An estimation of the variability in automated quantitative mineralogy measurements through inter-laboratory testing

Presently there are about 160 installations of automated mineralogy instruments such as QEM∗SCAN, MLA, TIMA, MINERALOGIC and INCAMineral. These instruments determine mineral quantities, and perform mineral liberation and mineral association analyses. Since the late 1990’s the scientific community of applied mineralogy and automated mineralogy has expressed the need of determining the variability in the measurements of mineral quantities and mineral liberation analyses through inter laboratory testing of suitable reference materials to improve confidence in the quantitative mineralogy measurements. A sample representing the −28+65 mesh size fraction of a sulphide flotation rougher concentrate from the Clarabelle mill in Copper Cliff, Vale (Sudbury, Ontario) was prepared for this testing. This rougher concentrate contains chalcopyrite, pyrrhotite, pentlandite, pyrite, quartz and feldspars as the dominant minerals. A 10-gram sub-sample was sent to the participating laboratories of the round robin. The sub-samples were analyzed by the participating laboratories to determine the mineral quantities, the liberation of chalcopyrite and the mineral association analysis for chalcopyrite. The findings indicate that there is a good agreement in the mineral quantities. The liberation analysis results also indicate a good agreement with the exception of two participants. These results indicate that correct mineral quantities do not necessarily imply correct mineral liberation. The testing also revealed that there was no consensus on how the mineral associations are treated and reported. We recommend that the modal analysis by liberation classes should be the preferred approach in reporting the mineral associations because it is more relevant to understand the mineral concentration operations.