Two-dimensional molybdenum carbide (MXene) as an efficient nanoadditive for achieving superlubricity under ultrahigh pressure

In this study, a robust macroscale liquid superlubricity with a coefficient of friction of 0.004 was achieved by introducing molybdenum carbide (Mo2CTx) MXene nanoparticles as lubricating additives in a lithium hexafluorophosphate-based ionic liquid at Si3N4—sapphire interfaces. The maximal contact pressure in the superlubricity state could reach 1.42 GPa, which far exceeds the limit of the superlubricity regime in previous studies. The results indicate that a composite tribofilm (mainly containing molybdenum oxide and phosphorus oxide) that formed at the interface by a tribochemical reaction contributed to the excellent antiwear performance. Furthermore, the extremely low shear strength of the tribofilm and the interlayers of Mo2CTx MXene contributed to the superlubricity. This work demonstrates the promising potential of Mo2CTx MXene in improving superlubricity properties, which could accelerate the application of superlubricity in mechanical systems.


Introduction
Friction and wear cause significant energy loss on a mechanical surface, serious material loss, shortened service life, and increased safety accidents involving machinery [1,2]. Therefore, many researchers have used various methods to reduce friction and wear [3]. In this context, a new concept of superlubricity, with a coefficient of friction (COF) of less than 0.01, was proposed to enhance lubrication performance at the nanoscale or macroscale, which could provide ultralow friction and wear to reduce energy loss. Currently, different superlubricity approaches exist, such as solid, liquid, and solid-liquid coupling superlubricity systems. For solid superlubricity, molybdenum disulfide (MoS 2 ) [4], graphite sheets [5], the graphene family of materials [6,7], and diamond-like carbon [8,9] could act as typical solid superlubricity materials at the nanoscale or macroscale.
Compared with solid superlubricity, liquid superlubricity is more efficiently realized at the macroscale with lubrication by various liquids, such as phosphoric acid [10], ionic liquid (IL) [11], oleic acid [12], polyalphaolefin oil [13], and biological liquid [14]. To enhance the load capacity and antiwear ability in the liquid superlubricity state, conventional layered materials, including black phosphorus [15], graphene materials [7,16], hexagonal boron nitride [17], and layered double hydroxides [18], have been used as lubricant additives because of their extremely low shear strength in their interlayers. These layered materials have a similar two-dimensional (2D) layered structure, which can be adsorbed on the worn-out zone after the friction process to reduce wear and simultaneously provide extremely low shear strength between friction pairs. However, one limitation of these liquid materials, related to the contact pressure (dividing the normal load by the contact area), must be improved. With lubrication by these liquids, even though the initial contact pressure between the frictional pairs is high (on the order of gigapascals), the final contact pressure during the superlubricity period is still at a much lower order of magnitude. This is caused by the severe wear occurring during the running-in period.
Recently, a novel member of the 2D layered material family, MXene, was formed by a few atoms with thick layers of transition metal carbides, nitrides, or carbon nitride based on MAX-phases [19][20][21]. This material exhibits excellent electrochemistry, thermoelectricity, lubrication, and other properties, making it suitable for use in energy storage [22], catalysis [23], lubrication [24], etc. Some studies on the MXene lubrication have been reported [25,26]. For example, in a number of studies, Ti 3 C 2 MXene or its composite materials were used as solid coatings to enhance wear resistance and tribological properties [27][28][29][30][31][32]. Other researchers have investigated MXene nanoparticles as lubricant additives that can reduce friction and wear [33][34][35]. Although previous research on the MXene lubrication has revealed that MXene has the potential to achieve low friction and wear, industry requirements (ultralow friction and ultrahigh contact pressure) still hinder the application of MXene. In this study, Mo 2 CT x MXene nanoparticles were synthesized and introduced into lithium hexafluorophosphatebased IL as a nanoadditive. Robust macroscale superlubricity has been achieved under an ultrahigh contact pressure of as much as 1.42 GPa, which is a 136.67% increase compared with previous results in Ref. [36]. The lubricating behavior related to this liquid superlubricity under ultrahigh pressure using Mo 2 CT x MXene as a lubricating nanoadditive was studied, and the related mechanism was revealed.

MXene and lubricant synthesis
The precursor Mo 2 Ga 2 C powders were synthesized from gallium ingots (Ga, 99.99 wt%, Beijing Xingrongyuan Co., Ltd., China) and molybdenum carbide powders (Mo 2 C, 99.5 wt%, 325 mesh, Sigma Aldrich Co., Ltd., USA). Mo 2 C and Ga were weighed in a molar ratio of 1:5 [37,38]. Subsequently, the resulting Mo 2 Ga 2 C powders were added to 40 mL of etching solution. The etching solution was prepared by mixing 2.0 g of LiF (99.9 wt%, Aladdin Industrial Co., Ltd., China) and 6-M HCl (11.8 mol/L, Shuangshuang Chemical Reagent Co., Ltd., China). Afterward, the mixture was washed until a pH of approximately 6.0 by using deionized water to obtain Mo 2 CT x MXene particles. Thereafter, the collected Mo 2 CT x MXene particles were mixed with deionized water at a ratio of 0.2-5 mg/mL to prepare an aqueous solution of Mo 2 CT x MXene and sonicated for 48 h. The IL formed in situ was synthesized by dissolving 0.1 g of LiPF 6 (purity of 97%, Aladdin Industrial Co., Ltd., China) in 2 g of ethylene glycol (EG) (purity of 99%, Aladdin Industrial Co., Ltd., China) via sonication for 8 h, followed by magnetic stirring at 50 °C for 8 h ( Fig. 1(d)). Once the IL was prepared, it was immediately mixed with a Mo 2 CT x MXene aqueous solution at a ratio of 1:5 by weight, and the mixtures were denoted as Mo 2 CT x -IL ( Fig. 1(e)). The concentration of nanoparticles varied from 0.021 to 0.416 wt%. For comparison, the graphene oxide (GO)-IL and MoS 2 -IL solutions were prepared with the same volume concentration as that of Mo 2 CT x -IL.

Material characterization
The dried Mo 2 CT x MXene particles were tested by the atomic force microscope (Icon, Bruker, Germany) to measure the sizes of the Mo 2 CT x MXene nanoparticles. The crystal structures of the Mo 2 CT x MXene nanoparticles were detected using the X-ray diffractometer (XRD) (Bruker, Germany) with 2θ ranging from 5° to 65°. The morphology of the Mo 2 CT x MXene nanoparticles was analyzed by the high-resolution transmission electron microscope (HRTEM) (2100F, JEM, Japan). The original chemical composition of the Mo 2 CT x MXene nanoparticles was characterized by the X-ray photoelectron spectroscope (XPS) (PHI Quantera II, Japan). The chemical groups of Mo 2 CT x MXene and IL were detected by the Fourier transform infrared spectroscope (FTIR) (Vertex 70V, Bruker, Germany) and the Raman spectroscope (HR-800, Horiba, Japan), www.Springer.com/journal/40544 | Friction respectively. The average size of the Mo 2 CT x MXene was measured using a commercial zeta potential instrument (Zetasizer Nano Series, Malvern, UK).

Tribological test
Friction and wear tests were conducted using a universal micro-tribotester in a ball-on-disk mode (UMT-5, Bruker, USA). A fixed Si 3 N 4 ball was loaded onto a rotatable sapphire disk. Subsequently, 20 μL of the lubricant was dripped onto the contact surfaces in each test. The applied load of the ball varied from 1 to 6 N, corresponding to the initial contact pressures of 1.31-2.40 GPa at the interface. The sliding speed of the disk ranged from 0.004 to 0.375 m/s. All tests were conducted at a room temperature of 25 °C and a relative humidity of 20%-25%.

Surface characterization
The diameter of the wear scar and the roughness of the worn-out zone on the ball and plate were observed using a 3D white-light interferometer (Nexview, ZYGO Lambda, USA). The topographies of the worn zones were acquired by the scanning electron microscope (SEM) (Quanta 200 FEG, FEI, USA). The wear zones on the surfaces were analyzed using the XPS and Raman spectroscopy after the friction test. The crosssectional structures of the worn zones on the surfaces were observed using HRTEM.

Mo 2 CT x MXene characterization
Mo 2 CT x MXene was synthesized by etching Ga from Mo 2 Ga 2 C powders using etchants, and subsequently delaminated by sonication in water (Fig. 1). The Mo 2 CT x -IL solution was prepared by mixing a lithium hexafluorophosphate-based IL and a Mo 2 CT x MXene aqueous solution with a mass ratio of 1:5. After the synthesis of Mo 2 CT x MXene, the morphology of Mo 2 CT x MXene was first observed using SEM ( Fig. 2(a)), which showed a highly crystalline structure of Mo 2 CT x MXene nanoparticles.
The lattice structures of Mo 2 CT x MXene were investigated by XRD, as shown in Fig. 2(b), showing the first peak at 2 = 8.6°, which was smaller than that of the precursor Mo 2 Ga 2 C [39]. This is an indication  [42], respectively. For C 1s, four main peaks were located at 283.2, 284.6, 286.1, and 289.3 eV. These peaks were assigned to C-Mo, C-C, C-O, and O-C=O bonds, respectively. FTIR was used to detect the reciprocities between IL and Mo 2 CT x ( Fig. S1(a) in the Electronic Supplementary Material (ESM)). The peaks at 1,087 and 1,109 cm −1 proved the existence of C-O bonds, and the peak at 1,642 cm −1 proved the presence of Mo. Furthermore, the peaks at 2,875 and 2,938 cm −1 indicated the existence of C−H, and the peak at 3,305 cm −1 indicated the existence of hydroxyl groups [43]. The peaks in the Raman spectra located at 488 and 546 cm −1 are the standard peaks of Mo 2 CT x MXene ( Fig. S1(b) in the ESM), confirming the successful preparation of the Mo 2 CT x MXene. The size distribution of Mo 2 CT x MXene in IL was measured using a Zetasizer, which showed that the MXene in IL had an average size of 531 nm (Fig. S1(c) in the ESM).

Tribological performances
To estimate the macroscale tribological properties of Mo 2 CT x -IL, the friction experiments were performed at the Si 3 N 4 -sapphire interface with lubrication by IL, Mo 2 CT x aqueous solution, and Mo 2 CT x -IL, respectively, under a load of 3 N and a sliding speed of 0.1 m/s ( Fig. 3(a)). The results showed that the COF first displayed an unstable oscillation in the range of 0.08-0.1 from 0 to 1,200 s with lubrication by the Mo 2 CT x MXene aqueous solution. Subsequently, the COF started a sudden drop from 0.1 to 0.058. However, this declining tendency of the COF was not maintained for a long time, and it started to rise after 1,500 s. This performance indicates that the Mo 2 CT x nanoparticles can reduce friction, but they cannot result in an ultralow COF. Thereafter, the IL (without MXene) as a lubricant was tested, showing a robust state with a COF of approximately 0.05 in the entire friction test. This phenomenon indicates that IL has good stability as a lubricant, but it still cannot achieve superlubricity. Based on these results, Mo 2 CT x MXene nanoparticles were selected as lubricating nanoadditives to be introduced into the IL, and the COF of the Mo 2 CT x -IL solution showed a similar tendency to that of Mo 2 CT x during the initial period of 900 s. Subsequently, the COF gradually dropped from 0.09 to 0.005 during the period of 900-1,350 s, which entered the superlubricity regime. Finally, the COF maintained stably at 0.004, and it could be sustained for at least 4.5 h ( Fig. 3(b)), corresponding to a sliding distance of 1,620 m.
After the friction test, the wear scar and surface roughness on the Si 3 N 4 and sapphire surfaces were characterized using 3D white-light interferometry. The results showed that the diameters of the wear scar on the Si 3 N 4 surface were 55. 46  MXene nanoparticles as additives in IL could reduce friction and suppress wear to enhance the contact pressure. Generally, a high contact pressure is detrimental to the achievement of superlubricity. However, the introduction of Mo 2 CT x MXene additives in ILs could make the contact pressure of superlubricity exceed 1 GPa, which is far higher than that achieved by elastohydrodynamic lubrication (< 300 MPa) [44]. Meanwhile, the surface roughness of the wear scar on the Si 3 N 4 surface after lubrication by Mo 2 CT x -IL was 23.2 nm, which was significantly smaller than that with lubrication by IL (56.4 nm). The corresponding surface roughnesses of the wear tracks on the sapphire surface were 18.9 and 33.4 nm with lubrication by Mo 2 CT x -IL and IL, respectively.
An additional investigation of the influence of the sliding speed on COF was performed with lubrication by Mo 2 CT x MXene nanoparticles as additives in the IL (Fig. 3(c)). The COF decreased from 0.37 to 0.018 when the sliding speed increased from 0.004 to 0.05 m/s, which was noticeably in the non-superlubricity regime. Subsequently, the COF decreased to 0.004 when the sliding speed further increased to 0.075 m/s, which entered the superlubricity regime. This was followed by maintaining the superlubricity state (the COF varied slightly from 0.004 to 0.008) when the sliding speed increased from 0.075 to 0.25 m/s. When the sliding speed exceeded 0.3 m/s, the COF was greater than 0.01, which was indicative of the failure of the superlubricity state. The evolution of the COF with sliding speed was consistent with the Stribeck curve, indicating that the hydrodynamic effect may also influence the superlubricity state. When the sliding speed was extremely low, the hydrodynamic effect was weak, leading to an increase in the COF. This illustrates that there is an optimal range of sliding speeds in which a robust superlubricity state can be achieved with lubrication by Mo 2 CT x -IL.
Subsequently, a friction test was carried out with lubrication using different concentrations of Mo 2 CT x MXene as lubricating additives (from 0.021 to 0.416 wt%) in IL (Fig. 3(d)). The extremely low concentration (< 0.042 wt%) clearly cannot lead to the achievement of superlubricity. The COF decreased from 0.042 to 0.021 when the concentration of Mo 2 CT x MXene increased from 0.021 to 0.042 wt%. However, when the  | https://mc03.manuscriptcentral.com/friction concentration of Mo 2 CT x MXene reached 0.083 wt%, the COF decreased to 0.007, which entered the superlubricity regime. When the concentration of Mo 2 CT x -IL further increased from 0.083 to 0.332 wt%, the results showed that the COF first decreased slightly from 0.008 to 0.004 when the concentration of Mo 2 CT x MXene reached 0.166 wt%, and then it had a slight increase (the COFs were always less than 0.01). However, when the Mo 2 CT x MXene concentration reached 0.416 wt%, the superlubricity state was damaged with a high COF of 0.035. Moreover, the lubrication state became unstable with a violent fluctuation of the COF (Fig. S3 in the ESM). These results indicate that the COF was significantly influenced by the concentration of Mo 2 CT x MXene, and an optimal concentration range (0.083-0.332 wt%) was necessary to achieve the superlubricity state. The diameter and wear volume of the wear scar on the Si 3 N 4 surface were also measured after the friction test (Figs. S4 and S5 in the ESM), showing that suitable concentrations of 0.083-0.25 wt% of Mo 2 CT x could also suppress the surface wear and meanwhile achieve the superlubricity state, which leads to a high contact pressure (> 900 MPa) for superlubricity. The maximum contact pressure in the superlubricity state could reach 1.24 GPa with a concentration of 0.166 wt% (Fig. 3(d)). However, when the concentration was too low or too high, superlubricity could not be achieved, and there was also severe wear (corresponding to a low contact pressure). This is because there were not enough Mo 2 CT x nanoparticles to enter the contact zone when the concentration was extremely low, and the agglomeration of the Mo 2 CT x nanoparticles occurred when the concentration was very high, both of which were harmful for achieving superlubricity and reducing surface wear.
Next, the influence of the normal load on the COF with lubrication by the 0.166 wt% of Mo 2 CT x MXene as additives in IL was investigated (Fig. 3(e)). When a normal load of 1 N was applied, the final COF decreased to 0.015, which belongs to the nonsuperlubricity regime. After the normal load was increased to 2 N, the COF directly dropped to 0.006, which entered the superlubricity regime. The lowest average value of the COF, 0.004, was achieved by further increasing the normal load to 3 N. However, when the normal load exceeded 3 N, the COF gradually increased with increasing load and finally reached 0.021 at a normal load of 6 N. These results indicate that the optimal normal load range is 2-4 N, in which superlubricity can be achieved with Mo 2 CT x -IL. When the load was too low or too high, superlubricity could not be achieved, which may be because the tribofilms could not be completely formed with an extremely low load, and the tribofilms were damaged with a very high load. Compared with the corresponding contact pressures after the friction test, high contact pressures (> 900 MPa) could be achieved when the normal load was in the range of 3-5 N, and the maximal contact pressure for superlubricity could reach 1.42 GPa at a normal load of 4 N (Fig. S6 in the ESM).
To verify the excellent lubrication performance of the Mo 2 CT x -IL, two other typical 2D materials (GO and MoS 2 ) were selected and mixed with IL at the same volume concentration of Mo 2 CT x , referred to as GO-IL and MoS 2 -IL, respectively. The COF results in Although the GO-IL could achieve superlubricity, the wear scar was much larger than those of Mo 2 CT x -IL and MoS 2 -IL, corresponding to a significantly lower contact pressure. This result confirms that molybdenum has much better antiwear ability than graphene.

Analysis of wear interface
After the friction test, the worn-out zone on the Si 3 N 4 surface was examined using SEM ( Fig. 4(a)). Some adsorbed substances can be observed on the Si 3 N 4 surface in the worn-out zone after lubrication by Mo 2 CT x -IL. To analyze the composition of this area, energy-dispersive X-ray spectroscopy (EDS) mapping was performed in the worn-out zone (Figs. 4(b) and 4(c)). Mo and P were detected on the Si 3 N 4 surface based on the strong contrast signals in the EDS diagram. The wear track on the sapphire surface was www.Springer.com/journal/40544 | Friction also examined using SEM with EDS (Figs. 4(d)-4(f)), in which adsorbed nanoparticles were observed in the wear track. The EDS results also showed a strong contrast in the Mo and P elements in the wear track. According to these results, it can be inferred that there may be Mo 2 CT x MXene nanoparticles and IL molecules adsorbed in the worn-out zone through physical adsorption or tribochemical reaction during the friction test.
To determine the change in chemical composition on the friction surfaces and possible tribochemical reaction in the friction process, the worn-out zones on the Si 3 N 4 and sapphire surfaces after lubrication by Mo 2 CT x -IL were also characterized by XPS (Fig. 5 and Fig. S8 in the ESM). On the worn-out surface of the Si 3 N 4 , the main element of C 1s contained four main peaks located at 283.0, 284.8, 286.6, and 288.4 eV (Fig. 5(a)), corresponding to C-Mo, carbon in aliphatic chains (C-C), C-O, and O-C=O bonds, respectively. The detected C-O bond may be derived from the IL or oxidation of the Mo 2 CT x MXene nanoparticles. The existence of the C-Mo and O-C=O bonds mainly originated from the Mo 2 CT x MXene (Fig. 2(d)), proving that there were Mo 2 CT x nanoparticles adsorbed on the Si 3 N 4 surface. For the Mo element ( Fig. 5(b)), there were four main peaks located at 228.6, 230.0, 232.1, and 234.4 eV. These peaks were assigned to Mo-O, Mo-C, Mo 3d 5/2 , and Mo 3d 3/2 , respectively. In comparison with the original XPS results of Mo 2 CT x MXene in Fig. 2(c), the existence of the Mo-O bond  showed that some Mo 2 CT x MXene participated in the oxidation reaction and produced molybdenum oxides (Mo x O y ) in the tribofilm. The production of Mo x O y improved the antiwear ability due to the molybdenum oxide, which is a wear-resistant material [45,46]. For the P 2p, strong peaks were located at 134.5 and 137.3 eV (Fig. 5(c)), which were assigned to P 2p and P-O bonds, respectively. This indicates that phosphorus oxides (P x O y ) were produced on the worn-out surface of Si 3 N 4 . There were also two peaks of F 1s located at 684.6 and 687.8 eV (Fig. S8(a) in the ESM), which corresponded to fluorides and C-F bonds, respectively [47]. This result was consistent with Ref. [11], in which the anion of AlF , respectively. The XPS results of Al 2p also prove the generation of Al-F bonds, indicating a tribochemical reaction between Al 2 O 3 and  6 PF anions (Fig. S8(d) in the ESM). Combining these XPS results and the EDS mapping in Fig. 4 indicates that there were tribofilms formed in the worn-out zone on the Si 3 N 4 and sapphire surfaces, which were mainly composed of Mo x O y , P x O y , and fluorides.
To reveal the structure of the tribofilm further, cross-sectional layers of the worn-out zone on the Si 3 N 4 and sapphire surfaces were prepared using a focused ion beam after lubrication with Mo 2 CT x -IL (the sampling regions are marked in Fig. S9 in the ESM). HRTEM was then conducted on the crosssectional samples of the tribofilm in the contact zone (Fig. 6). The thickness of the tribofilm on the Si 3 N 4 surface was approximately 45.6 nm ( Fig. 6(a)). Furthermore, EDS in line scan mode was performed to analyze the elemental distribution and determine the composition of the tribofilm (Fig. 6(c)). Along with the elemental distributions of P, Mo, and O, the production of P x O y and Mo x O y occurred in the tribofilm with approximate thicknesses of 25 and 20 nm, respectively. The formation of Mo x O y and P x O y confirms that the tribochemical reaction occurred for the Mo 2 CT x MXene nanoparticles and IL on the Si 3 N 4 surface, in accordance with the XPS results in Fig. 5. There was an overlapping region for Mo x O y and P x O y production, indicating that the tribofilm had a composite structure. Figure 6(c) shows that the Si 3 N 4 surface also participated in the tribochemical reaction (Si 3 N 4 + 6H 2 O → 3SiO 2 + 4NH 3 ), leading to the formation of silica on the Si 3 N 4 surface with an approximate thickness of 12 nm. Additionally, the structure of Mo 2 CT x MXene in the tribofilm was clearly observed in the enlarged HRTEM image (area A in Fig. 6(e)), which showed a multilayer structure with an interlayer spacing of approximately 0.7 nm. The detected layers of Mo 2 CT x MXene were parallel to the sliding direction, which had a beneficial effect on providing low shear strength. Raman detection was performed on the worn-out zone after the friction test (Fig. S10 in the ESM), and the Raman spectra of the worn-out zone of the Si 3 N 4 surface ( Fig. S10(a) in the ESM) exhibited the same bands as the original Raman results in Fig. S1(b) in the ESM, further demonstrating that Mo 2 CT x MXene nanoparticles were adsorbed on the worn-out zone of the Si 3 N 4 surface.
On the sapphire surface, a tribofilm with a thickness of 57.2 nm was also observed ( Fig. 6(b)). The EDS results show that Mo x O y and P x O y were produced in the tribofilm on the sapphire surface with thicknesses of 31 and 30 nm, respectively (Fig. 6(d)). The highermagnification HRTEM image (area B) displayed some layer structures with an interlayer spacing of 0.57-0.65 nm in the tribofilm, which corresponded to the Mo 2 CT x nanoparticles. The inherent thickness of Mo 2 C is approximately 5 Å [39], and the increase in the interlayer spacing is caused by surface termination or possible intercalated species. The Raman results also prove that the Mo 2 CT x MXene nanoparticles were www.Springer.com/journal/40544 | Friction adsorbed on the sapphire surface ( Fig. S10(b) in the ESM).

Mechanism of ultrahigh contact pressure under superlubricity
According to previous studies, it is difficult to achieve superlubricity under a contact pressure of more than 600 MPa (in the absence of coatings). However, in this work, the extended pressure limit of superlubricity was on the order of gigapascals. This implies a specific lubrication mechanism rather than the traditional elastohydrodynamic lubrication when the Mo 2 CT x MXene nanoparticles were used as lubricant additives, which was beneficial for achieving superlubricity under ultrahigh pressures. To determine the lubrication regime in the superlubricity state, the thickness (h c ) of the lubricating film at the Si 3 N 4 -sapphire interface was estimated using the Hamrock-Dowson formula [48]. The calculation in Section S12 in the ESM showed that the liquid film thickness in the superlubricity state was in the range of 41.95-45.67 nm (45.67 nm for 2 N, 43.42 nm for 3 N, and 41.95 nm for 4 N, respectively). The ratio of the liquid film thickness to the equivalent roughness of the two worn-out zones   52, which implies that the lubrication regime was located in the mixed lubrication regime (containing the boundary and elastohydrodynamic lubrication).
For the running-in period, the COF oscillated unsteadily at 0.09, and slight wear occurred on the Si 3 N 4 and sapphire surfaces with lubrication by Mo 2 CT x -IL. Although the COF value was larger than that with IL lubrication during the running-in period, the wear was significantly lower than that with IL lubrication. This phenomenon demonstrates that Mo 2 CT x MXene nanoparticles play a dominant role in the inhibition of wear during the running-in period. It is speculated that the Mo 2 CT x MXene nanoparticles are adsorbed in the worn-out zone of the friction pairs from the beginning of the friction test, which can prevent direct contact between the Si 3 N 4 and sapphire surface. To confirm that the Mo 2 CT x MXene nanoparticles were adsorbed on the friction surfaces within a very short period, XPS was performed on the contact zone of the Si 3 N 4 surface after lubrication by Mo 2 CT x -IL for running-in periods of 1 and 3 min. The XPS results show a strong Mo 3d and C 1s signal in the contact zone of the Si 3 N 4 surface (Fig. S11 in the ESM), which are typical peaks of Mo 2 CT x MXene nanoparticles (Figs. 2(c) and 2(d)). These results illustrate the excellent adsorption of Mo 2 CT x MXene nanoparticles from the beginning of the friction test, which prevented the friction pairs from direct contact and thereby suppressed surface wear. Furthermore, the existence of the Mo-O bond during the running-in period indicates that Mo 2 CT x -IL participated in the tribochemical reaction from the beginning of the test. A protective film containing Mo x O y formed, which could further decrease abrasion owing to the excellent antiwear property of Mo x O y [49]. Compared with other 2D materials, such as GO and MoS 2 , Mo 2 CT x MXene nanoparticles exhibited a much better reduction in wear (Fig. S7 in the ESM). Therefore, the antiwear properties of Mo 2 CT x MXene nanoparticles do not only depend on its layer structure, but also are related on the antiwear properties of the Mo 2 CT x MXene nanoparticles themselves, which are not presented in the traditional 2D materials, such as GO and MoS 2 . After the running-in period was complete, the COF was directly reduced from 0.09 to 0.004, entering the superlubricity regime.
During the superlubricity period, the lubrication state entered the mixed lubrication regime. A composite structure of tribofilms containing Mo x O y , P x O y , silica, and fluorides was formed on the friction surfaces by tribochemical reactions, which can provide extremely low shear strength for the boundary lubrication. Here, the ultralow shear strength occurred for three main reasons. First, Mo x O y and P x O y containing -OH groups are favorable for combining with water [50,51]; thus, a water layer can form through double hydrogen bonding between the -OH groups and water molecules, which could reduce the shear strength of the tribofilm. Second, the silica colloid layer could promote the boundary lubrication because it was an excellent boundary lubricating material in a water environment [52], which could further reduce the shear strength of the tribofilm. Third, the existence of Mo 2 CT x MXene nanoparticles in the tribofilm also provided low shear strength, which assisted in reducing friction and wear. Therefore, the COF of the boundary lubrication was very low owing to the combined action of these three principal factors.
For the elastohydrodynamic lubrication, the viscosity of Mo 2 CT x -IL increased from 1.26 to 14.36 mPa·s after the running-in period, which was beneficial for the formation of the elastohydrodynamic lubricating film. Consequently, under the synergistic effect of the boundary lubrication and elastohydrodynamic lubrication, an ultralow COF under an ultrahigh contact pressure was achieved. The results and analysis demonstrate that the introduction of Mo 2 CT x MXene additives into IL can effectively enhance the lubrication performance and suppress surface wear, which further enhances the contact pressure for superlubricity (exceeding 1 GPa). This magnitude of improvement indicates that ultrahigh load capacity and ultralow friction can be achieved simultaneously by MXene-liquid coupling lubrication. The outstanding friction performance of Mo 2 CT x MXene nanoparticles expands the feasibility of MXene materials in the field of lubrication and their potential application in the mechanical industry. Therefore, the synergistic lubrication effect between Mo 2 CT x MXene nanoparticles www.Springer.com/journal/40544 | Friction and ILs has important implications for the development of the liquid lubrication with superlubricity and super-wear-resistance performance.

Conclusions
Robust superlubricity can be achieved using Mo 2 CT x MXene nanoparticles in combination with IL at the Si 3 N 4 -sapphire interface. In this study, the upper limit of the contact pressure for the liquid superlubricity state was greatly improved to 1.42 GPa at the macroscale. The formation of the composite tribofilms induced by the tribochemical reaction, mainly containing Mo x O y , P x O y , and adsorbed Mo 2 CT x MXene, suppressed the surface wear and provided an extremely low shear strength, which is the origin of the superlubricity state under ultrahigh contact pressure. This work provides a new method for achieving macroscale superlubricity under ultrahigh contact pressure through the synergistic effect of MXene and liquid molecules. The method has great potential for the application of superlubricity in mechanical systems. you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.