In the laser cladding process, a multi-cladding layer with a large thickness is required to satisfy the requirements of industrial production. To improve the performance of the cladding layer, the powder metal is different from the matrix, and therefore, the elements in the cladding layer need to change from matrix to powder elemental composition. The properties of the cladding layer are affected by the distributions of elements. The faster the cladding elements change from matrix elements to powder elements, the more metal powder elements are contained in the cladding layer, which has better abrasion resistance. Therefore, it is of great significance to analyze the transient changes in the temperature field, flow field, and element distribution by numerical simulation of the laser cladding 316L powder multilayer stacking process as well as study the distribution mechanism of Cr elements in the cladding layer, providing a theoretical basis for the cladding layer to contain a higher proportion of powder elements and fewer matrix elements.
The multilayer laser cladding process of 316L powder on a 45-steel matrix is studied using a three-phase melting and solidification model based on the volume averaging method. The distribution mechanism of elements in the process of cladding layer stacking is clarified by comparing and analyzing the changes of temperature field, flow field, and solute field in the first three layers. The simulation results are verified from four aspects: the geometric morphologies and Cr concentrations of the first and second cladding layers.
The geometric morphologies of the molten pool and element distributions of the cladding layers are verified by comparing the experimental and simulation results of the first and second cladding layers in the stacking process (Figs. 4, 5, and 6). The Cr element is used as a tracer element to analyze the distribution mechanism of the cladding layer element (Fig. 10). The simulation results show that the molten pool morphologies and Cr element distributions of the first three layers are highly similar to the experimental results. During laser cladding, the matrix and powder continuously absorb energy, leading to a rapid increase in temperature and the formation of a small molten pool. As the laser beam moves, the molten pool continues to increase and becomes stable after a certain period (Fig. 7). Under the influence of heat accumulation, a W-shaped temperature field distribution is formed during the cladding of the second and third layers, forming longer and deeper molten pools (Fig. 8). The maximum flow velocity in the molten pool appears on the upper surface of the molten pool and decreases in the cladding process of the second and third layers. When cladding the second and third layers, the original cladding layer is partially remelted (Fig. 9), and the matrix elements in the remelted area enter the molten pool under the force of Marangoni and mix with powder elements. As the molten pool moves, the powder is continuously sent into the molten pool, leaving a stable area with a higher Cr concentration at the back end of the molten pool.
To study the element distribution mechanism in the cladding process of the first three layers, combined with experimental verification, we simulate the stacking process of multilayer laser cladding, and achieve an accurate prediction of element distribution after cladding layer stacking. The technological parameters to form the elements of the cladding layer similar to the metal powder elements with a minimum layer number can be subsequently studied. This provides a theoretical basis for the repair of high-end parts. The main conclusions are as follows. The reliability of the model is verified by comparing the molten pool morphologies and Cr element distribution results of the first and second layers obtained by simulation and experiment. The results indicate that the melting height error of the first layer is 5.81%, the melting depth error of the first layer is 3.23%, the melting height error of the second layer is 2.33%, and the melting depth error of the second layer is 3.23%. The slight errors in the molten pool morphology and the Cr distributions in the cladding layer obtained by the experiment and simulation are consistent, which proves that the current numerical model is reliable. In the first three layers during multilayer laser cladding, clockwise vortices exist in the front of the pool and counterclockwise vortices exist in the back of the pool, caused by the Marangoni effect in each layer. The length and depth of a molten pool increase because of heat accumulation. For the first three layers, the temperature gradients in the molten pool on the upper surface are G1>G2>G3. The decrease of the temperature change rate leads to the decrease of the maximum velocity. The original cladding layer partially remelts for the second and third layers, and the matrix elements in the remelting area enter the molten pool and are diluted by powder elements. Therefore, the cladding layer elements further transition from matrix elements to powder elements. After selecting Cr element as the tracer element of powder element, we find that the mass fraction of Cr element progressively increases with the height in the cladding layer, approximately 0.004 for each layer. Cr is easily enriched near the interface between the remelting and nonremelting areas, and the mass fraction of Cr increases by approximately 0.002 in the enrichment area.
