Environ. Sci. Technol. 2009, 43, 5857–5863
Dielectrophoretic Separation of Airborne Microbes and Dust Particles Using a Microfluidic Channel for Real-Time Bioaerosol Monitoring HUI-SUNG MOON,† YUN-WOO NAM,‡ J A E C H A N P A R K , ‡ A N D H Y O - I L J U N G * ,† Laboratory of Biochip Technology, School of Mechanical Engineering, Yonsei University, 262 Seongsanno, Seodaemun-gu, Seoul 120-749, South Korea, and Emerging Center, SAIT, Samsung Electronics Co., Ltd., Mt. 14-1, Nongseo-dong, Giheung-gu, Yongin-si Gyunggi-do 446-712, South Korea
Received January 16, 2009. Revised manuscript received May 23, 2009. Accepted June 18, 2009.
Airborne microbes such as fungi, bacteria, and viruses are a threat to public health. Robust and real-time detection systems are necessary to prevent and control such dangerous biological particles in public places and dwellings. For direct and real-time detection of airborne microbes, samples must be collected and typically resuspended in liquid prior to detection; however, environmental particles such as dust are also trapped in such samples. Therefore, the isolation of target bacteria or a selective collection of microbes from unwanted nonbiological particles prior to detection is of great importance. Dielectrophoresis (DEP), the translational motion of charge neutral matter in nonuniform electric fields, is an emerging technique that can rapidly separate biological particles in microfluidics because low voltages produce significant and contactlessforcesonparticleswithoutanymodificationorlabeling. In this paper, we propose a new method for the separation of airborne microbes using DEP with a simple and novel curved electrode design for separating bacteria in a solution containing beads or dust that are taken from an airborne environmental sample. Using this method, we successfully isolated 90% of the airborne bacterium Micrococcus luteus from a mixture of bacteria and dust using a microfluidic device, consisting of novel curved electrodes that attract bacteria and repel or leave dust particles. As there has been little research on analyzing environmental samples using microfluidics and DEP, this work describes a novel strategy for a rapid and direct bioaerosol monitoring system.
Introduction The real-time detection of microorganisms such as fungi, bacteria, and viruses in the air is an emerging and rapidly evolving field of research because of their ability to cause respiratory and other health disorders (1). The spread of airborne microbes such as measles, anthrax, and influenza * Corresponding author phone: +82-2-2123-5814; fax: +82-2-3122159; e-mail:
[email protected]. † Yonsei University. ‡ Samsung Electronics Co., Ltd. 10.1021/es900078z CCC: $40.75
Published on Web 07/07/2009
2009 American Chemical Society
are often regarded as major public health threats because they cause severe infectious diseases with high mortality rates (2, 3). Furthermore, several pathogenic microbes can be used as biological weapons capable of immense destruction (4). Rapid and real-time detection systems are required for use as early warning systems (EWS) to prevent and control such dangerous biological particles in public places and dwellings. Conventionally, the detection of airborne bacteria has been achieved by colony counting and various biochemical assays (5). However, these methods require a long incubation time (over 24 h) and are labor intensive. Because rapid detection and identification of airborne microorganisms has become more important, these methods are no longer considered suitable for monitoring airborne microbes. For rapid and effective detection, several methods including immunoassays, molecular biological tests, and optical, and electrical methods have been developed over the past several decades (6-8). Most of these methods, however, still require a long detection time and/or various expensive chemical reagents and complicated equipment. Recently, electrical methods have been widely used to detect bacteria because of their rapid, exact, and sensitive signals. Electrical methods are based on the insertion of electrodes into the sample solution and the measurement of several electrical properties such as impedance, conductance, and capacitance (9-11). A quartz crystal microbalance (QCM) immunosensing system also has been used for rapid detection of bacteria, which is based on perturbation of the sensor surface when the detection occurs that leads to a change in the resonant frequency (12). Our research group previously demonstrated that airborne bacterial growth on the surface of a thin solid medium could be rapidly detected by impedimetric analysis using imbedded electrodes (13) and that an ATP bioluminescence transducer could be used as a biosensor for the real-time detection of airborne bacteria (14). To detect microbes directly in the air, we collected samples and then typically resuspended the samples in a liquid prior to detection. However, environmental particles such as dust, which can result in false positives or mask the ability of sensors, are also trapped in such samples. Therefore, the isolation of target bacteria or a selective collection of microbes from unwanted or nonbiological particles prior to detection is of great importance (15) for more precise analysis. For example, if one can separate pollen, one of the major interfering materials for bioaerosol monitoring, and target bacteria before detecting, monitoring will be much easier. Particles that have a larger size than bacteria can be eliminated by filters, but particles with a similar size cause concern. Microfluidics technology has several advantages over conventional benchtop systems such as custom design, reduced consumption of reagents and sample, lower waste generation, and increased analysis speed and portability (16). Cell separation technology using microfluidic devices has emerged as an efficient technology that allows the user to purify target cells from a variety of environmental and biological samples, which can then be isolated and collected for downstream testing. Numerous techniques have been developed to achieve this purpose, including magnetophoretic (17, 18), hydrodynamic (19, 20), and acoustic (21) methods. Dielectrophoresis (DEP), the translational motion of charge neutral matter caused by polarization effects in nonuniform electric fields (22), is a recently emerging technique that can separate cells in microfluidic devices rapidly because low voltages produce significant and conVOL. 43, NO. 15, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5857
tactless forces on cells without any modification or labeling (23). This technique has been used successfully for separating components of a variety of different samples, including viable and nonviable yeast cells (24, 25), polymer beads and cells (26), and beads of different sizes (27); this technique has also been used to fractionate cells according to their cell cycle phase (28). Moreover, because different dielectric properties result in different behavior, this technique has been used to characterize biophysical properties of cells (29, 30). Besides, cell patterning in hydrogels (31) and in heterogeneous structures (32) are also interesting applications of this technique. A large number of medical, biological, and chemical applications currently utilize microfluidics and DEP. However, these techniques have not been applied in environmental research. In this paper, we examine the feasibility of using a microdevice with a simple and novel curved electrode design to separate bacteria from an airborne environmental sample containing beads or dust. This is the first description of a microfluidic approach to separate particle dust from bacteria. Typically, the main difficulties encountered in dealing with dust when monitoring air samples are the separation and quantification of microorganisms. These problems were successfully overcome using this microdevice together with absorbance measuring methods. To fully analyze airborne particles, we required collection and condensation, separation, and detection devices (Figure 1). The device that we developed for the purpose of particle separation is described below and will be integrated into future environmental monitoring systems.
Materials and Methods Principle of Dielectrophoresis. When a microparticle, nanoparticle, or a biological cell is subjected to an electric field, a dipole is induced in the particle, and particle polarization occurs. If the particle is more polarizable than the medium, the particle moves to higher electric field regions, and this process is called positive dielectrophoresis (DEP). If the particle is less polarizable than the medium, the particle is repelled from the higher electric field regions, and this process is called negative dielectrophoresis (33, 34). The DEP force acting on a spherical particle can be described by the following eq (21) FDEP ) 2πεmr 3Re[fCM]∇|E 2 |
(1)
where r is the radius of particle, εm is the medium permittivity, Re[fCM] is the real part of the Clausius-Mossotti (CM) factor, and ∇|E2| is the gradient of the square norm of the electric field. The CM factor is a function of permittivities and conductivities of the particles and medium and is given by (22) fCM )
ε*p - ε*m ε*p + 2ε*m
(2)
where ε*p and ε*m are the complex permittivity of the particles and the medium, respectively. The complex permittivity is given by
ε* ) ε - j
σ ω
(3)
where ε is relative permittivity (or dielectric constant), σ is the electric conductivity, and ω is the angular frequency of the alternating current (ac) electric field. The real part of the CM factor determines whether the dielectrophoresis is positive or negative because other terms of FDEP are always positive. Design of Electrodes Integrated in the Microfluidic Device. In a microsystem, ac electric potential applied to the microelectrodes generates nonuniform electric fields, and higher electric field regions are originated at the edge of the electrodes. The electric fields generated in our device are shown in Figure 2 as a cross sectional view; we can use the electrode edges to attract or repel particles. The microfluidic separator contains curved Au/Ti electrodes on a Si/SiO2 wafer and a polydimethylsilooxane (PDMS) channel (330 µm width, 30 µm height, and 11 mm length for the separation zone). The radius of curvature is 459.49 mm, and the tangential angle of curve varies from -0.79° (near the inlet) to 0.79° (near the outlet). The interelectrode distance is 25 µm, and the total length of the electrodes is 11 mm.The microchannel is composed of three inlets (one for the sample mixture and two for the buffer sheath flow) and two outlets. Figure 3 illustrates the working principle of the separator by showing the fluidic trajectory that the particles can take. Particles that experience a negative DEP will be repelled from the edge of the electrodes, and particles that experience a positive DEP will be attracted to the edge of the electrodes. Therefore, we can separate particle type I from particle types II and III. The sample flow stream is focused by sheath flows to meet the interelectrode area where the tangential angle is smallest (0°) to achieve most effective separation. The motion of complex particle mixtures like dust is difficult to characterize and predict using microfluidics. Our device was designed to specifically isolate target particles, namely, bacteria, using positive DEP from other particles such as dust, which are sorted to other outlets because they either show negative or negligible DEP. Other common electrode designs such as slanted electrodes, use only negative DEP forces to separate particles. However, if one type of particle shows a negative DEP while another shows a positive DEP, both particles move along the electrode edge causing difficulty in separation using the continuous flow-through method. Therefore, our curved shape electrode design provides more versatility and is more suitable for separating complex particles like dust from bacteria. Microfabrication of DEP Electrodes for the Integrated Microfluidic Device. The dielectrophoretic sorter is composed of a microfluidic channel and bottom electrodes. Bottom electrodes were fabricated on a 4 in. Si/SiO2 wafer using a conventional photolithography process. Ti/Au layers were deposited on the wafer for electrodes with a 20 Å /500 Å thickness using sputter. Photoresist (PR, AZ1512) was spincoated to 1 µm thickness, then exposed under UV light, and developed to define the electrode patterns. Au developer and BOE (buffered oxide etchant) were used successively to
FIGURE 1. General scheme for an indoor bioaerosol monitoring system. Air samples must first be collected and condensed using a negative pumping system to detect bioaerosols using a microsystem. After sample collection, target particles need to be isolated before sensing to enhance efficiency and specificity using a separation device. Finally, the detection device rapidly measures pretreated samples using optical, electrical, or chemical methods. The separation method is the focus of this paper. 5858
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 15, 2009
FIGURE 2. Numerical simulation of the electric field distribution over the electrodes at the cross section A-A′ of Figure 3. The E field was modeled using Comsol Multiphysics 3.2b with the electromagnetics module.
FIGURE 3. Schematic diagram of the DEP separator. Trajectories of particles that experience a positive DEP force, negative DEP force, and negligible DEP force are shown. Particle type I, which experiences a positive DEP force, enters the channel and deflects to outlet I because a positive DEP force guides particle type I through the edge of the electrodes (high electric field region). Particle type II, which experiences a negative DEP force, enters the channel and goes directly to outlet II because the negative DEP force repels particles from the edge of the electrodes. Particle type III, which receives a negligible DEP force, goes straight along the channel and exits to outlet II. define electrodes, and acetone was used to remove the patterned PR. Microfluidic channels were molded by a soft lithography process. SU-8 2025 (Microchem Corp) was spin-coated to a 30 µm height, which is equal to the channel depth, on Si wafers. After the soft-baking process, the wafer was exposed to UV light followed by additional heat treatment for the post exposure baking process. A SU-8 layer was etched using SU-8 developer and was rinsed using isopropyl alcohol. A polydimethylsilooxane (PDMS) prepolymer mixture (PDMS: curing agent ) 10:1 w/w, Dow Corning) was poured on the SU-8 mold and degassed using a vacuum chamber. The PDMS was then cured for 2 h at 80 °C in an oven and was peeled off. The PDMS molds were diced by razor blade, and inlet and outlet holes were made with a punch. Finally, the PDMS microchannel was assembled with bottom electrodes patterned on a Si/SiO2 substrate after O2 plasma treatment for 30 s and stored overnight in an oven to ensure permanent
bonding. Wires were attached to the electrodes using silver paste. The fabrication processes of electrodes and channel are shown in panels a and b of Figure S1 of the Supporting Information, respectively, and the fabricated device is shown in Figure 4. Sample Preparation. Polystyrene microbeads 0.71 µm in diameter and 2 µm in diameter (R700 and 4202A, respectively) were purchased from Duke Scientific Corporation. Deionized (DI) water with 0.18 mS/m conductance was used as media during separation experiments. The bacteria sample used in this experiment was obtained from the Korean Culture Center of Microorganisms (KCCM). Micrococcus luteus, a gram positive, spherical, commonly found airborne bacterium was obtained and grown in nutrient culture media at 37 °C. Bacterial growth was monitored using a spectrophotometer and harvested in midlog phase (optical density of 0.53), corresponding to about 1.33 × 108 cells per milliliter. Before being injected into the device, the bacterial sample was VOL. 43, NO. 15, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5859
FIGURE 4. Photograph of the microfabricated microfluidic device. The device contains a PDMS microchannel and Au/Ti electrodes on a Si/SiO2 wafer. Electrodes were wired using silver paste to receive an ac potential from an arbitrary/function generator. centrifuged, and the culture media was replaced with DI water. This process was repeated five times to ensure a complete change of media. Dust particles were obtained from an air conditioner filter and suspended in DI water (σ ) 0.18 mS/m). The supernatant of the suspension was used to obtain small particles only to prevent channel blocking. The concentration of the dust sample was measured by mass balance. Prepared dust suspension was kept in a freezer and aliquoted before use. Experimental Setup. Two syringe pumps (KDS 300, KD Scientific) were connected to the microfluidic channel to supply particle mixtures and DI water for each. The syringes and channel inlets were linked by Teflon tubes (Nano Port). The device was placed on an optical microscope and monitored by a CCD camera. An arbitrary/function generator (AGF310, Tektronix) delivered ac potential to the device (see Figure S2 of the Supporting Information). A hemocytometer was used to count bacteria and beads. A spectrophotometer (T60U Spectrometer, PG Instrument, Ltd.) was used to determine the concentration of dust.
Results and Discussion DEP Responses of Particles. To optimize separation conditions, we measured the DEP response of various particles using simple interdigitated electrodes. Sample particles in DI water with 0.18 mS/m conductance were dropped onto the interdigitated electrodes, and various electric field frequencies were applied with a 5 V amplitude. The response of several airborne bacteria, 0.71 µm polymer beads, 2 µm polymer beads, and dust particles against DEP forces were tested, and the results are shown in Table S1 of the Supporting Information. All bacterial cells showed a similar response, and from 500 kHz to 5 MHz they all showed a positive DEP. Therefore, M. luteus was used as the representative airborne biological particle in this paper. M. luteus showed a negative response (i.e., negative DEP) below 50 kHz and a positive response (i.e., positive DEP) from 70 kHz to 10 MHz. The 0.71 µm diameter polymer beads also showed a negative response below 50 kHz and a positive response from 100 kHz to 1 MHz. The 2 µm polymer beads showed a negative response below 10 MHz and a positive response at high frequencies (100 MHz). Dust did not show any significant and consistent movement at any frequency examined. 5860
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 15, 2009
These results indicate that 0.71 and 2 µm polymer beads can be separated using DEP at 1 MHz as the beads show different responses under this condition. Furthermore, M. luteus and 2 µm polymer beads can be separated in the same manner, and M. luteus can be separated from dust particles as well. Separation of Different Beads. The separation performance of the device was first characterized with a 0.71 and 2 µm polymer bead mixture in DI water. Results of the continuous separation are summarized in Figure 5. The initial mixture containing 3.3 × 106 particles/mL of each bead was injected into the channel (sheath flow was also calculated), and the 0.71 and 2 µm polymer beads were successfully separated into two different outlets. Flow rates of streams (sum of main flow rate and sheath flow rate) were tested at 0.5, 2.5, and 5 µL/min, and the applied voltage was 5 V, 1 MHz. To prevent particle trapping at the edge of the electrode, we applied the ac voltage with a 200 ms interval, which means ac voltage was applied for 100 ms and not applied for another 100 ms. Using this mode, we avoided particle adhesion on the electrode edge due to a high positive DEP force (24). As mentioned above, 0.71 µm diameter polymer beads showed a positive response and moved like particle type I in Figure 3, while 2 µm polymer beads moved like particle II. When the flow rate was low (0.1 µL/min for the main flow and 0.4 µL/min for the sheath flow), we could obtain 96.2% purity for 0.71 µm beads and 90.7% purity for 2 µm beads in each outlet. The purity of the separated beads decreased when the flow rate increased. Separation of Bacterial Cells and Beads. The feasibility of the device was tested with 2 µm beads and bacterial cells mixed in DI water. A particle mixture of 3.3 × 106 particles/ mL of beads and bacterial cells was introduced into the microchannel; these exited through different outlets successfully. The flow rates were 0.1 µL/min for the main flow and 0.4 µL/min for the sheath flow. The applied voltage was 5 V with 1 MHz in the burst gate mode with a 200 ms interval. Because the bacteria showed the same trajectory as particle type I in Figure 3, while the 2 µm beads had a similar trajectory to particle type II, we could separate bacteria from nonbiological particles. Figure 6 shows the separation results. The purity of the beads in outlet I was 96 ( 3.7%, and the purity of the bacteria in outlet II was 90 ( 4.0%. The overall recovery
FIGURE 5. Experimental results of dielectrophoretic separation of polymer beads under various flow rates. (Error bars indicate standard deviations.) Purity is defined by the number of target particles over the total number of particles in each outlet. Recovery is the percentage of particles successfully sorted from the total number of target particles within the initial fraction.
FIGURE 6. Experimental results of dielectrophoretic separation of bacteria and 2 µm polymer beads (Error bars indicate standard deviations). of bacteria and beads was 90 ( 4.5% and 92 ( 2.7%, respectively.
Separation of Bacteria and Dust. A particle mixture containing 0.03 g/mL of dust and 3.3 × 106 cells/mL of VOL. 43, NO. 15, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5861
the small amount of bacteria and dust particles that adhered to the fluidic tubing device and the microchannel walls. Our purpose in this research is to isolate the live airborne bacteria from the dust particles and deliver them onto the surface of sensor chip that is being developed by our group. We need to separate the bacteria from the dust sample to efficiently detect bacteria in the biosensor part as it is quite often observed that impurities such as dust interfere with the electric signal of the sensor. In conclusion, we could isolate bacteria in collected samples, including microbes and nonbiological materials such as dust, using the asysmetric electrode integrated microfluidic device. This system can be used as a continuous flow-through separation system for various particles and utilized as a pretreatment technique for microbe detection.
Acknowledgments This work was supported by grants from the Korea Institute of Environmental Science and Technology (Grant 101-082035), the Samsung Advanced Institute of Technology, and in part by the Basic Research Program of the Korea Science & Engineering Foundation (2008-05943). We also wish to thank Professor Hongseok (Moses) Noh at Drexel University for helpful discussions on DEP.
Supporting Information Available Investigation of the Joule heating effect and cell viability assay (Figure S1), fabrication processes (Figure S2), schematic of the experimental setup of the separation device (Table S1), and DEP response of various particles in DI water. This information is available free of charge via the Internet at http://pubs.acs.org.
Literature Cited FIGURE 7. (a) Regression of absorbance on dust concentration in various wavelengths. (b) Experimental results of bacteria and dust separation. Concentration of 1 Abs is equivalent to 0.03 g/ mL. (Error bars indicate standard deviations). bacteria was subjected to the microdevice for separation. As dust is a complex mixture of particles, these particles could not be quantified using a microscope and hemocytometer. Thus, a spectrophotometer was used to estimate the concentration of dust by measuring the absorbance. To determine the correlation between dust concentration and absorbance, we scanned various concentrations of dust over a range of wavelengths from 200 to 1100 nm. The absorbance at 400 nm was found to reflect the actual concentration of dust better than other wavelengths tested as the absorbance at 400 nm increased steeply as the concentration of dust increased (Figure 7a). The regression of absorbance on dust concentration at 400 nm is shown in Figure 7b where the Y axis and X axis indicate concentration and absorbance, respectively. Dots represent absorbance at a certain concentration, and the red colored line represents the regression line (slope of the regression is 4.512). The initial sample mixture was measured at 1.715 Abs, and the sample collected in outlet 1 showed 0.225 Abs. These results reflect that the dust mixture was diluted 7.6-fold as it passed through the DEP separator, and therefore, the recovery of dust was estimated to be 86.8% (Figure 7b). The recovery of bacteria was measured as 89.6% using a microscope and hemocytometer. Bacterial particles and dust particles were separated successfully, similar to bacterial particles and bead separation, although with slightly lower purity and recovery rates. However, the lower rates can be partially explained by the aggregation of bacterial cells and dust particles as well as a 5862
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 15, 2009
(1) Textbook of Diagnostic Microbiology, 2nd ed.; Mahon, C. R., Manuselis, G., Eds.; W. B. Saunders: Philadelphia, PA, 2000; pp 877-918. (2) Fiegel, J.; Clarke, R.; Edwards, D. A. Airborne infectious disease and the suppression of pulmonary bioaerosols. Drug Discovery Today 2006, 11, 51–57. (3) Henderson, D. A. The looming threat of bioterrorism. Science. 1999, 283, 1279–1282. (4) Hilleman, M. R. Overview: Cause and prevention in biowarfare and bioterrorism. Vaccine 2002, 20, 3055–3067. (5) Griffiths, W. D.; DeCosemo, G. A. L. The assessment of bioaerosols: a critical review. J. Aerosol Sci. 1994, 25, 1425–1458. (6) Bogdanovic, J.; Koets, M.; Sander, I.; Wouters, I.; Meijster, T.; Heederik, D.; Amerongen, A. V.; Doekes, G. Rapid detection of fungal R-amylase in the work environment with a lateral flow immunoassay. J. Allergy Clin. Immunol. 2006, 118, 1157–1163. (7) Pyankov, O. V.; Agranovski, I. E.; Pyankova, O.; Mokhonova, E.; Mokhonov, V.; Safatov, A. S.; Khromykh, A. A. Using a bioaerosol personal sampler in combination with real-time PCR analysis for rapid detection of airborne viruses. Environ. Microbiol. 2007, 9, 992–1000. (8) Sengupta, A.; Brar, N.; Davis, E. J. Bioaerosol detection and characterization by surface-enhanced Raman spectroscopy. J. Colloid Interface Sci. 2007, 309, 36–43. (9) Muhammad-Tahir, Z; Alocilja, E. C. A conductometric biosensor for biosecurity. Biosens. Bioelectron. 2003, 18, 813–819. (10) Sherry, A. E.; Patterson, M. F.; Kilpatrick, D.; Madden, R. H. Evaluation of the use of conductimetry for the rapid and precise measurement of Salmonella spp. growth rates. J. Microbiol. Methods 2006, 67, 86–92. (11) Suehiro, J.; Yatsunami, R.; Hamada, R.; Hara, M. Quantitative estimation of biological cell concentration suspended in aqueous medium by using dielectrophoretic impedance measurement method. J. Phys. D: Appl. Phys. 1999, 32, 2814–2820. (12) Lee, S. H.; Stubbs, D. D.; Cairney, J.; Hunt, W. D. Rapid detection of bacterial spores using a quartz crystal microbalance (QCM) immunoassay. IEEE Sens. J. 2005, 5 (4), 737–743. (13) Choi, A.; Park, J. S.; Jung, H. I. Solid medium-integrated impedimetric biosensor for real-time monitoring of microorganisms. Sens. Actuators, B. 2009, 137, 357–362.
(14) Lee, S. J.; Park, J. S.; Im, H. T.; Jung, H. I. A microfluidic ATPbioluminescence sensor for the detection of airborne microbes. Sens. Actuators, B 2008, 132, 443–448. (15) Fatoyinbo, H. O.; Hughes, M. P.; Martin, S. P.; Pashby, P.; Labeed, F. H. Dielectrophoretic separation of Bacillus subtilis spores from environmental diesel particles. J. Environ. Monit. 2007, 9, 87–90. (16) Ng, J. M.; Gitlin, I.; Stroock, A. D.; Whitesides, G. M. Components for integrated poly(dimethylsiloxane) microfluidic systems. Electrophoresis 2002, 23, 3461–3473. (17) Xia, N.; Hunt, T. P.; Mayers, B. T.; Alsberg, E.; Whitesides, G. M.; Westervelt, R. M.; Ingber, D. E. Combined microfluidicmicromagnetic separation of living cells in continuous flow. Biomed. Microdevices 2006, 8, 299–308. (18) Kim, H. S.; Son, O. T.; Kim, K. H.; Kim, S. H.; Maeng, S. L.; Jung, H. I. Separation of apoptotic cells using microfluidic device. Biotechnol. Lett. 2007, 29, 1659–1663. (19) Huang, L. R.; Cox, E. C.; Austin, R. H.; Sturm, J. C. Continuous particle separation through deterministic lateral displacement. Science. 2004, 304, 987–990. (20) Choi, S.; Park, J. K. Continuous hydrophoretic separation and sizing of microparticles using slanted obstacles in a microchannel. Lab Chip 2007, 7, 890–897. (21) Petersson, F.; Nilsson, A.; Holm, C.; Jo¨nsson, H.; Laurell, T. Continuous separation of lipid particles from erythrocytes by means of laminar flow and acoustic standing wave forces. Lab Chip 2005, 5, 20–22. (22) Pohl, H. A. Dielectrophoresis: The Behavior of Neutral Matter in Nonuniform Electric Fields. Cambridge University Press: Cambridge, U.K., 1978. (23) Voldman, J. Electrical forces for microscale cell manipulation. Annu. Rev. Biomed. Eng. 2006, 8, 425–54. (24) Doh, I.; Cho, Y. H. A continuous cell separation chip using hydrodynamic dielectrophoresis (DEP) process. Sens. Actuators, A 2005, 121, 59–65. (25) Li, Y.; Dalton, C.; Crabtree, H. J.; Nilsson, G.; Kaler, V. I. S. Continuous dielectrophoretic cell separation microfluidic device. Lab Chip 2007, 7, 39–248.
(26) Wang, L.; Flanagan, L. A.; Jeon, N. L.; Monuki, E.; Leea, A. P. Dielectrophoresis switching with vertical sidewall electrodes for microfluidic flow cytometry. Lab Chip 2007, 7, 1114– 1120. (27) Kralj, J. G.; Lis, M. T. W.; Schmidt, M. A.; Jensen, K. F. Continuous dielectrophoretic size-based particle sorting. Anal. Chem. 2006, 78, 5019–5025. (28) Kim, U.; Shu, C.; Dane, K. Y.; Daugherty, P. S.; Wang, J. Y. J.; Soh, H. T. Selection of mammalian cells based on their cell-cycle phase using dielectrophoresis. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 20708–20712. (29) Sanchis, A.; Brown, A. P.; Sancho, M.; Martinez, G.; Sebastian, J. L.; Munoz, S.; Miranda1, J. M. Dielectric characterization of bacterial cells using dielectrophoresis. Bioelectromagnetics 2007, 28, 393–401. (30) Hoettges, K. F.; Dale, J. W.; Hughes, M. P. Rapid determination of antibiotic resistance in E. coli using dielectrophoresis. Phys. Med. Biol 2007, 52, 6001–6009. (31) Suzuki, M.; Yasukawa, T.; Shiku, H.; Matsue, T. Negative dielectrophoretic patterning with colloidal particles and encapsulation into a hydrogel. Langmuir 2007, 23 (7), 4088– 4094. (32) Ho, C.; Lin, R.; Chang, W.; Chang, H.; Liu, C. Rapid heterogeneous liver cell on-chip patterning via the enhanced field-induced dielectrophoresis trap. Lab Chip 2006, 6, 724–734. (33) Hughes, M. P. AC electrokinetics: Applications for nanotechnology. Nanotechnology. 2000, 11, 124–132. (34) Wong, P. K.; Wang, T. H.; Deval, J. H.; Ho, C. M. Electrokinetics in microdevices for biotechnology applications. IEEE/ASME Trans. Mechatronics 2004, 9, 366–376. (35) Jones, T. B. Influence of scale on electrostatic forces and torques in AC particulate electrokinetics. IEE Proc.: Nanobiotech. 2003, 150, 39–46. (36) Castellanos, A; Ramos, A.; Gonzalez, A.; Green, N. G.; Morgan, H. Electrohydrodynamics and dielectrophoresis in microsystems: Scaling laws. J. Phys. D: Appl. Phys. 2003, 36, 2584–2597.
ES900078Z
VOL. 43, NO. 15, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5863
