Towards frictionless surfaces - ultra-low friction materials for engineering applications
Is it possible to produce frictionless material? Self-adaptive nanostructures could lead to huge friction reduction, say Dr Tomas Polcar from the University of Southampton, UK, and Professor Albano Cavaleiro from the University of Coimbra, Portugal, who explain their recent progress in designing ultra-low friction materials for engineering applications.
Friction is part of our everyday lives. We can use our fingers to estimate it. Everyone knows that friction should be high to avoid a slippery fall, or low to enjoy skiing, and we know that lubrication can significantly reduce friction, which is why ski edges are waxed for peak performance. Friction and wear also have a significant impact on society. It is estimated that the energy required to overcome friction and its associated maintenance costs totals 5% of the GDP of an industrialised country. The majority of this energy comes from non-renewable sources. Friction is directly related to the dissipated energy, which is mostly transformed into heat, causing local structural surface changes, such as oxidation, and changing the mechanical properties of materials in contact. Moreover, high friction is typically related to extensive wear, meaning most materials cannot be used in dry sliding conditions. As such, all problems associated with friction and wear are surface-related phenomena. There are only two ways to reduce the friction of bulk materials – surface treatment or use of lubricants, the latter being the widely used, traditional solution. However, lubricants are mostly produced from fossil fuels, and oil additives often contain dangerous and toxic chemicals that have significant impact on the environment. Furthermore, the wear of materials decreases their lifespan, which is associated with high production costs, extensive use of limited materials, and issues related to disposal and recycling. It is estimated that 16% of global fuel energy is wasted due to friction and wear – any improvement in this area immediately saves millions of tonnes of fuel and reduces the corresponding amount of CO2 released into atmosphere. Reducing friction and wear is therefore not just a way to save money, but also makes an important contribution towards a sustainable society.
The friction coefficient of traditional engineering materials such as steels, alloys, or ceramics is typically much greater than 0.5 for dry sliding, but lower than 0.1 for lubricated sliding. Over the last few years, this has posed a continual technical challenge to R&D teams worldwide, to find surface treatment that reduces dry friction to, or even below, the level typical for lubricated contact while also improving wear resistance.
At atomic level
The search for ultra-low friction materials must inevitably start at the atomic scale. Recent advances in atomistic simulations open new horizons to further optimise coating architecture by addressing fundamental issues of nanoscale sliding. There is a large range of materials with super-lubricity potential. However, by limiting the search to typical engineering conditions, such as moderate pressure, high sliding speeds and temperatures between -50–250°C, only two crystals exhibit almost frictionless sliding – graphite and transition metal dichalcogenides (TMD, which are molybdenum and tungsten disulphides and diselenides). Both these materials exhibit a highly anisotropic layered structure with very weak bonding between layers – writing with pencil is a nice demonstration of graphite easy slip – and atomistic simulations suggest that extremely low friction of these structures could be achieved.
Not surprisingly, carbon- and TMD-based materials, which have been used for decades as special additives to oil lubricants, are the most attractive ultra-low friction candidates. Diamond-like carbon (DLC) based films, the most widely applied for self-lubricating applications, exhibit low friction in humid air due to the formation of thin graphite layer on contact. However, their friction is typically much higher in dry air or vacuum. Although pioneering work on highly hydrogenated DLC by Professor Ali Erdemir at the Argonne National Laboratory, USA, showed that such coatings exhibited super-lubricity, their unique performance was limited to sliding in dry nitrogen. Doping with other elements could improve specific frictional properties, but it remains clear that multipurpose DLC coatings with low friction in all sliding regimes cannot be made.
TMDs are complementary to DLC coatings. In vacuum they exhibit ultra-low friction (the lowest ever measured for macroscopic contact) and they are often used in space applications. Their friction decreases with increasing load, violating the classical Amontons’ Law, and they withstand enormous contact pressures. On the other hand, the presence of oxygen and, in particular, air humidity in the sliding atmosphere hinders low-friction behaviour and accelerates wear. Similar to carbon, TMD can be prepared as bulk material, thin film, nanotube, fullerene-like particle or even monolayer. The latter three forms are extensively studied as oil additives or solid lubricants in small-scale contacts such as microelectromechanical systems (MEMS), although thin films, mostly those prepared by physical vapour deposition (PVD) methods, are still the most promising for engineering applications. While the mechanical properties (particularly load-bearing capacity) and wear resistance can be improved by doping or alloying of TMD with metals – indeed, molybdenum disulphide (MoS2) doped with titanium was successfully introduced to market about 10 years ago – the sensitivity to humidity remains.
The next step was to find a way of creating a film with the beneficial properties of both carbon and TMD. The answer was almost too obvious – combine these two materials together. The first such design suggested by Professor Andrey Voevodin from Air Force Research Laboratory, in the USA, was very complex. The deposited film comprised a nanocomposite containing nanograins of hard tungsten carbide and self-lubricant tungsten disulphide, embedded into an amorphous carbon matrix. Tests showed that the friction resembled that achieved for TMD in dry nitrogen and DLC in humid air. The concept was nicknamed chameleon, due to the expected reaction of the film to the sliding conditions. While the significantly lower friction in non-humid atmosphere was a great achievement, the anticipated benefits to aerospace applications were only marginal.
Because attempts to further improve TMD behaviour by different deposition methods or alloying with compounds did not prove successful, it was obvious that a radically new functional concept would be required to prepare ultra-low friction coatings. The conventional approach to coating design was to deposit an optimal nanostructure for specific conditions. We decided instead to prepare a self-adaptive coating that would produce ultralow friction surface almost regardless of sliding conditions.
We knew that well-oriented TMD crystals could be frictionless, but how could we produce these crystals on the surface and simultaneously replenish them during wear? We again started with a combination of carbon and TMD, but this time we prepared a hair-like nanostructure with separated molecular layers of dichalcogenides, which were randomly distributed in amorphous carbon matrix. During the sliding process, a thin tribolayer comprising near-optimally oriented TMD crystals was produced on the uppermost surface. To have such a stochastic, non-equilibrium process as macroscopic sliding resulting in the perfect formation of several nanometres-thick crystals covering the surface, smoothened to nanometre-scale roughness, was really something. This kind of tribolayer forms in all environments, from vacuum to humid air, and exhibits friction coefficients 0.05 or less. Interestingly, the tribolayer provides both low friction and resistance to environmental attack. Further research in this area has recently yielded ultra-low friction coatings with coefficients of less than 0.003, and these are currently being investigated by several groups in Europe as one of the most promising candidates for space applications.
While existing films were originally designed for dry sliding, lubrication is often unavoidable. A typical example is an automotive engine, where an increasing number of parts coated with diamond-like carbon (DLC) films are in contact with oil lubricants. However, the chemical inertness of these films, while beneficial for corrosion resistance, limits their interaction with oil additives.
Consequently, oil lubrication of DLC coated surfaces proves ineffective and the friction coefficient might be even higher than that of uncoated surfaces. To reduce friction, MoS2 containing anti-friction modifiers are often added into oils. Moreover, a thin MoS2 tribolayer could be chemically formed on the sliding surfaces from various compounds containing molybdenum and sulphur. By applying solid lubricant coating with ultra-low MoS2 (or TMD in general) tribolayer on automotive engine parts, it is possible to simultaneously decrease the friction and fuel consumption in the boundary lubrication system, and minimise – or even eventually eliminate – some of the oil additives. This in turn could further reduce both cost and environmental impacts.
This is just one example of nanostructured design of ultra-low friction coatings. Similar concepts, such as formation of an optimum sliding interface during the wear process, are intensively studied for other coating systems. In the USA, scientists at the Argonne National Laboratory are investigating coating systems that can replicate low-friction graphitic tribolayer typical for DLC coatings – in other words, the synthesis of a graphite-rich thin tribolayer during sliding via chemical reaction between the coating surface and oil.
It is apparent that all of these recent achievements in the field of super-low friction coatings are based on the ability to provide an optimum ultra-low friction interface. Self-adaptive coatings are here to stay.
For further information, contact Dr Tomas Polcar: T.Polcar@soton.ac.uk