Preparation of Mesoporous La0.8Sr0.2MnO3 Infiltrated Coatings in Porous SOFC Cathodes Using Evaporation-Induced Self-Assembly Methods
Robin Chao, John R. Kitchin, Kirk Gerdes, Edward M. Sabolsky and Paul A. Salvador, “Preparation of Mesoporous La0.8Sr0.2MnO3 Infiltrated Coatings in Porous SOFC Cathodes Using Evaporation-Induced Self-Assembly Methods”, ECS Transactions 35(1), 2387-2399 (2011). https://doi.org/10.1149/1.3570235
This summary was written by a language model from the paper and has not been checked line by line. Rely on the published paper for details.
Question
Cathode polarization limits the efficiency and power density of solid oxide fuel cells (SOFCs). Infiltrating an oxygen reduction catalyst into a pre-fired porous cathode by wet impregnation is known to help, and better performance has been correlated with more surface area and longer triple phase boundaries. Conventional wet-chemical infiltrates, however, have specific surface areas of only about 1 to 20 m2/g. Evaporation-induced self-assembly (EISA), in which a surfactant template directs the formation of mesopores during solvent evaporation, had already produced free-standing (La,Sr)MnO3 powders of about 50 m2/g. The paper asks whether an EISA solution, used without modification, can infiltrate porous cathode backbones, distribute catalyst through their thickness, and still form mesoporous coatings when the self-assembly is confined inside micron-sized pores (Chao et al. 2011).
Approach
The EISA solution contained the cationic surfactant CTAB (0.17 M) in a 1:1 water/ethanol mixture, with La, Sr and Mn nitrates in a La:Sr:Mn ratio of 0.8:0.2:1 at 2 M total metal, acidified with nitric acid to pH of about 1. Three backbones were infiltrated: pressed and fired porous 8 mol% YSZ pellets (about 0.75 mm thick, about 30% porosity), tape-cast La0.8Sr0.2MnO3 (LSM) cathodes on dense samaria-doped ceria electrolytes (about 27% porosity), and commercial anode-supported button cells with La0.6Sr0.4Co0.2Fe0.8O3 (LSCF) cathodes, including an LSCF/SDC functional layer next to the YSZ electrolyte (about 35% porosity). In each case 10 uL of solution was dropped on the cathode surface, wicked in by capillarity in about 15 s, evaporated for 8 h at about 60% humidity, and calcined at 800 C for 2 h, with the cycle repeated to raise loading. Cross sections were examined by SEM with EDS mapping, and a focused ion beam lamella from the LSCF cathode was imaged by HAADF STEM with EDS (Chao et al. 2011).
Key findings
A single infiltration put LSM nanoparticles throughout the thickness of all three backbones, from the free surface to the electrolyte, with no pressure or vacuum assistance. Loadings after one step were about 0.6 wt% (YSZ), 0.5 wt% (LSM) and 1.2 wt% (LSCF). Coverage was patchy at the pore scale, with roughly 20% of pores in YSZ coated, which the authors attribute to precursors remaining mobile and concentrating in the last pores to dry; but the pattern looked the same at every depth. EDS confirmed the target La0.8Sr0.2MnO3 composition, and the coating morphology did not depend on the backbone. The authors credit the surfactant with also acting as a wetting agent, so one non-optimized solution worked on all three materials (Chao et al. 2011).
Five infiltration and calcination cycles on the commercial LSCF cell gave about 6 wt% LSM. Mn and Fe maps showed Mn concentrated within about 10 to 20 um of the electrolyte interface (the functional layer) and depleted near large open pores, unlike most infiltration methods, which give flat profiles. Coatings in the functional layer were thin films on pore walls that did not visibly clog submicron pores. HAADF STEM revealed an interconnected framework of 10 to 20 nm solid regions alternating with pores, with a mean pore radius of about 17 +/- 8 nm (over 75 pores) and walls of about 25 nm, compared with 100 to 200 nm features reported for LSM infiltrates made by standard impregnation (Chao et al. 2011).
The authors propose that CTAB micelles form on the pore walls once evaporation pushes the surfactant past its critical micelle concentration, and that the templated precursor film becomes mesoporous when the surfactant burns off during calcination. They note that their free-standing CTAB-templated LSM lost much surface area after 100 h at 800 C whereas P123-templated material was more stable, so the durability of supported coatings, especially under electrochemical load, remains to be tested. No electrochemical performance data are reported; comparisons with conventional infiltrated cathodes were described as in progress (Chao et al. 2011).
The extracted text has some internal inconsistencies: the LSM cathode is given as about 50 um thick in the methods but 70 um in the results, the LSCF cathode as 70 or 75 um, and the mesopores are attributed to Figure 4 in one place although the TEM evidence is in Figure 5.
Significance
This is the first demonstration the authors know of that surfactant-templated EISA solutions can serve as infiltration vehicles, depositing mesoporous electrocatalyst coatings inside SOFC cathodes in a single step without tailoring the solution to the backbone. The finer walls and internal mesoporosity promise more catalyst surface area and triple phase boundary length per unit mass, and the preferential deposition near the electrolyte could put catalyst where it matters most while leaving the current collecting layer open for gas transport. It extends the group’s work on free-standing mesoporous LSM powders to real cathode microstructures (Chao et al. 2011).