Large scale energy storage is a necessary component of any electricity grid operating primarily on intermittent renewable power sources. Recent experimental and techno-economic studies on reversible solid oxide cell (ReSOC) technology suggest that this technology performs well compared to other available technologies in terms of round-trip efficiency, stable storage time, and cost per kilowatt-hour stored.[S.H. Jensen, Energy Environ Sci 2015 ,8, 2471-2479] However, SOCs are known to degrade by delamination at the oxygen electrode under electrolysis conditions; this must be better understood in order to operate SOCs over their desired lifetimes.  Here we investigate the conditions that cause delamination and degradation by carrying out life tests of symmetric cells, observing electrochemical behavior by impedance spectroscopy and by post-test SEM analysis. Symmetric cells are subjected to switching current operation with a 12 hour period at current densities ranging from 0.5 to 2.0 A/cm² – similar to that expected in a ReSOC energy storage application. Cells are typically operated for 1000 hours, long enough to detect degradation rates as low as 1 %/kh.  Three types of symmetric cells are investigated: single phase (La_0.6Sr_0.4)(Co_0.2Fe_0.8)O₃ (LSCF) on Ce_0.9Gd_0.1O₂ (GDC), (La_0.8Sr_0.2)MnO₃ (LSM) on Zr_0.92Y_0.08O₂ (YSZ) and La₂NiO₄ wet infiltrated into (La_0.9Sr_0.8)(Ga_0.8Mg_0.2)O₃ (LNO infiltrated LSGM) on LSGM.  Degradation rates are determined for each current density and corresponding initial electrode overpotential. Degradation rates < 1%/kh are observed for LNO infiltrated LSGM on LSGM at 1.0 A/cm² (0.20 V), and below ~0.9 A/cm² (0.1 V) for LSCF on GDC.  The results are also compared with prior results from similar experiments with (La_0.8Sr_0.2)MnO₃ (LSM) on Zr_0.92Y_0.08O₂ (YSZ).  Changing the temperature by 50 °C from 800 to 750 °C for a previously stable current density (0.5 A/cm²) for LSM-YSZ results in a higher over potential (0.55 V) and fast degradation. Comparing the critical current densities and overpotentials where the degradation rates become measurable, and where structural damage can be observed in post-test SEM images, for this range of different electrode and electrolyte materials, suggests that degradation is controlled by overpotential rather than by current density.