Tribological Properties of SiO2@Cu and SiO2@MoS2 Core–Shell Microspheres as Lubricant Additives

Herein, core–shell structural SiO2@Cu and SiO2@MoS2 microspheres were prepared using SiO2 as hard core, Cu and MoS2 as shell. As lubricant additives were introduced into base oil (PAO 40), their friction reduction and wear resistance were investigated in detail. Comparing with onefold additive (SiO2, Cu and MoS2), such core–shell structural additives can improve the tribological behaviors at the Hertz contact stress range of 1.26–2.72 GPa (SiO2@Cu reduces the friction and wear up to 32.47% and 67.86% at 2.72 GPa, respectively). Besides, the tribological properties of SiO2@Cu microspheres are superior to that of SiO2@MoS2 (the wear volume was reduced by 48.45% at 2.72 GPa). The excellent tribological behaviors of SiO2@Cu microspheres can be ascribed to its structural advantage, the synergistic effect of hard SiO2 core and Cu shell. The rolling effect of SiO2, easy-shearing and self-repairing of Cu shell offer a synergistic lubrication function and form a dense protection film, thereby contributing to the optimal lubrication performance.


Introduction
Energy losses across all mechanical systems are primarily attributed to friction and wear of moving elements. Therefore, lubricants are used to improve the energy efficiency of automobile engines and industrial machinery, extend service intervals, and enhance durability and reliability [1][2][3][4]. During the past decade, an important research topic in the field of tribology is to explore high performance solid lubricant additives with low friction and high wear resistance, in view of their small size and thermal stability [5][6][7]. With the development of nanotechnology, a considerable number of solid particles with different structures have been fabricated to regulate the tribological behaviors of lubricating oil [2,8]. But it cannot be ignored that onefold solid additive has some drawbacks. On the one hand, hard solid particles such as SiO 2 and ZnO will easily scratch the contact surface causing abrasive wear, and soft particles such as Cu and PMMA cannot be suitable for harsh conditions due to their poor mechanical strength. On the other hand, the traditional mechanical mixing method is easy to produce problems such as phase separation or uneven dispersion, which cannot achieve the technical requirements of low friction and wear resistance under harsh working conditions [9][10][11].
To compensate for the shortcomings of onefold solid additive, core-shell structure additives with soft shell and hard core were constructed recently [12][13][14][15]. The core-shell structure can not only perform the deformation ability of the soft shell, but also play the bearing function of hard core. Meanwhile, core-shell microspheres as solid lubricant additive can achieve long life, low friction and good service reliability under harsh working conditions, due to their stable interfacial adhesion between soft shell and hard core [16]. Recently, MoS 2 as a metal dichalcogenide has been used as the "soft shell" of core-shell particles, to solve dispersion problems and enhance the mechanical and chemical stability of hard core [17][18][19][20][21]. Abdullah et al. [22] suggested that carbon spheres coated with a MoS 2 nanolayer (CS-MoS 2 ) demonstrated a significant reduction in friction and wear (15-35%) relative to standard engine oil in the boundary and mixed lubrication regimes. But MoS 2 as soft shell cannot react with the contact surfaces to form robust tribo-film or reduce the surface roughness of the worn surface, which has a great negative impact on the tribological properties of core-shell microspheres [23]. Therefore, it is of great significance to find other soft materials to solve this problem.
Many investigations indicate that soft metal particles can act as friction modifiers [2]. Lubrication mechanisms of soft metallic nanoparticles could be explained by the sintering or repair effect, that is to say, the nanoparticles would be compacted on worn surface due to heat and pressure generated during friction process [24]. Although the soft metal has an excellent property in improving anti-wear performance, there are few researches regarding it as the soft shell of core-shell structure additives. Cu nanoparticles as a type of soft metal are often used as lubrication additives because of their good self-repairing property. Wang et al. [25] reported that the tribological performance of carbon nanotubes in base oil was improved via decorating with uniform copper nanoparticles. Qu et al. [26] demonstrated the good tribological properties of PTFE particles modified by Cu microparticles, because of the as-formed transfer film and self-repairing effect of Cu microparticles. Therefore, Cu is expected to be a kind of soft material to make up for the deficiency of MoS 2 as the shell of core-shell structure additives. However, the previous researches mainly used copper as a decoration, while the tribological performance of core-shell microspheres using Cu as the soft shell was rarely studied. Moreover, SiO 2 microspheres can be used to enhance the tribological properties of base oil [27,28]. Due to its rigid structure and perfect spherical morphology, SiO 2 microspheres are often used as the hard core of core-shell structure additives [29][30][31][32]. Unfortunately, the tribological performance of SiO 2 @Cu and SiO 2 @MoS 2 as lubricant additives were rarely studied. As mentioned earlier, of great significance is to find preferable soft material to enhance the tribological properties of core-shell microspheres. Therefore, it is necessary to investigate the tribological properties of core-shell structural SiO 2 @Cu and SiO 2 @MoS 2 microspheres.
Overall, this paper aims to explore the lubrication effect and mechanism of Cu and MoS 2 as the soft shell of SiO 2 microspheres. In this work, SiO 2 @Cu and SiO 2 @MoS 2 core-shell microspheres were successfully synthesized, and were added into polyalphaolefin 40 (PAO 40) to study their lubrication function under different loads. For comparison, the tribological properties of single SiO 2 , Cu and MoS 2 particles were also studied. In addition, the morphology of wear track and core-shell microspheres after test were investigated to reveal the anti-wear mechanism.

Materials
All solvents and chemicals used in this work were analytical grades, purchased from Chengdu Kelong Chemical Co., Ltd. (Chengdu, China). SiO 2 NPs, with single-particle sizes ranging from 560 to 700 nm, were used in this study. The base oil, polyalphaolefin 40 (PAO 40) was obtained from Liugong Machinery Co., Ltd. (Guangxi, China), whose specification is shown in Table 1.

Fabrication of SiO 2 @Cu and SiO 2 @MoS 2 Microspheres
The synthesis route of the SiO 2 @Cu and SiO 2 @MoS 2 microspheres is illustrated in Fig. 1. As described in Fig. 1, a typical synthesis procedure of the SiO 2 @Cu microspheres was as follows [20]. Firstly, 0.075 g SiO 2 was dispersed into 30 ml deionized water with sonication (KS-600 N, 250 W) for 10 min to ensure uniform dispersion, 0.1 g anhydrous copper sulfate was added to 6.25 ml deionized water, then clear and transparent blue solution was formed by stirring with a glass rod. Secondly, the above solution was mixed and stirred with a magnetic stirrer for 30 min to ensure that the copper ions in the solution were in full contact with the surface of the silica. Then adding 0.035 g iron powders were slowly added at room temperature and reacting for 2 h, the iron powders can be used as a reducing agent to reduce copper ions adsorbed on the surface of silica to copper. Lastly, the brown-red precipitation was collected by centrifugation and washed with distilled water and absolute ethanol until the excess reactants were removed. In addition, the SiO 2 @MoS 2 microspheres were synthesized by the following steps [32]. Firstly, 0.13 g SiO 2 was dispersed into 27 ml deionized water and mechanically stirred with ultrasonication. Secondly, 0.53 g sodium molybdate and 1.06 g thiourea were added and stirred with a magneton agitator for 1 h. Then the mixed solution was transferred into a 100 ml Teflon-lined stainless-steel autoclave and heated at 200 ℃ for 10 h. After that, the autoclave was cooled to room temperature naturally. Finally, the black products were collected through centrifugation and washed with deionized water and ethanol several times, then dried at 60 ℃ for 12 h. The preparation method of MoS 2 is as the same as the above method, but the preparation process does not add SiO 2 .

Characterization
The morphology of the as-obtained samples and friction surface were studied by using of a field-emission scanning electron microscope (SEM, FEI, Inspect F50, America) with an energy dispersive X-ray spectroscopic (EDS, acceleration voltage: 20 kV) detector. Raman spectra of wear tracks were obtained by Thermo-Fisher Scientific DXR Raman microscope with 532 nm laser excitation. Bruker Contour GT surface mapping microscope profilometer was used to examine the profile and wear volume of worn surfaces. An optical microscope (Zeiss Observer Z1m) was used to measure the wear track width and the wear scar diameter. PAO 40 was used as the lubricating fluid, and SiO 2 @Cu, SiO 2 @MoS 2 , SiO 2 , Cu and MoS 2 microspheres served as lubricating additives throughout this study. The as-prepared microspheres were added into the base oil with 1.0 wt.%. To make microspheres well-dispersed in the base oil, the core-shell material was stably dispersed in the base oil through ultrasound for 2 h. The tribological properties of asprepared lubricants were investigated using a ball-on-plate tribometer (Optimal-SRV-IV reciprocation friction tester) at different applied loads (20-200 N) for 1 h with the frequency of 2 Hz and displacement amplitude of 5 mm. The experiments were conducted at room temperature, with relative humidity between 55 and 70%. Before tribological tests, the steel ball and disk were thoroughly cleaned with ethanol and acetone. The GCr 15 bearing steel with a diameter of 10 mm was used as the upper steel balls (61 HRC, Ra = 53 nm) and the counterpart steel disc (62 HRC, Ra = 28 nm) was composed of AISI 52100 steel with a size of 24 × 7.9 mm. As a contrast, the tribological properties of pure base oil were tested under the same conditions.

Tribological Tests
The average Hertzian contact pressure at 20-200 N was calculated with Hertz's theory [33]: where p is the Hertz contact pressure; a is the Hertz contact diameter; W is the normal load; R is the radius of steel ball (R = 5 mm); and E ′ is the effective modulus of elasticity ( E ′ = 208 GPa). Calculated from the formula above, the corresponding contact pressures at 20-200 N are 1.264-2.724 GPa.
The boundary lubrication is especially significant for the form of tribo-film. Therefore, in order to compare the quality of the lubrication film formed by core-shell additives under different lubrication conditions, the corresponding lubrication conditions should be first determined according to the λ ratio in Eq. (3). Where, h min refers to the minimum film thickness and is evaluated by Eq. (4), and R q is the composite roughness calculated by Eq. (5).
where W is the applied load, R is the radius of the ball (5 mm), is the viscosity-pressure coefficient (3.59 × 10 −8 m 2 /N), U is the speed (0.01 m/s), 0 is the dynamic viscosity (357 N•s/m 2 ), R ball is the surface roughness of the ball (53 nm), R flat is the roughness of the flat (28 nm) and E ′ is elastic modulus (208 GPa). Usually, the value of is used to determine the lubrication regime: 0.1 < < 1 indicates boundary lubrication; 1 ≤ ≤ 3 indicates mixed lubrication, and > 3 indicates elastohydrodynamic lubrication [23,34,35]. Using the corresponding values of the fundamental constants of physics and material characteristics, the lambda ratio is 0.973 for 200 N, 1.065 for 100 N, and 1.199 for 40 N. Therefore, the tests under 200 N were in the boundary lubrication regime, and under 40-100 N were in mixed lubrication regime.
After tests, the ball and disk samples were cleaned with ethanol, then placed in a sealed bag for later characterization tests. Each oil was tested three times, and the standard deviation was calculated from the results of the three times friction tests. To illustrate the role of the two types of core-shell microspheres in tribo-tests, the microspheres in the oil were collected with cotton after the friction test and washed with petroleum ether, then observed by SEM. Figure 2 shows the SEM images of neat SiO 2 , SiO 2 @Cu and SiO 2 @MoS 2 microspheres at different magnifications.

Characterization of the core-shell materials
The morphology of SiO 2 spheres is shown in Fig. 2a, b, it can be seen that SiO 2 spheres possessed a perfectly spherical structure, and the average diameter is about 623 nm according to the measurement results (Fig. 2c, d). The average diameter of SiO 2 @Cu microspheres is about 694 nm. Apparently, copper microspheres are uniformly coated on the surface of SiO 2 spheres, forming a uniform copper shell layer with a thickness of about 35.5 nm. Likewise, as shown in Fig. 2e, f, SiO 2 microspheres are surround by irregular MoS 2 nanolayer. The results show that the average diameter of SiO 2 @MoS 2 microspheres is about 683 nm, and the thickness of MoS 2 nanolayer is about 30 nm. It is worth noting that the MoS 2 shell prepared by the hydrothermal method do not exhibit traditional floral crystal structure but an amorphous structure, which can be attributed to magnetic stirring during the synthesis process [13]. Consequently, the differences in morphology and shell thickness between two kinds of core-shell particles are basically eliminated, which was beneficial to compare lubrication performances between Cu shell and MoS 2 shell [17].
In order to further prove that the materials adsorbed on the surface of SiO 2 in the above figures are Cu and MoS 2 , the surface elemental compositions of spherical particles were analyzed by EDS. It is clear from Fig. 3a that Si, O and Cu elements are detected, and the peak value of each element is strong. Combined with the SEM images, SiO 2 @ Cu core-shell structure composites are successfully synthesized. Similarly, the characteristic peaks of Si, O, Mo and S elements are detected in the microspheres from Fig. 3b. These results demonstrate convincingly that SiO 2 @MoS 2 nanocomposites are successfully synthesized.
Homogeneous dispersion plays a vital role in lubrication. Visual observations were used to judge the dispersion stability of the core-shell microspheres in PAO 40. As shown in Fig. 3c, d, the 1.0 wt.% SiO 2 @Cu and 1.0 wt.% SiO 2 @MoS 2 nanospheres could be uniformly dispersed in oil after sonication. The oil containing SiO 2 @Cu appears brownish-yellow in color, and the oil containing SiO 2 @MoS 2 is black. After seven days of storage, the two core-shell microspheres still exhibited admirable dispersity due to their small size and the presence of oxygen-containing groups on their surface [26,32,36]. These results show that these nanoparticles have good dispersive stability, and can meet the requirements of the friction test.

Tribological Behavior
The friction coefficient curves and average friction coefficient of SiO 2 , Cu and SiO 2 @Cu core-shell microparticles as lubricating additives under mixed lubrication regimes are shown in Fig. 4. Clearly, the friction curve of PAO 40 lubricating oil is relatively unstable, and even fluctuates wildly under 40 N and 60 N loads. The friction coefficient of SiO 2 hybrid oil is stable due to the high bearing capacity of SiO 2 microspheres. However, the friction coefficient of SiO 2 hybrid oil increase gradually with the prolonging of experiment time due to the agglomeration of SiO 2 microspheres and serious abrasive wear. In addition, low and stable friction coefficient is obtained when using hybrid lubricants. In contrast, the friction curves of the lubricating oil with copper as an additive are smoother and lower than that of SiO 2 @Cu. Obviously, SiO 2 microparticles cannot effectively reduce wear under mixed lubrication conditions, but the copper layer of SiO 2 @Cu core-shell microspheres can make the friction coefficient tend to lower and more stable.
The tribological properties of MoS 2 , SiO 2 @MoS 2 core-shell composite lubricants under mixed lubrication regime were also investigated. As seen in Fig. 5, MoS 2 performs well under low load conditions and has the lowest friction coefficient. But its friction coefficient continues to rise with the increase of load, which is closely related to the characteristics of soft texture and low bearing capacity of MoS 2 . The friction coefficient of SiO 2 @MoS 2 core-shell microparticles is much lower than that of SiO 2 , indicating soft shell is beneficial to reduce abrasive wear caused by hard particles. Unfortunately, the friction reduction effect of the SiO 2 @MoS 2 core-shell microspheres as lubrication additives is not better than onefold materials, which maybe because it is difficult for core-shell particles to form stable tribo-film in the mixed lubrication state [17,36].
By comparing the size of wear tracks and scars generated at 40 N (Fig. 6), it is found that using Cu and MoS 2 as soft shells could effectively reduce the wear caused by SiO 2 microspheres under a mixed lubrication regime. Especially, when SiO 2 @Cu spheres are used as lubricating additives, the diameter of wear scar is 9.1% smaller than that of SiO 2 .
In addition, when Cu or MoS 2 particles are used as lubricating additives, the tribological properties of them are better than core-shell particles. This is caused by the competitive mechanism of the abrasive wear of SiO 2 and the selfrepairing property of soft shells. Specifically speaking, the self-repairing property of Cu is better than that of MoS 2 according to the comparison of the wear scar diameter of SiO 2 @Cu and SiO 2 @MoS 2 .  Figure 7 shows the friction coefficient curves and wear volume of the substrates lubricated by different oil samples at 100 N and 200 N. The friction experiments are still under mixed lubrication conditions at 100 N, so the friction coefficient and wear of the hybrid oil with various solid additives are greater than that of the pure oil. But the wear volume of SiO 2 @Cu hybrid oil is still less than that of SiO 2 @MoS 2 , which is consistent with the results in Fig. 6. By contrast, SiO 2 @Cu and SiO 2 @MoS 2 core-shell materials as lubrication additives have obvious anti-wear and anti-friction effects under boundary lubrication regimes (200 N). Compared with pure oil, the friction coefficient of SiO 2 @Cu and SiO 2 @MoS 2 is reduced by 32.47% and 30.98% respectively, and the wear volume is reduced by 67.86% and 52.24%. It is worth noting that the anti-wear effect of the  : a, b 20 N, c, d 40 N, and e, f 60 N core-shell materials is better than that of any onefold additives (Fig. 7d). Under boundary lubrication regime, the wear resistance of Cu microspheres and MoS 2 is very poor when they were used as single solid lubrication additive, even the wear volume of MoS 2 is much larger than that of pure oil. This may be because the insufficient carrying capacity of these soft particles causes the oil film to rupture, then causing adhesive wear under high contact pressure and high friction heat. In addition, the wear volume of SiO 2 @Cu and SiO 2 @MoS 2 decreased by 48.45% and 23.39% compared with SiO 2 , which indicates that the Cu shell has better wear reduction properties than SiO 2 . Especially, no matter SiO 2 @ Cu or SiO 2 @MoS 2 , the wear resistance of them is better than onefold additive (SiO 2 , Cu and MoS 2 ). Core-shell microspheres as solid lubricant additive can achieve low friction and good service reliability under harsh working conditions. Figure 8 shows the 3D surface morphologies and crosssectional profiles of the wear track at 200 N lubricated with PAO 40, SiO 2 , SiO 2 @Cu and SiO 2 @MoS 2 hybrid oil. From Fig. 8a we can see the cross-sectional profile from PAO 40 lubrication displays maximum wear track depth (1.364 μm), which could be attributed to contact fatigue and adhesive fatigue during the tribo-test. Whereas, the deep of wear track is decreased to 1.127 μm when 1 wt.% SiO 2 microspheres are added to the pure oil, suggesting that the SiO 2 microspheres fills the gaps in the worn surface and is beneficial for reducing wear. By contrast, the wear track depth of lubricant with SiO 2 @Cu and SiO 2 @ MoS 2 microspheres is reduced to 0.752 μm and 0.748 μm, respectively. Consequently, those results indicate that the soft shell of core-shell structure microspheres improves the anti-wear properties of SiO 2 , which are consistent with the change of friction coefficient and wear volume mentioned above.

Analysis of Worn Surface
After the friction test, the wear tracks and wear scar morphology were observed by optical microscope, then the EDS element diagram under mixed lubrication conditions (40 N) were analyzed. As can be seen from Fig. 9a, there are a lot of furrows on the PAO 40 lubricated worn surface, the width of the wear tracks reaches the maximum of 228 μm, which indicate that the poor bearing capacity of pure lubricating oil leads to seriously wear on the contact surface. Figure 9f shows that Si element signal is detected, which indicate SiO 2 particles are adsorbed on the surface of the wear track. There are obviously deeper furrows on the wear surface of SiO 2 lubricating oil (Fig. 9b), due to the three-body abrasion of SiO 2 microparticles, which have high hardness and brittle texture. Moreover, it can be seen from Fig. 9c, d that SiO 2 @ Cu and SiO 2 @MoS 2 core-shell materials hybrid lubricating oils have a lower wear track width than SiO 2 , and obvious black-brown transfer film is generated on the wear mark. Cu and S element signal are detected at their wear tracks even though the content of these elements is relatively low (Fig. 9g, h), which indicates the synergistic effect of SiO 2 and Cu or MoS 2 plays a role in the friction process. Additionally, the diameter of wear scar lubricated by SiO 2 @Cu hybrid oil is smaller that under SiO 2 hybrid oil lubrication, while the diameter of wear scar under SiO 2 @MoS 2 oil lubrication is almost same as using SiO 2 hybrid oil. Hence, the abrasive wear reduction of Cu shell under mixed lubrication conditions is better than MoS 2 .
Moreover, the worn surface after sliding for 1 h at 200 N was characterized by SEM and EDS (Fig. 10), the results can be used to analyze the effect of different additives under boundary lubrication conditions. The SEM results are basically consistent with the 3D morphology (Fig. 8). As shown in Fig. 10a, when pure oil is used for lubrication, there are large spalling pits and O element (14.95 wt.%) on the wear marks. That is because pure oil cannot reduce high contact pressure and interfacial flash temperature effectively, leading to serious adhesive wear and oxidation on the worn surface. Interestingly, spalling pits on the worn surface are fewer when SiO 2 microspheres are added, but deep furrows are produced owing to abrasive wear (Fig. 10b). EDS analysis shows that Si (0.69 wt.%) and O (11.10 wt.%) elements exist on the wear track, indicating the particles can be transferred and deposited on the worn surface. There are reasons to believe that SiO 2 microparticles could improve the anti-wear ability of base oil to a certain extent under boundary lubrication conditions, but will cause abrasive wear, because it is difficult to produce stable tribo-film [27]. Figure 10c, d show the morphology of the worn surface when SiO 2 @Cu and SiO 2 @MoS 2 microspheres are added, respectively. The worn surfaces lubricated by soft-shell@hard-core microparticles all display shallower and smoother wear tracks. Even more importantly, the friction contact surface is the smoothest when SiO 2 @Cu particles are added, and covered with a stable and robust tribo-film which containing a large amount of copper (37.13%). On the contrary, the wear track of SiO 2 @MoS 2 hybrid oil exhibited deeper furrows and fewer Mo element (16.29 wt.%), suggesting that the tribological transfer film formed by the MoS 2 shell is weaker than Cu shell. It implies that using Cu as shells are better than MoS 2 in the self-repairing property. On the other hand, the ratio of O element in SiO 2 @Cu lubricated surface decreased sharply than SiO 2 hybrid oil, due to the formation of a compact tribo-film in the contact area, which is conducive to preventing oxidation of the worn surface [37]. Thus, soft-shell@hard-core microparticles are proved to be an admirable composite material with better self-repairing property, and Cu as soft shells are likely better than MoS 2 for protecting contact surfaces [18].
Furthermore, in order to study the properties of the tribofilm formed by SiO 2 @MoS 2 microspheres hybrid oil, the worn surfaces at 200 N were investigated by Raman spectra analysis as shown in Fig. 11. According to the Raman spectrum, peaks at 223, 295, 411 and 663 cm −1 (Fig. 11a) can be assigned to the spectra of Fe 2 O 3 and Fe 3 O 4 [38], which indicate that there is a large amount of iron oxide on the worn surfaces. In fact, the occurrence of oxidational wear is one of the reasons for the poor anti-wear effect of pure oil under boundary conditions [39]. In Fig. 11b, two weak peaks at 373 and 410 cm −1 (E 1 2g and A 1g mode) of MoS 2 are discovered on the wear traces tested by SiO 2 @MoS 2 hybrid oil, indicating that SiO 2 @MoS 2 microspheres could enter the friction surfaces to form the tribo-film [40]. However, the intensity of E 1 2g and A 1g mode peaks are relatively low, this could be due to the weak reaction with the worn surfaces [41,42]. Besides, peaks at 292 and 670 cm −1 also can be assigned to the spectra of different iron oxides, the absence of peaks at 223 and 411 cm −1 may be because the decrease of oxidational wear [8]. As the main component of the tribo-film, iron oxides and MoS 2 play a vital role in boundary lubrication. Eventually, the analysis results of Raman spectra show that the soft-shell@hard-core microparticles are beneficial to the reduction of oxidational wear, but MoS 2 as the soft shell of SiO 2 cannot give aid to the formation of the robust tribo-film during the frictional process [43].

Lubrication Mechanism
The topographies of microspheres after test are related to the lubrication mechanism and follow a certain evolution law, which provides an effective weapon for us to study the effect of core-shell particles on different lubrication regimes. Therefore, the morphology and microstructure of SiO 2 @ Cu and SiO 2 @MoS 2 microparticles after the test could shed more insight into the lubrication mechanism. As shown in Fig. 12a, b, most the microspheres maintain original morphology after friction test at low contact pressure (1.593 GPa) for 1 h, and only a few of the SiO 2 @Cu and SiO 2 @ MoS 2 microspheres are broken. The spherical structure of the crushed core-shell microspheres is deformed into irregular spheres. Moreover, EDS results in Fig. 12a, b show that Cu and MoS 2 still exist on the surface of the SiO 2 particles, demonstrating that the part of the shell has not worn completely during the friction. Correspondingly, the morphology and EDS analysis of SiO 2 @Cu and SiO 2 @MoS 2 spheres after tribo-tests at 200 N for 1 h are shown in Fig. 12c, d, respectively. Original SiO 2 @Cu and SiO 2 @MoS 2 spheres are crushed into smaller spheres at high contact pressure (2.724 GPa). Surprisingly, EDS results show that the surface of the crushed particles has not Cu or MoS 2 , because the particles are severely worn. By observing the change of particle morphology after test under different loads, we found that whether SiO 2 @Cu or SiO 2 @MoS 2 particles, low pressure could make the surface of the spheres become rough, while high pressure could make the particles completely broken and the shell peeled off [43]. The breaking degree of SiO 2 @Cu microspheres is lower after the friction experiment, demonstrating that soft metal shell can equip the microspheres with a higher load-carrying capacity. Under boundary lubrication, using microparticles of soft-shell and rigid-core as lubricant additives can not only take advantage of the easy formation of a transfer film, but also bring the advantages of small size and bearing advantage of the hard nanoparticles into full play [38]. The two-phase composite particles are gradually peeled off during the rolling process, which promote the uniformity and integrity of the transfer film and enhance the self-healing capacity [14].
Combining with the analysis of the worn surface and microparticles after the test, the lubrication mechanism of two kinds of soft-shell@hard-core microspheres under different lubrication regimes is deduced. To begin with, the molecules of lubricant and core-shell microspheres are adsorbed on the frictional interfaces and formed a tribofilm during the friction process. During mixed lubrication conditions, the core-shell microspheres are acted as ballbearings on the nanometer scale, and only a few particles enter the worn surface to act as a third body material filling the gap on the worn surface [16]. The surface of the particles which have entered the contact area would be slightly deformed. Meanwhile, the core-shell microspheres are sheared off on the worn surface due to the role of shear force, and weak physical tribo-film is formed at the same time [2]. By contrast, during boundary lubrication conditions, the contact between the microbulges leads to the oil film piercing. Then core-shell particles broken up completely into smaller spheres and the shell of those particles peeled off under high load. Besides, the debris formed by the wear and core-shell particles would fill the pits on the worn surface to repair the damaged region and reduce the surface roughness. At the same time, for SiO 2 @Cu hybrid oil, microspheres gradually deposit to the friction interface during the friction process, then Cu shell become melt and spread to form a dense protective film under high temperature and pressure, and repair the damaged interface [37]. Then the crushed SiO 2 turn into a carrier to support worn surface [19], and form a composite boundary lubrication film with Cu shell [44]. For SiO 2 @ MoS 2 hybrid oil, the thermal induction during sliding can increase the size and crystallinity of MoS 2 . The parallel sliding of MoS 2 along the section induced by high pressure is beneficial to reduce the friction. However, the peeled MoS 2 nanosheet cannot form a stable adsorbed film and a robust tribo-film, that is the reason why the tribological properties of SiO 2 @MoS 2 are worse than that of SiO 2 @ Cu. On the whole, the physical deposition of core-shell microparticles, self-repairing of the Cu shell, and chemical reaction of base oil together construct a synergistic effect, which can contribute to the optimal lubrication performance of SiO 2 @Cu.

Conclusions
In this work, SiO 2 @Cu and SiO 2 @MoS 2 microspheres were synthesized to explore the lubrication properties and corresponding mechanism of Cu and MoS 2 as the soft shell of SiO 2 microspheres. Friction experiments were carried out at different loads, then the morphologies and elementary compositions of worn surface and wear debris were investigated to reveal the lubrication mechanism of SiO 2 @Cu and SiO 2 @MoS 2 microspheres. The following conclusions can be drawn from this work: (1) SiO 2 @Cu and SiO 2 @MoS 2 core-shell microspheres were prepared by iron reduction process and hydrothermal method. The core diameter and shell thickness were about 623 nm and 30 nm, respectively. (2) Core-shell structure hybrid oils display excellent friction reduction and wear resistance at high applied loads. The lubrication behaviors of SiO 2 @Cu are superior to that of SiO 2 @MoS 2 (the addition of 1.0 wt.% SiO 2 @Cu and SiO 2 @MoS 2 can reduce wear volume by 48.45% and 23.39% with SiO 2 as a comparison).
(3) Rolling effect of SiO 2 , easy-shearing and self-repairing of Cu shell together construct a synergistic effect, thereby contributing to optimal lubrication performance of SiO 2 @Cu.