PACROFI VI - Electronic Program
Carbonate - Silicate Equilibria in Granulites: Implications for the Formation of CO2-Rich Fluid Inclusions
Department of Geology and Geophysics, Texas A&M University, College Station, TX 77843-3115, U.S.A.
Rocks metamorphosed to granulite facies pressures and temperatures often differ from lower grade metamorphic rocks of similar bulk composition as peak metamorphic H2O activities are much < 1, and LILE contents are generally low. Various mechanisms have been proposed to explain the stabilization the granulite facies mineral assemblages, and they fall into one of three categories. (1) During partial melting granulites are generated as H2O is partitioned preferentially into the melt phase (Fyfe, 1973). Thus, melting can result in a fluid phase enriched in components with low solubilities in melts relative to H2O (e.g., CO2), or the rock may contain no free fluid phase. (2) Metamorphism of "dry" rocks, such as certain igneous lithologies or rocks previously metamorphosed to high grades, may also produce granulites (Lamb and Valley, 1988). (3) Pervasive infiltration of large amounts of CO2 (Newton et al., 1980; Touret, 1992), which dilutes H2O produced by dehydration reactions, could stabilize granulite facies mineral assemblages. The presence of CO2-rich fluid inclusions in many granulites may be evidence of a CO2-rich peak metamorphic fluid, and may support the CO2-infiltration hypothesis. However, this assumes that the fluid inclusions were formed at the peak of metamorphism, and that they have not undergone compositional changes during retrograde cooling and uplift.
While phase equilibria in granulites often requires aH2O < 1, H2O-buffering equilibria provide no direct information concerning CO2 fugacities (f CO2). Lamb and Valley (1984, 1985) showed that oxygen buffering equilibria can sometimes be used to constrain f CO2. However, direct determinations of f CO2 are still rare in granulites, with the possible exception of marbles and calc-silicates. A review of 39 publications that describe granulite facies fluid inclusions revealed that 14 of these studies report the presence of carbonate minerals in quartzo-feldspathic gneisses containing Qtz+Pl+Opx+/-Grt+/-Cpx+/-Kfs+/-Bt+/-Amph (Kretz, 1983). These carbonates occur within fluid inclusions (daughter crystals), and as solid inclusions within Grt, Pl, Qtz and Crd. In many cases the carbonates are thought to be part of the peak metamorphic assemblage and are intimately associated with CO2-rich fluid inclusions. A variety of CO2-buffering equilibria may be written between the carbonates and the various silicates that occur in these rocks. Thus, the presence of carbonates may provide a method for determining f CO2 in granulites
Carbonate Stability at Granulite P-T
Anovitz and Essene (1987) and Davidson (1994) have reviewed existing experimental data on the stability of carbonate phases in the CaCO3-MgCO3-FeCO3 system. According to these studies, at T ~ 700 to 750oC miscibility gaps exist between calcite (cal) and dolomite (dol), dol and magnesite (mgs), and cal and siderite (sd). There is complete solid solution between sd and mgs, and nearly complete solid solution between dol and ankerite (ank). At T ~ 750oC calcite contains < 20 mole % MgCO3 and < 40 mole % FeCO3. Given these compositional constraints, evaluation of the stability of carbonates in granulites can largely be accomplished by examining the stability of calcite, dolomite-ankerite solid solutions, and magnesite-siderite solid solutions. In the following, the stability of carbonates in equilibrium with various silicate phases was determined using the computer program THERMOCALC (Powell and Holland, 1988).
Many of the carbonate-bearing granulite facies samples described in the literature contain quartz and orthopyroxene (opx). Thus, the following equilibria are potentially important:Calcite + Enstatite + Quartz = Diopside + CO2 (1)
Calcite + Ferrosilite + Quartz = Hedenbergite + CO2 (2).
The P-T location of these two equilibria are shown on Figure 1, assuming XCO2 = 1.0. Addition of H2O to the fluid phase would not stabilize calcite at granulite facies temperatures, as reduced aCO2 would stabilize clinopyroxene (cpx) relative to cal + opx + qtz. Solid solution in cal will lower it's activity and increase the stability of cal + opx + qtz relative to cpx + CO2. In an extreme case involving the combination of Fe end-member pyroxenes and a cal activity of 0.6, there remains very little overlap between the stability of cal + opx + qtz and granulite facies P-T.
If the carbonates are not calcite, but rather dol or ank, then the equilibriaDolomite + Quartz = Diopside + CO2 (3)
Ankerite + Quartz = Hedenbergite + CO2 (4)
may be applicable to the rocks in question. The P-T stability of equilibria (3) and (4) are shown on Fig. 1, assuming XCO2 = 1 which yields the maximum stability of dol + qtz or ank + qtz. Given that the end-member dol and ank are not stable in the presence of quartz at granulite P-T, it is unlikely that solid solutions with intermediate values of Fe and Mg would be stable.
If the carbonate phase contained in a rock from the granulite facies is mgs or sd, thenMagnesite + Quartz = Enstatite + CO2 (5), and
Siderite + Quartz = Ferrosilite + CO2 (6)
are equilibria that may be relevant to the determination of CO2 activities. However, neither mgs nor sd is stable in the presence of quartz at the peak metamorphic pressure and temperature conditions reported for the carbonate-bearing granulite-facies samples (Fig. 1).
A more definitive determination of carbonate stability is possible for any given sample from the granulite facies if the compositions of all co-existing phases are known. Unfortunately, in most studies that report the presence of carbonates in granulites, the compositions of all pertinent phases have not been reported. Touret and Hansteen (1988) describe inclusions of Fe-rich Dol in garnets which also contain fluid inclusions (Doddabetta, S. India), and their published chemical analyses for Grt, Opx, and Pl provide the basis to calculate the stability of the ank. The stability of the equilibria Alm+Ank+Qtz=Fs+An+CO2 was calculated using their analyses and various activity models. The composition of the Fe-rich Dol is not reported, however, for ank activities ranging from 0.2 to 1.0 the equilibria in question lies approximately 125 to 180oC below peak T at peak P. Thus, the carbonate is not stable under peak conditions, and may be retrograde.
The calculations summarized above indicate that carbonates are part of the peak granulite facies equilibrium mineral assemblage. In some cases, it may be possible that small amounts of carbonates were included within minerals (e.g., garnet) as they grew prior to attainment of peak P-T conditions, and that the carbonates were then armored against further reaction with other phases. In most cases, however, it is likely that the carbonates were formed after the peak of metamorphism. These carbonate grains are often intimately associated with CO2-rich fluid inclusions, and so these fluid inclusions are also not part of the peak assemblage and, in many cases, were probably formed after the peak of metamorphism, perhaps during retrograde cooling and depressurization.
- Anovitz LM, Essene EJ (1987) Phase equilibria in the system CaCO3-MgCO3-FeCO3. J Petrology, 28, 389-414.
- Davidson (1994) Ternary, iron, magnesium, calcium carbonates: A thermodynamic model for dolomite as an ordered derivative of calcite-structure solutions. Am Mineral, 79, 332-339.
- Fyfe WS (1973) The granulite facies, partial melting and the Archean crust. Phil Trans Roy Soc Lond A. 273, 457-461
- Kretz R (1983) Symbols for rock-forming minerals. Am Mineral, 68, 277-279.
- Lamb WM, Valley JW (1984) Metamorphism of reduced granulites in low-CO2, vapor-free environment. Nature, 312, 56-58.
- Lamb WM, Valley JW (1985) C-O-H fluid calculations and granulite genesis. in Tobi A, Touret J (eds.) The deep proterozoic crust in the North Atlantic Provinces, Reidel, 119-131.
- Lamb WM, Valley JW (1988) Granulite facies amphibole and biotite equilibria, and calculated peak-metamorphic water activities. Contrib Mineral Petrol, 100, 349-360.
- Newton RC, Smith JV, Windley BF (1980) Carbonic metamorphism, granulites and crustal growth. Nature, 288, 45-50.
- Powell R, Holland TJB (1988) An internally consistent dataset with uncertainties and correlations: 3. Applications to geobarometry, worked examples and a computer program. J Meta Geol, 6, 173-204.
- Touret JLR (1992) CO2 transfer between the upper mantle and the atmosphere: temporary storage in the lower continental crust. Terra Nova, 4, 87-98.
- Touret JLR, Hansteen TH (1988) Geothermobarometry and fluid inclusions in a rock from the Doddabetta charnockite complex, Southwest India. Ren Soci Ital Mineral Petrol 43, 65-82.