Intercumulus Melt Research Articles

Crustal Growth Identified by High-δ18O Zircon and Olivine: A Perspective from Ultramafic Arc Cumulates in Southern Tibet

Abstract In recent studies of crustal growth using global zircon Hf–O isotopic datasets, high-δ18O zircons are typically attributed to intra-crustal reworking during which very little juvenile mantle-derived magmas were added to the crust. Although arc magmas may originate from a high-δ18O mantle wedge, it has been difficult to decipher the contribution of high-δ18O mantle to zircon-saturated felsic magma due to superimposed intra-crustal processes. We address this issue by combining the data from high-δ18O zircon-bearing ultramafic cumulates and coeval lavas from a Cretaceous magmatic arc in southern Tibet. The cumulates mainly consist of different proportions of cumulus olivine and intercumulus amphibole. Amphibole analyses show a transition from increasing to decreasing Zr with increasing SiO2 (50–74 wt.%) contents in the intercumulus melts, indicating zircon saturation in late-stage interstitial melts. The εNd(t) values (2.4 ± 1.4) of the apatite grains crystallized before and after zircon remain almost constant. Interstitial zircons have δ18O (6.1–7.2‰) values similar to the earliest crystallized olivine (δ18O = 6.3–7.1‰) in the cumulates. The coeval lavas may represent the intercumulus melts extracted from amphibole-rich cumulates at different depths. Both the lavas and cumulates were ultimately derived from high-δ18O arc mantle modified by small amounts (<12%) of subducted sediments, and crystallized zircon during intra-crustal magma evolution without involving crustal contamination or melting. These high-δ18O zircons therefore are not products of crustal reworking, but record crustal growth during their crystallization (110 ± 2 Ma). Our study shows that the combination of zircon and olivine oxygen isotopes for ultramafic to felsic rocks is more effective than zircon data alone in evaluating the role of crustal growth vs. reworking in an arc system. The implication is that global zircon-based crustal evolution models that attribute all high-δ18O zircons to crustal reworking may conceal recent crustal growth.

Behavior of critical metals in cumulates of alkaline ultramafic magmas in the Siberian large igneous province: Insights from melt inclusions in minerals

Ultramafic-alkaline-carbonatite complexes (UACC), which are formed from mantle-derived carbonated alkali-ultramafic melts in large igneous provinces (LIPs), are important resources of Fe, Ti, U, Th, Nb, rare earth elements (REEs), Cu, Ni and platinum group elements (PGEs). Concentration of these metals and ore formation is assumed to be largely controlled by magmatic differentiation and post-magmatic hydrothermal processes. Although basic patterns of the metals’ partitioning during formation of the UACCs are constrained by geochemical and mineralogical features of the rocks, including in experimental studies, our understanding of their pathway “from primitive melt to ore deposit” is far from complete. In order to further constrain the metals’ behavior during differentiation of a carbonated alkali-ultramafic melt, we studied multiphase inclusions in olivine, chromite, perovskite, pyroxene and magnetite in the ultramafic rocks from three UACCs (Guli, Bor-Uryakh and Odikhincha), located in the Siberian LIP. Examination of both unheated and experimentally heated and quenched inclusions reveals a variety of compositions from melanephelinitic through to highly differentiated nephelinitic to alkali-rich carbonatitic. In addition, sulfide minerals, which turn into immiscible sulfide liquids during heating experiments, are widely distributed in the inclusions’ assemblages. We consider these inclusions to be snapshots of intercumulus melts, which were entrained into olivine-rich cumulate mush, and use their compositions to delineate plutonic differentiation of a carbonated alkali-ultramafic melt. Highly differentiated silicate melts, entrapped in Fe-rich chromite (Guli dunites), were rich in U, Th and Nb and crystallized Os-Ir-Ru phases in proximity to the host chromite. Concentrations of U, Th, Nb and REEs in alkali-rich carbonatite liquids, which were present in the intercumulus of Odikhincha and Bor-Uryakh peridotites, approached levels similar to mineralized carbonatites and support the concentration of these metals in an immiscible alkali-carbonatite fraction, which was enriched in S, P and Cl. Minor sulfide liquids, which are closely associated with these carbonatite fractions, were strongly enriched in Cu and Ni, thus explaining the origin of the Phalaborwa-like sulfide ores in carbonatites as a result of magmatic differentiation. Finally, our study provides insights into the formation of peridotite-hosted types of mineralization (perovskite-magnetite ores, mineralized carbonatite veins and PGE-bearing chromitites) and shows that these ore-bearing assemblages can be formed due to the infiltration of the metal-bearing intercumulus melts through the ultramafic matrix.

Magnetite layer formation in the Bushveld Complex of South Africa

The great economic significance of layered mafic-ultramafic intrusions like the Bushveld Complex of South Africa results from the existence within them of some layers highly concentrated in valuable elements. Here we address the origins of the Main Magnetite Layer, a globally important resource of Fe-Ti-V-rich magnetite. Previous models of in situ fractional magnetite crystallization require frequent ad hoc adjustments to the boundary conditions. An alternative model incorporating compositional convection near the top of the pile and infiltration of the pile from beneath by migrating intercumulus melt fits observations without any adjustments. Lateral variations in Cr concentration formerly held as indisputable evidence for in situ crystallization can be accommodated better by models of reactive melt infiltration from below. The choice of models has pivotal ramifications for understanding of the fundamental processes by which crystals accumulate and layers form in layered intrusions.

No mantle residues in the Isua Supracrustal Belt

A critical component of our understanding of the evolution of Earth's mantle comes from rocks identified as direct mantle samples. Eoarchaean dunite lenses from the Isua Supracrustal Belt (ISB), North Atlantic Craton, Greenland, have been previously interpreted as depleted mantle wedge residues, complementary to arc-like volcanic rocks in the supracrustal sequence. This would place the ISB dunites among Earth's oldest mantle samples. We present new major element, platinum-group element (PGE) and Re-Os isotopic data for the ISB dunites, and critically assess the criteria previously used to invoke a mantle origin for the dunites. We find no evidence that uniquely supports a mantle origin. Instead, evidence of chromite and Os-Ir alloy fractionation, consistent Pt and Pd depletion, elevated Ni contents, and trace element systematics indicate that the dunites formed as olivine ± chromite cumulates with varying amounts of intercumulus melt. Their compositions indicate crystallisation from magmas represented by ISB volcanic rocks, and their Re-Os model ages overlap the ∼3720 Ma age of the volcanic sequence, consistent with the dunites representing magma chambers or conduits that fed the volcanic eruptions. Formation of the Isua dunites as cumulates removes an important line of evidence used to interpret the ISB as an ophiolite, and highlights the risks of using criteria that do not discriminate mantle residues from olivine-rich cumulates. Extending this reasoning to other Eoarchaean crustal peridotites previously identified as mantle rocks suggests there may be no mantle residues anywhere in the Itsaq Gneiss Complex, and that the oldest mantle samples may only be found as xenoliths in volcanic rocks.

Petrogenesis of mantle peridotite and cumulate peridotite rocks from the Nagaland Ophiolite Complex, NE India

Here, we report new geochemical data from mantle peridotite (spinel lherzolite, dunite, and harzburgite) and cumulate peridotite (websterite) rocks from the northernmost portion of the Nagaland Ophiolite Complex. Based on detailed textural, mineral (major element) chemistry, and whole‐rock (major and trace element) chemistry, we reveal the origin of these rocks in different tectonic settings and their modification via the melt–mantle interaction process. The spinel lherzolite rocks with low Cr# (0.16–0.25) chrome spinel are mantle residues formed after low degrees (~10–15%) of partial melting in a mid‐oceanic ridge setting, which has been subsequently modified in a subduction zone setting. However, the dunites and harzburgite samples are interpreted to be depleted mantle residue formed in a supra‐subduction zone (SSZ) setting. Additionally, the dunites are inferred to be of replacive origin representing former harzburgites. Based on Cr# of chrome spinel, the dunites are classified into low Cr# (0.53–0.57), medium Cr# (0.73–0.79), and high Cr# (0.89–0.90) dunites. High Cr# dunite with low‐TiO2 (0.01–0.06 wt%) possibly formed by interaction with a boninitic melt, whereas low‐medium Cr# dunite with high TiO2 (0.14–0.34 wt%) formed by interaction with a high‐Ti melt in an SSZ setting. Harzburgite samples with their highly resorbed orthopyroxene grain boundaries contain chrome spinels (Cr# = 0.84–0.88; TiO2 = 0.00–0.06 wt%) similar to the high Cr# dunite samples and are also interpreted to be formed by interaction with a boninitic melt. The websterite samples with their low Mg# (0.79) olivine and low Ni (345–413 ppm) content are interpreted to be of cumulate origin formed in an SSZ setting. However, the interstitial chrome spinel blebs with low Mg# (0.18–0.19), high Cr# (0.60–0.61), Fe3+ (0.20–0.21), and very high TiO2 (2.03–2.04 wt%) are interpreted to be a result of post‐cumulus interaction of interstitial chrome spinel with high‐Ti intercumulus melt. These mantle peridotites and cumulates of different tectonic origins were obducted and juxtaposed during the final stages of the India‐Asia collision.

MORB Melt Transport through Atlantis Bank Oceanic Batholith (SW Indian Ridge)

AbstractThe Atlantis Bank Oceanic Batholith is a 660 km2 gabbro massif representing the plutonic foundation of a major ridge magmatic center. It was continuously accreted, emplaced, and exposed in the rift mountains of the paleo-SW Indian Ridge from 13 to 10·3 Ma. Ocean Drilling Program (ODP) Hole 735B, drilled to 1508 m at Atlantis Bank, recovered evolved intercalated olivine and oxide gabbros representing the upper levels of the lower ocean crust. Within this section, ∼5·6 m of primitive chromian-spinel-bearing troctolites (0·1–2 m thick), with sharp modal contacts with the host gabbros, were cored between 410 and 500 m depth. Here we present new mineral (chromian spinel, clinopyroxene, olivine, and plagioclase) major and trace element and petrographic data for the troctolite suites (i.e. troctolites, clinopyroxene-rich troctolites, troctolitic gabbros) from 410 to 528 mbsf (meters below seafloor) and olivine gabbros in the 0–274 mbsf and 410–500 mbsf intervals, and examine the origin of these troctolite layers. Equilibrium melts for the troctolites are primitive to moderately evolved, with Mg# [= 100Mg/(Mg + Fe), 48–68 mol%] comparable with those of Atlantis Bank basalts. By contrast, equilibrium melts for both the 0–274 mbsf and 410–500 mbsf host olivine gabbros are highly to moderately evolved (Mg# 27–63 mol%). The 410–500 mbsf troctolite suites maintain high clinopyroxene Mg# (81–89 mol%) and mineral concentrations of compatible elements (i.e. clinopyroxene Cr2O3 0·1–1·2 wt%, Ni 151–330 μg g–1 and olivine Ni 1055–1559 μg g–1) with the increase in incompatible trace element abundances (i.e. clinopyroxene TiO2 0·3–2·1 wt%, Zr 5·8–112 μg g–1 and olivine Ti up to 293 μg g–1). Combined with abundant dissolution–reprecipitation textures, our results indicate that the troctolites were formed by the reaction between a spinel-bearing, troctolitic mush, from which they inherited the high Mg#, Ni and Cr, and percolating melts adding incompatible trace elements. Moreover, the spinel (NiO versus TiO2) and olivine (Ti versus Y) trace element compositions indicate that the troctolites were affected by low-degree Fe–Ti-rich melt metasomatism. In contrast, both the 0–274 mbsf and 410–500 mbsf olivine gabbros display a prominent decrease in clinopyroxene Mg# (from 88 to 66) and mineral compatible element concentrations (i.e. clinopyroxene Cr2O3 from 1·1 to ∼0 wt%, Ni from 208 to 34 μg g–1 and olivine Ni from 1200 to 136 μg g–1) with increasing incompatible trace element abundances (e.g. clinopyroxene Zr from 4·8 to 157 μg g–1). These features are compatible with the reactive porous flow of slightly to highly evolved melts through the cooling crystal mush zone. Our results, combined with the literature data, indicate that most olivine gabbros between 410 and 500 mbsf were formed prior to the troctolite layers. We document that the troctolites represent conduits for mid-ocean ridge basalt melt transport through the lower oceanic crust, whereas the olivine gabbros represent crystallization of a large crystal mush, recording initial gabbro emplacement, hyper- and sub-solidus deformation, and melt–rock reaction owing to upward penetrative flow of intercumulus melt.

Compositional Variations of Apatite and REE-Bearing Minerals in Relation to Crystallization Trends in the Monchepluton Layered Complex (Kola Peninsula)

Abstract ––We have investigated the compositional variations of apatite (Ap) and rare-earth element (REE) minerals in the Monchepluton layered complex on the Kola Peninsula. On the basis of large sets of pertinent analytical data, we have estimated geochemical trends involving major, minor, and trace elements and studied their relation with the compositions of rock-forming silicate and oxide minerals. The variations observed in Ap differ considerably from trends reported for other layered intrusions. The composition fields of Ap are not consistent with the variations in the chemical composition of the bulk rocks and their constituent minerals, as determined along the representative cross sections of the entire complex. The compositional variations of Ap are fairly similar in all units of the complex. Chlorapatite (>6 wt.% Cl) is invariably abundant. There is no relationship between the Cl content of Ap and the degree of magnesium enrichment of the coexisting early magmatic silicates. In the F–Cl–OH diagram, broad fields of ternary solid solution are observed. There are no compositions along the Cl–F axis. The compositions of Ap are notably poor in Cl in the marginal series (the Nyud massif) and correspond to hydroxylapatite with a high content of fluorapatite component. Two composition fields of Ap are recognized in the Monchepluton complex: ≤3 wt.% and >6 wt.% Cl; there are, however, extensive overlaps. Two generations of apatite are thus implied. The first nucleated at the early stage of crystallization of H2O-bearing intercumulus melt as a result of substantial increase in the contents of P, F, Cl, and other incompatible components. The following stage of degassing of the crystallizing melt caused a decoupling of Cl and F. Fluorine remained mostly in the melt; in contrast, Cl was partitioned efficiently into an H2O-bearing fluid phase. At the early stage, the apatite incorporated combinations of hydroxylapatite and fluorapatite, with a low content of Cl. At the late stage, chlorapatite crystallized from a Cl-rich fluid, and ferrochloropargasite (4.1 wt.% Cl) formed in the Poaz massif as a result of autometasomatic alteration via reactions of this fluid with plagioclase and pyroxene. The apatite has high Sr contents (up to 4.1 wt.% SrO) in the highly magnesian cumulates of the Dunite block and the massifs of mounts Kumuzh’ya, Nittis, and Travyanaya. This enrichment illustrates the accumulation of Sr in the intercumulus melt, in which Ap was the only Sr-bearing phase in the absence or scarcity of intercumulus plagioclase. The REE contents also increased in the intercumulus melt and led to the formation of monazite-(Ce), REE-bearing Ap, and allanite-(Ce) in the remaining microvolumes of melt. Loveringite and Ap crystallized as coexisting phases in Mt. Sopcha. For the first time in a layered intrusion, an extensive range of compositions is documented in the Ce–La–Nd diagram for the REE-bearing phosphates (monazite and REE-rich apatite), which display a predominant La ↔ Nd substitution at the constant contents of Ce.

A Fractional Crystallization Link between Komatiites, Basalts, and Dunites of the Palaeoproterozoic Winnipegosis Komatiite Belt, Manitoba, Canada

AbstractThe rock type most commonly associated with komatiite throughout Earth’s history is tholeiitic basalt. Despite this well-known association, the link between komatiite and basalt formation is still debated. Two models have been suggested: that tholeiitic basalts represent the products of extensive fractional crystallization of komatiite, or that basalts are formed by lower degrees of mantle melting than komatiites in the cooler portions of a zoned plume. We present major and trace element data for tholeiitic basalts (∼7·5 wt% MgO) and dunites (46–48 wt% MgO) from the Palaeoproterozoic Winnipegosis Komatiite Belt (WKB), which, along with previous data for WKB komatiites (17–26 wt% MgO), are utilized to explore the potential links between komatiite and basalt via crystallization processes. The dunites are interpreted as olivine + chromite cumulates that were pervasively serpentinized during metamorphism. Their major and immobile trace element relationships indicate that the accumulating olivine was highly magnesian (Mg# = 0·91–0·92), and that small amounts (4–7 wt% on average) of intercumulus melt were trapped during their formation. These high inferred olivine Mg# values suggest that the dunites are derived from crystallization of komatiite. The tholeiitic basalts have undergone greenschist-facies metamorphism and have typical geochemical characteristics for Palaeoproterozoic basalts, with the exception of high FeO contents. Their REE patterns are similar to Winnipegosis komatiites, although absolute concentrations are higher by a factor of ∼2·5. The ability of thermodynamic modelling with MELTS software to reproduce komatiite liquid lines of descent (LLD) is evaluated by comparison with the crystallization sequence and mineral compositions observed for Winnipegosis komatiites. With minor caveats, MELTS is able to successfully reproduce the LLD. This modelling is extended to higher pressures to simulate crystallization of komatiitic melt in an upper crustal magma chamber. All major and rare earth element characteristics of the tholeiitic basalts can be reproduced by ∼60 % crystallization of a Winnipegosis komatiite-like parental melt at pressures of ∼1·5–2·5 kbar at oxygen fugacities between QFM − 1 and QFM + 1, where QFM is the quartz–fayalite–magnetite buffer. Winnipegosis basalts have low Mg# values that are not in equilibrium with mantle peridotite. They therefore cannot represent primary mantle melts derived from cooler mantle than the komatiites, and require fractional crystallization processes in their formation. Furthermore, their trace element characteristics indicate a depth of melting indistinguishable from that of the Winnipegosis komatiites, and derivation from an identical depleted mantle source. All geochemical and geological evidence is therefore consistent with their derivation from a komatiitic melt, and the presence of a large komatiite-derived dunite body in the WKB provides evidence of extensive fractionation of komatiite in the upper crust. The observed uniform basalt compositions are interpreted as the result of a density minimum in the evolving komatiitic melt at temperatures between clinopyroxene and plagioclase saturation, with efficient extraction of melt from a mixture containing ∼60 % crystals. We conclude that the WKB basalts formed by fractional crystallization of a komatiitic parental melt, and suggest that this model may be more broadly applicable to other localities where komatiites and associated basalts show similar geochemical or isotopic characteristics.

Zircon formation in mafic and felsic rocks of the Bushveld Complex, South Africa: Constraints from composition, zoning, Th/U ratios, morphology, and modelling

Zircon is a potential petrogenetic indicator that can be used to derive physicochemical conditions during magma crystallization. In this study, such conditions are obtained from zircon of both felsic and mafic rocks of the Bushveld Complex (BC), which are characterized by a wide range in bulk-rock Zr contents (4–552 ppm). For that, information from bulk-rock compositions, petrography, zircon trace element data and morphologies are combined with results of thermodynamic modelling using the software packages rhyolite-MELTS and Perple_X. In felsic rocks (Lebowa Granite, Rashoop Granophyre), zircon is formed in rutile-free assemblages together with olivine-clinopyroxene-amphibole-biotite-ilmenite-titanite-apatite after ≥20% fractional crystallization at 761–935 °C (mean: 860 °C), based on Ti-in-zircon thermometry using aTiO2 = 0.3, in agreement with independent geothermometers and modelling results. The resulting zircon populations show {100}- and {101}-dominated morphologies as well as high ƩREE (mean: 651 ppm) and low Ti contents (mean: 9.5 ppm), and only minor zoning in Th/U (0.3–0.8) and Nb/Ta (1.2–4.4). Identical zircon characteristics in gabbros and diorites of the Upper Zone within the Rustenburg Layered Suite (RLS) suggest admixing of felsic melts during ingression of mafic parental magmas. In contrast, zircons in mafic rocks of the lower RLS (Basal Ultramafic Sequence to lower Main Zone) show significantly lower ƩREE (mean: 324 ppm), commonly higher Ti contents (8–60 ppm), as well as large variations in Ti, U, Th contents, Th/U (0.2–24) and Nb/Ta ratios (0.15–18) as well as in zircon morphology. These zircons mostly occur in rutile-bearing intercumulus domains associated with orthopyroxene-biotite-amphibole-plagioclase-quartz-rutile and are formed at 690–962 °C (mean: 835 °C; aTiO2 = 1.0) based on Ti-in-zircon thermometry. These temperatures are in good agreement with zircon morphologies, but mostly higher than those obtained by rhyolite-MELTS modelling, suggesting zircon growth at <810 °C after >75% fractional crystallization of high Mg andesitic (B1) and tholeiitic parental magma (B2). These lower temperatures perhaps result from oversimplified modelling parameters or may reflect variable mixing of parental magma with evolved resident magma. Zircon populations in rutile-bearing mafic rocks of the lower RLS reveal two distinct zoning trends: an early trend at high Ti (>20 ppm), characterized by increasing Th/U (0.5 to 18) at decreasing U (175 to 10 ppm) from core to rim and a reverse trend at lower Ti contents. Both trends require zircon formation in intercumulus melt pockets at high zircon/melt ratios. The high-Ti trend can be explained by Rayleigh-like fractionation due to zircon growth together with rutile, both having highly different partition coefficients for U ≫ Th. The low-Ti trend results from zircon growth after onset of the biotite-in reaction, causing breakdown of previously formed rutile, thereby releasing U but no Th. The absence of pronounced Th/U zoning of zircon in felsic rocks reflects zircon growth in less fractionated melts, resulting in ThU fractionation compensated by coeval crystallization of abundant rock-forming minerals, all being highly incompatible for Th and U.

SPINMELT-2.0: Simulation of Spinel–Melt Equilibrium in Basaltic Systems under Pressures up to 15 Kbar: III. The Effects of Major Components in the Melt on the Chrome-Spinel Stability and a Possible Solution of the Problem of Origin of Chromitites

The effect of variations in the fo, fa, en, fs, di, an and ab components in high-Mg basaltic melts on the topology of the spinel liquidus was quantified by conducting simulations with the new SPINMELT-2.0 model for the spinel–melt equilibrium. It has been established that enrichment of the melt in pyroxene components leads to an increase, and conversely, enrichment in plagioclase and olivine components, to a decrease in the solubility of chromite. This effect can be important at the gravitational compaction of cumulates accompanied by the extraction of intercumulus melt and its infiltration upward. In this case, it is reasonable to expect sequential reequilibration of the infiltrating melt with cumulate piles of different composition. This suggests that Cr-spinel can be transferred and concentrated again during the postcumulus solidification stage of layered intrusions. The concentrating process involves Cr-spinel extraction into the melt enriched in pyroxene components and the subsequent redeposition of the mineral during when this melt reacts with the feldspar-rich matrix of proto-anorthosite layers or horizons enriched in olivine. Some of these layers and horizons may be produced by additional injections of more primitive magma. The realism of the proposed mechanism is evident from the well-known spatial relations between the chromite layers with anorthosites and dunite–harzburgites of the Bushveld complex.

Formation of the Vergenoeg F–Fe–REE Deposit (South Africa) by Accumulation from a Ferroan Silicic Magma

AbstractSituated in the centre of the Paleoproterozoic Bushveld Large Igneous Province (LIP) of South Africa the Vergenoeg F–Fe–REE deposit is one of the largest, but at the same time most unusual, fluorite deposits on Earth. In situ major and trace element analyses of fayalite, magnetite, ilmenite, fluorapatite, fluorite and allanite from fayalite-rich rocks are combined with oxygen isotope data for fayalite, magnetite and ilmenite to unravel the complex evolution of the deposit. Textural and compositional characterization of the fayalite-rich rocks supports a magmatic formation as cumulates and an intense late hydrothermal overprint. Fayalite accumulated together with minor Ti-rich magnetite, ilmenite, fluorapatite and allanite from a highly evolved, H2O-poor felsic melt at low oxygen fugacity. Chondrite-normalized rare earth element (REE) patterns of fayalite and the recalculated parental melts, using fayalite–rhyolite partition coefficients, exhibit positive trends with strong enrichment of the heavy REE (HREE) relative to the light REE (LREE). Apart from the LREE depletion the patterns are similar to those of highly fractionated high-silica REE rhyolites that often occur in siliceous LIPs. We attribute the LREE depletion to crystallization of accessory allanite, the main host of the LREE in the cumulates. Chondrite-normalized REE patterns of the parental melt prior to fayalite accumulation, recalculated using allanite–rhyolite partition coefficients, resemble the composition of the rhyolites of the Rooiberg Group and therefore document a petrogenetic link to the Bushveld LIP. High δ18O values of fayalite (up to ≈7·4 ‰) are consistent with its crystallization in a rhyolitic melt that has formed by extensive fractionation from basic melts of the Rustenburg Layer Suite, the mafic member of the Bushveld LIP. Primary fluorite crystallized together with rare quartz, and a second generation of fayalite, magnetite and ilmenite from rare intercumulus melt in interstices between cumulate fayalite. Textural and mineral compositional data, as well as the generally negative δ18O values of magnetite (–2·9 to 0 ‰), are in agreement with the main magnetite–fluorite ore formation in Vergenoeg being related to a hydrothermal overprint, which was responsible for further F and Fe enrichments of the rocks. Fluorine-rich fluids, released from the crystallizing granites of the felsic member of the Bushveld LIP (Lebowa Granite Suite), caused the extensive alteration of fayalite to bowlingite and its replacement by Ti-poor magnetite and quartz. The hydrothermal overprint was associated with the widespread formation of secondary fluorite and minor fluorapatite. Our new petrogenetic model for the Vergenoeg deposit, as constrained from the primary fayalite cumulates, implies that the formation of the Vergenoeg deposit was directly linked to the evolution of the Bushveld LIP.

Petrologic History of Lunar Phosphates Accounts for the Water Content of the Moon’s Mare Basalts

We present reaction balancing and thermodynamic modeling based on microtextural observations and mineral chemistry, to constrain the history of phosphate crystallization within two lunar mare basalts, 10003 and 14053. Phosphates are typically found within intercumulus melt pockets (mesostasis), representing the final stages of basaltic crystallization. In addition to phosphates, these pockets typically consist of Fe-rich clinopyroxene, fayalite, plagioclase, ilmenite, SiO2, and a residual K-rich glass. Some pockets also display evidence for unmixing into two immiscible melts: A Si-K-rich and an Fe-rich liquid. In these cases, the crystallization sequence is not always clear. Despite petrologic complications associated with mesostasis pockets (e.g., unmixing), the phosphates (apatite and merrillite) within these areas have been recently used for constraining the water content in the lunar mantle. We compute mineral reaction balancing for mesostasis pockets from Apollo high-Ti basalt 10003 and high-Al basalt 14053 to suggest that their parental magmas have an H2O content of 25 ± 10 ppm, consistent with reported estimates based on directly measured H2O abundances from these samples. Our results permit to constrain in which immiscible liquid a phosphate of interest crystallizes, and allows us to estimate the extent to which volatiles may have partitioned into other phases such as K-rich glass or surrounding clinopyroxene and plagioclase using a non-destructive method.

Numerical modeling of the effects of major components in the melt on the chrome-spinel stability and a possible solution of the origin chromitites problem

Using a new model of the spinel-melt equilibrium SPINMELT-2.0, the effect of variations of fo-, fa-, en-, fs-, di-, an- and ab-components in high-Mg basaltic melts on the topology of the spinel liquidus was quantified. It has been established that enrichment of the melt in pyroxene components leads to an increase, and with plagioclase components, to a decrease in the solubility of chromite. This effect can be important during gravitational compaction of cumulates, accompanied by the extraction of intercumulus melt and its infiltration upward. In this case, one can expect a sequential re-equilibration of the infiltrating melt with cumulative piles of different composition. This suggests the possibility of transfer and new concentration of chrome-spinel at the postcumulus stage of solidification of layered intrusions. The nature of the concentration consists in the extraction of chrome-spinel into the melt enriched in pyroxene components, followed by its discharge during the reaction of this melt with a feldspar-rich matrix of proto-anorthosite layers. The realism of the proposed mechanism is evidenced by the well-known spatial association of chromite layers with anorthosites (intrusion of Ram island, Bushveld complex).

Numerical Modeling of the Effects of Major Elements on the Solubility of Chrome–Spinel and a Likely Solution of the Problem of the Origin of Chromitite

The influence of variations in fo-, fa-, en-, fs-, di-, an-, and ab-components in a high-Mg basaltic melt on the topology of the spinellide liquidus was analyzed using the new model of the spinellide–melt equilibrium SPINMELT-2.0. It is shown that enrichment of the melt in the clinopyroxene component results in an increase in chromite solubility, whereas the plagioclase component leads to a decrease in chromite solubility. This effect may be of key importance under the conditions of gravitational shrinkage of cumulates accompanied by removal of the intercumulus melt and its upward infiltration. The successive re-equilibration of the infiltrating melt with cumulative rocks of different compositions may be expected. This allows us to assume the probability of transportation and subsequent accumulation of chrome–spinel at the post-cumulus stage in consolidation of layered intrusions. The nature of accumulation consists in the extraction of chrome–spinel into a melt enriched in the pyroxene component, which is followed by its release during the reaction of this melt with the feldspar matrix of protoanorthosite interlayers. The realism of the mechanism proposed is evident from the spatial relationship of the chromite layers with anorthosite in the Ram Island intrusion and in the Bushveld Complex.

A Model for the High-temperature Origin and Paradoxical Distribution of Pegmatites in Mafic Plutons, Smartville Complex, California

The distribution of pegmatitic rocks in the zoned gabbro–diorite plutons of the late Jurassic Smartville Complex is counter-intuitive. Whereas pegmatitic gabbros are common in mafic cumulate olivine gabbros within the zoned plutons, they appear to be absent from the more evolved gabbros and diorites. We argue that this paradox is resolved by examining the crystallization and temperature history of the rocks in question. The evolved rocks in the plutons consist of two-pyroxene hornblende biotite gabbro and diorite. The amphibole in these rocks occurs as a partial replacement of pyroxene that forms as a consequence of the incongruent crystallization reaction generalized as amphibole ± quartz ± plag1 = pyroxene + hydrous melt ± plag2 [reaction (1)]. This is a vapor-absent reaction that can buffer melt water content during crystallization and conceivably preclude water saturation altogether. Experimental evidence places the onset of the incongruent formation of amphibole at c. 900°C. In contrast, amphibole and other hydrous phases occur only in trace amounts in the cumulate olivine gabbros. The water-buffering reaction is not encountered during crystallization, the intercumulus melt proceeds to water saturation, and high-temperature plagioclase–pyroxene pegmatites form. We argue that the primary role for undercooling favored for the formation of granitic pegmatites is not likely to be applicable to the Smartville gabbro pegmatites. As crystallization of the gabbro pegmatite proceeds, the amphibole-in boundary is eventually reached and incongruent crystallization of amphibole occurs. Nevertheless, once water saturation occurs, it is irreversible, and is eventually manifested by the low-temperature deuteric alteration associated with the pegmatites. In short, we argue that pegmatite formation in the Smartville gabbros, and probably many other gabbroic plutons, reflects water saturation of a magma at high temperature, estimated here at >900°C. In systems where reaction (1) is attained prior to water saturation, pegmatite formation may be delayed until late in the crystallization history or precluded altogether. MELTS modeling of basalt crystallization suggests a solution to the paradox, indicating that the difference between the two systems can reflect equilibrium versus fractional crystallization. Equilibrium crystallization of trapped, intercumulus melt favors high-temperature crystallization and water saturation and, hence, pegmatite formation. Fractional crystallization, for example in an open-system magma chamber, favors lower-temperature crystallization and permits reaction (1) to proceed, inhibiting water saturation and preventing pegmatite formation. The open-system nature of the fractionating magma chamber might also allow escape of volatiles at its margins, with the lower volatile content then communicated to the magma as a whole as the system convects or otherwise mixes.

Coronitic textures in the ferrogabbros of the Elet’ozero intrusive complex (Northern Karelia, Russia) as evidence for the existence of Fe-rich melt. 2. Origin of Fe-rich liquid

The study of coronitic textures in ferrogabbros and data on rhythmic layering of the Elet’ozero Massif supports the existence of a specific low-temperature Fe-rich liquid in nature. This liquid was formed during solidification of intrusion owing to the local multiple accumulation of Fe and Ti contents in a parental Fe-rich Fe–Ti basaltic melt. According to obtained data, this occurred on micro- and macroscale: 1) in the interglanular (intercumulus) space of the crystallization zone where intercumulus melt becomes rich in Fe and Ti owing to the crystallization of cumulus silicate minerals and is transformed into Fe-rich liquid, which concentrates residual components of an intergranular melt; 2) during formation of rhythmic layering when Fe-rich residual melt is accumulated before the upper part of the moving front of solidification; when Fe content reaches a certain limit, the melt is also transformed in a separate Fe-rich liquid, the interlayers of which form the upper (lowest temperature) members of rhythms. It was concluded that the emergence of a Fe-rich melt is related to its specific structure, which is formed when the Fe content reaches certain critical values in a liquid. Thus, this liquid is not a product of immiscible splitting of a melt, but represents a peculiar phenomenon. The preservation of primary textures and structures of the rocks is supposedly related to the lyophobic properties of surfaces, i.e., “repulsion” of nonwetting liquid by facets of cumulus crystals, especially plagioclase. Owing to this, the drops and even horizons of heavy Fe-rich liquid are retained in situ of their formation.

Ta and Sn concentration by muscovite fractionation and degassing in a lens-like granite body: The case study of the Penouta rare-metal albite granite (NW Spain)

The Penouta peraluminous low-phosphorous granite is the most important low-grade, high-tonnage Sn-Ta-Nb-bearing albite granite from the Iberian Massif. A sheet or laccolith shape, instead of a stock, is inferred for the Penouta granite, maybe in relation with the low viscosity and high mobility of a fluorine-bearing melt. Subhorizontal lateral extension of the magma is also inferred via vertical and horizontal geochemical variations. The absence of compositional gaps in variation diagrams, coupled with continuous evolutionary trends of compatible and incompatible elements with height, discard a multi-pulse intrusion and point to a single magma pulse. Mineral chemistry, trace element and least-squares mass balance modelling support a differentiation process from bottom to top in the emplacement place. The absence of switch from incompatible to compatible behaviour (bell-shaped trends) in Sn, Nb and Ta variation diagrams, coupled to experimental constraints on tantalite and cassiterite saturation, suggest that Nb-Ta oxides and probably cassiterite were not fractionated mineral phases, their crystallisation being related to concentration gradients within a trapped intercumulus melt. Major and trace element modelling support that the concentration upwards of Ta and the Ta/Nb ratio could be a consequence of mineral fractionation, with a key role of muscovite (mainly primary) for the Ta/Nb ratio, as this mineral has a higher partition coefficient for Nb than Ta. Our results suggest that fluorine and peraluminosity had a limited effect in the Ta/Nb ratio variations. Hence, Ta enrichment is mainly controlled by fractional crystallisation processes. In most cases, Sn enrichment was also concomitant with Ta, indicating that crystal-melt fractionation processes also played an important role in Sn concentration. Nevertheless, the strongest Sn enrichment in the granite (e.g., central part of the granite body) does not correspond to a significant Ta enrichment. The high affinity of Sn for fluids and the high partitioning of Ta for melt could explain this decoupling. Nevertheless, the magmatic signature of cassiterites in these strongly Sn-enriched zones (central part of the granite body) rules out a hydrothermal subsolidus origin for this fluid. By analogy with models carried out in sill-like bodies it seems likely that the Sn enrichment in the central part of the granite body is related to fluid saturation/degassing occurred in the lower margin, as a consequence of cooling and crystallisation of mostly anhydrous minerals (i.e. second boiling). The vapour exsolved migrated into the hotter melt up to the central part, where it probably was reabsorbed, yielding cassiterite with a magmatic signature. Moreover, we suggest that heat loss in the upper margin of the granite body might also contribute to the formation of a second fluid-saturated zone. As a result, pegmo-aplites and greisen were developed.

Electrochemical Processes in a Crystal Mush: Cyclic Units in the Upper Critical Zone of the Bushveld Complex, South Africa

The Upper Critical Zone (UCZ) of the Bushveld Igneous Complex displays spectacular layering in the form of cyclic units comprising a basal chromitite layer overlain by a sequence of silicate cumulates in the order, from bottom to top, pyroxenite–norite–anorthosite. Electron microprobe and laser ablation inductively coupled plasma mass spectrometry analyses of chromite and silicate minerals in layers between the UG2 chromitite and the Merensky Reef reveal variations in major and trace element compositions that defy explanation with existing models of cumulate mineral– melt evolution. The anomalous features are best developed at sharp contacts of chromitite with adjacent anorthosite and pyroxenite cumulates. Here, chromite compositions change abruptly from high and constant Mg/(Mg þ Fe 2þ ) and Fe 2þ /Fe 3þ ratios in chromitite layers to variable and generally lower values in chromite disseminated in silicate layers. Furthermore, the composition of disseminated chromites varies depending on the host silicate assemblage; for example, in Ti, V and Zn contents. Importantly, the abrupt change in chromite composition across the chromitite–silicate layer contacts is independent of the thickness of the chromitite layer and the estimated mass proportions of chromite to intercumulus liquid. Chemical variations in plagioclase are also abrupt and some are hard to reconcile with conventional models of re-equilibration with intercumulus liquid. Among those features is the decoupling of alkalis from other incompatible lithophile elements. In comparison with cumulus plagioclase, intercumulus poikilitic plagioclase in chromitite layers is enriched in rare earth elements but strongly depleted in equally incompatible Li, K and Rb. Strong alkali depletion is also observed in intercumulus pyroxene from ultramafic cumulates and chromitite layers. To explain these features, we propose a new model of post-cumulus recrystallization, which intensifies the modal layering in the crystal–liquid mush, producing the observed sequence of nearly monomineralic layers of chromitite, pyroxenite and anorthosite that define the cyclic units. The crucial element of this model is the establishment of redox potential gradients at contacts between chromite-rich cumulates and adjacent silicate layers owing to peritectic reactions between the crystals and intercumulus melt. Because basaltic melts are ionic electrolytes with Na þ as the main charge carrier, the redox potential gradient induces electrochemical migration of Na þ and other alkali ions. Selective mobility of alkalis can explain the enigmatic features of plagioclase composition in the cyclic units. Sodium migration is expected to cause remelting of previously formed cumulates and major changes in modal mineral proportions, which may eventually result in the formation of sharply divided monomineralic layers. The observed variations in ferric/ferrous iron ratios in chromite from the cyclic units and Fe distribution in plagioclase imply a redox gradient of

Composition and inner structure of the third layer in the oceanic crust in the subequatorial segment of the Mid-Atlantic Ridge (5°–7° N)

The paper presents data on the major-component, trace-element, and mineralogical composition of plutonic rocks, and the composition of their minerals, from the Sierra Leone region in the crest zone of the Mid-Atlantic Ridge between the Strakhov and Bogdanov fracture zones. According to their relations with seafloor structures, the rock associations are subdivided into those of rift valleys and nontransform offset zones. The troctolites and olivine gabbro composing the rift association were produced early in the fractionation course of oceanic tholeiite melt in unstationary and relatively small magmatic chambers. Most rocks beneath the nontransform offset zones crystallized during the long-lasting fractionation of the melt in large chambers hosted in serpentinized peridotites. This part consists of various cumulates, ranging from troctolites to gabbroids. Where deep tectonic detachments entered partly consolidated portions of the chambers, the melt interacted with the wall rocks. Fluid that was generated via the dehydration of serpentine and concentrated hydrophile elements, locally modified the composition of the melt and resulted in amphibole-bearing rocks. Under stress, the intercumulus melts were squeezed into tectonically weakened zones, mixed there, and also interacted with the wall rocks. These mix melts produced (with the participation of fractional crystallization) mineralized Fe-Ti gabbroids. Residual portions of the melts generated most of the diorites and plagiogranites. The high-Na diorites likely crystallized from acid melts that were derived via the partial melting of older gabbroids where aqueous fluids circulated; these fluids were generated by the deserpenitization of the host rocks in tectonized zones cutting through the chambers.

Origin of PGE-Poor and Cu-Rich Magmatic Sulfides from the Kalatongke Deposit, Xinjiang, Northwest China

The Kalatongke Cu-Ni sulfide deposit in the Paleozoic Altay orogenic belt, NW China, is hosted in a Per mian mafic intrusion consisting of norite, troctolite, gabbro, and diorite. Disseminated Ni-Cu, massive Ni-Cu, and massive Cu-rich sulfide ores are mainly hosted in norite and gabbro. Some massive Ni-Cu ores also occur in the Carboniferous sedimentary rocks. The geologic and compositional relationships between various sulfide ores and the rocks of Kalatongke offer a new interpretation of the sequence of emplacement of the magmas, which underpins an understanding of the compositions of the ores and the formation of the Kalatongke deposit. Olivine grains from disseminated Ni-Cu ores have Fo values ranging from 71.6 to 78.0 mol % and Ni con tents from 1,000 to 2,200 ppm. Typically, Ni decreases from the cores to the rims from 2,000 to 1,000 ppm at constant Fo content, indicating the reaction of early-formed olivine with later-segregated sulfide melt. Cr spinels at Kalatongke are highly enriched in Fe 3+ and Fe 2+ , with relatively low Cr, Al, and Ti, reflecting reac tion with evolved trapped intercumulus melt. Norites are depleted in Nb, Ta, Zr, Hf, and Th and enriched in Sr and Ba, whereas disseminated Ni-Cu sul fide ores have considerable depletion of Rb and enrichment of Sr and Ba and lack depletion of Nb, Ta, Zr, and Hf, indicating their different origins. Disseminated Ni-Cu sulfide ores have bulk compositions with variable Cu and Ni contents which are much lower than those of massive Cu-rich and Ni-Cu ores, but disseminated and massive Ni-Cu ores have similar PGE contents with relatively low Pd/Ir ratios. Massive Cu-rich ores have much higher Pd and Pt with very high Pd/Ir ratios. The Kalatongke Cu-Ni sulfide deposit appears to have formed from two different pulses of PGE-poor and Cu-rich basaltic magmas that underwent different degrees of assimilation and fractional crystallization. The first magma pulse gained sulfide saturation because of minor crustal contamination and fractionated a small amount of sulfide (<0.03%); the evolved melt then intruded and assimilated crustal materials to attain sulfide saturation again. Sulfide liquid segregated from the magma to form the massive sulfide melts and residual magma formed the noritic rocks in the shallow magma chamber. The segregated massive sulfide melts then un derwent further fractionation to form massive Ni-Cu and massive Cu-rich ores. The second pulse of magma after removal of sulfides (<0.02%) experienced more crustal contamination and re attained S saturation. This new S-saturated and phenocryst-laden magma intruded the earlier formed massive sulfide ores and norites and formed the disseminated sulfide ores.