I. Kushiro

Melting of hydrous upper mantle and possible generation of andesitic magma: An approach from synthetic systems

Phase equilibria in a portion of the system forsterite-plagioclase (An50Ab50 by weight)-silica-H2O have been determined at 15 kbar pressure under H2O-saturated conditions. The composition of the liquid pertinent to the piercing point forsterite + enstatite solid solution + amphibole + liquid + vapor is similar to that of calc-alkaline andesite. The electron microprobe analysis of the glass coexisting with the above three crystalline phases is very close to that of the piercing point determined by phase assemblage observations; however, the glass near (< 8 μm) forsterite crystals is significantly depleted in the normative forsterite component. With the addition of 10 wt.% KAlSi3O8, the composition of this piercing point becomes even closer to the compositions of calc-alkaline andesites. It is also shown that the liquid coexisting with forsterite and enstatite solid solution remains silica-rich (60–62 wt.%) over a wide (∼ 100°C) temperature range. The present experimental studies support the view that liquids similar in composition to calc-alkaline andesites can be generated by direct partial melting of hydrous upper mantle at least at or near 15 kbar.

Diffusivity of oxygen in jadeite and diopside melts at high pressures

Abstract The diffusivity of oxygen was determined in melts of Jadeite (NaAlSi2O6) and diopside (CaMgSi2O6) compositions using diffusion couples with 18O as a tracer. In the Jadeite melt, the diffusivity of oxygen increases from 6.87 −0.25 +0.28 × 10 −10 cm 2 / sec at 5 Kb to 1.32 ± 0.08 × 10 −9 cm 2 / sec at 20 Kb at constant temperature (1400°C), whereas in the diopside melt at 1650°C, the diffusivity decreases from 7.30 −0.18 0.29 × 10 −7 cm 2 / sec at 10 Kb to 5.28 −0.55 +0.60 × 10 −7 cm 2 / sec at 17 Kb. These results demonstrate that the diffusivity is inversely correlated with the viscosity of the melt. For the jadeite melt, in particular, the inverse correlation is very well approximated by the Eyring equation using the diameter of oxygen ions as a unit distance of translation, suggesting that the viscous flow is rate-limited by the diffusion of individual oxygen ions. In the diopside melt, the activation volume is slightly greater than the molar volume of oxygen ion, indicating that the individual oxygen ion is the diffusion unit. The negative activation volume obtained for the jadeite melt is interpreted as the volume decrease associated with a diffusive jump of an oxygen ion due to local collapse of the network structure.

Anionic Constitution of 1-Atmosphere Silicate Melts: Implications for the Structure of Igneous Melts

A structural model is proposed for the polymeric units in silicate melts quenched at 1 atmosphere. The anionic units that have been identified by the use of Raman spectroscopy are SiO44– monomers, Si2O76– dimers, SiO32– chains or rings, Si2O52– sheets, and SiO2 three-dimensional units. The coexisting anionic species are related to specific ranges of the ratio of nonbridging oxygens to tetrahedrally coordinated cations (NBO/Si). In melts with 2.0 < NBO/Si < ∼ 4.0, the equilibrium is of the type [See equation in the PDF file]. In melts with NBO/Si ∼ 1.0 to 2.0, the equilibrium anionic species are given by [See equation in the PDF file]. In alkali-silicate melts with NBO/Si <~ 1.3 and in aluminosilicate melts with NBO/T < 1.0, where T is (Si + Al), the anionic species in equilibrium are given by [See equation in the PDF file]. In multicomponent melts with compositions corresponding to those of the major igneous rocks, the anionic species are TO2, T2O5, T2O6, and TO4, and the coexisting polymeric units are determined by the second and third of these disproportionation reactions.

A possible effect of melt structure on the Mg-Fe2+ partitioning between olivine and melt

Partitioning of Mg and Fe2+ between olivine and mafic melts has been determined experimentally for eight different synthetic compositions in the temperature range between 1335 and 1425°C at 0.1 MPa pressure and at fo2 ∼1 log unit below the quartz-fayalite-magnetite buffer. The partition coefficient [KD = (Fe2+/Mg)ol/(Fe2+/Mg)melt] increases from 0.25 to 0.34 with increasing depolymerization of melt (NBO/T of melt from 0.25–1.2), and then decreases with further depolymerization of melt (NBO/T from 1.2–2.8). These variations are similar to those observed in natural basalt-peridotite systems. In particular, the variation in NBO/T ranges for basaltic-picritic melts (0.4–1.5) is nearly identical to that obtained in the present experiments. Because the present experiments were carried out at constant pressure (0.1 MPa) and in a relatively small temperature range (90°C), the observed variations of Mg and Fe2+ partitioning between olivine and melt must depend primarily on the composition or structure of melt. Such variations of KD may depend on the relative proportions of four-, five-, and six-coordinated Mg2+ and Fe2+ in melt as a function of degree of NBO/T.

Effect of pressure on the diffusivity of network-forming cations in melts of jadeitic compositions

Abstract The diffusivities of network-forming cations (Si4+, Al3+, Ge4+ and Ga3+) in melts of the jadeitic composition NaAl(Si, Ge)2O6 and Na(Al, Ga)Si2O6 have been measured at pressures between 6 and 20 kbar at 1400°C. The rates of interdiffusion of Si4+-Ge4+ and Al3+-Ge3+ increase with increasing pressure at constant temperature. The results are consistent with the ion-dynamics computer simulations of Jadeite melt by Angell et al. (1982, 1983). The coefficient measured for the Si4+-Ge4+ interdiffusion is between 8 × 10−10 and 2.5 × 10−8 cm 2 sec at 6 kbar, depending on the composition of the melt, whereas at 20 kbar it is between 7 × 10−9 and 2 × 10−7 cm 2 sec . The effect of pressure is greater for more Si-rich compositions (i.e., closer to NaAlSi2O6 composition). The coefficient measured for the Al3+-Ga3+ inter- diffusion is between 9 × 10−10 and 3 × 10−9 cm2/sec at 6 kbar and between 3 × 10−9 and 1 × 10−8 cm 2 sec at 20 kbar. The rate of increase in diffusivity with pressure of Al3+-Ga3+ (a factor of 3–4) is smaller than that of Si4+-Ge4+ (a factor of 7–17). The Si4+-Ge4+ interdiffusion in melts of Na2O · 4(Si, Ge)O2 composition has also been measured at 8 and 15 kbar for comparison. The effect of pressure on the diffusivity in this melt is significantly smaller than that for the jadeitic melts. The increase in diffusivity of the network-forming cations in jadeitic melts with increasing pressure may be related to the decrease in viscosity of the same melt. The present results, as well as the ion-dynamics simulations, suggest that the homogenization of partial melts and mixing of magmas would be more efficient at greater depths.

Pressure dependence of nickel partitioning between forsterite and aluminous silicate melts

Nickel partitioning between forsterite and aluminosilicate melt of fixed bulk composition has been determined at 1300°C to 20 kbar pressure. The value of the forsterite-liquid nickel partition coefficient is lowered from >20 at pressures equal to or less than 15 kbar to <10 at pressures above 15 kbar. n nPublished data indicate that melts on the join Na2O-Al2O3-SiO2 become depolymerized in the pressure range 10–20 kbar as a result of Al shifting from four-coordination at low pressure to higher coordination as the pressure is increased. This coordination shift results in a decreasing number of bridging oxygens in the melt. It is suggested that the activity coefficient of nickel decreases with this decrease in the number of bridging oxygens. As a result, the nickel partition coefficient for olivine and liquid is lowered. n nMagma genesis in the upper mantle occurs in the pressure range where the suggested change in aluminum coordination occurs in silicate melts. It is suggested, therefore, that data on nickel partitioning obtained at low pressure are not applicable to calculation of the nickel distribution between crystals and melts during partial melting in the upper mantle. Application of high-pressure experimental data determined here for Al-rich melts to the partial melting process indicates that the melts would contain about twice as much nickel as indicated by the data for the low-pressure experiments. If, as suggested here, the polymerization with pressure is related to the Al content of the melt, the difference in the crystal-liquid partition coefficient for nickel at low and high pressure is reduced with decreasing Al content of the melt. Consequently, the change ofDNiol-andesite melt is greater than that ofDNiol-basalt melt, for example.

Experimental studies of condensation processes of silicate materials at low pressures and high temperatures, I. Phase equilibria in the system CaMgSi2O6H2 in the temperature range 1200–1500°C and the pressure range (PH2) 10−6 to 10−9 bar

Phase equilibria in the system CaMgSi2O6H2 in the pressure and temperature ranges 10−9 to 10−6 bar and 1200–1500°C, respectively, have been experimentally determined. The vaporous curve was determined by monitoring evaporation rate and compositional changes as a function of temperature. Crystalline diopside condenses from a gas phase along a vaporous curve with a positive slope of approximately 16°C/log unitPH2 (bars) from about 1360°C atPH2 = 10−9 bar to 1388°C at 3 × 10−7 bar. Partially evaporated diopside at temperatures above its vaporous shows but systematic increase in Ca/Mg with time (and degree of evaporation). There were no chemical changes with time (and degree of evaporation) of diopside within its own stability field. At higher pressures, a liquid condenses from the vapor, with a slope of the liquid-vapor curve of about 50°C/log unitPH2 (bars). The liquidus temperature decreases from 1388°C near the triple point to about 1350°C atPH2 = 10−6 bar and is nearly vertical toPH2 = 1 bar. The Ca/Mg of the liquid exceeds 1 indicating incongruent melting behavior although no additional crystalline phases were observed. Liquid diopside probably also evaporates incongruently. n nProvided that the phase equilibria of the diopside composition mimic those of other refractory phases condensing from the early solar nebula and that only H2 gas was present, the experimental results indicate that the pressures in the nebula during condensation of crystals from vapor were significantly lower ( 10−4 bar). Alternatively, if the higher pressure estimates are correct, liquid silicate will condense from the vapor, and vapor-liquid and crystal-liquid equilibria were more important during the petrogenetic processes resulting in aggregated silicate materials than suggested previously.