Variational quantum circuit-based quantum machine learning approach for predicting corrosion inhibition efficiency of pyridine-quinoline compounds

Abstract

This work used a variational quantum circuit (VQC) in conjunction with a quantitative structure-property relationship (QSPR) model to completely investigate the corrosion inhibition efficiency (CIE) displayed by pyridine-quinoline compounds acting as corrosion inhibitors. Compared to conventional methods like multilayer perceptron neural networks (MLPNN), the VQC model predicts the CIE more accurately. With a coefficient of determination (R2), root mean square error (RMSE), mean absolute error (MAE), and mean absolute deviation (MAD) values of 0.989, 0.027, 0.024, and 0.019, respectively, VQC performs better. The established VQC model predicts the CIE with outstanding predictive accuracy for four newly synthesized pyrimidine derivative compounds: 1-(4-fluorophenyl)- 3-(4-(pyridin-4-ylmethyl)phenyl)urea (P1), 1-phenyl-3-(4-(pyridin-4-ylmethyl)phenyl)urea (P2), 1-(4-methylphenyl)-3-(4-(pyridin-4-ylmethyl)phenyl)urea (P3), and quaternary ammonium salt dimer (P4). It generates remarkably high CIE values of 92.87, 94.05, 94.96, and 96.93 for P1, P2, P3, and P4, respectively. With its ability to streamline the testing and production processes for novel anti-corrosion materials, this innovative approach holds the potential to revolutionize the market.