AN X-RAY DIFFRACTION STUDY OF ROCKS FROM SAN MIGUEL VOLCANO, EL SALVADOR

San Miguel is a composite cone that has dominantly erupted basaltic lava flows. Historic flank lavas and prehistoric crater lavas can be distinguished by their CaO, Al₂O₃, and Sr contents. Two small populations of basaltic andesites can be distinguished by their V contents. All San Miguel lavas contain phenocrysts of plag, ol, opx, ± mag, and the groundmass consists mostly of plag, cpx, mag, and glass. We believe that variations in major and trace element geochemistry among the different sample groups can be directly related to fractionation of varying proportions of phenocrysts. One method to test this hypothesis requires the accurate determination of modal mineralogy. This is typically done through exhaustive point-counting to determine both phenocryst and groundmass modes. However, with advances in X-Ray Diffraction technology and data reduction software, it may now be possible to make these determinations on rock samples. A major component of our study is to evaluate the feasibility of using an X-Ray Diffractometer (XRD) to determine the crystallinity and qualitative and quantitative mineralogy of volcanic rocks. In order to assess the ability of the XRD to determine crystallinities, several samples with varying proportions of crystals and glass were point-counted. Powders of these rock samples were then run on the XRD to acquire diffraction patterns. These patterns were processed using background and profile fitting techniques, and then crystallinity calculations were performed on the refined patterns. Results from 4 samples with crystallinities ranging from 47-91% indicate near perfect concurrence between crystallinities determined by point-counting and those from the diffraction patterns. Deviation between the techniques was <1.3%. Evaluating the ability of the XRD to accurately determine the qualitative mineralogy of rock samples containing solid-solution minerals could not be done using end-member mineral patterns from the Powder Diffraction File. Thus, we performed mineral separations on selected rocks and determined diffraction patterns for each mineral. These patterns will be used to determine whether mineralogy can be accurately determined from the diffraction pattern of the rock sample. We hope to eventually determine quantitative mineralogy from rock diffraction patterns also.