- Email: [email protected]
Polydimethylsiloxane material as hydrophobic and insulating layer in electrowetting-on-dielectric systems
Microelectronics Journal 45 (2014) 1684–1690
Contents lists available at ScienceDirect
Microelectronics Journal journal homepage: www.elsevier.com/locate/mejo
Polydimethylsiloxane material as hydrophobic and insulating layer in electrowetting-on-dielectric systems D. Caputo a,n, G. de Cesare a, N. Lo Vecchio a, A. Nascetti b, E. Parisi a, R. Scipinotti a a b
Department of Information, Electronic and Telecommunication Engineering, University of Rome “La Sapienza”, Via Eudossiana 18, 00184 Roma, Italy Department of Astronautics, Electrical and Energy Engineering, University of Rome “La Sapienza”, Via Salaria 851/881, 00138 Roma, Italy
art ic l e i nf o
a b s t r a c t
Article history: Received 27 December 2013 Accepted 14 May 2014 Available online 14 June 2014
Open and closed electrowetting-on-dielectric (EWOD) systems based on a spin coated polydimethylsiloxane (PDMS) layer are presented. The PDMS layer acts as both insulation and hydrophobic material. Characterization, through sessile drop experiments, shows the hydrophobic behaviors of the PDMS and saturation of the contact angle at negative bias voltage applied to the droplet. This behavior is ascribed to trapped carrier in the PDMS layer and explains the movement of the droplet toward the grounded electrode found in the EWOD experiments. An electronic board controls all the signals needed for the actuation and sensing functionalities of the EWOD systems. Detection of drop position along the electrode array is successfully achieved by implementing the time-constant method, which evaluates the variation of electrode capacitance induced by the droplet presence on the PDMS surface corresponding to the metal electrode. The microfluidic operations (movement, dispensing and splitting) in both open and closed configurations have been verified and accomplished at voltages around 200 V. & 2014 Elsevier Ltd. All rights reserved.
Keywords: Electrowetting-on-dielectric PDMS Microfluidic Contact angle saturation
1. Introduction Lab-on-Chip (LoC) devices are miniaturized systems able to perform complex bio-chemical analysis by integrating, in a small chip of few square centimeters, several modules that implement the functionalities of a standard laboratory for applications such as DNA and protein analysis, and biomedical diagnostics [1–4]. These analytical and diagnostic aspects imply the movement of solutions along microfluidic channels using continuous liquid flow that is achieved, in most cases, by using off-chip actuators, as peristaltic pumps and valves [5]. However, the complexity arising from the connections required to couple the mechanical parts to the LoC and the need of large mechanical or electrical energy sources to move solutions through the system have prompted the investigation of different fluid movement techniques [6,7]. An interesting alternative to continuous-flow microfluidics is electrowetting-on-dielectric (EWOD) technology. The EWOD technique is able to handle very small fluid quantity (down to picoliter) varying the contact angle of an electrically conductive liquid droplet placed on a hydrophobic surface by means of an external electric field [8–10].
n
Corresponding author: Tel:þ 39 0644585832, fax: þ 39 0644585918. E-mail address: [email protected] (D. Caputo).
http://dx.doi.org/10.1016/j.mejo.2014.05.016 0026-2692/& 2014 Elsevier Ltd. All rights reserved.
Fig. 1 shows the typical EWOD structure, constituted by a metal electrode and an insulation hydrophobic layer, without (Fig. 1a) and with (Fig. 1b) a voltage applied between the droplet and electrode. The contact angle change as a function of applied voltage is regulated by the Young–Lippmann equation [10] where γLG, γSG, and γSL are, respectively, the liquid–vapor, solid–vapor and solid– liquid surface tension, d is the thickness of the dielectric layer, ε0 is the permittivity of free space, εr is its relative dielectric constant and V is the applied voltage. θ is the contact angle at the applied voltage V, while θ is the contact angle without applied voltage [10,11]. Two different electrowetting configurations are usually implemented: open and closed electrowetting. In the open configuration, the droplet of liquid is placed over a plate consisting of a substrate (usually glass) covered by patterned electrodes coated with a stacked structure of dielectric and hydrophobic layers. The electrode array provides the movement path; the dielectric sustains the electric field needed for the technique avoiding electrolysis while the hydrophobic layer provides the surface properties for the contact angle variation. In the closed configuration, the liquid is placed between two coplanar plates, the bottom and the top plate, separated by a spacer. The bottom plate is identical to the plate of the open configuration. The top plate, instead, consists of a substrate
D. Caputo et al. / Microelectronics Journal 45 (2014) 1684–1690
1685
Fig. 1. Contact angle in a basic EWOD structure without (a) and with (b) an applied voltage between the droplet and the electrode.
Fig. 2. Cross section (a) and top view (b) of the open configuration.
Fig. 3. (a) Cross section of the EWOD closed configuration and (b) particulars of the EWOD electrodes in the closed configuration showing the reservoir and the first electrode of the electrode array.
(usually glass also in this case) covered by a conductive layer, which acts as a counter-electrode, in turn coated with a thin hydrophobic layer. The open configuration enables displacement and mixing operations, while the closed configuration adds dispensing of droplets from a reservoir and splitting of droplets in two parts [12–15]. In this paper, a spin coated polydimethylsiloxane (PDMS) layer is utilized as dielectric and hydrophobic material in both open and closed EWOD configuration. The thin film properties of PDMS as material for EWOD applications have already been investigated [16], and it has been shown, in a sessile drop experiment, that a large change of the contact angle can be induced and that the contact angle can almost return to the initial value [17]. This prompted us to develop our system where the EWOD functionalities are managed by an electronic circuit that generates all the control signals to achieve the droplet movement and detects the presence of the droplet over the electrodes in order to verify the correctness of the fluid manipulation.
The paper is organized as follows: Section II reports the whole system description including the EWOD systems and the electronic board. Section III presents the experimental results achieved in characterizing the PDMS material and the open and closed EWOD configurations. Section IV draws the conclusions.
2. System description 2.1. EWOD configurations As reported above, we have implemented both the open and closed configurations. The cross section and top view of the open system are reported in Fig. 2. The device substrate is a 5 5 cm2 glass on which an array of adjacent thin film metal electrodes has been fabricated in order to create the droplet path. The array is a line of 7 square-shaped electrodes of 1.5 1.5 mm2 area spaced by 60 mm gaps. Each electrode is connected to a 2 9 mm2 pad through a 100 mm wide metal line.
1686
D. Caputo et al. / Microelectronics Journal 45 (2014) 1684–1690
The cross section of the closed configuration is reported in Fig. 3a. It is constituted by two glass substrates coupled together through a kapton tape acting as a 100 μm thick spacer. The bottom glass substrate is identical to the glass substrate of the open configuration, except for the geometry of the electrodes, which in this case includes, in addition to the 7-squared electrode linear array, a circular reservoir with a star-shaped interior pattern and a rectangular electrode penetrating inside the reservoir to promote dispensing (a zoom of this electrode zone is reported in Fig. 3b). The top substrate is a 5 5 cm2 glass on which a transparent conductive oxide and a thin layer of hydrophobic material have been uniformly deposited.
The open configuration and the bottom glass substrate of the closed configuration have been fabricated with the following steps: 1. Deposition by magnetron sputtering of a 30 nm/200 nm/30 nm Cr/Al/Cr metal layer; 2. patterning by wet etching of the metal layers, deposited in step 1, for the definition of the electrodes; and 3. deposition by spin coating (30 s @ 6000 rpm) of a 1 μm PolyDimethilSiloxane (PDMS) layer that acts both as insulation and hydrophobic layer (Fig. 2a). The used PDMS is Rhodorsil RTV 90700 (Siliconi Padova, Italy, PD). The top glass substrate of the closed configuration has been fabricated with the following steps: 1. Deposition by magnetron sputtering of a 100 nm thick layer of Indium Tin Oxide (ITO) acting as transparent conductive; and 2. deposition by spin coating of Teflon AF 1600 (by DuPont) diluted in Fluorinert FC-77 (by 3 M) acting as hydrophobic layer. The spin-coating has been performed with a first step at 500 rpm for 10 s followed by a second step at 1000 rpm for 30 s. The curing has been performed with three steps: 107 1C for 10 min, 165 1C for 5 min and 330 1C for 15 min. This procedure ensures a final Teflon thickness of 66 nm.
Fig. 4. Schematic representation of the switches and their driving circuit.
Fig. 5. Equivalent electrical circuit of the EWOD device without (top) and with (bottom) the water drop between the electrodes. The shunt resistance has been neglected, due to the insulation behavior of the PDMS layer.
Fig. 7. Measurement of the contact angle of the PDMS layer performed with a sessile drop experiment.
6
Without droplet With droplet
Voltage (V)
5 4 3 2 1 Vtl
0
0
20
40 25 s
60
80
100
120
Time (ms) 50 s
Fig. 6. Circuit for the detection of the droplet over the electrode (left). The operation principle is based on the determination of the discharge time of the electrode capacitance and its comparison to a calibration value determined by a specified voltage Vt1 (right).
D. Caputo et al. / Microelectronics Journal 45 (2014) 1684–1690
1687
Fig. 8. Schematic representation of the droplet movement in our open EWOD configuration without (left) and with (right) activated electrode.
C
B
A
STEP 1 HV
STEP 2 HV
STEP 3 HV
STEP 4
STEP 5
F
F
F
D
F
F
F
F Fig. 10. Three picture frames demonstrating the droplet movement.
F
F
F
F
HV
F
Fig. 9. Voltage sequence applied for achieving the droplet movement in the open configuration.
For both configurations, an electronic board controls movement and position of the droplets on the EWOD devices. The electronic board is constituted by the following components: 1. A power supply, which provides the high voltage (up to 200 V) needed for the drop movement; 2. a microcontroller (PIC18F4550, from Microchip Technology Inc.) which rules the electrode activation timing, detects the presence of the droplet on the electrode and handles the data communication with a computer via USB interface; and 3. an array of switches (part number AQW210EH, GU-E PhotoMOS, PANASONIC EW), able to connect each electrode can be connected to high voltage, ground potential or floating potential (as shown in Fig. 4) depending on the status of the two photoMOS relays driven by two pins of the microprocessor. The output current of each pin has been limited to 1.2 mA by a 2.2 kΩ series resistance.
Fig. 11. Picture frames reporting mixing operation in open configuration.
In order to detect the presence of the droplet on the electrode, each electrode can be connected to an electronic circuit able to measure the capacitance between two adjacent electrodes. This capacitance changes with the drop presence over the electrode, since it can be considered as a capacitance (CW) in parallel to the capacitance (CEL) constituted by the two electrodes surrounded by the hydrophobic material (PDMS) (Fig. 5). The circuit measures the capacitance discharge time through a 10 MΩ resistance connected between ground potential and both a pin of the microprocessor and the electrode to measure (Fig. 6). During its operation, this circuit is electrically disconnected from the external high voltage power supply. During the drop movement the electrical insulation is guaranteed by another switch connected between the resistor and the microcontroller.
1688
D. Caputo et al. / Microelectronics Journal 45 (2014) 1684–1690
The discharge time is compared to a time value tcal previously determined in a calibration step. A discharge time higher than tcal indicates the presence of the droplet over the electrode. We have found that for the above-described EWOD device the value of the capacitance measured without the droplet is around 1 pF and it increases up to 2 pF when a droplet of 2 ml of water is placed over the electrode. These measurements have been performed with an HP Impedance Analyzer 4192. Typical values of the discharge time with and without the droplet are around 50 μs and 25 μs, respectively, which, taking into account the discharge resistance, are in fairly good agreement with the measured capacitance.
3. Experimental results 3.1. PDMS characterization In order to characterize the electrowetting behavior of the PDMS a preliminary sessile drop experiment has been performed. Over a single square-shaped metal electrode of 2 2 mm2 area fabricated on an ultrasonically cleaned glass, a 1 μm thick layer of PDMS has been deposited by spin coating and a water droplet of 2 μl has been placed over the PDMS surface aligned with the electrode. A tungsten tip has been partially inserted into the droplet and a variable bias voltage has been applied to the tip. The underneath electrode has been set to ground potential. The contact angle between the droplet and the PDMS surface as function of the applied voltage is reported in Fig. 7, where it is evident that the wetting behavior of the liquid depends on the polarity of the applied voltage. In particular, the contact angle exhibits a variation from 1041 to 501 when a positive voltage (up to 200 V) is applied, while a saturation level of 901 occurs for negative voltage as low as 40 V. The saturation of contact angle has been widely studied in literature and has been ascribed to different reasons. One of the mostly accepted descriptions is the
presence of trapped charge in the insulation and/or the hydrophobic layer upon voltage application [18]. Trapped charges lower the electric field at the solid–liquid interface and reduce the electrowetting forces. Differently from literature results [19], achieved on 1000-Å -thick amorphous fluoropolymer (TEFLON AF1601 from Dupont) covering various dielectric layers, in our case the saturation occurs for negative voltage and this implies that in our PDMS film the charges causing the angle saturation have an opposite polarity to the one causing contact angle saturation in TEFLON AF. In particular, from electrostatic considerations we can infer that charges trapped inside the PDMS layer have a negative sign. The contact angle saturation occurring for negative voltages causes, in our experiments, the movement toward the grounded electrode as opposite to literature cases where the movement is toward the activated (biased) electrode. Indeed, the movement of a droplet lying between two planar biased electrodes (left side of Fig. 8) is explained observing that the droplet assumes an intermediate potential between the two adjacent electrodes. The portion of the droplet placed over the electrode at ground potential experiments a positive voltage if compared to the droplet polarization in a sessile drop experiment. On the other hand, the portion of the droplet placed over the electrode at high voltage experiments a relatively negative voltage. From Fig. 7, we see that, at positive voltage higher than 40 V, the contact angle value is lower than the one achieved at negative voltage. Therefore the droplet experiments an asymmetrical and unbalanced shape (right side of Fig. 8) that leads to a movement toward the grounded electrode. 3.2. Results on open configuration The functionality of open configuration has been tested transferring droplets of liquid along the linear array of electrodes. The voltage sequence applied, at the different time steps, is illustrated in Fig. 9. When the high voltage (HV) is applied to A and
Fig. 12. Sequence of frames reporting the dispensing of a droplet from a reservoir.
D. Caputo et al. / Microelectronics Journal 45 (2014) 1684–1690
the electrode B is grounded, the drop changes its shape and moves toward B (step 1). Switching the B potential from ground to floating potential, the droplet returns to its original symmetrical hydrophobic shape (step 2). When electrode C is grounded (step 3) the liquid moves toward C overlapping electrodes A, B and C. The movement from B to C is complete when potential A is changed from HV to floating potential (step 4). Step 5 is the starting point for the movement from C to D. The time interval between each step of Fig. 9 (i.e. the duration of each step) can be chosen independently. The experiments have been performed, following the above procedure, using droplets of 2 μl of water. We have found that the system successfully moved single droplets along the linear path in a repeatable way for time intervals above 100 min. Fig. 10 reports three frames captured during the experiment. The maximum speed achieved is 5 cm/s.
1689
Fig. 11 reports three frames of the movement of two droplets have been moved contemporarily from the outer electrodes of the array toward the center. The resulting droplet has been moved along the whole path to mix the two drops. 3.3. Results on closed configuration The functionality of the closed configuration has been verified dispensing, moving and splitting a drop of liquid. Fig. 12 illustrates the dispensing from the reservoir. The numbers refer to the sequential times, while the ‘on’ label refers to the voltage activation. The reservoir has been initially filled with a 1 μl droplet. The control strategy can be summarized as follows: 1. Activation of the second electrode of the reserve, the one closest to the electrode array, while the first electrode is left
Fig. 13. Sequence of frames reporting the movement of a droplet from a reservoir.
Fig. 14. Sequence of frames reporting the splitting of a droplet in the closed configuration.
1690
D. Caputo et al. / Microelectronics Journal 45 (2014) 1684–1690
floating. In this way, the drop in the reserve begins to deform and moves toward the electrode array (Fig. 12, frame 2); 2. activation of the first electrode of the array (Fig. 12, frame 3), which allows the coverage of the electrode; and 3. as soon as the drop covers a significant portion of the rectangular electrode, the second electrode of the reservoir is switched off while the first electrode of the reserve is activated (Fig. 12, frame 3 and frame 4). This pulls back part of the liquid into the reserve (Fig. 12, frame 5) and separates the rest (Fig. 12, frame 6). The movement along the electrodes has been easily implemented as illustrated in Fig. 13 where the frames are sequentially reported and the ‘on’ label refers to the voltage activation. The movement of the droplet is toward the activated electrode that, in our case, means grounded electrode. Finally, Fig. 14 reports the droplet splitting. A 0.2 μl drop has been placed on the electrode array covering three different electrodes of the bottom plate. The top plate has been connected to the high voltage; the two external electrodes of the bottom plate have been connected to ground, while the central one has been left floating. The voltage to the top plate has been gradually increased to achieve the splitting, that has occurred at 175 V.
4. Conclusions In this paper we have presented and characterized the performances of both open and closed EWOD systems based on the use of a PDMS material acting as both insulation and hydrophobic layer. The material characterization in a sessile drop experiment shows that the saturation of the contact angle occurs for negative voltage (around 40 V), differently from literature results where the contact angle saturation occurs for positive voltage. As a consequence of this behavior the activated electrode is the grounded one. The EWOD systems comprise an electronic board, equipped with a microcontroller, a series of switches, a high voltage power supply (up to 200 V) and a circuit for the detection of the drop between and over the electrodes. The drop detection has been implemented measuring the discharging time of the EWOD electrode through a pin of the microcontroller. The discharge time, in open configuration, with and without the drop has been found to be around 50 and 25 μs, respectively.
The movement, splitting and dispensing is controlled by the sequence of voltage levels (high voltage, ground or floating) applied to the adjacent electrodes and by the activation time of the voltage applied to each electrode. We have implemented the voltage control strategies for successfully achieving all the microfluidic operations. We have found that a reliable movement occurs at 150 V, the splitting at 175 V and the dispensing at 200 V. References [1] A. Manz, H. Becker, Microsystem Technology in Chemistry and Life Sciences, Springer Verlag, 1998. [2] S.C. Jakeway, A.J. de Mello, E.L. Russell, Miniaturized total analysis systems for biological analysis, Fresenius J. Anal. Chem. 366 (6–7) (2000) 525–539. [3] D. Caputo, G. de Cesare, C. Manetti, A. Nascetti, R. Scipinotti, Smart thin layer chromatography plate, Lab Chip 7 (8) (2007) 978–980. [4] J.W. Hong, S.R. Quake, Integrated nanoliter systems, Nat. Biotechnol. 21 (10) (2003) 1179–1183. [5] J.-H. Tsai, L. Lin, Micro-to-macro fluidic interconnectors with an integrated polymer sealant, J. Micromech. Microeng. 11 (5) (2001) 577. [6] C.-M. Ho, Y.-C. Tai, Micro-electro-mechanical-systems (mems) and fluid flows, Annu. Rev. Fluid Mech. 30 (1) (1998) 579–612. [7] O.C. Jeong, S.S. Yang, Fabrication and test of a thermopneumatic micropump with a corrugated p þ diaphragm, Sens. Actuators A: Phys. 83 (1) (2000) 249–255. [8] M. Washizu, Electrostatic actuation of liquid droplets for micro-reactor applications, IEEE Trans. Ind. Appl. 34 (4) (1998) 732–737. [9] B. Shapiro, H. Moon, R.L. Garrell, C.-J. Kim, et al., Equilibrium behavior of sessile drops under surface tension, applied external fields, and material variations, J. Appl. Phys. 93 (9) (2003) 5794–5811. [10] G. Lippmann, Relations entre les phenomenes electriques et capillaires (Ph.D. dissertation), Gauthier-Villars, 1875. [11] Jean-Christophe Baret Frieder Mugele, Electrowetting: from basic to applications, J. Phys.: Condens. Matter 17 (2005) R705–R774. [12] F.E. Torres, P. Kuhn, D. De Bruyker, A.G. Bell, M.V. Wolkin, E. Peeters, J.R. Williamson, G.B. Anderson, G.P. Schmitz, M.I. Recht, et al., Enthalpy arrays, Proc. Natl. Acad. Sci. USA 101 (26) (2004) 9517–9522. [13] S.K. Cho, H. Moon, C.-J. Kim, Creating, transporting, cutting, and merging liquid droplets by electrowetting-based actuation for digital microfluidic circuits, J. Microelectromech. Syst. 12 (1) (2003) 70–80. [14] M. Pollack, A. Shenderov, R. Fair, Electrowetting-based actuation of droplets for integrated microfluidics, Lab Chip 2 (2) (2002) 96–101. [15] S.K. Cho, H. Moon, J. Fowler, and C.-J. Kim, Splitting a liquid droplet for electrowetting-based microfluidics, in: Proceedings of the 2001 ASME International Mechanical Engineering Congress and Exposition, November, 2001, pp. 11–16. [16] D. Caputo, M. Ceccarelli, G. de Cesare, A. Nascetti, and R. Scipinotti, Lab-onglass system for dna analysis using thin and thick film technologies, in: MRS Proceedings, vol. 1191, no. 1, Cambridge University Press, 2009. [17] W. Dai, Y.-P. Zhao, The nonlinear phenomena of thin poly- dimethylsiloxane (pdms) films in electrowetting, Int. J. Nonlinear Sci. Numer. Simul. 8 (4) (2007) 519–526. [18] H. Verheijen, M. Prins, Reversible electrowetting and trapping of charge: model and experiments, Langmuir 15 (20) (1999) 6616–6620. [19] H. Moon, S.K. Cho, R.L. Garrell, C.-J. Kim, et al., Low voltage electrowetting-ondielectric, J. Appl. Phys. 92 (7) (2002) 4080–4087.
Contents lists available at ScienceDirect
Microelectronics Journal journal homepage: www.elsevier.com/locate/mejo
Polydimethylsiloxane material as hydrophobic and insulating layer in electrowetting-on-dielectric systems D. Caputo a,n, G. de Cesare a, N. Lo Vecchio a, A. Nascetti b, E. Parisi a, R. Scipinotti a a b
Department of Information, Electronic and Telecommunication Engineering, University of Rome “La Sapienza”, Via Eudossiana 18, 00184 Roma, Italy Department of Astronautics, Electrical and Energy Engineering, University of Rome “La Sapienza”, Via Salaria 851/881, 00138 Roma, Italy
art ic l e i nf o
a b s t r a c t
Article history: Received 27 December 2013 Accepted 14 May 2014 Available online 14 June 2014
Open and closed electrowetting-on-dielectric (EWOD) systems based on a spin coated polydimethylsiloxane (PDMS) layer are presented. The PDMS layer acts as both insulation and hydrophobic material. Characterization, through sessile drop experiments, shows the hydrophobic behaviors of the PDMS and saturation of the contact angle at negative bias voltage applied to the droplet. This behavior is ascribed to trapped carrier in the PDMS layer and explains the movement of the droplet toward the grounded electrode found in the EWOD experiments. An electronic board controls all the signals needed for the actuation and sensing functionalities of the EWOD systems. Detection of drop position along the electrode array is successfully achieved by implementing the time-constant method, which evaluates the variation of electrode capacitance induced by the droplet presence on the PDMS surface corresponding to the metal electrode. The microfluidic operations (movement, dispensing and splitting) in both open and closed configurations have been verified and accomplished at voltages around 200 V. & 2014 Elsevier Ltd. All rights reserved.
Keywords: Electrowetting-on-dielectric PDMS Microfluidic Contact angle saturation
1. Introduction Lab-on-Chip (LoC) devices are miniaturized systems able to perform complex bio-chemical analysis by integrating, in a small chip of few square centimeters, several modules that implement the functionalities of a standard laboratory for applications such as DNA and protein analysis, and biomedical diagnostics [1–4]. These analytical and diagnostic aspects imply the movement of solutions along microfluidic channels using continuous liquid flow that is achieved, in most cases, by using off-chip actuators, as peristaltic pumps and valves [5]. However, the complexity arising from the connections required to couple the mechanical parts to the LoC and the need of large mechanical or electrical energy sources to move solutions through the system have prompted the investigation of different fluid movement techniques [6,7]. An interesting alternative to continuous-flow microfluidics is electrowetting-on-dielectric (EWOD) technology. The EWOD technique is able to handle very small fluid quantity (down to picoliter) varying the contact angle of an electrically conductive liquid droplet placed on a hydrophobic surface by means of an external electric field [8–10].
n
Corresponding author: Tel:þ 39 0644585832, fax: þ 39 0644585918. E-mail address: [email protected] (D. Caputo).
http://dx.doi.org/10.1016/j.mejo.2014.05.016 0026-2692/& 2014 Elsevier Ltd. All rights reserved.
Fig. 1 shows the typical EWOD structure, constituted by a metal electrode and an insulation hydrophobic layer, without (Fig. 1a) and with (Fig. 1b) a voltage applied between the droplet and electrode. The contact angle change as a function of applied voltage is regulated by the Young–Lippmann equation [10] where γLG, γSG, and γSL are, respectively, the liquid–vapor, solid–vapor and solid– liquid surface tension, d is the thickness of the dielectric layer, ε0 is the permittivity of free space, εr is its relative dielectric constant and V is the applied voltage. θ is the contact angle at the applied voltage V, while θ is the contact angle without applied voltage [10,11]. Two different electrowetting configurations are usually implemented: open and closed electrowetting. In the open configuration, the droplet of liquid is placed over a plate consisting of a substrate (usually glass) covered by patterned electrodes coated with a stacked structure of dielectric and hydrophobic layers. The electrode array provides the movement path; the dielectric sustains the electric field needed for the technique avoiding electrolysis while the hydrophobic layer provides the surface properties for the contact angle variation. In the closed configuration, the liquid is placed between two coplanar plates, the bottom and the top plate, separated by a spacer. The bottom plate is identical to the plate of the open configuration. The top plate, instead, consists of a substrate
D. Caputo et al. / Microelectronics Journal 45 (2014) 1684–1690
1685
Fig. 1. Contact angle in a basic EWOD structure without (a) and with (b) an applied voltage between the droplet and the electrode.
Fig. 2. Cross section (a) and top view (b) of the open configuration.
Fig. 3. (a) Cross section of the EWOD closed configuration and (b) particulars of the EWOD electrodes in the closed configuration showing the reservoir and the first electrode of the electrode array.
(usually glass also in this case) covered by a conductive layer, which acts as a counter-electrode, in turn coated with a thin hydrophobic layer. The open configuration enables displacement and mixing operations, while the closed configuration adds dispensing of droplets from a reservoir and splitting of droplets in two parts [12–15]. In this paper, a spin coated polydimethylsiloxane (PDMS) layer is utilized as dielectric and hydrophobic material in both open and closed EWOD configuration. The thin film properties of PDMS as material for EWOD applications have already been investigated [16], and it has been shown, in a sessile drop experiment, that a large change of the contact angle can be induced and that the contact angle can almost return to the initial value [17]. This prompted us to develop our system where the EWOD functionalities are managed by an electronic circuit that generates all the control signals to achieve the droplet movement and detects the presence of the droplet over the electrodes in order to verify the correctness of the fluid manipulation.
The paper is organized as follows: Section II reports the whole system description including the EWOD systems and the electronic board. Section III presents the experimental results achieved in characterizing the PDMS material and the open and closed EWOD configurations. Section IV draws the conclusions.
2. System description 2.1. EWOD configurations As reported above, we have implemented both the open and closed configurations. The cross section and top view of the open system are reported in Fig. 2. The device substrate is a 5 5 cm2 glass on which an array of adjacent thin film metal electrodes has been fabricated in order to create the droplet path. The array is a line of 7 square-shaped electrodes of 1.5 1.5 mm2 area spaced by 60 mm gaps. Each electrode is connected to a 2 9 mm2 pad through a 100 mm wide metal line.
1686
D. Caputo et al. / Microelectronics Journal 45 (2014) 1684–1690
The cross section of the closed configuration is reported in Fig. 3a. It is constituted by two glass substrates coupled together through a kapton tape acting as a 100 μm thick spacer. The bottom glass substrate is identical to the glass substrate of the open configuration, except for the geometry of the electrodes, which in this case includes, in addition to the 7-squared electrode linear array, a circular reservoir with a star-shaped interior pattern and a rectangular electrode penetrating inside the reservoir to promote dispensing (a zoom of this electrode zone is reported in Fig. 3b). The top substrate is a 5 5 cm2 glass on which a transparent conductive oxide and a thin layer of hydrophobic material have been uniformly deposited.
The open configuration and the bottom glass substrate of the closed configuration have been fabricated with the following steps: 1. Deposition by magnetron sputtering of a 30 nm/200 nm/30 nm Cr/Al/Cr metal layer; 2. patterning by wet etching of the metal layers, deposited in step 1, for the definition of the electrodes; and 3. deposition by spin coating (30 s @ 6000 rpm) of a 1 μm PolyDimethilSiloxane (PDMS) layer that acts both as insulation and hydrophobic layer (Fig. 2a). The used PDMS is Rhodorsil RTV 90700 (Siliconi Padova, Italy, PD). The top glass substrate of the closed configuration has been fabricated with the following steps: 1. Deposition by magnetron sputtering of a 100 nm thick layer of Indium Tin Oxide (ITO) acting as transparent conductive; and 2. deposition by spin coating of Teflon AF 1600 (by DuPont) diluted in Fluorinert FC-77 (by 3 M) acting as hydrophobic layer. The spin-coating has been performed with a first step at 500 rpm for 10 s followed by a second step at 1000 rpm for 30 s. The curing has been performed with three steps: 107 1C for 10 min, 165 1C for 5 min and 330 1C for 15 min. This procedure ensures a final Teflon thickness of 66 nm.
Fig. 4. Schematic representation of the switches and their driving circuit.
Fig. 5. Equivalent electrical circuit of the EWOD device without (top) and with (bottom) the water drop between the electrodes. The shunt resistance has been neglected, due to the insulation behavior of the PDMS layer.
Fig. 7. Measurement of the contact angle of the PDMS layer performed with a sessile drop experiment.
6
Without droplet With droplet
Voltage (V)
5 4 3 2 1 Vtl
0
0
20
40 25 s
60
80
100
120
Time (ms) 50 s
Fig. 6. Circuit for the detection of the droplet over the electrode (left). The operation principle is based on the determination of the discharge time of the electrode capacitance and its comparison to a calibration value determined by a specified voltage Vt1 (right).
D. Caputo et al. / Microelectronics Journal 45 (2014) 1684–1690
1687
Fig. 8. Schematic representation of the droplet movement in our open EWOD configuration without (left) and with (right) activated electrode.
C
B
A
STEP 1 HV
STEP 2 HV
STEP 3 HV
STEP 4
STEP 5
F
F
F
D
F
F
F
F Fig. 10. Three picture frames demonstrating the droplet movement.
F
F
F
F
HV
F
Fig. 9. Voltage sequence applied for achieving the droplet movement in the open configuration.
For both configurations, an electronic board controls movement and position of the droplets on the EWOD devices. The electronic board is constituted by the following components: 1. A power supply, which provides the high voltage (up to 200 V) needed for the drop movement; 2. a microcontroller (PIC18F4550, from Microchip Technology Inc.) which rules the electrode activation timing, detects the presence of the droplet on the electrode and handles the data communication with a computer via USB interface; and 3. an array of switches (part number AQW210EH, GU-E PhotoMOS, PANASONIC EW), able to connect each electrode can be connected to high voltage, ground potential or floating potential (as shown in Fig. 4) depending on the status of the two photoMOS relays driven by two pins of the microprocessor. The output current of each pin has been limited to 1.2 mA by a 2.2 kΩ series resistance.
Fig. 11. Picture frames reporting mixing operation in open configuration.
In order to detect the presence of the droplet on the electrode, each electrode can be connected to an electronic circuit able to measure the capacitance between two adjacent electrodes. This capacitance changes with the drop presence over the electrode, since it can be considered as a capacitance (CW) in parallel to the capacitance (CEL) constituted by the two electrodes surrounded by the hydrophobic material (PDMS) (Fig. 5). The circuit measures the capacitance discharge time through a 10 MΩ resistance connected between ground potential and both a pin of the microprocessor and the electrode to measure (Fig. 6). During its operation, this circuit is electrically disconnected from the external high voltage power supply. During the drop movement the electrical insulation is guaranteed by another switch connected between the resistor and the microcontroller.
1688
D. Caputo et al. / Microelectronics Journal 45 (2014) 1684–1690
The discharge time is compared to a time value tcal previously determined in a calibration step. A discharge time higher than tcal indicates the presence of the droplet over the electrode. We have found that for the above-described EWOD device the value of the capacitance measured without the droplet is around 1 pF and it increases up to 2 pF when a droplet of 2 ml of water is placed over the electrode. These measurements have been performed with an HP Impedance Analyzer 4192. Typical values of the discharge time with and without the droplet are around 50 μs and 25 μs, respectively, which, taking into account the discharge resistance, are in fairly good agreement with the measured capacitance.
3. Experimental results 3.1. PDMS characterization In order to characterize the electrowetting behavior of the PDMS a preliminary sessile drop experiment has been performed. Over a single square-shaped metal electrode of 2 2 mm2 area fabricated on an ultrasonically cleaned glass, a 1 μm thick layer of PDMS has been deposited by spin coating and a water droplet of 2 μl has been placed over the PDMS surface aligned with the electrode. A tungsten tip has been partially inserted into the droplet and a variable bias voltage has been applied to the tip. The underneath electrode has been set to ground potential. The contact angle between the droplet and the PDMS surface as function of the applied voltage is reported in Fig. 7, where it is evident that the wetting behavior of the liquid depends on the polarity of the applied voltage. In particular, the contact angle exhibits a variation from 1041 to 501 when a positive voltage (up to 200 V) is applied, while a saturation level of 901 occurs for negative voltage as low as 40 V. The saturation of contact angle has been widely studied in literature and has been ascribed to different reasons. One of the mostly accepted descriptions is the
presence of trapped charge in the insulation and/or the hydrophobic layer upon voltage application [18]. Trapped charges lower the electric field at the solid–liquid interface and reduce the electrowetting forces. Differently from literature results [19], achieved on 1000-Å -thick amorphous fluoropolymer (TEFLON AF1601 from Dupont) covering various dielectric layers, in our case the saturation occurs for negative voltage and this implies that in our PDMS film the charges causing the angle saturation have an opposite polarity to the one causing contact angle saturation in TEFLON AF. In particular, from electrostatic considerations we can infer that charges trapped inside the PDMS layer have a negative sign. The contact angle saturation occurring for negative voltages causes, in our experiments, the movement toward the grounded electrode as opposite to literature cases where the movement is toward the activated (biased) electrode. Indeed, the movement of a droplet lying between two planar biased electrodes (left side of Fig. 8) is explained observing that the droplet assumes an intermediate potential between the two adjacent electrodes. The portion of the droplet placed over the electrode at ground potential experiments a positive voltage if compared to the droplet polarization in a sessile drop experiment. On the other hand, the portion of the droplet placed over the electrode at high voltage experiments a relatively negative voltage. From Fig. 7, we see that, at positive voltage higher than 40 V, the contact angle value is lower than the one achieved at negative voltage. Therefore the droplet experiments an asymmetrical and unbalanced shape (right side of Fig. 8) that leads to a movement toward the grounded electrode. 3.2. Results on open configuration The functionality of open configuration has been tested transferring droplets of liquid along the linear array of electrodes. The voltage sequence applied, at the different time steps, is illustrated in Fig. 9. When the high voltage (HV) is applied to A and
Fig. 12. Sequence of frames reporting the dispensing of a droplet from a reservoir.
D. Caputo et al. / Microelectronics Journal 45 (2014) 1684–1690
the electrode B is grounded, the drop changes its shape and moves toward B (step 1). Switching the B potential from ground to floating potential, the droplet returns to its original symmetrical hydrophobic shape (step 2). When electrode C is grounded (step 3) the liquid moves toward C overlapping electrodes A, B and C. The movement from B to C is complete when potential A is changed from HV to floating potential (step 4). Step 5 is the starting point for the movement from C to D. The time interval between each step of Fig. 9 (i.e. the duration of each step) can be chosen independently. The experiments have been performed, following the above procedure, using droplets of 2 μl of water. We have found that the system successfully moved single droplets along the linear path in a repeatable way for time intervals above 100 min. Fig. 10 reports three frames captured during the experiment. The maximum speed achieved is 5 cm/s.
1689
Fig. 11 reports three frames of the movement of two droplets have been moved contemporarily from the outer electrodes of the array toward the center. The resulting droplet has been moved along the whole path to mix the two drops. 3.3. Results on closed configuration The functionality of the closed configuration has been verified dispensing, moving and splitting a drop of liquid. Fig. 12 illustrates the dispensing from the reservoir. The numbers refer to the sequential times, while the ‘on’ label refers to the voltage activation. The reservoir has been initially filled with a 1 μl droplet. The control strategy can be summarized as follows: 1. Activation of the second electrode of the reserve, the one closest to the electrode array, while the first electrode is left
Fig. 13. Sequence of frames reporting the movement of a droplet from a reservoir.
Fig. 14. Sequence of frames reporting the splitting of a droplet in the closed configuration.
1690
D. Caputo et al. / Microelectronics Journal 45 (2014) 1684–1690
floating. In this way, the drop in the reserve begins to deform and moves toward the electrode array (Fig. 12, frame 2); 2. activation of the first electrode of the array (Fig. 12, frame 3), which allows the coverage of the electrode; and 3. as soon as the drop covers a significant portion of the rectangular electrode, the second electrode of the reservoir is switched off while the first electrode of the reserve is activated (Fig. 12, frame 3 and frame 4). This pulls back part of the liquid into the reserve (Fig. 12, frame 5) and separates the rest (Fig. 12, frame 6). The movement along the electrodes has been easily implemented as illustrated in Fig. 13 where the frames are sequentially reported and the ‘on’ label refers to the voltage activation. The movement of the droplet is toward the activated electrode that, in our case, means grounded electrode. Finally, Fig. 14 reports the droplet splitting. A 0.2 μl drop has been placed on the electrode array covering three different electrodes of the bottom plate. The top plate has been connected to the high voltage; the two external electrodes of the bottom plate have been connected to ground, while the central one has been left floating. The voltage to the top plate has been gradually increased to achieve the splitting, that has occurred at 175 V.
4. Conclusions In this paper we have presented and characterized the performances of both open and closed EWOD systems based on the use of a PDMS material acting as both insulation and hydrophobic layer. The material characterization in a sessile drop experiment shows that the saturation of the contact angle occurs for negative voltage (around 40 V), differently from literature results where the contact angle saturation occurs for positive voltage. As a consequence of this behavior the activated electrode is the grounded one. The EWOD systems comprise an electronic board, equipped with a microcontroller, a series of switches, a high voltage power supply (up to 200 V) and a circuit for the detection of the drop between and over the electrodes. The drop detection has been implemented measuring the discharging time of the EWOD electrode through a pin of the microcontroller. The discharge time, in open configuration, with and without the drop has been found to be around 50 and 25 μs, respectively.
The movement, splitting and dispensing is controlled by the sequence of voltage levels (high voltage, ground or floating) applied to the adjacent electrodes and by the activation time of the voltage applied to each electrode. We have implemented the voltage control strategies for successfully achieving all the microfluidic operations. We have found that a reliable movement occurs at 150 V, the splitting at 175 V and the dispensing at 200 V. References [1] A. Manz, H. Becker, Microsystem Technology in Chemistry and Life Sciences, Springer Verlag, 1998. [2] S.C. Jakeway, A.J. de Mello, E.L. Russell, Miniaturized total analysis systems for biological analysis, Fresenius J. Anal. Chem. 366 (6–7) (2000) 525–539. [3] D. Caputo, G. de Cesare, C. Manetti, A. Nascetti, R. Scipinotti, Smart thin layer chromatography plate, Lab Chip 7 (8) (2007) 978–980. [4] J.W. Hong, S.R. Quake, Integrated nanoliter systems, Nat. Biotechnol. 21 (10) (2003) 1179–1183. [5] J.-H. Tsai, L. Lin, Micro-to-macro fluidic interconnectors with an integrated polymer sealant, J. Micromech. Microeng. 11 (5) (2001) 577. [6] C.-M. Ho, Y.-C. Tai, Micro-electro-mechanical-systems (mems) and fluid flows, Annu. Rev. Fluid Mech. 30 (1) (1998) 579–612. [7] O.C. Jeong, S.S. Yang, Fabrication and test of a thermopneumatic micropump with a corrugated p þ diaphragm, Sens. Actuators A: Phys. 83 (1) (2000) 249–255. [8] M. Washizu, Electrostatic actuation of liquid droplets for micro-reactor applications, IEEE Trans. Ind. Appl. 34 (4) (1998) 732–737. [9] B. Shapiro, H. Moon, R.L. Garrell, C.-J. Kim, et al., Equilibrium behavior of sessile drops under surface tension, applied external fields, and material variations, J. Appl. Phys. 93 (9) (2003) 5794–5811. [10] G. Lippmann, Relations entre les phenomenes electriques et capillaires (Ph.D. dissertation), Gauthier-Villars, 1875. [11] Jean-Christophe Baret Frieder Mugele, Electrowetting: from basic to applications, J. Phys.: Condens. Matter 17 (2005) R705–R774. [12] F.E. Torres, P. Kuhn, D. De Bruyker, A.G. Bell, M.V. Wolkin, E. Peeters, J.R. Williamson, G.B. Anderson, G.P. Schmitz, M.I. Recht, et al., Enthalpy arrays, Proc. Natl. Acad. Sci. USA 101 (26) (2004) 9517–9522. [13] S.K. Cho, H. Moon, C.-J. Kim, Creating, transporting, cutting, and merging liquid droplets by electrowetting-based actuation for digital microfluidic circuits, J. Microelectromech. Syst. 12 (1) (2003) 70–80. [14] M. Pollack, A. Shenderov, R. Fair, Electrowetting-based actuation of droplets for integrated microfluidics, Lab Chip 2 (2) (2002) 96–101. [15] S.K. Cho, H. Moon, J. Fowler, and C.-J. Kim, Splitting a liquid droplet for electrowetting-based microfluidics, in: Proceedings of the 2001 ASME International Mechanical Engineering Congress and Exposition, November, 2001, pp. 11–16. [16] D. Caputo, M. Ceccarelli, G. de Cesare, A. Nascetti, and R. Scipinotti, Lab-onglass system for dna analysis using thin and thick film technologies, in: MRS Proceedings, vol. 1191, no. 1, Cambridge University Press, 2009. [17] W. Dai, Y.-P. Zhao, The nonlinear phenomena of thin poly- dimethylsiloxane (pdms) films in electrowetting, Int. J. Nonlinear Sci. Numer. Simul. 8 (4) (2007) 519–526. [18] H. Verheijen, M. Prins, Reversible electrowetting and trapping of charge: model and experiments, Langmuir 15 (20) (1999) 6616–6620. [19] H. Moon, S.K. Cho, R.L. Garrell, C.-J. Kim, et al., Low voltage electrowetting-ondielectric, J. Appl. Phys. 92 (7) (2002) 4080–4087.