Abstract: In order to prepare a nickel-based WC composite coating with excellent forming quality and mechanical properties on 42CrMo, the Taguchi design experimental method was used to explore the influence of process parameters on the performance of nickel-based WC composite coatings, with laser power, powder feeding rate and scanning speed as influencing factors, and dilution rate and microhardness as response indicators. The results show that laser power has the greatest influence on the dilution rate and microhardness of the composite coating, followed by powder feeding rate, and scanning speed is the smallest. When the laser power is 2000 W, the scanning speed is 10 mm·s-1, and the powder feeding rate is 1.4 r·min-1, the microstructure of the coating is uniform and dense, and W, C, and Nb elements are segregated between grains, and a large amount of W elements are precipitated near the WC particles. The formation of other hard phase carbides such as WC, W2C, and NbC makes the microhardness of the composite coating about 1.7 times that of the substrate. Its wear mechanism evolves from adhesive wear to abrasive wear. The friction coefficient is about 0.12 lower than that of the substrate, which significantly improves the mechanical properties of 42CrMo.
Keywords: laser cladding; process optimization; composite coating; microstructure; mechanical properties
Laser cladding is an emerging technology to improve the surface properties of metals. Its principle is to fuse powder and substrate materials together through high-energy laser heat sources. The molten pool quickly solidifies on the surface of parts to form a dense and strong alloy coating, thereby improving the comprehensive performance of parts such as wear resistance. Compared with traditional surface strengthening technology, it has the advantages of small deformation, low dilution rate, and good morphology, and is widely used in the field of laser repair and remanufacturing. 42CrMo, which has excellent hardness, toughness and hardenability, is often used in engineering parts such as oil drill pipe joints, transmission shafts, gearbox gears, engine cylinders, etc. However, 42CrMo materials often suffer from wear and fatigue fracture under harsh working conditions. Due to the high cost of processing and replacing damaged parts, it is of great research significance to strengthen and repair the surface of parts through laser cladding technology.

Appropriate process parameters can prepare composite coatings with strong bonding, dense structure, excellent forming quality and performance. Hulka et al. studied the effect of laser cladding scanning speed on the microstructure and wear resistance of the coating, and obtained a dense, crack-free coating with good carbide distribution. Yang Kaixin et al. combined the Taguchi method with the gray theory for analysis, and improved the coating morphology and microstructure by controlling the process parameters. Wang Tao et al. [4] found that the IN718 coating had the best forming quality and wear resistance when the scanning speed was 14 mm/s, and the microstructure was dense and the grain growth distribution was uniform. Shi Xiaoshuai et al. [5-6] investigated the effects of laser power, spot diameter, and scanning speed on the morphology and quality of Fe-based alloy coatings based on the response surface method, and achieved accurate prediction of the geometric morphology model.

In this paper, a hard phase WC nickel-based composite coating was prepared on 42CrMo steel. The Taguchi method was used to analyze the effects of laser power, powder feeding rate, and scanning speed on the process optimization of the composite coating. The composite coating was subjected to phase and microstructure analysis and mechanical property testing, providing relevant theoretical references for the engineering application of nickel-based WC composite coatings.

1 Experimental design
1.1 Experimental equipment and materials
The laser cladding equipment uses a TruDisk disc laser, a TRUMPF circular laser cladding head with a spot size of 3 mm, an RC-PGF coaxial continuous powder feeder, a KUKA robotic arm, and other related workbenches. The substrate is made of 150mm×80mm×10mm 42CrMo steel, and the cladding powder is a mixed powder of 15wt% WC and IN625. The powder morphology is shown in Figure 1, and the chemical composition of the substrate and IN625 powder is shown in Table 1.

1.2 Experimental and test methods
The Taguchi design experimental method is used. With laser power P, scanning speed Vs and powder feeding rate Vf as the research objects, and coating dilution rate and microhardness as the response indicators, a three-factor five-level orthogonal experiment is designed to optimize the laser cladding process parameters. The experimental measurement results are shown in Table 2. The single-pass coating forming parameters are measured using a stereo microscope, and the dilution rate is calculated using the simplified formula (1). The Vickers hardness tester is used to test the hardness of the cladding layer cross section bypassing the WC particles. The optimized composite coating is analyzed for phase and microstructure using an X-ray diffractometer and a SEM electron microscope, and the substrate and cladding layer are subjected to friction tests. Formula (1) η=h/(h+H), where: η is the dilution rate; h is the melt height; H is the melt depth.

2 Experimental results and analysis
2.1 Analysis of macroscopic morphology of cladding layer
As shown in Figure 2, the lower laser energy cannot fully melt the metal powder, and the coating surface has local poor fusion, powder sticking and other molding quality problems; too high laser energy causes large thermal stress in the molten pool, micro cracks appear on the coating surface, and the substrate melts further. The lower scanning speed greatly increases the laser irradiation energy per unit time, and the coating surface burns; the too high scanning speed reduces the laser energy, and the coating has incomplete powder melting and reduced gloss. The increase in powder feeding rate increases the density of cladding powder and improves the coating height; the excessive powder feeding rate reduces the energy absorption of the substrate, and the decrease in the fluidity of the molten pool causes the coating to be uneven and uneven.

2.2 Analysis of factors affecting the dilution rate of cladding layer
2.2.1 Analysis of the main and secondary factors affecting the dilution rate
The dilution rate is an important basis for judging the bonding of the coating. The coating with a smaller dilution rate has poor bonding performance and is prone to slippage; the larger dilution rate is prone to defects such as inclusions, pores, and cracks. The coating dilution rate was analyzed by minimization analysis, and its mean and signal-to-noise ratio responses are shown in Tables 3 and 4. The range value Delta of each factor represents the weight of the horizontal fluctuation of the factor on the coating dilution rate response. The signal-to-noise ratio is the ratio between the response mean and the mean fluctuation, and is often used to measure the robustness of the experimental performance. The larger the value, the better. Through comprehensive analysis, it is found that the most important factor affecting the dilution rate of the nickel-based composite coating is the laser power, followed by the powder feeding rate, and the scanning speed has the smallest weight on the coating dilution rate.

2.2.2 Analysis of the main effect of dilution rate
As shown in Figure 3, the gradual increase in laser power leads to a rapid increase in the dilution rate of the cladding layer, the increase in the coating melting depth is greater than the increase in the melting height, and the area of ​​the substrate melting and the deposited cladding layer increases. The increase in scanning speed reduces the effective laser energy and powder mass per unit time, the coating height and depth decrease, and the dilution rate shows a slow upward trend. The increase in powder feeding rate leads to the continuous accumulation of coating melting height, the increase in powder melting absorption energy, the corresponding substrate melting area decreases, and the coating dilution rate shows an overall downward trend. As shown in Figure 4, when the laser power is 2000W, the coating dilution rate is relatively low, and its average dilution rate is 32.64%; when the laser power increases from 2100W to 2200W, the average coating dilution rate increases significantly, and its average dilution rate increases from 35.94% to 45.53%; when the laser power exceeds 2200W, the coating dilution rate tends to be flat.

2.2.3 Interactive analysis of factors affecting dilution rate
As shown in Figure 5 (color figure in the electronic version, the same below), when the dilution rate of the nickel-based composite coating is 20% to 30%, the reasonable range of laser power, scanning speed, and powder feeding rate should be controlled within the range of 2000 to 2200W, 9 to 12 mm·s-1, and 1.2 to 1.4 r·min-1; when the laser power is 2200 to 2400W and the powder feeding rate is 1.0 to 1.2 r·min-1, the composite coating
dilution rate is relatively large and the molding quality is poor. Comprehensive analysis shows that when the laser power is 2000W, the scanning speed is 10 mm·s-1, and the powder feeding rate is 1.4 r·min-1, the coating dilution rate is relatively small.

2.3 Analysis of factors affecting microhardness of cladding layer
2.3.1 Analysis of factors affecting microhardness
In order to obtain strong wear resistance, the microhardness of the coating is analyzed by the maximum value analysis, and its mean and signal-to-noise ratio response are shown in Tables 5 and 6. The ranking of the extreme values ​​of each factor shows that the laser power has the greatest influence on the microhardness of the coating, followed by the powder feeding rate, and the scanning speed has the smallest influence weight. The mean response is consistent with the signal-to-noise ratio response. By combining the signal-to-noise ratio response table, it can be judged that the laser power is both a divergence factor and a position factor, that is, the laser power determines the direction of the microhardness of the composite coating to a large extent.

2.3.2 Analysis of main effects of microhardness
The mean value of microhardness in Figure 6 shows that the increase of laser power leads to a rapid decrease in the microhardness of the cladding layer. Excessive laser energy affects the nucleation rate and fluidity of the molten pool, allowing more substrates to participate in the melting reaction, thereby affecting the number and type of coating grains. The increase in scanning speed causes the overall microhardness of the composite coating to decrease first, then increase, and then decrease. This is because the scanning speed has a great influence on the cooling rate of the molten pool. The temperature gradient generated at different cooling rates during the cladding process leads to different supercooling when the powder and the substrate are fused and crystallized, which affects the coating crystal nucleation state and the grain growth rate and growth direction, causing the coating microhardness to fluctuate. The increase in the amount of powder fed per unit time reduces the supercooling of the molten pool, and the density and mechanical properties of the cladding layer are greatly improved [7]. As shown in Figure 7, when the laser power is between 2000 and 2100 W, the coating microhardness is relatively large, and the highest microhardness of a single sample is 436.71 HV0.2; when the laser power is 2400 W, the coating microhardness is the lowest, which is 308.01 HV0.2.

2.3.3 Interactive analysis of factors affecting microhardness
As shown in Figure 8, when the microhardness of the coating is between 380 and 450 HV0.2, the reasonable range of laser power, scanning speed, and powder feeding rate should be controlled within the range of 2000 to 2200 W, 10 to 12 mm·s-1, and 1.2 to 1.4 r·min-1. Comprehensive analysis shows that when the laser power is 2000 W, the scanning speed is 10 mm·s-1, and the powder feeding rate is 1.4 r·min-1, the microhardness of the coating is relatively large.

3 Experimental verification and analysis
3.1 Microstructure and phase analysis
Figure 9 shows the line scan results from the top of the optimized composite coating to the substrate. It can be seen that Fe, Cr, Ni, Mo, Nb, W, and C are evenly distributed in the cladding area, the Ni content is relatively high, and the C content is the lowest. The strong fluidity at the bottom of the molten pool causes the element content at the bottom of the coating to change dramatically. The Fe content shows a dramatic increase trend toward the substrate, the C element increases sharply and then decreases, and the remaining elements all decrease rapidly. From the XRD spectrum of Figure 10, it is found that the composite coating phase contains γ-Ni, WC, W2C, NbC, M6C, and M23C6, where M represents Fe, Cr and other elements. Part of WC reacts with IN625 to generate carbides such as W2C, and W2C reacts with nickel-based powder and substrate to generate new carbides. The reaction process is shown in equations (2) to (4).

Figure 11 shows the microstructure of the upper, middle and lower regions of the composite coating. Due to the small temperature gradient G at the top of the coating and the large solidification rate R, a large number of fine equiaxed crystals and cellular crystals appear at the top of the coating, and there are a small number of irregular dendrites that are randomly arranged. As the temperature gradient G increases, the accumulation of heat increases the dendrite nucleation rate and grain growth rate. At the same time, the decrease in crystallization rate increases the G/R ratio, and the fine dendrites and eutectic structures gradually evolve into densely arranged coarse dendrite structures. Regularly arranged cellular crystals and dendrites are found in the middle of the coating, the grain growth trend is vertically crossed, the WC particles are relatively complete, and blocky precipitates exist on the surface and around the WC. Slender dendrites and cellular crystals that grow perpendicular to the bonding line appear at the bottom of the coating.

Table 7 shows the results of EDS analysis of intracrystalline, grain boundary and WC precipitates in the middle of the coating. By comparing the intracrystalline A with the grain boundary B, it is found that the grain boundary contains high Cr, Mo, Nb, C and W elements, and the intracrystalline is rich in Fe and Ni elements. In addition, the C content in the grain boundary and the intracrystalline is far higher than the W element, and the C and W contents in the grain boundary are higher than those in the intracrystalline. The W content in the carbon-containing precipitates is greatly increased, and the Cr, Mo, and Nb contents are somewhat increased compared to those far away from WC. From the scanning results of Figure 12, it can be seen that elements such as C, W, and Nb are segregated to the grain boundary, and C, Fe, Ni, and Cr elements account for a high proportion, while the remaining elements are relatively evenly distributed.

3.2 Microhardness analysis
As shown in Figure 13, the average hardness of the cladding zone is 441.6 HV0.2, and its highest microhardness is 471.2 HV0.2, which is about 1.7 times the hardness of the matrix. Since the solidification phase transformation is continuously pushed from the bottom of the molten pool to the top of the cladding layer during the cladding process, the crystal growth time at the bottom of the molten pool is greater than that at the top of the coating. The fine and dense grain structure on the top of the coating promotes the improvement of the microhardness of the coating, so that the hardness of the cladding layer decreases step by step as a whole [11]. The substantial improvement of the microhardness of the heat-affected zone is due to the rapid cooling of the laser, which produces needle-shaped quenched hard and brittle martensite near the melting zone of the substrate. A large amount of supersaturated alloy elements precipitate on the martensite in the form of metal carbides, which significantly improves the microhardness of the heat-affected zone. In actual measurements, it is found that the hardness of the area near the WC particles is higher than that of the area far away from the WC particles, and the hardness of the WC particles can reach 2804.3~3324.2HV0.2, which greatly improves the overall mechanical properties of the cladding layer.

3.3 Friction and wear analysis
Figure 14 is a friction coefficient curve of the substrate and the composite coating. It can be seen that when the friction time is 300 s, the friction coefficient of the substrate slowly rises to 0.73. Due to the high wear resistance of the hard phase WC, the friction coefficient of the composite coating reaches a maximum of 0.90 in the initial stage, and it tends to 0.61 relatively steadily after 800 s. By observing the friction and wear morphology Figure 15, it can be seen that the wear mechanism of the substrate is mainly adhesive wear and abrasive wear, accompanied by a large number of debris and pits. The adhesive wear of the composite coating is significantly reduced, and its wear mechanism is mainly abrasive wear. The uniformly distributed hard phase WC particles produce block peeling along the wear scar direction under the continuous shear sliding action of the friction pair. Combined with the XRD spectrum, it can be seen that under the action of hard phases such as WC, W2C, and NbC, the composite coating significantly improves the wear resistance of the substrate.

4 Conclusions
In this paper, the Taguchi design experimental method was used to clad the nickel-based WC composite coating on the surface of 42CrMo. The laser power, powder feeding rate and scanning speed were used as influencing factors, and the coating dilution rate and microhardness were used as response indicators. The influence of process parameters on the forming quality and mechanics of the nickel-based composite coating was explored, and the following conclusions were drawn:
(1) Combining the mean response and signal-to-noise ratio analysis, it was found that the laser power and powder feeding rate had a greater influence on the dilution rate and microhardness of the nickel-based WC composite coating, and the scanning speed had the lowest influence. The increase in laser power caused the coating dilution rate to increase significantly, which was in the opposite trend to the microhardness. When the laser power was 2000W, the scanning speed was 10 mm·s-1, and the powder feeding rate was 1.4 r·min-1, the comprehensive performance of the composite coating was better.
(2) Through the analysis of the phase composition and microstructure of the nickel-based WC composite coating, it was found that there were hard phase carbides such as WC, W2C, and NbC inside the coating. WC was well combined with the nickel-based alloy powder. There were dense eutectic structures such as irregular equiaxed crystals and cellular crystals on the upper part of the coating. Energy spectrum analysis of the middle part of the coating revealed that there were blocky precipitates with a high W content around the WC particles. The C and W contents in the grain boundaries were higher than those in the grains, and segregation of C, W, and Nb elements occurred.
(3) The average microhardness of the composite coating was 441.6 HV0.2, and the maximum microhardness was 471.2 HV0.2, which was about 1.7 times the hardness of the substrate. The surface hardness of WC could reach a maximum of 3324.2 HV0.2. By observing the wear morphology of the composite coating, it was found that its wear mechanism was abrasive wear, and the friction coefficient decreased by about 0.12 compared with the substrate, that is, the formation of WC particles and hard phase carbides can play a good role in improving mechanical properties.