A flow law for dislocation creep of quartz aggregates determined with the molten salt cell

We have used the molten salt cell to conduct an experimental study on the rheology of a natural quartzite containing ∼ 0.15 wt. % water. Co-axial deformation experiments were conducted at constant piston displacement rates, approximating constant strain rates at low strain. The strengths of our natural quartzite measured in the molten salt cell are approximately half those measured at the same conditions in solid media because, unlike solid confining media, molten salt does not contribute to the strength of the sample; it reduces the friction on the moving piston, and it allows clear identification of the 'hit' point. We have limited the experimental conditions to those required for dislocation creep, and have used only steady-state flow stresses measured during climb-accommodated dislocation creep to calculate the flow law parameters. Two flow laws were determined, one for samples containing minor amounts of melt (1-2%) and one for melt-free samples. In both cases, the power law stress exponent, n, is 4.0 ± 0.9, which is greater than that previously reported in flow laws for dislocation creep of quartz aggregates determined in solid media. The activation energy, Q, is 137 ± 34 kJ mol ^-1 for samples with melt and 223 ± 56 kJ mol ^-1 for those without, within the range of previously determined values for quartz aggregates containing ∼ 0.1 wt.% water. The pre-exponential term, A, is 1.1 × 10 ^{(-4 ± 2)} MPa ^{- n}s ^-1 for samples without melt and 1.8 × 10 ^{(-8 ± 2)} MPa ^{- n}s ^-1 for those with melt. The lower strengths measured in the molten salt cell indicate that previous piezometer relations for quartz experimentally determined in solid media are not correct. Extrapolation of the flow law for melt-free aggregates to natural strain rates predicts higher strengths than most previous quartz flow laws. However, accurate extrapolation requires determining the dependence of flow stress on f_{H 2O} and / or a_{H ⁺}.