International Journal of Advanced Robotic Systems

ARTICLE

A Dual-Axis Electrostatically Driven

MEMS Microgripper
Regular Paper

Yukun Jia1, Minping Jia2 and Qingsong Xu1*

1 Department of Electromechanical Engineering, Faculty of Science and Technology, University of Macau, Taipa, Macau, China
2 School of Mechanical Engineering, Southeast University, Nanjing, China
*Corresponding author(s) E-mail: qsxu@umac.mo

Received 21 March 2014; Accepted 22 October 2014

DOI: 10.5772/59677

© 2014 The Author(s). Licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License
(http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the
original work is properly cited.

Abstract 1. Introduction

This paper presents the design of a new monolithic two-
axis electrostatically actuated MEMS microgripper with
integrated capacitive position and force sensors working at
the micro-scale level. Each of the two jaws of the micro‐
gripper possesses two degrees-of-freedom (DOF) and is
capable of positioning in both x-and y-axes. Unlike existing
works, where one gripper arm is actuated and other one is
sensed, both arms of the proposed microgripper are
actuated and sensed independently. A sensing scheme is
constructed to provide the position and force signals in the
noncontact and contact phases, respectively. By applying a
120V driving voltage, the jaw can provide 70 μm x-axis and
18 μm y-axis displacements with the force of 190 μN. By
this design, the real-time position and grasping force
information can be obtained in the dual sensing mode. Both
analytical calculation and finite-element analysis (FEA)
were performed to verify the performance of the proposed
design. A scaled-up prototype is designed, fabricated and
tested through the experiment to verify the structure
design of the microgripper.
Keywords MEMS, Microgripper, Electrostatic actuator,
Capacitive sensor
In recent years, microgrippers have been widely applied in
microassembly and micromanipulation tasks. Microgrip‐
pers act as a key element for handling fragile objects, such
as living cells, micromechanical parts, etc. Besides, micro‐
electromechanical systems (MEMS) technology enables the
devices to achieve a compact size, low cost, and simple
fabrication process. Hence, MEMS microgrippers have
drawn considerable attention from both academia and
industry.
Different kinds of actuators such as shape-memory alloys
[1], electrostatic, electrothermal [2], piezoelectric [3],
pneumatic [4] and electromagnetic [5] actuators have been
designed to actuate the microgripper dedicated to various
applications. The review of recent developments on MEMS
microgrippers can be found in [6]. In particular, a micro‐
gripper driven by an electrostatic actuator was present in
[7]. This can provide a one-axis displacement of 100 μm
with 150 V applied voltage, and the gripper was fabricated
by deep-reactive ion etching (DRIE) on silicon-on-insulator
(SOI) wafer. On the other hand, an electrothermally driven
microgripper has been designed [8]. Owing to the benefits
of the SU-8 structure, such as high coefficient of thermal
expansion [9], relatively large elastic modulus [10], [11],

Int J Adv Robot Syst, 2014, 11:187 | doi: 10.5772/59677 1

 and relative large displacement with low operating realizations of actuation and sensing schemes enable easy
temperature, SU-8 based electrothermal actuation micro‐ operation of the microgripper. However, because only one
grippers have also been widely applied [12]. From the gripper arm is driven, the gripper exhibits a relatively small
actuation point of view, although an electrothermal gripping range. Some other multi-DOF grippers have also
actuator can generate a large output force and displace‐ been reported [19].
ment with a small applied voltage, the high operating
In this paper, a novel 2-DOF microgripper with both arms
temperature is the main adverse effect of this actuator.
actuated and sensed are devised to achieve a large gripping
Alternatively, the electrostatic comb-drive based actuator
range. Specifically, a dual-axis microgripper integrated
can provide a large displacement with a low operating
with electrostatic actuators and capacitive position/force
temperature [13].
sensors is designed. Since each of the two arms is actuated
In order to avoid damaging the fragile objects, a great independently, a relatively large gripping range can be
interest has been devoted to the real-time position/force generated. Each of the gripper arms is connected with two
sensors along with high resolution and sensitivity. For orthogonally attached electrostatic actuators, which are
instance, the optical focus method was reported in [14]. In electrostatic actuators, which are able to provide
instance, the optical focus method was reported in [14]. In able to provide displacement along the x-and y-axes,
[15] and [16], electrothermally driven microgrippers with displacement along the x- and y-axes, respectively.
[15] and [16], electrothermally driven microgrippers with respectively. Moreover, all of the actuators are integrated
piezoresistive force sensors
piezoresistive forcewere presented.
sensors However,
were presented. a
However,Moreover,
a all of the actuators are integrated with
with capacitive position/force sensors. Real-time position/
piezoresistive force sensor is easily affected by a variation
piezoresistive force sensor is easily affected by a variation capacitive position/force
force information of sensors. Real-time
the gripper position/force
jaw can be obtained for
in temperature,
in temperature, size, andsize,
so and
on. so
In on. In contrast,
contrast, capacitive force
capacitive information
both ofthethe gripper
x-and jawBoth
y-axes. can analytical
be obtained for both and
calculations
sensors
force sensors have widely
have been been widely
studiedstudied
due to due
theirtohigh
their high finite-element analysis (FEA) simulations have
the x- and y-axes. Both analytical calculations and finite-been
resolution
resolution and and sensitivity
sensitivity capabilitiescapabilities
[17]. For [17]. For example,
example, a a
element carried out to validate gripper performance. Furthermore,
analysis (FEA) simulations have been carried out
microgrippermicrogripper with an integrated
with an integrated capacitivecapacitive force sensor
force sensor a scaled-up
to validate prototype is . developed
gripper performance Furthermore,for aexperimental
scaled-
wasin
was presented presented in [7].
[7]. Simple Simple structure
structure and an inexpensive
and an inexpensive verification.
up prototype is developed for experimental verification.
fabrication fabrication
process are process are additional
additional meritsmerits of capacitive force
of capacitive
sensors.

l = 4 lm

lm

Figure 1. Schematic of the dual-axis microgripper

Figure 2. Detailed
Figure mechanism
2. Detailed parameters
mechanism parameters of microgripper
of microgripper
force sensors.
Figure 1. Schematic of the dual-axis microgripper 2. Mechanism Design design
2. Mechanism
Previously, a two-degree-of-freedom (2-DOF)
Previously,
microgripper withouta two-degree-of-freedom
integrated force sensor (2-DOF)was microgrip‐ Generally,
Generally, the electrostatic
the electrostatic actuators
actuators have have two different
two different
designed in [18]. It can offer the displacement up to 17in [18].types,
per without integrated force sensor was designed It types, i.e., transverse and lateral comb-drive
i.e., transverse and lateral comb-drive types. types.
The The
can offer the displacement up
μm and 11 μm along the x- and y-axes, respectively.to 17 μm and 11 μm along actuator usually consists of a large number of fingers,
actuator usually consists of a large number of fingers,
the x-and y-axes, respectively. Moreover, a monolithically
Moreover, a monolithically integrated two-axis which are parallel-plate capacitors. When a voltage is is
which are parallel-plate capacitors. When a voltage
integrated two-axis microtensile tester was developed [13].
microtensile tester was developed [13]. It consists of a appliedapplied
between between
the the movableand
movable and fixed
fixedplates, the actuator
plates, the
It consists of a two-axis electrostatic actuator with integrat‐ force is generated to move the comb. As shown in Figure
two-axis electrostatic actuator with integrated capacitive actuator force is generated to move the comb. As shown
ed capacitive position sensors and a two-axis capacitive 1, the microgripper is composed of compliant flexures [20].
position sensors and sensor.
micro-force a two-axis capacitive
The actuator micro-force
employs in Figure 1, the microgripper is composed of compliant
the lateral comb- The use of compliant elements allows the generation of a
sensor. Thedrive
actuator
typeemploys the lateralforce
and the capacitive comb-drive type
sensor uses the trans‐
flexures [20]. The use of compliant elements allows the
number of advantages [21]. In addition, the lateral comb-
verse comb-drive
and the capacitive force sensortype.uses
The the
actuator was designed
transverse comb- to drivegeneration of a number of advantages [21]. In addition,
drive type has been chosen for actuation because it can
drive type.oneThegripper arm, and
actuator wasthedesigned
other arm was connected
to drive oneto a two-
the lateral comb-drive
deliver relativelytype
largehas been chosen
displacement foraactuation
with simple struc‐
axis
gripper arm, capacitive
and the othermicroforce
arm wassensor. Such kinds
connected of decoupled
to a two- becauseture.
it can deliver relatively large displacement with a
axis capacitive microforce sensor. Such kinds of simple structure.
decoupled
2 realizations
Int J Adv Robot of actuation
Syst, and| doi:
2014, 11:187 sensing schemes
10.5772/59677
enable easy operation of the microgripper. However, 2.1 Design of Actuator
because only one gripper arm is driven, the gripper
exhibits a relatively small gripping range. Some other By applying a voltage, the actuator force is generated to
multi-DOF grippers have also been reported [19].
2.1 Design of actuator æ l ö
x = d çç ÷÷ = 4d (4)
By applying a voltage, the actuator force is generated to è lm ø
produce a linear motion. The actuator force can be calcu‐
lated as follows: where l/lm=4 is adopted in the present study. In addition, x
is the x-axis displacement, which is generated at the tip of
1 t the microgripper jaw by an applied voltage V.
Fac = V 2 × e × × n (1)
2 g
2.2 Design of sensor

where V is applied voltage, ε is permittivity of air, t is
material thickness, g is actuator electrode gap spacing
width, and n is number of electrode gaps per actuator.
In addition, four equal-sized flexures are connected with
the actuator. Hence, the overall spring constant k can be
calculated by:
The actuator with integrated capacitive position sensor
enables the generation of the real-time gripper jaw position.
As shown in (3), the gripper moves with a linear motion
once a certain voltage applied. Usually, when the gripper
touch the target object, an external force that is equal to the
grasping force is applied to the gripper jaw. The equation
can be expressed as (5):

E × t × w 3f
k = 4× F ac - Fgp
(2) d= (5)
L3f k
calculated by: F ac Fgp
 t  w3f wf is the width of d (5)
where E is Young’s modulus E where F gp is the grasping force.k
k  4  of silicon, (2)
flexure, and Lf is the length of flexure.
L3f F gp isforce
This external
where will change
the grasping the displacement increment
force.
Then,where
the displacement d can be of
E is Young’s modulus derived
silicon,for
wf each
is the axis
widthasof rate. As shown in Figure 3, the grasping process can be
follows.
flexure, and Lf is the length of flexure. classified into freeforce
This external phasewill
and change
contact phase. In the former
the displacement
phase, the capacitive
increment rate. As sensor
shownactsin as a position
Figure sensor,
3, the while
grasping
Then, the displacement d can be derived for each axis as inprocess
the latter
canphase, the capacitive
be classified into freesensor
phase functions
and contactasphase.
a force
Parameter Value
follows. sensor.
In the former phase, the capacitive sensor acts as a
Actuator position sensor, while in the latter phase, the capacitive
Parameter value
sensor functions as a force sensor.
V Actuator
Applied voltage 0-120V
V Applied voltage 0-120V Original displacement increment rate
ε Permittivity of air 8.85×10-12F/m
ε Permittivity of air -12
8.85×10 F/m
t Material thickness 50 μm
Displacement

t Material thickness 50 μm
g Actuator electrode gap
gActuator electrode gap spacing width 5 μm
5 μm
spacing width New displacement increment
n Number of electrode gaps per actuator 300 rate
Number of electrode gaps
n 300
E Young’s modulus of silicon
per actuator 169 GPa
E Young’s modulus of silicon 169 GPa
wf Width of flexure 6 μm
wf Width of flexure 6 μm
L f Length of flexure 900 μm t1 Time t2
Lf Length of flexure 900 μm
Capacitive position sensor
Capacitive position sensor Grasping force increment rate
Initial capacitive electrode overlapping
t1 L 0 Initial capacitive electrode 75 μm
t1L0 plate length
overlapping plate length
75 μm
Grasping force

gs Capacitive electrode gap

gCapacitive
s
electrode gap spacing width 5 μm
5 μm
spacing width
Number of capacitive electrode gap
ns Number of capacitive 150
ns spacing 150
electrode gap spacing
Table 1. Design
Table parameters
1. Design of the microgripper
parameters of the microgripper

F ac d F ac
d=  (3)(3)
k k t1 Time t2
Moreover, the displacement of the gripper jaw is Figure 3. The relationship between displacement and grasping
amplified by a lever. The output displacement of each Figure
force3. The relationship between displacement and grasping force
Moreover, the displacement of the gripper jaw is amplified
arm can be calculated as:
by a lever. The output displacement of each arm can be The blue line in the upper chart in Figure 3 denotes the
calculated as: l  The blue line in the upper chart in Figure 3 denotes the
original displacement without force applied. It describes
x  d    4d (4) original displacement without force applied. It describes
 lm  the free phase increment rate without force applied. The
Yukun Jia, Minping Jia and Qingsong Xu: 3
where l/lm = 4 is adopted in the present study. In addition, free phaseA Dual-Axis
occurs before time t1, Driven
Electrostatically in which
MEMStheMicrogripper
external
x is the x-axis displacement, which is generated at the tip force is not applied. At a certain time (t1), the gripper jaw
of the microgripper jaw by an applied voltage V. touches the object and the contact phase is induced.
During this phase, since the external force is applied, the
 the free phase increment rate without force applied. The
free phase occurs before time t1, in which the external force
is not applied. At a certain time (t1), the gripper jaw touches
the object and the contact phase is induced. During this
phase, since the external force is applied, the displacement
incremental rate is changed. It follows that the sensor
output signal is contributed to by both the driving dis‐
placement and the external force.
Therefore, the force signal needs to be extracted from the
combined signal. Specifically, a factor can be determined
below to give the relationship between the displacement
and grasping force:
D Fgp
h=
Dd
where △F gp and △d are the change values of the grasping
force and displacement, respectively, in the same time
interval.
In practice, before the microgripper is used for the first
time, it needs to be calibrated by the aforementioned
method to find the factor (6). Then, the real-time position
and force signal can be provided by the capacitive sensor
in the free and contact phases, alternately.
than 10%. Hence, the analytical models are confirmed by
the conducted FEA. Because of their computational
efficiency, the analytical models will be employed for an
optimal design of the gripper dimension in the future.
In addition, both analytical model and FEA simulation
reveal that the microgripper is capable of over 140 μm
gripping range. With an initial gap of 50 μm between the
two gripper jaws, the objects ranging between 50 and 190
μm can be grasped by adopting the designed microgripper.
From the dynamic modelling, some analytical calculations
and FEA-based simulations are performed to validate
natural frequency as follows. First, the stiffness of a leaf
flexure can be calculated as:
(6)
EI
Kq = 2g Kt (7)
Lf
where γ is the optimal value of the characteristic radius, the
average value of γ =0.85 is used in the research, K t =2.669 is
stiffness coefficient and I is the area moment of inertia:

3. Analytical computation and FEA simulation

In this section, both analytical method and FEA simulations
are performed to validate the gripper’s static and dynamic
performance.

Parameter Value

V 120V

Fac 190 μN

Size 6.9x6.5 mm2

Displacement (analytical)

x-axis 76.35 μm
Figure 4. The x-axis displacement result of FEA simulation
y-axis 19.09 μm

Displacement (FEA)

x-axis 70.88 μm

y-axis 18.19 μm

Table 2. Performance result of microgripper

Table 1 shows the designed parameters of the microgrip‐

per. The gripper owns a compact size of 6.9 x 6.5 mm2. By
employing a driving voltage of 120 V, the actuator force,
stiffness, and output displacement are calculated by
resorting to (1), (2), and (3), respectively.
FEA simulation is conducted using the calculated actuator
force. The simulation results of displacement along the x-
and y-axes are depicted in Figures 4 and 5, respectively.
Moreover, the performance of the microgripper is sum‐
marized in Table 2. It is observed that the discrepancy of
the displacements of the gripper in the two axes as predict‐ Figure 5. The y-axis displacement result of FEA simulation
ed by the analytical modelling and FEA simulations is less

4 Int J Adv Robot Syst, 2014, 11:187 | doi: 10.5772/59677

 tW f 3 é æ1ö ù
2 2
I= (8) æ1ö
ê r tAa ç ÷ + r tAb + r tAc ç ÷ ú
12 ê è4ø è4ø ú
mx = ê ú
ê æ1ö
2
r tAd ´ L2f ú
where the parameters L f , t, and W f are shown in Table 1. ê +4 r tAd ç 8 ÷ + 4 ú
( )
2
ê è ø 12 8 L f ú
The kinetic energy equation is shown as: ë û

1 2 where the components a, b, c, and d are denoted in Figure 6.

T= mq& (9)
2 In addition, the potential energy along the x-axis is ex‐
pressed as:
where q is displacement in working direction, for the x-axis,
the mass m is expressed in (10) below. 2
æ1 ö 2 Kq q
Vx = 4 ç Kq q Z2 ÷ »
(11)
è2
( )
2
ø 8L
é æ 1 ö2 f
æ1ö ù
2
ê maç ÷ +1mb + mc ç ÷ ú  d  T  T V
2 2
1
ê èm4 aø 
   è m4 bø ú m c         Fx (12)
mx = ê   42  ú 4  (10) (10) dt  x  x x
mx  ê  æ1ö 1 ú  Substituting (9) and (11) into the Lagrange’s equation:
ê +4 m ç ÷ + 4 I
1 8 L 2 úú
d d 2 where T, V, t, x represent the kinetic energy, the potential
ê è 4øm 
ë 
8
d 
( )
  f  û4I d
1 
 energy, time variable, and displacement in the working
 8
2
8Lf   d æ ¶T ö ¶T ¶V
direction, respectively:
=F
ç -÷ + x (12)
dt è ¶x ø
M¶qx2 ¶x Kq  F (13)
In addition, F is set to 0 to give the free motion dynamic
where T, V, t, x represent the kinetic energy, the potential
energy, time variable, and displacement in the working
direction, respectively:

Mq&&2 + Kq = F (13)

d c b

a In addition, F is set to 0 to give the free motion dynamic

model:

Figure Figure
6. Illustration of variousof
6. Illustration components of the microgripper
various components of the microgripper
Figure 7. Modal analysis result of natural frequency along x-axis

Parameter Value
Parameter Value
Lf 120V
L f 120V190μN
t
t Size 190μN
6.9x6.5 mm2
Size Frequency (analytical)
6.9x6.5 mm2
x-axis 592.3745 Hz
Frequency (analytical)
y-axis 2102.8 Hz
x-axis 592.3745 Hz
Displacement (FEA) Figure 7. Modal analysis result of natural frequency along x-axis
y-axis x-axis 2102.8531.8
Hz Hz
Displacement (FEA)
y-axis 1819 Hz
Mq&&2 + Kq = 0 (14)
Table 3.x-axis 531.8 Hz
Natural frequency result of microgripper
y-axis 1819 Hz
 Therefore, the natural frequency can be calculated as:
1 
2 2
1 of 
 tA
Table 3. Natural frequency result
a   
microgripper
 tA b   tA 
c  

 4 4 
mx   2  Figure 8. Modal analysis result of natural frequency along y-axis
 1 tAd  Lf2  Yukun Jia, Minping Jia and Qingsong Xu: 5
 4 tAd  8   4 
128Lf 
2 A Dual-Axis Electrostatically Driven MEMS Microgripper
    model:
where the components a, b, c, and d are denoted in Figure Mq2  Kq  0 (14)
6. Therefore, the natural frequency can be calculated as:
 The stiffness K y can be expressed as:

Kq
Kx = (19)
Lf 2

Thus, the natural frequency along the y-axis can be

calculated by substituting (17) and (19) into (15).

The performance of microgripper is shown in Table 3. Both

analytical calculation and FEA result show that the natural
frequencies along x-and y-axis are over 500 Hz and 1800
Hz, respectively.

Moreover, the gripper also enables a precise positioning in

the y-axis. As compared with conventional 1-DOF grip‐
Figure 8. Modal analysis result of natural frequency along y-axis
ping, the 2-DOF positioning allows a more dexterous
manipulation for practical applications. In the future work,
Kx a prototype of the designed microgripper will be fabricated
fx = (15) and a simultaneous position/force control [22], [23] will be
mx
conducted through experimental investigations.

where the stiffness K x is shown as: 4. Discussion on analytical results

Kq In this paper, the newly designed microgripper can provide

Kx = 2 (16) over 70 μm the x-axis anddesigned
18 μm y-axis displacements at the
16 L K In this paper, newly microgripper can
Kx f 2
(16) p r jaw
o v i dtip
e with
o v e r the
7 0 force
μ m ofx - a190
x i sμNa nby
d 1applying
8 μ m y -aa120V
x i s driving
16Lf voltage. Compared
displacements at the jaw tip withwithexisting
the force ofdesign
190 μNscheme,
by this
At the same time, the mass m and potential energy along concept
applying endows
a 120V driving thevoltage.
microgripper
Compared with
withthe capabilities of
existing
y-axisAtcan
thebe calculated
same time, the as shown
mass m andinpotential
(17) andenergy
(18): along achieving
design scheme, 2-DOF movement
this concept endows withthelarge displacement and
microgripper
y-axis can be calculated as shown in (17) and (18): withdelivering real-time force sensing simultaneously.
the capabilities of achieving 2-DOF movement with Two-
large displacement and delivering real-time force sensing
side actuation and two-side sensing enables the microgrip‐
150 800.00 simultaneously. Two-side actuation and two-side sensing
per a wider application in different field of environment.
Displacement (μm)

Natural Frequency (Hz)

enables the microgripper a wider application in different

f i e this
In l d design,
o f ethe n vlength
i r o nand m width
e n t . of the
I n flexure
t h i sconnected
600.00
100
400.00 with the electrostatic actuator are the key parameters of the
50
200.00
system performance. To achieve the desired displacement
(over 65 μm) and compact size (6.9x6.5 mm2), the range of
0 0.00
these two parameters are under restriction.
5 5.5 6 6.5 7
Width of Flexure (μm)
d_Analytical d_FEA
nf_Analytical nf_FEA
Figure Figure
9. Comparison of system
9. Comparison performance
of system performancewith
withdifferent
different flexure
parameters
flexure parameters
Figure 10. Experiment setup with actuator and sensor
 1
2

é tAa  tAb  2tAù 
æ1ö c    design, the length and width of the flexure connected
êr tA + r tAb + r tAc ç ÷ ú  2   (17)
m y  ê a 2
è2ø ú 
with the electrostatic actuator are the key parameters of
my = ê 2 1  2útA  L2f  (17) the system performance. To achieve the desired
ê  4 tA r
æ 1dö   d ´4L f ú d
tA
ê+4 r tAd ç 2 ÷  +24 ú 2L 2  displacement (over 65 μm) and compact size (6.9x6.5
è ø  12 2 L
( )
2 12
ê ú f  mm2), the range of these two parameters are under
ë f û restriction.
Figure 10. Experiment setup with actuator and sensor
1  2K  q 2
V y  4 K   Z2 2  (18) AsAsthe the
width
widthof flexure varies varies
of flexure betweenbetween
5 and 7 5μm,
anda 7 μm, a
æ1 2ö 2Kq q
Vy = 4 ç Kq qZ2 ÷ » 2Lf 2 (18)
comparison study of the system performance including
comparison study of the system performance including
è2
( f ) as:
2
ø 2L displacement and natural frequency is carried out. The
The stiffness K y can be expressed displacement and natural frequency is carried out. The
length of flexure is fixed at 900 μm due to the limitation
K length
of the wholeof size.
flexure is fixed at 900 μm due to the limitation of
K x  2 (19) the whole size.
Lf As shown in Figure 9, both analytical calculation and FEA
Thus, the natural frequency along the y-axis can be simulation are performed to test the system performance.
6 Int J Adv Robot Syst, 2014, 11:187 | doi: 10.5772/59677
calculated by substituting (17) and (19) into (15). The width of flexure is proportional to the displacement,
whereas it is inversely proportional to the natural
The performance of microgripper is shown in Table 3. frequency. The objective of the test is to find out the
Both analytical calculation and FEA result show that the maximum natural frequency when the displacement is
natural frequencies along x- and y-axis are over 500 Hz over 65 μm. As can be seen from the results, although 6.5
As shown in Figure 9, both analytical calculation and FEA
simulation are performed to test the system performance.
The width of flexure is proportional to the displacement,
whereas it is inversely proportional to the natural frequen‐
cy. The objective of the test is to find out the maximum
natural frequency when the displacement is over 65 μm. As
can be seen from the results, although 6.5 μm and 7 μm
width can provide larger natural frequency, the output
displacements of the microgripper under these two
parameters are less than 65 μm. Alternatively, when the
width of flexure is 6 μm, the system exhibits the maximum
natural frequency with over 70 μm displacement.

5. Experiment study

In order to verify the capability of the microgripper Figure 11. Experiment result of the gripper displacement
structure design, a scaled-up model has been developed
based on the design. Due to the relatively easier fabrication
process and faster manufacturing time, the prototype is
fabricated based on the scaled-up model. In this design, the
PZT is selected to produce the main displacement instead
of electrostatic actuator. Both analytical method and FEA
are tested to verify the static and dynamic performances.
The results have been compared with experimental results
to show the detailed structure capability of the microgrip‐
per design.

5.1 Experiment setup

The prototype has been fabricated based on the scaled-up

gripper design, the material is AL6061 and the actuator is
PZT, which can generate the maximum stroke of 9 μm.

In this experiment, the PZT amplifier gain is adjusted as 10. Figure 12. Experiment result of natural frequency evaluation
The real-time output sensing is carried out by both an
optimal microscope and a non-contact laser displacement
ment at the tip has a magnitude of 24.28 μm. At this
sensor. The experiment setup is shown in Figure 10.
moment, this result is confirmed by the conducted FEA
simulation.
5.2 Experiment results

Figure 11 gives the experimental result for the displace‐ Performance

ment of the gripper. With the applied voltage signal, which
Displacement
has a magnitude of 11V, a displacement of 20.88 μm has
been measured by the sensor. Figure 12 shows the natural FEA simulation 24.06 μm
frequency result when the input signal is a swept sine wave Analytical 27.48 μm
from 10 Hz to 1000 Hz frequency with 0.1V amplitude. The
Experiment 20.88 μm
system’s response is measured in decibels, and the phase is
measured in radians, versus frequency in Hertz. The Frequency

measured natural frequency is 870 Hz. FEA simulation 539.42 Hz

Table 4 shows the analytical calculation, FEA simulation Analytical 604.17 Hz

and experiment result of the scaled-up gripper. As can be Experiment 870 Hz

seen from the table, there are 13.22% and 24.02% difference
Table 4. The performance result comparison
on displacement between the experimental results and
analytical and FEA simulation results, respectively. The
Besides, the tested natural frequency conducted by the
reason is that the reference point of the laser sensor is
experiment has more than 300 Hz difference compared to
located at the arm of the gripper while the other two
the FEA simulation and analytical calculation. In this case,
methods are tested at the gripper tip. After the calculation
the groove is designed smaller than the PZT actuator to fix
correction of the displacement, it is found that the displace‐

Yukun Jia, Minping Jia and Qingsong Xu: 7

A Dual-Axis Electrostatically Driven MEMS Microgripper
 it tightly, which causes a deformation of the gripper
without voltage applied. This preloading effect enhances
the stiffness of the whole gripper. Consequently, the
natural frequency is increased after the assembly.
To conclude, the prototype based on the scale-up gripper
design has been verified in the experiment. The results
show that the microgripper structure design is feasible.
6. Conclusion
The design and analysis of an electrostatically actuated
monolithic two-axis microgripper with integrated capaci‐
tive position/force sensors has been presented in this paper.
For the purpose of grasping force sensing, the relationship
between the grasping force increment and displacement
ratio change is established. The performance of the micro‐
gripper has been verified with static and dynamic model‐
ling by both analytical method and simulation with FEA.
The results show that the displacement at the tip of each
gripper jaw is over 70 μm and 18 μm along the x-and y-
axes, respectively. The integrated capacitive sensor is
designed to provide the real-time position and grasping
force signals alternately. In addition, the natural frequency
along both axes is examined. Testing results show that the
parameters meet the desired requirement. A scaled-up
prototype is fabricated and tested through the experimen‐
tal study. The results show that the microgripper structure
design is successful. The development of a control scheme
is ongoing, and more experimental results will be present‐
ed in the future.
7. Acknowledgements
The work was support by the Macao Science and Technol‐
ogy Development Fund under Grant No.: 070/2012/A3 and
the Research Committee of the University of Macau under
Grant Nos.: MYRG083(L1-Y2)-FST12-XQS and MYRG078
(Y1-L2)-FST13-XQS.
8. . References
[1] C. M. Lin, C. H. Fan, C. C. Lan, “A shape memory
alloy actuated microgripper with wide handling
ranges,” in Proc. of Int. Conf. On Advanced Intelli‐
gent Mechatronics, pp. 12-17, 2009.
[2] R. Zhang, J. K. Chu, H. X. Wang, Z. P. Chen, “A
multipurpose electrothermal microgripper for
biological micromanipulation,” Microsystem
Technologies, vol. 19, no. 1, pp. 89-97, 2013.
[3] Q. Xu, “Adaptive discrete-time sliding mode
impedance control of a piezoelectric microgripper,”
IEEE Transactions on Robotics, vol. 29, no. 3, pp.
663-673, 2013.
[4] S. Butefisch, V. Seidemann, S. Buttgenbach, “Novel
micro-pneumatic actuator for MEMS,” Sensors and
Actuators A-Phys., vol. 91, no. 8, pp. 638-647, 2002.
8 Int J Adv Robot Syst, 2014, 11:187 | doi: 10.5772/59677
[5] D. H. Kim, M.G. Lee, N. Kim, and Y. Sun, “A
superelastic alloy microgripper with embedded
electromagnetic actuators and piezoelectric force
sensors: a numerical and experimental study,”
Smart Materials & Structures, vol. 14, no. 6, pp.
1265-1272, 2005.
[6] Y. Jia and Q. Xu, “MEMS microgripper actuators
and sensors: the state-of-the-art survey,” Recent
Patents on Mechanical Engineering, vol. 6, no. 2, pp.
132–142, 2013.
[7] F. Beyeler, A. Neild, S. Oberti, D. J. Bell, Y. Sun, J.
Dual, and B. J. Nelson, “monolithically fabricated
microgripper with integrated force sensor for
manipulating microobjects and biological cells
aligned in an ultrasonic field,” Journal of Microelec‐
tromechanical Systems, vol. 16, no.1, pp. 7-15, 2007.
[8] E. V. Bordatchev and S. K. Nikumb, “Microgripper:
design, finite element analysis and laser microfab‐
rication,” in Proc. of Int. Conf. On MEMS, NANO
and Smart Systems, pp. 308-313, 2003.
[9] H. Lorenz, M. Laudon, and P. Renaud, “Mechanical
characterization of a new high-aspect-ratio near
UV-photoresist,” Microelectronic Engineering, vol.
41–42, pp. 371-374, 1998.
[10] H. Lorenz, M. Despont, N. Fahrnl, N. LaBianca, P.
Renaud, P. Vettiger, “SU-8: A low-cost negative
resist for MEMS,” Journal of Micromech. Microeng.,
vol. 7, pp. 121-124, 1997.
[11] L. Dellmann, S. Roth, C. Beuret, G. A. Racine, H.
Lorenz, M. Despont, P. Renaud, P. Vettiger, and N.
F. de Rooij, “Fabrication process of high aspect ratio
elastic structures for piezoelectric motor applica‐
tions,” in Proc. Int. Conf. Solid State Sensors and
Actuators, pp. 641-644, 1997.
[12] T. C. Duc, G.-K. Lau, J. F. Creemer, and P. M. Sarro,
“Electrothermal microgripper with large jaw
displacement and integrated force sensors,” Journal
of Microelectromechanical Systems, vol. 17, pp.
1546-1555, 2008.
[13] S. Muntwyler, B. E. Kratochvil, F. Beyeler, and B. J.
Nelson, “Monolithically integrated two-axis
microtensile tester for the mechanical characteriza‐
tion of microscopic samples,” Journal of Microelec‐
tromechanical Systems, vol. 19, pp. 1223-1233, 2010.
[14] M. Lei, D. Zaili, and C. Hoyin, “Displacement
measurement of a thermally actuated polymer
microrobotic gripper using the optical focus
method,” Intelligent Control and Automation, vol.
5, pp. 3800-3804, 2004.
[15] G. Greitmann and R. A. Buser. “Tactile microgrip‐
per for automated handling of microparts,” Sensor
and Actuators A-Phys., vol. 53, no. 1, pp. 410-415,
1996.
[16] K. Molhave and O. Hansen, “Electro-thermally
actuated microgrippers with integrated force-
 feedback,” Journal of Micromech. Microeng., vol.
15, no. 6, pp. 1256-1270, 2005.
[17] Y. Zhang, X. Y. Liu, C. H. Ru, Y. L. Zhang, L. X.
Dong, and Y. Sun, ”Piezoresistivity characterization
of synthetic silicon nanowires using a MEMS
device,” Journal of Microelectromechanical Sys‐
tems, vol. 20, no, 4, pp. 959-967, 2011.
[18] G.-K. Lau, J. F. L. Goosen, F. van Keulen, T. C. Duc,
and P. M. Sarro, “Polymeric thermal microactuator
with embedded silicon skeleton: part i-design and
analysis,” Journal of Microelectromechanical
Systems, vol. 17, pp. 809-822, 2008.
[19] Q. Xu, “Mechanism design and analysis of a novel
2-DOF compliant modular microgripper,” in Proc.
of 7th IEEE Conference on Industrial Electronics
and Applications, pp. 1966-1971, 2012.
[20] Q. Xu, “New flexure parallel-kinematic microposi‐
tioning system with large workspace,” IEEE
Transactions on Robotics, vol. 28, no. 2, pp. 478-491,
2012.
[21] Q. Xu, “Design and development of a flexure-based
dual-stage nanopositioning system with minimum
interference behavior,” IEEE Transactions on
Automation Science and Engineering, vol. 9, no. 3,
pp. 554-563, 2012.
[22] K. Rabenorosoa, C. Clevy, Q. Chen, and P.
Lutz, ”Study of forces during microassembly tasks
using two-sensing-fingers grippers,” IEEE/ASME
Transactions on Mechatronics, vol. 17, no. 5, pp.
811-821, 2012.
[23] Q. Xu, “Precision position/force interaction control
of a piezoelectric multimorph microgripper for
microassembly,” IEEE Transactions on Automation
Science and Engineering, vol. 10, no. 3, pp. 503-514,
2013.
Yukun Jia, Minping Jia and Qingsong Xu: 9
A Dual-Axis Electrostatically Driven MEMS Microgripper