Column Leaching Procedures For Non-Homogeneous Ores

The leaching characteristics of non-homogeneous ores have traditionally been difficult to assess in laboratory tests. A result of the large sample size necessary to ensure representativity, this difficulty is most acute for ores containing nuggets or other irregular occurrences of precious minerals. Consequently, small-scale leaching tests are often regarded as indicative rather than conclusive. The predictive nature of leaching tests can be improved by applying the experimental design concepts of blocking and randomization. In the case of column leaching, experimental 'blocks" are constructed with the same number of columns as there are separate test conditions. These blocks are treated identically throughout the test, with differences between columns within each block being the variables of interest. To maximize the uniformity of the blocks, randomization procedures are employed in the preparation of the sample and conduct of the test. The initial ore sample is screened and individual column charges composited from randomly selected increments from each size fraction. The leaching order, daily solution analyses, and processing of leach residues are similarly randomized. The observed differences in leaching performance between the columns in each block are compared to determine the over-all outcome of the test. A statistical analysis using t distributions gives a quantitative measure of the reproducibility of the outcome.