An experimental investigation of corrosion resistance of mild steel by sol-gel process using TiO2 nanostructure coating and prediction of optimal parameters

Abstract

A TiO2 nanostructure coating has shown to significantly improve the thermal and electrical properties of steel plates and increase their resistance to oxidation, corrosion, and wear, especially in high temperature applications. In this research, the corrosion resistance properties of a mild steel substrate by applying TiO2 nanostructure coating using the sol-gel method were investigated. The quality of the coating, however, is notably affected by such process parameters as the dip-coating rate, drying time, heat-treatment rate, and the number of coating layers. Moreover, this article presents an integrated approach to the optimal parameter setting for the above process. Using experimental data from a coating process by the sol-gel method, an artificial neural network is trained to map the vector of process parameters onto a measure of corrosion resistance. An evolutionary search algorithm is then employed to find the optimum set(s) of process parameters. The efficiency of the proposed approach is demonstrated using a case study involving a 316L stainless steel substrate.