*2.2. Experimental Process Design and Related Parameters*

The optimum SLM process parameters of individual materials have been studied in previous experiments, and the optimized results were shown in Table 4.



At present, connection strategies of dissimilar materials forming include direct joining, gradient path method, and intermediate section method. Direct joining refers to the method of directly forming another material on one material. The gradient path method refers to that although the transition part is added between the two materials, a third material is not introduced in the transition part. The connection is improved by changing process parameters and other factors. The intermediate section method is to add a third material between the two to improve the connection. Direct joining is usually used where the

difference between the physical and chemical properties of the two materials is relatively low while the intermediate section method is opposite. The gradient path method is usually used when the situation is in between. Considering that the physical and chemical properties of NiTi and CuSn10 are quite different, direct joining is useless. The intermediate section method requires complex process exploration, so the gradient path method was selected here for forming. That is, the gradient path method was used to combine the two materials into a part. The strategy of forming CuSn10 alloy after NiTi alloy was adopted in this study. This is mainly because when CuSn10 is formed first, the heat of the laser beam will be rapidly dissipated by CuSn10 with good thermal conductivity when NiTi is formed so that enough heat cannot be gathered to form NiTi with high quality. At the same time, due to that NiTi is very sensitive to heat, the small part of NiTi formed first is easy to become separate from CuSn10 matrix due to thermal stress. The substrate was composed of NiTi alloy so the first layer could be easily connected to the substrate. Direct forming was carried out after remelting once in this experiment. An orthogonal scanning strategy was adopted to form all samples. That is, the laser scanning directions of adjacent layers were perpendicular to each other. The reason why this scanning method was adopted was that different heat flow directions would refine the grain so that the strength could increase [24]. In this study, small cuboids of dissimilar materials were formed, and the analysis of microstructure, element diffusion, microhardness, and phases was conducted on them. The process parameters of the first formed 20 layers of CuSn10 were taken as variables, the influence of which on the interfacial properties was studied. Considering that there was no relevant experiment on forming NiTi/CuSn10 dissimilar materials by SLM, the number of variables was set to only two. Laser power and scanning speed with small fluctuation were adopted to obtain different energy density, trying to preliminarily conclude how the forming of NiTi/CuSn10 dissimilar materials is affected by energy density. The calculation formula of energy density was shown in (1), in which *E* is energy density, *P* is laser power, *v* is scanning speed, *h* is scanning space, and *t* is layer thickness. The selected parameters were shown in Table 5. The scanning space is *h* = 0.08 mm and the layer thickness is *t* = 0.03 mm. Except for these layers, other parts were formed with the parameters in Table 4.

$$E = \frac{P}{vht} \tag{1}$$


**Table 5.** Process parameters of interlayers.

Small cuboids were 10 mm × 4 mm × 10 mm in size. A total of 5-mm high NiTi part was formed first, then the NiTi powders were cleared quickly, and CuSn10 powders were put in the powder chamber. Next 20 layers of CuSn10 (0.6 mm) was formed on the formed NiTi playing the role of transient inter-alloys. Finally, a 4.4-mm high CuSn10 was formed on the formed CuSn10. Tensile samples for the bonding strength test were formed in the same way as these small cuboids. For comparison, NiTi and CuSn10 single alloy tensile samples of the same size were formed with the parameters shown in Table 4. Figure 2 shows the diagram of forming the small cuboids and the tensile samples. Each parameter was used to fabricate three tensile samples to avoid occasionality. Figure 3 shows the sizes of the small cuboids and the tensile samples.

**Figure 2.** Diagram of forming the small cuboids and the tensile samples.

**Figure 3.** Sizes of the small cuboids and the tensile samples.

### *2.3. Microscopic Feature*

The cuboids were removed from the substrate using Wire Electrical Discharge Machining. The sides of the cuboids were sanded and polished. After polishing, the metallurgical defects at the interface of the dissimilar materials were observed by optical microscope (OM) (DMI3000M, Leica, Weztlar, Germany). After observation, samples were etched for 2 s with a mixed solution (6 g FeCl<sup>3</sup> + 10 mL HCl + 150 mL H2O) and 10 s with another mixed solution (1 mL HF + 2 mL HNO<sup>3</sup> + 10 mL H2O) to further observe the microscopic features. The microstructure at the interface was observed with SEM (QUANTA250, FEI, Hillsboro, OR, USA). Energy Dispersive Spectroscopy (EDS) (Noran System 7, Thermo Fisher Scientific, Waltham, MA, USA) was used for point scanning along the direction perpendicular to the interface to analyze the element distribution at the interfacial region. X-ray diffraction (XRD) tests were conducted in a step-by-step manner to analyse phases at the interface: step scan 2θ = 20–90◦ , scanning speed = 0.6◦/min, step width = 0.01◦ . The results were analyzed with jade 6.0.

#### *2.4. Mechanical Properties*

Vickers hardness tester (Wilson VH1202, Buehler, Lake Bluff, IL, USA) was used to measure the microhardness from CuSn10 alloy to NiTi alloy at the interfaces of the cuboids at 50 µm intervals. Tensile tests were performed on a universal testing machine (CMT5105, SUST) at a speed of 0.1 mm per min to measure the interfacial bonding strengths and the tensile strength of CuSn10 and NiTi single alloy.

#### **3. Results and Discussion**
