Objective Laser quenching has the advantages of small thermal deformation and thermal stress, short process cycle, stable and controllable quality, and high treatment efficiency. It can effectively improve the surface wear resistance, corrosion resistance, and fatigue resistance of mechanical parts. Recently, laser quenching has been widely used in many fields, such as automotive, aerospace, and mold. The macroscopic properties of a matrix material are the statistical results of all microscopic grains, and the mechanical properties are determined by the final state of all microscopic grains. The key to achieving precise control of the mechanical properties of the matrix is to optimize the microstructure during laser quenching. Therefore, to reveal the microevolution mechanism of the matrix in the laser quenching process is of great significance for optimizing the microstructural characteristics in the process. Laser quenching is a complex multi-field coupling process, and the microstructure changes instantaneously during the laser quenching process. It will consume a considerable amount of efforts to determine the laser quenching microevolution mechanism through repeatable experiments and traditional numerical simulation methods. In this study, a laser quenching model considering grain heterogeneity was developed on the ABAQUS platform with the Python script. This approach provided an effective method to reveal the laser quenching mechanism at the microcrystalline scale.

Methods First, a random microcrystalline structure model for the matrix was established by the Voronoi tessellation method. Then, the unquenched matrix nano-indentation test was conducted with a Keysight Nano Indenter G200 nano-indentation tester. The test results showed that the grains in the unquenched matrix were inhomogeneous (Fig. 3). The grain non-uniformity coefficient was calculated from the nano-indentation measurement results according to Eq. (8) and analyzed using statistical methods. The analysis results showed that the grain non-uniform coefficient obeys a normal distribution [Fig. 4(a)]. According to the grain non-uniformity coefficient, the grains in the unquenched matrix can be divided into seven types. After considering the sample points and the experimental errors, the grain non-uniformity coefficient distribution was standardized [Fig. 4(b)]. The mechanical properties of each type of grain were calculated according to the grain non-uniformity coefficient [Eq. (9)]. Finally, a Python script was used to randomly assign various material attributes to Voronoi cells in the unquenched matrix according to the grain non-uniformity coefficient after treatment (Fig. 5). A thermo-mechanical coupling model for the laser quenching process of SUS301L-HT stainless steels was established. The temperature field and thermal stress field were calculated.

Results and Discussions During the laser quenching process, the matrix rapidly produces temperatures and thermal stresses under the action of a high-energy laser. The temperature field of the matrix diffuses from the spot center to the surrounding. The matrix is simultaneously subjected to the combined effects of heat radiation, heat convection, and heat conduction during the transfer process. Thus, the matrix temperature decreases gradually from the heat source center to the outside. The temperature field distribution is approximately symmetrical, with the scanning track as the axis. The back of the heat source continuously inputted quantities of heat by heat conduction. Thus, the temperature gradient in the front of the heat source is larger than that at the back. After natural cooling for 300 s, the temperature of the matrix is close to room temperature. The change in the temperature field of laser quenching showed that the characteristics of rapid cooling of laser quenching are prominent (Fig. 7). The characteristics of the temperature field distribution of laser quenching are consistent with the experimental results (Fig. 8). The distribution of the thermal stress field in the matrix is similar to that of the temperature field. Due to the high temperature in the heat source center, the metal mechanical properties in the region are reduced. Thus, the value of the thermal stresses in the center of the heat source is relatively small. The thermal stress of each crystal grain in the matrix is different, and the thermal stress of adjacent grains in the matrix occurs a sudden change at the grain boundary. Therefore, the thermal stress of the entire matrix presents a non-uniform distribution similar to the random geometric structure of the grain boundary. A few grains in the matrix have high mechanical properties. Under the laser irradiation, the thermal stress of these grains is higher than the average level, reaching 1429 MPa. However, the stress level of most grains in the matrix is about 600 MPa. Therefore, it can be found that the part of the grains with a higher thermal stress only represents the thermal stress state of the grains themselves, and it has a little contribution to the thermal stress state of the entire matrix (Figs. 9 and 11).

Conclusions Because of the inhomogeneity of the grain mechanical properties in the matrix, a sudden change in the grain stresses occurs at the grain boundaries. The larger the difference in the mechanical properties between adjacent grains is, the more obvious the stress mutation at the grain boundaries is. The thermal stress isolines show a similar irregular distribution with the grain boundary random geometry for the entire matrix. The laser quenching model considering the grain inhomogeneity can effectively capture the temperature and thermal stress changes of each grain in the matrix during the quenching process.