Experimental confirmation of high temperature silicate liquid immiscibility

Friday, 19 December 2014
Tong Hou, State Key Laboratory of Geological Process and Mineral Resourse, China University of Geosciences, Beijing, Beijing, China; Helmholtz Centre Potsdam GFZ German Research Centre for Geosciences, Potsdam, Germany and Ilya V. Veksler, Perm State University, Geological Department, Perm, Russia; Technical University Berlin, Department of Mineralogy, Berlin, Germany
The existence of stable, super-liquidus silicate liquid immiscibility at temperatures up to 1200 °C has been proposed for some multicomponent ferrobasaltic-ferroandesitic compositions on the basis of centrifuge experiments (Veksler et al., 2007) but the evidence and interpretation of experimental results were challenged by Philpotts (2008) who argued that the products of centrifuge experiments were metastable phases formed during quenching. Here we report the results of static reverse experiments, which were aimed at resolving the debate.

The idea of the reverse experiments was to test miscibility between pre-synthesized pairs of silica-rich and Fe-rich immiscible melts at static conditions and long exposure times. Three pairs of the potentially immiscible compositions were taken from the original study by Veksler et al. (2007) and one more pair was taken from a recent report of liquid immiscibility in the Panzhihua intrusion in China. Experiments were carried out in one-atmosphere gas-mixing furnace (Ar-H2-CO2 gas mixture) at 1150 and 1200 °C and oxygen fugacity corresponding to that of the QFM buffer. Pairs of the silica-rich and Fe-rich starting compositions were loaded in Pt wire loops, fused separately at 1200 °C, and then brought in contact and kept at constant experimental temperature for more than 24 hours. Three pairs of compositions out of four used in this study did not mix. Some temperature-dependent chemical re-equilibration was observed in the Fe-rich melts but, in the cases of immiscibility, two liquids remained compositionally distinct and showed sharp compositional gradients at contacts. One pair of liquids crystallized some tridymite, whereas the other compositions were clearly above liquidus. Overall, the results of the reverse experiments are in good agreement with the earlier centrifugation study and confirm the existence of stable, super-liquids immiscibility in some Fe-rich basaltic-andesitic compositions at temperatures up to 1200 °C.


Veksler IV, Dorfman AM, Borisov AA, et al. 2007. Liquid immiscibility and the evolution of basaltic magma. J. Petrol., 48: 2187-2210.

Philpotts AR. 2008. Comments on: Liquid immiscibility and the evolution of basaltic magma[J]. J. Petrol., 49: 2171-2175.