Statistical design of experiments in briquette making of steel plant waste

Abstract

With changing environmental regulations, land-fill costs are increasing and sinter plants are becoming more difficult to operate within a newly established emission control standard. Alternate ways must be developed to dispose of, or recover, the valuable iron contained in the revert materials generated in the steel and iron making process. These materials are flue dust,, mill scale, iron ore fines and coke breeze, etc. Technologies have been developed to recover these materials. One of these technologies is cold briquetting. Hence, R&D studies have been undertaken to develop a suitable binder combination for producing briquettes from these wastes which can be charged into the blast furnace. The purpose of the present study is to determine the influence of the briquetting parameters on the quality of the product by means of experiments with a factorial design. A series of tests were carried out with different types of inorganic binders to determine their effects on briquette quality. In the present study, a full, two-level factorial experimental design was used to investigate the effect of three variables on the failure load of briquettes : amount of binder, briquetting pressure and amount of water addition. For three factors, the number of experiments required was 8 excluding replicates. A multi-linear mathematical model was successfully developed to predict the failure loaded. briquettes with correlation coefficient 'F' > 0.05 (where F = Fisher's coefficient). Abrasion tests established that the briquettes produced were sufficiently strong for handling outside storage and transportation. This was done in a rotating cylindrical drum containing four lifters which was used to determine abrasion indices of briquettes. Experimental evidence and mathematical analysis showed that the briquetting pressure and the amount of binder had marked effect on the strength of the briquettes.