Accelerated corrosion of nickel-200 walls of caustic evaporators causes contamination of the end product - caustic soda. An experimental design study was undertaken in order to understand the effect of process variables and liquor constituents on corrosion rate and other corrosion parameters of the evaporator walls. A matrix of experiments was set up with eight independent variables at the high and low ends of ranges expected in service. Corrosion parameters such as corrosion current densities, corrosion rate, corrosion potentials, etc were measured in polarization-resistance and potentiodynamic polarization measurements performed under conditions defined by the matrix. Additional sets of four experiments performed with the variables at mid values and four experiments in plant liquors provided data for estimation of errors and verification of empirical models. Empirical models accounting for variation ranging from 77 percent in Icorr to 99.5 percent in Ecorr were obtained. Values predicted from these models agreed well with those measured in experiments performed with plant liquors.
