Assessing the applied normal load during contact mode AFM: An issue with the conventional approach

Usually, the normal load applied to the tip of an Atomic Force Microscope (AFM) probe during its motion across the surface of a scanned sample is determined assuming the absence of any horizontal force. It is shown that this assumption, applied to tip-based nanomachining experiments, leads to wrong estimations of the applied normal force. During this study, the AFM was operated under the so-called “force-controlled” mode based on the optical lever method where a position sensitive photodiode (PSPD) is used to monitor the reflection of a laser beam from the back of the cantilever probe. Normally, the sensitivity of the PSPD is assumed as a known constant, hence, its voltage output in the vertical direction is utilised to determine the cantilever deflection and the normal load. We show that, due to the action of the horizontal force, the PSPD sensitivity is not a constant. Hence, even if the voltage output of the PSPD is kept constant via the feedback loop of the AFM instrument, the grooves produced with tip-based nanomachining are not equal to the depth of indents which are generated when the AFM stage is static. In particular, the difference was up to 55%