WO2001052997A1 - Method and apparatus for the separation of particles - Google Patents

Abstract

A method and apparatus for collecting and separating abiotic and/or biotic particles and/or chemicals in a liquid or gaseous sample. The apparatus includes a substrate (10) supporting at least one pair of electrodes (4, 6) and defining a fluid channel (8) over said electrodes. The electrodes (4, 6) are energised with at least two voltages having different predetermined frequencies which attract the particles, by dielectrophoresis, to the electrodes. By switching off one or more of the voltages, it is possible to selectively release one type of particle for subsequent collection and/or enumeration.

Description

METHOD AND APPARATUS FOR THE SEPARATION OF PARTICLES

The present invention relates to a method and apparatus for collecting and separating abiotic and/or biotic particles, such as latex beads, mammalian cells or microbial cells, viruses, prions and chemicals or biochemicals using dielectrophoresis. It also relates to a method and apparatus for enumerating a particular particle present in a test sample.

It is well known that when an AC voltage is applied to a pair of electrodes which have a suspension of particles between them, the particles may polarise and have a force exerted upon them where the electric field is non-uniform (see for example Pohl, 1978). This translational force (the dielectrophoretic force) may cause the particles to aggregate in areas of either high or low electric field gradient, dependant upon the polarisabilities of the particles and the suspending medium. The polarisabilities of the particle and medium are functions of their conductivity and permittivity, and varies with the frequency of the electric field (Pethig, 1991; Pethig et al., 1992; Betts, 1995). Measuring the number of particles collected as the frequency of the voltage generating the electric field changes allows a collection spectrum to be plotted as described by WO 91/08284. These spectra been shown to be characteristic for individual species of biological cells and for abiotic particles, since the polarisability of a particle type is dependant upon its individual, unique structure.

WO 98/04355 disclosed an apparatus and method for rapidly determining the dielectrophoretic spectrum of a particle suspension, characterising the properties of specified particle types within a fluid. This invention described an array of 20 electrodes separated from a common ground electrode, whereby a separate and distinct frequency of voltage could be applied between each electrode in the array and the common ground. By quantification of the number of the collected particles using a CCD camera, this enabled a complete dielectrophoresis spectrum to be obtained in a single experiment by applying electric fields of the same voltage but differing frequency to respective electrodes in the array. The disadvantage of this arrangement is that it is less convenient for separating different particle types and reducing the amount of background material. This AC electrokinetic technique, known as Dielectrophoresis (DEP), has been shown to be useful for particle and cell characterisation and also for the separation of a particle type from a mixed suspension (Hagedorn et al., 1992; Huang et al., 1993; Gascoyne et al., 1992; Gascoyne et al., 1994; Huang et al., 1992). Cells or particles become polarised by the action of AC electric fields and will experience a dielectrophoretic force when these fields are non- uniform. The dielectrophoretic force is a function of frequency, determined by the electrical properties of the cell, reflecting cell structure and morphology. Therefore cells with different electrical properties and polarisability will experience differential dielectrophoretic action, allowing separation of different cell types. By utilising selective differences in DEP response, the separation of live and dead yeast cells (Pohl & Hawk, 1966; Crane & Pohl, 1968; Pohl & Crane, 1971), cancerous and normal cells (Burt et al, 1990; Becker et al, 1994), and bacterial species (Markx et al., 1994; Markx et al., 1996) have all been achieved. Analyses of other micro-organisms, such as the water-borne protozoan Cryptosporidium parvum, have also shown that the determination and separation of different viability states is possible using dielectrophoretic methods (Archer et al. 1993; Quinn et al, 1995; Archer et al, 1995; Quinn et al, 1996; Goater et α/., 1997).

Many DEP methods of cell separation have relied upon the application of a single, fixed-frequency, AC voltage to an electrode structure. In particular, the frequency of the electric field and the dielectric constant and electrical conductivity of the suspending medium is selected to produce positive and negative dielectrophoretic forces, where the positive dielectrophoretic force acts upon some only of the particles in the suspension (to attract particles to electrode surfaces where the field gradient is high), and the negative dielectrophoretic force acts upon a different population of particles in the suspension (repelling these particles to a spatially separate region of low, normally zero, electric field gradient) (Pethig et al., 1992). Markx et al. (1994) also used castellated electrodes to bring about a localised separation of Saccharomyces cerevisiae and Micrococcus lysodeikticus by this method. Use of conductivity gradients or suspending media to facilitate dielectrophoretic separation has also been shown (Markx et al., 1996). Since negative DEP is a weaker force, a constant flow of the suspension can remove those particles undergoing negative DEP, whereas those undergoing positive DEP will remain in the areas of high field gradient and be separated from the suspension.

WO 91/11262 disclosed the application of electrical fields of different characteristics to several separate arrays of electrodes, energised independently, for the purposes of spatially separating particle and cell types from a mixture on the basis of dielectrophoretic properties. WO 94/22583 discloses a further separation device based upon the same localised spatial separation using dielectrophoretic action described in WO 91/11262. In this invention, a mixed suspension of two particle types is introduced into the centre of an electrode array and exposed to dielectrophoretic forces. The conditions are selected to cause positive DEP action upon one of the particle types, and negative DEP action upon the other particle type, resulting in a spatial separation. By alternation of pumping and dielectrophoretic action, the two components are separated and the individual components can be removed from either end of the electrode chamber.

GB 2,266,153 described a column array of interdigitating electrodes which could be energised to selectively retard cell populations within a mixture for subsequent elution of separated components, acting as a dielectrophoretic chromatographic column. A similar invention described by Markx et al. (1997) is that of field flow fractionation (FFF), whereby dielectrophoretic levitation of particles is used to displace particles into different regions of a parabolic flow profile travelling at different velocities. Patent WO 97/27933 discloses the combined use of field flow fractionation and phase shifted travelling wave DEP to bring about particle separations.

The disadvantage of these methods of separation is that the choice of frequency to be applied to the electrodes and the nature of the suspending medium often requires prior knowledge of the electrical properties of the particle and medium comprising the suspension, often using the related technique of electrorotation (Huang et al, 1992; Wang et al. 1992). Furthermore, the method only allows the separation of a limited number of cell or particle types from the suspension. In a complex mixture of unknown particle or cells species, significant differences in the dielectrophoretic frequency characteristics of each type maybe observed. Thus use of a single frequency may not result in successful separation or collection of many species within a mixture.

It has now been found that by superimposing two or more voltages of different frequencies on at least one pair or a set of electrodes, passing or circulating a liquid sample containing different particles suspended therein past the electrodes, then switching off one or more of the voltages it is possible to separate, identify and subsequently count the released particles. In contrast to WO 91/11262 and WO 98/04355 which independently energised separate and distinct electrode arrays to produce simultaneous positive and negative dielectrophoretic action within an electrode chamber, this present invention uses the principle of superposition to apply multiple voltages to a single electrode array.

It is an object of the invention to provide an improved dielectrophoretic method and apparatus for separating, collecting and counting abiotic and/or biotic particles.

According to one aspect of the invention there is provided a method of separating different particles present in a liquid or gaseous sample, the method comprising the steps of passing or circulating the sample through a region of non-uniform electric field density produced by at least one pair of electrodes, energising said electrodes with a first voltage having a first predetermined frequency selected to attract a first predetermined variety of particle in the sample to said electrodes, superimposing on said electrodes a second voltage having a second predetermined frequency selected to attract a second predetermined variety of particle in the sample, switching off either the first or the second voltage thereby releasing either the first or second variety of particle for subsequent collection and/or enumeration.

More than two different voltages having different predetermined frequencies may be superimposed on and applied to the electrodes in order to attract all the particles in the liquid sample to them. This can be applied to any number of cell or particle types within a mixture which may have wide variations in their DEP frequency spectrum. The particles can then be subsequently released en masse by switching off all of the voltages, thus permitting a total particle count to be determined. Alternatively, the particles may be released from the electrodes individually by type by switching off a selected voltage or voltages, thus facilitating separation of the particles for subsequent collection, identification and/or enumeration. Separation by this method may be enhanced by elution with specific suspending media of variable conductivity, pH or other physical characteristic.

The method may be used for separating different biotic particles such as microorganisms and/or different cell types and cell organelles including plasmids. The term micro-organism is intended to embrace bacteria, viruses, yeasts, algae, protozoa, fungi and prions. Abiotic particles which may be separated include for example latex beads, metal particles or any inorganic or organic material, chemical or biochemical species can also be separated.

According to a second aspect of the invention there is provided an apparatus for separating particles present in a liquid sample the apparatus comprising a support defining a fluid flow channel through a region of non-uniform electric field density produced by at least one pair of spaced electrodes, circulating means for passing or circulating said sample containing said particles through said channel, a first AC source for applying a first voltage at a first frequency to said electrodes, said frequency being selected to cause a first predetermined type of particle to be attracted to said electrodes, a second AC source for applying a second voltage at a second frequency to said electrodes, said second frequency being selected to cause a second predetermined type of particle to be attracted to said electrodes and means for determining the quantity of either the first or second predetermined type of particle when either the first voltage or second voltage is not applied.

The use of multiple frequencies may be simultaneously applied to any design of electrodes, with the electrodes being shaped to generate maximum DEP effect. Thus, for example, the electrodes may be castellated or they may comprise an electrode array which may or may not define more than one fluid flow channel. Several pairs of electrodes may be used thus forming a set of electrodes which may be linked together.

According to yet another aspect there is provided the use of the method defined above or the use of the apparatus defined above for separating eukaryotic cells, bacteria, yeast, viruses, algae, protozoa, fungi, prions, inorganic or organic abiotic particles, plasmids, cell organelles, chemicals and biochemicals including nucleic acids, and chromosomes.

A method and apparatus for separating particles such as micro-organisms will now be described, by way of example, with reference to the accompanying diagrammatic drawings, in which:

Figure 1 is an electrical and fluid circuit of an apparatus in accordance with the invention; and

Figure 2 is a perspective fragmentary section taken through the collection block of the apparatus of Figure 1; Figure 3 is a block diagram for the manufacture of a multiple frequency sinewave generator which can be used as an alternative to inductively coupled signal generators 28 and 30 of Figure 1;

Figure 4 is a graph showing dielectrophoretic frequency spectra of polystyrene latex spheres and E.coli bacteria demonstrating the differences in frequency of collection; Figure 5 is a graph showing the number of released latex beads counted over a period of time when different frequencies or voltage are applied individually and simultaneously using the multiple frequency sinewave generator;

Figure 6 is a graph showing the number of released E.coli counted over a period of time when different frequencies or voltage are applied individually and simultaneously using the multiple frequency sinewave generator;

Figure 7 is a graph showing the number of released latex beads and E.coli counted over a period of time when a 10 kHz voltage is applied;

Figure 8 is a graph showing the number of released latex beads and E.coli counted over a period of time when a 1 MHz voltage is applied; Figure 9 is a graph showing the number of released latex beads and E.coli counted over a period of time when both 10 kHz and 1 MHz voltages are applied; and

Figure 10 is a graph showing the number of released latex beads and E.coli counted over a period of time when both 10 kHz and 1 MHz voltages are applied and then disabled sequentially.

The apparatus shown in Figure 1 comprises a collection block 2 in which particles can be collected. The collection block 2 contains a pair of spaced electrodes 4 and 6 lying in a common plane and a fluid flow channel 8 positioned to cause a liquid to flow across the upper faces of the two electrodes.

The structure can be more clearly seen in Figure 2. As shown, the structure includes an electrically insulating substrate 10 on which two elongate electrodes 4 and 6 have been deposited in parallel but spaced relationship with each other. Two electrically insulating side walls 12 and 14 are attached to the substrate 10 and cover both the electrodes 4 and 6. The side walls can be composed of a polyimide. The side walls 12 and 14 define the channel 8 which extends at right angles to the electrodes 4 and 6. A further electrically insulating layer (not shown) extends over the side walls 12 and 14 and the channel 8 to form the roof of the channel 8. The exposed face of each electrode may be covered with a thin electrically insulating layer as required.

A reservoir 20, for containing a sample of liquid to be analysed, is connected by a duct 22 to the upstream end of the channel 8. A duct 24 connected to the downstream end of the channel 8 feeds liquid from the channel 8 through a pump 26 back to the reservoir 20.

The pump 26 is advantageously a peristaltic pump to prevent any contamination or damage to the sample liquid and particles therein. The liquid in the reservoir 20 may be agitated by bubbling air or other gas therethrough to keep the particles in suspension.

Two signal generators 28 and 30, which supply two voltages of different frequencies, are inductively coupled using an isolation transformer 32 and the voltages from the signal generators 28 and 30 are applied to the pair of electrodes 4 and 6. A multiple frequency sinewave generator may alternatively be used according to the schematic shown in Figure 3. The multiple frequency generator comprises a set of fixed frequency oscillators which can be energised to output a specific frequency sinewave. The outputs from these are mixed and outputted to energise electrodes 4 and 6 of Figure 1. Five fixed frequency oscillators are shown in the figure (at frequencies 1kHz, 10kHz, 100kHz, 1MHz and 10MHz) each providing an 8N peak output voltage. Thus it is possible to simultaneously superimpose voltages of all five of these frequencies on to electrodes 4 and 6.

Any number of signal generators may be inductively coupled to apply more frequencies of voltage to the pair of electrodes or superimposing using other methods. Alternatively the multiple frequency sinewave generator can be manufactured with any number of frequency oscillators. By using an appropriate number of frequencies it should be possible to collect every type of particle in a suspension even with widely differing frequency spectrum characteristics, such as those of E.coli and polystyrene latex as demonstrated in Figure 4. Following this it would be possible to determine the number of particles collected with the image analysis technique or another method. This would then be a measure of the total number of particles in the test suspension.

Impedance matching may be used to make the isolation transformers and the electrode structure match the 50Ω impedance of any coaxial cables used to connect the system. This will create a constant voltage across the electrode gap within the range of frequencies to be used.

The apparatus according to the invention may be computer controlled to enable rapid, easy control of the signal generators or multiple frequency generator. This will be of particular benefit if all particle types are to be collected from a sample and released individually by changing the applied frequencies.

To separate a suspension of particles, for example latex beads and E.coli bacteria, from a liquid sample, the liquid sample is placed in the reservoir 20 and pumped by pump 26 via duct 22 through the channel 8 over electrodes 4 and 6.

Latex beads are known to collect well on the electrodes when an output frequency of 10 kHz is applied to the electrodes 4 and 6. Signal generator 30 is therefore set to have an output frequency of 10 kHz (or the 10kHz oscillator on the multiple frequency sinewave generator is applied). Latex beads do not collect well at 1 MHz. In contrast, E.coli bacteria collect well at both frequencies as shown in Figure 4. Signal generator 28 is set to have an output frequency of 1 MHz (or the 1MHz oscillator on the multiple frequency sinewave generator is applied).

In order to assess any particle collection at the electrodes 4 and 6, they were mounted beneath a microscope to which a charge coupled device (CCD) camera was fitted. The microscope was focused slightly downstream of the electrodes. The camera fed into an image analysis card fitted in a computer, which labelled and counted the number of particles within the plane of focus five times per second. Other methods of counting the particles may be utilised here, eg spectrophotometric, impedance analysis.

To determine whether particle collection had occurred, one or both of the voltage inputs were applied for a period of time, while the image analysis program counted particles in order to determine the background level. The voltage was then turned off. If any particles had collected at the electrodes they would then be released and pass beneath the microscope, leading to an increase in the number of particles counted as the released packet of particles flowed by. The number of counted particles would then drop back to the background level. Changing the phase/contrast of the microscope allowed either the latex beads or the E.coli to be labelled and counted separately.

In an initial test which was performed using separate suspensions of latex beads and E.coli, the 1kHz frequency of the multiple frequency generator was applied for 20 seconds then disabled while the image analysis system counted the particles flowing beneath the microscope. This was then repeated for the other four frequencies of the multiple frequency generator and also with all five frequencies applied simultaneously. The results for both particle types are shown in Figures 5 and 6.

The latex beads collected well at 10 kHz, as there was a significant increase in the number counted once the voltage was disabled and the particles were released, before the count dropped back to a background level. At 1 MHz there was very poor collection, with the count remaining at a background level throughout the experiments. Application of the individual frequencies produced a frequency spectral profile very similar to that shown in Figure 5. The application of all voltages simultaneously resulted in good collection of the latex.

The E.coli collected well at both 10 kHz and 1 MHz, but there was an increase in collection with all voltages applied.

This demonstrates that by applying multiple frequencies simultaneously to a pair of electrodes it was possible to generate an increase in the number of E.coli cells collected at the electrodes, as compared with a decrease in latex collection when multiple frequencies were applied.

The initial test was repeated using a mixed suspension of latex beads and E.coli. Figures 7 to 9 show the results obtained using two inductively coupled signal generators The voltage was applied for 8 seconds then disabled while the image analysis system counted the particles flowing beneath the microscope. This was performed 5 times and the particle count averaged, and then repeated for 1 MHz alone and then 10kHz and 1MHz applied simultaneously.

At 10 kHz both particles collected well, as was seen from the individual suspensions, and at 1 MHz, as expected, the E.coli collected at a similar level as for 10 kHz, whereas the latex beads did not. When both voltages were applied both particles collected, with the E.coli reaching a greater level. In a further experiment, the effect of disabling the voltages subsequently was studied to find out how easily the collected particles could be separated.

Both voltages were applied simultaneously for 8 seconds. Then the 10 kHz voltage was disabled. Ten seconds later the 1 MHz voltage was disabled. The results are shown in Figure 10.

When the 10 kHz field was disabled and the 1 MHz field remained applied, the latex beads that collected on the electrodes were released (Time 30-35). No E.coli were released at this time. When the 1 MHz field was later disabled, the E.coli were released (Time 60-80). It should be noted that no latex was counted at this time, indicating that the beads were all released when the 10 kHz field was disabled.

This demonstrated the effectiveness of the technique as a method of particle separation, as all of the latex beads were released at a different time than the E.coli.

Apart from the image analysis technique described other methods including spectrophotometric (including fluorescence), laser, impedance analysis and radiometric may be used to count the number of particles collected at or released from the electrodes.

With appropriate choice of frequencies for separating particle types the method and apparatus defined above may be used to separate a single species from the test suspension by either collecting all cell types bar the desired one, or by only collecting the desired species. Another option may be to collect all of the cells in the suspension and then release them one at a time by either turning off one frequency or by changing one or more of the frequencies such that one cell type no longer collects. This method could be used to separate any number of species in the suspension, by collecting them all and selectively releasing them individually.

It may also be possible to collect any number of the species from a large suspension onto the electrodes by appropriate choice of applied frequencies. Following the collection it will be possible to elute the collected cells into a much smaller volume suspension, effectively concentrating the suspension. In addition, the collected cells could be eluted into a different suspending medium as a further form of sample preparation.

In addition, by applying more than one frequency of voltage it may be possible to change the DEP collection, generating a marked increase in the number of collected particles. Such a method could be used to enhance the collection of any number of particles in a suspension. It will also be possible to choose the frequencies such that a number of particle types experience negative DEP and are repelled from the electrodes when others are collected by positive dielectrophoresis, producing an additional form of sample separation.

REFERENCES

Archer G.P., Betts W.B. & Haigh T. (1993) Rapid differentiation of untreated and treated Cryptosporidium parvum oocysts using dielectrophoresis. Microbios 73, 165-172. Archer G.P., Quinn CM., Betts W.B., Allsopp D.W.E. & O'Neill J.G. (1995) Physical separation of untreated and ozone treated Cryptosporidium parvum oocysts using non-uniform electric fields. In Protozoan Parasites and Water. (Eds.) Betts W.B., Casemore D., Flicker C, Smith H.V & Watkins J., pp. 143-145. The Royal Society of Chemistry, Cambridge. Becker F.F., Wang X-B, Huang Y., Pethig R., Vykoukal J. & Gascoyne P.R.C. (1994) The removal of human leukaemia cells from blood using interdigitated microelectrodes. J. Phys. D: Appl. Phys. 27, 2659-2662.

Betts W.B. (1995) The potential of dielectrophoresis for the real-time detection of microorganisms in foods. Trends Food Sci. Technol. 6, 51-58.

Burt J.P.H., Pethig R., Gascoyne P.R.C. & Becker F.F. (1990) Dielectrophoretic characterisation of Friend murine erythroleukaemic cells as a measure of induced differentiation. Biochim. Biophys. Ada 1034, 93-101.

Crane J.S. & Pohl H.A. (1968) A study of living and dead yeast cells using dielectrophoresis. J. Electrochem. Soc. 115, 584-586.

Gascoyne P.R.C, Huang Y., Pethig R., Vykoukal J. & Becker F.F. (1992) Dielectrophoretic separation of mammalian-cells studied by computerized image-analysis. Meas. Sci. Technol. 3, 439-445.

Gascoyne P.R.C, Noshari J., Becker F.F. & Pethig R. (1994) Use of dielectrophoretic collection spectra for characterizing differences between normal and cancerous cells. IEEE Trans. Ind. Appl. 30, 829-834. Goater A.D., Burt J.P.H. & Pethig R. (1997) A combined travelling wave dielectrophoresis and electrorotation device: applied to the concentration and viability determination of Cryptosporidium. J. Phys. D: Appl. Phys. 30, L65-L69.

Hagedorn R., Fuhr G., Muller T. & Gimsa J. (1992) Travelling-wave dielectrophoresis of microparticles. Electrophoresis 13, 49-54. Huang Y., Holzel R., Pethig R. & Wang X-B. (1992) Differences in the AC electrodynamics of viable and nonviable yeast-cells determined through combined dielectrophoresis and electrorotation studies. Phys. Med. Biol. 37, 1499-1517.

Huang Y., Wang X-B., Tame J.A. & Pethig R. (1993) Electrokinetic behaviour of colloidal particles in travelling electric-fields - studies using yeast-cells. J. Phys. D - Appl. Phys. 26,

1528-1535 Markx G.H., Dyda P.A. & Pethig R. (1996) Dielectrophoretic separation of bacteria using a conductivity gradient. J. Biotechnol. 51, 175-180.

Markx G.H., Huang Y., Zhou X-F. & Pethig R. (1994) Dielectrophoretic characterisation and separation of micro-organisms. Microbiology 140, 585-591.

Markx G.H., Pethig R. & Rousselet J. (1997) The dielectrophoretic levitation of latex beads, with reference to field-flow fractionation. J. Phys. D: Appl. Phys. 30, 2470-2477.

Pethig R. (1991) Application of AC electrical fields to the manipulation and characterisation of cells. In Automation in Biotechnology. (Ed.) Karube I., pp. 159-185. Elsevier, Amsterdam.

Pethig R., Huang Y., Wang X-B. & Burt J.P.H. (1992) Positive and negative dielectrophoretic collection of colloidal particles using interdigitated castellated electrodes. J. Phys. D: Appl. Phys. 24, 881-888.

Pohl H.A. (1978) Dielectrophoresis. Cambridge University Press, Cambridge.

Pohl H.A. & Crane J.S. (1971) Dielectrophoresis of cells. Biophy . J. 11, 711-727.

Pohl H.A. & Hawk I. (1966) Separation of living and dead cells by dielectrophoresis. Science

152, 647-649. Quinn CM., Archer G.P., Betts W.B. & O'Neill J.G. (1996) Dose-dependent dielectrophoretic response of Cryptosporidium oocysts treated with ozone. Eett. Appl. Microbiol. 22, 224-228.

Quinn CM., Archer G.P., Betts W.B. & O'Neill J.G. (1995) An image analysis enhanced dielectrophoretic analysis of chlorine and ozone treated Cryptosporidium parvum oocysts. In

Protozoan Parasites and Water. (Eds.) Betts W.B., Casemore D., Flicker C, Smith H.V. & Watkins J., pp. 125-132. The Royal Society of Chemistry, Cambridge.

Wang X-B., Pethig R. & Jones T.B. (1992) Relationship of dielectrophoretic and electrorotational behaviour exhibited by polarised particles. Journal of Physics D: Applied

Physics. 25, 905-912.

Claims

1. A method of separating different particles present in a liquid or gaseous sample, the method comprising the steps of passing or circulating the liquid or gaseous sample through a region of non-uniform electric field density produced by at least one pair of electrodes, energising said electrodes with a first voltage having a first predetermined frequency selected to attract a first predetermined variety of particle in the sample to said electrodes, superimposing on said electrodes a second voltage having a second predetermined frequency selected to attract a second predetermined variety of particle in the sample, switching off either the first or second voltage, or both thereby releasing either the first or second variety of particle or both for subsequent separation, collection, identification and/or enumeration.

2. The method according to Claim 1 , wherein more than two different voltages having predetermined frequencies are superimposed on and applied to said pair of electrodes.

3. A method according to Claim 2, including the step of enhancing the separation by elution with specific suspending media of variable conductivity or pH.

4. Apparatus for separating particles present in a sample liquid the apparatus comprising a support defining a fluid flow channel through a region of non-uniform electric field density produced by at least one pair of spaced electrodes, circulating means for circulating said sample containing said particles through said channel, a first AC source for applying a first voltage at a first frequency to said pair of electrodes, said frequency being selected to cause a first predetermined type of particle to be attracted to said electrodes, a second AC source for applying a second voltage at a second frequency to said electrodes, said second frequency being selected to cause a second predetermined type of particle to be attracted to said electrodes and means for determining the quantity of either the first or second predetermined type of particle when either the first voltage or second voltage is not applied.

5. Apparatus and specification for application of multiple frequencies simultaneously by way of a multiple frequency sinewave generator.

6. Apparatus according to Claim 4 or to Claim 5, wherein the electrodes are castellated so as to generate maximum DEP effect.

7. Apparatus according to any one of Claims 4 to 6, wherein said electrodes define more than one fluid flow channel.