Micro-laboratory on a piezoelectric substrate: new application based on SAW streaming

Transcription

1 Micro-laboratory on a piezoelectric substrate: new application based on SAW streaming Jun Kondoh Graduate School pf Science and Technology, Shizuoka University, Hamamatsu-shi, Japan j-kondoh@sys.eng.shizuoka.ac.jp Abstract A longitudinal wave is radiated into a liquid when the liquid is placed on a surface acoustic wave (SAW) propagating surface. The radiated longitudinal wave induces liquid dynamics. This phenomenon is called SAW streaming. In this paper, SAW streaming phenomena are summarized. Then a novel microlaboratory on a piezoelectric substrate is proposed and demonstrated. On the micro-laboratory, a droplet is moved by the SAW streaming and measured by a sensor, which is integrated on a surface. Moreover, three-layer structure configuration is realized for a disposable application. 1. Introduction Surface acoustic wave (SAW) device has high potential for several applications, such as sensor [1], liquid actuator [2], and motor [3]. For the liquid actuator, Rayleigh-SAW () device is used. The is converted to a leaky-saw at liquid/solid interface, when a phase velocity of the R- SAW is faster than a sound velocity of liquid. Figure 1 shows the schematic illustration of radiation of the longitudinal wave from leaky-saw at liquid/solid interface. Radiation angle is a Rayleigh angle, θ R, and is expressed as 1 V l θ R = sin. (1) VR Here, V l and V R are the velocities of the longitudinal wave in the liquid and the, respectively. Shiokawa et al. discovered that a droplet on an R- SAW substrate is moved and jetted when the liquid volume is small and an acoustic power is raised above a certain threshold [2, 4, 5]. The phenomenon is called Longitudinal wave IDT Rayleigh SAW θ R 128XY-LiNbO 3 Figure 1. Schematic illustration of longitudinal wave radiation from leaky-saw due to liquid droplet loading on propagating surface. SAW streaming. Many applications based on the SAW streaming are reported [5-16]. In this paper, first, SAW at liquid/solid interface is discussed. Then, several SAW streaming phenomena are introduced. Finally, new application of a novel micro-laboratory on a piezoelectric substrate is described. 2. SAW at liquid-solid interface Liquid Leaky-SAW Behavior of the at liquid/solid interface is important role of this research. Particle displacement profiles at water/ interface were calculated using extending method of Campbell and Jones method [17]. Coordinate system and the result are shown in Fig. 2. Particle displacements of u1, u2 and u3 correspond to L (longitudinal), SH (shear horizontal), and SV (shear vertical) components, respectively. As propagating wave on 128YX- LiNbO3 is, SH component is small. The R- SAW has an elliptical displacement on the surface due to the combination of u1 and u3. Trajectory of the elliptical displacement was also calculated and is shown in Fig. 3. At x 3 = 0 λ (λ: wavelength), trajectory is ellipse. Increasing x 3, it changes and becomes linear polarization at x3 = λ. In other word, it

2 x 3 u3 y 0.8 Liquid x 2 x O u1 x Distance from surface (λ) u 1, u 2, u Normalized displacement Figure 2. Coordinate system and particle displacement profiles at and water interface. corresponds to transformation length from the leaky- SAW () to longitudinal wave. Polarization direction agrees with Rayleigh angle. Several liquid behaviors are occurred by the radiated longitudinal wave. Also, at the liquid/solid interface, the leaky- SAW attenuates as shown in Fig. 1. Decay length of the leaky-saw in the propagation direction is 8.74λ for shorted surface or 9.24λ for free surface [2]. Here, the decay length is defined the length when the leaky- SAW amplitude attenuates to 1/e. Streaming force, which is produced by the radiated longitudinal wave, was derived by Shiokawa et al. [4, 5] on the basis of Nyborg theory [18] ( ) 3 / k Water F = ρ 0 1+α iω A (2) Where F is SAW streaming force, ρ 0 is liquid density, α is attenuation constant of leaky-saw, k i is imaginary part of leaky-saw wave number, ω is angular frequency and A is leaky-saw amplitude. As the amplitude is proportional to an applied voltage [6, 8], F is proportional to square of the applied voltage. From the decay length of the amplitude, it for the SAW streaming force is estimated as 4.5λ. x3=0.005 λ x3=0.001 λ x3= λ x3=0 λ Figure 3. Trajectory of the u1 and u3 at different distance from the surface. 3. Experimental u3 y O u3 y O u3 y O A 128 rotated Y-cut X-propagating LiNbO 3 single crystal [19] was utilized as a SAW substrate in this study. As the crystal has the large electromechanical coupling coefficient of 5.5%[20], it was chosen. An interdigital transducer (IDT) with 32 finger pairs, u1 x u1 x u1 x

3 Signal generator Function synthesizer 50% 1 ms RF-power amplifier Antenna tuner mixed. Then the signal was amplified by an RF power amplifier (R&K A ). The amplified signal was fed to the IDTs. The amplitude of the input signal was measured using an oscilloscope (Agilent 54615B). Between the amplifier and the IDTs, an antenna tuner (Daiwa CNW-319II) was inserted to reduce the reflection from the IDTs. Impedance of the IDT was designed for 50 Ω. Using the antenna tuner, impedance of the system is adjusted at 50 Ω. 4. SAW streaming phenomena Figure 4. Experimental setup for generating in this study. apertures of 2 mm, and a center frequency of 48.6 MHz was designed and fabricated on the 128YX- LiNbO3. Experimental setup is shown in Fig. 4. An RF signal from a standard signal generator (Leader 3220) and a pulse signal from a multifunction synthesizer (NF Electronic Instruments 1940) were The SAW streaming phenomena are depended on the applied voltage. The dynamics of a water droplet of 10 µl were observed [8]. Droplet moves below 26V P-P of SAW input voltage. At 28V P-P, small droplets flew out from the droplet. When the voltage was increased above 30V P-P, the droplet atomized. Observed results are shown in Fig. 5. Diameter of generated microscopic droplets during atomization is less than 1 µm. Figure 6 shows the small droplet formation process [5]. Part of the water droplet is swelled by the SAW (c) Figure 5. Observed SAW streaming phenomena. Droplet moving, jetting, and (c) atomizing.

4 50 20V P-P Temperature ( o C) V P-P 10V P-P 5V P-P Time (s) (c) Figure 6. Observation results of small droplet formation [4]. Before SAW incidents into the water droplet, after 0.01s, and (c) after 0.014s. streaming force. Then water tube is rolled. The direction almost agrees with the Rayleigh angle. From the tip of water tube, small droplets are generated. For atomization, however, we have not observed the tube. Atomization is only observed from thin liquid layer [9, 14] and microscopic droplets are generated from whole surface of the layer as shown in Fig. 5 (c). Also, minimum applied voltage for atomization was 4.1 V P-P. But it is impossible to move the droplets. We have considered that the atomization is arisen from a capillary wave on the liquid thin layer. Using SAW streaming, liquid droplet temperature is controlled [21, 22]. Figure 7 shows the used Temperature ( o C) Y=0.085X X Applied voltage (V) (c) Figure 7. Experimental results for 10 µl water droplet. device, time responses, and (c) temperature obtained one minute after the voltage is applied [21]. device and responses. Water droplet of 10 µl was loaded on the center of the device and was generated from the both IDTs to avoid liquid moving. Applied voltage was changed and duty cycle was fixed at 50 %. The relationships between the applied voltage and the temperature at one minute are summarized in Fig. 7(c). The solid line in the figure indicates the fitting curve. The temperature was proportional to the square of the applied voltage. As the amplitude of SAW was proportional to the applied voltage [6, 9],

5 LCR meter (HP 4285A) IDE sensor Figure 8. Observation results of streaming in 80 wt% glycerol/water droplet [23]. The is exited from left IDT. C dl R sol C dl the temperature was proportional to the square of the amplitude. Similar measurements were performed by varying the duty cycle at the fixed applied voltage of 15 V P-P. Linear relationships between temperature and duty factor were obtained [21]. Therefore, liquid temperature is able to control. Observation of streaming is important for understanding the phenomena. In the droplet, acoustic streaming is generated. Figure 8 shows the visualization result of the streaming in the droplet [23]. (Detail results are written in Ref. [23].) Using the SAW streaming, uniform mixture is possible. 5. Development of a novel microlaboratory on a piezoelectric substrate 51. Fundamental research A droplet moving phenomenon has high potential for several applications [4, 6, 7, 8]. Those researches are mainly focused droplet moving. If a sensor is integrated onto the same surface, micro-laboratory using a piezoelectric substrate is realized. Yatsuda et al. proposed one-chip sensor using SAW and SH-SAW [24]. In this paper, an interdigital electrode (IDE) is used as a sensor. The IDE has the same structure as the IDT. Liquid impedance is obtained from the IDE sensor. Figure 9 shows the measurement system. Equivalent circuit model of the one pair in the IDE is shown in Fig. 9. The double layer capacitance, however, is ignored because frequency range of the LCR meter is from 75 khz to 30 MHz. The resistance and capacitance of the medium are represented as following equations [25], respectively. R sol πw 2K sin 1 1 = 2L nl σ πw K cos 2L (3) Figure 9. LCR meter with IDE sensor and equivalent circuit of one pair. Here, C dl is the double layer capacitance at each electrode, R sol is the resistance of the medium, and Cdi is the capacitance of the medium. Impedance (Ω) Concentration of glycerol/water (wt%) Figure 10. Impedance as a function of concentration of glycerol/water binary-mixture. C di C di Glass plate Theoretical value πw K cos 2L = nlε (4) πw 2K sin 2L Here, n is the number of the fingers, l is the length of the fingers, σ and ε are the conductivity and permittivity of the medium, w is the spacing between electrode, L=w+(electrode width), and K is the complete elliptic integral of the first kind. In our case, w and electrode width are equal, so L=2w. The IDE is normally fabricated on a glass substrate. In this paper, we fabricated the IDE sensor on the glass substrate and 128YX-LINbO3 and compared. Figure 10 shows the results. The theoretical values were calculated from

6 IDT IDE IDT IDE DW Glycerol SAW DW Slide glass IDT IDE Mixed liquid Figure 11. Schematic illustration of the measurements. Impedance (Ω) Mixed in the beaker Mixed on the SAW substrate Concentration of glycerol/water (wt%) Figure 12. Experimental results of the novel microlaboratory. eqs. (3) and (4). All results agree well. Therefore, The IDE sensor on is able to use as the detector of liquid impedance. The IDT and the IDE were integrated on the 128YX- LiNbO3. The IDT was used to generate the and the IDE was used as a detector. Figure 11 shows illustrations of the measurement method. Glycerol droplet was placed on the IDE and distilled water (DW) droplet was on the propagating surface. The water droplet was moved and mixed with glycerol. Then liquid impedance was measured. For comparison, the same concentration samples, which were mixed in a beaker, were prepared. The results are shown in Fig. 12. Both values agree well. From the results, we concluded that uniform mixture using the SAW streaming was achieved as be expected. A continuous flow is normally utilized at a microfluidic device. A Y-shaped flow channel is utilized for liquid mixing. For the Y-shaped micro-fluidic device, however, mixture of different liquids is difficult due to laminar flow. Also, quantity of sample increases due Figure 13. Schematic illustration of the novel microlaboratory with three-layer structure. to continuous flow, even if the device is small. For the SAW device, liquid droplet is moved. Minimum volume of liquid for 50 MHz device is less than 0.2 µl. When a SAW frequency is increased, minimum volume decreases. In other word, necessary quantity of a sample solution is a little. This is advantage to use SAW device for a micro-laboratory substrate. Other advantage is that the droplet is digitally moved. Therefore, position control of a droplet is performed by an electrical signal. Electrowetting actuation method was proposed for a droplet moving [26]. However, the structure is complex. For droplet moving, the use of the SAW streaming phenomenon is optimum selection Development of disposable microlaboratory on a piezoelectric substrate Using SAW device, the micro-laboratory is realized. Compared with other methods, it has many advantages. On the other hand, as price of a piezoelectric substrate is high, realization of a disposable micro-laboratory using a piezoelectric substrate is difficult. If a droplet is moved on a glass plate, low cost micro-laboratory is realized. When an acoustic wave is generated to move a droplet on the glass, a piezoelectric layer is necessary on the glass plate [27]. Instead of the piezoelectric layer, we propose simple and creative method to resolve the disadvantage. Figure 13 shows the idea of a novel and disposable micro-laboratory [28]. The IDT was fabricated on a piezoelectric substrate. Sensors are fabricated on a sensor plate. Using a matching liquid layer, two plates are bonded. At the measurement, the and a slide glass

7 Impedance (Ω) Mixed on the micro-laboratory Theoretical value Molar concentration (mol/l) Figure 14. Experimental results of the novel microlaboratory with three-layer structure. Glass Glass/liquid/ Water droplet were used as the piezoelectric substrate and the sensor plate. Distilled water was used as the matching liquid. The IDE was fabricated on the slide glass. The R- SAW generated from the IDT radiates the longitudinal wave into the matching liquid. The radiated wave generates an acoustic wave into the glass plate. A droplet on the glass plate is moved by the acoustic wave in the glass plate (see Fig. 15). One liquid was placed on the IDE and the other was moved. Potassium chloride aqueous solutions with different molar concentrations were used as sample solutions. Figure 14 shows the experimental results. The experimental results agree with the theoretical values. For the proposed structure, by changing a sensor plate, the piezoelectric substrate is able to reuse. In other word, disposable micro-laboratory is realized with three-layer structure. Figure 15 shows the droplet moving on the slide glass. As the acoustic wave is generated into the glass plate at three-layer structure region, droplet is moved to only glass plate region. The proposed method is able to combine with an optical measurement system, such as a microscope. 6. Conclusions (c) (d) Figure 15. Observation results of water droplet moving on the slideglass/liquid/128yx-linb3 structure. In this paper, fundamentals and applications based on the SAW streaming phenomena are describes. Several liquid behaviors are caused by the SAW streaming. The behaviors are grouped into two classes. Droplet moving, jetting, and heating are directly caused by the radiated longitudinal wave. On the other hand, atomization is non-direct phenomenon and it is mainly caused by the capillary wave. On the basis of the droplet moving, the microlaboratory is demonstrated. The proposed microlaboratory has three-layer structure. The droplet movement mechanism is fabricated on a piezoelectric substrate and a sensor is fabricated on a sensor plate. To insert a matching layer between them, disposable micro-laboratory is realized. In future, for optimization of the structure, wave calculation in the structure is necessary. Also, other sensors, such as acoustic wave and optical sensors, will be integrated on the sensor plate.

8 Acknowledgement Original work of the SAW streaming was started by Dr. Shiokawa, SAW&SPR-Tech. Co., who was supervisor of the author. The author would like to acknowledge the continuing guidance and encouragement of Dr. Shiokawa. This work is corroborated with Dr. Sugimoto and graduated and under-graduated students of the author s laboratory. References [1] J. Kondoh, Y. Okiyama, S. Mikuni, Y. Matsui, M. Nara, T. Mori, H. Yatsuda, Jpn. J. Appl. Phys., Vol. 47, No. 5B,pp , [2] S. Shiokawa, Y. Matsui, and T. Moriizumi, Jpn. J. Appl. Phys., Vol. 28, suppl. 28-1, pp , [3] T. Shigematsu, M. K. Kurosawa, IEEE Trans. UFFC, Vol. 55, No. 9, pp (2008). [4] S. Shiokawa, Y. Matsui, T. Ueda, Jpn. J. Appl. Phys., Vol. 29, suppl. 29-1, pp , [5] S. Shiokawa, Y. Matsui, Mat. Res. Soc. Symp. Proc., 360, pp.53-64, [6] A. Sano, Y. Matsui, S. Shiokawa, Jpn. J. Appl. Phys., Vol. 37, pp (1998). [7] S. Alzuaga, S. Ballandras, F. Bastien, W. Daniau, B. Gauthier-Manuel, J. F. Manceau, B. Cretin, P. Vairac, V. Laude, A. Khelif, R. Duhamel, Proc. IEEE Ultrasonic Symp., pp , [8] A. Wixforth, C. Strobl, Ch. Gauer, A. Toegl, J. Scriba, and Z. v. Guttenberg, Anal. Bioanal. Chem., Vol. 379, pp (2004). [9] K. Chono, N. Shimizu, Y. Matsui, J. Kondoh, S. Shiokawa, Jpn. J. Appl. Phys., Vol. 43, No. 5B, pp (2004). [10] J. W. Kim, Y. Yamagata, M. Takasaki, B. H. Lee, H. Ohmori, T. Higuchi, Sensors and Actuators B, 107, pp , [11] D. Beyssen, L. Le Brizonal, O. Elmazria, P. Alnot, Sensors and Actuators B, Vol. 118, pp (2006). [12] A. Renaudin, P. Tabourier, V. Zhang, J. C. Camart, C. Druon, Sensors and Actuators B, vol. 113, (2006). [13] M. K. Tan, J. R. Friend, L. Y. Yoe, Lab. Chip., Vol. 7, pp (2007). [14] N. Murochi, M. Sugimoto, Y. Matsui, J. Kondoh, Jpn. J. Appl. Phys., Vol. 46, pp , [15] J. R. Friend, L. Y. Yeo, D. R. Arifin, A. Mechler,, Nanotechnology, 19, (2008) [16] M. Cecchini, S. Girardo, D. Pisignano, R. Cingolani, F. Beltram, Appl. Phys. Lett., Vol. 92, (2008). [17] J. J. Campbell, W. R. Jones, IEEE Trans. Sonics & Ultrason., Vol. SU-17, pp (1970). [18] W. L. M. Nyborg, Physical Acoustics II part B (Academic Press, New York), W. P. Mason ed., pp , [19] K. Shibayama, Y. Yamanouchi, H. Sato, T. Meguro, Proc. of the IEEE, 64, 5, 1973, [20] Product catalogue, Yamaju Ceramics Co. Ltd. [21] J. Kondoh, N. Shimizu, Y. Matsui, M. Sugimoto, S. Shiokawa, Proc. the 2005 IEEE International Ultrasonics Symp., pp , (2005). [22] J. Kondoh, N. Shimizu, Y. Matsui, S. Shiokawa, IEEE Trans. on UFFC, Vol. 52, No. 10, pp , (2005). [23] S. Ito, M. Sugimoto, Y. Matsui, J. Kondoh, Jpn. J. Appl. Phys., Vol. 46, No. 7B, pp (2007). [24] H. Yatsuda, T. Kogai, S. Shiokawa, 2 nd Japan-Taiwan Workshop on Future Frequency Control Devices (2007). [25] Peter Van Gerwen et al., Sensors and Actuators B, Vol. 49, pp (1998). [26] H. Ren et al., Sensors and Actuators B, Vol. 98, pp (2004). [27] R. Ozaki, M. Aoki, H. Moritake, K. Yoshino, K. Toda, Jpn. J. Appl. Phys., Vol. 45, pp (2006). [28] N. Yasuda, M. Sugimoto, J. Kondoh, Proc. USE2008 (accepted for presentation).