Objective Due to the superior corrosion resistance of Ni and the good hot workability as well as low cost of 304SS, Ni and 304SS joints have been widely used in various industries, such as petrochemical, aerospace, and aviation. Laser welding has the advantages of high precision, high efficiency, and low residual stresses; therefore, it has been considered as a promising joining technology for dissimilar metals. The performance of a welded joint is affected by various factors. One of the most important factors is alloying element mixing within the weld pool (WP). However, processing parameters, such as laser power, scanning speed, and spot offset, can lead to a different element redistribution in the joint. Additionally, the physical mechanisms of element distribution affected by these processing parameters have not been fully investigated. Therefore, this study investigates the element-mixing process within the WP in the laser conduction welding of Ni and 304SS through numerical modeling. The effects of processing parameters on the element distribution in welded joints are investigated. The results show how processing parameters affect dissimilar-metal redistribution and how to obtain a more uniform element distribution with higher dilution by optimizing the process.
Methods The transient fluid-flow and element-mixing processes within the WP are difficult to observe directly through experiments. Numerical simulations can be used to predict the dynamic evolution of WP and dissimilar-metal redistribution during laser conduction welding. Therefore, this study develops a numerical model based on the Navier-Stokes equation, coupled with the temperature field, fluid-flow field, and concentration field to analyze the mixing process of the three main alloying elements (Fe, Ni, and Cr). Additionally, the flow characteristics inside the WP are investigated, which are closely correlated to the concentration dilution. The numerical model is validated by comparing the calculated WP profile and distribution of the three alloying element concentrations with the experimental results. Then, the influence of processing parameters on Fe element redistribution in welded joints is analyzed using orthogonal parameter design and range analysis. Additionally, the underlying physical mechanisms of parameters affecting the element distribution are explored based on the model.
Results and Discussions The order of magnitude of Peclet number in the transportation of the Fe element is estimated to be 10 4 in the laser welding of dissimilar metals, indicating that convection dominates the mass transfer process. The WP reaches a quasi-steady state for ~50 ms, and the fluid flow in the back section of the WP in the quasi-steady state facilitates the uniform distribution of elements along the z-axis (Fig. 6). Based on orthogonal simulation and range analysis, the range of each level of scanning speed is 9.45%; however, the range of spot offset and laser power is 9.17% and 1.11%, respectively. The scanning speed is negatively related with the average concentration of Fe, whereas the offset is positively correlated with it (Fig. 7). The model’s results show that scanning speed affects the dilution of the Fe element by changing the duration of WP (Table 5) and influences the mushy zone size of WP (Fig. 8). The offset affects the Fe redistribution by changing the longitudinal flow pattern of WP and the relative position of the cross-sectional branch flow to the joint interface (Fig. 9).
Conclusions This study establishes a three-dimensional numerical model coupled with the temperature, flow, and multicomponent concentration fields to investigate the WP behavior during the dissimilar welding of Ni and 304SS using laser. The calculated geometry of the fusion zone and the concentration distribution of the main alloying elements (Fe, Ni, and Cr) agree with the corresponding experimental results, verifying the model’s validity. Based on the dimensional analysis, it is found that the transportation of alloy elements is dominated by convection. In the initial stage of WP evolution, the dilution of Fe occurs mainly in the middle section of the WP and tends to stabilize as the WP to reach a quasi-steady state. The WP geometry and velocity show an asymmetric distribution owing to the difference in thermal properties between Ni and 304SS. After WP reaches a quasi-steady state, the fluid flow in the back section of the WP contributes to the uniform distribution of elements along the z-axis. To characterize the element distribution in the WP, the average content of the Fe element flowing into the Ni side is used to design L25(5 3) orthogonal simulation. The most important factors for the distribution of Fe elements are scanning speed (range R=9.45%), spot offset (R=9.17%), and laser power (R=2.11%). Additionally, the average concentration of Fe element flowing into the Ni side is negatively correlated with scanning speed and positively correlated with offset. It is shown that properly decreasing the scanning speed and shifting the spot toward the 304SS side are beneficial for the full dilution and uniform distribution of Fe elements.
