Publication Date

8-2022

Date of Final Oral Examination (Defense)

5-23-2022

Type of Culminating Activity

Dissertation

Degree Title

Doctor of Philosophy in Materials Science and Engineering

Department

Materials Science and Engineering

Supervisory Committee Chair

Eric Jankowski, Ph.D.

Supervisory Committee Co-Chair

Matthew D. King, Ph.D.

Supervisory Committee Member

Christopher R. Iacovella, Ph.D.

Supervisory Committee Member

Matthew L. Ferguson, Ph.D.

Abstract

We apply molecular simulation to predict the equilibrium structure of organic molecular aggregates and how these structures determine material properties, with a focus on software engineering practices for ensuring correctness. Because simulations are implemented in software, there is potential for authentic scientific reproducibility in such work: An entire experimental apparatus (codebase) can be given to another investigator who should be able to use the same processes to find the same answers. Yet in practice, there are many barriers which stand in the way of reproducible molecular simulations that we address through automation, generalization, and software packaging. Collaboration on and application of the Molecular Simulation and Design Framework (MoSDeF) features prominently.

We present structural investigations of organic molecule aggregates and the development of infrastructure and workflows that help manage, initialize, and analyze molecular simulation results through the following scientific applications (1) A screening study wherein we validate self-assembled poly-3-hexylthiophene (P3HT) morphologies show the same state dependency as in prior work, and (2) A multi-university collaborative reproducibility study where we examine modeling choices that give rise to differences between simulation engines. In aggregate, we reinforce the need for pipelines and practices emphasizing transferability, reproducibility, useability, and extensibility in molecular simulation.

DOI

https://doi.org/10.18122/td.1968.boisestate

Download

Files over 30MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS