The use of palladium and strontium in LaCrO3 perovskites for solid oxide fuel cell anodes is known to lead to improved performance; in part due to the formation of palladium at the surface under reducing conditions that can act as a catalyst, with regeneration of the catalyst possible by cycling between oxidizing and reducing conditions. Whether this cycling involves dissolution and exsolution of the palladium into the perovskite is unclear. We describe a detailed analysis of the perovskites La0.8Sr0.2Cr1 - xPdxO3 (-) (delta) (LSCrPd, nominal stoichiometry x = 0.1 and 0.2) under reducing and oxidizing conditions. A LSCrPd perovskite was found to be the main phase, further confirmed by transmission electron microscopy. Secondary phases including metallic Pd, PdO, and La4PdO7, as well as SrCrO4, were also present. Some phases, such as PdO, SrCrO4, and La4PdO7, were no longer present following reduction while other phases such as metallic Pd and La2O3, were found in increasing amounts. When used as active solid oxide fuel cell anode layers both with and without Gd0.1Ce0.9O2 (-) (beta) (GDC) in La0.9Sr0.1Ga0.8Mg0.2O3 (-) (epsilon)/La0.4Ce0.6O2 electrolyte-supported cells, SOFCs with anodes containing GDC and higher amounts of Pd demonstrated higher maximum power densities and lower anode polarization resistances compared to cells with GDC-free, lower Pd content anodes. While the Pd is important to improve the anode performance, the results indicate that cycling does not lead to simple dissolution/exsolution into the perovsldte, instead many other phases are present while a limited amount of Pd was actually observed in the perovskite. (C) 2016 Elsevier B.V. All rights reserved.