The effect of whole‐rock MnO content on the stability of garnet in pelitic schists during metamorphism

Abstract

Garnet‐bearing mineral assemblages are commonly observed in pelitic schists regionally metamorphosed to upper greenschist and amphibolite facies conditions. Modelling of thermodynamic data for minerals in the system Na2O–K2O–FeO–MgO–Al2O3–SiO2–H2O, however, predicts that garnet should be observed only in rocks of a narrow range of very high Fe/Mg bulk compositions. Traditionally, the nearly ubiquitous presence of garnet in medium‐ to high‐grade pelitic schists is attributed qualitatively to the stabilizing effect of MnO, based on the observed strong partitioning of MnO into garnet relative to other minerals. In order to quantify the dependence of garnet stability on whole‐rock MnO content, we have calculated mineral stabilities for pelitic rocks in the system MnO–Na2O–K2O–FeO–MgO–Al2O3–SiO2–H2O for a moderate range of MnO contents from a set of non‐linear equations that specify mass balance and chemical equilibrium among minerals and fluid. The model pelitic system includes quartz, muscovite. albite, pyrophyllite, chlorite, chloritoid, biotite, garnet, staurolite, cordierite, andalusite, kyanite. sillimanite, K‐feldspar and H2O fluid. In the MnO‐free system, garnet is restricted to high Fe/Mg bulk compositions, and commonly observed mineral assemblages such as garnet–chlorite and garnet–kyanite are not predicted at any pressure and temperature. In bulk compositions with XMn= Mn/(Fe + Mg + Mn) > 0.01, however, the predicted garnet‐bearing mineral assemblages are the same as the sequence of prograde mineral assemblages typically observed in regional metamorphic terranes. Temperatures predicted for the first appearance of garnet in model pelitic schist are also strongly dependent on whole‐rock MnO content. The small MnO contents of normal pelitic schists (XMn= 0.01–0.04) are both sufficient and necessary to account for the observed stability of garnet.