Optimization of Stereochemical Disorder in Polycaprolactone (PCL) Nanofibers Using Machine Learning Models

Date of Award

12-2024

Degree Name

Master of Science

Department

Engineering Design, Manufacturing and Management Systems

First Advisor

Robert Tuttle, Ph.D.

Second Advisor

Robert Makin, Ph.D.

Third Advisor

Smitha Rao Hatti, Ph.D.

Keywords

Biomedical application, electrospinning, machine learning optimization, nanofiber characterization, polycaprolactone, stereochemical disorder

Access Setting

Masters Thesis-Abstract Only

Restricted to Campus until

12-1-2034

Abstract

Polycaprolactone (PCL) is a biodegradable polymer known for its versatility, yet optimizing its properties through controlled fabrication processes remains challenging. This thesis focuses on the optimization of electrospun PCL nanofibers using machine learning models to predict and enhance molecular ordering (S²) for biomedical applications such as tissue engineering, drug delivery systems, and biodegradable scaffolds. The research employs random forest (RF), support vector machine (SVM), and neural networks (NN) to model the relationship between key process parameters like PCL concentration, rotational speed (RS), applied voltage, and flow rate and their effects on nanofiber characteristics. This work also investigates the relationship between stereochemical disorder in PCL and its impact on the polymer's physical properties, such as tensile strength. These models help reduce trial-and-error experimentation and offer precise insights into the optimal fabrication conditions for PCL nanofibers. This study highlights the efficacy of machine learning in optimizing PCL nanofiber production and presents a framework for improving the quality and functionality of nanomaterials in biomedical applications.

This document is currently not available here.

Share

COinS