How it works

Wexalia offers liberation simulation, allowing further geometallurgical analysis without ever having to break your valuable diamond drill cores or other mineralogical images.

The Basics

Mineral liberation occurs through comminution of ore, and is a prerequisite for making a saleable concentrate. The 'spectrum' of liberation is a function particle size and the ore texture, and can vary greatly from one deposit to the next, or even within deposits. Moreover, sometimes it is not the ore mineral, but the gangue or deleterious minerals that matter most.

The Current Status Quo

Liberation analysis has advanced a lot with the advent of MLA, QEMSCAN and other automated mineralogical techniques, but it is still a slow process involving milling of ore, sizing the fractions and waiting for 4-8 weeks for it to be analysed. It provides a great level of detail but only for a small sample, and the cost and turnaround time prohibit doing this work on the scale to fully understand the variability of a deposit. This leaves metallurgists and geologists short of the resolution of data they require for process design and production planning.

Wexalia's Solution

Approximation of liberation was done as long ago as the late '30s by trail-blazing mineral processing engineer A.M. Gaudin. Wexalia's solution builds on PhD research and a publication by van der Wielen and Rollinson (2016) showing the similarity between QEMSCAN liberation spectra and those simulated using Voronoi tessellations.

The original simulation process was painstakingly slow, but with the help of Deep Digital Cornwall it has been accellerated by over three orders of magnitude. This, and advances in core scanning, now allows fast, cost-effective image-based liberation simulation.

Benefits

Understanding the mineral liberation characteristics of an ore deposit forms the foundation of proper metallurgical design and operation. Wexalia offers a step-change, going from small and discrete datasets to large continous dataset.

It does this cost-effectively and faster than conventional liberation analysis techniques, enabling applications on a much larger scale such as liberation analysis of core scanning images.

The Prerequisites

For the Wexalia liberation simulation process to function, there are three fundamental requirements:

• Mineral identification - Wexalia's value proposition is around simulation of liberation. We do not offer mineral identification from images and, as such, will require images to be provided as a 'mineral map'. This means that each pixel must be of known mineralogy that has been determined through some other means such as automated mineralogy (thin sections) or by core scanning data analysis.

• Accurately defined scale - To simulate a given particle size distribution, it is crucial that the scale of the image is accurately defined. In practice, this means knowing exactly how big a single pixel is.

• Pixel size - The minimum particle size that can be simulated with confidence is around 8x the pixel size. So for a pixel size of 5 x 5 micron, the smallest particle size that can be reliably simulated is around 40 micron. Smaller sizes are possible but the variety of particles that can be made with less than 8 pixels becomes limited.