The development of deterministic models for predicting the accumulation of corrosion damage to high level nuclear waste (HLNW) canisters requires the acquisition of values for various model parameters. In the present work, we describe the extraction of values for various parameters in the point defect model (PDM) for the growth of passive films on Alloy-22, from electrochemical impedance data for this alloy measured in saturated NaCl brine (6.2 m, m = mol kg(-1)) at 80 degreesC, as described in Part 1. The barrier layer of the passive film on Alloy-22 in the passive region (V < 0.6 V-SHE) is postulated to be defective chromic oxide (Cr2+xO3-y) with n-type electronic character. The data suggest that the barrier layer is cation rich (x > y) with cation interstitials being the principal defect, but it is not possible to exclude oxygen vacancies, unequivocally, as being the principal defect. If the barrier layer is cation rich, the phase is estimated to have the composition Cr2.028O3 close to the metal\film interface, whereas if it is oxygen deficient, the stoichiometry is estimated to be Cr2O2.981 at the same location. The thickness of the barrier layer in the passive region is found to increase linearly with voltage while the passive current density is constant. In the transpassive range (V > 0.6 V-SHE), the passive film is found to be a p-type semi conductor, and the thickness of the barrier layer decreases, while the log of the passive current increases, with increasing applied potential. These observations are consistent with the predictions of the PDM for passive films on metals and alloys, including the transition from the passive to the transpassive states, which is predicted to coincide with a change in oxidation state of the chromium species being ejected from the barrier layer from Cr(III) to Cr(VI) and with the oxidative dissolution of the film. The principal defect in the barrier layer in the transpassive state appears to be the Cr(III) cation vacancy, which is generated by the oxidative ejection of cations from the barrier layer. Optimization of the PDM on the impedance data has yielded a set of parameter values for the passive state that can be used in deterministic models for predicting the accumulation of general corrosion damage to Alloy-22 in HLNW repositories. (C) 2004 Published by Elsevier B.V.