Precipitate phase selection and grain boundary morphology in Cu-Ni-Si-Mn alloys: A machine-learning interatomic potential study

Alloys inevitably contain interphase boundaries, whose energetics govern nucleation processes and precipitate morphology. In Cu-Ni-Si alloys, Mn addition markedly changes grain boundary (GB) precipitation behavior. While GB precipitation of stable Ni 2 Si in Mn-free alloys is associated with degraded mechanical properties, Mn addition instead promotes film-shaped Mn 6 Ni 16 Si 7 (G-phase) precipitation, which is correlated to improved mechanical properties. However, the atomic origin of the contrasting GB phase selection and morphology remains unclear. Here we perform machine-learning interatomic potential (MLIP) calculations to investigate the effect of interphase-boundary atomic structure on GB precipitates in Cu-Ni-Si alloys with and without Mn. The MLIP calculations reliably reproduce DFT-level energetics for interfacial bonding and microstructural configurations, and further predict that Mn addition favors GB precipitation of Mn 6 Ni 16 Si 7 rather than Ni 2 Si. Experimentally, Mn-free alloys are observed to exhibit irregularly shaped Ni 2 Si precipitates with open-boundary-like Cu/Ni 2 Si interfaces, whereas Mn-added alloys exhibit film-like G-phase at GBs. Large-scale atomistic interface calculations reveal that the coherent interface structure between Cu and Ni 2 Si favors the formation of plate-like strained Ni 2 Si precipitates in the matrix. Upon coarsening and stress release, an out-of-phase coherent-like interface can form at GBs, generating a local repulsive region that gives rise to surface-like open-boundary structures and explains the irregular morphology of GB stable Ni 2 Si precipitates. In contrast, Cu/Mn 6 Ni 16 Si 7 interfaces remain predominantly incoherent with moderate boundary energies and no pronounced repulsive regime, thereby stabilizing continuous interfacial contact and explaining film-shaped GB precipitation.