THE EFFECT OF LARGE MELT FRACTION ON THE RHEOLOGICAL PROPERTIES OF PERIDOTITE

Laboratory studies of the rheological properties of partially molten mantle rocks have reached melt fractions of φ < 0.15, which is smaller than thought to be appropriate for magma chambers and the interior of young terrestrial planets.  Therefore, we have performed a series of high-temperature, triaxial compressive creep experiments on dry, synthetic peridotites in a gas-medium apparatus at a confining P = 300 MPa and T = 1500 to 1553 K to determine the influence of large amounts of melt (0.15 < φ < 0.30) on the viscosity of partially molten rocks in both diffusion and dislocation creep.  Mechanical mixtures of San Carlos olivine (10 to 50 mm) plus MORB (~8 mm) were hot-pressed at 1523 K and 300 MPa for 4 to 10 h.  After hot-pressing, the melt is homogeneously distributed between grain-size melt pockets at triple junctions and smaller pockets at two-, three- and four-grain junctions.  Stress vs strain rate data from samples containing φ = 0.16 to 0.30 MORB with a grain size d ≈ 10 mm deformed at 1500 to 1553 K and differential stresses of s =1 to 70 MPa in the diffusion creep regime (stress exponent n = 1) reveal a drop in rock viscosity of several order of magnitude between φ = 0.25 and φ = 0.30, indicative of a rheologically critical melt fraction (RCMF).  We carried out a similar series of experiments on synthetic peridotites with d ≈ 30 to 60 mm at the same experimental conditions but with higher stresses.  Stress vs strain rate data from a sample with d ≈ 50 mm containing φ = 0.20 deformed at 1498 to 1550 K and s = 10 to 120 MPa combined with the data from the diffusion creep experiments indicates that the flow behavior in both creep regimes as well as the transition from diffusion to dislocation creep with φ = 0.20 is well described by the published flow laws for partially molten samples of olivine + basalt deformed under anhydrous conditions [e.g. Hirth & Kohlstedt, 2003] with strain rate dε/dt ~ exp(α φ) and α = 21 for both diffusion creep and dislocation creep.  Using these flow law parameters we can constrain the likely viscosity (η ), d and s for terrestrial magma chambers where φ has been determined via seismological or geochemical observations as well as the φ, d, and s in a terrestrial planetary interior where mantle η has been determined by spacecraft observation.