Search

Abstract: The acquisition rate of all scanning probe imaging techniques with feedback control is limited by the dynamic response of the control loops. Performance criteria are the control loop bandwidth and the output signal noise power spectral density. Depending on the acceptable noise level, it may be necessary to reduce the sampling frequency below the bandwidth of the control loop. In this work, the frequency response of a vacuum Kelvin force microscope with amplitude detection (AM-KFM) using a digital signal processing (DSP) controller is characterized and optimized. Then, the main noise source and its impact on the output signal is identified. A discussion follows on how the system design can be optimized with respect to output noise. Furthermore, the interaction between Kelvin and distance control loop is studied, confirming the beneficial effect of KFM on topography artefact reduction in the frequency domain. The experimental procedure described here can be generalized to other systems and allows to locate the performance limitations. [Copyright &y& Elsevier]

Abstract: We investigate the gap-voltage control loop in a Kelvin force microscopy setup with simultaneous non-contact topography imaging. The Kelvin controller electrostatically excites the second resonance of the cantilever at about 6.3 times the first resonance frequency and adjusts the DC component of the gap voltage to cancel the oscillation amplitude at this frequency, while the non-contact topography imaging is based on a frequency control loop that maintains a constant frequency of the mechanically excited first resonance of the cantilever by adjusting the tip–sample separation. Due to the self-excitation of the first resonance in our setup, it has to be considered that the electrostatic excitation at the second resonance frequency is applied to a closed feedback loop and cannot be considered as a simple superposition to the oscillation at the first resonance frequency. In particular, special care has to be taken about internal capacitive crosstalk between the tip bias and the cantilever deflection output signal. It is shown that such a coupling cannot be corrected by subtraction of a constant offset at the demodulator output since the crosstalk is sent into the self-excitation loop and is multiplied by the closed loop transfer function. We present a circuit that actively compensates, outside the vacuum environment, the internal crosstalk by adding to the deflection output a dephased fraction of the electrostatic excitation signal. [Copyright &y& Elsevier]

The measurement of local surface potentials by Kelvin force microscopy (KFM) can be sensitive to external perturbations which lead to artifacts such as strong dependences of experimental results (typically in a ∼1 V range) with KFM internal parameters (cantilever excitation frequency and/or the projection phase of the KFM feedback-loop). We analyze and demonstrate a correction of such effects on a KFM implementation in ambient air. Artifact-free KFM measurements, i.e., truly quantitative surface potential measurements, are obtained with a ∼30 mV accuracy. [ABSTRACT FROM AUTHOR]