Chitosan-Derived Green Corrosion Inhibitor for Carbon Steel in CO 2 -Saturated Media: A Graph Convolutional Network Approach

Creating environmentally friendly corrosion inhibitors is vital for sustainable operations in the oil and gas industry. This study presents an integrated green chemistry and deep learning approach to design and test chitosan-grafted polyacrylamide (CsAM) as a biodegradable corrosion inhibitor for carbon steel in CO 2 -rich environments. A graph convolutional network, trained on a curated data set of over 70 inhibitors, predicted that CsAM could achieve approximately 84% inhibition at 200 ppm, guiding experimental efforts. Four CsAM formulations with different chitosan-to-polyacrylamide ratios were synthesized and characterized by FTIR, SEM, and contact angle analysis to demonstrate beneficial functional and surface-active properties. Electrochemical tests showed that the 1:30 CsAM ratio achieved an impressive 98% inhibition efficiency, acting as a mixed-type inhibitor with physisorption as the main adsorption mechanism. These findings demonstrate that combining AI-based molecular prediction with sustainable polymer synthesis can significantly accelerate the development of effective green inhibitors while lowering reliance on toxic alternatives. The proposed computational–experimental approach offers a scalable pathway to creating high-performance, environmentally friendly corrosion control solutions for industry use.