq in the S-undersaturated and S-saturated runs are similar at both temperatures. This difference in behaviour is thought to arise because in the S-undersaturated and S-saturated runs the amount of S in the mss is controlled by the S content of the run. As the S content in the mss increases, the number of vacancies in the structure of the mss also increases, and Dmss/liq rises. In contrast, in S-oversaturated runs the mss has absorbed the maximum amount of S and thus the S content of the run no longer influences the structure of the mss and hence does not control Dmss/liq. Thus, the effect of temperature on Dmss/liqonly becomes apparent in the S-oversaturated runs.

The tendency for Os, Ir, Ru and Rh to partition into mss and the exclusion of Cu, Pt and Pd from mss maybe used to explain a number of phenomena; the zonation of massive sulphide bodies, the tendency for Os, Ir, Ru and to a lesser extent Rh to be enriched in cumulates with minor sulphides, and the presenceof two types of sulphides in mantle nodules (an Os-Ir-rich mss and Cu-Pd rich pentlandite).

The tendency of sulphide liquids to crystallize RuOsIr and Pt-Fe minerals at low f S2 may explain the enrichment of RuOsIr in ultramafic mafic cumulate rocks in the following manner. Sulphide solubility increases as pressures falls. Thus, sulphide droplets in rising basalt magma could be partly resorbed. The PGM could crystallize from this liquid. If these PGM survive long enough, then they could be incorporated into the early cumulate phases such as olivine and chromite. This would explain both the presence of PGM in many olivine and chromite cumulates and the tendency of more evolved magmas to have high Pd/Ir ratios.



The composition and mode of formation of the Pechenga nickel deposits, Kola Peninsula, northwestern Russia [pdf file, kb]

Sarah-Jane Barnes, Victor A. Melezhik and Stanislav V. Sokolov, 2001

Abstract
The Pechenga Ni-sulfide deposits, in the Kola Peninsula of Russia, are associated with ferropicrite flows and intrusions. The sulfides are divided into five types: 1) disseminated sulfides within the olivine cumulate portions of the ferropicrites, 2) massive sulfides, which occur at the contact between the ferropicrites and the country-rock black schists, 3) breccia-matrix sulfides, which occur at the contact between the ferropicrites and the schists, and in some cases continue for hundreds of meters subparallel to the contact but within the footwall, 4) chalcopyrite vein or stringer sulfides, which occur in the footwall, and 5) pyrite-rich layers, concretions and lenses in the sedimentary rocks. The disseminated sulfides have similarly shaped mantle-normalized patterns of chalcophile metal abundances to those of the ferropicrites, and could have formed in equilibrium with the ferropicritic magma at moderate R factors (250). The rocks are depleted in platinum-group elements (PGE), suggesting that the magma reached sulfide saturation prior to the formation of the ores. Sulfur isotope data indicate that the S was derived from the sediments onto and into which the ferropicrites were emplaced. The high As and Sb concentrations in the sulfides may have been derived from the sediments. The massive sulfides show a wider variety of composition than the disseminated sulfides; some sulfides are enriched in Os, Ir, Ru, Rh and depleted Pt, Pd, Ag, Au, Cu, Sb, As, Se. This pattern is attributed to the accumulation of monosulfide solid-solution during crystallization of a sulfide liquid. The sedimentary sulfides are richer in As and Sb than the disseminated and massive sulfides, but poorer in all the other chalcophile elements. The breccia-matrix sulfides consist of sedimentary and ultramafic fragments in a sulfide matrix. Compared with vein sulfides from other deposits, those from Pechenga have a very unusual composition. They are not only rich in Cu and Ag, but also in Os, Ir, Ru and Rh, and they are depleted in Pt, Pd, Au, As, Sb and Se.

Key words: nickel sulfide deposits, platinum-group elements, arsenic, antimony, gold, selenium, silver, rare-earth elements, ferropicrites, sulfur isotopes, Pechenga, Russia.

Résumé
Les gisements de nickel de Pechenga, dans la péninsule de Kola, en Russie, sont associés à des coulées et des massifs intrusifs de ferropicrite. Les sulfures de nickel se présentent en cinq associations: 1) sulfures disséminés dans les parties cumulatives des ferropicrites, 2) sulfures massifs au contact entre les ferropicrites et les schistes noirs encaissants, 3) sulfures formant la matrice de brèches de ferropicrites et de schistes, dans certains cas se prolongeant des centaines de mètres le long du contact mais invariablement situés au dessus de la paroi inférieure, 4) veines de chalcopyrite ou sulfures en veinules, dans la paroi inférieure, et 5) accumulations de pyrite formant concrétions et lentilles dans l'encaissant sédimentaire. Les sulfures disséminés possèdent des tracés de concentrations en métaux chalcophiles normalisées par rapport au manteau semblables à ceux des ferropicrites, et pourraient s'être formés en équilibre avec le magma ferropicritique à un facteur R moyen (250). Les roches sont appauvries en éléments du groupe du platine, ce qui indiquerait une saturation du magma en sulfures avant la formation du minerai. Les données isotopiques indiquent que le soufre a été dérivé à partir des sédiments sur et dans lesquels le magma ferropicritique s'est épanché. Les concentrations élevées d'arsenic et d'antimoine pourraient avoir une origine dans les sédiments. Les sulfures massifs montrent un intervalle de composition plus étendu que les sulfures disséminés. Dans certains cas, les sulfures montrent une enrichissement en Os, Ir, Ru et Rh, et un appauvrissement en Pt, Pd, Ag, Au, Cu, Sb, As et Se. Ces caractéristiques seraient attribuables à une accumulation de solution solide monosulfurée au cours de la cristallisation du liquide sulfuré. Les sulfures d'origine sédimentaire sont plus riches en arsenic et antimoine.que les sulfures disséminés et massifs, mais les autres éléments chalcophiles y sont appauvris. Les sulfures formant la matrice des brèches renferment des fragments sédimentaires et ultramafiques. En comparaison des sulfures en veines d'autres gisements, ceux de Pechenga ont une composition très inhabituelle. Ils sont enrichis non seulement en Cu et Ag, mais aussi en Os, Ir, Ru et Rh, et appauvris en Pt, Pd, Au, As, Sb et Se.

Mots clés : gisements de sulfures de nickel, éléments du groupe du platine, arsenic, antimoine, or, sélénium, argent, terres rares, ferropicrite, isotopes de soufre, Pechenga, Russie.



Re-Os isotopic study of komatiitic volcanism and magmatic sulfide formation in the southern Abitibi greenstone belt, Ontario, Canada [pdf file, kb]

Yann Lahaye, Sarah-Jane Barnes, Louise R. Frick and David D. Lambert, 2001

Abstract
We have investigated the Re–Os isotope geochemistry of 2.7-Ga metakomatiitic flows and associated Ni–Cu sulfide deposits from Alexo, Texmont and Hart in the Abitibi greenstone belt of Ontario in order to refine the thermal erosion model and evaluate the superimposed effects of metamorphism and hydrothermal alteration on ore environments and non-ore environments. Although the geochemical characteristics of these komatiites have led to the belief that these lavas were uncontaminated, radiogenic Os isotopic compositions (gOs = +20.2 and +33.2) obtained from well-preserved komatiites and olivine separates suggest that the Alexo flow has been contaminated by crust-derived material. These data are compatible with the trace-element enrichment observed in melt inclusions trapped within olivine. Redistribution of Os and Re did occur at least at the mineral scale and most likely during the Grenville orogeny. Hydrothermal fluids were channeled along the contact between the komatiites and their basement, and were responsible for the remobilization of Re or Os (or both) within the sulfides at Alexo and Hart. Matrix and disseminated sulfides from Texmont are located within the pile of cumulates and seem to have escaped this localized alteration. Although the Abitibi sulfides have experienced various degrees of metamorphism (from prehnite–pumpellyite to low amphibolite facies), the initial Re–Os isotopic composition of the flows appears to have been preserved at the whole-rock scale. Re–Os isotopic heterogeneity of the Abitibi sulfides is best explained by variable R-factor of the sulfides. Re and Os concentrations and Os isotopic heterogeneity of the Abitibi sulfides are consistent with the current model of nickel sulfide formation, which implies that the assimilation of sulfidic sedimentary rocks was the trigger for sulfide saturation.

Key words: komatiite, Ni–Cu sulfide, rhenium–osmium, melt inclusion, laser ablation, Alexo, Hart, Texmont, Abitibi greenstone belt, Ontario.

Résumé
Nous avons étudié la géochimie isotopique Re–Os des coulées métakomatiitiques et des gisements de sulfures de Ni–Cu associés, Alexo, Texmont et Hart, dans la ceinture de roches vertes de l'Abitibi, en Ontario, mis en place il y a 2.7 milliards d'années, afin d'évaluer le modèle de l'érosion thermique et des effets surimposés du métamorphisme et l'altération hydrothermale dans les milieux près des gisements ou non. Quoique les caractéristiques géochimiques de ces komatiites ont mené à la conclusion que ces laves ne sont pas contaminées, les proportions d'Os radiogénique (gOs = +20.2 et +33.2) obtenues à partir de roches komatiitiques et de fractions d'olivine bien conservées font penser que la coulée d'Alexo a été contaminée par un matériau d'origine crustale. Ces données sont compatibles avec les taux d'enrichissement décelés dans les reliquats magmatiques piégés dans l'olivine. Une redistribution de l'osmium et du rhénium a au moins eu lieu à l'échelle intergranulaire, et tout probablement au cours de l'orogenèse grenvillienne. Les fluides hydrothermaux ont été canalisés le long du contact entre les komatiites et le socle, et sont responsables de la remobilisation de Re ou de Os (ou des deux) parmi les sulfures à Alexo et Hart. Les sulfures de la matrice et disséminés à Texmont sont situés à l'intérieur d'un empilement de cumulats et semblent avoir échappé à cette altération localisée. Quoique les sulfures de ces sites dans l'Abitibi ont subi les effets variables d'une recristallisation métamorhique (alant du faciès prehnite–pumpellyite au faciès amphibolite inférieur), la composition isotopique Re–Os initiale des coulées semble avoir été conservée à l'échelle des roches globales. Les hétérogénéités isotopiques Re–Os des sulfures de ces suites résulteraient du facteur R variable des sulfures. Les concentrations en Re et Os et les rapports isotopiques hétérogènes de l'osmium dans les sulfures de ces suites de l'Abitibi concordent avec le modèle accepté de formation des sulfures de nickel, ce qui implique une assimilation de roches sédimentaires sulfurées pour déclencher la saturation des systèmes en sulfures.

Mots clés : komatiite, sulfure Ni–Cu, rhénium–osmium, reliquat magmatique, ablation au laser, Alexo, Hart, Texmont, ceinture de roches vertes de l'Abitibi, Ontario.



Platinum-group elements in the Pyroxenite Marker, Bushveld Complex: implications for the formation of the Main Zone

W.D. Maier, Sarah-Jane Barnes, and M. J. van der Merwe, 2001

Abstract
Concentrations of platinum-group elements and sulphides in the Pyroxenite Marker of the upper Main Zone are variable, but generally low (up to 100 ppb PGE and 0.2 weight % S). The metal patterns may mostly be explained by sulphide segregation from PGE depleted residual Upper Critical Zone magma, but they are inconsistent with sulphide segregation from a replenishing influx of undepleted Critical Zone magma. Instead, we favour a model whereby a relatively cool and dense Main Zone crystal mush intruded the Bushveld chamber during the later stages of the deposition of the Upper Critical Zone and displaced warmer and lighter residual magmadepleted in chalcophile metals (Sharpe, 1985). Based on the metal contents and textural evidence such as the occurrence of ophitic textures, we hypothesize that the Pyroxenite Marker formed in response to localized supercooling and the suppression of plagioclase crystallization. The model implies that the layer represents a poor target for PGE mineralization in the upper portions of the Bushveld Complex.



PGE-bearing mafic-ultramafic sills in the floor of the eastern Bushveld Complex on the farms Blaauwboschkraal, Zwartkopje, and Waterval

W.D. Maier, J. Sliep, Sarah-Jane Barnes, S.A. de Waal, and C. Li, 2001

Abstract
Mafic-ultramafic sills of up to 450m in thickness occur near the contact between sedimentary and volcanic rocks of the Silverton Formation, Transvaal Supergroup, on the farms Blaauwboschkraal, Zwartkopje and Waterval, some 10km north of Waterval Boven in the Mpumalanga Province. The sills consist of peridotite, harzburgite, pyroxenite and gabbro, and may locally contain up to about 10% Ni-Cu-PGE sulphides. Metal contents of the rocks reach 0.7% Cu, 0.8% Ni, and 2 ppm PGE. Sulphides are found at variable levels within the intrusions and are interpreted to have precipitated from distinct surges of magma streaming through a conduit. This model is analogous to that proposed for the Uitkomst Complex, which hosts sulphides of broadly similar composition to thepresent bodies and which consists of broadly similar lithologies,apart from the presence of chromitite in the latter. The lithologicaland compositional similarities between the intrusions raise the possibility of undiscovered economic sulphide concentrations in the present bodies. However, olivines in most of the ultramafic rocks are undepleted in Ni, suggesting that either sulphide segregation was minor and localized, or that any metal-depleted magmas were flushed out of the conduit by undepleted magma.



Origin of Cu-Ni-PGE sulphide mineralization in the Partridge River intrusion, Duluth Complex, Minnesota

Thériault, R.D., Barnes, S.-J., and Severson, M.J., 2000

Abstract
Four subeconomic Cu-Ni-PGE sulphide deposits occur near the base of the Partridge River intrusion, a mafic layered intrusion emplaced along the northwestern margin of the Duluth Complex. The host troctolitic rocks are in contact with sulphide-bearing metasedimentary rocks of the Virginia Formation. The origin of the sulphide mineralization has generally been link to contamination of the mafic magma through partial assimilation of the argillaceous country rocks.

Three main types of sulphide mineralization have been recognized within the four deposits. These are (1) PGE-poor disseminated sulfides; (2) PGE-rich disseminated sulfides; and (3) semimassive to massive sulfides. The PGE-poor disseminated sulfides typically occur within the lower 250 m of the intrusion and are hosted mainly by heterogeneous norite and olivine gabbro, both of wich contain abundant country-rock xenoliths. This type of mineralization shows numerous features, such as high proportions of pyrrhotite and arsenide minerals and high Cu/Pd, Ni/Pd, and Cu/Pt ratios, wich suggest that the magma had undergone substantial contamination. These sulfides appear to have formed at low to moderate ratios of silicate magma to sulphide melt (mean R factor = 600- 2400), as deduce from their metal-poor nature. The PGE-rich disseminated sulfides occur well within the intrusion directly beneath ultramafic layers and show little signs of contamination. They are composed mainly of chalcopyrite and pentlandite, with lesser amounts of pyrrhotite and cubanite. They appear to have formed at high R factors (mean = 6000-7700), wich explains their relatively high PGE and base metal contents. The semimassive to massive sulfides occur mainly as veins and lenses both along the basal contact and within the underlying sedimentary country rocks. They are typically zoned, being composed of both pyrrhotite-rich (Fe-rich) and chalcopyrite-cubanite-rich (Cu-rich) portions, the latter often forming along the base or top pf massive sulphide bodies. The zonation in the massive sulfides is interpreted to be the product of fractional crystallization of a sulphide melt, the pyrrhotite-rich sulfides representing the cumulate of this crystallization and the Cu-rich sulfides the fractionated liquid.

Based on the above evidence, we interpret the compositional variations observed between the different types of sulphide mineralization to originate from the combined action of three different processes that operated in sequence from magma emplacement until complete crystallization of the sulfides. These processes are (1) country-rock assimilation; (2) interaction between the sulphide melt and the silicate magma (R factor); and (3) fractional crystallization of the sulphide melt.



A reconnaissance study on the magmatic Cu-Ni-PGE sulphide potential of the Tete Complex, Mozambique

W.D. Maier, Sarah-Jane Barnes, L.D. Ashwal, and C. Li, 2001

Abstract
The Tete Complex of Mozambique is a composite mafic-ultramafic intrusion that consists largely of pyroxenites and gabbros, with lesser amounts of coarse-grained troctolitic and anorthositic rocks reminiscent of massif-type anorthosites. The Complex thus may have potential to host reef-type PGE ores and/or massive Ni-Cu sulphides analogous to those found in the gabbro-troctolite hosted Voisey’s Bay deposits. We analysed 8 isotopically and mineralogically well-characterized samples of anorthosite, leucotroctolite, clinopyroxenite, gabbro, olivine melagabbro, dolerite, and a pegmatitic orthopyroxenite for major and trace elements, including PGE. The samples show little evidence forsignificant crustal contamination and are mostly undepleted in PGE relative to Cu, and to a lesser degree, Ni. If the samples are representative of the intrusion, this may indicate that the Complex has limited potential to host economic Ni-Cu sulphide ores. The potential for reef-type PGE ores remains less clear, in view of the occurrence of such ores in seemingly uncontaminated intrusions elsewhere (e.g. the Great Dyke).



Concentrations of rare earth elements in silicate rocks of the Lower, Critical and Main Zones of the Bushveld Complex

W. D. Maier, and S. -J. Barnes, 1998

Abstract
Concentrations of rare earth elements in the Lower, Critical and Main Zones of the Bushveld Complex at Union Section show distinct variation with stratigraphic height, in that Lower and Lower Critical Zone cumulates are LREE and Th enriched (Ce/Sm: 10–25, Ce/SmN: 3.65, Th/SmN: 6.17) over Main Zone cumulates (Ce/Sm: 4–10, Ce/SmN: approximately 2, Th/SmN: 1.57). The Upper Critical Zone constitutes a transitional interval in terms of REE concentrations (Ce/Sm: 9–17). This pattern is broadly coupled, albeit in a contraposed manner, to that in initial Sr isotopic ratio and confirms earlier studies that the Bushveld Complex crystallized from at least two compositionally distinct parental magmas. Mixing between the two magmas occurred as early as in the Lower Zone and is sensitively reflected by variation in Ce/Sm, Th/Sm, modal plagioclase and chromite. While REE contents of the Lower and Critical Zones are approximately 15% those of the corresponding B1 and B2 marginal rocks, the Main Zone cumulates have similar REE contents to the B3 marginal rocks. This suggests that the Main Zone rocks either crystallized from a highly viscous crystal mush without appreciable fractionation, or from magma strongly REE enriched over the marginal rocks, which would imply that the Main Zone is not represented in the suite of marginal rocks.

Keywords: Rare earth elements; Cumulate rocks; Bushveld Complex; South Africa



Compositional variation of laurite at Union Section in the western Bushveld Complex

W. D. Maier, H. M. Prichard, P. C. Fisher, and S. J. Barnes, 1998

Abstract
One hundred and forty five grains of laurite in polished sections of samples from one borehole through the major chromitite layers and some chromite-bearing silicate rocks of the Lower and Critical Zones of the western Bushveld Complex at Union Section have been located and analysed by scanning electron microscope. Ninety per cent by number of laurite grains are included within chromite, with the remainder being located on chromite-silicate grain boundaries, and in interstitial silicates and sulphides. The composition of laurite shows considerable variation within individual samples. Furthermore, there is no apparent correlation between whole-rock Ru and Cr contents in our samples, arguing against a model whereby lauriteexsolved from the chromite lattice. Based on a well-defined correlation between whole-rock S, PPGE (Rh+Pt+Pd), and IPGE (Os+Ir+Ru) contents, we favour a mechanism whereby laurite crystallized from segregating sulphide melt and was subsequently entrapped by growing chromite grains.



Trace elements in sulfide inclusions from Yakutian diamonds

G. P. Bulanova, W. L. Griffin , C. G. Ryan, O. Y. Shestakova, S.-J. Barnes, 1996

Abstract
Sulfide inclusions in diamonds may provide the only pristine samples of mantle sulfides, and they carry important information on the distribution and abundances of chalcophile elements in the deep lithosphere. Trace-element abundances were measured by proton microprobe in >50 sulfide inclusions (SDI) from Yakutian diamonds; about half of these were measured in situ in polished plates of diamonds, providing information on the spatial distribution of compositional variations. Many of the diamonds were identified as peridotitic or eclogitic from the nature of coexisting silicate or oxide inclusions. Known peridotitic diamonds contain SDIs with Ni contents of 22-36%, consistent with equilibration between olivine, monosulfide solid solution (MSS) and sulfide melt, whereas SDIs in eclogitic diamonds contain 0-12% Ni. A group of diamonds without silicate or oxide inclusions has SDIs with 11-18% Ni, and may be derived from pyroxenitic parageneses. Eclogitic SDIs have lower Ni, Cu and Te than peridotitic SDIs; the ranges of the two parageneses overlap for Se, As and Mo. The Mo and Se contents range up to 700 and 300rppm, respectively; the highest levels are found in peridotitic diamonds. Among the in-situ SDIs, significant Zn and Pb levels are found in those connected by cracks to diamond surfaces, and these elements reflect interaction with kimberlitic melt. Significant levels of Ru (30-1300rppm) and Rh (10-170rppm) are found in many peridotitic SDIs; SDIs in one diamond with wustite and olivine inclusions and complex internal structures have high levels of other platinum-group elements (PGEs) as well, and high chondrite-normalized Ir/Pd. Comparison with experimental data on element partitioning between crystals of monosulfide solid solution (MSS) and sulfide melts suggests that most of the inclusions in both parageneses were trapped as MSS, while some high-Cu SDIs with high Pd-Rh may represent fractionated sulfide melts. Spatial variations of SDI composition within single diamonds are consistent with growth histories shown by cathodoluminescence images, in which several stages of growth and resorption have occurred within magmatic environments that evolved during diamond formation.



Partitioning of nickel, copper, iridium, rhenium, platinum, and palladium between monosulfide solid solution and sulfide liquid: Effects of composition and temperature

C. Li, S. -J. Barnes, E. Makovicky, J. Rose-Hansen and M. Makovicky, 1996

Abstract
Partitioning of Ni, Cu, and Pt-group elements (Ir, Rh, Pt, Pd) between monosulfide solid solution (Mss) and sulfide liquid has been investigated in the Fe-Ni-Cu-S system at 1000 and 1100°C and one atmosphere pressure. The Nernst partition coefficients (D = wt% in Mss/wt% in sulfide liquid) for Ni vary significantly from 0.19 to 1.17, while the values of DCu show a limited range of 0.17-0.27. The partition coefficients for Ir range from 1.06 to 13. Rhodium has a partition coefficient slightly lower than that of Ir under the same conditions, ranging from 0.37 to 8.23. The partition coefficients for Pt and Pd vary from 0.05 to 0.16, and from 0.08 to 0.27, respectively.

The partition coefficients depend strongly on the bulk S contents of the system. They increase with increasing S contents in both Mss and liquid. Platinum, Pd, and Cu behave incompatibly during Mss crystallization, strongly partitioning into sulfide liquid. Nickel is incompatible in S-undersaturated systems and S-saturated systems. It becomes compatible when the system is S-oversaturated. Rhodium is compatible in S-saturated and S-oversaturated systems, but incompatible in S-undersaturated systems. Iridium changes from highly compatible through moderately compatible to slightly compatible when the system changes from S-oversaturated through S-saturated to S-undersaturated. The effect of temperature on metal partitioning is observed only in S-oversaturated systems, in which the partition coefficients for Ni and Rh increase with decrease of temperature.

The compatible behavior of Ir and Rh, and incompatible behavior of Pt and Pd and Cu under S-saturated conditions appears to support the hypothesis that the observed metal zonation in many sulfide ore deposits such as Sudbury, Ontario and Noril'sk, Siberia resulted from sulfide liquid fractionation.



Thrusting, magmatic intraplating, and metamorphic core complex development in the Archaean Belleterre-Angliers Greenstone Belt, Superior Province, Quebec, Canada

E. W. Sawyer and Sarah-Jane Barnes, 1994

Abstract
The Belleterre-Angliers Greenstone Belt in the Pontiac Subprovince is the youngest and southernmost of the greenstone belts in the Archaean Superior Province of the Canadian shield. The greenstone belt has been disrupted into three fragments: from north to south the Baby, Belleterre and Lac des Bois groups, respectively.

The Belleterre-Angliers Greenstone Belt has sheared contacts with younger Pontiac metasediments and older tonalites. The Pontiac metasediments dip beneath the Baby Group and are tectonically interlayered with the tonalites. Thus the Belleterre-Angliers belt is interpreted to be a thrust sheet (6 km thick in the north, thinner in the south) above a mid-crustal duplex consisting of metasedimentary and tonalite imbricates. The thrust sheet was transported from the north, but age and geochemical constraints rule out derivation from the Abitibi Subprovince. It probably roots in a band of ultramafic rocks that mark a major south-vergent thrust in the northern Pontiac Subprovince.

After emplacement the Belleterre-Angliers sheet and the adjacent Pontiac metasediments were deformed by a system of linked, extensional shears and transfer faults that border dome or antiformal areas in a style similar to Phanerozoic metamorphic core complexes. The orientation of shear zones and foliations indicates that the regional far-field stresses were compressional both before and after extensional shearing. Since the system of extensional shears is intruded by a suite of monzodiorite, granodiorite, syenodiorite and syenite plutons (with high Mg number) it is concluded that crustal inflation caused by the injection (intraplating) of large volumes of magma perturbed the regional compressive stress field and facilitated extensional faulting.



The behaviour of platinum-group elements during partial melting, crystal fractionation, and sulphide segregation: An example from the Cape Smith Fold Belt, northern Quebec

S-J. Barnes and C. P. Picard, 1993

Abstract
The behaviour of the Pt-group elements (PGE) during both normal igneous processes and during the formation of PGE deposits is poorly understood. This is in part because of the limited data set available for nonmineralized rocks. Accordingly, PGE concentrations have been determined in komatiitic basalts associated with Ni-Cu sulphide deposits rich in PGE from the Proterozoic Cape Smith Fold Belt of northern Quebec.

The lavas have been divided into olivine-phyric, pyroxene-phyric, and plagioclase-phyric on the basis of the phenocrysts present. Olivine-phyric lavas have Mg#s greater than 0.66 and are thought to be primary partial melts. The bulk partition coefficients of the PGE during partial melting calculated from the average of olivine-phyric basalts are 6, 2, 0.6, and 0.2 for Ir, Rh, Pt, and Pd, respectively, which indicates a decrease in compatibility of the PGE in this order. The pyroxene and plagioclase-phyric lavas are thought to have formed from the olivine-phyric lavas by crystal fractionation of olivine and chromite in the pyroxene-phyric lavas and olivine, chromite, and pyroxene in the plagioclase-phyric lavas. Ir shows a strong positive correlation with Mg#, Cr, and Ni, and this is attributed to Ir having partitioned into olivine or chromite. If olivine, chromite, and clinopyroxene were the only phases to have crystallized, then Pd should have behaved as an incompatible element; however, it does not correlate with the lithophile incompatible elements. It does, however, correlate with Rh and Pt. This is interpreted to suggest that Rh, Pt, and Pd were controlled by sulphide segregation. The removal of a small amount of sulphide along with the olivine and chromite during crystal fractionation could have resulted in the mildly compatible behavior exhibited by Rh, Pt, and Pd.

The composition of the sulphides in the sulphide deposits may be modelled by assuming that the silicate magma from which they segregated was similar in composition to the spinifex-textured komatiites containing 15% MgO.



The use of metal ratios in prospecting for platinum-group element deposits in mafic and ultramafic intrusions

Sarah-Jane Barnes, 1990

Abstract
Sulphides act as collectors of platinum-group elements (PGE), therefore, if a magma is to contain sufficient PGE to form a deposit within an intrusion, it should not have experienced sulphide segregation prior to emplacement in its present position. Prior sulphide segregation will have scavenged the PGE from the magma and deposited them elsewhere, presumably at depth. If the PGE have not been scavenged from the magma by a prior sulphide segregation, then they may be concentrated within the intrusion at the site where the first sulphides formed. Platinum-group elements have much higher partition coefficients into sulphides than do Ni and Cu and therefore the Ni and Cu to PGE ratios are strongly influenced by the segregation of sulphides. Thus, in order to decide whether an intrusion has the potential to contain a PGE deposit, the Ni and Cu to PGE ratios of the chilled margins should be considered. If the chilled rocks are not depleted in PGE relative to Ni and Cu, the intrusion may contain a PGE deposit. In order to locate the PGE within the intrusion, the Ni and Cu to PGE ratios of the rocks across a stratigraphic section can be used to find the level where sulphide saturation first occurred. The situation is complicated by the fact the Ni and Cu to PGE ratios are also influenced by olivine and chromite chrystallization. To eliminate the effects of removal of PGE (Os, Ir, Ru) and Ni by olivine and chromite it is necessary to consider the ratios of Pd/Ir or Pd/Rh or Pd/Pt versus Ni/Cu.



The petrography and geochemistry of komatiite flows from the Abitibi Greenstone Belt and a model for their formation

Sarah-Jane Barnes, 1985

Abstract
The petrology and geochemistry of komatiites at four localities in the Abitibi Greenstone Belt are presented. The effects of alteration, crystal fractionation and partial melting on their geochemistry are considered.

It is concluded that Na, K, Ca, Sr and Rb have been disturbed during seafloor alteration, but the other major elements and the trace elements, Ni, Cr, Co, Sc, REE, Y, Zr and Hf have not been mobilized.

Variations in the geochemistry within the komatiite flows are the result of olivine and sulphide fractionation. The komatiites consisting of branching olivines (skeletal textured komatiites) and coarse parallel plate olivines (A3 spinifex textured zones) may represent crescumulates, rather than frozen liquids. Mass balance calculations indicate that the flows were open systems and could represent lava tubes.

26-34% partial melting of a primitive earth mantle (1.4 × C-1 chondrite) will produce melts similar to the komatiites, at each deposit, in terms of major elements, Ni, Cr, Co, Sc, HREE, and Y. However, on average, the concentrations of highly incompatible elements; LREE, Zr and Hf, in the komatiites only are 60–70% of the modelled concentrations, thus indicating that the komatiites were derived from a mantle depleted in highly incompatible elements.



The origin of the fractionation of platinum-group elements in terrestrial magmas

Sarah-Jane Barnes, A. J. Naldrett and M. P. Gorton, 1985

Abstract
The platinum-group elements (PGE's), when chondrite normalized, have been found to be fractionated in order of descending melting point (Os, Ir, Ru, Rh, Pt, Pd and Au). Mantle-derived material (garnet lherzolite and spinel lherzolite xenoliths and alphine peridotites) have essentially unfractionated PGE patterns. Periotitic komatiites have mildly fractionated patterns (Pd/Ir = 10), pyroxenitic komatiites are slightly more fractionated (Pd/Ir = 30). Both continental and ocean-floor basalts are highly fractionated (Pd/Ir = 100). Data from intrusive rocks show a large range in PGE fractionation from Pd-depleted chromities of ophiolites (Pd/Ir = 0.1) to the extreme Pd enrichment in the JM Reef of the Stillwater Complex (Pd/Ir = 865).

Some possible mechanisms for the origin of this fractionation are: alteration, partial melting and crystal fractionation. Carbonate alteration affects Au and Pt and hydrothermal alteration mobilizes Pd. Solid substitution of Ir (and associated Os and Ru) into olivine and chromite, during crystal fractionation or partial melting is rejected as a mechanism of fractionating the PGE's. It is suggested; that the major factor in PGE fractionation is the differences in solubility of the PGE's in a silicate magma, that Pd, Pt and Rh are more soluble than Os and Ir, which form an alloy and Ru which forms laurite. These differences in PGE solubility could fractionate the PGE's during partial melting or crystal fractionation. During crystal fractionation prior to Fe-Ni-Cu sulphur saturation the low solubility of Os, Ir and Ru leads to the formation of Os-Ir alloys and RuS2 in the magma. These may then be settled out of the magma by whatever phase is crystallizing and the remaining magma becomes fractionated in PGE's.



An alternative model for the Damara Mobile Belt: Ocean crust subduction and continental convergence

Sarah-Jane Barnes, and Edward W. Sawyer, 1980

Abstract
The Pan-African Damara Mobile Belt has previously been described as ensialic, possibly resulting from a modified aulacogen.

Three features of the Damara Mobile Belt are difficult to reconcile with ensialic models. Firstly, the complex asymmetrical structural pattern of linear zones, with up to 80% shortening across the belt. Secondly, the markedly asymmetric metamorphic pattern broadly follows the structural pattern forming two distinct, parallel metamorphic belts of relatively high (northern belt) and low (southern belt) geothermal gradients, respectively. Abundant granitic intrusions occur in the high-grade metamorphic belt. Thirdly, the evolution of the Damara igneous rocks; the early (Nosib) igneous rocks are alkali; mid-Damara (Matchless Member) amphibolites resemble oceanic-floor basalts. Depleted upper-mantle material representing oceanic lithosphere was tectonically emplaced into the Damara metasediments during early tectonism. An extensive calc-alkali suite (the Salem Suite) intruded the high-grade metamorphic belt during a long period spanning most of the Damara tectonism.

A model invoking the formation of alkali rocks, followed by the development of oceanic crust, initiation of northwestward subduction and ocean closure terminating in continental collision is considered to explain the major features.