Date of Award

2019

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Graduate Group

Chemical and Biomolecular Engineering

First Advisor

Raymond J. Gorte

Second Advisor

John M. Vohs

Abstract

Solid Oxide Fuel Cells (SOFCs) are energy-generating devices operating at elevated temperatures. At their cathode sides, oxygen molecules reduce into oxygen ions and then transport to the anode side through a solid-state ceramic membrane (typically, Yttria-Stabilized Zirconia (YSZ)). At their anode side, the fuels react with the oxygen ions and then release electrons to the external circuit. Always, it is the cathode that has the slowest reaction kinetics, such it contributes the largest resistance in the entire device.

The state-of-the-art cathode materials are typically perovskite-phase ceramic materials, such as Strontium-doped Lanthanum Manganate (LSM), Strontium-doped Lanthanum Ferrite (LSF) and Strontium-doped Lanthanum Cobaltite (LSC). Conventionally, they are just printed to the well-sintered solid-state electrolyte surface and then sinter again to a medium temperature (~1373K) to build contact with the electrolyte. This brings multiple problems, such as solid-state reactions, bad mechanical properties, limited reaction sites, etc. To resolve these problems, infiltrated cathodes, which has nanoscale cathode materials distributed in a porous matrix of electrolyte, was developed at Penn. Cathodes made with this method have multiple advantages such as a significantly larger number of reaction sites, the avoidance of solid-state reactions, strong mechanical properties and the match of coefficient of thermal expansion.

However, although infiltrated cathodes exhibit significantly better performances than traditional printed cathodes, there is still room for improvement. An important concern about the infiltration method is manufacturability. A single infiltration step takes about 30 mins, and the infiltrated material per cycle is restricted by the volume of material that can be added. Normally, 30 wt% loading of the nanoparticles in the cathode matrix is necessary to achieve a good electronic conductivity, with only one thirds of the nanoparticles are there for the surface reaction kinetics. Therefore, to improve the manufacturability of infiltrated SOFC cathodes, one must develop an electronically conductive cathode at a low loading of the cathode nanoparticles. In our study, we have demonstrated a promising approach of using composite materials, such as LSF-YSZ and LSCrF-YSZ, instead of single-phase YSZ as the matrix material. This allows the cathode to have a better performance at low nanoparticles loading due to the incorporated electronic conductivity in the porous matrix backbone.

Another approach to improve the cathode performance is to increase the surface reaction kinetics of the infiltrated cathode materials. In this study, we used Atomic Layer Deposition (ALD), a thin-film growth technology capable of making atomic-level composition tuning, to reveal the desired surface composition of the perovskite-phase cathodes for the sluggish oxygen reducing process. It was discovered that for the ABO3 perovskites, an AO layer on top of BO2 layer configuration is desired for fast oxygen reducing kinetics and it is very likely due to the increase in surface oxygen vacancies. Computational study performed by our collaborator investigated the thermodynamic nature of this configuration and suggested its thermal instability, which was proved by further experimental evidence.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS