- Email: [email protected]
A novel thermo-pneumatic peristaltic micropump with low temperature elevation on working fluid
Sensors and Actuators A 165 (2011) 86–93
Contents lists available at ScienceDirect
Sensors and Actuators A: Physical journal homepage: www.elsevier.com/locate/sna
A novel thermo-pneumatic peristaltic micropump with low temperature elevation on working fluid Bonnie Tingting Chia, Hsin-Hung Liao, Yao-Joe Yang ∗ Department of Mechanical Engineering, National Taiwan University, Taipei, Taiwan, ROC
a r t i c l e
i n f o
Article history: Available online 1 March 2010 Keywords: Thermo-pneumatic Peristaltic Micropump Microfluidic device MEMS
a b s t r a c t This work presents a novel thermo-pneumatic peristaltic micropump with low temperature elevation on working fluid. The proposed device, which consists of two separate zones for air-heating and fluidsqueezing, is realized by using micromachining techniques. Also, the device can be operated by using a small and simple actuation circuitry with low applied voltages. Under similar operational conditions, the proposed micropump shows similar fluid-pumping performance when compared with the conventional design of thermo-pneumatic micropumps. However, for the proposed design, the temperature elevation on the fluid-pumping area is as small as about 2.0 K, which is less than 8% of that generated by the conventional design. Furthermore, by applying higher voltages, larger flow rate can be achieved with relatively small increase in temperature elevation. Due to low temperature elevation on working fluid, the proposed device is suitable for the applications such as DNA chips or protein chips. In addition, because of its small size and simple actuation scheme, potentially the proposed device can be integrated into the devices for point-of-care applications. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Micro-electro-mechanical systems (MEMS) technologies are widely employed to realize microfluidic systems for biomedical applications. Typical microfluidic devices include micropumps for fluid driving, microvalves for fluid switching, and micromixers for fluid blending [1]. The advantages of these microfluidic devices include rapid response, small size, and significant reduction in reagents. In general, micropumps can be sorted into two types: mechanical type and non-mechanical type [2]. In terms of pumping-power-delivery schemes, micropumps can be classified as displacement type and dynamic type [3]. Peristaltic micropumps, which can be categorized as the mechanical type as well as the displacement type, utilize vibrating diaphragms in successive peristaltic sequence. Numerous actuation mechanisms have been proposed for continually generating peristaltic series on diaphragms to convey fluid in a desired direction. Piezoelectric actuation scheme is one of the well-known approaches for actuating displacement pumps [4–8], while it requires relatively complex, and possibly bulky, driving circuits. Pneumatic actuation by compressed air is another popular approach [9–14]. This method is considered to be robust and harmless for biomedical applications. However, an external compressor, which might be noisy and bulky, is indispensable for system operation. Electro-
∗ Corresponding author. Tel.: +886 2 33662712. E-mail address: [email protected] (Y.-J. Yang). 0924-4247/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.sna.2010.02.018
static actuation ensures smoother flow at high operating frequency, but relatively high driving voltages are needed [15,16]. Magnetic actuation with an electromagnetic motor is simple and robust, but requires a complex external mechanism for handling rotary motions [17,18]. On the other hand, thermal actuation possesses the advantages of simple device structure, simple actuation scheme, relatively low actuation voltage, and small system size. Therefore, it has the potential for incorporation in portable equipment, such as the devices for the point-of-care (POC) applications. A thermopneumatic peristaltic micropump was proposed and developed in [19]. In [20], the characterization and dynamic modeling of thermo-pneumatic peristaltic micropumps were presented. Also, the proposed pumps can be integrated with a tiny circuit module of small foot-print. In these previous works, on-chip thermal heaters are integrated into the system for device actuation. By heating air sealed in the chamber, the thermally induced volume change of air deforms the diaphragm to convey fluid toward a desired direction. However, when the thermo-pneumatic actuation scheme is used in microfluidic devices, high temperature elevation might damage the content of working fluid. For example, excess device temperature elevation might cause coagulation of protein. Therefore, temperature elevation is a critical issue for employing thermo-pneumatic pumps in biomedical applications. In this paper, we propose a thermo-pneumatic peristaltic micropump, which consists of two separate zones for air-heating and fluid-squeezing. The temperature elevation on working fluid can be significantly reduced because the fluidic channel surface is
B.T. Chia et al. / Sensors and Actuators A 165 (2011) 86–93
87
Fig. 1. Schematic illustrations of the proposed and the conventional designs: (a) the 3D model, (b) top view, and (c) exploded view.
away from the heater, which reduces the possibility of damaging the content in the working fluid by high temperature. Furthermore, the flow rate performance of the pump can be improved by increasing the applied voltage, while insignificant temperature elevation on working fluid is induced. The device can be realized by standard micromachining processes such as the soft-lithography process [21] and the lift-off process. Also, the device only requires a simple and small external circuit for independent operation [20]. Therefore, the proposed device can be potentially integrated in a portable system for POC applications [22].
2. Design 2.1. Configuration and operational principle In order to fully describe the design and operational principle of the proposed device, we start with the description of the conventional design of thermo-pneumatic peristaltic micropumps [19,20]. The left column of Fig. 1 shows the schematic view, top view and the exploded drawing of the conventional design. Fig. 2 is the corresponding cross-sectional view along line AA indicated in Fig. 1(b).
88
B.T. Chia et al. / Sensors and Actuators A 165 (2011) 86–93
Fig. 2. The conventional design in cross-sectional view along line AA .
As shown in Fig. 2, the conventional design consists of heaters, actuation chambers, pumping diaphragms, pumping chambers, and a microchannel. When a voltage is applied to the heater, the temperature of the air inside the actuation chamber increases, and consequently the air volume expands and the diaphragm is deformed to squeeze the working fluid in the pumping chamber. Note that the pumping chambers are parts of the microchannel. As the temperature of the air inside the actuation chamber decreases by natural cooling, the air volume shrinks and the deformation of the diaphragm decreases. By controlling the motions of the diaphragms in sequence, the working fluid can be conveyed. Fig. 3 is the timing diagram (actuation sequence) of the applied voltages on the heaters. In addition, a reversed sequence makes the working fluid flows in the opposite direction, so this type of devices is bi-directional. The schematic of the proposed design is shown in the right column of Fig. 1. Fig. 4(a)–(c) are the cross-sectional views of the proposed design along line BB , line CC , and line DD (shown in Fig. 1(b)), respectively. As shown in Fig. 1(a) and Fig. 4, the actuation chamber of the proposed design is composed of two connected subchambers: the air-heating zone (AH-zone) and the fluid-squeezing zone (FS-zone). Heaters are located below the AH-zone. The pumping diaphragm, which is above the FS-zone, is used to squeeze the fluid in the pumping chamber when the air in the actuation chamber is heated and expended. Fig. 5 illustrates the actuation sequence of the proposed design, which is the same as the conventional design (i.e., Fig. 3). When a voltage drop is established across the heater, the temperature of the air inside the AH-zone increases. Consequently, the air pressure in the AH-zone increases, which in turn increases the air pressure in the FS-zone because both zones are connected. Therefore, the diaphragm on the top of the FS-zone is deformed to squeeze the fluid. The major difference between these two designs is that the actuation chamber of the conventional design is in charge of airheating and fluid-squeezing at the same location. Therefore, the working fluid flows through the pumping diaphragm right above the heaters, which give rise to noticeable temperature elevation on the working fluid due to direct heating. On the other hand, for the proposed micropump, the microchannel (i.e., the working fluid) is relatively far away from the heaters when compared with the conventional design. Consequently, the proposed design does not induce much temperature elevation on working fluid. 2.2. Indirect heating In the previous work of the conventional design [20], the deformation amplitude (s) of a circular diaphragm due to thermal actuation can be approximated as: s ∝ T · V0 ,
(1)
where V0 is the initial volume of the air in the actuation chamber and T is the temperature elevation of the air.
Fig. 3. (a) Actuation sequence of the peristaltic micropump. (b) The timing diagram of the actuation signals (voltages) for each chamber.
For the proposed design, the radius of the FS-zone, which is also the radius of the pumping diaphragm, is assumed to be the same as the radius of the AH-zone. Also, we assume that the thermal energy, which is generated in the air-heating zone, is slowly transmitted to the FS-zone. Therefore, the temperature elevation T in
B.T. Chia et al. / Sensors and Actuators A 165 (2011) 86–93
89
Fig. 4. The cross-sectional views of the proposed micropump along (a) line BB , (b) line CC , and (c) line DD .
the AH-zone can be considered to be almost the same as that in the actuation chamber of the conventional design, when both designs are under the same heating condition. As a result, the deformation amplitude (s) in the proposed design could be almost the same as that of the conventional design under the same operating conditions, which in turn implies that the flow rates of both designs may be similar. Furthermore, by applying higher voltages, higher temperature elevation in the AH-zone could be obtained, which generates larger amplitude of diaphragm deformation, and may give higher flow rate. Therefore, by this way, the flow rates of the proposed design may increase without significantly increasing temperature elevation on the working fluid, since the FS-zone is not directly heated by heaters. 3. Fabrication As shown in Fig. 1(c), the proposed design consists of three different layers: the PDMS channel layer, the PDMS membrane layer,
Fig. 6. (a–n) Fabrication process of the micropump.
and the glass substrate layer. The PDMS channel layer and the PDMS membrane layer are fabricated by soft-lithography technology [21], and the glass substrate layer is realized by surface micromachining technique. Fig. 6 shows the fabrication process flow of the micropump. These three layers are fabricated individually and then bonded together. In order to compare the performances, both types of the micropumps are fabricated. The dimensions of the AH-zones, FS-zones, heaters, and microchannels are designed to be the same for both types. Dimensions of these devices are listed in Table 1. Fig. 6(a)–(d) show the fabrication process of the PDMS membrane layer. At first, a SU-8 master (GM 1070, Gersteltec Sarl) with patterns of 180-m thick is fabricated on a silicon wafer (Fig. 6(b)). PDMS prepolymer and curing agent (Sylgard® 184, Dow Corning Corp.) are mixed at 10:1 ratio. After stirred thoroughly and degassed in a vacuum chamber, the prepared PDMS mixture is poured onto a patterned SU-8 master (Fig. 6(c)). The prepared PDMS mixture is poured and spun at 300 rpm for 10 s. After cured at 90 ◦ C for 60 min, the cured PDMS layer is peeled off from the master substrate (Fig. 6(d)). Table 1 Dimensions of the proposed and the conventional micropumps (unit: m). Design Proposed
Depth
3000 (AH-zone) 3000 (FS-zone) 180
Diaphragm
Diameter Thickness
3000 50
Channel
Width Depth
100 50
Actuation chamber
Fig. 5. Actuation sequence of the proposed peristaltic micropump.
Diameter
Conventional 3000 180 3000 50 100 50
90
B.T. Chia et al. / Sensors and Actuators A 165 (2011) 86–93
Fig. 7. Photographs of the fabricated devices.
The PDMS channel layer is fabricated by the similar process (Fig. 6(e)–(i)). The patterned SU-8 master is 50 m in thickness (Fig. 6(f)). The chamber layer is peeled off from the master after softly cured at 90 ◦ C for 30 min (Fig. 6(h)). The inlet and the outlet are punched using stainless steel pipe (Fig. 6(i)). The fabrication process of the glass substrate layer is described in Fig. 6(j)–(m). Positive photoresist (AZ4620, Hoechst) is spin-coated (5,000 rpm, 50 sec) and patterned on a glass substrate (Fig. 6(k)). A 200 Å thick titanium and 3000 Å thick gold layer is deposited by electron beam evaporation (Fig. 6(l)), and then the heater pattern is formed by removing the photoresist (Fig. 6(m), the lift-off process). After oxygen plasma treatment (pressure: 350 mTorr; RF power: 10.5 W; 90 s), the PDMS channel layer and the PDMS membrane layer are bonded and then warmed at 80 ◦ C for 5 min. The glass substrate layer is also bonded to the other side of PDMS membrane layer after the same oxygen plasma treatment (Fig. 6(n)). Finally, polyethylene (PE) tubes are connected to the inlet/outlet holes. Fig. 7 shows the picture of fabricated 3-chamber devices (the proposed design and the conventional design). The dimension of the proposed 3-chamber micropump is 16 mm × 18 mm × 5 mm. 4. Experiment results and discussions Fig. 8 shows the measured transient temperature elevations of the conventional and the proposed designs. In this experiment, the PDMS channel layers of the devices were removed in order to expose the top surfaces of the PDMS membrane layers for temperature measurement by using an infrared thermometer. By this way, the temperature elevation, which is directly experienced by the fluid in the microchannel, can be obtained, because the PDMS membrane layer is right below the microchannel. For the conventional design, the measurement is made at Point-X indicated in Fig. 1(b). For the proposed design, the measurement is made at Point-Z (the FS-zone) indicated in Fig. 1(b). The measurement is conducted at room temperature (297.7 K), and the applied voltage is 5 V with a 40% duty ratio [20]. During repeated actuation cycles, the maximum value of temperature elevation on the diaphragm increases because each heating starts before the gas sealed in each chamber fully cools to ambient temperature. However, the average value of temperature elevation finally reaches a steady value since the cooling rate also increases with temperature. The measured temperature elevations above the FS-zone (Point-Z) are much lower than those on the AH-zone (Point-Y) which is similar to Point-X in the conventional design. Furthermore, at lower operating frequencies, the temperature variation becomes larger due to sufficient times for heating and cooling processes. Maximum values of the measured transient temperature elevations on pumping diaphragms are listed in Table 2. At Table 2 Maximum values of the measured temperature elevations on pumping diaphragms (unit: K). The applied voltage is 5 V. Devices
Operating frequency 0.05 Hz
Proposed design Conventional design
2.9 70.3
0.1 Hz 2.2 57.8
1.0 Hz 2.0 40.8
Fig. 8. Transient temperature elevations of the conventional and the proposed designs at various operating frequencies: (a) 0.05 Hz, (b) 0.1 Hz, and (c) 1 Hz.
1 Hz, the maximum temperature elevation measured on the pumping diaphragm above the FS-zone (i.e., Point-Z) is 2.0 K, which is a more than 90% reduction in temperature elevation when compared with those at Point-X and Point-Y. These experiment results also show that the aforementioned assumption that insignificant thermal energy flows from the AH-zone to the FS-zone is reasonable. Fig. 9 shows the temperature distribution on the upper surface of the PDMS membrane layer captured by an infrared thermal imager during pumping. Also, the shape of the micropump is sketched for indicating the location of each sub-component. This plot shows that the high temperature area is localized around the AH-zone. In Fig. 10, flow rates of the micropumps of the proposed and conventional designs are measured at the same condition as Fig. 8. The experiment setup is described in [20]. At low frequency, the flow rate increases with the frequency. However, the flow rate starts to decrease as frequency reaches to a certain value. It is because as
B.T. Chia et al. / Sensors and Actuators A 165 (2011) 86–93
Fig. 9. Infrared image captured on the upper surface of the proposed micropump without the PDMS channel layer. The shapes of each component are sketched.
the frequency increases, the actuation amplitude of the diaphragm decreases due to insufficient heating and cooling. With the same applied voltage, the maximum flow rate of the proposed micropump performs at about 0.3 Hz, while the maximum flow rate of the conventional device occurs at 1.5 Hz. We speculate that the shift of the maximum-flow-rate frequency is because the proposed design has larger thermal mass of air due to larger air volume (i.e., two zones), which results in slower thermal response time. Since the temperature measured on the pumping diaphragm of the proposed micropump is much lower than that of the conventional one, increasing applied voltage is a feasible approach to increase flow rate with low temperature elevation on the working fluid. Fig. 11 shows the measured flow rates with various applied voltages (heating duty ratio is 40%). The flow rate increases with the applied voltage. Besides, the maximum values of the flow rates occur at higher operating frequencies for larger applied voltages. It is reasonable to conclude that larger applied voltages can compensate for the slow thermal response arising from the enlarged actuation chambers.
Fig. 10. Measured flow rate vs. operating frequency for both types of micropumps.
91
Fig. 11. Measured flow rate vs. operating frequency with various applied voltages for the proposed micropump.
Fig. 12 shows the measured relationships between backpressure and flow rate with various applied voltages. The operating frequency is fixed at the value when maximum flow rate occurs at zero backpressure (see Fig. 11). As the backpressure increases, flow rate decreases linearly. Since higher applied voltage results in larger actuation amplitude of the pumping diaphragm, the cut-off backpressure increases with the applied voltage. Table 3 sums up the pumping performances of both the proposed and conventional micropumps. The maximum temperatures of the pumping diaphragm for different applied voltages are shown in Fig. 13. Higher applied voltages lead to higher flow rates but higher temperature elevations. When the applied voltage is 7 V, the temperature on the diaphragm is still below 30.2 ◦ C (at room temperature of 24.7 ◦ C). Note that it takes about 15 s for the temperature of the diaphragm to reach this steady-state maximum value. If the working fluid could stand higher temperatures, larger voltages may be applied to achieve
Fig. 12. Measured relationships between backpressure and flow rate with various applied voltages.
92
B.T. Chia et al. / Sensors and Actuators A 165 (2011) 86–93
Table 3 Pumping performances of the proposed and conventional micropumps. Devices
Pumping performance Applied voltage
Maximum flow rate
Operating frequency
Cut-off backpressure
Proposed design
5V 7V 9V
8.74 L/min 13.21 L/min 20.01 L/min
0.3 Hz 0.9 Hz 1.2 Hz
245 Pa 343 Pa 490 Pa
Conventional design
5V
9.18 L/min
1.2 Hz
400 Pa
References
Fig. 13. Maximum temperatures measured at Point-Z vs. operating frequency with various applied voltages.
even higher flow rate. When the applied voltage is 9 V and the driving frequency is 1.2 Hz, the proposed micropump generates a maximum flow rate of 20.01 L/min while the temperature on the diaphragm is still below 36.2 ◦ C. This temperature is still appropriate for many devices used in biomedical applications.
5. Conclusion In this paper, a novel thermo-pneumatic peristaltic micropump with low temperature elevation on the working fluid is developed. The proposed device has three actuation chambers, each of which consists of individual zones for air-heating and fluidsqueezing. Therefore, the working fluid will not be heated directly during operation, which gives tremendous flexibility in biomedical applications. The proposed micropump shows similar performance when compared with the conventional design under the same operational conditions. Also, it is demonstrated that increasing applied voltage can improve the flow rate, while only gives very mild temperature elevation on the working fluid.
Acknowledgements This research is partially sponsored by the Ministry of Education and the National Science Council, Taiwan, ROC (contract number: NSC-97-3114-E-02-007).
[1] C. Zang, D. Xing, Y. Li, Micropumps, microvalves, and micromixers within PCR microfluidic chips: advances and trends, Biotechnol. Adv. 25 (2007) 483–514. [2] N.T. Nguyen, X.Y. Huang, T.K. Chuan, MEMS-micropumps: a review, J. Fluids Eng. 124 (2002) 384–392. [3] D.J. Laser, J.G. Santiago, A review of micropumps, J. Micromech. Microeng. 14 (2004) R35–R64. [4] N.J. Graf, M.T. Bowser, A soft-polymer piezoelectric bimorph cantileveractuated peristaltic micropump, Lab Chip 8 (2008) 1664–1670. [5] M. Bu, T. Melvin, G. Ensell, J.S. Wilkinson, A.G.R. Evans, Design and theoretical evaluation of a novel microfluidic device to be used for PCR, J. Micromech. Microeng. 13 (2003) S125–S130. [6] D.-S. Lee, H.C. Yoon, J.S. Ko, Fabrication and characterization of a bidirectional valveless peristaltic micropump and its application to a flow-type immunoanalysis, Sens. Actuat. B 103 (2004) 409–415. [7] L.-S. Jang, K. Shu, Y.-C. Yu, Y.-J. Li, C.-H. Chen, Effect of actuation sequence on flow rates of peristaltic micropumps with PZT actuators, Biomed. Microdevices 11 (2009) 173–181. [8] M. Zhu, P. Kirby, M. Wacklerle, M. Herz, M. Richter, Optimization design of multi-material micropump using finite element method, Sens. Actuat. A 149 (2009) 130–135. [9] S.B. Huang, M.H. Wu, Z. Cui, G.B. Lee, A membrane-based serpentine-shape pneumatic micropump with pumping performance modulated by fluidic resistance, J. Micromech. Microeng. 18 (2008) 045008. [10] M.A. Unger, H.-P. Chou, T. Thorsen, A. Scherer, S.R. Quake, Monolithic microfabricated valves and pumps by multilater soft lithography, Science 288 (2000) 113–116. [11] P.A. Willis, B.D. Hunt, V.E. White, M.C. Lee, M. Ikeda, S. Bae, M.J. Pelletier, F.J. Grunthaner, Monolithic Teflon® membrane valves and pumps for harsh chemical and low-temperature use, Lab Chip 7 (2007) 1469–1474. [12] W.H. Grover, M.G. von Muhlen, S.R. Manalis, Teflon films for chemically-inert microfluidic valves and pumps, Lab Chip 8 (2008) 913–918. [13] F. Goldschmidtböing, A. Doll, M. Heinrichs, P. Woias, H.J. Schrag, U.T. Hopt, A generic analytical model for micro-diaphragm pumps with active valves, J. Micromech. Microeng. 15 (2005) 673–683. [14] W. Inman, K. Domansky, J. Serdy, B. Owens, D. Trumperm, L.G. Griffith, Design, modeling and fabrication of a constant flow pneumatic micropump, J. Micromech. Microeng. 17 (2007) 891–899. [15] M.M. Teymoori, E. Abbaspour-Sani, Design and simulation of a novel electrostatic peristaltic micromachined pump for drug delivery applications, Sens. Actuat. A 117 (2005) 222–229. [16] J. Xie, J. Shih, Q. Lin, B. Yang, Y.-C. Tai, Surface micromachined electrostatically actuated micro peristaltic pump, Lab Chip 4 (2004) 495–501. [17] T. Pan, E. Kai, M. Stay, V. Barocas, B. Ziaie, A magnetically driven PDMS peristaltic micropump, in: Proceedings of the 26th Annual International Conference of the IEEE EMBS, San Francisco, CA, USA, 2004, pp. 2639–2642. [18] L. Yobas, K.-C. Tang, S.-E. Yong, E.K.-Z. Ong, A disposable planar peristaltic pump for lab-on-a-chip, Lab Chip 8 (2008) 660–662. [19] O.C. Jeong, S.W. Park, S.S. Yang, J.J. Pak, Fabrication of a peristaltic PDMS micropump, Sens. Actuat. A 123–124 (2005) 453–458. [20] Y.-J. Yang, H.-H. Liao, Development and characterization of thermopneumatic peristaltic micropumps, J. Micromech. Microeng. 19 (2009) 025003. [21] Y. Xia, G.M. Whiteside, Soft lithography, Annu. Rev. Mater. Sci. 28 (1998) 153–184. [22] F. Amirouche, Y. Zhou, T. Johnson, Current micropump technologies and their biomedical applications, Microsyst. Technol. 15 (2009) 647–666.
Biographies Bonnie Tingting Chia received her BS degree in Mechanical Engineering at National Taiwan University, Taiwan, in 2006. She is now pursuing the PhD degree in Mechanical Engineering at National Taiwan University, Taiwan. She has been working on microfluidic system for biomedical applications. Hsin-Hung Liao received the BS degrees in Mechanical Engineering from National Chung Cheng University, Chiayi, Taiwan, in 2003. He is now pursuing the PhD degree in Mechanical Engineering at National Taiwan University, Taiwan. His research interests include optical measurement, micropump, drug delivery system and microelectromechanical digital-to-analog converter (MEMS-DAC) device.
B.T. Chia et al. / Sensors and Actuators A 165 (2011) 86–93 Yao-Joe Joseph Yang received his MS and PhD degrees in Electrical Engineering from the Massachusetts Institute of Technology (MIT) in 1997 and 1999, respectively. Also, he received his MS degree (1995) from UCLA and his BS degree (1990) from the National Taiwan University, both in Mechanical Engineering. From 1999 to 2000, he joined the Coventor Inc. (Cambridge, MA) as a senior application engineer. Since 2000, he joined the Department of Mechanical Engineering at the National Taiwan University, Taipei, Taiwan. Currently he is a professor, and serves as the associate chair of the department. He is also the director for CAD Technology in the NTU NEMS
93
Center. From 2005 to 2009, he serves as the deputy Secretary General of the Chinese Institute of Automation Engineering (CIAE). Currently he is the board member of the Institute. His research interests include microelectromechanical systems, nanotechnology, high-precision micromachining, flexible sensing arrays, sensor network, parallel processing, and semiconductor devices and vacuum microelectronics modeling. He has been consulted by more than three U.S-based companies and four Taiwan-based organizations. Dr. Yang is a member of IEEE. He is also the recipient of the Outstanding Young Researcher Award of the National Science Council.
Contents lists available at ScienceDirect
Sensors and Actuators A: Physical journal homepage: www.elsevier.com/locate/sna
A novel thermo-pneumatic peristaltic micropump with low temperature elevation on working fluid Bonnie Tingting Chia, Hsin-Hung Liao, Yao-Joe Yang ∗ Department of Mechanical Engineering, National Taiwan University, Taipei, Taiwan, ROC
a r t i c l e
i n f o
Article history: Available online 1 March 2010 Keywords: Thermo-pneumatic Peristaltic Micropump Microfluidic device MEMS
a b s t r a c t This work presents a novel thermo-pneumatic peristaltic micropump with low temperature elevation on working fluid. The proposed device, which consists of two separate zones for air-heating and fluidsqueezing, is realized by using micromachining techniques. Also, the device can be operated by using a small and simple actuation circuitry with low applied voltages. Under similar operational conditions, the proposed micropump shows similar fluid-pumping performance when compared with the conventional design of thermo-pneumatic micropumps. However, for the proposed design, the temperature elevation on the fluid-pumping area is as small as about 2.0 K, which is less than 8% of that generated by the conventional design. Furthermore, by applying higher voltages, larger flow rate can be achieved with relatively small increase in temperature elevation. Due to low temperature elevation on working fluid, the proposed device is suitable for the applications such as DNA chips or protein chips. In addition, because of its small size and simple actuation scheme, potentially the proposed device can be integrated into the devices for point-of-care applications. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Micro-electro-mechanical systems (MEMS) technologies are widely employed to realize microfluidic systems for biomedical applications. Typical microfluidic devices include micropumps for fluid driving, microvalves for fluid switching, and micromixers for fluid blending [1]. The advantages of these microfluidic devices include rapid response, small size, and significant reduction in reagents. In general, micropumps can be sorted into two types: mechanical type and non-mechanical type [2]. In terms of pumping-power-delivery schemes, micropumps can be classified as displacement type and dynamic type [3]. Peristaltic micropumps, which can be categorized as the mechanical type as well as the displacement type, utilize vibrating diaphragms in successive peristaltic sequence. Numerous actuation mechanisms have been proposed for continually generating peristaltic series on diaphragms to convey fluid in a desired direction. Piezoelectric actuation scheme is one of the well-known approaches for actuating displacement pumps [4–8], while it requires relatively complex, and possibly bulky, driving circuits. Pneumatic actuation by compressed air is another popular approach [9–14]. This method is considered to be robust and harmless for biomedical applications. However, an external compressor, which might be noisy and bulky, is indispensable for system operation. Electro-
∗ Corresponding author. Tel.: +886 2 33662712. E-mail address: [email protected] (Y.-J. Yang). 0924-4247/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.sna.2010.02.018
static actuation ensures smoother flow at high operating frequency, but relatively high driving voltages are needed [15,16]. Magnetic actuation with an electromagnetic motor is simple and robust, but requires a complex external mechanism for handling rotary motions [17,18]. On the other hand, thermal actuation possesses the advantages of simple device structure, simple actuation scheme, relatively low actuation voltage, and small system size. Therefore, it has the potential for incorporation in portable equipment, such as the devices for the point-of-care (POC) applications. A thermopneumatic peristaltic micropump was proposed and developed in [19]. In [20], the characterization and dynamic modeling of thermo-pneumatic peristaltic micropumps were presented. Also, the proposed pumps can be integrated with a tiny circuit module of small foot-print. In these previous works, on-chip thermal heaters are integrated into the system for device actuation. By heating air sealed in the chamber, the thermally induced volume change of air deforms the diaphragm to convey fluid toward a desired direction. However, when the thermo-pneumatic actuation scheme is used in microfluidic devices, high temperature elevation might damage the content of working fluid. For example, excess device temperature elevation might cause coagulation of protein. Therefore, temperature elevation is a critical issue for employing thermo-pneumatic pumps in biomedical applications. In this paper, we propose a thermo-pneumatic peristaltic micropump, which consists of two separate zones for air-heating and fluid-squeezing. The temperature elevation on working fluid can be significantly reduced because the fluidic channel surface is
B.T. Chia et al. / Sensors and Actuators A 165 (2011) 86–93
87
Fig. 1. Schematic illustrations of the proposed and the conventional designs: (a) the 3D model, (b) top view, and (c) exploded view.
away from the heater, which reduces the possibility of damaging the content in the working fluid by high temperature. Furthermore, the flow rate performance of the pump can be improved by increasing the applied voltage, while insignificant temperature elevation on working fluid is induced. The device can be realized by standard micromachining processes such as the soft-lithography process [21] and the lift-off process. Also, the device only requires a simple and small external circuit for independent operation [20]. Therefore, the proposed device can be potentially integrated in a portable system for POC applications [22].
2. Design 2.1. Configuration and operational principle In order to fully describe the design and operational principle of the proposed device, we start with the description of the conventional design of thermo-pneumatic peristaltic micropumps [19,20]. The left column of Fig. 1 shows the schematic view, top view and the exploded drawing of the conventional design. Fig. 2 is the corresponding cross-sectional view along line AA indicated in Fig. 1(b).
88
B.T. Chia et al. / Sensors and Actuators A 165 (2011) 86–93
Fig. 2. The conventional design in cross-sectional view along line AA .
As shown in Fig. 2, the conventional design consists of heaters, actuation chambers, pumping diaphragms, pumping chambers, and a microchannel. When a voltage is applied to the heater, the temperature of the air inside the actuation chamber increases, and consequently the air volume expands and the diaphragm is deformed to squeeze the working fluid in the pumping chamber. Note that the pumping chambers are parts of the microchannel. As the temperature of the air inside the actuation chamber decreases by natural cooling, the air volume shrinks and the deformation of the diaphragm decreases. By controlling the motions of the diaphragms in sequence, the working fluid can be conveyed. Fig. 3 is the timing diagram (actuation sequence) of the applied voltages on the heaters. In addition, a reversed sequence makes the working fluid flows in the opposite direction, so this type of devices is bi-directional. The schematic of the proposed design is shown in the right column of Fig. 1. Fig. 4(a)–(c) are the cross-sectional views of the proposed design along line BB , line CC , and line DD (shown in Fig. 1(b)), respectively. As shown in Fig. 1(a) and Fig. 4, the actuation chamber of the proposed design is composed of two connected subchambers: the air-heating zone (AH-zone) and the fluid-squeezing zone (FS-zone). Heaters are located below the AH-zone. The pumping diaphragm, which is above the FS-zone, is used to squeeze the fluid in the pumping chamber when the air in the actuation chamber is heated and expended. Fig. 5 illustrates the actuation sequence of the proposed design, which is the same as the conventional design (i.e., Fig. 3). When a voltage drop is established across the heater, the temperature of the air inside the AH-zone increases. Consequently, the air pressure in the AH-zone increases, which in turn increases the air pressure in the FS-zone because both zones are connected. Therefore, the diaphragm on the top of the FS-zone is deformed to squeeze the fluid. The major difference between these two designs is that the actuation chamber of the conventional design is in charge of airheating and fluid-squeezing at the same location. Therefore, the working fluid flows through the pumping diaphragm right above the heaters, which give rise to noticeable temperature elevation on the working fluid due to direct heating. On the other hand, for the proposed micropump, the microchannel (i.e., the working fluid) is relatively far away from the heaters when compared with the conventional design. Consequently, the proposed design does not induce much temperature elevation on working fluid. 2.2. Indirect heating In the previous work of the conventional design [20], the deformation amplitude (s) of a circular diaphragm due to thermal actuation can be approximated as: s ∝ T · V0 ,
(1)
where V0 is the initial volume of the air in the actuation chamber and T is the temperature elevation of the air.
Fig. 3. (a) Actuation sequence of the peristaltic micropump. (b) The timing diagram of the actuation signals (voltages) for each chamber.
For the proposed design, the radius of the FS-zone, which is also the radius of the pumping diaphragm, is assumed to be the same as the radius of the AH-zone. Also, we assume that the thermal energy, which is generated in the air-heating zone, is slowly transmitted to the FS-zone. Therefore, the temperature elevation T in
B.T. Chia et al. / Sensors and Actuators A 165 (2011) 86–93
89
Fig. 4. The cross-sectional views of the proposed micropump along (a) line BB , (b) line CC , and (c) line DD .
the AH-zone can be considered to be almost the same as that in the actuation chamber of the conventional design, when both designs are under the same heating condition. As a result, the deformation amplitude (s) in the proposed design could be almost the same as that of the conventional design under the same operating conditions, which in turn implies that the flow rates of both designs may be similar. Furthermore, by applying higher voltages, higher temperature elevation in the AH-zone could be obtained, which generates larger amplitude of diaphragm deformation, and may give higher flow rate. Therefore, by this way, the flow rates of the proposed design may increase without significantly increasing temperature elevation on the working fluid, since the FS-zone is not directly heated by heaters. 3. Fabrication As shown in Fig. 1(c), the proposed design consists of three different layers: the PDMS channel layer, the PDMS membrane layer,
Fig. 6. (a–n) Fabrication process of the micropump.
and the glass substrate layer. The PDMS channel layer and the PDMS membrane layer are fabricated by soft-lithography technology [21], and the glass substrate layer is realized by surface micromachining technique. Fig. 6 shows the fabrication process flow of the micropump. These three layers are fabricated individually and then bonded together. In order to compare the performances, both types of the micropumps are fabricated. The dimensions of the AH-zones, FS-zones, heaters, and microchannels are designed to be the same for both types. Dimensions of these devices are listed in Table 1. Fig. 6(a)–(d) show the fabrication process of the PDMS membrane layer. At first, a SU-8 master (GM 1070, Gersteltec Sarl) with patterns of 180-m thick is fabricated on a silicon wafer (Fig. 6(b)). PDMS prepolymer and curing agent (Sylgard® 184, Dow Corning Corp.) are mixed at 10:1 ratio. After stirred thoroughly and degassed in a vacuum chamber, the prepared PDMS mixture is poured onto a patterned SU-8 master (Fig. 6(c)). The prepared PDMS mixture is poured and spun at 300 rpm for 10 s. After cured at 90 ◦ C for 60 min, the cured PDMS layer is peeled off from the master substrate (Fig. 6(d)). Table 1 Dimensions of the proposed and the conventional micropumps (unit: m). Design Proposed
Depth
3000 (AH-zone) 3000 (FS-zone) 180
Diaphragm
Diameter Thickness
3000 50
Channel
Width Depth
100 50
Actuation chamber
Fig. 5. Actuation sequence of the proposed peristaltic micropump.
Diameter
Conventional 3000 180 3000 50 100 50
90
B.T. Chia et al. / Sensors and Actuators A 165 (2011) 86–93
Fig. 7. Photographs of the fabricated devices.
The PDMS channel layer is fabricated by the similar process (Fig. 6(e)–(i)). The patterned SU-8 master is 50 m in thickness (Fig. 6(f)). The chamber layer is peeled off from the master after softly cured at 90 ◦ C for 30 min (Fig. 6(h)). The inlet and the outlet are punched using stainless steel pipe (Fig. 6(i)). The fabrication process of the glass substrate layer is described in Fig. 6(j)–(m). Positive photoresist (AZ4620, Hoechst) is spin-coated (5,000 rpm, 50 sec) and patterned on a glass substrate (Fig. 6(k)). A 200 Å thick titanium and 3000 Å thick gold layer is deposited by electron beam evaporation (Fig. 6(l)), and then the heater pattern is formed by removing the photoresist (Fig. 6(m), the lift-off process). After oxygen plasma treatment (pressure: 350 mTorr; RF power: 10.5 W; 90 s), the PDMS channel layer and the PDMS membrane layer are bonded and then warmed at 80 ◦ C for 5 min. The glass substrate layer is also bonded to the other side of PDMS membrane layer after the same oxygen plasma treatment (Fig. 6(n)). Finally, polyethylene (PE) tubes are connected to the inlet/outlet holes. Fig. 7 shows the picture of fabricated 3-chamber devices (the proposed design and the conventional design). The dimension of the proposed 3-chamber micropump is 16 mm × 18 mm × 5 mm. 4. Experiment results and discussions Fig. 8 shows the measured transient temperature elevations of the conventional and the proposed designs. In this experiment, the PDMS channel layers of the devices were removed in order to expose the top surfaces of the PDMS membrane layers for temperature measurement by using an infrared thermometer. By this way, the temperature elevation, which is directly experienced by the fluid in the microchannel, can be obtained, because the PDMS membrane layer is right below the microchannel. For the conventional design, the measurement is made at Point-X indicated in Fig. 1(b). For the proposed design, the measurement is made at Point-Z (the FS-zone) indicated in Fig. 1(b). The measurement is conducted at room temperature (297.7 K), and the applied voltage is 5 V with a 40% duty ratio [20]. During repeated actuation cycles, the maximum value of temperature elevation on the diaphragm increases because each heating starts before the gas sealed in each chamber fully cools to ambient temperature. However, the average value of temperature elevation finally reaches a steady value since the cooling rate also increases with temperature. The measured temperature elevations above the FS-zone (Point-Z) are much lower than those on the AH-zone (Point-Y) which is similar to Point-X in the conventional design. Furthermore, at lower operating frequencies, the temperature variation becomes larger due to sufficient times for heating and cooling processes. Maximum values of the measured transient temperature elevations on pumping diaphragms are listed in Table 2. At Table 2 Maximum values of the measured temperature elevations on pumping diaphragms (unit: K). The applied voltage is 5 V. Devices
Operating frequency 0.05 Hz
Proposed design Conventional design
2.9 70.3
0.1 Hz 2.2 57.8
1.0 Hz 2.0 40.8
Fig. 8. Transient temperature elevations of the conventional and the proposed designs at various operating frequencies: (a) 0.05 Hz, (b) 0.1 Hz, and (c) 1 Hz.
1 Hz, the maximum temperature elevation measured on the pumping diaphragm above the FS-zone (i.e., Point-Z) is 2.0 K, which is a more than 90% reduction in temperature elevation when compared with those at Point-X and Point-Y. These experiment results also show that the aforementioned assumption that insignificant thermal energy flows from the AH-zone to the FS-zone is reasonable. Fig. 9 shows the temperature distribution on the upper surface of the PDMS membrane layer captured by an infrared thermal imager during pumping. Also, the shape of the micropump is sketched for indicating the location of each sub-component. This plot shows that the high temperature area is localized around the AH-zone. In Fig. 10, flow rates of the micropumps of the proposed and conventional designs are measured at the same condition as Fig. 8. The experiment setup is described in [20]. At low frequency, the flow rate increases with the frequency. However, the flow rate starts to decrease as frequency reaches to a certain value. It is because as
B.T. Chia et al. / Sensors and Actuators A 165 (2011) 86–93
Fig. 9. Infrared image captured on the upper surface of the proposed micropump without the PDMS channel layer. The shapes of each component are sketched.
the frequency increases, the actuation amplitude of the diaphragm decreases due to insufficient heating and cooling. With the same applied voltage, the maximum flow rate of the proposed micropump performs at about 0.3 Hz, while the maximum flow rate of the conventional device occurs at 1.5 Hz. We speculate that the shift of the maximum-flow-rate frequency is because the proposed design has larger thermal mass of air due to larger air volume (i.e., two zones), which results in slower thermal response time. Since the temperature measured on the pumping diaphragm of the proposed micropump is much lower than that of the conventional one, increasing applied voltage is a feasible approach to increase flow rate with low temperature elevation on the working fluid. Fig. 11 shows the measured flow rates with various applied voltages (heating duty ratio is 40%). The flow rate increases with the applied voltage. Besides, the maximum values of the flow rates occur at higher operating frequencies for larger applied voltages. It is reasonable to conclude that larger applied voltages can compensate for the slow thermal response arising from the enlarged actuation chambers.
Fig. 10. Measured flow rate vs. operating frequency for both types of micropumps.
91
Fig. 11. Measured flow rate vs. operating frequency with various applied voltages for the proposed micropump.
Fig. 12 shows the measured relationships between backpressure and flow rate with various applied voltages. The operating frequency is fixed at the value when maximum flow rate occurs at zero backpressure (see Fig. 11). As the backpressure increases, flow rate decreases linearly. Since higher applied voltage results in larger actuation amplitude of the pumping diaphragm, the cut-off backpressure increases with the applied voltage. Table 3 sums up the pumping performances of both the proposed and conventional micropumps. The maximum temperatures of the pumping diaphragm for different applied voltages are shown in Fig. 13. Higher applied voltages lead to higher flow rates but higher temperature elevations. When the applied voltage is 7 V, the temperature on the diaphragm is still below 30.2 ◦ C (at room temperature of 24.7 ◦ C). Note that it takes about 15 s for the temperature of the diaphragm to reach this steady-state maximum value. If the working fluid could stand higher temperatures, larger voltages may be applied to achieve
Fig. 12. Measured relationships between backpressure and flow rate with various applied voltages.
92
B.T. Chia et al. / Sensors and Actuators A 165 (2011) 86–93
Table 3 Pumping performances of the proposed and conventional micropumps. Devices
Pumping performance Applied voltage
Maximum flow rate
Operating frequency
Cut-off backpressure
Proposed design
5V 7V 9V
8.74 L/min 13.21 L/min 20.01 L/min
0.3 Hz 0.9 Hz 1.2 Hz
245 Pa 343 Pa 490 Pa
Conventional design
5V
9.18 L/min
1.2 Hz
400 Pa
References
Fig. 13. Maximum temperatures measured at Point-Z vs. operating frequency with various applied voltages.
even higher flow rate. When the applied voltage is 9 V and the driving frequency is 1.2 Hz, the proposed micropump generates a maximum flow rate of 20.01 L/min while the temperature on the diaphragm is still below 36.2 ◦ C. This temperature is still appropriate for many devices used in biomedical applications.
5. Conclusion In this paper, a novel thermo-pneumatic peristaltic micropump with low temperature elevation on the working fluid is developed. The proposed device has three actuation chambers, each of which consists of individual zones for air-heating and fluidsqueezing. Therefore, the working fluid will not be heated directly during operation, which gives tremendous flexibility in biomedical applications. The proposed micropump shows similar performance when compared with the conventional design under the same operational conditions. Also, it is demonstrated that increasing applied voltage can improve the flow rate, while only gives very mild temperature elevation on the working fluid.
Acknowledgements This research is partially sponsored by the Ministry of Education and the National Science Council, Taiwan, ROC (contract number: NSC-97-3114-E-02-007).
[1] C. Zang, D. Xing, Y. Li, Micropumps, microvalves, and micromixers within PCR microfluidic chips: advances and trends, Biotechnol. Adv. 25 (2007) 483–514. [2] N.T. Nguyen, X.Y. Huang, T.K. Chuan, MEMS-micropumps: a review, J. Fluids Eng. 124 (2002) 384–392. [3] D.J. Laser, J.G. Santiago, A review of micropumps, J. Micromech. Microeng. 14 (2004) R35–R64. [4] N.J. Graf, M.T. Bowser, A soft-polymer piezoelectric bimorph cantileveractuated peristaltic micropump, Lab Chip 8 (2008) 1664–1670. [5] M. Bu, T. Melvin, G. Ensell, J.S. Wilkinson, A.G.R. Evans, Design and theoretical evaluation of a novel microfluidic device to be used for PCR, J. Micromech. Microeng. 13 (2003) S125–S130. [6] D.-S. Lee, H.C. Yoon, J.S. Ko, Fabrication and characterization of a bidirectional valveless peristaltic micropump and its application to a flow-type immunoanalysis, Sens. Actuat. B 103 (2004) 409–415. [7] L.-S. Jang, K. Shu, Y.-C. Yu, Y.-J. Li, C.-H. Chen, Effect of actuation sequence on flow rates of peristaltic micropumps with PZT actuators, Biomed. Microdevices 11 (2009) 173–181. [8] M. Zhu, P. Kirby, M. Wacklerle, M. Herz, M. Richter, Optimization design of multi-material micropump using finite element method, Sens. Actuat. A 149 (2009) 130–135. [9] S.B. Huang, M.H. Wu, Z. Cui, G.B. Lee, A membrane-based serpentine-shape pneumatic micropump with pumping performance modulated by fluidic resistance, J. Micromech. Microeng. 18 (2008) 045008. [10] M.A. Unger, H.-P. Chou, T. Thorsen, A. Scherer, S.R. Quake, Monolithic microfabricated valves and pumps by multilater soft lithography, Science 288 (2000) 113–116. [11] P.A. Willis, B.D. Hunt, V.E. White, M.C. Lee, M. Ikeda, S. Bae, M.J. Pelletier, F.J. Grunthaner, Monolithic Teflon® membrane valves and pumps for harsh chemical and low-temperature use, Lab Chip 7 (2007) 1469–1474. [12] W.H. Grover, M.G. von Muhlen, S.R. Manalis, Teflon films for chemically-inert microfluidic valves and pumps, Lab Chip 8 (2008) 913–918. [13] F. Goldschmidtböing, A. Doll, M. Heinrichs, P. Woias, H.J. Schrag, U.T. Hopt, A generic analytical model for micro-diaphragm pumps with active valves, J. Micromech. Microeng. 15 (2005) 673–683. [14] W. Inman, K. Domansky, J. Serdy, B. Owens, D. Trumperm, L.G. Griffith, Design, modeling and fabrication of a constant flow pneumatic micropump, J. Micromech. Microeng. 17 (2007) 891–899. [15] M.M. Teymoori, E. Abbaspour-Sani, Design and simulation of a novel electrostatic peristaltic micromachined pump for drug delivery applications, Sens. Actuat. A 117 (2005) 222–229. [16] J. Xie, J. Shih, Q. Lin, B. Yang, Y.-C. Tai, Surface micromachined electrostatically actuated micro peristaltic pump, Lab Chip 4 (2004) 495–501. [17] T. Pan, E. Kai, M. Stay, V. Barocas, B. Ziaie, A magnetically driven PDMS peristaltic micropump, in: Proceedings of the 26th Annual International Conference of the IEEE EMBS, San Francisco, CA, USA, 2004, pp. 2639–2642. [18] L. Yobas, K.-C. Tang, S.-E. Yong, E.K.-Z. Ong, A disposable planar peristaltic pump for lab-on-a-chip, Lab Chip 8 (2008) 660–662. [19] O.C. Jeong, S.W. Park, S.S. Yang, J.J. Pak, Fabrication of a peristaltic PDMS micropump, Sens. Actuat. A 123–124 (2005) 453–458. [20] Y.-J. Yang, H.-H. Liao, Development and characterization of thermopneumatic peristaltic micropumps, J. Micromech. Microeng. 19 (2009) 025003. [21] Y. Xia, G.M. Whiteside, Soft lithography, Annu. Rev. Mater. Sci. 28 (1998) 153–184. [22] F. Amirouche, Y. Zhou, T. Johnson, Current micropump technologies and their biomedical applications, Microsyst. Technol. 15 (2009) 647–666.
Biographies Bonnie Tingting Chia received her BS degree in Mechanical Engineering at National Taiwan University, Taiwan, in 2006. She is now pursuing the PhD degree in Mechanical Engineering at National Taiwan University, Taiwan. She has been working on microfluidic system for biomedical applications. Hsin-Hung Liao received the BS degrees in Mechanical Engineering from National Chung Cheng University, Chiayi, Taiwan, in 2003. He is now pursuing the PhD degree in Mechanical Engineering at National Taiwan University, Taiwan. His research interests include optical measurement, micropump, drug delivery system and microelectromechanical digital-to-analog converter (MEMS-DAC) device.
B.T. Chia et al. / Sensors and Actuators A 165 (2011) 86–93 Yao-Joe Joseph Yang received his MS and PhD degrees in Electrical Engineering from the Massachusetts Institute of Technology (MIT) in 1997 and 1999, respectively. Also, he received his MS degree (1995) from UCLA and his BS degree (1990) from the National Taiwan University, both in Mechanical Engineering. From 1999 to 2000, he joined the Coventor Inc. (Cambridge, MA) as a senior application engineer. Since 2000, he joined the Department of Mechanical Engineering at the National Taiwan University, Taipei, Taiwan. Currently he is a professor, and serves as the associate chair of the department. He is also the director for CAD Technology in the NTU NEMS
93
Center. From 2005 to 2009, he serves as the deputy Secretary General of the Chinese Institute of Automation Engineering (CIAE). Currently he is the board member of the Institute. His research interests include microelectromechanical systems, nanotechnology, high-precision micromachining, flexible sensing arrays, sensor network, parallel processing, and semiconductor devices and vacuum microelectronics modeling. He has been consulted by more than three U.S-based companies and four Taiwan-based organizations. Dr. Yang is a member of IEEE. He is also the recipient of the Outstanding Young Researcher Award of the National Science Council.