Development of new models to predict the corrosion inhibition efficiency as functions of some molecular descriptors using statistical analysis

62 models have been proposed to predict the corrosion inhibition efficiency for a wide variety of organic and inorganic compounds using the statistical analysis. In this context, we used the experimental data of the corrosion inhibition efficiencies of 100 inhibitors (organic and inorganic) as an input data for a statistical analysis in order to determine the most adequate quantitative structure activity relationship (QSAR) models to predict the inhibition efficiency as function of some molecular descriptors. The experimental data used in this study are selected for the corrosion of steel in 0.5 M H2SO4 at 25 degrees C. Also, we choose the inhibition efficiencies determined by potentiodynamic polarization curves (IEp) and electrochemical impedance spectroscopy (IEEIS). Accordingly, the statistical analysis has been applied to collect the IEp and IEEIS models of corrosion inhibitors. The different models are compared to assess the performance of the best model. Global Performance Indicator (GPI) is computed to evaluate all models. The results show that the cubical model with a GPI of 1.27 is the best (M23) for IEEIS, and the linear model with a GPI of 0.64 for IEp.

Automated summary

62 models have been proposed to predict corrosion inhibition efficiency using statistical analysis. Experimental data for the corrosion inhibition efficiencies of 100 inhibitors were used to determine the most adequate quantitative structure activity relationship (QSAR) models. The best model was found to be the cubical model for IEEIS and the linear model for IEp.