Reproducibility of atomistic friction computer experiments: a molecular dynamics simulation study

ABSTRACT

The elementary processes of friction in contact processes of two solid bodies occur on the nanoscale and are difficult to study experimentally. Therefore, molecular dynamics simulations are often used for their elucidation. In these studies, usually, only a single simulation is carried out for each scenario and the resulting observables are evaluated. In the present work, the reliability and reproducibility of measured observables from such nanoscopic contact process simulations are assessed by means of their statistical uncertainties. Therefore, the computer experiment is carried out not only once, but it is repeated eight times, with individual runs that only differ in the initial thermal motion. This set of replicas enables an assessment of observations for distinguishing reproducible physical phenomena from stochastic coincidence. In this way, a dry and a lubricated nanoscale scratching process were studied, in which a cylindrical indenter carried out two sequential movements: it first penetrates a substrate vertically and then scratches laterally, which causes elastic and plastic deformation of the substrate. In the lubricated case, the indenter was fully immersed in the fluid. Substrate, indenter, and fluid were described by suitably parametrised Lennard–Jones potentials. Various mechanical and thermodynamic process properties were monitored in all simulation runs. The results are compared and evaluated statistically.

KEYWORDS:

• Friction simulation
• molecular dynamics
• statistical uncertainty
• dislocation analysis

Acknowledgments

The present research was conducted under the auspices of the Boltzmann-Zuse Society of Computational Molecular Engineering (BZS). The simulations were carried out on the ELWE at Regional University Computing Center Kaiserslautern (RHRK) under the grant TUKL-MTD as well as on the SUPERMUC at Leibniz Supercomputing Centre (LRZ) Garching within the computing project (AMSEL)² (pn56mo).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The authors gratefully acknowledge financial support by the Deutsche Forschungsgemeinschaft (DFG) within IRTG 2057 ‘Physical Modelling for Virtual Manufacturing Systems and Processes’ (project number 252408385) and CRC 926 ‘Microscale Morphology of Component Surfaces’ (project number 172116086). ) ${}^2$