*3.3. Phase Analysis*

Figure 5 shows the phase analysis of the wear test on the cladding surface by the XRD test. It was found that the cladding layer was mainly composed of TiC0.957(PDF#89-2716) and Ni3Fe(PDF#88-1715). The diffraction spectrum for different process conditions varied from each other only in the relative intensity of peaks. This was mainly due to the weight fraction of TiC in cladding powder, and the grain growth status occurred by process parameters. Specifically, for the 11# specimen, where TiC particles amount to 60% weight of deposited powder, the peak intensity of TiC0.957 was significantly strengthened as expected. It was also observed that the percentage of element C in TiC composition had decreased. This would be caused by the element loss during the heating process of laser-material interaction and replacement with Ni in the crystal structure of TiC at the molten pool [16].

**Figure 5.** XRD spectrum.

## *3.4. Microstructure and Element Distribution*

Figure 6 illustrates the element distribution of cladding layers at laser powers of 1.2 kW and 1.4 kW (Scanning speed of 7 mm/s, the gas-powder flow rate of 1000 L/h, and 60% of TiC). From the figure, it can be found the molten pool experienced strong convective flow during laser cladding. With higher laser input, the molten pool was prolonged, the element Fe was more diffused, by convection and diffusion, into the cladding layer from the substrate, especially to the boundary of three phases cladding layer-substrate-air and center of the molten pool. This, to a certain extent, diluted the cladding layer and decreased the fraction of the hard phase, thereby deteriorating the wear resistance.

**Figure 6.** Element distribution of cladding layers at different laser powers. (**a**) 13# sample: 1.2 kW, 7 mm/s, 1000 L/h, 60% TiC; (**b**) 3# sample: 1.4 kW, 7 mm/s, 1000 L/h, 60% TiC.

Figure 7 compares the microstructure of cladding layers at different laser powers. It can be seen that a large amount of dendritic crystal existed in the top, center, and bottom of the cladding layer. Element analysis by EDS showed that the fraction of TiC at the dark area was obviously greater than that of light area, where the percentage of Ni overweighed. Considering the XRD result, the microstructures in the dark and light areas were probably TiC and Ni-based alloy, respectively. The dendritic microstructure is probably produced from the growth of renucleared TiC grain after decomposition due to heating at the laser cladding process [17].

As the laser power increased, more energy input was imposed onto the cladding powder. The more hard phase was thus decomposed, resulting in a decrease of hard phase retaining in the cladding layer. The microstructure of Ni-based alloy at different laser powers was a flatten crystal. At higher laser power (1.4 kW), the grain size was greater with more apparent crystal boundary. This was because increased laser power had prolonged the span of the molten pool, increased thermal gradient, and enhanced heat convection. Part of Fe elements and Ti elements left by the decomposition of TiC was concentrated in the grain boundaries, and this grain boundary structure with relatively weak mechanical properties will further reduce the inter-grain bonding strength and increase the wear rate [18].

According to the literature [5,19], the microstructure of the cladding layer is mainly determined by the solidification rate of the molten pool. At a certain scanning speed, the ratio of G to R (G is the temperature gradient, and R is the solidification rate) declined from the bottom of the molten pool to the top, while the growth rate of crystals at the same cross-section increased. As illustrated by the microstructure in the middle area in Figure 7a,b, due to the variation of G/R, the grain size of dendritic TiC was much greater than that of the bottom. The microstructure at the molten pool top manifested similar circumstances as the top and middle areas shown in Figure 7a,b. The dark grey hard phase at the top region of Figure 7a (2# sample) accounted for 16.84%, and the dark gray hard phase at the top of Figure 7b (4# sample) accounted for 15.29%. As shown in Table 4, the hardness of the 2# sample was 59.7 HRC, the hardness of the 4# sample was 52.6 HRC. The laser cladding hard particle-reinforced composite coating had high hardness, uniform structure, and excellent resistance to abrasive wear. When the laser power was 1.2 kW, it was found that some secondary dendrite axis was fully grown at the top area of the cladding layer, which facilitates anchoring in the Ni-based alloy and avoids hard phase being sloughed as abrasive grains [20]. Figure 8 shows the friction and wear diagram of the cladding layer with different laser powers. As shown in Table 6, the content of element Fe varied slightly through the top, middle, and bottom parts of the cladding layer.

**Figure 7.** Microstructure of cladding layers at different laser powers: (**a**) 2# sample: 1.4 kW, 7 mm/s, 1000 L/h, 20% TiC; (**b**) 4# sample: 1.2 kW, 7 mm/s, 1000 L/h, 20% TiC.

**Figure 8.** Friction and wear diagram of cladding layer with different laser powers: (**a**) 2# sample: 1.4 kW, 7 mm/s, 1000 L/h, 20% TiC; (**b**) 4# sample: 1.2 kW, 7 mm/s, 1000 L/h, 20% TiC.

**Table 6.** The element content of cladding layers at different laser powers (Labels 1–6 belong to 2# sample, Labels 7–12 belong to 4# sample).


The microstructure of the cladding layer obtained with 60% TiC indicated that the energy demand for decomposition and renuclearation of all the TiC particles grew increasingly with more TiC added in the cladding powder. The undecomposed TiC became nucleation sites and grew into the microstructure, as illustrated in Figure 9b. According to the distribution of thermal gradients along the cladding layer, TiC clustered more at the top and grew fully, while the TiC at the middle part grew with exceptional size, and at the bottom, more dendrite was obtained. The dark grey hard phase at the top region of Figure 9a (12# sample) accounted for 18.57%, and the dark gray hard phase at the top of Figure 9b (11# sample) accounted for 71.8%. As shown in Table 4, the hardness of the 12# sample was 55.4 HRC, and the hardness of the 11# sample was 70 HRC. From the energy spectrum by EDS, the percentage of TiC in the dark area increased obviously up to 70% as the weight fraction increased. Since TiC contained a more hard phase, the wear resistance was thus improved. With more TiC in the cladding powder, the effect of second phase enhancement became more apparent, which further strengthened the wear resistance. Figure 10 shows that the wear rate of the cladding layer decreased as the weight fraction of TiC increased in the cladding powder. This was due to the characteristic of high hardness and good wear resistance for the ceramic powder TiC. It was reported that the wear resistance of alloys was simply proportional to the area percentage of the hard phase. Increasing TiC percentage significantly improved the hardness and wear resistance of the cladding layer and declined the wear rate per unit time. As shown in Table 7, the content of element Fe varied slightly through the top, middle, and bottom parts of the cladding layer. This is owing to the fact that most of the imposed energy is absorbed by the TiC for decomposition and nucleation [21]. It also resulted from the decreased flowability of the melt pool as a more hard phase of the high melting point was mixed. Therefore, the di ffusion of Fe was not widely observed, and the dilution e ffect was restrained.


**Table 7.** Element content table of micro TiC cladding layer structure determination. (Labels 1–6 belong to 12# sample, Labels 7–12 belong to 11# sample).

**Figure 9.** Microstructure of the top, bottom, and middle of the cladding layer with different TiC powder ratios: (**a**) 12# sample, 1.2 kW, 5 mm/s, 1000 L/h, 20% TiC; (**b**) 11# sample, 1.2 kW, 5 mm/s, 1000 L/h, 60% TiC.

**Figure 10.** Friction and wear diagram of cladding layer with different TiC powder ratios: (**a**) 12# sample: 1.2 kW, 5 mm/s, 1000 L/h, 20% TiC; (**b**) 11# sample: 1.2 kW, 5 mm/s, 1000 L/h, 60% TiC.

Figure 11 reflects the interaction effect of scanning speed and the weight ratio of TiC on the wear rate of the cladding layer. It can be seen that the wear rate increased as the scanning speed increased. This was because scanning speed determined the time span of the melt pool affected by the laser beam. The larger the speed, the shorter the energy effect time, which was adverse to thorough decomposition and renucleation of TiC. Specifically, at a higher scanning speed of 7 mm/s, as shown in Figure 11a, a large amount of fine TiC grain was observed at the top region of the cladding layer due to the more laser energy input. Compared with a needle or dendritic microstructure, near sphere structure was much preferable as it mitigated the possibility of stress concentration. The contact area between the spherical particles and the substrate was small, and the degree of fit between them was low, so the spherical particles were more likely to fall off during the friction process, resulting in abrasive wear, which reduced the wear resistance and increased the wear rate per unit time. As the scanning speed was decreased, the effect time of laser beam on the cladding powder and substrate was increased, and the undercooling was thus enlarged. The dendritic TiC crystal structure was obtained with finer sizes at the bottom. Furthermore, the smaller the speed, the larger amount the secondary phase at the crystal boundary. From the results of EDS, as listed in Table 8, the content of element Fe in Figure 12a was much greater than that in Figure 12b. The increase in Fe content, to some extent, diluted the fraction of the hard phase, thereby deteriorating the wear resistance. However, this was comparably less influencing the effect of the profile and distribution of the hard phase. It was presumed that the above-mentioned two effects counteract with each other so that the influence of scanning speed on the wear resistance is weakened even though the microstructure had been significantly changed [22]. The dark grey hard phase at the top region of Figure 12a (6# sample) accounted for 15.89%, and the dark gray hard phase at the top of Figure 12b (8# sample) accounted for 16.52%. As shown in Table 4, the hardness of the 6# sample was 58.7 HRC, and the hardness of the 8# sample was 59.7 HRC. Figure 13 shows the friction and wear diagram of the cladding layer at different scanning speeds.

**Figure 11.** (**a**) 3D response curve of the interaction between *SS* and *PR* against wear rate; (**b**) Contour map of the interaction between SS and PR against wear rate.

**Figure 12.** Microstructure of the top, middle, and bottom of the cladding layer at different scanning speeds: (**a**) 6# sample: 1.4 kW, 7 mm/s, 1400 L/h, 20% TiC; (**b**) 8# sample: 1.4 kW, 5 mm/s, 1400 L/h, 20% TiC.

**Figure 13.** Friction and wear diagram of cladding layer at different scanning speeds: (**a**) 6# sample: 1.4 kW, 7 mm/s, 1400 L/h, 20% TiC; (**b**) 8# sample: 1.4 kW, 5 mm/s, 1400 L/h, 20% TiC.


