The difficulty of predicting the wear performance of materials is on one hand due to the multitude of physical phenomena taking place at contacting surfaces, but is also a result of missing connections between these mechanisms and the macroscopic progression of wear. The simple case of dry contact between materials of comparable hardness leads to adhesive wear by the formation of wear particles. Recent computer simulations demonstrated that a critical length scale for the local contacts between rough surfaces determines if wear particle detachment can take place [1]. This length scale can be modified by local geometry and the strength of adhesion [2]. This mechanistic view alone, though, is insufficient to arrive at reasonable wear laws.

The first complication is the evolution of the system. The contact map of two sliding surfaces is modified not only by the wear particle formation, but the evolution of the particles themselves. Using molecular dynamics simulations, we find that rolling wear particles grow by wearing the contacting bodies and leave behind fractal roughness [3]. The wear rate due particle formation is expected to be higher than the wear rate due to particle growth [4], but how these relate to each other is still an unsolved question, since this depends on the dynamics of the wear particles over long sliding distances inaccessible to atomistic simulations. We can nevertheless provide nanoscale wear equations for short sliding distances [4].

The second complication is to understand how the aforementioned critical length scale translates into macroscopic wear laws. Since this length is inversely proportional to the square of the material’s hardness, hard materials can form wear particles at small contacts and soft materials only at large contacts. This would mean, however, that hard materials can form wear particles more easily and should thus wear more – a contradiction of typical experimental evidence. The solution is found by treating adhesive wear as a process that happens during sliding: hard materials will never form large contacts and thus large wear particles, because they are worn off before growing big. Soft materials form bigger wear particles and thus wear more. We present a simple numerical model of this based on elastoplastic contact solutions [5]. While this model is likely overly simplistic for quantitative predictions, it does not require any empirical parameters, provides for the first time a connection between nanoscale wear mechanisms and mesoscale contact maps, recovers qualitative trends, and represents a big step towards modeling wear from first principles.

[1] Aghababaei et al., Nat. Commun. 7 (2016)
[2] Brink and Molinari, PR Mater. 3 (2019)
[3] Milanese et al., Nat. Commun. 10 (2019)
[4] Brink et al., PR Mater. 6 (2022)
[5] Brink et al., JMPS 147 (2021)