Beta track autoradiography and infrared spectroscopy bearing on the solubility of CO2 in albite melt at 2 GPa and 1450° C

Albite glasses with 0.5–2.0 wt.% CO₂ synthesized at 2 GPa and 1450° C, previously analyzed by beta track autoradiography (Tingle 1987), have been analyzed by infrared spectroscopy. Values of CO₂ concentration determined by the beta track technique are 10–50% greater than the amount of CO₂ added to the sample. Values determined by infrared spectroscopy match the loaded CO₂ contents, substantiating the existence of a previously unrecognized error in the beta track technique. Concentrations, and hence solubilities, in albite determined by infrared spectroscopy are more accurate than those determined by beta track autoradiography. Values obtained by the beta track technique for diopside are 10% lower than the loaded CO₂ contents (Tingle 1987) and suggest that the systematic error may be dependent on composition and/or density.     

Temperature dependence of CO2 solubility in high pressure quenched glasses of diopside composition

The temperature dependence of carbon dioxide solubility in glasses of diopside composition, quenched from 20 kbar, has been investigated using a combination of high-temperature mass spectrometry and Raman spectroscopy.CO₂-charged diopside glasses were synthesized in a piston-cylinder apparatus. Because of diffusion of hydrogen through the platinum capsules, significant amounts of H₂O, CH₄ and CO were detected along with CO: in the diopside glasses. All three carbon species show a bimodal release pattern in the mass pyrograms. The CO₂ solubility shows a linear and negative temperature dependence. We do not observe any maxima in the solubility curve as was reported previously (Mysen and virgo, 1980a).None of the additional bands observed in Raman spectra of CO₂-charged diopside glasses compared to those in the spectrum of diopside glass can be assigned to molecular CO₂. These bands are caused by CO^?2₃ ions and indicate that the physical solubility of molecular carbon dioxide is negligible. The bimodal release pattern observed for CO₂ in the mass pyrograms, is consistent with the Raman data which strongly suggests that CO^?2₃ ions are present in at least two distinct sites in the glass.     

Argon and CO2 on the race track in silicate melts: A tool for the development of a CO2 speciation and diffusion model

We have analysed the kinetics of Argon and CO₂ diffusion in simplified iron free rhyolitic to hawaiitic melts using the diffusion couple technique. The concentration distance profiles of Ar and CO₂ were measured with electron microprobe analysis and Fourier Transform Infrared Spectroscopy, respectively. Error functions were fitted to the symmetrical concentration distance profiles to extract the diffusion coefficients.In the temperature range 1373 to 1773 K the activation energies for Ar diffusion range from 169 ± 20 to 257 ± 62 kJ mol⁻¹. Ar diffusivity increases exponentially with the degree of depolymerisation. In contrast, the mobility of total CO₂, that is identical to Ar mobility in rhyolitic melt, keeps constant with changing bulk composition from rhyolite to hawaiite. CO₂ speciation at 1623 K and 500 MPa was modeled for the range of compositions studied using the diffusion data of Ar and total CO₂ in combination with network former diffusion calculated from viscosity data. Within error this model is in excellent agreement with CO₂ speciation data extrapolated from temperatures near the glass transition temperature for dacitic melt composition. This model shows that even in highly depolymerised hawaiitic and tholeiitic melts molecular CO₂ is a stable species and contributes 70 to 80% to the total CO₂ diffusion, respectively.     

Solution of H2O in diopside melts: a thermodynamic model

Structural similarities between dry diopside melt and superhydrous albite melt (X _w >0.5) — both lack three-dimensional silicate units — suggest that thermodynamic relations may be similar. A model based on that assumption successfully predicts diopside melting relations and H₂O solubilities. For the model, the three partial differential equations describing solution of H₂O in albite melt for X _w >0.5 have been integrated for diopside melt from X _w =0 to X _w at least as large as 0.76, with two exceptions: an alternative partial differential equation for Henrian solution of H₂O in dilute melts was applied for X _w <0.20, and an alternative differential equation for the pressure dependence of a _w at pressures below 2 kbar was developed. The latter alternative equation yields relatively small ¯V_w's at low pressures rather than the large ¯V_w's calculated from the equation from the albite system. Available experimental solubility data are not precise enough to offer a choice between the small-¯V_w and large-¯V_w equations. Integration of all the partial differential equations was constrained solely by the P and T of a single experimentally-determined point on the H₂O-saturated solidus.Solubilities calculated by a Henrian-analogue solution model (a _di=X _di ² ) from the experimental H₂O saturated solidus lie outside experimental solubility constraints for dilute melts. On the other hand, a Henrian model (a _di=X_di) successfully predicts solubilities in dilute melts. The formulation of the Henrian model and magnitudes of model molar entropies of solution are consistent with the hypothesis that H₂O dissolves in diopside melt as an essentially undissociated species with little ordering on melt structural sites. That species could in turn be consistently, if not uniquely, interpreted to be molecular H₂O or a hydroxylation (OH^–) complex formed from nonbridging oxygens.     

Phase-equilibrium geobarometers for silicic rocks based on rhyolite-MELTS. Part 1: Principles,procedures, and evaluation of the method

Constraining the pressure of crystallization of magmas is an important but elusive task. In this work, we present a method to derive crystallization pressures for rocks that preserve glass compositions (either glass inclusions or matrix glass) representative of equilibration between melt, quartz, and 1 or 2 feldspars. The method relies on the well-known shift of the quartz–feldspar saturation surface toward higher normative quartz melt compositions with decreasing pressure. The critical realization for development of the method is the fact that melt, quartz and feldspars need to be in equilibrium at the liquidus for the melt composition. The method thus consists of calculating the saturation surfaces for quartz and feldspars using rhyolite-MELTS over a range of pressures, and searching for the pressure at which the expected assemblage (quartz+1 feldspar or quartz+2 feldspars) is found at the liquidus. We evaluate errors resulting from uncertainties in glass composition using a series of Monte Carlo simulations for a quartz-hosted glass inclusion composition from the Bishop Tuff, which reveal errors on the order of 20–45 MPa for the quartz+2 feldspars constraint and on the order of 25–100 MPa for the quartz+1 feldspar constraint; we suggest actual errors are closer to the lower bounds of these ranges. We investigate the effect of fluid saturation in two ways: (1) By applying our procedure over a range of water contents for three glass compositions; we show that the effect of fluid saturation is more important at higher pressures (~300 MPa) than at lower pressures (~100 MPa), but reasonable pressure estimates can be derived irrespective of fluid saturation for geologically relevant H₂O concentrations >3 wt% and (2) by performing the same type of pressure determinations with a preliminary version of rhyolite-MELTS that includes a H₂O–CO₂ mixed fluid phase; we use a range of H₂O and CO₂ concentrations for two compositions characteristic of early-erupted and late-erupted Bishop Tuff glass inclusions and demonstrate that calculated pressures are largely independent of CO₂ concentration (for CO₂ <1,000 ppm), at least for relatively high H₂O contents, as expected in most natural magmas, such that CO₂ concentration can be effectively neglected for application of our method. Finally, we demonstrate that pressures calculated using the rhyolite-MELTS geobarometer compare well with those resulting from H₂O–CO₂ glass inclusion barometry and Al-in-hornblende barometry for an array of natural systems for which data have been compiled from the literature; the agreement is best for quartz-hosted glass inclusions, while matrix glass yields systematically lower rhyolite-MELTS pressures, suggestive of melt evolution during eruptive decompression.     

Ion dynamics studies of liquid and glassy silicates,and gas-in-liquid solutions

We give a brief review of ion dynamics studies of liquid and glassy states of SiO₂ and silicate colutions which have been carried out in recent years in this laboratory. We summarize studies on SiO₂, Na⁺ migration in Na₂O·SiO₂ in the “glassy state”, and ionic coordination in multicomponent framework silicates. We present new results on the coordination of Al³⁺ in albite as a function of pressure and show that it is consistent with results of laboratory studies on albite glasses formed at high pressure. We compare calculated PVT data for jadeite, albite and diopside and relate the behavior of the low pressure compressibility to the spinodal limit at negative pressures. Some preliminary studies of inert gas solution in jadeite and of CO₂ solution in a glass having a composition of approximately Na₂O·3SiO₂ are described.     

Experimental study of magmatic melt oxidation by CO2

The process of CO₂ flashing through hydrous albite-hedenbergite melt was experimentally examined at a temperature of 1100°C and a pressure of 2 kbar. Carbon dioxide was generated when the melt interacted with calcite, and wollastonite was the predominant synthesized phase. Mafic components were introduced into the hydrous albite melt via the dissolution of natural hedenbergite. Raman spectroscopic data on bubbles of the fluid phase in the quench glass indicate that the CO₂/H₂O proportions of the bubbles vary. IR spectroscopic data on the glass prove that the water concentration after CO₂ flashing decreased from 5.5 to approximately 3 wt %. The comparison of the composition of the recrystallized clinopyroxene in contact with melt (with and without CO₂ blowing) indicates that CO₂ oxidizes Fe in the melt. The redox effect of CO₂ is quantified by the empirical clinopyroxene tool for metering oxygen fugacity (oxometer), which was calibrated based on experimental data. The oxygen fugacity in our experiments with CO₂ flashing (estimated by the clinopyroxene oxometer) was NNO + (3.0?C3.5). Our estimates with the application of the clinopyroxene oxometer indicate that the maximum oxygen fugacity in the magmatic chambers of Vesuvius and Stromboli volcanoes (which are bubbled with CO₂) is also close to NNO + (3.5 ± 0.5).     

Electron microprobe measurement of pyroxenes coexisting with H2O-undersaturated liquid in the join CaMgSi2O6-Mg2Si2O6-H2O at 30 kilobars,with applications to geothermometry

Mixtures of synthetic crystalline enstatite and diopside were reacted with small water contents in sealed capsules in piston-cylinder apparatus at 30 kb between 1000° C and 1700° C. The compositions of coexisting enstatite and diopside solid solutions were measured with an ARL-EMX electron microprobe between 1000° C and 1500° C. Between 1100° C and 1500° C the pyroxenes coexisted with H₂O-undersaturated liquid which quenched to inhomogeneous pyroxene crystals. The presence of liquid facilitated growth of pyroxene crystals suitable for microprobe determinations. The solvus of Davis and Boyd (1966) is generally used in geothermometry; our enstatite solvus limb is a few mol-% richer in Mg₂Si₂O₆ in the temperature range 1000–1400° C; our diopside solvus limb is a few mol-% richer in Mg₂Si₂O₆ below 1100°C, in close agreement between 1100° C and 1200° C, but richer in CaMgSi₂O₆ between 1200° C and 1500° C. Estimated equilibration temperatures for a diopside with composition 78.7% Di is 1300° C according to our results compared with 1210° C for the Davis and Boyd solvus.     

Experimental determination of argon solubility in silicate melts: An assessment of the effects of liquid composition and temperature

The argon solubility of 38 liquids in the system Na₂O-CaO-MgO-Al₂O₃-SiO₂ (NCMAS) has been determined at 1873 K and 1 bar, the argon concentration of presaturated glasses being measured using a static mass spectrometer. For compositions in the subsystem diopside (CaMgSi₂O₆), nepheline (NaAlSiO₄), albite (NaAlSi₃O₈), anorthite (CaAl₂Si₂O₈), argon solubility is generally a linear function of the relative proportion of each end member, solubility being lowest in diopside melt (1.53 10⁻⁵ cm³ STP · g⁻¹ · bar⁻¹) and highest in albite melt (2.88 10⁻⁴ cm³ STP · g⁻¹ · bar⁻¹). For the tectosilicate joins studied (SiO₂-Na₂Al₂O₄, SiO₂-CaAl₂O₄, SiO₂-MgAl₂O₄) solubility decreases with decreasing silica content in all cases, being highest for Na-bearing liquids and lowest for Mg-bearing liquids at constant molar silica content. Where comparison is possible our results are in good agreement with data from the literature. When our data are considered in isolation we find that argon solubility shows an excellent correlation with calculated ionic porosity. The covariation of argon solubility and liquid density is also reasonable, that with molar volume less convincing and that with polymerization state (as defined by the ratio of the number of nonbridging oxygens and tetrahedral network forming cations; NBO/T) nonexistent. However, when our data are combined with those from the literature no well constrained correlation between argon solubility and ionic porosity is apparent. Based upon this observation and consideration of the temperature dependence of noble gas solubility it is concluded that ionic porosity is not a universally applicable parameter which may be used to predict noble gas solubility as a function of composition, temperature and pressure. Two new models for calculating argon solubility are proposed, both employing the notion of partial molar argon solubilities. The first uses oxide components, for which partial molar argon solubility is directly proportional to partial molar ionic porosity calculated at 1873 K, irrespective of the temperature of experimental equilibration. The second model, which offers the best fit to the available data, employs tetrahedral units rather than oxides as the proposed melt components. This latter model successfully accounts for reported argon solubilities in simple Al-free systems, in simple Al-bearing systems and in natural liquids. This is interpreted to infer that argon is incorporated in large sites in the liquid structure (such as the space within rings of n-tetrahedra) although further work is required to understand the quantitative links between melt structure and noble gas solubility.     

Calcite solubility in supercritical CO2H2O fluids

An extraction-quench apparatus was used to measure calcite solubilities in supercritical CO₂H₂O mixtures. Experiments were conducted at 1 kbar and 2 kbar, between 240°C and 620°C and from X_CO2 = .02 toX_CO2 = .15 in order to determine the solubility behavior as a function of pressure, temperature and CO₂ content. The results indicate that calcite solubilities under these conditions behave similarly to previously investigated calcite solubilities at lower pressures and temperatures (SHARP and Kennedy, 1965). At constant X_CO2, the solubility increases with increasing pressure, but it decreases with increasing temperature. When the temperature and pressure are constant, the calcite solubility rises with increasing X_CO2 to a maximum value at X_CO2 between 0.02 and 0.05. For higher CO₂ contents, up to X_CO2 = .15, the calcite solubility decreases, probably due to the decrease of H₂O activities to values significantly below unity.The solubility behavior can be successfully modeled by making the assumption that Ca⁺⁺ is the dominant calcium species and that the carbon-bearing species are CO_2(aq) and HCO⁻₃. Since for these dilute H₂OCO₂ fluids, all activity coefficients can be assumed to not differ significantly from unity, ionization constants for the reaction H₂O + CO_2(aq) H⁺ + HCO⁻₃ can be calculated at 1 and 2 kbar between 250°C and 550°C. These calculated values are in good agreement with the low temperature determinations of the ionization constants for this reaction determined by Read (1975). Values of the molal Gibbs free energy of CO_2(aq) obtained in our study exhibit a much greater positive departure from ideality than those calculated with the modified Redlich-Kwong equations of either Flowers (1979) or Kerrick and Jacobs (1981) for dilute CO₂ aqueous solutions.     

CO2 in haplo-phonolite melt: solubility, speciation and carbonate complexation

CO₂ solubility was measured in a synthetic iron-free phonolite (haplo-phonolite) by equilibrating melt with excess CO₂ fluid in a piston cylinder apparatus for a range of pressures (1.0- 2.5 GPa) and temperatures (1300 to 1550°C). The quenched glasses were then analysed using a bulk carbon analytical method (LECO). The measured solubilities are between 0.65 and 2.77 wt.% for the range of conditions studied and show a negative correlation with temperature as reported for most other silicate melt compositions.A range of carbonate species are present within the glass, as well as minor amounts of molecular CO₂. FTIR and NMR analyses suggest that carbonate is present as both ‘network’ and ‘depolymerised’ units as shown for relatively highly polymerised compositions in the model of Brooker et al. (2001b). The bulk CO₂ analyses were used to calibrate the IR extinction coefficient for the carbonate groups. However, the results show that the values obtained for the glasses vary with the melt equilibration conditions, presumably because the ratio of the different carbonate species changes as a complex function of run pressure, temperature and quench rate. Thus the use of IR may not be a reliable method for the quantification of dissolved CO₂ concentrations in natural glasses of ‘intermediate’ composition.     

Polymerization of silicate and aluminate tetrahedra in glasses,melts, and aqueous solutions—V. The polymeric structure of silica in albite and anorthite composition glass and the devitrification of amorphous anorthite

²⁹Si MAS NMR experiments have been carried out to determine the silica species distribution (Q distribution) in albite, NaAlSi₃O₈, and anorthite, CaAl₂Si₂O₈, composition glasses (designated albite and anorthite glass). Our results indicate that the Q distribution of albite glass contains all five possible silica species and shows a tendency towards high Q₃ and Q₄ concentrations, whereas anorthite glass does not contain Q₄ and has a high Q₀ concentration. Rationalizations are made in terms of the observed Q distributions to explain differences in devitrification behavior of these two glasses. ²⁷Al MAS NMR data for these glasses suggest that differences in devitrification behavior between these two glasses should be ascribed to small growth rates rather than small nucleation rates of crystalline albite from albite glass.     

Diffusion-controlled growth of albite and pyroxene reaction rims

The growth rates of albite and pyroxene (enstatite + diopside + spinel) reaction rims were measured at 1000°C and ˜700 MPa and found to be parabolic indicating diffusion-controlled growth. The parabolic rate constants for the pyroxene (+ spinel) rims in samples with 0.5 wt% H₂O added or initially vacuum dried at 25°C and 250°C are 1.68 ± 0.09, 0.54 ± 0.05 and 0.25 ± 0.06 μm²/h, respectively. The values for albite rim growth in samples initially dried at 60°C and with 0.1 wt% H₂O added are 0.25 ± 0.04 and 0.33 ± 0.03 μm²/h, respectively. The latter values were used to derive the product of the grain boundary diffusion coefficient D′_A, where A = SiO₂, NaAlO₂, or NaAlSi₋₁, and the grain boundary thickness δ in albite. The calculated D′_SIO2δ in the albite aggregate for the situations of two different water contents are about 9.9 × 10⁻²³ and 1.4 × 10⁻²² m³ s⁻¹, respectively. Both the rate constants and the calculated D′_Aδ demonstrate that the effect of water content on the grain boundary diffusion rate in monomineralic albite and polymineralic pyroxene (+ spinel) aggregates is small, consistent with recent studies of monomineralic enstatite and forsterite rims. Received: 1 July 1995 / Accepted: 1 August 1996     

Microthermometrical and chemical studies of fluid inclusions in minerals from Alpine veins from the penninic rocks of the central and western Tauern Window (Austria/Italy)

The fluid inclusions in samples of quartz, apatite, epidote, diopside, beryl and phenakite from Alpine veins in gneisses, amphibolites and mica schists from the western Tauern Window were analysed by microthermometrical, chemical and neutron activation methods. The inclusions of the eclogites contain a high density CO₂ phase without optically detectable H₂. In the Greiner Schieferserie the fluid inclusions show high CO₂/H₂O ratios and low salt contents. In the Zentralgneis area inclusions with low CO₂/H₂O ratios and high salt contents are typical. In the calcareous mica schists of the lower Schieferhülle, in the eastern part of the investigated area, generally no CO₂ could be detected in the inclusions. These inclusions contain aqueous solutions showing a low salt content. The only CO₂ bearing inclusions observed here were in the graphite-rich rocks of the so-called Habachzungen and in the eclogites from south of the Großvenediger. Trapping pressures estimated from the fluid inclusions are up to 7.5 kbar in the eclogites, but in general the pressures are between 2 and 4 kbar. These pressure data are in good agreement with the pressure data of mineral equilibria. The chemically analysed elements in the fluid inclusions are Na, K, Cs, Mg, Ca, Mn, As, Cl and Br. From the K/Na ratios temperatures between 435 and 490°C can be deduced. The very low Cl/Br ratios (<110) suggest that the dissolved elements came from the country rocks. The alkali/chlorine ratios (～1) indicate that the positive loadings of the cations are balanced by Cl.     

Causes and consequences of errors in determining sorption capacity of coals for carbon dioxide at high pressure

Recent comparisons of CO₂ sorption by coals at high pressures have shown major differences between the results obtained by different laboratories. These need to be resolved for laboratory estimation of CO₂ sequestration in coal seams to be useful. A compilation of potential sources of error in determination of sorption characteristics and their impact on sorption measurements is provided here. A series of tests is also provided that can be used to identify and reduce such errors in measurement. For example, an error in temperature produces a characteristic distortion of the sorption curve for carbon dioxide, which can be corrected to some extent. A negative value for excess sorption at high pressure is almost certainly diagnostic of either a cell volume that has been overestimated or that some part of the substrate that is inaccessible to the gas is accessible to helium. The major source of variation between results from the different laboratories that supplied the closest sorption values was found to be variations in the assumed free space volume, which could be due to discrepancies in determined helium density or measured cell volume. Including a term in the sorption model that is proportional to gas density will markedly reduce the influence of such errors in estimating sorption capacity or heats of sorption. The influence of swelling and moisture on sorption isotherms is also quantified here. Correction for swelling of coals in carbon dioxide changes the estimated sorption capacity by less than 1%, if a term that is proportional to gas density is included as a free parameter in the model fitting the isotherm.     

Skeletal crystallization and residual glass compositions in a cellular alkalic pyroxenite nodule from Oldoinyo Lengai

The alkalic pyroxenite nodule consists of megacrysts of diopside, apatite, perovskite and titanomagnetite in a groundmass consisting of diopside, apatite, titanomagnetite, nepheline, melilite, garnet and vishnevite crystals of various shapes, including previously undescribed skeletal and dendritic shapes, together with vesicles and residual glass. The residual glass is poor in SiO₂ (38–40 wt%), and extraordinarily rich in Na₂O (12.8–15 wt%), SO₃ (1–1.5 wt%), and Cl (0.25–0.7 wt%), as a result of rapid, non-equilibrium crystallization of groundmass phases from a CO₂-rich nephelinite melt.The Oldoinyo Lengai alkalic carbonatite lavas do not represent extreme products of the fractional crystallization of pyroxene, wollastonite, nepheline and alkali feldspar from the carbonated nephelinite melt. The most likely connection between the carbonatite and silicate magma types is one of liquid immiscibility, probably involving phonolite melt.     

Solubility of carbon dioxide in melts of andesite,tholeiite, and olivine nephelinite composition to 30 kbar pressure

Carbon dioxide solubilities in H₂O-free hydrous silicate melts of natural andesite (CA), tholeiite (K 1921), and olivine nephelinite (OM1) compositions have been determined employing carbon-14 beta-track mapping techniques. The CO₂ solubility increases with increasing pressure, temperature, and degree of silica-undersaturation of the silicate melt. At 1650° C, CO₂ solubility in CA increases from 1.48±0.05 wt % at 15 kbar to 1.95±0.03 wt % at 30 kbar. The respective solubilities in OM1 are 3.41±0.08 wt % and 7.11±0.10 wt %. The CO₂ solubility in K1921 is intermediate between those of CA and OM1 compositions. At lower temperatures, the CO₂ contents of these silicate melts are lower, and the pressure dependence of the solubility is less pronounced. The presence of H₂O also affects the CO₂ solubility (20–30% more CO₂ dissolves in hydrous than in H₂O-free silicate melts); the solubility curves pass through an isothermal, isobaric maximum at an intermediate CO₂/(CO₂+H₂O) composition of the volatile phase. Under conditions within the upper mantle where carbonate minerals are not stable and CO₂ and H₂O are present a vapor phase must exist. Because the solubility of CO₂ in silicate melts is lower than that of H₂O, volatiles must fractionate between the melt and vapor during partial melting of peridotite. Initial low-temperature melts will be more H₂O-rich than later high-temperature melts, provided vapor is present during the melting. Published phase equilibrium data indicate that the compositional sequence of melts from peridotite +H₂O+CO₂ parent will be andesite-tholeiite-nephelinite with increasing temperature at a pressure of about 20 kbar. Examples of this sequence may be found in the Lesser Antilles and in the Indonesian Island Arcs.     

Synthesis of kosmochlor and phase equilibria in the join CaMgSi2O6-NaCrSi2O6

Kosmochlor (NaCrSi₂O₆) was synthesized by the flux method from melts along the join Na₂O·2 SiO₂-Na₂O·Cr₂O₃·4 SiO₂ at 1000° C in air, and isolated by dissolving the glassy matrix with hydrofluoric and perchloric acids. The join CaMgSi₂O₆-NaCrSi₂O₆ was studied at 1 atmosphere in air by the quenching technique at temperatures between 900° and 1450° C, using mixtures of kosmochlor and diopside crystals or diopside glass as starting materials. The phases are diopside solid solution, kosmochlor, spinel (Mg-chromite), eskolaite (Cr₂O₃) and glass. The maximum solubility of kosmochlor in diopside is 24 wt percent at 1140° C, while diopside is not soluble at all in kosmochlor, resulting in the existence of a wide range of immiscibility. Petrologic significance of the results is discussed.     

Calorimetry of silicate melts at 1773 K: measurement of enthalpies of fusion and of mixing in the systems diopside-anorthite-albite and anorthite-forsterite

Transposed-temperature-drop calorimetry, using a Setaram HT 1500 calorimeter, was used to study directly the melting at 1773 K of mixtures of crystalline albite, anorthite, and diopside and of anorthite and forsterite. The enthalpy of albite at 1000–1773 K, starting with both crystalline and glassy samples, was also measured. The results confirm previously measured enthalpies of fusion of albite, diopside and anorthite (Stebbins et al. 1982, 1983; Richet and Bottinga 1984,1986). The new results use thermochemical cycles which completely avoid the glassy state by transforming crystals directly to melts. The enthalpy of fusion of forsterite is estimated to be 89±12 kJ/mol at 1773 K and 114±20 kJ/mol at its melting point of 2163 K. The data allow semiquantitative evaluation of heats of mixing in the molten silicates. Along the Ab-An join, enthalpies of mixing in the liquid at 773 K are the same or somewhat more negative than those in the glass at 986 K, whereas along Ab-Di and An-Di, enthalpies of mixing in the liquid are distinctly more positive than in the glass. These differences correlate with excess heat capacities in the liquids suggested by Stebbins et al. (1984).