The laser cladding experiments were conducted using commercial M2 high-speed steel powders with a particle size distribution of 75 - 120 μm and #45 steel as the substrate. The chemical compositions of both the powder and the substrate are provided in Table 1. Prior to processing, the M2 powders were dried in a vacuum oven at 120˚C for 2 hours to reduce moisture, while the substrate surface was thoroughly cleaned with absolute ethanol to eliminate contaminants.

Laser cladding experiments were performed using a self-developed system equipped with a continuous-wave IPG YLR-6000 fiber laser (6 kW). A schematic diagram of the experimental setup is presented in Figure 1. All samples were produced under consistently optimized processing parameters. For the CLC coating, the laser power was set to 2400 W, with a spot diameter of 3 mm, a scanning speed of 1 m/min, an overlap rate of 50%, and a powder feed rate of 8 g/min. The HSLC process differed primarily in the scanning speed, which was increased to 10 m/min. All samples were allowed to cool naturally at room temperature through heat dissipation.

Samples sectioned from the coatings were prepared for microstructural analysis through grinding, polishing, and etching with aqua regia for 5 seconds. Microstructural characterization was carried out using scanning electron microscopy (SEM). The microhardness profiles across the coating cross-sections were measured with a Vickers microhardness tester (model: Vickers-1000) under a load of 1 kg applied for a dwell time of 15 seconds.

To evaluate tribological performance, reciprocating sliding wear tests were conducted at room temperature under dry conditions. Wear test specimens with dimensions of 10 × 10 mm² in cross-section and 6 mm in height were extracted from the coating surface. An Al₂O₃ ball with a diameter of 6 mm was used as the counter body. The test parameters included an applied load of 30 N, a reciprocating speed of 10 mm/s, a stroke length of 5 mm, and a total duration of 30 minutes. The wear tracks were subsequently analyzed using a three-dimensional confocal laser scanning microscope (CLSM) to quantify wear volume and examine surface morphology.

Figure 2 displays SEM micrographs of the coatings produced by different laser cladding techniques. The microstructure of all samples is composed of martensite (appearing in grey) and an interconnected network of carbides (exhibiting bright contrast). Notably, the sizes of both martensitic structures and carbides decrease with increasing laser scanning speed. In addition, the morphology of martensite transitions from coarse and bulky to a finer, needle-like form.

Figure 3 summarizes the Vickers hardness profiles of the coatings along the depth direction. The uniformity of hardness distribution reflects the structural stability of the coating under wear conditions, while the average surface hardness correlates directly with wear resistance. As shown in Figure 3, all samples exhibit a relatively uniform hardness distribution along the depth, indicating good structural stability against wear. However, the average surface hardness decreases as the laser scanning speed increases from 1 m/min to 10 m/min. The measured average surface hardness values are (692.25 ± 23.82) HV for the lower speed and (657.29 ± 17.03) HV for the higher speed. This reduction is primarily attributed to the incomplete transformation from austenite to martensite, which results from the higher cooling rates associated with increased scanning speed.

Figure 4 presents the coefficient of friction (COF) curves obtained during dry sliding wear tests. The COF evolution can be divided into two distinct stages: a running-in stage and a steady-state stage. The contact area between the Al₂O₃ ball and the sample surface significantly influences the frictional behavior, with a larger contact area generally corresponding to a higher COF.

During the running-in stage, initial contact causes preferential wear of the softer martensitic phase compared to the harder carbides. The resulting wear-induced expansion of the contact area leads to a rapid rise in the COF. As sliding continues, plastic deformation generates wear debris. The wear process enters the steady-state stage when a dynamic equilibrium is established between debris generation and its expulsion from the contact zone. The average steady-state COF values for the samples are 0.44 ± 0.08 and 0.42 ± 0.08, respectively.

Figure 5 and Figure 6 present the three-dimensional morphology and two-dimensional cross-sectional profiles of the wear tracks, respectively. The 3D images reveal the presence of furrows aligned parallel to the sliding direction in all samples. As the scanning speed increases, the maximum width and depth of the wear tracks decrease. This trend, however, contrasts with the decrease in hardness observed at higher scanning speeds. The wear volume of the tracks can be calculated using the following equation [16].

V = W_q.avg × S

where V represents the wear volume, W_q.avg denotes the average cross-sectional wear area, and S is the total reciprocating sliding distance. The calculated wear volumes for the samples are 0.29 mm³ and 0.24 mm³, respectively. Contrary to trends commonly reported in previous studies, the wear volume does not increase with decreasing hardness. The sample produced by HSLC exhibits the lowest volume loss. This behavior can be attributed to the refined martensitic matrix and carbide distribution, which enhance load-bearing capacity and mitigate the micro-cutting effect during wear.