A pioneering study from the East China University of Science and Technology has unveiled an AI-driven materials-genome approach (MGA) that accelerates the development of high-performance polyimide films, critical for aerospace, flexible electronics, and micro-display technologies. Published on 2 September 2025 in the Chinese Journal of Polymer Science, the research introduces a machine-learning framework that optimizes the mechanical properties of thermosetting polyimides, achieving a breakthrough formulation, PPI-TB, with superior strength, stiffness, and toughness compared to existing benchmarks.
Polyimide films are prized for their thermal stability and insulation but face challenges in balancing mechanical properties like modulus, tensile strength, and elongation at break. Traditional synthesis methods, reliant on trial-and-error, are slow and costly, limiting exploration of complex molecular designs. The new AI-assisted approach overcomes these hurdles by integrating computational modeling, experimental data, and machine learning to predict and optimize multiple mechanical properties simultaneously.
The research team developed Gaussian process regression models trained on over 120 experimental polyimide datasets, treating structural components—dianhydride, diamine, and end-capping units—as „genes.“ This framework enabled the analysis of 1,720 phenylethynyl-terminated polyimides (PPIs), predicting Young’s modulus, tensile strength, and elongation at break with high accuracy (R² ? 0.70–0.74). Through virtual screening, the team identified PPI-TB, a formulation that outperformed established polyimides like PETI-1 and O-O-3, achieving a modulus of 3.48 GPa and exceptional toughness. Experimental validation confirmed the model’s predictions, showing strong alignment between predicted and measured performance.
Analysis of molecular „genes“ revealed key design insights: conjugated aromatic structures enhance stiffness, heteroatoms and heterocycles strengthen molecular interactions, and flexible silicon- or sulfur-containing units improve elongation. These findings provide a roadmap for tailoring polyimides with balanced mechanical properties.
The AI-driven MGA offers a scalable, efficient framework for polymer design, drastically reducing development time and costs. Its applications extend beyond polyimides, promising advancements in lightweight, durable, and thermally stable materials for aerospace composites, flexible electronics, and microelectronics. Funded by the National Key R&D Program of China and the National Natural Science Foundation of China, this work marks a significant leap in materials innovation, paving the way for next-generation high-performance polymers.
DOI
