Modelling and experiments of metal interconnect degradation in solid oxide fuel cells

Oum, Melissa (2021). Modelling and experiments of metal interconnect degradation in solid oxide fuel cells. University of Birmingham. Ph.D.

Preview

Oum2021PhD.pdf
Text - Accepted Version
Available under License All rights reserved.
Download (4MB) | Preview

Abstract

Chromium is a key component of ferritic stainless steels developed for solid oxide fuel cell (SOFC) interconnect components. The properties of these steel alloys change over time due to the formation of a chromia scale and subsequent volatilisation of chromium gaseous species. As the operational temperature of SOFC systems is continuously reduced, the present thesis reports research aimed at investigating the behaviour of chromium in stainless steels, at 700°C¬¬ – from an experimental and modelling point of view.
The research consisted of three stages. The first stage involved a detailed study of the sinterability of pure chromium at 1150°C and the isothermal oxidation of the sintered pellets at 700°C in dry and humidified atmospheres. The aim was to optimise the sinterability of chromium and improve the understanding of pure chromium behaviour at a reduced operating temperature. The results showed that for both binder-free and binder containing chromium pellets, sintering time was the determining factor to achieve low porosity and high-density pellets. No volatile chromium was found in dry oxidation at this temperature and the study revealed that the duration of exposure and the chromium collection method used influenced the values of oxidation and evaporation rate constants obtained.
The second stage involved a study of the isothermal humidified oxidation at 700°C of commercial ferritic steels K41 and Crofer 22H. The study aimed primarily at deriving oxidation and volatilisation rate constants from the oxide growth for use in the modelling of the lifetime of the binary system FeCr. Crofer 22H, whilst equally growing a spinel oxide surface layer, showed less chromium evaporation than K41. Besides this, effects of sample preparation method were also investigated, which demonstrated that compared to other methods commonly used, a refined grinding was the most effective pre-treatment method in achieving a thinner oxide.
The final stage involved designing a one-dimensional thermodynamic-kinetic diffusion and oxidation model for the FeCr system. The developed model can predict the sequential layering oxide phases that are not always differentiable experimentally as well as quantify oxide growth thickness and depletion profiles of chromium in single-phase Fe-Cr binary alloys, using the Calphad approach and finite difference method. The thermodynamic modelling predicted the successive oxide phases formation at 700°C: (Fe,Cr)2O3, spinel (Fe,Cr)3O4 and halite (FeO) and highlighted the inconsequential effect of the oxygen partial pressure at values higher than 10-8 atm. The kinetic calculations applied to Fe17Cr and Fe22Cr inferred that in the long term, for the Fe17Cr alloy, it would take 26,500 hours of exposure to a cathodic SOFC atmosphere to reach the critical minimum chromium concentration of 11 wt.%, whereas the same concentration would be reached after 53,500 hours for the Fe22Cr alloy.

Type of Work:

Thesis (Doctorates > Ph.D.)

Award Type:

Doctorates > Ph.D.

Supervisor(s):