A Micromachined 2DOF Nanopositioner with

Integrated Capacitive Displacement Sensor

Lujun Ji, Yong Zhu, S. O. Reza Moheimani, Mehmet Rasit Yuce
School of Electrical Engineering and Computer Science, the University of Newcastle
University Drive, Callaghan, NSW, 2308, Australia
Email: {lujun.ji, yong.zhu, reza.moheimani, mehmet.yuce}@newcastle.edu.au

Abstract—This paper presents the design, fabrication and parallel plate) are employed to drive the stage to move
characterization of a micromachined two degrees-of-freedom independently along the XYZ directions. The drawback of this
(2DOF) nanopositioner. The proposed micro-electro-mechanical design is that there is no sensing implemented for closed-loop
system (MEMS) stage, consisting of comb-drive actuators and feedback control to achieve important objectives such as
on-chip capacitive displacement sensors in both X and Y improved dynamic behaviour of the actuator with fast
directions, can simultaneously actuate the microstage and sense response time, precise position control and continuous tuning
the corresponding displacements. A commercial capacitive of position [6].
readout IC (MS3110) is used for the open-loop capacitive
sensing. The first resonance frequency of the stage is measured To provide functional improvement to [12], this paper
to be 4.24 kHz. The positioner has a dynamic range from −6.27 presents a MEMS-based 2DOF nanopositioning stage, which
μm to +6.64 μm at an actuation voltage of 100 V. utilizes four sets of electrostatic linear comb drives for jointly
driving the stage to produce motions in the X and Y directions,
I. INTRODUCTION and four sets of integrated comb-drive displacement sensors in
conjunction with a commercial capacitive readout IC
With the development of the MEMS technology, MEMS- (MS3110) for simultaneously reading out the corresponding
based positioning stages have attracted more and more displacements.
attention in numerous applications in micro-/nano-scale
positioning and manipulation systems due to their small size, II. DEVICE DESIGN
low cost, fast response, and flexibility for system integration.
The applications, such as microlens-array based optical cross The schematic diagram of the MEMS-based 2DOF
connect (OXC) [1], micro-confocal imaging [2], scanning nanopositioner is depicted in Fig. 1. On the silicon layer,
probe microscopy (SPM)-based high-density data storage [3], distributed around the center stage, four sets of in-plane comb-
and optical and magnetic pickup heads [4][5], demand the drive actuators are employed to drive the center stage along
capability of positioning over a motion rage of micrometers the X and Y directions, with each actuator set consisting of
and with a resolution of nanometers. four banks of combs. And, four sets of comb-drive sensors are
designed to sense the corresponding displacements of the
Piezoelectric stages have been widely used for center stage, with each sensor set consisting of two banks of
nanopositioning applications [6]-[9]. SPM is a typical example, combs. Four tethering beams are used to suspend the center
wherein piezoactuators are used for X-, Y-, and Z-positioning. stage and transmit in-plane motions from the comb-drive
Although piezoactuators are capable of providing nanometer actuators.
resolutions, the inherent hysteresis and creep that are
characteristics of piezoelectric material can cause significant To minimize the cross-coupling of motions among
open-loop positioning errors. Thus, piezoactuators require different directions, orthogonal configuration of the XY
sophisticated nonlinear compensation techniques [10]-[11]. In actuators is chosen [12]. For instance, when the actuators (I)
addition, the relatively large sizes of most commercially and (II) drive the center stage in the X direction, the two
available piezoelectric stages (usually around 10 cm) limit tethering beams in the X direction are tensile, thus no
their use in micro-scale system integration. MEMS-based displacements in the Y direction are generated. The X-
positioning stages can offer a good alternative to piezoelectric directional actuation forces not only deflect the two tethering
stages due to their small size, high resonance frequency, beams in the Y direction but also introduce X-directional loads
precise positioning control, and flexibility in system to the suspension beams of actuators (III) and (IV). In order
integration. not to interfere with Y-directional positioning of the stage, the
suspension beams of actuators (III) and (IV) must have a high
Liu et al. have reported on the design, fabrication and lateral stiffness in the X direction to minimize X-directional
testing of a MEMS-based 3-axis positioning stage [12]. In- displacements of the movable comb fingers of actuators (III)
plane and out-of-plane electrostatic actuators (comb-drive and and (IV). In this design, four fixed-guided beams are designed

This research is funded by Australian Research Council (ARC) discovery

grant – DP0774287.

978-1-4244-8168-2/10/$26.00 ©2010 IEEE 1464 IEEE SENSORS 2010 Conference

 where the +/− signs represent forward and backward motions.
Based on (3), the design parameters of the 2DOF
nanopositioner are determined and summarized in Table 1.

TABLE I. DESIGN PARAMETERS OF THE 2DOF NANOPOSITIONER.

Structural parameters
Suspension beams ls = 700 μm, ws = 6 μm, hs = 25 μm

Tethering beams Lt = 700 μm, Wt = 6 μm, Ht = 25 μm

Center stage 520 μm × 520 μm

Center shaft lsh = 640 μm, wsh = 80 μm

Device size 3.75 mm × 3.75 mm

Actuation and sensing parameters
Comb-drive actuator Nf = 256, lf = 30 μm, wf = 3 μm, hf = 25 μm,
(one set) df = 2 μm, overlapf = 10 μm
Fig. 1. Schematic diagram of the micromachined 2DOF nanopositioner. Comb-drive sensor Nf = 128, lf = 30 μm, wf = 3 μm, hf = 25 μm,
(one set) df = 2 μm, overlapf = 10 μm
to suspend each comb-drive actuator.
To determine device parameters, in-plane stiffness of the III. FABRICATION
2DOF nanopositioner is analyzed, and the relationships The device was fabricated using the silicon-on-insulator
between actuation voltages and displacements in the XY (SOI) multi-user MEMS processes (SOIMUMPs) in a
directions are derived. In the analysis, it is assumed that, under commercial foundry (MEMSCAP). The SOIMUMPs process
small deformations, the stiffness of beams in one direction is is a 4-mask level SOI patterning and etching process, which
not significantly affected by the structural deformations along offers a 25 μm thick device layer and a 2 μm thick buried
other directions. oxide (BOX) layer supported on a 400 μm thick handle layer,
Neglecting the high lateral stiffness of the suspension and a minimum gap of 2 μm [14]. The fabrication process, as
beams, according to symmetry, the in-plane stiffness of the illustrated in Fig. 2 [15], is briefly described as follows:
nanopositioner is [12][13]: 1) Microfabrication starts with a highly doped n-type 25 µm
8Ehs ws3 2 EH tWt 3 silicon device layer, on which a metal stack consisting of
K x, y = + (1) 20 nm Cr and 500 nm Au is patterned to allow for ohmic
ls3 L3t contact through a liftoff process;
where E = 129.5 GPa is Young’s modulus of silicon, ls, ws, 2) Silicon is lithographically patterned with the second mask
and hs are the length, width and thickness of the suspension level, SOI, and etched using deep reactive ion etch (DRIE)
beams, and Lt, Wt, and Ht are the length, width, and thickness to define both the movable and anchored features of the
of the tethering beams. structure;
The center stage is actuated by applying the driving 3) A protection polyimide layer is applied to the top surface
voltages to the driving electrodes along the X direction with of the silicon layer;
differential AC voltages (±VAC) superimposed on DC bias 4) A deep trench underneath the movable structure is created
(VDC). Since the driving voltages applied to electrodes which by etching through the substrate layer using DRIE;
are placed on either side of the X-directional actuators are V1 = 5) The exposed buried oxide layer is removed using a wet
VDC + VAC and V2 = VDC − VAC respectively (where VAC = HF etch;
Vd·sinωt), the driving force in the X direction is the subtraction
of the electrostatic forces [13]: 6) The front side polyimide layer is removed by oxygen
plasma, thereby allowing the movable structure to be fully
1 ∂C 2 1 ∂C 2 ∂C 4 N f ε 0h released. Then, a blanket metal layer, consisting of 50 nm
F= V1 − V2 = 2 VDCVAC = VDCVAC (2 ) Cr and 600 nm Au, is deposited and patterned using a
2 ∂x 2 ∂x ∂x d
shadow masking technique. The shadow mask is removed
where ε0 = 8.85×10-12 C2/(Nm2) is the permittivity of air, d, h after evaporation, leaving a patterned metal layer on the
and Nf are the gap, thickness and the number of comb fingers, substrate for electrical contact.
respectively. Therefore, the in-plane displacements are [13]: The images of the whole device and the center stage taken
under a Scanning Electron Microscope (SEM) are shown in
F
x, y = ± (3) Fig. 3. As depicted in Fig. 3 (b), patterned with blanket metal,
K x, y the micron-arrays on top of the center stage were designed for
Atomic Force Microscope (AFM) image scanning.

1465
 the stage along the X direction against the applied DC voltage.
The stage was actuated bi-directionally by applying static
voltages ranging from 0 V to 100 V to the driving electrode
on either side of the X direction, and the displacement was
measured by the PMA under different driving voltages. The
stage has a dynamic range from −6.27 μm to +6.64 μm along
the X direction, which falls into the range of AFM scanning
applications.
A capacitive readout IC (MS3110) was chosen for sensing
the capacitance change and providing an output voltage
proportional to that change. The advantages of MS3110 are
high resolution, stability, low drift and more importantly
adjustable internal balancing capacitors [16]. To minimize the
stray capacitance and electromagnetic interference, the
Fig. 2. Fabrication process for the 2DOF nanopositioner.
MEMS device chip is positioned very close to MS3110 chip.
Fig. 5 demonstrates the experimental setup for capacitive
sensing using the evaluation board MS3110BDPC [16]. Pin 4
and Pin 6 supply the capacitance bridge with AC carrier
signals which are 100 kHz square wave differential signals
with a peak-to-peak amplitude of 2.25 V. Pin 5 is kept at 2.25
V DC potential and connected to a common electrode that is
the movable part of the MEMS structure. The feedback
capacitor of the capacitance transimpedance amplifier, the
output buffer gain, and the LPF bandwidth were selected to
be 1.007 pF, 4, and 0.5 kHz respectively. At every actuation
voltage for the displacement measurement, the output of
MS3110 (Pin 14) was measured using a digital oscilloscope
(Tektronix TDS 3024B). Fig. 6 shows the measured sensor
output voltage versus the stage displacement.
The dynamic behaviour of the 2DOF nanopositioner was
characterized using a spectrum analyzer (HP 35670A).
Superimposed on a 40 V DC bias, a 3 V sinusoidal actuation
signal with different frequencies ranging from 10 Hz to 8 kHz
was applied to the actuator and the output signal reflecting
changes in the amplitude of vibration, detected by the
capacitive displacement sensor, is sent back to the spectrum
analyzer. Fig. 7 presents the frequency response of the
fabricated nanopositioner. The first resonance frequency is
measured at 4.24 kHz, which is 15.5% higher than the
simulated first undamped natural frequency of 3.67 kHz. This
discrepancy is due to the fabrication imperfections. The phase
delay from 10 Hz to 8 kHz is 200°.

Fig. 3. SEM image of the 2DOF nanopositioner. (a) whole structure;

(b) magnified view (center stage).

IV. CHARACTERIZATION
Accurate characterization of the device is important to
verify the design as well as the fabrication quality. Prior to
testing, the device was glued and wire-bonded onto a PCB
packaging board. The static behaviour of the nanopositioner
was measured using a PolytecTM Planar Motion Analyzer
Fig. 4. Static X-directional displacement as a function of applied voltage.
(PMA). Fig. 4 shows the static bi-directional displacement of
1466
 be designed to precisely control the developed
nanopositioning stage. Furthermore, the 2-axis closed-loop
feedback control of the nanopositioner will be implemented to
allow for AFM imaging of the micron-arrays patterned on top
of the center stage.

ACKNOWLEDGMENT
The authors would like to thank Dr. Andrew Fleming for
his valuable suggestions on device packaging, and Dave
Phelan from the School of Medical Science at the University
of Newcastle for his help with taking SEM photographs.

REFERENCES
[1] H. Toshiyoshi, G. J. Su, J. LaCosse, and M. C. Wu, “A surface
Fig. 5. Capacitive sensing measurement setup using MS3110. micromachined optical scanner array using photoresist lenses fabricated
by a thermal reflow process,” Journal of Lightwave Technology, vol.
21, no. 7, pp. 1700–1708, Jul 2003.
[2] S. H. Kwon and L. P. Lee, “Stacked two dimensional micro-lens
scanner for micro confocal imaging array”, in Proc. IEEE MEMS 2002,
pp. 483–486, Jan 2002.
[3] P. F. Indermühle, V. P. Jaecklin, J. Brugger, C. Linder, N. F. De Rooij,
and M. Binggeli, “AFM imaging with an XY-micropositioner with
integrated TIP,” Sens. Actuators A, Phys., vol. 47, no. 1, pp. 562–565,
Mar 1995.
[4] C. H. Kim, H. M. Jeong, J. U. Jeon, and Y. K. Kim, “Silicon micro XY
stage with a large area shuttle and no-etching holes for SPM-based data
storage,” J. Microelectromech. Syst., vol. 12, no. 4, pp. 470–478, Aug
2003.
[5] X. H. Huang, R. Horowitz, and Y. F. Li, “A comparative study of
MEMS microactuators for use in a dual-stage servo with an
instrumented suspension,” IEEE/ASME Trans. Mechatronics, vol. 11,
no. 5, pp. 524–532, Oct 2006.
[6] Y. K. Yong, S. S. Aphale, and S. O. R. Moheimani, “Design,
identification, and control of a flexure-based XY stage for fast
Fig. 6. Sensor output voltage as a function of X-directional displacement.
nanoscale positioning,” IEEE Trans. Nanotechnol., vol. 8, no. 1, pp.
46–54, Jan 2009.
[7] C. L. Chu and S. H. Fan, “A novel long-travel piezoelectric-driven
linear nanopositioning stage,” Precis. Eng., vol. 30, no. 1, pp. 85–95,
Jan 2006.
[8] D. Croft and S. Devasia, “Vibration compensation for high speed
scanning tunneling microscopy,” Rev. Sci. Instrum., vol. 70, no. 12, pp.
4600–4605, Dec 1999.
[9] C. W. Lee and S. W. Kim, “An ultraprecision stage for alighment of
wafers in advanced microlithography,” Precis. Eng., vol. 21, no. 2/3, pp.
113–122, Nov 1997.
[10] S. Devasia, E. Eleftheriou, and S. O. R. Moheimani, “A survey of
control issues in nanopositioning,” IEEE Trans. Control Syst. Technol.,
vol. 15, no. 5, pp. 802–823, Sep 2007.
[11] C. Ru, B. Shao, L. Chen, W. Rong, and L. Sun, “Design, identification,
and control of piezoactuated positioning mechanism based on adaptive
inverse method,” Proc. Inst. Mech. Eng., J. Syst. Control Eng., vol. 222,
no. 16, pp. 501–509, 2008.
[12] X. Liu, K. Kim, and Y. Sun, “A MEMS stage for 3-axis
nanopositioning,” J. Micromech. Microeng, vol. 17, no. 9, pp. 1796–
1802, Jul 2007.
Fig. 7. Frequency response of the X-directional motion, both magnitude [13] R. Legtenberg, A. W. Groeneveld, and M. Elwenspoek, “Comb-drive
(upper trace) and phase (lower trace). actuators for large displacements,” J. Micromech. Microeng, vol. 6, no.
3, pp. 320–329, Jun 1996.
V. CONCLUSION [14] http://www.memscap.com/en_mumps.html
[15] Y. Zhu, M. R. Yuce, and S. O. R. Moheimani, “A low-loss MEMS
An electrostatically actuated 2DOF nanopositioner with tunable capacitor with movable dielectric,” in Proc. IEEE Sensors
on-chip capacitive displacement sensor is presented in this 2009, pp. 651–654, Oct 2009.
paper. With the characterization results shown in Fig. 4 and [16] http://www.irvine-sensors.com/pdf/MS3110%20Datasheet%20USE.pdf
Fig. 6, a closed-loop proportional-integral (PI) controller can

1467