AI-designed epoxy nanocomposite coatings offer fire safety and self-sensing features

Researchers at Hanoi University of Industry have developed an advanced epoxy nanocomposite coating that integrates fire safety with self-sensing capabilities. This study, published in Polymer Bulletin, highlights the use of machine learning to optimize a hybrid nanocomposite composed of epoxy resin, multi-walled carbon nanotubes (MWCNTs), and nanoclay. The MWCNTs provide electrical conductivity and environmental responsiveness, while nanoclay enhances flame retardancy and mechanical strength.

The team, led by Tuan Anh Nguyen, employed a Box–Behnken design and a Random Forest model to navigate the complex formulation space. This approach allowed them to predict material properties with high accuracy, achieving R² values of 0.986 for limiting oxygen index, 0.995 for tensile strength, and 0.998 for electrical conductivity. The model's predictions were validated through laboratory experiments, confirming the coating's performance across multiple criteria.

The optimized coating demonstrated a limiting oxygen index of 28.0%, a peak heat release rate of 720 kW/m², and a tensile strength of 80 MPa. Its electrical conductivity and self-sensing capabilities were attributed to the percolating network of carbon nanotubes, which respond to temperature changes and ammonia exposure. This multifunctional performance positions the coating as a promising candidate for applications in construction and industrial safety.

The study's broader impact lies in its reproducible framework, which combines efficient experimental design, machine learning, and multi-objective optimization. This methodology can be applied to other multifunctional materials, offering a strategic advantage in navigating complex design spaces. The research received support from Hanoi University of Industry's Advanced Materials and Sustainable Technologies research group.