Atomic force microscopic scans of a crystal surface containing an atomic defect were simulated in both the contact and non-contact regimes. When scanning in contact mode near to a defect, the tip/sample interaction force experienced a bifurcation of the lines of constant force. When the load force was small, the bifurcation caused the tip to be pushed out of the defect. However, if the scan force was greater than some critical value (depending upon the composition of the tip and the sample), the atomic force microscope tip became trapped in the vicinity of the defect. The trapped tip remained at the level of the vacancy and thus crashed into the sample as the scan continued. This resulted in the tip apex being destroyed, or in disruption of the crystal lattice around the defect. Both effects resulted in the disappearance of the defect from the scan images. The trap was intrinsic and could not be avoided. In the case of the non-contact mode, the tip position was driven by the scan force gradient rather than by the force. Simulations showed that, for this case, the trap did not exist and atomic defects would not be destroyed. This explained why atomic defects were generally not observed when using contact-mode atomic force microscopy, but were observed in non-contact atomic force microscopy.

Simulation of the Observability of Atomic Defects by Atomic Force Microscopy in Contact and Non-Contact Modes. I.Y.Sokolov, G.S.Henderson: Surface Science, 2002, 499[2-3], 135-40