Prediction of grain structure after thermomechanical processing of U-10Mo alloy using sensitivity analysis and machine learning surrogate model.

Hot rolling and annealing are critical intermediate steps for controlling microstructures and thickness variations when fabricating uranium alloyed with 10% molybdenum (U-10Mo), which is highly relevant to worldwide nuclear non-proliferation efforts. This work proposes a machine-learning surrogate model combined with sensitivity analysis to identify and predict U-10Mo microstructure development during thermomechanical processing. Over 200 simulations were collected using physics-based microstructure models covering a wide range of thermomechanical processing routes and initial alloy grain features. Based on the sensitivity analysis, we determined that an increase in rolling reduction percentage at each processing pass has the strongest effect in reducing the grain size. Multi-pass rolling and annealing can significantly improve recrystallization regardless of the reduction percentage. With a volume fraction below 2%, uranium carbide particles were found to have marginal effects on the average grain size and distribution. The proposed stratified stacking ensemble surrogate predicts the U-10Mo grain size with a mean square error four times smaller than a standard single deep neural network. At the same time, with a significant speedup (1000×) compared to the physics-based model, the machine learning surrogate shows good potential for U-10Mo fabrication process optimization.

Extended Shear Deformation of the Immiscible Cu-Nb Alloy Resulting in Nanostructuring and Oxygen Ingress with Enhancement in Mechanical Properties.

Deformation processing of immiscible systems is observed to disrupt thermodynamic equilibrium, often resulting in nonequilibrium microstructures. The microstructural changes including nanostructuring, hierarchical distribution of phases, localized solute supersaturation, and oxygen ingress result from high-strain extended deformation, causing a significant change in mechanical properties. Because of the dynamic evolution of the material under large strain shear load, a detailed understanding of the transformation pathway has not been established. Additionally, the influence of these microstructural changes on mechanical properties is also not well characterized. Here, an immiscible Cu-4 at. % Nb alloy is subjected to a high-strain shear deformation (∼200); the deformation-induced changes in the morphology, crystal structure, and composition of Cu and Nb phases as a function of total strain are characterized using transmission electron microscopy and atom probe tomography. Furthermore, a multimodal experiment-guided computational approach is used to depict the initiation of deformation by an increase in misorientation boundaries by crystal plasticity-based grain misorientation modeling (strain ∼0.6). Then, co-deformation and nanolamination of Cu and Nb are envisaged by a finite element method-based computational fluid dynamic model with strain ranging from 10 to 200. Finally, the experimentally observed amorphization of the severely sheared supersaturated Cu-Nb-O phase was validated using the first principle-based simulation using density functional theory while highlighting the influence of oxygen ingress during deformation. Furthermore, the nanocrystalline microstructure shows a >2-fold increase in hardness and compressive yield strength of the alloy, elucidating the potential of deformation processing to obtain high-strength low-alloyed metals. Our approach presents a step-by-step evolution of a microstructure in an immiscible alloy undergoing severe shear deformation, which is broadly applicable to materials processing based on friction stir, extrusion, rolling, and surface shear deformation under wear and can be directly applied to understanding material behavior during these processes.