Additive manufacturing (AM) is a revolutionary digital manufacturing technology. Compared with traditional manufacturing techniques, components fabricated using AM exhibit high geometric freedom and excellent performance. They have been widely used in aviation, aerospace, nuclear energy, and other fields. Laser directed energy deposition (LDED) is one of the most widely employed manufacturing technologies. Currently, problems, such as porosity, cracking, deformation, and the precipitation of brittle phases, still exist in certain L-DED components, especially high-entropy alloy (HEA) components. Revealing the formation mechanism of the L-DED process in HEAs is of great significance for macro-defect and microstructure regulation. However, the laser directed energy deposition process is accompanied by complex physicochemical phenomena such as extremely high temperatures, vaporization, rapid melting, and solidification. However, it is challenging to examine this mechanism experimentally. To reveal the formation mechanism of HEAs, a numerical model for the L-DED process of AlCoCrFeNi HEAs is developed. This model makes it possible to analyze the molding process from melt pool data, which are difficult to observe directly. We hope that our findings will aid in the study of the physical processes of L-DED and the regulation of the microstructure of components.

In this study, a numerical model was developed based on the coupled method of the Lagrange discrete-phase model (DPM), volume fraction of fluid (VOF) method, and adaptive laser body heat source model. The physical processes involved in the model mainly include the interactions among air, powder, laser, and metal. The interface between the deposited layer and the air was calculated and tracked using the VOF method. The enthalpy?porosity method was used to simulate the melting and solidification of the metal phase. Simultaneously, the DPM was used to characterize the motion of metal powders, and the collisions between particles were neglected. When the particles came in contact with the melt pool, the momentum, energy, and mass carried by the particles were transformed into the momentum, energy, and mass of the metal phase in the form of source terms using the coupled DPM-VOF model. Additionally, a Gaussian body heat source with adaptive adjustment of the range of action was used to simulate the interaction between the laser and the deposited layer. Subsequently, the developed model was solved using the finite-volume method. Finally, single- and four-layer samples were fabricated for the model validation and microstructural observations.

A comparison between the simulation and experimental results shows that the developed model exhibits high reliability (Fig. 7). The simulation results show that the melt pool is mainly affected by gravity, Marangoni force, and powder particle impact effects during the L-DED process. The impact of the particles interrupted the convection mode at the center of the melt pool (Figs. 10 and 11), which has an important effect on the distribution of the flow and temperature fields in the melt pool. Additionally, during the formation of the thin-walled AlCoCrFeNi HEA samples, a columnar-to-equiaxial transition (CET) is observed in the microstructure from the bottom to the top of the sample (Fig. 17). According to the simulation results, the maximum temperature gradient at the solid?liquid interface at the back of the melt pool decreases from 1.26×10⁶ K/m in the first layer to 5×10⁵ K/m in the fourth layer (Fig. 14), whereas there is no significant change in the distribution of the solidification rate (Fig. 16). Hence, the average G/R value at the solid?liquid interface at the back of the melt pool decreases from 5.16×10⁹ to 1.06×10⁹ (Table 7). According to the theory of non-equilibrium solidification of alloys, when the G/R value decreases, the tendency of the alloy to form columnar crystals during solidification decreases, and the tendency to form equiaxial crystals increases, which explains the phenomenon of CET in the AlCoCrFeNi samples.

In this study, a numerical model of the L-DED forming process is developed to simulate the machining of single-track and thin-walled structures of an AlCoCrFeNi high-entropy alloy. The model shows high reliability based on a comparison of the geometric data between the simulation and experimental results. According to the simulation results, the flow field of the melt pool is mainly affected by gravity, the Marangoni effect, and the impact of the particles. The intermittent impact of the particles on the melt pool is mainly concentrated near the center of the melt pool, which interrupts the convection mode in this region and thus affects the temperature distribution of the melt pool. Consequently, the influence of particles on the temperature and flow distributions of the melt pool cannot be ignored. Additionally, the microstructure of L-DED AlCoCrFeNi thin-walled samples shows a transition from columnar to equiaxial crystals from the bottom to the top, which is mainly affected by the solidification rate (R) and temperature gradient value (G) along the back edge of the melt pool. According to the simulation results, the average G/R value at the solid?liquid interface at the back of the melt pool decreases from 5.16×10⁹ in the first layer to 1.06×10⁹ in the fourth layer. The change in the solidification conditions is the main reason for the CET transformation of the samples. Our study provides a method for exploring the relationship between the forming process and microstructure of parts fabricated via L-DED.