Selection of rock depressants based on laboratory kinetic studies

"Rock rejection during the flotation of a copper-nickel ore was improved by adding carboxymethyl cellulose to both a rougher and cleaner flotation stage. The selection of the rock depressant was based on laboratory rates studies. Plant operating results are presented which confirm the laboratory selection of two particular brands of carboxymethyl cellulose.The rock assay of the concentrate was reduced from more than 40% to less than 15%.IntroductionThe traditional method of selecting flotation reagents relied upon batch bench scale tests, conducted under arbitrary conditions, to identify the effect of the reagent under examination. Too often these were ""one-off"" tests with little or no regard paid to the reproducibility of the results. Sophisticated researchers used statistical methods to design and evaluate the experiments but they were still obliged to rely on the arbitrary and often unreliable batch tests to generate the numbers that were analyzed statistically. An interesting side-effect of the enhanced sophistication of the statistical analysis was often that no significant effects would be uncovered in a series of experiments that would have generated many positive conclusions and much further experimentation if the same results had been evaluated by the traditional method.In the recent past, at least two groups of researchers a have directed their attention to the reliability of the batch experiments with a view toward improving the reproducibility of the test data. In this work the best of the improvements advocated by both of these groups have been adopted in order to generate kinetic data from batch tests that are as reproducible as possible.The objective of the work was to reduce the rock content of a bulk sulphide concentrate produced at Inco's Shebandowan mill. During 1982 the Shebandowan concentrate analyzed 3.5% Cu, 7.4010 Ni, 25% pyrrhotite and more than 40% rock, which resulted in an MgO content of 10.4%. Besides the expense of shipping all of this rock to Sudbury, the rock also contributed to the poor filterability of the concentrate. On favourable occasions the moisture content of the filter cake was slightly over 20% by weight. The magnesia content of the concentrate was deleterious to the reverberatory furnaces used by Inco for smelting nickel concentrate because the slag viscosity increased as the magnesia content increased, and the nickel lost in the slag went up."