Glue-free tuning fork shear-force microscope

REVIEW OF SCIENTIFIC INSTRUMENTS 77, 016105 共2006兲 Glue-free tuning fork shear-force microscope P. Mühlschlegel, J. Toquant, D. W. Pohl, and B. Hechta兲 Nano-Optics group, National Center of Competence for Research in Nanoscale Science, Institute of Physics, University of Basel, Klingelbergstrasse 82, CH-4056 Basel, Switzerland 共Received 17 October 2005; accepted 14 December 2005; published online 25 January 2006兲 A scanning near-field optical microscope without any glued parts is described. Key elements are the optical fiber probe/tuning fork junction and the piezotube scanner assembly. In both cases, fixation is achieved by means of controlled pressure and elastic deformation. The avoidance of glued connections was found to improve the Q factor of the shear-force sensor as well as to facilitate the replacement of the fiber probe and other parts of the scanner head. We present approach curves and shear-force images that demonstrate the performance and stability of the system. © 2006 American Institute of Physics. 关DOI: 10.1063/1.2165548兴 Piezoelectric shear-force sensors are widely used for probe-sample distance control in scanning near-field optical microscopy1–7 共SNOM兲 and other scanning probe techniques 共see Ref. 8 for further reference兲. In SNOM the optical probe is usually attached to a quartz tuning fork 共TF兲, of the type used in watches. The interaction of the probe with the surface induces a shift of the TF’s resonance frequency. This shift or the resulting impedance change can be used for distance control by means of a feedback mechanism acting on the so-called z piezo, usually a piezoelectric ceramic tube. The TF is excited either mechanically by a piezoelectric element, or electrically by a driving voltage applied directly to the TF.2 The stiffness and fragility of fiber SNOM probes require a sensitive and fast feedback to prevent probes from crashing during approach and scanning. Therefore a sensor with high-quality factor 共Q兲, in combination with a phaselocked loop 共PLL兲 feedback system, is desirable. The standard method of connecting a fiber probe to the TF is gluing with epoxy.1,4,7 This process is somewhat problematic since the adhesive tends to form a thin cushion between fiber and TF. The latter forms a “soft” connection because the elastic properties of quartz and epoxy are widely different, the elastic modulus of the latter being 20 times smaller than that of quartz 共see, e.g., Ref. 9兲 and also having appreciable loss at the typical tuning fork frequencies 共=32 kHz兲. The losses generate damping and result in reduced Q-factor values 关e.g., ⬃1000 共Ref. 1兲兴. To minimize the resulting damping, the gluing process has to be controlled carefully which requires considerable skill and experience. Once glued, a probe cannot be removed easily, e.g., for characterization; neither can a poorly glued probe be readjusted after curing of the epoxy. To circumvent probe gluing, mechanical fixation by clamping the fiber probe between the two arms of a TF 共Ref. 5兲 or connecting the probe to a homebuilt piezoceramic TF 共Ref. 6兲 has been suggested. The present scheme is based on the same principle; however, it relies on an improved mechanical design. It provides adjustable Q factors as high as ⬃4000, resulting in high force sensitivity and in conjunction a兲 Electronic mail: bert.hecht@nano-optics.ch 0034-6748/2006/77共1兲/016105/3/$23.00 with a PLL short feedback response time. The very simple compact implementation allows for particularly fast, easy, and reproducible probe exchange. The elimination of glued parts from the z piezo results in similar advantages as described for the shear-force sensor. Therefore the piezoceramic tube used as z piezo is mounted between two adaptors kept together by means of an appropriately designed screw instead of an adhesive. The main improvements here are the ease of mounting and replacement, the well-defined mechanical parameters, and the higher Q. Figure 1 shows the relevant parts of the SNOM head. Fixation of probe holder base plate 共BP兲 and z piezotube 共PT兲 is illustrated in Fig. 1共a兲. Screw S 共steel兲 exerts a compression force on the assembly that is adjusted by means of a sensitive torque wrench. Torques in the range of 3 – 8 cN m were found to provide stable fixation without damage of the brittle piezoceramic material 共PZT Staveley EBL3, length = 35 mm, D = 6.25 mm, and w = 0.5 mm兲. The electromechanical response of the PT is influenced marginally only by the screw. Since the elastic moduli of steel and standard piezoceramics are roughly the same, the longitudinal 共z direction兲 piezoelectric expansion is reduced in proportion to the ratio of the cross-sectional areas which is ⬍17%, hence irrelevant in most applications. With respect to bending 共lateral scanning兲, the influence is completely negligible since here the area moments of inertia are the relevant parameters which differ by more than three orders of magnitude. Figure 1共b兲 displays the mechanical resonances of the PT, measured with a spectrum analyzer, the excitation voltage being applied to one pair of the quadrant electrodes of the PT, and the signal being picked up from the other pair. The measurement was performed for two different torques applied to screw S and for a conventionally glued z piezo. Apparently, the lowfrequency resonances of the glued piezotube move to higher frequencies for the unglued, clamped piezo. TF and the fiber probe with removed coating are mounted in separate blocks B1 and B2 关Figs. 1共c兲 and 1共d兲兴 such that the fiber is pressed against one prong of the slightly tilted TF near its end. The blocks are screwed to the insulating BP. The relative position of the blocks determines the position H of mechanical contact between TF and fiber and 77, 016105-1 © 2006 American Institute of Physics Downloaded 30 Mar 2006 to 131.152.107.160. Redistribution subject to AIP license or copyright, see http://rsi.aip.org/rsi/copyright.jsp 016105-2 Mühlschlegel et al. Rev. Sci. Instrum. 77, 016105 共2006兲 FIG. 2. 共a兲 Resonance curves for a repeatedly mounted fiber. 共b兲 Resonance of the freely oscillating tuning fork. FIG. 1. Scanner assembly and tuning fork mount. 共a兲 Sketch of the clamped piezotube 共PT兲. S: fixation screw, D: PT outer diameter, w: width of PT walls, and d: diameter of the screw. 共b兲 Resonance curves of the clamped piezotube for different torques exerted on the screw compared to a glued piezotube 关共top phases shift兲 bottom: normalized amplitudes兴. 共c兲–共e兲 Sketch of the tuning fork mount. 共c兲 共side view兲 PT: piezotube, BP: insulating base plate, TF: tuning fork, F: fiber, CL: clamp, B1: block with clamp and screws for fiber fixation, and B2: block with incorporated TF. 共d兲 Rotated view without PT and BP. 共e兲 Magnification of the indicated position in 共d兲. G: grove for fiber guidance, S1 , 2: screws for clamp fixation, ⌬xF: fiber bending amplitude, l1: distance between fixed fiber end and TF, l2: length of free fiber end, H: distance between TF base and fiber, and L: TF length. the amount of fiber bending ⌬xF that we adjusted to ⬇100 ␮m 关Fig. 1共e兲兴. Bending of the fiber results in a force F = ⌬xFkF acting on the TF. The spring constant of the slightly bent cylindrical fiber, anchored on one side, was calculated as10 kF = 共3␲ER4兲 / 共4l31兲共⬇600 N / m兲. Here E, R, and l1 are Young’s modulus, radius, and length of the bent part of the fiber, respectively. With E = 6 ⫻ 1010 N / m2 共SiO2兲,10 R = 62.5 ␮m, and l1 = 1.5 mm 关Fig. 1共e兲兴, the total force exerted by the fiber on the TF amounts to 0.06 N. For stable operation it is instrumental that the probe follows the motion of the tuning fork prong without loosing contact at any moment in time during the tuning fork oscillation. The acceleration ẍF共l1兲 of the free fiber towards the prong hence has to be larger than the maximum acceleration ẍT共H兲 of the TF prong during vibration. To calculate ẍ j; j苸 兵F, T其 the solution to the equation of motion of a vibrating lever 共fiber or TF prong兲 has to be separated:10 x j共z , t兲 = x j共z兲exp共i␻ jt兲. The maximum acceleration for the fundamental resonance frequency ␻ j follows as 兩ẍ j共z兲兩max = ␻2j 兩⌬x j共z兲兩, 共1兲 where 兩⌬x j共z兲兩 is the maximum bending amplitude of the respective levels. The fundamental resonance frequency of an oscillating fiber with a circular cross section fixed on one end is10 ␻F = 共1.76冑E / ␳兲R / l2 ⬇ 2␲共15 kHz兲, where ␳ = 2.2 ⫻ 103 kg/ m3 is the specific mass density of SiO2,10 and l = l1 + l2 ⬇ 2.5 mm is the length of the oscillating free fiber 关Fig. 1共e兲兴. This results in 兩ẍF共l1兲兩max ⬇ 8 ⫻ 105 m / s2. The vibration amplitude 兩⌬xT共H兲兩 of the free TF was estimated from the driving voltage VD = 0.8 mV and resulting current signal Imax ⬇ 1.4 nA of our TF 关length: L = 4.0 mm, thickness:T = 0.63 mm, width: W = 0.35 mm, resonance frequency: ␻T = 2␲ 共32 768 Hz兲, and static spring constant: kstat = 26.9 ␮N / nm兴 by comparison with data for a TF of a similar type2 where VD = 2 mV, Imax = 2.9 nA and a vibration amplitude of 0.4 nm were measured. This yields 兩⌬xT共H兲 兩 ⬍ 1 nm and hence 兩ẍT共H兲兩max ⬍ 40 m / s2. Thus 兩ẍF共l兲兩max Ⰷ 兩ẍT共H兲兩max which ensures save contact between fiber and TF. The ratio R = L / H between total TF length L and fiber mounting position H 关Fig. 1共e兲兴 shows a strong influence on the Q factor of the coupled system fiber/TF, also reported by Crottini et al.7 The position H can be tuned to optimum condition in our arrangement by appropriate adjustment of B1 and B2 with respect to BP 关Figs. 1共c兲 and 1共d兲兴. Once B1 and B2 are fixed in a favorable position, the probe 共fiber兲 can be replaced within a few minutes by loosening screws 共S1,2兲. Clamp CL opens up such that the old fiber can easily be pulled out from grove G and be replaced by a fresh one 关Fig. 1共e兲兴. To characterize the influence of the fiber on the TF, a fiber was mounted and unmounted ten times and the resonance was measured each time 共Fig. 2兲. An average resonance frequency f̄ 0 = 32047± 250 Hz and an average Q̄ = 2032± 1612 factor were determined. Since the static spring constant kstat of the TF is hardly altered by the fiber, we assign the observed shift 共⬇−700 Hz兲 of the resonance for the coupled fiber/TF system with respect to the free TF 关Fig. 2共b兲兴 and to the increase in effective mass m0 due to the attached fiber. The scattering of the data is attributed to the uncertainty in length l2. The deviation in Q is due to slight variations in clamp fixation 关Fig. 1共e兲兴. Samples were mounted on a regulated x-y scan stage 共Physik Instrumente, P-733兲 for imaging. The TF current signal was preamplified and converted to an oscillating voltage by a current-to-voltage amplifier 共amplification factor: 2 ⫻ 107兲. Gap width feedback control was established by means of a phase-locked loop 共PLL, Nanosurf AG兲 and the preamplified TF signal. The use of a PLL reduces the response time of the feedback system.4 Fiber probes were pro- Downloaded 30 Mar 2006 to 131.152.107.160. Redistribution subject to AIP license or copyright, see http://rsi.aip.org/rsi/copyright.jsp 016105-3 Notes FIG. 3. Approach curve towards a SiO2 surface. Voltage oscillation amplitude 共black兲 and frequency shift dF 共gray兲 are recorded during approach 共filled triangles兲 and retract 共open triangles兲 starting from position z = 0. duced by tube etching,11 followed by hot sulfuric acid 共96% 兲 removal of the fiber coating on the tip end side. Approach curves and shear-force images of a SiO2 calibration sample were recorded to demonstrate the functionality and robustness of the glue-free TF shear-force microscope. Figure 3 shows the damping of the voltage oscillation amplitude and the frequency shift when the tip is approached to and retracted from the SiO2 surface. First the z piezo approaches the tip to the surface. The active feedback system stopped the approach when gentle shear-force contact was established 共amplitude damping ⬍3%兲. From this point on 共z = 0 in Fig. 3兲 the feedback system was turned off and further approach was started. Retraction was triggered when the oscillation amplitude reached a predefined lower limit. The deviation between approach and retraction curves is caused by hysteresis of the z piezo. The fit curves were obtained from a shear-force model12 describing the damping of the resonance amplitude for an approaching tip. The amplitude drops within 3 nm to 70% of its undisturbed value, comparable to the data reported by others.6,7 The resonance frequency was found to increase simultaneously with the damping in amplitude. Figure 4共a兲 depicts a shear-force image of a SiO2 calibration sample, consisting of a two-dimensional lattice of inverted square pyramids with 200 nm pitch etched into a silicon chip. Figures 4共b兲–4共d兲 show the data recorded along the line marked in Fig. 4共a兲, topography, the voltage oscillation amplitude, and the frequency shift. At a scan speed of 1 ␮m / s the topography profile is slightly asymmetric due to the feedback response time 关Fig. 4共b兲兴, however, the amplitude never dropped below 90% 关Fig. 4共e兲兴. A tip radius of ⬃23 nm was determined by comparing Fig. 4共b兲 with the specifications of the calibration sample. The presented tuning fork microscope with unglued exchangeable probe and piezotube achieves a performance, Rev. Sci. Instrum. 77, 016105 共2006兲 FIG. 4. 共a兲 Shear-force image of a SiO2 calibration sample 共2D200 by Nanosensors, 200 nm scale bar兲. 共b兲–共d兲 Indicated line profile from 共a兲. 共b兲 z: topography signal. 共c兲 A: voltage oscillation amplitude. 共d兲 dF: frequency shift 共error signal兲. e.g., imaging quality, and approach stability comparable to the best conventionally glued designs. Probe mounting and replacement, as well as mounting of the piezotube, however, is much easier, more reproducible, and extremely fast. The absence of glue will be of particular advantage for operation at nonambient conditions. The adjustability of the Q factor finally allows for an optimal tuning of the feedback loop. The authors gratefully acknowledge H. -J. Güntherodt for his continuous support and thank J. Y. P. Butter, J. N. Farahani, B. W. Hoogenboom, S. Karotke, A. Lieb, Y. Lill, V. Thommen, and A. Tonin for help and fruitful discussions. Financial support by the Swiss National Science Foundation via the National Center of Competence in Research 共NCCR兲 in Nanoscale Science and a research professorship for one of the authors 共B. H.兲 is gratefully acknowledged. 1 K. Karrai and R. D. Grober, Appl. Phys. Lett. 66, 1842 共1995兲. K. Karrai and R. D. Grober, Ultramicroscopy 61, 197 共1995兲. 3 J. Barenz, O. Hollricher, and O. Marti, Rev. Sci. Instrum. 67, 1912 共1996兲. 4 A. G. T. Ruiter, K. van der Werf, J. Veerman, M. F. Garcia-Parajo, W. H. J. Rebsen, and N. F. van Hulst, Ultramicroscopy 71, 149 共1998兲. 5 J. Salvi, P. Chevassus, A. Moufard, S. Davy, M. Spajer, D. Courjon, K. Hjort, and L. Rosengren, Rev. Sci. Instrum. 69, 1744 共1998兲. 6 D. Mulin, C. Vannier, C. Bainier, D. Courjon, and M. Spajer, Rev. Sci. Instrum. 71, 3441 共2000兲. 7 A. Crottini, J. L. Staehli, B. Deveaud, X. L. Wang, and M. Ogura, Ultramicroscopy 90, 97 共2002兲. 8 J. Rychen, T. Ihn, P. Studerus, A. Herrmann, K. Ensslin, H. Hug, P. Schendel, and H.-J. Güntherodt, Rev. Sci. Instrum. 71, 1695 共2000兲. 9 C. Zilg, R. Mülhaupt, and J. Finter, Macromol. Chem. Phys. 200, 661 共1999兲. 10 D. Sarid, Scanning Force Microscopy, revised ed. 共Oxford University Press, New York, 1994兲. 11 R. Stöckle, C. Fokas, V. Deckert, R. Zenobi, B. Sick, B. Hecht, and U. P. Wild, Appl. Phys. Lett. 75, 160 共1999兲. 12 K. Karrai and I. Tiemann, Phys. Rev. B 62, 13174 共2000兲. 2 Downloaded 30 Mar 2006 to 131.152.107.160. Redistribution subject to AIP license or copyright, see http://rsi.aip.org/rsi/copyright.jsp