Modeling of morphological evolution of columnar dendritic grains in the molten pool of gas tungsten arc welding

A macro–micro coupled model for epitaxial nucleation and the subsequent competitive dendrite growth was developed to study the morphological evolution of both dendrite and grain structures in molten pool of the gas tungsten arc welding (GTAW) for Fe–C alloy. The simulation of heat and mass transfer in molten pool was conducted by the three-dimensional finite element (FE) model to obtain the transient solidification conditions. The process of epitaxial nucleation and the competitive dendrite growth was simulated by a two-dimensional cellular automata (CA) model. The size and random preferential orientations of substrate grains were considered in this model. The transient thermal conditions used in the CA model were obtained from the results of FE model through the interpolation method. In addition, the effects of the substrate grain size and the welding speed on the morphologies of both dendrite and grain structures were investigated. The simulated results indicate that dendrites with the preferential orientations parallel to the direction of the highest temperature gradient are more competitive during the competitive dendrite growth, and the morphology of resulting columnar grains is determined by the competition between different dendritic arrays. Under the same welding conditions, with the increase of substrate grain size, the average width of resulting columnar grains becomes larger, and the characteristics of dendrite structure within the columnar grains do not change obviously. Without considering the new nucleation in the melt, with the increase of welding speed, the dendrite structure in weld seam becomes much finer, and the average columnar grain width within the calculation domain of the CA model does not change obviously. The trend of the simulated results of dendrite arm spacing under various welding conditions are consistent with the analytical and experimental data.