Statistical scaling of elemental image by X-ray fluorescence spectrometry of magnesium alloy

Abstract

A technique incorporating the statistical methods into colour range scale of a 2D elemental image was developed to enable visual evaluation of significant differences in the data - a challenging process with conventional elemental images. The normal interquartile range was used as the basis for segmenting the colour range scale of the 2D elemental image, setting the median concentration as the central value at the centre of the scale. By visually scrutinizing the image, noteworthy disparities in the data become discernible. This method is unique and has not been reported in 2D elemental images. Applying this method to X-ray fluorescence spectrometry, images of Al, Si, Mn, and Zn in two Mg alloys were observed. Using the developed method, it was easily evaluated that one sample showed segregation in the central part of the image of Mn, but the difference was insignificant.