HR: 1340h
AN: V13D-1587    [Abstracts]
TI: Reproducibility of Raman Spectroscopy Measurements for Carbonaceous Material in Metamorphic Rocks
AU: * Wong, E
EM: elizabeth.wong@yale.edu
AF: Yale University, Kline Geology Laboratory 210 Whitney Avenue, New Haven, CT 06511, United States
AU: Brandon, M
EM: mark.brandon@yale.edu
AF: Yale University, Kline Geology Laboratory 210 Whitney Avenue, New Haven, CT 06511, United States
AU: Pasteris, J
EM: pasteris@levee.wustl.edu
AF: Washington University, Department of Earth and Planetary Sciences Campus Box 1169, St. Louis, MO 63130-4899, United States
AU: Wopenka, B
EM: bwopenka@levee.wustl.edu
AF: Washington University, Department of Earth and Planetary Sciences Campus Box 1169, St. Louis, MO 63130-4899, United States
AU: Dunn, S
EM: sdunn@mtholyoke.edu
AF: Mount Holyoke College, Department of Earth & Environment 50 College Street, South Hadley, MA 01075, United States
AU: Karabinos, P
EM: pkarabin@williams.edu
AF: Williams College, 127 Bronfman Science Center, Williamstown, MA 01267, United States
AB: Raman spectroscopy shows that that carbonaceous material in metamorphic rocks shows a systematic increase in crystallinity with increasing metamorphic grade. Beyssac et al. (2002) reported a calibration curve for the "Raman spectrum of carbonaceous material" (RSCM) thermometer, from 350 to 650 C. Rahl et al. (2005) extended that calibration down to ~100 C. We report here the results of a comparative study using laser microRaman systems at Yale and Washington University to understand the reproducibility of RSCM measurements. We have started a comparative study using Raman systems at Washington and Yale Universities. Reproducible results are possible but several factors can cause problems. (1) We find that the background for the Raman spectra are commonly curved and that a second-order polynomial is needed to fully remove the background. (2) Operator differences in the peak- fitting process may produce a significant bias in results. The peak-fitting process requires the operator to separately fit for peaks in each of the ~10 spectra collected for a sample. These results are then averaged to get a composite result for the sample. We have found it is better to fit peaks to an integrated spectrum obtained from individual spectra of that particular sample. The F test is used to test for the homogeneity of a sample, and hence eliminating anomalous data. This method also allows a direct comparison of results between different labs and operators, as well as spot homogeneity estimates. (3) Laser-induced heating has been shown to distort peaks in the Raman spectra, hence, producing an upward bias in estimated metamorphic temperatures. This problem is most acute when making measurements on very low-grade metamorphic rocks. We use the Raman spectra to measure in-situ temperatures and found that heating effects can be minimized by using thin sections rather than rock chips and by greatly reducing the laser power. Our inter-lab comparison shows that RSCM measurements can be accurately reproduced, if care is given to the problems above.
DE: 3660 Metamorphic petrology
DE: 3934 Optical, infrared, and Raman spectroscopy
DE: 8094 Instruments and techniques
SC: Volcanology, Geochemistry, Petrology [V]
MN: 2007 Fall Meeting