Chapter 3 Fabrication and Experimental
3.1 UV Lithography of SU-8
3.1.7 PDMS Casting
Figure 3.5 Supersonic cleaner
Figure 3.6 Schematic illustrate of PDMS casting (a) Infusion PDMS (b) Make PDMS mold (c) Lift off PDMS (d) O2 plasma
3.1.7 PDMS Casting
To construct complex 3-D structure made of PDMS membrane shown in figure 3.6, microfabrication technique for stacking multi layers of SU-8 mold was used. A curing agent and PDMS prepolymer are mixed in a 1:10 weight ratio. Degassing was done after mixing in the mechanical vacuum chamber for enough time to remove any air bubbles by vacuum device (Figure 3.7) in the mixture. The mixture was poured onto master mold made of SU-8 layers, and excessive PDMS was removed using razor knife. The PDMS premixture was cured for hours at 80~100˚C. Temperature was ramped up to set point and cooled
32
down to room temperature again.
Figure 3.7 Vacuum device for remove the bubble from mixture of PDMS 3.1.8 Chip Bonding
Plasma ashing is a standard procedure in semiconductor manufacture, used to remove photoresist from substrate wafer. Oxygen is introduced to the vacuum process chamber as the reactive gas. The plasma is formed by exposing the reactive gas to electromagnetic radiation such as microwave or radio frequency. Free radicals generated in the plasma state react with the photoresist and substrate material. In addition to using the oxygen plasma to clean the surface (stripping photoresist), we also use the system to facilitate the PDMS bonding process.
The process usually takes 5~10 minutes to achieve permanent irreversible bonding. Because the bonding takes effect immediately, it is suitable for quick fabrication of devices. It is less than suitable if the device requires careful alignment of multiple layers. Finally, the PDMS structures and the glass substrate are bonded together by utilizing an oxygen plasma treatment (Figure 3.8) to form the complete microfluidic chip.
33
Figure 3.8 O2 plasma machine
3.2 Fabrication processes of proposed micropump
Figure 3.9 shows a schematic representation of the microfabrication process by a CNC machining process and a PDMS replication process. Master molds with microstructures on polymethylmethacrylate (PMMA) plates are first formed by using a CNC machine (EGX-400, Roland Inc., Japan) equipped with a 0.5 mm drill bit (Figure 3.9(a)). The rotational speed and feed rate of the drill bit are 27000 rpm and 15 mm/min, respectively. The high rotational speed and low feed rate forms a smooth surface on PMMA. It is then followed by a PDMS casting process (Figure 3.9(b)) to form inverse images of the air chamber mold and the fluidic channel mold (Figure 3.9(c)). Note that another PMMA plate is place on top of the PDMS such that a flat surface can be formed. Finally, the PDMS structures and the glass substrate are bonded together by utilizing an oxygen plasma treatment to form the complete microfluidic reaction chip (Figure 3.9(d)). Figure 3.10(a) and 3.10(b) are photographs of the PMMA molds for the air chambers and the fluidic channels. The dimensions of the micropump chip are measured to be 6 mm in length, 2 mm in width, 1.0 mm in depth and 0.5mm in thickness, respectively.
34
Figure 3.9 Fabrication processes of micropump
Figures 3.10 and 3.11 showed the (a) top-layer and (b) bottom-layer photographs of master mold and PDMS replication of the proposed micropump. Figure 3.12 shows the (a) top-side view and (b) up-side down view photograph of completed micropump.
35
(a)
(b)
Figure 3.10 (a) Top-layer and (b) bottom-layer photographs of master mold for the proposed micropump.
36
(a)
(b)
Figure 3.11 (a) Top-layer (b) bottom-layer photographs of PDMS replication for the proposed micropump after fabrication processing.
37
(a)
(b)
Figure 3.12 (a) Top-side view and (b) up-side down view photographs of completed micropump.
3.3 Experimental setup
Figure 3.13(a) provides the illustration of the control system developed for the current micropump. The system comprises an air compressor (JUN-AIR Inc., MDR2-1A/11, Japan), which supplies compressed air to the micropump, a functional control circuit, and three EMVs
38
(SMC Inc., S070M-5BG-32, Taiwan) to control the pneumatic pump. Tests were performed to explore the pumping rates of the operational frequency, pneumatic driving pressure, and operation modes of the current pneumatic micropump. A constant current was supplied to the microflow sensor throughout the tests, and the electrical signal output from the sensor was recorded by an ADC (analog-to-digital converter, ATMEL Corp., ATMEGA8535, USA) connected to a personal computer. The relationship between the flow rate and the output voltage of the microflow sensor was calibrated using a syringe pump (KDScientific, KDS 100, USA). During the tests, the current pneumatic micropump was positioned under an optical microscope (Olympus, BH2-UMA, Japan) and the motion of the fluid in the microchannel observed and recorded using an image capturing system (Camdio, CA-8M3N, Japan) (Figure 3.5(b)).
(a) Microprocess
Controller EMV
Air compressor Power supply
Micropump
Control System
Detection
39
(b)
Figure 3.13 (a) Illustration of the controlled system for the proposed micropump. (b) Photograph of the experiment setup for the measured observations.
Micropump
Controller
Air compressor Detection
40
Chapter 4
Results and discussion
4.1 Membrane deflection simulation
This study aimed to investigate the transporting performance of the developed micro-pump by establishing the optimum design by utilizing the active check-valve structures activated pneumatically to transport the bio-sample. A numerical simulation was applied on the design of micro-pumps and investigation of deformation of PDMS membrane. The deformation was simulated numerically using a commercial code (CFD-ACE+, CFD-RC, USA). In the simulations, the moving boundary condition was calculated by using the stress module as well as the deformation module. The moving boundary of the membrane was discretely separated to allow smooth motion. The deformed grid of the moving boundary was accomplished using the auto-remesh function in the deformation process. Dense grids were used in the regions of the moving boundary, where the deformation was induced by the buried side chamber. The material properties described as: The density (ρ)’ Young’s module (E) ,and Poisson’s ratio (ν) of PDMS are 970 kg/m3, 740 kPa, and 0.5, respectively.
In order to simulate the micropump, 3-D numerical domains were discretized into approximately 200,000 cells (100×100×20) with structured hexahedral meshes. Note that non-uniform mesh grids were used for numerical simulations. A stringent residuals criterion (less than 10−8) and nonlinear stress residuals criterion (less then 10-4) between each iterative solution step was used to guarantee the convergence of the solution. The boundary conditions of numerical simulation are depicted in figure 4.1(a).
41
Figures 4.1(b)~(e) show that applied on the primary PDMS moving membrane at the external pressures of 5psi (34.4kPa), 10psi (69.0kPa), 15psi (103.4kPa), and 20 psi (137.8 kPa), respectively. As we expected, the membrane deflection increases with the increasing air pressure. A maximum deflection of membrane of 0.9 mm is obtained at an applied air pressure of 10 psi (69.0kPa). This value can be increased up when the external air pressures are to be increased. This numerical calculation is reasonable agreement with the experimental observations since an applied pressure higher to 10 psi (68.0 kPa) can push the PDMS moving membrane to completely touch down to the glass substrate.
Figure 4.2 shows the active check valve lifts up by the moving membrane for the different air pressures. As we expected, the lift-up performance of active check valve increase with the increasing air pressure. In the design principle of the proposed micropump, the higher lift-up performance will obtain the higher pumping rate. Conversely, the lower lift-up performance will obtain the lower pumping rate.
(a)
uniform load of air pressure on the top displacements of x,
y and z are 0
free surface on the bottom
displacements of x, y and z are 0
displacements of x, y and z are 0
displacements of x, y and z are 0
42
(b)
(c)
(d)
43
(e)
Figure 4.1 (a) Boundary conditions of numerical simulation and deflection of the PDMS membrane with the different air pressures of (b) 5psi, (c) 10psi, (d) 15psi, (e) 20 psi, respectively.
(a)
44
(b)
(c)
(d)
Figure 4.2 Deflection of the PDMS membrane with an active check valve with the different air pressures of (b) 5psi, (c) 10psi, (d) 15psi, (e) 20 psi, respectively.
4.2. Pumping effect
A major contribution of the current study is to design a new pneumatic micropump using an active check valve, which still provides reasonable pumping performance, even at a smaller scale. Using an active check valve makes the fluid-driven system more compact when
45
compared to the previous peristaltic micropump. There are many factors affecting the flow rate of the micropump including the driving frequency, the applied air pressure and the location of the active check-valve. For all of the experiments, the primary moving membrane is set to be 500μm for the thickness of the membrane, 2000μm for the width, 6000μm for the length and 1000 μm for the channel depth.
4.2.1 Operation of external pressure
The effect of the applied air pressure affected the flow rate at different driving frequency is explored and the results are shown in figure 4.3. During the test process, the EMV operating frequency is varied sequentially to test the effect of the driving frequency. In general, the flow rate increases with increasing driving frequency (at a constant pneumatic pressure). However, the maximum flow rate at a constant applied pressure is limited by the fill and release times of the compressed air. If the operating frequency is too higher, the air chamber cannot be completely filled and released, and the flow rate will not increase, but it will start to fall, as shown in figure 4.3. Specifically, the optimum operational frequencies at air pressures of 5psi (34.4kPa), 10psi (69.0kPa) and 20psi (136.1kPa) are found to be 30, 26 and 24 Hz, respectively. Experimental data also show that the higher driving pressures, the higher flow rates for the applied pressures of 5 psi (34.4kPa), 10 psi (69.0 kPa), and 20 psi (138kPa). A maximum flow rate of 532μl/min is obtained at a driving frequency of 24 Hz and at an applied air pressure of 20 psi (136.1kPa). This observation is reasonable since an applied pressure higher than 10 psi (69.0 kPa) can push the PDMS membrane to completely touch down to the glass substrate. It can also explain the same flow rates appear at the pressure of 10 psi and 20 psi. In this study, the applied pressure of 20 psi can be used for the propose of valve to block the fluids to pass through the microchannel.
46
Frequency (Hz)
P ump in g rate ( u L /min )
0 10 20 30 40
0 100 200 300 400 500 600
5 (psi) 10 (psi) 20 (psi)
Figure 4.3 Relationship of the pumping rate and the applied air pressure at different driving frequencies.
4.2.2 Locations of active check-valve
The effect of the different location of active check valve affected the pumping rate was also tested. The distance from active check valve to outlet of channel with 1.8 (Case 1), 2.0 (case 2) and 2.2mm (case 3), respectively, were tested with a constant pneumatic pressure of 10 psi. Figure 4.4 plots the relationship between the pumping rate and the operational frequency as a function of the different location of active check valve. It can be seen that the pumping rates first increases with increasing frequency, certainly, that also increase with an increasing distance of active check valve. However, the optimum operational frequency, f, i.e.
the frequency corresponding to the maximum pumping rate, reduces as the distances of active check valve increase. Specifically, the optimum operational frequencies for the channels with
47
distance of check valve 1.8, 2.0 and 2.2mm are found to be 26, 24 and 26 Hz, respectively.
Conversely, the maximum pumping rate increases with an increasing distance of active check valve. Apparently, the maximum pumping rates obtained for the distance of check valve 1.8, 2.0 and 2.2mm are 296, 373 and 488 μL/mm, respectively. This observation is reasonable since the peristaltic driving effect is enhanced when the distance of active check valve increases. However, the time required to increase for the compressed air to completely release with the increasing distance of active check valve. Therefore, the optimum driving frequency reduces with the increasing distance of active check valve.
The relationship between the pumping rate and the different locations of active check valve are also experimentally explored. Figure 4.5 shows the maximum flow rates under the optimal driving frequency (26 Hz) for different locations of valve of channel, i.e. the distance from active check valve to the outlet of reservoir. Again, the membrane thickness is 500μm and the applied air pressure is 10 psi (69.0kPa). The results indicate that the micropump can be operated for various location of active check valve. In general, the higher distance of active check-valve induces higher pumping rates since more stroke volume is squeezed by the defected PDMS membranes. However, the maximum flow rate is experimentally obtained at a distance of 600μm. The flow rates will increase with the increasing distance of active check-valve. However, the ones will decrease when the distance is higher than 600μm. A smaller distance of active check-valve obtained the lower pumping rates of the proposed micropump.
Conversely, the longer distance of active check valve obtained higher pumping rates. The reason is that the active check valve lifted up by the moving structure deflected. The smaller distance of active check valve caused a smaller lift to let fluids pass through the microchannel.
Certainly, the longer distance will interrupt the deflection of moving structure. Therefore, it exists an optimum distance of active check valve to achieve a higher pumping rate. The data
48
in figure 4.5 are fitted to a curve line, which can be found the optimum location of active check valve to be 600μm (i.e, the distance is 2.2 mm) for this study.
Frequency (Hz)
P ump in ra te (u L/min )
0 10 20 30 40
0 100 200 300 400
500 Case 1
Case 2 Case 3
Figure 4.4 Relationship between pumping rate and operational frequency as a function of different location of active check valve for a pneumatic pressure of 10 psi.
49
Distance ( m)
P umpin g rate (uL /m in)
0 250 500 750 1000
0 100 200 300 400 500
μ
Frequecy=26 (Hz)
Figure 4.5 The pumping rates of the micropumps with different locations of active check valve of channels under the optimal driving frequency of 26 Hz.
50
Chapter 5 Conclusions
This study has proposed a novel pneumatic pump which is based on the peristaltic-like actuation effect caused by the sequential deflection of membrane with an active check valve.
The pneumatic micropump provides improved pumping rates and can be controlled using a EMV. By using conventional MEMS techniques, the microfluidic mold was produced from negative thick-film photoresist (SU-8), while the microchannels and membrane structures were fabricated from elastic polymer PDMS.
The performance of the pneumatic pump has been thoroughly investigated for the various effects. It has been shown that the pumping rate is determined by the pneumatic pressure, the operational frequency, the location of active check valve. The optimum parameters of air pressure, driving frequency, and location position of active check valve are experimentally 10 psi, 26 Hz, and 600 μm, respectively. The maximum pumping rate was found to be 532μL/min. Furthermore, the proposed micro-pump could be crucial for the transportation of the microfluidics and could be incorporated with the micro-total-analysis-systems.
References
51
[1] Laser, D. J. and Santiago, J. G., “A review of micropumps,” J. Micromech. Microeng., 14 35–64, (2004).
[2] Fu, A.Y., Chou, H.P., Spence. C., Frances, H.A., and Stephen R.Q., “An Integrated Microfabricated Cell Sorter,” Anal. Chem., 74, 2451-2457, (2002).
[3] Fu, A.Y., Spence, C., Scherer. A., Frances H.A., and Stephen R.Q., “A microfabricated fluorescence-activated cell sorter,” NATURE BIOTECHNOLOGY VOL 17 NOVEMBER 1999, 1109-1111.
[4] Maillefer, D., Gamper, S., Frehner, B., Balmer, P., “A high-performance silicon micropump for disposable drug delivery systems,” IEEE, 413-417, (2001).
[5] Hong, J. W., Studer, V., Hang, G., Anderson, W.F., Quake, S., “A nanoliter-scale nucleic acid processor with parallel architecture,” Nature Biotechnology, 22(4), 435-439, (2004).
[6] Liu, J., Enzelberger, M., Quake, S., “A nanoliter rotary device for polymerase chain reaction,” Electrophoresis, 23, 1531–1536, (2002).
[7] Reichmuth, D.S., Chirica, G.S., Kirby, B.J., “Increasing the performance of high-pressure, high-efficiency electrokinetic micropumps using zwitterionic solute additives,” Sensors and Actuators B., 92, 37–43, (2003).
[8] Studer, V., Pe´pin, A., Chen, Y. and Ajdari, A., “An integrated AC electrokinetic pump in a microfluidic loop for fast and tunable flow control,” Analyst, 129, 944–949, (2004).
[9] S. Zehg, C.H., Chen, J. C., Mikkelsen, J.G., Santiago, “Fabrication and Characterization of electroosmotic micropump,” Sensors and Actuators B., 79, 107–114, (2001).
[10] Zeng, S., Chen, C.H., Juan, G.S., Chen, J.R., Richard, N.Z., Jennifer, A.T., Frantisek, S., Jean, M.J.F, “Electroosmotic flow pumps with polymer frits,” Sensors and Actuators B., 82, 209–212, (2002).
[11] Nguyen, N.T. and White, R.M., “Design and optimization of an ultrasonic flexural plate wave micropump using numerical simulation” Sensors Actuators., 77, 229–236, (1999).
[12] Zhong, J.H., Yi, M.Q. and Bau, H.H.,” Magneto hydrodynamic (MHD) pump fabricated with ceramic tapes,” Sensors Actuators A., 96, 59–66, (2002).
[13] Vestad, T., Marr, D.W.M. and Oakey, J., “Flow control for capillary-pumped microfluidic systems,” J. Micromech. Microeng., 14, 1503–1506, (2004).
[14] Lee, J., and Kim, C.J., “Surface-tension-driven microactuation based on continuous electrowetting” J. Microelectromech. Syst., 9, 171–80, (2000).
52
[15] Lee, J., Moon, H., Fowler, J., Schoellhammer, T. and Kim, C.J., “Electrowetting and electrowetting-on-dielectric for microscale liquid handling” Sensors Actuators., 95, 259-68, (2002).
[16] Bourouina, T., Bosseboeuf, A. and Grandchamp, J. P., “Design and simulation of an electrostatic micropump for drug-delivery applications” J. Micromech. Microeng., 7, 1868, (1997).
[17] Saif, M.T.A., Alaca, B.E., and Sehitoglu, H., “Analytical Modeling of Electrostatic Membrane Actuator for Micro Pumps,” IEEE JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 8, NO. 3, SEPTEMBER, 335-345, (1999).
[18] Francais, O., Dufour, I. and Sarraute, E., “Analytical static modelling and optimization of electrostatic micropumps” J. Micromech. Microeng., 7, 183–5, (1997).
[19] Gong, Q.L., Zhou, Z.Y., Yang, Y.H. and Wang, X.H., “Design, optimization and simulation on microelectromagnetic pump” Sensors Actuators A., 83, 200–7, (2000).
[20] Capanu, M., Boyd, J.G. and Hesketh, P.J., “Design; fabrication; and testing of a bistable electromagnetically actuated microvalve” J. Microelectromech. Syst., 9, 181–9, (2000).
[21] Husband, B., Bu, M., Evans, A. G. R. and Melvin, T., “Investigation for the operation of an integrated peristaltic micropump,” J. Micromech. Microeng., 14 ,64–69, (2004).
[22] Lin, Q., Yang, B., Xie. J. and Tai, Y.C., “Dynamic simulation of a peristaltic micropump considering coupled fluid flow and structural motion,” J. Micromech.
Microeng., 17, 220–228, (2007).
[23] Jeong, O.C. and Konishi, S., “Fabrication and drive test of pneumatic PDMS micro pump,” Sensors and Actuators A., (2006).
[24] Unger, M.A., Chou, H.P., Thorsen, T., Scherer, A., Quake, S.R., “Monolithic Microfabricated Valves and Pumps by Multilayer Soft Lithography,” SCIENCE, 288, 113–116, (2000).
[25] Thorsen, T., Maerkl, S.J., Quake, S.R., “Microfluidic Large-Scale Integration,”
SCIENCE, 298, 580–584, (2002).
[26] Chou, H.P., Unger. M.A., Quake, S.R., ”A microfabricated Rotary Pump,”
Biomedical Microdevices., 3:4, 323–330, (2001)
[27] Jeong, O.C. and Yang, S.S., “Fabrication and test of a thermopneumatic micropump
53
with a corrugated p+ diaphragm” Sensors and Actuators A: Physical, 83, 249-255, (2000).
[28] Van, D.P.F.C.M., Van, L.H.T.G., Elwenspoek, M., and Fluitman, J.H.J.,
“Therpneumatic micropump based on microengineering techniques” Sensors and Actuators A: Physical, 21, 198-202, (1990).
[29] Spencer, J.G. “Piezoelectric micropump with three valves working peristaltically”
Sensor and Actuators A: Physical, 21-23, 203-206, (1990).
[30] Morris, C.J. and Forster, F.K., “Optimization of a circular piezoelectric bimorph for a micropump driver” J. Micromech. Microeng., 10, 459–65, (2000).
[31] Wang, C.H. and Lee, G.B., “Automatic bio-sampling chips integrated with micropumps and microvalves for multiple disease detection” Biosensors and Bioelectronics, 21(3), 419-425, (2005).
[32] Tseng, H.Y., Wang, C.H., Lin, W.Y. and Lee, G.B., “Membrane-activated microfluidic rotary devices for pumping and mixing,” Biomedical Microdevices., 9(4), 545-554, (2007).
[33] Wang, C.H. and Lee, G.B., “Pneumatically-driven Peristaltic Micropumps Utilizing Serpentine-shape Channels,” J. Micromech. Microeng., 16, 341-348, (2006).
[34] Timoshenko, S. and Krieger, S.W., “Theory of plates and shells,” second Edition, 415-420, (1959).