Objective To improve the wear resistance and corrosion resistance of ER8 high-speed wheel steel and extend the service life of the wheel. Methods Two coatings, Fe-based alloy powder and Co-based alloy powder, were prepared on the surface of ER8 wheel steel. The metallographic structure, phase type and nanohardness of the coatings were analyzed by SEM, XRD, nanoindenter and other characterization equipment. The samples were placed in acid rain solution for friction and wear tests and electrochemical corrosion tests using MFT-EC4000 reciprocating electrochemical friction and wear tester. Results The coating surface structure was dense and uniform, forming a good metallurgical bond. The Fe-based and Co-based alloy coatings showed “cellular” and “honeycomb” shapes, respectively, without obvious defects such as holes and cracks. The substrate underwent slight abrasive wear at low frequency (1 Hz), and severe spalling and pitting occurred at medium and high frequencies (2, 4 Hz). The wear mechanism was mainly adhesive wear, oxidative wear and abrasive wear, and there was no obvious corrosion and spalling in the wear area of the coating. At high frequency, the wear rates of Fe-based coating and Co-based coating are reduced by 46.10×10’–5 mm’3/(N·m) and 39.85×10’–5 mm’3/(N·m) respectively compared with the substrate. At the same time, the impedance value of the coating is significantly improved. The polarization curve test results show that the self-corrosion potentials of Fe-based coating, Co-based coating and substrate are –0.522, –0.381, –0.603 V, respectively, and the corrosion density is 3.916, 0.312, 5.483 μA/cm2, respectively. Conclusion The wear resistance and corrosion resistance of the repaired wheel steel samples are improved to varying degrees. In comparison, the Fe-based alloy coating is more excellent in wear resistance; in terms of corrosion resistance, the Co-based alloy coating is slightly stronger than the Fe-based coating.
The rapid increase in the mileage of high-speed railways and the number of EMUs in my country is accompanied by a rapid increase in operating and maintenance costs, part of which comes from the maintenance and replacement of high-speed train wheels. Generally, wheel damage is mainly caused by wear, rolling contact fatigue and complex service environment, among which the harsh acid rain environment is a typical example. If wheel damage is not discovered and treated in time, it will affect the riding experience at the least, and may even lead to catastrophic consequences such as derailment, resulting in heavy casualties and huge property losses. According to statistics, the annual steel loss caused by corrosion and rust accounts for about 10% to 20% of the annual steel production. The economic loss caused by corrosion in my country is as high as more than 230 billion yuan each year [1]. The severe acidic environment will further aggravate the occurrence of corrosion. The commonly used turning repair process can remove wheel surface damage and restore the wheel to its usable state, but it will quickly reduce the wheel size, resulting in a significant shortening of the wheel service life. Therefore, in-situ repair of wheel damage is a more ideal repair process.
Due to its corrosion resistance, wear resistance, heat resistance and oxidation resistance, laser cladding is widely used in aerospace, mining machinery, automobile, petrochemical, railway, power and other industries [2-3]. Compared with surface modification technologies such as submerged arc welding, plasma welding, and gas-electrode welding, laser cladding technology has the characteristics of low dilution rate, low coating porosity, high bonding strength with the base material, wide material selectivity, and no environmental pollution [4]. At present, some scholars have tried to apply laser cladding to the surface strengthening of train wheels and rails to improve their wear resistance, rolling contact fatigue resistance, and corrosion resistance, and have done a lot of research in this regard [5-6]. Lewis et al. [7] clad a Co-based alloy coating and a martensitic stainless steel coating on a full-size R260 rail and tested them. The results showed that after cladding treatment, the fatigue strength and wear resistance of the rail were significantly improved. Gamon et al. [8] used laser cladding technology to prepare manganese bronze and aluminum bronze coatings on railway buffer heads and found that the cladding layer could significantly reduce the sliding wear of the buffer head. Wang et al. [9] conducted relevant tests on the wear of heavy-load wheel-rail materials after cladding. The results showed that the surface of the wheel-rail material before and after laser cladding treatment changed from obvious adhesive wear and peeling damage to plowing trend. Mu Xinpeng et al. [10] used a CO2 multi-simulation laser to prepare Co-based and Fe-based materials on the surface of wheel-rail materials, and conducted rolling friction and wear tests. The experimental results showed that the wear rate of the sample was significantly reduced after laser cladding treatment. Zhu et al. [11] used rolling contact tests to study the wear and rolling contact fatigue behavior of the local cladding stainless steel coating on the wheel, and found that the wear resistance and rolling contact fatigue performance of the coating were better than those of the substrate. It can be seen that current research mainly focuses on using laser cladding technology to strengthen the surface of wheels and rails, and research on laser cladding repair of wheels still needs to be deepened. This paper examines the wear resistance and corrosion resistance of the repaired wheel steel, and compares and evaluates the consistency of the tribological properties of the interface area and the substrate. The research results can provide certain technical guidance for the engineering application of laser cladding in wheel damage repair. Due to the acid rain environment and the variability of train running speed, the systematic research on train wheels facing harsh service environments is still insufficient. In this paper, Fe-based and Co-based alloy local repair coatings widely used in the field of laser cladding are prepared on the surface of ER8 wheel steel using traditional laser cladding technology. This is mainly because Fe-based and Co-based alloy powders are self-soluble alloy powders with deoxidation and slag-making capabilities, and are simple to prepare and have excellent performance. Using acid rain solution as the third medium, the sliding friction and wear test of wheel samples in acid rain environment was simulated at different running speeds using the MFT-EC4000 reciprocating electrochemical friction and wear tester. At the same time, electrochemical corrosion tests were carried out in acid rain solution. By analyzing the metallographic structure, microhardness, wear morphology, electrochemical corrosion, etc. of the coating, the damage mechanism and electrochemical corrosion behavior of the substrate, Fe-based and Co-based alloy coatings in the sliding friction and wear process under acid rain environment were revealed.
1 Experiment
1.1 Sample preparation
A number of samples with a size of 30 mm × 20 mm × 5 mm were cut from 5 mm below the tread of the train wheel (ER8 wheel steel) by wire cutting, and a circular arc notch with a width of 10 mm and a maximum depth of 2 mm was cut and removed in the middle of the sample, as shown in Figure 1. Under the conditions of laser power of 1 600 W, spot diameter of 4 mm, scanning rate of 7.5 mm/s, overlap rate of 50%, and preheating of the wheel steel substrate at 200 °C, the circular arc notch was repaired by laser cladding technology, and the working surface was polished step by step to 2 000# sandpaper with metallographic sandpaper, and ultrasonic cleaning was performed after grinding and polishing, and then dried, sealed and stored for later use. The chemical composition of the wheel steel is shown in Table 1, and the chemical composition of the Fe-based and Co-based alloy powders is shown in Table 2 and Table 3 respectively.
1.2 Materials and methods
The composition and content of the acid rain solution used in the experiment are shown in Table 4. The corresponding chemical reagents are scientifically and rationally selected in strict accordance with the composition and content in the table. An electronic balance with an accuracy of 0.005 g is used to weigh 17.76 g (NH4)2SO4, 10.00 g NaNO3, 1.50 g Ca(NO3)2.6H2O,
1.00 g NaCl, 1.00 g KNO3, and 0.96 g Mg (SO)4. The error is specified to be 0.01 g. The weighed drugs are placed in a 500 mL beaker, 500 mL of deionized water is added, and stirred with a glass rod. After the solution is dissolved, it is moved to a 5 L container and the same deionized water is added until
5 L. The pH value of the solution is controlled in the range of 3.1~3.5. If it does not meet the standard, glacial acetic acid and NaOH solid can be used for adjustment.
In the friction and wear test, the test sample was placed on the test bench of the MFT EC4000 tester, the fixed load was 10 N, the flow rate of the acid rain solution was 4.5 mL/min, the companion specimen was a φ6 mm Si3N4 sphere, the wear scar length was 5 mm, and the friction stroke of the sphere was adjusted to be just divided equally by the coating and the substrate. Different sliding friction frequencies in the acid rain environment were set to low frequency (1 Hz), medium frequency (2 Hz) and high frequency (4 Hz), and the friction time was 90 min. The hardness of the sample surface was measured using a Vickers hardness tester (Qness Q60). After the test, the results were characterized and analyzed with the help of characterization equipment such as SEM, XRD and OM. To ensure the reliability of the test results, all tests were repeated twice.
In the electrochemical corrosion test, the polished sample and the prepared acid rain solution were taken out for use. The analysis equipment used was an electrochemical wear instrument workstation. The polarization curve and AC impedance of the sample were tested in an acid rain solution with a pH value of 3.5. The traditional three-electrode system was used, with the coating and substrate samples as working electrodes, the Pt electrode as the auxiliary electrode, and the saturated calomel electrode as the reference electrode (filled with saturated KCl solution). First, the sample was immersed in the acid rain solution for a period of time to stabilize the open circuit. The scanning range of the polarization curve was –1.5~1 V relative to the open circuit potential, and the scanning rate was 1 mV/s. The corrosion parameters were obtained from the polarization curve. After the open circuit was stabilized, the electrochemical impedance spectroscopy of the sample was measured with a scanning frequency of 10’5~0.01 Hz. All tests were carried out at room temperature (26±1) ℃. At room temperature, the electrochemical impedance spectroscopy (EIS) and polarization curve tests were carried out to study the corrosion behavior of the coating and substrate samples in the acid rain solution.
2 Results and discussion
2.1 Metallographic structure and microhardness of wheel samples
The microstructures of the two different alloy coatings are shown in Figure 2. It can be seen that the microstructure of the Fe-based alloy coating (see Figure 2a) and the Co-based alloy coating (see Figure 2b) is quite different. This is because the temperature gradient (G) and the solidification rate (R) play a decisive role in the microstructure, and the shape control factor (G/R) value has a significant effect on the microstructure of the coating [12-13]. No obvious pores or cracks were found on the surface of the coating. The microstructure is uniform and dense, and the substrate and the coating form a good metallurgical bond. The alloy coating is composed of dendrites and eutectic structures. During the laser cladding process, when the laser beam leaves the molten pool, the surface temperature of the coating drops sharply, and the alloy powder quickly melts with the substrate, resulting in the formation of dendrites. The dendrites are closely arranged, and a large number of dispersed substances appear on the surface of the coating. These substances are solid alloys formed by the re-nucleation and solidification of the unsolidified liquid metal when the laser beam leaves the molten pool, that is, the eutectic structure. The solidification rate of the liquid metal determines the morphological characteristics of the eutectic structure. Compared with the “cellular” Fe-based alloy coating structure, the “honeycomb” Co-based alloy coating structure is more compact. In addition, the heat-affected zone structure of the two alloy coatings shows fine-flaked troostite, which is caused by the high temperature of the laser beam [14].
The XRD spectra of the two alloy coatings are shown in Figure 3. It can be clearly seen from the analysis of Jade6.5 software that the diffraction peaks of different coatings are quite different. The Fe-based alloy coating phase is mainly composed of austenite and carbide Cr7C3, in which Ni element is dissolved in austenite to form (Fe, Ni) solid solution. At the same time, the high content of Cr element in the powder is easy to react with C at high temperature to form carbide Cr7C3 [15], and these carbides can effectively improve the hardness and strength of the material. The Co-based alloy coating is mainly composed of γ-Co phase and Cr23C6. Similarly, at high temperatures, the high Cr content in the Co-based alloy powder is easy to form carbide Cr23C6 with C. The cross-sectional hardness of the two cladding coatings and the substrate is shown in Figure 4. The thickness of the cladding coating is about 1 mm. The hardness of the coating is significantly improved compared with the substrate, and the hardness of the Fe-based alloy coating is much higher than that of the Co-based alloy coating, with an average hardness of 711.4HV. During the coating formation process, the rapid cooling of the molten pool liquid will cause solid solution strengthening, which significantly improves the hardness of the coating [10]. The hardness of the interface area decreases rapidly with the increase of the cross-sectional depth until it decreases to about 300.0HV and then remains stable, close to the average hardness value of the substrate (287.0HV). This is because the deeper the depth of the cross-section, the less energy the material absorbs, so that there is not enough energy to cause the structure to undergo phase transformation. This also shows that the structural organization of the interface area is closer to that of the substrate. The increase in the thickness of the Co-based alloy coating and the heat-affected zone did not have a significant effect on the hardness of the Co-based coating, and the hardness of the coating cross-section was close to that of the substrate.
2.2 Friction coefficient
The experimental friction coefficient curves of Fe-based and Co-based alloy coatings under different sliding frequency conditions are shown in Figure 5. It can be seen that the friction coefficient is inversely proportional to the sliding speed, and is mainly divided into a violent friction stage and a stable friction stage. Since the sample surface is smooth and has an adsorption film before the test [16], the friction coefficient fluctuates greatly in the initial stage when the friction pair and the sample surface just contact. As time goes by, the surface and the adsorption film will gradually be damaged to varying degrees during the sliding process, and the sample surface will gradually become rough, resulting in a sharp increase in the friction coefficient. At the same time, the friction heat generated during the friction process will further increase the friction coefficient. After running-in, the surface of the wheel steel sample reaches a stable wear state, and the friction coefficient tends to be stable. With the increase of sliding frequency, the friction coefficient gradually decreases, mainly because the higher the sliding speed, the more oxides are produced on the friction surface, and the oxides can play a certain role in lubrication and wear reduction, thereby reducing the friction coefficient [17].
2.3 Wear scar morphology
Figures 6 and 7 show the surface wear morphology SEM images and OM images of the coating and substrate bonding area under different sliding speeds. From Figures 6 and 7, it can be seen that the degree of wear increases significantly from the interface area to the substrate position, and the damage morphology of the friction surface generally changes from obvious adhesive wear and severe peeling damage to abrasive wear. From the wear morphology of the interface area of the Fe-based alloy coating at sliding frequencies of 1, 2, and 4 Hz in Figures 6 and 7, it can be seen that the wear morphology between the coating and the substrate is quite different. When the sliding frequency is 1 Hz, the friction surface of the Fe-based coating is smooth in the acid rain environment, and a small amount of wear debris is found to adhere to the contact surface, which mainly manifests as slight abrasive wear. With the increase of sliding speed, the furrows become more obvious and the wear debris particles become more. The direct result is that the black oxide on the grinding surface increases significantly and the width of the wear mark increases. From the wear morphology of the interface area of the Co-based alloy coating at sliding frequencies of 1, 2, and 4 Hz in Figures 6 and 7, it can be seen that its friction and wear mechanism is similar to that of the Fe-based alloy coating, but the pitting and oxidation degree are more minor. Compared with the two coatings, the corrosion and wear of the substrate are more serious. Obvious spalling pits, oxide layers, and furrows appear on the friction surface, indicating that the wear between the wheel materials is mainly caused by three wear mechanisms: adhesive wear, oxidative wear, and abrasive wear. This phenomenon becomes more serious with the increase of sliding speed. This is because during the sliding friction process, the surface material of the sample breaks and peels off from the surface of the sample to form hard particles. These furrows are formed by the relative sliding between the particles with higher hardness and the surface of the sample. At the same time, repeated extrusion deformation causes the sample to eventually peel off at the part pitted by the acid rain solution under the combined action of friction, wear and shear force, forming spalling pits. The reciprocating friction between the friction pairs generates high friction heat. Under the combined action of friction heat and acidic environment, black oxides are formed on the grinding surface. In addition, the chloride ions and some acidic ions in the acid rain solution will destroy the formation of the oxide film on the surface of the sample, and penetrate into the depth of the sample along the microcracks, promoting the formation of pitting corrosion and accelerating the corrosion rate [18]. From Figures 6 and 7, it can be observed that the two ends of the dividing line present two completely different friction and wear morphologies. The damage degree of the substrate is much higher than that of the coating, while the wear morphology of the interface area is more inclined to the substrate part, which shows that the tribological properties of the interface area are similar to those of the substrate to a certain extent.
Take the EDS plane scanning analysis results at a sliding frequency of 2 Hz as an example, as shown in Figure 8. The distribution of Cr elements between the coating and the substrate is very different. A small amount of Cr is detected on the substrate because the substrate carries a small amount of coating wear debris during friction. The O element is mainly concentrated in the peeling pits and the black oxide layer, while the O element on the coating surface is evenly distributed, and no concentration phenomenon is found, which just shows the high corrosion resistance and high oxidation resistance of the coating.
2.4 Wear rate
Figures 9 and 10 are the 3D optical profile morphology of the wear scar after the sliding friction test and the two-dimensional profile curves of the wear scar of the coating and the ER8 substrate at different frequencies, respectively. The wear scar depth and width of the substrate at the three frequencies vary significantly. At 1 Hz, the wear scar depth and width of the substrate are approximately 5.26 μm and 540 μm, respectively. With the increase of frequency, the wear scar depth and width also increase. At 4 Hz, the wear scar depth and width are approximately 43.54 μm and 1 100 μm, respectively, showing the poor wear resistance of medium and high carbon steel. The wear scar depth and width of the coating at the three frequencies are slightly improved compared with the substrate. The wear scar depth of the Fe-based alloy coating at a sliding frequency of 1 Hz is approximately 4.1 μm, and the wear scar depth at a frequency of 4 Hz increases to 10.3 μm. Under the same conditions, the wear scar depth and width of the Co-based alloy coating are slightly greater than those of the Fe-based alloy coating, but are both less than the depth and width of the substrate material. Figure 11 shows the wear rate of the ER8 wheel steel substrate and coating at different frequencies, and its calculation formula is shown in formula (2).
Where: k is the wear rate, mm’3/(N·m); V is the wear volume, cm3; S is the sliding single-stroke length, cm; n is the number of cycles; W is the load, N. The substrate material shows a high wear rate at different frequencies and the wear rate is proportional to the frequency. The wear rate of the cladding coating under low and medium frequency conditions is slightly lower than that of the substrate material. With the increase of speed, the wear rate of the coating remains basically unchanged, while the wear rate of the substrate increases significantly. Generally, the wear rate of the cladding coating is proportional to the sliding speed. This is because with the increase of sliding speed, it is easy to have a significant friction heat effect at the interface, which leads to the rupture of the coating, the loss of protective effect, the formation of more serious peeling, and the increase of wear rate [20]. However, due to the different number of cycles n in this study, there is a phenomenon that the coating wear rate decreases with the increase of frequency. In addition, the wear rate of the substrate sample increased significantly under high frequency. The reason for this may be that during the high-speed reciprocating friction process, the substrate material was partially broken due to the excessive sliding speed, and the broken and detached parts formed hard wear debris, which increased the wear amount.
2.5 Electrochemical corrosion
The electrochemical impedance spectra of the three samples are shown in Figure 12. The impedance spectrum shows a semicircular shape, and the semicircles are all capacitive. The diameter of the semicircle directly reflects the impedance size of the sample surface. It can be clearly seen that the Co-based coating has the largest diameter, followed by the Fe-based coating, and the smallest radius is the substrate material. According to the characteristics of electrochemical corrosion, the equivalent circuit shown in Figure 13 is used to fit the impedance test results of the substrate and coating, where Figure 13a and Figure 13b are the equivalent circuits of the substrate and coating, respectively. Phase elements are represented by Q and n, where CPEbl and CPEdl are constant phase elements related to the substrate interface capacitance and double electric layer capacitance, respectively. CPE is a constant phase element of non-ideal capacitance formed due to system reasons. It is often introduced as a pure capacitor in simulation, so that the experimental data and simulation data are well matched [21-23]. EIS measurement is usually analyzed by analogy with the equivalent circuit (EEC). In Figure 13a, Rs is the resistance of the acid rain solution, and Rct is the resistance when the solution-substrate interface is connected in parallel with CPEdl; in Figure 13b, Rs is the resistance of the acid rain solution, Rb is the resistance of the corrosion product film, and Rct is the barrier layer resistance of the coating parallel to the constant phase element CPEbl. The larger the R value, the stronger the corrosion resistance of the material. ZSimpWin software was used to obtain the EIS data of the substrate and coating, and the fitting data are shown in Table 5. Combining the data analysis of Figure 12 and Table 5, the impedance value of the substrate is the smallest among the three samples, and the corrosion resistance is the lowest. The Co-based coating has the largest impedance value, followed by the Fe-based coating, but the difference is not large. It is preliminarily determined that the corrosion resistance of the cladding coating is significantly better than that of the substrate.
The potentiodynamic polarization curves of the ER8 wheel steel substrate, Fe-based and Co-based alloy coatings were tested using an electrochemical workstation, as shown in Figure 14. The potentiodynamic polarization curves of the three samples in acid rain solution were analyzed using the Tafel method and linear polarization method, and parameters such as corrosion potential and corrosion current density were determined to characterize the corrosion resistance of the cladding coating in acid rain solution [24]. The electrochemical corrosion parameters are shown in Table 6, and the polarization resistance RP is calculated using the existing parameters. The polarization resistance formula is shown in formula (2) [25].
Where: Ecorr is the corrosion potential; Jcorr is the corrosion current density; βc is the slope of the anodic polarization curve; βa is the slope of the cathodic polarization curve. Combining Figure 14 and Table 6, it can be seen that no obvious passivation area appears on the ER8 wheel steel substrate and the cladding coating in the acid rain solution, indicating that no passivation film is formed on the substrate and coating surface under the condition of open circuit potential of -1.5~1 V. At the same time, it can be found that the Ecorr and Jcorr of the substrate are significantly lower than those of the two coatings, indicating that in the acid rain solution, the corrosion tendency and corrosion rate of the cladding coating are much lower than those of the substrate [26]. With the increase of corrosion aging time, the corrosion potential Ecorr of the alloy coating shows a trend of moving in the positive direction, the corrosion current density Jcorr tends to decrease, and the polarization resistance RP increases. According to Faraday Ohm’s law [27], the electrochemical corrosion rate is proportional to the corrosion current density Jcorr and inversely proportional to the polarization resistance RP. This shows that the larger the corrosion current density Jcorr of the coating, the faster the corrosion rate of the coating and the weaker the corrosion resistance, while the larger the polarization resistance RP of the coating, the better the corrosion resistance of the coating. At the same time, the difference in corrosion resistance between Fe-based alloy coating and Co-based alloy coating is not obvious, mainly because the content of Cr element in the two alloy coatings is not significantly different, and the presence of Cr element can effectively improve the corrosion resistance of the material.
3 Conclusions
In this paper, local Fe-based and Co-based alloy coatings were prepared in ER8 wheel steel material by laser cladding technology, and the friction, wear and corrosion mechanism of the coating in acid rain solution were studied by reciprocating friction and wear and electrochemical corrosion tests. The following conclusions can be drawn:
1) The friction coefficient is inversely proportional to the sliding speed, and under the same parameter conditions, the friction coefficient of the Co-based alloy coating is slightly smaller than that of the Fe-based alloy coating.
2) The damage morphology on the grinding surface of the substrate material in the acid rain solution generally changes from obvious adhesive wear and severe peeling damage to the coating abrasive wear mechanism, while the coating grinding surface is slightly abrasive wear.
3) Compared with the substrate, both alloy coatings have relatively low wear rates, and with the increase of sliding speed, the wear resistance of the coatings is more excellent; under high frequency (4 Hz) conditions, the wear rates of Fe-based and Co-based alloy coatings can be reduced by up to 46.10× 10’–5 mm’3/(N·m) and 39.85×10’–5 mm’3/(N·m) respectively compared with the substrate.
4) The polarization resistances of the substrate, Fe-based alloy coating and Co-based alloy coating are 511.4, 7 148.8 and 30 984.4 Ω respectively, so the coatings show excellent corrosion resistance. In comparison, the Co-based alloy coating is better, but because the Fe-based alloy coating has a lower cost and shows better wear resistance, the corrosion resistance is not significantly different from that of the Co-based alloy coating. In practical engineering applications, the Fe-based alloy coating is still the preferred choice.
