Numerical simulation on macrosegregation of ingots with dynamic shrinkage porosity and dendrite microstructure

Macrosegregation and shrinkage porosity are common and interrelated defects that occur during ingot solidification. Existing models typically simulate these phenomena separately rather than in a coupled manner. To address this limitation, this study employed an improved cellular automaton-lattice Boltzmann Method (CA-LBM) model to investigate the mechanisms behind macrosegregation and shrinkage effects during solidification process. Key improvements to the model include the optimization of the migration model in the LBM framework for solid-liquid-gas three-phase mixing to ensure the strict conservation of momentum, mass, and heat, and the incorporation of a shrinkage field during dendrite growth to dynamically track shrinkage evolution. Firstly, the stability and accuracy of the CA-LBM model were validated through simulations of dendrite growth involving coupled multiphysical fields. Subsequently, to investigate the formation mechanisms of macrosegregation and shrinkage porosity defects in both layered and mushy-zone solidification processes, solidification simulations were performed separately for an Fe-0.34wt.%C ingot (layered solidification) and an Al-4.7wt.%Cu alloy ingot (mushy-zone solidification) using the CA-LBM model. For the Fe-0.34wt.%C ingot solidification simulation, the results reveal defects such as axis porosity band, surface sink, hot-top segregation, negative base segregation, A-segregation, and V-segregation. A comparison with simulations that neglect shrinkage effects shows that shrinkage significantly alters both the distribution and morphology of macrosegregation. For the Al-4.7wt.%Cu ingot, the results indicate notable inverse segregation, surface sink, and widespread interdendritic shrinkage. By comparison with scenarios that do not consider shrinkage, it is evident that shrinkage markedly intensifies the degree of inverse segregation.