Experimental examination of two-pyroxene graphical thermometers using natural pyroxenes with application to metaigneous pyroxenes from the Adirondack Mountains, New York

Abstract

Clinopyroxene with exsolved orthopyroxene and coexisting orthopyroxene with exsolved clinopyroxene (inverted pigeonite) in metaigneous rocks from the Adirondacks, New York, were experimentally homogenized at temperatures near those inferred for their original crystallization. The purposes were several: (1) to test the graphical two-pyroxene geothermometer of Lindsley (1983); (2) to test the hypothesis of Bohlen and Essene (1978) that these were originally igneous pyroxenes; and (3) to test whether modal recombination of complexly exsolved pyroxenes yields realistic compositions. Experiments on Fe-rich compositions at 930° and 870° C (1 GPa) are compatible with the graphical thermometer of Lindsley (1983); however, this graphical thermometer yields apparent temperatures approximately 50° C too high for experiments at 1050° C and 1100° C (0 MPa). This suggests that at intermediate Mg/Fe the augite isotherms for these temperatures lie at lower wollastonite compositions than shown by Lindsley. The results are, however, in good agreement with isotherms derived from the solution model of Davidson (1985). When these isotherms are applied to a variety of terrestrial and lunar igneous rocks and the metaigneous rocks from the Adirondacks, temperatures given by augite and pigeonite compositions from coexisting pairs are similar. Comparison of the experimentally homogenized compositions with modally recombined compositions of Bohlen and Essene (1978) show that discrepancies between augite and pigeonite temperatures may nevertheless arise if pyroxene grains formed by granular exsolution are not correctly reintegrated.