AI-guided design of green corrosion inhibitors for carbon steel in acidic media: integrating DFT, molecular dynamics, and electrochemical validation

• M.M. Hegazi¹, Mai A. Khaled² and A.S. Fouda³

¹ Department of Pharmaceutical Chemistry, College of Pharmacy, Al-Rafidain University, Palestine Street, 10052 Baghdad, Iraq
² Basic Science Department, Faculty of Engineering, Horus University, New Damietta 34511, Egypt
³ Department of Chemistry, Faculty of Science, Mansoura University, Mansoura-35516, Egypt

Abstract: The corrosion of carbon steel in acidic environments remains a critical challenge in industrial processes such as acid pickling, oil well acidizing, and chemical cleaning. Traditional effective synthetic inhibitors often contain toxic components that pose significant environmental and health hazards. The development of environmentally benign corrosion inhibitors derived from natural sources has emerged as a promising alternative aligned with green chemistry principles. In this study, an artificial intelligence-guided screening approach was employed to identify promising green corrosion inhibitors from a library of naturally occurring flavonoids. Machine learning models trained on molecular descriptors predicted inhibition efficiencies, and the top three candidates – quercetin, caffeic acid, and hesperidin – were subjected to comprehensive computational and experimental validation. Density functional theory (DFT) calculations at the B3LYP/6-311++G(d,p) level provided quantum chemical parameters, while molecular dynamics (MD) simulations evaluated adsorption behavior on the Fe(110) surface. Electrochemical techniques, including potentiodynamic polarization (PDP) and electrochemical impedance spectroscopy (EIS), were used to validate the inhibition performance in 1.0 M HCl solution. Quercetin exhibited the highest inhibition efficiency of 94.2% at 300 ppm and 95.7% at 500 ppm, followed by caffeic acid (89.7%) and hesperidin (86.5%) at 300 ppm. DFT analysis showed that quercetin possessed the highest HOMO energy (–5.42 eV) and the highest fraction of electron transfer (ΔN=0.721), indicating superior electron-donating capability and adsorption affinity. MD simulations confirmed strong adsorption energies ranging from –485.6 to –567.3 kJ/mol at 300 ppm. Electrochemical measurements demonstrated that all three compounds functioned as mixed-type inhibitors, with charge-transfer resistance increasing markedly upon inhibitor addition. Surface characterization via SEM, AFM, and XPS confirmed the formation of a protective adsorbed film on the carbon steel surface. The integration of AI screening with computational chemistry and experimental validation provides an efficient and reliable framework for discovering green corrosion inhibitors. The identified flavonoid compounds offer excellent protection for carbon steel in acidic media and represent viable, environmentally sustainable alternatives to conventional toxic inhibitors.

Keywords: green corrosion inhibitors, DFT calculations, molecular dynamics simulation, carbon steel, acidic media, electrochemical validation

Int. J. Corros. Scale Inhib., 2026, 15, no. 3, 633-653
doi: 10.17675/2305-6894-2026-15-3-30

Download PDF (Total downloads: 17)

Back to this issue content: 2026, Vol. 15, Issue 3 (pp. 1-... (in progress))