Preheated light gas gun shock experiments: hot Molybdenum and diopside-anorthite liquid Hugoniots revisited

We have extended the techniques for pre-heated Hugoniot equation of state measurements for use on Caltech's 25 mm light gas gun at flyer velocities up to 7.5 km/s. Previous data on Mo at 1400°C and on a variety of silicate liquids were collected on a 40 mm propellant gun up to a maximum flyer velocity of 2.6 km/s. Higher impact velocities open up a range of new opportunities, including tests of previous extrapolations of low-pressure data and direct probing of the properties of molten silicates at lower mantle pressure. Our preheated liquid experiments are conducted in sealed Mo capsules and therefore we need to know the Hugoniot of Mo initially at elevated temperature, which may differ by several percent from the principal Hugoniot of Mo. Miller et al. [1] measured the Hugoniot EOS of Mo initially at 1400°C up to a particle velocity (Up) of 1.5 km/s and applied a linear fit with shock velocities slower than the principal Hugoniot in the measured range, but implying a crossover when extrapolated above 1.8 km/s (i.e., about 100 GPa pressure). Molodets [2] fit these data to a parameter-free theoretical form for the volume dependence of the Grüneisen parameter that predicts a concave-downward high-temperature Hugoniot that runs below and approaches parallel with the principal Hugoniot. Our data point at Up = 2.5 km/s (204 GPa) is coincident with Molodet's theory within error. However, our data point at Up = 3.24 km/s (302 GPa) is not; we are investigating this discrepancy. The silicate liquid composition consisting of 64 mol % anorthite and 36 mol % diopside is a simplified analogue for basalt and was chosen for study by Rigden et al. [3]. This earlier study found the expected linear Us-Up Hugoniot (with molar volume intermediate between anorthite and diopside end members) up to 25 GPa, followed by two data points that suggested a dramatic stiffening to a nearly incompressible Hugoniot. We now have three experiments at higher pressure (44, 81, and 110 GPa) that clearly show that this extrapolation was incorrect. All the data on this composition can be fit with a single linear Hugoniot. Although basaltic liquids of this composition are not expected in the lower mantle, the implication is that silicate liquids remain more compressible than solids at compressions approaching 50%. This is consistent with results from our laboratory on SiO2, MgSiO3, and Mg2SiO4 systems showing that melts in these systems become denser than coexisting solids at pressures similar to the base of the mantle. 1. Miller, G.H., T.J. Ahrens, and E.M. Stolper, The Equation of State of Molybdenum at 1400 °C. J. Appl. Phys., 1988. 63(9): p. 4469-4475. 2. Molodets, A.M., Shock compression of preheated molybdenum. High Pressure Research, 2005. 25(3): p. 211-216. 3. Rigden, S.M., T.J. Ahrens, and E.M. Stolper, Shock compression of molten silicate - results for a model basaltic composition. J. Geophys. Res., 1988. 93(B1): p. 367-382.