BSE and Mineral Montage Kelly

Introduction

Scanning Electron Microscope Energy Dispersive Spectroscopy (EDS) analysis can be used to help with ore sample characterisation within the mining industry. Over the years newer techniques for characterisation and analysis of samples have been developed and introduced. Automated mineralogical techniques are being introduced and allow for more automatic characterisation of samples, requiring less specialist input. Carl Zeiss have produced their own automated mineralogy solution called Mineralogic. The aim of this project was to test the capabilities Mineralogic in respect to resource recovery by analysing the software as well as critically comparing the data produced with standard microscopical techniques that are currently being used.

Methods

Three rock samples taken from different ore deposits, were characterised under Scanning Electron Microscope EDS and analysed using two different techniques. Standard Electron Microscopy techniques were used by analysing the samples with a JEOL 7001 FE-SEM, Oxford Instruments detectors and AZtec analysis software. Zeiss’s Mineralogic software is an automated mineralogy solution designed for mineral characterisation and understanding. Quantification of elements and mineral characterisation based on stoichiometry allows for mineral maps to be created along with quantified data on Bulk composition, elemental assay, grain texture and mineral associations (Graham et al, 2014). A Carl Zeiss Sigma 300 VP SEM, and Bruker detectors was used along with Mineralogic to analyse samples and acquire data. The data collected could then be compared and analysed in respect to resource recovery.

Sample 1 – Copper, lead and zinc deposit from New Mexico

This sample has an assemblage that is made up of chemically similar minerals. This meant that mineral separation for abundance calculations using standard electron microscopy solutions like AZtec was not possible and chemically similar minerals were grouped together. Using Mineralogic, mineral groups can be split up easily by creating personalised minerals lists based on elemental weight percentages in minerals, allowing for a more in depth mineralogical characterisation when creating a mineralogical map. I.e. Malachite (Cu2(CO3)(OH)2) and Azurite (Cu3(CO3)2(OH)2) were able to be identified as separate minerals when analysing the sample with Mineralogic. This allows for a more in depth and accurate sample characterisation, which can allow for more informed decisions about the extraction process for resource recovery.

Mineralogic Vs AZtec Mineral abundance
Mineral Abundance comparison graph
Kelly Mineralogic image
Mineralogic mineral map
AZtec EDS montage map
AZtec EDS map

Sample 2 – Nickel deposit from Western Australia

The mineralogy of this deposit sample is fairly simple and is made up of minerals that are chemically distinct from each other. This allows for easier mineral distinction and identification with both standard techniques using Aztec and Mineralogic. The mineral assemblages calculated with both Aztec and Mineralogic do not vary significantly, suggesting that for mine deposits that have a simple mineralogy, with chemical distinct minerals, Mineralogic does not necessarily have a major advantage over standard techniques, and automated mineralogy may not provide a benefit in these circumstances.


Mineralogic vs Aztec Mineral assemblages
Mineral Abundance comparison graph
Mineralogic mineral map
Mineral Map of Sinclair
AZtec EDS montage map
EDS montage map Sinclair

Sample 3 – European gold deposit

This sample is low grade, preliminary data collected using standard electron microscopy techniques calculated an economic minerals (gold, silver and copper) abundance of 0.07%. Many of these grains were not picked up and identified using standard EDS mapping, on both Aztec and Mineralogic. However, on Mineralogic a specific work flow known as a Bright Phase Search allowed for targeted analysis. Only gold, silver and copper grains (which are made up of heavier elements, therefore are white on a Back Scatter Electron map) were analysed, allowing for a higher resolution (magnification) run to be undertaken at a faster time frame, as compared to analysing the whole sample. This allows for a large amount of data to be produced about the economic minerals within the sample, that can help to determine an efficient and economic extraction process.

BPS of gold sample
Mineral Maps of grains analysed using a Bright Phase Search

Conclusions

The ability to create personalised mineral lists specific to each sample depending on their assemblage and chemical makeup, enables more accurate and in depth characterisation of samples, which is automatically quantified without the need for additional data manipulation after acquisition. It also allows the separation of chemically similar minerals, rather than grouping them together, consequently allowing for more accurate modal abundances to be created. This additional control in mineral identification can be used to inform decisions within the resource recovery process. However, when looking at samples that have a simple mineralogy with chemically distinct minerals, the automated mineralogy software may not be necessary, as standard techniques produce the same results.

The ability to set up individualised work flows, such as the Bright Phase Search used with the European gold mine sample, allows for tailored analysis, reducing the running time of data collection, as well as creating more detailed data of desirable minerals and grains to be collected from low grade ores, but this can also be applied to higher grade ore types. This additional data that is specific to the desirable minerals in the samples saves time as well as helping to inform any decisions made in the extraction process

Mineralogic overall is a powerful piece of software that allows for quick, in depth quantitative characterisation whilst requiring less human input. The large amounts of varied quantitative data that can be collected is an advantage as having more ‘Information reduces uncertainty about decisions which have economic consequences’ (Gu et al, 2014). In mining, reducing the uncertainties can allow for a more efficient and economic extraction process to be determined and put in place.

References

Graham, S.D, Hill, E, Dominy, S.C, and Spratt, J., 2014. The application of Mineralogic, an automated mineralogy solution in mineral exploration. Society of Exploration Geologists – Building Exploration Capability for the 21st Century.

Gu, Y, Schouwstra, R.P, Rule, C., 2014. The value of automated mineralogy. Minerals Engineering. 58, 100-103.