This paper reports two types of basalt that have different ages (Carboniferous and Permian) but occur next to each other in the northern part of the Changning–Menglian accretionary wedge, southwestern China. We use the geochronology and geochemical data to evaluate the tectonic evolution of the Palaeo‐Tethys during this period. Zircon grains from the mafic‐ultramafic rocks associated with the Permian basalts in the Xiaomengtai area yield a U–Pb age of 281 Ma and ε Hf (t) values from +9.2 to +12.8. The Permian mafic‐ultramafic rocks and the associated basalts are all characterized by normal mid‐ocean ridge basalt (N‐MORB)‐like chondrite‐normalized rare earth elements patterns, moderately negative Nb‐Ta anomalies in the mantle‐normalized immobile incompatible trace element patterns, and positive ε Nd (t) values from +4.2 to +6.5, which are consistent with the geochemical characteristics of mafic‐ultramafic rocks in supra‐subduction zone (SSZ)‐type ophiolite. On the contrary, the Carboniferous basalts, which are associated with marine carbonates, are characterized by light REE enrichments, slightly positive Nb‐Ta anomalies, and ε Nd (t) values from +2.8 to +4.0. These features are similar to those of typical oceanic island basalts (OIBs) worldwide. Modelling results using REEs show that the parental magmas for the Carboniferous OIBs and the Permian basalts were likely derived from mantle peridotites at the depths of garnet and spinel stability, respectively, consistent with the formation depth of these two different types of basalt globally. The occurrence of these two different types of mafic‐ultramafic rocks with significantly different ages in the same area supports the view that they are the remnants of the accreted Palaeo‐Tethys oceanic crust. The Carboniferous OIBs are considered to be parts of an OIB‐carbonate seamount chain that originally formed in the southern part of the Palaeo‐Tethys. The Permian mafic‐ultramafic rocks are regarded as fragments of SSZ‐type ophiolites that were present in the northern part of the Palaeo‐Tethys. These different pieces of oceanic crust were accreted to the Simao–Indochina continental Block by subduction between the Late Permian and the Triassic.

Some mafic–ultramafic intrusions in the North American Midcontinent Rift System host disseminated to massive sulfides of magmatic origin. Massive sulfides are also present in the immediate sedimentary country rocks to some of these intrusions, such as Partridge River, Tamarack, and Eagle. Our working hypothesis is that the country rock-hosted massive sulfides are also of magmatic origin. To test this hypothesis, we have carried out an integrated mineralogical, chalcophile elements, and isotopic (S-Os-Pb) study of the country rock-hosted massive sulfide samples from Partridge River, Tamarack, and Eagle. Data for the intrusion-hosted sulfides from previous studies are used for comparison. Like the intrusion-hosted massive sulfides, the country rock-hosted massive sulfides are mainly composed of pyrrhotite, pentlandite, chalcopyrite, and cubanite and have high Ni, Cu, and PGE tenors, consistent with the crystallization products of magmatic sulfide liquids. These two different types of sulfide occurrences at Partridge River are different in some chalcophile element ratios and S-Os-Pb isotopes, but such differences can be explained by different parental magmas with different degrees of crustal contamination and different R-factors during sulfide segregation. At Tamarack and Eagle, these two different types of sulfide occurrences have similar S-Os-Pb isotope compositions, but the similarity in chalcophile element compositions between them is restricted to only some of the samples. Negative Pt anomalies are more common for the country rock-hosted massive sulfide than the intrusion-hosted sulfide ores. Positive Pt anomalies are not observed in the country rock-hosted massive sulfide samples but are present in some of the intrusion-hosted sulfide ore samples. Our modeling results show that the observed similarities and differences between these two different types of sulfide occurrences in each of the deposits can be explained by a common parental magma, variable R-factors during sulfide-liquid segregation, and variable degrees of fractional crystallization of monosulfide solid solution from sulfide liquids. Given the fact that positive Pt anomalies are present in some of the intrusion-hosted sulfides ores, we suggest that the negative Pt anomalies in the country rock-hosted magmatic sulfides are due to a nugget effect or removal of early-crystallized platinum group minerals, such as sperrylite (PtAs2), from the sulfide liquids prior to their infiltration into the surrounding country rocks.

Previous Cu isotope work has documented a clear disparity between δ65Cu values of sheet-style (–0.5 to 0.5‰) and conduit-style (0.5–2.0‰) intrusions associated with the Midcontinent Rift System. The application of metal isotopes to the study of magmatic Ni-Cu-platinum group element (PGE) deposits is in an early stage, and very little is known regarding isotope distributions and mechanisms of fractionation at high temperatures. In order to resolve the previously mentioned Cu isotope disparity, to determine metal sources for the intrusions, and to assess sources of high-temperature metal isotope fractionation, we have measured Cu and Ni isotope ratios from a suite of exceptionally well characterized Ni-Cu-PGE massive sulfides that occur in sedimentary country rocks near intrusions within the Midcontinent Rift System in Michigan and Minnesota. Previous mineralogic and S, Pb, and Os isotope measurements indicate that the massive sulfides are of magmatic origin and provide a framework for the interpretation of the Cu and Ni isotope data in terms of magmatic processes, including assimilation of Proterozoic country rocks. Copper and Ni isotope ratios were determined for massive sulfides as well as local sedimentary sulfides, and these results were compared with available Cu and Ni isotope results of magmatic and sedimentary sulfides in the Midcontinent Rift and elsewhere. Nickel isotope ratios of the sulfides have been modeled in terms of the effects of variable silicate/sulfide ratios, or R-factors (the mass ratio of silicate magma to sulfide magma), crustal contamination, and olivine fractional crystallization. The near-zero and slightly negative δ60Ni values of country rock-hosted magmatic sulfides (–0.45 to 0.17‰) near Tamarack and Eagle can be explained by minor degrees of crustal contamination of a mantle-derived melt at variable R-factors. Igneous-sourced, Cu-poor sulfides from below the Partridge River intrusion generally have lower δ60Ni values (−0.77 to –0.52‰) that require substantially more contamination from a low δ60Ni source, similar to some of the local sedimentary rocks. Copper isotope ratios of country rock-hosted massive sulfides near Eagle and Tamarack are lower than those reported by previous workers and are mostly about 0‰, similar to those expected for unaltered mantle. Tamarack samples require very little crustal contamination to explain their isotope ratios, whereas Eagle samples require no contamination from local sedimentary rocks to explain their Cu isotope compositions. Copper isotope ratios of samples beneath the Partridge River intrusion are similar to those from previous analyses of igneous-hosted sulfides, supporting their origin from a magmatic sulfide liquid. The low δ65Cu values (–1.14 to 0.25‰) from samples below the Partridge River intrusion cannot be explained by contamination from the Virginia and Thomson Formations, which are characterized by mostly positive δ65Cu values (–0.33 to 3.12‰), unless a lighter reservoir remains to be discovered in the local Proterozoic rocks. We suggest these values may have been produced by a combination of incomplete sulfide melting during partial melt generation and fractionation related to sulfide segregation at variable R-factors.

The Tulaergen magmatic Ni-Cu sulfide ore deposit is located along the southern margin of the Central Asian Orogenic Belt in northwestern China. This deposit is hosted by a small mafic–ultramafic complex composed of a Late Carboniferous (~ 301 Ma) gabbroic member at the margin and a younger ultramafic dyke in the center. Net-textured and semi-massive sulfides are mainly concentrated in the steeply dipping, widened parts of the dyke. Zircons from the ultramafic member yield a U–Pb age of 281 ± 2 Ma, ~ 20 myr younger than the gabbroic member. The average ε Hf (t) value of the zircons is ~ 16. The Tulaergen mafic–ultramafic rocks are characterized by light REE enrichments, pronounced negative Nb–Ta anomalies, ( 87 Sr/ 86 Sr) i ratios from 0.7034 to 0.7036, and ε Nd (t) values from 5.1 to 6.9. The isotope data indicate negligible bulk contamination with the crust. The δ 34 S values of sulfide ores are from − 0.3 to 1.5‰, similar to the values of the country rocks. The γ Os values of the sulfide ores are from + 605 to + 954, much higher than typical mantle values. The Os-S isotope data together support the view that the addition of Os-bearing organic matter from the country rocks may have played a critical role in triggering sulfide saturation. In the sub-vertical segment of the deposit, the upper zone has lower PGE tenors coupled with lower and rather constant olivine Fo contents compared to the lower zone. Based on the geometry of the dyke and sulfide distribution, we conclude that the Tulaergen deposit formed in a dynamic magma conduit.

Magma-carbonate interaction and recycled oceanic crust in the mantle source are thought to have played a critical role in the genesis of magmatic Fe-Ti-V oxide ore deposits in the Emeishan mantle plume-related large igneous province in south-western China. To test these hypotheses, we have carried out a combined study of zircon Hf-O isotopes and whole-rock Sr-Nd isotopes on three representative magmatic Fe-Ti-V oxide ore deposits associated with different types of wall-rock in the province: the Hongge and Taihe deposits with granitoid wall-rocks in most places, and the Panzhihua deposit with marble wall-rocks. Diopside-garnet marble xenoliths are common in the host gabbros of the Taihe deposit. The zircon separates used in this study are all from the mineralized units of the host layered mafic-ultramafic intrusions. The selected zircon grains yield U-Pb isotopic ages of 258.2 2.8 Ma for Hongge intrusion, 263.3 +/- 2.2 Ma for Taihe intrusion, and 257.6 +/- 2.1 Ma for Panzhihua intrusion. The Hf-O isotopes of the zircon separates are epsilon(Hf)(t) from 3.1 to 4.7 and delta O-18 from 5.77 to 6.45 parts per thousand for Hongge, epsilon(Hf)(t) from 8.3 to 9.8 and delta O-18 from 4.12 to 4.56 parts per thousand for Taihe, and epsilon(Hf)(t) from 4.6 to 6.3 and delta O-18 from 5.49 to 6.10% for Panzhihua. The zircon delta O-18 values of the Taihe deposit are significantly lower than the mantle zircon values (5.3 +/- 0.3%), whereas those of the Hongge and Panzhihua deposits are higher than the mantle values. The initial Sr-87/Sr-86 and epsilon(Nd)(t) of the whole rock samples from the deposits are similar, from 0.7047 to 0.7052 and from 2.3 to 3.1, respectively. The observed decoupling of zircon Hf-O isotopes (low delta O-18 and high epsilon(Hf)(t) values) in the Taihe deposit is most consistent with a high-temperature altered, low delta O-18 oceanic gabbroic component in the plume source for this deposit. All of the isotope data together show that the parental magmas for these deposits experienced different types of crustal contamination. Contamination with the Precambrian metamorphic country rocks is much more significant in the parental magma for the Hongge deposit than those for the Panzhihua and Taihe deposits. On the other hand, contamination with marble country rocks is more pronounced in the parental magmas for the Panzhihua and Taihe deposits than in the parental magma for the Hongge deposit. An important new finding from this study is that these deposits are not linked to the same component of the mantle plume nor a single type of crustal contamination. (C) 2020 Elsevier Ltd. All rights reserved.

The link of the widespread Late Permian-Triassic magmatism in the southwestern corner of the South China block to a subduction zone or the Emeishan mantle plume is debated. We have carried out an integrated geochronological-petrological-geochemical study of four mafic dikes in the Gulinjing-Nanxi area of this region to shed new light on the issue. Zircon grains from the Gulinjing-Nanxi mafic dikes yield U-Pb ages from 254 +/- 2 to 250 +/- 2 Ma, indicating that they are coeval with abundant felsic intrusive-extrusive rocks in the region, but are 6-7 myr younger than the nearby Emeishan flood basalts and associated mafic-ultramafic intrusions. The Gulinjing-Nanxi mafic dikes are characterized by light rare- earth element enrichments and pronounced negativeNb-Ta anomalies, similar to typical arc basaltsworldwide, plus enriched Sr-Nd-Hf isotope compositions (initial Sr-87/Sr-86 from 0.7088 to 0.7103, epsilon(Nd)(t) from-6.6 to-4.6, and zircon epsilon(Hf)(t) from-3.7 to-0.3). The isotopes and incompatible trace element ratios indicate that the parental magmas for the GulinjingNanxi mafic dikes are consistent with contaminated arc mafic magmas, but not contaminated Emeishan flood basalts in the region. This, together with a significant age difference between the mafic dikes and the Emeishan flood basalts and associated mafic-ultramafic intrusions in the nearby areas, support the interpretation that the Gulinjing-Nanxi mafic dikes are unlikely to be parts of the Emeishan mantle plume. Based on regional igneous association and the distribution of known paleo subduction zones in the surrounding regions, we conclude that the Gulinjing-Nanxi mafic dikes are linked to Paleo-Pacific subduction to the south, and not Paleo-Tethys subduction to the west. (C) 2021 Elsevier B.V. All rights reserved.

The Jinchuan Ni-Cu deposit is one of the largest magmatic sulfide deposits in the world, but the cause of sulfide saturation remains unclear. It has been suggested that sulfide saturation in the Jinchuan magma was triggered by oxidation through assimilation of fluids related to de-carbonation of marbles. This study uses Mg and C-O isotopes to evaluate the process of carbonate assimilation. The measured values of delta O-18 and delta C-13 for carbonate minerals from country rock marbles range from 13 parts per thousand to 22 parts per thousand and from -1.7 parts per thousand to 1.3 parts per thousand, respectively. Carbonate minerals from xenoliths and hybrid rocks in the contact zone have lower C-O isotope ratios, with delta O-18 from 11 parts per thousand to 16 parts per thousand and delta C-13 from -4.7 parts per thousand to -2.3 parts per thousand, respectively. The carbonate delta Mg-26 values of xenoliths and hybrid rocks are from -1.71 parts per thousand to -1.36 parts per thousand, similar to the values of the marbles (-1.74 parts per thousand to -1.34%.). The whole-rock delta Mg-26 values of hybrid rocks samples associated with xenoliths vary from from -1.69 parts per thousand to -0.92%. The delta O-18 values of pyroxene at the margin and the center of the Jinchuan intrusion vary from 5.1 parts per thousand to 7.8 parts per thousand and from 5.3 parts per thousand to 6.5 parts per thousand, respectively. Despite varying distances from contacts with country rocks, pyroxene in these samples are characterized by delta Mg-26 values from -0.32 parts per thousand to -0.16 parts per thousand, which are similar to the typical value of the Earth's mantle (-0.25 parts per thousand +/- 0.07 parts per thousand). The C-O isotopic variations are consistent with de-carbonation. Rayleigh fractionation models suggest from 40% to 80% CO2 loss from xenoliths and hybrid rocks. The Mg isotopic compositions in the contact zone reflect reactive assimilation between xenoliths and mafic magma. Product olivine and diopside inherited the delta Mg-26 value of reactant dolomite. The transport of released CO2 from xenoliths into magma is consistent with increasing oxygen fugacity and potentially triggering sulfide saturation. However, Mg isotope exchange was confined to the contact zone, and appears to have been unrelated to the ore-forming process.

The East Eagle Ni-Cu-platinum group element deposit is a conduit-type deposit located in northern Michigan, in close spatial proximity to the currently producing Eagle deposit. Massive and semimassive (net-textured) sulfide mineralization at East Eagle occurs approximately 800 m lower in the stratigraphic sequence than that at Eagle and only similar to 200 m above the contact between Proterozoic and Archean rocks. Although sulfide mineralogy and textural types are similar at the two occurrences, there are important differences in their S isotope systematics. Massive sulfide mineralization at East Eagle is characterized by a relatively narrow range of delta S-34 values from 1.5 to 3.2 parts per thousand. Semimassive sulfides show a similar range from 2.1 to 3.8 parts per thousand. In strong contrast to these values, those from disseminated sulfides that border the massive and semimassive mineralization define a much larger range from -4.3 to 22.8 parts per thousand. The much more restricted range in delta S-34 values recorded in the massive and semimassive sulfide mineralization compared to that of the disseminated mineralization is thought to reflect isotopic exchange reactions in the conduit involving accumulated sulfide and pulses of magma containing S of mantle origin. The Delta S-33 values of all three major types of sulfide mineralization at East Eagle are near 0 parts per thousand, with most values between -0.03 and 0.03 parts per thousand. Unlike Delta S-33 values from semimassive sulfide mineralization at Eagle, the Delta S-33 values at East Eagle show no, or very limited, evidence for the involvement of S derived from Archean sedimentary rocks. The wide range in delta S-34 values recorded in the disseminated mineralization provides strong evidence that S from Proterozoic sedimentary host rocks was involved in the mineralization; in some cases, as much as 85% of the S may have been of external origin. In addition to the wide range in delta S-34 values, the disseminated mineralization is characterized by spatially heterogeneous delta S-34 values. Meter-scale S isotope variations, as well as variations in Pt and Pd tenor, are consistent with multiple inputs of magma, each characterized by distinct S isotope ratios. Heterogeneity of several per mill at the centimeter scale indicates that the degree of supercooling exceeded the S diffusivity, preserving small-scale S isotope variability inherited from the sedimentary country-rock source. Elongate, branching plagioclase grains in many of the gabbroic rocks that host the disseminated sulfide mineralization are consistent with a rapid second stage of cooling.

The temporal and spatial variations of Late Paleozoic basaltic lavas in Baoshan, the northern part of a Gondwana-derived micro-continental block called Sibumasu, are important for Gondwana reconstruction. Magmatic zircon crystals from three selected dolerite dykes in the Baoshan region yield U-Pb ages from similar to 295 to similar to 310 Ma. These new ages, together with previous zircon U-Pb ages for this type of rock, define a protracted (similar to 30 myr) episode of basaltic magmatism from similar to 310 to 280 Ma in a small area of this region, which is inconsistent with the typical temporal-spatial distribution of mantle plume magmatism. The trace element compositions of the Baoshan dolerite dykes and associated lavas are similar to arc basalts as well as continental flood basalts worldwide, showing light REE enrichments and negative Nb-Ta anomalies. Mixing calculations using the Sr-Nd-Hf isotope data of the Baoshan mafic rocks indicate that their intriguing trace element characteristics can be explained by contamination of mantle-derived magmas with crustal materials. Our new data, together with the lack of Late Paleozoic are related calcalkaline rocks and granitoids in the Sibumasu block and other contemporaneous Gondwana-derived micro-continental blocks, strongly support the premise that the 310-280 Ma basalts and dolerites in the Baoshan region are the products of continental rift-related magmatism rather than arc magmatism. Based on the temporal correlation of the 310-280 Ma rift-related magmatism in several related Gondwana-derived micro-continents (Sibumasu, South Qiangtang, Lhasa and Himalaya), plus other independent constraints such as paleoclimate biotas and paleolatitudes from the literature, we provide an improved model for the configuration of the Gondwana supercontinent in the Early Permian. Based on the results from this study, we conclude that further investment in the exploration of magmatic Ni-Cu sulfide deposits associated with the 310-280 Ma mafic-ultramafic intrusions in northern Baoshan and the other related Gondwana-derived micro-continental blocks is warranted. (C) 2019 International Association for Gondwana Research. Published by Elsevier B.V. All rights reserved.

The Neoarchean Taoke mafic-ultramafic complex is located in the eastern block of the North China Craton. We report the first geochronological and geochemical data for the complex, and use these data to unravel the fundamental tectonic-magmatic processes involved in the formation of this complex. The Taoke complex comprises a large elongated gabbronorite intrusion, a small sulfide-mineralized olivine websterite body in the center, and many dolerite dikes. The U-Pb isotope ages of magmatic zircon from the mafic and ultramafic intrusions are 2520.3 +/- 6.5 Ma and 2537 +/- 12 Ma, respectively, which are contemporaneous with abundant granitoids and boninitic basalts but similar to 620 myr older than the collision-related, granulite-facies metamorphism in the region. Like the coeval boninitic basalts, the Taoke mafic-ultramafic rocks are characterized by light-REE enrichments and negative Nb-Ta anomalies, which are characteristic of modern arc basalts. The epsilon(Nd)(t) values of the Taoke mafic-ultramafic rocks are from 1.0 to 3.3, within the range of the coeval basalts in the region. The epsilon(Hf)(t) values of zircon from the Taoke mafic-ultramafic complex are from 0.2 to 2.3. The integrated Nd-Hf isotope data indicate that crustal contamination is absent to negligible. No correlation between Th/Nb ratios and epsilon(Nd)(t) values are present, implying that the negative Nb-Ta anomalies of these rocks are not the result of crustal contamination but are original features of the magma. Our new data support the interpretation that the complex is a Neoarchean arc-type mafic-ultramafic complex. Liquid inversion using the La/Yb and Gd/Yb ratios of the associated dolerite dikes and coeval boninitic basalts indicates a garnet-bearing lherzolite mantle source. The depth of partial melting is estimated to be > 85 km, which is within the range of mafic magma generation in modern arc settings. The results from this study demonstrate that by Neoarchean the style of arc basaltic magmatism already closely resembled the modern counterpart. Based on the presence of sulfide mineralization in the Neoarchean arc-type mafic-ultramafic complex at Taoke and the association of major magmatic Ni-Cu sulfide ore deposits with post-Archean arc-type mafic-ultramafic complexes elsewhere in the world, we suggest that exploration for this type of deposit in the Neoarchean arc-type mafic-ultramafic complexes including the Taoke complex in western Shandong, should continue.

The North Qiangtang continental block in central Tibet is a critical piece of the Pangea puzzle. This paper uses integrated geochronological and geochemical data for selected mafic dykes and dioritic enclaves in this block to evaluate its tectonic evolution in the Triassic. Zircons from two mafic dykes and the dioritic enclaves of a large arc granodiorite pluton in eastern North Qiangtang yield indistinguishable U-Pb ages from 248 +/- 2 to 251 +/- 3 Ma, contemporaneous with widespread arc basaltic andesites and crust-derived rhyolites in the region. The mafic dykes and coeval arc basaltic andesites have almost identical Sr-Nd isotopes (initial Sr-87/Sr-86 = 0.707 to 0.708, epsilon(Nd) = -4.4 to -3.6), and are all characterized by light REE enrichments and pronounced negative Nb-Ta anomalies. The dioritic enclaves and the hosts have indistinguishable zircon U-Pb ages, almost identical Sr-Nd isotopes (initial Sr-87/Sr-86 = 0.709 to 0.711, epsilon(Nd) = -7.4 to -5.9), and similar zircon epsilon(Hf) (-13.7 to -5.7), but contrasting chondrite-normalized REE patterns due to hornblende fractionation. The Sr-Nd isotope data indicate that the dioritic enclaves formed from the hybrid melts produced by mixing at depth between the arc basaltic andesites and the crust-derived rhyolites. We propose that the Early Triassic arc igneous suites are related to the northward subduction of the southern Paleo-Tethys beneath the North Qiangtang block from Early to Middle Triassic. The occurrence of several Late Triassic porphyry Cu deposits plus a VMS Ag-Pb-Zn deposit in the Yidun arc, which is the product of the southward subduction of the northern Paleo-Tethys beneath the North Qiangtang block in the Late Triassic, indicates that the arc magmas generated during the subduction of the Paleo-Tethys are fertile in ore metals. Therefore, exploration for Early-Middle Triassic porphyry Cu and VMS deposits in the southern part of the North Qiangtang block is warranted. (C) 2020 International Association for Gondwana Research. Published by Elsevier B.V. All rights reserved.

Abstract Economically valuable magmatic Ni-Cu sulfide deposits in China include the large (515 million tonnes [Mt] of sulfide ore) Jinchuan deposit and 12 smaller ones (3–100 Mt), including Xiarihamu (100 Mt), Huangshanxi (80 Mt), and Poyi (40 Mt). These deposits occur in two principal tectonic environments: within continental plates and at convergent plate margins. The former group can be further divided into those that are hosted in the feeders of a flood basalt province, such as Limahe, and those that occur in rifted continental margins, such as Jinchuan. The latter group can be further divided into those that formed during active subduction, such as Xiarihamu, and those that formed shortly after subduction (20–40 m.y. later), such as Poyi. Despite different geodynamic settings, the Chinese magmatic Ni-Cu sulfide deposits are all characterized by low tenors of platinum group elements (PGEs), indicating PGE depletions in the parental magmas. The small Jinbaoshan deposit (15 Mt, 3 ppm Pt + Pd) is the only significant magmatic PGE deposit in China. It is hosted in a small sill-like ultramafic intrusion associated with the Permian Emeishan flood basalts in southwestern China. The Chinese magmatic Ni-Cu and Pt-Pd sulfide deposits collectively show a positive correlation between Ni/Cu ratios in sulfide ores and forsterite contents in olivine, indicating that fractional crystallization is an important control on Ni/Cu in the parental magmas. The Os isotope data for these deposits support the premise that addition of external sulfur is essential for the generation of ore-quality magmatic sulfide deposits. More detailed geochronological studies are necessary to detect intrusion targets that may be temporally related to other known ore-bearing intrusions in a given area or region. The recent discovery of the subduction-related Xiarihamu Ni-Cu sulfide deposit is consistent with the idea that convergent plate tectonic settings in the world have potential for world-class magmatic Ni-Cu sulfide deposits. We suggest that the Paleozoic Kunlun orogenic belt in the northern part of the Tibet-Qinghai plateau, where the Xiarihamu deposit is situated, is a new frontier for Ni-Cu exploration in China. Furthermore, the search for economic magmatic Ni-Cu and Pt-Pd sulfide deposits associated with the subvolcanic intrusions of the Permian Emeishan flood basalt province in southwestern China should continue.

•Qixin prospect formed at ~286.5 Ma, coeval with regional Ni-Cu sulfide deposits.•Qixin parent magma experienced ≤15 wt% contamination with siliceous country rocks.•Integrated modeling implies that crustal sulfur is essential for inducing sulfide saturation.•Further exploration in Qixin area is warranted due to its great economic potential.

Like many continental flood basalt (CFB) provinces in the world, the source mantle compositions (peridotites or pyroxenites) of the Emeishan CFB province in SW China are still debated. We have used the combination of olivine O isotopes determined in situ using SIMS and trace element abundances measured in situ using LA-ICP-MS to address this fundamental problem. The newly discovered picrites in the Emeishan CFB province at Yumen are used in this study. In addition, we have also analyzed primitive olivine phenocrysts with known O isotope ratios from four other picrite occurrences (Maoniuping, Tanglanghe, Wuguijing and Wulongba) in the Emeishan CFB province for trace element abundances. Based on whole rock compositions the picrite samples from Wulongba belong to the low-Ti group (Ti/Y < 500) and those from the other locations belong to the high-Ti group (Ti/Y > 500). Mineral chemical data reveal that the two different groups of picrites can be distinguished by using Ti/Al of olivine phenocrysts and Cr-spinel inclusions enclosed in olivine. Specifically, the high-Ti picrite group is characterized by 10Ti/Al (molar) >0.9 and >0.5 for olivine phenocrysts and Cr-spinel inclusions, respectively. In the mantle source discrimination diagrams based on olivine 100Mn/Fe versus 100Ni/Mg or 100Ca/Fe, both groups of picrites plot between the suggested fields of peridotite and pyroxenite mantle sources. In the same type of diagrams based on olivine Mn/Zn and 10000Zn/Fe, both groups of picrites plot exclusively in the fields for a peridotite mantle source. The mean delta O-1(8) value of olivine phenocrysts from the Yumen high-Ti picrites is 5.19 +/- 0.25 parts per thousand (1 sigma, n = 35), which is significantly lower than that for the Wulongba low-Ti picrites (5.6 +/- 0.15%0 but is similar to values (5.1 +/- 0.15 to 5.3 +/- 0.15 parts per thousand) for the high-Ti picrites from elsewhere (Maoniuping, Talanghe and Wuguijing) in the Emeishan CFB province. The mean delta O-1(8) values of olivine phenocrysts from the Emeishan high-Ti picrites at different locations are all within the range of typical mantle olivine values (5.1 +/- 03 parts per thousand. Based on these results, we conclude that the high-Ti picrites in the Emeishan CFB province were primarily derived from a modally enriched peridotite mantle source. Our new olivine Zn-Mn-Fe data are consistent with the view that the Wulongba low-Ti picrites were derived from a lithospheric peridotite mantle source. (C) 2019 Elsevier B.V. All rights reserved.

We report the first measurements of Fe isotope fractionation between coexisting pyrrhotite and chalcopyrite from the Lengshuiqing Ni-Cu magmatic sulfide deposit, Southwest China. The delta Fe-56 values of chalcopyrite (Cp) range from 0.24 parts per thousand to 1.25 parts per thousand, much higher than those in the coexisting pyrrhotite (Po, -1.25 parts per thousand to -0.13 parts per thousand). The delta Fe-56 values of pyroxene separates from the Lengshuiqing intrusions are in a range from -0.03 parts per thousand to 0.18 parts per thousand. Pyrrhotite in sulfide-mineralized rocks with greater than similar to 2 wt.% S has delta Fe-56 values of -0.3 +/- 0.2 parts per thousand, indicating Delta Fe-56(po-sil) (pyrrhotite - silicate melt) in the range of -0.4 +/- 0.2 parts per thousand with parental silicate magma delta Fe-56 of 0.1 parts per thousand as calculated from the least evolved pyroxene delta Fe-56 value. This fractionation is similar to that experimentally determined by Schuessler et al. (2007) for pyrrhotite and a rhyolitic melt, suggesting that Delta Fe-56(po-sil) of approximately -0.4 parts per thousand represents a near-equilibrium value for sulfide-rich mineralization in the Lengshuiqing intrusions. Pyrrhotite-dominated sulfide separates in samples with less than similar to 2 wt.% S show delta Fe-56 values as low as -1.25 parts per thousand. The low delta Fe-56 values can be modeled by a non-equilibrium process, where the fractionation between immiscible sulfide liquid and silicate magma decreases from -0.4 to -1.4 parts per thousand as the R-factor (mass ratio of silicate melt to coexisting sulfide liquid) increases from 1 to 1000. We suggest that the relative proportion of externally introduced Fe from sulfidic sedimentary country rocks may be higher in low-S samples, controlling the low delta Fe-56 values of the sulfide assemblage in the mineralized rocks. The trend of Fe isotope enrichment at Lengshuiqing is consistent with Fe-S bond strengths, but the difference between coexisting pyrrhotite and chalcopyrite is not consistent with equilibration at high temperatures. The Delta Fe-56(Cp-po) values at Lengshuiqing (1.06-1.77 parts per thousand) are similar to those measured for samples from the Voisey's Bay and Eagle Ni-Cu deposits, and suggest that in a cooling sulfide liquid Fe isotope equilibration may continue to temperatures less than 200 degrees C. (C) 2019 Elsevier Ltd. All rights reserved.

The Tamarack Intrusive Complex (1105 +/- 1.2 Ma), in northeastern Minnesota, occurs within the Midcontinent rift system and hosts potentially economic Ni-Cu-(PGE) mineralization. The system represents "conduit-style" mineralization and with 1.3 wt % Ni, 0.7 wt % Cu, 0.3 ppm Pt, and 0.25 ppm Pd, is similar in many aspects to the Eagle deposit in Michigan. Sulfur, O, and Os isotopes have been used to evaluate the role of crustal contamination in promoting sulfide liquid saturation. All of the types of mineralization in the Tamarack Intrusive Complex are characterized by delta S-34 values between -0.2 and 2.8 parts per thousand, values that are not strongly anomalous relative to uncontaminated mantle values near 0 parts per thousand. The values are very similar to those from the Eagle deposit, but contrast sharply with values of disseminated sulfides in intrusions of the Duluth Complex and Crystal Lake Gabbro, which may be as elevated as 17 parts per thousand. Initial Os-187/Os-188 ratios in the Tamarack intrusive Complex are between 4 and 44% higher than the same ratio of the undepleted primitive mantle at 1105 Ma and correspond to gamma Os values for all magmatic sulfide types from the Tamarack Intrusive Complex ranging from 10 to 92. These values are consistent with crustal contamination but for S, the isotopic ratios are remarkably lower than those from mineralization in the Duluth Complex, where initial Os-187/Os-188 ratios are more than 110% higher than that of primitive mantle and gamma os values may be in excess of 1,000. Olivine from an unmineralized but sparsely serpentinized portion of the Tamarack Intrusive Complex has O isotope compositions from 5.2 to 5.5 parts per thousand, indicating a fraction of a percent crustal contamination of the parental magma. The striking contrast between Os-S isotope systematics in conduit-type deposits associated with the intrusions that formed during the Main stage in the history of the rift is a primary characteristic of these ore systems and as such is important to explain by a geologic model. High Re and Os contents are found in sulfide minerals and organic matter in the sedimentary country rocks. Selective contamination via partial melting and devolatilization involving these phases could have been responsible for S and Re-Os transfer from country rocks to the sulfides without affecting the silicate magma. Therefore, the contamination of the system may have been an order of magnitude greater than that recorded by the silicate portion of the system. Alternatively, the rather low degrees of crustal contamination indicated by the Re-Os and S systems in the Eagle intrusion and the Tamarack Intrusive Complex may also be related to isotopic exchange between contaminated and pristine magmas in the conduit system during the Early stage of the rift development. The high Ni grades of the complex are in part related to the dynamic conduit environment and contamination of picritic magmas promoting sulfide saturation before large quantities of Ni were sequestered by olivine.

Ore deposits that are associated with mafic to ultramafic igneous rocks include Ni-Cu-PGE deposits, chromite deposits, Fe-Ti-V deposits (magnetite/titanomagnetite), and Fe-Ti±P (ilmenite, titanomagnetite±apatite) deposits. Ni-Cu-PGE deposits may be sulfide-rich, with PGEs held as sulfides, alloys, arsenides, telluirdes, bismuthinides, or selenides. Deposits are normally relatively small and occur as part of conduit systems, including extrusive rocks in the case of komatiites. The world's largest PGE deposits occur in large layered intrusions (e.g., Bushveld Complex, Great Dyke Stillwater Complex) where sulfide volume percentage is low, generally less than five. Chromite deposits occur as stratiform layers in large intrusions, and as irregular podiform deposits in ophiolite sequences. V-bearing magnetite may occur as massive layers in large layered intrusions like the Bushveld Complex, or may be disseminated in oxide-rich gabbros in layered intrusions of various sizes. Titanomagnetite and ilmenite occur in gabbroic rocks in the Emeishan Large Igneous Province as massive layers or as disseminations. Fe-Ti±P deposits often occur in gabbroic rocks that are associated with Proterozoic anorthosite suites and in layered intrusions. Ti-rich deposits also occur with ultramafic rocks in the Duluth Complex and within Ural-Alaskan intrusions. Genetic models for the origin of the various deposits must account for the concentration of metals via either chemical or physical processes. Immiscible sulfide liquids strongly sequester Ni, Cu, and PGEs due to the very chalcophile nature of these elements. Due to their density, gravitational settling may be important as a concentration mechanism, as well as accumulation in conduits where restrictions may cause decreases in magma velocity. For chromite and Fe-Ti-V-P deposits, processes such as fractional crystallization accompanied by gravitational crystal accumulation and removal of buoyant residual material are important concentration mechanisms. In some cases the separation of immiscible Si-rich and Fe-rich melts may be an important process, with the Fe-rich melts giving rise to Fe-Ti-P ores. The tectonic settings of the deposits are varied, ranging from intraplate settings where hydrous mantle plumes may have been important, to divergent zones of intracontinental rifting, to convergent zones where asthenospheric mantle-derived melts produced layered bodies and conduit-related intrusions where interaction with sulfide-bearing country rocks has occurred.

The Daxueshan deposit is the first magmatic Ni-Cu sulfide deposit that has been discovered in the eastern part oldie Tethyan orogenic belt, which stretches from southwest China to Turkey. Although the size of the deposit is small, containing similar to 0.52 million tonnes of sulfide ore with grades of 0.67 wt % Ni and 0.46 wt % Cu, it provides a unique opportunity to learn more about nickel metallogeny in arcs. The host intrusion of this deposit is composed of gabbro, harzburgite, and lherzolite. Sulfide mineralization occurs as disseminated and massive sulfides (pyrrhotite, pentlandite, and chalcopyrite) in the basal zone of the ultramafic rocks. Sensitive high-resolution ion microprobe U-Pb dating of zircon crystals from the gabbro yields a crystallization age of 300.5 +/- 1.6 Ma. The eiff(t) values of these zircon crystals are from -2 to -11. The forsterite contents of olivine from sulfide-bearing (>0.3 wt % S) and sulfide-barren (<0.3 wt % S) ultramafic rocks are from 80 to 83 mol % and from 76 to 80 mol %, respectively. Coexisting pyroxenes are bronzite and augite. The Al/Ti ratios of augite from the Daxueshan intrusion and global arc cumulates are similar. Coeval arc basalts in the area are characterized by light rare earth element (REE) enrichments relative to heavy REEs, pronounced negative Nb-Ta anomalies, elevated initial Sr-87/Sr-86 ratios from 0.7065 to 0.7071, and slightly negative epsilon(Nd)(t) values from -0.8 to -0.3. The Daxueshan mafic-ultramafic rocks have higher initial Sr-87/Sr-86 ratios from 0.7116 to 0.7139, lower epsilon(Nd)(t) values from -5.7 and -7.1, and higher degrees of light REE enrichments. These differences can be explained by higher degrees of crustal contamination (up to 20% more) for the mafic-ultramafic intrusive rocks than the coeval basalts. The (delta S-34 and gamma os values of sulfide separates from the deposit are from -2.6 to 1.2% and from 28 to 482, respectively. The former are similar to the typical mantle value (0 +/- 2%), whereas the latter are significantly different from the primitive mantle value, indicating contamination with organic matter-bearing (and hence Os-rich) sedimentary rocks. Olivine chemistry and Sr-Nd-Hf-Os isotope data indicate that fractional crystallization and crustal contamination played a role in triggering sulfide saturation in the Daxueshan magma, although their relative significance is unclear. Like most arc-type magmatic sulfide deposits worldwide, the platinum group element (PGE) tenors of the Daxueshan deposit are extremely low, indicating a severe PGE depletion of the parental magma due to previous sulfide segregation at depth, including the lower part of the arc crust, to form sulfidebearing, Cu-PCE-rich cumulates. This finding supports the notion that the formation of sulfide-bearing cumulates in the lower part of the arc crust may be a critical step in continent building or the genesis of porphyry ore deposits because new magma or volatiles may cannibalize sulfides from the previous cumulates in the pathway.

Integrated zircon–olivine O–Hf isotope data have been successfully used to unravel the nature of the source mantle for the early Permian post-collisional mafic–ultramafic intrusive rocks in the southern margin of the Central Asian Orogenic Belt in NW China. Olivine crystals with forsterite (Fo) contents varying from 91 to 87 mol% from the Permian Pobei mafic–ultramafic complex in the region yield highly elevated δ18O from 6.0 to 7.2‰. These values are much higher than typical mantle values (~ 5.3‰) and are apparently at odds with the mantle-like εNd(t) values of whole rocks (4.9–5.4). Magmatic zircon crystals from troctolite and gabbroic rocks show divergent oxygen and hafnium isotopic compositions: mantle-like εHf(t) values from 5.1 to 11.9 and crust-like δ18O values from 7.6 to 10.1‰. The observed increase of δ18O values from olivine (an early crystallizing phase) to zircon (a late crystallizing phase) in the mafic–ultramafic rocks is generally consistent with an AFC process. However, this process cannot fully explain the highly elevated δ18O values (6–7‰) for the most primitive olivine containing Fo as high as mantle olivine (> 90 mol%) and the mantle-like Hf isotope composition of zircon. Mixing calculation indicates that such highly unusual isotope compositions can be explained by the previous source mantle contamination with subducted sediment-derived melts and slab-derived fluids. Our results show that the combination of zircon O–Hf isotopes and olivine oxygen isotopes is more effective than the data of zircon or olivine alone to distinguish the effect of AFC process from source contamination. The results from this study provide a new line of evidence that the sub-arc mantle is not homogeneous in oxygen isotopes.