Pyrometry in the Multianvil Press: New approach for temperature measurement in large volume press experiments

Temperature measurement in large volume press experiments has been based on thermocouple emf, which has well known problems: unknown pressure dependence of emf [e.g., 1], chemical reaction between thermocouple and other materials, deformation related texture development in the thermocouple wires [2], and so on. Thus, different techniques to measure temperatures in large volume press experiments other than thermocouples are required to measure accurate temperatures under high pressures. Here we report a new development using pyrometry in the multianvil press, where temperatures are derived on the basis of spectral radiometry. Several high pressure runs were conducted using the 1000 ton press with a DIA module installed at 13 ID-D GSECARS beamline at Advanced Photon Source (APS) [3]. The cubic pressure medium, 14 mm edge length, was made of soft-fired pyrophyllite with a graphite furnace. A moissanite (SiC) single crystal was built inside the pressure medium as a window for the thermal emission signal to go through. An MgO disk with 1.0 mm thickness was inserted in a gap between the top of the SiC crystal and thermocouple hot junction. The bottom of the window crystal was in direct contact with the tip of the anvil, which had a 1.5 mm diameter hole drilled all the way through the anvil axis. An optical fiber was inserted in this hole and the open end of fiber was in contact with the SiC crystal. Thermal spectral radiance from the inner cell assembly was obtained via the fiber and recorded by an Ocean Optics HP2000 spectrometer. The system response of spectrometer was calibrated by a tungsten ribbon ramp (OL550S, Optronic Laboratories, Inc.) with standard of spectral radiance. The cell assembly was compressed up to target value of 15 tons and then temperature was increased up to 1573 K. Radiation spectra were mainly obtained above 873 K and typical integration time was 1 ms or 10 ms. Data collection was done in the process of increase and decrease of temperature. In one in-situ X-ray experiment, pressures near the thermocouple were derived from the equations of state for gold [4] or MgO [5], and the pressure values varied up to approximately 1 GPa at various temperatures. Two different calculations were made to infer temperature based on pyrometry. One was made based on Planck radiation function and the other was made by the J-function, which was defined based on the law of Wien [6]. Calculated temperatures above about 1000 K were generally consistent with those obtained from the thermocouple emf. However, below about 1000 K, temperatures based on pyrometry in the process of increase of temperature were way off from those obtained by thermocouple emf. Recent results and future improvements will be discussed. Reference [1] I. C. Getting and G. C. Kennedy J. Appl. Phys., 41, 4552-4562, 1970. [2] J. Li, C. Hadidiacos, H.-K. Mao, Y. Fei and R. J. Hemley High Pressure Research, 23, 389-401, 2003. [3] Y. Wang, M. Rivers, S. Sutton, N. Nishiyama, T. Uchida, T. Sanehira Phys. Earth Planet. Inter. in press. [4] S.-H. Shim, T. S. Duffy and K. Takemura Earth Planet. Sci. Lett., 203, 729-739, 2002. [5] M. Matsui, S. C. Parker and M. Leslie Am. Mineral., 85, 312-316, 2000. [6] T. Yagi and J. Susaki, in High-Pressure Research: Application to Earth and Planetary Sciences, edited by Y. Syono and M. H. Manghnani, pp.51-54, TERRAPUB, Tokyo / AGU, Washington D.C., 1992.