Zircon equilibria in metamorphic rocks

Abstract

This study defines a number of fundamental principles of zircon equilibria in metamorphic rocks using a combination of traditional metamorphic petrology, trace element geochemistry, experimental petrology, thermobarometry and geochronology. It is demonstrated that solid-state zircon formation in metamorphic rocks is more likely to occur during retrograde rather than prograde processes. Zircon equilibria in metamorphic rocks is also found to be highly temperature dependent; phases such as garnet and rutile incorporate more Zr with increasing temperature, thereby resorbing coexisting zircon. These findings have profound implications for the interpretation of ages derived from metamorphic zircons. Experimental investigation of the reaction Zr-rutile + quartz -> rutile + zircon reveals a strong temperature dependence on the Zr content of rutile. From experiments, a new geothermometcr is defined based on Zr02 in rutile, in the presence of zircon and quartz: T(OC) = 89297.49 + 0.63(P - 1) / RinK + 33.46 - 273 This thermometer can be used to define temperatures for successive stages of metamorphism by comparing the Zr contents of rutile inclusions in major phases, with the composition of rutile in a quartz matrix. In the 1 apier Complex, East Antarctica, a rutile inclusion in sapphirine records the temperature of peak metamorphism, at 1100°C, 10 kbar. A similar inclusion in cordierite gives a tenlperature of 1060°C, recording the temperature of sapphirine + quartz breakdown. Rutile grains in the matrix in contact with quartz give lower temperatures, as low as - 700°C. Simple calculations using the above formula show that during high grade metamorphism of a rutile-bearing rock, the Zr-content of rutile increase to such an extent that a significant proportion of pre-existing zircon will be resorbed, inhibiting new zircon growth and removing evidence of prior zircon forming events. Any metamorphic zircon present in such a rock must therefore have formed during cooling rather than at peak metamorphic conditions. A similar relationship is evident for the reaction Zr-almandinc + quartz -> almandine + zircon, in which the Zr content of garnet increases with increasing temperature. Because Zr diffusion in garnet has a higher closure temperature than major clements such as Fe, Mg, Ca and Mn, such a trace element thermometer can be used to investigate successive periods of heating and cooling, or zircon growth and resorption, through Zr zonation patterns in garnet where the major clement compositions have been reset during subsequent high temperature metamorphism. Two natural examples of zircon-forming reactions have been studied in detail. In the Napier Complex of East Antarctica, zircon formed at 700°C and 10 kbar via the reaction Zr-rutile + quartz -> rutile + zircon, during isobaric cooling from UHT metamorphism. In situ SHRIMP dating of zircons in thin section from the resulting reaction textures gives an age of 2452 ± 17 Ma. In contrast, zircon in the Rogaland metamorphic aureole of SW Norway, formed via the reaction Zr-garnet + sillimanite + quartz -> cordierite + zircon. This reaction occurred at 710°C and 5.6 kbar during regional decompression and has a U-Pb isotopic age of 955 = 8 Ma. For the first time, this study has defined the boundaries for which precise P-T-t constraint may be placed on zircon formation in metamorphic rocks.