Frequency-dependent dielectric constant prediction of polymers using machine learning

The dielectric constant (epsilon) is a critical parameter utilized in the design of polymeric dielectrics for energy storage capacitors, microelectronic devices, and high-voltage insulations. However, agile discovery of polymer dielectrics with desirable epsilon remains a challenge, especially for high-energy, high-temperature applications. To aid accelerated polymer dielectrics discovery, we have developed a machine-learning (ML)-based model to instantly and accurately predict the frequency-dependent epsilon of polymers with the frequency range spanning 15 orders of magnitude. Our model is trained using a dataset of 1210 experimentally measured epsilon values at different frequencies, an advanced polymer fingerprinting scheme and the Gaussian process regression algorithm. The developed ML model is utilized to predict the epsilon of synthesizable 11,000 candidate polymers across the frequency range 60-10(15) Hz, with the correct inverse epsilon vs. frequency trend recovered throughout. Furthermore, using epsilon and another previously studied key design property (glass transition temperature, T-g) as screening criteria, we propose five representative polymers with desired epsilon and T-g for capacitors and microelectronic applications. This work demonstrates the use of surrogate ML models to successfully and rapidly discover polymers satisfying single or multiple property requirements for specific applications.