New image acquisition algorithms for atomic force microscope

The behavior of the digital feedback control system of an atomic force microscope (AFM) and the development of novel AFM scan algorithms are described. A digital AFM feedback control system model is developed in which the system is constructed by cascading the AFM with a digital proportional-integral (PI) compensator. The PI compensator is composed by a proportional controller and an integral controller in parallel and is software implemented inside a digital signal processor, DSP 56001 from Motorola. The PI compensator manipulates the bicell detector signal received from the AFM and feeds back the manipulated signal to the AFM to adjust the vertical position of the sample accordingly via a piezo tube scanner. The rate of adjusting the vertical position of the sample can be increased by increasing the proportional gain and integral gain of the PI compensator. However, the system will become unstable if these two gains are too high. By adjusting the proportional gain and integral gain manually, we usually cannot find the value of these two gains that give the system the fastest and stable system response to adjust the vertical position of the sample. An automated scanner is developed which uses genetic algorithm to search for the optimal gains that gives the system the fastest stable system response. In addition, the automated scanner is able to vary the scan speed according to the surface roughness of the sample surface and the image data acquired has a finite known error margin. Attempts are made to find an improved alternative control algorithm to the proportional-integral-derivative (PID) compensator. Frequency response techniques and root-locus design techniques are used but they both do not produce an improved compensator. Instead, two alternative scan algorithms, predictor corrector and ramp scanner, are developed which improve the performance of the AFM system using PID compensator.