Limited understanding of damage nucleation, particularly in cases where nucleation probability is spatially variable and evolves with deformation, is a critical source of uncertainty in existing models of material failure. Most current models link damage nucleation to microstructure at best implicitly, with stochastic treatments that can be tuned to reproduce known failure behavior but give little insight into the underlying process. Although such models are computationally tractable, they are inadequate to explain or predict variability in performance across different material samples and structural designs, principally because they assume a static, spatially uniform field of equally potent nucleation sites. In reality, material heterogeneity promotes highly variable propensity for damage nucleation, often due to the presence of secondary phases or localized deformation. In addition, the dominant mechanisms of damage nucleation can vary with triaxiality and Lode angle, complicating the task of formulating a universally applicable model. Here, we propose a damage nucleation model for aluminum alloys that directly links void nucleation to micromechanical effects from second phase particles. Simulation results indicate that damage nucleation is sensitive to certain features of particle morphology, and these are included as material parameters in the model. The resulting formulation captures the influence of particles on local damage fields across a range of triaxialities and Lode angles. SNL is managed and operated by NTESS under DOE NNSA contract DE-NA0003525.