Alexander Blümel1,2, Harald Plank3, Evelin Fisslthaler3, Meltem Sezen3, Werner Grogger3, Emil J.W. List-Kratochvil1,2
1NanoTecCenter Weiz Forschungsgesellschaft mbH, 2Institute for Solid State Physics, Graz University of Technology, 3Institute for Electron Microscopy, Graz Univ
15:20 - 17:30 Thursday 24 October 2013 Foyer Alte Technik Kelvin Probe Force Microscopy (KPFM) is an Atomic Force Microscopy (AFM)-based technique and it is used in many fields of application for its capability to reveal the lateral variations of the surface potential on a flat sample.
However, proper interpretation of the real data is often difficult due to a well-known artefact originating from the long range behavior of electrostatic forces: the measured surface potential is crucially affected by the interaction of the cantilever area with the sample surface, leading to a possible misinterpretation of the experimental data.
In order to get a better understanding of the measurement process, both the AFM tip and a rectangular shaped sample surface potential distribution were modelled and the tip-sample interaction was simulated, leading to a calculated KPFM image of this test structure. At the same time rectangular test structures were fabricated and measured by means of KPFM. In particular, spin-cast polyfluorene films were selectively irradiated by an electron beam in a Scanning Electron Microscope (SEM).
Although simplified, the calculation is capable of showing the influence of the cantilever in the correct qualitative manner: both the simulated and the measured KPFM images show an asymmetrical broadening along the cantilever’s axis of symmetry instead of a pronounced border between the two areas of different surface potential.
However, despite the accordance of calculation and experiment, in everyday lab life the more interesting question is, if it is possible to recalculate the real potential distribution from the measured and convoluted images. In this work a deconvolution was tried on the simulated images, showing that for simple geometries it is possible to reveal the "real" surface potential values in principle: areas of equal potential were assigned to the sample surface in the simulated KPFM image. In practice this information could be gathered by e.g. the AFM height or phase image or additional information on the sample. The original potential values are then calculated over a multiple regression, thus for simple situations a possible way towards a deconvolution of KPFM data could be shown within this work.
|
