Method and Apparatus for Isolating and Detecting Biological and Other Particles

Abstract

An apparatus and method for isolating bacterial particles in a sample using a container with material in temporary fluid blocking position to lower orifice in the container, a separation medium having an electrical conductivity lower than and physical density greater than that of the sample above the material that supports a sample concentrate after passing through the separation medium when exposed to centrifugal force, a heating element for liquefying the material to permit flow into a chamber past an electrode array that attracts and holds subject particles. The system allows rapid detection and isolation of particles from samples from animal, human, environmental sites, a bio-industrial reactor or a food or beverage production facility requiring relatively small volumes, short incubation times resulting in structurally intact particles for further analysis. Testing may be completed in a single unit that requires decreased technician manipulation, fewer steps and a decrease in cross-contamination.

Claims

1-20. (canceled)

21. An apparatus for collection and separation of microorganisms, cells, and biological particles in a sandwiched configuration comprising: a. a non-conductive base layer; b. a second non-conductive layer disposed on the base layer where the second layer has a void; c. a primary array of electrodes positioned between the layers where the void provides a fluid flow path in which biological particles are capable of flowing past the arrays; and d. a ground plane electrically connected to the primary array.

22. The apparatus as claimed in claim 21 wherein the electrodes are interdigitated.

23. The apparatus as claimed in claim 21 wherein the electrodes are disposed in fluid communication with the void.

24. The apparatus as claimed in claim 21 wherein the electrodes are coated with biological or chemical materials to promote or discourage adhesion or biological particles, microorganisms, or cells.

25. The apparatus as claimed in claim 21 wherein the electrodes are configured in a parallel, sawtooth, staggered, semi-circular, or castellated formation.

26. The apparatus as claimed in claim 22 wherein the interdigitated electrodes have spacing, width, height, and angle relative to the direction of fluid flow that are varied along either the length, width, or height of the void.

27. The apparatus as claimed in claim 21 further comprising a secondary array of electrodes facing the primary array of electrodes.

28. The apparatus as claimed in claim 21 wherein the electrodes have a cross sectional shapes from a list comprising a rectangle, circle, triangle, polygon, or semi-circle.

29. The apparatus as claimed in claim 27 wherein the secondary array is at least 2 dimensional and is positioned on the top layer in a configuration chosen from the following orientation: opposite or perpendicular to the primary array.

30. The apparatus as claimed in claim 21 wherein the ground plane is positioned below the base layer.

31. The apparatus as claimed in claim 21 wherein the ground plane encloses the base layer, second layer and primary array.

32. An apparatus for collection and separation of microorganisms, cells, and biological particles in a sandwiched configuration comprising: a. a non-conductive base layer; b. a primary array of electrodes positioned on the layer to provide a fluid flow path in which biological particles are capable of flowing past the arrays; and c. a ground plane electrically connected to the primary array positioned below the array.

33. The apparatus as claimed in claim 32 further comprising a second non-conductive layer disposed on the base layer where the second layer has a void that exposes a portion of the primary array.

34. The apparatus as claimed in claim 33 wherein the void is laser etched, die-cut, machined, molded, or embossed into the second layer.

35. The apparatus as claimed in claim 32 wherein the ground plane is positioned below the base layer.

36. The apparatus as claimed in claim 33 wherein the ground plane is positioned on top of the second layer.

37. The apparatus as claimed in claim 32 wherein the ground plane encloses the base layer and the primary array.

38. An apparatus for collection and separation of microorganisms, cells, and biological particles in a sandwiched configuration comprising: a. a base layer; b. a primary array of interdigitated electrodes positioned on the layer; c. a second layer disposed on the base layer creating a sandwich for the primary array where the second layer has a void to provide a fluid flow path; and d. a secondary array of electrodes positioned on the second layer facing the primary array of electrodes where the void provides a fluid flow path in which biological particles are capable of flowing past the arrays.

39. The apparatus as claimed in claim 38 further comprising a ground plane in electrical contact with one of the arrays and positioned beneath the base layer.

40. The apparatus as claimed in claim 38 wherein the fluid flows past the arrays under pressure.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The drawings constitute a part of this specification and include exemplary embodiments to the invention, which may be embodied in various forms. It is to be understood that in some instances various aspects of the invention may be shown exaggerated or enlarged to facilitate an understanding of the invention.

[0033] FIG. 1 shows a graph of conductivity versus time according to a preferred embodiment of the present invention.

[0034] FIG. 2 shows an exploded schematic of a testing apparatus according to a preferred embodiment of the present invention.

[0035] FIGS. 3A, 3B and 3C show a side schematic view of a testing apparatus in successive steps of operation according to a preferred embodiment of the present invention.

[0036] FIGS. 4A and 4B show a side view schematic of a microchamber for fluid flow of a testing apparatus according to a preferred embodiment of the present invention. FIG. 4C shows an alternative side view schematic of a microchamber for fluid flow of a testing apparatus according to a preferred embodiment of the present invention.

[0037] FIG. 5 shows a perspective view of a microchamber assembly and electrode arrangement for a testing apparatus according to a preferred embodiment of the present invention.

[0038] FIGS. 6A, 6B, and 6C show cross sectional views of a microchamber assembly with alternative electrode arrangements according to preferred embodiments of the present invention.

[0039] FIG. 7 shows a block diagram of the method and system for detecting and analyzing bacterial samples according to a preferred embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0040] Detailed descriptions of the preferred embodiment are provided herein. It is to be understood, however, that the present invention may be embodied in various forms. Therefore, specific details disclosed herein are not to be interpreted as limiting, but rather as a basis for later filed claims and as a representative basis for teaching one skilled in the art to employ the present invention in virtually any appropriately detailed system, structure or manner.

[0041] In Vitro Diagnostic Tool for Bacterial and Fungal Bloodstream Infections

[0042] A preferred embodiment of the present invention device is essentially a “lab-in-a-tube” about the size of a standard 1, 5, 15, or 50 mL conical tubes, depending on the needs of the application. In the application of processing blood specimens from adult patients (typically 10 mL), a technician would mount a standard blood collection tube on top of the present invention. Optionally, the blood collection tube may be one that contains lysis buffer that selectively disrupts host blood cells but leaves microorganisms relatively unaffected, in terms of their physical properties and physiological status. Blood is automatically drawn into the present invention chamber by negative pressure in the lab-in-a-tube device, or by creating a pressure gradient using a system external to the tube device, or by manually loading the sample into the device. Then following a brief centrifugation step, the technician places the present invention device into a custom benchtop instrument and is free to perform other lab tasks. Meanwhile, the centrifugation step is performed by the benchtop instrument, which may require only a few minutes or several hours, depending on the application. In the case of a detecting bacteria or fungi in a 10-mL blood specimen, the lab tube might be centrifuged between 10-45 minutes, depending on the rotor diameter, the angular velocity of the centrifuge motor, and the dimensions of the lab-in-tube device. Overall the present invention device and instrument i) use centrifugation, fluid-flow, and electrokinetic phenomena to non-destructively isolate microorganisms within a microfluidic chamber, and ii) use built-in microelectrode sensors and/or optically-clear walls to achieve rapid detection and accurate quantification of a broad range of microbes using a non-disruptive transduction method such as electrochemical impedance spectroscopy, electroenzymatic biosensing, calorimetry, and/or microscopy. An alarm is triggered (sound and electronic notification) once microorganisms are detected or the analysis is complete. The process will require less than 10 minutes of hands-on time and the tube is disposable. Once detected, microbes can be retrieved from the device for further analysis as further described with FIGS. 2 through 6 below.

[0043] Non-Destructive Isolation of a Broad Range of Biological Particles

[0044] Detecting low level biological particles quickly demands that target biological particles are first concentrated. Furthermore, many analytical techniques may require that biological particles of interest are first separated from the primary specimen to minimize interference from inhibiting substances. In the case of isolating microorganisms from blood, the present invention device and system achieves this by 1) selective lysis of all blood cells or undesired biological particles, 2) density-gradient centrifugation to separate and pre-concentrate biological particles of interest from the sample, and 3) further concentration of target biological particles using electrokinetic phenomena, such as dielectrophoresis (DEP) in a microfluidic chip integrated at the base of the centrifuge tube. DEP is the motion of electrically polarized particles due to a non-uniform electrical field. Any intact biological cell will experience DEP, which allows the present invention to isolate all bacteria and fungi without a priori knowledge of the target. In this way, the device provides unbiased isolation of biological particles. Other effects, such as electrohydrodynamic phenomena, may also influence the trajectory of biological particles. Diffusion may be an important force governing the motion of biological particles in the present invention, depending on the so-called Stoke's diameter of the particle and the temperature of the system. In the case of other sample types, such as urine, the lysis of host cells may be unnecessary at a first step in the process. In some embodiments, the density gradient separation medium may prevent the transport of undesired particles or substances into the lower layers of the chamber, effectively removing them from further analyses. The characteristics of the electrical signals that give rise to electrokinetic phenomena within the present invention may too enable the separation of target biological particles from undesired particles and substances.

[0045] Electrical Polarization in Biological Matter

[0046] Electrical polarization and conduction processes occur in biological systems and are important in understanding the electrokinetic behavior of cells. Various mobile charge carriers that can contribute to electrical conduction and polarization in biological systems, including ions, protons and other small molecules. It can be appreciated that the structural, composition, and conformation attributes may all contribute to the electrical properties of biological particles. Biological cells are assemblies of interacting biological systems encapsulated within a membrane structure that separates the internal processes of life from the environment. Some cells also have internal organelles and compartments with distinct compositions and structural features. The size, shape and composition of biological particles and cells vary immensely, giving rise to tremendous diversity in the physical and electrical properties of cells. In general, cells can be approximated as spheres having a diameter on the order of one to ten micrometers. The cell cytosol of bacteria, archaea, algae, viruses, parasites, and fungi are all encapsulated by barriers that are more complex than the plasma membrane of mammalian cells and that are composed of different types of interwoven biopolymers. Algae typically possess external cell walls made of glycoproteins and polysaccharides which are similar in composition to the plant cell wall like cellulose, alginate and agarose. Diatoms have a cell wall composed of biogenic silica. Fungi are differentiated from plant cells by the fact that their cell walls contain the unbranched chains of the glucosamine polymer chitin that are cross-linked by various sugar molecules. The other major constituent of the fungal wall are glycosylated proteins with branching mannose sugars. Bacteria have a cell wall that contains a structure called the peptidoglycan, also known as the polysaccharide-peptidoglycan wall. Gram-negatives possess a second outer phospholipid bilayer which envelops a thin peptidoglycan mesh. There are also bacteria that lack cell walls. Some prokaryotes produce an extracellular, hydro-gelatinous polymer called the glycocalyx. Most archaea possess an outer proteinaceous layer called the surface-layer (S-layer). Most S-layers are 5-25 nm thick. It follows that these regions of fixed charge contribute to the electrical response of these interwoven structural biopolymers. The diversity produced by nature confers unique and intrinsic properties upon the structural variants of prokaryotes, potentiating opportunities for isolating and sorting different subpopulations contained within microbiomes by exploiting polarization and conductance phenomena.

[0047] Each phase or layer contributes to the overall electrical properties of the biological particle or cell. These different material properties cause charges to move with differing ease and/or at different rates following the application of an excitation field. For example, the phospholipid bilayer represents a barrier to mobile charges and can lead to the accumulations of charge at a membrane interface in the presence of an applied electric field. This interfacial polarization arises from dissimilar permittivity and/or conductivity values on either side of the interface. For spherical particles like biological cells, the accumulation of charge at the interface between the dissimilar conductive, dielectric materials results in an induced dipole. The time constant for such a particle may be written,

[00001]$\tau =\frac{{\epsilon }_p+2{\epsilon }_m}{{\sigma }_p+2{\sigma }_m}$

[0048] where subscripts “p” and “m” refer to the particle and suspending medium, respectively. In aqueous media a shell of counter ions attracted from the bulk medium by coulombic forces to the ionized groups also influence other ions that are proximal to the surface of the particle. The balance electrostatic attraction and thermal dispersion produces an effective macroscopic charge density ρ.sub.DL in the diffuse double layer of counterions. For a binary electrolyte with a volume density of ions n.sub.0 having equal valency z, a convenient measure of the extent of the electric double layer may be written as

[00002]${\lambda }_{DL}={\left(\frac{{\epsilon }_m{k}_{B}T}{2z^2{q}^2{n}_0}\right)}^{\frac{1}{2}},$

[0049] corresponding to the distance between the Stern plane of the particle and where the electrostatic potential has decreased by 63.2 percent. Physically, the electric double layer is an ionic atmosphere distributed around the particle in its suspending electrolyte. In the case of bacteria, an expression for the dielectric increment Δε due to the counter ion concentration within the porous region of the cell wall N.sub.+ may be written,

[00003]$\Delta \epsilon =\frac{q^{2}a{\delta }_0}{{\epsilon }_0{k}_{B}T}$

[0050] Here, δ.sub.0 is the counter ion concentration per unit area. In the case where the counter ion density within the porous region is far greater than the bulk suspending medium ion concentration (N.sub.+>>n.sub.0), the surface counter ion density may be calculated from,

[00004]${\delta }_0\approx \frac{1}{2q}{\left(N_{+}{\epsilon }_w{k}_{B}T\right)}^{\frac{1}{2}}\ln \left(\frac{.Math.N_{+}.Math.k_{B}T}{{\epsilon }_w}\right)$

[0051] where ε.sub.w is the electrical permittivity of the porous region. An expression for the effective particle conductivity that incorporates this tangential current pathway about a spherical surface may be written,

[00005]${\sigma }_p={\sigma }_{\perp }+\frac{2{\sigma }_{\tan }}{a}$

[0052] where again a is the particle diameter. The first term is the charge flux perpendicular to the particle surface and σ.sub.tan is the charge that moves along the surface in response to the tangential field. The characteristic response time of surface charge polarization may be written,

[00006]${\tau }_{\mathrm{surf}}=\frac{a^2}{2D_{\mathrm{eff}}}\frac{1}{M}.$

[0053] to account for the contributions of fluxes of both mobile charge and perturbed bound charge, the term M can be found from

M=1+z.sub.+z.sub.−(z.sub.++z.sub.−)n.sub.ba/n.sub.0λ.sub.DL

[0054] where z.sub.+ and z.sub.− are the electrovalencies of positive and negative charge carriers and n.sub.b the total density of counter ions at equilibrium. The time response of charge carriers in the surface layers of the Stern model incorporates the particle size and an effective diffusivity of surface counter charges,

D.sub.eff=u.sub.sk.sub.BT/q

u.sub.s=u.sub.0e.sup.−U.sup.C.sup./k.sup.B.sup.T,

[0055] where u.sub.0 represents the mobility of ions in the bulk. The fixed charges on the particle surface are assumed to create a set of periodically spaced, overlapping coulombic wells separated by potential energy barriers. The mean charging time τ.sub.surf is,

[00007]${\tau }_{\mathrm{surf}}=\frac{q{a}^2}{2u_0{k}_{B}T}{e}^{U_C/k_{B}T}$${\overline{\tau }}_{\mathrm{surf}}={{\tau }_{\mathrm{surf}}\left[e^{-\Delta U_c/k_{B}T}\right]}^{-\frac{1}{2}}$

[0056] These distinct and time-dependent polarization phenomena are important when the system is energized by a time-varying electrical field,

E=E.sub.0 cos ωt+jE.sub.0 sin ωt=E.sub.0e.sup.jωt

[0057] where E.sub.0 is the maximum amplitude of the harmonic wave and j=√{square root over (−1)}, and the electric excitation field switches polarity slowly compared to the response rate of ions, then the polarization will be able to respond completely to the field changes and reach a maximum value in accordance with the field strength and direction. Conversely, if the applied field varies while charges are still in motion, there will be insufficient time for the polarization to reach its equilibrium state. As a result, the induced conduction and polarization will not be in phase with the applied field. Nomadic and molecular dipolar polarizations occur over short time scale whereas interfacial polarization and surface effects in particles require longer charging times. When time-varying electric fields are applied it is useful to define the complex conductivity {tilde over (σ)}=σ+jωε and the complex permittivity as

[00008]$\tilde{\epsilon }=\epsilon -\frac{\sigma }{j\omega }.$

The complex permittivity and conductivity parameters reflect the fact that the bioelectric response to electrical excitation in cells is not instantaneous.

[0058] To account for the different layers of biological particles and the different polarization phenomena, it is common to mathematically model cells using the so-called multi-shell model. Through this approach, a multilayered particle can be transformed into an equivalent dielectric model. For example, a two-layer particle with corresponding complex permittivities {tilde over (ε)}.sub.1 and {tilde over (ε)}.sub.2 will have an effective complex permittivity given by,

[00009]${\tilde{\epsilon }}_p={\tilde{\epsilon }}_2\left[\frac{{\gamma }^3+2\left({\tilde{\epsilon }}_1-{\tilde{\epsilon }}_2\right)}{{\gamma }^3-\left({\tilde{\epsilon }}_1+2{\tilde{\epsilon }}_2\right)}\right]$

[0059] where γ=a.sub.2/a.sub.1. Bacteria have been modeled in terms of two and three concentric shell models, respectively, including a shunt admittance element representing counter ions in the porous wall moving parallel to the surface. The effective complex permittivity of a Gram-positive bacterium may be written

{tilde over (ε)}.sub.G+={tilde over (ε)}.sub.core+ε.sub.wall

[0060] This equation is equivalent to a model for shunt admittance with the inclusion of a conducting shell enclosing an electrically homogeneous inner core which, for example, may represent the effective electrical properties of the cell membrane and cytosol. This term is derived by solving the electrostatic potential outside a sphere but with a variant form of the boundary condition for charge continuity,

jω[{tilde over (ε)}.sub.m{tilde over (E)}.sub.m−{tilde over (ε)}.sub.p{tilde over (E)}.sub.p]+∇.sub.s{tilde over (G)}.sub.s

[0061] where ∇.sub.s is the surface del operator, or divergence of the complex surface conductance {tilde over (G)}.sub.s induced by the tangential component of the applied electric field. For Gram-negative bacteria an additional shell is included to account for the outer plasma membrane according to the admittance model,

[00010]${\tilde{\epsilon }}_{G-}={\tilde{\epsilon }}_{om}\frac{{\gamma }_2^3+2\left(\frac{{\chi }_1-{\tilde{\epsilon }}_{om}}{{\chi }_1+2{\tilde{\epsilon }}_{om}}\right)}{{\gamma }_2^3-\left(\frac{{\chi }_1-{\tilde{\epsilon }}_{om}}{{\chi }_1+2{\tilde{\epsilon }}_{om}}\right)}$

[0062] with χ.sub.1={tilde over (ε)}.sub.core+{tilde over (ε)}.sub.wall.

[0063] For concentric heterogeneous, layered systems like biological cells this iterative process proves to be a very useful model for predicting the electrical response of different cell types under different conditions. The effective permittivity and conductivity of the particle can be easily calculated by taking the real and imaginary parts of the effective complex permittivity,

[00011]${\epsilon }_p=\frac{\mathrm{Re}\left[{\tilde{\epsilon }}_p\right]}{{\epsilon }_0}$$\mathrm{and}{\sigma }_p=\mathrm{Im}\left[-\omega .Math.{\tilde{\epsilon }}_p\right]$

[0064] When time-varying external fields are used, the finite charging time associated with the various conduction and polarization phenomena is important in determining how significantly each process contributes to the overall electrical responses of biological cells. It has been demonstrated that cells exhibit four different regimes of dielectric dispersions the frequency range between 10 Hz-100 MHz associated with the various compartments and barriers defining the cellular structure. For example, in the case of bacteria, at low frequencies the electrical properties are dominated by the relatively high conductivity of the cell wall with the electrical properties of the cell interior being entirely screened from low frequency imposed electric fields by the insulating inner plasma membrane. The electrical responses of many bacteria exhibit two response regimes as the frequency of the applied field is increased from 1 kHz to around 100 kHz. The first response is interpreted as the ionic conduction at the cell wall, the other by dipolar and ionic losses occurring at the cell membrane. In the 100 kHz to 1 MHz frequency range, the well-known Maxwell-Wagner polarization effect dominates the polarization response. As the frequency is increased still further above 1 MHz the electric field penetrates the cell and internal structures and properties therefore govern the bioelectrical responses of cells in this higher frequency range.

[0065] For a spherical biological cell, the polarization field can be treated as being equivalent to the field emanating from a single induced effective dipole moment. For a conducting dielectric particle, it is well known that the complex dipole moment of the particle is therefore given by,

{tilde over (p)}.sub.eff=4πε.sub.m{tilde over (K)}.sub.CMa.sup.3E.sub.0

[0066] where a is the particle radius, ε.sub.m is the relative permittivity (not complex) and,

[00012]${\tilde{K}}_{CM}=\left(\frac{{\tilde{\epsilon }}_p-{\tilde{\epsilon }}_m}{{\tilde{\epsilon }}_p+2{\tilde{\epsilon }}_m}\right)$

[0067] The complex number {tilde over (K)}.sub.CM, which is typically referred to as the Clausius-Mosotti factor, contains the dynamic electrical response of the particle. If the particle is far less polarizable than the suspending fluid {tilde over (ε)}.sub.p<<{tilde over (ε)}.sub.m and the induced dipole is antiparallel to the applied field and the particle is driven away from the high field region, a process referred to as negative dielectrophoresis. When the particle is more electrically polarizable than the medium {tilde over (ε)}.sub.p>>{tilde over (ε)}.sub.m the particle moves towards the high field region, a phenomenon called positive dielectrophoresis (DEP). The dielectrophoretic force arises from the interaction of a non-uniform electric field and the asymmetrically distributed space charges at the particle boundary. If the electric field is non-uniform in space the particle experiences a translational force. The magnitude and direction depend on the electrical properties of the particle and surrounding medium, and on the magnitude and frequency of the harmonic electric field. The time-averaged dielectrophoretic force for a dipole in the time-varying electric field E(ω) is given by,

F=({tilde over (p)}.sub.eff.Math.∇)E

[0068] For a spherical particle this expression becomes,

[00013]$.Math.F_{\mathrm{DEP}}.Math.=\frac{1}{2}\mathrm{Re}\left[\left(p.Math.\nabla \right)E^{*}\right]=\frac{1}{4}V{\epsilon }_m\mathrm{Re}\left[{\tilde{K}}_{CM}\right]\nabla {.Math.E.Math.}^2-\frac{1}{2}V{\epsilon }_m\mathrm{Im}\left[{\tilde{K}}_{CM}\right]\left(\nabla \times \left(\mathrm{Re}\left[E\right]\times \mathrm{Im}\left[E\right]\right)\right)$

[0069] where V is the particle volume, E is the RMS electric field strength, and ∇ is the vector gradient operator; the symbol * indicates complex conjugation; Re and Im are the real and imaginary operators; the polarizability of the cell is {tilde over (K)}.sub.CM and |E|.sup.2=E.Math.E*. It follows that the DEP force is determined by the divergent field created by the geometric configuration of the electrodes. The term ∇|E|.sup.2 underscores the importance for using microelectrodes for DEP manipulation of biological cells because the DEP forces scales with the square of the system characteristic length. Various electrode geometries can be used to maximize the spatial inhomogeneity of the electric field. A common pattern is the castellated geometry and its variants. A staggered variant of the castellated geometry may also be employed to further enhance the field inhomogeneity.

[0070] The simplest microelectrode array used in DEP microsystems is the co-planar, parallel bar pattern. This creates homogeneous fields in the plane of the microelectrodes but strongly inhomogeneous fringing fields above the plane of the electrode array. The highest electric field region is located at the electrode edges. In the castellated pattern a local low field occurs in the pockets between adjacent electrode tips. Similarly, a local field minimum may be found at the midpoint between parallel bars. In all cases, the magnitude of the electric field strength and inhomogeneity decreases rapidly with distance above the energized surface.

[0071] To effectively harness DEP forces there is a need to physically confine particles close to the surface. It is for this reason that virtually all DEP devices incorporate microfluidics with channel feature sizes on the order of 1-100 μm. Microelectrodes embedded on one or more wall of microfluidic channels enable the use of DEP for working with small particles like cells. Internal fluid flow is governed by the well-known Navier-Stoke's equation describing the mechanical force balance that may be written,

[00014]$\stackrel{\stackrel{\mathrm{Intertia}}{︷}}{{\rho }_m\left(\frac{\partial v}{\partial t}+\underset{\underset{\begin{array}{c}\mathrm{Convective}\\ \mathrm{acceleration}\end{array}}{︸}}{\left(v.Math.\nabla \right)v}\right)}=\stackrel{\stackrel{\mathrm{Divergence}\mathrm{of}\mathrm{stress}}{︷}}{\underset{\underset{\begin{array}{c}\mathrm{Pressure}\\ \mathrm{gradient}\end{array}}{︸}}{-\nabla p}+\underset{\underset{\mathrm{Viscosity}}{︸}}{\nabla .Math.\left(n\nabla v\right)}}+\stackrel{\stackrel{\begin{array}{c}\mathrm{Body}\\ \mathrm{forces}\end{array}}{︷}}{f}$

[0072] In the absence of body forces (f=0) the fully developed

[00015]$\left(\frac{\partial v}{\partial t}=0\right)$

flow velocity between semi-infinite, parallel walls separated a distance H may be written,

[00016]$v\left(h\right)=\frac{H^2}{2\eta }\left(\frac{\partial p}{\partial x}\right)\left[{\left(\frac{h}{H}\right)}^2-\frac{h}{H}\right]$

[0073] where

[00017]$\frac{\partial p}{\partial x}$

is the rate of change of pressure along the length of the channel and h is the height above the bottom wall. It is useful to represent this fluid flow velocity field in terms of the average velocity v,

[00018]$v\left(h\right)=6\overline{v}\frac{h}{H}\left(1-\frac{h}{H}\right)$

[0074] This mathematical form allows one to calculate the velocity distribution for parameters that may be changed during device fabrication or controlled during operations, such as the width of the channel W and the volumetric flow rate Q. This is possible using the relationship

[00019]$\overline{v}=\frac{Q}{WH}.$

Considering the governing equations of fluid flow within microfluidic devices and DEP reveals important limitations to microfluidic-based DEP systems. Most DEP microsystems are designed to handle very small volumes, usually on the order of 0.01 mL. Slow processing speed is a major roadblock to DEP applications in clinical and industrial applications. Typical DEP devices are operated at ˜30 L/min, which equates to a processing time of about 12.5 days for a single 10 mL specimen of blood. The reason for this slow processing rate is readily apparent when the estimated time required for collecting bacteria by DEP is compared to the average time microbes spend within the microfluidic chamber. The residence time of the cell suspension is determined by the geometry of the duct and the specimen processing rate. The DEP velocity may be written,

[00020]$u_{DEP}=\frac{F_{DEP}}{\gamma }=\frac{a^2{\epsilon }_m}{6\eta }\mathrm{Re}\left[{\stackrel{\sim }{K}}_{CM}\left(\omega \right)\right]\nabla {.Math.E.Math.}^2,$

[0075] It follows that the trapping efficiency of electrokinetic microdevices can be improved by increasing the magnitude of the electric field. However, Joule heating can raise the temperature of the system and adversely affect the physiology of particles such as cells and proteins. The membrane structures of most biological cells become permeabilized around ˜45° C. within ˜30 seconds. From an engineering design perspective, it is useful to estimate the temperature changes within DEP microdevices to understand the operational limitations. The electrical, mechanical, and energy equations are coupled and related to the spatial-temporal temperature distribution within the suspending fluid T.sub.f by,

[00021]$\underset{\begin{array}{c}\mathrm{Energy}\\ \mathrm{storage}\end{array}}{\underbrace{{\rho }_f{c}_{f,p}\frac{\partial T_f}{\partial t}}}=\stackrel{\begin{array}{c}\mathrm{Internal}\\ \mathrm{advection}\end{array}}{\overbrace{{\rho }_f{c}_{f,p}u\frac{\partial T_f}{\partial x}}}+\stackrel{\begin{array}{c}\mathrm{Heat}\\ \mathrm{diffusion}\end{array}}{\overbrace{k_f\frac{{\partial }^2{T}_f}{\partial z^2}}}+\underset{\begin{array}{c}\mathrm{Electric}\mathrm{heat}\\ \mathrm{generation}\end{array}}{\underbrace{{\sigma }_f{E}^2}}$

[0076] c.sub.f,p is the specific heat of the fluid at constant pressure, u is the fluid velocity, ρ.sub.f is the density of the suspending medium, μ.sub.f is the dynamic viscosity of the fluid, {dot over (γ)} is the fluid shear rate, E is the time-averaged magnitude of the applied electric field, and σ.sub.f is the electrical conductivity of the suspending medium. Endothermic and exothermic reactions are assumed to not significantly impact the energy balance within the device during cell separations. Under most conditions, advection is negligible compared to conduction. The fluid layer thermal time constant is computed by τ.sub.f=(H/Ak.sub.f)(ρ.sub.fVc.sub.f,p)˜10 milliseconds, and the fluid is therefore assumed to be at steady-state. Heat generated in the fluid layer is dissipated into the chamber walls, which are assumed to also have no phase change, nor any internal heat generation or consumption reactions. The energy balance for this material may be written,

[00022]$\underset{\begin{array}{c}\mathrm{Energy}\\ \mathrm{storage}\end{array}}{\underbrace{{\rho }_b{c}_{b,p}\frac{\partial T_b}{\partial t}}}=\stackrel{\begin{array}{c}\mathrm{Heat}\\ \mathrm{diffusion}\end{array}}{\overbrace{\frac{\partial }{\partial z}.Math.\left(k_b\frac{\partial T_b}{\partial z}\right)}}$

[0077] where T.sub.b is the temperature at the board (or substrate layer). Convective and radiative heat transport are excluded from the analysis because the microfluidic device may be placed in thermal contact with a heat sink made of a material such as a metal. The thermal time constant of the chamber walls depends on the choice of material and thickness of the board layer. Glass, molded silicones like polydimethylsiloxane (PDMS), and thermoplastics like polymethylmethacrylate (PMMA) are common selections for microfluidic devices. In the worst case, where thick (˜1 mm) and poor thermally conducting materials (k.sub.b≈0.15 W.Math.m.sup.−1.Math.K.sup.−1) are chosen, the thermal time constant of the board layer would be τ.sub.b˜100 sec, shorter than the duration of most separation processes, which typically require about 30 minutes for milliliter scale specimens. Therefore, this material is assumed to be at thermal equilibrium. Proceeding with these assumptions and approximations, the energy balance for the fluid and board layers comprising a typical continuous-mode micro-electrokinetic separation device is,

[00023]$\mathrm{Fluid}:\frac{{\partial }^2{T}_f}{\partial z^2}=-\frac{{\sigma }_f{E}^2}{k_f}$$\mathrm{Board}\mathrm{layer}:0=\frac{{\partial }^2{T}_b}{\partial z^2}$

[0078] It is further assumed that heat generation is uniform within the microfluidic layer. Embedded electrodes do not contribute significantly to heat transfer or thermal energy storage because they are very thin relative to the other layers and conduct heat much faster. Within typical operating temperature ranges, the thermal conductivity values of the fluid and board layers do not vary significantly. However, the electrical conductivity of the electrolyte may significantly vary during operation. To find the maximum fluid temperature at steady-state such that cells are not damaged during operation, it is common to approximate the electrical conductivity temperature-dependence as being linear, with σ.sub.f(T)=σ.sub.f,0[1+α(T.sub.f−T.sub.f,0)], where σ.sub.f,0 is the electrical conductivity at some reference temperature T.sub.f,0 and α≈0.02 K.sup.−1. The mathematical analysis is further simplified by taking advantage of symmetry about the mid-plane within the fluid layer. Applying the Neumann boundary condition at the mid-plane (z=0) and letting T.sub.f(z=H/2)=T.sub.i, the solution to the fluid energy balance may approximated as,

[00024]$T_f=\frac{{\sigma }_0{V}^2\left(1+\alpha \Delta T\right)}{2k_f}\left(\frac{1}{4}-\frac{z^2}{H^2}\right)+T_i\mathrm{for}0\le z\le \frac{H}{2}$

[0079] with ΔT=(T.sub.f,max−T.sub.f,0). The maximum fluid temperature can be calculated from,

[00025]$T_f\left(z=0\right)=T_{f,\max }=\frac{{\sigma }_0{V}^2\left(1+\alpha \Delta T\right)}{8k_f}+T_i$

[0080] Solving for the maximum fluid temperature therefore reduces to working out the equivalent thermal circuit for half the composite assembly, with constant heat flux per unit area q.sub.ET′={dot over (Q)}.sub.ETH and thermal resistance per unit area R′. The temperature at the electrolyte/board interface is T.sub.i=T.sub.∞+q.sub.ET′R′, and the expression for estimating the maximum temperature change, after arranging terms, may be written as,

[00026]$\Delta T\approx \frac{\beta }{8k_{f}H-\alpha \beta }$

[0081] Where β=σ.sub.0V.sup.2(H+8k.sub.fR′). This final expression is useful because it directly relates the maximum fluid temperature change to engineering design and operational parameters, including the geometry of the microfluidic assembly, the properties of the construction materials, the magnitude of the applied voltage, and the temperature-dependent electrical conductivity of the fluid.

[0082] How the Present Invention Detects a Broad Range of Microorganisms

[0083] The present invention uses electrochemical impedance spectroscopy (EIS) to measure changes in the electrical properties of the fluid caused by microbial metabolism and proliferation. In general, microorganisms consume non-ionic compounds (e.g. carbohydrates) to power metabolism. They produce ionic species (hydrogen ions, ammonium, lactate, etc.) that are expelled into the surrounding medium and increase its conductivity. This rate of change in the conductivity will increase as the microorganisms proliferate and become greater in numbers. By placing microelectrodes in culture medium, the gradual rise in conductivity can be measure over time [27], [28].

[0084] We adapted this method for rapid detection by embedding two microelectrodes in a 0.5 μL (0.05 mm height×2.5 mm length×5 mm width) microfluidic incubator built by inventors. It was possible to detect the presence of 100 colony forming units (CFU) after 5 hours of incubation.

[0085] The present invention uses this proven detection method and uses automated optical microscopy to provide quantitative and clinically actionable results. In this way, any metabolically active microbes will be detected and enumerated in the chamber.

[0086] Turning to FIG. 1, there is shown a graph 100 of the change of electrical conductivity in siemens per meter or S/m on the y-axis 110 versus time in hours after starting the experiment on the x-axis 120. Three different 130, 140 and 150 are shown for Colony Forming Units of bacteria of 1000, 100 and 0 CFU respectively. Testing results were averaged at each hour with ranges 160 shown for example on each curve. As demonstrated on graph 100, microfluidic EIS detection of E. coli with increasing conductivity over time for higher concentrations was experienced, where curves for conductivity values were averaged over five separate runs. Note: Time-to-detection depends on the initial concentration.

[0087] Turning now to FIG. 2, there is shown an exploded schematic of a testing device 200 according to a preferred embodiment of the invention. A blood sample is taken in a conventional manner from a patient typically in a standard collection tube of 1, 5, 10, 15 or 50 mL and mounted on needle 220 for transference of the sample into container 245. Alternatively, the blood collection tube may contain lysis buffer that disrupts host blood cells but leaves subject particles unaffected for processing according to the present invention. Lid 225 has an opening through which needle 220 is pre-affixed to permit fluid flow from tube 205 into container 245 through stopper 215 that is punctured upon placement of tube 205 onto needle 220. Flow may be accomplished by negative pressure, such as from a vacuum in container 245 or external pressure through port 230 which is operatively attached to lid 225 and may be affixed to container 245 with mating threads 235 and 240.

[0088] As more fully described below, container 245 has low electrical conductivity medium 252 across the entire section of container 245 and extended downward and positioned above meltable medium 258 in fluid blocking position at the bottom of container 245 blocking orifice 255. As a separate module or assembled in one piece along with container 245, microchamber 270 is positioned in fluid communication with orifice 255 for fluid flow of subject particles as further described below. Microchamber 270 is composed of channel 290 that allows fluid flow past electrode array 280 in a void created by the sandwiching of heat sink 275 to thermally-conductive adhesive layer 296 to electrode array 282 and adhesive strip 285 having laser etched opening 340 as shown in FIG. 5. Thermally conductive adhesive layer 296 may also be a thermally conductive paste.

[0089] Turning to FIGS. 3A, 3B, and 3C there is shown a side schematic view of the cartridge and electrode chamber according to a preferred embodiment of the present invention operation (side view of the device) in stages of operation.

[0090] FIG. 3A shows a fully mounted container 245 with sample tube 205 and subject particles 210 within sample fluid 218. Upon fixing tube 205 onto needle 220 the sample fluid flows through needle 220 into container 245. In a preferred embodiment, needle 220 is angled so that its output is gently streamed against an inside wall of container 245 so that the sample and its subject particles initially fall onto low ionic separation medium 252 for later centrifugation. FIG. 3B shows subject particles 246 in low ionic medium 252 that is above impermeable but meltable medium 258. At this stage container 245 is spun in a conventional centrifuge that causes the subject particles 246 to be collected on top of meltable material 258 in a more concentrated form. FIG. 3C shows container 245 and microchamber 270 mounted to a base that permits action by heat 264 upon meltable material 258.

[0091] Stage 1 Sample Transfer: In a preferred embodiment, a technician simply mounts a standard Wampole blood collection tube, here shown as tube 205, on the top of the device. Blood is automatically drawn through a needle 220 toward the wall of the device tube so that the blood is layered over an isotonic iodixanol-mannitol density medium (˜1070 kg/m3) 252 without mixing. The density of the iodixanol-mannitol mixture can be changed according to the needs of the application.

[0092] Stage 2 Separate Microbes from Sample: Efficient electrokinetic manipulation of biological particles requires that the ion levels in the suspending fluid be reduced below levels typically observed in biological fluids or environmental samples. This is because the electrical conductivity of the suspending medium can i) reduce the magnitude of the applied electric field because of an impedance mismatch between the suspending fluid and the signal generating electronics; ii) so-called Joule heating can adversely impact the physiology of many biological cells; iii) DEP trapping is less effective at higher conductivities because many biological particles will experience negative dielectrophoresis and be repelled away from microelectrodes and into faster moving fluid flow streams. The physical densities of bacteria and yeast are greater than the density medium, so centrifuging the device causes them to accelerate into the density medium that contains minimal ion levels and onto a 10% (w/v) gelatin plug. Other gel materials may be included in the present invention, depending on the application. The operational conditions required to separate target biological particles into the density medium layer depend upon the sedimentation coefficient of the particle

[00027]$s=\frac{v_t}{a},$

where v.sub.t is the terminal sedimentation velocity and a is the applied acceleration, typically due to the centrifugal acceleration where a=ω.sup.2R with R being the distance of the object from the axis of rotation and ω the angular velocity of the object. The biological particle reaches terminal velocity when the applied centrifugal force balances the opposing force of viscous drag exerted on the particle by the suspending medium, so the terminal velocity for a spherical particle may be written mathematically

[00028]$v_t=\frac{m_b{\omega }^{2}R}{6\mathrm{\pi \eta }r}$

with m.sub.b being the buoyant mass of the particle, η the viscosity of the suspending medium, and r the radius of the spherical particle. These equations can be used to guide the design and operation of the centrifuge system. Because the density medium and the sample do not mix, ions remain in the sample and only slowly diffuse into the density medium. In this way, target biological particles that are denser than both the fluid in the original specimen and denser than the density gradient medium will be separated from the original specimen and travel toward the bottom of the density gradient medium 258 when inertial forces are applied. In the present invention, there may be a gelatin plug which provides a semi-soft landing pad for target particles.

[0093] Alternatively, the plug may be made of any of a variety of impermeable meltable materials and may be a coating on the bottom of the device tube that covers the lower orifice. In some embodiments, it may be feasible to have a permanent coating and use a valve in the lower orifice before flushing the concentrated microbes into the microchamber. The meltable material may be turned into a flowing material by any of a number of approaches that apply heat 264 to the meltable material 258 so it, and the subject particles may flow out of orifice 255. Meltable material 258 may be formed of a gelatin plug. Heating may be accomplished by placement of the entire apparatus in an incubator, not shown, or by use of heating elements external to container 245 for application of sufficient heat.

[0094] In other embodiments, the meltable material and/or the separation medium may also contain reagents necessary for biochemical reactions to occur with target particles, including but not limited to: biochemical dyes and stains: aptamers; functionalized micro- and nanoparticles; bioengineered tags like antibodies conjugated to enzymes; fluorophores or phosphors; oligonucleotide probes; or stimuli-responsive vesicles. The meltable material and separation medium may also contain chemical agents that alter the dielectric properties of biological particles, including but not limited to antimicrobial agents; ionophores; enzymes; detergents; chaotropes; kosmotropes; stimuli-responsive vesicles that containing any of the above agents (including agents needed for biochemical reactions) or ionic species; and other substances that are known to alter the electrical properties of biological materials.

[0095] Stage 3 Concentrate Microbes: Once microbes are positioned on gelatin plug 258, the testing apparatus is mounted in a control instrument. A needle punctures a septum blocking orifice 255 to draw fluid into microchamber 270, electrical contact is made with the device, and a tube connects the device to the control instruments pneumatics. The device is brought up to at least 32 Degrees Celsius to make a gelatin plug liquid. The target temperature may vary, depending on the composition of the plug. The instrument pressurizes the present invention to pump the biological particle concentrate into the detection microchamber, where they are temporarily immobilized by the phenomenon of ac DEP or dielectrophoresis.

[0096] In other embodiments the phenomenon of dc-biased DEP may be employed to temporarily immobilize biological particles of interest. In the case where the biological particle concentrate contains mixtures of different types of particles, the frequency of the applied ac field, the polarity of the dc-bias, and the magnitude of the dc-bias and the ac signal all operational variables that can be tuned to preferentially immobilize a fraction of biological particles while the remaining fraction of particles continue to flow through the channel and/or are repelled from the electrode elements. It is further possible to change the characteristics of the electrical signals in subsequent and individually addressable electrode arrays to immobilize biological particles that are contained in the fraction that is repelled from the first array. It is understood that any number of individually addressable arrays may be configured for the purposes of differentially manipulating with particles present in the concentrate.

[0097] Stage Four Detect Biological Particles: FIG. 4A shows fluid flow from orifice 255 down tube 267 in the direction of the electrode array 272 which is activated. The fluid flow is shut off and the microbes are incubated in microchamber 270, which may have a volume of just a few picoliters to as much as a few hundred microliters, depending on the application. In the case of microorganisms, the detection time depends on the i) the volume of the detection chamber, ii) the metabolic and growth rates of the organisms, iii) the initial concentration of target biological particles, and iv) the performance of the signal processing electronics connected to the device. The addition of reagents in either the separation medium of the meltable plug may accelerate the detection of target particles, too. Reagents may expedite the conductivity change by enabling a biochemical reaction between the target biological particles and said reagents to either change the electrical conductivity of the surrounding medium in the presence of and in proportion to the target particles, or by changing the optical properties of the suspending medium or the reagents or of the target particles. Signal acquisition and processing may be conducted in a multitude of ways using analog and/or digital circuitry.

[0098] The lab-in-tube device that is positioned within an incubator is connected to signal acquisition and processing circuitry using at least four electrical connectors that are adequately grounded from interfering electromagnetic signals, perhaps using a combination of coaxial cables, ground planes built into multilayered circuit boards, and/or the device is within a Faraday Cage. The EIS method involves the application of an excitation electrical signal of known magnitude E.sub.0 (usually only a few hundred millivolts to achieve so-called pseudo-linearity) and frequency ω, which can be expressed as a function of time: E(t)=E.sub.0 sin (ωt) and measuring the current flowing through the system, said to be the response current. The response current will have a different magnitude I(t) and phase ϕ than those of the applied signal, and can also be written as a function of time: I(t)=I.sub.0 sin(ωt+ϕ). The impedance is calculated using Ohm's law:

[00029]$Z=\frac{E\left(t\right)}{I\left(t\right)}=\frac{E_0\sin \left(\omega t\right)}{I_0\sin \left(\omega t+\varphi \right)}.$

The electronic hardware is programmed to execute this measurement over a range of frequency values at desired points in time to obtain the impedance spectra. In the present invention, the frequency range that can be monitored is expanded to lower frequencies (<300 kHz) using a so-called tetrapolar or four-terminal impedance setup, and in some embodiments coating the electrodes within the detection chamber with so-called blackened platinum or polypyrrole to increase the surface area of the electrodes. In the present invention, many electrode configurations and geometries may be used depending on the application of the technology. For instance, the electrode size, thickness, and materials may be varied for different uses of the invention. The number of electrodes included in the four-terminal configuration might change in that one or more of the four terminals may branch within the invention to create a microelectrode array having electrodes elements that are castellated, staggered (symmetrically or asymmetrically), sawtooth, parallel bar, or three-dimensional shapes like cylinders or pyramids or bumps, as is needed for the application of the device. Once the EIS signal reaches a threshold, determined by comparing the time evolving impedance spectra to a baseline value and/or to impedance spectra contained in computer memory, the alarm is set off notifying operators that biological cells are present within the device. The technician may then mount the present device on a custom microscope (not shown) to enumerate microbes using video processing software or by manually counting particles in the field of view.

[0099] Stage Five Recover Microbes: Isolated microbes can be eluted from the device for downstream analysis. FIG. 4B show microchamber 270 with outlet tube flushing out subject particles past septum 266 in a first and second state of fluid flow past a single array of electrodes.

[0100] FIG. 4C shows an alternative microchamber 270 having electrode arrays 280 and 282 on both sides of the collection segment of flow path 289.

[0101] FIG. 5 shows a perspective view of sandwiched fluid chamber 300 and two arrays of electrodes 280 and 335. In a preferred embodiment of the invention, two separate collection arrays are positioned in the fluid path for selective collection of subject particles as each set of electrodes are successively activated. Sandwiched fluid chamber 300 is formed by at least two layers. Adhesive strip 285 has void 340 laser etched to create an opening that faces a portion of the flow path in microchamber 270. Adhesive strip 285 may be dual sided adhesive and is stuck to the microchamber face 293 shown in FIG. 2 and electrode mount 345 that has embedded within it first array of electrodes 280, second array of electrodes 282, contact bus 315 and common ground 310. This arrangement permits subject particles to flow past first array of electrodes 280 and be held for a pre-determined period of time, and then past second array of electrodes 282 and held for a pre-determined period of time. Particles may be held against an array or held in the microchamber near or at an array.

[0102] The dimensions of the void may be of any shape, depth, and volume so long as sufficient area is presented for the fluid flow to accomplish the desired purpose. In a preferred embodiment, adhesive strip 285 may have a width 355 in the range of 25 microns to approximately 500 microns and electrode mount 345 may have a width 350 between 25 microns and 3 mm, depending on the construction materials. In another preferred embodiment the smaller range of values may be used to facilitate heat dissipation. In a preferred embodiment, etched width 325 of opening 340 is in the range of 1-30 mm, and length of upper triangle 320 of opening 340 is approximately 1.5 times the value of etched width 325. This facilitates a more even flow through the channel. Other dimensions may be used and the shape of triangle 320 may also be altered to a variety of shapes including a half circle, half oval, rectangle, square, or arched mitre configuration. Similarly, the opening 340 formed by the etched portion of wafer 305 may have other edge features so long as the sandwiched wafer when affixed to the chamber face forms a flow channel for the concentrate.

[0103] FIGS. 6A, 6B and 6C shows three different arrangements of electrodes in the preferred embodiments of the invention. FIG. 6A shows a cross sectional split view of sandwiched fluid chamber 300 along A-A of FIG. 5 with electrode array 400 having electrodes 425 on one side and ground 420 on the other side. FIG. 6B shows a cross sectional split view of sandwiched fluid chamber 300 with electrode array 405 having electrodes 425 on both sides. FIG. 6C shows a cross sectional split view of sandwiched fluid chamber 300 with electrode array 415 having electrodes 425 on one side and only. These configurations show how they might be used for electrokinetic isolation and electrochemical impedance spectroscopic analysis of the detection chamber within the lab-in-tube device.

[0104] Depicted diagrammatically in FIG. 7 is a block diagram and flow chart of a preferred embodiment of the present invention for apparatus setup that includes other external instrumentation helpful in conducting tests and analysis. System 600 has a first step 605 using primary sample container that is in turn flowed within lab in a tube device 610 using the apparatus heretofore described with centrifuge 615. This lab in a tube device 610 interfaces with the system 600 via custom device holder 635, and is controlled by controller 620 having an operator control panel 625, a temperature control system 630 both operated by computer 645 and visually shown on monitor 640. Computer 645 may acquire information from lab in a tube device 610 once it is in electrical and fluidic contact with customer device holder 635 via signal acquisition and processing 637 and frequency response analyzer 639. Further, computer 645 controls signal generator electronics 641 to produce the electrical waveforms that are transmitted to signal amplifier electronics 643 to produce a new signal of sufficient electrical power as is required by various applications of the invention. Signals transmitted from the signal amplifier electronics 643 to customer device holder 635 give rise to electrokinetic phenomena within lab in a tube device 610. In this way, the testing apparatus of the present invention may be fully automated and controlled using computer systems for sequencing the various steps of operation already described herein.

[0105] Examples of results using a system according to a preferred embodiment of the invention are discussed below.

Example 1

[0106] Firstly, the use of a density gradient medium as a barrier to the ions in the original sample was tested and shown to work successfully. Briefly, a substance called OptiPrep is created, which is about 60% iodixanol, was diluted with 5% D-mannitol to create a solution having a physical density of 1080 kg/m3. About 500 microliters of this density gradient medium (OptiPrep) is placed in a 2 mL centrifuge tube. Then, about 500 microliters of whole blood, previously treated with saponin (final concentration of saponin was ˜5 mg/mL), is spiked with bacteria (Enteroccocus faecalis) and is carefully layered on top of the density gradient medium using a pipettor equipped with a wide-bore pipette tip, so as to not disturb the heavier bottom fluid layer and cause mixing of the sample and that layer. The tube containing the sample and the separation density medium is then centrifuged at 8,000 RPM for 15 minutes using a standard bench top centrifuge having a rotor diameter of about 20 cm. If the target biological particles were smaller than bacteria, such as virus particles, the angular velocity and centrifugation interval would need to be increased to perhaps 20,000 RPM and several hours, depending on the sample volume and dimensions of the tube.

[0107] Following this procedure, the tube is removed from the centrifuge, the top fluid layer is carefully extracted manually using a pipettor equipped with a wide-bore pipette tip. A small volume of the density gradient nearest to the interface is additionally withdrawn. The remaining fluid is vortexed to resuspend any particulates, including bacteria and debris, that had become immobilized at the bottom of the vessel during the centrifugation process. The remaining fluid is then withdrawn, and a small sample is placed on a microscope slide or in a microfluidic well for further examination or processing. In either case, the slide or the device is mounted on a microscope stage so that the bacteria and other debris could be observed. The microfluidic well is typically made by gluing a rubber O-ring to a glass substrate having photopatterned, co-planar, and interdigitated metal microelectrode array on the top surface. Within the microfluidic well it is possible to manipulate bacteria recovered from the centrifuge tube using positive and negative dielectrophoresis by applying a time-varying electrical signal to the microelectrode array. Interestingly, some debris are observed in the suspension as well. However, those quasi-transparent objects do not seem to respond to the applied electric field. The signal is produced using a bench-top electronic signal generator, and the signal is monitored using an oscilloscope.

Example 2

[0108] In another set of experiments, one mL of blood is placed above one mL of the density gradient medium. In this separate experiment, a larger volume allows for the electrical conductivity of the density gradient medium to be measured off-line using a flow-through probe and a bench-top conductivity meter. In three separate experiments, the highest conductivity value measured was 62.5 mS/m, which is about 24× lower than the electrical conductivity of whole blood (˜1.5 S/m). The density gradient not only presents the migration of some biological particles to the bottom of the device while still allowing sufficiently dense particles to be transported through the density medium, it also provides a barrier to the diffusion of ions from the sample into lower fluid layers.

Example 3

[0109] In yet another set of experiments, a small hole was bored in the bottom of the centrifuge tube and a rubber septum having dimensions of 11 mm in diameter and about 3 mm in thickness was positioned at the bottom of the centrifuge tube. This allowed the operator to access the bottom layer without disturbing the sample and the fluid interface between the density gradient medium and the sample when extracting a sub-sample from the density gradient medium. To test this device, E. coli was grown overnight in lysogeny broth, and 100 microliters of the culture was harvested and layered on top of a density gradient medium. The gradient medium was made by mixing Optiprep and a mannitol solution to adjust the density to approximately 1080 kg/m.sup.3. After centrifuging the tube at 10,000 RPM for 10 minutes, the sample liquid appeared clarified, and a pellet formed along the surface of the septum. A needle was used to puncture the septum the plunger was withdrawn and depressed repeatedly to resuspend cells near the surface of the septum. On the final withdrawal, the needle was removed, and the liquid expelled onto a glass slide for examination. Bacteria could be detected using a microscope.

Example 4

[0110] In other experiments, the invention was tested utilizing the combined use of dielectrophoresis to concentrate and position Escherichia coli particles before monitoring their metabolic activity and growth using electrochemical impedance spectroscopy. In these experiments, a microfluidic chamber was created by laser cutting the pattern of the desired fluid path into a double-sided pressure-activated adhesive sheet 0.002″ thick. A plastic top was then bonded to a glass substrate that had co-planar, photopatterned, and interdigitated microelectrodes on its surface using said laser cut adhesive. The width of this microfluidic channel was about 3 mm. Several different test suspensions containing varying concentrations of bacteria suspended in a low electrical conductivity medium (about 15 mS/m NaCl and 5% mannitol) were injected into the device and the target particles were first concentrated on a small section of microelectrodes about 5 mm in length and 5 mm in width (spanning the entire width of the channel) using positive dielectrophoresis. Then, the bacteria were flushed from that segment by injecting the culture medium Lysogeny Broth into the chamber. The high conductivity of the Broth (>1 S/m) caused bacteria to be released from the microelectrodes by negative dielectrophoresis, and they were transported by the fluid flow into a separate chamber of about one microliter in volume and having embedded in the bottom wall two microelectrode bars that spanned the width of the microfluidic chamber. These two microelectrodes were connected to a bench-top frequency-response analyzer, which measured the in-phase and out-of-phase electrical impedance of the device every 20 minutes for a 12-hour period. The electrical impedance of the fluid containing the bacteria could be deduced using an equivalent circuit model, which was developed by measuring the parasitic impedance of individual components (the coaxial cables that connected the microfluidic device to the frequency-response analyzer, the metal contact pads on the device itself) and finally the impedance of the microchamber having the two microelectrodes embedded within. The magnitude and phase of the electrical impedance of this system could be displayed on a personal computer monitor that was connected to and controlling the frequency-response analyzer. Using this experimental setup, it was possible to show that the rate of change of the electrical impedance of the suspending fluid within the device was dependent upon the initial concentration of E. coli. Furthermore, it was confirmed that the signal did not appreciably change over time when no bacteria were present in the initial sample.

Example 5

[0111] A 1/32″ hole was drilled at the base of a 50 mL conical centrifuge tube and covered using several sort pieces of tape. A 10% w/v gelatin solution was brought to 40 degrees Celsius in a hot water bath and 3 mL was transferred to the conical tube using a pipette and allowed to cool to room temperature, until the liquid solidified. Then 10 mL of a mixture of Optiprep and 5% mannitol having a physical density of 1080 kg/m3 was layered over the gelatin plug. A 1 mL sample of urine was spiked with 10 microliters of E. coli (optical density at 600 nm of 0.5), and that mixture was gently pipetted with the tip of the pipette placed at the side wall of the conical tube to not disturb the interface between the sample and the density gradient medium. The conical tube was capped and the tube was centrifuged in a swinging-bucket rotor (Rotor radius=10 cm) at 5000 RPM for 20 minutes. The tube was then transferred to a stationary incubator set to 37 degrees Celsius for about 10 minutes, until the gelatin melted. Then the tape was removed, and the cap was slightly unscrewed to create a pressure differential from the proximal and distal ends of the tube. Fluid droplets collected into three different 2 mL vessels and flow was stopped by screwing the cap back on the tube tightly. Bacteria could be detected in second vessel, and a small volume (˜10 microliters) was transferred to a DEP device to confirm that bacteria could be collected by positive DEP at 5 MHz and 3 V.sub.pp. The DEP device was made by securing a rubber o-ring above a microelectrode array microfabricated on a glass microscope slide with quick-set epoxy. The array had a castellated electrode geometry with a feature size of about 50 micrometers. The DEP signal was generated using a Red Pitaya device which was under computer control and amplified using an RF signal generator procured from Microcircuits.

Example 6

[0112] We demonstrated that DEP could discriminate between bacteria and blood cells and concentrate microbes into a microchamber where they could be detected. To achieve this, co-planar and interdigitated microelectrodes embedded in the chamber that is 50 μm high are energized with an AC electric signal having a frequency of 5 MHz and amplitude of at least 10 V.sub.pp. The electrodes are 50 μm wide and spaced and are about 2 μm thick. In accordance with a preferred embodiment of the present invention a custom centrifuge tube so that clinically-relevant volumes (>10 mL) of biological specimens like blood required for diagnosing BSI can be analyzed.

[0113] In one preferred embodiment, the present invention includes a larger machine that holds many individual devices, within which the specimen is processed. One novel feature of the device is that it includes a density medium that is layered atop a gel plug that separates a larger compartment from smaller compartments located at the bottom of the device. The gel plug is included so that biological particles that migrate to the bottom of the device during centrifugation come to rest on the semi-rigid gel, thereby minimizing the risk that target biological particles become immobilized on the surface of the device and cannot be detected or recovered.

[0114] In another preferred embodiment, the present device allows for inertial forces generated by centrifugation, drag generated by fluid flow, and electrokinetic forces to be applied along one or more spatial dimensions with time-varying magnitude defined by the operator. As is illustrated in the drawings of the invention, the device enables operators to fractionate and separate mixtures of biological particles by controlling the angular velocity of the device, the pressure exerted on the suspending fluid, the frequency of the electric signal, the magnitude of the applied electrical potential, the geometric configuration of the physical chambers defining the fluid path, and the geometric configuration of electrode structures embedded within the chambers. Importantly, the device is designed such that fractionation may occur in one or more of the compartments within the device, depending on the intended use of the device.

[0115] In another preferred embodiment, the device that is the present invention manifests biological particles of interest that can be extracted from the device by way of a small access port that is sealed so that there is no need for an active valve mechanism to be built into the device; flow out of the port can be controlled upstream in the device by applying pressure and the effluent can be collected in another vessel.

[0116] Yet another preferred embodiment, the device harbors biological particles of interest that can, in a single device, be separated from the original sample matrix using centrifugation, concentrated using the combined action of fluid flow and electrokinetic trapping, detected using at least one non-disruptive transduction mechanism, such as optical microscopy, calorimetry, electroenzymatic reactions, bio- or chemiluminescence or electroluminescence, or electrochemical impedance spectroscopy, for example, and then finally the biological particles can be recovered from the device.

[0117] While the invention has been described in connection with a preferred embodiment, it is not intended to limit the scope of the invention to the particular form set forth, but on the contrary, it is intended to cover such alternatives, modifications, and equivalents as may be included within the spirit and scope of the invention as defined by the later issued claims.