Interpretable Machine Learning for Investigating the Molecular Mechanisms Governing the Transparency of Colorless Transparent Polyimide for OLED Cover Windows

Songyang Zhang,Xiaojie He,Peng Xiao,Xuejian Xia,Feng Zheng,Shuangfei Xiang,Qinghua Lu
DOI: https://doi.org/10.1002/adfm.202409143
IF: 19
2024-09-07
Advanced Functional Materials
Abstract:Innovative data‐driven approach to develop a complete pathway for transparent polyimide films with excellent properties for use in OLED, elucidating the molecular characteristics of the model and its interpretability based on molecular descriptors. With the rapid development of flexible displays and wearable electronics, there are a substantial demand for colorless transparent polyimide (CPI) films with different properties. Traditional trial‐and‐error experimental methods are time‐consuming and costly, and density functional theory based prediction of HOMO‐LUMO gap energy also takes time and is prone to varying degrees of error. Inspired by machine learning (ML) applications in molecular and materials science, this paper proposed a data‐driven ML strategy to study the correlation between microscopic molecular mechanisms and macroscopic optical properties. Based on varying degrees of impact of various molecular features on the cutoff wavelength (λcutoff), the ML algorithm is first used to quickly and accurately predict the λcutoff of CPI. Several new CPI films are then designed and prepared based on the key molecular features, and the predicted values of their λcutoff are effectively verified within the experimental error range. The interpretability provided by the model allows to establish correlations between the nine key descriptors identified and their physicochemical meanings. The contributions are also analyzed to the transparency of polyimide films, thereby giving insight into the molecular mechanisms underlying transparency modulation for CPIs.
materials science, multidisciplinary,chemistry, physical,physics, applied, condensed matter,nanoscience & nanotechnology
What problem does this paper attempt to address?