Predictive Sensitization Modeling for AA5xxx Aluminum Alloys Including Non-Isothermal Cases

Sensitization of marine grade AA5xxx aluminum alloys results from precipitation of the intermetallic β phase along grain boundaries and is typically assessed by the destructive ASTM G67 procedure. Recently it has been shown that a Johnson-Mehl-Avarami-Kolmogorov model of grain boundary precipitation kinetics can be used to predict sensitization under isothermal aging conditions. This paper addresses the viability of expanding the isothermal model to cases with thermal history functions, utilizing parameters determined from controlled isothermal cases. Some fundamental assumptions required for non-isothermal expansion are examined using the special case of cyclic thermal treatments, suggesting the model may be applied to more complex thermal histories, especially in the form of environmental daily cycles. In this extended investigation, it is demonstrated that the grain boundary precipitation model predicts the kinetic behavior of sensitization with a remarkable 0.965, although experimental deviations in microstructure and application of the ASTM Standard G67 contribute to a relatively large variability in the raw sensitization measurements. The model is then compared to one contemporary case study, confirming a direct correlation with β-phase grain boundary coverage (R2 = 0.985) that substantiates certain fundamental assumptions.