Home Search Collections Journals About Contact us My IOPscience

Characterization of electrowetting actuation on addressable single-side coplanar electrodes

This content has been downloaded from IOPscience. Please scroll down to see the full text.

2006 J. Micromech. Microeng. 16 2053

(http://iopscience.iop.org/0960-1317/16/10/018)

View the table of contents for this issue, or go to the journal homepage for more

Download details:

IP Address: 130.220.71.25
This content was downloaded on 01/07/2015 at 09:55

Please note that terms and conditions apply.

INSTITUTE OF PHYSICS PUBLISHING JOURNAL OF MICROMECHANICS AND MICROENGINEERING
J. Micromech. Microeng. 16 (2006) 2053–2059 doi:10.1088/0960-1317/16/10/018

Characterization of electrowetting
actuation on addressable single-side
coplanar electrodes
Ui-Chong Yi1 and Chang-Jin Kim
Mechanical and Aerospace Engineering Department, University of California,
Los Angeles (UCLA), CA, USA
E-mail: uyi@coremicrosolutions.com

Received 27 April 2006, in final form 22 June 2006

Published 25 August 2006
Online at stacks.iop.org/JMM/16/2053

Abstract
This paper studies and characterizes electrowetting-on-dielectric (EWOD)
actuations on coplanar electrodes with an electrode-free cover plate or no
cover plate. By arranging driving and reference electrodes on one plate,
such an EWOD configuration can accommodate more sensing mechanisms
from above and thus allows increased flexibility for system development.
Various coplanar electrodes are tested for contact angle changes by EWOD
with a focus on the effect of the percentage gap between electrodes and are
found to be in good agreement with a simple analytical model. The
droplet-moving devices demonstrate the successful moving, cutting and
merging of droplets (∼0.1 µl) in a parallel-plate configuration, i.e., between
the driving plate with coplanar electrodes and the passive plate with no
electrode. EWOD actuation in single plate configuration, i.e. no cover plate,
is also demonstrated, adding additional flexibility for the system design.
(Some figures in this article are in colour only in the electronic version)

1. Introduction wave [5], electrostatic [6], dielectrophoretic [7] and, most

commonly, electrowetting [8, 9]. A common feature of
Microfluidic technology plays a critical role in lab-on-a-chip the droplet-driving schemes is that the actuations are local.
systems, or micro-total analysis systems (µ-TAS) [1, 2], Since no pressure is needed, digital microfluidic systems
by delivering chemical or biological samples and reagents can be built without microchannels, pressure sources (e.g.,
to processing components for data collections or on-chip micropumps) or regulatory elements (e.g., microvalves). This
sample preparations. Currently, the majority of lab-on- greatly simplifies the device and system both in design and
a-chip systems rely on a continuous microfluidics flowing fabrication. The simplicity further presents additional unique
liquid through fixed passages (e.g., microchannels) by an opportunities such as the reduction of dead volumes and the
electrokinetic driving or by a pressure source on or off chip. possibility of portable systems.
A new approach in microfluidics, at times called ‘digital For electrowetting actuation, the configuration of
microfluidics’, uses discrete volumes of fluids rather than electrowetting-on-dielectric (EWOD), as opposed to the
channel-based continuous flows. Although first demonstrated historically known electrowetting on a metal, has become
by discrete vapor bubbles in a liquid-filled microchannel [3], the choice for aqueous liquids for its reversible operations
the complementary state of liquid droplets in a gas environment [8]. Demonstrated to manipulate aqueous droplets in air
is far more common today for digital microfluidics, as the [8, 10] and oil [9], EWOD-based microfluidics development
demands for microfluidics have been dominated by biomedical has reported many physical functions: creating, dividing and
and chemical applications. The droplet can be moved by merging droplets [10]; mixing different droplets [11, 12];
various actuation methods, including thermal [4], surface separating and concentrating particles in a droplet [13]; and
droplet printing [14]—all with little additional complications
1 Currently with Core Microsolutions, Inc. LA, CA USA. to the device. With the ability to create EWOD chips capable

0960-1317/06/102053+07$30.00 © 2006 IOP Publishing Ltd Printed in the UK 2053

U-C Yi and C-J Kim

Liquid
droplet γ LG

Dielectric
layer θ0
γ SL γ SG

Substrate
Electrode
(a) (a)

Activated
electrode
V
θV

Substrate

(b)

(b) Figure 2. Contact angle changes by electrical potential on the

EWOD surface: (a) initial state; (b) activated state.
Figure 1. Popular two-plate configuration EWOD chip with
asymmetric electrode arrangement. Once voltage is applied, the
change in the meniscus shape (from the dotted shape) caused by no cover plate) [17], the idea has not been studied or tested in
contact angle changes induces a gradient inside the droplet. (a) Top any detail.
view; (b) cross-section view. By putting all the electrodes necessary for EWOD
actuation on one plate, the other plate can independently
of manipulating multiple droplets on a two-dimensional grid accommodate other various mechanisms, such as
array [15] and to build an entire system on a printed circuit electrochemical, mechanical and optical sensing. Or the
board [16], a complete hand-held lab-on-a-chip system seems EWOD device can be designed without a cover plate for
within the horizon. some applications. Furthermore, we expect the coplanar
A popular configuration of EWOD chip today uses two electrode design will help building stacked EWOD chips,
parallel plates: a top plate with a fixed blanket referencing which we envision for future EWOD devices with increased
electrode and a bottom plate with patterned addressable driving capacities. With these future directions in mind, we verify the
electrodes (figure 1). Both electrodes are covered with an coplanar electrodes, analyze the EWOD mechanism on them,
insulating layer (e.g., SiO2 and Parylene R
), which is, in turn, characterize the performance for various electrode designs
coated with a hydrophobic layer (e.g., Teflon R
and Cytop R
). A and demonstrate common droplet operations, expanding from
liquid droplet is squeezed in between the two plates, forming our preliminary result [18].
the initial droplet shape drawn in dotted lines in figure 1.
Once electrical potential is applied between the top reference 2. EWOD actuation without the top reference
electrode and the bottom driving electrode at the droplet edge, electrode
the contact angles change accordingly, generating a pressure
gradient inside the droplet and driving it [8, 9]. The top 2.1. Contact angle change on coplanar electrodes
reference electrode is essential for stable actuation since it
provides the ground for the potential drops across insulating Electrowetting enables the control of surface wettability by
layers. changing the surface free energy through applied potentials.
Although the two-plate configuration provides many Figure 2 illustrates the case of a sessile drop for EWOD,
advantages (e.g., reliable droplet volume control, gravity where electric potential is applied between the liquid and the
insensitivity), one drawback is the difficulty in integrating electrodes embedded underneath a dielectric layer. When a
additional functions on the chip (e.g., sensing devices) voltage is applied, the contact angle of the liquid is decreased
without impacting electrode area/arrangement or interfering (figure 2(b)) from its initial contact angle (figure 2(a)). The
with EWOD actuation. While optical sensing or actuation relation between applied voltages and contact angle changes
is compatible because it can be added off chip by using can be formulated by combining Lippmann’s equation and
transparent plates (e.g., glass coated with indium tin oxide Young’s equation:
(ITO) electrodes), other mechanisms face steeper challenges εr ε0 2
cos θV − cos θ0 = V (1)
in integration onto the EWOD chip. It would be most desirable 2γLG t
if the top plate has no role in EWOD actuation, so that other where θ V and θ 0 are corresponding contact angles at applied
functions can be built on or operate through the plate. To obtain voltage V and 0 V, respectively, ε r and ε 0 are the dielectric
the above goal, in this paper, we develop and characterize constant of the insulating layer and the permittivity of vacuum,
a EWOD plate containing both the driving and reference respectively, γ LG is liquid–gas interfacial energy and t is the
electrodes coplanar. Although coplanar placement of both thickness of the dielectric layer [19].
electrodes on one plate has been known, as mentioned for a If a voltage is applied between two electrodes placed
two-plate configuration [9] or demonstrated for one-plate (i.e., coplanar on the surface (figure 3(b)), equation (1) needs to

2054
 Characterization of electrowetting actuation on addressable single-side coplanar electrodes

V
Liquid
Gap area droplet
Dielectric
layer θV
Electrode
(driving)
Electrode Substrate
(reference)
Electrode Electrode
(a) (reference) (b) (driving)

(c)

Figure 3. EWOD of a sessile drop on single-side coplanar electrodes: (a) electrode arrangement (top view); (b) cross-section view with
voltage applied; (c) contact angle change on coplanar electrodes.

be modified to account for the gap areas of insulating material 2.2. Driving a droplet on coplanar electrodes in a
between the adjacent electrodes (figure 3(a)). By assuming parallel-plate configuration with a passive top plate
that the area of each sub-electrode (e.g. ground or signal
electrode) covered by the droplet is much smaller than the total Similar to the modification made to equation (1) to reach
area covered by the droplet and thus averaging the changes equation (3) for the sessile drop cases of figures 2 and 3,
in interfacial energy above the electrodes and the gap area, the equation describing the EWOD driving pressure on the
the dependence of the solid–liquid interfacial tension γ SL at coplanar electrodes under the parallel-plate configuration of
applied voltage V is modified from the original Lippmann’s figure 1 needs to be modified accordingly. In the coplanar
equation as electrode configuration, one driving unit is composed of

 multiple coplanar sub-electrodes, as shown in figure 4(a). By
εr ε0 Ad 2 Ar 2 Ag 2
γSL (V ) − γSL (0) = − Vd + Vr + Vg (2) applying potentials among the sub-electrodes within a driving
2t At At At
unit, the contact angle change occurs only at the bottom right
where γ SL(0) is the interfacial tension with no voltage applied,
solid-liquid interface (as shown in figure 4(b)), unlike a typical
Ad, Ar and Ag are the areas over the driving electrodes, the
top–bottom electrode configuration of figure 1(b), where the
reference electrodes and the gap areas between the electrodes,
respectively, At is the total combined area, and Vd, Vr and Vg top two end interfaces and right bottom interface have contact
are voltages across the dielectric layer above each respective angle changes. The driving pressure in the droplet on the
area. The values of Vd and Vr are inversely proportional to their co-planar electrodes under the parallel-plate configuration can
relative ratio of the area. In equation (2), Vg = 0 V because no finally be related to the contact angle changes through the
electrode is embedded underneath the gap area. Combining Laplace equation as

2
2 
equation (2) with Young’s equation, we obtain the relation εr ε0 Ad Ar Ar Ad
between the applied voltage and the contact angle change for P = + V 2 (4)
2td At Ad + Ar At Ad + Ar
the coplanar electrode design2 :

2
εr ε0 Ad Ar where P is the internal pressure difference and d is the height
cos θV − cos θ0 = of the channel defined by the gap between the two plates.
2γLG t At Ad + Ar

2 
Ar Ad 2.3. Driving a droplet on coplanar electrodes in an
+ V2 (3)
At Ad + Ar open-plate configuration with no cover plate
As seen in equation (3), the contact angle changes are related Driving of a sessile droplet on coplanar electrodes in an
to the area ratio among driving electrodes, reference electrodes open-plate configuration (i.e., no cover plate) is schematically
and the gaps between them. To maximize the angle changes, illustrated in figure 4(c). The maximum driving force is
the modifying geometric factor in the large bracket in the achieved through the same optimizing scheme for the electrode
equation needs to be optimized; for a given voltage, the design described above, since the internal pressure gradient is
maximum angle changes are obtained when Ad and Ar are directly proportional to the contact angle changes. However,
equal and Ag is minimized. This maximum value is 25% of the droplet changes its shape into complex geometries
the figure 2 case and about 50% of what the two-plate with varying internal pressure distribution, and a simplified
configuration of figure 1 would allow. expression for the net driving pressure cannot easily be found,
2 A similar expression was reported by Kwon and Lee [20]. calling for a numerical simulation to obtain a solution.

2055
U-C Yi and C-J Kim

Driving electrode unit

(activated)
0V

(a) V
70 V
Substrate

Driving electrode unit Droplet

(float)
120 V
Droplet
Top cover plate
(a) (b) (c) (d )
Hydrophobic Insulating
(b) layers layers V Figure 5. Contact angles of sessile drops on coplanar electrodes of
figure 3 for four different gap areas under EWOD actuation.
Substrate (a) 2% gap area, (b) 20% gap area, (c) 40% gap area, (d ) 60% gap
area.
Driving electrode unit
(float) to define the electrodes and the gaps between the electrodes.
For the insulating layer, ∼3000 Å silicon dioxide (SiO2) was
deposited by plasma-enhanced chemical vapor deposition,
Droplet
followed by spin-coating of ∼2000 Å Cytop R
to form the top
hydrophobic layer. For the passive top plate in figure 4(b), no
V
conductive layer was deposited and only Cytop R
(∼2000 Å)
(c) was spin-coated as a hydrophobic layer.
Substrate
Driving electrode unit 3. Experiment and result
(activated)
3.1. Contact angle measurement
Figure 4. Droplet moving by EWOD on coplanar electrodes.
(a) Top view showing driving electrode units, each consisting of The contact angles of a small sessile drop (∼5 µl) of DI water
sub-electrodes (four in this figure). (b) Driving in parallel-plate under EWOD actuation by coplanar electrodes were measured
configuration with the top plate having no electrode. (c) Driving in optically using a First Ten Angstroms (FTA R
4000A) system.
open-plate configuration (i.e., no cover plate). Note the initial
Unlike the tests in Moon et al [21], no platinum referencing
shapes of droplets before contact angle changes for the two cases,
drawn with broken lines. wire was inserted into the droplet because the reference
electrodes are included in the coplanar electrodes on the chip.
2.4. Experimental design and fabrications Only direct current (dc) voltages were used in the tests for
accurate measurements, since alternating current (ac) voltages
To demonstrate and characterize EWOD on single-side would change the menisci continuously and obstruct clear
coplanar electrodes, we designed two main experiments: pictures. The pictures and data obtained from the experiments
contact angle measurement and droplet actuation. For the are shown in figures 5 and 6. Parabolic decreases in contact
contact angle measurement tests, a sessile drop of deionized angles by the increase in applied voltages were observed, as
(DI) water was placed on a radial pattern of electrodes coated expected from equation (3). The resulting contact angles of
with dielectrics, as shown in figure 3. The radial pattern keeps each type of the chips are in relatively good agreement with
the contact angles along the edge of a droplet symmetric while the equation. The highest and lowest changes in the contact
voltages are applied. Keeping the areas of the two electrodes angle were observed with 2% and 60% gap areas, respectively,
equal, the gap areas between them were varied. Chips with as predicted, and the results quantified the influence of the gap
2%, 10%, 20%, 40% and 60% gap areas were designed to find area on the EWOD actuation. For 120 Vdc applied, the contact
the effect of varying ratios, the 2% gap mimicking the limiting angles of water on 2% chips dropped from ∼108◦ to ∼68◦ ,
case of no gap within the measurement error. For the droplet while those on 60% chips dropped from ∼108◦ to only ∼92◦ .
actuation tests, three different types of chips were designed, The results show that it is important to minimize the gap area
each with a set of two, four or six individually addressable to design effective coplanar EWOD devices.
sub-electrodes comprising a driving unit (or pad), as shown in During the contact angle measurement tests, we have not
figure 4(a) for a set of four. For each design, the gap areas observed the contact angle saturation phenomenon, previously
were minimized within our fabrication limitation (typically reported through several papers [21–25]. The resulting contact
∼4 µm line gap) to maximize the EWOD driving force. Each angles (∼68◦ for 2% gap) at ∼120 Vdc were below the
actuation pad is 1.4 mm by 1.4 mm in size, and the sizes of saturation values reported by Moon et al (∼75◦ with ∼1000 Å)
sub-electrodes are adjusted to fit in the pad size. [21], which used similar dielectric layers. Several physical
Both the contact angle measurement chips and the droplet mechanisms for the saturation phenomenon, including a zero
movement chips were designed and fabricated to have identical solid/liquid surface energy limit [22], a charge trapping
layers. The chips were micromachined with 4 inch Borofloat R
[23], a material deficiency [24] and ionization of air [25],
glass wafers. For the metal layer, ∼50 Å chromium and were proposed. However, the models proposed are mostly
∼1000 Å gold were evaporated successively and patterned inconclusive or specific to the systems studied, and the

2056
 Characterization of electrowetting actuation on addressable single-side coplanar electrodes

110 (a)
V

100
Contact angle (°)

90

(b)
V

80

60%
40%
20%
70 2%

Figure 7. Droplet movement by EWOD on single-side coplanar

electrodes, each unit consisting of two sub-electrodes in this figure:
0 20 40 60 80 100 120
(a) with top cover plate (i.e., droplet squeezed to a disk shape); and
Voltage (V) (b) with no cover plate (i.e., droplet formed to a spherical shape).
Figure 6. Contact angle changes for EWOD on coplanar electrodes
with 2%, 20%, 40% and 60% gap areas. Theoretical values (by
equation (3)) are plotted with solid lines while actual data are
plotted with geometric dots with error bars (±σ ). The data from (1)
chips with 10% gap area were also consistent but not plotted in this
graph for the readability of data.

(2)
phenomenon is yet to be clearly understood. Rather than trying
to fit our result to a saturation model, we apply our results to
the confirmed dependence on a dielectric layer thickness [21]
and suspect the lower contact angles could have been from (3)
the thicker dielectric layers (∼3000 Å). The applied voltages
simply might not have been high enough to cause saturations
in our contact angle measurement tests. Higher voltages
(4)
to reach the possible saturation could not be applied to the
chips due to the electrolysis starting around ∼125 V. Further
investigations including the contact angle measurement with
various dielectric thicknesses would be needed to clarify the (5)
deviation.
(a) (b) (c)
3.2. Moving droplets by EWOD
Figure 8. Transportation of a droplet by EWOD on coplanar
Coplanar electrode devices with three different numbers (2, 4 electrodes under parallel-plate configuration: (a) transportation on
two-sub-electrode pads; (b) transportation on six-sub-electrode
and 6) of sub-electrodes were tested for droplet manipulations pads; (c) cutting and merging of a droplet on two-sub-electrode pads.
with a setup consisting of (1) an intermediate testing stage,
which provides the electric connections for programmed
electrical signals from a custom-made EWOD control circuit The tests were done with and without a top cover plate, as
board [26] and (2) an optical microscope for visualization. schematically illustrated in figure 7. Note that the bottom glass
Only ac signals were used for the droplet movement tests to plate, containing both driving and reference electrodes covered
reduce electrolysis on the chip surfaces and for more reliable with insulating layers, provides all the EWOD actuations. In
actuations under varying experiment conditions. Although the parallel-plate configuration of figure 7(a), the top glass
only ac signals were used, the equations derived previously are plate, containing no electrodes and only hydrophobically
still valid for estimating the transportation by EWOD. Using coated, has only a passive role of squeezing DI water droplets
Vrms, contact angle changes can be estimated since equivalent (∼0.1 µl) into a disk shape. The channel height (∼100 µm)
values of Vdc brings the same energy to the system. It was is controlled by spacers between the plates.
also observed that about the same Vrms are required for EWOD Operations were demonstrated as summarized in figure 8.
actuations compared to the actuations by Vdc. All the values All devices with three different numbers of sub-electrodes
of ac voltages are reported in rms in this paper. The frequency (two and six sub-electrodes shown in the figure) were able
of the signal was fixed at 1 kHz for all the tests for EWOD to move the droplets with minimum applied voltages of
actuations of droplets with and without cover plates. ∼60 Vac. With ∼65 Vac, droplet movements were found

2057
U-C Yi and C-J Kim

a liquid cannot be cut passively by this instability in the

parallel-plate configuration even from a columnar shape.
(1) For an open-plate configuration, the evaporation of liquid
could be a concern for future device developments. While
evaporation is minimized in a closed-plate configuration that
exposes only the droplet sides to air [8], it is much faster
for a sessile drop in an open-plate configuration. The
(2) evaporation can be minimized by operating the droplets in an
oil environment [9] or in a humid environment. The humidity
can be provided by sealing the plate [16] or placing the device
in a chamber with a humidity control. In a sealed device, it
was observed that a ∼1.2 µl droplet loses ∼15% of its volume
(3) after 8 days.
Without the need for accurate spacing between the top
(a) (b) and bottom plates, the droplet volumes are determined only
by the electrode sizes, which are defined by photolithography,
Figure 9. Manipulation of a droplet by EWOD on single-side allowing smaller droplets, if necessary. Be reminded, however,
coplanar electrodes on the open surface (i.e., no cover plate). The
that the volume of sessile droplets varies much more than
droplet volume is ∼2 µl, which is ∼20 times that of figure 8.
(a) Transportation on two-sub-electrode pads; (b) transportation on that of disk-shape droplets in the parallel-plates. We envision
six-sub-electrode pads. that a microfluidic system incorporating both open-plate and
parallel-plate configurations can enhance the detection and
more reliable over the entire driving pads without a failure. liquid handling by allowing much greater flexibility in system
Keeping the ratios of the gap areas to the electrode areas in design and moving various volumes of liquid with the same
all three cases of sub-electrodes to minimal (i.e. from ∼0.2% electrode designs.
for two-sub-electrode pads to ∼1.4% for six-sub-electrode
pads), we found no effect of the number of sub-electrodes per 4. Conclusions
pad on EWOD operations; all three types performed without
detectable differences and moving at ∼4.5 mm s−1 under given We have characterized the EWOD actuation on coplanar
voltages (∼65 Vac). In addition to droplet transportation, electrodes through the measurement of contact angle changes
splitting and merging of the droplets were also tested and demonstrated the operations with devices. The
(figure 8(c)), as described in Cho et al [10]. Although the same contact angle changes with corresponding electrode designs
voltages as transportation tests (∼65 Vac) worked for most were obtained through an optical measurement system and
cases, the split/merge operations in the figure were obtained compared with an analytical model. The basic digital fluidic
by applying higher voltages (75–80 Vac) for reliable repeated operations (e.g. moving, splitting and merging) have been
operations. demonstrated in a parallel-plate configuration, where one plate
Lastly, devices with interdigitated fingers [8, 9] between has no electrodes. The driving of droplets in the open-plate
two adjacent actuation pads were also tested to see the effect configuration with no cover plate has also been demonstrated.
of the fingers on actuations. However, we have not observed The results verify that the EWOD devices can be made with all
any difference between normal electrode patterns (i.e. straight the electrodes on one plate, opening the door for lab-on-a-chip
edges) and interdigitated patterns. systems incorporating various sensing schemes.
Operations on an open-plate configuration (i.e. no cover
plate) were confirmed, as shown in figure 9. The spherical Acknowledgments
droplet (DI water) on an electrode unit in this configuration is
∼2 µl, much larger than that in the parallel-plate configuration This work was supported by NASA through the Institute for
(∼0.1 µl). All devices with three different numbers of sub- Cell Mimetic Space Exploration (CMISE) at UCLA. The
electrodes (two and six sub-electrodes shown in the figures) authors would like to thank Dr Hyejin Moon and Professor
were able to move the droplets at ∼10 mm s−1 with ∼65 Vac. Robin L Garrell for their valuable discussions and suggestions
Droplets move faster in the open-plate configuration because for contact angle measurement tests. We also would like
the resistance by the top plate does not exist and the droplets to thank Jian Gong for sharing his experimental data about
can roll. In the open-plate configuration, a droplet cannot be droplet evaporation.
split in the same fashion. A droplet needs to be squeezed
enough between two plates to be cut actively by EWOD References
actuation [10], which is not possible with one plate. Instead, a
passive method may be used on the open surface. A column of [1] Erickson D and Li D 2003 Integrated microfluidic devices
liquid should first be formed on the surface, and the column is Anal. Chim. Acta 507 11–26
then cut into multiple droplets by hydrodynamic instability. [2] Reyes D R, Lossifidis D, Auroux P-A and Manz A 2002 Micro
total analysis systems: 1. Introduction, theory and
Typically employed in the dielectrophoretic (DEP) droplet technology Anal. Chem. 74 2623–36
microfluidics [7], this passive cutting method is limited to [3] Jun T K and Kim C-J 1998 Valveless pumping using traversing
liquid columns and cannot be applied to droplets. Conversely, vapor bubbles in microchannels J. Appl. Phys. 83 5658–64

2058
 Characterization of electrowetting actuation on addressable single-side coplanar electrodes

[4] Sammarco T S and Burns M A 1999 Thermalcapillary [16] Gong J, Fan S K and Kim C-J 2004 Portable digital
pumping of discrete drops in microfabricated analysis microfluidics platform with active but disposable
devices AIChE J. 45 350–66 lap-on-chip Proc.Conf. IEEE MEMS (Maastricht, the
[5] Strobl C J, Rathgeber A, Wixforth A, Gauer C and Scriba J Netherlands) pp 355–8
2002 Planar microfluidic processors Proc. IEEE [17] Kim C-J 2001 Integrated digital microfluidic circuits operated
Ultrasonics Symp. (Munich, Germany) vol 1 pp 255–8 by electrowetting-on-dielectrics (EWOD) principle
[6] Washizu M 1998 Electrostatic actuation of liquid droplets for BIOFLIPS Program Summary Book: DARPA/MTO
microreactor applications IEEE Trans. Ind. Appl. Principle Investigators’ Meeting (Isle of Palms, SC, USA)
34 732–7 pp 32–3
[7] Jones T B, Gunji M, Washizu M and Feldman M J 2001 [18] Yi U-C and Kim C-J 2005 EWOD actuation with
Dielectrophoretic liquid actuation and nanodroplet electrode-free cover plate Tech. Dig. Transducers 2005:
formation J. Appl. Phys. 89 1441–8 13th Int. Conf. on Solid-State Sensors, Actuators and
[8] Lee J, Moon H, Fowler J, Schoellhammer T and Kim C-J 2002 Microsystems (Seoul, Korea) vol 1 pp 89–92
Electrowetting and electrowetting-on-dielectric for [19] Berge B 1993 Electrocapillarity and wetting of insulator films
microscale liquid handling Sensors Actuators by water C. R. Acad. Sci., Paris II 317 157–63
A 95 259–68 [20] Kwon S and Lee L 2001 Focal length control by
[9] Pollack M G, Shenderov A D and Fair R B 2002 microfabricated planar electrodes-based liquid lens
Electrowetting-based actuation of droplets for integrated (µPELL) Tech. Dig. Transducers 2001: 11th Int. Conf. on
microfluidics Lab Chip 2 96–101 Solid-State Sensors, Actuators and Microsystems (Munich,
[10] Cho S K, Moon H and Kim C-J 2003 Creating, transporting, Germany) pp 1348–51
cutting, and merging liquid droplets by [21] Moon H J, Cho S K, Garell R L and Kim C-J 2002 Low voltage
electrowetting-based actuation for digital microfluidic electrowetting-on-dielectric J. Appl. Phys. 92 4080–7
circuits J. Microelectromech. Syst. 12 70–80 [22] Peykov V, Quinn A and Ralston J 2000 Electrowetting: a
[11] Fowler J, Moon H and Kim C-J 2002 Enhancement of mixing model for contact-angle saturation Colloid Polym. Sci.
by droplet-based microfluidics Proc. Conf. IEEE MEMS 278 789–93
(Las Vegas, NV, USA) pp 97–100 [23] Verhdijen H J J and Prins M W J 1999 Reversible
[12] Paik P, Pamula V K and Fair R B 2003 Rapid droplet mixers electrowetting and trapping of charge: model and
for digital microfluidic systems Lap Chip 3 253–9 experiments Langmuir 15 6616–20
[13] Cho S K and Kim C-J 2003 Particle separation and [24] Seyrat E and Hayes R 2001 Amorphous fluoropolymers as
concentration control for digital microfluidic systems insulators for reversible low-voltage electrowetting J. Appl.
Proc. Conf. IEEE MEMS (Kyoto, Japan) pp 686–9 Phys. 90 1383–6
[14] Yi U-C and Kim C-J 2004 Soft printing of droplets [25] Vallet M, Vallade M and Berge B 1999 Limiting phenomena
pre-metered by electrowetting Sensors Actuators for the spreading of water on polymer films by
A 114 347–54 electrowetting Eur. Phys. J. B 11 583–91
[15] Fan S K, de Guzman P-P and Kim C-J 2002 EWOD driving of [26] Lee J 2000 Microactuation by continuous electrowetting and
droplet on NxM grid using single-layer electrode patterns electrowetting: theory, fabrication, and demonstration
Tech. Dig. Solid State Sensor, Actuator, and Microsystems PhD Dissertation, University of California, Los Angeles,
Workshop (Hilton Head Island, SC, USA) pp 134–7 CA, USA

2059