It would appear that primitive man had a clear understanding of the importance of friction and lubrication — transportation of massive objects by the Egyptians in 2000 BC, for example, was facilitated by using sledges dragged over lubricated wooden boards. The role of a lubricant is similar to that of a peace-keeping force: its main function is to prevent the opposing surfaces from coming into close contact with each other at the atomic level. Most lubricants are liquids, but when the operating conditions (such as high or low temperature, or vacuum) are beyond the liquid realm, attention turns to solid materials. The most common solid lubricants are graphite and the transition-metal dichalcogenides, MX2(where M is molybdenum or tungsten, and X is sulphur, selenium or tellurium). But both graphite and the metal dichalcogenides have certain inherent deficiencies. The synthesis of hollow nanoparticles of tungsten disulphide (WS2) similar to fullerenes and nanotubes, reported by Tenne and co-workers on page 791 of this issue1, offers some exciting possibilities for a new generation of solid lubricants.

First, how do traditional solid lubricants work? The low friction of both graphite and metal dichalcogenides is usually due to interplanar mechanical weakness, intrinsic to their crystal structures. For example, WS2crystallizes in the hexagonal structure, in which a sheet of tungsten atoms is sandwiched between two hexagonally packed sulphur layers (Fig. 1). The bonding within the S-W-S sandwich is covalent, whereas the sandwiches themselves are held together by weak Van der Waals forces, resulting in interplanar mechanical weakness. Under the action of a shear force, intracrystalline slip occurs in the weak interplanar regions. This mechanism is responsible for the formation of smooth transfer films by wear: the new surfaces, created by separating the weakly bonded sandwiches, are quite inert. They can easily slide back and forth over one another (by intercrystalline slip), thereby providing lubrication.

Figure 1
figure 1

Crystal structure of hexagonal tungsten disulphide, 2H-WS2.

But there is a major obstacle to lubrication by metal dichalcogenides: the presence of unsaturated or dangling bonds. In a typical layered structure (2H-WS2), the obvious source of dangling bonds is an edge plane. Furthermore, the propagation of a pre-existing crack in a WS2film will break a few covalent bonds and create more unsaturated bonds. If sliding takes place in humid air, such activated surfaces can instantly react with moisture and oxygen in the surrounding environment, forming WO3. Thus, the tribological (friction and wear) behaviour of metal dichalcogenides is strongly dependent on their environment. For instance, when the test environment is switched from dry nitrogen to humid air2, the friction coefficient of a typical WS2film can rise from 0.03-0.04 to 0.15-0.20, decreasing its wear life by several orders of magnitude. In contrast, graphite loses its lubricating property in a vacuum, as the complete absence of adsorbed vapours makes it difficult to shear the layers. According to our current understanding, no material can act as a solid lubricant in all these environments. Has such a material now been discovered?

Tenne and co-workers3 first observed closed structures of WS2, ranging in size from 10 to 100 nm, when they annealed tungsten films at 1,000 °C in a reducing H2S atmosphere (see Fig. 2). The distance between two fringes in their lattice images is 0.615 nm, the same as the distance between two neighbouring sandwiches in an open 2H-WS2structure. Tenne and co-workers3 argue that the curling and closure of WS2networks was probably induced by thermal stresses (1,000 °C is a fairly high temperature) and facilitated by structural defects such as edge dislocations, or by contaminant atoms, such as oxygen, that came from the quartz substrate. The authors have also made nanoparticles of WS2(HN-WS2) in gram quantities by a sol-gel reaction1.

Figure 2
figure 2


High-resolution electron micrograph of a WS2nanoparticle.

The dangling bonds that are responsible for causing oxidation in layered 2H-WS2structures are no longer present in HN-WS2particles. Does this mean that we now have a super solid that can effectively lubricate in all environments? The intracrystalline slip that is responsible for creating smooth transfer films is no longer present in this powder, as appreciable shear between inner and outer closed layers is almost impossible. By the same token, HN-WS2powder cannot be burnished like the traditional layered structures. Indeed, Tenne and colleagues1 had to roughen the substrate and use a drop of mineral oil to avoid runaway of the nanopowder during friction measurements. Yet the friction coefficient and wear rate for nano WS2powder measured in oil were superior to those of traditional WS2and MoS2powders1. This suggests that a different mechanism is controlling the tribological behaviour of nanoparticles.

The mechanism of lubrication in the case of hollow nanopowder appears to be one of rolling and not sliding. By virtue of their extremely small size, the HN particles can fill the submicrometre-sized valleys between asperities and prevent asperity contact between mating surfaces. This, coupled with their rounded shape, opens the possibility that they undergo rolling friction at the nanometre level. Furthermore, elimination of dangling bonds makes the HN particles chemically inert, so the particles are less likely to stick to the substrate, or to one another. Their hollow-cage structure imparts high elasticity, and elastic deformations (as opposed to inelastic deformations) diminish the energy dissipation associated with friction. These are some of the characteristics that make the hollow nanoparticles of metal dichalcogenides potential candidates for miniature rolling-element bearings.

Perhaps, in the coming decades, we might see a micrometre-sized rolling-element bearing using hollow nanotubes of metal dichalcogenide. But even in the near future, these small, round, inert nanoparticles should find some applications in powder lubrication, and as additives in lubricating oils.