Evaluation of Charge Drives for Scanning Probe Microscope Positioning Stages Andrew J. Fleming School of Electrical Eng. and Computer Science The University of Newcastle Callaghan, NSW 2308, Australia Kam K. Leang Department of Mechanical Engineering Virginia Commonwealth University Richmond, Virginia, 23284-3015, USA Abstract— Due to hysteresis exhibited by piezoelectric actua- tors, positioning stages in scanning probe microscopes require sensor-based closed-loop control. Although closed-loop control is effective at eliminating non-linearity at scan speeds below 10Hz, it also severely limits bandwidth and contributes sensor- induced noise. The need for high-gain feedback is reduced or eliminated if the piezoelectric actuators are driven with charge rather than voltage. Charge drives can reduce hysteresis to less than 1% of the scan range. This results in a corresponding increase in bandwidth and reduction of sensor induced noise. In this work we review the design of charge drives and compare them to voltage amplifiers for driving lateral SPM scanners. The first experimental images using charge drive are presented. I. I NTRODUCTION A key component of Scanning Probe Microscopes (SPM’s) [1] is the nanopositioning system required to manoeuvre the probe or sample. Piezoelectric actuators are universally employed in positioning systems due to their high stiffness, compact size and effectively infinite resolution. A major disadvantage of piezoelectric actuators however, is the hys- teresis exhibited at high electric fields. This causes imaging artefacts in scanning probe microscopes. Techniques to elim- inate this non-linearity include feedback, feedforward and image-based compensation, which are reviewed in references [2], [3] and [4]. The most popular technique for compensation in commer- cial scanning probe microscopes is sensor-based feedback using Proportional-Integral (PI) control. Such controllers are simple, robust to modeling error, and due to high loop-gain at low-frequencies, they effectively mitigate piezoelectric non- linearity. The foremost disadvantages of closed-loop control, however, are the cost, additional complexity, bandwidth limi- tations, and sensor-induced noise. In this work, the technique of charge control is evaluated for linearization of SPM Email: Andrew.Fleming@newcastle.edu.au Email: kkleang@vcu.edu positioning stages. The aim is to eliminate the requirement for feedback control and alleviate the associated problems. Since the late 80’s, it has been known that driving piezo- electric transducers with current or charge rather than voltage significantly reduces hysteresis [5]. Simply by regulating the current or charge, a five-fold reduction in hysteresis can be achieved [6]. Although the circuit topology of a charge or current amplifier is much the same as a simple voltage amplifier, the uncontrolled nature of the output voltage typically results in the load capacitor being lin- early charged. Recent developments have eliminated low- frequency drift and permitted grounded loads, which are necessary in nanopositioning systems [7]. In the following section, charge drives are briefly re- viewed, then applied to imaging experiments in Section III. A critical evaluation of charge drives for open- and closed- loop SPM applications is discussed in Sections IV and V, followed by conclusions. II. CHARGE DRIVES Consider the simplified diagram of a grounded load charge drive shown in Figure 1 [7]. The piezoelectric load, modeled as a capacitor and voltage source v p , is shown in gray. The high-gain feedback loop works to equate the applied refer- ence voltage v ref , to the voltage across a sensing capacitor C s . Neglecting the resistances R L and R s , at frequencies well within the bandwidth of the control loop, the load charge q L is equal to q L = V ref C s , (1) i.e., the gain is C s Coulombs/V . When connected to a capacitive load, the equivalent voltage gain is C s /C L . As discussed in [7], the existence of R L and R s introduces error at low-frequencies. Essentially they draw charge away from the load and form a voltage feedback loop at DC and low-frequencies. This source of error can be eliminated 2008 American Control Conference Westin Seattle Hotel, Seattle, Washington, USA June 11-13, 2008 ThA08.2 978-1-4244-2079-7/08/$25.00 ©2008 AACC. 2028