Tantalum, which is used in bone implants in the medical field, must be processed with optimal process parameters according to individual differences. Selective laser melting (SLM) technology helps meet this requirement, and the forming process of SLM determines the shaping quality and performance of tantalum. To date, few studies have been conducted on the forming process and performance of tantalum by SLM, and it is particularly important to study the process of SLM-formed tantalum. In this study, to obtain tantalum samples with excellent forming performances, the SLM forming process parameters of tantalum are explored using the control variable method, and tantalum samples with a smooth surface and dense inner are obtained. The microstructures and mechanical properties of the tantalum samples are analyzed. This provides a reference for further optimization of the SLM forming process of tantalum.

In this study, three identical tantalum samples were fabricated under the same conditions by varying the laser power and scanning speed for SLM forming with pure tantalum powder as the material. The surface morphologies and internal defects of the samples were characterized by optical microscope (OM). The densities of the samples were measured using the Archimedes drainage method, and the microhardness of the bottom surface of the samples was characterized by a Vickers hardness tester. The optimal process window and parameters were obtained through comparative analysis. As the next step, cross-sectional erosion and electrolysis of the tantalum samples formed by the optimal process were carried out, and the microstructures were characterized by OM. The crystal structure and grain orientation were analyzed by scanning electron microscope (SEM) with an electron backscatter diffraction (EBSD) probe. Further, the tensile samples were prepared with the optimal process parameters, and the tensile test was performed using an electromechanical universal testing machine. The fracture morphology was characterized using SEM.

When the laser power is 350 W and the scanning speed is 750 mm/s or 850 mm/s, the cooling rate is faster. At this time, the energy density is low and insufficient to completely melt the powder. The temperature gradient along the scanning direction is high; therefore, the surface of the sample presents a discontinuous scanning trace, and there are many pores inside. The microhardness values are 247.15 HV_{and 224 HV,respectively. When the scanning speed is changed in the range of 350?550 mm/s, the time that the laser beam remains on the powder increases, the energy density of the input molten pool increases, and the powder melting degree improves. Although the density of the sample is higher at this time, the powder melting speed is faster; therefore, argon gas remains inside the sample and forms small holes. Small droplet splashes and pores are observed on the surface, and the local hardening effect on the surface results in higher microhardness values. When the scanning speed is 650 mm/s, the melting and solidification of the powder reach equilibrium. No obvious pores or splashes are observed inside the sample, and the surface of the melt channel is flat and pinnate. The microhardness of the sample is 260.92 HV (Figs. 3, 5, 7, and 8).}

When the scanning speed is fixed at 650 mm/s and the laser power varies within the range of 150?300 W, the laser energy density is relatively low, the amount of powder melted per unit volume is lower, surface spheroidization and agglomeration are obvious, and unfused defects appear inside. The density is less than 99.4%, and the microhardness increases from 148.05 HV to 253.62 HV. When the laser power is increased to 350 W, the laser energy density increases, which effectively suppresses the spheroidization and agglomeration. The surface of the sample forms a regular melting channel, and the internal bonding is tight, without obvious defects. When the laser power is 400 W, the laser energy is high, molten pool is unstable, and surface is continuously flat with splashes and pores. Although the sample is dense, there are pores inside the sample, and there are cracks and unfused defects at the edges (Figs. 4, 6, 7, and 8).

The optimal process window and parameters are determined by analyzing the surface morphology, internal defects, and density. The lengthwise-section microstructure of the tantalum samples prepared with the optimal process parameters has relatively large axis-to-diameter ratios and grows across layers along the forming direction. The crystal orientation is not obvious, which is related to the interlayer rotation angle of 67° (Figs. 2, 9, and 10). The mechanical properties are better than those of tantalum samples formed by traditional methods, with a yield strength of 668 MPa. The fractures of tantalum tensile samples prepared using the optimal process parameters have macroscopic plastic deformation characteristics, and there are dimples with different sizes, indicating that the fracture mechanism is ductile fracture (Figs. 11 and 12 and Table 3).

In the process of forming tantalum by SLM, a higher scanning speed or lower laser power produces a larger temperature gradient, which causes obvious spheroidization on the surface of the sample, accompanied by pores or unfused defects inside, and the density of the sample is low. A lower scanning speed or higher laser power produces excess laser energy, resulting in excessive melting of the powder. At this time, the cooling rate of the molten pool is low, resulting in cracks at the edge, a hardening effect on the surface with a small amount of spatter, and pores inside.