Date of Award


Document Type

Open Access Dissertation


Mechanical Engineering

First Advisor

Kevin Huang


Fossil fuels are the dominant energy source powering our modern society. However, burning fossil fuels emits carbon dioxide (CO2), a greenhouse gas that can cause climate change and ultimately threaten the survival of humanity. Effectively mitigating CO2 emissions from the use of fossil fuels has become an intense subject of scientific research as well as political debate in recent years. The current mainstream technical approach to achieving that goal is to curb the emission of CO2 by capturing CO2 at point-sources and geologically storing it. The CO2 separation and capture process, the first step toward the ultimate storage of CO2, can be applied to various stages of the combustion of fossil fuels: post-combustion, pre-combustion and oxy-combustion Significant technical progress in materials development and engineering design of carbon capture systems has been made over the past decades, but the major challenge to the commercial deployment of these technologies remains to be the energy penalty associated with CO2 capture, compression and storage (CCS) that considerably lowers the overall plant efficiency and ultimately increases the cost-of-electricity produced.

The state-of-the-art technologies for CO2-capture are principally based on reversible chemical/physical sorption processes using liquid solvents and solid sorbents as a CO2 scrubber and on size-exclusion permeation using membranes as a CO2 molecular filter. However, as aforementioned, the solvents and sorbents based technology is costive, cumbersome and inefficient, and the membrane technology is susceptible to poor selectivity and incompatibility to high temperatures. As of today, only a few of these technologies are considered commercially viable for large-scale application.

In this research, a new type of electrochemical CO2 separation membranes that are in theory exclusively permeable and can continuously separate CO2 from either a pre-combustion or post-combustion industrial stream have been investigated. This category of membranes functions as a mixed conductor, through the surface electrochemical reactions of which CO2 molecules can be electrochemically transported from the targeted stream to an isolated and concentrated stream that can be readily pressurized, transported and stored. Since these membranes only allow the electrochemically active species to transport through the membrane under a gradient of electrochemical potential, its selectivity is exclusive. Furthermore, this type of membrane normally operates at elevated temperatures in a continuous fashion, presenting excellent compatibility with high-temperature process streams. There are two types of mixed conductors envisioned for CO2 separation: 1) mixed carbonate-ion and oxide-ion conductor (MOCC); 2) mixed carbonate-ion and electron conductor (MECC). The former is more suited for pre-combustion CO2 separation while the latter is more applicable to post-combustion CO2 separation. These CO2 separation membranes should also be technically and economically more attractive than conventional electrically driven, molten carbonate fuel cell based CO2 concentrator since no external electronics are needed.

The objective of this proposed research is to systematically study the chemical and physical properties of both MOCC and MECC membranes for selective electrochemical CO2 separations. First, a combined "co-precipitation" and "sacrificial template" technique has been demonstrated to produce a highly efficient porous ceramic matrix containing a vast number of three-dimensional intra- and interconnected pathways as revealed by 3D XCT for fast-ion transport. The performance of thus synthesized MOCC membrane is remarkable, showing a permeation CO2 flux density from a simulated fuel gas stream two orders of magnitude higher than ceramic-carbonate systems fabricated with other methods. In addition, the CO2 flux density measured was found to increase with the partial pressure of hydrogen in the feedstock, further verifying the CO2 transport mechanism understood. Second, a surface-modified dense silver-MC dual-phase MECC membrane for CO2 separation from flue gas has been also demonstrated with improved stability. Two pore formers were investigated to make the porous silver matrix: microcrystalline methylcellulose and carbon black. The surface modifier is Al2O3, which was prepared by coating Al2O3 colloidal onto the exposed surfaces of a porous silver matrix. The results show that the use of 5% Al2O3 colloidal gives MECC the best flux density and stability compared to the unmodified sample. Approximately 90% of the original flux values can still be maintained after 130-hour testing for the modified membrane with microcrystalline methylcellulose pore former, whereas only one-third of the original flux values can be retained even after 60 hours for the unmodified membrane. For the surface modified MECC membrane with carbon black pore former, the CO2 flux was found to increase with time for the first 160 hours by 200%, followed by decreasing for the next 90 hours. At the 250-hour marker, the flux is still 160% of the original value. Overall, silver-molten carbonate MECC with carbon black as a pore former and Al2O3 as a surface modifier demonstrates great potential to separate CO2 from flue gas.

With the stabilized MECC membranes, we also demonstrate that the CO2 flux follow a linear relationship with reciprocal thickness in a thickness greater than 0.84 mm, suggesting a bulk diffusion controlling mechanism. Below 0.84 mm, the flux remains flat, suggesting that the rate-limiting step has shifted to the surface exchange kinetics of CO2 and O2. We also found that the CO2 flux is proportional to the linear chemical gradient of CO2 and O2, implying that the conductivity of CO32- is dependent of PCO2 and PO2 with a unity reaction order.

Establishing CO2 transport models and developing flux theory for MOCC and MECC membranes is another task of this thesis project. Multifunctional CO2 transport models encompassing 3PBs and 2PBs pathways are proposed for the first time. The CO2 flux equations suitable for dual-phase mixed conductors are developed from classical flux theory and verified by experimental results.

Finally, we have discovered for the first time the existence of pyrocarbonate C2O52- species on the surface of a eutectic Li2CO3 and Na2CO3 melt subject to CO2 atmosphere through a combined "DFT" and "Raman Spectroscopy" methodology. This discovery lays the foundation for the CO2 transport models established.