Date on Master's Thesis/Doctoral Dissertation

5-2024

Document Type

Master's Thesis

Degree Name

M. Eng.

Department

Chemical Engineering

Degree Program

JB Speed School of Engineering

Committee Chair

Fried, Joel R.

Committee Member

Jaeger, Vance

Committee Member

Berson, Eric

Committee Member

Narayanan, Badri

Author's Keywords

polymers; simulations; mechanical properties; molecular dynamics; Materials Studio

Abstract

This work explores mechanical properties of membranes made of polybenzimidazole utilizing molecular simulations. These simulations researchers to narrow down polybenzimidazole membrane compositions to viable candidates. These simulations are quicker than current purely experimental methods. The increase in throughput would be achieved by eliminating compositions that have unfavorable mechanical properties through preliminary simulation work before moving onto experimental methods. Determining polybenzimidazole membranes’ mechanical properties experimentally would, therefore, allow for researchers to speed up research in areas such as fuel cells and electrolyzers that make use of polybenzimidazole membranes. This study uses Materials Studio and the COMPASS III forcefield to perform these molecular dynamics simulations. An amorphous cell made up of m/p-polybenzimidazole, phosphoric acid, and water was generated [matching the makeup of polybenzimidazole membranes created experimentally by Xin Li, Laura Ann Murdock, and Brian C. Benicewicz] then dynamics simulations were performed on these cells to equilibrate pressure and volume. Finally, mechanical properties’ simulations were performed on the cells to determine their mechanical properties, specifically their density, Young’s, bulk and shear modulus, and their Poisson’s ratios. The values generated by these simulations were compared to the values determined experimentally. Strengths and weaknesses of the

computational approach are discussed and the viability of using molecular dynamics for screening candidate molecules is explored. The results generated in this study were determined to accurately simulate the density of all explored polymer compositions. The Young’s modulus for the compositions of neat polybenzimidazoles were also determined to be accurately simulated. The study failed to determine the Young’s modulus of other polymer compositions. Potential explanations for this failure are discussed and include the simulation model not adequately accounting for the hydrogen bonding in the compositions that contain water and phosphoric acid in addition to polybenzimidazole. Finally, future steps that could be taken to advance mechanical property simulations of polybenzimidazoles are discussed including the use of more robust software and different techniques that could be explored.

Share

COinS