ABSTRACT
Quantitative measurements of the friction force are possible once the torsional spring constant of the A F M cantilever is known. The cantilever torsional spring constant was determined according to two recently developed calibration protocols. In the method of Feiler etal. [29], a glass fibre attached orthogonally to the end of the cantilever provides torque as the cantilever is bought into contact with a flat silicon substrate. The second calibration procedure was a method proposed by Bogdanovic et al. [30], in which a second cantilever with integrated tip glued upside on the lower surface acts as a pivot which twists the cantilever as it is brought into contact with the pivot. The procedure was simplified by pressing the cantilever at three contact points only, one in the centre line of the cantilever (where the probe was to be mounted) and one at either side as close as possible to the outer edge. Since the width of the cantilever could be determined accurately using optical microscopy, this avoided the use of the piezo movement to measure the distance. Both calibration methods were in quantitative agreement and yielded values for the torsional spring constant varying between 2.7 and 4.1 x 10~8 Nm/rad. There was a linear correlation between the normal and torsional spring constants.
