2098
IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control ,
vol. 61, no. 12,
December
2014
The Breakup of Digital Microfluids on a Piezoelectric Substrate Using Surface Acoustic Waves An-Liang Zhang and Yan Zha Abstract—A new method for the breakup of a digital microfluid (a discrete droplet) is presented and a device for splitting the digital microfluid is fabricated on a 128° yx-LiNbO3 piezoelectric substrate using microelectronic technology. Together with the surface tension of the digital microfluid, the inertia of acoustic streaming caused by the sudden disappearance of the electric signal for generating the surface acoustic wave breaks up the digital microfluid. The escape angle of the daughter digital microfluids is calculated. A sound-absorption film is coated on the acoustic path to prevent the further breakup of the daughter digital microfluids. Droplet breakups are demonstrated using red dye solution digital microfluids. Results show that digital microfluids can be broken up by suddenly decreasing the power of the electrical signal from 12.3 dBm to −3.98 dBm, and the average escape angle of daughter digital microfluids is 68.5° for 4 µL of initial digital microfluid. The results also show that the escape angle is affected by the initial volume of the digital microfluid.
I. Introduction
W
ith the development of MEMS technology, several laboratories have been working toward miniaturized devices. A microfluidic analysis system is one of the typical applications of MEMS technology. Such a system can integrate analytical processes for sequential operations such as sampling, sample pre-treatment, chemical reaction, and detection in a single microfluidic device [1], [2]. It can also be coupled with a sensitive detection method [3], such as chemiluminescence, to provide high sensitivity, wide dynamic range, and low detection limit for biochemical analysis. Compared with traditional analysis, chip-based microfluidic analysis has advantages including low reagent volume, short reaction time, portability for in situ use, low cost, versatility in design, and potential for parallel operation, leading to chip-based microfluidic analysis being widely applied for DNA sequencing, protein
Manuscript received August 21, 2013; accepted August 27, 2014. This work was supported by the Science and Technology Department of Zhejiang Province and the Natural Science Foundation of Ningbo Municipality in China under award numbers 2009R50025 and 2011A610108. This work was also partly sponsored by the K. C. Wong Magna Fund of Ningbo University. The authors are with the Department of Electric Engineering, Ningbo University, Ningbo, Zhejiang, P. R. China (e-mail: zhanganliang@nbu. edu.cn). A.-L. Zhang is also with the School of Electronic Information and Electrical Engineering, Changzhou Institute of Technology, Changzhou, Jiangsu, P. R. China. DOI http://dx.doi.org/10.1109/TUFFC.2013.005997
0885–3010
analysis, single-cell analysis, drug screening, cell sorting, and food safety [4]–[6]. Most current microfluidic systems involve the continuous flow of liquid and require networks of microchannels and valves combined with pressure sources, resulting in complicated fabrication and component integration, reduced portability, and dead volumes of samples [7]. Alternatively, droplet-based microfluidic systems can solve these problems in a microfluidic system working with a continuous flow of fluid, and can independently control each digital microfluid (a discrete droplet) to implement microfluidic operations such as transportation, mixture, fusion, separation, and analysis [8]–[10]. Thus, digital microfluidic systems have obtained more widespread attention in recent years. Trace-level determination is frequently required for analysis of biological samples. The operation of the breakup of a digital microfluid is also usually indispensable. Several methods for the breakup of digital microfluids have been reported. Generally speaking, they are mainly used to reduce the volume of digital microfluids [11], control the concentration of chemicals inside the digital microfluids [12] and produce arrays of digital microfluids for high throughput [13]. Cubaud et al. [14] reported a viscous stratification method for breaking up a digital microfluid on glass and silicon substrates. The advantage of the method is that it is simple to set up; however, it is only applicable to flowing fluids with high viscosity. The breakup of digital microfluids can also be implemented by shear force on fluid in a channel. A shear strain on a digital microfluid will arise when the digital microfluid in immiscible fluid in a channel flows toward a bifurcating junction. If the shear force is greater than the interface tension of a digital microfluid, the breakup of the digital microfluid will occur [15]. By changing only the inlet width of the bifurcating junction, the volume of daughter digital microfluids can be controlled and concentration differences in the daughter digital microfluids can be produced. Link et al. [16] reported a new method to break up a digital microfluid in a channel. Fluid is driven in a channel with isolated obstacles by a pressure. The shear force of the digital microfluid is increased when the digital microfluid passes through isolated obstacles in the channel, and the digital microfluid is broken up when the shear force is greater than the interface tension of the digital microfluid. The size of daughter digital microfluids
© 2014 IEEE
zhang and zha: the breakup of digital microfluids on a piezoelectric substrate using SAWs
can be controlled by changing the position of the obstacle from the channel center. These breakup techniques for digital microfluids have their advantages, and have been applied in microfluidic biochemical analysis systems. However, the digital microfluids are all suspended flowing fluids in channels, which are not suitable for an open microfluidic system. A SAW is an elastic wave which can be transported along the surface of a piezoelectric substrate. When an electrical signal with an appropriate frequency is applied to an interdigital transducer (IDT) on a piezoelectric substrate, SAWs can be generated. The amplitude of the SAW is determined by the power of the electrical signal. The frequency of electric signal should be near the sound synchronization frequency of the IDT, where SAWs can be efficiently excited. When a SAW encounters a digital microfluid on the piezoelectric substrate, the major part of the incident wave energy radiates into the digital microfluid by Rayleigh angle θR [17]:
θ R = arcsin
VW , (1) VR
where VW is the velocity of sound propagation in the fluid and VR is that on the piezoelectric substrate. The Rayleigh angle θR is about 22° when a SAW is radiated into an aqueous digital microfluid. The properties of SAWs radiating into a microfluid have been used in microfluidic systems. Especially, this is frequently used in open microfluidic systems, in which microfluidic operations, including digital microfluid transportation [18], ultra-fast homogenous mixing [19], particle separation, cell concentration, and protein purification [20]–[23] have been implemented. Recently, SAWs have been used for breaking up digital microfluids [24]. However, an electric signal of more than 60 V is usually used to excite SAWs with large enough amplitude, which will probably crack some piezoelectric substrates, such as 128° yx-LiNbO3 substrate, in the piezoelectric microfluidic system.
Fig. 1. Acoustic wave radiation into a digital microfluid.
2099
In this paper, an innovative microfluidic device for breaking up digital microfluids was designed and made using a 128° yx-LiNbO3 piezoelectric substrate. Only 12.3 dBm of electric signal power was used to successfully implement the breakup of digital microfluids. The soundabsorption film could efficiently prevent the further breakup of daughter digital microfluids. In addition, the escape angle of the daughter digital microfluids was calculated. The influence of electric signal power, turn-off time, and the initial volume of digital microfluids on the breakup of the microfluids were studied. II. Experiment A. Principles of the Breakup of Digital Microfluids Using SAW As soon as a SAW encounters a digital microfluid, the major part of the SAW’s energy is radiated into the microfluid. The principle of a SAW radiating into the digital microfluid is diagrammed in Fig. 1. A SAW is radiated into the digital microfluid, leading to internal acoustic streaming in the digital microfluid, which causes SAW streaming force per unit volume given by [25]
Fs = −ρ(1 + αl2) 3/2A 2ω 2K i exp 2(K ix + αlK iz ), (2)
where ρ is the density of the microfluid, A is the SAW amplitude, ω is angular frequency, ki is the imaginary part of the wave number of the leaky SAW (kl), αl = jα with α = 1 − (V R/V W)2. The wave number of leaky SAWs can be obtained by the angular frequency divided by leaky SAW velocity (VL). The direction of the SAW streaming force is the same angle as the Rayleigh angle θR. When the SAW amplitude is not large enough, the digital microfluid rotates around the center of the microfluid in its original position on the piezoelectric substrate. At this moment, no fluid escapes from the initial digital mi-
2100
IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control ,
vol. 61, no. 12,
December
2014
transducer to the digital microfluid, H is the height of the microfluid on the piezoelectric substrate, and R is the radius of the digital microfluid. The shape of the microfluid on the piezoelectric substrate is almost spherical, which can be verified by experiments in the next section. The microfluid can be expressed as (x − x 0)2 + y 2 + z 2 = R 2. (3)
According to Fig. 2(e), the radius r of a circular section at the height z can be calculated by
r =
R 2 − (H − R − z )2, z ≤ H − R (4)
r =
R 2 − (R + z − H )2, z > H − R. (5)
or
The circle equation at the height z can be expressed as (x − x 0)2 + y 2 = r 2. (6)
Fig. 2. The diagram of forces on a digital microfluid.
crofluid. However, when the SAW is continuously radiated into the digital microfluid for about one second and then suddenly disappears, part of the microfluid escapes from the initial digital microfluid because of inertial forces. The main reason is that the surface tension seeks to restore the equilibrium shape of the microfluid after the acoustic signal is turned off. The digital microfluid oscillations arise and at this stage the inertia breaks up the digital microfluid. There are several forces on the digital microfluids: one is the surface tension of the piezoelectric substrate to the digital microfluid, one is the gravitational force, and the other is the SAW streaming force. A diagram of forces on the digital microfluid is shown in Fig. 2. In Fig. 2, Fig. 2(a) shows the initial digital microfluid on a piezoelectric substrate, and the SAW is not excited. The parameter γ1 is the solid–gas interfacial energy, γ2 is the liquid-gas interfacial energy, and γ3 is the solid–liquid interfacial energy. These parameters represent the adhesion strength of the substrate to the digital microfluid, which is called the adhesion work. Fg is the gravitational force acting on the digital microfluid. The adhesion work and the gravitational force stabilize the microfluid on the substrate. Fig. 2(b) shows the forces when the SAW streaming force is also acting on the microfluid as a result of SAW radiation. Fig. 2(c) shows the forces when the SAW is suddenly removed. At this time, the upper part of the microfluid is mainly acted on by the inertial SAW streaming force, whereas the lower part of the microfluid is attracted by adhesion work and its gravitational force, leading to part of the microfluid escaping from the initial digital microfluid, as shown in Fig. 2(d). Fig. 2(e) is a diagram for analyzing the forces on the digital microfluid. In Fig. 2(e), x0 is the center distance from the interdigital
Then, according (2)–(6), we can derive the acoustic radiation force on the upper part of the microfluid, which is higher than the height of z0. N
F =
H x 0 + r 2 −y i2
∑∫
∫
r ⋅ Fs ⋅ dx dz , (7) N 2
i =1 z x − r 2 −y 0 0 i
where r is divided into N equal segments, and yi = r (i − 1)/N. Using (7), we can calculate the acoustic radiation force on the upper part of the microfluid. The surface tension of the digital microfluid on the piezoelectric substrate can be calculated using 2π
Fsurface =
∫ γ (1 + cos φ) ⋅ r0 dt , (8) 2
0
where ϕ is the contact angle of the microfluid to the Teflon-modified piezoelectric substrate. The value of ϕ can be measured by a contact angle measuring apparatus (Digidrop, GBX Etude des Technologies Avancees Co. Ltd., Bourg de Peage, France), which was measured to be 87.5° in our experiment. B. Experimental Setup The experimental setup for the breakup of digital microfluids is shown in Fig. 3. In Fig. 3, an interdigital transducer and a reflector are fabricated on the 128° yx-LiNbO3 substrate using microelectronic technology. The aperture and period of the interdigital transducer are 4.32 mm and 144 µm, respectively. The center frequency of the interdigital transducer is 27.5 MHz. The number of finger pairs in the interdigital transducer is 35. The area of the acoustic path is coated
zhang and zha: the breakup of digital microfluids on a piezoelectric substrate using SAWs
Fig. 3. The experimental setup for the breakup of digital microfluids.
with Teflon AF 1600 (DuPont Corp., Wilmington, DE). Poly(dimethylsilicone) (PDMS) film with 4.8 mm length in the transportation direction is used for attenuating the SAW to avoid the further breakup of daughter digital microfluids. Filter paper, which will be placed on the surface of the piezoelectric substrate at the position AA′, is used to measure the escape angle of the ejected daughter digital microfluids. A signal generator (SP1461, Nanjing Sample Instrument Technologies Co. Ltd., Nanjing, China) supplies a radio-frequency sine signal. The radio-frequency sine signal is amplified by a power amplifier (TSA002A, Tianjing Zeland Electronic Technology Co. Ltd., Tianjing, China) with a gain of 48 dB and maximum unsaturated output power of 30 W. A highly sensitive charge-coupled imager (DCE-2, Ningbo Yongxin Optics Co. Ltd., Ningbo, China) is used to monitor the breakup of the digital microfluid. MDVNT software (Ningbo Yongxin Optics Co. Ltd.) is used for camera control and image processing. To observe the breakup of a digital microfluid, red dye solution digital microfluid was broken up for demonstration. Red dye solution digital microfluid was first pipetted onto the piezoelectric substrate using a micro-syringe. A radio-frequency signal was applied to the interdigital transducer and then suddenly decreased to lower amplitude. The breakup of daughter digital microfluids at the right side of the sound-absorption film was also studied. The escape angle of the daughter digital microfluids was calculated by measuring the height of the daughter digital microfluid on the filter paper at the position of AA′. III. Results and Discussion A 4-µL red dye solution digital microfluid was broken up on the 128° yx-LiNbO3 piezoelectric substrate. After a digital microfluid was pipetted onto the piezoelectric substrate, a 12.3-dBm radio-frequency signal was applied to the interdigital transducer. By suddenly decreasing the radio-frequency signal power to −3.98 dBm, the digital microfluid was broken up as a result of inertial force. Video snapshots for the breakup of the digital microfluid are shown in Fig. 4.
2101
In Fig. 4, Fig. 4(a) shows the state of a 4-µL red dye solution microfluid on the piezoelectric substrate. According to Fig. 4(a), the shape of the microfluid is nearly a spherical shape. Fig. 4(b) shows the escape of a daughter digital microfluid from the initial digital microfluid when the power of radio-frequency signal is suddenly reduced from 12.3 dBm to −3.98 dBm. Fig. 4(c) shows the state of the microfluid after the initial breakup. The states of the microfluid after the second to fourth breakup times are shown in Figs. 4(d)–4(f). In Fig. 4, the sound-absorption film with powdered carbon was used for observation. If necessary, the daughter microfluids can be further broken up by actuating them to the left of the sound absorption by the interdigital transducer array. However, if the daughter microfluid does not need to be further broken up, the sound-absorption film is indispensable. Attenuation of the SAW causes a decreased SAW streaming force which is not enough to break the microfluid up, as shown in Fig. 5. In Fig. 5, Fig. 5(a) shows the state of a 4-µL red dye solution microfluid on the piezoelectric substrate at the right side of the PDMS sound-absorption film. Fig. 5(b) shows the state of the microfluid after five sudden changes of electric signal from 12.3 dBm to −3.98 dBm. Compared with Fig. 5(a), no breakup occurs during this period. Increasing the power of the electric signal to 24.3 dBm, the microfluid was actuated by a step in which the electric signal power is suddenly decreased to 8.5 dBm, as shown Fig. 5(c). Figs. 5(d)–5(f) show another three steps of transportation of the microfluid by the same power change of electric signal. According to Fig. 5, the PDMS sound-absorption film can partly attenuate the SAW and prevent the daughter digital microfluid from being further broken up. Digital microfluids with particles such as proteins, DNA, or drugs will also be broken up using the same method, but only if the inertial SAW streaming force is greater than that of the sum of intermolecular forces and the gravitational force of the digital microfluid. After the digital microfluid has been broken up, the daughter microfluid with small volume can be used for further microfluidic operations for microfluidic biochemical analysis on the piezoelectric substrate. The acoustic radiation force on the 4-µL red dye solution microfluid can be calculated. The peak-to-peak values of the electric voltage are 3.661 V and 0.548 V, respectively, when the electric signal power is 12.3 dBm and −3.98 dBm. Because the amplitude of the SAW is proportional to the electric voltages applied to the interdigital transducer [25], [26], the amplitude of the SAW can be obtained. The radius of the 4-µL red dye solution microfluid can be measured by the MDVNT software. The height H of the 4-µL digital microfluid on the piezoelectric substrate can be calculated using the relation of the microfluidic volume to its radius, which will be expressed in following text. Taking VL = (3931 + j67.7) m/s and VW = 1500 m/s [25], the acoustic radiation force on the 4-µL red dye solution microfluid can be obtained by (2)–(7). The value is decreased from 0.031 mN to 0.00068 mN when
2102
IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control ,
vol. 61, no. 12,
December
2014
Fig. 4. The video snapshots for the breakup of a 4-µL red dye solution digital microfluid.
the electrical signal power is decreased from 12.3 dBm to −3.98 dBm. The surface tension of the 4-µL red dye solution to the piezoelectric substrate is 0.495 mN and its gravitational force is 0.039 mN. The sum of the surface tension and the gravitational force of the digital microfluid is larger than the endured acoustic radiation force when the continuous electric signal with 12.3 dBm power is used to excite SAW. Thus, the microfluid is stable on the piezoelectric substrate, although a centripetence occurs inside the digital microfluid. However, when the electric signal is suddenly decreased to −3.98 dBm, the lower part of the digital microfluid is stuck on the substrate because the sum of surface tension and gravitational force is far larger than 0.0006 mN. At this time, the upper part of the digital microfluid endures not only the inertia of the acoustic radiation force, but also intermolecular forces
and part of the gravitational force of the digital microfluid. The intermolecular forces between the upper part and lower part of the microfluid is decreased along with their decreasing interface area resulting from surface tension seeking to restore the equilibrium shape of the lower part of digital microfluid. As soon as the sum of intermolecular forces and the gravitational force of the upper part of the microfluid is less than the inertial force, the breakup of the microfluid occurs. The escape angle is calculated by measuring the height at which the daughter digital microfluids hit a filter paper which is vertically placed on the piezoelectric substrate at the position of AA′ as shown in Fig. 3. A 4-µL red dye solution microfluid is pipetted on the piezoelectric substrate at the left side of the sound-absorption film. The microfluid is then broken up by suddenly decreasing the power
Fig. 5. The PDMS sound-absorption film prevents the daughter digital microfluid from being further broken up.
zhang and zha: the breakup of digital microfluids on a piezoelectric substrate using SAWs
2103
Fig. 7. The diagram for calculating the escape angle of the daughter digital microfluids.
where V is the volume of the digital microfluid, and R is the radius of the digital microfluid. Because of the hydrophobic surface of the piezoelectric substrate, the value of H is greater than the radius of the digital microfluid and H2 can be calculated by
H 2 = H 1 − H + R, (10)
then, the escape angle β can be calculated:
Fig. 6. Position of daughter microfluids hitting the filter paper.
of electric signal from 12.3 dBm to −3.98 dBm. Fig. 6 shows a filter paper hit by escaped daughter digital micro fluids after three breakups. In Fig. 6, the words one, two, and three indicate the breakups of the 4-µL red dye solution microfluid three times in succession. The trajectory of escaped daughter digital microfluid is a parabola; however, the horizontal distance from the escaped daughter digital microfluid to the initial microfluid D is far longer than that from the initial microfluid to the filter paper L. The starting trajectory of an escaped daughter digital microfluid is approximately linear. A diagram for calculating the escape angle of daughter digital microfluids is shown as Fig. 7. From the height at which the daughter microfluid hits the filter paper and the distance from the initial digital microfluid to the filter paper, the escape angle β can be calculated. For simplicity, the center of the microfluid can be regarded as the starting point during calculation of the escape angle. The height of the daughter microfluid on the filter paper H1 can be measured. To calculate the H2, the value of H should be obtained. The volume of a digital microfluid on the piezoelectric substrate can be calculated using the following formula:
V =
π ⋅ H 2(3R − H ), (9) 3
tan β =
H2 . (11) L
After the breakup of the 4-µL red dye solution microfluid, H1, L, and R can be measured from Figs. 4 and 6; they are 17.5 mm, 6.7 mm, and 1.0 mm, respectively. According to (9), (10), and (11), the escape angle β is 68.8°. The breakup experiments of the 4-µL red dye solution microfluid were repeated eight times with the same experimental conditions. Their escape angles were calculated as shown in Table I. In Table I, the average value of the escape angles is 68.5°, and that of θ is 21.5°, which is near the Rayleigh angle θR. With respect to the mean value, the maximum error of the escape angles is 3.6°. The difference between the highest and lowest value of the escape angle is 6.4°. The error would arise mainly from several factors: 1) The slope angle of the filter paper at the position AA′ to the surface of the piezoelectric substrate will lead to a distance error from the filter paper to the digital microfluid to be broken up. Although we try to place the filter paper vertically on the surface of the piezoelectric substrate at the position AA′ with the help of a PDMS block, there would still be a slight error. 2) The gain of the power amplifier fluctuates during the experiment, which will affect the initial power of the electrical signal. 3) There exists a measurement error in measuring the distance from the filter paper to the digital microfluid and the height at which the daughter digital microfluid hits the filter paper. To decrease the error, the experiment is repeated eight times, and the average value is used in this work. Different volumes of the initial red dye solution microfluid were broken up using the same power variation of
2104
IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control ,
TABLE I. Escape Angles of the Breakup of a 4-µL Red Dye Solution Microfluid. L (mm) 6.7 6.6 6.6 6.3 6.5 6.6 6.8 6.7
H1 (mm)
β (°)
θ = 90° − β (°)
16.0 21.0 17.8 16.8 15.0 15.6 20.0 17.8
66.5 72.1 69.0 68.8 65.7 66.3 70.7 68.8
23.5 17.9 21.0 21.2 24.3 23.7 19.3 21.2
Volume of initial digital microfluid (µL) 1 2 4 6 8
Volume of escaped daughter microfluid (µL)
β (°)
θ (°)
0.684 1.082 1.99 2.39 3.28
63.4 65.5 68.5 63.1 62.9
26.6 24.5 21.5 26.9 27.1
2014
Volume of escaped daughter microfluid (µL)
β (°)
θ (°)
1.99 2.03 1.89 2.05
68.5 68.8 70.5 72.7
21.5 21.2 19.5 17.3
12.3 16.0 17.7 19.3
TABLE II. The Parameters of the Digital Microfluid Breakup at Different Initial Volume With the Same Power Variation of the Electrical Signal.
December
TABLE III. Escape Angles and Escaped Daughter Microfluid Volume of a 4-µL Red Dye Solution Microfluid at Different Initial Electrical Signal Powers. Initial electrical signal power (dBm)
the electrical signal. The parameters of the breakup of the microfluids are listed in Table II. According to Table II, we can deduce that the volume of the escaped daughter microfluid increases with the initial volume of the digital microfluid. The escape angle is affected by initial volume of the digital microfluid. When the volume of initial digital microfluid is smaller—for example, 1 and 2 µL—the escape angle is also smaller because of the smaller SAW streaming force. However, when the volume of initial digital microfluids is greater— for example, 6 and 8 µL—the escape angle is gradually decreased. The reason for this is that the gravitational force of the escaped daughter digital microfluid cannot be ignored and affects its movement after breakup. The phenomenon is also verified by Fig. 6. To observe the influence of initial electrical signal power on the parameters of the microfluidic breakup, a 4-µL red dye solution microfluid was broken up at different initial electrical signal powers. Table III shows the escaped daughter digital microfluid volume and escape angle at four different initial electric signal powers. Table III shows that the influence of initial electric signal power on the volume of the escaped daughter microfluid is not obvious. However, the escape angle is slightly affected by the initial electrical signal power. The reason is that the surface tension is greater than the acoustic radiation force, which determines the volume of the daughter microfluid. The acoustic radiation force is the main factor to affect the movement of the escaped daughter microfluid, and the escape angles of microfluids were slightly increased with initial electrical signal power. The influences of turn-off time on the escape angles and the volume of the escaped daughter microfluids were also studied, as shown in Table IV.
vol. 61, no. 12,
In Table IV, we can observe that the escape angles and the volume of the escaped daughter microfluids are not obviously affected by the turn-off time when it is greater than 0.533 s, although the digital microfluid cannot be broken up at a very short turn-off time. The reason is that the inertial force is determined by SAW streaming force, which is hardly affected by the duration at the same electrical signal power after the inertial force has been formed. We define transfer efficiency as the ratio of the volume of the escaped daughter microfluid to the initial volume of the digital microfluid. According the Tables II, III, and IV, we can deduce that the transfer efficiency is affected by the initial volume of the digital microfluid at the same power variation of the electrical signal. However, the influence of initial electric signal power and turn-off time on the transfer efficiency is not obvious. The reasons have already been mentioned. Although flow-based microfluidics have obvious advantages in terms of speed and throughput, a piezoelectric microfluidic device based on a digital microfluid does not require networks of microchannels and valves, and it can independently control each digital microfluid to implement microfluidic operations such as droplet transportation, mixture, and separation. Thus, the presented method is especially applied to single-droplet operation or analysis. For example, the presented method can be easily applied for diluting single droplet sample or reagent on a piezoelectric substrate. Because the electrochemical electrodes can be integrated into the piezoelectric device without other processing steps, the presented method can also be applied for electrochemical micro-analysis based on a digital microfluid. However, high-throughput microTABLE IV. Escape Angles and Escaped Daughter Microfluid Volume of a 4-µL Red Dye Solution Microfluid at Different Turn-Off Times. Turn off time (s) 0.533 1.333 1.6 2.067 2.6 3.8 5.467 6.133
Volume of escaped daughter microfluid (µL)
β (°)
θ (°)
1.98 2.02 1.87 2.01 2.11 1.89 1.83 1.94
68.7 68.8 68.5 68.7 69.1 68.7 68.9 69.1
21.3 21.2 21.5 21.3 20.9 21.3 21.1 20.9
The initial electrical signal power is 16 dBm.
zhang and zha: the breakup of digital microfluids on a piezoelectric substrate using SAWs
fluidic analysis, including cell analysis or high-throughput screening for drug detection, should avoid using the presented technique. IV. Conclusions A new method for the breakup of digital microfluids was presented and implemented on a 128° yx-LiNbO3 substrate using microelectronic technology. PDMS sound-absorption film was used to prevent the daughter digital microfluids from being further broken up. The escape angle was calculated by measuring the height of a daughter digital microfluid hit on a filter paper. Droplet breakups were demonstrated using red dye solution digital microfluids. According to the work, several conclusions can be drawn: 1) The breakup of digital microfluids can be implemented by suddenly decreasing the power of an electrical signal with appropriate amplitude; 2) sound-absorption film can effectively avoid further breakup of the daughter digital microfluids; and 3) the initial volume of digital microfluids can affect the escape angle. The presented breakup method for digital microfluids provides a new sample preparation technique which is helpful for microfluidic biochemical analysis on a piezoelectric microfluidic device. References [1] Z. L. Xiao, M. L. Niu, and B. Zhang, “Droplet microfluidics based microseparation systems, ” J. Sep. Sci., vol. 35, no. 10–11, pp. 1284– 1293, 2012. [2] A. Arora, G. Simone, G. Salieb-Beugelaar, J. Kim, and A. Manz, “Latest developments in micro total analysis systems,” Anal. Chem., vol. 82, no. 12, pp. 4830–4847, 2010. [3] M. Kamruzzaman, A. Alam, K. M. Kim, S. H. Lee, Y. H. Kim, G. M. Kim, and T. D. Dang, “Microfluidic chip based chemiluminescence detection of L-phenylalanine in pharmaceutical and soft drinks,” Food Chem., vol. 135, no. 1, pp. 57–62, 2012. [4] E. Jang, S. Kim, and W. G. Koh, “Microfluidic bioassay system based on microarrays of hydrogel sensing elements entrapping quantum dot–enzyme conjugates,” Biosens. Bioelectron., vol. 31, no. 1, pp. 529–536, 2012. [5] J. Autebert, B. Coudert, F. C. Bidard, J. Y. Pierga, S. Descroix, L. Malaquin, and J. L. Viovy, “Microfluidic: An innovative tool for efficient cell sorting,” Methods, vol. 57, no. 3, pp. 297–307, 2012. [6] G. Grenci, G. Birarda, E. Mitri, L. Businaro, S. Pacor, L. Vaccari, and M. Tormen, “Optimization of microfluidic systems for IRMS long term measurement of living cells,” Microelectron. Eng., vol. 98, no. 1, pp. 698–702, 2012. [7] K. N. Han, C. A. Li, M. P. N. Bui, X. H. Pham, B. S. Kim, Y. H. Choa, E. K. Lee, and G. H. Seong, “On-chip electrochemical detection of bio/chemical molecule by nanostructures fabricated in a microfluidic channel,” Sens. Actuators B, vol. 177, no. 2, pp. 472–477, 2013. [8] P. Paik, V. K. Pamula, and R. B. Fair, “Rapid droplet mixers for digital microfluidic systems,” Lab Chip, vol. 3, no. 4, pp. 253–259, 2003. [9] Y. Ai and B. L. Marrone, “Droplet translocation by focused surface acoustic waves,” Microfluid. Nanofluid., vol. 13, no. 5, pp. 715–722, 2012. [10] Y. Okabe, Y. L. Chen, R. Purohit, R. M. Corn, and A. P. Lee, “Piezoelectrically driven vertical cavity acoustic transducers for the convective transport and rapid detection of DNA and protein binding to DNA microarrays with SPR imaging—A parametric study,” Biosens. Bioelectron., vol. 35, no. 1, pp. 37–43, 2012.
2105
[11] U. Brosa and S. Grossman, “Droplet fission: A performance of surface tension,” Phys. Lett. B, vol. 126, no. 6, pp. 425–427, 1983. [12] Y. C. Tan, J. S. Fisher, A. I. Lee, V. Cristini, and A. P. Lee, “Design of microfluidic channel geometries for the control of droplet volume, chemical concentration, and sorting,” Lab Chip, vol. 4, no. 4, pp. 292–298, 2004. [13] D. N. Adamson, D. Mustafi, J. X. J. Zhang, B. Zheng, and R. F. Ismagilov, “Production of arrays of chemically distinct nanolitre plugs via repeated splitting in microfluidic devices,” Lab Chip, vol. 6, no. 9, pp. 1178–1186, 2006. [14] T. Cubaud, B. M. Jose, S. Darvishi, and R. Sun, “Droplet breakup and viscosity-stratified flows in microchannels,” Int. J. Multiph. Flow, vol. 39, no. 1, pp. 29–36, 2012. [15] H. Song, J. D. Tice, and R. F. Ismagilov, “A microfluidic system for controlling reaction networks in time,” Angew. Chem. Int. Ed. Engl., vol. 42, no. 7, pp. 768–772, 2003. [16] D. R. Link, S. L. Anna, D. A. Weitz, and H. A. Stone, “Geometrically mediated breakup of drops in microfluidic devices,” Phys. Rev. Lett., vol. 92, no. 5, art. no. 054503, 2004. [17] T. Uchida, T. Suzuki, and S. Shiokawa, “Investigation of acoustic streaming excited by surface acoustic waves,” in IEEE Ultrasonics Symp., 1995, pp.1081–1084. [18] A. Renaudin, P. Tabourier, V. Zhang, J. C. Camart, and C. Druon, “SAW nanopump for handling droplets in view of biological applications,” Sens. Actuators B, vol. 113, no. 1, pp. 389–397, 2006. [19] Q. Zeng, F. Guo, L. Yao, H. W. Zhu, L. Zheng, Z. X. Guo, W. Liu, Y. Chen, S. S. Guo, and X. Z. Zhao, “Milliseconds mixing in microfluidic channel using focused surface acoustic wave,” Sens. Actuators B, vol. 160, no. 1, pp. 1552–1556, 2011. [20] S. Oberti, A. Neild, R. Quach, and J. Dual, “The use of acoustic radiation forces to position particles within fluid droplets,” Ultrasonics, vol. 49, no. 1, pp. 47–52, 2009. [21] C. J. Myeong and G. Rasim, “Active density-based separation using standing surface acoustic waves,” Sens. Actuators A, vol. 187, no. 1, pp. 22–28, 2012. [22] H. Y. Li, J. R. Friend, and L. Y. Yeo, “Surface acoustic wave concentration of particle and bioparticle suspensions,” Biomed. Microdevices, vol. 9, no. 5, pp. 647–656, 2007. [23] R. Hahn, “Methods for characterization of biochromatography media,” J. Sep. Sci., vol. 35, no. 22, pp. 3001–3032, 2012. [24] C. Y. Lee, H. Y. Yu, W. Pang, and E. S. Kim, “Droplet-based microreactions with oil encapsulation,” J. Microelectromech. Syst., vol. 17, no. 1, pp. 147–156, 2008. [25] S. Shiokawa, Y. Matsui and T. Ueda, “Liquid streaming and droplet formation caused by leaky Rayleigh wave,” in IEEE Ultrasonics Symp., 1989, pp. 643–646. [26] J. R. Doherty, G. E. Trahey, K. R. Nightingale, and M. L. Palmeri, “Acoustic radiation force elasticity imaging in diagnostic ultrasound,” IEEE Trans. Ultrason. Ferroelectr. Freq. Control, vol. 60, no. 4, pp. 685–701, 2013. An-Liang Zhang received the Ph.D. degree in microelectrics from Zhejiang University, Zhejiang, China, in 2004. He joined the Key Laboratory of Analytical Chemistry for Life Science, School of Chemistry and Chemical Engineering, Nanjing University, in 2008 as a postdoctoral fellow. He is an associate professor in the Department of Electrical Engineering, Ningbo University. In June 2014, he joined the Changzhou Institute of Technology. His scientific interests include surface acoustic waves, lab on a piezoelectric substrate, and SAW sensors.
Yan Zha received her B.S. degree in electrical engineering in 2011 from Anhui Engineering University, Anhui, China. She is now a master’s degree student.
IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control ,
vol. 61, no. 12,
December
2014
The Breakup of Digital Microfluids on a Piezoelectric Substrate Using Surface Acoustic Waves An-Liang Zhang and Yan Zha Abstract—A new method for the breakup of a digital microfluid (a discrete droplet) is presented and a device for splitting the digital microfluid is fabricated on a 128° yx-LiNbO3 piezoelectric substrate using microelectronic technology. Together with the surface tension of the digital microfluid, the inertia of acoustic streaming caused by the sudden disappearance of the electric signal for generating the surface acoustic wave breaks up the digital microfluid. The escape angle of the daughter digital microfluids is calculated. A sound-absorption film is coated on the acoustic path to prevent the further breakup of the daughter digital microfluids. Droplet breakups are demonstrated using red dye solution digital microfluids. Results show that digital microfluids can be broken up by suddenly decreasing the power of the electrical signal from 12.3 dBm to −3.98 dBm, and the average escape angle of daughter digital microfluids is 68.5° for 4 µL of initial digital microfluid. The results also show that the escape angle is affected by the initial volume of the digital microfluid.
I. Introduction
W
ith the development of MEMS technology, several laboratories have been working toward miniaturized devices. A microfluidic analysis system is one of the typical applications of MEMS technology. Such a system can integrate analytical processes for sequential operations such as sampling, sample pre-treatment, chemical reaction, and detection in a single microfluidic device [1], [2]. It can also be coupled with a sensitive detection method [3], such as chemiluminescence, to provide high sensitivity, wide dynamic range, and low detection limit for biochemical analysis. Compared with traditional analysis, chip-based microfluidic analysis has advantages including low reagent volume, short reaction time, portability for in situ use, low cost, versatility in design, and potential for parallel operation, leading to chip-based microfluidic analysis being widely applied for DNA sequencing, protein
Manuscript received August 21, 2013; accepted August 27, 2014. This work was supported by the Science and Technology Department of Zhejiang Province and the Natural Science Foundation of Ningbo Municipality in China under award numbers 2009R50025 and 2011A610108. This work was also partly sponsored by the K. C. Wong Magna Fund of Ningbo University. The authors are with the Department of Electric Engineering, Ningbo University, Ningbo, Zhejiang, P. R. China (e-mail: zhanganliang@nbu. edu.cn). A.-L. Zhang is also with the School of Electronic Information and Electrical Engineering, Changzhou Institute of Technology, Changzhou, Jiangsu, P. R. China. DOI http://dx.doi.org/10.1109/TUFFC.2013.005997
0885–3010
analysis, single-cell analysis, drug screening, cell sorting, and food safety [4]–[6]. Most current microfluidic systems involve the continuous flow of liquid and require networks of microchannels and valves combined with pressure sources, resulting in complicated fabrication and component integration, reduced portability, and dead volumes of samples [7]. Alternatively, droplet-based microfluidic systems can solve these problems in a microfluidic system working with a continuous flow of fluid, and can independently control each digital microfluid (a discrete droplet) to implement microfluidic operations such as transportation, mixture, fusion, separation, and analysis [8]–[10]. Thus, digital microfluidic systems have obtained more widespread attention in recent years. Trace-level determination is frequently required for analysis of biological samples. The operation of the breakup of a digital microfluid is also usually indispensable. Several methods for the breakup of digital microfluids have been reported. Generally speaking, they are mainly used to reduce the volume of digital microfluids [11], control the concentration of chemicals inside the digital microfluids [12] and produce arrays of digital microfluids for high throughput [13]. Cubaud et al. [14] reported a viscous stratification method for breaking up a digital microfluid on glass and silicon substrates. The advantage of the method is that it is simple to set up; however, it is only applicable to flowing fluids with high viscosity. The breakup of digital microfluids can also be implemented by shear force on fluid in a channel. A shear strain on a digital microfluid will arise when the digital microfluid in immiscible fluid in a channel flows toward a bifurcating junction. If the shear force is greater than the interface tension of a digital microfluid, the breakup of the digital microfluid will occur [15]. By changing only the inlet width of the bifurcating junction, the volume of daughter digital microfluids can be controlled and concentration differences in the daughter digital microfluids can be produced. Link et al. [16] reported a new method to break up a digital microfluid in a channel. Fluid is driven in a channel with isolated obstacles by a pressure. The shear force of the digital microfluid is increased when the digital microfluid passes through isolated obstacles in the channel, and the digital microfluid is broken up when the shear force is greater than the interface tension of the digital microfluid. The size of daughter digital microfluids
© 2014 IEEE
zhang and zha: the breakup of digital microfluids on a piezoelectric substrate using SAWs
can be controlled by changing the position of the obstacle from the channel center. These breakup techniques for digital microfluids have their advantages, and have been applied in microfluidic biochemical analysis systems. However, the digital microfluids are all suspended flowing fluids in channels, which are not suitable for an open microfluidic system. A SAW is an elastic wave which can be transported along the surface of a piezoelectric substrate. When an electrical signal with an appropriate frequency is applied to an interdigital transducer (IDT) on a piezoelectric substrate, SAWs can be generated. The amplitude of the SAW is determined by the power of the electrical signal. The frequency of electric signal should be near the sound synchronization frequency of the IDT, where SAWs can be efficiently excited. When a SAW encounters a digital microfluid on the piezoelectric substrate, the major part of the incident wave energy radiates into the digital microfluid by Rayleigh angle θR [17]:
θ R = arcsin
VW , (1) VR
where VW is the velocity of sound propagation in the fluid and VR is that on the piezoelectric substrate. The Rayleigh angle θR is about 22° when a SAW is radiated into an aqueous digital microfluid. The properties of SAWs radiating into a microfluid have been used in microfluidic systems. Especially, this is frequently used in open microfluidic systems, in which microfluidic operations, including digital microfluid transportation [18], ultra-fast homogenous mixing [19], particle separation, cell concentration, and protein purification [20]–[23] have been implemented. Recently, SAWs have been used for breaking up digital microfluids [24]. However, an electric signal of more than 60 V is usually used to excite SAWs with large enough amplitude, which will probably crack some piezoelectric substrates, such as 128° yx-LiNbO3 substrate, in the piezoelectric microfluidic system.
Fig. 1. Acoustic wave radiation into a digital microfluid.
2099
In this paper, an innovative microfluidic device for breaking up digital microfluids was designed and made using a 128° yx-LiNbO3 piezoelectric substrate. Only 12.3 dBm of electric signal power was used to successfully implement the breakup of digital microfluids. The soundabsorption film could efficiently prevent the further breakup of daughter digital microfluids. In addition, the escape angle of the daughter digital microfluids was calculated. The influence of electric signal power, turn-off time, and the initial volume of digital microfluids on the breakup of the microfluids were studied. II. Experiment A. Principles of the Breakup of Digital Microfluids Using SAW As soon as a SAW encounters a digital microfluid, the major part of the SAW’s energy is radiated into the microfluid. The principle of a SAW radiating into the digital microfluid is diagrammed in Fig. 1. A SAW is radiated into the digital microfluid, leading to internal acoustic streaming in the digital microfluid, which causes SAW streaming force per unit volume given by [25]
Fs = −ρ(1 + αl2) 3/2A 2ω 2K i exp 2(K ix + αlK iz ), (2)
where ρ is the density of the microfluid, A is the SAW amplitude, ω is angular frequency, ki is the imaginary part of the wave number of the leaky SAW (kl), αl = jα with α = 1 − (V R/V W)2. The wave number of leaky SAWs can be obtained by the angular frequency divided by leaky SAW velocity (VL). The direction of the SAW streaming force is the same angle as the Rayleigh angle θR. When the SAW amplitude is not large enough, the digital microfluid rotates around the center of the microfluid in its original position on the piezoelectric substrate. At this moment, no fluid escapes from the initial digital mi-
2100
IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control ,
vol. 61, no. 12,
December
2014
transducer to the digital microfluid, H is the height of the microfluid on the piezoelectric substrate, and R is the radius of the digital microfluid. The shape of the microfluid on the piezoelectric substrate is almost spherical, which can be verified by experiments in the next section. The microfluid can be expressed as (x − x 0)2 + y 2 + z 2 = R 2. (3)
According to Fig. 2(e), the radius r of a circular section at the height z can be calculated by
r =
R 2 − (H − R − z )2, z ≤ H − R (4)
r =
R 2 − (R + z − H )2, z > H − R. (5)
or
The circle equation at the height z can be expressed as (x − x 0)2 + y 2 = r 2. (6)
Fig. 2. The diagram of forces on a digital microfluid.
crofluid. However, when the SAW is continuously radiated into the digital microfluid for about one second and then suddenly disappears, part of the microfluid escapes from the initial digital microfluid because of inertial forces. The main reason is that the surface tension seeks to restore the equilibrium shape of the microfluid after the acoustic signal is turned off. The digital microfluid oscillations arise and at this stage the inertia breaks up the digital microfluid. There are several forces on the digital microfluids: one is the surface tension of the piezoelectric substrate to the digital microfluid, one is the gravitational force, and the other is the SAW streaming force. A diagram of forces on the digital microfluid is shown in Fig. 2. In Fig. 2, Fig. 2(a) shows the initial digital microfluid on a piezoelectric substrate, and the SAW is not excited. The parameter γ1 is the solid–gas interfacial energy, γ2 is the liquid-gas interfacial energy, and γ3 is the solid–liquid interfacial energy. These parameters represent the adhesion strength of the substrate to the digital microfluid, which is called the adhesion work. Fg is the gravitational force acting on the digital microfluid. The adhesion work and the gravitational force stabilize the microfluid on the substrate. Fig. 2(b) shows the forces when the SAW streaming force is also acting on the microfluid as a result of SAW radiation. Fig. 2(c) shows the forces when the SAW is suddenly removed. At this time, the upper part of the microfluid is mainly acted on by the inertial SAW streaming force, whereas the lower part of the microfluid is attracted by adhesion work and its gravitational force, leading to part of the microfluid escaping from the initial digital microfluid, as shown in Fig. 2(d). Fig. 2(e) is a diagram for analyzing the forces on the digital microfluid. In Fig. 2(e), x0 is the center distance from the interdigital
Then, according (2)–(6), we can derive the acoustic radiation force on the upper part of the microfluid, which is higher than the height of z0. N
F =
H x 0 + r 2 −y i2
∑∫
∫
r ⋅ Fs ⋅ dx dz , (7) N 2
i =1 z x − r 2 −y 0 0 i
where r is divided into N equal segments, and yi = r (i − 1)/N. Using (7), we can calculate the acoustic radiation force on the upper part of the microfluid. The surface tension of the digital microfluid on the piezoelectric substrate can be calculated using 2π
Fsurface =
∫ γ (1 + cos φ) ⋅ r0 dt , (8) 2
0
where ϕ is the contact angle of the microfluid to the Teflon-modified piezoelectric substrate. The value of ϕ can be measured by a contact angle measuring apparatus (Digidrop, GBX Etude des Technologies Avancees Co. Ltd., Bourg de Peage, France), which was measured to be 87.5° in our experiment. B. Experimental Setup The experimental setup for the breakup of digital microfluids is shown in Fig. 3. In Fig. 3, an interdigital transducer and a reflector are fabricated on the 128° yx-LiNbO3 substrate using microelectronic technology. The aperture and period of the interdigital transducer are 4.32 mm and 144 µm, respectively. The center frequency of the interdigital transducer is 27.5 MHz. The number of finger pairs in the interdigital transducer is 35. The area of the acoustic path is coated
zhang and zha: the breakup of digital microfluids on a piezoelectric substrate using SAWs
Fig. 3. The experimental setup for the breakup of digital microfluids.
with Teflon AF 1600 (DuPont Corp., Wilmington, DE). Poly(dimethylsilicone) (PDMS) film with 4.8 mm length in the transportation direction is used for attenuating the SAW to avoid the further breakup of daughter digital microfluids. Filter paper, which will be placed on the surface of the piezoelectric substrate at the position AA′, is used to measure the escape angle of the ejected daughter digital microfluids. A signal generator (SP1461, Nanjing Sample Instrument Technologies Co. Ltd., Nanjing, China) supplies a radio-frequency sine signal. The radio-frequency sine signal is amplified by a power amplifier (TSA002A, Tianjing Zeland Electronic Technology Co. Ltd., Tianjing, China) with a gain of 48 dB and maximum unsaturated output power of 30 W. A highly sensitive charge-coupled imager (DCE-2, Ningbo Yongxin Optics Co. Ltd., Ningbo, China) is used to monitor the breakup of the digital microfluid. MDVNT software (Ningbo Yongxin Optics Co. Ltd.) is used for camera control and image processing. To observe the breakup of a digital microfluid, red dye solution digital microfluid was broken up for demonstration. Red dye solution digital microfluid was first pipetted onto the piezoelectric substrate using a micro-syringe. A radio-frequency signal was applied to the interdigital transducer and then suddenly decreased to lower amplitude. The breakup of daughter digital microfluids at the right side of the sound-absorption film was also studied. The escape angle of the daughter digital microfluids was calculated by measuring the height of the daughter digital microfluid on the filter paper at the position of AA′. III. Results and Discussion A 4-µL red dye solution digital microfluid was broken up on the 128° yx-LiNbO3 piezoelectric substrate. After a digital microfluid was pipetted onto the piezoelectric substrate, a 12.3-dBm radio-frequency signal was applied to the interdigital transducer. By suddenly decreasing the radio-frequency signal power to −3.98 dBm, the digital microfluid was broken up as a result of inertial force. Video snapshots for the breakup of the digital microfluid are shown in Fig. 4.
2101
In Fig. 4, Fig. 4(a) shows the state of a 4-µL red dye solution microfluid on the piezoelectric substrate. According to Fig. 4(a), the shape of the microfluid is nearly a spherical shape. Fig. 4(b) shows the escape of a daughter digital microfluid from the initial digital microfluid when the power of radio-frequency signal is suddenly reduced from 12.3 dBm to −3.98 dBm. Fig. 4(c) shows the state of the microfluid after the initial breakup. The states of the microfluid after the second to fourth breakup times are shown in Figs. 4(d)–4(f). In Fig. 4, the sound-absorption film with powdered carbon was used for observation. If necessary, the daughter microfluids can be further broken up by actuating them to the left of the sound absorption by the interdigital transducer array. However, if the daughter microfluid does not need to be further broken up, the sound-absorption film is indispensable. Attenuation of the SAW causes a decreased SAW streaming force which is not enough to break the microfluid up, as shown in Fig. 5. In Fig. 5, Fig. 5(a) shows the state of a 4-µL red dye solution microfluid on the piezoelectric substrate at the right side of the PDMS sound-absorption film. Fig. 5(b) shows the state of the microfluid after five sudden changes of electric signal from 12.3 dBm to −3.98 dBm. Compared with Fig. 5(a), no breakup occurs during this period. Increasing the power of the electric signal to 24.3 dBm, the microfluid was actuated by a step in which the electric signal power is suddenly decreased to 8.5 dBm, as shown Fig. 5(c). Figs. 5(d)–5(f) show another three steps of transportation of the microfluid by the same power change of electric signal. According to Fig. 5, the PDMS sound-absorption film can partly attenuate the SAW and prevent the daughter digital microfluid from being further broken up. Digital microfluids with particles such as proteins, DNA, or drugs will also be broken up using the same method, but only if the inertial SAW streaming force is greater than that of the sum of intermolecular forces and the gravitational force of the digital microfluid. After the digital microfluid has been broken up, the daughter microfluid with small volume can be used for further microfluidic operations for microfluidic biochemical analysis on the piezoelectric substrate. The acoustic radiation force on the 4-µL red dye solution microfluid can be calculated. The peak-to-peak values of the electric voltage are 3.661 V and 0.548 V, respectively, when the electric signal power is 12.3 dBm and −3.98 dBm. Because the amplitude of the SAW is proportional to the electric voltages applied to the interdigital transducer [25], [26], the amplitude of the SAW can be obtained. The radius of the 4-µL red dye solution microfluid can be measured by the MDVNT software. The height H of the 4-µL digital microfluid on the piezoelectric substrate can be calculated using the relation of the microfluidic volume to its radius, which will be expressed in following text. Taking VL = (3931 + j67.7) m/s and VW = 1500 m/s [25], the acoustic radiation force on the 4-µL red dye solution microfluid can be obtained by (2)–(7). The value is decreased from 0.031 mN to 0.00068 mN when
2102
IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control ,
vol. 61, no. 12,
December
2014
Fig. 4. The video snapshots for the breakup of a 4-µL red dye solution digital microfluid.
the electrical signal power is decreased from 12.3 dBm to −3.98 dBm. The surface tension of the 4-µL red dye solution to the piezoelectric substrate is 0.495 mN and its gravitational force is 0.039 mN. The sum of the surface tension and the gravitational force of the digital microfluid is larger than the endured acoustic radiation force when the continuous electric signal with 12.3 dBm power is used to excite SAW. Thus, the microfluid is stable on the piezoelectric substrate, although a centripetence occurs inside the digital microfluid. However, when the electric signal is suddenly decreased to −3.98 dBm, the lower part of the digital microfluid is stuck on the substrate because the sum of surface tension and gravitational force is far larger than 0.0006 mN. At this time, the upper part of the digital microfluid endures not only the inertia of the acoustic radiation force, but also intermolecular forces
and part of the gravitational force of the digital microfluid. The intermolecular forces between the upper part and lower part of the microfluid is decreased along with their decreasing interface area resulting from surface tension seeking to restore the equilibrium shape of the lower part of digital microfluid. As soon as the sum of intermolecular forces and the gravitational force of the upper part of the microfluid is less than the inertial force, the breakup of the microfluid occurs. The escape angle is calculated by measuring the height at which the daughter digital microfluids hit a filter paper which is vertically placed on the piezoelectric substrate at the position of AA′ as shown in Fig. 3. A 4-µL red dye solution microfluid is pipetted on the piezoelectric substrate at the left side of the sound-absorption film. The microfluid is then broken up by suddenly decreasing the power
Fig. 5. The PDMS sound-absorption film prevents the daughter digital microfluid from being further broken up.
zhang and zha: the breakup of digital microfluids on a piezoelectric substrate using SAWs
2103
Fig. 7. The diagram for calculating the escape angle of the daughter digital microfluids.
where V is the volume of the digital microfluid, and R is the radius of the digital microfluid. Because of the hydrophobic surface of the piezoelectric substrate, the value of H is greater than the radius of the digital microfluid and H2 can be calculated by
H 2 = H 1 − H + R, (10)
then, the escape angle β can be calculated:
Fig. 6. Position of daughter microfluids hitting the filter paper.
of electric signal from 12.3 dBm to −3.98 dBm. Fig. 6 shows a filter paper hit by escaped daughter digital micro fluids after three breakups. In Fig. 6, the words one, two, and three indicate the breakups of the 4-µL red dye solution microfluid three times in succession. The trajectory of escaped daughter digital microfluid is a parabola; however, the horizontal distance from the escaped daughter digital microfluid to the initial microfluid D is far longer than that from the initial microfluid to the filter paper L. The starting trajectory of an escaped daughter digital microfluid is approximately linear. A diagram for calculating the escape angle of daughter digital microfluids is shown as Fig. 7. From the height at which the daughter microfluid hits the filter paper and the distance from the initial digital microfluid to the filter paper, the escape angle β can be calculated. For simplicity, the center of the microfluid can be regarded as the starting point during calculation of the escape angle. The height of the daughter microfluid on the filter paper H1 can be measured. To calculate the H2, the value of H should be obtained. The volume of a digital microfluid on the piezoelectric substrate can be calculated using the following formula:
V =
π ⋅ H 2(3R − H ), (9) 3
tan β =
H2 . (11) L
After the breakup of the 4-µL red dye solution microfluid, H1, L, and R can be measured from Figs. 4 and 6; they are 17.5 mm, 6.7 mm, and 1.0 mm, respectively. According to (9), (10), and (11), the escape angle β is 68.8°. The breakup experiments of the 4-µL red dye solution microfluid were repeated eight times with the same experimental conditions. Their escape angles were calculated as shown in Table I. In Table I, the average value of the escape angles is 68.5°, and that of θ is 21.5°, which is near the Rayleigh angle θR. With respect to the mean value, the maximum error of the escape angles is 3.6°. The difference between the highest and lowest value of the escape angle is 6.4°. The error would arise mainly from several factors: 1) The slope angle of the filter paper at the position AA′ to the surface of the piezoelectric substrate will lead to a distance error from the filter paper to the digital microfluid to be broken up. Although we try to place the filter paper vertically on the surface of the piezoelectric substrate at the position AA′ with the help of a PDMS block, there would still be a slight error. 2) The gain of the power amplifier fluctuates during the experiment, which will affect the initial power of the electrical signal. 3) There exists a measurement error in measuring the distance from the filter paper to the digital microfluid and the height at which the daughter digital microfluid hits the filter paper. To decrease the error, the experiment is repeated eight times, and the average value is used in this work. Different volumes of the initial red dye solution microfluid were broken up using the same power variation of
2104
IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control ,
TABLE I. Escape Angles of the Breakup of a 4-µL Red Dye Solution Microfluid. L (mm) 6.7 6.6 6.6 6.3 6.5 6.6 6.8 6.7
H1 (mm)
β (°)
θ = 90° − β (°)
16.0 21.0 17.8 16.8 15.0 15.6 20.0 17.8
66.5 72.1 69.0 68.8 65.7 66.3 70.7 68.8
23.5 17.9 21.0 21.2 24.3 23.7 19.3 21.2
Volume of initial digital microfluid (µL) 1 2 4 6 8
Volume of escaped daughter microfluid (µL)
β (°)
θ (°)
0.684 1.082 1.99 2.39 3.28
63.4 65.5 68.5 63.1 62.9
26.6 24.5 21.5 26.9 27.1
2014
Volume of escaped daughter microfluid (µL)
β (°)
θ (°)
1.99 2.03 1.89 2.05
68.5 68.8 70.5 72.7
21.5 21.2 19.5 17.3
12.3 16.0 17.7 19.3
TABLE II. The Parameters of the Digital Microfluid Breakup at Different Initial Volume With the Same Power Variation of the Electrical Signal.
December
TABLE III. Escape Angles and Escaped Daughter Microfluid Volume of a 4-µL Red Dye Solution Microfluid at Different Initial Electrical Signal Powers. Initial electrical signal power (dBm)
the electrical signal. The parameters of the breakup of the microfluids are listed in Table II. According to Table II, we can deduce that the volume of the escaped daughter microfluid increases with the initial volume of the digital microfluid. The escape angle is affected by initial volume of the digital microfluid. When the volume of initial digital microfluid is smaller—for example, 1 and 2 µL—the escape angle is also smaller because of the smaller SAW streaming force. However, when the volume of initial digital microfluids is greater— for example, 6 and 8 µL—the escape angle is gradually decreased. The reason for this is that the gravitational force of the escaped daughter digital microfluid cannot be ignored and affects its movement after breakup. The phenomenon is also verified by Fig. 6. To observe the influence of initial electrical signal power on the parameters of the microfluidic breakup, a 4-µL red dye solution microfluid was broken up at different initial electrical signal powers. Table III shows the escaped daughter digital microfluid volume and escape angle at four different initial electric signal powers. Table III shows that the influence of initial electric signal power on the volume of the escaped daughter microfluid is not obvious. However, the escape angle is slightly affected by the initial electrical signal power. The reason is that the surface tension is greater than the acoustic radiation force, which determines the volume of the daughter microfluid. The acoustic radiation force is the main factor to affect the movement of the escaped daughter microfluid, and the escape angles of microfluids were slightly increased with initial electrical signal power. The influences of turn-off time on the escape angles and the volume of the escaped daughter microfluids were also studied, as shown in Table IV.
vol. 61, no. 12,
In Table IV, we can observe that the escape angles and the volume of the escaped daughter microfluids are not obviously affected by the turn-off time when it is greater than 0.533 s, although the digital microfluid cannot be broken up at a very short turn-off time. The reason is that the inertial force is determined by SAW streaming force, which is hardly affected by the duration at the same electrical signal power after the inertial force has been formed. We define transfer efficiency as the ratio of the volume of the escaped daughter microfluid to the initial volume of the digital microfluid. According the Tables II, III, and IV, we can deduce that the transfer efficiency is affected by the initial volume of the digital microfluid at the same power variation of the electrical signal. However, the influence of initial electric signal power and turn-off time on the transfer efficiency is not obvious. The reasons have already been mentioned. Although flow-based microfluidics have obvious advantages in terms of speed and throughput, a piezoelectric microfluidic device based on a digital microfluid does not require networks of microchannels and valves, and it can independently control each digital microfluid to implement microfluidic operations such as droplet transportation, mixture, and separation. Thus, the presented method is especially applied to single-droplet operation or analysis. For example, the presented method can be easily applied for diluting single droplet sample or reagent on a piezoelectric substrate. Because the electrochemical electrodes can be integrated into the piezoelectric device without other processing steps, the presented method can also be applied for electrochemical micro-analysis based on a digital microfluid. However, high-throughput microTABLE IV. Escape Angles and Escaped Daughter Microfluid Volume of a 4-µL Red Dye Solution Microfluid at Different Turn-Off Times. Turn off time (s) 0.533 1.333 1.6 2.067 2.6 3.8 5.467 6.133
Volume of escaped daughter microfluid (µL)
β (°)
θ (°)
1.98 2.02 1.87 2.01 2.11 1.89 1.83 1.94
68.7 68.8 68.5 68.7 69.1 68.7 68.9 69.1
21.3 21.2 21.5 21.3 20.9 21.3 21.1 20.9
The initial electrical signal power is 16 dBm.
zhang and zha: the breakup of digital microfluids on a piezoelectric substrate using SAWs
fluidic analysis, including cell analysis or high-throughput screening for drug detection, should avoid using the presented technique. IV. Conclusions A new method for the breakup of digital microfluids was presented and implemented on a 128° yx-LiNbO3 substrate using microelectronic technology. PDMS sound-absorption film was used to prevent the daughter digital microfluids from being further broken up. The escape angle was calculated by measuring the height of a daughter digital microfluid hit on a filter paper. Droplet breakups were demonstrated using red dye solution digital microfluids. According to the work, several conclusions can be drawn: 1) The breakup of digital microfluids can be implemented by suddenly decreasing the power of an electrical signal with appropriate amplitude; 2) sound-absorption film can effectively avoid further breakup of the daughter digital microfluids; and 3) the initial volume of digital microfluids can affect the escape angle. The presented breakup method for digital microfluids provides a new sample preparation technique which is helpful for microfluidic biochemical analysis on a piezoelectric microfluidic device. References [1] Z. L. Xiao, M. L. Niu, and B. Zhang, “Droplet microfluidics based microseparation systems, ” J. Sep. Sci., vol. 35, no. 10–11, pp. 1284– 1293, 2012. [2] A. Arora, G. Simone, G. Salieb-Beugelaar, J. Kim, and A. Manz, “Latest developments in micro total analysis systems,” Anal. Chem., vol. 82, no. 12, pp. 4830–4847, 2010. [3] M. Kamruzzaman, A. Alam, K. M. Kim, S. H. Lee, Y. H. Kim, G. M. Kim, and T. D. Dang, “Microfluidic chip based chemiluminescence detection of L-phenylalanine in pharmaceutical and soft drinks,” Food Chem., vol. 135, no. 1, pp. 57–62, 2012. [4] E. Jang, S. Kim, and W. G. Koh, “Microfluidic bioassay system based on microarrays of hydrogel sensing elements entrapping quantum dot–enzyme conjugates,” Biosens. Bioelectron., vol. 31, no. 1, pp. 529–536, 2012. [5] J. Autebert, B. Coudert, F. C. Bidard, J. Y. Pierga, S. Descroix, L. Malaquin, and J. L. Viovy, “Microfluidic: An innovative tool for efficient cell sorting,” Methods, vol. 57, no. 3, pp. 297–307, 2012. [6] G. Grenci, G. Birarda, E. Mitri, L. Businaro, S. Pacor, L. Vaccari, and M. Tormen, “Optimization of microfluidic systems for IRMS long term measurement of living cells,” Microelectron. Eng., vol. 98, no. 1, pp. 698–702, 2012. [7] K. N. Han, C. A. Li, M. P. N. Bui, X. H. Pham, B. S. Kim, Y. H. Choa, E. K. Lee, and G. H. Seong, “On-chip electrochemical detection of bio/chemical molecule by nanostructures fabricated in a microfluidic channel,” Sens. Actuators B, vol. 177, no. 2, pp. 472–477, 2013. [8] P. Paik, V. K. Pamula, and R. B. Fair, “Rapid droplet mixers for digital microfluidic systems,” Lab Chip, vol. 3, no. 4, pp. 253–259, 2003. [9] Y. Ai and B. L. Marrone, “Droplet translocation by focused surface acoustic waves,” Microfluid. Nanofluid., vol. 13, no. 5, pp. 715–722, 2012. [10] Y. Okabe, Y. L. Chen, R. Purohit, R. M. Corn, and A. P. Lee, “Piezoelectrically driven vertical cavity acoustic transducers for the convective transport and rapid detection of DNA and protein binding to DNA microarrays with SPR imaging—A parametric study,” Biosens. Bioelectron., vol. 35, no. 1, pp. 37–43, 2012.
2105
[11] U. Brosa and S. Grossman, “Droplet fission: A performance of surface tension,” Phys. Lett. B, vol. 126, no. 6, pp. 425–427, 1983. [12] Y. C. Tan, J. S. Fisher, A. I. Lee, V. Cristini, and A. P. Lee, “Design of microfluidic channel geometries for the control of droplet volume, chemical concentration, and sorting,” Lab Chip, vol. 4, no. 4, pp. 292–298, 2004. [13] D. N. Adamson, D. Mustafi, J. X. J. Zhang, B. Zheng, and R. F. Ismagilov, “Production of arrays of chemically distinct nanolitre plugs via repeated splitting in microfluidic devices,” Lab Chip, vol. 6, no. 9, pp. 1178–1186, 2006. [14] T. Cubaud, B. M. Jose, S. Darvishi, and R. Sun, “Droplet breakup and viscosity-stratified flows in microchannels,” Int. J. Multiph. Flow, vol. 39, no. 1, pp. 29–36, 2012. [15] H. Song, J. D. Tice, and R. F. Ismagilov, “A microfluidic system for controlling reaction networks in time,” Angew. Chem. Int. Ed. Engl., vol. 42, no. 7, pp. 768–772, 2003. [16] D. R. Link, S. L. Anna, D. A. Weitz, and H. A. Stone, “Geometrically mediated breakup of drops in microfluidic devices,” Phys. Rev. Lett., vol. 92, no. 5, art. no. 054503, 2004. [17] T. Uchida, T. Suzuki, and S. Shiokawa, “Investigation of acoustic streaming excited by surface acoustic waves,” in IEEE Ultrasonics Symp., 1995, pp.1081–1084. [18] A. Renaudin, P. Tabourier, V. Zhang, J. C. Camart, and C. Druon, “SAW nanopump for handling droplets in view of biological applications,” Sens. Actuators B, vol. 113, no. 1, pp. 389–397, 2006. [19] Q. Zeng, F. Guo, L. Yao, H. W. Zhu, L. Zheng, Z. X. Guo, W. Liu, Y. Chen, S. S. Guo, and X. Z. Zhao, “Milliseconds mixing in microfluidic channel using focused surface acoustic wave,” Sens. Actuators B, vol. 160, no. 1, pp. 1552–1556, 2011. [20] S. Oberti, A. Neild, R. Quach, and J. Dual, “The use of acoustic radiation forces to position particles within fluid droplets,” Ultrasonics, vol. 49, no. 1, pp. 47–52, 2009. [21] C. J. Myeong and G. Rasim, “Active density-based separation using standing surface acoustic waves,” Sens. Actuators A, vol. 187, no. 1, pp. 22–28, 2012. [22] H. Y. Li, J. R. Friend, and L. Y. Yeo, “Surface acoustic wave concentration of particle and bioparticle suspensions,” Biomed. Microdevices, vol. 9, no. 5, pp. 647–656, 2007. [23] R. Hahn, “Methods for characterization of biochromatography media,” J. Sep. Sci., vol. 35, no. 22, pp. 3001–3032, 2012. [24] C. Y. Lee, H. Y. Yu, W. Pang, and E. S. Kim, “Droplet-based microreactions with oil encapsulation,” J. Microelectromech. Syst., vol. 17, no. 1, pp. 147–156, 2008. [25] S. Shiokawa, Y. Matsui and T. Ueda, “Liquid streaming and droplet formation caused by leaky Rayleigh wave,” in IEEE Ultrasonics Symp., 1989, pp. 643–646. [26] J. R. Doherty, G. E. Trahey, K. R. Nightingale, and M. L. Palmeri, “Acoustic radiation force elasticity imaging in diagnostic ultrasound,” IEEE Trans. Ultrason. Ferroelectr. Freq. Control, vol. 60, no. 4, pp. 685–701, 2013. An-Liang Zhang received the Ph.D. degree in microelectrics from Zhejiang University, Zhejiang, China, in 2004. He joined the Key Laboratory of Analytical Chemistry for Life Science, School of Chemistry and Chemical Engineering, Nanjing University, in 2008 as a postdoctoral fellow. He is an associate professor in the Department of Electrical Engineering, Ningbo University. In June 2014, he joined the Changzhou Institute of Technology. His scientific interests include surface acoustic waves, lab on a piezoelectric substrate, and SAW sensors.
Yan Zha received her B.S. degree in electrical engineering in 2011 from Anhui Engineering University, Anhui, China. She is now a master’s degree student.
