Dissertation/Thesis Abstract

Understanding Thermally Induced Flow in Porous Media for Electrochemical Energy Applications
by Shum, Andrew D., Ph.D., Tufts University, 2020, 111; 27735801
Abstract (Summary)

Today, fossil fuels are a primary source of energy due to low prices and technology maturity. However, it is generally accepted that continued use of fossil fuels is not sustainable due to environmental impacts and eventual resource depletion. Many proposed solutions regarding renewable sources and a revised electric grid require dispatchable, sometimes distributable, energy storage/generation. Polymer-electrolyte fuel cells (PEFCs) are a promising energy-conversion technology not only for this purpose but also for the transportation industry. Using hydrogen gas and oxygen gas, PEFCs produce electric current and exhaust water.

One of the remaining issues with PEFCs is water management. Achieving effective water management relies on a detailed understanding of the various water transport phenomena at work. Pressure gradients have been studied thoroughly through both experimental (including x-ray computed tomography (CT)) and computational means. Evaporative transport, however, has been studied primarily through computational means due to its complexity. In this work, synchrotron-based micro-scale x-ray CT is used to perform in-situ observation and measurement of evaporation in the porous carbon media used for PEFCs. Infrared thermography is used in addition to x-ray CT in order to observe the impact of various factors on the formation of cracks in the catalyst layer during fabrication. Lastly, out-of-the-box machine learning, including convolutional neural networks, are explored as a means for improving the speed and accuracy with which x-ray CT data can be processed for these materials.

Indexing (document details)
Advisor: Zenyuk, Iryna V
Commitee: Chiesa, Luisa, Xiao, Xianghui, Shimpalee, Sirivatch
School: Tufts University
Department: Mechanical Engineering
School Location: United States -- Massachusetts
Source: DAI-B 81/8(E), Dissertation Abstracts International
Source Type: DISSERTATION
Subjects: Alternative Energy, Materials science, Mechanical engineering
Keywords: Convolutional neural network, Evaporation, Fuel cell, Machine learning, Phase-change-induced flow, X-ray computed tomography
Publication Number: 27735801
ISBN: 9781658416566
Copyright © 2020 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy
ProQuest