Internally Heated Pressure Vessel Research Articles

Stability of hydrous basaltic melts at low water fugacity: evidence for widespread melting at the lithosphere-asthenosphere boundary

The upper mantle low velocity zone is often attributed to partial melting at the lithosphere-asthenosphere boundary. This implies that basaltic melts may be stable along plausible geotherms due to the freezing point depression in the presence of water and other incompatible impurities. However, the freezing point depression (ΔT) as a function of water content in the near-solidus basaltic melt (cH2O) cannot be precisely determined from peridotite melting experiments because of difficulties in recovering homogeneous basaltic glasses at high pressures. We therefore used an alternative approach to reinvestigate and accurately constrain the ΔT–cH2O relationship for basaltic melts at the low water fugacities that are expected in the upper mantle. Internally heated pressure vessel (IHPV) experiments were performed at water-saturated conditions in the anorthite-diopside-H2O system at confining pressures of 0.02 to 0.2 GPa and temperatures between 940 and 1450 ℃. We determined the water-saturated solidus, and obtained ΔT by combining our data with reports of dry melting temperatures in the anorthite-diopside system. In another series of experiments, we measured water solubility in haplobasaltic melts and extrapolated cH2O to pressures and temperatures of the water-saturated solidus. By combining the results from these two series of experiments, we showed that the effect of water on ΔT was previously underestimated by at least 50 ℃. The new ΔT–cH2O relationship was then used to revise predictions of melt distribution in the upper mantle. Hydrous melt is almost certainly stable beneath extensive regions of the oceanic lithosphere, and may be present in younger and water-enriched zones of the subcontinental mantle.

A method for the direct analysis of quenched, magmatic-hydrothermal fluids recovered from high-pressure, high-temperature experiments

A method for analyzing low-salinity fluids (up to ~11 wt% NaCl equivalent) that have been equilibrated with felsic melts in high-pressure, high-temperature experiments is reported. Experiments were performed at a pressure of 200 MPa and 800 °C in internally-heated pressure vessels, wherein magmatic fluids were quenched and recovered in a multi-step process for analyses of major (Na, K, Ca, Cl, and F) and trace (Li, Be, Mn, Cu, Zn, Rb, and Sr) elements by ion chromatography and inductively coupled plasma mass spectrometry (ICP-MS). The quenched melts were analyzed for major and trace elements by electron microprobe and laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS). Such analyses permitted the composition of the fluid to be directly measured without the need for mass balance calculations or trapping fluids within fluid inclusions while still enabling the precise determination of partition coefficients. Mass balance calculations tend to overestimate Na, K, Ca, and Be; underestimate Cu, Zn, and F; and are in agreement for Li, Mn, Rb, Sr, and Cl when compared to direct measurements. Additionally, a comparison of partition coefficients calculated from measured and mass balance derived concentrations to available literature data is consistent with mass balance calculations being less accurate than measured values. Thus, direct measurements of quenched fluids by ion chromatography and ICP-MS offer a robust method for characterizing the composition of magmatic-hydrothermal fluids that have been equilibrated with silicate melts.

Nitrogen solubility in basaltic silicate melt - Implications for degassing processes

The distribution of nitrogen between the different terrestrial reservoirs (core-mantle-atmosphere) and how this may have changed since the earliest planetary stages is uncertain. In particular, the primordial degassing processes of the magma ocean and its role in the formation of the atmosphere remains to be quantified. Since no geological samples can capture this early degassing process, we need to go through the thermodynamic modeling of the nitrogen solubility in silicate melt. We hence performed experiments on basaltic samples at fluid saturation in the C-H-O-N system, using an Internally Heated Pressure Vessel (IHPV) and Piston Cylinder (PC) in the pressure range 0.8 kbar to 10 kbar, temperature between 1200 and 1300 °C, and a wide range of fO2 conditions from IW + 4.9 to IW-4.7 (IW standing for the Iron-Wustite redox buffer). The nitrogen concentration in the quenched silicate melts at fluid saturation was analysed by secondary ion mass spectrometry (SIMS), and the speciation of the dissolved C-O-H species was determined by Fourier transform infrared spectroscopy (FTIR). We identified two nitrogen species in the silicate melt: N2 dominating at fO2 > IW and N3− at lower fO2. Using these data and a database constraining nitrogen concentration at fluid saturation from 1 bar to 10 kbar pressure, we calibrated a solubility law for nitrogen in basalts defining its P-T-fO2 dependences. This model expands the model of Libourel et al. (2003) to high pressure and higher C-O-H activities. It can be used to investigate the nitrogen degassing processes for different pressure, temperature and fO2 conditions relevant to planetary accretion and modern volcanism.

Retrieving dissolved H2O content from micro-Raman spectroscopy on nanolitized silicic glasses: Application to volcanic products of the Paraná Magmatic Province, Brazil

Determining volatile content in silicate glasses has critical importance to evaluate the original water content of magmas and its evolution from storage to eruption. Here we present a Raman spectroscopic study devoted to defining an optimized analytical procedure for the determination of dissolved water content of nano-crystallized ancient silicic glasses, belonging to the Paraná Magmatic Province (PMP) volcanism in South Brazil (131 to 134 Ma). A set of anhydrous glasses was prepared under atmospheric pressure by melting and quenching of natural rocks. These glasses were later hydrated under high pressure in an Internally Heated Pressure Vessel (IHPV). Even after rapid quenching, most of the dry and hydrous glasses showed prominent Raman peaks at ca. 670–690 cm−1, which are associated with the presence of Ti-magnetite nanolites. Although large Raman peaks are observed, the nanolites are invisible to standard Scanning Electron Microscopy (SEM) due to their small size and not quantifiable by X-ray diffraction (XRD) because of their low abundance. Available models and procedures to estimate water content in glasses are calibrated with crystal-free glasses and were proved inadequate for water measurement in nanolite-bearing glasses. Here we provide new calibration strategies to determine the water content in nanolite-bearing glasses. These strategies are based on the normalization of the intensity and area of the water band (3550 cm−1) to the individual vibrational bands of the silicate framework of the glass, at 350–500 cm−1 and 860–1240 cm−1. Our new calibrations were used to estimate the dissolved water content of glass matrixes of natural dacitic vitrophyres from the Caxias do Sul volcanic suite, whose Raman spectra show nanolite peaks of variable intensity. Our work constitutes a further step ahead toward the definition of strategies to determine the volatile content in natural multicomponent and multiphase glasses.

Magmatic Processes at the Oman Ophiolite Paleoridge – Perspectives on the Role of Water

AbstractThe Oman ophiolite is regarded as best proxy for accreted oceanic crust from typical fast‐spreading ridge systems on land. However, the Oman ophiolite is influenced by initial subduction zone initiation, and the nature of the details of the subduction zone setting is still under controversial debate. While a first magmatic phase shows features of magmatic accretion very similar to those known from the East Pacific Rise, except that the primary melts were slightly water‐enriched, a second type of magmatism is characterized by an apparent subduction‐zone related imprint, producing rocks like FAB basalts and boninites in the upper crust, as well as cross‐cutting gabbronorites and wehrlites in the deeper crust. In this paper, we apply diverse experimental studies in wet tholeiitic and peridotitic systems performed at lower pressures (100 to 500 MPa) in the experimental lab of the University Hannover, in order to constrain the details of the magmatic processes proceeded at the Oman ophiolite paleoridge during the Cretaceous, with special focus on the influence of water on the phase stabilities and phase relations. The experiments were performed in vertically oriented internally heated pressure vessels (IHPV) (see Berndt et al., 2002; Fig. 1). This facility uses as pressure medium mixtures of Ar and H2 in order to adjust the required fH2 in the vessel, enabling us to control the redox conditions. The fH2 prevailing in the IHPV at high P and T was measured with a Shaw‐membrane made of platinum. The overall variation in fO2 in all experimental series was in the range between ∼FMQ‐1 and ∼FMQ+3.2, thus covering the range of oxygen fugacities prevailing in natural MORB magmas (Bezos and Humler, 2005).For understanding the magmatic processes during the Oman ophiolite paleoridge accretion, transects through the lower (GT1) and middle (GT2) crust have been drilled in the frame of ICDP (International Continental Scientific Drilling Program). Drill sites have been selected in the Wadi Tayin massif, which is known that the influence of magmatic phase 2 characterized by subduction‐related primary melts is minimal. Details and progress obtained in the Oman Drilling Project (OmanDP) can be found here: (https://www.omandrilling.ac.uk/). Regarding the first magmatic phase of the processes at the Oman ophiolite paleoridge, a characteristic observation made during the description of the drilled cores GT1 (lower crust) and GT2 (mid‐curst) was that quite often layers in the layered gabbro series occur which show the presence of clinopyroxene joining olivine instead of plagioclase (under near liquidus conditions). In terms of lithologies this could be interpreted as presence of wehrlitic crystal mushes as early cumulates instead of troctolitic, which are the typical ones for primary magmatism at typical fast‐spreading ridges. This situation could be experimentally simulated by adding a moderate to high water activity to primitive MORB at pressures ≥ 200 MPa, resulting in a shift of the clinopyroxene‐in curve to higher temperatures above the plagioclase‐in curve (Feig et al. 2006; see Fig. 2).Regarding the second, late‐stage magmatic phase, the formation of typical Oman high‐Ca‐boninites could be experimentally simulated by water‐saturated partial melting of Oman harzburgite at 200 MPa and relatively low temperatures between 1100 and 1200°C. Depleted gabbronorites crosscutting layered gabbros of phase 1 magmatism can be regarded as cumulates formed in these boninitic melts. Late wehrlites crosscutting layered gabbro could be produced by accumulation of olivine and clinopyroxene at temperatures between 1040 and 1080°C in a hydrous gabbroic system at pressures > 100 MPa with bulk water content of 2–3 wt%.

Experimental study on the effects of H2O upon crystallization in the Lower and Critical Zones of the Bushveld Complex with an emphasis on chromitite formation

Mafic–ultramafic layered rocks of the Lower and Critical Zones of the Bushveld Complex are believed to have formed from a parental magma usually labelled B1 with SiO2 content at 55 wt.%, MgO at about 12 wt.% and Cr2O3 concentration at 0.15 wt.%. Previous experimental study at 300 MPa and nominally dry conditions showed that this composition had orthopyroxene on liquidus between 1250 and 1300 °C, and the mineral was joined by plagioclase and clinopyroxene at and below 1166 °C. The objective of this study was to investigate the effects of variable additions of H2O (2, 4 and 6 wt.% in starting compositions) on liquidus and sub-liquidus phase equilibria. Main attention was given to the solubility of Cr-spinel and Cr distribution between melt and crystalline phases. Six experiments were carried out in the internally heated pressure vessel (IHPV) at 300 MPa and in the temperature interval 1100–1300 °C. Redox conditions were not controlled and thermal dissociation of the H2O component followed by hydrogen diffusion through container walls resulted in a noticeable oxidation of run products. The effective values of fO2 in experiments were calculated using SPINMELT 2.0 software bases on the regression of experimental data on the spinel-melt equilibrium. According to the calculations, H2O additions in isothermal series of experiments increased fO2 by 0.8–1.6 log-units from the level in a nominally dry run. Overall, the experiments showed that H2O additions suppressed crystallization of silicates, shifted olivine-orthopyroxene peritectic equilibrium towards higher olivine stability (olivine first appears in H2O-bearing compositions at 1175 °C) and increased the proportion of magnetite component in spinel solid solution. The addition of 4 wt.% H2O to the starting composition at 1200 °C resulted in complete melting of orthopyroxene, which left Cr-spinel as the only liquidus phase. Equal amount of H2O at 1125 °C leads to complete melting of plagioclase, orthopyroxene and clinopyroxene, which leaves only olivine and spinel stable and increases the liquid mass fraction from 0.62 in the nominally dry compositions to 0.9. The results of this study appear to support the idea that chromitite layers in the Critical Zone of the Bushveld Complex were formed by episodes of hydration melting of silicate cumulates.

Features of basaltic melt-hydrogen interaction at hydrogen pressure 10–100 MPa and temperature 1100–1250 °C

New experimental data on interaction of basaltic melt with hydrogen at high temperature (1100–1250 °C) and hydrogen pressure (10–100 MPa) have been obtained to gain insight into the possible role of hydrogen in the processes occurring in basaltic melts in the earth's crust under reducing conditions (f(O2) = 10−13–10−15). The experiments were carried out in a unique in house built internally heated pressure vessel (IHPV). This apparatus is equipped with an internal device, which allows long-term experiments under hydrogen pressure at high temperatures without losing hydrogen. Two types of experiments were carried out: 1 - kinetic experiments with isobaric quenching, 2 - crystallization experiments with subsequent isobaric quenching. It was found that, despite the high reducing potential of the H2 - basaltic melt system, the hydrogen oxidation reactions and the Fe oxides reduction in the melt are not complete. Initially homogeneous basaltic melts become heterogeneous. H2O is formed in the fluid phase (initially consisted of pure hydrogen); H2O (0.34–0.9 wt%) also dissolves in the basaltic melts, and small metal blobs with an emulsion-like texture are formed in the melts at temperature significantly lower than the melting temperature of the metal phase (Fe). The texture and dimensions of the metal segregations resemble those reported from natural magmatic rocks. Some new features of the crystallization process of basaltic melt under very reducing conditions are also discussed. The assemblage of subliquidus minerals (olivine + clinopyroxene + plagioclase) formed in the crystallization experiment at T = 1100 °C, P(H2) = 10 MPa with run duration 2 h, and their chemical compositions match closely those described in natural lavas (e.g. Kamchatka volcanoes). This result can be considered as an experimental confirmation of the participation of hydrogen in the magmatic process, which also corroborates with the composition of volcanic gases.

Experimental Study of Amphibole Interaction with H2O–HCl Fluid at 650–750°C, 5–7 Kbar: Implications for High-Temperature Metasomatism of Metabasites

Interaction between amphibole and H2O–HCl fluid was experimentally studied in an internally heated pressure vessel (IHPV) at 650–800°C and 5–7 kbar. The stability of the anorthite–amphibole association was approved to be constrained at T = 650–800°C, (P = 5–7 kbar) within a range of HCl fugacity (fHCl) 50 100 µm) corundum and quartz crystals. Based on experimental data, it is demonstrated that interaction between metabasite with acidified fluid resulted in Ca, Fe, and no so much Mg removal from the metabasite (as inferred for rocks cropping out on Kii Island in the White Sea). The HCl fugacity in the metasomatizing fluid was approximately 50–200 bar at T = 650 – 800°C, P = 5–7 kbar. Rock depletion in bases led to that the residual matrix enriched in Al2O3 and SiO2, and this was favorable for the origin of corundum, on the one hand, and highly silicic rocks, like quartzite, on the other.

Interaction of highly saline fluid and olivine gabbro: Experimental simulation of deep hydrothermal processes involving amphibole at the base of the oceanic crust

Natural occurrences of Cl-bearing amphibole indicate crystallization in the presence of a highly saline fluid. Amphibole with varying Cl contents (0.1 to 5 wt% Cl) found in lower oceanic crustal rocks demonstrates the activity of saline hydrothermal fluids at depth. Here we present an experimental study done in cold seal pressure vessels (CSPV) and internally heated pressure vessels (IHPV) to illustrate the process by which gabbro-hosted amphibole-rich parageneses evolve in the presence of a hydrothermal saline fluid. The starting material used was a natural gabbroic rock containing high-Ti magnesio-hastingsite with the addition of a moderately to highly saline fluid (NaCl-H2O) with XNaCl = 0.02, 0.07 and 0.24 (6, 20 and 50 wt% NaCl). We evaluated a range of conditions from hydrothermal (500–750 °C) to magmatic (900 °C) at pressures of 200 MPa and fO2 close to NNO. Fluid/rock mass ratios used were 0.2 and 1 in subsolidus (hydrothermal) experiments, and 0.07 in partial melting (magmatic) experiments. New amphibole was formed on rims of the starting high-Ti magnesio-hastingsite, with product compositions corresponding to magnesio-hastingsite, high-Si magnesio-hornblende, tschermakite, edenite, hastingsite and ferro-pargasite with varying Cl contents up to 0.47 wt% Cl. Our results from subsolidus experiments demonstrate a decrease in olivine, plagioclase and clinopyroxene in the starting rock, and formation of new amphibole, with decreased IVAl and Ti with respect to starting amphibole. Product amphibole does not display any correlation between Cl and Fe2+, IVAl and K, in contrast to natural highly Cl-rich amphibole, suggesting that these correlations hold true only at Cl contents higher than those attained in our experiments. In addition, we found that increasing NaCl in the fluid correlates with increased maximum Cl contents in amphibole. Compositional variations found in product amphibole highlight the heterogeneities in fluid infiltration and Cl activity that account for the complexity of hydrothermal fluid/rock interactions in deep oceanic geological systems.

Crystallization of the lunar magma ocean and the primordial mantle-crust differentiation of the Moon

We present crystallization experiments on silicate melt compositions related to the lunar magma ocean (LMO) and its evolution with cooling. Our approach aims at constraining the primordial internal differentiation of the Moon into mantle and crust. We used graphite capsules in piston cylinder (1.35–0.80 GPa) and internally-heated pressure vessels (<0.50 GPa), over 1580–1020 °C, and produced melt compositions using a stepwise approach that reproduces fractional crystallization. Using our new experimental dataset, we define phase equilibria and equations predicting the saturation of liquidus phases, magma temperature, and crystal/melt partitioning for major elements relevant for the crystallization of the LMO. These empirical expressions are then used in a forward model that predicts the liquid line of descent and crystallization products of a 600 km-thick magma ocean. Our results show that the effects of changes in the bulk composition on the sequence of crystallization are minor. Our experiments also show the crystallization of a silica phase at ca. 1080 °C and we suggest that this phase might have contributed to the building of the lower anorthositic crust. Calculation of crustal thickness clearly shows that a thin crust similar to that revealed by GRAIL cannot have been generated through solidification of whole Moon magma ocean. We discuss the role of magma ocean depth, trapped liquid fraction (with implication for the alumina budget in the mantle and the crust), and the efficiency of plagioclase flotation in producing the thin crust. We also constrain the potential range of pyroxene compositions that could be incorporated into the crust and show that delayed crustal building during ca. 4% LMO crystallization on the nearside of the Moon may explain the dichotomy for Mg-number. Finally, we show that the LMO can produce magnesian anorthosites during the first stages of plagioclase crystallization.

DryCO2–COfluid as an important potential deep Earth solvent

AbstractTransport properties of reduced carbonic fluid have been studied experimentally at P = 2 kbar andT = 700–1000°C in internally heated pressure vessel (IHPV). Synthetic FeCO3and natural siderite were used to generate fluid during experiments using a platinum double‐capsule technique. A natural CaTiSiO5aggregate was placed into the inner capsule as an additional source of trace elements. The outer capsule was loaded with albite glass. No water was introduced to the system and oxygen fugacity was established near to graphite–oxygen (CCO) buffer due to transformation of FeCO3into a magnetite aggregate during decarbonation to yieldCOandCO2. The carbonates decomposed during initial heating of the experiments, causing their some constituent components to be dissolved in and transferred by the fluid to the pore space of the albite glass matrix. After temperature reached 1000°C glass, the shards annealed and then melted, as evidenced by a vesiculated glass in the quench products. Micro‐Raman investigation of the fluid in bubbles in the albite glass in experiments with decomposition of natural siderite yieldedCO–CO2mixture whereCOmole fraction was 0.15–0.16. We observe significant concentrations of Pt, Mn, P, andREEin the albite glass; in contrast, no Fe or Mg transfer was detected.LA‐ICP‐MSanalysis of the albite glass product yielded the average Pt content of 2 ppm. Such high Pt signal came from Pt particles (100–500 nm in size), which were observed on the walls of the bubbles embedded in the glass. Olivines and aluminous spinel were observed in the Fe‐oxide aggregate, demonstrating transfer of SiO2and Al2O3from the albite melt by the reduced carbonic fluid from the albite glass (large capsule). Our results demonstrate that dryCO–CO2 fluid can be important agents of dissolution and transport, especially for Pt and other metals. The data imply that metals are chiefly dissolved as carbonyl complexes.

The origin of nelsonite constrained by melting experiment and melt inclusions in apatite: The Damiao anorthosite complex, North China Craton

Models for the nelsonite formation are currently highly contentious, with liquid immiscibility and fractional crystallization as frequently proposed formation mechanisms. The nelsonites in the Damiao massif anorthosite complex in the North China Craton are revisited here together with experimental evidence for the existence of silica-free CaO-FeO-Fe2O3-TiO2-P2O5 immiscible nelsonitic liquids. Our results of differential scanning calorimetry (DSC) and internally heated pressure vessel (IHPV) demonstrate that nelsonite with the composition of one-third apatite and two-thirds Fe-Ti oxides by weight 1) completely melts well above 1450°C at dry condition, which is in good agreement with numerous experimental studies of the CaO-P2O5-FexO system for metallurgical purposes; 2) does not melt at the temperature up to 1200°C with presence of considerable amount of volatiles, e.g., fluorine and water. Therefore, the composition of the nelsonite cannot be molten at temperatures relevant for crystallization of the Damiao magma. A review of experimental studies of liquid immiscibility and analyses of natural immiscible glasses show that all the liquids on the Fe- and P-rich side of the miscibility gap have at least 20wt.% of aluminosilicate components.Analyses of apatite-hosted melt inclusions from Damiao nelsonite provide further constraints on the nelsonite formation. The inclusions range from ~3 to 200μm in diameter. They are ubiquitous and meet all the morphological criteria of primary melt inclusions crystallized into assemblages of daughter minerals. Almost all assemblages contain vermiculite and/or chlorite, and some contain biotite, amphibole and Fe-Ti oxide. Melt compositions of individual inclusions has been estimated based on modal proportions and mineral compositions of daughter phases. The inclusions show large variations in SiO2 (20.8–50.2wt.%) and FeOt (13.4–32.8wt.%). With a few exceptions, the inclusions are very low in CaO (<1.5wt.%) and high in Al2O3 (10.0–21.2wt.%). The low total values indicate very hydrated nature of the melt inclusions. The melt inclusions were also re-heated at 1200°C by IHPV. Glasses from partly to fully homogenized melt inclusions show similar compositional characteristics as the bulk compositions from the crystallized melt inclusions but incorporated excessive melting of the host apatite. It appears that the cumulus apatite crystallized from hydrated late-stage immiscible Fe-rich melt formed by liquid immiscibility occurred in the relatively early stage, i.e. prior to the arrival of aqueous fluid phase. We propose that the inclusions at Damiao record a trend of late-stage melt evolution, which have been through silicate liquid immiscibility but most likely have been strongly affected by separation of a magmatic fluid phase and loss of alkali and Ca silicate components from the melt into the fluid.

Experimental investigation of the S and S-isotope distribution between H2O–S ± Cl fluids and basaltic melts during decompression

Decompression experiments (from 400 to 70MPa) were conducted to investigate sulfur (S) distribution and S-isotope fractionation between basaltic melts and coexisting fluids. Volatile-bearing [~3 to ~7wt.% water (H2O), ~300 to ~1200ppmS, 0 to ~3600ppm chlorine (Cl)] basaltic glasses were used as starting materials. The MgO content in the melt was either ~1wt.% (Mg-poor basalt) or ~10wt.% (alkali basalt) to investigate the possible role of compositional changes in basaltic systems on fluid-melt distribution of S and S-isotopes. The experiments were performed in internally heated pressure vessels (IHPV) at 1050°C to 1250°C, variable oxygen fugacities (fO2; ranging from log(fO2/bar)~QFM to ~QFM+4; QFM=quartz–fayalite–magnetite buffer) and at a constant decompression rate (r) of 0.1MPa/s. The annealing time (tA) at final pressure (p) and temperature (T) after decompression was varied from 0 to 5.5h to study the fluid–melt equilibration process.Sulfur and H2O contents in the melt decreased significantly during decompression, while the Cl contents remained almost constant. No changes in H2O and Cl content were observed with tA, while S concentrations decreased slightly with tA <2h; i.e., near-equilibrium fluid–melt conditions were reached within ~2h after decompression, even in experiments performed at the lowest T of 1050°C. Thus, fluid–melt partitioning coefficients of S (DSfl/m) were determined from experiments with tA ≥2h.The MgO (~1 to ~10wt.%), H2O (~3 to ~7wt.%) and Cl contents (<0.4wt.%) in the melt have no significant effect on DSfl/m. Consistent with previous studies we found that DSfl/m decreased strongly with increasing fO2; e.g., at ~1200°C DSfl/m≈180 at QFM+1 and DSfl/m≈40 at QFM+4. A positive correlation was observed between DSfl/m and T in the range of 1150 to 1250°C at both oxidizing (QFM+4; DSfl/m=52±27 to 76±30) and intermediate (QFM+1.5; DSfl/m=94±20 to 209±80) redox conditions. Data compiled at 1050°C and relatively reducing conditions (~QFM; DSfl/m=58±18) indicate that the trends may be extrapolated to lower T, at least for intermediate to reducing conditions (~QFM+1.5 to ~QFM).The S-isotope composition in glasses of selected samples was measured by secondary ion mass spectrometry (SIMS). Gas–melt isotopic fractionation factors (αfl–m) were calculated via mass balance. At 1200°C an average αfl–m of 0.9981±0.0015 was determined for oxidizing conditions (~QFM+4), while an average αfl–m of 1.0025±0.0010 was found for fairly reducing conditions (~QFM+1). Furthermore, at lower T (1050°C) an average αfl–m of 1.0037±0.0009 was determined for reducing conditions (~QFM). The data showed that equilibrium fractionation effects during closed-system degassing of basaltic melts at T relevant for magmatic systems (1050 to 1250°C) can induce a S-isotope fluid–melt fractionation of about +4‰ in relatively reduced systems and of about −2‰ in relatively oxidized systems.The reported experimental results are valuable for the interpretation of S and S-isotope signature in magmatic systems (e.g., in volcanic gasses or melt inclusions) and will help to elucidate, for instance, volatile transport processes across subduction zones and Earth's S cycle.

Sulfur isotope fractionation between fluid and andesitic melt: An experimental study

Glasses produced from decompression experiments conducted by Fiege et al. (2014a) were used to investigate the fractionation of sulfur isotopes between fluid and andesitic melt upon magma degassing. Starting materials were synthetic glasses with a composition close to a Krakatau dacitic andesite. The glasses contained 4.55–7.95wt% H2O, ∼140 to 2700ppm sulfur (S), and 0–1000ppm chlorine (Cl). The experiments were carried out in internally heated pressure vessels (IHPV) at 1030°C and oxygen fugacities (fO2) ranging from QFM+0.8 log units up to QFM+4.2 log units (QFM: quartz–fayalite–magnetite buffer). The decompression experiments were conducted by releasing pressure (P) continuously from ∼400MPa to final P of 150, 100, 70 and 30MPa. The decompression rate (r) ranged from 0.01 to 0.17MPa/s. The samples were annealed for 0–72h (annealing time, tA) at the final P and quenched rapidly from 1030°C to room temperature (T).The decompression led to the formation of a S-bearing aqueous fluid phase due to the relatively large fluid–melt partitioning coefficients of S. Secondary ion mass spectrometry (SIMS) was used to determine the isotopic composition of the glasses before and after decompression. Mass balance calculations were applied to estimate the gas–melt S isotope fractionation factor αg-m.No detectable effect of r and tA on αg-m was observed. However, SIMS data revealed a remarkable increase of αg-m from ∼0.9985±0.0007 at >QFM+3 to ∼1.0042±0.0042 at ∼QFM+1. Noteworthy, the isotopic fractionation at reducing conditions was about an order of magnitude larger than predicted by previous works. Based on our experimental results and on previous findings for S speciation in fluid and silicate melt a new model predicting the effect of fO2 on αg-m (or Δ34Sg–m) in andesitic systems at 1030°C is proposed. Our experimental results as well as our modeling are of high importance for the interpretation of S isotope signatures in natural samples (e.g., melt inclusions or volcanic gases).

Kinetic vs. thermodynamic control of degassing of H2O–S ± Cl-bearing andesitic melts

The sulfur (S) distribution between andesitic melts and fluids, both under near-equilibrium conditions and during fast decompression (disequilibrium conditions), has been experimentally investigated. Isothermal decompression experiments were conducted at ∼1030°C and variable oxygen fugacity (fO2; log(fO2) from ∼QFM+0.8 to ∼QFM+4.2; QFM=quartz–fayalite–magnetite buffer) in internally heated pressure vessels (IHPV) using synthetic, H2O- and S-bearing andesitic melts (∼4–8wt% H2O, ∼140–2700ppm S). Selected glasses were doped with chlorine (Cl; 500–1000ppm) to study the influence of Cl on S partitioning. The starting pressure varied from 300 to 500MPa, and pressure (p) was released continuously to reach 150, 100, 70, or 30MPa. The decompression rate (r) ranged from 0.0005 to 0.17MPa/s and samples were either directly quenched to preserve disequilibrium conditions or annealed for various times (annealing time (tA)=1–72h) at final p and 1030°C to achieve near-equilibrium conditions.The directly quenched experiments revealed a strong increase of the S(fluid)/S(melt) ratio (S(fluid)=wt% S in the fluid; S(melt)=wt% S in the melt) with increasing r, from ∼30 at 0.02MPa/s to ∼300 at 0.2MPa/s at oxidizing conditions (log(fO2)>QFM+3), i.e., when sulfate (S6+) was the only S species. After fast decompression (∼0.1MPa/s) subsequent annealing for ≥5h resulted in a decrease of S(fluid)/S(melt) by a factor of ∼6, indicating that part of the S present in the fluid at tA=0h diffuses back into the melt during annealing. This behavior is explained by the positive correlation between p and sulfate solubility in aqueous fluids. In contrast to oxidizing conditions, the S content in the melt remained almost constant with varying r and was independent of tA at low fO2 (QFM+1 to QFM+1.5), when sulfide (S2−) became abundant. Thus, the different behaviors of S2− and S6+ during kinetically-controlled degassing need to be considered when modeling the volatile release of ascending magma.The addition of >500ppm Cl to the system slightly increased the S(fluid)/S(melt) under near-equilibrium conditions by a factor of ∼2 at QFM+1.8. Furthermore, Cl(fluid)/Cl(melt) (wt% Cl in fluid/wt% Cl in melt) showed a positive correlation with initial Cl content of the melt and ranged from 1 to 13, largely independent of r and tA. The interaction between S and Cl as well as the dependence of Cl(fluid)/Cl(melt) on Cl content in the melt may have a significant influence on S/Cl ratios in volcanic gasses.

Hydrogen concentration in plagioclase as a hygrometer of arc basaltic melts: Approaches from melt inclusion analyses and hydrous melting experiments

The partition coefficients of hydrogen between plagioclase and basaltic melt were determined by two approaches. For the first part of this study, plagioclase-hosted melt inclusions in mid-ocean ridge basalt (MORB) from the Rodriguez Triple Junction in the Indian Ocean were analyzed. The hydrogen concentration in plagioclase is less than 60wtppm water, and the average H2O concentration in melt inclusions is 0.3wt%. Therefore, the apparent partition coefficient of hydrogen between plagioclase and melt is ≈0.01 on a molar basis. For the second part of this study, hydrous melting experiments of arc basaltic magma were performed at 0.35GPa using an internally-heated pressure vessel at fO2≈NNO+3. The starting material was hydrous basaltic glass with H2O ranging from 0.8 to 5.5wt%. A grain of Ca-rich plagioclase (≈1mg) and 10mg of powdered basaltic glass were sealed in a Au80Pd20 alloy capsule, and then kept at near the crystallization temperature of plagioclase as a liquidus phase to attain an equilibrium of hydrogen between plagioclase and melt. Combining the results of these two parts of this study, we formulated two linear equations to correlate the hydrogen concentration in plagioclase and the H2O concentration in melt. When H2O in melt is ≤1wt%,hydrogen in plagioclase (wtppm water)≈80×H2O in melt (wt%).When H2O in melt is≥4wt%,hydrogen in plagioclase (wtppm water)≈40×H2O in melt (wt%).Hydrogen concentration in plagioclase lies between two equations when H2O in melt ranges from 1 to 4wt%. In accordance with these formulations, the partition coefficients of hydrogen between plagioclase and basaltic melt switches from 0.01±0.005 under H2O-poor conditions (≤100wtppm water in plagioclase, ≤1wt% H2O in melt) to 0.005±0.001 under H2O-rich conditions (≥150wtppm water in plagioclase, ≥4wt% H2O in melt). Such switches of hydrogen partitioning with an increase in H2O can be related to change of the atomic site for hydrogen in the crystal structure of plagioclase.

A new type of high-pressure low-flow metering valve for continuous decompression: First experimental results on degassing of rhyodacitic melts

A novel type of high-pressure low-flow metering valve has been designed for experiments at continuous decompression in internally heated pressure vessels (IHPV) at pressures up to 500 MPa and temperatures up to 1500 °C. It consists of a modified high-pressure valve coupled with a piezoelectric nanopositioning system. The piezoelectric actuator (1) positions the needle of the valve statically with very high precision on a nanometer scale for infinitesimally slow decompression and, alternatively, (2) opens and closes the valve completely, within milliseconds, to achieve a fast pressure drop only limited by the diameter of the high-pressure tubing and the entire volume of the autoclave system. The valve connected to an IHPV has been successfully tested to simulate continuous decompression of volatile-bearing magma. Water saturated rhyodacitic melt was synthesized at 1050 °C and 300 MPa and subsequently decompressed to 50 MPa at an integrated decompression rate of 0.28 MPa/s using continuous decompression as well as single-step and multi-step decompression techniques. The experimental results of the three methods show significant differences, having important implications for the interpretation of textures observed in natural volcanic rocks. Bubble number density (BND) values increase from continuous to multi-step and to single-step decompression by two orders of magnitude. Bubble size distribution (BSD) also differs significantly. The BSD curve of the single-step decompression shows largest variation both in size and in population number density, whereas the BSD trend for the sample of multi-step decompression shows the smallest variation between size and population density of bubbles. The BSD of the continuously decompressed sample indicates similar proportions of bubbles with different size. Channel formation in the sample of single-step decompression may indicate fluid escape from over-pressurized bubbles through the melt. Single-step and continuous decompression style represent two extreme cases of magma degassing in which bubble nucleation and bubble growth are predominant processes, respectively, whereas multi-step decompression resembles the intermediate case.

Sulfur K-edge XANES analysis of natural and synthetic basaltic glasses: Implications for S speciation and S content as function of oxygen fugacity

XANES analyses at the sulfur K-edge were used to determine the oxidation state of S species in natural and synthetic basaltic glasses and to constrain the fO 2 conditions for the transition from sulfide (S 2−) to sulfate (S 6+) in silicate melts. XANES spectra of basaltic samples from the Galapagos spreading center, the Juan de Fuca ridge and the Lau Basin showed a dominant broad peak at 2476.8 eV, similar to the spectra obtained from synthetic sulfide-saturated basalts and pyrrhotite. An additional sharp peak at 2469.8 eV, similar to that of crystalline sulfides, was present in synthetic glasses quenched from hydrous melts but absent in anhydrous glasses and may indicate differences in sulfide species with hydration or presence of minute sulfide inclusions exsolved during quenching. The XANES spectra of a basalt from the 1991 eruption of Mount Pinatubo, Philippines, and absarokitic basalts from the Cascades Range, Oregon, USA, showed a sharp peak at 2482.8 eV, characteristic of synthetic sulfate-saturated basaltic glasses and crystalline sulfate-bearing minerals such as hauyne. Basaltic samples from the Lamont Seamount, the early submarine phase of Kilauea volcano and the Loihi Seamount showed unequivocal evidence of the coexistence of S 2− and S 6+ species, emphasizing the relevance of S 6+ to these systems. XANES spectra of basaltic glasses synthesized in internally-heated pressure vessels and equilibrated at fO 2 ranging from FMQ − 1.4 to FMQ + 2.7 showed systematic changes in the features related to S 2− and S 6+ with changes in fO 2. No significant features related to sulfite (S 4+) species were observed. These results were used to construct a function that allows estimates of S 6+/ΣS from XANES data. Comparison of S 6+/ΣS data obtained by S Kα shifts measured with electron probe microanalysis (EPMA), S 6+/ΣS obtained from XANES spectra, and theoretical considerations show that data obtained from EPMA measurements underestimate S 6+/ΣS in samples that are sulfate-dominated (most likely because of photo-reduction effects during analysis) whereas S 6+/ΣS from XANES provide a close match to the expected theoretical values. The XANES-derived relationship for S 6+/ΣS as a function of fO 2 indicates that the transition from S 2− to S 6− with increasing fO 2 occurs over a narrower interval than what is predicted by the EPMA-derived relationship. The implications for natural systems is that small variation of fO 2 above FMQ + 1 will have a large effect on S behavior in basaltic systems, in particular regarding the amount of S that can be transported by basaltic melts before sulfide saturation can occur.

Pressure dependence of viscosity of rhyolitic melts

Viscosity of silicate melts is a critical property for understanding volcanic and igneous processes in the Earth. We investigate the pressure effect on the viscosity of rhyolitic melts using two methods: indirect viscosity inference from hydrous species reaction in melts using a piston cylinder at pressures up to 2.8 GPa and direct viscosity measurement by parallel-plate creep viscometer in an internally-heated pressure vessel at pressures up to 0.4 GPa. Comparison of viscosities of a rhyolitic melt with 0.8 wt% water at 0.4 GPa shows that both methods give consistent results. In the indirect method, viscosities of hydrous rhyolitic melts were inferred based on the kinetics of hydrous species reaction in the melt upon cooling (i.e., the equivalence of rheologically defined glass transition temperature and chemically defined apparent equilibrium temperature). The cooling experiments were carried out in a piston-cylinder apparatus using hydrous rhyolitic samples with 0.8–4 wt% water. Cooling rates of the kinetic experiments varied from 0.1 K/s to 100 K/s; hence the range of viscosity inferred from this method covers 3 orders of magnitude. The data from this method show that viscosity increases with increasing pressure from 1 GPa to 3 GPa for hydrous rhyolitic melts with water content ⩾0.8 wt% in the high viscosity range. We also measured viscosity of rhyolitic melt with 0.13 wt% water using the parallel-plate viscometer at pressures 0.2 and 0.4 GPa in an internally-heated pressure vessel. The data show that viscosity of rhyolitic melt with 0.13 wt% water decreases with increasing pressure. Combining our new data with literature data, we develop a viscosity model of rhyolitic melts as a function of temperature, pressure and water content.

The connection between high K melts and Au deposits: Evidence from natural and experimental systems

It has long been suspected that K-enriched melts are efficient carriers of Au. This assumption is the result of clear spatial/temporal associations between Au deposits and High-K (or shoshonitic) intrusives in the deposits of Lihir Island (Muller et al., 2001), Bajo de la Alumbrera, Argentina (Muler and Forrestal, 1998), Quesnilla, BC, Canada (Barrie, 1993) and in the Lower Yangtze region of China (Zhou et al., 1996) and others. However, no direct link has been made between the high K nature of these melts and their Au endowment. As a test of this hypothesis, we analysed Au by LA-ICPMS in clinopyroxene-hosted melt inclusions from an ore-associated dyke at the Masbate Gold Deposit in the Phillipines. Pyroxenes were re-equilibrated at 1100 C for 24 h in a controlled gas mixing furnace, quenched in water and polished to expose glassy, homogenised ultrapotassic (8–11wt% K2O) inclusions. All inclusions analysed contain high levels of dissolved gold (up to 20 ppm Au). To independently test this empirical link, we equilibrated hydrous rhyolitic glasses of variable K contents (plus either FeS or FeS and NaCl) with Au capsules in internally-heated pressure vessels. At constant water content, temperature, pressure, fO2 (controlled by H–Ar mixtures) the addition of K2O increases Au solubility by nearly an order of magnitude. In runs with FeS only, Au content increased from 0.08 ppm Au (@1.6% K2O) to 0.66 ppm Au (@ 6.8%K2O), while those with FeS and NaCl contained 0.45 ppm @ 1.6% K2O and 2.2 ppm Au @ 6.1%K2O. These analyses and experiments do not explain the mechanism for increased Au solubility, but they do strengthen the link. Furthermore, in terranes where pervasive alterration obscures primary igneous compositions, melt inclusions may provide the best information, and may provide a valuable exploration tool. References Barrie, C.T., 1993. J. Geochem. Expl. 48, 225–258. Muller, D., Franz, L., Herzig, P.M., Hunt, S., 2001. Lithos 57, 163–186. Muler, D., Forrestal, P., 1998. Min. Pet. 64, 47–64. Zhou, J.C., Zhou, J.P., Liu, J., Zhao, T.P., Chen, W., 1996. J. Geochem. Expl. 57, 273–283. doi:10.1016/j.gca.2006.06.815 A404 Goldschmidt Conference Abstracts 2006