Compositional Variability of Rutile in Hydrothermal Ore Deposits
Rutile is a relatively common accessory phase in many geological environments, and although it is almost always composed dominantly of TiO2, it is also associated with a wide range of minor and trace element substitutions. The most prominent minor elements that occur in rutile are Fe, Cr, V, Nb and Ta. Like Ti, the latter two elements are essentially immobile in most non-magmatic metallic ore deposits, and their concentrations in rutile are largely influenced by precursor mineral compositions. Iron, Cr and V concentrations vary considerably in rutile hosted by ore deposits, and reflect combinations of precursor mineral composition and the bulk chemistry of the local mineralized or altered rock environment. However, in hydrothermal alteration zones, rutile compositions are clearly anomalous compared to those in unaltered host rocks, and have distinctive elemental associations and substitutions in different types of ore deposits. We have evaluated the mineral chemistry of rutile in >40 ore deposits worldwide. In general, rutile in volcanogenic massive sulfide deposits contains Sn (and locally W, Sb and/or Cu). Rutile from mesothermal and related gold deposits invariably contains W, and in some of the larger and more important deposits, also contains Sb and/or V. Tungsten-bearing detrital rutile grains from the Witwatersrand suggest that paleoplacer mineralization may have had a mesothermal/orogenic gold source. In some magmatic-hydrothermal Pd-Ni-Cu deposits, rutile contains Ni and Cu. Rutile associated with granite-related Sn deposits has strongly elevated concentrations of Sn and W, and granite-pegmatite W-Sn deposits contain rutile with these elements plus Nb and Ta. The Olympic Dam deposit hosts rutile that is enriched in W, Sn and Cu. Rutile associated with porphyry and skarn Cu and Cu-Au deposits tends to contain elevated W, Cu (and sometimes V). Although many ore deposits have well-defined and diagnostic rutile compositions, there are some compositional overlaps between mineralization types. Nevertheless, element combinations and ratios can be used to distinguish qualitatively between rutile compositions for most ore deposit types, and statistical methods can be used to provide more quantitative evaluation. Rutile occurs in significant abundance (typically 0.05 to 0.5 vol%) in most metallic ore deposits and is most plentiful in sulfidic systems where high fS2 and/or fO2 conditions stabilize rutile in the presence of minerals such as pyrite and hematite. Rutile is also persistent in weathering environments, and is likely to survive transport by glacial and fluvial processes. As a common component of heavy mineral sands, rutile is readily separable by routine magnetic, heavy liquid, and other density methods. These features, combined with the sensitive compositional variations in altered and mineralized rocks noted above, and the relative ease of analyses by routine electron microprobe methods, suggest that rutile has considerable potential as a geochemical indicator mineral for hydrothermal ore deposits, analogous to the kimberlite indicator minerals such as Cr-pyrope, magnesiochromite and picroilmenite that are used regularly in diamond exploration.
Contrasting Platinum-Group Element and Chalcophile Element Contents in Pyrrhotite, Pentlandite and Chalcopyrite From Different Environments
It is now possible to determine the platinum-group element (PGE) and chalcophile element contents of pyrrhotite (Po), pentlandite (Pn) and chalcopyrite (Ccp). This information may be used to: a) Improve recovery of important economic elements from ore; b) Consider the petrogenesis of the rocks. We have determined the PGE and other chalcophile element contents of Po, Pn and Ccp from a meteorite, a subvolcanic sill, the Merensky Reef Bushveld, Great Dyke, JM Reef Stillwater, AP and PV Reefs Penikat and Creighton Mine at Sudbury. The aims of these studies are to determine which phases host the elements and what implications the host phases have for the petrogenesis of the rocks involved. Sulfides from the meteorite and the subvolcanic sill have been chilled fairly rapidly and experienced very little subsolidus re-equilibration. The Bushveld and Great Dyke sulfides cooled slowly and thus had longer to exsolve. The Stillwater and Penikat sulfides were metamorphosed post-intrusion thus these sulfides have been reheated. The Sudbury sulfides have been deformed and metamorphosed and are the major phases whereas in all other cases the sulfides were minor phases in the rocks. In all cases Pt and Au are not present in the sulfides. Platinum is generally found as Pt-arsenides, Pt-bismuth- tellurides or Pt-alloys as inclusions in the sulfides. Palladium is generally hosted principally by Pn. Osmium and Ru are present in Po and Pn. In the meteorite, the subvolcanic sill and the unmetamorphosed intrusions Re, Ir and Rh are also present in Po and Pn. However in the metamorphosed and deformed sulfides Ir and Rh are present largely as sulfarsenides and as various Re minerals included in the sulfides. Thus in the least metamorphosed and deformed rocks all the PGE except Pt are present in the sulfides. In the metamorphosed rocks Re, Ir and Rh tend to form inclusions in the sulfides. We suggest that this is the result of exsolution during reworking of the sulfides. The concentration of Re, Os, Ir, Ru and Rh in Po and Pn is consistent with the formation of Po and Pn from monosulfide solid solution. However the mechanism for the concentration of Pd in pentlandite is not understood.
Incorporation of platinum-group elements and cobalt into subsidiary pyrite in alkalic Cu-Au porphyry deposits: significant implications for precious metal distribution in felsic magmatic-hydrothermal systems
Certain alkalic porphyry Cu-Au systems contain significant concentrations of the platinum-group elements (PGE) Pd and Pt, and may serve as important unconventional resources for these metals. Bulk rock analyses of ore styles from these deposits show no correlation between the PGE and Cu-Au abundance, suggesting that the timing/mechanisms of introduction and precipitation for the PGE and Cu-Au were not the same. To elucidate some uncertainties concerning PGE enrichment, we have performed a mineralogical evaluation of two PGE-bearing porphyry systems in British Columbia (the Afton and Mount Milligan deposits) with the aid of a variety of microanalytical techniques (LA-ICPMS, SEM, EMP). Discrete PGE mineralogy in these systems is predominantly represented by Hg-rich Pd-Pt-As-Sb species (naldrettite-stibiopalladinite-sperrylite) and Pd-Te-Hg species (kotulskite-temagamite). However, these mineral phases are unambiguously late- stage (with carbonate-chlorite alteration) and contribute insignificantly (<5-10%) to the total Pd+Pt grade based on mass balance calculations. Similarly, LA-ICPMS analyses of chalcopyrite, bornite, oxides and various common sulfosalts show that these do not contribute any Pd+Pt to the bulk grades. Suprisingly, pyrite is the predominant carrier of PGE. It occurs in trace to minor abundances and predates both the Cu-Au mineralising event and the late stage carbonate-chlorite alteration. LA-ICPMS analyses of pyrite show that at least 90% of the bulk Pd+Pt occurs within this atypical host mineral. The PGE are highly enriched in the cores of the pyrite grains (up to 90 ppm and 20 ppm, respectively) and their abundance correlates well with Co (up to 4 wt%). The rims of the pyrite grains are Co-PGE-poor but Ni-rich. Early Co- PGE enrichment in these deposits may implicate a mafic contributor to the PGE tenor of the deposits, but also a potential crystallographic (substitutional) control on Pd partitioning into pyrite. At Afton, the former is supported by LA-ICPMS analyses of primary silicate melt inclusions within leucititic, high Mg basalt flows that are coeval with the porphyry rocks. These glassy inclusions preserve Pd/Pt ratios that are of the same order of magnitude as the bulk mineralization, suggesting possibly a common parental melt source for these metals coupled with limited fractionation of Pd from Pt during melt fractionation or mixing, melt ascent, fluid exsolution, and pyrite precipitation. The results of this study show that minor pyrite is an important repository of PGE in alkalic (shoshonitic) Cu- Au porphyry deposits. PGE precipitation in pyrite was coincident to Co enrichment during a hydrothermal event that pre-dated Cu and Au precipitation. From the perspective of resource evaluation, this unusual association may lead to misrepresentation of PGE grades if analyses of only Cu concentrates are obtained. From the point of view of exploration, recognition of bulk Co anomalies will not be useful because pyrite is not the only mineral controlling Co abundance in the system (spinels, chalcopyrite, mafic silicates) and is present in abundances of only 2-5 vol%. On the other hand, recognition of Co-rich pyrites in heavy mineral separates or thin section (detectable by routine microprobe) may be a viable tool for discriminating between PGE-barren and PGE- bearing systems since it is only the Co-rich pyrite that is associated with Pd and Pt enrichment.
The Distribution of Platinum-Group Elements and Other Chalcophile Elements Among Mineral Phases of the Ni-Cu-PGE Deposit of Creighton Mine, Sudbury, Canada: A Quantitative Mass Balance
The distribution of PGE and other chalcophile elements among minerals of the Creighton Ore Body of the Sudbury Ni-Cu-PGE deposit has been investigated in order to: 1) better understand the petrogenesis of the ores and 2) aid efficient extraction of the PGE which are recovered as a by-product of mining these ores. A mass balance has been carried out by determining the concentrations of the elements in the; whole rock, the base metal sulphides (BMS) and the associated accessory minerals, including platinum-group minerals (PGM). Calculations show that the BMS host the majority (70-95%) of Co, Os, Pd, Ru and Se, that BMS host some (∼ 50%) Ag, Bi, Cd, Te and Zn but very little (< 20%) Ir, Rh, Re, Pt, Au, As, Pb, Sn and Sb. Discrete PGM and associated accessory minerals (e.g. galena (PbS), sphalerite (Zn(Cd)S), Ag-pentlandite and numerous Bi-Te phases) account for the trace elements that are 'missing' from the sulfide mass balance. Iridum, Rh and Pt are hosted by sulfarsenide-bearing PGM within the BMS. These comprise irarsite (IrAsS)- hollingworthite (RhAsS) cores, containing some Pt, Os and Ru, and Ni-cobaltite (NiCoAsS) rims. Rhenium- phases, such as (CuFe)(ReMoOs)S and RePb(Bi)S, are also hosted by the BMS. Palladium-bearing PGM (michenerite PdBiTe), Pt-bearing PGM (sperrylite PtAs2) and Au (electrum AuAg) are present but rare and are hosted by both BMS and silicates. The Ni-cobaltite is estimated to have equilibrated at 550-600°C, and thus the irarsite-hollingworthite cores probably formed at > 600°C possibly by exsolution from the BMS during cooling. Michenerite was observed to infill fractures in amphibole indicating that some Pd can remobilise during metamorphism and deformation at lower temperatures (∼ 490°C).
Chromite Does not Control IPGE (Os, Ir, Ru)
It has long been recognized that chromite-rich rocks are enriched in Ir-group Platinum-group elements (IPGE; Os, Ir, Ru). Experiments have shown that IPGE partition into spinel under oxidizing conditions. On the other hand small grains of laurite (Ru,Os,Ir)S2 are found associated with chromite in many cases. Thus a debate has arisen as to which phase controls the IPGE. We have determined the concentrations of PGE in both the whole rocks and the chromites from a number of different rock types (boninites, MORB and ophiolites chromitite), to investigate how much of the IPGE could be present in chromite. In all cases the concentrations of the IPGE in the chromite were less than the LA-ICP-MS detection limits of 9 Os, 3 Ir and 40 Ru ppb. Using the whole rock values and the detection limit for chromite as the maximum amount of IPGE possible in chromite and we calculated the maximum amount of IPGE that chromite could host. Chromite controls less than 24 weight percent of Os, less than 10 percent of the Ir, less than 33 percent of the Ru. In view of the absence of IPGE in the chromite and the presence of laurite it is logical to assume that the main host for the IPGE is laurite. Although laurite occurs as inclusions in chromite it is not thought to be an exsolution product from chromite, because chromites from both lavas and plutonic rocks do not appear to control IPGE. In the lavas laurite would not have an opportunity to exsolve. Furthermore laurite contains S which not be expected to be present in chromite and finally in many cases the laurite occurs with silicate phases that could not have exsolved from the chromite. Thus the correlation of Cr and IPGE is probably due to co-precipitation of laurite and chromite, rather than partitioning of IPGE into chromite.
A Comparison of Fracture and Matrix Indicator Minerals in the Dispersion Halo Around the Dibs Rare-Metal Pegmatite, Manitoba
It is well known that rare-metal pegmatites are surrounded by metasomatic halos enriched in alkali elements, notably Li, Rb and Cs. However false lithogeochemistry anomalies are also common and these may be caused by minerals along fractures. The Dibs LCT-type pegmatite is located approximately 3 km east of the world-class Tanco pegmatite and is an ideal test site for investigating metasomatic halos because it is completely buried, intersected by diamond drill holes at a depth of approximately 80 m below the surface. The Dibs pegmatite is roughly 100x500x65 m and zoning includes an Upper Intermediate zone with petalite and a Central Intermediate zone with Ta-Sn mineralization. It intruded the Bernic Lake Formation metabasite that prior to the intrusion was metamorphosed to amphibolite facies (amphibole-plagioclase-biotite-quartz- ilmenite). Three diamond drill holes were examined in detail. Whole-rock compositions were determined by ICP-MS and matrix and fracture minerals were analysed by EMP and LA-ICP-MS. Analyses of whole-rocks indicate that Li, as expected, was the most mobile alkali; Li contents at the tops of the holes were 28-73 ppm, which is at least twice the threshold limit. By contrast Rb and Cs whole-rock contents at the tops of the holes were at or below threshold values, 11-18 ppm and 3.6-7.0 ppm, respectively. Li, Rb and Cs values increase toward the upper contact with the Dibs pegmatite and at the contact the metabasite contains up to 1048 ppm Li, 169 ppm Rb and 68 ppm Cs. Matrix minerals are typically 0.5 to 3 mm across and the alkali elements are most concentrated in biotite; it contains 75-4132 ppm Li, which is almost an order of magnitude higher than whole- rock Li contents and nearly two orders of magnitude more than Li in amphibole. The Rb and Cs whole rock budget is controlled by biotite, which contains 186-6081 and 15-6138 ppm, respective. Also of note, the Nb-Ta contents of biotite are low, approximately 12 and 0.5 ppm, respectively, indicating a lack of Nb-Ta metasomatism. Finally, Tl in biotite is also a good indicator of rare-metal pegmatites. Its abundance in biotite is strongly correlated to Rb and ranges from 1-45 ppm. Two main fractures/vein types were observed. Early late-kinematic shear veins are comprised almost entirely of tremolite. Younger, planar, open-spaced-filled fractures consist of chlorite-titanite-carbonate-clinozoisite. At the pegmatite-metabasite contact, matrix and shear-vein amphiboles contain similar amounts of Li, approximately 400 ppm, whereas at the top of the drill holes shear-vein amphibole contains slightly more Li that the matrix amphibole, approximately 15 versus 5 ppm, respectively. This indicates that the Dibs pegmatite was emplaced during the late kinematic stage of regional deformation. However, the pegmatite is non- deformed, emplaced during brittle deformation in an en echelon system, thus the later chlorite-bearing fractures may also be coeval with the Dibs pegmatite. Near the pegmatite contact fracture chlorite attain a maximum Li value over 6000 ppm and at the top of drill holes still contain approximately 200-400 ppm Li, which is an order of magnitude higher than whole-rock values. Fracture chlorite is thus one of the dominant controls of dispersion around the Dibs pegmatite. It is also worth noting that titanite associated with fracture chlorite at a distance of greater than 1 m from the pegmatite contains < 10 ppm Sn, < 100 ppm Nb and 10-20 ppm Ta indicating that these elements were not mobile during chlorite alteration.
Apatite as a paleohydrothermal fluid recorder in Carlin-type gold deposits
Apatite is a common accessory mineral in most rocks. A variety of trace elements can be substituted into apatite, meaning that apatite has the potential to record changes in the chemistry of ore-forming hydrothermal fluids. This study focuses on variations in apatite texture and chemistry around the world-class Carlin-type Au deposits of NE Nevada. These deposits are characterized by cryptic alteration of calcareous and siliciclastic sedimentary rocks induced by acidic, low-temperature (150-220 °C) ore fluids. A large database of apatite fission track (AFT) samples collected from NE Nevada (Hickey, unpublished data) is being used to examine relationships between apatite fission track ages, textural zonation within apatite crystals and apatite trace element composition. AFT data from "background" samples collected away from hydrothermal mineralization and Cenozoic igneous stocks reveal that regional uplift occurred in the Cretaceous at ~70-60 Ma. In comparison, AFT data from samples around gold mineralization reveal an Eocene heating event, interpreted as the result of hydrothermal reheating by the Carlin Au-forming system (Cline et al., 2005). Optical cathodoluminescence observations reveal that some apatite from Au-bearing material (Eocene AFT age) has embayed cores, and at least four generations of overgrowths (typically ∼10 μm wide overgrowths), which may also be embayed. In comparison, apatite from hydrothermally altered, but unmineralized, material has a single overgrowth generation. 'Background' apatites do not have significant overgrowths. Current research is characterizing the trace element composition of apatite cores and overgrowths via SIMS. We propose that apatite textures and trace element composition record hydrothermal fluid interactions. Applications include using apatite to detect the signature of hydrothermal fluids in rocks cryptically altered by low-temperature hydrothermal systems, or detecting mineralization by examining detrital apatites deposited in sedimentary environments. References Cline, J et al. 2005. Econ. Geol. 100th anniversary volume. Carlin type gold deposits in Nevada: critical geological characteristics and viable models. pp 451-484