13 Computer Simulations of Nanometer-Scale Indentation and Friction 697
case the stick-slip was over a much longer length scale and may be due to the fact
that the surfaces were not hydrogen-terminated.
Mulliah et al. [210] used MD simulations with bond-order potentials [211] to
model interactions between indenter atoms, EAM potentials [212] to model interac-
tions between substrate atoms, and the Ziegler–Biersack–Littmack potential [213]
to model interactions between indenter and substrate atoms to study the atomic-
scale stick-slip phenomenon of a pyramidal diamond tip interacting with a silver
surface at several sliding rates and vertical support displacements. These simulat-
ions showed that dislocations are related to the stick events emitting a dislocation in
the substrate near the tip. The scratch in the substrate is discrete due to the tip jump-
ing over the surface in the case of small vertical displacements. In contrast, large
displacements of 15Å or more result in a continuous scratch. These simulations
also showed how the dynamic friction coefficient and the static friction coefficient
increase with increasing tip depth. The tip moves continuously through a stick and
slip motion at large depths,whereas it comes to a halt in the case of shallow indents.
Although the sliding rate can change the exact points of stick and slip, the range of
sliding rates over the range of values considered in this study (1.0 to 5.0m/s) has
no influence on the damage to the substrate, the atomistic stick-slip mechanisms, or
the calculated friction coefficients.
The effect of the way in which the tip is rastered across the surface in MD
simulationswas consideredby Cai and Wang[209,214]using bond-orderpotentials.
In particular, they dragged silicon tips across several silicon surfaces, as illustrated
in Fig. 13.41, in two different ways. In the first, they moved the tip every MD step
while in the other they advanced the tip every 1000 steps. In both cases, the overall
sliding rate is the same and equals 1.67m/s. In both cases, wear of the tip such as is
illustrated in Fig. 13.41 occurs. However, the mechanisms by which the wear occurs
are found to depend on the approach used, and the latter approach is found to be in
better agreement with experimental data.
In many studies, diamond tips or diamond-decorated tips are used in friction
measurements. Diamond is an attractive material for an FFM tip because of its high
mechanical strength and the belief that such tips are wear-resistant. However, dia-
mond tips that were used to scratch diamond and silicon surfaces and then imaged
showed significant wear that increased with the increasing hardness of the tested
material [215,216]. This wear altered the shape of the tip and hence influenced the
contact area that is used to determine friction coefficients.
In summary, MD simulations provide insight into dry sliding friction and the
sliding of metal tips across clean metal, crystalline ceramics, and layered ceramics
surfaces. Stick-slip friction or wear can occur depending on the sliding conditions.
The good lubricating properties of layered ceramics are observed in the simulations
along with localized fluctuations in atomic-scale friction. Crystalline ceramics, such
as diamond, exhibit relatively low friction and the simulations show how stick-slip
atomic-scale motion changes with the conditions of sliding and the way in which
the simulation is performed.