An analysis method for atomistic abrasion simulations featuring rough surfaces and multiple abrasive particles

• Published in: Computer physics communications 2014-10, Vol.185 (10), p.2456-2466
• Main Authors: Eder, S.J., Bianchi, D., Cihak-Bayr, U., Vernes, A., Betz, G.
• Format: Article
• Language: English
• Subjects: Abrasion; Abrasion resistance; Abrasive wear; Atomistic abrasive wear; Computer simulation; Contact; Mathematical models; Molecular dynamics; Multiple particle abrasion; Wear; Wear particles
• ISSN: 0010-4655; 1879-2944
• DOI: 10.1016/j.cpc.2014.05.018

We present a molecular dynamics (MD) model system to quantitatively study nanoscopic wear of rough surfaces under two-body and three-body contact conditions with multiple abrasive particles. We describe how to generate a surface with a pseudo-random Gaussian topography which is periodically replicable, and we discuss the constraints on the abrasive particles that lead to certain wear conditions. We propose a post-processing scheme which, based on advection velocity, dynamically identifies the atoms in the simulation as either part of a wear particle, the substrate, or the sheared zone in-between. This scheme is then justified from a crystallographic order point of view. We apply a distance-based contact zone identification scheme and outline a clustering algorithm which can associate each contact atom with the abrasive particle causing the respective contact zone. Finally, we show how the knowledge of each atom’s zone affiliation and a time-resolved evaluation of the substrate topography leads to a break-down of the asperity volume reduction into its components: the pit fill-up volume, the individual wear particles, the shear zone, and the sub-surface substrate compression. As an example, we analyze the time and pressure dependence of the wear volume contributions for two-body and three-body wear processes of a rough iron surface with rigid spherical and cubic abrasive particles.