Mineral processing plant design and operation are underpinned by a knowledge of ore mineralogy and grain size, but the influences of ore textures on mineral processing are very poorly understood. This is exacerbated by present geological data communicated to the mine metallurgists as guides to the likely mineral processing behaviour of the ore. The textural variation and character of ore is rarely noted on a systematic basis.
The first aim of this study was to define the textural variation in a group of ore lenses constituting the as yet undeveloped Hilton North deposit, a stratiform silver-lead-zinc ore body, 22km north of Mount Isa, northwest Queensland. This work was undertaken to determine whether it was possible, by visual logging of drill core, firstly, to characterise persistent mesotextural types, and secondly, identify any systematic mesotextural distribution throughout the ore lenses.
The third aim of the study was to investigate the microtextural composition of the mesotextures to determine any systematic relationships between mesotextures and microtextures.
The fourth aim of the study was to investigate possible relationships between ore textures and the liberation of ore minerals by grinding, as a first step in an attempt to better define the links between ore mineralogy, ore texture, and minerals processing behaviour of the ore. At the outset, it became apparent that much confusion existed in the literature in the use of the terms structure, texture, and fabric. A comprehensive review of previous applications for these terms and a recommended practice for their future use is presented in Part A Chapter 1 of this thesis.
Liberation is strongly influenced by the arrangement of ore minerals at a microscopic scale (microtextures). However, the study of microtextures is a laborious and time consuming one, and a faster approach was needed. To this effect, a new logging method looking at the arrangement of ore minerals at a mesoscopic scale (mesotextures) was conceived, to use in conjunction with normal mine logging. A relationship between the mesotextures and the microtextures was hypothesised, enabling a geologist to approximate microscopical composition of the ore during logging. The influence of the microtextures on liberation would provide the final link in estimating liberation from a mesoscopic scale.
Detail logging of cores from Hilton North established that seven mesotextures could be recognised in the ores. The seven mesotextures identified were logged in five of the twelve recognised mineralised zones, and their abundances were modelled on strike sections (or long sections). Reflected light microscopic study of polished samples of the mesotextures revealed twelve microtextures. The hypothesised relationship between certain microtextures and mesotextures was observed. The distribution of mesotextures in C and D ore bodies (the two principal ore bodies) showed an influential relationship with preliminary metallurgical testing results As a consequence, liberation data was obtained for mesotextures dominated by sphalerite and galena. A relationship between the liberation results and the mesotextural zoning was observed. A correlation between liberation results and recovery results from bulk flotation tests (done by Mount Isa Mines Limited) was also observed. The chemical properties of the ore were not considered in this study.
The ultimate goal of research in this subject would be to increase recovery by enabling the geologist to convey information pertaining to metallurgical performance to the metallurgist. The data would be used in conjunction with grades as a tool for selective mining, increasing recovery by selecting or blending ores of known liberation behaviour, and treating them. It can also be used as a prevention tool against sudden changes in metallurgical behaviour of the ore in the plant, in the sense that a knowledge of textural change, and its likely impact on mill recovery, can be predicted from drill hole samples. Changes in metallurgical behaviour commonly result in a decrease in recovery, and their prediction from textural data may enable selective mining or blending to minimise change.
As a result of its positive findings, this research concludes with a methodology for the textural analysis of an ore body as an indicator of its liberation characteristics.