A Step Forward in Quantitative Automated Mineralogy in 2D and 3D

Abstract

Automated mineralogy is a software addition usually seen on scanning electron microscopes designed to provide rapid insight into sample chemistry and texture in routine petrology workflows. The specific purpose of automated mineralogy is to provide mineral classifications to uniquely identified phases typically using energy dispersive spectroscopy, thus removing laborious and time-consuming human input for routine tasks. These mineral classifications can then be applied to image data to quantify which mineral is associated with any particular texture. Automated mineralogy systems were primarily designed to generate quantitative textural analysis of particle samples to the mining industry and have remained a critical technique in this setting for the last several decades. Automated mineralogy has become more widely used in academia, and this has changed the focus of the technique, applying it to a broader range of workflows and applications. Here we show petrology examples focussing where combined geochemical and textural analysis are widely used. Critically, the use of quantitative geochemical data means that mineral classifications are based on their quantitatively measured chemistry. By making both the chemical and textural analysis quantitative, automated mineralogy can become highly flexible and provide a unique system for petrologists in both industry and academia.