Empirical Correction Factors for the Electron Microanalysis of Silicates and Oxides

Abstract

Given suitable correction factors, the use of pure oxides and binary oxide phases as standards would make electron microanalysis independent of chemical analyses and problems of sample inhomogeneity. Ziebold and Ogilvie (1964) have shown that the calibration curve in a binary metal alloy system can, within the variance of data points, be described by the linear expression , where is the concentration of element A in alloy Ârelative to pure A, and is the background-corrected intensity of a characteristic radiation line of A in the alloy relative to that of pure A. This linear variation of the correction factor with composition can be extended to multicomponent systems by using the weighted average of the binary correction factors. Correction factors have been determined empirically for characteristic lines of ten major elements (Na, Mg, Al, Si, K, Ca, Ti, Cr, Mn, and Fe) in the corresponding oxides using phases on binary and pseudobinary joins and extrapolating into more complicated systems. Where available, synthetic minerals were used as standards, but natural minerals, verified to be nearly stoichiometric by the absence of other elements as established in wavelength scans and to be nearly homogeneous by step and electron-beam scanning, were also used. The results obtained for complicated minerals such as amphiboles and micas, using oxides and simple silicates as standards, are comparable to those of standard analytical techniques.