{"@context":"https://schema.org","@type":"ScholarlyArticle","headline":"Development of Robust and Durable Fuel Electrodes Using Composite La\n _0.20\n Sr\n _0.25\n Ca\n _0.45\n TiO\n ₃\n :3YSZ Supports","author":[{"@type":"Person","name":"Robert Price"},{"@type":"Person","name":"Eleanor Gladders"},{"@type":"Person","name":"John Irvine"}],"datePublished":"2025","isPartOf":{"@type":"Periodical","name":"ECS Meeting Abstracts"},"identifier":"https://doi.org/10.1149/ma2025-03130mtgabs","description":"\n Introduction\n \n \n La\n 0.20\n Sr\n 0.25\n Ca\n 0.45\n TiO\n 3\n (LSCT\n A-\n ) is perovskite-structured material that has been successfully employed as redox stable fuel-electrode backbone in electrolyte-supported Solid Oxide Fuel Cells (SOFCs) for more than a decade, in collaboration with industrial partners.\n 1–4\n This perovskite material previously exhibited an effective conductivity of 17 S cm\n -1\n at 850 °C and\n \n p(O\n 2\n )\n \n = 10\n -19\n atm, in thick-film DC conductivity testing,\n 5,6\n and has demonstrated comparable degradation rates (3.3 mV/kh\n -1\n )\n 3\n to state-of-the-art (SoA) SOFCs, containing Ni/CGO fuel electrodes, at stack scales when impregnated with Rh and CGO electrocatalysts.\n 3,4\n \n \n Research into incorporation of LSCT\n A-\n into fuel electrode-support microstructures of SOFCs has also been successfully performed at small scales (<1 cm\n 2\n active area). In particular, when co-impregnating the fuel electrode with 6 weight % (wt. %) CeO\n 2\n and 4 wt. % Ni\n 7\n or singly impregnating with 3 wt. % NiFe alloy.\n 8\n In addition, further optimisation of the fuel electrode support microstructures and processing methods required to produce support-electrolyte bilayers was recently undertaken by the authors.\n 9\n Although the aforementioned studies show the potential of the LSCT\n A-\n material to negate degradation associated with agglomeration and migration of Ni in SoA Ni-stabilised zirconia (Ni-SZ) supports, scaling of SOFCs above the 1-2 cm\n 2\n active area has remained challenging, due to the reduced mechanical strength of this material in comparison to that of the Ni-SZ composites.\n 1,10\n \n \n In this research, the authors present data pertaining to the ceramic processing of composite fuel electrode-supports comprising LSCT\n A-\n and 3 molar % yttria-stabilised zirconia (3YSZ) in order to increase the robustness of such supports. Particle size and dilatometric analysis of a variety of LSCT\n A-\n and YSZ materials will be provided and examined to identify the best matched materials for possible co-sintering of fuel electrode-support, fuel electrode-functional and electrolyte tri-layers. Additionally, microstructural evaluation and DC conductivity testing of composite supports (under reducing conditions) will be outlined.\n \n \n \n Experimental\n \n \n Table I summarises the materials used in this research, including composition, supplier and any additional treatments the powders underwent. For the LSCT\n A-\n and 8YSZ powders containing 1.3 wt. % NiO and 1.0 wt. % Co\n 3\n O\n 4\n , respectively, the required mass of Ni or Co nitrate hexahydrate was dissolved in deionized water and mixed with the ceramic powder under magnetic stirring, before the nitrates were decomposed in a muffle furnace. Calcined powders were deagglomerated in ethanol for 2 hours using a Dispermat LC55 high shear mixer, equipped with a nylon milling impeller and 2 mm zirconia milling media. The solvent was allowed to evaporate post-milling.\n \n \n The resultant ceramic powders and organic pore formers were subjected to triplicate particle size analysis (PSA) with an Anton Paar 1190L particle size analyser, using ethanol as a solvent. Ceramic powders were also pressed into 10 mm diameter pellets (at a mass of 1 tonne for 60 seconds) before undergoing dilatometric analysis using a Netzsch DIL 402 C dilatometer. The sintering curves of each material were measured whilst heating to 1350 °C for 6 hours, at a ramp rate of 3 °C min\n -1\n , in a 50 mL min\n -1\n flow of compressed air.\n \n \n Fuel electrode-supports were produced using an aqueous slurry formulation through a roller ball milling procedure. Ceramic powder(s), along with poly(methyl methacrylate) (MX180TA, 1.8 μm, Soken) and rice starch (Sigma-Aldrich), were added to a milling vessel with 10 mm diameter zirconia milling media, deionised water, a Hypermer KD6 dispersant (Croda) and a DF002 defoamer (Polymer Innovations). Dispersion milling was performed at 160 rpm for 24 hours. Next, poly(ethylene glycol) (PEG 300, average M\n n\n = 300, Aldrich), glycerol (99.5 %, ACS, Alfa Aesar), poly(vinyl alcohol) (PVA, 87–89 % hydrolysed, M\n w\n = 88,000–97,000, Alfa Aesar) and extra DF002 defoamer were added for mixing at 100 rpm for 24 hours.\n \n \n Slurries were subsequently poured into the reservoir of a doctor blade until the underlying Mylar® carrier film was completely coated. The blade gap was set to 650 µm and the slurries were tape cast onto the carrier film using a Richard E. Mistler TTC1200 Benchtop Caster. A casting rate of 0.48 cm s\n -1\n was used and the tapes were allowed to dry in air under ambient conditions (17 °C to 20 °C). The dried, green fuel electrode-support tapes were cut into 20 mm x 120 mm strips, 6 of which were hot laminated together, before being cut into 15 mm x 15 mm squares for sintering up to 1350 °C for 6 hours in a muffle furnace.\n \n \n After sintering, fuel electrode-support substrates (~10 mm x 10 mm) had Au paste (Sun Chemicals) and Au wire electrodes applied to the surface in a linear array, before being fired up to 750 °C in a muffle furnace, to ensure good contact with the surface. Fuel electrode-supports with varying ratios of LSCTNi:3YSZ were subjected to conductivity testing in a single atmosphere tube furnace, with an inline pO\n 2\n sensor. The Au wires of the samples were connected to a Keithley 2401 SourceMeter via the wiring of the DC conductivity rig. Measurements were carried out in an atmosphere of 5% H\n 2\n :95% N\n 2\n up to 900 °C, with a reduction period of 18 hours at this temperature, before data collection during cooling.\n \n Scanning Electron Microscopy (SEM) was carried out using a JEOL JSM-IT800 SEM. Cross sections of the tested fuel electrode-support microstructures were embedded in epoxy resin, before being cut and diamond polished after curing. Subsequently, porosity and thickness measurements were performed on obtained SE micrographs using the threshold and length scale calibration functions of the ImageJ software.\n \n Results and Discussion\n \n \n Figure Ia displays the particle size distributions of the studied materials and a summary of the average d\n 10\n , d\n 50\n and d\n 90\n values determined from these distributions is presented in Table I. These data show that the Treibacher LSCT\n A-\n exhibits a monomodal distribution with a d\n 50\n of 1.65 μm, agreeing with our previous characterization of this ceramic powder.\n 5,11\n After powder infiltration with Ni(NO\n 3\n )\n 2\n ·6H\n 2\n O, calcination and deagglomeration, the LSCTNi powder exhibits a very similar distribution and d\n 50\n value of 1.69 μm, implying that only surface modification of the perovskite material occurred and did not lead to severe agglomeration of the powder. The PSD of the PMMA pore former (d\n 50\n = 1.97 μm) chosen in this work, therefore, matches the particle size of the LSCT/LSCTNi very well, which is important for the mechanical rigidity of the support after sintering. Similarly, the 8YSZ and 8YSZCo powders show similar particle size distributions before and after Co(NO\n 3\n )\n 2\n ·6H\n 2\n O infiltration, calcination and deagglomeration, though in this case, a bimodal distribution is observed, with one centred around 0.5 μm and another between 1 and 2 μm. Interestingly, the particle size distribution of the 3YSZ powder obtained from Tosoh shows comparability to the 8YSZ (from DKKK) distribution, whereas the 3YSZ from DKKK shows a monomodal distribution centred at 0.6 μm, indicating that the 8YSZ powder may well be better matched, in terms of sintering and shrinkage, to the 3YSZ powder from Tosoh. This is corroborated by the dilatometric data presented in figure Ib. Here, in fact, both of the 3YSZ powders and the 8YSZ powder exhibit almost analogous sintering profiles, though the addition of 1 wt. % Co\n 3\n O\n 4\n to a sample of the 8YSZ powder shifts the sintering onset temperature from ~1000 °C to ~900 °C with only minimal gains in shrinkage (i.e. 2 % increase in densification). In comparison, addition of 1.3 wt. % NiO increases the shrinkage of the LSCT\n A-\n material by more than 10 %, forcing the sintering onset and final shrinkage of LSCTNi closer to those of the 3YSZ and 8YSZ powders to provide a good match in shrinkage properties required for co-sintering of anode support, anode functional and electrolyte layers.\n \n \n In order to ensure good mechanical strength of the support and further match shrinkages of the aforementioned layers, fuel-electrode supports containing 40:60, 50:50 and 60:40 volume percent LSCTNi:T3YSZ were prepared by aqueous slurry formulation and tape casting. Figures IIa, b and c show the microstructures of the supports, respectively. DC conductivity data, collected during cooling from 900 °C, are displayed in figure IId. At 850 °C (average\n \n p(O\n 2\n )\n \n ~10\n -20\n atm), the 60:40 ratio shows highest conductivity (0.65 S cm\n −1\n ), whilst 50:50 and 40:60 ratios exhibit conductivities of 0.48 S cm\n -1\n and 0.21 S cm\n -1\n , respectively. Although superior mechanical stability of the supports was achieved by substituting a proportion of LSCT\n A-\n with 3YSZ, the electrical conductivity drops by 2 orders of magnitude in comparison to ‘pure’ LSCT\n A-\n layers tested under the same conditions on electrolyte supports in our previous research. It should be noted, however, that during slurry formulation and tape casting, the addition of the rice starch and PMMA pore former reduced the adhesive properties of the tapes to both the Mylar® carrier film and to the tapes themselves during hot lamination. Therefore, sub-optimal connectivity between the layers may have resulted in such a large decrease in conductivity, compared to the literature. Our current focus is to evaluate the effect of this reduction in conductivity in full SOFC tests using these supports.\n \n \n \n \n Conclusions\n \n \n Characterisation of a variety of La\n 0.20\n Sr\n 0.25\n Ca\n 0.45\n TiO\n 3\n (LSCT\n A-\n ) and yttria-stabilised zirconia (YSZ) powders via particle size analysis and dilatometry indicated that the shrinkages of LSCT\n A-\n with 1.3 wt. % NiO (LSCTNi), Tosoh 3YSZ and DKKK 8YSZ were well-matched in order to facilitate potential co-sintering of fuel electrode support, fuel electrode functional and electrolyte tri-layers. Manufacture of composite fuel electrode supports, with LSCTNi:T3YSZ ratios of 40:60, 50:50 and 60:40 volume %, was performed through slurry formulation, tape casting and sintering to 1350 °C for 6 hours and the microstructures were confirmed to be well-processed through scanning electron microscopic analysis (though improvements in slurry formulation could improve the connectivity of layers during tape lamination). DC conductivity testing of these samples illustrated that ‘effective’ conductivities of 0.21, 0.48 and 0.65 S cm\n -1\n could be achieved in these composite substrates; lower than those exhibited by ‘pure’ LSCT\n A-\n -based microstructures. Therefore, the improvement in mechanical strength offered by addition of 3YSZ to LSCTNi supports must be balanced with the loss of electrical conductivity, which is of paramount importance in fuel electrode-supported SOFCs.\n \n \n Acknowledgements\n \n We would like to thank Dr Cristian D. Savaniu, Dr Aaron B. Naden and Dr Paul A. Connor for support during this research, as well as the EPSRC projects: EP/R023751/1 – “Electron Microscopy for the Characterisation and Manipulation of Advanced Functional Materials and their Interfaces at the Nanoscale” and EP/X025322/1 – “ENSIGN: ENergy System dIGital twiN” for funding.\n \n References\n \n \n 1. M. C. Verbraeken, B. R. Sudireddy, V. Vasechko, M. Cassidy, T. Ramos, J. Malzbender, P. Holtappels and J. T. S. Irvine,\n J. Eur. Ceram. Soc.\n , 2018,\n 38\n , 1663–1672.\n \n \n 2. P. Holtappels, J. Irvine and B. Iwanschitz,\n ECS Trans.\n , 2013,\n 57\n , 1175–1184.\n \n \n 3. R. Price, M. Cassidy, J. G. Grolig, G. Longo, U. Weissen, A. Mai and J. T. S. Irvine,\n Adv. Energy Mater.\n , 2021,\n 11\n , 2003951.\n \n \n 4. R. Price, H. Bausinger, G. Longo, U. Weissen, M. Cassidy, J. G. Grolig, A. Mai and J. T. S. Irvine,\n Fuel Cells\n , 2023,\n 23\n , 482–491.\n \n \n 5. R. Price, M. Cassidy, J. A. Schuler, A. Mai and J. T. S. Irvine,\n ECS Trans.\n , 2017,\n 78\n , 1385–1395.\n \n \n 6. R. Price, M. Cassidy, J. G. Grolig, A. Mai and J. T. S. Irvine,\n J. Electrochem. Soc.\n , 2019,\n 166\n , F343–F349.\n \n \n 7. L. Lu, C. Ni, M. Cassidy and J. T. S. Irvine,\n J. Mater. Chem. A\n 2016,\n 4\n , 11708–11718.\n \n \n 8. C. Ni, L. Lu, D. N. Miller, M. Cassidy and J. T. S. Irvine,\n J. Mater. Chem. A\n , 2018,\n 6\n , 5398–5406.\n \n \n 9. R. Price, C. D. Savaniu, M. Cassidy and J. T. S. Irvine,\n ECS Trans.\n , 2021,\n 103\n , 1625–1639.\n \n \n 10. V. Vasechko, M. Ziegner and J. Malzbender,\n Ceram. Int.\n , 2014,\n 40\n , 13179–13189.\n \n \n 11. R. Price, M. Cassidy, J. A. Schuler, A. Mai and J. T. S. Irvine,\n ECS Trans.\n , 2015,\n 68\n , 1499–1508.\n \n \n \n \n Figure 1\n ","url":"https://innscience.ca/es/community/literature/499833ff-ab2a-4c5c-a0ad-bec7a700263d"}