Testing thermal gradient driving force for grain boundary migration using molecular dynamics simulations

Strong thermal gradients in low-thermal-conductivity ceramics may drive extended defects, such as grain boundaries and voids, to migrate in preferential directions. In this work, molecular dynamics simulations are conducted to study thermal-gradient-driven grain boundary migration and to verify a previously proposed thermal gradient driving force equation, using uranium dioxide as a model system. It is found that a thermal gradient drives grain boundaries to migrate up the gradient, and the migration velocity increases under a constant gradient owing to the increase in mobility with temperature. Different grain boundaries migrate at very different rates owing to their different intrinsic mobilities. The extracted mobilities from the thermal-gradient-driven simulations are compared with those calculated from two other well-established methods, and good agreement between the three different methods is found, demonstrating that the theoretical equation of the thermal gradient driving force is valid, although a correction of one input parameter should be made. The discrepancy in the grain boundary mobilities between modeling and experiments is also discussed.