Redox-controlled chalcophile element geochemistry of the Polaris Alaskan-type mafic-ultramafic complex, British Columbia, Canada

ABSTRACT The Early Jurassic Polaris Alaskan-type intrusion in the Quesnel accreted arc terrane of the North American Cordillera is a zoned, mafic-ultramafic intrusive body that contains two main styles of magmatic mineralization of petrologic and potential economic significance: (1) chromitite-associated platinum group element (PGE) mineralization hosted by dunite (±wehrlite); and (2) sulfide-associated Cu-PGE-Au mineralization hosted by olivine (±magnetite) clinopyroxenite, hornblendite, and gabbro-diorite. Dunite-hosted PGE mineralization is spatially associated with thin discontinuous layers and schlieren of chromitite and chromitiferous dunite and is characterized by marked enrichments in iridium-subgroup PGE (IPGE) relative to palladium-subgroup PGE (PPGE). Discrete grains of platinum group minerals (PGM) are exceedingly rare, and the bulk of the PGE are inferred to reside in solid solution within chromite±olivine. The absence of Pt-Fe alloys in dunite of the Polaris intrusion is atypical, as Pt-enrichment of dunite-hosted chromitite is widely regarded as a characteristic feature of Alaskan-type intrusions. This discrepancy appears to be consistent with the strong positive dependence of Pt solubility on the oxidation state of sulfide-undersaturated magmas. Through comparison with experimentally determined PGE solubilities, we infer that the earliest (highest temperature) olivine-chromite cumulates of the Polaris intrusion crystallized from a strongly oxidized ultramafic parental magma with an estimated log f(O2) > FMQ+2. Parental magmas with oxygen fugacities more typical of volcanic arc settings [log f(O2) ∼ FMQ to ∼ FMQ+2] are, in turn, considered more favorable for co-precipitation of Pt-Fe alloys with olivine and chromite. More evolved clinopyroxene- and hornblende-rich cumulates of the Polaris intrusion contain low abundances of disseminated magmatic sulfides, consisting of pyrrhotite and chalcopyrite with minor pentlandite, pyrite, and rare bornite (≤12 wt.% total sulfides), which occur interstitially or as polyphase inclusions in silicates and oxides. The sulfide-bearing rocks are characterized by strong primitive mantle-normalized depletions in IPGE and enrichments in Cu-PPGE-Au, patterns that resemble those of other Alaskan-type intrusions and primitive arc lavas. The absolute abundances and sulfur-normalized whole-rock concentrations (Ci/S, serving as proxy for sulfide metal tenor) of chalcophile elements, including Cu/S, in sulfide-bearing rocks are highest in olivine clinopyroxenite. Sulfide saturation in the relatively evolved magmas of the Polaris intrusion, and Alaskan-type intrusions in general, appears to be intimately tied to the appearance of magnetite. Fractional crystallization of magnetite during the formation of olivine clinopyroxenite at Polaris resulted in reduction of the residual magma to log f(O2) ≤ FMQ+2, leading to segregation of an immiscible sulfide melt with high Cu/Fe and Cu/S, and high PGE and Au tenors. Continued fractionation resulted in sulfide melts that were progressively more depleted in precious and base chalcophile metals. The two styles of PGE mineralization in the Polaris Alaskan-type intrusion are interpreted to reflect the evolution of strongly oxidized, hydrous ultramafic parental magma(s) through intrinsic magmatic fractionation processes that potentially promote sulfide saturation in the absence of wallrock assimilation.

Geochronology, Petrogenesis and Oxidation State of the Northparkes Igneous Suite, New South Wales, Australia: Implications for Magma Fertility

New geochronological and geochemical data for the barren and ore-associated suites from the Northparkes porphyry Cu-Au deposits, Australia, have implications for magma fertility. The Goonumbla and Wombin Volcanics and intrusions are barren in the Northparkes area. A sample from Wombin suite yielded a zircon U-Pb age of 433.8 ± 3.1 Ma, whereas the ore-associated porphyries yielded ages between 441.8 ± 3.7 and 436.3 ± 4.5 Ma. The bulk of the mineralization at Northparkes is associated with a K-feldspar-phyric quartz monzonite porphyry (K-QMP), which gave U-Pb zircon ages of 441.8 ± 3.7 and 441.1 ± 2.5 Ma. Whole-rock Sr-Nd isotope compositions of the Goonumbla, Wombin, and ore-associated suites are similar, with (87Sr/86Sr)i = 0.704112 to 0.704424 and εNd = 5.6 to 6.9, which is typical of primitive intraoceanic island arcs, and their Pb isotope values lie within the MORB array. Most of the zircons from the Wombin and ore-associated suites have arc mantle-like O-Hf isotope compositions, with δ18O values that vary from 6.13 to 4.95, and εHf(t) from 11.5 to 6. These results suggest that the Goonumbla, Wombin, and ore-associated suites originated from typical arc mantle. The magmas that produced the Goonumbla and Wombin suites were dominated by plagioclase-pyroxene fractionation, and the Wombin suite has a low oxidation state with ΔFMQ between ~0 and 1.5. They were relatively reduced and dry. This combination resulted in early sulfide saturation, probably without reaching fluid saturation. Trace element modeling shows that plagioclase-amphibole dominated the later stages of fractionation of the ore-associated suite, implying that it had a higher water content than the barren suites. It was also more oxidized (ΔFMQ from ~0 to 4). The result was late sulfide saturation, which was followed shortly thereafter by voluminous fluid release. As a consequence, the ore-forming fluid effectively transferred Cu and Au from the magma to the site of hydrothermal ore deposition. We suggest that the higher water content and oxidation state of the ore-associated suite was due to the deep underlying magma chamber, which was recharged by many more pulses of magma than the chamber that underlay the barren suites. This is more effective in raising the concentration of incompatible water and ferric iron in the residual melt than straight fractional crystallization. High oxygen fugacities and water contents played a significant role in determining the timing of sulfide and fluid saturation, respectively, and as a result, they had a critical influence on magma fertility.

Modeling the sulfide saturation in continuously assimilating magmatic systems with the Magma Chamber Simulator

<p>The timing and degree of immiscible sulfide precipitation in a magma effectively controls the formation of magmatic sulfide deposits and the budget of degassing sulfur species in volcanic systems. Besides the absolute sulfur (S) content, sulfide precipitation is strongly affected by the sulfur content at sulfide saturation (SCSS) in the host silicate melt. Assimilation of S-rich wall-rocks, such as black shales, effectively increases the S content in the magma, while simultaneously lowering the SCSS. Accordingly, assimilation has been identified as the most important process in the formation of many economically significant magmatic base metal sulfide deposit, especially in continental tectonic settings. Detailed understanding of the relation between wall-rock assimilation and sulfide saturation requires accurate thermodynamic models for open magmatic systems experiencing assimilation-fractional crystallization (AFC).</p><p>The Magma Chamber Simulator (MCS) is currently the only geochemical modeling software that considers the thermodynamic phase equilibria in open magmatic systems involving magma and wall-rock (and recharge) subsystems. We utilized the MCS to explore how assimilation affects the SCSS and S content of the magma. With the current lack of thermodynamic data for sulfides, we tentatively modeled S as a trace element and varied its compatibility to wall-rock in the different models. For a case study, we chose the mafic layered intrusions of Duluth Complex, Minnesota, which host some of the largest Cu-Ni sulfide deposits in the world. Assimilation of the adjacent black shale has been established as the main source for S in the deposits.</p><p>Our MCS models show in detail how continuous assimilation of the black shale lowers the SCSS of the melt. Partial melt from the black shale enriches the magma in SiO<sub>2</sub>, Al<sub>2</sub>O<sub>3</sub>, K<sub>2</sub>O, and H<sub>2</sub>O, while depleting FeO, MgO, CaO, and Na<sub>2</sub>O, which causes a first order decrease in the SCSS. The compositional change also replaces troctolitic cumulates (plagioclase, olivine ± clinopyroxene) with norite (plagioclase and orthopyroxene), which leads to more pronounced FeO depletion in the melt, further lowering the SCSS. On the other hand, the assimilated partial melt also increases the melt mass in the magma subsystem, which counteracts the S enrichment. Accordingly, in the model where S is compatible to the wall-rock residual, the degree of sulfide saturation only slightly increases relative to the same magma experiencing FC without assimilation.</p><p>More than half of the wall-rock S must partition to the assimilated partial melt in order to meet the S isotopic criteria of the modeled Cu-Ni-deposits. The main stage of sulfide precipitation is associated with ~30 wt.% crystallization of the assimilating host magma. The proportion of sulfides relative to silicates in these models is smaller than observed in the Duluth Complex deposits, which underlines the role of dynamic processes in concentrating sulfides from the silicate magma.</p>