Abstract

ABSTRACT Permian metapegmatite garnets from the Koralpe region (Eastern Alps, Austria) contain abundant submicrometer- to micrometer-sized inclusions of rutile, corundum, Fe-Mn phosphate, ilmenite, xenotime, zircon, and apatite. Variations in inclusion abundance, phase assemblage, habit, and size define sector and concentric zones in the garnets, tracing low-indexed garnet facets. Zoning resulted from a process occurring at the garnet-melt interface, homogeneous along each facet, but sensitive to its crystallographic plane. Furthermore, inclusion and host lattices interacted, generating host-inclusion crystallographic orientation relationships (CORs). These phenomena exclude inclusion formation via overgrowth of pre-existing phases, infiltration of fluids/melts, or dissolution-reprecipitation. Magmatic garnet rims contain rutile needles up to 100 μm long, showing an interface-dependent shape-preferred orientation (SPO) that cannot be explained by exsolution models. Furthermore, the COR distribution for needles is unique, and implies large 3D lattice mismatches. These phenomena suggest that needles originated via oriented heterogeneous nucleation at the garnet interface and subsequent simultaneous growth of both phases. The origin of equant inclusions in core domains is less clear. With some assumptions, integrated compositions remain compatible with closed system exsolution or open system precipitation (OSP) involving divalent cation loss. Still, the oriented interface nucleation hypothesis seems to better explain the fact that the frequency of rutile-garnet CORs varies strongly not only between cores and rims but also between garnet core domains. Inclusion formation by oriented interface nucleation and simultaneous growth can explain many observations commonly attributed to exsolution, making distinguishing between these two mechanisms a challenge. We suggest interface-dependence of SPOs and COR frequencies as criteria for identifying inclusion formation via oriented nucleation at an interface and subsequent simultaneous growth.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call