Deep learning-assisted design of eco-friendly intrinsic self-healing biopolymer/ZnO nanocomposite coatings for corrosion protection of aluminium 7075

Microcracking and corrosion significantly reduce the lifetime of aerospace alloys such as aluminium 7075. This work presents an eco-friendly intrinsic self-healing coating formulated from chitosan–gelatin biopolymers reinforced with ZnO nanoparticles, crosslinked with citric acid, and plasticized with glycerol. The coatings were applied by dip and brush coating methods and characterized using SEM, EDS, FTIR, XRF, tensile testing, microhardness, contact angle, and salt spray analysis. Dip-coated films demonstrated uniform morphology, enhanced ZnO incorporation, strong adhesion, ∼99% tensile recovery, and delayed corrosion initiation up to 48 h. Brush-coated samples, though thicker, exhibited non-uniformity and localized degradation. Scratch healing was attributed to reversible esterification-hydrolysis reactions supported by ZnO crack bridging. A CNN–LSTM hybrid model was trained on experimental datasets, achieving R2 > 0.95 in predicting crack closure, tensile recovery, and corrosion onset. Sensitivity analysis identified ZnO loading and coating thickness as key parameters. This study integrates sustainable materials and deep learning, offering a robust pathway for next-generation corrosion-resistant aerospace coatings.