Article pubs.acs.org/ac
Continuous Cell Separation Using Dielectrophoresis through Asymmetric and Periodic Microelectrode Array Siang Hooi Ling, Yee Cheong Lam,* and Kerm Sin Chian School of Mechanical and Aerospace Engineering, Nanyang Technological University, Singapore 639798 S Supporting Information *
ABSTRACT: The study presents a dielectrophoretic cell separation method via three-dimensional (3D) nonuniform electric fields generated by employing a periodic array of discrete but locally asymmetric triangular bottom microelectrodes and a continuous top electrode. Traversing through the microelectrodes, heterogeneous cells are electrically polarized to experience different strengths of positive dielectrophoretic forces, in response to the 3D nonuniform electric fields. The cells that experience stronger positive dielectrophoresis are streamed further in the perpendicular direction to the fluid flow, leaving the cells that experience weak positive dielectrophoresis, which continue to traverse the microelectrode array essentially along the laminar flow streamlines. The proposed method has achieved 87.3% pure live cells harvesting efficiency from a live/dead NIH-3T3 cells mixture, and separation of MG-63 cells from erythrocytes with a separation efficiency of 82.8%. The demonstrated cell separation shows promising applications of the DEP separator for cell separation in a continuous mode.
D
allowed continuous operation to separate live and dead yeast cells.15 Other conventional DEP separation techniques were Dielectrophoresis-Field Flow Fractionation (DEP-FFF)26−28 and Traveling-wave Dielectrophoresis (TwDEP).29−32 In the DEP-FFF technique, bioparticles of different DEP mobility were pushed by nDEP from the underlying electrode array onto different elevations within a microchannel and subsequently transported downstream by a parabolic flow field. Bioparticles at higher elevations were thus transported by higher flow velocity to elute downstream at a shorter time, compared to bioparticles at lower elevations. However, in order to produce a prominent elution time lag between bioparticles, this technique requires a long (∼30 cm) separation electrode array to achieve an effective separation, which could severely affect device portability. The TwDEP technique separates bioparticles in stationary29,30 or moving31,32 fluid by employing a row of underlying microelectrodes, with each adjacent microelectrode connected to different potential inputs having different phases. However, this technique requires complicated manufacturing processes to fabricate the independent excitable underlying microelectrode array. There are recent DEP developments targeted in the separation of bioparticles or engineering particles in lateral mode under continuous operation. Various microelectrode designs and configurations were explored, including, but not limited to, trapezoidal electrode array,33 slanted lines electrode array,34 and physical35 or virtual36 circular pillar arrays.
evelopment of dielectrophoresis (DEP) as a particle manipulation technology has progressed rapidly since its discovery.1 DEP technology is advantageous not only for its ability to manipulate various dielectric bioparticles, especially cells2,3 and nano bioparticles,4,5 but also for its simplicity to integrate into a microfluidic system. To date, the versatility of DEP applications have expanded to include bioparticle trapping,6−8 transportation,9,10 and patterning.11,12 Nevertheless, there are continued advances in the traditionally recognized application of DEP as the bioparticle separation technique.13,14 One of the DEP applications has provided a nonstaining method to differentiate between live and dead cells.15,16 This serves as an alternative to conventional methods such as flow cytometry,17 fluorescence technique,18 dye exclusion,19,20 and capillary electrophoresis based on laserinduced fluorescence,21 which require prestaining/labeling of cell samples. As such, DEP provides new possibilities for bioparticle separation based on their DEP properties, rather than cell surface antigen, membrane integrity, or density. Conventional DEP bioparticle separation was achieved by inducing strong positive dielectrophoresis (pDEP) on one type of bioparticles and trapping them. Meanwhile, the bioparticles of another type were induced with either weak pDEP or negative dielectrophoresis (nDEP), and these were eluted downstream with the flowing fluid. The trapped bioparticles were released from the trapping regions after the elution of the other bioparticles. This DEP separation technique has been demonstrated to separate cancer22,23/stem24 cells from blood cells, and yeasts from bacteria.25 However, prolonged exposure of trapped bioparticles under high electric fields could result in an adverse biological effect. Early development of this technique was limited to batch mode, but its later modification © 2012 American Chemical Society
Received: February 1, 2012 Accepted: July 19, 2012 Published: July 19, 2012 6463
dx.doi.org/10.1021/ac300079q | Anal. Chem. 2012, 84, 6463−6470
Analytical Chemistry
Article
distance of 80 μm in the x-direction and 11 μm in the ydirection between adjacent microelectrodes of different column), relative to the flow direction. The performance of the proposed DEP separator was demonstrated through the separation of live and dead Swiss mouse fibroblast (NIH-3T3) cells, and human osteosarcoma (MG-63) cells from horse erythrocytes. The proposed DEP separator inherits the merits of our previous design, such as simple in fabrication and robust in electrode design.44 The proposed 3D electrode configuration is also advantageous to generate strong 3D nonuniform electric fields that distribute throughout the microchannel height, thus allowing all incoming cells (regardless of their height positions) to experience sufficiently strong dielectrophoretic effects, and increasing the effectiveness of the proposed DEP separator in cell manipulation. This advantage compensates the DEP decaying effects that could result in those employing 2D or 3D microelectrodes along the microchannel sidewalls. Furthermore, alternating power on/off duty cycles is not required in the operation to achieve cell separation with low cell adhesion. This is due to the nature of the periodic discrete microelectrode array in generating cyclic nonuniform electric fields that resembles the nature of nonuniform electric fields generated by the power on/off duty cycles, as demonstrated by using the slanted line microelectrodes.
However, these electrode arrays were mainly demonstrated in the separation of particles under nDEP, possibly unwanted particle adhesions at microelectrodes are always an issue for the separation of pDEP particles/cells. To overcome the adhesion of pDEP particles/cells, one possible solution is to power the microelectrode array with alternating on/off duty cycles,37,38 e.g., powering on and off the microelectrodes with intervals of 0.5 ms. Although the proposed solution avoid particle adhesions yet promote separation of pDEP particles/cells, but additional external electronic controller is required to control the alternating on/off duty cycles. Other proposed DEP separators employed two-dimensional (2D)39,40 or threedimensional (3D)41−43 microelectrodes along the sidewalls of the microchannel. Since the dielectrophoretic effect decays exponentially with distance from these sidewall microelectrodes, the particle separation performance of these separators was only effective with the particles flowing near to these sidewall microelectrodes. We have previously demonstrated the employment of 3D electrode to generate 3D nonuniform electric fields for continuous size-sorting of microparticles.44 The reported 3D electrode consists of discrete periodic “isosceles” triangular bottom microelectrodes, and a continuous top electrode. This “isosceles” triangular microelectrode array has been demonstrated to work well in the separation of microparticles under nDEP, but not biological cells under pDEP. This is mainly attributed to the selection of “isosceles” triangular microelectrodes that create two steep edges (with an acute angle of 24° for the apex of the microelectrode), with respect to the flow direction. When a cell traverses such an “isosceles” triangular microelectrode, two extreme situations can result: (i) for a cell that experiences predominant hydrodynamic force but intermediate strength or weak pDEP, the steep edges of the microelectrodes fail in their roles to promote effective streaming of cells along the edges, and, as a result, the cell is driven across the microelectrode with a minimal change in yposition; or (ii) for a cell that experiences predominant pDEP, the strong electric field gradient only causes cell trapping at the corners of “isosceles” triangular microelectrodes. Furthermore, the row arrangement between these discrete periodic “isosceles” triangular bottom microelectrodes posed at a steeper tilting angle of ∼37.6° (with an offset distance of 35 μm in the x-direction and 27 μm in the y-direction between adjacent microelectrodes of different column), relative to the flow direction, is not favorable for cell separation. Given such a steeper angle between these microelectrodes arrangement, a targeted cell could leave the previous microelectrode without being effectively guided to flow along the next nearest microelectrode of the same row but different column. Failure in guiding the cell to this preferential arrangement of the microelectrode array voids the function of microelectrode arrangement to promote effective cell separation under pDEP. In view of the drawbacks in our previous electrode design, we thus proposed a different triangular microelectrode design with the following modifications to promote effective streaming of targeted pDEP cells along both the microelectrode edges and the row arrangement between microelectrodes:
■
MATERIALS AND METHODS Cell Preparation. Two different cell samples were prepared as the separation targets, namely, a mixture of live and dead NIH-3T3 cells, and a mixture of MG-63 cells and horse erythrocytes. NIH-3T3 cells (CRL-1658, ATCC) and MG-63 cells (CRL1427, ATCC) were cultured in a 5% CO2 incubator at 37 °C with Dulbecco’s Modified Eagle Medium (DMEM) containing 10% fetal bovine serum and 1% antibiotic antimycotic solution. To prepare the live NIH-3T3 and MG-63 cells sample, the respective cultured cells were harvested at ∼90% confluence and resuspended in DEP buffer (containing 8.5% sucrose and 0.3% dextrose solution). The conductivity of the DEP buffer was adjusted to 122.3 μS/cm with NaCl solution and measured using a conductivity meter (AR20, Fisher Scientific). The dead NIH-3T3 cells sample was prepared by heating live cells at 70 °C for 18 min. To prepare the erythrocytes sample, a stock of horse erythrocytes (IC100-0110, Innovative Research) was washed once and resuspended in DEP buffer. To prepare a live/dead cells mixture, live and dead NIH-3T3 cells were individually stained with Calcein AM (Molecular Probes) and Ethidium homodimer-1 (Molecular Probes) for 15 min in darkness, then washed once and resuspended in fresh DEP buffer to create two cell suspensions of ∼3.5 × 106 cells/ mL. The stained live and dead cells were then mixed at a nominal ratio of 1:1 and could be distinguished by their green and red fluorescence, respectively. To prepare a cancer cells/erythrocytes mixture, MG-63 cells were prestained with Calcein AM (Molecular Probes) for 15 min in darkness, then washed once before being resuspended in fresh DEP buffer. Cell suspensions of MG-63 cells and erythrocytes were then mixed at a nominal ratio of 1.3:1 at cell concentration of ∼2.32 × 106 cells/mL. The stained MG63 cells emitted green fluorescence, thus differentiating them from the unstained small white-colored erythrocytes.
• Each discrete triangular bottom microelectrode was locally asymmetric in its design, and all its edges with angles FDEP). The FDEP is sufficiently high to retard the cell’s incoming velocity and cause it to flow along the edges of the bottom microelectrode. Upon being released from the previous microelectrode, it will be streamed toward the corner of the next-nearest microelectrode of the same row but located at a higher y-position (where the strong electric field gradient exists), and flows along its edges. This action continues and the cell will continuous to traverse along the row of the microelectrode array to result in a higher exit y-position at the downstream. The other cell is assumed to experience intermediate strength of positive FDEP, and it is relatively weak, compared with FStokes (illustrated as the solid circles in Figure 4). The exception is that the cell may occasionally experience a strong positive FDEP at the microelectrode edge where the strong spatial strength of ∇|E|2 exists. As such, it will just flow past the respective row of the microelectrode array with only a moderate change in y-position. This streaming pattern is repetitive as the cell traverses the microelectrode array and will exit at a lower streaming position toward the positive ydirection, compared to the cell with strong positive FDEP. For another cell that experiences a negligible FDEP, illustrated as dotted circles in Figure 4, it will just be carried by the flowing fluid to traverse the microelectrode array without a trajectory change. Given a properly designed microelectrode array, cell separation can be achieved by manipulating these cells to exit at different bands of y-positions downstream from the microelectrode array.
Figure 3. Re[fcm] as a function of electric field frequency for (a) live and dead NIH-3T3 cells and (b) MG-63 cells and erythrocytes in DEP buffer of 122.3 μS/cm. Electrical and geometrical properties of these cells were taken from literature9,47,48 (see the SI).
as a function of the electric field frequency for live and dead NIH-3T3 cells, and MG-63 cell and erythrocyte in a DEP buffer of 122.3 μS/cm. When a cell experiences Re[fcm] > 0, the cell will translate toward the electric field intensity maxima. This is termed as pDEP. When a cell experiences Re[fcm] > 0, the cell will translate toward the electric field intensity minima. This is referred as nDEP. Under a special condition in which a cell has Re[fcm] = 0, the cell experiences no DEP even in the presence of the nonuniform electric fields. Traversing the particle separation electrode, the cells which experience pDEP will be acted upon by different strength of FDEP to achieve separation. However, it is required that the positive FDEP on the cells is smaller than FStokes. Given that positive FDEP > FStokes, the cell will be caged at the electric field intensity maxima on the bottom microelectrode as it travels near the microelectrode. No spatial movement for this cell will be observed thereafter. To better illustrate the separation concept, a schematic showing only the x−y plane view is drawn in Figure 4. Assuming a cell that experiences strong positive FDEP travels near the microelectrode (illustrated as the double-enclosed solid circle in Figure 4), it tends to be attracted toward the
■
RESULTS AND DISCUSSION Separation of Live/Dead NIH-3T3 Cells Mixture. Upstream of the particle separation electrode, the incoming cells were prefocused to transport near the microchannel sidewall and with incoming y-positions of ≤0.45 mm (results not shown). In the absence of an electric field (VSEP = 0), these cells were observed to transport from upstream to downstream along streamlines under laminar flow. At the exit, all the cells remained at y-positions ≤0.45 mm and therefore, no cell separation occurred (see Figure 5a). The live and dead cells were recorded to have average exit y-positions of 0.143 ± 0.093 mm and 0.149 ± 0.089 mm, respectively. With VSEP = 3.4 VPP at 150 kHz or 300 kHz, separation between the live and dead cells (with a throughput of ∼1302 cells/min) was observed, as shown in Figures 5b and 5c. As they traversed the particle separation electrode, the live cells that experienced pDEP (with Re[fcm] > 0) were progressively deviated from the dead cells and moved up in the positive ydirection. On the other hand, the dead cells that experienced negligible DEP (with Re[fcm] = 0), just traversed the particle 6467
dx.doi.org/10.1021/ac300079q | Anal. Chem. 2012, 84, 6463−6470
Analytical Chemistry
Article
Figure 5. Exit distributions of live and dead NIH-3T3 cells downstream of particle separation electrode: (a) without VSEP; (b) VSEP of 3.4 VPP at 150 kHz; (c) VSEP of 3.4 VPP at 300 kHz. Micrographs showing actual exit distributions of cells are placed next to the histograms. Each dot in the micrographs indicates the exit location of either a live or dead cell downstream from the particle separation electrode.
and 1.7% dead cells at the upper separation stream. On the other hand, the value VSEP = 3.4 VPP at 300 kHz increased the average exit y-position of live cells, resulting in an improved cell separation, with 7.7% live cells and 98.0% dead cells at the lower separation stream, and 92.3% live cells and 2.0% dead cells at the upper separation stream. Using the same set of experimental results but setting 0.45 mm (the maximum y-coordinate a cell can reach in the absence of electric field, as shown in Figure 5a) as the effective separation distance for pure live cell harvesting criterion, a value of VSEP = 3.4 VPP at 300 kHz yielded 87.3% live cell harvesting efficiency, compared to 82.7% at 150 kHz. The stated live cell harvesting efficiency was indeed a conservative estimate. The claim was made as there was always a fraction of live cells dying during the sample preparation and experiment that were still being counted as live cells, because of the presence of their
separation electrode without noticeable difference between their incoming and exit y-positions. As a result, the live cells left preferentially at higher exit y-positions, compared with the dead cells at both frequencies. Comparing the two frequencies, the average lateral exit y-position of live cells at 300 kHz (0.964 ± 0.306 mm) was found to be higher than that at 150 kHz (0.902 ± 0.349 mm). In contrast, the difference between the average lateral exit y-position of dead cells from 150 to 300 kHz was less noticeable (0.130 ± 0.073 mm vs 0.157 ± 0.086 mm). A higher exit y-position at increased frequency can be explained by the higher pDEP force (associated with higher Re[fcm], as shown in Figure 3a) experienced by the live cells, which leads to a better separation of live cells from dead cells. Given an effective separation distance of 0.329 mm, a value of VSEP = 3.4 VPP at 150 kHz yielded 12.7% live cells and 98.3% dead cells at the lower separation stream, with 87.3% live cells 6468
dx.doi.org/10.1021/ac300079q | Anal. Chem. 2012, 84, 6463−6470
Analytical Chemistry
Article
Figure 6. Distributions of MG-63 cells and erythrocytes: (a) upstream; (b) downstream of particle separation electrode with VSEP = 4.4 VPP at 600 kHz. Micrographs showing actual incoming/exit distributions of cells are placed next to the respective histograms. Each dot in the micrographs indicates an incoming/exit location of either a MG-63 cell or a erythrocyte upstream/downstream of particle separation electrode.
higher average lateral exit y-position (0.930 ± 0.291 mm), compared to the erythrocytes (0.274 ± 0.187 mm). With an effective separation distance of 0.647 mm, this experimental condition led to cell separation efficiencies of 17.2% MG-63 cells and 94.4% erythrocytes at the lower separation stream, with 82.8% MG-63 cells and 5.6% erythrocytes at the upper separation stream. For cells collected at the upper separation stream, this translates to enrichments of MG-63 cells by a factor of ∼11.4, compared to the 1.3:1 mixture of MG-63 cells and erythrocytes in the initial preparation.
green fluorescent labeling. The “disguised” live cells (which were actually dead cells and exhibited similar DEP behavior as the prepared dead cells) thus left the electrode area like other dead cells at exit y-positions ≤0.45 mm. Therefore, taking these “disguised” live cells into account would thus result in a lower estimate of the live cell harvesting efficiency. Separation of MG-63 Cells/Erythrocytes Mixture. Our device could separate not only live and dead cells but also two live-cell populations. At frequencies as high as 600 kHz, both MG-63 cells and erythrocytes exhibited strongly pDEP response in a 122.3 μS/cm DEP buffer, with Re[fcm] values of 0.930 for MG-63 cells and 0.857 for erythrocytes (see Figure 3b). As FDEP ∝ R3Re[fcm], the R3Re[fcm] value of MG-63 cells (with an average cell diameter of 20.61 μm) is determined to be ∼23-fold higher than that of erythrocytes (with an average cell diameter of 7.40 μm). The large difference in R3Re[fcm] values suggests the possible separation of these two live-cell populations. As shown in Figure 6a, all MG-63 cells or erythrocytes were prefocused to transport near the microchannel sidewall with incoming y-positions ≤0.45 mm upstream. The average incoming y-positions of MG-63 cells and erythrocytes were 0.230 ± 0.108 mm and 0.235 ± 0.122 mm, respectively. With VSEP = 4.4 VPP at 600 kHz, separation between these cells (with a throughput of ∼1260 cells/min) was observed as they traversed the particle separation electrode. The MG-63 cells with a larger cell diameter experienced a dominant pDEP force and their flow paths deviated continuously toward a morepositive y-position, compared to the erythrocytes. Downstream, as shown in Figure 6b, the MG-63 cells left preferentially at a
■
CONCLUSION
This study presents a microfluidic chip for continuous cell separation based on the positive dielectrophoresis (pDEP) principle. It utilizes three-dimensional (3D) nonuniform electric fields generated by a periodic array of discrete but locally asymmetric triangular bottom microelectrodes and a continuous top electrode. The device has been demonstrated to achieve 87.3% pure live cells harvesting efficiency from a live/ dead NIH-3T3 cells mixture, and to separate MG-63 cells from erythrocytes with a separation efficiency of 82.8%. With its simplicity and robustness of the electrode design, the proposed device can be envisioned as a front-end technology to integrate into a multifunctional lab-on-chip platform for biomedical analysis. 6469
dx.doi.org/10.1021/ac300079q | Anal. Chem. 2012, 84, 6463−6470
Analytical Chemistry
■
Article
(27) Wang, X. B.; Yang, J.; Huang, Y.; Vykoukal, J.; Becker, F. F.; Gascoyne, P. R. C. Anal. Chem. 2000, 72, 832−839. (28) Yang, J.; Huang, Y.; Wang, X. B.; Becker, F. F.; Gascoyne, P. R. C. Biophys. J. 2000, 78, 2680−2689. (29) Talary, M. S.; Burt, J. P. H.; Tame, J. A.; Pethig, R. J. Phys. D: Appl. Phys. 1996, 29, 2198−2203. (30) Morgan, H.; Green, N. G.; Hughes, M. P.; Monaghan, W.; Tan, T. C. J. Micromech. Microeng. 1997, 7, 65−70. (31) Cheng, I. F.; Froude, V. E.; Yingxi, Z.; Hsueh-Chia, C.; HsienChang, C. Lab Chip 2009, 9, 3193−3201. (32) Cheng, I. F.; Chung, C.-C.; Chang, H.-C. Microfluid. Nanofluid. 2011, 10, 649−660. (33) Choi, S.; Park, J.-K. Lab Chip 2005, 5, 1161−1167. (34) Han, K.-H.; Han, S.-I.; Frazier, A. B. Lab Chip 2009, 9, 2958− 2964. (35) Beech, J. P.; Jonsson, P.; Tegenfeldt, J. O. Lab Chip 2009, 9, 2698−2706. (36) Chang, S.; Cho, Y.-H. Lab Chip 2008, 8, 1930−1936. (37) Doh, I.; Cho, Y.-H. Sens. Actuators, A 2005, 121, 59−65. (38) Moon, H.-S.; Kwon, K.; Kim, S.-I.; Han, H.; Sohn, J.; Lee, S.; Jung, H.-I. Lab Chip 2011, 11, 1118−1125. (39) Demierre, N.; Braschler, T.; Muller, R.; Renaud, P. Sens. Actuators, B 2008, 132, 388−396. (40) Valero, A.; Braschler, T.; Demierre, N.; Renaud, P. Biomicrofluidics 2010, 4, 022807. (41) Kang, Y.; Cetin, B.; Wu, Z.; Li, D. Electrochim. Acta 2009, 54, 1715−1720. (42) Lewpiriyawong, N.; Yang, C.; Lam, Y. C. Electrophoresis 2010, 31, 2622−2631. (43) Wang, L.; Flanagan, L. A.; Jeon, N. L.; Monuki, E.; Lee, A. P. Lab Chip 2007, 7, 1114−1120. (44) Ling, S. H.; Lam, Y. C.; Kua, C. H. Microfluid. Nanofluid. 2011, 11, 579−591. (45) Jones, T. B. Electromechanics of Particles; Cambridge University Press: Cambridge, U.K., New York, 1995. (46) Morgan, H.; Green, N. G. AC Electrokinetics: Colloids and Nanoparticles; Research Studies Press: Philadelphia, PA, 2001. (47) Fuhr, G.; Glasser, H.; Muller, T.; Schnelle, T. Biochim. Biophys. Acta, Gen. Subj. 1994, 1201, 353−360. (48) Jen, C. P.; Chen, T. W. Biomed. Microdevices 2009, 11, 597−607.
ASSOCIATED CONTENT
S Supporting Information *
This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*Tel.: +65 6790 6957. Fax: +65 6795 4634. Email: MYClam@ ntu.edu.sg. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS L.S.H. gratefully acknowledges the financial support of Nanyang Technological University, in the form of a NTU Research Scholarship. The authors also thank Dr. Chan Wing Yue for valuable discussions on cell preparation.
■
REFERENCES
(1) Pohl, H. A. Dielectrophoresis: The Behavior of Neutral Matter in Nonuniform Electric Fields; Cambridge University Press: Cambridge, U.K., New York, 1978. (2) Gascoyne, P. R. C.; Vykoukal, J. Electrophoresis 2002, 23, 1973− 1983. (3) Hughes, M. P. Electrophoresis 2002, 23, 2569−2582. (4) Hughes, M. P. Nanotechnology 2000, 11, 124−132. (5) Lapizco-Encinas, B. H.; Rito-Palomares, M. Electrophoresis 2007, 28, 4521−4538. (6) Park, S.; Koklu, M.; Beskok, A. Anal. Chem. 2009, 81, 2303− 2310. (7) Yang, L.; Banada, P. P.; Bhunia, A. K.; Bashir, R. J. Biol. Eng. 2008, 2, 1−14. (8) Voldman, J.; Gray, M. L.; Toner, M.; Schmidt, M. A. Anal. Chem. 2002, 74, 3984−3990. (9) Gascoyne, P.; Mahidol, C.; Ruchirawat, M.; Satayavivad, J.; Watcharasit, P.; Becker, F. F. Lab Chip 2002, 2, 70−75. (10) Kua, C. H.; Lam, Y. C.; Rodriguez, I.; Yang, C.; Youcef-Toumi, K. Anal. Chem. 2007, 79, 6975−6987. (11) Albrecht, D. R.; Underhill, G. H.; Wassermann, T. B.; Sah, R. L.; Bhatia, S. N. Nat. Methods 2006, 3, 369−375. (12) Lin, R.-Z.; Ho, C.-T.; Liu, C.-H.; Chang, H.-Y. Biotechnol. J. 2006, 1, 949−957. (13) Kang, Y.; Li, D. Microfluid. Nanofluid. 2009, 6, 431−460. (14) Michelle, M. M.; Sarah, J. R. S.; Mark, A. H. Electrophoresis 2009, 30, 852−865. (15) Markx, G. H.; Pethig, R. Biotechnol. Bioeng. 1995, 45, 337−343. (16) Wang, X. B.; Huang, Y.; Burt, J. P. H.; Markx, G. H.; Pethig, R. J. Phys. D: Appl. Phys. 1993, 26, 1278−1285. (17) Ormerod, M. G. Flow Cytometry: A Practical Approach; Oxford University Press: Oxford, U.K., New York, 2000. (18) Breeuwer, P.; Abee, T. Int. J. Food Microbiol. 2000, 55, 193−200. (19) Kucsera, J.; Yarita, K.; Takeo, K. J. Microbiol. Methods 2000, 41, 19−21. (20) Sarma, K. D.; Ray, D.; Antony, A. Cytotechnology 2000, 32, 93− 95. (21) Armstrong, D. W.; He, L. Anal. Chem. 2001, 73, 4551−4557. (22) Becker, F. F.; Wang, X. B.; Huang, Y.; Pethig, R.; Vykoukal, J.; Gascoyne, P. R. C. J. Phys. D: Appl. Phys. 1994, 27, 2659−2662. (23) Becker, F. F.; Wang, X. B.; Huang, Y.; Pethig, R.; Vykoukal, J.; Gascoyne, P. R. C. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 860−864. (24) Talary, M. S.; Mills, K. I.; Hoy, T.; Burnett, A. K.; Pethig, R. Med. Biol. Eng. Comput. 1995, 33, 235−237. (25) Markx, G. H.; Huang, Y.; Zhou, X.-F.; Pethig, R. Microbiology 1994, 140, 585−591. (26) Yang, J.; Huang, Y.; Wang, X.-B.; Becker, F. F.; Gascoyne, P. R. C. Anal. Chem. 1999, 71, 911−918. 6470
dx.doi.org/10.1021/ac300079q | Anal. Chem. 2012, 84, 6463−6470
