Previous page     Contents     Next page

Seismologists require laboratory derived thermoelastic data on candidate minerals for interpretation of raw seismic velocities in the Earth. In particular, elastic constant data in conjunction with high-precision isothermal compression (bulk moduli) from X-ray diffraction provide equation of state parameters necessary for the extrapolation of laboratory data to deep Earth conditions for direct comparison with seismological observation.

In recent years, considerable progress has been made towards adapting the diamond anvil cell (DAC) for in-situ elastic property measurements and X-ray diffraction. Despite this success, ultrasonic measurements in the DAC at high temperature still face two major challenges: 1) maintaining acoustic coupling between the sample and the diamond at high temperature in a liquid pressure medium; and 2) measuring pressure changes in the cell induced by high temperature. Recently, we maintained an acoustic coupling between the buffer-rod/diamond assembly and a natural olivine sample at modest conditions of 2.6 GPa and 250°C in an ethanol-methanol pressure medium, although we cannot yet directly measure the pressure at high temperature.

High temperatures are achieved by two independently controlled Mo-wire resistance heaters in contact with the tungsten-carbide seats. Temperature is measured by a pair of chromel-alumel thermocouples attached to the lower and upper diamonds. The DAC frame and diamonds are cooled by a directed stream of argon gas. We cut and polished (perpendicular to [001]) discs of single-crystal San Carlos olivine (Fo₈₉) measuring 200 µm in diameter and 90 µm thickness. One of these discs was loaded in an ethanol-methanol mixture pressure medium and brought to 2.0 GPa at room temperature (measured by ruby-fluorescence). No bonding agent between the sample and the diamond anvil was required to achieve coherent transmission of acoustic energy. We made several heating excursions up to 200ºC and found that the acoustic coupling to the sample was maintained at these temperatures. Next, we increased the pressure to 2.6 GPa and measured round trip P-wave travel times through the sample at 25, 100, 150, and 250°C. Figure 3.8-4 shows the interference amplitude versus frequency for acoustic waves reflected at the buffer-rod/diamond interface and from the free end of the sample at 2.6 GPa and 250°C. The reduced travel time data for this experiment are plotted in Fig. 3.8-5.
 

Fig. 3.8-4: Reduced interference amplitude data for olivine at 2.6 GPa and 250°C. This plot illustrates that the interference between two acoustic wave packets reflected from the buffer rod/diamond interface and the free end of the sample are still measurable at high temperature.

Fig. 3.8-5: Plot of the round-trip travel time (ns) through the olivine sample as a function of frequency (GHz), measured at 2.6 GPa and 250°C.