Shivank Shukla

Advisor: Prof. Rampi Ramprasad

will defend a doctoral thesis entitled,

ADVANCED POLYMER DESIGN FOR EXTREME ENVIRONMENTS: 

HIGH-ENERGY DENSITY CAPACITORS AND AI-DRIVEN PROPERTY OPTIMIZATION

On

Tuesday, Feb. 17, 2026

11: 30 am - 1:30 pm

MRDC Room 3515

or virtually via Teams:

Teams link

Committee:

Prof. Rampi Ramprasad- School of Materials Science and Engineering (advisor)

Prof. Naresh Thadani- School of Materials Science and Engineering (co-advisor)

Prof. Chaitanya Deo- School of Materials Science and Engineering

Prof. Roshan Joseph- School of Industrial and Systems Engineering

Prof. Guoxiang (Emma) Hu- School of Materials Science and Engineering

Abstract:

Designing functional polymers for targeted applications remains a highly complex and non-trivial challenge due to the vastness of chemical design space and the strong coupling between molecular structure and macroscopic properties. This challenge is further amplified when engineering polymers for extreme operating conditions, where few or no existing materials satisfy performance requirements. This thesis addresses one such regime: polymer dielectric membranes for high-energy-density capacitors operating at elevated temperatures, where most commercially available polymer dielectrics exhibit a pronounced decline in energy density.

In this work, a comprehensive and scalable polymer design framework is developed that integrates data-driven modeling, machine learning, synthesis aware polymer generation, and high-throughput molecular simulations, followed by experimental validation. The proposed pipeline enables the rapid exploration and screening of millions of chemically feasible polymer candidates, reducing the materials discovery timeline from years to months. Machine learning models are trained across multiple polymer classes to enable fast and accurate prediction of dielectric and thermomechanical properties. In parallel, a robust molecular dynamics–based simulation protocol is established to predict the glass transition temperature of polymers, serving both as a validation tool and as a secondary data source to augment model training.

Collectively, this integrated approach demonstrates an efficient pathway for the discovery and design of polymer dielectrics with improved high-temperature performance and establishes a generalizable framework for polymer materials design under extreme operating conditions.