BULK FLOW STRENGTH OF FORSTERITE-ENSTATITE COMPOSITES AS A FUNCTION OF FORSTERITE CONTENT

Abstract

Creep experiments have been conducted to investigate the effect of varying forsterite content (VFo) on the bulk flow strength of dry forsterite-enstatite (Fo-En) aggregates in order to evaluate the applicability of existing theoretical models to two-phase rocks, as well as to understand the rheology of polyphase systems in general. The experiments were performed at temperatures of 1423-1593 K, stresses of 18-100 MPa, oxygen fugacities of 10-14-10-2.5 MPa and 0.1 MPa total pressure. The fine-grained (Fo: 10-17 μm; En: 14-31 μm) composites of various Fo volume fractions (VFo=0, 0.2, 0.4, 0.5, 0.6, 0.8 and 1) were synthesized by isostatically hot-pressing in a gas-medium apparatus at 1523 and 350 MPa. Our experiments show that flow strength contrasts between Fo and En are in the range of 3-8 at the given experimental conditions, with Fo as the stronger phase. The measured stress exponent (n) and activation energy (Q) values of the Fo-En composites fall between those of the end-members. The n values show a nearly linear increase from 1.3 to 2.0, while the Q values display a non-linear increase from 472 to 584 kJ/mol with En volume fraction from 0 to 1.0. There is no clear dependence of creep rates on oxygen fugacity for the Fo-En composites. The mechanical data and TEM microstructural observations suggest no change in deformation mechanism of each phase when in the composites, compared to when in a single-phase aggregate, the En deformed mainly by dislocation creep while the Fo deformed by dislocation-accommodated diffusion creep for our grain sizes and experimental conditions.Comparisons between the measured composite strengths and various theoretical models indicate that none of the existing theoretical models can give a precise predication over the entire VFo range from 0 to 1. However, the theoretical models based on weak-phase supported structures (WPS) yield a good prediction for the flow strengths of the composites with VFo<0.4, while those based on strong-phase supported structures (SPS) are better for the composites with VFo>0.6. No model gives a good prediction for the bulk strength of two-phase composites in the transitional regime (VFo=0.4-0.6). Applications of the WPS- and SPS-based models in the transitional regime result in under- and over-estimations for the composite flow strength, respectively. Thus, the effect of rock microstructure should be taken into consideration in modeling the bulk flow strengths of the crust and upper mantle using laboratory-determined flow laws of single-phase aggregates.