Search

Carbonatite-related rare earth element (REE) deposits, the most significant source of REEs globally, are normally generated in extensional settings, such as intracontinental rifts, mantle plume-related environments, or postcollisional orogens. Syncollisional orogens represent overall compressional regimes, so carbonatites and related REE deposits are rarely identified in such a setting. However, this study reports an anomalous syncollisional carbonatite-related REE deposit, Dong Pao, in the India-Asia collision zone in northwestern Vietnam. The Dong Pao deposit is dated at ca. 52 to 51 Ma through zircon and bastnäsite Th-U-Pb chronometers. The ore-hosting carbonatites were emplaced as stocks with associated syenite. The carbonatite-syenite complex is significantly enriched in light REEs, Ba, and Sr and depleted in high-field strength elements, and has high (87Sr/86Sr)i ratios (>0.707) and low εNd(t) values (–6.5 to –5.6). These geochemical signatures imply that the carbonatite-syenite complex was derived from partial melting of subcontinental lithospheric mantle previously metasomatized and fertilized by REE- and CO2-bearing fluids. Timing of the REE-rich carbonatite-syenite complex indicates that it was related to a far-field stress within the early Eocene main-collision stage at 52 to 51 Ma rather than the late-collision stage at 42 to 35 Ma as previously thought. Collisional tectonism involving block rotation and fault activation are interpreted to have induced disturbance of the lithosphere mantle and created localized, transtensional/extensional environments oblique to the trend of the orogen that facilitated emplacement of the REE-rich carbonatitic magmas. Dong Pao appears to be the first identified, high-tonnage REE deposit that formed in the syncollisional geodynamic setting. Such a finding highlights that tectonic disturbance of an REE-rich lithosphere mantle distal to collision sutures has the potential to generate REE deposits, even during prominent convergence and collision of continents. As such, it defines additional search spaces for exploration of other REE orebodies of this style in complex collisional orogens.

This study presents a complete trace element data set of chromite from the world’s largest podiform chromite deposit at Kempirsai, Kazakhstan, together with platinum group element (PGE) and Re-Os isotope compositions, to explore the potential of trace elements in the study of podiform chromite deposits. Chromite orebodies in the Karaagash area of the Kempirsai Main ore field are hosted by dunites in the mantle sequence of the ophiolite. Dunites slightly distal to the orebodies are composed of olivine (forsterite content = 91.5–93.4) and minor euhedral chromite with variable Cr# (0.73–0.83). These dunites have primitive mantle-normalized Pt-Pd–poor PGE patterns and subchondritic 187Os/188Os ratios (0.12174–0.12220). Negative correlations between Cr# and incompatible trace elements (Sc and Ti) of the chromite are consistent with melt-rock reaction, whereas positive correlations between Cr# and compatible trace elements (V and Ga) reflect the role of fractional crystallization. It is therefore proposed that chromite in the distal dunites was crystallized from magmas after reaction with mantle peridotites. In comparison, chromite grains in the ores and adjacent dunites exhibit relatively uniform Cr# numbers (0.81–0.86) and higher 187Os/188Os values (0.12319–0.12414). Both compatible and incompatible trace elements of chromite exhibit negative correlations with Cr#, which are best explained by mixing between the modified magma after melt-rock reaction and a later injected magma from a more depleted source. After the ore formation, subsolidus reequilibration between olivine and chromite reduced Mg# and modified divalent trace elements of chromite in the order of Zn > Co = Mn > Ni but did not affect distribution of nondivalent trace elements, except V and Fe3+. The decoupling of redox-sensitive and redox-insensitive trace elements suggests that the redox states have been modified during subsolidus reequilibration. The existing iron equilibrium between olivine and chromite records high fO2 values (quartz-fayalite-magnetite buffer [∆QFM] 0.8–2.4), whereas correction of subsolidus reequilibration results in a decrease of the fO2 values toward the QFM buffer (∆QFM 0.1–0.6), indicating that increase in fO2 was not a critical factor in the ore formation. Principal component analysis and factor analysis were further carried out to identify differences in trace elements between high-Cr and high-Al podiform chromite deposits worldwide. High-Cr chromite deposits have variable divalent trace element compositions and show complex distributions of nondivalent trace elements. The former reflects longer and slower cooling histories of the high-Cr chromite deposits with respect to the high-Al chromite deposits, whereas the latter indicates more variable magma compositions and complex processes during the formation of the high-Cr chromite deposits.

Mantle plume rich in gold is considered to be important for the formation of giant epigenetic gold deposits. The Youjiang basin, SW China, is the world’s second largest Carlin-type gold province, but the ultimate source of its gold remains enigmatic. In this study, we report that the Middle–Late Permian basaltic rocks in the basin are rich in native gold grains. These gold grains are scattered in the interstices of pyrite and marcasite and in the amorphous silica cavities. Mineralogy and S-Pb isotope geochemistry of the auriferous sulfides suggest that the gold was largely derived from the gold-rich Late Permian (ca. 260 Ma) Emeishan plume and was released to a near-surface volcanogenic massive sulfide (VMS) metallogenic system, where it accumulated. The native gold grains from the basalts may have been inherited by the younger (ca. 140 Ma) Carlin-type ores in the Youjiang basin, which are indicative of gold pre-enrichment in the basin. Our study highlights that golden plume upwelling could carry abundant gold into the upper crust, even into shallow-level metallogenic systems, and thus provides an alternative view on the source of gold in the Youjiang basin.

Rare earth elements (REEs) mobilize, fractionate, and are re-distributed during supergene processes, and thus provide a powerful tool with which to quantitatively reconstruct the effects of chemical weathering. Moreover, under certain conditions, the REEs can concentrate to levels in the regolith sufficient to form giant regolith-hosted REE deposits, that are now responsible for much of the World’s production of heavy REEs (HREEs). Understanding the supergene behavior of the REEs is an important first step towards applying the REEs as a geochemical tool and meeting the growing global demand of the REEs. Thermodynamic calculations predict that dissolution of the main REE minerals in REE-rich protoliths, namely synchysite-(Y), gadolinite-(Y), hingganite-(Y), yttrialite-(Y), allanite-(Ce), eudialyte, chevkinite-(Ce), britholite, euxenite and loparite-(Ce), should occur spontaneously during weathering. It therefore follows that a high abundance of these minerals in the protolith implies high mobility of the REEs during weathering and consequently a high potential for the discovery of economic REE resources. Dissolution of apatite is promoted by metamictization or structural distortion and could be also important at low pH. In contrast, some LREE-fluorocarbonate minerals, notably bastnasite-(Ce) and parisite-(Ce), and monazite-(Ce) are likely thermodynamically stable in acidic environments. Thus, they would be preserved in the regolith. Zircon, titanite, aeschynite, fergusonite, and xenotime-(Y) are resistant to acidic dissolution, consistent with their common occurrence as residual minerals. In the cases of the world-class regolith-hosted REE deposits in South China, the groundwater is mildly acidic to circumneutral and carbonate-rich. The REEs are consequently transported dominantly as hydrated cations and carbonate complexes, depending on the pH. The general inheritance of the REE pattern of the regolith groundwater in the clay-sorbed fraction in the regolith indicates that the REEs in regolith are scavenged from the regolith groundwater. Elemental anomalies of specific REEs in the clay-sorbed fraction are very likely caused by an anomalously high REECO3+ fraction in the corresponding regolith groundwater, suggesting a preferential uptake of the REECO3+ complexes by the clay minerals, feasibly by halloysite through intercalation as interlayer complexes. This reaction is expected to be particularly important for the sorption and enrichment of the HREEs in the regolith. Depending on the pH and carbonate concentration of the water, surface complexation on clay minerals or interlayer intercalation particularly in halloysite control the pattern of REE enrichment. Mixing of the regolith groundwater with the alkaline and carbonate-rich aquifer groundwater increases the pH and carbonate concentration and, in turn, affects the ability of the mixed water to transport the HREEs. Interplay of aqueous complexation with the regolith mineralogy significantly affects the REE fractionation and re-distribution during groundwater-regolith interaction.

Skarn ores have recently been identified beneath the historically mined placer Sn deposit at Kanbauk of the Dawei region, southern Myanmar. A large-tonnage skarn ore reserve at Kanbauk is estimated to be over 100 million tonnes, with reported ore grades of 0.17% WO3, 0.26% Sn, and 15.4% CaF2, potentially making it one of the largest W-Sn skarn deposits in the Southeast Asian tin belt. The mineralized skarns lie between marbles to the east and metasediments of the Mergui Group to the west. The timing of the mineralization is unclear, and thus the genetic relationship with regional magmatic events is not known. We report laser ablation-inductively coupled plasma-mass spectrometry U-Pb ages of garnet and cassiterite from the mineralized skarns. Garnet grains from the massive prograde skarns are typically subhedral to euhedral and show both sector and oscillatory zoning. They have 15 to 23% andradite (Ad), 55 to 67% grossularite (Gr), and 16 to 30% pyralspite (Py) (Ad15-23Gr55-67Py16-30) and contain 0.08 to 306 ppm U with a lower intercept 206Pb/238U age of 56.0 ± 1.5 Ma. Cassiterite grains from retrograde veinlets are subhedral to anhedral and have U contents from 110 to 12,000 ppm with a lower intercept 206Pb/238U age of 54.2 ± 1.7 Ma. Garnet and cassiterite have ages consistent within error and can be taken to indicate the formation of the Kanbauk W-Sn(-F) skarn deposit at around 55 Ma. Together with published ages of primary Sn-W deposits in the Dawei region, our study confirms a westwardly younging trend of mineralization toward the coast and provides support for rollback of the Neo-Tethyan subducting slab since the Late Cretaceous, which is considered as the main mechanism for the regional, extensive Sn-W mineralization.

Despite a consensus view that Mars has been cold and icy throughout the Amazonian (3 billion years ago until present), the Zhurong rover has provided striking evidence for aqueous resurfacing processes in and around the landing site during this time period. Here, we report a close association of magmatism (i.e. dikes) and several distinctive landforms (e.g. mud volcanoes and troughs) indicating that ground ice in the Vastitas Borealis Formation (VBF) was thawed by the magma heating causing the collapse of VBF to form distinct landforms and alteration mineralogy. This collapse released pressurized meltwater and mud, producing mud volcanoes and muddy water. Upon freezing, the slurry materials were further buried by younger porous sediments. Freezing of the upwelled chemical-rich groundwater potentially circulated by cryosuction produced large crystals of sulphate (probably selentite) unlike sulphates previously identified on Mars, evidencing relatively young aqueous alteration on Mars driven by magma-ice interactions.

Scandium is important in modern technology and is regarded as a strategic metal in many countries. It is highly dispersed in Earth’s crust and rarely forms independent minerals. Clinopyroxene is the most important Sc-bearing mineral in some world-class deposits hosted in mafic–ultramafic intrusions, which are also the major source of laterite-hosted Sc deposits. However, the factors controlling Sc distribution in minerals have been little explored, impeding the understanding of the geochemical behavior of Sc and why it is common in some clinopyroxene grains. The newly discovered Mouding Sc deposit in SW China is hosted in a zoned intrusion composed, from core to rim, of monzogabbro, syenogabbro, gabbro, magnetite clinopyroxenite, and clinopyroxenite. Clinopyroxene in the intrusion is diopsidic in composition with high Sc contents (80–105 ppm). In-situ trace element mappings of diopside crystals reveal homogeneous, zoned, swallow-tailed, and hourglass internal Sc distribution patterns. These patterns can be produced through kinetically controlled incorporation of Sc on different crystal faces. The preferential substitution of Sc can take place on the {1 0 0}, {1 1 0} and {0 1 0} prism faces because of the high flexibility of the octahedral M1 protosites. The fast growth of diopside, which facilitates kinetically controlled crystallization, is dominated by textural coarsening and promoted by the hydrous parental magmas with low viscosities and active convection. The active flow and efficient interstitial communication of the magma can direct compatible elements from the magma into clinopyroxene, thus favoring formation of Sc-rich grains. Our study provides a feasible way to study intra-grain variations of Sc in minerals and emphasizes that kinetic effects may play a critical role in Sc distribution and enrichment in hydrous magmatic Sc deposits. We also show that disequilibrium crystallization may be more pervasive than previously thought, and the hourglass zoning of clinopyroxene can provide valuable information on this process.

Regolith-hosted rare earth element (REE) deposits predominate global resources of heavy REEs. Regoliths are underlain by various types of igneous rocks and do not always host economically valuable deposits. Thus a feasible and convenient method is desired to identify REE mineralization in a particular regolith. This study presents a detailed visible short-wave infrared reflectance (VSWIR) spectroscopic study of the Renju regolith-hosted REE deposit, South China, to provide diagnostic parameters for targeting REE orebodies in regoliths. The results show that the spectral parameters, M794_2nd and M800_2nd, derived from the VSWIR absorption of Nd3+ at approximately 800 nm, can be effectively used to estimate the total REE concentrations in regolith profiles. M1396_2nd/M1910_2nd ratios can serve as proxies to evaluate weathering intensities in a regolith. Abrupt changes of specific spectral features related to mineral abundances, chemical compositions, and weathering intensities can be correlated with variations of protolith that formed a regolith. These VSWIR proxies are robust and can be used for exploration of regolith-hosted REE deposits.

Microbial and hydrothermal venting activities on the seafloor are important for the formation of sediment-hosted stratiform sulfide (SHSS) deposits. Fe isotopic compositions are sensitive to both microbial and hydrothermal activities and may be used to investigate the formation of these deposits. However, to the best of our knowledge, no Fe isotopic studies have been conducted on SHSS deposits. In the Devonian Dajiangping SHSS-type pyrite deposit (389 Ma), South China, laminated pyrite ores were precipitated from exhalative hydrothermal fluids, whereas black shales were deposited during intervals with no exhalation. Pyrite grains from black shales mostly display positive δ56Fe-py (0.01–0.73‰), higher than marine sediments (ca. 0‰), due to pyrite deriving Fe from basinal shuttled Fe(III) (hydr-)oxides and slowly crystallizing in pores of sediments with equilibrium fractionation, except for negative δ56Fe-py (−0.17‰ to −0.24‰) of two samples caused by mixing of Fe from underlain laminated ores. The positive δ34S-py (3.50–24.5‰) of black shales reflect that sulfur of pyrite originated from quantitative reduction of sulfate in closed pores of sediments. In contrast, pyrite grains of laminated ores have negative δ56Fe-py (−0.60‰ to −0.21‰), which were not only inherited from the negative δ56Fe of hydrothermal fluids but also caused by kinetic fractionation during rapid precipitation of a pyrite precursor (FeS) in hydrothermal plumes. These ores have negative δ34S-py (−28.7‰ to −1.82‰), because H2S for pyrite mineralization was produced by bacterial sulfate reduction (BSR) in a sulfate-rich seawater column or shallow sediments. The δ56Fe-py values of laminated ores co-vary positively with δ34S-py and δ13C-carbonate along the ore stratigraphy, with δ13C-carbonate values ranging from −12.0‰ to −2.50‰. However, they correlate negatively with aluminum-normalized total organic carbon (TOC/Al2O3). Organic carbon is thus considered to enhance the production of H2S by BSR activities, increase pyrite precipitation rates and promote the expression of kinetic fractionation of Fe isotopes. Intriguingly, in the ore units with vigorous hydrothermal venting activities, δ56Fe-py, δ34S-py and δ13C-carbonate values display a consistently increasing trend. Such results suggest that venting hydrothermal fluids significantly inhibited the H2S production of BSR, which then reduced the pyrite crystallization rate and decreased the kinetic fractionation of Fe isotopes. Our study reveals that the formation of SHSS deposits relies on H2S from microbial activities and metals from hydrothermal exhalation on the seafloor, but that vigorous exhalation can inhibit microbial activities and thus sulfide precipitation rates. The integrated use of Fe, S, and C isotopes can effectively elucidate these dynamic interactions between hydrothermal venting and microbial activities during the formation of SHSS deposits.

Precambrian iron oxide copper-gold (IOCG) deposits are generally encountered with multistage hydrothermal overprints and hence have complex isotopic records. Precise dating of ore-forming and overprinting events and assessment of time-resolved metal sources are fundamental for understanding ore genesis. Here, we quantify the evolution history by integrating in situ U-Pb dating of texturally constrained allanite and Sm-Nd isotope data of ores and major rare earth element (REE) minerals in the breccia-hosted Lanniping Fe-Cu deposit in Kangdian region, southwestern China. The economically mineralized breccia in Lanniping Fe-Cu deposit is characterized by pervasive and texturally destructive replacement of polymictic clasts, including host metasedimentary packages, the intruded dolerite, and pre-ore halokinetic breccia. Ore minerals in cements are mainly composed of magnetite, chalcopyrite, bornite, and variable amounts of REE-rich minerals (e.g., apatite and allanite/epidote). Two types of allanite were identified in ores. Type I prismatic allanite texturally intergrown with magnetite has a SHRIMP U-Pb age of 1728 ± 20 Ma (1σ), which matches a zircon U-Pb age of 1713 ± 14 Ma (2σ) for the dolerite clasts and provides the direct age constraint on the Fe-Cu mineralization event. Type II anhedral allanite shows complex zoning and is spatially associated with, but texturally later than, magnetite, apatite, and chalcopyrite. This type of allanite yields significantly younger SHRIMP dates of 1015 ± 33 (1σ) and 800 ± 16 Ma (1σ) for cores and rims, respectively, which correspond to discrete regional magmatic events and hence record hydrothermal overprint/remobilization events of ore minerals in the deposit. Integrated Sm-Nd isotope compositions of type I allanite, apatite, and whole ores generally align along the reference Sm-Nd isochron of 1728 Ma, further confirming the primary ore formation at ~1.7 Ga. Corresponding εNd(1728 Ma) values ranging from –2.8 to 0.3 are significantly higher than those of the host metasedimentary rocks (–9.5 to –6.2) but comparable to those of contemporaneous igneous intrusions (–0.3 to 5.3) in the region, demonstrating that REE components of the primary ores were dominantly sourced from rocks of mantle-derived affinity. Both cores and rims of the younger type II allanite grains have Nd isotope compositions consistent with the unique time-evolved line of the ~1.7 Ga ores, implying that REEs incorporated into type II allanite were ultimately sourced from the primary ores in this deposit. The combined texture, chemical, U-Pb, and Sm-Nd isotope data thus demonstrate that REE remobilization was localized during post-ore hydrothermal overprint with negligible external inputs of REEs to the primary ores in the Lanniping deposit. In this contribution, we not only date primary ore formation but also recognize several younger allanite generations that record internal metal redistributions in response to post-ore tectonothermal events. Our study highlights the potential of ore-associated REE minerals such as allanite for resolving the age of multiple stages of hydrothermal events in complex ore deposits by ion probe, provided that careful examination of textural and paragenetic relationship of ores is conducted. Our finding of these younger allanite generations also exemplifies the significance of evaluation on time-resolved metal input for better characterizing the evolution history of the IOCG deposits.

Significant perturbations in ocean redox states have been thought to be one of the major causes for the biological evolution from the Ediacara biota to the Cambrian fauna. However, the driving force of these changes in the ecosystem across the Ediacaran–Cambrian (E–C) boundary is still a matter of debate. In this study, we reported stratigraphic variations of Hg content and Hg isotopes in E–C sediments from the Mussoorie section, Lesser Himalaya, India. Dolomites from the late Ediacaran Krol D and E members have increasing Hg/Al ratios and decreasing Δ199Hg values from the base upwards, which were most likely originated from a prolonged and enormous terrestrial-Hg input into the E–C ocean. The overlying cherty phosphorites in the lowest Tal Group are highly rich in Re relative to Mo, indicative of a suboxic-anoxic depositional environment on the earliest Cambrian shelf that was coincident with the disappearance of the Ediacara biota. Such an O2-deficient condition was possibly led by a consumption of atmospheric O2 during the oxidative weathering of sedimentary organic matter and shallow-water eutrophication, with evidence from the coupled δ202Hg and δ13C values in late-Ediacaran dolomites and high Cd/TOC and Zn/TOC ratios in early Cambrian phosphorites. The overlying black shales have decreasing contents of Hg and micronutrients (Cd and Zn) and increasing values of Δ199Hg, representing notable declines of terrestrial flux and seawater nutrient at the Cambrian Stage 2 before the major phase of Cambrian diversification. Therefore, we conclude that enhanced continental erosion and pervasive upwelling had led to major changes in redox state and nutrient content in shallow seawaters on shelves, which in turn had promoted the biological evolution across the E–C boundary.

Exogenic deposits are an important source of rare earth elements (REEs), especially heavy REEs (HREEs). It is generally accepted that microorganisms are able to dissolve minerals and mobilize elements in supergene environments. However, little is known about the roles of microorganisms in the formation of exogenic deposits such as regolith-hosted REE deposits that are of HREE enrichment and provide over 90% of global HREE demand. In this study, we characterized the microbial community composition and diversity along a complete weathering profile drilled from a regolith-hosted REE deposit in Southeastern China and report the striking contributions of microorganisms to the enrichment of REEs and fractionation between HREEs and light REEs (LREEs). Our results provide evidence that the variations in REE contents are correlated with microbial community along the profile. Both fungi and bacteria contributed to the accumulation of REEs, whereas bacteria played a key role in the fractionation between HREEs and LREEs. Taking advantage of bacteria strains isolated from the profile, Gram-positive bacteria affiliated with

The magmatic processes beneath the active continental margins are very complicated and affect structures and compositions of the arc roots. Neoproterozoic igneous rocks are widely distributed around the margins of the Tarim Block in NW China. The Xingdier mafic-ultramafic intrusion is a composite body, located at the northern margin of the Tarim Block, and consists of gabbro, pyroxenite, and peridotite units. The gabbro unit has a secondary ion mass spectrometry zircon U-Pb age of 727 ± 5 Ma. Rocks from the Xingdier intrusion have a large range of MgO (12.9–32.8 wt%) and SiO2 (43.0–57.9 wt%), and low K2O + Na2O (0.11–2.25 wt%) contents. They have right inclined chondrite-normalized rare earth element patterns with (La/Yb)N ratios of 2.2–8.6. Their primitive mantle normalized trace element patterns show arc-affinity geochemical features characterized by enrichment in Rb, Ba, Th, U, and Pb and depletion in Nb, Ta, and Ti. They have variable initial 87Sr/86Sr ratios (0.7063–0.7093), εNd(t) values (−2.9 to −7.8), 206Pb/204Pb (17.08–17.80), 207Pb/204Pb (15.42–15.49), and 208Pb/204Pb ratios (37.48–38.05), forming an evolution trend from the peridotite unit to the gabbro and pyroxenite units. Clinopyroxene in the three units is chemically similar to those formed in hydrous magmas. The spinel inclusions in olivine from the peridotite unit show unmixing texture and have high Al contents and oxygen fugacity of ~FMQ+1. Therefore, the parental magma was probably derived from a lithospheric mantle enriched by slab-derived fluids. Rocks from the gabbro and peridotite units are proposed to have been derived from olivine-normative melts, whereas rocks from the pyroxenite unit are cumulates from the quartz-normative melts. Such contrasting parental magmas resulted from variable degrees of crustal contamination and fractional crystallization in the arc root.

Isotopic dating studies have revealed that many carbonatite-related REE ore systems have protracted and complex temporal evolution histories. The mechanisms that cause these complex dating results remain poorly understood. Here, we present an in-situ Sm-Nd and Th-Pb isotopic investigation of monazite and bastnasite from a geologically-complicated deposit, the world-class Bayan Obo REE deposit of China, with the aim of defining the general mechanisms accounting for the protracted age spread observed in this and many other REE deposits. Monazite and bastnasite are two principle REE ore minerals in the Bayan Obo deposit. Most of the studied monazite and bastnasite grains have experienced fluid-aided modification. The modified grains have generally homogeneous Sm and Nd contents and Sm-Nd isotopic compositions, implying low mobility of Sm and Nd during the modification process. Thus, the Sm-Nd chronometer can be used to constrain the primary crystallization timing of the REE minerals at 1293 ± 48 Ma. However, altered REE minerals have apparent 208Pb-232Th ages (1248–261 Ma) that are variably younger than the primary crystallization age. In most samples, Th contents have no correlations with 208Pb-232Th ages, while Pb contents tend to decrease along with the younging trend of 208Pb-232Th ages. NanoSIMS mapping shows that 208Pb and 208Pb/232Th are heterogeneous, and irregularly distributed in the altered grains. Such results are indicative of variable Pb loss during the fluid-aided modification process. Thus, many Th-Pb ages represent partial isotopic resetting and, hence, are geologically meaningless. Completely reset ages cluster at 450–400 Ma or 280–260 Ma, representing two discrete episodes of post-ore hydrothermal activity. Small amounts of monazite and bastnasite that were newly crystallized during these two overprinting episodes have initial Nd isotopic signatures approaching the time-evolved isotopic compositions of primary ores, which suggests insignificant REE input to the primary ores during these events. This study provides critical evidence to resolve complex dating results and disparate ore genesis models for the giant Bayan Obo deposit. Moreover, our results indicate the resistance of Sm-Nd isotopes, but the susceptibility of Th-(U)-Pb isotopes, to fluid-aided modification. Therefore, younger Th-U-Pb dates in geologically-complicated deposits should be treated with caution as they may represent partial resetting results and/or may not represent events of increased metal tenor.

Gold (Au) is largely hosted by pyrite in a variety of hydrothermal systems, but the incorporation of Au into pyrite under disequilibrium conditions remains poorly understood. We integrate synchrotron X-ray fluorescence microscopy, electron backscatter diffraction, nanoscale secondary ion mass spectrometry, and laser ablation–multicollector–inductively coupled plasma–mass spectrometry to constrain the processes that sequester Au into zoned pyrite in the hydrothermal cement of breccia ores from the world-class Daqiao orogenic Au deposit, central China. Euhedral pyrite cores with oscillatory and sector zoning, variable δ34S values, and lower Au-As contents than the mantles are attributed to crystallization during oxidation of metal-depleted ore fluids with local variation in fluid conditions. The isotopically uniform colloform mantles are formed by pyrite crystallites separated by low-angle boundaries and are characterized by unusual decoupling of Au and As. Mantle formation is attributed to rapid disequilibrium precipitation from a metal-rich FeS2-supersaturated fluid. Incorporation of Au into the pyrite mantles was facilitated by abundant lattice defects produced by rapid nucleation. Gold-As–poor pyrite rims were deposited from an evolved ore fluid or other metal-depleted fluids. These results show that chemical variations recorded by fine layering within minerals can provide valuable insights into disequilibrium mass transfer and ore formation. The decoupling between Au and As in pyrite mantles indicates that As is not always a reliable proxy for Au enrichment in rapidly crystallized porous pyrite.

Boninites in many supra-subduction zone ophiolites are thought to mark the early stage of subduction. The Troodos ophiolite in Cyprus contains a variety of volcanic rocks including the lower pillow lavas and the boninite-bearing upper pillow lavas (UPL). Here we present major, trace element and Pb isotope data for melt inclusions in olivine from the UPL on the northern flank of the Troodos ophiolite. Melt inclusions hosted by high-Mg (Fo = 89.4–93.5) olivine grains have typical high-Ca boninitic compositions with high MgO (11.2–18.4 wt%) and SiO2 (51.2–55.0 wt%) and low TiO2 (0.2–0.6 wt%) contents, in line with the existence of boninites in the northern part of the ophiolite. The melt inclusions have trace element compositions similar to the Troodos boninitic lavas and can be regarded as representing the parental magma of the boninites. The boninitic magmas were derived from high-degrees of partial melting of a refractory source at 1400 °C and 1.5 GPa based on thermobarometers and phase equilibria. The melt inclusions have significantly variable Pb isotopic compositions (208Pb/206Pb = 2.059–2.122; 207Pb/206Pb = 0.828–0.862) and are enriched in fluid-mobile trace elements (e.g., U, Sr, and Pb), indicating that the refractory source was metasomatized by slab fluids. In comparison, recalculation of literature data suggests that the lower pillow lavas were derived from a less refractory and shallower source. We propose a subduction initiation model in which melting of two unique sources was needed to form the chemostratigraphy of lavas in the Troodos ophiolite.

Although the involvement of hydrous fluids has been widely invoked in formation of podiform chromitites in ophiolites, there is lack of natural evidence to signify the role and mechanism of fluids. In this study, a new model for the genesis of podiform chromitite is proposed on basis of revisits of comprehensive petrological, mineralogical and geochemical results of the well-preserved Kizildag ophiolite and the well-characterized Luobusa chromite deposit. In this model, ascending magmas intruding oceanic lithospheric mantle would presumably form a series of small magma chambers continuously connected by conduits. Tiny chromite nuclei would collect fluids dispersed in such magmas to form nascent droplets. They tend to float upward in the magma chamber and would be easily transported upward by flowing magmas. Chromite-rich droplets would be enlarged via coalescence of dispersed droplets during mingling and circulation in the magma chamber and/or transport in magma conduits. Crystallization of the chromite-rich liquid droplets would proceed from the margin of the droplet inward, leaving liquid entrapped within grains as precursor of mineral inclusions. With preferential upward transportation, immiscible chromite-rich liquids would coalesce to a large pool in a magma chamber. Large volumes of chromite would crystallize in situ , forming podiform chromitite and resulting in fluid enrichment in the chamber. The fluids would penetrate and compositionally modify ambient dunite and harzburgite, leading to significant fractionations of elemental and isotopic compositions between melts and fluids from which dunite and chromitite respectively formed. Therefore, fluid immiscibility during basaltic magma ascent plays a vital role in chromitite formation.

In order to understand the effect of fluid-induced alteration on the Sm-Nd isotope systematic in apatite, a series of fluid/apatite reaction experiments, which involve reacting the well-characterized Durango fluorapatite with CO2 ± CaCO3-, NaF- or HCl ± CaCl2-bearing solutions with a known 143Nd/144Nd ratio, were conducted at 800 or 600 °C at 200 MPa. In experiments involving CO2-H2O ± CaCO3, the fluorapatite grains did not react with the solution, such that the Sm-Nd isotopic system was undisturbed. In experiments involving NaF, the fluorapatite grains were partially to completely altered. During the alteration process, REE mobilization was retarded via the coupled substitution Na+ + REE3+ = 2Ca2+ due to the high activity of Na in the fluid. Because the REE were not mobilized, the 147Sm/144Nd ratios remained constant. However, the 143Nd/144Nd ratios were slightly altered due to small degrees of Nd isotopic exchange between the fluid and fluorapatite. In experiments involving HCl ± CaCl2, the fluorapatite grains were partially altered, and the REE were variably leached from the altered fluorapatite. Leaching of REE was accompanied by an increase in the 147Sm/144Nd ratio, which is related to the higher compatibility of Sm in the fluorapatite structure and the lower mobility of Sm in Cl-bearing fluids. Although the 147Sm/144Nd ratios were strongly affected, the 143Nd/144Nd ratios experienced only a small change, which is related to the slow rate of transport for Nd between reaction-front fluid and bulk fluid. In general, the experimental results indicate that fluid chemistry is the main factor controlling the response of the Sm-Nd isotopic system to fluid-induced alteration. The 147Sm/144Nd ratio of apatite can be highly modified, and in turn the Sm-Nd isotopic system is disturbed when the fluids are rich in ligands that are able to facilitate REE mobilization and fractionation. Therefore, a thorough evaluation of the fluid-rock history, along with the conjectured fluid chemistry, is necessary before using apatite Sm-Nd isotopes as geological indicators.

The role of the Cathaysia Block, South China, and its linkage with orogenesis in the Rodinia and Gondwana supercontinents remain unresolved because of uncertainties in its paleoposition in supercontinental reconstructions. The lithostratigraphic, geochronologic, geochemical (including isotopic), and paleocurrent data on late Neoproterozoic to early Paleozoic sub-basins in the Cathaysia Block reveal spatio-temporal, tectono-sedimentary, and provenance diversity that show linkages with previously adjacent terranes and orogens in East Gondwana. An abundance of siliceous and conglomeratic rocks, local unconformities, and pinch-out in certain layers indicate a late Cryogenian proximal deposition, late Ediacaran transgression, and Middle Cambrian uplift. Cryogenian to early Ordovician strata contain predominantly 1000–900 Ma (late Grenvillian age) detrital zircon population, whereas 1300–1050 Ma (early Grenvillian age) zircons are only dominant in strata earlier than late Ediacaran. Besides, 850–700 Ma zircons are the most prominent group in the Middle Cambrian strata along with an occasional increase in the 650–500 Ma (Pan-African age) zircons.The Grenvillian age zircon groups exhibit significant sediment input from the eastern Indian terrane (990–950 Ma) and western Australia (1300–1050 Ma), underlining the fact that the Cathaysia Block was located between these two terranes in the northern part of East Gondwana. The diminishing contribution of early Grenvillian components in the late Ediacaran strata can be linked to the Pinjarra Orogeny (550–520 Ma), which led to uplift that blocked the transport of detritus from Australia. Middle Cambrian provenance variation with high abundance of 850–700 Ma components indicates the presence of intrinsic sediment from the Wuyishan terrane of South China. Given that the Cathaysia Block was a passive continental margin, this change was caused by the uplift of the southeastern Cathaysia Block, which was related to the far-field stress effects of the late phase of the Kuunga Orogeny (530–480 Ma). The decrease in abundance of early Grenvillian and Pan-African zircons in space implies that they were transported into the basins through independent drainage systems. This is consistent with the local and temporal variations in paleocurrent orientations during the Cambrian, further implying diverse and complex drainage systems in the southwestern Cathaysia Block during this period.

Subtropical weathering of granitic catchments in South China has led to the formation of numerous giant regolith-hosted rare earth element (REE) deposits that currently account for more than 15% of global REE production and more than 95% of global heavy REE (HREE) production. Understanding the controls on mobilization and redistribution of the REEs during subtropical weathering in these granitic catchments is crucial for efficient exploration for this type of deposit in the world. As exemplified by the Bankeng light REE (LREE) deposit in South China, the key factors controlling the mobilization and redistribution of the REEs, especially the easily exchangeable REEs, are soil pH and primary REE mineralogy. The nature of the primary REE minerals, apatite, monazite-(Ce), and subordinate bastnäsite-(Ce), parisite-(Ce), and xenotime-(Y) places an important control on the behavior of the REEs during incipient weathering. Dissolution of these minerals is slow during incipient weathering, and, therefore, enrichment in REEs in this stage results largely from the removal of major elements during the decomposition of albite, K-feldspar, and biotite. Dissolution of the primary REE minerals higher in the profile liberates the REEs, which are then transported to locations where the soil pH abruptly increases due to water-regolith interaction, such as the pedolith-saprolite interface, and adsorption on kaolinite-group minerals efficiently fixes the REEs in regolith. Geomorphologically, the Bankeng deposit, like most of the other regolith-hosted REE deposits in South China, is located on concave-convex hillslopes, where erosion is prevalent at the ridgetop and decreases in intensity downslope. Results of this study show that strong erosion, coupled with intense chemical weathering at the ridgetop, is responsible for the enrichment in REEs by releasing the REEs, especially the LREEs, from their primary sources and supplying kaolinite and halloysite needed for the REE adsorption by decomposing albite, K-feldspar, and biotite. Decomposition of these major rock-forming minerals also leads to an enrichment of the REEs through the removal of components. The HREEs are lost preferentially to the groundwater and transported downslope, resulting in the enrichment of these elements in the lower part of the weathering crust at the footslope. Significant lateral Ce transport is also probable. A series of oxic fronts were developed at the footslope, with the most persistent one along the saprolite-saprock interface, due to seasonal fluctuations of the groundwater table. Cerium was immobilized there, predominantly through adsorption on Fe-Mn oxyhydroxides, causing enormous accumulation. Therefore, hillslope processes and groundwater flow could redistribute the REEs across the entire catchment, preferentially enriching the LREEs at the ridgetop and the HREEs at the footslope. Also, intense erosion facilitates chemical weathering and the accumulation of REEs, but the development of a thick weathering crust is favored by weak erosion. Repeated periods of high and low erosion rates in South China have enabled the gradual development of thick weathering crusts at the ridgetops that are sufficiently enriched in REEs to now constitute a major resource of these economically important elements.

Rare earth elements (REE) and scandium (Sc) have become strategical and critical resources to nations worldwide. Currently, regolith-hosted REE deposits in South China and the neighbouring regions provide more than 90% of the global heavy REE (HREE) production. Recent discoveries highlight the potential of exploitable Sc resources from regolith-hosted deposits. Hence, knowledge about the nature and ore-forming processes of these REE and Sc deposits is important to facilitate identification of favourable geological and geomorphological environments and thus to guide exploration of these deposits. Previous studies have confirmed that felsic and mafic-ultramafic rocks are favourable parent rocks for the formation of REE and Sc deposits, respectively. Apparently, these parent rocks are already enriched in REE and Sc, both geochemically and mineralogically, during their petrogenesis. Favorable protolith for the formation of regolith-hosted REE deposits includes granites, syenites, and volcanic rocks with most of the REE hosted in weathering susceptible minerals, for example, synchysite, gadolinite, and hingganite for the HREEs, and allanite, titanite, apatite for the LREEs. Clinopyroxene-rich mafic-ultramafic rocks are ideal protolith for the formation of regolith-hosted Sc deposits. During weathering, REE are liberated during decomposition of the parental minerals and mainly sorbed on clay minerals, notably kaolinite and halloysite. A critical transformation from poorly crystallized, nano-sized halloysite and kaolinite to much more crystalline and larger vermicular kaolinite is observed with progressive weathering. Adsorption capacity of the clay minerals significantly decreases in this transformation, thereby, REE adsorption becomes unfavorable and desorption dominates in the shallow soils, whereas efficient adsorption on the clay minerals could take place in deep regolith. Weathering behavior of Sc is different from the REE. A case study from Australia reveals that Sc is initially sorbed on smectite after decomposition of parental Sc-rich clinopyroxene. Subsequently, Sc is sorbed on neoformed Fe oxyhydroxides, thereby efficiently retain in the shallow Fe-rich duricrust, when smectite is dissolved in more advanced weathering. Mobilization, re-distribution, and accumulation of REE and Sc to form deposits are closely related to the groundwater flow and adsorption-desorption processes of these elements on the supergene minerals. Transport of REE would be dominated as free cations in shallow, acidic soils but the importance of REE-carbonate complexes dramatically increase with depth, facilitating the mobilization of HREE. Moreover, external elemental input from local groundwater would result in precipitation of supergene REE minerals, for example chernovite-(Y) at the Zudong deposit in South China. Supergene processes are key in controlling the formation and preservation of ore bodies. Alternating episodes of intense and weak erosion are favourable to form giant regolith-hosted deposits, whereas an equilibrium state between weathering and erosion also plays a role in the ore formation. Intense erosion would accelerate chemical weathering and liberation of REE and Sc from the protolith for ore formation, whereas intervals of minimal erosion are also required for preservation of thick ore bodies. Pre-concentration of REE and Sc in parent rocks, and favourable supergene environments for the ore formation and preservation remain for further studies. Key scientific questions awaited to be solved include the petrogenesis of geochemically and mineralogically favorable protolith as ground preparation for the supergene ore formation and the concentration mechanism of REE and Sc in this process. Moreover, it is crucial to decipher the occurrence of REE and Sc in different clay minerals and Fe oxyhydroxides in these deposits, and the associated sorption-desorption mechanism. Application of visible light-infrared spectroscopy to the exploration of regolith-hosted REE and Sc deposits deserves particular attention in the future endeavour.

Carbonaceous matter (CM) plays a critical role in the formation of sediment-hosted ore deposits, but metal partitioning between CM and associated sulfides remains unclear. Here we use synchrotron X-ray fluorescence microscopy (SXRF), laser ablation–inductively coupled plasma–mass spectrometry (LA–ICP–MS), and nanoscale secondary ion mass spectrometry (NanoSIMS) to characterize the distribution and deportment of trace elements between CM and sulfides from the Daqiao orogenic gold deposit, China. Four types of CM co-existing with different generations of sulfides are recognized: sedimentary CM1 with pyrite polyframboids in black shale host; indigenous CM2 with early- to main-ore pyrite and marcasite; CM3 with main-ore pyrite; CM4 with pyrite porphyroblasts in underlying graphitic schists. The SXRF results reveal that As, Se, Ni, Pb, and Cu are enriched in pyrite polyframboids relative to CM1 in black shales, although a small proportion of As, Pb, Bi, and U is accommodated in the latter. NanoSIMS images show that Au and As are associated with the CM1 matrix within pyrite framboids that are overgrown by recrystallized pyrite rims. Most trace elements are enriched in the newly-formed aggregates of pyrite and marcasite relative to the CM2 rims that formed via the interactions between ore fluids and pre-existing CM2 cores. Gold, As, Tl, Se, and Hg are highly enriched in the main-ore pyrite relative to co-precipitated CM3 veinlets, and this deportment might reflect changes in fluid redox conditions in response to hydraulic fracturing that is the main mechanism of gold deposition. The pyrite porphyroblasts in the schists are enriched in As, Se, Ni, and Cu relative to CM4, but have lower Au, As, Co, Ni, and Pb than sedimentary pyrite framboids. Taken together, we suggest that biological activity causes accumulation of As and Au within CM interstitial to the framboidal pyrite, and that these elements are released during metamorphic recrystallization and hydrothermal replacement. Trace element deportment in hydrothermal CM and sulfides is controlled by a variety of factors such as the fluid chemistry and interaction between indigenous CM and fluids. Our quantitative results provide significant new insights into the complex processes that formed carbonaceous sediment-hosted ore systems worldwide.

Permian-Triassic granites and associated tin deposits are widespread in the Eastern and Western belts of Peninsular Malaysia. The ages and key controlling factors of tin mineralization, however, are poorly constrained. Cassiterite separates from the Sintok and Rahman tin deposits in the Western belt, and Bandi, Setahum, Lembing, and Cherul tin deposits in the Eastern belt have U-Pb ages of 218.9 ± 3.4 and 226.8 ± 7.6 Ma, and 213.1 ± 3.9, 270.6 ± 4.6, 282.7 ± 4.6, and 281.3 ± 3.5 Ma, respectively. These ages directly constrain the tin mineralization in Peninsular Malaysia to two separate periods: 290 to 270 Ma and 230 to 210 Ma. Zircon crystals from tin-bearing granites in the Cherul and Sintok deposits have U-Pb ages of 276.0 ± 1.9 and 221.9 ± 0.6 Ma, respectively, consistent with the cassiterite U-Pb ages within uncertainties. Zircon crystals from barren granites of the Kuantan pluton in the Eastern belt have a U-Pb age of 260.5 ± 0.7 Ma, which is between the two tin mineralization periods. Zircon from these barren granites have εHf(t) values from −5.4 to 3.6, two-stage Hf model ages (TDM2) from 1.4 to 1.0 Ga, and Ce4+/Ce3+ ratios from 40 to 120. By comparison, zircon crystals from the tin-bearing granites have low εHf(t) values (−9.7 to −3.2) and Ce4+/Ce3+ ratios (4–67) and high TDM2 (1.8–1.4 Ga). Zircon ages, Hf isotopes, and trace elements indicate that the tin-bearing granitic magmas in Peninsular Malaysia had relatively low oxygen fugacity and were derived from reworking of Paleo- to Mesoproterozoic sedimentary rock-dominated crust in response to the Paleo-Tethyan subduction and continental collision. This study confirms that the nature of magma sources and redox states of magmas were key in the formation of the tin-rich granites and associated tin deposits and that the granite-related tin mineralization in Peninsular Malaysia was closely related to the evolution of the eastern Paleo-Tethys.

Rare earth elements (REEs) have become increasingly important to our modern society due to their strategical significance and numerous high technological applications. Regolith-hosted heavy rare earth element (HREE) deposits in South China are currently the main source of the HREEs, but the ore-forming processes are poorly understood. In these deposits, the REEs are postulated to accumulate in regolith through adsorption on clay minerals. In the Zudong deposit, the world's largest regolith-hosted HREE deposit, clay minerals are dominated by short, stubby, nanometer-scale halloysite tubes (either 10 or 7 Å) and microcrystalline kaolinite in the saprolite and lower pedolith and micrometer-sized vermicular kaolinite in the humic layer and upper pedolith. A critical transformation of the clay minerals in the upper pedolith is coalescence and unrolling of halloysite to form vermicular kaolinite. Microcrystalline kaolinite also transformed to large, well-crystalline vermicular kaolinite. This transformation could result in significant changes in different physicochemical properties of the clay assemblages. Halloysite-abundant clay assemblages in the deep regolith have specific surface area and porosity significantly higher than the kaolinite-dominant clay assemblages in the shallow soils. The crystallinity of clay minerals also increased, exemplified by decrease in Fe contents of the kaolinite group minerals (from ~1.2 wt% in the lower saprolite to ~0.35 wt% in the upper pedolith), thereby indicative of less availability of various types of adsorption sites. Hence, halloysite-abundant clay minerals of high adsorption capacity in deep regolith could efficiently retain the REEs released from weathering of the parent granite. Reduction in adsorption capacity during the clay transformation in shallow depth partially leads to REE desorption, and the released REEs would be subsequently transported to and adsorbed at deeper part of the soil profile. Hence, the clay-adsorbed REE concentration in the lower pedolith and saprolite (~2500 ppm on average) is much higher than the uppermost soils (~400 ppm on average). Therefore, weathering environments that favor the release of the REEs in the shallow soils but preservation of halloysite in the deep regolith can continuously adsorb REEs in the clay minerals to form economically valuable deposits.

The Toba volcanic system in Indonesia has produced two of the largest eruptions (>2,000 km(3) dense-rock equivalent [DRE] each) on Earth since the Quaternary. U–Pb crystallization ages of zircon span a period of ∼600 ky before each eruptive event, and in the run-up to each eruption, the mean and variance of the zircons’ U content decrease. To quantify the process of accumulation of eruptible magma underneath the Toba caldera, we integrated these observations with thermal and geochemical modeling. We show that caldera-forming eruptions at Toba are the result of progressive thermal maturation of the upper crustal magma reservoir, which grows and chemically homogenizes, by sustained magma influx at average volumetric rates between 0.008 and 0.01 km(3)/y over the past 2.2 My. Protracted thermal pulses related to magma-recharge events prime the system for eruption without necessarily requiring an increased magma-recharge rate before the two supereruptions. If the rate of magma input was maintained since the last supereruption of Toba at 75 ka, eruptible magma is currently accumulating at a minimum rate of ∼4.2 km(3) per millennium, and the current estimate of the total volume of potentially eruptible magma available today is a minimum of ∼315 km(3). Our approach to evaluate magma flux and the rate of eruptible magma accumulation is applicable to other volcanic systems capable of producing supereruptions and thereby could help in assessing the potential of active volcanic systems to feed supereruptions.

Clay-bearing sedimentary rocks at Gale crater are more consistent with subaerial chemical weathering than formation in a lake., After over 8 years of successful surface operations on Mars, the Curiosity rover has revealed much about the environment in Gale crater. Despite early observations of a lacustrine environment, few of the subsequent deposits exhibit demonstrable lacustrine character. We suggest instead that most of the stratigraphic section explored to date can be best explained as eolian and/or volcaniclastic sediments subaerially chemically weathered by acidic precipitation in a reduced atmosphere. Most of the deposits in Gale crater seemingly did not form in an ancient lake, but the results nonetheless shed considerable light on ancient climate, environmental change, and the astrobiology of Mars. Discoveries by Curiosity provide a critical piece to Mars’ global alteration puzzle.

Combined Fe isotopic and olivine fabric analyses have been conducted on the nodular chromitites from the Luobusa ophiolite, Tibet, to study the physico-chemical effects of plastic flow in modifying chromitites and their associated mantle peridotites below spreading centers. The Luobusa nodular chromitites are generally composed of fresh dunitic matrices and well-aligned chromite nodules. Olivine grains in both the dunitic matrices and chromite nodules underwent pressure solution, and those in the dunitic matrices have prevalent Fe-rich stripes parallel to their kink bands and sometimes show lattice dislocation-induced preferred orientations of (1 0 0). Chromite grains in the chromite nodules have Cr# ∼ 75–80 and δ56Fe values from −0.122 to 0.067. The Fo and δ56Fe values of olivine in the dunitic matrices vary from 96 to 99 and 0.165 to 0.294, respectively, distinguishably higher than those of olivine in ordinary ophiolitic dunites (mostly 90–95 and 0.050–0.150). The calculated Mg-Fe exchange temperatures between olivine in the dunitic matrices and chromite in the chromite nodules range from 700 to 900 °C, overall higher than those for ordinary dunites and chromitites (∼600–800 °C). Combined with numerical modeling of diffusion of Fe in olivine, our study reveals that the abnormally high Fo and δ56Fe values of olivine in the dunitic matrices of nodular chromitites were not controlled by their parental magmas but caused by significant loss of Fe from olivine under subsolidus conditions. Intensive plastic deformation in nodular chromitites largely promoted the migration of Fe from olivine and facilitated the Mg-Fe exchange between the chromite nodules and dunitic matrices, maximizing the Fo and δ56Fe values of olivine. However, such an effect is expected to be negligible in chromite-barren dunites and harzburgites because such rocks do not have enough chromite to generate effective chemical-potential gradients for Fe migration. The high Fo values of olivine in nodular and massive chromitites require fast cooling rates (∼0.05–0.1 °C/yr) in the surrounding mantle, consistent with settings below intermediate-fast spreading ridges. Boudinization of small chromitite bodies in less competent dunites under 900–1200 °C is a potential mechanism for generating the nodular texture, also accounting for the en-echelon or bead-like occurrences of chromitite bodies in ophiolites. Effective mantle flow below spreading centers plays an important role in shaping the textures and compositions of mantle rocks, explaining the absence of deformation-related nodular chromitites in layered mafic-ultramafic intrusions.

Titanium (Ti) is typically hosted in detrital minerals in marine sediments and has long been considered to be immobile during diagenesis. In this study, the authigenic titania minerals, brookite and anatase, are observed in early Cambrian carbonaceous shales from the Meishucun and Zhajin sections of South China, respectively. Black shales in the Meishucun section have total organic carbon (TOC) contents from 1.6 to 3.9 wt% and HI (hydrogen index) values from 3.8 to 20 mg HC/g TOC, whereas black shales in the Zhajin section have much higher TOC (7.1–15.6 wt%) but lower HI (

The Li-O isotopes of olivine and pyroxene from the mantle sequence of the Luobusa ophiolite, a fragment of Neo-Tethyan forearc lithosphere in Tibet, reveal a series of magmatic processes and geochemical evolution in a nascent mantle wedge during subduction initiation. Olivine grains from the ophiolitic mantle sequence display δ18O values similar to those of normal mantle (+5.18 ± 0.2‰), but have large δ7Li variations, ∼+4‰ to +13‰ in harzburgites and ∼+1‰ to +11‰ in high-Cr chromitites. Clinopyroxene grains in cpx-bearing harzburgites have finger-like protrusions and LREE-depleted patterns comparable to those of clinopyroxene in abyssal peridotites, and also show high Li concentrations (∼2–8 μg/g) and negative δ7Li values (∼−15‰ to −8‰), which are beyond the ranges of normal upper mantle (

Apatite is present within both the hosting lherzolite and sulfide ore at the Jinchuan magmatic Ni-Cu sulfide deposit of northwest China. Apatite grains within the lherzolite are generally large and hexagonal (>200 μm) and are associated with interstitial phlogopite and amphibole. These apatite grains contain ~0.9 wt % F, ~1 wt % Cl, 6,800 to 14,500 ppm rare earth elements (REE) and have in situ δ18OV-SMOW values of 5.10 to 6.38‰, all of which are indicative of crystallization from an evolved silicate magma. In comparison, the massive and disseminated sulfide ores contain fine-grained apatite ( The apatite within the lherzolite and sulfide ore crystallized from two conjugate immiscible silicate and sulfide melts, respectively. Rare earth elements and F were preferentially partitioning into silicate melts, whereas most volatile components were mainly partitioned into the sulfide melts. The silicate magmas from which apatite crystallized were rich in light REE (LREE) relative to heavy REE (HREE). Volatile components in the sulfide melts changed the physicochemical conditions to enable such high-density melts to migrate upward and finally settle in the shallow chamber with silicate rocks.

Tables A1–A7; Figures A1–A4.