SiO2-rich Melt Research Articles

Calcium isotopic compositions of eclogite melts and negligible modification during reaction with lithospheric mantle

Subduction-driven recycling of crustal materials plays a substantial role in the generation of geochemical and lithological heterogeneities in the Earth’s mantle. Because the distinct compositions and/or mineral-specific equilibrium isotopic fractionation effects are present in eclogitized crustal materials, stable isotopes of major elements that exert significant impacts on geochemical and petrologic properties of the mantle may serve as powerful tracers for linking the geochemical anomaly to lithological heterogeneity within the Earth’s interior. Here we present high-precision Ca isotope data, combined with previously reported Mg-Fe isotopes, for a suite of intraplate basalts with a large contribution from recycled crustal material in their mantle sources. The δ44/40Ca in these potassic basalts are consistently lower (0.66–0.78 ‰) than those of mid-ocean ridge basalt (MORBs; 0.85 ± 0.09 ‰), with an average of 0.72 ± 0.09 ‰. The co-variations of major elements with radiogenic isotopes (e.g., MgO-ɛNd) reflect the mixing of two distinct endmembers, one of which is the low-MgO primitive melts and the other is the lithospheric mantle. The low-MgO, SiO2-rich primitive melts are characterized by low CaO/Al2O3 (<0.4) and high Dy/Yb (>5). Remarkably, the primitive melts exhibit low δ44/40Ca (ca. 0.70 ‰)-δ26Mg (ca. –0.60 ‰), and elevated δ57Fe (ca. 0.30 ‰) relative to MORBs. These features are consistent with their origin as partial melts of eclogitic crustal materials. During ascending to the surface, these melts reacted with the surrounding lithospheric peridotite. This process resulted in a noticeable shift of the Mg-Fe isotopes towards the typical mantle values but very limited δ44/40Ca modification. These results suggest that Ca isotopes largely hold the signature of primary melts and are a promising tracer of eclogite in the mantle source of basalts. We further show that the combined isotopes of Ca-Mg-Fe, which are stoichiometrically incorporated into mantle minerals, offer substantial potential to establish the links between lithological and geochemical heterogeneity in the mantle.

Sulfide-sulfate metasomatism and nickel release in the suprasubduction mantle

Metasomatic replacement of olivine by orthopyroxene under the influence of SiO2-rich melts is a widespread process in mantle conditions responsible for the formation of pyroxenite from mantle peridotite. However, the behavior of nickel in this process remains unclear. The reaction of Ni-containing olivine with silica- and sulfur-rich metasomatic agents in the suprasubduction mantle may release nickel from olivine and subsequently form a rock containing orthopyroxene and nickeliferous sulfides. Xenoliths of mantle wedge harzburgite from the Shiveluch volcano (Kamchatka continental arc) contain abundant globules of Ni-rich MSS, pentlandite and smaller amount of copper sulfides, totaling up to 0.75 wt.% (∼2600 ppm S2–), in association with Ca-sulfates (up to 7900 ppm S6+). The process accounting for the observed mineral association can be described via simplified reactions (the coefficients are approximate): (Mg,Fe,Ni)2SiO4 + 2xH+ +SiO2 = 2(Mg,Fe)SiO3 + xNi2+ + xH2O; Ni2+ + S2– = NiS. The influx of both oxidized and reduced sulfur may account for the coexistence of sulfide and sulfate phases. The presence of low-Ti chrome spinel with high Fe(II)/Fe(III) ratios, low Al, Ca and Ti in olivine, and no evidence of deserpentinization confirms the mantle origin of the sulfide/sulfate-bearing xenoliths. The mean isotopic composition of sulfur in the sulfide-sulfate assemblage, δ34S = +4.5‰, supports the contribution of slab-derived sulfur. The accumulations of Ni-(Cu)-rich sulfides and Ca-sulfate in metasomatized peridotites can serve as an intermediate host for sulfur and chalcophile metals, thereby generating magmas containing sulfur at the level of sulfide saturation and simultaneously enriched in Ni and Cu during the subsequent partial melting of the mantle.

Field and petrological characteristics of PGE-rich podiform chromitites: Implications for efficient exploration

Two distinct types of podiform chromitites have been extensively described within the mantle sections of Neoproterozoic ophiolites in Saudi Arabia and Phanerozoic ophiolites in Oman. The first type is concordant with the foliation plane of the harzburgite host, generally lensoidal in shape and is characterized by its low platinum-group elements (PGE) contents. The second type occurs as discordant or dike-like bodies and predominantly rich in PGE. In addition to their mode of occurrence, several lines of evidence can be employed to differentiate these two types of chromitites, both in the field and during laboratory investigations. Concordant PGE-poor chromitites are typically homogeneous in appearance (with regards to chromite grains distribution) and massive, featuring chromian spinel (Cr-spinel) with olivine as the dominant silicate inclusions. In contrast, discordant PGE-rich chromitites are often heterogeneous in appearance, exhibiting a semi-massive to disseminated, and are rich in both hydrous (amphiboles) and anhydrous (clinopyroxene) silicate inclusions. Discordant dunite veins, sharing similar petrological characteristics, are consistently associated with PGE-rich chromitites, with Cr-spinel containing a notable abundance of clinopyroxene inclusions. Compared to podiform chromitites found worldwide, concordant chromitites have lower to intermediate PGE contents (typically < 500 ppb), display a gently sloping negative trend in PGE patterns, contain very few platinum-group mineral (PGM) grains, and exhibit a wide compositional range of Cr-spinel (Cr# = Cr/(Cr + Al) atomic ratio, varying from 0.5 to 0.9). In contrast, discordant chromitites have higher PGE contents (commonly > 500 ppb), show a strong decoupling between the IPGE (Os, Ir, Ru) and PPGE (Rh, Pt, Pd) subgroups of the PGE distribution patterns, contain numerous PGM grains, and consist of high-Cr varieties (Cr# mostly > 0.6). In-situ analysis of clinopyroxene inclusions within Cr-spinel of discordant chromitites and associated dunite veins revealed characteristic U-shaped rare-earth elements (REE) patterns. This suggests the involvement of a hydrous SiO2-rich melt in the formation of PGE-rich chromitites and their associated discordant dunites. The wide range of Os-isotope composition of the Os-rich PGM, spanning from sub-chondritic to supra-chondritic 187Os/188Os values in the PGE-rich chromitites serves as further evidence for the heterogeneous hydrous melting process that led to the formation of these chromitites. These features, encompassing field and petrological characteristics, can be attributed to variations in the tectonic setting during the formation of podiform chromitites and can serve as valuable guidelines for PGE exploration.

Seismic signals induced by the Metasomatism of mantle wedge by siliceous melts: Insights from the elasticity of orthopyroxene at high pressure and temperature

Large amounts of silica held by slabs during subduction prefer to concentrate into the eclogite melts and further lead to the metasomatism of the overlying mantle wedge. However, the metasomatism process mediated by siliceous melts in the mantle wedge and the seismic signals associated remain poorly constrained. Peridotite in the mantle wedge is sufficiently silica undersaturated. Therefore, SiO2-rich melts would react with olivine in the peridotite to form orthopyroxene (opx). The local enrichment of opx has also been proposed to account for the observed low VP/VS ratio in the mantle wedges of a few subduction zones, revealing the metasomatism by silica saturated melts there. But the anomalously low VP/VS regions in a few subduction zones are inconsistent with the widespread metasomatism of the mantle wedge. In this study, we determined the elasticity of Fe-bearing opx (Mg0.875Fe0.125)SiO3 at high pressure and temperature (P-T) conditions. The VP/VS ratio of opx has a strong dependence on the pressure and temperature conditions, and only exhibits low VP/VS at low pressure and high temperature. Although the enrichment of opx may be widespread, it can only produce a discernible low VP/VS ratio at subduction zones with proper thermal conditions. Together with the elasticities of other major minerals in the upper mantle, we investigated the effect of opx enrichment on VP/VS ratio at mantle wedge conditions. The VP/VS ratio linearly decreases with the opx volume fraction (Ω), and the relationship can be described as VP/VS = 1.772 − 0.142Ω. Combining our high-quality mineral elasticity data with seismic tomography and thermal structure, we constrained the mineral proportions in the northern wedge of the Alaska subduction zone. The opx enriched area coincides well with the anomalously low VP/VS domain, indicating the strong metasomatism by siliceous melts in this mantle wedge.

Multi-stage garnet formation and destruction in Kimberley harzburgitic xenoliths, South Africa

Thirty-nine garnet harzburgites from Kimberley in the Kaapvaal Craton (South Africa) were studied to constrain the origin, age and evolution of sub-cratonic lithospheric mantle (SCLM). In order to avoid chemical overprinting by recent metasomatism, only garnet harzburgites that appeared clinopyroxene-free to the naked eye were sampled. The majority of garnets were, however, in equilibrium with clinopyroxene (24 of 39). Whole rock and mineral major–trace element geochemistry and garnet Sr–Nd–Hf isotope data are presented. Equilibration pressures range from 3.8–6.1 GPa, indicating the harzburgites were derived from a large portion of the SCLM (~115–185 km). High olivine Mg# (~93.4, n = 39) and low whole rock heavy rare earth elements (HREE) contents are consistent with large degrees of partial melting (>45%) and garnet exhaustion leaving a dunitic residue with olivine ≥90%, orthopyroxene ≤10% and HREE <0.01 times chondrite. Mineral modes, whole rock Al2O3 (0.5–3.2 wt%) and SiO2 (43.1–49.1 wt%), however, indicate heterogeneous re-introduction of garnet (≤13%) and orthopyroxene (≤50%). Harzburgites with high garnet and relatively low orthopyroxene modes (mostly ~7–13% and ~ 9–30%; n = 6) are characterised by mildly sinusoidal garnet REE patterns (Tb-Dy minimum and high HREE) and Archaean depleted Hf TDM ages (2.7–3.3 Ga; εHfe: +190 to +709). In contrast, harzburgites with high orthopyroxene and relatively low garnet and modes (~1.5–7.5% and ~ 25–50%; n = 19) are characterised by highly sinuous REE patterns (Ho-Yb minimum and low HREE) and Proterozoic enriched Hf TDM ages (0.7–1.6 Ga; εHfe: −16 to +6). It is inferred that Archaean G10 garnet re-introduction caused a significant increase in HREE, making melt depletion models based on HREE inaccurate. Orthopyroxene addition, a few hundred million years later, most likely at ~2.7 Ga and associated with Ventersdorp magmatic activity, caused partial consumption of garnet and olivine, and changed garnet compositions leading to: 1) Cr/Al ratio increase; 2) HREE decrease; 3) more sinusoidal REE patterns; and 4) un-radiogenic 176Hf/177Hf. Garnets define a Lu-Hf isochron age of 2702 ± 64 Ma (εHfi = +44, n = 31), which is interpreted as a consequence of partial isotopic equilibrium within the SCLM and mixing of the garnet- and orthopyroxene-rich metasomatic components. The low LILE contents and absence of Nb-Ta anomalies are consistent with modal metasomatism caused by intra-plate magmatism. In addition, the REE signatures of metasomatic agents in equilibrium with the garnets suggest that carbonatitic melts and SiO2-rich hydrous melts were responsible for re-introduction of garnet and orthopyroxene, respectively. Sr-Nd isotope systematics were disrupted associated with kimberlite magmatism (Nd isochron: 217 ± 58 Ma, εNdi = +4; n = 34), consistent with recent G10 garnet transformation into G9 garnets (Ca + Fe-enriched). This event may have caused garnet addition (up to 1%), suggesting that garnet was formed or destroyed in at least 4 different events: i) initial extensive polybaric melting, ii) asthenospheric melts re-introducing the bulk of the garnet, iii) orthopyroxene addition and garnet loss, all in the Archaean, and iv) minor garnet addition possibly related to recent kimberlite magmatism prior to eruption.

Partial melting of a depleted peridotite metasomatized by a MORB-derived hydrous silicate melt – Implications for subduction zone magmatism

Recent geodynamic models and geothermometers suggest that slabs in intermediate to hot subduction zone cross the water-saturated basalt solidus, indicating that hydrous silicate melts are important agents of mass transfer from slab to mantle wedge beneath arcs. Yet the effects of basaltic crust-derived hydrous melt fluxing on mantle wedge melting are poorly known. Here we present the melting phase relations of a depleted peridotite + a MORB-derived hydrous silicate melt at a melt:rock mass ratio of 0.1 and 0.05 (3.5 and 1.7 wt.% H2O, respectively) to simulate fluid-present partial melting of a depleted peridotite, which has been metasomatized by a hydrous silicate melt derived from subducting basaltic crust. Experiments were performed at 2–3 GPa and 900–1250 °C in a piston cylinder, using Au and Au75Pd25 capsules. Amphibole (7–10 wt.%) is stable up to 1000 °C at 2 and 3 GPa coexisting with an assemblage dominated by olivine and opx and with minor fractions of cpx and garnet at 3 GPa. The apparent fluid-saturated solidus of our bulk composition is located at 1000–1050 °C, coinciding with the exhaustion of amphibole at 2 and 3 GPa. Amphibole is exhausted between 0 and 5 wt.% melting at 2 and 3 GPa and dominates the melting reactions in this melting interval along with opx, generating SiO2 and Al2O3-rich, and FeO*- and MgO-poor primitive andesites under fluid-saturated conditions. The melting reactions during low-degree, fluid-saturated melting are incongruent, consuming opx and producing olivine + SiO2-rich melts and is observed over a wide range of starting compositions and pressures from this study and others. As extent of melting increases and the free fluid phase is consumed, a spectrum of basaltic andesites to basanites are produced. Comparison of experimental partial melts from this and other hydrous peridotite melting studies with natural primitive arc magmas suggests that melting of peridotites with varying bulk compositions but with 2.5 – 4.2 wt.% H2O can reproduce the major oxide spread and trends of primitive arc magmas globally. From this comparison, it is clear that differences solely in the pressure of hydrous mantle melting, where the partial melts are fluid-under saturated, can account for the first order trends observed in experimental and natural data, with differences in temperature and composition contributing to the compositional spread within these trends. The ubiquity of andesite genesis over a wide range of pressures and bulk compositions during aqueous fluid-saturated melting suggests that the relative rarity of primitive andesitic melt flux through the crust could be related to the fact that such melts are only produced at the base of the mantle wedge where temperatures are relatively low. As fluid-saturated andesitic melts ascend into the hotter core of the mantle wedge, they are likely consumed by higher-degree, fluid-undersaturated melting generating more common hydrous basaltic melts.

Tectono-magmatic Interplay and Related Metasomatism in Gabbros of the Chenaillet Ophiolite (Western Alps)

Abstract The Jurassic Chenaillet ophiolite in the Western Alps consists of a gabbro–mantle association exhumed to the seafloor through detachment faulting and partly covered by basaltic lavas. One of the Chenaillet gabbroic bodies includes mylonites that are transected by a network of felsic veins, thereby testifying to the interplay of ductile shearing and magma emplacement. The deformed gabbros preserve clinopyroxene porphyroclasts of primary magmatic origin, which are typically mantled by amphibole (titanian edenite) and minor secondary clinopyroxene. Titanian edenite and secondary clinopyroxene also occur as fine-grained syn-kinematic phases locally associated with fine-grained plagioclase. The felsic veins are made up of anorthite-poor plagioclase and minor titanian edenite. Geothermometric investigations document that the ductile gabbro deformation and the crystallization of the felsic veins occurred at 765 ± 50 °C and 800 ± 55 °C, respectively. With respect to undeformed counterparts, the deformed gabbros are variably enriched in SiO2 and variably depleted in Mg/(Mg + Fetot2+) and Ca/(Ca + Na). In addition, the deformed gabbros show relatively high concentrations of incompatible trace elements such as rare earth elements (REE), Y, Zr and Nb. The felsic veins are characterized by low Mg/(Mg + Fetot2+) and Ca/(Ca + Na), high SiO2 and high concentrations of incompatible trace elements. Relict clinopyroxene porphyroclasts from the deformed gabbros display a rather primitive, mid-ocean ridge-type geochemical signature, which contrasts with the trace element fingerprint of titanian edenite from both the deformed gabbros and the felsic veins. For instance, titanian edenite typically has relatively high REE abundances, with chondrite-normalized REE patterns characterized by a pronounced negative Eu anomaly. A similar trace element signature is shown by secondary clinopyroxene from the deformed gabbros. Amphibole from both the deformed gabbros and the felsic veins displays high F/Cl values. We show that the SiO2-rich hydrous melts feeding the felsic veins were involved in the high-temperature gabbro deformation and that melt–gabbro reactions led to major and trace element metasomatism of the deforming gabbros.

High-temperature structural change and microtexture formation of sillimanite and its phase relation with mullite

Abstract Synchrotron powder X-ray diffraction (XRD) experiments and transmission electron microscopy (TEM) observations of heat-treated sillimanite at various pressures were conducted to clarify the detailed phase relation between sillimanite and mullite. Under TEM, heat-treated sillimanite frequently showed anti-phase boundary (APB)-like textures with a displacement vector of ½[001]sil. Additional scanning TEM energy-dispersive X-ray spectroscopy analysis of regions with APB-like texture showed that they were clearly enriched in Al and accompanied by very fine, Si-rich glass inclusions, which indicates that the APB-like textures are composed of fine mullite. Moreover, synchrotron XRD patterns of these samples clearly showed double peaks of newly formed mullite and remnant sillimanite, indicating that the compositional transformation from sillimanite to mullite and glass is discontinuous. We separately determined the cell parameters of the sillimanite and mullite from the XRD pattern and found that the b axial length of the sillimanite increased with the treatment temperature, reflecting disordering of tetrahedral Al and Si in the sillimanite. In contrast, the positions of the deconvoluted mullite peaks indicated that the a axial length of mullite decreased as experimental pressure increased, owing to enrichment of the Si component. By projecting the cell parameters onto the a–b axial plane, the detailed changes in the crystallographic state of the sillimanite and mullite could be easily and comprehensively identified. On the basis of our results, we propose a new P-T diagram for the Al2SiO5 system that shows the transformation boundary between sillimanite and mullite + SiO2-rich melt and the contour of the Al/Si order parameter of sillimanite.

Crustal structure of the Dabie orogenic belt (eastern China) inferred from gravity and magnetic data

In order to better characterize the crustal structure of the Dabie orogen and its tectonic history, we present a crustal structure along a 500km long profile across the Dabie orogenic belt using various data processing and interpretation of the gravity and magnetic data. Source depth estimations from the spectral analysis by continuous wavelet transform (CWT) provide better constraints for constructing the initial density model. The calculated gravity effects from the initial model show great discrepancy with the observed data, especially at the center of the profile. More practical factors are then incorporated into the gravity modeling. First, we add a high density body right beneath the high pressure metamorphic (HPM) and ultrahigh pressure metamorphic (UHPM) belt considering the exposed HPM and UHPM rocks in the mid of our profile. Then, the anomalous bodies A, B, and C inferred from the CWT-based spectral analysis results are fixed in the model geometry. In the final crustal density structure, two anomalous bodies B and C with high density and low magnetization could possibly be attributed to metasomatised mantle materials by SiO2-rich melt derived from the foundering subducted mafic lower crust. Under the extensional environment in the early Cretaceous, the upwelling metasomatised mantle was partially melted to produce the parental magma of the post-collisional mafic-ultramafic intrusive rocks. As for the low density body A with strong magnetization located in the lower crust right beneath the HP and UHP metamorphic belt, it is more likely to be composed of serpentinized mantle peridotite (SMP). This serpentinized mantle peridotite body (SMPB) represents the emplacement of mantle-derived peridotites in the crust, accompanying the exhumation of the UHP metamorphic rocks.

Study of diffusion and local structure of sodium-silicate liquid: the molecular dynamic simulation

A systematic analysis on sodium-silicate melt with various silica contents was carried out. The simulation revealed two diffusion mechanisms occurred in the melt: the bond-breaking and hopping between sites. The local structure was analyzed through T-simplexes. It was revealed that T-clusters have a non-spherical shape and represent the diffusion channel, in which Na atoms are dominant, but no any O atoms are located. The SiO2-poor melt acquires a long channel. In contrast, the SiO2-rich melt consists of unconnected short channels. The simulation also revealed the immobile and mobile regions which differ in local structure and constituent composition. We propose a new CL-function to characterizing the spatial distribution of different atom component. The spatial distribution of mobile and immobile atoms is found quite different. In particular, the immobile atoms are concentrated in high-density regions possessing very large density of immobile atoms. The spatial distribution of mobile atoms in contrast is more homogeneous.

The jotunite of the Korosten AMCG complex (Ukrainian shield): Crust- or mantle-derived?

The Korosten Plutonic Complex (KPC) of Palaeoproterozoic age is made up of an AMCG suite of rocks. Jotunitic/ferromonzodioritic magmas have been emplaced at the margins or within massive anorthosites and in hypabyssal dykes coeval with the AMCG plutonism. Major and trace element geochemistry shows a calc-alkalic to alkali calcic trend with large variations of composition with SiO2 ranging from 44% to 62% and Mg# from 19 to 36, high concentrations of Fe, Ti and P, up to 20% Fe2O3t, 4% TiO2 and 2% P2O5. The Korosten jotunitic sequence of rocks matches the jotunite liquid line of descent that has been defined in the Rogaland anorthosite province. Both trends are compared to experimental data in dry conditions and medium to high pressure on the evolution of olivine tholeiite melt. It is shown that the initial step of the jotunite evolution characterized by a decrease in SiO2 correlated with increases in Fe and P is not reproduced in the experiments at medium pressure. At high pressure such evolution is possible but cannot yield SiO2 rich melts with further fractionation because of the existence of a thermal barrier. Derivation of jotunite melts from mantle-derived olivine tholeiite would be possible in a polybaric evolution of the melt. However, the polybaric evolution of the anorthosite mush with extraction of the interstitial jotunite melt to form the mafic marginal intrusion is not supported by field evidence of filter press mechanism. Derivation of the jotunite from melting of gabbroic rocks in the crust and their evolution at medium pressure appears as a more convincing mechanism. The ages of formation of the Korosten and Rogaland anorthosites are separated by 800Myr. It is speculated that the secular cooling of the earth has modified the generation process of the anorthosite. In the KPC a hot zone developed in the crust due to an asthenosphere uprise concomitant with the development of Ni-rich tholeiitic dykes, and in Rogaland, the cooler and more rigid crust was able to subduct and form crustal tongues that were potential source of melts.

On progress and rate of the peritectic reaction Fo + SiO2 → En in natural andesitic arc magmas

Using a high resolution ion microprobe with SCAPS imaging, the peritectic reaction of forsterite+silica to enstatite was studied down to submicron level in a natural andesitic tephra from the Central Plateau of North Island, New Zealand. The fayalitic component of natural olivines is stable in high-silica melts, and therefore the reaction is in fact a two-step progress: 1. Dissolution of Mg-rich olivine, rate-limited by Fe–Mg interdiffusion at the crystal rim, results in enrichment of Fe in the crystal rim and of Mg in the c. 1μm wide melt boundary layer around the crystal. 2. Magnesian pyroxenes preferentially but not exclusively nucleate in the melt boundary layer and grow; as soon as these microlites touch the rim of the dissolving olivine, they shield the crystals from the silica-rich melt, thereby preventing further olivine dissolution. At this point, Fe–Mg interdiffusion begins to destroy the Fe-enrichment of the olivine rim. The reaction is completed when the dissolving olivine crystal is completely mantled by magnesian pyroxene microlites. Thick pyroxene mantles are likely the result of pyroxene overgrowth rather than due to peritectic transformation. The morphology of the olivine rim preserves information about the reaction history of the grain. Modeling of Fe–Mg interdiffusion in the olivine rim following its shielding from the melt by pyroxene overgrowth may yield the rates of olivine dissolution and the rates of pyroxene growth if temperature is known. For the tephra we have studied, microlite thermometry yields a temperature of 1137 (±41)°C, indicating an olivine dissolution rate of 3–6×10−11ms−1 and an initially volumetric pyroxene growth rate of 2.2–5.6×10−21m3s−1. This points to timescales between olivine crystal uptake into the SiO2-rich melt and explosive eruption at the surface of a few hours to at most a day.

The magmatic–hydrothermal transition in the lower oceanic crust: Clues from the Ligurian ophiolites, Italy

The gabbroic bodies from the Jurassic Ligurian ophiolites are structurally and compositionally similar to the gabbroic sequences from the oceanic core complexes of the Mid Atlantic Ridge. Initial cooling of the Ligurian gabbros is associated with local development of hornblende-bearing felsic dykes and hornblende vein networks. The hornblende veining is correlated with the widespread development of hornblende as coronas/pseudomorphs after the clinopyroxene in the host gabbros. In addition, the studied gabbroic body includes a mantle sliver locally containing hornblende gabbros and hornblendite veins. The hornblendes from the felsic dykes and the hornblende-rich rocks within the mantle sliver show a similar geochemical signature, characterized by low Mg#, CaO and Al2O3, negligible Cl, and high TiO2, K2O, REE, Y, Zr and Nb concentrations. The whole-rock Sm–Nd isotopic compositions of the felsic dykes and the hornblende-rich rocks define a Sm–Nd isochron corresponding to an age of 154±20Ma and an initial εNd of 9.2±0.5. The δ18O of the hornblendes and coexisting zircons from these rocks (about +4.5‰ and +5.8‰, respectively) do not indicate the presence of a seawater component in these melts. The formation of the felsic dykes and of the hornblende-rich rocks within the mantle sliver involved SiO2-rich silicate melts with negligible seawater component, which presumably were derived from high degree fractional crystallization of MOR-type basalts. The vein and the coronitic/pseudomorphic hornblendes show high Mg# and CaO, significant Cl (0.02–0.17wt%) and low TiO2 and K2O concentrations. The coronitic/pseudomorphic hornblendes have trace element compositions similar to those of the clinopyroxenes from the gabbros and δ18O values (+1.0‰ to 0.7‰) close to seawater, suggesting an origin by reaction between migrating seawater-derived fluids and the host gabbros. The vein hornblendes commonly show slight LREE enrichment, relatively high concentrations of Nb (up to 2.5ppm) and δ18O ranging from +3.7‰ to +0.8‰. The crystallization of these hornblendes most likely involved both seawater and magmatic components.

Melting and Phase Relations of Carbonated Eclogite at 9-21 GPa and the Petrogenesis of Alkali-Rich Melts in the Deep Mantle

The melting and phase relations of carbonated MORB eclogite have been investigated using the multi-anvil technique at 9–21 GPa and 1100–1900°C. The starting compositions were two synthetic mixes, GA1 and Volga, with the CO2 component added as CaCO3 (cc): GA1 + 10%cc (GA1cc) models altered oceanic crust recycled into the convecting mantle via subduction, and Volga + 10%cc (Volga-cc) models subducted oceanic crust that has lost some of its siliceous component in the sub-arc regime (GA1 minus 6·5 wt % SiO2). The subsolidus mineral assemblage at 9 and 13 GPa includes garnet, clinopyroxene, magnesite, aragonite, a high-pressure polymorph of TiO2 (only at 9 GPa) and stishovite (only at 13 GPa). At 17–21 GPa clinopyroxene is no longer stable; the mineral assemblage consists predominantly of garnet with subordinate magnesite (only at 17 GPa), Na-rich aragonite, stishovite, Ca-perovskite (mostly at 21 GPa), and K-hollandite (mostly at 17 GPa). Na-carbonate with an inferred composition (Na,K)2(Ca,Mg,Fe)(CO3)2 was present in Volga-cc at 21 GPa and 1200°C. Diamond (or graphite) crystallized in most runs in the GA1cc composition, but it was absent in experiments with the Volga-cc composition. In Volga-cc, the solidus temperatures are nearly constant between 1200 and 1300°C over the entire pressure range investigated. In GA1cc, the solidus is located at similar temperatures at 9–13 GPa, but at higher temperatures of 1300–1500°C at 17–21 GPa. The difference in solidi between the GA1cc and Volga-cc compositions can be explained by a change in Na compatibility between 13 and 17 GPa as omphacitic clinopyroxene disappears, resulting in the formation of Na-carbonate or Na-rich melt in Volga-cc. The solidus temperature in GA1cc also increases with increasing pressure as a consequence of carbonate reduction and diamond precipitation, possibly brought on either via progressive Fe2+–Fe3+ transition in garnet at higher pressures or by a decrease of the activity of the diopside component in clinopyroxene. The low-degree melts are highly alkalic (K-rich at 9–13 GPa and Na-rich at 17–21 GPa) carbonatites, changing towards SiO2-rich melts with increasing temperature at constant pressure. The solidi of both compositions remain higher than typical subduction pressure–temperature (P–T) profiles at 5–10 GPa; however, at higher pressures the flat solidus curve of carbonated eclogite may intersect the subduction P–T profile in the Transition Zone, where carbonated eclogite can produce alkali- and carbonate-rich melts. Such subduction-related alkali-rich melts can be potential analogues of kimberlite and carbonatite melt compositions and important agents of mantle metasomatism and diamond formation in the Transition Zone and in cratonic roots. Melting of carbonated eclogite produces a garnet-bearing refractory residue, which could be stored in the Transition Zone or lower mantle.

Oceanic Material Recycled within the Sub-Patagonian Lithospheric Mantle (Cerro del Fraile, Argentina)

A detailed petrological study of mafic and ultramafic xenoliths from Cerro del Fraile (Southern Patagonia, Argentina) was developed to highlight (1) the mineralogical and geochemical composition of the lithospheric mantle beneath the area, (2) the nature of the metasomatizing agents that infiltrate the mantle wedge above the Antarctic plate, (3) the processes that allow the mantle to be refertilized, and (4) the nature of the material dragged down in the subduction zone and recycled within the South Patagonian sub-arc mantle. Major and trace element analyses of clinopyroxene and orthopyroxene in peridotitic and pyroxenitic rocks suggest that a proto-adakite, deriving from the melting of the subducting Antarctic plate, was responsible for the metasomatic features of the peridotitic rocks and the crystallization of the pyroxenites. A few composite xenoliths bridge the two processescperidotite enrichment and pyroxenite crystallizationcindicating that the variously depleted mantle reacts with the incoming melt to generate a newly fertile mantle domain. This reaction occurs under reducing conditions [log fO2 (QFM) ^1·21 to ^0·34], unusual for mantle wedge settings. Heavy rare earth elements and Al2O3 and MgO contents in pyroxenes indicate a partial melting degree varying from 10 to 25%. The peculiar enrichment in Zr (^Hf), Th and U of the pyroxenes indicates the melting of oceanic sediments, which include manganese nodules and, possibly, organic matter, in agreement with the estimated low oxygen fugacity conditions. Some geochemical analogies have been found between the calculated metasomatic melts and the Austral Volcanic Zone adakites. In this case, the amount of sediment involved in the genesis of the infiltrating melts is larger than that previously proposed for the genesis of the erupted Patagonian adakites. Chemical^physical conditions favouring the upward percolation through the mantle wedge of these SiO2-rich and viscous melts are also discussed.

Trace Elements in the Si Furnace. Part I: Behavior of Impurities in Quartz During Reduction

Quartz and carbonaceous materials, which are used in the production of silicon as well as electrodes and refractories in the silicon furnace, contain trace elements mostly in the form of oxides. These oxides can be reduced to gaseous compounds and leave the furnace or stay in the reaction products—metal and slag. This article examines the behavior of trace elements in hydrothermal quartz and quartzite in the reaction of SiO2 with Si or SiC. Mixtures of SiO2 (quartz or quartzite), SiC, and Si in forms of lumps or pellets were heated to 1923 K and 2123 K (1650°C and 1850°C) in high purity graphite crucibles under Argon gas flow. The gaseous compounds condensed in the inner lining of the tube attached to the crucible. The phases present in the reacted charge and the collected condensates were studied quantitatively by X-ray diffraction (XRD) and qualitatively by Electron Probe Micro Analyzer (EPMA). Contaminants in the charge materials, reacted charge and condensate were analyzed by Inductively Coupled Plasma-Mass Spectroscopy (ICP-MS). Muscovite in the mineral phase of quartz melted and formed two immiscible liquid phases: an Al-rich melt at the core of the mineral, and a SiO2-rich melt at the mineral boundaries. B, Mn, and Pb in quartz were removed during heating in reducing atmosphere at temperature above 1923 K (1650°C). Mn, Fe, Al and B diffused from quartz into silicon. P concentration was under the detection limit. Quartzite and hydrothermal quartz had different initial impurity levels: quartzite remained more impure after reduction experiment but approached purity of hydrothermal quartz upon silica reduction.

Textures and geochemistry of the Saramta peridotites (Siberian craton): Melting and refertilization during early evolution of the continental lithospheric mantle

The Saramta peridotite massif is located within the Sharyzhalgai complex, SW margin of the Siberian craton. The Saramta massif was formed in the Archean and then juxtaposed with granulites of crystalline basement of the Siberian craton. The Saramta harzburgites are highly refractory in terms of lack of residual clinopyroxene, olivine Mg-number (up to 0.937), and spinel Cr-number (∼0.5), suggesting high degree of partial melting. Detailed study of their microstructures shows that they have extensively reacted with a SiO2-rich melt, leading to the crystallization of orthopyroxene, clinopyroxene, amphibole and spinel at the expense of olivine. The major element compositions of the least reacted harzburgites are similar to the residues of refractory peridotites produced by the fractional melting (initial melting pressures >3GPa and melt fractions ∼40%). Moreover, non-residual clinopyroxenes are highly depleted in Yb, Zr and Ti, but highly enriched in LREE. A two-stage history is proposed for the Saramta peridotite: (1) primitive mantle underwent depletion in the garnet stability field followed by melting in the spinel stability field; (2) refractory harzburgites underwent refertilization by SiO2-rich melt in supra-subduction zone. Rare Saramta lherzolites probably formed from more refractory harzburgites as a result of such a melt–rock reaction. The Saramta peridotites are similar to low-T coarse-grained peridotites of subcratonic mantle. Processes of their formation, as reflected by textures and composition of minerals of the Saramta peridotites, are characteristic of the early stages of subcratonic mantle formation.

Textures of Peritectic Crystals as Guides to Reactive Minerals in Magmatic Systems: New Insights from Melting Experiments

Peritectic crystals in igneous rocks may be derived from either the source or country rocks, or may have formed by reactive assimilation of source-inherited solids, primary magmatic minerals during self- or magma mixing, or country-rock xenoliths or xenocrysts. Identifying such peritectic crystals is important for constraining the components and textures of igneous rocks and the underlying processes of magmatic evolution. In this study we demonstrate that peritectic olivine formed in melting experiments crystallizes as clusters of euhedral to subhedral crystals. Olivine replacing orthopyroxene, amphibole, and phlogopite forms crystal clusters with distinct crystal to melt ratios, 2D surface area, grain boundary segmentation, and inclusion relations. In our experiments the textures of peritectic crystals are primarily controlled by the stability temperature and decomposition rate of reactive minerals. High-temperature minerals such as orthopyroxene slowly decompose to form high-density clusters of large crystals with long grain boundary segments. The SiO2-rich peritectic melt produced favours formation of melt inclusions. Low-temperature minerals such as amphibole and phlogopite rapidly decompose to form low-density clusters of small crystals with short grain boundary segments. The relatively SiO2-poor peritectic melt produced results in the formation of fewer melt inclusions. Host melt composition has a minor effect on the textures of peritectic olivine formed in the melting experiments of this study and previous contamination experiments, but affects the assemblages of the peritectic crystal clusters. Cluster density and 2D surface area of peritectic olivine tend to decrease, whereas grain boundary segment length increases with increasing experimental temperature and H2O content. Using textural criteria that distinguish olivine formed after different minerals in our melting experiments, we hypothesize that two olivine populations from a basaltic^andesitic lava flow of the Tatara^San Pedro volcanic complex, Chile, may be peritectic crystals formed after amphibole and orthopyroxene. Both amphibole and orthopyroxene are common in xenoliths preserved in some Tatara^San Pedro lava flows. One notable difference between the experimental and natural olivine crystals is that the natural olivine crystals have 2D surface areas and 2D grain boundary segments up to � 1000 and � 100 times larger, respectively, than those produced in our experiments. We propose that this size difference is primarily controlled by comparatively slow heating and decomposition of reactive crystals and textural coarsening of peritectic crystals during prolonged magma residence in the natural system.

Depressurized Cavities within High-strain Shear Zones: their Role in the Segregation and Flow of SiO2-rich Melt in Feldspar-dominated Rocks

We observe void growth and coalescence into cavity-bearing shear bands during deformation of wet synthetic anorthite aggregates containing53 vol. % silica-enriched melt. Samples were deformed in the Newtonian creep regime to high strain during torsion experiments at 11008C and 400MPa confining pressure. Localized cavity-bearing shear bands show an S^C’-geometry: the bands (C’) are oriented at about 308 to the compression direction of the imposed simple shear and the internal foliation (S) of the bands is rotated towards the horizontal external shear plane. Cavity-bearing shear bands started to nucleate in the sample periphery above a shear strain threshold of 2. Quartz crystallized from the water-saturated SiO2-rich melt within large cavities inside these bands, which requires that the melt is decompressed by4200MPa during their formation. The dynamically evolving cavities are sites of locally reduced pressure that collect the melt distributed in the adjacent matrix.Therefore, cavitation damage under ductile conditions may result in the development of an efficient melt channelling system controlling SiO2-rich melt flow in the lower crust. Electron backscatter diffraction analysis shows that the quartz inside the cavity bands has a crystallographic preferred orientation (CPO). The development of the CPO is explained by the preferred dissolution of crystals oriented with the rhombohedra and trigonal dipyramids orthogonal to the compression direction and by preferential growth of crystals aligned with their500014axis in the extension direction of the externally applied simple shear deformation.

Critical cooling rates for glass formation in levitated Mg2SiO4-MgSiO3 chondrule melts

Using a gas-jet levitator, Mg-silicate glasses containing forsterite (Mg2SiO4) or enstatite (MgSiO3) are obtained, with a particle diameter of 2 mm. The levitator aids in floating the melt droplets using a gas-jet. It can transform the levitated droplets into crystals or glass during cooling. The critical cooling rate for glass formation (Rc) for levitated Mg-silicate melts («10-1-102 K/s) is considerably lower than that for non-levitated melts (>103 K/s) because of the inhibition of crystallization by the sample holder in the former. Furthermore, the crystallization temperature of the levitated Mg-silicate melts is very low (∼ 900 °C). Chondrules in chondrites are assumed to have been formed from the melt droplets in such a levitation environment. Several researchers have concluded that Rc for a chondrule melt held by a platinum wire is equal to the upper limit of the cooling rate for the chondrule melt, estimated to be ∼ 1000 K/h. However, in a levitation environment, SiO2-rich melts do not crystallize at this cooling rate. We conclude that the previously proposed maximum cooling rate for chondrule melts has been greatly overestimated. We infer that spontaneous crystallization of a completely melted chondrule precursor is extremely difficult, and almost all chondrules turn into glass, despite a low cooling rate. For crystallization of chondrule melts, it is necessary that the precursor of the chondrules should be partially melted or the completely melted precursor should be in contact with nebular dust.