2022 年 63 巻 8 号 p. 1151-1158
Although plasma electrolytic oxidation (PEO) coatings possess high hardness and good adhesion, they usually contain lots of pores and exhibit high friction coefficient. In this work, MoS2 was synthesized via hydrothermal reaction to fill the pores of PEO coatings and improve their tribological performance. Microstructure observation of coatings shows that the porous PEO coating formed on the titanium was basically composed of Al2TiO5, and MoS2 in-situ nucleated and grew on the PEO surface and in the PEO pores. The fabricated PEO/MoS2 composite coating had a dense surface and consisted of two layers: the upper MoS2 surface layer and the under PEO-MoS2 interlocking layer. The tribological tests of coatings revealed that the friction coefficient of the PEO/MoS2 composite coating (below 0.2) was much lower than that of the PEO coating (up to 0.6). The PEO coating failed after a short sliding distance of about 56 m under the normal load of 10 N. In contrast, the PEO/MoS2 composite coating reached a much longer sliding distance up to 1000 m without failure. The excellent tribological performance of the PEO/MoS2 composite coating was attributed to the synergistic action of PEO and MoS2 components.
Fig. 8 Schematic diagrams of the friction and wear behavior of the PEO/MoS2 composite coating.
Titanium and its alloys are widely used in automotive, aerospace and shipping industry due to their high specific strength, low density and excellent corrosion resistance.1,2) However, the low surface hardness and poor anti-wear property limited the tribological applications of titanium alloys. Plasma electrolytic oxidation (PEO) technology can form a hard and adherent oxide coating on the surface of titanium alloy.3–6) But the formed PEO ceramic coating contains numerous micro-pores caused by micro-plasma discharges, leading to high friction coefficient.7,8) Therefore, filling the PEO pores with solid lubricants and fabricating the self-lubricating PEO composite coatings are getting more and more attention.9–12)
Currently, there are two major methods to fill the PEO pores with solid lubricants.13–16) One method is to decrease the porosity of PEO coatings by performing the PEO treatment in the electrolyte containing solid lubricant particles. For example, F.C. Chang et al.13) performed the plasma electrolytic oxidation of Ti6Al4V alloy in a MoS2-dispersed phosphate electrolyte and a part of MoS2 particles embedded in the PEO pores, resulting in decreased friction coefficient from about 0.7 to 0.5. Similarly, plasma electrolytic oxidation of Ti6Al4V alloy in a BN-dispersed electrolyte produced a composite coating with lower porosity and friction coefficient.14) However, it is worth noting that despite the decreased porosity, this method cannot eliminate the pore structure completely due to the constant occurring of micro-plasma discharges during the PEO treatment. Furthermore, it is difficult to control the incorporation amount of solid particles in the PEO coating with this method.
Another method is to introduce solid lubricants into the PEO pores using the post-PEO techniques such as burnishing process, magnetron sputtering and spray deposition. A. Shirani et al.15) applied a graphite–MoS2–Sb2O3 chameleon coating to the surface of the PEO coating by the burnishing process. Although the thermo-mechanical stimulus during high-temperature applications of the PEO-chameleon composite coating can help the chameleon to fill the PEO pores, the burnishing process is not very suitable for the workpieces with complex shape. Y.Y. Li et al.16) deposited the diamond-like carbon (DLC) on the PEO coating with the magnetron sputtering technology. Results show that the deposited DLC mainly presented on the PEO surface and very few in the PEO pores. This is because the magnetron sputtering technology is difficult to transport solid lubricants deep into tortuous PEO pores due to its directional nature. Y.M. Wang et al.17) adopted the spray technology to deposit the graphite on the PEO coating. Results show that the sprayed graphite film covered the PEO pores and the fabricated PEO/graphite composite coating exhibited good antifriction property. But similar to the solid lubricants deposited with magnetron sputtering technology, the sprayed materials were also difficult to sink deep into the PEO pores.
In order to achieve good pore-filling effect, in-situ generating solid lubricants inside the pores could be a better method. J.J. Yang et al.18) generated Ag/Ag2MoO4 nanoparticles in the pores of plasma-sprayed zirconia coatings with the aid of vacuum impregnation and hydrothermal reaction. The generated nanoparticles in the pores dramatically reduced the friction coefficient and wear rate of zirconia coatings. Y. Su et al.19) fabricated an Al2O3–MoS2 composite via the hydrothermal synthesis of MoS2 inside a sintered porous Al2O3 ceramic matrix. This Al2O3–MoS2 composite exhibited excellent self-lubricating property and low wear rate in a high-vacuum environment. Above researches illustrate that the hydrothermal reaction solution can easily infiltrate into irregular pores with the aid of vacuum impregnation, and then solid lubricants can in-situ generate via hydrothermal reaction to fill the pores.
In this paper, PEO coatings were prepared on the titanium substrate and then PEO pores were filled with hydrothermal synthesized MoS2. The structure and tribological performance of the fabricated PEO/MoS2 composite coating were investigated in depth and compared with that of the PEO coating.
The substrate material used for the PEO treatment was commercially pure titanium with the chemical composition of Fe $ \leqslant $ 0.80, C $ \leqslant $ 0.10, N $ \leqslant $ 0.09, H $ \leqslant $ 0.08, O $ \leqslant $ 0.10 and Ti balance (mass%). The titanium plate was cut into samples with the dimension of 30 mm × 25 mm × 3 mm. Prior to the PEO treatment, the titanium samples was successively ground using a series of SiC abrasive sandpapers (150, 400, 800 and 1200-grit), and then ultrasonically cleaned in an alcoholic solution for 10 min at room temperature to remove surface residues and dried. A pulsed asymmetric bipolar AC power supply (MAO120HD-III, Xi’an University of Technology, China) was employed to perform the PEO treatment under the mode of constant voltage with the electrical parameters listed in Table 1. A dilute alkaline electrolyte containing 0.1 mol/L NaAlO2 and 0.05 mol/L NaOH was used, and the titanium substrate and stainless steel plate served as the anode and cathode respectively. During the PEO treatment, the temperature of the electrolyte was maintained below 40°C using a stirring and cooling system. The fabricated PEO coating samples were cleaned with deionized water and dried at ambient temperature.
MoS2 was hydrothermal synthesized in the PEO coating to fabricate the PEO/MoS2 composite coating. First, the PEO coating sample was immersed in the 60 mL hydrothermal reaction solution containing 0.1 mol/L Na2MoO4, 0.4 mol/L SC(NH2)2 and deionized water. Ultrasonic agitation and vacuum impregnation were carried out successively to ensure that the pores in the PEO coating were filled with the reaction solution. Then, the hydrothermal reaction solution together with the PEO coating sample was transferred into a 100 mL PPL-lined stainless steel autoclave. The autoclave was sealed and heated at 220°C for 48 h in a drying oven (DHG-9030A, Wuhan Zhongke Wantong Instrument Co., Ltd., China). After that, the autoclave was cooled to room temperature, and the fabricated PEO/MoS2 composite coating sample was taken out, rinsed with deionized water and dried at 40°C for 24 h.
2.2 Characterization and tribological test of coatingsThe microstructure and elemental composition of coatings were characterized by FEI-Scios scanning electron microscopy (SEM) equipped with energy dispersive spectroscopy (EDS). The phase composition of coatings was determined by X-ray diffraction (XRD, D/MAX-2500/PC, RIGAKU, Tokyo, Japan). The XRD measurement was conducted in the 2θ range of 10°∼80° at a scanning speed of 2°/min.
Friction and wear behavior of coatings were tested at room temperature under dry sliding condition by using a reciprocating ball-on-plate friction and wear tester (MS-M9000, Lanzhou Huahui Instrument Technology Co., Ltd., China). The GCr15 stainless steel ball, with a diameter of 6 mm, was used as the counterpart. The tribo-test was performed under a normal load of 10 N with a sliding velocity of 10 cm/s. Triplicate tests were carried out for each coating sample to ensure repeatability. After the test, the wear track of coating samples was measured using a surface profiler (MarSurf, Mahr, Germany). The morphology and elemental composition of wear tracks were characterized by SEM/EDS. The wear rate K was determined by the following equation: K = (S · l)/F · L, where S is the cross-sectional area of wear track (mm2), l is the length of wear track (mm), F is the applied load (N) and L is the sliding distance (m).
The microstructure and composition of the PEO coating are shown in Fig. 1. It can be seen from Fig. 1(a) that the PEO coating exhibited a porous and rough surface. The crater-shaped pores with varied size were scattered over the whole coating surface and these pores corresponded to the positions where vigorous micro-plasmas were generated and discharged. Micro-projections were formed around the relatively bigger pores because of the abundant accumulation of molten oxide erupted out of the pore.20,21) Figure 1(b) shows the cross-sectional morphology of the PEO coating. It can be noted from Fig. 1(b) that the eruption of molten oxide left lots of cavities inside the coating. Although these cavities possessed irregular and various shape, they basically all had small openings on the coating surface and large internal volume. Such a structural feature caused great difficulty in filling these cavities with solid lubricants. The solid lubricants usually tended to congregate on the PEO coating surface rather than entering deep into cavities. Figure 1(c) and (d) present the EDS elemental maps corresponding to the cross-sectional morphology shown in Fig. 1(b). These maps suggest that the PEO coating consisted mainly of the oxides of aluminum and titanium. The Al and Ti elements made up of the coating came from the PEO electrolyte and the titanium substrate respectively.
Surface and cross-sectional morphologies (a), (b) of the PEO coating and EDS elemental maps (c), (d) corresponding to the cross-section.
The porous and rough PEO surface caused by widespread pores and micro-projections usually produces high and unstable friction coefficient, which is harmful to the tribological application of the PEO coating. Besides, these cavities inside the coating are stress concentrators and promote the fatigue fracture of the PEO coating under severe loading conditions.22) Therefore, filling these pores with functional materials is crucial for improving the performance of the PEO coating. Figure 2 shows the surface morphology and chemical composition of the PEO coating after the pore-filling with hydrothermal synthesized MoS2. As can be seen from Fig. 2(a), the original porous PEO coating had been covered by a dense and nodular surface layer after the hydrothermal treatment. The high magnification surface morphology shown in Fig. 2(b) reveals that this surface layer consisted of a large amount of clustered microspheres. The microspheres had flower-like structure and were composed of many lamellae, which is in agreement with the structural characteristics of hydrothermal synthesized MoS2.23–25) Furthermore, the EDS element analyses of the surface layer revealed uniform distribution of Mo and S elements, as shown in Fig. 2(c) and (d). So it can be inferred that MoS2 had generated on the PEO coating via hydrothermal reaction and the synthesized MoS2 played a successful role in sealing the PEO pores.
Surface morphologies (a), (b) and corresponding EDS elemental maps (c), (d) of the PEO/MoS2 composite coating.
X-ray diffraction analysis was performed to verify the generation of MoS2 on the PEO coating. The XRD patterns of the PEO coating before and after the hydrothermal treatment are shown in Fig. 3. It can be seen from Fig. 3(a) that the PEO coating was basically composed of Al2TiO5 phase. This XRD result agreed well with the EDS results shown in Fig. 1 and the previous reported data.26) After the hydrothermal treatment, significant diffraction peaks corresponding to MoS2 emerged and the diffraction peaks from the PEO coating became weak, as shown in Fig. 3(b). Based on the results shown in Fig. 2 and Fig. 3, it can be concluded that lots of MoS2 were formed on the PEO coating. In the process of hydrothermal treatment of the PEO coating in the solution containing Na2MoO4 and SC(NH2)2, Na2MoO4 reacted with H2S resulting from the hydrolysis of SC(NH2)2 to generate MoS2.27)
XRD patterns of PEO (a) and PEO/MoS2 (b) coatings.
In order to achieve good pore-filling effect, hydrothermal synthesized MoS2 should not only generate on the PEO surface, but generate in the PEO pores too. Figure 4 shows the cross-sectional morphologies of the PEO/MoS2 composite coating and the corresponding EDS elemental maps. Among these elemental maps, the Mo and S maps depicted in Fig. 4(c) and (d) displayed the distribution of MoS2 component in the coating, while the Al and Ti maps shown in Fig. 4(e) and (f) sketched the PEO component. It can be noted from Fig. 4 that the PEO pores were well filled by hydrothermal synthesized MoS2 despite the complex interior shape of these pores. The fabricated PEO/MoS2 composite coating basically consisted of two layers: the upper MoS2 surface layer and the under PEO-MoS2 interlocking layer. The upper MoS2 surface layer was relatively thin and formed via the hydrothermal reaction occurred on the PEO surface. The under PEO-MoS2 interlocking layer was much thicker and included the PEO layer and the MoS2 filled in the PEO pores. In the process of hydrothermal treatment, the reaction solution first infiltrated into PEO pores with the aid of ultrasonic agitation and vacuum impregnation, and then MoS2 was in-situ generated in these pores via the hydrothermal reaction.
Cross-sectional morphologies (a), (b) and corresponding elemental maps (c), (d), (e), (f) of the PEO/MoS2 composite coating.
Friction and wear tests were performed to evaluate the tribological performance of PEO and PEO/MoS2 coatings. Dry sliding tests of coatings were carried out at ambient temperature under the normal load of 10 N with the GCr15 ball as the counterpart. Figure 5(a) presents the friction coefficient versus sliding distance curve for the PEO coating. It can be seen from Fig. 5(a) that the friction coefficient of the PEO coating displayed a continuous increase until a peak value of about 0.6. After that, it dropped abruptly at the sliding distance of about 56 m and then reached a low average value of about 0.47 with significant fluctuations, corresponding to the friction coefficient of the GCr15 ball sliding against titanium substrate.28,29) Figure 5(b) shows the wear track profile of the PEO coating after the sliding distance of 100 m. It can be noted that the wear track depth reached about 58 µm and this value was much larger than the average thickness of the PEO coating (about 20 µm, as shown in Fig. 1). These results indicate that the PEO coating had been worn through and removed completely from the titanium substrate after the tribo-test. The reason for this phenomenon is related to the porous microstructure of the PEO coating. At the beginning of the friction test, the GCr15 counterpart ball contacted with the rough and porous PEO surface, and the micro-protrusions on the PEO surface impeded the relative motion between them, resulting in the increasing friction coefficient up to about 0.6. With the proceeding of the friction test, the micro-protrusions gradually fractured under the action of shear force and the formed cracks propagated into the coating interior and coalesced with internal cavities, inducing the coating fragmentation and detachment from the titanium substrate. After the PEO coating had been removed, the friction coefficient began to decline and reached a low value with significant fluctuations. These fluctuations indicated the severe adhesive friction and wear between the GCr15 counterpart ball and titanium substrate.
Friction coefficient (a) and wear track profile (b) of the PEO coating.
Compared with the PEO coating, the PEO/MoS2 composite coating exhibited a totally different friction coefficient curve when sliding against the GCr15 ball under the same conditions, as shown in Fig. 6. First, a much lower friction coefficient varying in the range from about 0.1 to 0.2 was achieved by the PEO/MoS2 coating, and the maximum friction coefficient (∼0.2) reduced by about 66% compared to that of the PEO coating (∼0.6). Since the major difference between the PEO and PEO/MoS2 coatings lies in that the latter contained hydrothermal synthesized MoS2, it is reasonable to believe that MoS2 functioned as solid lubricants and played a lubricating role during the tribo-test. It’s also worth noting that the maximum friction coefficient (∼0.2) was still much lower than the friction coefficient of the GCr15 ball sliding against the titanium substrate (around 0.47), indicating that the GCr15 ball did not wear through the PEO/MoS2 coating. Secondly, a much longer sliding distance up to 1000 m was reached for the PEO/MoS2 coating, meaning that this coating had remarkable durability. Thirdly, the friction coefficient curve of the PEO/MoS2 coating presented a different variation tendency from that of the PEO coating. Specifically, this curve can be divided into two stages, as marked in Fig. 6. At the stage I, the friction coefficient remained at a relatively low and steady value of about 0.12 before the sliding distance of about 580 m. At the stage II, the friction coefficient presented a modest jump up to about 0.15 and then slowly increased up to about 0.2 with slight fluctuations.
Friction coefficient of the PEO/MoS2 composite coating.
In order to figure out the friction and wear mechanism of the PEO/MoS2 composite coating at the stage I and II, the worn surface of this coating was investigated by SEM/EDS after the sliding distance of 240 m (stage I) and 720 m (stage II) respectively. As shown in Fig. 7(a), the worn surface of the PEO/MoS2 coating after the sliding distance of 240 m was mostly smooth and flat except for small areas with grainy appearance. According to the corresponding EDS elemental maps shown in Fig. 7(b), it can be known that the smooth and flat surface was basically made up of MoS2 and the small portion of grainy surface was mostly Al2TiO5, the main component of the PEO coating. This result demonstrates that there existed a relatively uniform and continuous self-lubricating MoS2 layer on the worn surface at the stage I, which explains why the PEO/MoS2 coating had the lowest friction coefficient at this stage. With the increasing of the sliding distance up to 720 m, the formed self-lubricating MoS2 layer became less continuous although it still covered most of the worn surface, and more Al2TiO5 was exposed, as shown in Fig. 7(c) and (d). The reduced MoS2 and increased Al2TiO5 on the worn surface were responsible for the relatively higher friction coefficient at the stage II.
Worn surfaces and corresponding EDS elemental maps of the PEO/MoS2 composite coating after the sliding distance of 240 m (a), (b) and 720 m (c), (d).
Based on the coating structure and tribo-test results, schematic diagrams of the friction and wear behavior of the PEO/MoS2 composite coating are depicted in Fig. 8. As shown in Fig. 8(a), the PEO/MoS2 coating possessed a two-layer structure, including the upper MoS2 surface layer and the under PEO-MoS2 interlocking layer. At the beginning of the tribo-test, the GCr15 counterpart ball first contacted with the upper MoS2 surface layer, and a relatively uniform and continuous self-lubricating MoS2 layer was very easy to form on the worn surface due to the abundance of hydrothermal synthesized MoS2, which guaranteed the low friction coefficient of the composite coating at this stage. The small amounts of Al2TiO5 exposed on the worn surface corresponded to the micro-projections of the PEO coating. After the upper MoS2 surface layer had been worn away, the GCr15 counterpart ball began to contact the under PEO-MoS2 interlocking layer. At this stage, although the self-lubricating MoS2 layer on the worn surface became less continuous (as shown in Fig. 7(c) and 8(b)), it still covered most of worn surface and could constantly regenerate because the PEO/MoS2 coating had reservoir of MoS2 in the PEO pores. Therefore, the PEO/MoS2 coating maintained its self-lubrication property throughout the test. In addition, more Al2TiO5 was exposed on the worn surface, which is beneficial to the wear resistance of the composite coating.
Schematic diagrams of the friction and wear behavior of the PEO/MoS2 composite coating.
Figure 9 shows the wear volume and wear rate of the PEO/MoS2 composite coating at the sliding distance of 240 and 720 m. It can be noted that the wear volume of the PEO/MoS2 coating increased with the sliding distance. However, the increasing amount of wear volume in the range of sliding distance from 0 to 240 m (about 57 × 10−3 mm3) was much larger than that in the range from 240 to 720 m (about 29 × 10−3 mm3). As a result, the wear rate of the PEO/MoS2 coating at the sliding distance of 240 m was higher than that at the sliding distance of 720 m. That is to say, the PEO/MoS2 coating exhibited better wear resistance in the sliding distance from 240 to 720 m. When the sliding distance was smaller than 240 m, the higher wear rate was reasonable because the friction and wear at this stage occurred mainly between the hard GCr15 counterpart ball and the soft MoS2 surface layer. In contrast, more Al2TiO5 was exposed on the worn surface when the sliding distance was increased to 720 m, as shown in Fig. 7(c) and (d). The hard Al2TiO5 component in the composite coating withstood the load applied by the counterpart and avoided the rapid consuming of MoS2. In the meantime, the soft MoS2 component in the composite coating decreased the shear stress exerting on the worn surface via the formation of self-lubricating layer and thus alleviated the wear of friction pair. Therefore, the synergistic action of hard Al2TiO5 and soft MoS2 led to the excellent tribological performance of the PEO/MoS2 composite coating.
Wear volume and wear rate of the PEO/MoS2 composite coating at the sliding distance of 240 and 720 m.
Hydrothermal synthesized MoS2 successfully filled the pores of plasma electrolytic oxidation coatings on titanium. The fabricated PEO/MoS2 composite coating possessed a two-layer structure, including the upper MoS2 surface layer and the under PEO-MoS2 interlocking layer. Compared with the PEO coating, the PEO/MoS2 coating exhibited much lower friction coefficient and longer durability when sliding against the GCr15 counterpart. The excellent tribological performance of the PEO/MoS2 composite coating was attributed to the synergistic action of PEO and MoS2 components.
This work is supported by the Natural Science Foundation of Hebei Province (E2020203057), Foundation for Science and Technology Research in Universities of Hebei Province (QN2019013) and Fundamental Research Foundation of Yanshan University (020000904).
