Sensing methods for dielectrophoresis phenomenon - IEEE Xplore

Feature

Sensing Methods for Dielectrophoresis Phenomenon:

From Bulky Instruments to Lab-on-a-Chip Yehya Ghallab and Wael Badawy

Introduction n recent decades, dielectrophoresis (DEP) [1] has been a fairly well-known phenomenon in which a spatially nonuniform electric field exerts a net force on the field-induced dipole of a particle. Particles with higher polarizability than the surrounding medium experience positive dielectrophoresis, they move toward regions of the highest electric field concentration. Particles less polarized than the surrounding medium experience negative dielectrophoresis, and move towards regions of low electric field concentration. The force depends on the induced dipole and the electric field gradient, not on the particle’s charge. Thus, dielectrophoresis has been used to precipitate DNA

BACKGROUND ©EYEWIRE; LAB TOOLS © DIGITAL STOCK; DNA ©1999 PHOTO DISC, INC.; BEAKERS ©DIGITAL VISION; MICROSCOPE LENS ©CORBIS CORP.

I

THIRD QUARTER 2004

1531-6364/04/$20.00©2004 IEEE

IEEE CIRCUITS AND SYSTEMS MAGAZINE

5

and proteins [2], to manipulate bacteria [3], and to manipulate and separate cells [4], and subcellular components such as microtubules [5]. Also, DEP is technologically important in its own right, as evidenced by the number of applications in such scientific and technical fields as biophysics, bioengineering, and mineral separation [6]. As an example can serve the cell fusion [7], which is important in cancer treatment; in this process, the nonuniform electric field collects some fractions of cells on electrode surfaces where cells of the two types inevitably encounter one another and form chains. Then a serious of short DC pulses is applied to the electrodes. The strong DC field disturbs the membranes in the region of contact between cells and initiate their merger or fusion. A potential application of this technique is the production of antibodies useful in cancer research and treatment. Moreover, the utility of DEP has also been extended to some other research areas such as environment (mineralogical separation) and food industry [8]. On the other hand, lab-on-a-chip technology is exciting the interest of scientists in many areas. This technology can be used not only to synthesize efficiently and economically chemicals, biological or cancer cells, and DNA but also to carry out biological and clinical analyses, to perform combinatorial chemistry, and to carry out fullscale analyses from sample introduction to cell separation, manipulation and detection, on a single, miniaturized device [9, 10]. Miniaturization of biological analysis and synthesis will improve throughput, performance and accessibility, and lead to significantly reduced costs [11–13]. To merge the robustness of both lab-on-a-chip and DEP phenomenon, recently, lab-on-a-chip based on DEP phenomenon has appeared [14–16]. This lab-on-a-chip needs to integrate functions such as: actuation, sensing, and processing, to increase their effectiveness. However, to date there is still an unmet need for lab-on-a-chip to deal effectively with biological systems at the cell level. The rest of the paper is organized as follows: second part reviews some fundamental aspects of the dielectrophoresis phenomenon. Sensing methods for DEP phenomenon and the state of the art of the lab-ona-chip based DEP phenomenon are presented in the third part. Dielectrophoresis Fundamentals Dielectrophoresis (DEP) [1, 17] is defined as the motion of an uncharged (neutral) particle caused by polarization effect in a nonuniform electric field. The main features of the DEP phenomenon can be summarized as follows:

I. Particles experience DEP force only when the electric field is nonuniform. II. The DEP force does not depend on the polarity of the applied electric field and is observed with AC as well as DC excitation. III. There are two kinds of DEP forces: • Positive DEP for εm < εp , where εm is the permittivity of the suspended medium and εp is the permittivity of the particles. In this case, particles are attracted to regions of stronger electric field. • Negative DEP for εm > εp . In this case, particles are repelled from regions of stronger electric field. IV. DEP is most readily observed for particles with diameters ranging from approximately 1 to 1000 µm. DEP should be contrasted with electrophoresis, where one manipulates charged particles in a dissipative medium with electric fields [18], as there are several important differences. First, DEP does not require the particle to be charged in order to manipulate it; the particle must only differ electrically from the medium that it is in. Second, DEP works with AC fields, whereas no net electrophoretic movement occurs in such a field. Thus, with DEP one can avoid problems such as electrode polarization effects [13] and electrolysis at electrodes. Even more importantly, the use of AC fields reduces membrane charging of biological cells [17]. Membrane charging is due to the potential developed across cell membranes in electric fields. This potential, which can impact cell physiology, can be diminished by applying of high-frequency fields. Finally, DEP forces increase with the gradient of the 2 (described square of the electric field, i.e. ∇| E| below), whereas electrophoretic forces increase linearly with the electric field. Figures 1 and 2 illustrate these differences. Fig. 1 shows a uniform electric field applied to two bodies, one of them is neutral and the second is charged. In a uniform electric field, a charged body is pulled toward the electrode carrying the charge opposite to that on the particle. In the same field, a neutral body will be polarized merely. The result may produce a torque, but not a translational force, i.e. the body will not move toward either electrode. In Fig. 2, where a nonuniform electric field is shown, we can observe different behavior of the charged and neutral bodies. The charged one is still attracted toward the electrode of opposite polarity. The neutral body, in this case, will find a translational force act upon it.

The authors are with the Department of Electrical and Computer Engineering, University of Calgary, 2500 University Drive, N.W, Alberta, T2N 1N4, Canada. Email: {ghallab, badawy}@enel.ucalgary.ca

6


THIRD QUARTER 2004

This occurs as a result of the polarization, polarization causes a physical displacement of charges in the dielectric [1], which happens under the influence of the nonuniform electric field. The polarization has the effect of putting a negative charge upon the side nearer the positive electrode, and a positive one on the side nearer the negative electrode. Because the particle is neutral, the two charges on the body are opposite but equal. Dielectrophorises Force (DEP Force) To obtain an estimated expression of the net force upon a small physical dipole [17], let us consider a dipole consisting of equal and opposite charges +q and −q located a vector distance d apart, under the effect of a nonuni r ), as shown in Fig. 3. The dipole form electric field E( moment p is defined as the result of the multiplication of Assuming the charge q and the distance d (i.e. p = qd). is very small in comparison to the characteristhat |d| tic dimension of the electric field nonuniformity, and the r ) includes no contributions due to the electric field E( dipole itself. r ) is nonuniform, Because the electric field applied E( then, in general, the two charges (+q and −q) will expe r ) and rience different values of the vector field, E( r + d), respectively, and the dipole will experience a E( net force. The charge +q experiences r + d) F+ = qE(

Neutral Body Merely Polarized

Figure 1. Uniform electric field applied to neutral and charged bodies.

A

Charged Body Moves Along Field Lines

+

(+)

(–) – – –

Neutral Body Polarized and Pulled Towards Strongest Field Region

+ + +

B

(1)

while the charge −q experiences r) F− = −qE(

Charged Body Moves Along Field Lines

Figure 2. Nonuniform electric field applied to neutral and charged bodies.

(2) d3 , and higher, in (4) have been neglected. Putting (4) into (3), we can simplify (3); that is:

The net force on the dipole is: E F = qd.∇ r + d) − qE( r) F = qE(

(3)

(5)

As p = qd (the dipole moment), so the force acting on an infinitesimal dipole is expressed by the formula [19]:

Using Taylor series expansion,

r + d) = E( r ) + d.∇ E( r ) + higher additional terms (4) E(

is very small in Taking into consideration that |d| comparison to the characteristic dimension of the electric field nonuniformity, the additional terms, of order d2 , THIRD QUARTER 2004

Fdipole = p · ∇ E

(6)

Similarly, if we have a neutral particle suspended in some dielectric fluid and polarized by a nonuniform electric field E , due to polarization, E will induce a moment in the particle and a net electric force will be given as: IEEE CIRCUITS AND SYSTEMS MAGAZINE

7

F = (peff · ∇) E

where peff is the effective dipole moment, and it is defined as the moment of an equivalent, charge-free, point dipole that when immersed in the same dielectric fluid and placed at the same location as the center of the original particle, produces the same dipolar electrostatic potential. In the case of a sphere in a medium, both of which are lossless dielectrics, i.e. both materials have finite dielectric constant and zero conductivity (ideal case), the induced effective dipole moment peff is given by: peff = 4π a3 εm

εp − εm εp + 2εm

E

(8)

where a is the radius of the sphere, εm is the permittivity of the medium, εp is the permittivity of the sphere, and E is the strength of the applied electric field. Putting (8) into (7), we get that the net force on the sphere is:

F = 2π a3 εm

εp − εm εp + 2εm

∗ εm = εm − j

(7)

2 ∇| E|

(9)

ε −ε The term εpp+ 2εmm is known as the Clausius-Mossotti factor (K). In the case of a non-ideal situation, when ohmic losses are taken into consideration, a homogenous spherical particle with radius a and complex permittivity ω, when immersed in a medium of complex per∗ and exposed to a specially nonuniform AC mittivity εm electric field, the time average DEP force FDE P is defined as follows:

σm , ω

(13)

where σp and σm are the conductivity of the particle and the medium, respectively, and ω is the angular frequency of the applied electric field. From equation (10), we can observe the following: I. The DEP force is linearly related to the volume of the immersed particle (i.e. FDEP ∝ a 3 ). II. FDEP is related directly to the permittivity of the suspended medium (i.e. FDEP ∝ εm ). III. The DEP force depends upon the magnitude and sign of the Clausius-Mossotti factor. IV. The DEP force vector is directed along the gradient of 2 , and the time averthe electric field intensity ∇ Erms age DEP force is independent of the field polarity, as we are taking the rms value of the intensity of the electric field. Positive and Negative DEP Forces According to equation (10), the real part of the ClausiusMossotti factor (Re[K ]) determines the direction of the DEP force. When Re[K ] is positive, (i.e. εm < εp ), particles are attracted to the electric field intensity maxima and repelled from the minima. This force is known as positive dielectrophoresis. The negative dielectrophoresis occurs when Re[K ]< 0 (i.e. εm < εp ), particles are attracted to the electric field intensity minima and repelled from the maxima. The most interesting dielectrophoretic behavior is exhibited by particles with frequency-dependent dispersion due to dielectric or conductive loss. Putting (12) and (13) into (11), we get the frequency dependent form of Re[K ] [17]:

Re[K ] =

εp − εm 3(εm σp − εp σm ) (14) + 2 εp + 2εm τMW (σp + 2σm )2 1 + ω2 τMW ε +ε

2 FDE P = 2π a3 εm Re[K ]∇ Erms

(10)

where τMW = σpp+ 2σmm and it is the Maxwell-Wagner charge relaxation time. The high and low frequency limits for Re[K ] will be identified as:

where K=

∗ εp∗ − εm ∗ ∗ εp + 2εm

8


σp , ω

σp − σm , for ω → 0 σp + 2σm

(15)

Re[K ] =

εp − εm , for ω 0 εp + 2εm

(16)

(11)

∗ are is the complex Clausius-Mossotti factor, and εp∗ and εm defined by equations (12) and (13), respectively,

εp∗ = εp − j

Re[K ] =

(12)

From (15) and (16) we obtain that the conductivity dominates the low frequency DEP behavior, while the permittivity dominates the high frequency behavior. Fig. 4 and Fig. 5 show schematic diagrams representing THIRD QUARTER 2004

the relations between Re[K ] and the radian frequency ω for two different cases. The first case is presented in Fig. 4, where σp < σm and εp > εm ; Re[K ] is negative at low frequencies and positive at high frequencies. The reverse case is shown in Fig. 5, where σp > σm and εp < εm , then Re[K ] becomes positive at low frequencies and negative at high frequencies. Obviously, the frequency dependence of Re[K ] gives a frequencydependent, time average DEP force FDE P acting on a homogenous particle, [10]. In other words, both the magnitude and sign of FDE P are functions of the electric field frequency ω. For example, in Fig. 4, where σp > σm and εp < εm , the particle will be attracted to the electric field intensity minima at low frequencies and to the maxima at high frequencies. The crossing frequency ωc (see Fig. 4 and Fig. 5), at which no DEP force acts on a particle (i.e. Re[K ] = 0), and the DEP force changes from positive to negative or vice versa, can be calculated as: [17] ωc =

(σm − σp )(σp + 2σm ) (εp − εm )(εp + 2εm )

E(r+d) Dipole +q +

E(r) –q

–

y

x z

Figure 3. Representation of an elementary dipole in a nonuniform electric field.

(17) Re [K (ω)]

It can be deduced that ωc is defined, (i.e. the DEP spectrum crosses the x-coordinate axis), only when (σm − σp ) (εp − εm ) > 0.

+DEP ωc 0

Lab-On-A-Chip Based DEP Phenomenon As we have mentioned before, lab-on-a-chip based on the DEP phenomenon, is one of the hottest areas of research these days. It has many applications in biology, pharmaceutics, medicine, and environmental sciences. These applications are characterized by complex experimental protocols, which require both microorganism detection and manipulation. Hence lab-on-a-chip needs to integrate such functions as actuation, sensing, and processing to increase their effectiveness. On the other hand, lab-on-a-chip holds the promise of cheaper, better and faster biological analysis. However, until now there is still an unmet need for lab-on-a-chip to deal effectively with biological systems at the cell level. There are many techniques that have been used for sensing, analyzing and monitoring the behavior of the cells under the influence of a DEP force. They are optical technique [20–22], fluorescent labeling (e.g., microfabricated fluorescence-activated cells sorter (µFACS)) technique [23, 19], and impedance sensing technique [24–28]. The sensing parts in the currently used labson-a-chip are based on impedance sensing technique [29, 14], and optical technique (i.e., using photo detectors) [15, 16]. THIRD QUARTER 2004

σp < σm εp > εm –DEP ω Figure 4. Dielectrophoretic spectra for σp < σm and

εp > εm .

Re [K (ω)] +DEP ωc 0 σp > σm εp < εm –DEP ω Figure 5. Dielectrophoretic spectra for σp > σm and

εp < εm .


9

force [20]. The optical system used is shown in Fig. 6. It has an optical chamber, which incorpoSample Laser rates two microelectrode arrays Photodiode between which cell suspensions Read Only are subjected to nonuniform elecMemory tric fields, and consequently a Reference Photodiode DEP force. Dielectrophoretically, 8 Bits the induced motion of the colCounter Optical Chamber loidal particles is monitored as a Sample AC to DC change in the optical density of and Hold Converter Clock Log Ratio the suspension by measuring the Amplifier intensity of a light beam that passAC to DC Sample Converter and Hold es through the gap between the two electrode arrays. Conditions + Gain of positive DEP, where particles Set Digital to Analog Amplifier Chart Converter are removed from the bulk soluRecorder tion to collect at the edges of the +5V electrodes, are thus monitored as Digital to Analog a decrease in optical absorbance Converter DC Offset of the bulk solution. Negative DEP, where the particles are directed –5V into the low field regions in the Figure 6. The overall electronic design of the dual DEP spectrometer [20]. bulk solutions away from the electrodes, is detected as an increase Optical Technique in the optical absorbance. The light from a laser diode is In this method, a dual beam optical spectrometer has split into two parallel beams using half and full-mirrors as been used for rapid measurement of the dielectrophoretic shown in Fig. 6. The beam emerging from the half mirror is behavior of cells and other particles subjected to a DEP focused down by a means of a standard microscope into the chamber containing the colloidal sample, while the other beam is similarly focused into a second (reference) chamber that contains fluid of the same composition as Flourescence Detector Cell Suspension that used for the colloid suspending medium. This dual beam arrangement minimizes effects associated with fluctuations of light intensity and thus results in an improved signal-to-noise ratio compared with the other systems [21, Difference Amplifier

Difference Amplifier

Optical Chamber

Laser

Detector of Forward Scattered Light Electrodes

Figure 7. Schematic representation of the fluorescenceactivated cell sorter (FACS) [23].

10


Figure 8. Optical micrograph of the µFACS device [19].

THIRD QUARTER 2004

22]. The advantage of the optical technique is the ability to characterize the positive and negative DEP, which in turn provides a technique for deriving the practical conditions required for the selective manipulation and separation of cells using dielectrophoretic forces. The disadvantages of this technique from the lab-on-a-chip point of view can be summarized as follows: (a) it requires bulky and expensive equipment, (b) it needs complex sampling preparation, and (c) it is not suitable for miniaturization.

Pt Electrodes High Voltage Amplifiers

Microfabricated Cell-Sorting Device Computer

Coherent Innova Ar Laser 488 nm

60X, 1.4NA Oil Immersion Lens

Dichroic

Fluorescent Labeling Technique Focusing Optics Based on this technique, cell suspension Beam Splitter CCD Camera containing cells labeled with a fluorescent dye has been used. Cells can be characterized and can be sorted by detecting VIDEO Preamplifier their fluorescence [23]. In this technique, the fluorescence-activated cell sorter Photomultiplier Tube (FACS) machine is used [19]. This machine can rapidly separate cells in a suspension Figure 9. Schematic diagram of the cell sorting apparatus [19]. on the basis of their size and the color of their fluorescence; a schematic representation of the FACS is shown in Fig. 7. The FACS machine works as follows: a cell suspension containing cells labeled with a fluorescent dye is directed into a thin stream so that all the cells pass in single file. This stream Flow Profile emerges from a nozzle vibrating at some 40,000 cycles per Cell AC second which breaks the stream into 40,000 discrete Current Lines droplets each second, some of these may contain a cell. A laser beam is directed at the stream just before it breaks up into droplets. As each labeled cell passes through the Electrodes C B A beam, its resulting fluorescence is detected by a photocell. If the signals from the two detectors (i.e., fluorescence and forward scattered light detectors) meet either of the criteria set for fluorescence and size, an electrical Figure 10. Side schematic view of the microchannel [25]. charge (+ or −) is given to the stream. The droplets retain this charge, then as they pass between a pair of charged metal plates positively charged drops are attracted to the negatively charged plate and vice versa. Uncharged droplets (those that contain no cell or a cell that fails to Cell Signal ZAC - ZBC meet the desired criteria of fluorescence and size) pass ttr straight into a third container and are later discarded. The process does not damage the cells. In fact, because the machine can be set to ignore droplets con0.0 0.5 1.0 1.5 2.0 t(ms) taining dead cells, the percent viability of the sorted cells can be higher than that in the original suspension. Although FACS provide impressively efficient sorting, they are expensive, and mechanically complex. Inexpensive devices that rapidly sort live cells, particles, and Figure 11. Impedance difference signal [25]. even single molecules would greatly facilitate the screenTHIRD QUARTER 2004


11

ing of combinatorial chemistry libraries or cell populations during in vitro molecular Membrane evolution. Moreover, such devices would Cell Flow Profile Cm R R have wide applications in clinical mediSal2 c Cm cine and basic biological and materials Cytoplasm research. Recently, a disposable microfabCdl ricated fluorescence-activated cell sorter Cdl Electrodes Cdl RSal1 (µFACS) has been presented [24]. Compared with conventional FACS machines, C the µFACS provides higher sensitivity, no B A cross-contamination, and lower cost. The µFACS device is a silicone elasFigure 12. An electrical model of the impedance change [25]. tomer chip with three channels joined at a T-shaped junction (Fig. 8). The channels are sealed with a glass coverslip. A buffer solution is introduced at the input channel and fills the device by PDMS Microflodic Channel capillary action. The pressure is equalized by adding buffer to the two output ports and then adding a sample Inlet containing the cells to the input port. The cells are manipulated with an electro-osmotic flow, which is controlled Electrode1 Electrode1 L by three platinum electrodes at the input and output wells. The chip is mounted on an inverted optical microd V0 scope, and fluorescence is excited near the T shaped junction with a focused laser beam. The fluorescent emission is collected by the microscope and measured with a Outlet photomultiplier tube (PMT). A computer digitizes the PMT signal and controls the flow by the electro-osmotic Substrate potentials (Fig. 9). The advantages of the fluorescent labeling technique Figure 13. Schematic illustration of the microfluidic device [26]. are high sensitivity and impressive efficient sorting. While the disadvantages from the lab-on-a-chip point of view are (a) it requires cell modification by markers or antibody, and (b) the equipment is rather expensive, bulky, and complex to operate. Thus it’s not suitable for miniaturization. Upper Electrode

Micro-chamber

DEP Cage

Electrodes

Substrate

Electrode Potential Phases Figure 14. The proposed approach (in (2002)) for establishing DEP cage [16].

12


Impedance Sensing Technique Recently, it has been demonstrated that micro-machining technologies may alleviate some of the problems encountered in ‘macroscale’ flow-cytometry [24]. On-chip integration of microfluidics and microelectrodes has already been reported for stationary cell measurements, trapping and sorting [25]. The common feature of impedance flow cytometry devices is that they are all liable to measure the size of the cell under consideration. Thus, it has to be determined with a maximum precision as it can significantly influence the estimation of the other parameters of interest. Essentially, the measurement consists in determining the impedance change R due to the particle passing through an aperture placed between two electrodes. In a microchannel the aperture is inbuilt and channel walls appear to be the best solution to achieve a high R/R ratio by placing the electrodes as close as possible to each THIRD QUARTER 2004

RESCOL

other and concentrating the current lines on the cell. The vin measurements are realized by sensing the differential variation of impedance Z AC − Z BC into two successive channel segments as the cell passes consecutively into RF each one (Fig. 10). This method was preferred over the CM RM use of a separate reference channel because cell’s properties can be measured directly against its surrounding CF media. Another advantage derives from the fact that the measurement and reference electrodes are inherently – vout switched; any uneven drift of the electrode properties can + be sensed and corrected, as both signal maxima should be of the same amplitude. Furthermore, the speed of the particle can be determined, as the distance between the two measurement areas and time ttr separating both signal spikes are known, as shown in Fig. 11. An electric model is Figure 15. Impedance sensing part [29]. shown in Fig. 12, where Cdl is the double layer capacitance, Rc is the resistivity of the cell, Cm is the capacitance of the surrounded non-conductive membrane, and RSoL1 disadvantage of this technique is that it doesn’t provide is the resistance of the fluid surrounding the cell. Based on integration actuation capabilities and requires microfluthis model, the impedance change can be determined ana- idics to move cells in the device. lytically or simulated using 3D finite elements [25]. Also recently, another method based also on the A CMOS Lab-on-a-Chip Based DEP micro-machining technology has been developed; it is Medoro et al. in 2002 [14, 29] proposed the 1st lab-on-acalled a “capacitance cytometry” technique [26]. Using chip approach to electronic manipulation and detection this technique, one can quantify the DNA content of sin- of microorganism. The proposed approach combines gle eukaryotic cells from a diverse set of organisms, dielectrophoresis with impedance measurement to trap ranging from yeast to mammals. In addition, capacitance cytometry can be used as an assay for Conductive Glass abnormal changes in DNA content, such as are encountered frequently in neoplastic cells. In Metal 3 this technique, Because DNA is a COLS SELP highly charged molecule, in an applied low-frequency AC electric field, its polarization SELM response, in combination with ROWS the motion of the surrounding COLW counter ions, can be substantial. n-Well SELP Juction One measures this response as a ROWS Diode ROWW change in the total capacitance, SELP CT , across a pair of microelecB Vout trodes as individual eukaryotic Vorow cells suspended in buffer soluP-Substrate tion flow one by one through a B SELM microfluidic channel, as shown ROWS RMUX in Fig. 13. The advantages of the impedance sensing technique are that Actuation Sensing it can be used in many tasks, Micro-Site e.g., counting, sizing, and population study, and that it is suitFigure 16. Schematic presentation of a microsite [15]. able for miniaturization. The THIRD QUARTER 2004


13

and move particles, while monitoring their location and quantity, into the device. The prototype has been realized using the standard printed circuit board (PCB) technology. The actuation part in this approach is based on the DEP moving cage [16]. The proposed approach for establishing DEP cage is shown in Fig. 14. The DEP cage in the spatial region above an electrode can be obtained by connecting the associated electrode and the microchamber lid to an out of phase (180◦ phase) sinusoidal voltage, while the neighboring electrodes are connected to an in-phase sinusoidal voltage. A field minimum is then obtained in the liquid corresponding to a DEP cage in which one or more particles are trapped and levitated. By changing the pattern voltage applied to the electrodes, DEP cage can be independently moved around the device plan. The sensing part in this approach can be performed for any electrode by switching it from the electrical stimulus to a transimpedance amplifier, while all the other electrodes are connected to ground and a sinusoidal stimulus is applied to the lid (Fig. 15). During the sensing task the actuation is switched-off. The presence of a particle above electrodes induces a perturbation of the electric field that can be detected by impedance sensing. Manaresi et al. in 2003 [15, 16], proposed another CMOS lab-on-a-chip microsystem for cell manipulation and detection based on the standard 0.35 µm CMOS technology. This lab-on-a-chip microsystem consists of two main units, the actuation unit, which creates the DEP Cage, and the sensing unit. The chip surface implements a 2D array of microsites, each consisting of a superficial electrodes and embedded photodiode sensor and logic. The actuation part is based on the same DEP cage technique [16], which we have mentioned before. While the sensing part depends on that particles in the sample can be detected by the changes in the optical radiation impinging on the photodiode associated with each microsite. During the sensing, the actuation voltages are halted, to avoid coupling with the pixel readout. However, due to inertia, the cells keep their position in the liquid. A schematic diagram for a single microsite is shown in Fig. 16 [15]. When the site is addressed and WRITE is activated, the metal 3 electrode can be switched between Vphip and Vphim by programming the memory element addressed by ROWW and COLW. The sensing circuit is an active-pixel optical sensor implemented with a 2 × 17 µm2 well-junction photodiode placed in correspondence to the 1.2 µm wide gap of each electrode to its right neighbor. The sensor array is read row-wise. Activating sense, a pixel is addressed through ROWS and COLS, and the voltage after integration is sampled by the readout circuit through Vorow , While the pixel is still addressed, RESET goes high, and the reset voltage is sampled also by 14


the readout circuit. The difference between the two voltages is amplified and output as a differential voltage Voutp − Voutm . The subtraction of the reset voltage allows one to compensate the pixel’s fixed pattern noise. The advantages of these lab-on-a-chip microsystems are that these are the first PCB and CMOS labs on a chip that contain actuation as well as sensing parts that deal with the cell level, and that they can trap, concentrate, and quantify biocells. The disadvantage of these CMOS lab-on-a-chip microsystems can be summarized as follows: 1. Based on these two systems, we can detect the position of the levitated cells. However, we cannot sense the actual intensity of the nonuniform electric field that produces the DEP force. 2. The measurements are indirect. In other words, there is no real time detection of the cell response under the effect of the nonuniform electric field, as we halt the actuation part and activate the sensing part. 3. The sensing part in theses two microsystems depends on the inertia of the levitated cells. In other words, this sensing approach depends on an external factor, which is the inertia of the levitated cells. Thus, only cells with higher inertia can be sensed and detected by using these two microsystems. Summary Recently, the sensing methods for dielectrophoriese (DEP) have been changed from bulky instruments to labon-a-chip. Lab-on-a-chip based the dielectrophoresis phenomenon holds the promise to give biology the advantage of miniaturization for carrying out complex experiments. However, until now, there is an unmet need for lab-on-a-chip to effectively deal with the biological systems at the cell level. Acknowledgement The authors want to acknowledge National Science and Engineering research Council (NSERC) strategic grant, STPGP 258024-02, Canadian Microelectronics Corporation (CMC), Macralyne for funding this work and Dr. Karan Kaler, University of Calgary, for his advice and academic help. References [1] H.A. Pohl, Dielectrophoresis, Cambridge: Cambridge University Press, 1978. [2] M. Washizu and O. Kurosawa, “Electrostatic manipulation of DNA in microfabricated structures,” IEEE Transactions on Industry Applications, vol. 26, no. 6, pp. 1165–1172, 1990. [3] R. Casanella, J. Samitier, A. Errachid, C. Madrid, S. Paytubi, and A. Juarez, “Aggregation profile characterisation in dielectrophoretic structures using bacteria and submicron latex particles,” IEE ProceedingsNanobiotechnnology, vol. 150, pp. 70–74, 2003. [4] D.J. Bennett, B. Khusid, C.D. James, P.C. Galambos, M. Okandan, D. Jacqmin, and A. Acrivos, “Combined field-induced dielectrophoresis and phase separation for manipulating particles in microfluidics,” Journal of THIRD QUARTER 2004

Applied Physics Letters, vol. 83, pp. 4866–4868, 2003. [5] E.V. Tsiper and Z.G. Soos, “Electronic polarization at surfaces and thin films of organic molecular crystal: PTCDA,” Chemical Physics Letters 360(1–2), pp. 47–52, 2002. [6] M.P. Hughes et al., “Strategies for dielectrophoretic separation in laboratory-on-a-chip systems,” Electrophoresis, vol. 23, no. 16, pp. 2569–2582, 2002. [7] P.T. Gaynor and P.S. Bodger, “Electrofusion processes: Theoretical evaluation of high electric field effects on cellular transmembrane potentials,” IEE Proceedings-Science, Measurement and Technology, vol. 142, no. 2, pp. 176–182, 1995. [8] L. Benguigui, A.L. Shalom, and I.J. Lin, “Influence of the sinusoidal field frequency on dielectrophoretic capture of a particle on a rod,” Journal of Physics D (Applied Physics), vol. 19, no. 10, pp. 1853–1861, 1986. [9] P. Fortina, S. Surrey, and Lj. Kricka, “Molecular diagnostics: Hurdles for clinical implementation,” Trends Mol. Med., vol. 8, pp. 264–266, 2002. [10] Available: www.lab-on-a-chip.com. [11] K.K. Jain, “Pharmacogenomics,” in Cambridge Healthtech Inst. Third Annual. Conf. Lab-on-a-Chip and Microarrays, vol. 2, Zurich, Switzerland, pp. 73–77, 2001. [12] Lj. Kricka, “Microchips, microarrays, biochips and nanochip: Personal laboratories for the 21st century,” Clin. Chim. Acta, vol. 307, pp. 219–223, 2001. [13] Available: http://www.healthtech.com/2003/mfe/index.asp, [14] G. Medoro, N. Manaresi, M. Tartagni, and R. Guerrieri, “CMOS-only Sensors and Manipulation for microorganisms,” Proc. IEDM, pp. 415–418, 2000. [15] N. Manaresi, A. Romani, G. Medoro, L. Altomare, A. Leonardi, M. Tartagni, and R. Guerrieri, “A CMOC Chip for Individual Manipulation and Detection,” IEEE International Solid-State Circuits Conf., ISSCC 03, pp. 486–488, 2003. [16] G. Medoro, N. Manaresi, A. Leonardi, L. Altomare, M. Tartagni, and R. Guerrieri, “A Lab-on-a-Chip for Cell Detection and Manipulation,” IEEE Sensors Journal, vol. 3, no. 3, pp. 317–325, 2003. [17] T.B. Jones, Electromechanics of Particles, Cambridge: Cambridge Univ. Press, 1995. [18] J. Voldman, “A Microfabricated Dielectrophoretic Trapping array for Cell-based Biological assays,” PhD thesis, Massachusetts Institute of Technology, 2001. [19] A.Y. Fu, C. Spence, A. Scherer, F.H. Arnold, and S.R. Quake, “A microfabricated fluorescence-activated cell sorter,” Nat. Biotech., vol. 17, 1999. [20] M.S. Talary and R. Pethig, “Optical technique for measuring the positive and negative dielectrophoretic behavior of cells and colloidal suspensions,” Proc. Inst. Elect. Eng.—Sci. Meas. Technol., vol. 14, no. 5, 1994. [21] J.P.H Burt, T.A.K. Al-Ameen, R. Pethig, and X. Wang, “An optical dielectrophoresis spectrometer for low frequency measurements on colloidal suspensions,” J. Physics E: Sci. Instr., vol. 22, pp. 952–957, 1989. [22] J.A.R. Price, J.P.H. Burt, and R. Pethig, “Applications of a new optical technique for measuring the dielectrophoretic behavior of microorganism,” Biochim. Biophy. Acta, vol. 964, pp. 221–230, 1988. [23] S. Eyal and S.R. Quake, “Velocity-independent microfluidic flow cytometry,” Electrophoresis, vol. 23, pp. 2653–2657, 2002. [24] S. Gawad, L. Schild, and Ph. Renaud, “Micromachined impedance spectroscopy flow cytometer for cell analysis and particle sizing,” Lab on a Chip, vol. 1, pp. 76–82, 2001. [25] K.C. Fuller, J. Hamilton, H. Ackler, P. Krulevitch, B. Boser, A. Eldredge, F. Becker, J. Yang, and P. Gascoyne, “Microfabricated multi-frequency particle impedance characterization systems,” in Micro Total Analysis Systems. Enschede, The Netherlands: Kluwer, 2000. [26] L.L. Sohn, O.A. Saleh, G.R. Facer, A.J. Beavis, R.S. Allan, and D.A. Notterman, “Capacitance cytometry: measuring biological cells one by one,” in Proc. Nat. Acad. Sci. USA, vol. 97, pp. 10 687–10 690, 2002. [27] H.E. Ayliffe, A.B. Frazier, and R.D. Rabbitt, IEEE J. Microelectromech. Syst., vol. 8, no. 1, pp. 50–57, 1999. [28] S. Gawad, L. Schildb, and Ph. Renauda, “Micromachined impedance spectroscopy flow cytometer for cell analysis and particle sizing,” Lab on a Chip, vol. 1, pp. 76–82, 2001. [29] G. Medoro, N. Manaresi, A. Leonardi, L. Altomare, M. Tartagni, and THIRD QUARTER 2004

R. Guerrieri, “A lab-on-a-chip for cell detection and manipulation,” Proc. of IEEE SENSORS 2002, vol. 1, pp. 472–475, 2002.

The authors want to acknowledge National Science and Engineering research Council (NSERC) strategic grant, STPGP 258024-02, Canadian Microelectronics Corporation (CMC), Macralyne for funding this work and Dr. Karan Kaler, University of Calgary, for his advice and academic help. Yehya Ghallab received the B.Sc. and M.Sc. degrees from Electronics and Communication Department, Ain Shams University, Cairo, Egypt in 1995 and 2000, respectively. He is currently a Research Associate in the Department of Electrical and Computer Engineering, University of Calgary, Calgary, Alberta, Canada. From 2000 till 2002, he worked as an Assistant Lecturer in the Department of Biomedical Engineering, Helwan University, Cairo, Egypt. His research interests include the application of microelectronics in biomedicine, dielectrophoresis, microelectronic bio-manipulators, sensors and currentmode devices.

Wael Badawy is currently an associate professor at the Department of Electrical and Computer Engineering, University of Calgary. Dr. Badawy received his B.Sc. and M.Sc. degrees from Department of Computer Science and Automatic Control Engineering, University of Alexandria, M.S. and Ph.D. degrees from the Center for Advanced Computer Studies, University of Louisiana at Lafayette. Dr. Badawy authors and co-authors more than 60 refereed Journal/Conference papers and about 20 technical reports. He is the Guest Editor for the special issue on System on Chip for Real-Time Applications of the Canadian Journal on Electrical and Computer Engineering, the Technical Chair for the 2004 International Workshop on SoC for real-time applications, and a technical reviewer in several IEEE journals and conferences. He is currently a member of the IEEE-CAS Technical Committee on BioCas, VLSI and Communications. Dr. Badawy is honored with the “2002 Petro Canada Young Innovator Award”, “2001 Micralyne Microsystems Design Award” and the “1998 Upsilon Pi Epsilon Honor Society and IEEE Computer Society Award for Academic Excellence in Computer Disciplines”.

1540-7977/04/$20.00©2004 IEEE


15