Date Published: January 03, 2018
Publisher: John Wiley and Sons Inc.
Author(s): Jinjin Li, Tianyang Gao, Jianbin Luo.
2D or 3D layered materials, such as graphene, graphite, and molybdenum disulfide, usually exhibit superlubricity properties when sliding occurs between the incommensurate interface lattices. This study reports the superlubricity between graphite and silica under ambient conditions, induced by the formation of multiple transferred graphene nanoflakes on the asperities of silica surfaces after the initial frictional sliding. The friction coefficient can be reduced to as low as 0.0003 with excellent robustness and is independent of the surface roughness, sliding velocities, and rotation angles. The superlubricity mechanism can be attributed to the extremely weak interaction and easy sliding between the transferred graphene nanoflakes and graphite in their incommensurate contact. This finding has important implications for developing approaches to achieve superlubricity of layered materials at the nanoscale by tribointeractions.
Friction and wear are the two major sources of considerable energy consumptions in mechanical sliding systems, especially in the nano and micromachines.1 The superlubricity, a physical regime in which the friction between two sliding surfaces nearly vanishes (or the sliding friction coefficient is in the level of 0.001),2 is one of the most effective approaches to reduce the frictional energy dissipation and meanwhile provide a near‐wearless condition. There are a series of solid lubricants, such as the diamond‐like carbon (DLC) film and 2D or 3D layered materials, including graphene, graphite, boron nitride, and molybdenum disulfide (MoS2), that have been observed to achieve the superlubricity state under certain special conditions.3 For example, Martin et al. observed the superlow friction of a MoS2 coating in the ultrahigh vacuum, where the atomic origin of superlubricity was from the incommensurate contact of MoS2 basal planes.4 Dienwiebel et al. studied the friction between the graphite nanoflake and sheet, observing the rotation angle dependent superlubricity phenomenon.[[qv: 3c,5]] Liu et al. found that the micrometer‐scale graphite flakes were retracted back to their initial positions after displacement from the equilibrium configuration due to the structural superlubricity between graphite layers in the sliding direction.[[qv: 3d,6]] Feng et al. observed the facile translational and rotational motions between graphene nanoflakes (GNFs) and graphene surface at an extremely low temperature.7 Only recently, superlubricity has also been achieved for dissimilar surfaces, like amorphous antimony or crystalline gold nanoparticles sliding on graphite,8 graphene nanoribbons sliding on gold,9 and in tribometer sliding experiments between DLC and graphene by the formation of graphene nanoscrolls surrounding the nanodiamond particles.10
The silica particle with a radius of 11.5 µm was first glued onto a rectangular tipless cantilever end by using epoxy glue, yielding a SiO2 probe (Figure1a). The topography on the top region of the probe was obtained by the atomic force microscopy (AFM), as shown in Figure 1b,c. The surface of the probe was not atomically smooth, and instead a series of asperities appeared on the surface with a roughness of ≈1 nm over an area of 90 000 nm2. From the cross‐sectional height profile in Figure 1d,e, the radii (r) of these asperities were in the range of 50–80 nm, with a peak‐to‐peak distance of ≈30 nm. Meanwhile, a highly ordered pyrolytic graphite (HOPG, 0.4° mosaic spread) substrate was freshly cleaved by using the adhesive tape to give a clean atomically flat surface. The frictional forces as functions of normal loads were measured on AFM, by driving the probe sliding on the freshly cleaved HOPG as the load was increased gradually, as shown in Figure 1f.
In summary, our work has demonstrated that the superlubricity of graphite can be achieved by the formation of multiple transferred GNFs on the asperities of silica surfaces via the tribointeractions in the presliding process. The friction coefficient can reduce to as low as 0.0003 and remain very stable at a maximal local contact pressure of up to 700 MPa. The superlubricity state is independent of the sliding velocity, rotation angle, and surface roughness, which can be attributed to the extremely weak interaction and easy sliding between the transferred GNFs and graphite in their incommensurate contact. Our finding provides a possible method to achieve the robust superlubricity state of layered materials at the nanoscale by taking advantage of the tribointeractions that enables the transfer of 2D layered nanoflakes onto the opposite surface.
Frictional Force Microscopy Tests: The friction measurements were performed using the Asylum Research MFP‐3D AFM in contact mode. A rectangular tipless cantilever (TL‐CONT) with a normal spring constant of 0.02–0.77 N m−1 and a lateral spring constant of 0.01–0.1 N m−1 was used, and the silica particles (R = 11.5 µm, elastic modulus ≈ 50 GPa) were glued onto the cantilever ends by using epoxy glue, yielding a SiO2 probe. Meanwhile, a HOPG (0.4° mosaic spread, elastic modulus ≈ 18 GPa) substrate was freshly cleaved by using the adhesive tape to give a clean atomically flat surface. The spring constant was determined by the frequency method,20 and the lateral detector sensitivity was obtained by using a diamagnetic lateral force calibrator.21 The frictional forces as functions of normal loads were measured by driving the probe sliding on HOPG as the load was increased (loading from 0 to 500 nN) under ambient conditions, at a temperature of 25 ± 2 °C and a relative humidity of 40–70%. The maximal diameter of contact area is about 130 nm at the applied load of 460 nN (estimated by the Hertz contact theory). Thus, the lateral scanning area was set in the range of 0.6 × 0.6 µm2 to 20 × 20 µm2 (it was equally divided into 16 parts as measured in the loading process), and the scanning velocity was set in the range of 0.3–30 µm s−1. The presliding process was performed at an applied load of 185 nN, and a scanning velocity of 10 µm s−1. The scanning area was set to 10 × 10 µm2 on different positions, and the presliding was stopped when the frictional force was observed to have a significant reduction. The distance for the presliding was varied from 102 to 104 µm, which was dependent on the SiO2 probe. The frictional forces before and after the presliding process were determined from the difference in the lateral force detector signal in one complete friction loop (20 friction loops were averaged for every load) and the lateral detector sensitivity.
The authors declare no conflict of interest.