Prediction methods in coal and coke microscopy

SUMMARY Prediction of coke strength based on measurements of vitrinite reflectance and data from group‐maceral analyses requires information about the coking power of the different coal macerals in relation to rank. A range of bases are considered by different individuals to be important in prediction studies, e.g. the optimum ratio of reactive to inert constituents for each ‘V‐step’ (vitrinite oil reflectance step × 10); the proportion of vitrinite and clarite and rank; the amount of inertinite with no consideration given to an optimum ratio of reactives to inerts; the rank with little regard given to varying maceral composition. Entirely satisfactory results can be obtained by all investigators for coke‐strength predictions from certain coals under given coking conditions. For other coals, some of the determined values lie well outside the permissible limits of error. Reasons for these deviations may lie in an erroneous, or at least in an inadequate evaluation, of the different microscopical parameters. With this introduction, different predictive methods are discussed and tested. Limits of applicability are demonstrated. Possible improvements are suggested and questions, which have so far not been answered entirely satisfactorily, are posed. Above all, the underlying importance of light microscopy to a major industrial field is stressed.