Exploring Mineralogic

Modern mining places a higher emphasis on liberation of key resources from more complex and lower return ores, and an in-depth understanding of the material is vital. Mineralogic allows us to characterise the target minerals in complicated samples with reduced active time as the analysis can be left unattended. This analysis not only reveals the distribution of minerals but also how the minerals are associated with each other and various useful statistics such as porosity, mineral association, grain size, and so on that can add to assays carried out by the mine as part of production. 

We challenged the Mineralogic software with a local mining sample where we identified most minerals prior using EDS to create a mineral classification scheme. A Mineralogic analysis was then carried out which generated a mineral map, and proved very successful.

In Mineralogic, we aim for less than 5% of pixels that are unclassified across the mapped area. In this instance, the whole sample surface was mapped with only 0.83% of pixels left unclassified. If this number was higher, we can also do a "retrospective analysis" on the sample where any unclassified or misclassified pixels can be sorted through and corrected with small changes to the mineral classification scheme which produces an updated mineral classification and data for that sample. This ensures that the data is as accurate as possible with the step size and conditions chosen and means additional mineral phases that could have been missed analysis set-up.

Not only can classification maps and an abundance of assay data be collected using Mineralogic, we can also produce heat maps for individual elements across the sample. This allows the user to identify the areas of greatest concentration for this element and thus specific areas of interest within a sample for further investigation. 

Northwest Africa 1110 is meteorite produced by volcanic eruptions on Mars approximately 110 million years ago called a shergottite. This group of meteorites are known for their large, often zoned, olivine crystals surrounded by smaller crystals. Mineralogic allows us to identify the different mineral phases present across the sample by running compositional data collected using EDS through a mineral classification scheme that defines each mineral by a set of compositional criteria, created by the user during the set-up of the Mineralogic run. Using the heat map function on Northwest Africa 1110, we are also able to see that the sample is richest in iron at the rims of the large olivine crystals and small grains of ilmenite that are distributed throughout the groundmass.

Northwest Africa 8266 is a brecciated eucrite that contains a large amount of melt created during impact events in the early Solar System. As this impact melt is a mixture of pyroxene and plagioclase, it has an average composition somewhere between the two minerals. Mineralogic allows us to separate this out as a distinct phase and can then be quantified by the software. Previously, the amount of impact melt in a sample would be calculated using pixel counting techniques from either backscattered electron or layered EDS images of the whole sample which can take significantly longer, particularly in complicated samples. In this case, Mineralogic revealed that this piece of NWA 8266 contains 8% impact melt.