This research is focused on improving the mechatronic design and control strategies of Atomic Force Microscopes (AFM) in order to increase the imaging speed of these instruments, while maintaining the resolution and accuracy of AFM. Different techniques are developed which have led to a faster lateral scanning speed, higher bandwidth control of the tip sample force, and improved accuracy of the topography estimate.
An Atomic Force Microscope (AFM) is a mechanical microscope in which the sample is probed by a very sharp tip. AFM allows measuring the sample topography with (sub-) nanometer resolution. As in AFM the sample is probed point by point, AFM imaging is a relatively slow process, which limits the applicability of AFM in fields where high throughput is important.
In AFM de sample is scanned in the lateral plane by use of a piezoelectric scanning stage. The lateral scanning speed is limited by the weakly damped resonances of the scanning stage, which may cause strong oscillations when excited. In this research a cost efficient method is developed to dampen the resonances of the scanning stage, resulting in a 30 times faster lateral scanning motion.
During scanning the force between the tip and the sample is measured and controlled in a feedback loop to prevent damage of the tip and the sample. Moreover, this feedbackloop provides an estimate of the sample topography via a topography estimator. In this research an approach is presented for integrated design of the feedback controller and topography estimator. This approach explicitly addresses the uncertainty in the dynamical behavior of the instrument. It is shown that due to the uncertainty in the dynamical behavior of the instrument, a design trade-off has to be made between the speed and the accuracy of AFM instruments.
In order to further improve the imaging speed of AFM, a dual actuated control approach is investigates to control the force between the tip and the sample via a combination of a long-stroke and a short-stroke actuator. This dual actuated control approach has shown to allow 20 times faster imaging as compared to the optimally controlled single actuated AFM, without compromising the effective positioning range.