THE STATE OF PHASE EQUILIBRIA MODELING OF METAMORPHOSED HYDROTHERMAL ALTERATION ZONES IN THE VMS ENVIRONMENT

Pre-metamorphic hydrothermal alteration zones are ideal locations for testing metamorphic phase equilibrium predictions because they present varying bulk-system compositions over a small spatial footprint, eliminating variability of P-T history between samples. Phase equilibria predictions from 2 VMS deposit environments are used to illustrate the utility of phase equilibria modeling in a mineral exploration context and to identify problems to be addressed.

Rocks near Sherridon, MB, locally contain assemblages involving garnet, cordierite, sillimanite, biotite, gedrite, quartz, and feldspar, were metamorphosed to 680 – 700 °C, and experienced variable amounts of partial melting. The observed assemblages are common in metamorphosed alteration zones, but can also occur in restite from zones of partial melting. It is shown using phase equilibria modeling that the variability in rock chemistry is not consistent with melt extraction; the rocks experienced moderate amounts of pre-metamorphic K (sericite) and Fe-Mg (chlorite) alteration in the VMS environment.

Rocks in the Lalor VMS deposit alteration system at Snow Lake, MB, contain sillimanite after kyanite and were metamorphosed to 600 °C & 5 kbar. Phase equilibria modeling of a variety of rock compositions leads to a metamorphic fluid evolution model that links dehydration and hydration/decarbonation reactions in the alteration system. Modeling of chlorite alteration zones indicates a significant amount of H₂O production at temperatures that overlap with hydration/decarbonation reactions that produced significant amounts of CO₂ from dolomite + quartz bearing rocks.

Phase equilibria modeling of rocks in the VMS environment can produce results useful in ore deposit studies, but significant problems remain. Recent thermodynamic models for sulfides and COHS fluids allow testing phase equilibria predictions for Fe-sulfide bearing rocks. Predictions on rocks containing more than a few percent sulfide remains problematic, primarily due to the inability to predict the amount of Fe-sulfide required to reduce the effective bulk-rock Fe/(Fe+Mg) to a value that allows prediction of observed Fe-Mg silicate and oxide mineralogy. Another significant problem is the lack of thermodynamic models for Zn-bearing minerals in VMS environments.