Evaluation of different non-metallic inclusions in steel chips by using electrolytic extraction : Evaluation of a methodology for electrolytic extraction and scanning electron microscopy - energy dispersive spectroscopy (SEM-EDS) analysis

Abstract: Analysing non-metallic inclusions (NMI) by conventional microscopy is prone to errors.  Better imaging can be achieved by using electrolytic extraction (EE) to effectively dissolve the metal matrix, freeing the inclusions so that they can be collected on a filter.  This method of studying NMI was tested on a 157C steel chip, with EE taking place three times on the same surface with increasing levels of  charge  applied.  The  relationship  between  charge  and  extracted  layer depth was examined, so as to facilitate the targeting of NMI from specific depths.  A selection of the extracted inclusions were micrographed with scanning electron microscopy (SEM) and had their compositions measured with energy dispersive spectroscopy (EDS). Based on this data, two methods of classifying NMI were briefly  examined:  manual  classification,  mostly  based  on images,  and  a  semi-automated process based on a sorting algorithm applied to the compositions. The study implies that it is possible to dissolve a 157C steel to a desired depth by applying a charge proportional to it, but the current method introduces an error which might limit the resolution of depth by a significant amount.  Also in the current method, there was no systematic way to select NMI for micrography, and no solution to this problem was found.  The semi-automated classification algorithm  was  compromised  by  inaccurate  readings  of  compositions  from the EDS, and could not be thoroughly tested.