IEEE SENSORS JOURNAL, VOL. 13, NO.

2, FEBRUARY 2013

601

PDMS Microcantilever-Based Flow Sensor

Integration for Lab-on-a-Chip
Amir Sanati Nezhad, Mahmood Ghanbari, Carlos G. Agudelo, Muthukumaran Packirisamy,
Rama B. Bhat, and Anja Geitmann

Abstract In this paper, a simple practical method is presented

to fabricate a high aspect ratio horizontal polydimethylsiloxane
(PDMS) microcantilever-based flow sensor integrated into a
microfluidic device. A multilayer soft lithography process is
developed to fabricate a thin PDMS layer involving the PDMS
microcantilever and the microfluidics network. A three-layer fabrication technique is explored for the integration of the microflow
meter. The upper and lower PDMS layers are bonded to the thin
layer to release the microcantilever for free deflection. A 3-D
finite element analysis is carried out to simulate fluid-structure
interaction and estimate cantilever deflection under various flow
conditions. The dynamic range of flow rates that is detectable
using the flow sensor is assessed by both simulation and experimental methods and compared. Limited by the accuracy of the
1.76-m resolution of the image acquisition method, the present
setup allows for flow rates as low as 35 L/min to be detected.
This is equal to 0.8-N resolution in equivalent force at the tip.
This flow meter can be integrated into any type of microfluidicbased lab-on-a-chip in which flow measurement is crucial, such
as flow cytometry and particle separation applications.
Index Terms Flow sensor, lab-on-a-chip, multilayer soft
lithography, polydimethylsiloxane (PDMS) microcantilever.

I. I NTRODUCTION

PPLICATION of microfluidic technology on lab-on-achip for biological and chemical processes has tremendous potential. Chemical analysis, mixing, bio-sensing, and
particle separation are just a few examples of these applications. Among those, particle sorting/separation [1] and cytometry [2] are extremely sensitive to the flow rate within the
microfluidic network. Therefore, flow meters integrated within
a lab-on-a-chip to enable real time monitoring of the flow rate
significantly improve the functionality of such a device.
Various flow meters have been reported in the literature
by employing micro-electro-mechanical systems (MEMS) to

Manuscript received May 8, 2012; revised August 19, 2012; accepted

October 1, 2012. Date of publication October 9, 2012; date of current version
January 11, 2013. The associate editor coordinating the review of this paper
and approving it for publication was Prof. Boris Stoeber.
A. S. Nezhad, M. Ghanbari, C. G. Agudelo, and M. Packirisamy are with the
Mechanical Engineering Department, Optical Bio-Microsystem Laboratory,
Concordia University, Montreal, QC H3G 1M8, Canada (e-mail: a_sana@
encs.concordia.ca; ghanbari_m1984@yahoo.com; carlosgustavoagudelo-@
yahoo.ca; pmuthu@alcor.concordia.ca).
R. B. Bhat is with the Mechanical Engineering Department, Concordia University, Montreal, QC H3G 1M8, Canada (e-mail: rbhat@vax2.concordia.ca).
A. Geitmann is with the Dpartement de Sciences Biologiques, Institut de
Recherche en Biologie Vgtale, University of de Montreal, Montreal, QC
H3T 1J4, Canada (e-mail: anja.geitmann@umontreal.ca).
Color versions of one or more of the figures in this paper are available
online at http://ieeexplore.ieee.org.
Digital Object Identifier 10.1109/JSEN.2012.2223667

miniaturize the flow sensor and to integrate it within the

microfluidic device in order to monitor locally the flow rate
[3], [4]. Although these flow sensors have impressive performance, they usually involve complex fabrication including
multi-step layer patterning which requires expensive facilities
[5], [6]. Silicon or SU8 have mostly been used as a micromechanical element in the form of a microcantilever to sense
the flow drag force. Since these materials have high elastic
modulus, the deflection of the mechanical element is too
small and has to be detected by employing advanced detection
methods such as optical or electrical [7], [8]. Integration of
optical or electrical components and their calibration is often
elaborate and hence limits the application of these sensors for
general lab. In addition, living cells are sensitive to electrical
fields and the use of electrical detection methods has the
potential to cause artifacts that make the interpretation of
studies difficult [9]. Similarly, thermal flow sensors have the
potential to change the medium temperature [10][12] and are
therefore equally unsuitable for certain applications despite
their simple readout system.
Besides conventional materials such as silicon or SU8 [13],
Polydimethylsiloxane (PDMS) is a favorable alternative material due to its low elasticity modulus (ranging between 0.5 MPa
to 2 MPa depending upon the ratio of curing agent) [14]. It
simplifies the detection system by measuring the deflection of
the microcantilevers tip using vision based optical microscopy
due to its large deflection in the range of few micrometers [15].
Other PDMS attributes such as good biocompatibility, optical
transparency, low chemical reactivity and non-toxicity in a
microfluidic environment are also important factors in cell
analysis [16]. The fabrication of a PDMS chip is also cost
effective, making it suitable for disposable uses in biological
applications.
Few vertical PDMS microcantilevers have been presented
in the literature to measure the force within microfluidic
devices. Sasoglu et al. [17] used vertical PDMS microbeams
to sense micro-scale cell forces. Liu et al [18] employed
a visual algorithm for tracking the PDMS beam displacements to measure forces in the range of a few nanoNewton.
Tan et al. [19] applied pillar structures while visually inspecting their deflection. Since fabrication limits the length of
vertical cantilevers, they typically have a low aspect ratio
which limits the range of forces that can be measured accurately. A high aspect ratio PDMS microcantilever is promising
since it may provide flow meters with higher sensitivity and
improved linear response for LOC applications.

1530437X/$31.00 2012 IEEE

602

IEEE SENSORS JOURNAL, VOL. 13, NO. 2, FEBRUARY 2013

In order to deduce flow drag force from microcantilever

deflection, it is necessary to know the mechanical stiffness
of the beam. Several experiments have been carried out to
measure the Youngs modulus of PDMS for bulk and thin
PDMS layers. Depending upon the cantilever length, the small
deflection beam theory for beam aspect ratios greater than
10 [20] or large deflection beam theory [21] have been used
to determine beam stiffness. In addition, the MooneyRivlin
(MR) constitutive model [22], the hyperelastic model [23] and
finite element analysis have been used in order to consider the
nonlinear properties of PDMS [24].
In this paper, a high aspect ratio PDMS micro-cantileverbased flow sensor is presented. The micro-cantilever is
integrated into a microfluidic device and fabricated using
multilayer soft lithography. The microcantilever and the
microchannels network are embedded into a thin PDMS layer.
The flow sensor is made of PDMS material identical to the
material of the microfluidic network in order to avoid any
difficulties in bonding non-similar multi-layers. A three layer
fabrication technique is used for the integration of the micro
flow sensor. A thin middle layer is sandwiched between two
other PDMS layers in order to release the microcantilever
and to seal the microfluidic device. The performance of the
flow sensor is tested by introducing various flow rates into
the microfluidic device and measuring the deflection of the
cantilevers tip using an optical microscope. The elasticity
modulus of the cantilever is measured using a precision
balance method developed in house to measure the sensitivity
of the microcantilever to the flow drag force. Based on the
elastic modulus of the cantilever and the deflection, the drag
force applied on the cantilever is determined.
3D finite element analysis (FEM) is carried out using
COMSOL Multiphysics 3.5 to simulate the fluid-structure
interaction and to estimate the cantilever deflection under
various fluid forces. The comparison of simulation and experimental results helps to identify the linear range of the cantilever response and to find the dynamic range of flow rates
detectable by the flow meter. Due to the high aspect ratio
of the cantilever (> 10), the calibration of the cantilever is
simple and the effect of shear force on the deflection can
be ignored. In addition, since the cantilever deflection is in
the range of few micrometers, it can be measured effectively
using vision based optical microscopy and it does not require
advanced detection methods such as AFM or integrated optical
or electrical systems. In spite of the off-line measuring method,
the proposed flow sensor can be integrated into any kind
of microfluidic-based LOC to accurately measure the local
flow rate. As the sensor does not introduce any potentially
disturbing signal, it is particularly useful for applications in
which external electrical or thermal signals at the point of
care should be avoided.
II. S ENSOR D ESIGN
The proposed flow sensor can be integrated with LOC in
two different ways to measure the flow rate such as monolithic
integration and hybrid integration. For microfluidic-based LOC
devices fabricated by multilayer PDMS fabrication technique

flow sensor
Inlet

LOC

Outlet

Platform
(a)

Outlet

Inlet

Flow sensor

LOC

Platform
(b)
Fig. 1. Two different setups for integrating the flow sensor in a LOC device.
(a) Monolithic integration. (b) Hybrid integration.

such as multilayer micromixers, the flow sensor can be monolithically integrated into the design as shown in Fig. 1a.
If the desired LOC is not a multilayer PDMS device, the flow
sensor can be hybrid integrated as a stand-alone unit as shown
in Fig. 1b.
The 3D schematic of the flow sensor is shown in Fig. 2a.
The flow enters the device through the inlet, faces the
microcantilever and then exits towards the outlet. Since the
microcantilever is free, the flow induces the loading to
the microcantilever and deflects it. The flow induced loading
causes the cantilevers deflection (Fig. 2b). The high aspect
ratio PDMS microcantilever is suspended within the microfluidic device and is sandwiched between the bottom and top
thick PDMS layers (Fig. 2c). As the microcantilever deflects
within the horizontal plane, the deflection is readily detected
using vision based optical microscopy.
III. FABRICATION OF I NTEGRATED PDMS
M ICROCANTILEVER W ITHIN A M ICROCHANNEL
The fabrication process of the integrated PDMS microcantilever within the microfluidic network is detailed in Fig. 3.
A 4-inch silicon wafer is cleaned with acetone and DI water,
the surface is blow-dried with pure nitrogen, followed by
baking for 5 min at 200 C to dehydrate the surface (Fig. 3a).
A SU8 mold is made using a standard photolithographic
technique. The negative photoresist SU8 (MicroChem Corp.)
is spin coated at 1500 rpm for 1 min to reach the thickness
of 80 m (Fig. 3b). The SU-8 is next soft-baked for 3 min at
65 C and for 7 min at 95 C in a hotplate. The resist is then
cooled down at room temperature and exposed to UV light
for 30 s using a photo mask (Fig. 3c). Post-exposure bake
step (PEB) is then carried out, baking the resist for 3 min at
65 C and for 6 min at 95 C in order to cross-link the SU-8.
The SU-8 layer is then developed to obtain the SU8 mold
(Fig. 3d).
In order to employ flexible microstructures for bio-sensing,
a multi-layer technique is developed. The microfluidic device
involves a middle thin layer and two more top and bottom

NEZHAD et al.: PDMS MICROCANTILEVER-BASED FLOW SENSOR INTEGRATION FOR LAB-ON-A-CHIP

603

PDMS microcantilever
Outlet

Top layer

Inlet
Bottom layer
Middle layer
(a)

Hollow feature

PDMS microcantilever

Hollow feature

Microcantilever
Top layer

Inlet

Thin layer

Outlet

Fluid loading

Bottom layer
A-A
A
(c)

(b)
Fig. 2.

(a)

Schematic of the PDMS microcantilever-based flow sensor. (a) 3-D schematic of the design. (b) Top view. (c) Cross section.

(e)

(b)
(f)

(c)

(g)

(d)
(h)
Fig. 3. Fabrication process of a multilayer PDMS device. (a) Cleaning.
(b) SU8 coating. (b) UV exposure. (c) SU8 exposure to light. (e) Thin-layer
PDMS loading. (f) Thin layer peeled off. (g) Thin- and bottom-layer bonding.
(h) Top-layer bonding.

PDMS layers sealing the middle layer. The thickness of the

middle layer is critical as it determines the thickness and
hence the stiffness of the microcantilever. To fabricate the
thin middle layer, a mixture of PDMS prepolymer (Sylgard

184 Dow Corning) with a curing agent at a volume ratio of

10:1 is poured onto the SU8 mold and placed in a vacuum
chamber for degasification. A plastic film is then carefully
placed on the PDMS to avoid introducing any bubble underneath. It is then loaded with a clamp to remove the extra
PDMS (Fig. 3e). By applying pressure, the plastic film comes
into physical contact with the mold features and the PDMS
is removed from the contact area. Next, the whole setup is
kept in an oven to cure the PDMS for 2 hours at 60 C. After
curing, the thin layer is peeled off from the mold (Fig. 3f).
Similarly, top and bottom PDMS layers are fabricated by
pouring PDMS onto the mold, curing in the oven, and then
peeling off from the mold. The resulting thin middle layer
and bottom layer are then subjected to plasma treatment and
bonded using microscope alignment (Fig. 3g). Last, the plastic
layer is separated from the package and the top PDMS layer is
placed on top (Fig. 3h). The alignment is done manually under
the microscope, achieving an accuracy of less than 10 m.
The PDMS squeezed between the plastic film and the mold
tends to cover the mold features with a very thin blocking film
in the range of a few micrometers, leaving a residual PDMS
layer (Fig. 4a and b).
Various methods have been presented to remove this residual PDMS layer such as squeezing the thin layer [25], blowing
the PDMS residue away [26] and rupturing the residue by
a sharp needle [27]. Although the sharp needle or blowing
method might be useful for fabrication of conventional thin
PDMS layers, when a high aspect ratio microcantilever is
embedded within the thin PDMS layer, these methods will lead
to rupture of the microcantilever. The squeezing method seems
more applicable for an integrated PDMS microcantilever and

604

IEEE SENSORS JOURNAL, VOL. 13, NO. 2, FEBRUARY 2013

Blocking
PDMS

Clamping force

SU8

SU8

PDMS

PDMS

SU8

SU8
Si

Si

(b)

(a)

10 m

50 m
(c)

(d)

(e)

Fig. 4. Blocking PDMS film as challenge in fabricating thin PDMS layer. (a) Squeezing the thin layer. (b) Blocking PDMS film. (c) PDMS microcantilever.
(d) Cross section of PDMS microcantilever with blocking layer. (e) Cross section of fully released PDMS microcantilever without blocking layer.

has been used in this report to remove the blocking PDMS

layer using the clamp method. Fig. 4c shows the top view
image of the fabricated PDMS microcantilever. The cross
section of the cantilever with blocking PDMS layer attached is
seen in Fig. 4d. By applying sufficient clamping force on the
thin PDMS layer, the blocking layer is removed and only two
small residues are left at the bottom edge of the cantilever
which does not have significant effect on its performance
(Fig. 4e).
IV. M ECHANICAL C HARACTERIZATION OF THE
PDMS M ICROCANTILEVER
To be able to use the microcantilever in quantitative manner,
the exact stiffness needs to be determined to allow for calibration. A precision balance method is employed to measure
the force on the tip and hence the stiffness of PDMS microcantilever (700 m length, 70 m width, 80 m thickness)
(Fig. 5). In order to provide free space around the cantilever
tip, the thin PDMS layer is bonded to a support PDMS layer
in such a way that the cantilever could freely face the balance
(Denver instrument S1-114, sensitivity of 0.1 mg (1 N)).
Fig. 5a illustrates the schematic of the calibration setup. In
this technique, a thin PDMS layer is bonded to a 2 mm thick
PDMS substrate such that the cantilever has a free tip to touch
the balance. It is then mounted on a positioner (ULTRAlign
Metric Linear 3 Axis Stag, resolution of 1 m) and is moved
toward the balance at a distance of 250 m from its clamped
end. The positioner is gradually moved down to force the glass
plate. The glass plate is positioned on the balance to serve as
a rigid body. The force is directly transferred to the balance
(Fig. 5 b and c). Based on the displacement of the positioner
() and the recorded force (F) from the precision balance, the
variation of force against deflection is obtained. The deflection

of the glass placed on the balance due to the cantilever force

can be neglected. In order to consider the effect of out of center
error in the precision balance, a sample mass is positioned
on the center and out of center on the glass. A deviation
of 2 N was detected due to the out of center error for a
sample mass of 500 mg (5 mN). Since the maximum force
detected for the cantilever deflection is 80 N (60 times less
than the force of sample mass), the out of center error would be
negligible.
In order to determine the stiffness, the force (F) is measured
for the deflections () of 50 m, 100 m, 150 m, 200 m of
the microcantilever. As the sensitivity of the balance setup is
not high enough to find the stiffness of a long microcantilever
with 700 m length, this microcantilever meets the balance at
250 m away from the cantilevers support (Fig. 5a). Due to
large deflections, slight non-linearity is found in the stiffness.
As shown in Fig. 6, linear regression is used to estimate the
beam stiffness using the equation of F = K. This test was
repeated three times and the average stiffness of 0.46 N/m
is estimated for the cantilever. The elastic modulus is then
calculated
from the deflection equation of the beam K =


3E I
as E = 802 kPa.
L3
The force-deflection response of the microcantilever confirms the linear behavior of the PDMS microcantilever even
for large deflection. The value of the Youngs modulus is close
to that reported in the range of 4001000 kPa for thin PDMS
layer [28]. The authors reasoned that the Youngs modulus
of PDMS is dependent on the thickness of the layer and
the fabrication process. They reported that thin layer PDMS
fabricated by spin coating has greater mechanical strength
than the layer created in soft lithography by pouring the
PDMS mixture on the mold due to the difference in the
cross-linking.

NEZHAD et al.: PDMS MICROCANTILEVER-BASED FLOW SENSOR INTEGRATION FOR LAB-ON-A-CHIP

605

Support
Microcantilever

250 m
(a)

F
700 m

500 m
(b)

(c)

Force (N)

Fig. 5. Characterization of the PDMS microcantilever using a precision balance. (a) Schematic of the balance setup. (b) Experimental setup. (c) Contact
between the PDMS microcantilever and glass plate.

100
90
80
70
60
50
40
30
20
10
0

y = 0.46x

50

100

150

200

250

Deflection (m)
Fig. 6.

Microcantilever characterization force against deflection.

V. F LOW S ENSOR T ESTING

To assess the performance of the PDMS micro-cantilever
suspended within the microfluidic channel, it was subjected
to fluid flow. It was observed that the microcantilever often
stuck to the microchannel floor or ceiling during fabrication
in plasma bonding, most likely due to the large area of the
structure (700 m length by 70 m width). In order to
overcome stiction, the size of the microcantilever was modified
to reduce its surface area. The new microcantilever has a length
of 510 m and a width of 40 m. The thickness is maintained
at 80 m in order to avoid any change on the elastic modulus
due to thickness alteration. By keeping the microcantilevers
width less than its height, the bending of the structure towards
the top or bottom is minimized whereas the bending in the

horizontal plane along the actual flow direction is promoted.

Fig. 7a shows the schematic of experimental testing of flow
sensor. Water with a density of = 1000 kg/m3 and dynamic
viscosity = 0.001 Pas is injected into the microdevice
through the inlet using a syringe pump with various flow
rates while the cantilever deflection is monitored through
the optical microscope (Fig. 7b). The PDMS microdevice
fabricated with PDMS multilayer technique consists of a thin
PDMS layer with integrated microcantilever and microfluidics
network sealed by two other top and bottom PDMS layers as
shown in Fig. 7c. The flow introduced into the chip through
the inlet, induces the loading force on the microcantilever
to bend it and is transferred out toward the outlet. The
inlet microchannel has a width of 840 m and a depth
of 80 m which is equal to the thickness of thin PDMS
layer. The microfluidic chip has a depth of 240 m at the
location of the flow sensor since there are hollow features
above and below the microcantilever in order for it to be
free. The deflection of the modified micro-cantilever is shown
in Fig. 7d.
The performance of the flow meter was tested within the
typical range of fluid flows in microfluidic devices. The deflection of the microcantilever was measured for various flow rates
between 0.21.3 ml/min using image processing techniques
and the results are shown in Fig. 8. Images were taken with
a 4 objective (calibrated pixel size: 1.76 m 1.76 m)
during the flow experiment for a duration of 40 sec with an
interval of 100 ms. The start and stop point indicate the instant
when the water pumping is turned on and off and correspond
to 1 s and 30 s, respectively.

606

IEEE SENSORS JOURNAL, VOL. 13, NO. 2, FEBRUARY 2013

Microscope

CCD

Lens
Inlet

Image
processing

Outlet

Syring pump
Microcantilever

Flow sensor
(a)

(b)

Outlet

Inlet

Thin layer

100 m

Top layer
Bottom layer
(c)

(d)

Fig. 7. (a) Schematic of experimental setup for the fluid flow test of the PDMS microcantilever. (b) Experimental setup. (c) Fabricated microdevice.
(d) Deflected cantilever under fluid flow.

Stop

Cantilever deflection (m)

Start

incompressible. The fluid domain is governed by the incompressible Navier-Stokes equations and the continuity equations
[29] as follows

100
80
0.2 ml/min

60

0.5 ml/min

40

0.7 ml/min
1.0 ml/min

20
0

1.3 ml/min
0

Fig. 8.

10

20
30
Time (s)

40

50

Deflection of the cantilever tip for different flow rates.

VI. S IMULATION
3D numerical analysis is performed to estimate the deflection of microcantilever under various flow rates. The proposed
model consists of the main microchannel with one inlet,
one outlet and the microcantilever suspended within the
microchannel (Fig. 9a). Water with density of = 1000 kg/m3
and dynamic viscosity = 0.001 Pas is used as the medium
to test the cantilever performance.
The sequential coupled method is used to numerically solve
the interactions between the fluid and the solid structure.
The flow is assumed to be laminar Newtonian, viscous and

u
(pI + (u + (u)T ) + (u ) u) = F
t
u = 0

(1)
(2)

where I is the unit diagonal matrix, u = (u, v) is the velocity

field, p is the fluid pressure and F is the volume force
affecting the fluid. Since the gravitation and other volume
forces affecting on the fluid are negligible, so F = 0.
For each flow rate, the fluid velocity at the entrance of the
microchannel is provided as the boundary condition for the
inlet. The pressure at the outlet is set to atmospheric pressure.
At all other boundaries, no-slip condition and defined as u = 0.
The fluid flow loading acting on the microcantilever is
defined as the force per area, similar to [30]:
FT = n (p I + (u + (u)T ))

(3)

where n is the normal unit vector of the boundary, u is the

velocity field on the cantilever surface pointing out from fluid.
FT is the fluid loading consists of pressure and viscous forces.
The first term in equation (3) is the pressure gradient extracted
from the fluidic simulation results. The second term is the viscous component of the force depending on the velocity and the
dynamic viscosity of the fluid. For the structural deformation,

NEZHAD et al.: PDMS MICROCANTILEVER-BASED FLOW SENSOR INTEGRATION FOR LAB-ON-A-CHIP

607

Outlet
Microchannel

Microcantilever

500 m

Inlet
(a)
Fig. 9.

(b)

3-D numerical result of fluid structure interaction for a water flow rate of 0.2 mL/min. (a) CAD model. (b) Deflected condition of microcantilever.

the end point of microcantilever is considered as fixed boundary condition. The fluid loading is applied on the cantilever
surface. The PDMS microcantilever is considered isotropic
with Youngs modulus of estimated K value of 802 KPa
(Fig. 6) and Poissons ratio of 0.45. The microchannels cross
section has 840 m width and 240 m depth.
COMSOL Multiphysics 3.5 software is used to model the
fluid flow within the channel and to solve the governing
equations of fluid structure interaction. The iterative coupling
with individual systems of equations is solved for the fluid
and the structure. Coupling is achieved through the transfer of
fluid structure interaction between the fluid and the structure
within a nonlinear iteration loop. Since the microcantilever
has a large deformation and the fluid flow domain also
changes considerably, the moving mesh application mode is
used to model the fluidic domain deformation along with the
mechanical structure. The deflection of microcantilever under
the flow rate of 0.2 ml/min is shown in Fig. 9b.
The numerical results for the microcantilever response to
various flow rates are compared with the experimental data
in Fig. 10. The comparison of simulation and experimental
results confirm the reasonable accuracy of the model presented
for the PDMS microcantilever for the range of flow rates
considered. For more linear response, one could use the linear
range of the response (00.6 ml/min).
Since the accuracy of the measurement is limited by the
1.76 m resolution of the image taken by optical microscopy,
the minimum flow rate that can be detected in the present
combination of flow meter and optical camera is 35 l/min as
estimated by FE simulation. This corresponds to a 0.8 N
resolution in equivalent force at the tip. By employing a
thinner top PDMS layer, higher resolution optical objectives
can be used to increase the quality of the image taken by the
microscope and improve the minimum detectable flow rate.
Additionally, the sensitivity of the cantilever can be increased
by modifying the geometry of the cantilever. The present
flow sensor has special application in environments with no
electric field hence, integration of electrical elements such as
piezoresistive or piezoelectric elements to the sensing element
should be avoided. However, this flow sensor can be integrated
with optical waveguides within the PDMS microcantilever in

Fig. 10. Comparison of simulation and experiment results for the deflection
of the PDMS microcantilever for different flow rates.

order to provide an electrical readout for online monitoring of

the flow rate [31]. The performance of the flow sensor was
tested by a syringe pump. However, for practical use in low
flow rates, it has to be calibrated with a more accurate flow
instrument.
VII. C ONCLUSION
A high aspect ratio PDMS microcantilever-based flow sensor has been fabricated using a multilayer PDMS fabrication
process. The stiffness of the cantilever has been found by a
precision balance setup and the elastic modulus is estimated
by linear beam theory. The performance of the flow sensor
has been tested by introducing different flow rates into the
microfluidic device and monitoring the cantilever response to
the fluid force using optical microscopy method. The practical
flow range of 0-600 l/min was achieved for operation with
water. The minimum flow rate of 35 l/min was estimated
using finite element simulation. The sensitivity of the proposed
flow sensor can be increased either by modifying the dimension of the microcantilever or by employing higher resolution
optical objectives.

608

IEEE SENSORS JOURNAL, VOL. 13, NO. 2, FEBRUARY 2013

R EFERENCES
[1] S. H. Cho, C. H. Chen, F. S. Tsai, J. Godin, and Y. H. Lo, Human
mammalian cell sorting using microfabricated fluorescence activated cell
sorter, Lab Chip, vol. 10, no. 12, pp. 15671573, 2010.
[2] A. Chandrasekaran and M. Packirisamy, Integrated biophotonic micro
total analysis systems for flow cytometry and particle detection, Proc.
SPIE, Photon. North, vol. 7386, pp. 783603-1783603-8, Jan. 2009.
[3] M. Dijkstra, M. J. Boer, J. W. Berenschot, T. S. Lammerink,
R. J. Wiegerink, and M. Elwenspoek, Miniaturized flow sensor with
planar integrated sensor structures on semicircular surface channels, in
Proc. IEEE Micro Electro Mech. Syst. Conf., Jan. 2007, pp. 123126.
[4] Y. Wang, C. Lee, and C. Chiang, A MEMS-based air flow sensor with
a free-standing microcantilever structure, Sensors, vol. 7, no. 10, pp.
23892401, 2007.
[5] S. Hossain, M. Packirisamy, and S. Rakheja, Dynamic behavior of
microcantilever under periodic flow of viscous fluids, in Proc. Vib. Eng.
Tech Mach. Conf, New Delhi, India, Dec. 2010.
[6] A. Quist, A. Chand, S. Ramachandran, D. Cohena, and R. Lal, Piezoresistive cantilever based nanoflow and viscosity sensor for microchannels, Lab Chip, vol. 6, no. 11, pp. 14501454, 2006.
[7] V. Lien and F. Vollmer, Microfluidic flow rate detection based on
integrated optical fiber, Lab Chip, vol. 7, no. 10, pp. 13521356, 2007.
[8] N. Q. Balaban, U. S. Schwarz, D. Riveline, P. Goichberg, G. Tzur,
I. Sabanay, D. Mahalu, S. Safran, A. Bershadsky, L. Addadi, and B.
Geiger, Force and focal adhesion assembly: A close relationship studied
using elastic micropatterned substrates, Nature Cell Biol., vol. 3, pp.
466472, Apr. 2001.
[9] A. Kuoni, R. Holzherr, M. Boillat, and N. Rooij, Polyimide membrane
with ZnO piezoelectric thin film pressure transducers as a differential
pressure liquid flow sensor, J. Micromech. Microeng., vol. 13, no. 4,
pp. 103107, 2003.
[10] M. Terao, T. Akutsu, and Y. Tanaka, Non-wetted thermal micro flow
sensor, in Proc. SICE Annu. Conf., Sep. 2007, pp. 20842088.
[11] A. Rasmussen, C. Mavriplis, M. E. Zaghloul, O. Mikulchenko, and
K. Mayaram, Simulation and optimization of a microfluidic flow
sensor, Sensors Actuat. A, Phys., vol. 88, no. 2, pp. 121132, Feb.
2001.
[12] M. Ashauer, H. Glosch, F. Hedrich, N. Hey, H. Sandmaier, and W. Lang,
Thermal flow sensor for liquids and gases based on combinations of
two principles, Sensors Actuat. A, Phys., vol. 73, nos. 12, pp. 713,
Mar. 1999.
[13] I. K. Lin, Y. M. Liao, Y. Liu, K. S. Ou, K. S. Chen, and X. Zhang,
Viscoelastic mechanical behavior of soft microcantilever-based force
sensors, Appl. Phys. Lett., vol. 93, no. 25, pp. 251256, Dec. 2008.
[14] J. C. Ltters, W. Olthuis, P. H. Veltink, and P. Bergveld, The mechanical
properties of the rubber elastic polymer polydimethylsiloxane for sensor
applications, J. Micromech. Microeng., vol. 7, no. 3, pp. 145152, 1997.
[15] M. A. Greminger and B. J. Nelson, Vision-based force measurement,
IEEE Trans. Pattern Anal. Mach. Intell., vol. 26, no. 3, pp. 290293,
Mar. 2004.
[16] S. Pintoa, P. Alvesa, C. M. Matosb, A. C. Santosb, L. R. Rodriguesc,
J. A. Teixeirac, and M. H. Gil, Poly(dimethyl siloxane) surface modification by low pressure plasma to improve its characteristics toward
biomedical applications, Colloids Surf. B, Biointerf., vol. 81, no. 1, pp.
2026, Nov. 2010.
[17] F. Sasoglu, A. Bohl, and B. Layton, Design and microfabrication of a
high-aspect-ratio PDMS microbeam array for parallel nanonewton force
measurement and protein printing, J. Micromech. Microeng., vol. 17,
no. 3, pp. 623632, Mar. 2007.
[18] X. Liu, K. Kim, Y. Zhang, and Y. Sun, Nanonewton force sensing and
control in microrobotic cell manipulation, Int. J. Robot. Res., vol. 28,
no. 8, pp. 10651076, Aug. 2009.
[19] J. L. Tan, J. Tien, D. M. Pirone, D. S. Gary, K. Bhadriraju, and
C. S. Chen, Cells lying on a bed of microneedles: An approach to
isolate mechanical force, Proc. Nat. Acad. Sci., vol. 100, no. 4, pp.
14841493, Jan. 2003.
[20] Y. Zhang, C. W. Lo, J. A. Taylor, and S. Yang, Replica molding of highaspect-ratio polymeric nanopillar arrays with high fidelity, Langmuir,
vol. 22, no. 20, pp. 85958601, Sep. 2006.
[21] S. Timoshenko and J. M. Gere, Mechanics of Materials. New York: Van
Nostrand Reinhold, 1972.
[22] T. K. Kim, J. K. Kim, and C. Jeong, Measurement of nonlinear
mechanical properties of PDMS elastomer, J. Microelectron. Eng.,
vol. 88, no. 8, pp. 19821985, Aug. 2011.

[23] L. C. S. Nunes, Mechanical characterization of hyperelastic polydimethylsiloxane by simple shear test, Mater. Sci. Eng., vol. 528, no. 13,
pp. 17991804, Jan. 2011.
[24] Y. Xiang and D. A. LaVan, Analysis of soft cantilevers as force
transducers, Appl. Phys. Lett, vol. 90, no. 13, pp. 133901-1133901-3,
Mar. 2007.
[25] J. R. Anderson, D. T. Chiu, R. J. Jackman, O. Cherniavskaya,
J. C. McDonald, H. Wu, S. H. Whitesides, and G. M. Whitesides,
Fabrication of topologically complex three-dimensional microfluidic
systems in PDMS by rapid prototyping, Anal. Chem., vol. 72, no. 14,
pp. 31583164, Jun. 2000.
[26] J. H. Kang, E. Um, and J. Park, Fabrication of a poly(dimethylsiloxane)
membrane with well-defined through-holes for three-dimensional
microfluidic networks, J. Micromech. Microeng., vol. 19, no. 4, pp.
045027-1045027-6, 2009.
[27] U. Kloter, H. Schmid, H. W. Michel, and D. Juncker, High-resolution
patterning and transfer of thin PDMS films: Fabrication of hybrid selfsealing 3D microfluidic systems, in Proc. IEEE MEMS Conf., Aug.
2004, pp. 745748.
[28] M. Liu, J. Sun, Y. Sun, C. Bock, and Q. Chen, Thicknessdependent mechanical properties of polydimethylsiloxane membranes,
J. Micromech. Microeng., vol. 19, no. 3, pp. 035028-1035028-4, 2009.
[29] K. R. Stein and R. J. Benney, Fluid-structure interactions of a round
parachute: Modeling and simulation techniques, J. Aircraft, vol. 38,
no. 5, pp. 800808, 2001.
[30] S. Basak and A. Ramana, Hydrodynamic loading of microcantilevers
vibrating in viscous fluids, J. Appl. Phys., vol. 99, no. 11, pp. 1149061114906-10, Jun. 2006.
[31] A. S. Nezhad, M. Ghanbari, C. G. Agudelo, M. Packirisamy, and
R. Bhat, A new polydimethylsiloxane (PDMS) microcantilever with
integrated optical waveguide for biosensing application, in Proc. Photon. North Conf., Jun. 2012, pp. 68.

Amir Sanati Nezhad received the B.Sc. degree in

mechanical engineering from the Isfahan University
of Technology, Isfahan, Iran, and the M.Sc. degree
in mechanical engineering and mechatronics from
the Amirkabir University of Technology (Tehran
Polytechnic), Tehran, Iran. He is currently pursuing
the Ph.D. degree at the Optical Bio-Microsystems
Laboratory, Mechanical and Industrial Engineering
Department, Concordia University, Montreal, QC,
Canada.
His current research interests include optical bioMEMS, lab-on-a-chip, and single-cell analysis within integrated microfluidic
chips.

Mahmood Ghanbari received the B.Sc. degree

from the Isfahan University of Technology, Isfahan, Iran, and the M.Sc. degree from Tarbiat
Modares University, Tehran, Iran, in 2006 and 2009,
respectively, both in mechanical engineering. He is
currently pursuing the Ph.D. degree at the Optical Bio Microsystems Laboratory, Mechanical and
Industrial Engineering Department, Concordia University, Montreal, QC, Canada.
His current research interests include integrated
microsystems for biological applications, lab-on-achip, single-cell analysis, microfluidics, biophotonics, and micro-sensors and
actuators.

NEZHAD et al.: PDMS MICROCANTILEVER-BASED FLOW SENSOR INTEGRATION FOR LAB-ON-A-CHIP

Carlos G. Agudelo received the B.Sc. degree in

electrical engineering from the Universidad Nacional
de Colombia, Cundinamarca, Colombia, and the
M.S. degree in electrical engineering from the cole
Polytechnique de Montral, Montral, QC, Canada,
in 2001 and 2008, respectively. He is currently
pursuing the Ph.D. degree at the Department of
Mechanical and Industrial Engineering, Concordia
University, Montreal.
He was a Graduate Research Assistant with Andes
University, Bogot, Colombia, from 2000 to 2002.
From 2003 to 2004, he was a Research Fellow with the Technological University of Cartagena, Cartagena, Colombia. In 2008, he was a Software Developer
with CAE, Montreal. His current research interests include BioMEMS, control
systems, and software development.

Muthukumaran Packirisamy received the B.S.

degree from the University of Madras, Chennai,
India, the M.S. degree from the Indian Institute of
Technology, Chennai, and the Ph.D. degree from
Concordia University, Montreal, QC, Canada.
He is currently a Professor and the Concordia Research Chair of optical bioMEMS with the
Department of Mechanical and Industrial Engineering, Concordia University. He has experience with
many microelectromechanical systems industries in
Canada. He is currently involved in the development
of bioMEMS devices in collaboration with industry. He has authored or coauthored more than 225 articles published in journals and conference proceedings. He holds nine patents. His current research interests include optical
bio-MEMS, integration of microsystems, and micro- and nanointegration.
Prof. Packirisamy was a recipient of the I. W. Smith Award from the Canadian Society for Mechanical Engineers, the Concordia University Research
Fellowship, the Petro Canada Young Innovator Award, and the ENCS Young
Research Achievement Award. He is a fellow of the Canadian Society for
Mechanical Engineers.

609

Rama Bhat received the Bachelors degree in engineering from Karnataka Regional Engineering College, Srinivasanagar, India, (National Institute of
Technology, Karnataka) in 1966, and the M.Tech.
and Ph.D. degrees in mechanical engineering from
IIT Madras, Chennai, India, in 1968 and 1972,
respectively.
He is a Professor of Mechanical and Industrial
Engineering with Concordia University, Montreal,
QC, Canada. He has trained many Ph.D. students
in these areas since he joined the Department of
Mechanical and Industrial Engineering in 1979. His current research interests
include mechanical vibrations, vehicle dynamics, structural acoustics, rotor
dynamics, dynamics of micro-electro-mechanical systems.
Dr. Bhat is a fellow of the Canadian Society for Mechanical Engineering,
the Engineering Institute of Canada, the American Society of Mechanical
Engineers, and the Indian Institution of Engineers. He served as the President
of the Canadian Society for Mechanical Engineering from 2004 to 2006. He
received the prestigious NASA Award for Technical Innovation for his contribution in developing "PROSSSProgramming Structured Synthesis System."
He proposed the use of Boundary Characteristic Orthogonal Polynomials for
use in the Rayleigh Ritz Method in 1985.

Anja Geitmann received the B.Sc. degree in

exchange studies from Oregon State University, Corvallis, and Stockholm University, Stockholm, Sweden, the M.Sc. degree in biology from the University
of Konstanz, Konstanz, Germany, the Ph.D. degree
from the University of Siena, Siena, Italy, in 1997.
She was a Post-Doctoral Fellow with Laval University, Quebec City, QC, Canada, and Wageningen University, Wageningen, The Netherlands. She
joined the Institut de Recherche en Biologie Vgtale
and the Department of Biological Sciences, University of Montreal, Montreal, QC, in 2001, where she supervises a team of
biologists and engineers. Her current research interests include mechanical
aspects of plant cell growth and morphogenesis.
Dr. Geitmann is the Vice-President of the Microscopical Society of Canada
and an Executive Member of the Canadian Society of Plant Biologists. Her
publications and outreach activities can be accessed at www.geitmannlab.org.