Amorphization-assisted nanoscale wear during the running-in process

Abstract

Atomistic simulations were used to study the nanoscale wear of crystalline silicon with a native oxide sliding against amorphous silicon dioxide. The size, shape and crystallographic orientation of the model were defined to be comparable to those in a corresponding atomic force microscope experiment, where the tip was imaged before and after 40 nm of sliding using ex situ transmission electron microscopy. Tip wear was quantified in the simulation as the volume of silicon atoms removed from the tip at intervals up to 40 nm sliding distance. We also quantified amorphization during sliding as the change of tip material from crystalline to amorphous. The rate of amorphization decreased with sliding distance, and this trend was analyzed in the context of a previously-proposed analytical model for crystalline-to-amorphous transitions and explained qualitatively by local strain distributions within the tip. Both wear and amorphization were found to exhibit similar trends, but the wear rate was faster than the rate of amorphization. The results suggest that amorphization may be a dominant mechanism of nanoscale wear during the early stages of running-in, but less so during steady-state sliding.

Mechanisms of near-surface structural evolution in nanocrystalline materials during sliding contact

“…However, the sample surface also moves downward, showing how material is removed by the indenter during the sliding process. This scenario is in sharp contrast to that of diamond and related materials in which wear occurs through surface amorphization followed by material removal [34,43,44].…”

Tribochemistry: A Review of Reactive Molecular Dynamics Simulations

“…Further, it was suggested that shear-induced amorphous Si is denser than diamond-cubic Si, while amorphous C is less dense than diamond, so the amorphization rates of Si and C exhibit opposite pressure dependence trends. In another study, simulations with the COMB potential were used to complement atomic force microscope experiments of a nanoscale silicon probe with an oxidized surface sliding on an amorphous SiO 2 substrate [70]. The simulations underpredicted the volume of probe wear after 40 nm of sliding compared to the experiment, but the trends were consistent.…”

Measuring and Understanding Contact Area at the Nanoscale: A Review

“…For crystalline materials, the response of a nanocontact to load is dependent on the crystallographic orientation of the materials relative to the loading direction [107]. Where crystallographic information is known-for example, from experimentally measured diffraction patterns or visual analysis of a TEM image-then the model can be designed correspondingly [103,112]. Figure 5(c) shows a cross section of a model silicon SPM tip, where the crystallographic and loading directions are matched to those of the experimental images in Fig.…”

“…Some models explicitly include grain boundaries [89,113], which further improves the accuracy of the simulation for some materials. Other features that have been included in nanocontact simulations are near-surface material amorphization [103] and oxygen or hydrogen surface termination [106,107,[112][113][114]. Finally, in cases where the tip and/or substrate contain roughness, this too may affect nanocontact behavior.…”

“…In a molecular dynamics simulation, the loading and unloading process can be modeled with a set maximum load. However, to model load accurately it requires sufficient hold time at the maximum load to allow the contact to relax to a stable configuration [35,95,103,112,115]. Also, the stiffness of an experimental cantilever/tip cannot be explicitly captured in a tip apex model, but is important in some dynamic simulations.…”