FERROFLUIDS

BACKGROUND OF THE INVENTION

Early diagnosis of diseases involving rare cells in blood (such as metastatic cancer or low-level bacteremia) and accurate monitoring of certain genetic conditions (such as sickle cell anemia) require rapid and accurate separation, sorting, and direction of target cell types toward a sensor surface. In that regard, ceilular manipulation, separation, and sorting are increasingly finding application potential within various bioassays in the context of cancer diagnosis (Dittrich et al. _; 2006, Nat Rev Drug Discovery 5:210 ^{^} 218), pathogen detection (Beyor et al., 2008, Biomed Microdevices 10:909-917), and genomic testing (Kamei et al. 2005, Biomed Microdevices 7: 147-152; Cheong et al., 2008, Lab Chip 8:810-813).

A variety of contactless micromanipulation methods exist, including optical tweezers (Ashkin et al., 1987, Nature 330:769-771 ; Chiou et al., 2005, Nature 436:370-372), dielectrophoresis (DEP) (Hughes, 2002, Electrophoresis 23:2569-2582), magnetic bead- based separators (Lee et al., 2001, Appl Phys Lett 79:3308-3310; Yan et al., 2004, Phys Rev E 70:01 1905), and deterministic hydrodynamics (Davis et al., 2006, Proc Natl Acad Sci USA 103: 14779-14784). However, most existing methods have been unable to reliably achieve fast speed, high throughput and resolution, simultaneously with low costs (Dufresne et al., 1998, Rev Sci Instntm 69: 1974-1977; Kremser et al., 2004, Electrophoresis 25:2282-2291 ; Cabrera et al., 2001 , Electrophoresis 22:355-362). Optical tweezers offer high resolution and sensitivity for manipulating single cells, although such manipulation may cause sample heating (Liu et al, 1995, Biophys J 68:2137-2144), and is typically limited to a very small area (Ashkin et al, 1987, Science 235: 1517 1520). Holographic schemes have recently extended the reach of optical tweezers to several tens of cells simultaneously (Applegate et al., 2004, Optical Express 12:4390-4398), although the overall throughput remains quite low. Schemes based on electric fields, such as DEP, offer the potential to realize integrated, cost- effective devices for the simultaneous manipulation of multiple cells; nevertheless, their performance depends sensitively on the electrical properties of the specific liquid medium, the particle shape, and its effective dielectric constant (Pethig et al., 1 97, Trends BioteclmoJ 15:426-432). DEP device operating regimes and the working ionic medium need to be carefully optimized for each different cell type so as to reach a workable compromise between the need to reduce heating (Menachery et al., 2005, NanoBiotechnology 152: 145- 149; Muller, et al., 2003, IEEE Eng Biol Med Mag 22:5 1-61) and minimize ceil polarization (Sebastian et al., 2006, JMicromech Microetig 16: 1769-1777). Using functionalized magnetic beads to separate target molecules and cells overcomes these challenges through the use of magnetic fields instead of electric. However, the downside of this technique is the lengthy incubation times and wash cycles, and the difficulty of removing the label post priori (Gijs 2004, Microfluidics Nan fluidics 1 :22-40), The deterministic hydrodynamics approach, as demonstrated by Davis et al. (Davis et al., 2006, Proc Nail Acad Sci USA 103 : 14779- 14784), is capable of achieving high resolution of separation without the use of any electromagnetic fields, However, high throughput with this device requires high-resolution lithography on a large area, keeping the cost per device high,

Most common applications of ferrofluids in biomedicine involve highly dilute colloidal suspensions of magnetic nanoparticles. Their widest commercial use is as MRI contrast agents (Kim et al., 2005, J Magn Magn Mater 289:328-330). When properly coated with targeting antibodies, they can also be used in hyperthermia therapy for cancer or as sensors to detect pathogens (Scherer et al., 2005, Brazilian J Phys 45:718-727). While these advances in the use of ferrofluids provide many opportunities in medicine and diagnostics, there remains a need in the ait for a microfluidic platform that uses biocompatible ferrofluids for the controlled manipulation and rapid separation of both micropaiticles and live cells. The present invention satisfies this need.

SUMMARY OF THE INVENTION

A device for separating a sample of particles suspended in a biocompatible ferrofiuid is described. The device includes a microfluidic channel having a sample inlet, at least one output, and a length between the sample inlet and the at least one output, wherein a sample can be added to the sample inlet and flow along the length to the at least one outlet. The device includes a plurality of electrodes, wherein the microfluidic channel length transverses the plurality of electrodes, and further includes a power source for applying a current to the plurality of electrodes to create a magnetic field pattern along the length of the microfluidic channel. In one embodiment, the spacing between electrodes is gradually increased. In another embodiment, the spacing between electrodes is gradually decreased. In another embodiment, the plurality of electrodes comprise at least one electrode layer. In another embodiment, the plurality of electrodes comprises a plurality of electrode layers. In another embodiment, the plurality of electrode layers is in a substantially orthogonal pattern. In another embodiment, the plurality of electrodes comprises a pattern of concentric circles. In another embodiment, the walls of the microffuidic channel length include a pocketed, a ridged, a grooved, a trenched or a sloped region. In another embodiment, the microfluidic channel length transverses at least a portion of the plurality of electrodes at an angle between about 1 -90 degrees. In another embodiment, tlie particles are living cells.

Also described is a system for separating at least one target from a sample suspended in a biocompatible ferrofluid. The system includes a micrbfiuidic channel having a sample inlet, at least one output, and a length between the sample inlet and the at least one output, wherein a sample can be added to the sample inlet and flow along the length to the at least one outlet. The system also includes a plurality of electrodes, wherein the microfluidic channel length transverses the plurality of electrodes, and further generates a magnetic field pattern along the length of the microfluidic channel when a current is applied to the electrodes, The system further includes at least one target in a sample suspended in a biocompatible ferrofluid, wherein the at least one target is separated from the remaining sample as the at least one target passes along at least a portion of the microfluidic channel length. In one embodiment, the biocompatible ferrofluid comprises a suitable amount of ionic species to control the osmotic pressure on the cells to promote celt sustainability. In another embodiment, the biocompatible ferrofluid comprises a citrate concentration of between about 5-200 niM, In another embodiment, the biocompatible ferrofluid comprises a citrate concentration of about 40 Mm, In another embodiment, the biocompatible ferrofluid has a pH of about 7.4. In another embodiment, the at least one target is separated based on target size. In another embodiment, the at least one target is separated based on target shape. In another embodiment, the at least one target is separated based on target elasticity. In another embodiment, the target is separated by being directed to a selected outlet. In another embodiment, the target is trapped based on the spacing of electrodes, In another

embodiment, the at least one target is a cell. In another embodiment, the at least one target is a particle.

Also described is a method for separating at least one cell type. The method includes the steps of suspending two or more cell types in a biocompatible ferrofluid to form a sample, passing the sample through a microfluidic channel that transverses a plurality of electrodes, applying a current to the plurality of electrodes to create a magnetic field pattern along the length of the microfluidic channel, and sorting the cells into at least one output channel based on a variation of at least one of cell size, shape and elasticity, In one embodiment, the cells are separated at an efficiency of at least about 90%. In another embodiment, the size resolution in separating is less than about 10 μιη. In another embodiment, the cells are separated in less than about I minute.

BRIEF DESCRIPTION OF THE DRAWINGS

For the purpose of illustrating the invention, there are depicted in the drawings certain embodiments of the invention. However, the invention is not limited to the precise arrangements and instrumentalities of the embodiments depicted in the drawings,

Figure 1 , comprising Figures 1 A- 1D, illustrate a ferromicrofiuidic device and particle manipulation platform. Figure I A is a schematic of the experimental setup displaying the microfluidic channel and the underlying electrodes (not drawn to scale). Two output channels from an amplifier provide sinusoidal currents f/t and / ₂ ) phase-locked 90° with respect to each other. The neighboring electrodes on the substrate are connected in a manner to carry sinusoidal currents in quadrature and support a traveling wave magnetic field within the microfluidic channel. The magnetic field gradient generated pushes the nonmagnetic microspheres or cells within the ferromicrofiuidic channel up and into the gap between electrodes ( ); the traveling field also causes the cells to rotate and roll along the channel ceiling, resulting in continuous translation along the length of the channel at frequencies above a threshold (if), The resulting microparticle motion is observed with an upright microscope from above and captured with a CCD camera at 18 frames per second for further analysis. Figure I B depicts a COMSOL simulation of a magnetic field (dark arrows) and magnitude of magnetic flux density (color) across the cross-section of the ferromici fluidic device at a given instant in time. Fainter arrows depict the field at every 30° within one period. Simulation is for 12-A peak-to-peak current input at 1 ,670 Hz. Figure 1 C is a graph of computed force and torque on a 6- ιη diameter microsphere along the length of the microchamiel with 7-A peak-to-peak input excitation at 4.6 kHz. Figure I D is a graph of computed magnetic force and torque as a function of frequency for the same particle located between electrodes on the channel ceiling. Input current amplitude is 7 A peak to peak;

assumed slip factor for all simulations depicted here is 1 . Figure 2, comprising Figures 2A-2C characterizes biocompatible ferrofhiid. Figure 2A is a graph depicting the distribution of cobaltferrite nanoparticle sizes within the ferrofluid, as obtained by TEM, Mean nanoparticle core diameter is 1 1 ,3 ± 4.4 nm. (Scale bar: 50 nm.) Figure 2B is a graph depicting ac susceptibility and dc magnetization curve (Inset) of the ferrofluid. A fit to the ac susceptibility data assuming a log-normal size distribution indicates moderate particle aggregation with a mean hydrodynamic diameter of 72.5 nm. Figure 2C is a chart depicting live cell count vs, citrate concentration. As shown herein, 40mM citrate concentration (stabilized with citric acid to yield a pH of 7,4) is found to be optimum for cell viability and ferrofluid stability combined. The dashed line shows the cell count in the original blood sample. Count 3 corresponds to cells spending ~1 h in the citrate solution.

Figure 3, comprising Figures 3A and 3B, is a demonstration of particie velocity as a function of input frequency and current amplitude. Figure 3A (middle) is a spatial distribution of instantaneous average ^-velocities for 6-jtm-diameter particles at 7-A input current amplitude (peak to peak) at two different frequencies. Because of repulsive forces from magnetic field gradients, mtcroparticles either slow down or completely stop in between electrodes. Zero crossings with negative slope correspond to stable equilibrium points (i.e., particle trapping); Figure 3 A (top) shows at 10 Hz, particle trajectories terminate in between electrodes, resulting in trapping. Figure 3A (bottom) shows at 4,640 Hz, particles move continuously throughout the length of the channel. This is the regime where magnetic torque from the locally rotating component of the traveling wave dominates over the repulsive forces. The black dots at the end of each trajectory indicate where particles eventually stop. Figure 3B shows that above a critical frequency (f _c the 6-μιη microspheres roll continuously along the top channel surface without getting trapped.

Figure 4, comprising Figures 4A-4D, are illustrative of frequency-dependent particle separation. Figure 4A depticts the particle size dependence of critical frequency (f _c ), Discrete fc values for different diameters of particles enable size-based separation by tuning to the right frequency, The solid curve corresponds to the simulation result with a slip factor of 1 and a particle-wall gap of 1 nm. Figure 4B depicts the average manipulation speed as a function of input frequency for two different particie sizes; 2.2- and 9.9-μηι particles can be separated at 400 Hz. Figure 4C is a fluorescent microscopy image from a section of the microfluidic channel containing 2,2- and 9.9-μιη microspheres randomly dispersed within the channel right before the excitation. Vertical lines indicate electrode borders. Figure 4D is a snapshot of the channel from the same location as in Figure 4C, 45 seconds after the excitation (6 A peak to peak, 400 Hz) is turned on. The 9.9-μηι particles quickly localize within the nearest spacing between electrodes, whereas 97% of the 2.2-μ ^η ι microspheres continuously travel from right to left without being trapped. Almost all of the smaller microspheres within the field of view in Figure 4D have entered from the right as a fresh batch.

Figure 5, comprising Figures 5A and 5B, depict cellular separation with bacteria and blood cells. Figure 5A depicts the spatial distribution of ^-velocities at 200 Hz in a sample containing E. coli bacteria and red blood cells. At this frequency, most red blood cells are trapped between the electrodes (indicated by their zero local speeds), whereas E. coli can slowly but continuously move through that region. Fluctuations in the red blood cell data are statistical in nature, as explained herein. Figure 5B depicts sickle cell separation. Sickle cells, which have an elongated shape and altered elasticity compared with normal red blood cells, are trapped and concentrated between the electrodes, whereas the healthy cells are still able to circulate within the microfluidic channel at 300Hz. Electrode spacing of the device in Figure 5A is different from that in Figure 5B, resulting in different f _c s for red blood cells within each channel.

Figure 6, comprising Figures 6A-6C, Figure 6A depicts microparticle average speed, normalized by the square of the current amplitude (peak-to-peak) and to the maximum value at 5A, depicted as a function of excitation frequency. Particle velocities are proportional to the square of current until about 7 A. Figure 6B depicts the average velocity vs. frequency at 6A peak-to-peak for 9.9 μιη microspheres travelling over electrodes with different spacings. A smaller spacing leads to higher particle velocity and a smaller critical frequency. Figure 6C depicts a conceptual sketch of a microparticle sorter based on the effect observed in 6C. At a given excitation frequency, smaller spacings trap larger particles, while letting smaller ones pass through. Eventually, even the smallest particles can be trapped in the larger gaps, Here, it is assumed that particles move from left to right, and the channel over the electrodes depicted is initially cell-free.

Figure 7, comprising Figures 7A and 7B depict alternative embodiments of the electrode and channel components of the separating device. Figure 7A depicts electrode formations of 100, 1 50, 200 and 300 μηι spacing in the channel. Figure 7B depicts channel formations of between about 0.17-0.19 cm, and having four inlets and four outlets.

Figure 8, comprising Figures 8A and 8B, depict alternative embodiments of electrode patterns. Figure 8A depicts a multi-layer electrode pattern that is substantially orthogonal. Figure 8B depicts electrodes in a pattern of concentric circles. Figure 9 depicts a continuous flow device allowing a sample suspended within the ferrofhiid to enter the device inlet and pass through the separation chamber, and exit via multiple outlets designed for capturing particles of a particular size. The 2μι ^η particles flow to outlet A, the 5μιη particles flow to outlet B, and the remaining sample flows to a waste outlet C. The flow device depicted in Figure 9 is thus suitable for sorting two or more particle types based on one or a combination of size, shape and elasticity.

DETAILED DESCRIPTION

The present invention relates to a microfluidic platform that uses biocompatible ferrofhiids for the controlled manipulation and rapid separation of both microparticles and live cells. More particularly, the present invention relates to the high-throughput manipulation, label-free sorting and separating of eel Is via a concentrated ferrofhiid that is bio-compatible. Bio-compatiblity of the ferrofhiid is based on an effective balance, or concentration, of ionic surfactant, such as citrate. Bio-compatibility generally requires a neutral pH, a sufficient osmotic pressure on the cells, and a stable ferrofhiid (too much ionic content destabilizes the suspension). This low-cost platform exploits differences in particle size, shape, and elasticity to achieve rapid and efficient separation. Using microspheres, size- based separation is demonstrated with about 99% separation efficiency and sub- 10-μηι resolution in less than about 45 seconds. The present invention also provides for the continuous manipulation and shape-based separation of live red blood cells from sickle cells and bacteria. These demonstrations highlight the ability of fenOmicrofhiidics to significantly reduce incubation times and increase diagnostic sensitivity in cellular assays through rapid separation and delivery of target cells to sensor arrays. Definitions

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Although any methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, the preferred methods and materials are described,

As used herein, each of the following terms has the meaning associated with it in this section. The articles "a" and "an" are used herein to refer to one or to more than one (/.(?.,■ to at least one) of the grammatical object of the article. By way of example, "an element" means one element or more than one element.

"About" as used herein when referring to a measurable value such as an amount, a temporal duration, and the like, is meant to encompass variations of ±20% or ±10%, more preferably ±5%, even more preferably ± I %, and still more preferably ±0.1 % from the specified value, as such variations are appropriate to perform the disclosed methods.

Throughout this disclosure, various aspects of the invention can be presented in a range format. It should be understood that the description in range format is merely for convenience and brevity and should not be construed as an inflexible limitation on the scope of the invention. Accordingly, the description of a range should be considered to have specifically disclosed all the possible subranges as well as individual numerical values within that range. For example, description of a range such as from 1 to 6 should be considered to have specifically disclosed subranges such as from I to 3, from 1 to 4, from 1 to 5, from 2 to 4, from 2 to 6, from 3 to 6 etc., as well as individual numbers within that range, for example, 1 , 2, 2.7, 3, 4, 5, 5.3, 6 and any whole and partial increments therebetween. This applies regardless of the breadth of the range.

Separation Systems and Devices

In one aspect, the present invention includes a microfluidic system, based on ferrohydrodynamics for the label-free manipulation and separation of cells and

microorganisms within biocompatible ferrofluids. In one embodiment, the system includes a water-based ferrofluid ts used as a uniform magnetic environment that surrounds the cells or other particles within a microfluidic channel. Cells and other nonmagnetic particles within the ferrofluid act as "magnetic voids" (Kashevsky, 1997, Phys Fluids 9: 1 81 1-1 818), in a manner analogous to electronic holes in a semiconductor. An externally applied magnetic field gradient can attract magnetic nanoparticles, which causes nonmagnetic microparticies or cells to be effectively pushed away (Rosensweig RE (1997) Ferrohydrodynamics (Dover, New York); Odenbach S (2002) Ferrofluids: Magnetically Controllable Fluids and Their Applications (Springer, New York)). Recently, this principle has been applied to capture nonmagnetic microbeads between magnetic film islands in a microchannel filled with ferrofluid (Ye!len et a!., 2005, Proc Natl Acad Sci USA 102:8860-8864). In contrast to this, the present invention can utilize a microfluidic device with integrated copper electrodes that carry currents to generate programmable magnetic field gradients locally, as depicted in Figure 1 A.

For example, the particle manipulation device of the present invention includes two primary components: 1 ) a microfhiidic channel; and 2) underlying electrodes. At least one electrode layer may sit atop a standard, insulated metai substrate. For example, an aluminium substrate coated with an insulating polymer may be used, which al lows efficient heat sinking, and enables AC currents up to 10A at low voltages through the electrodes. In one exemplary embodiment, a single electrode layer is used, as may be seen in Figure I A, In other exemplary embodiments, multiple electrode layers are used to provide multi-dimensional control, for example, such as the orthogonal electrode layers depicted in Figure 8A. Within a given electrode layer, the electrodes may be about 30 μιη high, about 300 μιη wide and about 2 cm in length. In alternative embodiments, the electrodes within a given layer may range anywhere from about 5- 100 μηι high, about 0.01 - 1 mm wide, and about 0.1 -10 cm in length, and any whole or partial increments therebetween. Thus, it should be appreciated that the size of electrodes utilized is not limited, and can vary across multiple electrode layers.

Further, the electrodes may include any shape, curvature or pattern, and may include variable gap sizes between electrodes. For example, Figure 7A depicts electrode formations of 100, 150, 200 and 300 μιη spacing in the channel region. In another exemplary embodiment, a multi-layer, orthogonal pattern may be used, as depicted in Figure 8A. In yet another exemplary embodiment, a concentric circle electrode pattern may be used, as depicted in Figure 8B, such that traveling waves effectively move the particles or cells into the circles, out of the circles, or trap them in various portions of the circles. In yet another exemplary embodiment, the electrodes may be "wiggly", or generally non-linear, so as to introduce disturbance forces and torques on nearby particles or cells. These wiggly regions may be uniform, non-uniform, random and/or periodic in nature throughout the electrodes, Again, it should be appreciated that the shape, spacing and pattern of electrodes may vary within a given electrode layer, and may further vary across multiple layers, such that any combination of shape, size, spacing and patterning may occur with an electrode layer and across multiple layers to create the desired magnetic field. The electrodes may be composed of any suitable conductive material, such as copper, as would be understood by those skilled in the art. The electrodes can be fabricated by wet etching the copper layer of a thermal-clad printed circuit board (on an insulated metal substrate) through a photoresist mask. It should be appreciated that any type of etching or other suitable fabrication method may be used in creation of the electrodes, as would be understood by those skilled in the art. The channel can include at least one inlet and at least one outlet, and may run at an angle such that the channel ultimately transverses the electrodes. For example, in one embodiment, the microfhiidic channel can be rotated by about 90 degrees so that the electrodes of the device are substantially parallel to its length, in other embodiments, the channel transverses the electrodes at an angle between about 1 -90 degrees, and any whole or partial increments therebetween. In a further exemplary embodiment, the channel transverses the electrodes in a substantially straight line. In still other exemplary embodiments, the channel transverses the electrodes in a curved, bent or generally non-linear pattern. The microfluidic channel can range from 20- 100 μιη high, 1 -3 mm wide and 2-3 cm in length, and any whole or partial increments therebetween. In other exemplary embodiments, the channel may include any number and size of pockets, ridges, grooves, trenches, and/or slopes within the channel walls, such that the particles or cells traveling within the channel can be locally concentrated or dispersed, based on the conformational effects of the contours of the channel walls. Additional exemplary channels, having multiple inlets and outlets, are depicted in Figure 7B. The channels may be composed of any suitable material as would be understood by those skilled in the art. In one embodiment, the channel may be prepared from polydimethylsiloxane (PDMS) stamps through soft lithography, and bonded to an insulating layer of very thin PDMS covering the electrodes (Mao et al., 2006, Nanotechnolog 17:34- 47). In certain embodiments, the channel height may be selected to be well below the optimum for localized fei ohydrodynamic flow, in order to minimize its potential effects on particle migration. While not required, the channel can be washed with a 1 % triton-X solution for about 10 minutes before introducing the ferrofluid/microsphere mixture into the microfluidic device, to minimize particle attachment to the PDMS walls, It should be appreciated that the substrate, insulating layer and channel can each alternatively be composed of materials having similar features and/or properties, as would be understood by those skilled in the art. Thus, the devices of the present invention can be constructed on an inexpensive printed circuit board that features an insulated copper layer etched via a single, low-resolution transparency mask to define the electrodes. The microfluidic channel can be constructed via soft lithography using a low-resolution mold. In certain embodiments, device fabrication does not necessitate a clean room, and hence, can be fabricated rapidly and inexpensively.

A travelling magnetic field may be generated within the channel from a power source by applying current to the electrodes to create a magnetic field pattern along the length of the microfluidic channel. Alternating currents up to about 7 A peak to peak in amplitude and with frequencies from about 10 Hz to 100 kHz, which correspond to a maximum magnetic field strength of about 90 Oe within the ferrofluid, can be applied to the electrodes, in other exemplary embodiments, the generated magnetic field strength may range between 1 -200 Oe, and any whole or partial increments therebetween. In one example, the magnetic field is generated by applying alternating currents in quadrature to a single layer of electrodes, to create a periodic magnetic field pattern that travels along the length of the microchannel, With this configuration, the device is able to create both magnetic field gradients, resulting in a time-average force on the cells or particles, and local rotation of ferrofluid magnetization, which eventually results in torque on the nonmagnetic particles, as illustrated in Figure IB.

When the current is turned on, the cells or particles are pushed away from the electrodes to the top of the channel, due at least in part to magnetic force, where they start to rotate and roll along its length, due at least in part to magnetic torque. The device behavior can thus mimic the frequency-dependent susceptibility of the particular ferrofluid used. For a given particle size, its speed may depend on the local force and torque values along the channel length, as illustrated in Figure 1 C. For example, at low frequencies, the force dominates, pushing the nonmagnetic micropai icles up to the channel ceiling and into the space between the electrodes. In another example, at high frequencies, the rolling microparticles can overcome the diminishing repulsion caused by magnetic force and move continuously along the channel, as illustrated in Figure I D.

Using the aforementioned microfluidic setup, the typical magnetic force that can be applied on a particle several micrometers in diameter can be on the order of tens of piconewtons, which is significantly larger than what is typical with optical tweezers on μηι- size particles. In certain embodiments, this actuation force can be increased by applying larger excitation currents, For example, a simple heat sink can maintain the channel contents at room temperature up to 10-A peak-to-peak input current (Mao L, oser H (2006) Toward ferrofluidics for μ-TAS and lab on-a-chip applications. Nanotecfmology 17:34— 7).

Biocompatible Ferrofhi ds

FeiTofluids are colloidal mixtures of nanometer sized magnetic particles, such as cobaltferrite, covered by a surfactant, suspended in a carrier medium that is compatible with the surfactant material. For example, a sample reaction that results in magnetite particles is as follows:

2 FeCB + FeC!2 + 8 NH3 + 4H20 → Fe304 + 8 NH4CI A 10% by volume suspension of magnetite has a saturation magnetization of around 560 G. The magnetization of each single-domain particle responds to a high magnetic field with a time constant on the order of 10 \is. High magnetic field gradients can be used to position the ferro fluid. "Spikes" and other interesting features may appear at the ferro fluid surface in the presence of such high fields,

In one embodiment, particle diameters can range from about l nm to about 100 nm, and any whole or partial increments therebetween. For example, and without limitation, the particle diameters can range between 1- 10 nm, 1 -20 nm, 5-50 nm, or 10- 1 00 nm . In a preferred embodiment, particle diameters average about 10 nm. Volume fractions may range from 0.1 % to about 10%, and any whole or partial increments therebetween.

In another embodiment, the ferrofluids of the present invention are biocompatible and can sustain live ceils for several hours without deterioration in physical properties or sustainabiiity, allowing for extended examination of the target sample. The biocompatible ferrofluid can be suitable for sustaining any living cell type and/or shape, such as any animal or plant tissue cell type, any microorganism, or any combination thereof, for example. Of course, the ferrofluid is also suitable for suspending any type of particle, and for any sized or shaped particle, or particle clusters or clumps, whether living or non-living.

Citrate is an effective surfactant in ferrofluids, and it is mostly biocompatible in cell cultures as well. Therefore, in one embodiment, citrate was utilized both to stabilize the ferrofluid and to provide an ionic medium for the cells to survive in. In this context, determining a novel and optimum citrate concentration was necessary, as too little or too much citrate would result in particle aggregation and precipitation within the ferrofluid. Further, cell survival within the ferrofluid depends on having enough ionic species to control the osmotic pressure on the cells to promote sustainabiiity. In one exemplary embodiment, a citrate concentration within the ferrofluid resulting in a stable colloidal suspension of magnetic nanoparticles was about 40 mM. While higher citrate concentrations might begin to gradually destabilize the ferrofluid, concentrations of citrate may range anywhere between 5- 200mM, and any whole or partial increments therebetween, depending on the characteristics and type of ferrofluid used, in the exemplary embodiment depicted in Figure 2C, it was also determined that the minimum concentration of citrate, stabilized with citric acid to result in a pH of about 7.4, that first resulted in substantial cell survival over the course of several hours was about 40 mM. At this ionic concentration, the cells were viable and the ferrofluid was stable. Hence, in a preferred embodiment, a citrate concentration of about 40 mM can be used as an effective biocompatible ferrofiuid. As contemplated herein, and depending on the type of ferroflitid used, the ferrofluids of the present invention may also be stabilized at pH ranges from between about 2- 1 1 , and any whole or partial increments therebetween. In certain embodiments, the ferrofiuid is biocompatible, such that ceils can survive within the ferrofiuid for at least about 1 hour, about 2 hours, about 3 hours, about 5 hours, about 10 hours, and even up to about 24 hours.

Calculations for Particle Manipulation

Provided herein is an analytical approach that enables the estimation of particle velocities and critical frequencies observed in the ferromicrofhiidic devices of the present invention. To simplify the calculations, a perfectly spherical, incompressible microparticie, a magnetically linear ferrofiuid, and a slip factor (see below) that is independent of field magnitude and frequency are assumed. Further assumed is that the spherical particle radius (R) is small compared to the wavelength of the travelling magnetic Field ( H ) ~ as determined by electrode dimensions and spacing, so that VH « H holds true. Under these assumptions, the ferrofiuid magnetization in the immediate vicinity of the microparticie - and, the virtual magnetization {M _ej j within the particle's volume (V _p ) - can be approximated as uniform. It is also approximated that any field value (but not its gradient) is constant within the interior extent of the microparticie.

The total instantaneous force on that dipole is then given by

( 1 )

where B _in is the magnetic flux density within the spherical microparticie and the integration is over the internal volume of the particle (Zahn et al,, 1995, J of Magnetism and Magnetic Materials 149: 165-173). Under these assumptions, the surface terms in ( 1 ) due to discontinuities in and B create pressure terms that integrate out to 0, and a simple vector expansion of the integrand reveals that this term is the same as the Kelvin force density. Hence, the instantaneous force expression can be simplified to

^ V^M^ - B^) (2)

To obtain an eventual analytical expression for the magnetic force, it is helpful to express M^-md ¾ in terms of the external magnetic field (H _exl ) in the absence of the microparticie, since this field value can be easily obtained from simple simulations. The net magnetization of the particle with a magnetic permeability μ _ρ (essential ly μά) within the ferrofluid (with a complex, frequency-dependent permeability ^ = μ^ + χ _; )) depends on as follows

Determ ining the magnetic flux density and field within the particle requires consideration of the demagnetization field inside it, The overall field inside the particle is

Hi» - H- e i ^~ H<{mag > ^w ' ^tn Hjmag ⁼ ^ iff ^ f° ^{r a} sphere. Hence, in the linear regime where particle magnetization can be written &s M = X _tif H _in , one finds

Xeff + Z<ff Comparing (3) and (4) reveals the effective susceptibility of the particle in terms of the ferrofluid susceptibi lity:

„ ^~ Xf

^{Xiff ~} + x _f

(5)

Note that the effective magnetic susceptibility depends on that of the ferrofluid, since the ni icroparticle responds to magnetic forces only because it displaces ferroflu id and creates a "magnetic hole". In that regard, the magnetic medium in which the hole resides determ ines the strength of the interactions between that magnetic hole and applied fields. The negative sign in (5) indicates that the effective magnetization of the niicroparticle is in the opposite direction of the local ferrofluid magnetization under static conditions, While x _(ff « -χ _} - for X _j « 1 , the effective susceptibility of the magnetic hole approaches - 1 in a strongly magnetisable medium , In effect, using too strong a ferrofluid for microparticle manipulation could be counter-productive.

The instantaneous magnetic force on the particle can be expressed as cos (9 + . (6)

Here, # is the angle between M _e!} and H _j , given by the angle of the complex susceptibility % _tff . Using Ad H _exl , we get

3 + 2χ

In the presence of traveling fields, the local magnetic field varies sinusoidally, and the time-average force is just half the maximum value of the instantaneous force:

Similarly, the instantaneous torque on a magnetic dipole is given by

= ίΚ, (Λ¾ x ¾). (Zahn et al., 1995, J of Magnetism and Magnetic Materials 149: 165- 173), Here, the possibility that the rotation of the nonmagnetic microparticle may be subject to a slip s between 0 and 1 has been allowed for, This slip factor represents the ratio of the torque that a non-magnetic microparticle experiences in the ferro-microfluidic devices of the present invention to the value of the torque that would be felt by an isolated particle of the same size and effective magnetization,

The remaining details of the derivation for magnetic torque mirror those for magnetic force. Substituting for the magnetic flux density, one obtains r _im = 5V _p (M x B _i ) = sV _{p (i} {M _eff x (H,„ + M _eff ))

(10)

Since M^ x M _eff = 0 and _t:// = ^H,,, , one

And time average torque is given by

A finite element analysis program (COMSOL) was used to calc late H _ex , for given input current amplitude using a realistic, two-dimensional cross-section of the

fei omicrofhiidic channel and the electrodes underneath. The Reynolds' number associated with the motion of micron-scale beads and cells in a quiescent ferrofluid is very small. In this regime, inertia! effects can be neglected and Stokes flow equations dominate hydrodynamics. Hence, the equilibrium between viscous drag and magnetic forces determine microparticle dynamics. Since Stokes flow equations are linear, all liydrodynamic coefficients involved can be combined into a resistance matrix:

(Happel J, Brenner H (1983) Low Reynolds Number Hydrodynamics with special applications to particulate media. ( artinus Nijhoff; Dordrecht)). Here, v is the linear velocity of the microparticle along the channel length, ω is its angular velocity, is the ferrofluid viscosity, R is microsphere radius, and J? is a resistance factor that depends on particle radius and its distance (Λ) from the channel ceiling. Assuming « R, these resistance factors can be obtained from standard lubrication theory as ^■ j \n(h/ R) + 0.9588; f ₂ * ~ )n(h I R) - 0.2526

(14) — In( h! R) - 0.1895; /, * -- ln(/;/ ?) + 0.3817.

(Goldman et al., 1967, Client Eng Sci 22:637-65 1 ) In general, it is possible to estimate h through Derjaguin, Landau, Verwey and Overbeek theory (DLVO theory) (Ise, 2007, Proc Jpn Acad B Phys Biol Sci 83: 192- 198) using the surface charge density on the microparticle and the channel surfaces, given the ionic conditions within the ferrofluid.

Interestingly, the vertical force that pushes the microparticles up to the channel ceiling is on the order of nN's and they are expected to be close to touching the channel wall.

Equation ( 1 3) can be solved for v and ω through a simple matrix inversion,

( 1 5)

where

- I _

Here, G≡ f _t f ₄ - _{2 3} has been defined for notational convenience. Hence, particle linear velocities due to magnetic force and torque alone can be determined: forci.x (16)

6nnRG ^F " ^{yf X} sf ₂

lorqiie.x 2^-, ' <nv,y' ( 1 7)

SrrnR'G rticle velocity is then given by

V = V , + V

x ' /οην,τ ( 1 8)

Both maguetic force and torque scale with particle volume (R ³ ) from (16) and ( 17), it is clear that particle velocity due to magnetic force depends on R ² , whereas that due to torque scales with R. This observation indicates that torque effects on smaller particles is relatively more significant, and explains why smaller microparticles display smaller critical frequencies in their dynamics. The preceding theoretical approach explains the experimental results very well (e.g., Fig. 4A) for a slip factor of 1 and h of about I nm, confirming the expectation that the microparticles are indeed pushed strongly towards the channel ceiling. The slip factor of 1 implies that the microspheres rotate under no-slip conditions.

Force on a Magnetic Dipole

In general, the magnetic force on a magnetic dipole can be found using the Kelvin force expression, i.e.,

F = μ _ύ \ {Μ - V)HdV (19) This expression is approximately equivalent to equation ( 1 ).

A key assumption is that the applied field is not too inhomogeneous and the particle radius (R) is small enough, such at 7?|vH « H| in any direction. Under this assumption, the magnetization of the feriOfluid immediately surrounding the microparticle can be taken as uniform. It is further approximated that any field value (but not its gradient) is constant within the interior extent of the microparticle,

With these simplifying assumptions in mind, vector identities can be used to rewrite the integrand of ( I ) as follows:

V( . B) = M x ( V x B) + B x (V x M) + (M■ V)£ + (B■ V)M. (20)

The first term on the right-hand-side (RHS) of (20) involves the curl of the magnetic flux density, which could be expanded as

M x (v x B ) = M x (v x H ) + M x (v x M ) (21 )

The curl of the magnetic field is 0 everywhere within the integration volume of ( 1 ), since both the ferro fluid and the plastic microparticle are insulating and do not support electrical currents. Hence, the first term on the RHS of (21) vanishes. The curi of the magnetization is 0 inside the microparticle, but across its surface, the effective magnetization changes as a step. Therefore, surface contributions to the force density should, in general, be considered, However, since the magnetization of the ferrofluid immediately surrounding the

microparticle is assumed to be constant, the second term of (21 ) also vanishes when integrated around a sphere (due to symmetry). The same logic could be applied to the B x (V x M) term in (20): (V x M) is 0 inside the m icrosphere and B x (V x M) integrates to 0 around the sphere surface with the magnetic flux density and ferrofluid magnetization assumed constant in the immediate vicinity of the microsphere.

With the same reasoning, the (B■ V)M term in (20) will also be 0 inside the microsphere and integrate to 0 around it. The only term that involves the non-zero gradient of a field vector then is

The integral is valid over the volume of the microparticle, which is nonmagnetic. Hence, inside the microparticle, B = _j u ₀ Fl and (22) becomes the same as ( 19). In other words, under the assumptions outlined above, ( 1 ) and (19) are equivalent in the case of the setup presented herein.

The equation in ( 1) is preferrably used as the force expression instead of that in ( 19), since the former leads to a force whose direction is determined by the gradient operator - requiring taking a single derivative along a given spatial direction to determine the force along that direction.

Also considered is what happens to surface gradients associated with the expression in

( 1). Once again, using the assumption that the ferrofluid magnetization and the magnetic flux density are constant inside and just around the microsphere surface (but not their derivatives), it can be determined that

Here, the field and magnetization vectors are taken to be in the x-z plane due to the symmetry of the ferro-microfiuidic channel. In the previous section, the expression in (23) is evaluated only within the interior of the microparticle to calculate the x-directed force. Now, without any loss of genera lit)', the microsphere center is taken as the origin. Both M and B are discontinuous at the particle-ferrofluid boundary, so their derivatives result in an impulse; when integrated across the microsphere surface, each contribution M _{X tlut} B^ _mil - A4 _{l n} B _{I n} \o the integral from a surface patch at Λ ^' = ^R ² - y ² - z ² gets cancelled out by the negative of that contribution at the opposite patch at x = . The resulting surface integral is 0. By spherical symmetry, the same is true for the terms in F _z . Hence, under the described assumptions, evaluating equation (1) within tiie interior of the microparticle yields the magnetic force on it,

System and Methods of Separation

The present invention also relates to a flow-based assay system incorporating the aforementioned biocompatible ferrofluids. In one embodiment, a much higher throughput can be achieved by conducting the bioferrofluidic separation while fluid flow continuously introduces fresh cells into the inlet of the channel. At the outlet, the incoming beads or cells are sorted into different output channels. From there, the cells can be collected for inspection or directed towards an external or internal (i.e., integrated) sensor. For example, the flow device allows a sample suspended within the ferrofluid to enter the device inlet and pass through the separation chamber, and exit via multiple outlets designed for capturing particles of a particular size. For example, as depicted in Figure 9, 2μηι particles can flow to outlet A, 5μηι particles flow to outlet B, etc., and the remaining sample flows to a waste outlet C, In certain embodiments, flow is not needed to direct cells. Instead, magnetic excitation can be used to direct them. Further, the sensor may be directly integrated into a side pocket along the flow channel, for example.

Manipulation is not only dependent on cell size, but also on the shape and elasticity of the cells. For example, the variable size, shape and elasticity of bacteria and sickle cells allows them to be separated from healthy blood cells. According to an aspect of the present invention, particle separation can be dependent on size and frequency, in another embodiment, critical frequency, as described hereinthroughout, may also depend on the electrode gap. For example, larger microspheres can be trapped first in smaller gaps. By utilizing this phenomenon, sorting can be performed based on particle or target size. Further, the system may alternate between the manipulation excitation at the chosen frequency and another frequency that helps break possible nanoparticle chains that may form due to the magnetic excitation. According to an aspect of the present invention, "wiggly" electrodes can be used to prevent beads or ceils from clustering into large chunks as they flow down the channel. The "wiggly" electrodes introduce disturbance forces and torques on the beads or cells to break clusters apart, allowing for larger individual beads or cells to line up like pearls on a necklace. By periodically breaking the nanoparticle chains, the ferrofluid physical properties are kept constant over time. Further, target cells can be concentrated, trapped, localized, or simply directed toward sensor surfaces efficiently, rapidly, and in a label-free fashion. For example, the method of separation can direct a cell or particle type based on size, shape or elasticity into an outlet, or by trapping a cell or particle type based on size, shape or elasticity via increasing or decreasing the spacing between electrodes.

Thus, the present invention includes a method for separating at least one cell type from a sample. The method includes the steps of suspending cells in a biocompatible ferrofluid to form a sample, passing the sample through a microfHtidic channel that transverses a plurality of electrodes at about 90°, such that the plurality of electrodes are substantially parallel to the length of the microfluidic channel, applying a current to the plurality of electrodes to create a magnetic field pattern along the length of the microfluidic channel, and sorting the cells into at least one output channel based on a variation of at least one of cell size, shape and elasticity. Separation can occur via concentrating, trapping, localizing, or simply directing toward sensor surfaces efficiently, rapidly, and in a label-free fashion.

The present invention also provides for a method for separating ceils and/or particles based on side. This size-based separation can be demonstrated with about 50% efficiency, about 60% efficiency, about 70% efficiency, about 80% efficiency, about 90% efficiency, about 92% efficiency, about 94% efficiency, about 96% efficiency, about 97% efficiency, about 98% efficiency, and about 99% separation efficiency. Size resolution in the separation process can be less than about 10 μηι, less than about 9 μιη, less than about 8 μιη, less than about 7 μηι, less than about 6 μηι, less than about 5 μιη, less than about 4 μηι, less than about 3 μιη, less than about 2 μιη, less than about l μιη, less than about 0.5 μιη, less than about 0.1 μιη, and less than about 10 nm, Such separation can be accomplished in less than about 2 m, less than about I m, less than about 45 s, less than about 30 sec, less than about 20 s, and less than about 10 s.

The present invention also provides for the continuous manipulation and shape-based separation of cells, such as live red blood cells from sickle cells and/or bacteria. These demonstrations highlight the ability of ferrom icrofluidics to significantly reduce incubation times and increase diagnostic sensitivity in cellular assays through rapid separation and delivery of target cells to sensor arrays.

The microfluidic system of the present invention has a number of unique advantages, in that it provides a laminar flow platform for use with tiny sample sizes. The system further provides fast diffusion and fast results, can be portable, and can be integrated with other existing sensors. For example, the present invention can be used for the sterilization of adult stem ceils obtained from blood samples, for use in the context of wound healing and organ regeneration for soldiers and marines in combat. The present invention can also be used for the rapid detection (i.e., < 1 min) of low-level bacterial contamination in donated blood. This can be particularly useful in battlefield trauma emergency situations. The present invention can also be utilized for "needle-in-a-haystack" applications that require the detection of ultra- low concentrations of cells in blood, such as searching for circulating tumor cells in blood.

EXPERIMENTAL EXAMPLES

The invention is now described with reference to the following Examples. These Examples are provided for the purpose of illustration only and the invention should in no way be construed as being limited to these Examples, but rather should be construed to encompass any and ail variations which become evident as a result of the teaching provided herein.

Without further description, it is believed that one of ordinary skill in the art can, using the preceding description and the following illustrative examples, make and utilize the present invention and practice the claimed methods. The following working examples therefore, specifically point out the preferred embodiments of the present invention, and are not to be construed as limiting in any way the remainder of the disclosure.

Ferrofluid preparation

A co-precipitation method was used to synthesize cobalt-ferrite nanoparticles that were eventually incorporated into a water-based ferrofluid with a 20% solid content

( halafalia SE, Reimers GW (1973) U.S. Patent No. 3,764,540). Cobalt-ferrite nanoparticles were precipitated out of a boiling solution of 1 M sodium hydroxide by adding a mixture of cobalt (il)-chloride hexahydrate and iron (lll)-chloride. The magnetic precipitate was washed twice using DI water. 2 M nitric acid and a 0.35 M solution of iron (lil)-nitrate were added to the precipitate (Massart, mi, IEEE Tram Magn 17: 1247- 1248; Fischer et al., 2008, IEEE In! Confon Ήαηο/Micro Eng and Molecular Syst China, 907-910). Th is mixture was then stirred at 80°C for 20 minutes. The nitric acid solution was then decanted while the precipitate was held in place with a magnet. Cobalt-ferrite particles within the precipitate were later dispersed in DI water, and the resulting ferrofluid was dialyzed for one week against a 40 niM sodium citrate and citric acid solution at a pH level of 7.4. The solution was refreshed on a daily basis during dialysis. The resulting ferrofluid had a viscosity of 1.5 cP at 20°C. TEM Sample Preparation

The TEM images were taken using a Tecnai 12 electron microscope from Philips ( 120 keV). A copper/rhodium grid (from Electron Microscopy Sciences) was covered with a thin carbon film and dipped into a ferrofluid sample diluted with ethanol. After TEM images were obtained, particle sizes in the images were characterized using ImageJ software

(http://rsbweb.nih.gov/ij/), The distribution of magnetic nanoparticle core sizes, as obtained from the TEM images (around 200 particles counted), was fitted with a lognormal probability density function as where D is the random variable depicting core diameter, while Z¼ find (rare the mean and standard deviation of ln(D), respectively.

AC susceptibility measurements

The frequency-dependent AC susceptibility of the ferrofluid can be obtained by measuring the changes in the mutual inductance of an electromagnetic coil pair, with and without the presence of a ferrofluid (Maiorov, 1979, Magnetohydrodynamics 15 : 135- 139). In this regard, a pick-up coil (of 200 turns, with an average diameter of 9,76 mm) was centered within a solenoidal excitation coil (of 340 turns with an average diameter of 13,34 mm), and the mutual inductance of the two coils was characterized via an LCR meter from Agilent (E4980A). The ferrofluid sample was introduced within the two sets of coils in a 1 cc plastic syringe. The symmetry in the setup ensured parallel field lines at the location of the pick-up coil and enabled an analytical calculation of mutual inductance, and eventually, AC susceptibility, from measured data.

The magnetization relaxation equation, assuming no fluid motion or convection

(Rosensweig RE ( 1997) Feirohydrodyna ics (Dover: New York)), is

where ω is the local vorticity within the ferrofluid, χ ₀ is the DC susceptibility value of the ferrofluid, and r is the magnetic relaxation time constant associated with the magnetic nanoparticles. The uniform magnetic field within the cylindrical setup leads to a symmetry that makes vorticity (and hence, tiie second term in (25)) negligible within the measurement volume. The magnetic relaxation time constant represents a combination of two physical relaxation processes. If the magnetic cores of the particles are small enough, their magnetic moment will simply rotate inside the iianoparticles (Νέεΐ relaxation) (Rosensweig RE ( 1997) Ferrohydwdy tmics (Dover: New York)) with a characteristic time constant given by

( J'f k.T)

(26;

0

whereto is a precession frequency (typically in the range 10 ⁸ -10 ¹² Hz), K _a is the magnetic anisotropy energy density, V _core is the magnetic core volume of the nanopa ticle, and k^T is the thermal energy. Particles with larger cores will have higher magnetic anisotropy energies, leading to fixed magnetic moments within the cores, and the particles themselves will rotate in solution to orient with the applied field (Brownian relaxation), with a characteristic time constant given by

2k _B T

Here, η is the dynamic viscosity of the fluid, k _B is the Boitzmann's constant, T is the absolute temperature (in Kelvins), and D _hyci is the hydrodynamic diameter of the particle, including its surfactant layer. The faster of the two mechanisms dominates the relaxation process. Cobalt- ferrite possesses a high magnetic anisotropy energy density (between ! .8x l 0 ⁵ and 3.0x l0 ⁵ J/m ³ for bulk material and up to 3. 15 x 10 ⁶ J/m ³ for nanoparticles (Tung et at., 2003, JAppl Phys 93:7486-7488)), and ferrofluids based on this material relax primarily by particle rotation (Brownian mechanism) above a critical nanoparticle size of about 5 nm in diameter, Since most of the nanoparticles observed in the ΊΈΜ pictures were larger than this critical size, only the Brownian time constant was considered in interpreting our AC susceptibility measurements.

The sinusoidal steady-state solution to (25) in the absence of vorticity yields the concept of an effective susceptibility that describes the magnitude and phase relationship between ferrofiuid magnetization and the a lied field as a function of frequenc : (Debye PJW ( 1 29) Polar Molecules, (Dover: New York)) Here, χ ₀ is the DC susceptibility value of the ferrofluid and r is the Brownian relaxation time constant associated with the magnetic nanoparticles.

Ferrofluids consist of particles with a size distribution (typically !ognormat), which leads to a distribution of the relaxation times as well. To take this into account, we describe the overall AC susceptibility as a linear combination of all susceptibility spectra that would result from the particle sizes present in the ferrofluid, weighed by the !ognormal probability density function F(D _hy d) associate ith a given particle size:

The magnitude of the total magnetization within a nanoparticle is proportional to its core volume; so is its individual contribution to the susceptibility spectrum. Hence, the probability density function in (24) is scaled by ^ _w , The normalization factor A is iven by

The AC susceptibility data (Fig. 2B) can be fit with the sinusoidal steady-state solution to the magnetic relaxation equation assuming a log-normal distribution of hydrodynamic diameters (equations (27) through (30)). Simultaneously, the relative shape of the DC magnetization data (as depicted in Fig. 2B-inset) can be fit with the Langevin equation, once again assuming the same log-normal distribution of hydrodynamic diameters (and particle concentration as a free parameter). The simultaneous fits explain the experimental results very well, yielding an average hydrodynamic diameter of 72.5 nm. This value is much larger than the average core diameter of the nanoparticles as obtained with TEM. A reasonable explanation for this discrepancy is that, in equilibrium, the nanoparticles have formed moderate-sized aggregates that respond as single units to the magnetic fields that were applied during measurements. Dynamic light scattering experiments were also conducted on diluted samples of the same feiTofluid. Those results confirmed that the hydrodynamic diameters were much larger than the core diameters, supporting the explanation presented herein.

Typically, the surfactant concentration used is high enough to prevent continuous degradation in colloidal stability (at least over several months). Therefore, it is likely that the particle aggregates may have formed during one of the brief precipitation stages of the feiTofliitd synthesis protocol, which often involves the use of a permanent magnet to speed the process. The surfactant is added later, and cannot break aggregates that have already formed. Dynamic Light Scattering

The dynamic light scattering experiments were conducted using a ZetaPALS instrument from Brookhaven Instruments Group. For these measurements, the ferrofluid was diluted with DI water to avoid multiple scattering, The hydrodynamic particle diameter was found to be 64.9 nm.

Device Fabrication

The particle manipulation device (Fig. 1 A) used in the experiments presented herein consists of two parts: the microflitidic channel and the underlying copper electrodes. The electrodes (30 μιη high, 300 μιη wide and 2 cm in length) were fabricated by wet etching the copper layer of a thermal-clad printed circuit board (on an insulated metal substrate) through a photoresist mask. A travelling magnetic field was generated in the channel by applying alternating currents in quadrature to a single layer of electrodes. The microfluidic channel (20 μιη to 100 μιη high, ! mm to 3mm wide and 2 cm to 3cm long) was prepared from polydiniethylsiloxane (PDMS) stamps through soft lithography and was bonded to an insulating layer of very thin PDMS covering the electrodes (Mao et ai., 2006,

Nanotechnology 1 7:34-47). The channel height was chosen to be well below the optimum for localized ferrohydrodynamic flow in order to minimize its potential effects on particle migration, in separate experiments with sub-micron tracer particles, no discernible hydrodynamic flow was observed. The insulated metal substrate allows efficient heat sinking, enabling AC currents up to 3 OA at low voltages through the electrodes. Before introducing the ferrofluid/microsphere mixture into the microfluidic device, the channel was washed with a 1% triton-X solution for about 10 minutes in order to minimize particle attachment to the PDMS walls. Microspheres

Different sizes ( 1 .2 μηι, 1.9 μηι, 2,2 μιη, 3.1 μιη, 5,0 μηι, 6.0 μιη, 9.9 μηι in diameter) of green fluorescent polystyrene microspheres were obtained from Duke Scientific (Fremont, CA, USA). The coefficient of variation on the microsphere diameters was about 1 %. These custom produced microspheres had a very low porosity and carried a minimal amount of charged groups on their surfaces. Microspheres were suspended in deionized (DI) water and kept at 4°C until they were used in ferrofluid experiments.

ΡΚΉ cell staining

To make the cells visible within the ferrofl id, blood cells were stained with green fluorescent membrane dye P H67 (obtained from Sigma-Aidrich), This dye has an excitation peak at 490 nm and emission at 502 nra (Horan et ai., 1989, Nature 340: 167 - 168). Cell staining was performed by following the manufacturer's protocol with some modifications for our study.

General Preparation Protocol

Blood was drawn from donors prior to the experiments and kept at 4°C prior to staining. Approximately 10 million cells were centrifuged and the plasma was subsequently removed. The cells were then suspended in 500 μ Ι RPMl 1640 culture medium without serum (obtained from Invitrogen, Carlsbad, CA, USA) and mixed well to remove any adherent and bound cells. The resulting suspension of cells was centrifuged again for 5 minutes at 1000 ipm.

The supernatant was carefully aspirated and the pellet was suspended in 500 μΙ

Diluent C (supplied with the staining kit). Immediately after this, 4 micromolar PKH67 dye in Diluent C was prepared. Equal volumes of dye and cell solutions were mixed. The resulting cell suspension was incubated for 4 minutes, avoiding exposure to light. The staining reaction was stopped by adding an equal volume of fetal bovine serum (FBS) and the cell suspension was further incubated for 1 minute. The cells were then centrifuged for 5 minutes at 1200 rpm to remove the staining solution. They were washed three times in cell culture containing 10% FBS to remove any remaining dye in the solution. After washing was complete, the cells were suspended in culture medium. The brightness of labelled cells was tested with fluorescence microscopy. Before mixing with ferrofluid, the stained cells were washed with Dulbecco's phosphate buffered saline (PBS) buffer containing 10% FBS, Cell Viability Test and Cell Counting

Citrate is an effective surfactant in ferrofluids, and it is mostly biocompatible in cell cultures as well. Therefore, citrate was utilized both to stabilize the ferrofluid and to provide an ionic medium for the cells to survive in. In this context, determining a novel and optimum citrate concentration was necessary, as too little or too much citrate would result in particle aggregation and precipitation within the ferrofluid. Further, celt survival within the ferrofluid depends on having enough ionic species to control the osmotic pressure on the cells to promote sustainability.

The highest citrate concentration within the ferrofluid that still resulted in a stable colloidal suspension of magnetic nanoparticles was determined to be about 40 mM. Higher citrate concentrations would begin to gradually destabilize the ferrofluid.

Cell viability was monitored using the Trypan Blue (obtained from Invitrogen) staining technique. Trypan Blue is a dye that selectively stains dead cells blue, allowing live and dead cells to be distinguished (The Sigma-Aldrich Handbook of Stains, Dyes &

Indicators, Green, F. J., ed., Aldrich Chemical Co. (Milwaukee, WI: 1990), 721 -722).

Following the manufacturer's protocol; 90 μ Ι of 0.4% Trypan Blue Stain was added to 10 μ ΐ of the cell suspension with a concentration of 5 x 10 ^s cells per 1 ml of culture medium. After incubation for 5 minutes at room temperature, a small sample from the mixture was placed onto a hemocytometer to count live cells. It was determined that the minimum concentration of citrate (stabilized with citric acid to result in a pH of 7.4) that first resulted in substantial cell survival over the course of several hours was 40 mM (Fig. 2C). At this ionic

concentration, the cells were viable and the ferrofluid was stable. Hence, in all experiments involving cells suspended within our ferrofluid, a citrate concentration of 40 mM was used. Example 1 : Ferrofluid Properties and Device Characterization

Using highly concentrated ferrofluids with live cells has traditionally proven to be a challenge, because it requires a carefully engineered colloidal system. The ferrofluid parameters that are most relevant to sustaining live cells include pH, ionic strength, and nanoparticle-surfactant combination, together with their overall and relative concentrations. Finding the right nanoparticle-surfactant combination is crucial in this regard: the ferrofluid needs to be stable at a pH of 7.4, and colloidal stability has to be maintained up to an ionic strength that can sustain live cells. One also needs to pay special attention to the size distribution of the nanoparticles within the ferrofluid. If there exist nanoparticles only a few nanometers in diameter, they could pass through the cell membrane and cause direct cytotoxicity (Scherer et al, 2005, Brazilian J Phys 45:718-727). For this reason, the present invention includes a magnetic precipitation step in the synthesis of the biocompatible ferrofluids to specifically leave the smallest nanoparticles behind.

Traditional approaches to improving ferrofluid biocompatibility typically involve covering the magnetic nanoparticles permanently with a thick polymer layer, such as dextran (Bautista, et al., 2004, Ncmo/eclmo/og}' 15:S I 54-S 159), because the surfactant molecules reduce toxicity by impeding direct contact with the surface of the inorganic nanoparticles. However, such an approach leads to a significant reduction in the volume content of the magnetic nanoparticles within the ferrofluid, and a corresponding decline in its susceptibility. Higher ferrofluid susceptibility typically translates to faster particle manipulation, so the ferrofluid of the present invention has been optimized by using a short surfactant molecule.

In one embodiment, the ferrofluid of the present invention comprises cobaltferrite nanoparticles suspended in wa er and stabilized with citrate. Mean nanoparticle core diameter within the ferrofluid, as determined with transmission electron microscopy (TEM), was found to be about 1 1.3 ± 4.4 nm (Fig, 2A). From simultaneous fits to ac susceptibility and dc magnetization data (Fig. 2B), the average hydrodynaniic diameter was determined to be about 72,5 nm. The discrepancy between the average hydrodynamic diameter and the individual core sizes observed in TEM images points to a certain degree of particle aggregation within the colloidal suspension of the ferrofluid, This finding was also confirmed through dynamic light scattering measurements, which yielded an average hydrodynamic diameter of about 64.9 nm on highly diluted samples of ferrofluid. Nevertheless, compared with the μηι-sized microspheres and cells, the magnetic nanoparticles were still small enough to approximate the ferrofluid as a continuous magnetic medium.

During synthesis, it was determined that the optimum ionic concentration within the ferrofluid to provide a good compromise between cell viability (as determined by the trypan blue test) and ferrofluid stability was about 40 iiiM (Fig. 2C). During the course of a given experiment, cells retained their viability. We observed that 75% of cells remained viable, even after being suspended in the ferrofluid for several hours, enabling extended tests involving live cell manipulation and separation.

Before the cell manipulation experiments, the feriOmicrofluidic devices were characterized by using fluorescent polystyrene microspheres (Duke Scientific; monodisperse sets with diameters ranging from 1 .2 to 9.9 μηι). To understand the influence of excitation frequency and current amplitude on the behavior of nonmagnetic microparticles dispersed in ferrofluid, a series of experiments were performed using different sizes of microspheres at various excitation frequencies and current amplitudes. Microspheres of a given size were mixed with the ferrofluid in small quantities (up to 1.1 ^χ I 0 ⁶ microspheres per mL for the smallest microparticle diameter) and subsequently added to the microfluidic channel. The channel inlet and outlet were clamped at both ends to prevent transient fluid motion.

Microspheres near the roof of the microchannel were imaged from above with an upright fluorescent microscope (Zeiss Axioimager A I ) and a high-sensitivity video camera (Retiga 2000R) using StreamPix software. Image analysis was performed offline in MATLAB (Math Works) via an optical flow algorithm. The program could automatically track the trajectory and determine the size of thousands of individual microspheres within the field of view in <1 min.

During these experiments, two types of particle dynamics were observed. At low frequencies, the microspheres localized between the electrodes, where repulsive forces caused by magnetic field gradients form local minima (Fig. 3A). Frequencies above a critical value, f _c> led to continuous translation of the microspheres along the length of the channel roof; this critical frequency depended only on particle size and electrode spacing, not on input current amplitude (Fig. 3B), The average velocity of microspheres of a given size depended on the excitation frequency, current amplitude, and their location with respect to the underlying electrodes.

In experiments with various microsphere diameters, a monotomc increase was found in critical frequency with increasing particle size (Fig. 4A), demonstrating the potential for size-based particle separation through excitation frequency control. This phenomenon may be explained through a simple hydrodynamic reasoning. Both magnetic force and torque scale with particle volume (Λ ³ ); the hydrodynamic drag that resists linear particle motion scales with R against force and R ² against torque that rolls the particle. Hence, linear particle velocity caused by magnetic force alone depends on R ² , whereas that caused by torque scales with R. This observation indicates that torque effects on smaller particles are relatively more significant and explains why smaller microparticles can overcome the repulsion of magnetic force traps and propagate continuously within the channel at lower frequencies. The solid curve depicted in Fig. 4A represents simulation results for critical frequency and explains the data very well for an average microsphere-wall gap of «1 nm and no-slip conditions applied to the rotation of the microspheres,

Fig. 4B shows the average velocity of 2.2- and 9.9-μιη microspheres (mixed in an 8 : 1 ratio within the same ferrofiuid) under excitation frequencies ranging from 10 Hz to 100 kHz. For a wide frequency range, the smaller particles translated continuously, whereas the larger particles were trapped between the electrodes. In this particular experiment and others, a mixture of particles/cells was eventually separated into two groups, e.g., those trapped vs, those cleared from channel. Assuming that the target particles/cells are those that are intended for trapping, the trapping efficiency can be defined as the ratio of the number of target moieties within the trapped group to their corresponding number in the initial mixture. Similarly, separation efficiency is defined as the ratio of the number of nontarget moieties within the cleared group to their corresponding number in the i itial mixture. However, particle/cell purity is simply the ratio of the number of target cells within the trapped group to the total number of cells in that group. At an excitation frequency of 400 Hz, 96.5% of the 9.9-μηι microspheres (167 of 173) were trapped within 10 s, whereas the 2,2-μι ^η particles (1,285 of 1 ,294) continued to translate along the channel and were cleared out of the observation window (45 s) without being trapped (Fig. 4 C and 4D) with a 99.3% separation efficiency. The particle purity in the trapped group was 94.9% (167 targets of 176 total trapped particles). Most of the small microspheres that failed to clear the channel were stuck on the polydimethylsiloxane (PDMS) wall in random locations, instead of being trapped between the electrodes. With better channel preparation, the separation efficiency and particle purity can be even higher.

Particle motion was also determined to depend on electrode spacing, with a smaller spacing leading to faster microsphere travel and a reduction in critical frequency (Fig. 6B). This phenomenon may be used in a device featuring regions of electrodes with different gaps to use the same excitation frequency to separate particle mixtures with more than two distinct sizes. One could also create an electrode pattern with a gradually increasing gap to sort particles based on size (Fig. 6C). In this context, it was observed that small nonuniformities in actual electrode spacing (caused by fabrication) partly determined the resolution of separation, defined as the minimiun size difference in particles that can still be separated with high efficiency (e.g., >90%). Given a range of particle sizes, this resolution of separation is directly related to the difference in the corresponding critical frequencies; under ideal conditions (i.e., perfectly controlled electrode gaps and a very dilute cell concentration), the resolution of separation could be arbitrarily small. However, critical frequencies tend to show slight local variations around each nonuniform electrode gap. As depicted in Fig. 4A, the ideal critical frequency depends nonlinear!}' on the particle radius; hence, a l -μιη difference in diameter between 9- and 10-μηι microspheres is easier to resolve (with slight random variations in electrode spacing) than one between 1 - and 2-μιτι particles. Ultimately, the resolution of separation that was achieved in the experiments presented herein was «1 μηι for particles 2 μιη or larger.

Example 2: Effects of Current Amplitute on Microparticle Speed

In additional experiments, the dependence of microparticle manipulation speed as a function of input current amplitude was determined, According to the calculations outlined in herein, and assuming the ferrofluid remains magnetically linear, the particle speed scales with the square of the input current. As illustrated in Fig, 6A, this assumption begins to break above 7A peak-to-peak input current amplitude.

Electrode spacing was also varied (with electrode width fixed at 210 μι ^η ) to determine its effect on particle manipulation. A smaller electrode spacing resulted in faster average particle speeds and lower critical frequencies (Fig. 6B). This observation could be explained by noting that electrodes spaced closer reduce the local magnetic field gradient that produces the magnetic force on the microparticles; closer electrodes also pack more energy into the fundamental of the travelling wave that produces the magnetic torque on these microparticles, A lower magnetic force and a higher torque result in faster microparticle rotation and overall travel speed; they also result in the critical frequency being lowered (Tung LD, et al. (2003) Magnetic properties of u!trafine cobalt ferrite particles, JAppl Phys 93:7486-7488). This observation supports the present invention for use as a microparticle sorting device in which the spacing between the electrodes could be gradually increased to trap increasingly smaller particles or cells (Fig. 6C).

Example 3: Separation of Live Cells

With the physical behavior of the ferromicrofluidic platform characterized, manipulation and separation experiments were conducted with live human red blood cells and bacteria to demonstrate the utility and practicality of the ferromicrofluidic devices of the present invention for biomedical applications. Red blood cells and Escherichia coli bacteria [K12 strain (Blattner et al., 1997, Science 277: 1453-1474)] were stained with a green fluorescent marker and mixed before suspension in ferrofluid. The average velocity for cells and bacteria within the channel was measured with 6 A of peak-to-peak current amplitude for frequencies from 10 Hz to 100 kHz. The critical frequencies for ceils and bacteria were found to be 21 5 and 77 Hz, respectively, These c values are somewhat lower than those found for comparably sized polystyrene microspheres, likely due to a combination of compliant shapes and nonspherical geometries that lead to increased difficulty of rolling along the channel roof. Moreover, bacteria and cells, with their complex surface chem istries, interacted with the PDMS channel more strongly (resulting in more prevalent cellular attachment) than the bare microspheres, indicating potentially higher effective kinetic friction coefficients between the cells and the PDMS surface. Fig. 5A depicts the spatially averaged linear velocity of cells and bacteria along the channel for an excitation frequency of 200 Hz. The smaller E. coli moved continuously along the channel (velocity points in Fig. 5 A do not cross zero) and eventually left the observation window, whereas blood cells were localized between electrodes (velocity points reach zero). Note that the larger variation observable in the red blood cell data of Fig. 5A stems from a statistical fluctuation: there are only a few red blood cells that passed through a given ,Y- location during the observation window, and their nonspherical shapes mean that each cell would be at a random angular orientation (and slightly different instantaneous velocity) as it rolled down the channel at that location. Bacteria, although varying in length and nonspherical, had enough numbers (several hundred through a given x- location) to result in good average statistics. In the end, ~6,750 of the 7,050 E. coli bacteria initially present within the field of view of the sample were cleared (95.7% separation efficiency) within 45 s. Of the 1 ,01 8 red blood cells initially present, 954 were trapped, corresponding to a trapping efficiency of 93,7% and cell purity of 76.1 %.

In a different experiment, healthy red blood cells were separated from those afflicted with sickle cell anemia by exploiting the shape and elasticity differences between them (Fig, 5B). A blood sample containing approximately a 4: 1 ratio of healthy-to-sickle red blood cells was added to the ferrofluid and introduced into the microchannel. At 300 Hz, sickle cells were trapped, whereas the healthy blood cells were cleared continuously from the channel (fluctuations in each dataset depicted in Fig, 5B are statistical in nature). In a sample initially containing «501 healthy red blood cells and 145 sickle cells, 300 healthy cells were cleared, whereas 109 sickle ceils were trapped. Assuming that the goal is to clear the sample from sickle cells, these numbers correspond to a separation efficiency of 75.2% ( 109 of 145 sickle cells were separated from healthy ones) and a healthy cell purity of 89.3% (300 healthy cells and 36 sickle cells were cleared).

These Examples demonstrate the use of ferromiciOfluidics in significantly reducing incubation times and increasing diagnostic sensitivity in cellular assays through rapid separation and selective delivery of target cells to sensor arrays. While manipulation and separation of micropaiticles and live cells within microfluidic devices is also possible through established techniques (such as DEP and magnetic label-based methods), the

ferroinicrofluidic approach of the present invention offers numerous advantages over existing methods. For example, target cells can be concentrated, trapped, localized, or simply directed toward sensor surfaces efficiently, rapidly, and in a label-free fashion. The biocompatible ferrofluid of the present invention can sustain live blood cells for several hours without deterioration in physical properties, allowing extended examination of the target sample.

When combined with a simple photodiode, ferromtcrofluidic separation of cells can provide a rapid, automated, and disposable blood assay that can count and estimate the concentration of any target cell type (such as bacteria or sickle cells) within 1 min, without the need for a microscope, pumps, or lengthy sample preparation steps. The present invention can also be used to selectively concentrate rare cells, such as circulating tumor cells in blood samples, by exploiting the differences in Young's modulus of the subject cell types (Lekka et a!., 1999, Eur Bioph s J 28:3 12-3 16). Applied in this manner, the present invention can increase detection sensitivity of existing cellular assays.

The present invention thus includes a cellular manipulation and separation platform using biocompatible ferrofluids within lowcost micro flu id ic devices. We have demonstrated highly efficient particle separation that is achievable in < 1 min. As an example, bacteria can be separated from live blood cells, and sickle cells can be separated from healthy red blood cells. In the case of a flow-based device, separation can be achieved with particle

manipulation perpendicular to the flow direction. By varying electrode geometry and input excitation frequency, the devices of the present invention can be tailored for different size ranges of particles and cells, Together with control of microchannel surface chemistry, the present invention can be integrated within lab-on-a-chip sensors and diagnostic systems to direct target cells toward active regions. In this manner, the present invention can

significantly reduce incubation times and increase the practical detection sensitivities achieved in existing sensors and diagnostic platforms. The disclosures of each and every patent, patent application, and publication cited herein are hereby incorporated herein by reference in their entirety.

While this invention has been disclosed with reference to specific embodiments, it is apparent that other embodiments and variations of this invention may be devised by others skilled in the art without departing from the true spirit and scope of the invention. The appended claims are intended to be construed to include all such embodiments and equivalent variations.