WO2005033654A2 - Micro flow cytometer with multiangular waveguide detectors having spectral capabilities - Google Patents

Abstract

A micro flow cytometer having the ability to distinguish light emanating from a particle in the interrogation zone (6) of the cytometer contains a plurality of optical waveguides (10, 11, 12, 13) for receiving emanating light, the optical waveguides (10, 11, 12, 13) positioned at different angles from the axis of the interrogation zone and in the same plane. Photodetectors measuring emanated light provide an electrical signal which is processable to reveal an angular dependence of wavelength with respect to the interrogation channel (6).

Description

MICRO FLOW CYTOMETER WITH MULTIANGULAR WAVEGUIDE DETECTORS HAVING SPECTRAL CAPABILITIES

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims the benefit of U. S. provisional application Serial No. 60/498,453 filed August 28, 2003., herein incorporated by reference.

BACKGROUND OF THE INVENTION

1. Field of the Invention

The present invention pertains to micro flow cytometers having spectral capabilities.

2. Background Art

Micro flow cytometers are known, for example as described in PCT/US02/22903, incorporated herein by reference. Such flow cytometers are distinguished from conventional cytometers by their small size and inexpensive construction. The small channel sizes of such devices encourage laminar fluid flow, and their small volume requires correspondingly small amounts of analyte and focusing fluids, when the latter are used. In PCT/US02/22903, a unique detection system employing PIN diodes as detectors has been described. By modulating the incident light and amplifying the detector signal synchronously with the frequency of modulation, a detector output having lower noise than conventional photomultiplier tubes is achieved. PCT/US02/22903 also describes the use of multiple wavelengths of light with suitably filtered detectors, so as to detect fluorescence from more than one species. In PCT published application WO 85/05680 is disclosed a conventional flow cytometer of macroscopic size in which the interrogation channel is interrupted by an orifice plate transverse to the axis of the interrogation channel. The orifice plate contains a plurality of optical waveguides which are perpendicular to the channel. However, WO 85/05680 does not disclose any variation of spectral response at the different waveguides, which rather are for the purpose of gathering a plurality of differing kinds of information based on interaction of particles in the channel with an exciting light source which flows through one of the waveguides.

It would be desirable to provide new modes of detection in micro flow cytometers, particularly enhanced ability to gather wavelength-dependent spectral data. Gathering such data may provide means to further increase detection characteristics such as signal to noise ratio, and to provide information on particles being detected which is not currently available.

SUMMARY OF THE INVENTION It has now been surprisingly discovered that interrogation waveguides having different angular relationships with the interrogation channel of a micro flow cytometer can provide increased signal to noise ratio in particle detection, as well as allowing for investigating spectral response characteristics. In such devices, a plurality of detectors for a single color, color band, etc. may be used, but at different angles to the flow channel. The detector response exhibits an angular dependence on wavelength.

BRIEF DESCRIPTION OF THE DRAWINGS

FIGURE 1 illustrates one embodiment of a micro flow cytometer having microgrooves adapted to receive optical waveguides. FIGURE 2 illustrates a test setup used to investigate detector response in a micro flow cytometer employing red and blue light sources. FIGURE 3 illustrates a block diagram for detection and signal amplification of fluorescence emitted by a particle in an interrogation zone of a MFC.

FIGURES 4a and 4b represent typical traces from a MFC using a PMT detector and PIN detector with lock-in, respectively.

FIGURE 5 illustrates one relationship between signal to noise ration (S/N) at various detector angles for various colored microspheres.

FIGURE 6 illustrates the dependency of S/N ratio on wavelength at various detector angles. FIGURE 7 illustrates actual and theoretical S/N variation with different sample column widths.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

The micro flow cytometers ("MFC") of the subject invention contain at least one optical waveguide which supplies incident light to an interrogation zone of a MFC, and a plurality of collecting optical waveguides to gather light from the interrogation zone, the collecting optical waveguides positioned at different included angles with respect to the axis of the interrogation zone, and in the plane of the axis of the interrogation zone. The incident and collecting optical waveguides are preferably optical fibers. The flow cytometer itself may be made of any convenient material, including but not limited to micromachined glass, quartz, silicon, stainless steel, ceramic, etc., injection molded plastics, and the like. However, the material of construction is preferably a cast silicone elastomer, preferably a transparent elastomer. Such MFCs can be readily cast from a micromachined master. They may also be fabricated as multiple layer structures which are subsequently bonded together. One embodiment of such a device is shown in Figure 1, where a three layer construction is used. In leftmost layer 1, the inner face contains gas channels 2a which communicate with leftmost portion of focusing zone 3a. A middle layer 4 contains the bulk of the focusing zone 3 and a sample fluid passage 5 in communication therewith. The focusing zone communicates with the interrogation zone or "channel" 6 which communicates with the outlet reservoir 7 and outlet passage 8. The rightmost layer 9 is, in this case, a mirror image of leftmost layer 1, and contains gas passages 2b and, rightmost focusing zone portion 3b.

The terminations of the various channels external to the device are advantageously adapted to be single plug-in connectors as are common for supply of fluids, or in the case of the connectors for the light supply and detector passageways, simple snap fittings which allow suitable coupling to external light emitters (lasers, laser diodes, etc.) and detectors (photomultiplier tubes, PIN diodes, etc). In preferred devices, these latter companies may be incorporated into the body of the cell, such that only electrical connections (electrical contacts, jacks, coaxial cable), etc. , may be required. The central layer 4 is also configured to contain microgrooves for receiving minimally one incident waveguide and two collector waveguides. The device shown contains microgrooves for two incident waveguides 10, 11 and two collector waveguides, 12 and 13.

As indicated previously, the micro flow cytometer ('MFC") devices may be constructed of various materials. For example, the entire device may be constructed of a hydrophobic polymer such as polydimethylsiloxane (PDMS), or may be constructed of alternative materials, including conventional thermosetting polymers, thermoplastics, glass, ceramic, metal, and the like, provided that the portions of the internal passageways of the device preceding and including the interrogation zone where both liquid and optionally gas are simultaneously present are transparent to the wavelengths of interest in an area between the light source and the detector within the interrogation zone, so that light scattered by particles passing through the interrogation zone can be analyzed appropriately as in conventional flow cytometers. The walls of the internal passages are preferably of hydrophobic material, or are coated or otherwise treated to render them hydrophobic. A convenient method of constructing the focusing chamber, interrogation zone, and further channels and reservoirs is the use of soft lithography as disclosed by Y.N. Xia et al. , SCIENCE, V. 273 p.347 ff (1996). The use of this technique also allows a simple lens or even a grating to be fabricated in the device. Y.N. Xia et al., "SOFT LITHOGRAPHY" Annu. Rev. Matter SCI V. 28, pp 153 - 184, (1998). The lens and grating may form part of the wall of the interrogation zone, for example, for light focusing and/or collimation or spectral dispersion.

The devices may be unfocused, i.e. configured such that the only fluid flowing through the interrogation zone of the device is the fluid being analyzed. This type of configuration is particularly useful at low flow rates in channels of very small dimension. However, in general, it is preferable that the devices be focused, either by means of a surrounding sheath of liquid, or of gas. As is well known, focused devices maintain the fluid being analyzed in a narrower and more well defined flow through the interrogation channel. The optical waveguides used are preferably optical fibers of glass, and may be cast in place in the device or positioned by any other suitable means. Most preferably, MFC devices are constructed of layers of materials, with one or more layers, preferably a layer also containing the interrogation channel or a portion thereof, containing grooves or channels sized to receive and emplace the optical waveguides. Providing of such channels simplifies the construction process and renders the devices uniform from device to device. Since the MFCs are preferably designed to be low cost and disposable, uniformity is particularly desirable. Traditional large scale flow cytometers are both expensive as well as specifically designed for multiple use. The concepts expressed herein can be used with conventional flow cytometers as well.

The geometry of the focusing chamber, interrogation zone, etc., is such that the sample stream is focused, either two dimensionally or multiply dimensionally, without substantial breakup of the sample stream when used as a flow cytometer. Numerous geometries are possible, and the various channel chambers and zones may take numerous cross-sections. A suitable geometry is provided in Figure 7 of PCT/US02/22903, which should not be construed as limiting. Dimensions are in mm. Focusing of the sample fluid is illustrated in Figure 8 of PCT/US02/22903, wherein sample fluid stream 41 enters focusing chamber 42 together with air sheaths or "focusing streams" 43a and 43b. As the focusing chamber narrows, the sample stream is altered in shape and constricted 44 to a narrow and coherent stream 45 within the interrogation zone. The geometry may be determined by standard commercially available fluid dynamics software.

In a typical device, the diameter of the liquid sample channel may be from 1 - 1000 μm, preferably 10-300 μm, and most preferably about 20 μm at the downstream end where it communicates with the focusing chamber. The gas channels at this point are preferably somewhat larger in dimension, for example 200-600 μm, preferably 250-500 μm, and more preferably 300-400 μm, and converge with the fluid sample channel at an acute included angle, preferably from 15-45°, more preferably 20-40°, and most preferably ca. 30°. The walls of the focusing chamber are defined by the outer walls of the gas inlet channels at its "upper" end, for example 1000 μm to 2000 μm, preferably about 1500 μm, and at the "lower" end, by the diameter of the interrogation zone into which the focusing chamber transitions, i.e. from 100 μm to 500 μm, preferably 200 μm to 400 μm, and most preferably about 300 μm. The transition of the upper end of the focusing chamber to the lower end, the beginning of the interrogation zone, is roughly V- shaped in plan, and the side walls may be slightly concave, as described in the figures, the actual geometry being straight, circular, elliptical, hyperbolic, parabolic, or other shape which suitably minimizes breakup of the sample stream.

The liquid sample and focusing gas flow rates (when used) are selected so as to form a focused but coherent liquid sample stream within the interrogation zone of the desired diameter. It should be noted that when gas focusing is used, the fluid sample is not necessarily completely enveloped by gas, although complete envelopment is preferred. The diameter of the sample stream in the interrogation zone will be a function of the gas flow rate, the liquid sample flow rate, and the vacuum (if any) applied at the outlet. Liquid focusing in the same manner may also be used. The lower end of the interrogation zone, for example the outlet reservoir, is preferably maintained at less than atmospheric pressure, for example from 10 mm Hg to 700 mm Hg. A pressure of 45 mm Hg has been found quite suitable. Alternatively, the inlet reservoirs could be maintained at positive pressure without requiring application of vacuum at the outlet. The physical geometry of the device may also affect sample stream diameter.

For example, when gas is used for focusing, the gas flow rate may range from 40 mL/hr to 15,000 mL/hr, preferably ca. 800 mL/hr. At these gas flow rates, with a 45 mm Hg vacuum, and in a channel with a 100 μm height, the diameter of the focused liquid sample stream in the interrogation zone may be varied from as little as 15 μm at a liquid sample flow rate of 6 mL/hr to about 100 μm at a flow rate of 20 mL/hr. Calculations based on fluid dynamics indicate that in a device with a 300 μm wide interrogation zone, a gas flow rate of about 700-900 mL, and a vacuum at the outlet of 45 mm Hg, the maximum focused liquid stream flow rate is approximately 50 mL/hr, at which point the liquid will contact the walls of the interrogation chamber. These same calculations indicate a minimum sample flow rate of 6.6 mL/hr without losing stream coherency. A stable 6 mL/hr stream has been observed. However, lowering the flow rate to 5 mL/hr caused the stream to be discontinuous. The geometry and dimensions of the channel can be tailored to achieve stability at different flow rates. For further discussion of the necessary calculations, reference may be made to Huh, et al. , "Use of Air-Liquid Two-Phase Flow in Hydrophobic Microfluidic Channels for Disposable Flow Cytometers," BlOMEDiCAL MlCRODEViCES, 4 (2), 141-49 (2002), based on the work of the present inventors.

The light source and light impingement system may comprise a conventional laser, numerous types of which are commercially available. With such lasers, optical systems are generally necessary to properly direct the laser beam through the interrogation zone. In this type of light supply, conventional auxiliary components can be used. "Auxiliary" components include conventional optics including but not limited to lenses, lens systems, filters, gratings, polarizers, quarter wave plates, beam splitters, etc. These auxiliary components comprise the light impingement system when standard light sources are used. The light source may emit a range of light, may be monochromatic, or may be multichromatic. In one embodiment, a plurality of light sources of different spectral characteristics are employed. These, for example, may emit light in the UV, visible, or IR portions of the spectrum. The wavelengths are often chosen such that they maximize fluorescing modes of particles to be detected. For purposes of detection, it is also desirable that the light sources be of a wavelength other than that to be detected.

Similarly, conventional light collection systems and light detection systems can be used. These include, generally, a photomultiplier tube or equivalent device as part of the detector system, generally also associated with auxiliary components such as lenses or lens systems, beam splitters, filters, gratings, etc., which also comprise the light collection system. The output of the photomultiplier and of photodiodes, when used, are generally connected to standard data acquisition devices and processed using analog or digital circuitry to maximize the signal to noise ratio. These techniques are easily implemented by those skilled in the art based on the concepts presented herein. For example, a noise baseline may be established, and only peaks projecting above this baseline may be reported.

When lenses or lens systems are employed, these may be external to the flow cell, or may be cast or inserted into the flow cell to minimize the number of external auxiliary components. The impingement and collection pathways terminate internally such that light impinges upon the collection zone, and is detectable following passage through the collection zone by the detector. The pathways for conventional impingement and collection pathways generally terminate "proximate" to the interrogation zone. For transparent devices, for example of acrylic resin or polyorganosiloxane, the path termini may be at an outside wall of the device, or may be located in close proximity to the interrogation zone, with an optically transparent passage, for example a hole, transparent rod or window, etc., whose innermost closed end is close to the interrogation zone. The actual construction of the impingement and collection zones is not critical so long as their function is achieved. It should be noted that the material of construction, if optically transparent at the desired wavelengths, may be structured in the form of a lens element or grating. While light transmission from the interrogation zone will be conveyed by optical waveguides, and while it is preferably that the impinging light also be so conveyed, the lenses, gratings, etc. heretofore described may precede or follow the optical waveguides per se. Most preferably, the optical waveguides communicate directly with the interrogation channel or an optically transparent wall thereof, as described more fully below.

To reduce size and cost, it is particularly preferred that the conventional light source be supplanted by one or more solid state "laser" diodes, which are preferably located within or on the flow cell or a housing thereof. Use of such devices is made practical by a light impingement system which employs optical waveguides, preferably fiber optics. In this embodiment, the light from a laser diode light source is transmitted to the proximity of the interrogation zone by fiber optics. The optical fiber(s) terminate as close as possible to the interrogation zone to minimize scattering effects, for example terminating from 10 to 100 μm, preferably 30-80 μm from the interrogation zone. Due to the nature of the signal carrying capabilities of the optical fibers, lens systems which are commonly used in conventional systems may be eliminated, thus providing significant savings in size and cost, to the degree that in a preferred embodiment, the laser diode light source(s) may be incorporated within the body of the flow cell itself or intimately associated therewith, for example in a compact case, such that the entire device is portable, handholdable, and/or disposable. In addition to focusing fluid and sample fluid connections, only standard electrical connections need be provided, which can constitute metal contacts, jacks, etc.

The light impingement system can also be used with a remotely located light source, either of the conventional laser type or a laser diode type. In such cases, the optical fibers embedded within the body of the flow cell are preferably terminated at the exterior of the cell body, for example by conventional fiber optics coupling devices.

In like manner, the light collection system preferably comprises fiber optics as well, and preferably does so for purposes of miniaturization and disposability. As in the case of the light impingement systems, the optical fibers may be connected to an external detector, for example a photomultiplier tube and auxiliary components desired to maximize the signal to noise ratio. However, the fiber optics may also be connected to a detector system "on board," which allows for an integrated transportable and even handheld device. The waveguide configuration is also self-locating and fixtured, which minimizes the potential for interfering vibrations during measurements with a miniaturized device.

The detector system of the present invention is used in conjunction with an optical waveguide or fiber optic collection system, and preferably comprises a semiconductor detector, preferably a PIN diode. Such diodes are commercially available, are capable of being biased by relatively low voltages, and are of relatively small size and cost. The PIN diode preferably includes one or more filters grown directly on the PIN structure or attached thereto. Such a filter is useful for filtering out extraneous light. For example, the filter may allow light of a fluorescence wavelength to pass while filtering out the wavelength used to excite the fluorescent species, for example a dye marker. Alternatively, an optional filter may be contained in the collection pathway as a component of a collimating lens system.

The PDMS observation channel illustrated in Figure 1 was designed to have a cross-sectional are of 100x300μm² and three fluid entry compartments. Design parameters for the channel such as lengths, angles, and volumes were determined by developing a microfluidic model within the computational fluid dynamics (CFD) software Fluent 5.5 (Fluent Inc., NH). A three-dimensional computational model was utilized to simulate the hydrodynamic focusing behavior of the sample at the interrogation point of the laser under different scenarios. Specifically, models which could predict the sample focusing effectiveness of sheath fluid injected at different impingement angles (20° - 60°) were constructed to find an optimal configuration for the impingement angle of the sheath fluid channels on the sample channel. Similarly, various ratios of sample flow rate to sheath fluid flow rate (1:1-1: 10) were simulated to observe the effectiveness of sample fluid focusing. The final dimensions and operating parameters for the channel were selected such that the sample fluid is well focused with a constant velocity in the center of the channel well before the interrogation point. The design was also set such that the sample fluid in entire thickness direction (lOOμm) would be uniformly illuminated as necessary to minimize device performance variation.

The optical sub-system of the micro flow cytometry system must serve 1) to deliver light, preferably laser light, to a well-defined observation zone, and 2) to collect fluorescent light from the observation zone and deliver it to an optical filter for separation from scattered laser light prior to electronic detection. In contrast to conventional flow cytometers, the present invention uses integrated optical fibers. The fiberoptic system has a more compact volume, yet is still capable of transmitting excitation and fluorescence light with high efficiency. In addition, by embossing groves that serve as fiber receptacles directly into the PDMS observation cell as shown in Figure 1 , the optical fibers can be fit tightly in the system without separate alignment steps necessary during assembly. Figure 1 also illustrates that multiple light impingement and collection angles are also easily possible (45° and 90° shown), which is not the case in conventional flow cytometers. The detection system can therefore be configured to collect information from any or all angles simultaneously as desired in a given application.

In addition to the consideration of microfluidics and materials of construction, much effort has been directed to develop micro-optical components that could potentially be utilized within a flow cytometer microsystem. In a flow cytometer, the optical system usually consists of a laser, photodetectors, and optical components such as lenses, mirrors, and filters. In conventional flow cytometers photomultiple-tubes (PMTs), and occasionally avalanche photodiodes (APDs), are used for fluorescence detection in the wavelength range of 400 to lOOOnm. These detectors are utilized in flow cytometry and other applications due to their high sensitivity, high internal gain (> 10⁶), and fast response (10^"7-10^"9s). However, PMTs are bulky, fragile, consume large amounts of power, and are generally sensitive to environmental conditions. APDs are also somewhat bulky and require high-power, resulting from their need for active thermal management to maintain high S/N. Although both PMT and APD photodetectors yield a large S/N due to their internal gain, optical systems incorporating these devices into flow cytometers are costly and difficult to miniaturize. For this reason, PIN-based photodetectors which are 1) low-cost, 2) low power, and 3) tightly integratable with other optical components are preferably used.

Since the fluorescent signals generated from the detection of biological cells only generate a few nA of current in PIN photodiodes, a lock-in amplification technique is applied to increase the S/N of the micro flow cytometer. While common in other technical applications, the lock-in amplification approach, which is also described in PCT/US02/22903, has not been previously discussed in the literature for flow cytometry. The lock-in technique is effective because it acts as a narrow band-pass filter which removes much of the unwanted noise while allowing the signal which is to be measured to pass through. To achieve this, the laser excitation is modulated at a reference frequency which induces fluorescence emission from the interrogated sample at the same frequency. The modulated signals detected at the photodiode are then monitored with high sensitivity using the lock-in circuit. For reliable performance, the modulation frequency is set to be two orders of magnitude higher than the bandwidth required for detecting the fluorescence emission from moving sample particles. The lock-in circuit parameters such as the time constant and sensitivity are selected to maximize S/N at the given modulation frequency.

The spectral characteristics can best be understood by consideration of an MFC having two monochromatic light sources of different peak wavelengths, for example blue and red. The term "monochromatic" is used in its ordinary sense, and is meant to include light sources which emit the majority of their energy over a relatively narrow wavelength band. Examples of such monochromatic light sources include, but are not limited to, LEDs, lasers, etc. Spectral emission lamps such as mercury vapor lamps, which typically emit a line spectrum, may also be used, preferably with a filter to isolate one or more wavelengths of interest.

The two light sources of the present illustration may be applied to the interrogation channel through a single optical waveguide, for example through a combining prism, rotating mirror, etc, or may be applied through two different optical waveguides. When two or more optical waveguides are used for the incident light, the waveguides are preferably positioned with their ends which are proximate the interrogation channel as close as possible to each other, and may be angled slightly with respect to each other to try to obtain a condition where a single particle is illuminated by both incident wavelengths in the same position in the channel, as shown in Figure 1. Otherwise, the output signals will be staggered from each other in time.

At least two output optical waveguides are employed in this illustration, one, for example, at an included angle of 90° to the axis of the interrogation channel, and one at an included angle of 45° to the axis of the interrogation channel. Typically, the two output optical waveguides will not receive the same light energy at any given wavelength. Rather, due to refractive, diffractive, and reflective effects, the signal strengths will differ. Moreover, the difference in signal strength will be wavelength dependent. Therefore, by measuring the difference in signal strengths at a first wavelength and the difference in signal strengths at a second wavelength, by comparing the relative differences of these "comparison" signals at the two different wavelengths, a plot of wavelength versus difference in comparison signal can be made. By measuring a difference in comparison signal strength during operation, each difference will correspond to a different wavelength. Thus, the wavelength emitted by a fluorescing particle may be measured by comparing the difference in output signal strength between two output waveguides at different included angles to the interrogation channel.

For example, once the output signal and wavelength relationship has been established, signals may be selected with regard to their comparison output to distinguish between multiple particles having different spectral signatures, i.e. , different fluorescent wavelengths. Thus, in a sample fluid containing two different microorganisms, each receiving a stain which will cause fluorescence at a different wavelength, the detectors, for example one or more photodiodes or PMTs may be filtered to remove the incident light, and the outputs sampled at two different angles with the interrogation channel. The signal strength at any given output waveguide will not yield any information which can distinguish between the different microorganisms. Rather, a signal merely indicates that a particle is pressing through the interrogation zone. However, if at this instant, the difference between the two differently angled output waveguides is measured, the difference will indicate which type of particle is traversing the zone. Thus, a low difference in signal strength will correspond to one wavelength of fluorescence, attributable to one type of particle, while a higher difference will correspond to a second wavelength and hence a second type of particle. By synchronizing particle detection with output signal strength difference between the two output waveguides, a count of each type of particle can be made. In a like manner, the S/N ratio may be used in lieu of a difference or "comparison" signal. As indicated in the Example and as illustrated in Figure 6, the S/N ratio is also wavelength dependent when expressed as a ratio between variously angled detectors.

The means for measuring signal strength, amplifying signals, measuring signal strength differences, etc. , are all accomplished by means well known in the electronic arts. For example, signal strengths may be amplified by conventional operational amplifiers which are readily available, or amplifiers constructed of discrete components. Differences in signal strength may be made by means of comparators or differential amplifiers. The signals may be exclusively analog, or may be digitized by analog-to-digital converters. Once digitized, the signals, their difference (comparison signals), etc., may be processed by suitable software. An output may be presented as a trace on a CRT as shown in Figures 4a, 4b, or in tabular form, etc.

The micro flow cytometers must have an inlet for supplying test fluid, and an outlet. However, these need not be made in such a fashion that fluid necessarily needs to enter or exit the device during measurement. For example, the inlet may comprise a well or reservoir holding fluid until fluid flow is desired to be established. In like fashion, the outlet may continually remove fluid from the device, or again may comprise a well or reservoir to receive fluid from the interrogation channel. In disposable MFCs, outlet wells or reservoirs may be highly desirable, to avoid the necessity of disposing of microorganism containing fluids. In preferred devices, photodiodes are employed as the detector, for reasons advanced earlier. Due to the lack of internal amplification and noise associated with such devices, it is preferable also as earlier indicated, to modulate the incident light and to amplify detected output light synchronously. For example, a function generator may be used to modulate the incident light source such as an LED, and also to provide a signal to the amplifier for synchronous amplification. The function generator may produce any type of output, for example sinusoidal, square wave, sawtooth waves, etc. When two or more incident wavelengths are desired to be used, they are most preferably driven substantially out-of-phase with respect to each other. For two sources, for example, they may be 180° out-of- phase. Some overlap of the light intensities at multiple wavelengths may be tolerated. However, the greater the overlap, the less will be the signal-to-noise ratio. A function generator may be easily arranged to provide signals which would produce no overlap. The term "substantially out-of-phase" implies that the peak intensities of the various wavelengths can be distinguished from each other with respect to time.

Figure 3 illustrates a schematic of a synchronous ("lock in") detection system which can be assembled with conventional components. A 1 MHZ oscillator 20 is chosen to provide signal to both the laser driver 21 and synchronous detector 22. The laser driver provides a suitable driving signal for laser diode or LED 23, causing it to emit pulses of light a the 1 MHZ signal rate. The laser light, upon contacting a particle 24 in the interrogation zone causes a fluorescent output which is detected by PIN photodiode 25. The resulting electrical signal is amplified by a low noise differential amplifier 26, and optionally further amplified by additional amplifier (s) 27. The amplified signal is routed to a high Q (e.g. Q= 100) bandpass filter 28, which may, for example, limit the output signal to the range of 995 KHz to 1.005 MHZ, only a 10 KHz bandwidth. This output signal from the bandpass filter is routed to the synchronous detector 22, the output of which may be digitized by analog to digital (A/D) converter 29, and the data subsequently processed and/or collected by microcontroller or microcomputer 30. The term "emanating" is used to describe light from a particle having been illuminated by incident light. Light emanating from such a particle may have the same wavelength as the incident light, for example by the incident light having been reflected from or scattered by the particle. In most cases, however, the particles will fluoresce, i.e. , emit one or more wavelengths different from the incident light wavelength. Fluorescence at a different wavelength provides the opportunity to eliminate the rather strong incident light from the output, resulting in a significant increase in the signal-to-noise (S/N) ratio which may be obtained. The term "comparator" means a device or plurality of devices which are able to compare relative signal strengths. A comparator need not be a discrete device, but may, for example, include detectors, amplifiers, analog to digital converters, and a microprocessor. The output of the comparator may be processed further to provide any desired type of signal.

Thus, in one aspect, the subject invention pertains to a spectral micro flow cytometer device with a micro flow cytometer having an inlet, an outlet, and an interrogation channel in fluid communication with the inlet and outlet, the interrogation channel defining an interrogation plane in which the channel is positioned; niinimally three optical waveguides teπninating proximate to the channel and in the interrogation plane, each optical waveguide directed toward the channel at a different angle with respect to the axis of the channel, the inlet direction of the channel being 0°; minimally one incident light source coupled to an optical waveguide and illuminating the channel; and a comparator which compares a light intensity or a signal to noise ratio at one optical waveguide positioned at a first included angle, with a light intensity or a signal to noise ratio at a further optical waveguide positioned at another included angle, the difference in intensity or in signal to noise ratio being proportional to the wavelength of light emanating from a particle passing through the channel.

In another aspect, the micro flow cytometer contains a modulated, first monochromatic light source which directs light into an incident optical waveguide such that modulated monochromatic light impinges upon particles within an interrogation channel; optionally, a modulated, second monochromatic light source of a different color whose modulation is substantially out-of-phase with the modulation of the first monochromatic light source, and which directs light to the interrogation channel through a second incident light optical waveguide, which may be the same as or different from the first incident light waveguide; a first photodiode detector sensitive to light from the first monochromatic light source, the second monochromatic light source, or both, or sensitive to a fluorescent wavelength emitted by a particle in the channel as a result of the particle being illuminated by one or more light sources, the photodiode detector communicating with the channel through a first output waveguide in a plane with the first incident waveguide and the channel, and positioned at a first included angle with respect to the channel; a second photodiode detector communicating with the channel through a second output waveguide in a plane with the interrogation channel and the first incident waveguide, but at a different included angle to the channel than the first output optical waveguide; at least one amplifier which amplifies an output from the photodiode detector(s); a means for selectively sampling an amplified output from the amplifier synchronously with a respective modulation frequency of the first or second monochromatic light sources to provide an output proportional to incident light fluorescent light impinging on the detector, and a second output proportional to light of a different incident color or fluorescent color; a means for comparing the signal strength obtained from the first detector with the second detector at at least two different wavelengths to obtain an angularly dependent comparison signal or signal to noise ratio for each wavelength; and a means for comparing the angularly dependent comparison signals or signal to noise ratios to each other to provide spectral differentiation of light collected by the respective output waveguides. A yet further aspect of the invention pertains to a method for providing spectral information for particles in an interrogation channel of a micro flow cytometer, by illuminating the particles with incident light directed to the channel through an incident optical waveguide; transmitting reflected, scattered, or fluorescent light from a particle in the channel through separate output optical waveguides positioned in a plane which includes the interrogation channel, the output optical waveguides positioned at different included angles with respect to the axis of the channel; measuring light intensity or signal to noise ratio from each output optical waveguide at two different wavelengths, and providing comparison output signals proportional to the difference between the intensity of each respective wavelength from the output optical waveguides, thus providing a comparison output signal for a first wavelength and a second comparison output signal for a second wavelength; and comparing the comparison output signals or signal to noise ratios, by means of which spectral information relating to the particles being observed can be collected, and if desired, further processed.

Example

The PDMS observation cell with a microfluidic channel and optical fiber microgrooves is fabricated using replica molding. The process begins with the patterning of a negative tone SU-8 photoresist (SU-8 2050, Microchem Co., MA) mold on a silicon wafer using lithography. This is followed by silanization of the wafer surface using 3-(trimethoxysilyl)propyl methacrylate (Sigma- Aldrich Co.,

MO), which facilitates the parting of a PDMS layer during the next process step. A PDMS precursor liquid is then well mixed with a curing agent (Sylgard 184

Silicone Elastomer Base, Dow Corning Co., MI) and poured onto the SU-8 micromold on the silicon wafer. After curing for 24 hours at room temperature, the

PDMS slab with embossed observation channel features is peeled off. The final size of the fabricated flow cell is 4x2cm². The microfluidic channel has a length of 1 cm and cross-sectional area of 100 x 300μm².

The flow profile of the hydrodynamically-focused sample liquid is observed using optical and confocal microscopy. These methods permit characterization of the hydrodynamic focusing performance of the PDMS microfluidic channel and provide cross-sectional images of the sample fluid. For the confocal microscopy study, sample water containing fluorescein (Sigma- Aldrich, MO) is utilized as the characterized fluid. The sample fluid is excited by a blue laser contained within the confocal microscope (LSM-510, Zeiss, Germany).

Figure 2 illustrates the experimental testbed used to evaluate the performance of the microfluidic and optical design of the PDMS micro flow cytometer previously described. FluoSpheres^® polystyrene microspheres (Molecular Probes, OR) are used as a reference sample to illustrate the performance of the micro flow cytometer, using scarlet fluorescent (F-8843) microspheres with 15.5μm average diameter, and green fluorescent (F-21010) microspheres with 15 μm average diameter, as listed in Table 1. Significant variation in the absorption and emission wavelengths and intensities is observed experimentally.

TABLE 1

The semiconductor laser diodes used to excite the reference fluorescent microspheres include 1) a 6mW red iode laser (SCFC2-0635-0006-T-25, Optical Fiber System Inc. , MA) with a peak wavelength of approximately 635nm, and 2) a 5mW blue diode laser (NDHB500APAE1 Engineering Sample, Nichia Co. , Japan) with a peak wavelength of approximately 440nm. The blue and red laser excitations are modulated at 100 kHz with differing phase angles (e.g. , signals 180 degrees apart in phase). Multimode optical fibers (F-MCB-T, Newport Co., CA) with lOOμm core diameter and 1 lOμm cladding diameter are utilized to transmit energy to and from the interrogation zone as shown in Figure 2. The optical filters are selected according to the transmission spectral characteristics listed in Table 2 in order to separate the high-intensity excitation light from the relatively weak fluorescent light emitted by the above-described microspheres. The photocurrent generated at the PIN photodiode (S7329-01, Hamamatsu Photonics Co., NJ) is amplified using an OP-amplifier-based circuit (10⁴X, OPA655, Texas Instruments Inc., TX). The circuit converts the photocurrent signal to an amplified voltage signal. The voltage signal is processed using a lock-in amplifier (SR-840, Stanford Research, Inc. , CA) and extracted from the background noise. A block diagram is illustrated in Figure 3.

TABLE 2

Two independent experiments are conducted to fully validate the performance of the PIN-based signal detection process. First, fluorescence detection with and without the lock-in amplifier was performed. To achieve this, the signal detected at the PIN photodiode is guided to a pre-amplifier circuit and then identically divided into two cables. One of the divided signals is directly monitored and the other was processed by the lock-in amplifier prior to monitoring. The signals are then compared with each other.

Second, signals simultaneously obtained from a PMT (P30CWAD5- 01 , Electron Tubes Inc. , NJ) and PIN photodiode are compared with each other to validate the performance of the signal detection system using the PIN photodiode and the lock-in amplifier. The PMT was selected as a reference detector, as it is widely used in cytometry and known to perform well in this application. The PMT is powered by a 900-1200V voltage source to obtain maximum S/N. For this experiment, one excitation and two detection optical fibers are embedded into the observation cell at the same angle. The detection fibers, one of which is connected to the PIN photodiode and the other connected to the PMT, simultaneously capture the fluorescent signals and send them to the photodetectors. The signals from the

PIN photodiode with lock-in amplifier and the PMT (without signal processing) are then compared.

To better understand the influence of experimental conditions on the PDMS micro flow cytometer, scarlet/green inorganic and biological particles are tested under variable sample/sheath flow rate ratios, and multiple optical detection angles. Three major parameters are investigated: 1) S/N, 2) S/N variation, and 3) recovery rate. The recovery rate is estimated by injecting a sample fluid with polystyrene microspheres of known concentration (10⁶/30 beads/ml) into the microchannel. The final microsphere concentration is achieved by a 1:30 dilution in buffer solution (Tris Buffered Saline with bovine serum albumin (BSA), T6789, Sigma- Aldrich Co., MO).

To demonstrate two-color excitation and detection using the developed micro flow cytometer, two pairs of optical fibers are arranged at an angle of 135° between the excitation and detection fibers as shown in Figure 2. A mixed sample consisting of scarlet and green microspheres (each diluted 1:30 with buffer solution) was injected into the microfluid channel and hydrodynamically focused. The sample was excited at a single interrogation point using two lasers with 440nm and 635nm wavelengths. In order to prevent the coupling of the two excitation light sources, the red and blue laser diodes were driven 180° out-of-phase (100kHz).

To demonstrate the feasibility of using the micro flow cytometer to detect fluorescent labeled biological samples, an experiment with the fungus Saccharomyces cerevisiae is performed. To prepare the fungus sample, a powder sample of dried cells is mixed into warm water and maintained at 38°C for about 20 minutes to revive the culture. After full activation of the cells, a 2μl yeast solution is stained using Syto^® 62 (or Syto^® 44) (Molecular Probes, OR) and incubated for 20 minutes at room temperature. The yeast cells, with fluorescently labeled nucleic acid, are injected into the flow cell at a sample flow rate of 5ml/hr with total flow rate of 15ml/hr and excited using the 635nm red (or 440nm blue) laser diode. The optical signals are separated from the excitation radiation using the 695nm band-pass (or 480 long-pass) filter.

When examined by optical and confocal microscopy, the sample fluid is well confined in the center of the channel in the horizontal direction. It is found, as predicted, that the width of the focused sample fluid can be controlled by changing the ratio of sample/sheath flow rate. It is observed that by tuning the ratio of flow rates between the sample fluid and the sheath fluid (maintaining fixed total flow rate of 15ml/hr), experimental results conferred with CFD simulation results to within 10% error.

Figure 4a demonstrates the simultaneous detection by the PMT (without advanced signal processing) and the PIN photodiode detection (4b) with lock-in technique. It is observed that all particles detected with PMT are observed with the PIN photodiode with equal or higher S/N.

The micro flow cytometer performance was evaluated for the case of fluorescent microsphere detection. Three parameters were used as performance metrics: S/N, S/N variation, and recovery rate. These performance parameters are discussed in the following paragraphs.

S/N: Figure 5 shows the average S/N corresponding to various angles between the excitation and detection fibers. The S/N values obtained from the experiments show that the device can successfully detect the fluorescence emission in different angles for 15μm polystyrene microspheres. The detection angle resulting in maximum S/N ( — 9.5) is 180°. For the scarlet polystyrene microspheres, the S/N is slightly reduced when measuring at 45° and 135° off the excitation direction. In contrast, the S/N at 45° and 135° for green polystyrene microspheres measurement is reduced to about 35% of that in 180°. These results have useful and significant ramifications for estimating fluorescence color that will be discussed later.

The S/N variation (expressed as the coefficient of variation percentage) varies at different angles for numerous ratios of sample/sheath flow rates. The minimum variation occurs at 180° detection and at the lowest ratio of sample/sheath flow rate. The S/N variation for the 180° detection is observed to be approximately 25% and 35% for the scarlet and green polystyrene microspheres respectively over most of the flow rate ratios investigated. The variation over this range is believed to be in part due to the variation in the dimensions and intensity of the microspheres, and in part due to the variation in the vertical position of the microspheres in the microfluidic channel. Furthermore, the S/N variation is observed to increase as the ratio of sample/sheath flow rate became higher. This variation increase is more obvious at 45° and 135° detection than it was at 180°. As will be discussed later, this is likely due to the wider column of sample fluid at higher at higher sample/sheath flow rate ratios, and the increased possibility of multiple particles in the interrogation zone at the same time under this condition.

The recovery rate also varies for different angles at different ratios of sample/sheath flow rate. The maximum recovery rate (75 %) is obtained at 180° detection. Higher recovery rates generally are observed with increasing sample/sheath flow rate ratio. Deviations from 100% recovery are discussed later.

Experiments with simultaneous two-color excitation/detection demonstrates the feasibility of the compact optical arrangement as shown in Figure 2. The high S/N ratio (6.4 and 2.1 for scarlet and green microspheres, respectively) of the detected signal shows that the system is capable of simultaneous two-color excitation/detection with a single interrogation zone. Moreover, the recovery rates of the two-color excitation/detection system are observed to be only slightly lower than those of the single-color results. The differences are approximately 10-20% , which may be due to experimental variation, or correctable interference between the detection systems of the two colors. In observing the electronic signal corresponding to the detection of

Saccharomyces cerevisiae labeled with Syto 62^®, the concentration of the yeast cells calculated from the experimental results is about 1.90xl0⁴ cells/ml, with a S/N of approximately 2.8. The results show that the micro flow cytometer is effective in detecting yeast cells, with high potential for application to other biological cells.

S/N: The data in Figure 5 indicate that the average S/N is almost invariant regardless of the detection angles for the scarlet fluorescent microspheres, whereas it is highly influenced by angle for the green fluorescent microspheres. Moreover, it is observed that the signal intensity is reduced at the 45° and 135° angles compared to that at the 180° angle. The reduced off-angle signal intensities are likely due to numerous opportunities for scattering losses at the off angles. For instance, when the detection optical fiber is not aligned normal to the channel, the optical path can be significantly bent due to the difference in index of refraction at the interfaces between the PDMS channel wall and the fluid, and between the PDMS flow cell and the air gap in front of the optical fiber. In addition to optical path bending, the distance between the end of the detector fiber and the center of the interrogation zone, d_c, highly influences the magnitude of the detected light intensity. In the current flow cell design, the 180° detection arrangement has a shorter distance between these two points (d_c~20Q μm) than the 45° and 135° detection arrangements ( _c~330μm). The longer distance leads to weaker detected signals at the 45° and 135° angles, as would be predicted if the sample particles were assumed to be point light sources positioned at the center of the interrogation zone. In this case, the intensity of the light collected by the optical fiber is proportional to d_c ^~2, and it would be predicted that the signal intensity at the 45° and 135° angles should be about 37% as that at 180°. From the experimental results, the average signal intensities at 45° and 135° were 36-42% of that at 180°, which is in reasonable agreement with the prediction.

Moreover, experimental results show that the relative S/N between various detection angles varies for sample particles with different spectra.

Consequently, by taking advantage of simultaneous multiple-angle detection, a micro flow cytometer with spectrometer characteristics is possible. To test this hypothesis, both scarlet and green fluorescent microspheres (F-2101 and F-8843) and

Saccharomyces cerevisiae labeled by Syto^® 62 and Syto^® 44 are examined, and the ratio of average S/N at 180° and 45° detection is calculated for both green and scarlet particle/cell detection. As shown in Figure 6, both microsphere and biological detection yield a distinct ratio of 180° to 45° S/N depending on the fluorescent color of the interrogation target. The difference of the ratio is consistent with color. Thus, spectral flow cytometers for specialized biological assays targeted at simultaneous multi-species detection are possible with the devices of the subject invention. To determine whether particles smaller than 15 μm can be detected using this micro flow cytometer design, scarlet fluorescent polystyrene microspheres with 6 μm average diameter are also tested in the micro flow cytometer using the 180° optical fiber arrangement. The S/N ( ~ 1.5) is low but detectable, suggesting that the signal will be prohibitively weak if the particle size becomes smaller than 6 μm. Ideally, the intensity of fluorescence emission is proportional to the surface area of the fluorescence particles, and therefore the fluorescence intensity of the 15 μm microspheres is expected to be about 6.25 times larger than that of the 6 μm microspheres. The experimentally observed results are in close agreement with this prediction, showing a S/N difference of 6.3 (9.5/1.5).

S/N Variation: The variation of the S/N at a given detection angle can be correlated with the shape and area of the interrogation zone, Ω_;, that is defined by the intersection of the sample fluid column, the optical excitation zone, and the optical detection zone as shown in Figure 12. To account for the S/N variation observed experimentally, it is assumed that:

• The positional variation of the particle passing through Ω₍- is negligible in the vertical direction. • The particle can pass through any point in Ω, at the same velocity, with the same probability.

• The optical excitation in Ω, is spatially uniform.

• The detected particle is a point light source.

• The optical path bending can be ignored at any interfaces.

• The optical attenuation loss in the optical fiber is negligible.

Based on these assumptions, the intensity of the fluorescence signal / detected from the sample particle at an arbitrary point r = (x,y) in Ω₍- is given by.

(1)

where I_Q is the original total intensity of the excited fluorescent microsphere, and r_d represents the position of the end of the detection optical fiber in the x-y coordinate system in Figure 12. The maximum and n inimum signal intensities are given by the maximum and minimum of the maximum intensity calculated at each x position within the interrogation zone, Ω_;-, which can be expressed as

(2) f₁ (χ) = -J- + . * cos θ cos θ (3)

d f₂ (^χ) +- cosø cos# (4)

where d_f is the excitation and detection optical fiber diameter, w_c is the width of sample fluid column, and θ is the angle between excitation and detection direction. In this case, the S/N variation ΔS/N is estimated as

7 17. ΔS / N = max — 7 min

(5)

where (I) is the averaged signal intensity given by / r\ max min

(6)

Figure 7 shows the calculated S/N variations at varying sample column width and detection angle. Due to the geometrical symmetry, the calculations for 45° and 135° detection angles yield the same predictions.

As can be seen from Figure 7, the estimated S/N variation of 180° detection is close to the observed experimental result obtained from green fluorescence detection, with the difference within 10% . In contrast, the predicted S/N variation of 45° or 135° detection is much lower than that observed experimentally. This contradiction is a likely result of ignoring optical path bending in the model, which makes 180° detection less sensitive to the position of the sample particles in the microfluidic channel when compared with 45° and 135° detection.

Recovery Rate: At a low sample flow rate relative to the sheath flow rate ( — 3:12), the sample stream becomes unstable due to the large pressure difference between the sample and sheath flows near the channel inlets, resulting in discontinuous fluid motion. The discontinuous sample flow causes particles to become trapped in the inlet reservoir, which is one of the major reasons for the low detection recovery rate at a low sample flow rate (10-30%). As the sample flow rate increases to nearly 50% of the sheath flow rate, the sample stream can establish a stable liquid column by overcoming the pressure difference at the channel inlets. The stable liquid column is more effective at centering the sample particles in the center of the interrogation zone in a consistent manner. The stable flow of the sample particles leads to a higher recovery rate (55-75%) at the higher sample/sheath flow rate ratio. On the other hand, larger liquid column width resulting from a higher sample flow rate is likely to prohibit a single-file sample flow, allowing multiple particles to simultaneously enter into the interrogation zone. Fluorescence signal intensities detected from the scarlet polystyrene microspheres at sample/sheath flow rate ratio of 6:9 and 7:8, respectively (total flow rate of 15ml/hr) show that the signal intensity distribution is approximately divided into two subgroups at a flow rate ratio increases to 7:8. The middle intensity of each subgroup relates to the others in a whole number ratio. For instance, the middle intensities in Figure 14(b) are approximately in the ratio of 1 : 2 : 3 (0.45 : 0.90 : 1.35) . Assuming that the signal intensity is proportional to the number of the simultaneously detected beads, these results appear to indicate that multiple fluorescent particles are detected in the interrogation zone at the same time. To test this hypothesis, the detection algorithm is modified to account for the possibility of multiple beads passing through the interrogation zone. With this modification to the detection algorithm, the recovery rate increases by about 25-30% (for flow rate ratio 6:9 and 7:8), leading to about 95% recovery for the fluorescence detection in the micro flow cytometer.

Even with the intensity correction, 100% recovery is not observed due to some loss of microspheres in the microfluidic interconnects. To minimize such losses, a buffer fluid containing BSA is utilized. BSA is found to be effective in coating the PDMS channel wall surface to prevent particle adhesion. The BSA buffer doubles recovery rates relative to experiments with other buffers such as PBS (phosphate buffered saline) or de-ionized water, both of which have little effect on changing the surface chemistry of the channel wall. Lower recovery rates are also observed at the 45° and 135° detection angles. This suggests that the fluorescence signal may become weak enough to be indistinguishable from the background noise when the sample particles take a fluidic path far away from the detection optical fiber. The observation is likely due to the fact that optical path bending can shift the detection zone from the center of excitation zone defined by the laser diodes. This shifting can generate a partial or total blind spot for the 45° or 135° detector fibers, resulting in a low recovery rate. This is readily correctable by modifying the optical pathways in the system through the use of micro optical components in a next generation design of the system.

Single-Color vs. Two-Color Excitation/Detection: For two-color excitation/detection setup, the S/N ratios are about 16% (for λ=635nm) and 50% (for λ-440nm) lower than that observed for the single-color excitation/detection setup. For the most part, this is because the background noise level in the two-color system is approximately 30% and 100% larger for scarlet and green fluorescence detection respectively. The background noise increase for the 2-color system is likely caused by the excitation light of one color leaking into the detection optical fiber. The leakage of excitation light and imperfect filtering can be solved by use of customized optical filters based upon the specific wavelength spectra of the excitation and detection targets. This will yield a two-color excitation/detection microcytometer possessing nearly the same S/N as the single-color system discussed previously.

While the inventive microsystem is highly successful to demonstrate the feasibility of producing a two-color micro-flow cytometer with PIN photodiodes and color differentiation ability, a number of design modifications would be achieved by designing the flow cell to have a channel width more comparable to the sample particle diameter. This channel design would result in a larger ratio of the sample flow rate to the sheath flow rate at a fixed combined total flow rate. In this case, a stable sample liquid column could be established while keeping its width narrow enough to confine the particle in the center of the interrogation zone with high precision. The modified sample liquid column width would improve the recovery rate and reduce the S/N variation, and promote a single-file flow in the microfluidic channel.

With respect to the optics design, micro optical components such as micro lenses and apertures could be integrated into the flow cell to prevent the optical bending and reduce background noise. This would also help increase the amount of collected light and better define the shape of the excitation area, leading to a higher S/N. These optical components would be designed to allow optical excitation to be more spatially uniform within a smaller area in the channel, which would reduce the S/N variation. Moreover, the blind spot that appears for the off- angle detection can be eliminated with a new optics design, which will naturally lead to an improved recovery rate at the off-angles. In addition, a new optics system could incorporate a larger number of detection optical fibers embedded into the observation cell. These optical fibers can simultaneously collect the fluorescence emission from a single particle within the interrogation zone. The signals collected at each optical fiber can be combined to amplify the signal and to cancel the random components of the background noise. It appears that practical and affordable particle detection below 6 microns will be readily achievable with this design approach.

While embodiments of the invention have been illustrated and described, it is not intended that these embodiments illustrate and describe all possible forms of the invention. Rather, the words used in the specification are words of description rather than limitation, and it is understood that various changes may be made without departing from the spirit and scope of the invention.

Claims

WHAT IS CLAIMED IS:

1. A spectral micro flow cytometer device, comprising: a) a micro flow cytometer having an inlet, an outlet, and an interrogation channel in fluid communication with said inlet and said outlet, said interrogation channel defining an interrogation plane in which said interrogation channel is positioned; b) a plurality of minimally three optical waveguides terminating proximate said interrogation channel and positioned in said interrogation plane, each optical waveguide of said plurality of waveguides directed toward said interrogation channel at a different angle with respect to the direction of said interrogation channel, with the inlet direction of said interrogation channel being 0° ; c) minimally one incident light source coupled to one of said optical waveguides and illuminating said interrogation channel; d) a comparator which compares a light intensity or a signal to noise ratio at one optical waveguide at a first included angle, with a light intensity or a signal to noise ratio at a further optical waveguide at a further included angle, the difference in intensity or in signal to noise ratio being proportional to the wavelength of light emanating from a particle passing through said interrogation channel.

2. The device of claim 1 , wherein a first included angle of a first optical waveguide with the interrogation channel and a further included angle of a further optical waveguide with the interrogation channel are such that the change in intensity of a first wavelength of light measured at said first optical waveguide and said further optical waveguide is different from the change in relative intensity of a second wavelength of light measured at said first optical waveguide and said further optical waveguide.

3. The device of claim 1, wherein said optical waveguides comprise optical fibers.

4. The device of claim 1, wherein two or more incident light sources of differing spectral characteristics are employed.

5. The device of claim 4, wherein at least two incident light sources are supplied to two different optical waveguides.

6. The device of claim 1, wherein at least one incident light source is monochromatic.

7. The device of claim 1, wherein at least two monochromatic incident light sources are employed.

8. The device of claim 7, wherein said two incident monochromatic light sources are pulsed light sources out-of-phase with respect to each other.

9. The device of claim 1, comprising first and second output optical waveguides positioned at different included angles with respect to the axis of the interrogation zone, and means for calculating the signal to noise ratio of the electrical signal from photodetectors receiving light from the respective output optical waveguides.

10. The micro flow cytometer of claim 1, comprising: a modulated, first monochromatic light source having a first peak wavelength directing light into a first incident optical waveguide such that modulated monochromatic light impinges upon particles within said interrogation channel; optionally, a modulated, second monochromatic light source having a second peak wavelength different from said first peak wavelength, whose modulation is substantially out-of-phase with the modulation of said first monochromatic light source, said second monochromatic light source directing light to said interrogation channel through a second incident light optical waveguide, said first and second incident waveguides being the same or different; a first photodiode detector sensitive to light from said first monochromatic light source, said second monochromatic light source, or both said first and said second monochromatic light sources or to a wavelength corresponding to a fluorescence emitted by a particle in said interrogation channel as a result of said particle being illuminated by one or more of said first and second light sources, said at least one photodiode detector communicating with said interrogation channel through a first output optical waveguide in a plane with said first incident waveguide and said interrogation channel, and having a first included angle with said interrogation channel; at least a second photodiode detector communicating with said interrogation channel through a second output optical waveguide in a plane with said interrogation channel and said first incident waveguide, but at a different included angle to said interrogation channel than said first output optical waveguide; at least one amplifier which amplifies an output from said photodiode detector(s); means for selectively sampling an amplified output from said at least one amplifier synchronously with a respective modulation frequency of said first and said second monochromatic light sources to provide a first output proportional to light of an incident wavelength or a fluorescent wavelength impinging on said detector, and a second output proportional to light of a different incident wavelength or a different fluorescent wavelength; means for comparing the signal strength obtained from said first photodiode detector with said second photodiode detector at at least two different wavelengths to obtain an angularly dependent comparison signal for each wavelength, and/or means for calculating a signal to noise ratio from said first photodiode detector and said second photodiode detector; means for comparing the angularly dependent comparison signals or the signal to noise ratios to each other to provide spectral differentiation of light collected by said first output waveguide and said second output waveguide.

11. A method for providing spectral information for particles in an interrogation channel of a micro flow cytometer of claim 1, comprising: illuminating said particles with incident light from at least one incident optical waveguide; transmitting reflected, scattered, or fluorescent light from a particle in said interrogation channel through at least first and second output optical waveguides, all of said optical waveguides in a plane including said interrogation channel, said first and second output optical waveguides at different included angles with respect to the axis of said interrogation channel; measuring light intensity or signal to noise ratio from each output optical waveguide at each of at least two different wavelengths and providing comparison output signals proportional to the difference between the intensity of each respective wavelength from the first and second output optical waveguides, thus providing a first comparison output signal for a first wavelength and a second comparison output signal for a second wavelength; comparing said first and said second comparison output signals or signal to noise ratios, a difference being spectrally dependent.

12. The process of claim 11 , wherein an output signal is amplified by a photomultiplier tube.

13. The process of claim 11 , wherein incident light is modulated, and output signals are amplified in phase with modulated incident light.

14. The process of claim 11, wherein said incident light comprises two or more modulated, monochromatic light sources of different peak wavelengths, each of said monochromatic light sources modulated out-of-phase with all other monochromatic light sources.

15. A method for spectrally distinguishing output signals from a device of claim 1 , comprising illuminating particles in said interrogation channel with an incident light wavelength which causes said particles to fluoresce with a fluorescent wavelength different from the incident light wavelength; measuring the signal to noise ratio of a photodetector signal from a first output optical waveguide and a photodetector signal from a second output optical waveguide, said first and second output optical waveguides positioned at different angles with respect to said interrogation channel; identifying the fluorescent wavelength emitted by a particle by the ratio of signal to noise ratio at one photodetector to the signal to noise ratio at the other photodetector.