Aluminum alloys contain alloying elements such as copper that are added to enhance their properties, making them attractive for numerous engineering applications. In particular, aluminum alloys are an important structural engineering material in the aerospace industry and other applications where light-weight is a key design requirement. Presently, laser welding, a process of joining metals or thermoplastics using a laser beam, is extensively used in manufacturing industries. Other than improved mechanical and structural properties, laser welded alloys exhibit many intrinsic advantages like high flexibility and productive efficiency. Based on the literature, the mechanical properties and performance of laser welds highly depend on the solidified structures, both the grain and dendritic structures.
The solidification process involves a complex interaction between the microscale solid-liquid interface and the macroscale heat-mass transfer dynamics in the melt pool. Therefore, optimization of the laser welding process to improve the properties of the resulting microstructures requires knowledge of the dynamic solidification behavior. The introduction of computational simulations has made it possible to simulate the process during laser welding. However, most of the existing research on the dynamic behavior of the solidification process focuses on the evolution of mesoscale grain structures, the localized solidification behavior, and the initial solidification stage in the melt pool with little attention to the subsequent stages. Additionally, the knowledge on the solute redistribution and the morphology evolution of the sub-grain dendritic structure is missing.
Understanding the formation mechanisms of the solidified structure requires accurate modeling of the intricate solidification behavior of the whole weld, taking into account the effects of the heat-mass transfer behaviors. Equipped with this knowledge, researchers at Huazhong University of Science and Technology: Professor Ping Jiang, Graduate student Song Gao, Postdoctoral researcher Shaoning Geng, Graduate student Chu Han and Assistant professor Gaoyang Mi developed a multi-scale multi-physics model to invstigate the solidification behavior during laser welding of 5083 aluminum alloys. They aimed to simulate the entire weld pool to understand the underlying mechanism behind the dendritic structures’ formation. The work is currently published in the International Journal of Heat and Mass Transfer.
In their approach, the proposed model combined three different models: the macroscale-heat and mass transfer model, Gaussian nucleation distribution model and microscale phase-field model. A detailed solidification process covering the planar and cellular growth as well as columnar dendritic and equiaxed dendritic growth in the weld pool was analyzed through temperature distributions, constitutional undercooling fields, and solute concentrations. The model was validated by comparing the simulation and experimental results with a particular interest in the grain structure, weld profile and dendritic structure.
The authors accurately simulated dynamic solidification behavior for the entire weld pool. The Marangoni convection effect induced by surface tension variations was vital to the characterization of the laser molten pool. The three-dimensional realistic solidification process could be simplified into a two-dimensional process despite the negligible effects of the heat flow along the thickness direction. Moreover, the authors predicted the distribution of constitutional undercooling of the liquid before the solidification front. They noted the selective growth of both the equiaxed and columnar grains at different weld locations.
In summary, a simulation of the entire melt pool, using a multi-physics multi-scale model, during laser welding of an aluminum sheet was reported. The model combined different models, thereby allowing for accurate simulation of the dynamic solidification behavior of the welds. A good agreement between the simulation and experimental results was reported, an indication of the capability of the model to quantitatively predict weld microstructure. In a statement to Advances in Engineering, the authors explained their new model provided a comprehensive understanding of the laser welding solidification process and would help improve the properties and performance of laser-welded aluminum.
Jiang, P., Gao, S., Geng, S., Han, C., & Mi, G. (2020). Multi-physics multi-scale simulation of the solidification process in the molten pool during laser welding of aluminum alloys. International Journal of Heat and Mass Transfer, 161, 1-12.