LANL201684
LANL_040_004
Techset Composition Ltd, Salisbury, U.K.
1/19/2007
Analytical Letters, 40: 749–764, 2007 Copyright # Taylor & Francis Group, LLC ISSN 0003-2719 print/1532-236X online DOI: 10.1080/00032710601017896
NANOTECHNOLOGY
Design and Fabrication of an Automated Microchip-Based Cell Separation Device Chengjun Huang and Jun Yu Department of Electronic Science and Technology, Huazhong University of Science and Technology, Wuhan, China and National Engineering Research Center for Beijing Biochip Technology (NERCBBT), Beijing, China
Jiang Zhu and Lei Wang Medical Systems Biology Research Center, Tsinghua University, School of Medicine, Beijing, China and National Engineering Research Center for Beijing Biochip Technology (NERCBBT), Beijing, China
Min Guo National Engineering Research Center for Beijing Biochip Technology (NERCBBT), Beijing, China
Abstract: A microchip-based cell separation system was developed and fabricated by integrating traveling-wave dielectrophoresis and laminar flow on a single biochipbased device. Numerical simulations regarding the geometric model of the planar electrode and microfluidic channel were performed primarily for electric-field and field-flow analysis. The fabrication processes of microelectrodes and microfluidic channels were investigated, and an automated method of measurement was developed for cell characterization. The function of the device was demonstrated by separating viable human myelogenous HL-60 cells from non-viable ones. The preliminary data achieved with this device indicated that cells in different physiological states Received 25 August 2006; accepted 5 September 2006 The authors would like to thank Drs. Ailiang Chen, Xiaosheng Guan, Lihua Huang, and Guanbin Zhang for helpful discussions and Dr. Keith Mitchelson for his assistance with manuscript presentation. This work was supported by the National Hi-Tech Program (No. 2002AA2Z2011) from the Department of Science and Technology, China. Address correspondence to Jun Yu, Department of Electronic Science and Technology, Huazhong University of Science and Technology, Wuhan 430074, China. E-mail:
[email protected] or
[email protected] 749
LANL201684
LANL_040_004
750
Techset Composition Ltd, Salisbury, U.K.
1/19/2007
C. Huang et al.
could be effectively separated. The present chip, which is capable of separating and manipulating various kinds of bio-particle mixtures with different traveling-wave dielectrophoresis responses, shows promise for applications in high-throughput integrated biological analysis systems. Keywords: Traveling-wave dielectrophoresis, laminar flow, cell separation
INTRODUCTION New and improved techniques to characterize or sort cells and other bioparticles are in high demand for a wide range of applications in areas such as biomedical research, clinical diagnosis, and environment analysis. The identification and characterization of individual cells and the purification of cell subpopulations from mixed suspensions are of particular importance to medical research and to systems biology. It is also of great importance to separate and isolate pathogenic bacteria from non-pathogenic bacteria and to separate viable cells from non-viable cells for biomedical analysis. The current methods commonly used for manipulation, concentration, and separation of biological particles employ several kinds of physical forces, including mechanical, hydrodynamic, ultrasonic, optical, and electromagnetic forces (Wang et al. 2000). Among these methods, dielectrophoresis (DEP) on microchips has proved especially suitable, due to the relative ease of structuring an electric field with microfabricated electrodes and microscale generation of an appropriate DEP field. It offers several other advantages over conventional separation and characterization techniques, including label-free separation, portability, and ease of integration with other physical forces. A number of different methods based on DEP have been developed for on-chip separation and characterization of cells in recent years, such as DEP migration (Cheng et al. 1998; Fiedler et al. 1998), DEP retention (Li and Kaler 2004), and dielectrophoretic field-flow fractionation (DEP-FFF) (Wang et al. 2000), yet all of these approaches have some disadvantages, such as being noncontinuous separation processes and requiring manual identification of the resulting cell fractions. Here, we developed and fabricated a high-throughput continuous particle separation system based on traveling-wave dielectrophoresis (TWDEP) and also provided an automated process for cell motion measurements, based on computer-assisted image analysis methods, to overcome both of the previously mentioned drawbacks. In manufacturing the microdevice, we fabricated a transparent indium-tin-oxide (ITO) electrode array substrate and adhesively bonded the electrode array substrate onto a poly(dimethylsiloxane) (PDMS) cover. A recognition rate of more than 90% of the cells on ITO electrodes was achieved during the fractionation experiments, allowing rapid and successful cell tracking to be performed. A mixture of viable and non-viable HL-60 cells was fractionated by about 60 mm within 45 s. The device is also
LANL201684
LANL_040_004
Techset Composition Ltd, Salisbury, U.K.
Microchip-Based Cell Separation Device
1/19/2007
751
capable of being readily integrated with other microscale components, such as electrorotation chips (Holzel et al. 1999) and microreactor chips (Shen et al. 2005), to form a “sample-to-answer” system, i.e., an integrated “lab-on-achip” (Borgatti et al. 2005; Gambari et al. 2003).
DEVICE DESIGN AND SIMULATION Microelectrode Design A detailed outline of the theory governing TWDEP has been given elsewhere (Fuhr et al. 1991; Hagedorn et al. 1992; Wang et al. 1994; Hughes et al. 1996; Talary et al. 1996; Morgan et al. 1997). Briefly, a series of parallel planar microelectrodes is energized with an AC voltage in phase quadrature to produce a nonuniform traveling electric field. At an appropriate frequency and medium conductivity, polarizable particles will be levitated above the electrodes under negative DEP forces and be pushed to move perpendicularly to the electrode array by the TWDEP forces. Figure 1 illustrates the basic principle of our device. The cell separation chip is fabricated by bonding a PDMS cover piece with pre-molded microstructures tightly onto a glass substrate. The upper surface of the glass substrate is plated with an array of parallel electrodes fabricated using micro-electro-mechanical systems (MEMS) technology. A mixture of cells suspended in a medium with suitable conductivity is introduced into the central channel of the device and carried toward the outlet by laminar flow of the bulk medium. When the traveling electric field is not applied, cells will be carried by the flow and leave the channels via the central outlet. After applying an AC signal on the electrodes, TWDEP forces are generated, which will deflect the cells in the mixture to migrate perpendicularly to the flow direction. The magnitude and direction of TWDEP forces are dependent on the electric polarization properties of the cell (Hughes et al. 1996; Wang et al. 1995). Cells that respond to the different TWDEP force magnitude and direction will move continuously to different locations across the channels as they flow and thus be continuously separated toward the different outlets. The electric potential, applied to the microelectrodes in the microchannels, generates symmetric traveling electric fields across the microchannels. Figure 2 shows the results of simulation of the electric field distribution across the separation channel (A-A’ in Fig. 1). A symmetric traveling electric field was generated in the channels (Fig. 2a). Above a height of the travelingwave length, the TWDEP force points in the negative X-direction, perpendicularly to the electrode array (Fig. 2b). Closer to the electrodes, the vector exhibits a more complicated pattern over the electrode edges, where the magnitude of this vector function is maximal (Green et al. 2002; Li et al. 2004). The simulation illustrated in Fig. 2 shows that the magnitude
LANL201684
LANL_040_004
Techset Composition Ltd, Salisbury, U.K.
752
1/19/2007
C. Huang et al.
Figure 1. The basic principle of the device. The device was a typical sandwich structure. The bottom layer was glass substrate; the middle layer was microelectrode array, and the top layer was microchannels formed with PDMS. A special entrance port with bifurcation structure was designed. Cell mixture entered the channels from the central inlet; meanwhile, separation buffer entered the channels from the other two side inlets.
of the TWDEP forces are estimated to be in the range between several pN to tens of pN, generating forces sufficient to move the bio-particles and cells (Hughes et al. 1996; Wang et al. 1995; Doh and Cho 2005).
Microfluidic Channels Design A special entrance port of the microfluidic channels was designed based on a combination of the principles of laminar flow and hydrodynamic focusing to guarantee that the input cell suspension stays confined within a certain range, using a concept borrowed from flow cytometry (Fiedler et al. 1998; Huang et al. 1997). As shown in Fig. 1, the microfluidic channels include a rectangular flow chamber with a particular microstructure that connects 3 inlets with 7 outlets. The central inlet in the middle of the spacer functions as the passage for the through-flowing cell suspension, and the other two inlets at either side function as the passages for the separation buffer. The channels for the separation buffer begin from one common inlet port and are then bifurcated into two streams. Each line is again bifurcated into two narrower lines, until the terminal width of each channel is 50 mm. With this symmetric design, only one syringe pump is required to generate the laminar flow, and the flow rate is stable and uniform across the flow bed because of the uniform fluidic
LANL201684
LANL_040_004
Techset Composition Ltd, Salisbury, U.K.
Microchip-Based Cell Separation Device
1/19/2007
753
Figure 2. The distribution of traveling-wave electric field and TWDEP force. The distribution of traveling electric field at the cross-section (A-A’ of Fig. 1) (a) and the vector distribution of the TWDEP force component in the unit solution cell at the cross-section (A-A’ of Fig. 1) (b). (The “X” and “Y” stand for the normalized position coordinates.)
resistance in each laminar flow. The laminar flow in the microfluidic channels was analyzed by the finite element analysis (FEA) method and validated by introducing red ink from the central inlet and pure water from the other two inlets, as shown in Fig. 3. We can see that a steady laminar-flow field was generated in the most area of the flow chamber if without TWDEP influence (Fig. 3a). The directions and colors of the arrows in Fig. 3b stand
LANL201684
LANL_040_004
Techset Composition Ltd, Salisbury, U.K.
754
1/19/2007
C. Huang et al.
Figure 3. The simulation of flow field and the validation of experiment. A steady laminar-flow field was generated in the most area of the flow chamber without TWDEP influence (a). The directions and colors of the arrows stand for the directions and magnitudes of the flow, respectively. The laminar flow was validated by introducing red ink from the central inlet and pure water from the others (b).
for the directions and magnitudes of the flow, respectively. The laminar flow was achieved in the experiment by using red ink and pure water.
MATERIALS AND METHODS HL-60 Cell Preparation The HL-60 cells were obtained from ATCC (Rockville, MD) and were grown at 378C for 24 h in RPMI 1640 culture medium (HyClone, Logan, UT) in a cell
LANL201684
LANL_040_004
Techset Composition Ltd, Salisbury, U.K.
Microchip-Based Cell Separation Device
1/19/2007
755
incubator, washed two times with the separation buffer composed of 8.5% (w/v) sucrose þ0.3% (w/v) dextrose and PBS with a conductivity of 35.5 mS/m, and then re-suspended in the separation buffer. Non-viable HL-60 cells were prepared by adding 50 nM bufalin, which induces cell necrosis, to the cell culture solution and incubating for 24 h. The separation sample for the test was prepared by mixing equal amounts of viable and non-viable cells in the separation buffer. Device Fabrication and Experimental Setup The fabrication processes for producing the TWDEP chip are shown in Fig. 4. We used a 4-inch ITO conductive glass wafer (CSG Holding Co., Ltd., Shenzhen, China) as a substrate to fabricate the microelectrode array. It was coated with 1.2-mm thick PR (Photoresist: AZ 1512) and etched with the following etching solution: 37% HCl: 67% HNO3: H2O ¼ 50: 3: 50, volume ratio. Next, we used a 4-inch silicon wafer and coated it with 50mm thick SU-8 2050 PR layer (Microlithography Chemical Corporation, USA) and then defined it for pouring the microchannel mold. Then, a PDMS prepolymer mixture (PDMS: curing agent ¼ 10:1) (Sylgard 184, Dow Corning Co., Ltd., Midland, MI, USA) (Duffy et al. 1998) was poured to form the microchannels by using the mold. After 4 h of curing at 658C, the PDMS microchannels were peeled from the mold and bonded together with the ITO glass substrate by O2 plasma treatment. Figure 5 shows the fabricated device and the enlarged view of the ITO electrodes. To separate the cell mixture, the electrodes were energized by four sine waves (Vpp ¼ 8 V) in phase quadrature with frequencies of 1 kHz and 10 kHz, respectively, by a signal generator (DEP Power B, Capitalbio, Beijing, China).
Figure 4.
The fabrication processes of the microelectrodes and microchannels.
LANL201684
LANL_040_004
756
Techset Composition Ltd, Salisbury, U.K.
1/19/2007
C. Huang et al.
Figure 5. The photograph of the fabricated device and the enlarged view of the ITO electrodes. Width of the electrode and the gap were both 20 mm. (Scale bar ¼ 50 mm).
A separation buffer was delivered at 20 mL/min by a syringe pump (PHD 2000, Harvard Ltd., Holliston, MA, USA); meanwhile, the cell suspension was pumped into the central inlet at 5 mL/min by another smaller syringe pump (Kd Scientific, Holliston, MA, USA). Both the separation buffer flows and the cell sample flow were maintained at the same terminal flow rates within the rectangular chamber. Cell motion was visualized by using phase contrast microscopy (Leica DMRE, Leica Microsystems, Wetzlar, Germany) with a color CCD (Panasonic WVGP410, Osaka, Japan) and recorded to a videotape (Panasonic NV-HD 500 video recorder, Osaka, Japan) for later image analysis. The temperature of the separation channel was maintained at 21 + 0.18C. During the experiments, no obvious Joule heating effects were detected in the channels by using a low signal voltage (Vpp ¼ 8 V) (Ramos et al. 1998; Green et al. 2000) and a medium with low conductivity (,1000 mS/m) (Schnelle et al. 1999). To confirm that the cell motion in the microchannel was due to TWDEP effect, we measured the velocity of the viable and non-viable cells as a function of traveling electric fields with different directions and magnitudes, respectively.
RESULTS Cell Tracking An important characteristic of the proposed approach was the possibility of the recognition of cell motion through analysis of data using software compiled on a computer, which would overcome the drawbacks of manual methods. For
LANL201684
LANL_040_004
Techset Composition Ltd, Salisbury, U.K.
Microchip-Based Cell Separation Device
1/19/2007
757
manual methods, the velocity of motion of the cells was usually calculated by determining the displacement of the cells with a micrometer from the image on a TV monitor, and the time duration of this displacement was measured with a stopwatch. In contrast, our image analyses using software involved three main automated processes: cell recognition, cell tracking, and velocity calculation. In order to test the function of our software, the motion of cells over the ITO electrodes was captured by a CCD camera. Then, each color image, as shown in Fig. 6a, was converted into a gray-scale image by summing the red, green, and blue components of the image and calculating the overall intensity of each pixel in the image. Then, a binary image was obtained by performing a threshold operation on the gray-scale image to segment the regions of background and cells. This segmentation process could help extract positional information about each of the cells. A suitable level for the threshold could be determined by the mean and standard deviation in the intensity of the background pixels, as shown in Fig. 6b. Because the ITO electrodes used were transparent, the background of the image was easily subtracted, and more than 90% of the cells were recognized by labeling the gravity center of each cell region. During the cell recognition process, the sequence of video frames took the form of coordinates of the cells, which were used to track the cells in a sequence of video frames. The cell tracking was performed by selecting one cell from the first image and searching for the corresponding cell in the next image, as shown in Fig. 6c, and the translational velocity of each cell was determined by connecting the positions of each cell and calculating the rates of position changes between sequential images. The difference in the velocity determined between automated and manual measurements was no more than 5%. These data indicate that an automated, real-time, online TWDEP measurement method could be developed by using the ITO electrode-based device.
Continuous Separation Figure 7 shows sequential video images of the continuous separation on the ITO electrode array, which illustrates the distinct separation efficiency of our device. When a signal voltage is applied at the electrodes, a traveling electric field is generated and the TWDEP forces deflect the cells perpendicularly to the flow direction. Figure 7 a, b, and c shows a section of the TWDEP electrode array with HL-60 cells moving across the field of view from left to right, with the device energized with a TWDEP at 1 kHz frequency. The three images were taken at 15-s intervals, and they show that the viable cells had moved perpendicularly to the direction of flow by approximately 190 mm and that non-viable cells had moved 130 mm at 1 kHz frequency. The electrodes are out of focus in this figure, owing to the fact that the cells were levitated by about 7 mm by negative DEP forces. The software captured the movement of several cells for over 45 s and gave average
LANL201684
LANL_040_004
758
Techset Composition Ltd, Salisbury, U.K.
1/19/2007
C. Huang et al.
Figure 6. Cell tracking. The color image (a) was captured from a CCD and was performed by a threshold operation (b) after it was converted into a gray-scale image. The cell tracking (c) was performed by selecting one cell from the first image and searching for the corresponding cell in the next image.
LANL201684
LANL_040_004
Techset Composition Ltd, Salisbury, U.K.
Microchip-Based Cell Separation Device
1/19/2007
759
Figure 7. A sequence of 3 images taken at 15-s intervals (a, b, c), showing that the viable HL-60 cell moved faster perpendicularly to the electrodes than the non-viable HL-60 cells.They differed from each other by 60 mm within 45 s. Width of the electrode and the gap were both 20 mm. (Scale bar ¼ 50 mm).
deflected velocities of 4.3 mm/s for the viable cells and 3.0 mm/s for the nonviable cells, as shown in Fig 8. The result was also confirmed by the velocity of cell motion after increasing the frequency of the electric field to 10 kHz, as also shown in Fig. 8. At both of these frequencies, the velocity of viable cells was faster than that of the non-viable cells. DISCUSSION Effect of Microelectrode and Microfluidic Channels There are design variables that had significant influence on the performance of the devices. The efficiency of separation of the viable and non-viable cells depends mainly on the performance of the microelectrode array and microfluidic channel. Compared with the Cr/Au electrodes used previously by others (Talary et al. 1996; Morgan et al. 1997), the use of optically transparent ITO electrodes provided much better optical accuracy and much lower background noise, which was helpful for a computer-assisted image analysis. The ITO electrodes had an optical transparency of more than 85% over the completely optically opaque Cr/Au electrodes. Figure 9 shows the motion of cells on the two types of electrodes. Cells with a radius about 6 mm could
LANL201684
LANL_040_004
760
Techset Composition Ltd, Salisbury, U.K.
1/19/2007
C. Huang et al.
Figure 8. The transversal migration velocities of viable and non-viable HL-60 cells.Both at 1 kHz and 10 kHz, the viable HL-60 cells had a higher velocity perpendicular to the electrodes than that of the non-viable HL-60 cells. The values were expressed as the mean from three independent experiments; bars, +SD.
easily be observed on the ITO electrodes. The opaque Cr/Au electrodes with a width of 20 mm partially prevented the view of the motion of cells inside the microfluidic channels, making it difficult for the accurate calculation of the velocity of the cell motion, because they could not be discerned when the cells were submerged in the darkness of the Cr/Au electrodes under a transmission microscope. Thus, despite the higher impedance, we chose the ITO electrodes to realize our automated particle separation system. The epoxy-based photoresist SU-8 was used for microchannel molding. This could be deposited with a thickness of up to 500 mm and patterned with high resolution. Figure 10 shows a photograph of the microstructure mold fabricated by SU-8 2050. The sizes and the surface of the channel molds were measured by using a scanning electron microscope (SEM) (Hitachi, S-3000N, Tokyo, Japan). It could be seen that the reproducibility and lift-off was excellent, with little residue remaining on the substrate surface and with perfectly sharp side walls. These uniform features were critical for forming an effective seal between the electrode array and the PDMS microchannels and for the subsequent generation of a steady laminar flow in the microchannels. Separation Efficiency Biological cells consist of adjacent structures of materials that have very different electrical properties and exhibit large, induced boundary
LANL201684
LANL_040_004
Techset Composition Ltd, Salisbury, U.K.
Microchip-Based Cell Separation Device
1/19/2007
761
Figure 9. The motion of HL-60 cells on the ITO electrodes (a) and on the Cr/Au electrodes. (b) Compared with the Cr/Au electrodes, the motion of HL-60 cells on the ITO electrodes had much lower background noise.
polarizations that are highly dependent on the applied field frequency, as well as their physiological states. For example, the cell membrane consists of a very thin lipid bi-layer with many proteins and is highly insulating, with a conductivity of around 1027 S/m, whereas the cell interior contains many dissolved charged molecules, leading to a conductivity as high as 1 S/m. Upon death, the cell membrane becomes permeable, and its conductivity can increase by a factor of 104 due to the cell contents freely exchanging materials with the external medium through small pores on the membrane (Li and Bashir 2002). This large change in the dielectric properties upon
LANL201684
LANL_040_004
762
Techset Composition Ltd, Salisbury, U.K.
1/19/2007
C. Huang et al.
Figure 10. A SEM photograph of the microstructure mold fabricated by SU-8 2050 photoresist. The mold had little residue remaining on the substrate surface and had perfectly sharp side walls. (Scale bar ¼ 25 mm).
cell death indicates a large change in the dielectric polarizability. Hence, a large difference in DEP responses and a selective separation can be achieved between viable and non-viable cells. The length of the separation of the microfluidic channels was directly proportional to the maximum flow rate that the system could handle. When the separator element was longer, the particles would experience the TWDEP force for a longer time. Therefore, less displacement per unit of time would be required for separating in a longer channel. A simple analysis of the separation efficiency showed that, assuming both cell types began traveling at the same time and from the same end of the array, there would be a path difference of 1 mm between the viable cells and non-viable cells by the time they reached the end of the 4-cm long electrodes. For the separation of small numbers of cells, as required for diagnostic purposes, smaller band separations would be acceptable. Large TWDEP arrays, extending over the length of a glass microscope slide (5 cm), would be capable of separating cells with extremely small differences in their respective dielectric properties. For a 5-cm long array, an equivalent difference in dielectric properties (Im (fCM)) between particles of as little as 0.2% would translate into a band separation of more than 1 mm (Morgan et al. 1997).
CONCLUSION Here, we presented the development of a high-throughput continuous cell separation chip that uses field-flow and TWDEP forces to effect cell separation.
LANL201684
LANL_040_004
Techset Composition Ltd, Salisbury, U.K.
Microchip-Based Cell Separation Device
1/19/2007
763
The fabrication processes of the microelectrode and the microfluidic channels were investigated, and an automated measurement method was developed for cell characterization. In the experimental study, we obtained the cell separation conditions under which viable and non-viable HL-60 cells showed different TWDEP responses. Cells would be fractionated from each other in the separation channels, and continuous cell separation was achieved with improved performance over other existing methods. The present chip, capable of separating and manipulating various kinds of bio-particles mixtures having different TWDEP responses, has promise for applications in high-throughput integrated systems biological analyses and for biomedical and diagnostic testing.
REFERENCES Borgatti, M., Altomare, L., Abonnec, M., Fabbri, E., Manaresi, N., Medoro, G., Romani, A., Tartagni, M., Nastruzzi, C., Croce, S.D., Tosi, A., Mancini, I., Guerrieri, R., and Gambari, R. 2005. Dielectrophoresis-based ‘lab-on-a-chip’ devices for programmable binding of microspheres to target cells. Int. J. Oncol., 27: 1559– 1566. Cheng, J., Sheldon, E.L., Wu, L., Uribe, A., Gerrue, L.O., Carrino, J., Heller, M.J., and O’Connell, J.P. 1998. Preparation and hybridization analysis of DNA/RNA from E. coli on microfabricated bioelectronic chips. Nat. Biotechnol., 16: 541 –546. Doh, I. and Cho, Y.H. 2005. A continuous cell separation chip using hydrodynamic dielectrophoresis process. Sens. Actuators A, 121: 59 – 65. Duffy, D.C., McDonald, J.C., Schueller, O.J.A., and Whitesides, G.M. 1998. Rapid prototyping of microfluidic systems in poly(dimethylsioxane). Anal. Chem., 70: 4974– 4984. Fiedler, S., Shirley, S.G., Schnelle, T., and Fuhr, G. 1998. Dielectrophoretic sorting of particles and cells in a microsystem. Anal. Chem., 70: 1909– 1915. Fuhr, G., Hagedorn, R., Muller, T., Benecke, W., Wagner, B., and Gimsa, J. 1991. Asynchronnous traveling-wave induced linear motion of living cells. Studia. Biophysica, 140: 79 – 102. Gambari, R., Borgatti, M., Altomare, L., Manaresi, N., Medoro, G., Romani, A., Tartagni, M., and Guerrieri, R. 2003. Applications to cancer research of ‘lab-on-achip’ devices based on dielectrophoresis. Technol. Cancer Res. Treat, 2: 31 – 40. Green, N.G., Ramos, A., and Morgan, H. 2000. AC electrokinetics: A survey of submicrometre particle dynamics. J. Phys. D: Appl. Phys., 33: 632– 641. Green, N.G., Ramos, A., and Morgan, H. 2002. Numerical solution of the dielectrophoretic and travelling wave forces for interdigitated electrode arrays using the finite element method. J. Electrostat., 56: 235– 254. Hagedorn, R., Fuhr, G., Muller, T., and Gimsa, J. 1992. J. Traveling-wave dielectrophoresis of microparticles. Electrophoresis, 13: 49 – 54. Holzel, R. 1999. Non-invasive determination of bacterial single cell properties by electrorotation. Biochim. Biophy. Acta, 1450: 53 – 60. Huang, Y., Wang, X.B., Becker, F.F., and Gascoyne, P.R. 1997. Introducing dielectrophoresis as a new force field for field-flow fractionation. Biophys. J, 73: 1118– 1129. Hughes, M.P., Pethig, R., and Wang, X.B. 1996. Dielectrophoretic forces on particles in traveling electric fields. J. Phys. D: Appl. Phys., 29: 474– 482.
LANL201684
LANL_040_004
764
Techset Composition Ltd, Salisbury, U.K.
1/19/2007
C. Huang et al.
Li, H. and Bashir, R. 2002. Dielectrophoretic separation and manipulation of live and heat-treated cells of Listeria on microfabricated devices with interdigitated electrodes. Sen. Actuators B, 86: 215– 221. Li, W.H., Du, H., Chen, D.F., and Shu, C. 2004. Analysis of dielectrophoretic electrode arrays for nanoparticle maniplulaton. Comp. Mater. Sci., 30: 320– 325. Li, Y. and Kaler, K.V.I.S. 2004. Dielectrophoretic fluidic cell fractionation system. Anal. Chim. Acta, 501: 151– 161. Morgan, H., Green, N.G., Hughes, M.P., Monaghan, W., and Tan, T.C. 1997. Largearea traveling-wave dielectrophoresis particle separator. J. Micromech. Microeng., 7: 65 – 70. Ramos, A., Morgan, H., Green, N.G., and Cstellanous, A. 1998. AC electrokinetics: A review of forces in microelectrode structures. J. Phys. D: Appl. Phys., 31: 2338– 2353. Schnelle, T., Muller, T., Gradl, G., Shirley, S.G., and Fuhr, G. 1999. Paired microelectrode system: Dielectrophoretic particle sorting and force calibration. J. Electrostat., 47: 121– 132. Shen, K.Y., Chen, X.F., Guo, M., and Cheng, J. 2005. A microchip-based PCR device using flexible printed circuit technology. Sens. Actuators B, 105: 251– 258. Talary, M.S., Burt, J.P.H., Tame, J.A., and Pethig, R. 1996. Electromanipulation and separation of cells using traveling electric fields. J. Phys. D: Appl. Phys., 29: 2198– 2203. Wang, X.B., Huang, Y., Becker, F.F., and Gascoyne, P.R. 1994. A unified theory of dielectrophoresis and traveling-wave dielectrophoresis. J. Phys. D: Appl. Phys., 27: 1571– 1574. Wang, X.B., Hughes, M.P., Huang, Y., Becker, F.F., and Gascoyne, P.R. 1995. Nonuniform spatial distributions of both the magnitude and phase of AC electric fields determine dielectrophoretic forces. Biochim. Biophys. Acta, 1243: 185– 194. Wang, X.B., Yang, J., Huang, Y., Vykoukal, J., Becker, F.F., and Gascoyne, P.R. 2000. Cell separation by dielectrophoretic field-flow fractionation. Anal. Chem., 72: 832– 839.
