Exploring structural transitions at grain boundaries in Nb using a generalized embedded atom interatomic potential

The advancement in experimental techniques, like the atom probe tomography and high resolution electron microscopy, is fueling interest in studying structural transformations of grain boundaries in metal and alloys to uncover correlations between mechanical properties and solute or impurity segregation to grain boundaries. Atomistic modeling is an important tool that can pinpoint the intricate dynamics of grain boundary phase transitions, but the lack of accurate interatomic potentials needed to simulate the complex dynamics of grain boundary structural transitions and identify different metastable phases has been the bottleneck. To this end, we use niobium as a model body centered cubic (BCC) metal and develop an interatomic potential to study grain boundary phase transitions. The potential for Nb is based on a generalization of the embedded atomic method potential and has sufficient flexibility to learn complex energy landscapes using a small set of training structures. We systematically test and validate the using data from ab initio density functional theory calculations and experiments. Using this potential, we calculate energies of multiple symmetric-tilt grain boundaries spanning a wide range of misorientation angles. Further, we explore different metastable structures of the ⠆27(552) [110] grain boundary and use molecular dynamic simulations to study the coexistence of metastable phases and grain boundary transitions at finite temperature.

Automated summary

The study of grain boundary structural transformations and their correlation with mechanical properties is important for the development of new materials. However, the lack of accurate interatomic potentials has been a bottleneck in simulating such transformations. This study focuses on niobium as a model material and develops an interatomic potential to investigate grain boundary phase transitions, validating the potential using ab initio calculations and experiments.