Artificial Intelligence-Guided Electrospray Deposition of Polymeric Films

This grant supports research into an innovative manufacturing technique for producing thin polymer films. Polymeric materials play an essential role in the functionality of many devices and are vital in the microelectronics, healthcare, and energy industries. Electrospray deposition is increasingly recognized as an effective, low-cost method for creating versatile polymeric films.

However, a significant gap remains in understanding how the electrospray deposition process influences the characteristics of the resulting films, which has limited the widespread adoption of this technology in manufacturing. This research aims to bridge this knowledge gap by integrating experiments, computational modeling, and artificial intelligence/machine learning methods to develop a comprehensive framework for the electrospray deposition process.

This knowledge enables the development of a method for the production of high-quality, customizable polymeric films at scale. The ultimate goal is to establish electrospray deposition as a viable manufacturing tool that benefits various industries and, thus, the U.S. economy and prosperity. In collaboration with the Alliance for Manufacturing and Technology, the project also aims to grow the manufacturing workforce through student training programs and community engagement.

This research supports innovations in electrospray deposition, artificial intelligence/machine learning integration with physics-based models, and manufacturability. Electrospray deposition has the potential to be a powerful tool for polymeric film manufacturing, but the relationship between processing conditions and film characteristics, in particular the role of charge accumulation and decay in governing microstructure development in the film, remains unclear.

The complex coupling between charge transport and heterogeneous film connectivity (porosity) makes it difficult to use experimental testing alone to elucidate the nature of the processing-structure-property relationship. This research aims to discover how charge transport influences the characteristics of an electrospray-deposited film by combining experimental methods with physics-informed machine learning.

In particular, this work addresses the gap in physics-based modeling on the interplay between charge transport and structure evolution in dielectric granular films. Secondly, the artificial intelligence/machine learning methods, including the physics-informed multivariate Gaussian process and 3D porosity profiling, in this project help uncover the unobservable physics of electrospray deposition.

This enhances the physics-based model by integrating simulation and experimental data to enable deeper insights into the deposition physics with less time and effort. Regarding manufacturability, this methodology improves the efficiency of the models by refining the physics of particle behavior and charge decay. This reduces processing costs and enables the prediction of outcomes under various conditions, while also allowing for the customization of the deposition process.

This approach facilitates rapid design, inverse mapping, and the creation of polymer films with unique functionalities.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.