Effects of La on the clustering process of B2-NiAl nanoparticles in maraging steels: Atomic-scale characterization and molecular dynamics simulation using deep learning potential

Maraging steels strengthened by coherent B2-NiAl precipitates exhibit an exceptional combination of strength and toughness, and trace amounts of rare earth (RE) elements can further enhance their mechanical performance through molten steel purification and microstructure tuning, yet the atomic-scale mechanisms governing RE effects on the NiAl precipitation remain unclear. Here, we unveil how La accelerates NiAl clustering and ordering in Fe-Ni-Al system through a synergistic combination of atomic-scale characterization, and a novel deep learning potential enabling multi-component atomic simulations. Atom probe tomography (APT) and high-angle annular dark field-scanning transmission electron microscopy (HAADF-STEM) reveal La enhances NiAl cluster density and growth kinetics, yielding earlier peak hardness during aging. Molecular dynamics simulations demonstrate La increases thermodynamic driving forces for precipitation and promotes Ni-Al pair ordering through strong Ni-La atomic interactions, as validated by first-principles binding energy calculations. This work establishes a paradigm integrating machine-learning enhanced atomistic simulations with advanced characterization to decode RE effects in complex alloys, and offers transformative strategies for high-performance alloy development targeting extreme service environments.