Tsinghua Science and Technology Tsinghua University, China ISSN: 1007-0212 Vol. 6, Num. 3, 2001, pp. 277-280

Tsinghua Science and Technology, Vol. 6, No. 3, August 2001 pp. 277-280

Fluidic Control System Employing Micropump and Microvalves for Cell Isolation*

ZHANG Jian ¹ , JING Gaoshan ¹ , XU Junquan ¹ , CHENG Jing ^1,2 , ZHOU Yuxiang ^1,2

1. Beijing National Biochip Research and Engineering Center; Jia 2# Qinghua West Road, Beijing 100084, China;
2. State Key Laboratory of Biomembrane and Membrane Biotechnology, School of Life Science and Engineering,  Tsinghua University, Beijing 100084, China

* Supported  by the National Key Project (No.39889001); the National Outstanding Young Scientist Fund (No.39825108) from the National Natural Science Foundation of China; and the National “863” High Technology Program (No.103-13-05-02)

Received: 2000-11-20

Code Number: ts01082

Abstract:   

A micro fluidic system was developed and used for cell isolation on biochips. The flow speed could be precisely adjusted by a micropump, and the flow direction could be controlled by switching valves. The complete operation was computer-controlled. The control system for the micropump and microvalves is discussed here. This control system can be combined with other biochip devices to construct a micro total analysis system (mTAS).

Key  words:  microfluidics; micropump; microvalve; micro total analysis system (mTAS)

Introduction   

Many chip-based devices (e.g., micropumps[1-7]  and microvalves[8-15] ) have been developed after numerous efforts in the past decades. Developments in microfabrication technology have greatly accelerated the progress towards a micro total analysis system (mTAS) in which microfluidics plays a significant role including reagent mixing, sample isolation, and the control of fluidic flow rate and direction. The fluidic control system described in this paper uses a miniaturized off-chip micropump and several microvalves for cell separation and isolation because they are easily applicable and reliable. The fluid flow rate was controlled by the pump and the flow direction by valves. A data acquisition (DAQ) board and LabVIEW software were utilized with an electronic circuit designed to interface with the board to generate control signals applied to the pump and valves.

The system was used for isolating blood cells and blue alga cells on a dielectrophoretic chip. The theory for cell separation by dielectrophoresis is well established and will not be discussed here. Examples of cell separation using dielectrophoresis have been given previously[16-20].

1 Material and Methods  

1.1 Valving and pumping

The valves and pump were purchased from the Lee Company (Westbrook, Connecticut, U.S.A.). Two types of valves, i.e., LFVA 1210120H a 2-way valve and LHDA 1221111H, a 3-way valve are used in the system. They both have low internal volume and zero dead volume, and can stand high pressure (LFVA 1210120H, 2.1 x10⁵ Pa; LHDA 1221111H, 1.1 x 10⁵Pa). The LFVA valve is normally closed, and changes to open when the actuating signal (12 V/1.5 W) is applied. The 3-way LHDA valve has one common port, one normally closed port and one normally open port. When the switching signal (12 V/550 mW) is applied, the closed and the open ports are reversed.

The pump is model LPVX 0502600BB with a full stroke volume of 50 mL. This pump is driven by an internal stepper motor controlled by digital pulses. The motor rotates 7.5° per digital pulse allowing the pump to input or output fluid with a fixed volume of 0.1 mL. The pumping rate is controlled by the frequency of the applied digital pulses.

The pump input or output is controlled by two LFVA valves connected to the pump as shown in Fig.1. Only one valve is open at a time for the pumping assembly to work. The pumping assembly cannot input or output continuously since at the end of each full stroke the piston must return before starting the next cycle.

1.2 Control circuit

The pump and valves are controlled using a DAQ board by the National Instruments (Austin, Texas, U.S.A.). Its main parameters are listed in Table 1.

The  DAQ board uses two built-in timers to generate a finite pulse train, which is split into two pulse trains using the method shown in Fig.2. The second pulse train is produced by phase inversion of the pulse sequence generated directly from the DAQ board, which results in a 180° phase shift between the second sequence and the original one. Frequency division changes the phase shift to 90° followed by level adjustment and power amplification. The two pulse trains obtained using this method can then be used to drive the stepper motor.

A  user application program was developed using LabVIEW. Figure 3 shows the user interface which can be used to select one of two working modes: continuous mode or fixed volume mode. In continuous mode, the pump will keep on running until the user stops it. In fixed volume mode the pump input/output a pre-determined volume of fluid followed by a break and then input/output again. The interval between adjacent cycles, the number of cycles and the volume per cycle can be selected by the user. The two modes are suitable for different requirements. If the fluid is to be pumped to a destination regardless of volume, the first mode is preferred. The cell separation study presented here uses the fixed volume mode. For a single run, a fixed volume of liquid sample is introduced into the flow chamber (usually equal to the flow chamber volume) then the pump stops. During the pauses, the suspended cells migrate to different positions on the electronic chip due to the positive and negative dielectrophoretic forces. Once the target cells are captured the pump is started again and the next batch is introduced into the flow chamber. The first mode can be used for a continuous cell separation process if only the flow rate delivered is slow enough to guarantee that the positive dielectrophoretic forces are able to dynamically capture the cells before they flows out of the flow chamber.

The DAQ board has 8 digital I/O ports. However, with 4 CD4099B (a CMOS 8-bit addressable latch) chips a total of 32 valves can be controlled (one pump uses 2 valves). The 74HC138 chip is used for addressing and 5 TDA2003A chips for driving the valves.

1.3 Fluidic system set-up

The cell isolation fluidic system is shown in Fig.4. A sample containing two types of cells can be added from the reservoir. Both the undesired cells and the target cells can be introduced into different containers by switching valves. The pump flow rate can be controlled from 50 mL/min to 1 mL/min.In the current set-up, the sample collector and the waste collector are placed between the pump and the flow chamber. Therefore, no liquid passes through the pump and thus avoiding contamination of the pump.

1.4 Sample preparation

A fresh blood sample (500 mL) was added to an Eppendorf tube containing EDTA for anticoagulation. Blue alga cells were introduced into 500 mL of cell separation buffer (0.05 x TBE,  8.5%  W/V sucrose) then the tube was centrifuged at 500 g for 2 min. After removing the supernatant the cell pellet was resuspended in 500 mL of cell separation buffer. Afterwards, fresh blood sample (20 mL) was added to the above blue alga cell suspension and mixed gently.

1.5 Cell separation

The flow chamber was cleaned using ethanol added into the reservoir and pumped into the flow chamber followed by the cell separation buffer. A mixture of blood cells and alga cells was then introduced into the flow chamber. The cell mixture flow was monitored through a CCD camera. The pump was switched off once the flow chamber was full of the sample. A sinusoidal signal (10 V, peak to peak) was applied to the electrodes with the frequency carefully adjusted starting from 10 kHz. At the frequency of 70 kHz alga cells migrated to the electrodes with the maximum field and the blood cells migrated to the minimum field. In a short time the two types of cells were completely separated. The pump was then switched on at low speed (100 mL/min) to wash away the blood cells in the minima field. When the mixture was nearly drained from the sample/buffer reservoir, separation buffer was added with the electronic signal still on. After the mixed sample was pumped through the chamber the separation signal was switched off. The separation buffer was then drained and the blue alga cells were allowed to flow into the target cell-collecting container by switching the valves.

2 Results and Discussion  

Figure 5 shows the flow in the chamber as the sample mixture was passing through the biochip. The blue alga cells were captured around the edge of the electrodes whereas the blood cells were driven to areas where the field was minimized andeventually washed out. Figure 6 shows the flow in the chamber after the blood cells were washed away leaving only the blue alga cells retained on the chip.

The indicated flow rate refers to the rate while the pump was sucking fluid without calculating the discharge time.

For cell separation and isolation the flow chamber volume was not a constant so the indicated flow rate makes sense only in this experiment. In other experiments flow rate should be carefully analyzed.

While  flushing the blue alga cells from the dielectrophoresis chip, some alga cells were found adhering to the surface of the chip which could not be removed by the separation buffer. To overcome this problem, the electronic signal was kept on during flushing with the frequency adjusted to subject the alga cells to a negative dielectrophoretic force. The negative dielectrophoretic force counteracted the adhesion force allowing the cells to be washed away.

The electric circuit can simultaneously drive a maximum of 30 microvalves, so the fluidic system can be extended for other biochemical reactions or detection on a biochip with additional reagents and flow control in more complicated fluidic channels.

Acknowledgements

The blue alga cells were kindly provided by Dr. Wu Qingyu of the Department of Biological Sciences and Biotechnology, Tsinghua University.HT

References

1. Seiler K, Fan Z H, Fluri K, et al. Electroosmatic pumping and valveless control of fluid flow within a manifold of capillaries on a glass chip. Analytical Chemistry, 1994, 66: 3485-3491.
2. Schasfoort R B M, Schlautmann S, Hendrikse L, et al. Field-effect flow control for microfabricated fluidic networks. Science, 1999, 286: 942-945.
3. McBride S E, Moroney R M, Chiang W. Electrohydrodynamic pumps for high-density microfluidic arrays. In:  Proceedings of the mTAS'98 Workshop, 1998. 45-48.
4. Benard W L, Kahn H, Heuer A H, et al. Thin-film shape-memory alloy actuated micropumps. Journal of Microelectromechanical Systems, 1998, 7:   245-251 .
5. Luginbuhl P, Collins S D, Racine G A, et al. Microfabricated lamb wave device based on PZT sol-gel thin film for mechanical transport of solid particles and liquids. Journal of Microelectromechanical Systems, 1997, 6:   337-346 .
6. Grunze M. Surface science: driven liquids. Science, 1999, 283:  41CD*242.
7. Duffy D C, Gillis H L, Lin J, et al. Microfabricated centrifugal microfluidic systems: characterization and multiple enzymatic assays. Analytical Chemistry, 1999, 71:  4669-4678.
8. Hartshorne H, Ning Y B, Lee W E, Backhouse C. Development of microfabricated valves for mTAS. In:  Proceedings of the mTAS'98 Workshop, 1998. 379-381.

Copyright 2001 - Tsinghua Science and Technology

The following images related to this document are available:

Photo images