Influence of Axle Weight and Frequency on the Tribological Properties of Laser-Repaired 316L Stainless Steel Coatings in Railway Wheel Tread Braking

Abstract

The impact of the complex braking environment on the service performance of the repair fusion cladding was studied, which is of great significance to improve the ability of the train wheel track system to resist the extremely harsh environment. In this paper, a 316L stainless steel coating was prepared using laser fusion cladding repair technology for the local damage location of the train wheel tread. The MS-HT1000 high-temperature wear tester was used for the experiment. Then, the influence of different braking conditions on the friction and wear performance of the repaired specimens at room temperature and high temperature was analyzed. The results showed that the microstructure of the laser-repaired 316L stainless steel coating was dendritic and eutectic, and its physical phase was mainly composed of austenite, Fe-Cr, and carbides. The wear rate increases with the rise in the shaft weight load, indicating that the higher the contact load, the more severe the wear. In contrast, the influence of the friction coefficient in a room temperature environment is less variable. With an increase in the braking frequency, the wear of the specimen firstly rises and then decreases, and when the frequency is 1 Hz, the value of the wear rate at room temperature is the largest, and the wear surface appears as more peeling layers, and a large amount of wear debris is randomly distributed, which manifests as the wear mechanism being characterized by abrasive wear and adhesive wear. Therefore, different factors affect the wear level of the material differently, with the axle weight load having the greatest influence. The relevant results help to provide corresponding theoretical references for the optimization of parameters under the braking condition of the wheel tread, which ensures the normal operation of the braking system when driving.

1. Introduction

As a train speeds, the axle weights and capacity increase, and trains in the repeated acceleration, braking, bending, and occasionally complete sliding process of frictional heat generation will make the wheels produce a high thermal load [1]. This will lead to high temperatures up to 600 °C in the braking process between the brake tile and the tread surface, which could cause local damage to the surface of the wheel [2]. To ensure the safety of vehicles, the wheel surface damage found in daily maintenance is mainly removed through wheel turning; however, wheel turning leads to a rapid reduction in the wheel size, which greatly shortens the service life of the wheel, resulting in a huge waste of materials and economic losses [3]. Therefore, there is an urgent need to develop more efficient and economical train-wheel-repair technology.

The use of laser cladding technology in restoration is very extensive, and laser cladding is one of the laser processing techniques [4]. During the laser cladding process, the surface of the material is fused with different alloy powders using a high-energy beam and rapidly solidified to form a coating with a good forming quality [5]. In addition, unlike other metal surface technologies, the repair coatings prepared using laser cladding technology are characterized by a high laser beam energy density, low heat input, low thermal impact on the substrate, dense and fine grain structure, and good bonding with the substrate [6,7]. At the same time, different alloy cladding powders can be selected to achieve high wear resistance, high corrosion resistance, high strength, and other high-performance effects, for applications to complex and harsh environments. This technology has been widely used in aerospace, rail transport, petrochemical, machinery manufacturing, and other fields [8,9] and will become a new type of surface-modification technology.

In the field of the local damage repair of railway wheel rails, scholars have carried out much exploratory work, with most researchers focusing on the consideration of the preparation of the high-strength, wear-resistant performance of different alloy cladding layers on the rail wheel track material to enhance its service life and achieve significant results. Seo et al. [10] evaluated the wear characteristics of Stellite 21, Inconel 625, and Hastelloy C laser cladding layers and found that Hastelloy C is the most appropriate when considering both the wear and rolling contact fatigue (RCF) of the cladding boundary. Guo et al. [11], in the wheel substrate preparation of Co-based alloy coatings, compared and analyzed the coating’s wear resistance by using a rolling fatigue tester. The results showed that the wear rate of the repaired coating was only 1/2 that of the substrate, indicating that the laser cladding technology can repair the damage defects in the wheel and rail materials. Lewis et al. [12] investigated the composite coating of Co-based and martensitic stainless steel, and the fatigue characteristics of the specimens were significantly improved after laser cladding processing, and the wear of the wheel steel was also attenuated to a greater extent. Wang et al. [13] conducted in-depth research on the application of Fe-based laser cladding coatings on train wheels. After fatigue rolling tests, they found that the wear surface damage was relatively minor, the wear rate was low, and the fatigue resistance performance was the best. However, the surface hardness of the coatings prepared via laser cladding of the alloy powders used in the above studies is generally too large, such as the Fe-based alloy coating hardness of about 800 HV [14], which is significantly higher than the hardness of steel rails (about 300 HV), which is difficult to coordinate with the wheel substrate and accelerates the abrasion and damage of steel rails. In view of this, Zhu et al. [15] tried to choose 316L, 410, and 420, three kinds of stainless steel powders, on the surface of the local wheel damage laser cladding repair and found that the hardness of the three repair layers and the substrate is close to the hardness of the coating, or it runs for a period of time after the hardness of the coating and the substrate hardness tends to be coordinated and shows excellent friction wear and contact fatigue performance compared to the substrate, which indicates that stainless steel powder may be more suitable for local wheel damage repair.

Stainless steel powder is an important part of iron-based materials, and it has excellent corrosion resistance at the same time, both iron-based materials, low manufacturing costs, and good compatibility with a large number of steel materials used in industrial production, and it is thus favored by researchers [16,17].

It is well known that the railway train, in the process of braking, and the rail contact at the tread surface not only bear the compressive stress on the brake tile and the contact pressure under load but also bear the rapid rise in the surface temperature caused by thermal stress. The cyclic mechanical load and thermal stresses together cause the wheel tread to undergo thermal-mechanical fatigue damage. In order to verify the main factors affecting the cyclic dynamic fatigue damage of the tread, Moyar et al. [18] adopted the critical plane fatigue damage theory analyzing the change rule of the thermal stresses generated by the axle load and frequent starting/braking. Teimourimanesh et al. [19] confirmed that the friction occurring during the braking of the tread of the train affects the mode of heat transfer in the contact position among the wheel, brake pads, and steel rails. Therefore, the use of laser cladding technology for the repair of local damage to wheels should focus on the interaction between thermal and mechanical loads at the contact location under train-braking conditions and their effects on the tread of the wheel.

This work is intended to repair the 316L stainless steel coating through fusion cladding in the locally damaged notch of ER8 wheel material using laser additive repair technology. Under different service conditions (axle load and braking frequency), sliding friction wear tests were carried out on the repaired specimens using the ball and disc wear tester. An analysis, at room temperature and high temperature, of the repair coating structure, friction coefficient, wear rate, surface damage morphology and other evolutionary laws has important theoretical significance. Investigating the complex braking environment in the repair of the fusion cladding service performance, to improve the train wheel and rail system to withstand the extremely harsh environment, has important practical significance.

2. Materials and Methods

2.1. Materials and Specimen

The substrate material selected for the experiment was taken from the rim part of ER8 wheel steel, and its chemical composition is shown in

Table 1. Chemical composition of ER8 (wt.%).

Material	C	Si	Mn	P	S	Cr	Ni	Fe
ER8 wheel steel	0.52	0.26	0.73	0.016	0.002	0.25	≤0.30	Bal.

. Figure 1 shows the flow chart of specimen preparation; firstly, the substrate material was cut in the lower 2–3 mm of the tread surface using a wire cutting machine, and the arc groove notch was pre-machined in the middle position of the substrate surface (which represents the local damage defects). The width (W) of the groove to be repaired was 7 mm, the thickness (H) was 1.5 mm, the diameter of the arc length (Φ) was 15 mm, and the friction disc specimen of Φ53 mm × 4 mm was cut after completion of the cladding repair. Before laser cladding, the rust layer on the surface of the groove was first polished and cleaned using ordinary sandpaper, then cleaned using an ultrasonic wave with an acetone solution for about 15 min, and finally cleaned by water and dried before labeling as a sample for use. The cladding material was 316L stainless steel powder, and the specific composition is shown in

Table 2. Chemical composition of stainless steel powder.

Powder	Composition(wt.%)
	C	Si	Mn	P	Cu	Mo	S	Cr	Ni	Fe
316L stainless steel	<0.03	<0.75	<2.0	<0.025	<0.5	2.25–2.5	<0.01	17.5–18.0	12.5–13.0	Bal.

.

Before the test, the surface and interface of the sample were studied and analyzed via SEM and EDS. The samples’ elemental composition and the size and mass ratio of the impurity atoms were determined and are shown in Figure 1. The results show that the main elements in the interface region of 316L stainless steel are Fe, C, Cr, C, Si, and Mo, and the mass ratio of the O element is almost zero. This indicates that the surface finish of the repair coating is better, so the repaired specimen meets the test requirements.

The ZF-R6000-60 laser cladding system was integrated, and the experimental processing parameters were selected to optimize the process parameters: laser power 1500 W, scanning speed 500 mm/min, powder feeding rate 15 g/min, spot diameter 4 mm, overlap rate 50%. Alloy repair coatings of different materials were prepared at the groove position in the middle of the disk sample through synchronous powder feeding. The repair work was completed under inert gas (Ar) protection in the entire system. The schematic diagram of the repair process is shown in Figure 2.

2.2. Friction Test

In this experiment, the high-temperature wear tester (MS-HT1000, Huahui, Lanzhou, China) was used. The friction and wear diagram is shown in Figure 3, and the specific test parameters are shown in

Table 3. Friction and wear test parameters of the cladding layer under different service conditions.

Test Parameters	Value
Test duration (min)	30
Rotation radius (mm)	20
Contact stress (MPa)	695–1103
Temperature condition (°C)	25 (RT); 600 (HT)
Axle weight (N)	10; 20; 40
Braking frequency (Hz)	0.5; 1; 2

. The selected grinding pair was a silicon nitride ceramic ball (diameter of Φ6 mm, hardness of about 1700 HV, Ra ≤ 0.2 μm). The test time was 30 min, and the wear trace diameter Φ was 20 mm. Before the test, the previously polished surface was cleaned with an alcohol rag to remove the surface dust.

The contact between the grinding ball and the grinding disc was not purely rigid, but was an elastic–plastic contact. The maximum Hertzian contact stress σ (MPa) by the ball disc was calculated using the formula shown in Equation (1). Using a load of 10–40 N, the resulting Hertzian contact stress was 695–1103 MPa, and the range of this value was approximately equal to the contact stress between the wheel and the rail. It can be seen that a small load added to the grinding ball is sufficient to produce a large contact stress, which was used for the comparative study of friction and wear of different metallic materials, and the load was selected to be 10–40 N.

$${\mathsf{\sigma }}_{\max }=\sqrt[3]{\frac{3\mathit{Fr}}{4}\left(\frac{1-{\mu }_1^2}{E_1}+\frac{1-{\mu }_2^2}{E_2}\right)},$$

(1)

where F is the test force (N); r is the diameter of the ball (mm); μ₁ is Poisson’s ratio of the grinding ball; μ₂ is the Poisson’s ratio of the grinding disc material; E₁ is the elastic modulus of the grinding ball; E₂ is the elastic modulus of the disc material.

2.3. Characterization

In this test, the coating cross-section was qualitatively and quantitatively analyzed using a scanning electron microscope (SEM, SU8010, Hitachi, Tokyo, Japan) with an energy spectrometer (EDS), the prepared fusion-coated coatings were physically analyzed using an X-ray diffractometer (XRD-6000X, Shimadzu, Tokyo, Japan), and the hardness of the coatings was determined using a Vickers hardness tester (Qness 10A+, Wien, Austria) for the determination of the coating hardness. Different parts of the contour on the wear track were scanned three times using a 3D optical profilometer (ZeGage^TM Pro HR, ZYGO, CT, USA) to measure the wear track, and the average value was taken as the test result.

The volumetric wear rate is a key index to evaluate the friction and wear properties. The wear rate is calculated based on Archard’s law [20], and its calculation formula is as follows (2):

$$W=\frac{V}{F_{N}d},$$

(2)

where W is the wear rate (mm³/(N·m)); V is the wear volume before and after the test (mm³); F_N is the contact load (N); d is the total sliding displacement (m).

3. Results

3.1. Microstructure

Figure 4a shows the microstructure of the cross-section of the repair cladding layer of the 316L stainless steel. Under the microscope, it can be clearly seen that the morphology is divided into three different areas, namely the coating, the heat-affected zone, and the matrix. Among them, the microstructure of the coating surface is dense, with no pores and cracks and uniform grains. Figure 4b shows the line scan composition distribution at the interface between the 316L stainless steel coating and the interface, and the line scan direction is from the top of the coating to the substrate, so the size of the coating dilution rate can be characterized. It can be seen that there is no obvious abrupt change in the distribution curve of Fe elements in the fusion cladding layer, indicating that the Fe elements are uniformly distributed in the fusion cladding layer. At the interface, the Cr element shows a non-cliff transition distribution between the coating and substrate surfaces, because the concentration of the Cr element is different in the two regions, which leads to diffusion during the fusion-coating process. It can be further judged that the bonding mode between the substrate and the coating bonding region is metallurgical bonding, which improves the coating performance.

Selecting the cladding layer in the middle area for a high-magnification observation (Figure 5), it was found that the size of the coating grain structure is 7 to 10 μm and the cells are in the same crystallographic direction. According to Hunter’s crystal solidification theory, the temperature gradient (G) and solidification rate (R) are the most important indicators of structural characteristic changes [21,22], and the G/R ratio directly affects grain formation. During the preparation process, heat is dissipated through the external environment and the nearby solidified coating and substrate conduct heat. During the solidification process, the G/R value decreases, the cooling rate of the molten pool accelerates, the growth rate increases, and cellular crystals are easily generated, resulting in a highly refined grain size. In addition, the structure of the heat-affected zone is fine martensite, with a large number of strengthening phases dispersed in it. This is caused by the high temperature of the laser beam [23], forming an overheated structure, which has a quenching effect and can easily form a martensitic organization.

The elemental segregation behavior is strongly influenced by the solidification growth rate, and the ratio of the elemental content in the dendrite stem (C_s) to the inter-dendrite (C₀) is defined as the coefficient of segregation K; thus $K=\frac{C_s}{C_0}$, and the closer the value of K is to 1, the lighter the segregation [24]. The elemental EDS maps of the microstructure of the coating surface are given in Figure 5, and the chemical compositions of the dendritic core region determined via energy spectrometry (EDS) are shown in

Table 4. The chemical composition of the dendritic core region of the coating microstructure was determined via EDS (wt.%).

Element	Fe	Cr	Ni	C	Mo	Mn	Si
Dendrite core (Cs)	81.89	7.96	4.81	2.78	0.80	1.20	0.56
K = Cs/C0	1.10	0.73	0.71	0.85	0.44	0.69	0.62

, which reveals the distribution of the different elements. the Fe element, with a value of K > 1, is mainly distributed in the dry region of the dendrites, whereas the Cr, Ni, C, and Mo elements have a value of K < 1, which suggests that these elements are mainly enriched in the inter-dendritic region. The elemental content of Fe in dendrites is slightly higher than that of the internal cytosol crystals, while the content of C and Mo in interdendrites is relatively high. Both elements Cr and Mo very easily form carbides with C. However, elemental Mo has a stronger affinity with C, so that in interdendrites, Mo’s carbide is preferentially precipitated, and thus, the content of Mo and C is higher in interdendrites [25].

The degree of change in the composition of the cladding layer caused by the fusion of the cladding material into the cladding matrix is called the dilution rate. The size of the dilution rate directly affects the organization and properties of the cladding layer; too large a dilution rate will reduce the performance of the cladding material, while too small a dilution rate reflects the lack of metallurgical bonding between the substrate and the cladding material, which is prone to problems, such as spalling and cracking. The dilution rate D is calculated using Formula (3) [26]:

$$D=\frac{h}{H+h}\times 100\%$$

(3)

in Equation (3), H denotes the height of the fusion-coated layer and h denotes the depth of the melt pool. The measurements are illustrated in Figure 6, and the calculation based on Equation (3) shows that the coating dilution is 17.38%.

3.2. Phase Analysis and Microhardness

Figure 7 shows the X-ray diffraction spectrum of the repair coating on the 316L stainless steel. According to the phase composition analysis, the phase of the 316L coating is mainly composed of austenite (A), Fe-Cr, and carbides (M₂₃C₆, M represents Fe, Cr, etc.). The Ni element in the composition can expand the austenite phase area, forming stable austenite [27]; the Cr element content is high, and it can easily form Fe-Cr and other intermetallic compounds with the element Fe. At the same time, there are carbide substances in the phase. This is because, during the cladding process, the carbon and alloying elements enter the large dilution molten pool and solidify to form a small portion of the carbide phase [28], which helps to increase the hardness of the 316L stainless steel repair layer and achieve a strengthening effect.

Figure 8 demonstrates the microhardness distribution of the specimen, and it can be seen that the hardness of the repair coating is 372.5 HV_0.1, which indicates that the hardness of the metal surface has been enhanced after the laser fusion coating repair. According to the above micro-morphology and XRD analysis, the coating area shows a high hardness effect, which is due to the segregation of the elements in the molten pool at the grain boundaries during the cooling and solidification process, which produces a large number of solid solutions and high-strength carbide substances, thus providing a strengthening effect and giving the coating area a high microhardness.

3.3. Influence of Axle Weight

The axle load has an important influence on the friction coefficient and wear rate, so the influence of load conditions (10, 20, and 40 N) on the friction coefficient and wear rate was evaluated. Figure 9 shows the variation pattern in the friction coefficient under different contact loads at room temperature/high temperature. As shown in the figure, the friction coefficient curves of the three contact load conditions at a normal temperature all had an extreme value before the start of the test, with the value being 0, and then gradually increased to the mean value, and the increase was more obvious as the friction time increased. On the contrary, the above rules did not appear in the high-temperature test results. In the changing pattern at high temperatures, the growth trend for the friction coefficient under the three contact loads was relatively slow. Under 20 N and 40 N, fluctuations always existed during the friction process, among which the friction coefficient at 20 N fluctuated the most.

Figure 10 shows the comparison of the average friction coefficient and wear rate of the repaired specimens at room temperature/high temperature with different contact loads. It can be observed from Figure 10a that the average coefficient of friction of the repaired specimens at room temperature was obviously larger than the value of the high-temperature environmental test, and there was no significant difference in the average coefficient of friction of the repaired specimens at room temperature, with values of about 0.66. From Figure 10b, it is observed that the wear rate of the fused coating area of the specimen with 10 N at room temperature was 1.35 × 10⁻⁵ mm³/(N·m), and both of which were smaller than the corresponding values under 20 N and 40 N conditions (1.66 × 10⁻⁵ mm³/(N·m) and 2.36 × 10⁻⁵ mm³/(N·m), respectively), and it can be seen that with an increase in the contact load, the wear rate of the repaired specimen presented a gradual increase in the law of change, and the law in the high-temperature test was also the same, when the contact load was 10 N; at this time, the wear rate of the specimen reached the minimum value of 1.21 × 10⁻⁵ mm³/(N·m). From the comparison of the ambient temperature, the wear rate of the repaired specimen at a high temperature changed in the range of 1.2–1.8 × 10⁻⁵ mm³/(N·m), and for room temperature, the maximum wear was 2.36, indicating that room temperature has a great influence on the wear degree of the repaired specimens.

Figure 11 shows a comparison of the cross-sectional profiles of the wear traces of the repaired specimens under different axial heavy loads. The two-dimensional cross-sectional profile was taken from the middle profile of the wear zone of each specimen. According to the observation of three-dimensional morphology, it was found that under different contact loads, different wear conditions occurred at the interface of the repaired specimens, and the wear morphology of the coating and substrate areas became more serious as the load increased. This is because the increase in normal pressure increases the lateral friction and intensifies the mechanical wear at the interface. More and more accumulation of wear debris and spalling pits deteriorate the contact surface, thereby increasing the roughness of the material surface, and the friction and wear will continue to increase [29]. In addition, from the cross-sectional profile traces of the wear traces, it was observed that the surface wear depth of the coatings was deeper at room temperature than at high temperature. All the coating profiles showed different heights of “bumps”, but the number of “bumps” at room temperature was significantly higher than that at a high temperature and decreased with an increase in the contact load, mainly because the large contact load could better flatten the third-body abrasive chips and cause them to expand [30].

Figure 12 shows the scanning morphology of the wear area on the surface of the repaired specimen under different contact loads at room temperature/high temperature. It is clearly observed that after friction and wear tests with different contact loads, different wear conditions appeared at the interface of the sample. It is obvious that the wear width of the coating was narrower than that of the substrate, indicating that the coating has better wear resistance than the substrate under the same conditions. The laser cladding process has immediate heating and cooling characteristics that increase the nucleation rate and grain growth rate of the material. Compared to the substrate, the grains of the coating became finer, the organization was denser, and there was a fine austenitic organization on the surface of the fused cladding, which provides better toughness and strength. The coating zone is significantly more resistant to plastic deformation, which effectively reduces the phenomenon of fatigue cracks caused by contact stress and avoids spalling of the material. To further analyze the characteristics of the element composition and content in the wear track of repaired samples under different loads, the interface morphology of the sample and the distribution of each component element at the interface position were tested. The results are shown in

Table 5. EDS analysis results in the yellow selected areas in Figure 12 (wt.%).

Content	Fe	Cr	C	Ni	Si	O
Zone A	70.94	8.96	4.53	6.14	1.16	4.13
Zone B	72.70	8.15	3.93	5.19	1.19	5.07
Zone C	72.95	8.75	2.54	5.17	1.25	6.47
Zone D	66.00	6.65	3.06	4.16	0.36	12.29
Zone E	68.32	6.56	3.07	4.00	0.60	14.06
Zone F	65.25	7.98	2.23	4.80	0.40	16.31

. It can be seen from the table that the oxide layer at the high-temperature wear interface was enriched in a large amount of the O element, and the O element accounted for as high as 16.31% of the total mass fraction. Among them, the O element increased as the load increased. The increase in load and constant contact wear make the grinding surface easily oxidized, and the wear becomes more severe.

As can be seen from Figure 12, under the 10 N load condition, furrows and slight delamination were clearly visible on the worn surface of the coating. The main reason for this is that the 316L stainless steel coating has high hardness and a large tangential force on the contact position of the grinding surface, which causes the surface layer to peel off. At this time, the wear mechanism is abrasive wear and a small amount of adhesive wear. When the contact load is 20 N, it can be observed that the surface is severely damaged, with more adhesive delamination and abrasive wear debris remaining. It can be inferred that the wear mechanism is abrasive wear and oxidative wear. When the load increased to 40 N, the grinding surface experienced severe plastic deformation and some spalling pits appeared. It was analyzed that the wear mechanism at 40 N was the coupling effect of adhesive wear and abrasive wear. Therefore, the increase in the load will cause plastic deformation and peeling of the material, making the worn surface rough and causing serious wear. When the contact load is small, the coating wear surface undergoes gradual wear failure, and the surface layer is simultaneously subjected to load extrusion deformation and tangential force, forming wear debris. The wear debris during the sliding wear process can easily form three-body abrasive wear. At the same time, under the action of stress and frictional heat, the wear debris repairs the wear surface of the sample to form a third-body layer (tribochemical film) [31]. With the input of a larger axle load, the failure mechanism of the coating surface changes to fatigue fracture and peeling. Under large loads, repeated extrusion and adhesion cause the coating to peel off in sheets, and then, the wear debris fills the peeling pits, causing uneven hardness on the worn surface, leading to aggravation of adhesive wear and abrasive wear [32].

3.4. Influence of Braking Frequency

Figure 13 shows the changes in the friction coefficient of different test frequencies at room temperature/high temperature. It can be seen from the figure that the friction coefficient at room temperature first increased and decreased the final rise and tended to stabilize, and there was no above phenomenon in the comparison diagram of the high-temperature test results. When the test frequency was 2 Hz, the overall fluctuations of the friction coefficient were obvious, and over time, the increase in the friction coefficient was also small. When the test frequency was smaller than 1 Hz, the fluctuation of the friction coefficient curve was not obvious. At high temperatures, the friction coefficient at the frequency of the three tests was not large over time, and the overall fluctuations were consistent.

Figure 14 shows the average friction coefficient and wear rate comparison diagram of different test conditions at room temperature/high temperature. From the average friction coefficient values in Figure 14a, the friction coefficient at room temperature was generally higher than that at high temperature, and the larger the braking frequency was, the smaller the average friction coefficient at room temperature was, whereas the average friction coefficient at a high temperature underwent a small change and was almost unchanged, with a stable mean of 0.36, which indicates that the effect of the different test frequencies at high temperatures is small. From Figure 14b, it can be seen that the wear rates of the coated area under different test frequency conditions at room temperature were 1.28 × 10⁻⁵ mm³/(N·m), 1.66 × 10⁻⁵ mm³/(N·m), and 1.36 × 10⁻⁵ mm³/(N·m), respectively. While the wear rates at high temperature were 1.19 × 10⁻⁵ mm³/(N·m), 1.44 × 10⁻⁵ mm³/(N·m), and 1.30 × 10⁻⁵ mm³/(N·m), it is clear that at both temperatures, the wear rate increased and then decreased with the braking frequency. Among them, at room temperature, there was a large increment in the wear rate when increasing from 0.5 Hz to 1 Hz conditions.

Figure 15 shows the comparison diagram of the cross-section of the abrasion marks of the sample under different brake frequencies. The two-dimensional cross-section contour was taken from the outline of the middle part of each sample. The 3D morphology shows that the substrate was more worn than the coating in the wear track area. The fine grain organization in the fusion cladding layer promotes the increase in grain boundaries in the unit area, which can effectively impede the dislocation movement of the material in the wear process. From the Hall–Petch theory, it can be seen that the smaller the grain size, the increase in the microhardness of the material can, to a certain extent, improve the performance of the material resistance to ploughing [33]. The reduction in the grain size and the increase in hardness can effectively enhance the wear resistance of the material. The two-dimensional contour abrasion trace diagram shows the bulge of the abrasive debris squeezed out to the sides at the three test frequencies, which ultimately creates a bulge in the edge region of the abrasion trace under high-temperature conditions, and as the test frequency increases, the two sides of the peak become lower. The increase in the speed of movement makes the specimen more prone to jumping and impact phenomena. Therefore, the abrasive chips generated by friction are squeezed and then weakened by the shear effect on both sides of the peaks.

Figure 16 shows the scanned morphology of the surface area of the repaired specimen at different test frequencies at room temperature/high temperature. It can be observed that regardless of whether it was a room temperature or high-temperature environment, at a test frequency of 1 Hz, the abrasion width of the coating and the substrate area was approximately the same. In the case of 2 Hz, the size of the abrasion width of the coating and the substrate was significantly different. The greater the relative motion speed of the mating pair, the more intense the phenomenon of beating and impact between the specimen and the Si₃N₄ ceramic ball and the frequency of the occurrence of the phenomenon, which has a greater impact on the interface state of the contact pair in the friction and wear process. To further analyze the characteristics of the element composition and content in the wear track of the repaired samples under different loads, the interface morphology of the sample and the distribution of each component element at the interface position were tested. The results are shown in

Table 6. EDS analysis results in the yellow selected areas in Figure 16 (wt.%).

Content	Fe	Cr	C	Ni	Si	O
Zone G	77.69	7.84	2.36	4.74	0.96	5.50
Zone H	72.70	8.15	3.93	5.19	1.19	3.07
Zone I	74.44	9.79	2.12	5.86	0.93	5.69
Zone J	68.34	5.16	2.30	3.00	0.44	17.71
Zone K	68.32	6.56	3.07	4.00	0.60	14.06
Zone L	64.01	5.65	3.52	3.77	0.47	19.36

. It can be seen from the table that at high temperatures, the amount of the O element in the interface area was about three times higher than that at room temperature. At both temperatures, the O element content first decreased and then increased. Combined with the aforementioned change in the wear rate, this indicates that the more oxygen there is, the more oxides there are, helping to form an oxide layer to act as a protective film and reduce wear.

It can be seen from Figure 16 that the peeling phenomenon and abrasion marks appeared on the surface of the coating at a low frequency. This is due to the decrease in the frequency, corresponding to the decrease in speed, leading to an increase in the contact time between the convex micro-bodies of the bonding material, as the micro-convex bodies are stretched for a longer time during frictional wear. This process is reflected by longer peeling traces. When the braking frequency increases, corresponding to the increase in speed, it results in a shift from long and shallow to short and deep adhesion spalling pits. The increase in speed also leads to a reduction in the contact time between the micro-convex bodies, which are subjected to instantaneous fracture under the influence of shear. Abrasive debris is generated, which crushes the surface under contact loads and forms a stripping layer [34]. At the same time, the continued increase in the test frequency will also lead to an increase in the surface temperature. Asperities are more likely to tear off (a type of adhesive wear) under the action of friction, resulting in deep and small adhesion pits. When the test frequency is 0.5 Hz, the coating wear surface appears as peeling and abrasion marks, and its wear mechanism is abrasive wear; when the frequency is 1 Hz, the abrasive surface appears as more peeling layers and a large amount of randomly distributed wear debris, and its wear mechanism is abrasive wear and adhesive wear. As the test frequency was increased to 2 Hz, a large number of flaking pits and oxides appeared on the wear surface, and the wear mechanism was the coupling effect of adhesive wear and abrasive wear. With an increasing frequency, the surface temperature and adhesive wear increase, and the detached flaking pits are crushed to form an oxide layer under the load. These oxides act as a lubricating film between the abrasive material and the Si₃N₄ ceramic ball, helping to reduce wear rates [35].

4. Conclusions

(1)

The microstructure on the surface of the 316L stainless steel cladding repair coating is dense and uniform, without defects, such as pores and cracks. The cladding layer is composed of uniform cellular crystals, and the crystalline phase is mainly composed of austenite, Fe-Cr, and carbide.

(2)

As the heavy load of the contact shaft increases, the impact on the friction coefficient under a room temperature environment changes slightly. Regardless of whether it is room temperature or high temperature, the wear rate of the repaired sample gradually increases, with the maximum values in the coating and substrate areas being approximately 2.36 × 10⁻⁵ mm³/(N·m) and 6.23 × 10⁻⁵ mm³/(N·m), respectively. Increasing the load from 10 N to 40 N, the coating wear surface switches from slight delamination and plowing at the beginning to severe plastic deformation with some flaking pits on the surface, the wear becomes severe, and the wear mechanism changes from abrasive wear with a small amount of adhesive wear to the joint coupling of adhesive wear and abrasive wear.

(3)

As the braking frequency increases, the impact on the friction coefficient in high-temperature environments changes slightly. In the two temperature environments, the wear rates of the repaired samples show a pattern of first increasing and then decreasing. The lowest wear rates of the coating area at room temperature and high temperature were 1.28 × 10⁻⁵ mm³/(N·m) and 1.19 × 10⁻⁵ mm³/(N·m), respectively. When the frequency is 0.5 Hz, peeling and wear marks appear on the worn surface of the repair coating area, which is manifested as abrasive wear. At a frequency of 1 Hz, the worn surface has a peeling layer and a large amount of wear debris, and the wear mechanism is abrasive wear and adhesive wear. When the test frequency is increased to 2 Hz, the grinding surface shows many peeling pits and some oxides, which together show the characteristics of adhesive wear and severe abrasive wear.

(4)

The axle load and braking frequency conditions of the train wheel tread in service have a certain impact on the friction and wear performance of the local repair cladding layer.