A mineralogical approach to quantifying ore variability within a polymetallic Cu-Pb-Zn Broken Hill-type deposit and its implications for geometallurgy

Gordon, Henry Justinian Jnr (2019-12)

Thesis (MSc)--Stellenbosch University, 2019.

Thesis

ENGLISH ABSTRACT: A detailed two-part mineralogical study was undertaken on a polymetallic Cu-Pb-Zn ore from the Aggeneys-Gamsberg Ore District to quantify which mineral characteristics of the lithological ore types are problematic during flotation. The first part involved quantifying the bulk mineralogy, grain size distribution and textural characteristics from five primary lithological units (Quartz-Magnetite, Amphibole-Magnetite, Mineralised Schist, Sulphidic-Quartzite and Garnet-Quartzite) in order to propose early-stage geometallurgical domains prior to flotation testing. Strong bulk mineralogical and chalcopyrite grain size distribution patterns were selected as the partition between three early-stage geometallurgical domains. These domains were, the least variable Cu-rich quartz dominated Garnet-Quartzite domain, the Cu-Pb-rich quartz-dominated Lower Ore Body domain (consisting of the Mineralised Schist and Sulphidic Quartzite) and the most variable Cu-Pb-Zn rich magnetite dominated – Upper Ore Body domain (consisting of the Amphibole Magnetite and Quartzite-Magnetite). Six bulk samples that represent the magnetite-dominated Upper Ore Body domain were selected to test the process mineralogical variability of the Quartz-Magnetite and Amphibole-Magnetite subordinate lithological units, and thereby prove or disprove the proposed early-stage Upper Ore Body geometallurgical domain. Distinct metal zonation (head grade), liberation and mineral association configurations within these ores proved to be the defining variables that influenced their respective flotation responses. From this approach, the Cu-Pb Quartz-Magnetite domain (medium grade, best liberated and most straightforward), the Zn-Cu-Pb Pyroxmangite-Quartz-Magnetite domain (highest grade, moderately liberated and most complex) and the Cu-Pb Amphibole-Magnetite domain (lowest grade, poorest liberated and poorest quality) were resultant. However, provisions should be made to counteract the negative implications associated with the economic sulphide -, gangue sulphide - and non-sulphide gangue mineralogies of the three magnetite-dominated geometallurgical domains, if processed individually. Problematic process mineralogical features of the magnetite-dominated ores are: Slow floating sphalerite minerals (implications for Zn recovery); chalcopyrite disease (implications for selectivity between chalcopyrite and sphalerite); varying chalcopyrite and galena head grades (implications for Cu and Pb recovery); varying hardness and concentration ratios (implications for ore throughput); locked economic sulphide minerals (implications for recovery of gangue minerals) and fine-grained manganese minerals (implications for manganese entrainment). These problematic process mineralogical characteristics can potentially be neutralised through finer grinding (increased liberation and recovery of economic sulphide minerals), blending of Mn-rich and Mn-poor ores, and follow up quantitative mineralogical test work to ascertain the lower limit grain size at which manganese entrainment can be minimized. Integration of the above-mentioned domain considerations back into the geological block model is challenging due to limited mineralogy data and the incompatibility of the mineralogy and chemical assay data that define geological block models. As a result, an elemental proxy, the Cu:S ratio was presented as a quantitative variable that could be used to illustrate the differences between these domains in a 2-D and 3-D manner to inform the geologist and metallurgists about the expected variability. Secondly, a geometallurgical matrix was presented as a qualitative approach to ensure that the domains are accurately identified by the underground geologists and the variability of the mined feeds are efficiently communicated to the metallurgists. The approach of this study can contribute to the way in which the geometallurgical domains within other deposits are formulated.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/107227
This item appears in the following collections: