Characterizing experimental pressure and temperature conditions in multi-anvil apparatus

Abstract

Octahedral media made of MgO–5%Cr2O3, with edge lengths of 18, 14, and 10 mm are used as pressure cells in experiments in a multi-anvil solid media apparatus at pressures of 4 to 27 GPa and temperatures to > 2700 °C. Calibrations of press-load versus sample pressure are based on room-temperature and high-temperature phase transitions, and are accurate to within ± 0.5 GPa. Calibrations of the temperature distribution were made in the central portion of the furnaces (graphite or LaCrO3) in the various sample assemblies used routinely in this laboratory. The following gradients away from the furnace midlines were observed: 18 mm: high-T straight graphite (−100 °C mm−1), high-T stepped graphite (+ 25 °C mm−1), low-T stepped graphite (−20 °C mm−1), high-T stepped LaCrO3 (−50 °C mm−1); 14 mm: high-T stepped LaCrO3 (−70 °C mm−1); 10 mm: straight LaCrO3 (−200 °C mm−1). The effect of increasing the wall thickness of the central segment of the furnace ("stepping") is to reduce the temperature gradient relative to a straight design. The relative effect of pressure on W3Re–W25Re and Pt–Pt13Rh thermocouples was measured by comparison of apparent temperatures recorded by each type in a given experiment. Corrections for the effect of pressure on thermocouple emf depend on the temperature distribution in the gasket regions surrounding the pressure cell, where pressure is reduced to ambient conditions. The temperature of this pressure seal controls the magnitude of the effect of pressure on thermocouple emf. Because this temperature will vary depending on the assembly, no universal pressure correction can be derived.