Experimental determination of activities in FeTiO3-MnTiO3 ilmenite solid solution by redox reversals

 Solid solutions of (Fe,Mn)TiO₃ were synthesized, mostly at 0.10 X_Mn intervals, at 1 bar, 900°C and log f _{O 2} = –17.50. Analysis by EMP indicate an ideal stoichiometry for the Fe-Mn ilmenites with (Fe+Mn) = Ti = 1.000 when normalized to 3 oxygens. Their unit cell volume increases linearly with X_Mn. The composition of Fe-Mn ilmenite coexisting with metallic Fe and rutile was reversed at 1 bar, 700–900°C and fixed f _{O 2} in a gas-mixing furnace. Oxygen fugacity was controlled by mixing CO₂ and H₂ gas and was continuously monitored with an yttrium-stabilized zirconia electrolyte. Solution properties of Fe-Mn ilmenite were derived from the experimental data by mathematical programming (Engi and Feenstra, in preparation) including notably the results of Fe-Mn exchange experiments between ilmenite and garnet (Feenstra and Engi, submitted) and anchoring the standard state properties to the updated thermodynamic dataset of Berman and Aranovich (1996). The thermodynamic analysis resulted in positive deviations from ideality for (Fe,Mn)TiO₃ ilmenite, which is well described by an asymmetric Margules model with W^H _FeFeMn = –9.703 and W^H _FeMnMn = –23.234 kJ/mol, W^S _FeFeMn = –19.65 and W^S _FeMnMn = –22.06 J/(K·mol). The excess free energy for Fe-Mn ilmenite derived from the redox reversals is larger than in the symmetric ilmenite model (W^G _FeMn = +2.2 kJ/mol) determined by O'Neill et al. from emf measurements on the assemblage iron-rutile-(Fe,Mn)ilmenite. Received: 10 January 1996 / Accepted: 11 July 1996     

An empirical model for the calculation of spinel-melt equilibria in mafic igneous systems at atmospheric pressure: 2. Fe-Ti oxides

In order to develop models simulating the crystallization of Fe-Ti oxides in natural lavas, we have processed published experimental data on magnetite-melt and ilmenite-melt equilibria. These data include 62 Mt-melt and 75 Ilm-melt pairs at temperatures 1040–1150 °C, oxygen fugacities from IW to NNO+2, and bulk compositions ranging from ferrobasalts to andesites and dacites. Five major cations (Fe³⁺, Fe²⁺, Ti⁴⁺, Mg²⁺ and Al³⁺) were considered for the purpose of describing Fe-Ti oxide saturation as a function of melt composition, temperature and oxygen fugacity at 1 atmosphere pressure. The empirically calibrated mineral-melt expression based on multiple linear regressions is: ln D _i  = a/T + blog f _O2 + c + d ₁ X _Na + d ₂ X _K + d ₃ X _P, where D _i represents molar distribution coefficients of the given cations between Mt/Ilm and melt; X _Na, X _K, and X _P are the molar fractions of Na, K, and P in the melt. The empirically calibrated Mt-melt and Ilm-melt equilibria equations allowed us to develop two models for calculating crystallization temperatures of the Fe-Ti oxides in the melts with an accuracy of 10–15 °C, and compositions with an accuracy of 0.5–2 mol%. These models have been integrated into the COMAGMAT-3.5 program, improving our ability to study numerically the effects of temperature and oxygen fugacity on the stability and phase equilibria of Fe-Ti oxides. Application of this approach to the tholeiitic series of Chazhma Sill from Eastern Kamchatka (Russia) indicates oxygen fugacity conditions near NNO + 0.5. Numerical simulation of fractional crystallization of an iron-enriched basaltic andesite parent at these oxidizing conditions accurately reproduces the FeO-SiO₂ relations observed in the Chazhma suite. Received: 3 March 1998 / Accepted: 7 August 1998     

Partitioning of Fe and Mn between ilmenite and olivine at 1100 °C: constraints on the thermodynamic mixing properties of (Fe, Mn)TiO3 ilmenite solid solutions

The partitioning of Fe²⁺ and Mn²⁺ between (Fe, Mn)TiO₃ and (Fe, Mn)₂SiO₄ solid solutions in the system FeO-MnO-TiO₂-SiO₂ has been experimentally investigated at 1100 ^∘C and pressures of 1 bar and 25 kbar, over a wide range of Fe/Mn ratios, using electron microprobe analysis of quenched run products. The ilmenite solid solution in this system is within analytical uncertainty a simple binary between FeTiO₃ and MnTiO₃, but the olivine solid solution appears to contain up to 2.5 wt% TiO₂. The Fe-Mn partitioning results constrain precisely the difference in the thermodynamic mixing properties of the two solid solutions. If the mixing properties of (Fe, Mn)₂SiO₄ solid solutions are assumed to be ideal, as experimentally determined by Schwerdtfeger and Muan (1966), then the ilmenite is a regular, symmetric solution with W ^ilm _Fe-Mn=1.8±0.1 kJ mol⁻¹. The quoted uncertainty does not include the contribution from the uncertainty in the mixing properties of the olivine solution, which is estimated to be ±1.8 kJ mol⁻¹, and which therefore dominates the uncertainty in the present results. Nevertheless, this result is in good agreement with the previous experimental study of O'Neill et al. (1989), who obtained W ^ilm _Fe-Mn=2.2±0.3 kJ mol⁻¹ from an independent method. The results provide another item of empirical evidence supporting the proposition that solid solutions between isostructural end-members, in which order-disorder effects are not important, generally have simple thermodynamic mixing properties, with little asymmetry, modest excess entropies, and excess enthalpies approximately proportional to the difference in the molar volumes of the end-members. Received: 11 February 1998 / Accepted: 29 June 1998     

Experimental study of the stability of cordierite and garnet in pelitic compositions at high pressures and temperatures

The stability of cordierite and garnet has been studied experimentally in complex, silica oversaturated compositions (in the systems MgO-FeO-Al₂O-CaO₃-Na₂O-K₂OSiO₂) in which the molecular ratio Al₂O₃/FeO+MgO<1. Compositions with 100 Mg/Mg+Fe²⁺ ratios (X) of 0, 30, 50, 70 and 100 have been used to investigate the role of this ratio in determining phase assemblages and P, T coordinates of reactions. The minimum pressure for appearance of garnet at a given temperature is strongly dependent on X _{total rock}.The X-values of co-existing phases (chiefly garnet, cordierite, hypersthene) in divariant equilibrium are a function of temperature and pressure and have been experimentally determined at 900° C, 1000° C and 1100° C. At high temperature (>1050° C) the phases sapphirine and spinel are stable with quartz in Mg-rich and Fe-rich compositions respectively. Experiments in the system MgO-FeO-Al₂O₃-SiO₂ show that for a given X-value and temperature the pressure required to produce Ca-free garnet from hypersthene-cordierite assemblages is 1–2 kb greater than that required to produce garnet containing 6±2 mol percent grossular solid solution in the more complex Ca-bearing system.     

The influence of crystal field stabilization energy on Fe2+ partitioning in paragenetic minerals

In order to explore possible quantitative relations between crystal field stabilization energy, CFSE, and partitioning behaviour of the 3d ⁶-configured Fe²⁺ ion, a suite of 29 paragenetic rock-forming minerals from 12 high-grade metamorphic rock samples of the Ukrainian shield, including the parageneses garnet/orthopyroxene/clinopyroxene (2x), orthopyroxene/clinopyroxene, garnet/clinopyroxene, garnet/orthopyroxene/biotite, garnet/biotite, garnet/cordierite, garnet/cordierite/biotite, garnet/orthopyroxene/clinopyroxene/Ca-amphibole, Ca-amphibole/biotite (retrograde), was studied by electron microprobe analysis to obtain the respective K _D Fe^{2+ (Ph1/Ph2)} values and by polarized single crystal electronic absorption spectroscopy to evaluate the respective CFSE_Fe2+ values. Other than in the case of Cr³⁺, a clear quantitative relation between K _D ^(Ph1/Ph2) and the ΔCFSE^(Ph1/Ph2) was only observed when geometrical factors, mainly the volume of crystallographic sites and ionic radii of ions competing in the partitioning process, are similar in the respective two paragenetic phases to within 15–20%. In such cases, the ΔCFSE_Fe2+ contribution to K _D ^(Ph1/Ph2) amounts to 0.1 to 0.2 log K _D per 100 cm⁻¹ΔCFSE. The conclusion is that ΔCFSE_Fe2+ plays only a secondary role after geometrical factors, in the partitioning behaviour of Fe²⁺. The reason for this is seen in the facts that, compared to the 3d ³-configured Cr³⁺ ion, CFSE of the 3d ⁶-configured Fe²⁺ amounts only to 20–25%, and that the former ion enters only octahedral sites with similar geometrical properties in the paragenetic mineral phases. Received: 17 November 1998 / Accepted: 28 June 1999     

Magnetite, ilmenite and ulvite in rocks and ore deposits: petrography, microprobe analyses and genetic implications

Summary ?Detailed petrographic studies and microchemical analyses of titanomagnetite from igneous and metamorphic rocks and ore deposits form the basis of this investigation. Its aim is to compare the data obtained and their interpretations with the experimentally deduced subsolidus oxidation-exsolution model of Buddington and Lindsley (1964). The results are also considered relevant for the interpretation of compositional variations in black sands which are recovered for titanium production. The arrangement of the samples investigated is in accordance with textural stages C1 to C5 caused by subsolidus exsolution with increasing degrees of oxidation (Haggerty, 1991). Stage 1 is represented by two types of optically homogeneous TiO₂-rich magnetite: a. An isotropic type considered to represent solid solutions of magnetite and ulvite containing between 5.2 to 27.5 wt% TiO₂ corresponding to about 14.7 to 77.7 mol% Fe₂TiO₄ in solid solution with magnetite. The general formula of this type is Fe²⁺ _1+x Fe³⁺ _2−2x Ti _x O₄ (x = 0.0–1.0). b. The second type which has not been reported so far is anisotropic and shows complex internal twinning resembling inversion textures. It is thus attributed to inversion of a high-temperature ilmenite modification (with statistical distribution of the cations) which forms solid solutions with magnetite. TiO₂ varies between 9.3 and 24.5 wt% corresponding to about 17.2 to 43.6 mol% ilmenite in solid solution with magnetite. This type is interpreted as a cation-deficient spinel with the general formula Fe²⁺ _{12/12 + 1/4x}Fe³⁺ _{24/12 − 3/2x} □ _{0 + 1/4x} Ti _x O₄ (x = 0.0–16/12). Isotropic and anisotropic homogeneous magnetites occur in volcanic rocks only; the homogeneity of the solid solutions was explained by fast cooling which prevented the development of exsolution textures. Stages 2 and 3 are represented by magnetite with or without ulvite. The magnetite host contains ilmenite lamellae forming trellis and sandwich textures. In contrast to the requirement of the oxidation-exsolution model, the ilmenite lamellae are concentrated exclusively in the cores of the host crystals. The reverse host-guest relationship may also occur. Stages 4 and 5 are identical with thermally generated martite (= martite due to heating). The textures are characterized by very broad lamellae of ferrian ilmenite or titanohematite dominantly concentrated along the margins of the host crystals. Thermally generated martite is restricted to subsolidus-oxidation reactions. The ilmenite lamellae of trellis and sandwich textures contain low Fe₂O₃-concentrations (average 4.8 mol%; to a maximum of 8.3), whereas the Fe₂O₃-content of thermally generated martite is between 32 to 71 mol%. With respect to the Fe₂O₃-concentrations in the ilmenite lamellae, no transition between the two types was observed. The results of this paper show that the widely accepted oxy-exsolution model of Buddington and Lindsley (1964) which is based on experimental results can – with the exception of thermally generated martite – not explain the tremendous variety of magnetite–ilmenite–ulvite relationships in natural rocks and ore deposits. Received October 16, 2001; accepted May 2, 2002     

Hydrothermal synthesis of pumpellyite–okhotskite series minerals

Summary ?Hydrothermal experiments to synthesize pumpellyite group minerals of the pumpellyite–okhotskite series and to investigate their stability have been carried out at 200, 300 and 400 MPa P _fluid and 250–500 °C by using cold-seal pressure vessels and solid buffers of MnO₂–Mn₂O₃, Cu₂O–CuO and Cu₂O–Cu buffer assemblages. Okhotskite and pumpellyite rich in the okhotskite component crystallized from an oxide mixture starting material of Ca₄MgMn³⁺ ₃Al₂Si₆O_24.5-oxide+excess H₂O at P _fluid of 200, 300 and 400 MPa and temperatures of 300 and 400 °C. However, a single phase of okhotskite was not produced, and associated piemontite, hausmannite, wollastonite, clinopyroxene, corundum, braunite–neltnerite solid solution and alleghanyite also formed. Mn-pumpellyite of the okhotskite–pumpellyite join occurs as aggregates of needle crystals, rounded grains or flaky crystals. Chemical compositions are variable and range from pumpellyite-(Mn²⁺) to okhotskite: 31–36 SiO₂, 13–21 Al₂O₃, 12–25 total Mn₂O₃, 0.6–4 MgO and 20–24 wt.% CaO. Reconnaissance experiments using a starting material of synthetic Ca₂Mn³⁺Al₂Si₃O₁₂(OH)-piemontite at 300 MPa and temperatures of 250, 300, 400 and 500 °C indicate that Mn-rich pumpellyite can crystallize from piemontite at lower temperatures than the stability field of piemontite. The Mn-rich pumpellyite was accompanied by garnet, wollastonite and alleghanyite. The chemical compositions of the Mn-pumpellyites are 32–36 SiO₂, 18–27 Al₂O₃, 8–18 total Mn₂O₃ and 20–23 wt.% CaO. This study shows that the stability fields of piemontite, piemontite+Mn-pumpellyite, and Mn-pumpellyite range in this order with decreasing temperature under high fO₂ conditions. The maximum stability temperature of Mn-rich pumpellyite lies between 400 and 500 °C at 200–400 MPa in high fO₂ conditions. Received March 3, 2000; revised version accepted December 28, 2001     

An experimental study of Ca-(Fe,Mg) interdiffusion in silicate garnets

Ca-(Fe,Mg) interdiffusion experiments between natural single crystals of grossular (Ca_2.74Mg_0.15 Fe_0.23Al_1.76Cr_0.04Si_3.05O₁₂) and almandine (Ca_0.21Mg_0.40 Fe_2.23Mn_0.13Al_2.00Cr_0.08Si_2.99O₁₂ or Ca_0.43Mg_0.36Fe_2.11 Al_1.95Si_3.04O₁₂), were undertaken at 900–1100 °C and 30 kbar, and pressures of 15.0–32.5 kbar at 1000 °C. Samples were buffered by Fe/FeO in most cases. Diffusion profiles were determined by electron microprobe. Across the experimental couples the interdiffusion coefficients (D˜) were almost independent of composition. The diffusion rates in an unbuffered sample were significantly faster than in buffered samples. The temperature dependence of the D˜ (Ca-Fe,Mg) interdiffusion coefficients may be described by

Variation of the oxygen content and point defects in olivines, (Fe x Mg 1−x )2SiO4, 0.2 ≤ x ≤ 1.0

 The variation of the oxygen content in olivines, (Fe _x Mg_{1− x} )₂SiO₄, with 0.2 ≤ x ≤ 1.0, was investigated by thermogravimetric measurements. Mass changes occurring upon oxygen activity changes were measured as a function of oxygen activity and cationic composition at 1130 and 1200 °C. During the measurements the samples were in direct contact with gases containing CO, CO₂ and N₂ and, at a few spots at the bottom of the sample stack, also with SiO₂. By fitting experimental data of mass changes to equations derived using point defect thermodynamics, it was shown for olivines with 0.2 ≤ x ≤ 1.0 at 1130 °C and 0.2 ≤ x ≤ 0.7 at 1200 °C within the oxygen activity ranges investigated that the observed variations in the oxygen contents are compatible with cation vacancies and Fe³⁺ ions on M sites and Fe³⁺ ions on silicon sites as majority defects if it is assumed that only three types of point defects occur as majority defects. Different cases were considered, closed systems, taking into account that ξ=[Si]/([Si]+[Fe]+[Mg]) is not necessarily equal to 1/3, and olivines in equilibrium with SiO₂ or pyroxenes. The oxygen content variations observed in this study are significantly smaller than those reported previously in the literature. It is proposed that these differences are related to the dissolution of Fe into noble metal containers used as sample holders in earlier studies and/or to the presence of secondary phases. Received: 1 November 1995 / Accepted: 15 September 2002 Acknowledgements This work was supported by the Cornell Center for Materials Research (CCMR), a Materials Research Science and Engineering Center of the National Science Foundation (DMR-0079992). The authors thank Mr. Daniel M. DiPasquo and Mr. Jason A. Schick for helping in experimental work.     

Fe-rich tholeiitic liquids and their cumulate products in the Pleasant Bay layered intrusion, coastal Maine

Fe-rich tholeiitic liquids are preserved as chilled pillows and as the chilled base of a 27 meter thick macrorhythmic layer in the Pleasant Bay mafic-silicic layered intrusion. The compositions of olivine (Fo₁) and plagioclase (An₁₃₋₈) in these extremely fine grained rocks suggest that they represent nearly end stage liquids that formed by fractionation of tholeiitic basalt. Their major element compositions (∼17.5 wt% FeO_T and 54 wt%SiO₂) closely resemble highly evolved glasses in the Loch Ba ring dike and some recent estimates of end-stage liquids related to the Skaergaard layered intrusion, and are consistent with recent experimental studies of tholeiite fractionation. Their trace element compositions are consistent with extensive earlier fractionation of plagioclase, olivine, clinopyroxene, ilmenite, magnetite and apatite. The mineral assemblage of the chilled rocks (olivine, clinopyroxene, quartz, ilmenite and magnetite), apatite saturation temperatures, and very low Fe³⁺/Fe²⁺indicate conditions of crystallization at temperatures of about 950 °C and f _{O 2} about two log units below FMQ. Cumulates that lie about 3 meters above the chilled base of the macrorhythmic layer contain cumulus plagioclase, olivine, clinopyroxene, ilmenite, apatite and zircon. This mineral assemblage and the Fe-Mg ratio in clinopyroxene cores suggest that this cumulate was in equilibrium with a liquid having a composition identical to that of the chilled margin which lies directly beneath it. The high FeO_T and low SiO₂ concentrations of this cumulate (23.3 and 45.8 wt%, respectively) are comparable to those in late stage cumulates of the Skaergaard and Kiglapait intrusions. This association of a chilled liquid and cumulate in the Pleasant Bay intrusion suggests that late stage liquids in tholeiitic layered intrusions may have been more SiO₂-rich than field-based models suggest and lends support to recent experimental studies of tholeiite fractionation at low f _O2 which indicate that saturation of an Fe-Ti oxide phase should cause FeO_T to decrease in the remaining liquid. Received: 17 January 1997 / Accepted: 10 June 1997     

The CaGeO3–Ca3Fe2Ge3O12 garnet join: an experimental study

Germanate garnets are often used as isostructural analogues of silicate garnets to provide insight into the crystal chemistry and symmetry of the less accessible natural garnet solid solutions. We synthesised two series of germanate garnets at 3 GPa along the join^VIIICa₃^VI(CaGe)^IVGe₃O₁₂–^VIIICa₃^VIFe₂^IVGe₃O₁₂ at 900 °C and 1,100 °C. Samples with compositions close to the CaGeO₃ end-member consist of tetragonal garnet with a small amount of triclinic CaGe₂O₅. Samples with nominal compositions between X_Fe=0.4 and 1.0 consist of a mixture of tetragonal and cubic garnets; whereas, single-phase cubic garnets were obtained for compositions with X_Fe>1.2 (X_Fe gives the iron content expressed in atoms per formula unit, and varies between 0 and 2 along the join). Run products which were primarily single-phase garnet were investigated using Mössbauer spectroscopy. Spectra from samples synthesised at 1,100°C consist of one well-resolved doublet that can be assigned to Fe³⁺ in the octahedral site of the garnet structure. A second doublet, present primarily in samples synthesised at 900°C, can be assigned to Fe²⁺ at the octahedral sites of the garnet structure. The relative abundance of Fe²⁺ decreases with increasing iron content. Transmission electron microscopy analyses confirm this tendency and show that the garnets are essentially defect-free. The unit-cell parameters of tetragonal ^VIIICa₃^VI(CaGe)^IVGe₃O₃ garnet decrease with increasing synthesis temperature, and the deviation from cubic symmetry becomes smaller. Cubic garnets show a linear decrease of unit-cell parameter with increasing iron content. The results are discussed in the context of iron incorporation into ^VIIIMg₃^VI(MgSi)^IVSi₃O₃ majorite.     

Incongruent melting of biotite to spinel in a quartz-free restite at El Joyazo (SE Spain): Textures and reaction characterization

In the Grt-Bt-Sil restitic xenoliths of El Joyazo (Cerro de Hoyazo), hercynitic spinel is a minor phase commonly associated with biotite. The possible reaction relationships among biotite and spinel are studied in reaction textures developed around biotites at their contact with patches of fibrolitic sillimanite and rhyolitic melt. In these textures, resorbed biotite crystals about 1 mm long are rimmed by a layer of glass <200 μm thick containing spinel and ilmenite; the same glass also fills embayments in biotite. Spinel forms euhedral crystals <100 μm in size, and ilmenite occurs as smaller anhedral crystals or needles, often intergrown with spinel. The homogeneous felt-like melt-sillimanite aggregate (“mix”) is richest in glass close to the reaction rim around biotite. Plagioclase and garnet are located >5 mm away from the reaction texture. Biotite is chemically zoned. Cores (Bt ₁ ) have X_Mg=0.35 ± 0.02 and Ti=0.58 ± 0.01 atoms; whereas the outer rims (Bt ₂ ) have X_Mg=0.45 ± 0.01 and Ti up to 0.68 atoms. The hercynite-rich spinel (Spl) has low ZnO content (<0.80 wt%) and X_Mg=0.26 ± 0.04. The chemical compositions of the mix aggregate represent linear combinations between sillimanite and a silica-rich melt. This melt (melt ₁ ) is different from that of the layer around biotite (melt ₂ ), which is also richer in Ca and alkalis. Garnet rims (Grt) have low Ca and Mn, and X_Mg=0.14. Plagioclase is characterized by large homogeneous cores (Pl ₁ , An_31 ± 2) and more calcic rims (Pl ₂ , An_49 ± 6). Matrix analysis in the 9-component (Al-Ca-Fe-K-Mg-Mn-Na-Si-Ti), 9-phase (Bt₁-Bt₂-Grt-Spl-Ilm-melt₂-mix-Pl₁-Pl₂) system provides the mass balance (in mole units):

Elasticity of the pyrope (Mg3Al2Si3O12)-majorite (MgSiO3) garnets solid solution

Dense isotropic polycrystalline specimens of majorite-rich garnets (Py₁₀₀, Py₆₂Mj₃₈, Py₅₀Mj₅₀, Py₂₁Mj₇₉ and Mj₁₀₀) along the pyrope (Mg₃Al₂Si₃O₁₂ = Py₁₀₀)-majorite (MgSiO₃ = Mj₁₀₀) join were fabricated in a 2000-ton uniaxial split-sphere anvil apparatus (USSA-2000) at pressures from 10 to 18.5 GPa and temperatures from 1200 to 1850 °C, within their stability fields in runs of 2–4-h duration, using hot-pressing techniques developed by Gwanmesia et al. (1993). These specimens are single-phased, fine-grained (≤5 mm), free of microcracks, and have bulk densities greater than 99% of the corresponding single-crystal X-ray density. Elastic compressional (P) and shear (S) wave velocities were determined at room pressure and temperature for these polycrystalline garnet specimens by phase comparison ultrasonic interferometry. For Mj₁₀₀, the P and S wave velocities are within 1% of the Hashin-Shtrikman averages calculated from the single crystal elastic moduli measured by Brillouin spectroscopy. Both the elastic bulk modulus (K) and the shear modulus (G) decrease continuously with increasing majorite content from pyrope garnet (Py₁₀₀) to pure majorite garnet (Mj₁₀₀). The compositional dependence of K and G are given by K = 172.3 (40) − 0.085X, and G = 91.6 (10) − 0.038X, where X = mol% majorite), respectively, indicating that substitution of Si for Mg and Al decreases both K and G by about 5% along the solid solution series. Received: 25 March 1999 / Accepted: 12 July 1999     

Fluid flow in marbles at Jervois, central Australia: oxygen isotope disequilibrium and zoning produced by decoupling of mineralogical and isotopic resetting

The Jervois region of the Arunta Inlier, central Australia, contains para- and orthogneisses that underwent low-pressure amphibolite facies metamorphism (P = 200–300 MPa, T = 520–600 °C). Marble layers cut by metre-wide quartz + garnet ± epidote veins comprise calcite, quartz, epidote, clinopyroxene, grandite garnet, and locally wollastonite. The marbles also contain locally discordant decimetre-thick garnet and epidote skarn layers. The mineral assemblages imply that the rocks were infiltrated by water-rich fluids (XCO₂ = 0.1–0.3) at ∼600 °C. The fluids were probably derived from the quartz-garnet vein systems that represent conduits for fluids exsolved from crystallizing pegmatites emplaced close to the metamorphic peak. At one locality, the marble has calcite (Cc) δ¹⁸O values of 9–18‰ and garnet (Gnt) δ¹⁸O values of 10–14‰. The δ¹⁸O(Gnt) values are only poorly correlated with δ¹⁸O(Cc), and the δ¹⁸O values of some garnet cores are higher than the rims. The isotopic disequilibrium indicates that garnet grew before the δ¹⁸O values of the rock were reset. The marbles contain  ≤15% garnet and, for water-rich fluids, garnet-forming reactions are predicted to propagate faster than O-isotopes are reset. The Sm-Nd and Pb-Pb ages of garnets imply that fluid flow occurred at 1750–1720 Ma. There are no significant age differences between garnet cores and rims, suggesting that fluid flow was relatively rapid. Texturally late epidote has δ¹⁸O values of 1.5–6.2‰ implying δ¹⁸O(H₂O) values of 2–7‰. Waters with such low-δ¹⁸O values are probably at least partly meteoric in origin, and the epidote may be recording the late influx of meteoric water into a cooling hydrothermal system. Received: 29 April 1996 / Accepted: 12 March 1997     

Diffusion along interphase boundaries and its effect on retrograde zoning patterns of metamorphic minerals

In this study we use two dimensional chemical patterns and numerical modeling to estimate the relative rates of chemical transport along interphase boundaries (ib) and through grain (s) interiors during retrograde Fe–Mg exchange between garnet and biotite at a garnet–biotite–quartz triple junction. We demonstrate that systematic variations in garnet–rim compositions, which are frequently observed along garnet–quartz interfaces, and deviations from concentric retrograde zoning patterns start to develop when chemical transport along the interphase boundaries becomes slow during cooling. The capacities for chemical transport along an interphase boundary depend on the product D ^ib K ^ib/s a, where D ^ib is the diffusion coefficient of the exchangeable components within the interphase boundary medium, K ^ib/s is the equilibrium partitioning coefficient between the cation exchange partners and the interphase boundary medium and a is the interphase boundary width. The model is applied to analyze the retrograde zoning patterns in garnets from the Mozambique belt (SE-Kenya), which cooled from 820°C at a rate of ca. 2°C/my. It is found that non-equilibrated compositions in garnet along garnet/quartz interphase boundaries started to develop below 700°C due to insufficient rates of chemical transport along these boundaries. The transport capacities of garnet/quartz interphase boundaries was estimated to have been between about 1E-23 cm³/s (575°C) and 1E-20 cm³/s (700°C) from modeling the observed X _Fe pattern in garnet close to a garnet–quartz–biotite triple junction and relying on published data on the diffusivity of Fe²⁺ in garnet. Similar transport capacities are obtained; when the interphase boundary is assumed to be filled with a material that has the transport properties and chemical composition of a free melt in equilibrium with garnet, biotite and quartz at the respective conditions. In contrast, if the transport properties of the interphase boundary medium are related to the diffusivities and solubility of Fe²⁺ and FeOH⁺ within a free aqueous solution, chemical transport along the interphase boundaries would be much more efficient, and exchange equilibrium would have been maintained during the entire cooling history of the rocks. The observation of systematic deviations from local equilibrium along the garnet–quartz interphase boundaries leads us to exclude the presence of an aqueous fluid along the interphase boundary at any time during cooling.     

Biotite and garnet compositional variation and mineral equilibria in Grenville gneisses of the Otter Lake area, Quebec

Garnet-biotite gneisses, some of which contain sillimanite or hornblende, are widespread within the Otter Lake terrain, a portion of the Grenville Province of the Canadian Shield. The metamorphic grade is upper amphibolite to, locally, lower granulite facies. The atomic ratio Fe²⁺/(Fe²⁺+ Fe³⁺) in biotite ranges from 0.79 to 0.89 (ferrous iron determinations in 10 highly pure separates), with a mean of 0.86. Mg and Fe²⁺ atoms occupy 67–78% of the octahedral sites, the remainder are occupied by Fe³⁺, Ti, and Al, and some are vacant. Mg/(Mg + Fe²⁺), denoted X, in the analysed samples ranges from 0.32 to 0.65. Garnet contains 1–24% grossular, 1–12% spessartine and X ranges from 0.07 to 0.34. Compositional variation in biotite and garnet is examined in relation to three mineral equilibria: (I) biotite + sillimanite + quartz = garnet + K-feldspar + H₂O; (II) pyrope + annite = almandine + phlogopite; (III) anorthite = grossular + sillimanite + quartz. Measurements of X (biotite) and X (garnet) are used to construct an illustrative model for equilibrium (I) which relates the observed variation in X to a temperature range of 70°C or a range in H₂O activity of 0.6; the latter interpretation is preferred. In sillimanite-free gneisses, the distribution of Mg and Fe²⁺ between garnet (low in Ca and Mn) and biotite is adequately described by a distribution coefficient (KD) of 4.1 (equilibrium II). The observed increase in the distribution coefficient with increasing Ca in garnet is ln K_D= 1.3 + 2.5 × 10^?2 [Ca] where [Ca] = 100 Ca/(Mg + Fe²⁺+ Mn + Ca). The distribution coefficient is apparently unaffected by the presence of up to 12% spessartine in garnet. In several specimens of garnet-sillimanite-plagioclase gneiss, the Ca contents of garnet and of plagioclase increase in unison, as required by equilibrium (III). The mean pressure calculated from these data (n= 17) is 5.9 kbar, and the 95% confidence limits are ±0.5 kbar.     

Trace-element partitioning between garnet,clinopyroxene and Fe-rich picritic melts at 3 to 7 GPa

We have determined mineral-melt partition coefficients (D values) for 20 trace elements in garnet-pyroxenite run products, generated in 3 to 7 GPa, 1,425–1,750°C experiments on a high-Fe mantle melt (97SB68) from the Paraná-Etendeka continental-flood-basalt (CFB) province. D values for both garnet (∼Py₆₃Al₂₅Gr₁₂) and clinopyroxene (∼Ca_0.2Mg_0.6Fe_0.2Si₂O₆) show a large variation with temperature but are less dependent on pressure. At 3 GPa, D ^cpx/liq values for pyroxenes in garnet-pyroxenite run products are generally lower than those reported from Ca-rich pyroxenes generated in melting experiments on eclogites and basalts (∼Ca_0.3–0.5Mg_0.3–0.6Fe_0.07–0.2Si₂O₆) but higher than those for Ca-poor pyroxenes from peridotites (∼Ca_0.2Mg_0.7Fe_0.1Si₂O₆). D ^grt/liq values for light and heavy rare-earth elements are ≤0.07 and >0.8, respectively, and are similar to those for peridotitic garnets that have comparable grossular but higher pyrope contents (Py_70–88Al_l7–20Gr_8–14). 97SB68 D _LREE^grt/liq values are higher and D _HREE^grt/liq values lower than those for eclogitic garnets which generally have higher grossular contents but lower pyrope contents (Py_20–70Al_10–50Gr_10–55). D values agree with those predicted by lattice strain modelling and suggest that equilibrium was closely approached for all of our experimental runs. Correlations of D values with lattice-strain parameters and major-element contents suggest that the wollastonite component and pyrope:grossular ratio exert major controls on 97SB68 clinopyroxene and garnet partitioning, respectively. These are controlled by the prevailing pressure and temperature conditions for a given bulk-composition. The composition of co-existing melt was found to have a relatively minor effect on 97SB68 D values. The variations in D values displayed by different mantle lithologies are subtle and our study confirms previous investigations which have suggested that the modal proportions of garnet and clinopyroxene are by far the most influential factor in determining incompatible trace-element concentrations in mantle melts. The trace-element partition coefficients we have determined may be used to place high-pressure constraints on garnet-pyroxenite melting models.     

Ca–Mn exchange between grossular and MnCl2 solutions at 2 kbar and 600°°C: reaction mechanism and evidence for non-ideal mixing in spessartine-grossular garnets

 The hydrothermal reaction between grossular and 1 molar manganese chloride solution was studied at 2 kbar and 600 °C at various bulk Ca/(Ca+Mn) compositions: Ca₃Al₂Si₃O₁₂+3Mn²⁺(aq) ⇔ Mn₃Al₂Si₃O₁₂+3Ca²⁺(aq) The reaction products are garnets of the spessartine-grossular solid-solution series which discontinuously armour the dissolving grossular grains. The first garnet to crystallize is spessartine rich (X ^gt _Mn≥0.95), reflecting the high Mn content of the solution, but as the reaction proceeds more calcium-rich garnets progressively overgrow the initial products. The armouring product layer is detached from the dissolving grossular, which allows the progressive overgrowth to occur on both its external and internal surfaces and results in the development of a two directional Ca/(Ca+Mn) zoning pattern in the product grains. The compositional changes in the run products are consistent with attainment of heterogeneous equilibrium between the external rims of the spessartine-grossular garnets and the bulk solutions in runs of duration ≥24 hours. Plots of ln K_D versus X ^gt _Ca maxima show linear variations that are not consistent with the ideal mixing that has been proposed for spessartine-grossular garnets at temperatures of 900 to 1200 °C. The data rather fit a regular solution model with the parameters ΔG° (600 °C, 2 kbar)=−8.0±0.8 kJ/mol and w ^gt _CaMn=2.6±2.0 kJ/mol. Existing solubility measurements and thermodynamic data from other Ca and Mn silicates support the calculated data. Grossular activities calculated using the w ^gt _CaMn parameter indicate that even in manganese-rich metapelites pressure estimates calculated using the garnet-plagioclase-Al₂SiO₅-quartz barometer will not be increased by more than 0.2 kbar. Received: 18 January 1995/Accepted: 4 June 1996     

Phosphorus – an omnipresent minor element in garnet of diverse textural types from leucocratic granitic rocks

Summary Elevated P contents of up to 0.086 apfu (1.21 wt.% P₂O₅) were found in garnet from leucocratic granitic rocks (orthogneisses, granites, barren to highly evolved pegmatites) in the Moldanubicum and Silesicum, Czech Republic, and in complex granitic pegmatites from southern California, USA, and Australia. Minor concentrations (0.15–0.55 wt.% P₂O₅) appear ubiquitous in garnet from leucocratic granitic rocks of different origins and degrees of fractionation. Concentrations of P are not related to Mn/(Mn + Fe) that vary from 0.12–0.86 and to textural types of garnet (i.e., isolated anhedral to euhedral grains and nodules, graphic and random garnet–quartz aggregates, subsolidus veins of fine-grained garnet). Garnet compositions exhibit negative correlations for P/Si and P/R²⁺ where R²⁺ = Fe + Mn + Mg + Ca, while Al is constant at ∼2.05 apfu. Concentrations of Na are largely below 0.02 apfu but positively correlate with P. The main substitution may involve A-site vacancy and/or the presence of some light element(s) in the crystal structure. The substitution □P₂ R²⁺ ₋₁Si₋₂ and/or alluaudite-type Na□P₃ R²⁺ ₋₁Si₋₃ seem the most likely P-incorporating mechanisms. The partitioning of P among garnet and associated minerals in granitic systems remains unclear; however, it directly affects the distribution of Y and REEs.     

An experimental investigation on diffusion of water in haplogranitic melts

  The diffusivity of water has been investigated for a haplogranitic melt of anhydrous composition Qz₂₈Ab₃₈Or₃₄ (in wt %) at temperatures of 800–1200°C and at pressures of 0.5–5.0 kbar using the diffusion couple technique. Water contents of the starting glass pairs varied between 0 and 9 wt %. Concentration-distance profiles for the different water species (molecular water and hydroxyl groups) were determined by near-infrared microspectroscopy. Because the water speciation of the melt is not quenchable (Nowak 1995; Nowak and Behrens 1995; Shen and Keppler 1995), the diffusivities of the individual species can not be evaluated directly from these profiles. Therefore, apparent chemical diffusion coefficients of water (D _water) were determined from the total water profiles using a modified Boltzmann-Matano analysis. The diffusivity of water increases linearly with water content <3 wt % but exponentially at higher water contents. The activation energy decreases from 64 ± 10 kJ/mole for 0.5 wt % water to 46 ± 5 kJ/mole for 4 wt % water but remains constant at higher water contents. A small but systematic decrease of D _water with pressure indicates an average activation volume of about 9 cm³/mole. The diffusivity (in cm²/s) can be calculated for given water content (in wt %), T (in K) and P (in kbar) by