Potentiostatic anodic polarization data show that a transition from active to passive state is observed for Ni-Cu and Ni-Cu-Zn alloys only when the 3d band of energy levels in the alloy is unfilled. At and above a critical alloy composition, electrons from Cu or Zn completely fill the d band and the passive state disappears. Previous data showed that the calculated critical composition comes close to the observed value assuming that alloyed Cu donates one electron and alloyed Zn two electrons per atom. Present measurements for the 45% Ni-Cu-A1 ternary alloys show that A1 contributes three electrons per atom. Secondary passivity, which is observed for pure nickel in a potential region noble to that for primary passivity, is similarly observed in the Cu-Ni-A1 alloys, but again only when the d band of energy levels is unfilled. Potential decay curves confirm that a passive film, either in the primary or secondary region, exists only on alloys with an unfilled d band. Decay potentials corresponding either to secondary or to primary passivity (Flade potentials) are not greatly sensitive to aluminum content nor to the absolute number of electron vacancies. However, alloyed aluminum affects the passive current density for the 45% Ni ternary alloys accounting for a minimum at 2.35% At.The primary and secondary passive films are considered to have an adsorbed structure consisting mainly of oxygen in mono-or multilayers which increase the overvoltage for anodic dissolution of the alloy. The film corresponding to secondary passivity is less stable and presumably thicker than that corresponding to primary passivity. Both films are considered to be intermediates in the formation of stoichiometric metal oxides.The present work on passivity in the Cu-Ni-A1 alloys continues a research program on the general mechanism of passivity in metals and alloys. Osterwald and Uhlig (1) confirmed previous reports (2, 3) that there is a critical composition for passivity in the binary Cu-Ni alloys; Stolica and Uhlig (4) found later that there also exists a critical composition m the ternary Cu-Ni-Zn alloys. In both cases the critical compositions for passivity were determined by measuring anodic potentiostatic polarization curves in 1N H2SO4 at 25~According to the electron configuration theory of passivity (3), a metal or alloy should be capable of becoming passive so long as it has unfilled d-orbitais or an unfilled d band which favors chemisorption of oxygen. It is suggested that the chemisorbed oxygen markedly decreases the exchange current density thereby increasing the overvoltage for metal dissolution in the passive region. When the d band is filled, as in the nontransition metals, the overvoltage for anodic dissolution remains small, and the metal or alloy tends to form stoichiometric oxides in preference to chemisorbed oxygen films. Oxide films, if they form, are not thought to be a major source of corrosion protection.From this it follows that the critical composition for passivity in Cu-Ni alloys should appear when the d electron v...