Experimental study of amphibolite and eclogite stability

Abstract

The amphibolite-eclogite transition is often observed in metamorphic rocks and may be expected to occur in the lower crust or upper mantle if P H 2O ≈ P S. Many simple amphiboles become unstable at high water pressures due to backbending of dehydration reactions or to solid-solid transitions involving sheet silicates. Contrary to Yoder and Tilley (1962), eclogite is expected to be stable relative to amphibolite at sufficiently high water pressures because of the formation of dense garnet and the high compressibility of water vapor. Experiments in progress on several basaltic compositions have produced eclogite at the expense of amphibolite at high water pressures. Seeding has been found to be important as hornblende crystallizing from glass will persist at least 4 kb above its stability field defined by reversals with garnet-pyroxene-amphibole mixtures. In reconnaissance experiments on a quartz tholeiite and an alkali olivine basalt, hornblende disappears at high water pressures between 15–25 kb and 700–900°C. Biotite appears to be stable in the alkali olivine basalt to at least 40 kb at these temperatures. Basaltic amphibolites do not appear to be stable very far into the low-velocity zone even at P H 2O = P S confirming Lambert and Wyllie's (1968) earlier experiments. Biotite however seems to be stable deeper in the mantle (at least for potassic basalts) than previously considered possible if P H 2O ≈ P S and may be a source of water for magmas when partially melted.