The Fundamental Relationship Between Sample Mass and Sampling Variance in Real Geological Samples and Corresponding Statistical Models

One of the most cited papers in the geological sciences has been that of Albee and Bence on the use of empirical " α -factors" to correct quantitative electron microprobe data. During the past 25 years this method has remained the most commonly used correction for geological samples, despite the facts that few investigators have actually determined empirical α-factors, but instead employ tables of calculated α-factors using one of the conventional "ZAF" correction programs; a number of investigators have shown that the assumption that an α-factor is constant in binary systems where there are large matrix corrections is incorrect (e.g, 2-3); and the procedure’s desirability in terms of program size and computational speed is much less important today because of developments in computing capabilities. The question thus exists whether it is time to honorably retire the Bence-Albee procedure and turn to more modern, robust correction methods. This paper proposes that, although it is perhaps time to retire the original Bence-Albee procedure, it should be replaced by a similar method based on compositiondependent polynomial α-factor expressions.

Download Full-text

Quantitative Analysis Of X-Ray Images From Geological Materials

Proceedings, annual meeting, Electron Microscopy Society of America ◽

10.1017/s042482010013482x ◽

1990 ◽

Vol 48 (2) ◽

pp. 244-245

Author(s):

J. M. Paque ◽

R. Browning ◽

P. L. King ◽

P. Pianetta

Keyword(s):

Quantitative Analysis ◽

Bulk Composition ◽

Principal Component ◽

Analytical Description ◽

Bulk Sample ◽

Geological Samples ◽

X Ray ◽

Software And Hardware ◽

And Storage ◽

Insight Into

Geological samples typically contain many minerals (phases) with multiple element compositions. A complete analytical description should give the number of phases present, the volume occupied by each phase in the bulk sample, the average and range of composition of each phase, and the bulk composition of the sample. A practical approach to providing such a complete description is from quantitative analysis of multi-elemental x-ray images.With the advances in recent years in the speed and storage capabilities of laboratory computers, large quantities of data can be efficiently manipulated. Commercial software and hardware presently available allow simultaneous collection of multiple x-ray images from a sample (up to 16 for the Kevex Delta system). Thus, high resolution x-ray images of the majority of the detectable elements in a sample can be collected. The use of statistical techniques, including principal component analysis (PCA), can provide insight into mineral phase composition and the distribution of minerals within a sample.

Download Full-text