In this chapter we will discuss a selection of experimental observations of friction on the nanoscale, obtained by atomic force microscopy and related techniques. After presenting high resolution friction maps on different materials, we will compare the load, velocity and temperature dependence of friction detected in the experiments to the predictions of the Prandtl–Tomlinson model. The comparison will be extended to simple experiments showing the effect of contact vibrations and friction anisotropy on crystalline samples.
Friction measurements on the atomic scale
The first lattice resolved maps of stick–slip were acquired by Mate et al.  just one year after the atomic force microscope was invented . In their experiment, Mate et al. used a tungsten wire as a probing tip and detected lateral forces on a graphite surface using non-fiber interferometry. Since graphite is stable, chemically inert and easy to cleave along atomic planes, it is an ideal material for this kind of measurement. The pioneer work by Mate et al. was followed by experiments on ionic crystals (NaCl, KBr etc.), metals (Cu, Au, Al, W, Pt, Pd and Ag) and covalent materials: semiconductors, carbon-based materials (e.g. graphite, diamond, and diamondlike carbon), organic materials and many oxides.
The ultra-high vacuum (UHV) environment reduces the influence of contaminants on the sample surfaces and results in more precise and reproducible results. An atomic-scale friction map, acquired with a silicon tip sliding on a NaCl(001) cleavage surface in UHV (lattice constant a = 0. 564 nm), is shown in Fig. 18.1(a). The spring force F grows up to a maximum value, corresponding to the static friction Fs ≈ 0.4 nN at which the tip suddenly slips. After that, the tip quickly rebinds to a neighboring unit cell on the crystal surface. The process is repeated several times along each scan line, reproducing the structure of the surface lattice.