EPIDOTE RAMAN PEAK SHIFT AS A FUNCTION OF PRESSURE AND COMPOSITION: CALIBRATION AND PRELIMINARY APPLICATIONS

In this study, we assess the usefulness of epidote as an inclusion barometer. We ran a diamond anvil cell experiment from 0.05 GPa to 4.5 GPa and measured the epidote Raman spectrum at sixteen pressures. We also measured the ambient Raman spectra of eight natural epidote samples from the Rensselaer Polytechnic Institute collection in order to determine the composition effect on the epidote Raman spectrum.

Two Raman peaks appear to be best suited for GeoBaRamantry applications: the 568 cm^-1 and 600 cm^-1 peaks. The ambient peak positions appear to shift with changes in X_Fe, but analysis of more Fe-rich samples is necessary before the composition dependence can be quantitatively determined. Pressure dependence of these peaks, however, is roughly linear and can be represented by the equations

P_int (GPA)=0.3371Δs₅₆₈-0.0195

P_int (GPA)=0.4282Δs₆₀₀+0.0224

where P_int is the internal pressure of the epidote inclusion (precision = ±0.09GPa) and Δs is the shift of the respective Raman peak relative to the peak position at 1 atmosphere.

These calibrations have been applied to measurements from two samples to determine the internal pressure of epidote inclusions in garnet. A thermoelastic model was then applied to determine the pressure of epidote entrapment and hence garnet growth. The composition effect on the epidote peak positions can be mitigated by using the peak positions of an epidote grain in the matrix as a baseline. This approach assumes the composition of the epidote inclusion is the same as that of the matrix grain, which may not always be the case.

Sample TM-730 is a garnet-staurolite-kyanite pelitic schist from central Vermont that yields an entrapment pressure of 10 Kbar at 550°C. This pressure is higher than previously published results of 7 Kbar but is consistent with inferred pressures of entrapment from QuiG barometry on nearby samples. Sample SPH99-7 is a blueschist from Sifnos, Greece that yields a pressure of 32 Kbar at an assumed temperature of 500°C. This result is much higher than the previous 22 Kbar estimate using traditional geothermobarometry and QuiG barometry. These discrepancies may be a function of the overstepping of garnet producing reactions as a result of thrust faulting in central Vermont and subduction in Greece.