Application of near infrared sensors to preconcentration of hydrothermally-formed copper ore

Abstract

Near infrared sensors can be a very useful technique for the qualitative analysis of complex ores, and thus could be useful for the preconcentration of ores. In this paper, individual particles of hydrothermally-formed copper ore sampled from a mine in the Los Pozos mining district, northern Chile, were classified as product, middling and waste based on their near infrared response. The classification of copper bearing minerals (product) from gangue (waste) was considered for vibration combination bands at longer wavelengths from 2000 to 2400 nm. This region exhibits characteristic features for carbonates and hydroxyl gangue bearing minerals. The near infrared features at 1400 and 1900 nm were not considered favourable for classification and subsequent discrimination because they can be influenced by moisture and other environmental factors and are easily suppressed by iron-rich minerals. Two near infrared preconcentration strategies were applied for particle discrimination. Results indicate that targeting only the calcite (carbonate) dominated particles for discrimination as waste provided the best option for preconcentration. The near infrared discrimination analysis correlates well with mineralogical (QEMSCAN® and XRD) and elemental (XRF) data classification. The results indicate that near infrared spectroscopy is a suitable preconcentration method for supergene copper ore.