Nickel cathodes have been found to become deactivated under long-term polarization in the H-2 evolution region during alkaline water electrolysis. The cause of deactivation was examined using steady state polarization and measurement of hydrogen permeation through nickel foil in 8 mol/l KOH at 70 degreesC and 100 mA/cm(2). The long-term (over 50 h) permeation behaviour was explained by formation and growth of a nickel hydride phase. The rise in hydrogen overpotential was ascribed to an increase of the hydrogen surface coverage on the newly formed hydride. The effect of an electrolyte additive (a vanadium salt) on the hydrogen overpotential and permeation rate was also investigated. Upon addition of dissolved V2O5, the permeation rate was found to increase quickly and then slowly decrease to a steady value close to that measured for hydride-free nickel. Meanwhile, the hydrogen overpotential was observed to recover back to nearly its initial value for fresh nickel. The exhibited behaviour was attributed to decomposition of the hydride phase after deposition of a vanadium-bearing compound. The prolonged contact between Ni and V was proposed as the main reason for hydride decomposition. The addition of more vanadium had no further result on the hydrogen overpotential.