Schematic view of a STM system. The STM controller adjusts the z-position of the tip such that, ideally, the tunneling current matches the setpoint for all x, y-positions and generates the x, y positions. 4 The bottom inset shows the ideal tip trajectory, the top inset displays the actual trajectory that suffers from overlaid noise.
Schematic of the scanning probe microscope simulator (SPMS). The SPMS consists of (from left to right) input resistors, limitation diodes, an impedance converter, and the tunneling current diode. The following current-to-voltage converter can be part of the SPMS or separately attached to the feedback system. Optionally, the dynamic response of the mechanical loop of the microscope can be added by closing switch S, where an LC series is connected to ground. Choosing L = 25 mH and C = 1 μF leads to a scanner resonance of 1 kHz.
The roughness R a of the topography as a function of the P-Gain is shown. Low P-Gain values show a strong increase of R a (Region I) as the feedback system cannot fully zero the error signal. In Region II, the P-Gain values show a saturation of R a . For high P-Gain values (Region III), the feedback system starts to oscillate and the apparent roughness R a strongly increases.
(a) Schematic of the rectangular perturbation signal and the output response. The settling time t settling is the time between the rectangular perturbation signal rise and the output response is within a defined error. 19 (b) Topographic image of the rectangular perturbation signal, with V r = 50 mV, f = 36.6 Hz, and P-Gain = 450 pm. (c) Cross section from (b) to extract t settling .
The optimal P-Gain value was determined by measuring the minimum settling time for a rectangular perturbation signal. The two outer points of the parabola were estimated as there settling times were not within the time according to a plateau of the rectangular perturbation signal. The minimum P-Gain is around 340 pm.
The roughness values R a are shown for two STM controllers, A and B. The extracted R a for controller A is 1.12 pm, which is approximately two times smaller than R a = 1.99 pm for controller B. Controller B was a 20 year old feedback electronic compared to a modern feedback electronic called controller A.
Article metrics loading...
Full text loading...