Quantification of the influence of chemical composition on the hot metal viscositty

The present work aims at quantifying the influence of chemical composition and temperature on the hot metal viscosity. The experiments were conducted on specially assemble apparatus and data were analyzed with the help of artificial neural networks. Samples utilized were collected from CSN's blast furnaces main runners and some had their chemical composition modified to allow the evaluation of silicon, manganese, phosphorus and titanium, besides temperature, on the hot metal viscosity. Results were satisfactory for the available conditions. The viscosity was, in this work, indicated by the electric current through the motor, rotating which drove at constant speed, the spindle immersed in a molted iron sample. The work opens the field that is a little explored in spite of its importance for blast furnace operation and campaigns term.