Uranium-niobium alloys exhibit complex crystal phases and unique mechanical behaviors under various thermodynamic states and external loads. However, due to the lack of accurate interatomic potentials, the atomic-scale phase behaviors and dynamical processes in this important alloy are still unclear. In recent years, the development of machine-learning-based force fields has provided a systematic way to generate accurate interatomic potentials on large and complex first-principle-based datasets. However, this crucial nuclear material has received limited attention from researchers in the field of machine-learning potentials.

In this work, based on our previous researches on the neural-network potential training and evaluation framework, which we called NNAP (neural-network atomic potential), a new neural network potential is constructed for the uranium-niobium alloy system. A combination of random structure search and active learning algorithms is utilized to enhance coverage of the chemical and structural space of the alloy system. Testing of the generated potential demonstrates high generalization performance and accuracy. On the testing set, the mean absolute error of the energy and the force are 5.6 meV/atom and 0.095 eV/Å, respectively. Further calculation results of crystal structure parameters, equation of state, and phonon dispersions coincide well with the results from the first-principle or experimental references.

The atomic-scale evolution of the spinodal decomposition process in the U-Nb alloys is investigated based on the newly trained potential. It is shown that the atom-swapping hybrid Monte Carlo can be a powerful tool to understand the thermodynamic evolution of the systems. By using the atom-swapping hybrid Monte Carlo method, the decrease of potential energy due to phase segregation is observed within 5000 steps, while no significant energy reduction is found after 3-ns MD simulation. Finally, the stress-strain curves under shear load for different initial states are obtained. It is found that the Nb precipitation generates strengthened phases in the alloy and the deformation behavior of U-Nb alloys is significantly changed, where a disorder shear band emerges in the deformation path of the \mathrm\gamma -phase alloys. Our work lays a foundation for understanding the mechanical processes in this important alloy system.