Modelling mineralogy from whole rock assay data – a case study from Productora Cu-Au-Mo deposit, Chile

Mineralogy is a fundamental characteristic for a given rock mass throughout the mining value chain. Understanding the mineralogy of the bulk rock, including both the gangue and ore components, is critical when predicting processing behavior and waste characteristics. Throughout the exploration and resource development process, mineralogical data are collected mostly in a qualitative manner through visual logging. These datasets are subjective and commonly inconsistent. Current methods for quantitative estimates of bulk mineralogy (e.g. X-ray point counting using SEM-EDS based software packages, and QXRD) are expensive and often very slow.

We present two new approaches to predicting bulk mineralogy using commonly available 33-element geochemical assay data. Firstly, we demonstrate qualitative estimation from assay using geochemical discrimination plots and validate this approach using quantitative XRD data. Secondly, we demonstrate quantitative estimation by calculated mineralogy. These approaches are presented using the Productora Cu-Au-Mo deposit, Chile, as a case study. Productora provides an example of an early stage project where geometallurgical models of mineralogy have been used to mitigate risk and uncertainty.

Our results indicate that robust, deposit-wide, predictions of bulk mineralogy can be generated quickly and cost effectively from geochemical assay data. These methods provide semi-quantitative mineralogical data for every multi-element geochemistry sample interval across the whole deposit. The number and spatial distribution of mineralogical data points can be significantly increased, by orders of magnitude. This data is then available to construct geometallurgical models and provides truly representative, deposit wide, mineralogical variability data. In turn, this provides significant benefits for the selection of representative samples for metallurgical test work, or the siting of specific metallurgical test work holes. Quantitative estimates of mineralogy on each assay interval can also be directly incorporated into resource block models, for mine planning purposes.