Thermochemistry of sulfide liquids IV: density measurements and the thermodynamics of O–S–Fe–Ni–Cu liquids at low to moderate pressures

We present the results of a series of density experiments in the system O–S–Fe–Ni–Cu. These experiments were designed to extend our understanding of the physical properties of sulfide liquids, and to extend one-bar thermochemical models for sulfide liquids to apply to low to moderate pressures. Density measurements indicate both positive and negative deviations from linear mixing of partial molar volumes across this five-dimensional composition space. In terms of the homogeneous speciation model of Kress (in Contrib Mineral Petrol 154:191–204, 2007), the best fit to experimental data can be achieved by starting with a model where the volume of formation reaction for associated species initially is set to zero. Further refinement of this first-order fit yields a volume mixing model which reproduces experimental data to within nearly the estimated experimental uncertainty. Experimental ultrasonic and X-ray absorption data from the literature, along with the bulk modulus–volume relation of Anderson and Nafe (J Geophys Res 16:3951–3963, 1965), allow the estimation of the pressure dependence of partial molar volumes for sulfide liquid species. The resulting combined thermochemical model should be valid to about 2,000 K and 3 GPa. Application of this thermochemical model in a simple adiabatic magma ascent scenario confirms earlier work suggesting that the pressure dependence of sulfur solubility in sulfide-saturated magma will decrease with increasing pressure along geologically reasonable paths in P–T–– space.     

Determining the composition of C–H–O liquids following high-pressure and high-temperature diamond-trap experiments

Here, we present the first analytical technique (the quartz tube system technique—QTS) to directly analyze H₂O and CO₂ contents in liquids following high-pressure, high-temperature experiments in capsules containing mantle minerals and a diamond layer serving as a fluid/melt trap. In this technique, the capsule is frozen prior to opening; the diamond trap is cut out of the capsule and placed inside a N₂-filled quartz tube. The diamond trap is heated up to 900 °C to release the gases to an Infrared Gas Analyzer, which determines the CO₂ and H₂O contents. Three sets of experiments containing SiO₂ and CaCO₃ powders were performed at 6 GPa and 1,000 °C in order to calibrate and validate the technique. These experiments demonstrated that when samples are prepared in a N₂ environment, CO₂ and H₂O can be directly measured with an accuracy and precision of 2–3 and 3–4 %, respectively. The QTS technique (for H₂O and CO₂ determination) together with the cryogenic technique (total dissolved solids content) can be applied to diamond-trap capsules following HP–HT experiments in order to provide direct and complete liquid compositions coexisting with mantle material. The principal advantage of the QTS technique of direct analysis of volatile content in liquids over the indirect approach of mass balance calculations is the possibility of studying carbonated and hydrous liquid compositions in equilibrium with mantle material regardless of chemistry and pressure–temperature experimental conditions.     

Do carbonate liquids become denser than silicate liquids at pressure? Constraints from the fusion curve of K<Subscript>2</Subscript>CO<Subscript>3</Subscript> to 3.2 GPa

Brackets on the melting temperature of K₂CO₃ were experimentally determined at 1.86 ± 0.02 GPa (1,163–1,167°C), 2.79 ± 0.03 GPa (1,187–1,195°C), and 3.16 ± 0.04 GPa (1,183–1,189°C) in a piston-cylinder apparatus. These new data, in combination with published experiments at low pressure (<0.5 GPa), establish the K₂CO₃ fusion curve to 3.2 GPa. On the basis of these experiments and published thermodynamic data for crystalline and liquid K₂CO₃, the high-pressure density and compressibility of K₂CO₃ liquid were derived from the fusion curve. The pressure dependence of the liquid compressibility (K′₀ = dK ₀/dP, where K ₀ = 1/β₀) is between 16.2 and 11.6, with a best estimate of 13.7, in a third-order Birch–Murnaghan equation of state (EOS). This liquid K′₀ leads to a density of 2,175 ± 36 kg/m³ at 4 GPa and 1,500°C, which is ∼30% lower than that reported in the literature on the basis of the falling-sphere method at the same conditions. The uncertainty in the liquid K′₀ leads to an error in melt density of ± 2% at 4 GPa; the error decreases with decreasing pressure. With a K′₀ of 13.7, the compressibility of K₂CO₃ at 1,500°C and 1 bar (K ₀ = 3.8 GPa) drops rapidly with increasing pressure ( ), which prevents a density crossover with silicate melts, such as CaAlSi₂O₈ and CaMgSi₂O₆, at upper mantle depths.     

Sound velocity and elastic properties of Fe–Ni and Fe–Ni–C liquids at high pressure

The sound velocity (V _P) of liquid Fe–10 wt% Ni and Fe–10 wt% Ni–4 wt% C up to 6.6 GPa was studied using the ultrasonic pulse-echo method combined with synchrotron X-ray techniques. The obtained V _P of liquid Fe–Ni is insensitive to temperature, whereas that of liquid Fe–Ni–C tends to decrease with increasing temperature. The V _P values of both liquid Fe–Ni and Fe–Ni–C increase with pressure. Alloying with 10 wt% of Ni slightly reduces the V _P of liquid Fe, whereas alloying with C is likely to increase the V _P. However, a difference in V _P between liquid Fe–Ni and Fe–Ni–C becomes to be smaller at higher temperature. By fitting the measured V _P data with the Murnaghan equation of state, the adiabatic bulk modulus (K _S0) and its pressure derivative (K _S ^′ ) were obtained to be K _S0 = 103 GPa and K _S ^′  = 5.7 for liquid Fe–Ni and K _S0 = 110 GPa and K _S ^′  = 7.6 for liquid Fe–Ni–C. The calculated density of liquid Fe–Ni–C using the obtained elastic parameters was consistent with the density values measured directly using the X-ray computed tomography technique. In the relation between the density (ρ) and sound velocity (V _P) at 5 GPa (the lunar core condition), it was found that the effect of alloying Fe with Ni was that ρ increased mildly and V _P decreased, whereas the effect of C dissolution was to decrease ρ but increase V _P. In contrast, alloying with S significantly reduces both ρ and V _P. Therefore, the effects of light elements (C and S) and Ni on the ρ and V _P of liquid Fe are quite different under the lunar core conditions, providing a clue to constrain the light element in the lunar core by comparing with lunar seismic data.     

Decoupling the mechanical role of pore liquids in soils to liquid viscous effect and solid–liquid adhesion effect

Acta Geotechnica - This paper presents a novel conceptual approach for evaluating the mechanical effect of pore liquids on the overall geotechnical behavior. The approach is based on empiric...     

A method for estimating the activity of titania in magmatic liquids from the compositions of coexisting rhombohedral and cubic iron–titanium oxides

A method is described for estimating the activity of titania (TiO₂) in a magmatic liquid from the compositions of coexisting cubic oxide (spinel) and rhombohedral oxide (ilmenite). These estimates are derived from the thermodynamic models of Ghiorso and Evans (Am J Sci 308:957–1039, 2008; see also Sack and Ghiorso in Contrib Mineral Petrol 106:474–505, 1991a; Am Mineral 76:827-847, 1991b) and may be computed self consistently along with temperature and oxygen fugacity for an assumed pressure. The method is applied to a collection of 729 naturally occurring oxide pairs from rhyolites and dacites. For this suite of oxides, values of titania activity relative to rutile saturation range from 0.3 to 0.9. Genetically related groups of oxide pairs display activity–temperature trends with negative slopes at higher activities (0.6–0.9) or positive slopes at lower activities (0.3–0.7). Thermodynamic analysis supports the assumption of two-oxide, liquid equilibrium for the former group, but suggests that such an interpretation for oxide sequences with positive activity–temperature trends may be problematic. Application of the estimation method to oxide pairs from the Shiveluch Volcano and the Bishop Tuff reveals that the former are consistent with having equilibrated with known matrix glass compositions, whereas the latter pairs are inconsistent with equilibration with pre-eruptive liquids trapped in quartz inclusions.     

Experimental investigation of H2O and Cl− solubilities in F-enriched silicate liquids; implications for volatile saturation of topaz rhyolite magmas

To interpret the degassing of F-bearing felsic magmas, the solubilities of H₂O, NaCl, and KCl in topaz rhyolite liquids have been investigated experimentally at 2000, 500, and ≈1 bar and 700° to 975 °C. Chloride solubility in these liquids increases with decreasing H₂O activity, increasing pressure, increasing F content of the liquid from 0.2 to 1.2 wt% F, and increasing the molar ratio of ((Al + Na + Ca + Mg)/Si). Small quantities of Cl⁻ exert a strong influence on the exsolution of magmatic volatile phases (MVPs) from F-bearing topaz rhyolite melts at shallow crustal pressures. Water- and chloride-bearing volatile phases, such as vapor, brine, or fluid, exsolve from F-enriched silicate liquids containing as little as 1 wt% H₂O and 0.2 to 0.6 wt% Cl at 2000 bar compared with 5 to 6 wt% H₂O required for volatile phase exsolution in chloride-free liquids. The maximum solubility of Cl⁻ in H₂O-poor silicate liquids at 500 and 2000 bar is not related to the maximum solubility of H₂O in chloride-poor liquids by simple linear and negative relationships; there are strong positive deviations from ideality in the activities of each volatile in both the silicate liquid and the MVP(s). Plots of H₂O versus Cl⁻ in rhyolite liquids, for experiments conducted at 500 bar and 910°–930 °C, show a distinct 90° break-in-slope pattern that is indicative of coexisting vapor and brine under closed-system conditions. The presence of two MVPs buffers the H₂O and Cl⁻ concentrations of the silicate liquids. Comparison of these experimentally-determined volatile solubilities with the pre-eruptive H₂O and Cl⁻ concentrations of five North American topaz and tin rhyolite melts, determined from melt inclusion compositions, provides evidence for the exsolution of MVPs from felsic magmas. One of these, the Cerro el Lobo magma, appears to have exsolved alkali chloride-bearing vapor plus brine or a single supercritical fluid phase prior to entrapment of the melt inclusions and prior to eruption. Received: 6 November 1995 / Accepted: 29 January 1998     

Diffusive equilibration between hydrous metaluminous-peraluminous haplogranite liquid couples at 200 MPa (H<Subscript>2</Subscript>O) and alkali transport in granitic liquids

This study examines the systematics and rate of alkali transport in haplogranite diffusion couples in which a chemical potential gradient in Al is established between near water-saturated metaluminous and peraluminous liquids that differ only in their initial content of normative corundum. At 800°C, measurable chemical diffusion of alkalis occurs throughout the entire length (∼1 cm) of the diffusion couples in 2–6 h, indicating long range diffusive communication through melt. Alkali transport results in homogenization of initially different Na/Al and ASI [=mol. Al₂O₃/(CaO + Na₂O + K₂O)] throughout the couples within ∼24 h, whereas initially homogenous K* evolves to become uniformly different between metaluminous and peraluminous ends. Calculated effective binary diffusion coefficients for alkalis in experiments that do not significantly violate the requirement of a semi-infinite chemical reservoir (0- to 2-h duration at 800°C) are similar to those observed in previous studies: in the range of (1–8) × 10⁻¹² m²/s. Such a magnitude of diffusivity, however, is inadequate to account for the observed changes of alkali concentrations and molecular ratios throughout the couples in 2- to 6-h experiments. The latter changes are consistent with diffusivities estimated via the x ² = Dt approximation, which yields effective values around 10⁻⁹ m²/s. These observations suggest that Fick’s law alone does not adequately describe the diffusive transport of alkalis in granitic liquids. In addition to simple ionic diffusion associated with local gradients in concentration or chemical potential of the diffusing component described by Fick’s second law (local diffusion), alkali transport through melt involves system-wide diffusion (field diffusion) driven by chemical potential gradients that also include components with which the alkalis couple or complex (e.g., Al). Field diffusion involves the coordinated migration of essentially all alkali cations, resembling a positive ionic current that drives the system to a metastable state having a minimum energy configuration with respect to alkali distribution. The net result is effective transport rates perhaps three orders of magnitude faster than simple local alkali diffusion, and at least seven to eight orders of magnitude faster than the diffusive equilibration of Al and Si.     

Experimental evidence for the coexistence of two liquids in the H2O-SiO2-NaF-Na2SO4 System at T = 700°C and P = 2 kbar

The phase state of fluid in the H₂O-NaF-Na₂SO₄ system in the presence of silicates (quartz and albite) was experimentally explored using the method of synthetic fluid inclusions in quartz at 700°C and pressures of 1 and 2 kbar. Parallel experiments were conducted under identical conditions with either two silicates (quartz and albite) or quartz only. The presence of albite affects heterogeneous fluid equilibria both at different pressures and at different solution compositions. This indicates high solubilities of silicates in a saltwater fluid containing NaF and Na₂SO₄. The absence of inclusions homogenizing to a gas phase in the experimental products provides compelling evidence that liquid-liquid rather than liquid-vapor equilibria are characteristic of the H₂O-SiO₂-NaF-Na₂SO₄ and H₂O-SiO₂-NaF-Na₂SO₄-NaAlSi₃O₂ systems in the heterogeneous region. It can be concluded that critical equilibria in saturated solutions can exist in these systems. In addition, it was shown that the phase diagrams of these systems are complicated by the formation of immiscible liquids in the presence of vapor. This allowed us to conclude that there are two critical curves describing equilibria with two different salts. Fluids containing two salts (NaF and Na₂SO₄) are similar to fluids containing only one of these salts: (a) two liquids are in equilibrium under the parameters of the upper heterogeneous region, (b) each of them can in turn undergo unmixing at decreasing temperature and pressure, and (c) owing to chemical interaction between silicate and fluid components, a glassy phase can be formed and trapped in inclusions.     

Contrasting interactions of sodium and potassium with H<Subscript>2</Subscript>O in haplogranitic liquids and glasses at 200 MPa from hydration–diffusion experiments

This study examines hydration–diffusion in the metaluminous haplogranite system at 200 MPa H₂O and 800–300°C. At 800°C hydration is accompanied by melting and uphill diffusion of sodium from anhydrous glass toward the region of hydration and melting, whereas potassium diffuses away from the hydration front and into anhydrous glass. Silicon and aluminum are simply diluted upon hydration. There is no change in molecular Al/(Na + K) throughout the entire hydration-diffusion aureole and, therefore, (1) there is no loss of alkalis to the vapor, and (2) K migrates to replace Na in order to maintain local charge balance required by ^IVAl. Alkali diffusion occurs over a viscosity contrast from 10^4.1 Pa s in hydrous liquid to 10^11.8–10^13.5 Pa s in anhydrous glass. From these results, we interpret that: (1) Na is structurally or energetically favored over K as a charge-balancing cation for ^IVAl in hydrous granitic liquids, whereas the opposite behavior has been observed for anhydrous melts, and (2) the diffusion of alkalis through silicate melts is largely independent of viscosity. Results from 600°C are similar to those at 800°C, but hydration at 300°C involves a loss of Na and concomitant increase in molar Al/(Na + K) in the hydration zone due to hydrogen-alkali exchange between fluid and glass. Hydration behavior at 400°C is transitional between those at 300°C and 600°C, suggesting that the change in hydration mechanism occurs near the glass transition.     

A study of rare earth element (REE)–SiO2 variations in felsic liquids generated by basalt fractionation and amphibolite melting: a potential test for discriminating between the two different processes

The origin of felsic magmas (>63% SiO₂) in intra-oceanic arc settings is still a matter of debate. Two very different processes are currently invoked to explain their origin. These include fractional crystallization of basaltic magma and partial melting of lower crustal amphibolite. Because both fractionation and melting can lead to similar major element, trace element and isotopic characteristics in felsic magmas, such lines of evidence have been generally unsuccessful in discriminating between the two processes. A commonly under-appreciated aspect of rare earth element (REE) solid–liquid partitioning behavior is that D _REE for most common igneous minerals (especially hornblende) increase significantly with increasing liquid SiO₂ contents. For some minerals (e.g., hornblende and augite), REE partitioning can change from incomptatible (D < 1) at low liquid SiO₂ to compatible (D > 1) at high liquid SiO₂. When this behavior is incorporated into carefully constrained mass-balance models for mafic (basaltic) amphibolite melting, intermediate (andesitic) amphibolite melting, lower or mid to upper crustal hornblende-present basalt fractionation, and mid to upper crustal hornblende-absent basalt fractionation the following general predictions emerge for felsic magmas (e.g., ∼63 to 76% SiO₂). Partial melting of either mafic or intermediate amphibolite should, regardless of the type of melting (equilibrium, fractional, accumulated fractional) yield REE abundances that remain essentially constant and then decrease, or steadily decrease with increasing liquid SiO₂ content. At high liquid SiO₂ contents LREE abundances should be slightly enriched to slightly depleted (i.e., C _l/C _o ∼ 2 to 0.2) while HREE abundances should be slightly depleted (C _l/C _o ∼ 1 to 0.2). Lower crustal hornblende-bearing basalt fractionation should yield roughly constant REE abundances with increasing liquid SiO₂ and exhibit only slight enrichment (C _l/C _o ∼ 1.2). Mid to upper crustal hornblende-bearing basalt fractionation should yield steadily increasing LREE abundances but constant and then decreasing HREE abundances. At high liquid SiO₂ contents LREE abundances may range from non-enriched to highly enriched (C _l/C _o ∼ 1 to 5) while HREE abundances are generally non-enriched to only slightly enriched (C _l/C _o ∼ 1 to 2). Hornblende-absent basalt fractionation should yield steadily increasing REE abundances with increasing liquid SiO₂ contents. At high SiO₂ contents both LREE and HREE are highly enriched (C _l/C _o ∼ 3 to 4). It is proposed that these model predictions constitute a viable test for determining a fractionation or amphibolite melting origin for felsic magmas in intra-oceanic arc environments where continental crust is absent. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.

Formation and properties of hydrosilicate liquids in the systems Na2O-Al2O3-SiO2-H2O and granite-Na2O-SiO2-H2O at 600°C and 1.5 kbar

In order to determine the mechanisms of formation and properties of natural hydrosilicate liquids (HSLs), which are formed during the transition from magmatic to hydrothermal mineral formation in granitic pegmatites and rare-metal granites, the formation of HSLs was experimentally studied in the Na₂O-SiO₂-H₂O, Na₂O-Al₂O₃-SiO₂-H₂O, and Na₂O-K₂O-Li₂O-Al₂O₃-SiO₂-H₂O systems at 600°C and 1.5 kbar. It was shown that the sequential extension of composition does not suppress HSL formation in the systems and expands the stability field of this phase. However, HSLs formed in extended chemical systems have different structure and properties: the addition of alumina induces some compression of the structure of the silicate framework of HSLs, which results in a decrease in water content in this phase and probably hinders the reversibility of its dehydration. It was demonstrated that HSL can be formed by the coagulation of silica present in a silica-oversaturated alkaline aqueous fluid. It was supposed that the HSL formed during this process has a finely dispersed structure. It was argued that anomalous enrichment in some elements in natural HSLs can be due to their sorption by the extensively developed surface of HSL at the moment of its formation.     

Hydrosilicate liquids in the system rare-metal granite–Na<Subscript>2</Subscript>O–SiO<Subscript>2</Subscript>–H<Subscript>2</Subscript>O as accumulators of ore components at high pressure and temperature

Experimental investigations in the system rare-metal granite–Na₂O–SiO₂–H₂O with the addition of aqueous solutions containing Rb, Cs, Sn, W, Mo, and Zn at 600°C and 1.5 kbar showed that the typical elements of rare-metal granites (Li, Rb, Cs, Be, Nb, and Ta) are preferentially concentrated in hydrosilicate liquids coexisting with aqueous fluid. The same behavior is characteristic of Zn and Sn, the minerals of which are usually formed under hydrothermal conditions. In contrast, Mo and W are weakly extracted by hydrosilicate liquids and almost equally distributed between them and aqueous fluids. Liquids similar to those described in this paper are formed during the final stages of magmatic crystallization in granite and granitepegmatite systems. The formation of hydrosilicate liquids in late magmatic and postmagmatic processes will be an important factor controlling the redistribution of metal components between residual magmatic melts, minerals, and aqueous fluids and, consequently, the mobility of these components in fluid-saturated magmatic systems enriched in rare metals.