Date of Award

Fall 2025

Document Type

Open Access Dissertation

Department

Mechanical Engineering

First Advisor

Xingjian Xue

Abstract

Oxygen plays a central role in numerous industrial processes. Several technologies have been reported for producing oxygen from air. Among them, oxygen transport membrane (OTM) technology—based on mixed-conducting, gas-tight ceramic membranes—has attracted significant attention due to its high oxygen selectivity, relatively low capital and operating costs, and versatility for both ex-situ and in-situ applications. Mathematical modeling of OTMs offers a powerful tool to investigate internal multi-physics transport phenomena, providing deeper insight into fundamental mechanisms while serving as a cost-effective approach for optimizing membrane stack designs.

After a comprehensive introduction, in the first part of this dissertation (chapter 2), a comprehensive hollow fiber membrane model was developed, grounded in an experimental system as the physical basis. The model couples Multiphysics transport processes within the membrane with large-scale thermal–fluid transport in the test assembly and furnace. Systematic parametric studies are performed to investigate fundamental mechanisms and assess membrane performance. The results show that the oxygen partial pressure on the permeate side increases asymptotically from the inlet to the outlet of the hollow fiber membrane. In contrast, the longitudinal distribution of oxygen vacancy concentration decreases along the same direction, while the oxygen flux distribution follows the profile of oxygen vacancy concentration at the permeate surface. High feed-side air pressure and low permeate-side gas pressure are found to enhance oxygen permeation performance. Among the transport resistances, surface exchange resistance at the permeate side dominates, whereas bulk diffusion resistance contributes minimally in substrate-supported thin-film hollow fiber membranes. Overall, the modeling study provides valuable insight into the underlying mechanisms and offers practical guidance for membrane design and operation to improve oxygen production efficiency.

To enable practical applications, upscaling from a single hollow fiber membrane to stacks and modules is essential. However, experimental methods for evaluating upscaling strategies are both time-consuming and costly. Mathematical modeling, by contrast, offers a cost-effective and flexible tool for this purpose. In the second part of the dissertation (chapter 3), building upon experimental results from a proof-of-concept hollow fiber membrane stack, a computational fluid dynamics (CFD)-based Multiphysics stack model was developed and validated. Extensive simulations were performed to examine stack behavior under varying operating conditions, and different design strategies were evaluated to optimize stack performance. The oxygen permeation process is thermally activated. Increasing the argon sweep gas flow rate lowers the oxygen partial pressure around the shell sides of the hollow fibers and enhances the permeation flux. Along the lumen side, oxygen partial pressure rises from the inlet to the outlet, with flux significantly higher in the upstream region than downstream. A distinct gradient of oxygen vacancy concentrations is observed across upstream fiber sections, from shell to lumen surfaces, but this gradient diminishes downstream. For a fixed stack length, adding more hollow fibers increases the overall permeation rate but reduces flux. Oxygen partial pressure decreases radially from the periphery to the center of the stack, leading to lower average flux in inner layers. An appropriate packing density is therefore required to achieve compact design while limiting pressure losses. For a given total fiber length, an optimal fiber number exists that maximizes average permeation performance. Diffusive oxygen flux dominates near fiber walls, while convective flux becomes increasingly important toward the fiber center. A higher sweep gas flow rate reduces the region where diffusion dominates.

To maintain the elevated temperatures required for membrane operation, a high-temperature furnace is typically used. However, this results in low heating power efficiency and makes rapid temperature changes difficult due to the large volume of the furnace. Recently, a novel strategy has been employed in which external electrical power is directly applied to a hollow fiber membrane, enabling compact self-heating. To better understand the fundamental mechanisms, a mathematical model is developed in the third part of the dissertation (chapter 4) for a self-heated hollow fiber oxygen separation membrane, assisted by vacuum conditions applied at the lumen-side outlet. Comprehensive simulations are conducted to study the effects of self-heating on Multiphysics transport processes and oxygen permeation performance. Additional simulations are performed to investigate the influence of electrical field orientations applied to the hollow fiber membrane and the vacuum levels at the lumen-side outlet. The associated fundamental mechanisms are discussed and elaborated. A higher applied voltage increases the average membrane temperature. The longitudinal temperature profile is non-uniform, with a maximum in the mid-region and steep decreases toward both ends. Oxygen permeation flux rises with applied potential and is further enhanced by higher vacuum levels (lower permeate-side oxygen partial pressure). The flux distribution shows a domed shape along the fiber length, approaching zero near the ends. At low potentials, the effect of vacuum level is negligible but becomes significant at higher potentials. On the feed side, oxygen concentration decreases from the bulk to the surface, while on the permeate side it decreases from the surface toward the lumen outlet. Both gradients intensify with increasing voltage and/or vacuum level. Across the membrane bulk, oxygen vacancy concentration increases from feed to permeate surfaces, with steeper gradients under higher potentials and stronger vacuum. The orientation of the applied potential strongly influences performance. Alignment with the permeation direction, with positive and negative electrodes connected to permeate and feed surfaces respectively, greatly enhances flux. The opposite configuration suppresses it, while a perpendicular potential has little effect on radial ion transport.

Building on these advances, Chapter 5 integrates the modeling and heating strategies into a Joule-heated hollow fiber ion transport membrane reactor for methane oxidative coupling (OCM). Results demonstrate that methane plays a dual role: it is both the feedstock for conversion and a promoter of oxygen transport by lowering surface oxygen partial pressure and sustaining higher vacancy-driven flux. The coupled transport–reaction model reveals that gradual, membrane-mediated oxygen delivery significantly improves C₂ selectivity compared with conventional co-feed reactors. Methane–vacancy interactions, heterogeneous surface reactions, gas-phase chemistry, and localized Joule heating jointly shape flux, product selectivity, and thermal profiles. This chapter demonstrates the dual functionality of MIEC membranes as both oxygen separators and catalytic reactors, providing a path toward intensified, energy-efficient chemical production.

Finally, Chapter 6 synthesizes the insights, highlighting how Multiphysics modeling bridges the gap between laboratory observations and industrial-scale application. By clarifying oxygen transport mechanisms, optimizing stack designs, enabling compact Joule-heating strategies, and extending membranes into reactive processes, this dissertation contributes both fundamental knowledge and practical guidance. The results position MIEC hollow fiber membranes not only as efficient oxygen separators but also as versatile platforms for low-carbon energy, 𝐶𝑂₂ management, and sustainable chemical synthesis.

Rights

© 2025, Hamed Abdolahimansoorkhani

Download

Share

COinS