Mineralium Deposita (v.51, #7)

The chromitites of the Bushveld Complex in South Africa contain vast resources of platinum-group elements (PGE). However, knowledge of the distribution and the mineralogical siting of the PGE in the lower group (LG) and middle group (MG) chromitite seams of the Bushveld Complex is limited. We studied concentrates from the LG-6 and MG-2 chromitites of the western Bushveld Complex by a variety of microanalytical techniques. The dominant PGM are sulfides, namely laurite, cooperite-braggite, and malanite-cuprorhodsite, followed by PGE-sulfarsenides, sperrylite, and Pt-Fe alloys. Laurite is the most abundant PGM (vol%). The matching sets of PGM present in the LG and MG chromitites of both the western and the eastern Bushveld Complex, and in the UG-2 chromitite, show strong similarities which support the assumption of a characteristic and general chromitite-related PGM assemblage. Palladium and Rh contents in pentlandite are low and erratic although maximum contents of 7730 ppm Pd and 6020 ppm Rh were detected. Rare thiospinels of the polydymite-linnaeite-greigite series have PGE contents of 1430 ppm Pt, 5370 ppm Rh, and 1460 ppm Pd. The various PGE occur in different deportment: Platinum is generally present in the form of discrete PGM (sulfides, arsenides, alloys). Palladium is present as a large variety of discrete PGM and also incorporated in pentlandite. Rhodium forms discrete PGM and is occasionally present in pentlandite. The IPGE (Os, Ir, and Ru) are dominantly incorporated in laurite (often as inclusions in chromite) and also occur as sulfarsenides.

Platinum-group element concentrations in pyrite from the Main Sulfide Zone of the Great Dyke of Zimbabwe by R. Piña; F. Gervilla; S.-J. Barnes; T. Oberthür; R. Lunar (853-872).
The Main Sulfide Zone (MSZ) of the Great Dyke of Zimbabwe hosts the world’s second largest resource of platinum-group elements (PGE) after the Bushveld Complex in South Africa. The sulfide assemblage of the MSZ comprises pyrrhotite, pentlandite, chalcopyrite, and minor pyrite. Recently, several studies have observed in a number of Ni-Cu-PGE ore deposits that pyrite may host significant amounts of PGE, particularly Pt and Rh. In this study, we have determined PGE and other trace element contents in pyrite from the Hartley, Ngezi, Unki, and Mimosa mines of the Great Dyke by laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS). Based on the textures and PGE contents, two types of pyrite can be differentiated. Py1 occurs as individual euhedral or subhedral grains or clusters of crystals mostly within chalcopyrite and pentlandite, in some cases in the form of symplectitic intergrowths, and is PGE rich (up to 99 ppm Pt and 61 ppm Rh; 1.7 to 47.1 ppm Ru, 0.1 to 7.8 ppm Os, and 1.2 to 20.2 ppm Ir). Py2 occurs as small individual euhedral or subhedral crystals within pyrrhotite, pentlandite, and less frequently within chalcopyrite and silicates and has low PGE contents (<0.11 ppm Pt, <0.34 ppm Rh, <2.5 ppm Ru, <0.37 ppm Ir, and <0.40 ppm Os). Py1 contains higher Os, Ir, Ru, Rh, and Pt contents than the associated pyrrhotite, pentlandite, and chalcopyrite, whereas Py2 has similar PGE contents as coexisting pyrrhotite and pentlandite. Based on the textural relationships, two different origins are attributed for each pyrite type. Py1 intergrowth with pentlandite and chalcopyrite is inferred to have formed by late, low temperature (<300 °C) decomposition of residual Ni-rich monosulfide solid solution, whereas Py2 is suggested to have formed by replacement of pyrrhotite and pentlandite caused by late magmatic/hydrothermal fluids.
Keywords: Pyrite; Base metal sulfides; PGE; LA-ICP-MS; Main sulfide zone; Great Dyke; Zimbabwe

Alto de la Blenda is a ∼6.6-Ma intermediate-sulphidation epithermal vein system in the Farallón Negro Volcanic Complex, which also hosts the 7.1-Ma porphyry–Cu–Au deposit of Bajo de la Alumbrera. The epithermal vein system is characterised by a large extent and continuity (2 km × 400 m open to depth × 6 m maximum width) and an average gold grade of ∼8 g/t. The vein is best developed within an intrusion of a fine-grained equigranular monzonite, interpreted as the central conduit of a stratovolcano whose extrusive activity ended prior to porphyry–Cu–Au emplacement at Bajo de la Alumbrera, which is in turn cut by minor epithermal veins. The Alto de la Blenda vein consists predominantly of variably Mn-rich carbonates and quartz, with a few percent of pyrite, sphalerite, galena and other sulphide and sulphosalt minerals. Four phases of vein opening, hydrothermal mineralisation and repeated brecciation can be correlated between different vein segments. Stages 2 and 3 contain the greatest fraction of sulphide and gold. They are separated by the emplacement of a polymictic breccia containing clasts of quartz feldspar porphyry as well as basement rocks. Fluid inclusions in quartz related to stages 2 to 4 are liquid rich with 2–4 wt% NaCl(eq). They homogenise between 160 and 300 °C, with very consistent values within each assemblage. Vapour inclusions are practically absent in the epithermal vein. Quartz fragments in the polymictic breccia contain inclusions of intermediate to vapour-like density and similar low salinity (∼3 wt% NaCl(eq)), besides rare brine inclusions containing halite. Laser ablation–inductively coupled plasma–mass spectrometry (LA–ICP–MS) analyses of epithermal inclusions indicate high concentrations of K, Fe, As, Sb, Cs, and Pb that significantly vary within and through subsequent vein stages. Careful consideration of detection limits for individual inclusions shows high gold concentrations of ∼0.5 to 3 ppm dissolved in the ore fluid, which contains variably high sulphur concentrations in excess over Fe and other chalcophile metals. Compositional variations are interpreted to reflect cooling and contraction of lower-density magmatic fluids at depth, like those preserved in porphyry clasts that were mechanically transported up by the polymictic breccia. Ore mineral precipitation from the magmatic fluid occurred by further cooling and possibly minor mixing with surface-derived water, leading to sulphide saturation, de-sulphidation of the magmatic fluid and consequent gold precipitation. The absence of flash boiling and/or reduction by carbonaceous host rocks has led to relatively modest but constant gold grades in the carbonate–base metal–gold veins of Alto de la Blenda.
Keywords: Epithermal; Gold; Farallón Negro Volcanic Complex; Alto de la Blenda; Fluid evolution; Ore mineralogy; LA–ICP–MS

The Duke Island Complex is one of the several “Ural-Alaskan” intrusions of Cretaceous age that occur along the coast of SE Alaska. Significant quantities of magmatic Ni–Cu–PGE sulfide mineralization are locally found in the complex, primarily within olivine clinopyroxenites. Sulfide mineralization is Ni-poor, consistent with petrologic evidence which indicates that sulfide saturation was reached after extensive olivine crystallization. Olivine clinopyroxenites were intruded by magmas that produced sulfide-poor, adcumulate dunites. As part of a study to investigate the potential for Ni-rich sulfide mineralization in association with the dunites, a Re–Os and S isotope study of the dunites, as well as sulfide mineralization in the olivine clinopyroxenites, was initiated. Importantly, recent drilling in the complex identified the presence of sulfidic and carbonaceous country rocks that may have been involved in the contamination of magmas and generation of sulfide mineralization. γOs (110 Ma) values of two sulfidic country rocks are 1022 and 2011. δ34S values of the country rocks range from −2.6 to −16.1 ‰. 187Os/188Os ratios of sulfide minerals in the mineralization hosted by olivine clinopyroxenites are variable and high, with γOs (110 Ma) values between 151 and 2059. Extensive interaction with Re-rich sedimentary country rocks is indicated. In contrast, γOs (110 Ma) values of the dunites are significantly lower, ranging between 2 and 16. 187Os/188Os ratios increase with decreasing Os concentration. This inverse relation is similar to that shown by ultramafic rocks from several arc settings, as well as altered abyssal dunites and peridotites. The relation may be indicative of magma derivation from a sub-arc mantle that had experienced metasomatism via slab-derived fluids. Alternatively, the relation may be indicative of minor contamination of magma by crustal rocks with low Os concentrations but high 187Os/188Os ratios. A third alternative is that the low Os concentrations and elevated 187Os/188Os ratios denote subsolidus interaction with seawater or meteoric water. δ34S values of the dunites range between −6.4 and 6.6 ‰, and are consistent with the addition of S during fluid–rock interaction and serpentinization. The sharp contrast between the Os isotope ratios of the dunites and those of the sulfide mineralization illustrate that magmas that were spatially part of the same intrusive system may have experienced very different histories of interaction with country rocks. An important corollary is that because of the concentrations of Os and S, elevated Os isotope ratios (a function of high Re concentrations) and variable sulfur isotope ratios of sulfidic and carbonaceous country rocks, both S and Os isotope data from the olivine clinopyroxenite-hosted sulfide mineralization, are consistent with less than ∼2 % of bulk rock contamination. Even lower fractional abundance values may be indicated if the contaminant was a S–C–Os-rich fluid or partial melt derived from the sulfidic–carbonaceous metasedimentary country rocks. Despite the low degrees of contamination, the amounts of Os and S in the sulfide mineralization that may have been derived from country rocks often exceed 50 %.
Keywords: Duke Island Complex; Ni–Cu–PGE sulfides; Dunite; Olivine clinopyroxenite; Os isotopes; S isotopes

Assimilation by mafic to ultramafic magmas of sulfur-bearing country rocks is considered an important contributing factor to reach sulfide saturation and form magmatic Ni–Cu–platinum group element (PGE) sulfide deposits. Sulfur-bearing sedimentary rocks in the Archean are generally characterized by mass-independent fractionation of sulfur isotopes that is a result of atmospheric photochemical reactions, which produces isotopically distinct pools of sulfur. Likewise, low-temperature processing of iron, through biological and abiotic redox cycling, produces a range of Fe isotope values in Archean sedimentary rocks that is distinct from the range of the mantle and magmatic Fe isotope values. Both of these signals can be used to identify potential country rock assimilants and their contribution to magmatic sulfide deposits. We use multiple S and Fe isotopes to characterize the composition of the potential iron and sulfur sources for the sulfide liquids that formed the Hart deposit in the Shaw Dome area within the Abitibi greenstone belt in Ontario (Canada). The Hart deposit is composed of two zones with komatiite-associated Ni–Cu–PGE mineralization; the main zone consists of a massive sulfide deposit at the base of the basal flow in the komatiite sequence, whereas the eastern extension consists of a semi-massive sulfide zone located 12 to 25 m above the base of the second flow in the komatiite sequence. Low δ56Fe values and non-zero δ34S and Δ33S values of the komatiitic rocks and associated mineralization at the Hart deposit is best explained by mixing and isotope exchange with crustal materials, such as exhalite and graphitic argillite, rather than intrinsic fractionation within the komatiite. This approach allows tracing the extent of crustal contamination away from the deposit and the degree of mixing between the sulfide and komatiite melts. The exhalite and graphitic argillite were the dominant contaminants for the main zone of mineralization and the eastern extension zone of the Hart deposit, respectively. Critically, the extent of contamination, as revealed by multiple S and Fe isotope systematics, is greatest within the deposit and decreases away from it within the komatiite flow. This pattern points to a local source of crustal contamination for the mantle-derived komatiitic melt and a low degree of homogenization between the mineralization and the surrounding lava flow. Coupled S and Fe isotope patterns like those identified at the Hart deposit may provide a useful tool for assessing the potential of a komatiitic sequence to host Ni–Cu–(PGE).
Keywords: Komatiite; Nickel; Isotope; Sulfur; Iron; Geochemistry

The Tamarack magmatic sulfide deposit is hosted by the Tamarack Intrusive Complex (1105.6 ± 1.2 Ma) in the Midcontinent Rift System. The most important sulfide mineralization in the Complex occurs in the northern part, which consists of two separate intrusive units: an early funnel-shaped layered peridotite body containing relatively fine-grained olivine (referred to as the FGO Intrusion) at the top, and a late gabbro-troctolite-peridotite dike-like body containing relatively coarse-grained olivine (referred to as the CGO Intrusion) at the bottom. Disseminated, net-textured, and massive sulfides occur in the base of the FGO Intrusion as well as in the upper part of the CGO Intrusion. The widest part of the CGO Intrusion also hosts a large semi-massive (net-textured) sulfide ore body locally surrounded by disseminated sulfide mineralization. Small massive sulfide veins occur in the footwall of the FGO Intrusion and in the wall rocks of the CGO dike. The sulfide mineralization is predominantly composed of pyrrhotite, pentlandite, and chalcopyrite, plus minor magnetite. Pyrrhotite containing the highest Ni and Co contents occurs in the FGO disseminated sulfides and in the CGO semi-massive sulfide ores, respectively. The most important platinum-group minerals associated with the base metal sulfides are sperrylite (PtAs2), sudburyite (PdSb), and michenerite (PdBiTe). Nickel shows a strong positive correlation with S in all types of sulfide mineralization, and Cu shows a strong positive correlation with S in the disseminated sulfide mineralization. At a given S content, the concentrations of Pt, Pd, and Au in the CGO disseminated sulfides are significantly higher than those in the FGO disseminated sulfides. The semi-massive sulfide ores are characterized by significantly higher IPGE (Ir, Os, Ru, and Rh) concentrations than most of the massive sulfide ores. With few exceptions, all of the various textural types of sulfide mineralization collectively show a good positive correlation between Pt and Pd, and between individual IPGE. At a given Pt or Pd content, however, the semi-massive sulfide ores have higher IPGE contents than the disseminated sulfide samples. Modeling results show that the variations in PGE tenors (metals in recalculated 100 % sulfide) in the Tamarack magmatic sulfide deposit are mainly controlled by variable R factors (magma/sulfide-liquid mass ratios) during sulfide-liquid segregation and subsequent monosulfide solid solution (MSS) fractionation during cooling. The initial contents of Ir, Pt, and Pd in the parental magma, estimated from the metal tenors of the disseminated sulfides, are 0.2, 2, and 1.8 ppb, respectively, which are ∼1/5 of the values for the PGE-undepleted primitive basalts of the Midcontinent Rift System. The variations of PGE tenors in the semi-massive and massive sulfide ores can be explained by MSS fractional crystallization from sulfide liquids. Extreme variations in the PGE contents of the massive sulfides may also in part reflect metal mobility during post-crystallization hydrothermal processes. The higher PGE tenors for the disseminated sulfides in the CGO dike relative to those in the FGO Intrusion are consistent with formation in a dynamic conduit where the early sulfide liquids left in the conduit by the FGO magma were subsequently upgraded by the subsequent surge of the CGO magma. The relatively low PGE tenors for the semi-massive and massive sulfides can be explained by lack of such an upgrading process for the sulfide due to their distal locations in a migrating conduit.
Keywords: Nickel; Platinum-group elements; R factor; Sulfide-liquid fractionation; Conduit-type deposit; Mafic magma

The Baishan porphyry Mo deposit (0.72 Mt; 0.06 % Mo) is located in the interior of the eastern Tianshan orogenic belt in Xinjiang, NW China. The deposit comprises 15 orebodies that are associated with monzogranite and granite porphyry stocks and are structurally controlled by roughly EW-trending faults. Secondary ion mass spectrometry (SIMS) zircon U–Pb dating of the monzogranite and granite porphyry yielded the Middle Triassic age (228 ± 2 to 227 ± 2 Ma), which coincide with the molybdenite Re–Os model ages ranging from 226 ± 3 to 228 ± 3 Ma. The Triassic monzogranite and granite porphyry belong to high-K calc-alkaline series and are characterized by high SiO2 and Al2O3 and low MgO, TiO2, and P2O5 concentrations, with negative Eu anomalies (δEu = 0.55–0.91). The least-altered monzogranite and granite porphyry yield uniform ε Nd(t) values from +1.6 to +3.6, and wide (87Sr/86Sr) i ratios ranging between 0.7035 and 0.7071, indicating that they were derived from the lower crust. In situ O–Hf isotopic analyses on zircon using SIMS and laser ablation multi-collector inductively coupled plasma mass spectrometry (LA-MC-ICP-MS) indicate that the δ18O and ε Hf(t) values of zircon from a monzogranite sample vary from 6.1 to 7.3 ‰ and +8.0 to +11.7, respectively, whereas zircon from a granite porphyry sample vary from 6.2 to 6.9 ‰ and +7.3 to +11.2, respectively. The geochemical and isotopic data imply that the primary magmas of the Baishan granite were likely derived from partial melts from the lower crust involving some mantle components. The Baishan Mo deposit and granitic emplacement were proposed to be most likely related to post-orogenic lithospheric extension and magmatic underplating. An extensional event coupled with the rising of hot mantle-derived melts triggered partial melting of the lower crust, as well as provided metals (Mo).
Keywords: Zircon U–Pb age; Molybdenite Re–Os dating; Sr–Nd–O–Hf isotopes; Baishan porphyry Mo deposit; Eastern Tianshan