HK1097039A

HK1097039A - Method and apparatus for multiwavelength imaging spectrometer

Info

Publication number: HK1097039A
Application number: HK07102035.6A
Authority: HK
Inventors: 约翰．S.．麦尔; 帕特里克．J.．特瑞多; 王晨晖
Original assignee: 凯米映像公司
Priority date: 2003-07-18
Filing date: 2004-07-19
Publication date: 2007-06-15

Description

Method and apparatus for multi-wavelength imaging spectrometer

RELATED APPLICATIONS

[0001] This application claims benefit of filing date priority from U.S. provisional application No. 60/488,246, filed on 18/7/2003, the entire disclosure of which is incorporated herein by reference. Cross-reference to patent applications filed concurrently herewith entitled Method and Apparatus for Compact birefringence Interference Imaging Spectrometer (Attorney Docket No. che 01009), Method and Apparatus for Compact Dispersive Imaging Spectrometer (Attorney Docket No. che 01010), and Method and Apparatus for Compact resonance Imaging Spectrometer (Method and Apparatus for Compact resonance Imaging Spectrometer) (Attorney Docket No. che 01012), the descriptions of each of which are incorporated herein in their entirety.

Background

[0002] Spectral imaging, which combines digital imaging and molecular spectroscopy techniques, can include Raman scattering, fluorescence, photoluminescence, ultraviolet, visible and infrared absorption spectroscopy. When applied to chemical analysis of materials, spectroscopic imaging is commonly referred to as chemical imaging. Instruments that perform spectroscopic (i.e., chemical) imaging typically include image collection optics, a focal plane array imaging detector, and an imaging spectrometer.

[0003] Typically, the sample size determines the choice of image collection optics. For example, microscopes are typically used for analysis of sub-micron to millimeter spatial dimension samples. For larger objects, macro lens optics are appropriate in the millimeter to meter size range. For samples located in relatively inaccessible environments, a soft fiberscope or a rigid borescope may be used. For very large objects, such as planetary objects, telescopes are suitable image collection optics.

[0004] For the detection of images formed by various optical systems, two-dimensional imaging Focal Plane Array (FPA) detectors are typically used. The choice of FPA detector is governed by the spectroscopic technique used to characterize the sample of interest. For example, silicon (Si) Charge Coupled Device (CCD) detectors or CMOS detectors are typically used with visible wavelength fluorescence and Raman spectral imaging systems, while indium gallium arsenide (InGaAs) FPA detectors are typically used with near infrared spectral imaging systems.

[0005] A variety of imaging spectrometers have been designed for spectral imaging systems. Examples include, but are not limited to, grating spectrometers, filter wheels, Sagnac interferometers, Michelson interferometers and tunable filters such as acousto-optic tunable filters (AOTFs) and Liquid Crystal Tunable Filters (LCTFs).

[0006] Many imaging spectrometers, including acousto-optic tunable filters (AOTFs) and Liquid Crystal Tunable Filters (LCTFs), are polarization sensitive, passing one linearly polarized light and rejecting the orthogonal linearly polarized light. An AOTF is a solid state birefringent crystal that provides an electronically tunable spectral notch passband in response to an applied acoustic field. The LCTF also provides a notch passband that can be controlled by incorporating a liquid crystal retarder into a birefringent interference filter, such as a Lyot filter. Conventional systems are typically bulky and immobile. Hand-held chemical imaging sensors capable of performing transient chemical analyses would represent an advance in size, weight and cost reductions. Therefore, there is a need for a tunable filter that is portable, lightweight, and more efficient.

Disclosure of Invention

[0007] In one embodiment, the disclosure relates to an optical filter that passes photons. The filter includes a first filtering stage and a second filtering stage. The first filter stage may comprise a first retarder element and a first liquid crystal cell. The first element may comprise an input face and an output face. One of the first element faces is oriented not substantially perpendicular to the trajectory of the photons passing through the filter.

[0008] In another embodiment, the disclosure is directed to an optical filter having a first liquid crystal cell including a first region having a first photon retardation value and a second region having a second photon retardation value. The optical filter may also include a second liquid crystal cell including a first region having the first photon delay value and a second region having the second photon delay value. Each of the first and second delay values may be determined by a first excitation source and a second excitation source in communication with each of the first and second regions of the first and second cells, respectively.

Drawings

[0009] FIG. 1 is a schematic representation of a conventional line scan Raman imaging system;

[0010] FIG. 2 is a schematic diagram of a conventional wide field of view scanning Raman imaging system;

[0011] FIG. 3 is a schematic representation of a three-stage conventional Lyot liquid crystal tunable filter;

[0012] FIG. 4A schematically illustrates a uniform liquid crystal cell of an electronically controlled birefringent cell;

[0013] FIG. 4B schematically illustrates a wedge-shaped liquid crystal cell of an electronically controlled birefringent cell;

[0014] fig. 4C is a schematic illustration of an optical device and an optical stage according to one embodiment of the disclosure;

[0015] fig. 5 is a schematic illustration of an optical stage according to another embodiment of the disclosure;

[0016] fig. 6 schematically illustrates a stepped LCTF device for use in unshaped Raman imaging or other chemical imaging applications, in accordance with another embodiment of the disclosure;

[0017] FIG. 7 is a schematic diagram of a columnar LCTF device for Raman imaging or other chemical imaging applications, in accordance with another embodiment of the disclosure;

[0018] FIG. 8 is a schematic diagram of a columnar LCTF device for Raman or other chemical imaging applications, according to one embodiment of the disclosure;

[0019] FIG. 9 is a schematic view of an unshaped Raman imaging or other chemical imaging application system, according to another embodiment of the disclosure;

[0020] FIG. 10 is a schematic diagram of an imaging system using a tunable Fabry-Perot;

[0021] 11A-11C are schematic diagrams of a hand-held chemical imaging threat assessment device detection apparatus (CHITA) according to one embodiment of the present invention;

[0022] FIG. 12 is an auxiliary illumination source according to one embodiment of the disclosure;

[0023] FIG. 13 schematically illustrates operation of an example detection apparatus;

[0024] FIG. 14 provides an example package selection in accordance with one embodiment of the disclosure; and

[0025] fig. 15 shows dispersive Raman spectroscopy examination of a sample through PMMA.

Detailed Description

[0026]Fig. 1 is a schematic representation of one type of conventional line-scan Raman imaging system, particularly a scatter-scan Raman imaging system using a "push-scan" scanning scheme. The sample 105 is illuminated from a source 110 and light energy is reflectively scattered back from the sample and collected at optics 115. Images of the sample are accumulated from successive images at adjacent parallel lines in the image. An image of the sample is obtained for each such row and at each of a plurality of specific wavelengths in the spectrum. Thus, a row or one-dimensional set of light amplitudes is at a given position X₁，Y₁…Y_nAnd (6) sampling. The light is filtered for a particular wavelength using a spectrometer 120 that functions as a variable wavelength bandpass filter. The continuous amplitude illustrated in plot 140 is converted to a given wavelength λ by advancing from one row to the next in a "push-and-sweep" sequence₁Of the array 160. By sampling at different wavelengths, an independent substantially monochromatic image 160 is produced for each wavelength value λ₁～λ_nAnd collected. The goal of this technique is conventionally to produce a set of monochromatic images that can be compared to each other to help identify features in the image that can be characterized by a contrast in their intensity compared to adjacent features and other wavelengths.

[0027] The source 110 may be a laser, a fluorescent source, or another source. The reflected scattered photons are received by the optical objective 107 and directed to the spectrometer 120 via optics 115. Particular schemes for collecting an image of a row of pixels may include focusing the image on a linear array of photosensors such as a CCD, or scanning and sampling using one photosensor. The spectrometer 120 needs to be tunable in some way to selectively pass one wavelength bandpass at a time. It is possible to use grating or prism or birefringent crystal schemes for wavelength selection in different configurations. Collecting the amplitude for each pixel location and each wavelength can take considerable time while tuning from one wavelength to another and while proceeding from one row to another (in any order).

[0028] FIG. 2 is a schematic diagram of a conventional two-dimensional Raman imaging system. The system operates somewhat the same as the system of fig. 1, namely illuminating the sample 205 with a source 210, collecting the reflected image via optics 215, selectively passing the bandpass wavelengths via a tunable spectrometer 220 and sampling the image at a photosensitive array 230. However, this embodiment collects a two-dimensional array of pixel amplitudes during each sampling period.

[0029] The spectrometer 220 is effectively an imaging or two-dimensional tunable wavelength bandpass filter. By repeatedly sampling and tuning to one wavelength or color and then the other, any number of wavelength-specific images of the sample can be collected and compared. The spectrometer or tunable filter may be arranged to select one or more specific bandpass wavelengths or reject specific wavelengths. The two-pixel size spectral image 225 may be collected by the CCD camera 230 to produce several spectral images 240 that may be sampled for each wavelength as a pixel data image 260.

[0030] Advantageous tunable wavelength bandpass filters may include birefringent crystals and polarizers that are variably spaced and/or rotationally tuned to select particular wavelengths and reject other wavelengths. A Liquid Crystal Tunable Filter (LCTF) is an advantageous device for this purpose because its birefringence can be tuned electronically. The liquid crystal filter may be nematic or smectic.

[0031] Birefringent materials have different refractive indices for light energy polarized along two orthogonal axes, sometimes referred to as the fast and slow axes. This has the effect of retarding light polarized along one axis compared to light polarized along the other axis. The degree of retardation depends on factors including the refractive index of the material and the thickness of the material and the path along which the light passes. The difference in the travel time of light polarized on one axis compared to the other axis is the time difference. When considering different wavelengths of light, a given time difference is converted to a phase angle difference for any given wavelength, but the phase angle difference is a different angle for two different wavelengths.

[0032] Retardation of light polarized in one axis more than light polarized in another axis may have the effect of changing or realigning the polarization state of the light. The degree of realignment likewise varies with wavelength. For these reasons, polarization and birefringence are useful considerations for wavelength bandpass filters.

[0033] The polarization state of light entering or exiting the birefringent crystal can be selectively controlled. If a planar polarizing filter (or "polarizer") is placed to randomly filter polarized light, for example on the input side of the crystal, the light passed may be limited to light energy more or less aligned to one or the other of the birefringent axes. If the light is aligned to one axis, then rotating the polarizer by 90 degrees causes the polarizer to be exclusively aligned to the other birefringent axis. Polarization filters may be used on the input and output sides of the birefringent crystal to select the characteristics of the input signal applied to the crystal or to selectively pass only as many outputs as are aligned to the output polarizer.

[0034] Assuming that the light is initially polarized to a given orientation angle, for example by an input polarizer, an oriented birefringent crystal at 45 degrees to the orientation angle of the polarizer splits the polarized light into equal vector components, one aligned to each of the fast and slow axes of the crystal. The retardation of the component on the slow axis relative to the component on the fast axis then changes the polarization state of the light by a rotation angle that is dependent on the wavelength. If the output polarizer is aligned at the proper angle for a given wavelength, that wavelength is transmitted while the other wavelengths are not.

[0035] By manipulating the arrangement of the polarizers and birefringent crystals, the distribution of light energy to the vector components aligned with the fast and slow axes of each stage of birefringent crystal can be controlled. The birefringence of each stage of the crystal is delayed by one of the two vector components relative to the other, causing a phase difference between the components on the fast and slow axes of the crystal. The phase delay between the two components corresponds to a change in the polarization orientation of the optical signal, i.e. a change in the angle of alignment of the vector sum of the two components relative to a reference angle. This transformation of polarization state or vector and angle is wavelength specific.

[0036] Light and polarizers and birefringent crystals that exist in these areas can be used as different approaches to wavelength bandpass filters. In the solution involving polarizers, a particular wavelength passing through the birefringent crystal may undergo a polarization change of a particular rotation angle. Other wavelengths are rotated in polarization by different amounts. By placing a polarizer aligned at this angle on the output of the birefringent crystal stage, only certain wavelengths pass. Successive stages increase the resolution of the filter.

[0037] Some types of wavelength-specific filters that may use liquid crystal tunable elements include Lyot, Solc, Evans and Fabry-Perot wavelength filter structures, as well as hybrids using combinations of related or additional elements. An LCTF Raman imaging system designed using a Lyot filter for selecting the wavelength spectrum 225 is shown in fig. 3. The Lyot architecture is an example and it should be understood that the present invention is applicable to other LCTF architectures including, but not limited to those described above.

[0038] In addition, a controller for controlling the LCTF detector and the shutter, as well as a CPU, display unit, keyboard and software are conventionally used with the system of fig. 3. Both the time to collect the CCD images and the time to tune the wavelength bandpass between images contribute to the time required to collect many wavelength-specific images. The image from the detector may be a 2D image (X, Y) given the selected wavelength (λ). The LCTF wide field of view can produce, for example, a 512 x 512 pixel image.

[0039] As mentioned, a conventional tunable filter may use a Lyot filter. A typical Lyot filter, shown in figure 3, comprises a set of birefringent crystals between two polarizers placed at 45 ° to the optical axis of the birefringent material. The bandpass wavelength is a function of the crystal thickness, among other factors. Incoming light from the polarizer is divided uniformly between normal and extraordinary polarizations by 45 ° alignment. Polarization propagates at different phase velocities due to birefringence. This alters the polarization orientation of the light energy as a function of wavelength. For a given thickness of birefringent crystal, only one wavelength (or a set of spaced wavelengths) is aligned to pass through the exit polarizer. Thus, the filter produces a set of comb frequencies. The separation between the bandpass wavelengths and the "comb teeth" depends on the length of the birefringent crystal. In liquid crystals, the effective optical length is adjustably variable for polarization components aligned to the extraordinary axis. In this way, the bandpass wavelength can be tuned.

[0040] Lyot filters use a plurality of birefringent crystals of different lengths, in particular R, 2R, 4R, etc., with a polarizer between each crystal. The Solc filter uses equal crystal thicknesses, only input and output polarizers (without interleaved polarizers) and relative angular orientations between the crystals, which equally divide the relative orientation between the input and output polarizers between the crystals.

[0041] In fig. 3, a sample 315, which may have a multi-wavelength Raman image, receives the emitted photons 310 to form scattered photons. Photons scattered by the sample enter the Lyot filter 300, which includes four polarizers 320 and three birefringent optical elements 330 that define successive stages. At each stage, the entrance side polarizer 320 acts as an optical filter to pass polarization-oriented light aligned with the polarizer and to block vertically-oriented light. The subsequent birefringent element 330 is oriented at an angle, in particular 45, to the previous polarizer 320. Thus, equal vector components of light passing through the polarizer are aligned with each of the ordinary and extraordinary axes of the birefringent element 330. The orientation of the optical axis 320 of the example Lyot structure of the wavelength band-pass filter is shown in fig. 3. Other configurations are also known.

[0042] The polarization components aligned with the ordinary and extraordinary axes of the birefringent element 330 propagate at different phase velocities due to the birefringence of the element 330. Also, the birefringent elements 330 of each stage have different thicknesses. By delaying the orthogonal components of the optical signal, the polarization orientation of the light is re-aligned to an angle that depends on the wavelength of the light. At the next polarizer encountered, only one set of wavelengths of light is aligned to pass through the next polarizer, which acts as the output polarizer or selector for the previous stage and as the input polarizer for the next stage. The thickness of the individual birefringent elements 320 and the alignment of the birefringent elements are selected so that each stage further identifies light of the same bandpass wavelength.

[0043] The bandpass wavelength is tunable by applying a control voltage 335, preferably comprising a liquid crystal, to the birefringent element 330. The effect of changing the birefringence of the liquid crystal is to shorten or lengthen the effective optical path encountered by the component of light aligned to the extraordinary axis, while keeping the effective length constant with respect to the ordinary axis. This is much the same as controllably adjusting the effective thickness of the birefringent element 330. Each birefringent liquid crystal element 330 of the Lyot filter is connected to a voltage source 335 for tuning the bandpass of the birefringent element 330. In the Lyot structure, the thickness is an integer multiple (R, 2R, 4R, etc.) and can be controlled in a coordinated manner, for example connected to the same control voltage source 335, in order to keep the effective thickness equal to the desired multiple. In other similar configurations, the thicknesses may be otherwise related (e.g., as in Solc configurations of equal thickness) and controlled so as to maintain a desired relationship, e.g., in parallel to the same drive voltage source.

[0044] In multispectral imaging, it is conventional to collect individual images, where the entire image is collected at one wavelength band pass, compared to other images at a different wavelength band pass. According to one embodiment of the invention, the new tunable filter is arranged to tune to different wavelengths across the X-Y image field. The filter may have one or more stages, at least one stage having a wedge-shaped liquid crystal cell or other shape in a tunable structure with a bandpass wavelength that is not uniform across the filter surface, examples being shown in fig. 4B, 4C, 6, 7, etc. In these structures, the wedge-shaped or similarly configured birefringent element has a thickness that varies across its operational region. The wedge-shaped birefringent element may be adjustable and optionally associated with a uniform (non-adjustable) retarder. Alternatively, a non-adjustable wedge birefringent retarder may be combined with an adjustable birefringent element having a uniform thickness. Different locations in the field are tuned to different bandpass wavelengths by a combination of controllable birefringence and fixed birefringence elements with uniformly varying thicknesses along the optical path.

[0045] Figure 4A schematically shows an element having a liquid crystal cell of uniform thickness and electronically controllable birefringence. The uniform thickness liquid crystal shown in fig. 4A may form a sub-unit or cell of a liquid crystal filter having multiple elements stacked along the optical path or arranged adjacent to each other in a pixel-like structure. Each subunit has an associated transparent plate 450, for example made of fused silica or glass. On the side of the plate 450 facing the liquid crystal cell 465, the glass plate 450 has a conductive coating (not shown) such as Indium Tin Oxide (ITO) which is nearly transparent but sufficiently conductive to apply an electric field to the liquid crystal material 465 in the element. The conductive surfaces are connected to wires that provide the drive voltage for each subunit by operating LCTF controller 470. Between the ITO transparent electrode and the liquid crystal 465 the plate 455 is further coated with an alignment layer. The alignment layer is typically physically treated by rubbing or polishing to induce a direction in which the molecules of the liquid crystal material tend to align. This orientation direction determines the director orientation of the liquid crystal and serves to orient the liquid crystal element 465 with respect to the transmitted polarized light.

[0046] Spacers 460 are provided to maintain the thickness of the regions between the alignment layers 455 occupied by the liquid crystal material 465. The spacers 460 are shown as spheres but may be of different shapes such as cylinders, etc. The spacers may be polymer or quartz glass of the type produced in large quantities having a relatively uniform size. Spacers 460 may be mixed into the liquid crystal material to provide a minimum thickness to which liquid crystal 465 may be compressed. The spacer material may be distributed around the periphery of the liquid crystal material in the cell (inside the glue bead) or throughout the liquid crystal material. The spacer may be used to keep the cell gap uniform (although it may alternatively be deliberately made slightly slanted to avoid interference fringes due to coherent laser light). The volume ratio of the spacer material to the liquid crystal material is sufficiently low to minimize the effect on light propagating through the liquid crystal material. When a liquid crystal material is contained between the alignment layers, molecules in the liquid crystal layer near the plate 450 are aligned parallel to the alignment direction of the plate 450 and the alignment layer 455. The liquid crystal cell may be configured to Electronically Control Birefringence (ECB). The liquid crystal may also be a nematic or smectic liquid crystal.

[0047]In contrast to the homogeneous liquid crystal cell of fig. 4A, fig. 4B schematically illustrates a wedge-shaped liquid crystal cell of an electronically controlled birefringent cell. In the embodiment of fig. 4B, spacers 460 and 461 have different dimensions such that plate 450 is not parallel and the entire unit has a wedge shape. I.e. the spacers of the support plate on opposite sides of the liquid crystal element have different thicknesses, as a result of which the retarder in fig. 4B defines a wedge having a thickness that varies in relation to the point at which light passes through the retarder element. Light passing through the retarder element will encounter different thicknesses at points closer to one spacer or the other, and will encounter linearly varying thicknesses across the distance between the spacers (since the plate is flat in this case). Thus, the delay is the relative diameter (d) of the spacer₁，d₂) And the effective refractive index (n) of the cell_eff) As a function of (c).

[0048]Fig. 4C is a schematic illustration of an optical device and an optical stage according to one embodiment of the disclosure. In the embodiment of fig. 4C, the optical device 400 is shown with three optical stages. At least one of the stages 450 is arranged to provide different effective retarder thicknesses across the X-Y operating region where light can pass through the stage. Stage 450 is shown having a uniform retarder 410 coupled to a wedge of liquid crystal 415. The wedge shaped liquid crystal 415 receives a complementary concave wedge shape 420 to form a geometric cube. The wedge shaped liquid crystal is shown with two spacers 417. Conventional liquid crystal materials may be used for this application. A voltage source 435 may be connected to the liquid crystal segments of stage 400 to fine tune the birefringence of stage 405. The uniform retarder 410 may be made of quartz, lithium niobate (LiNbO)₃) Or a polymeric material having the desired birefringent optical properties. The concave wedge 420 may be homogeneous optically clear glass or a polymeric material with similar optical properties.

[0049] The placement of the concave wedge is optional. Although in the example embodiment of fig. 4C, only one stage of filter 400 is shown having a tapered liquid crystal segment, it should be noted that the disclosure is not so limited and more than one stage may be constructed in accordance with the principles of the disclosure. Moreover, the principles of the disclosure are not limited to having three stages of filters and may include more (or fewer) stages than shown in FIG. 4C. Indeed, the principles illustrated in the example implementation of fig. 4C may be used with pixilated and/or columnar LCTFs. The number of liquid crystal cells may be a function of the application. For example, for a Lyot filter, each stage may typically contain one liquid crystal cell, while for an Evans-type filter, each stage may contain multiple liquid crystal cells.

[0050] In one embodiment, the disclosure relates to a multi-stage filter, wherein each stage comprises a wedge-shaped liquid crystal cell, and optionally, a complementary optically transparent wedge of glass. Also, the liquid crystal cell may be connected to a voltage source to enable further fine tuning of the cell. The filter can be tuned by using a voltage source and a controller.

[0051] In embodiments having wedge-shaped elements as shown in fig. 4 and 5 and stepped thickness elements as shown in fig. 6, there is a difference in the thickness of the operative retardation element along the light propagation axis Z for different points in the X-Y field across the X-Y image field. The thickness difference may occur in a controlled birefringence liquid crystal cell. The thickness difference may alternatively or additionally occur in a retarder of fixed birefringence used with another element controllable for tuning. In each of these cases, light passing through the element at a given point in the X-Y image field along a line parallel to the Z axis experiences a different phase delay than the phase delay of other points in the X-Y image field due to the thickness difference Z. As a wavelength bandpass filter, the result of the wedge element is that the center wavelength passed by the filter is different at different points across the X-Y field.

[0052] In the embodiment of fig. 4-6, the wedge shape has a thickness that varies linearly or stepwise from a minimum thickness at one end of one of the X or Y axes to a maximum thickness at the other end. The minimum thickness may taper to some non-zero minimum thickness or may taper to a sharp edge. It is also possible that the thickness may vary in both X and Y, for example with minimum and maximum thicknesses occurring at diagonal corners rather than adjacent corners. This configuration corresponds to rotating the wedge member 45 from the orientation shown. Other thickness variations are potentially applicable for varying the wavelength passband at different points in the field, such as conical, pyramidal, truncated conical or pyramidal, etc.

[0053] The thickness variation of the wedge shape according to this aspect should be distinguished from a technique of tilting a birefringent element used with a monochromatic (laser) light source as a method of preventing an edge effect of a monochromatic image due to an interference effect. According to the present invention, unlike the anti-fringe effect technique, thickness variations are introduced into the tunable or fixed birefringent element to allow the device to be tuned to different wavelengths at different X-Y locations on the image field at the same time. This is achieved in example embodiments using a continuous wedge adjustable or fixed retarder (potentially resulting in a wedge of the inclined surface or interface), or a stepped wedge (e.g., fig. 6).

[0054] Fig. 5 is a schematic illustration of an optical stage according to another embodiment of the disclosure. According to the example embodiment of fig. 5, the optical stage 500 includes a uniform liquid crystal 510 connected to a wedge retarder 515 and a concave wedge 520. Spacers 517 are shown along the axis of the liquid crystal 510 device to maintain a uniform thickness. As with the example embodiment of fig. 4, the concave wedge 520 is complementary to the wedge retarder 510, and may optionally be used. In addition, a voltage source 535 is provided to provide a programming voltage to the uniform liquid crystal segment 510.

[0055] In another embodiment, the disclosure is directed to a tunable filter comprising several uniform stages. Each uniformity stage can include a fixed wedge retarder and, optionally, a homogeneous optically transparent wedge of glass. The wedge of glass may be configured to complement the wedge of retarder such that when combined, the two form a three-dimensional rectangle. Each stage may also include a first polarizer that affects photons entering the retarder and a second polarizer that affects photons exiting the optional transparent glass. The tunable filter may include a stack of N stages. In another embodiment, a stage according to the principles disclosed herein may be used as part of a stack configured as a Lyot filter, a Solc filter, an Evan filter, or a hybrid filter. As known to those skilled in the art, a hybrid filter may comprise many stages (at least two stages). Some of the stages in the hybrid filter may include a Lyot filter, a Solc filter, an Evans filter or a Fabry-Perot interferometer.

[0056] Fig. 6 schematically illustrates a stepped LCTF device for use in unshaped Raman imaging or other chemical imaging applications, according to another embodiment of the disclosure. In the example implementation of fig. 6, stage 600 includes a uniform liquid crystal 610 in optical communication with a retarder 615 having a stepped profile (here a "stepped retarder"). The stepped retarder 615 receives a complementary concave wedge shape 620. As with the example embodiments of fig. 4 and 5, stage 600 may be connected to voltage source 635 for better optical modulation. The embodiment of fig. 6 may be modified to include a symmetric uniform retarder, a stepped liquid crystal and a complementary concave wedge shape. As before, the concave wedge shape may be homogeneous optically clear glass or a composition having similar optical properties.

[0057] In one embodiment, the tunable filter may be configured to include N stages. Each stage may have a convex stepped retarder and a complementary concave homogeneous transparent glass. The polarizer may cover each face of a three-dimensional rectangle formed by combining the retarder and the complementary glass. Multiple stages may be combined in order of increasing thickness to form a tunable filter. In this embodiment, the filter resolution is determined by the number of steps and the number of stages in the filter.

[0058]Fig. 7 is a schematic diagram of an LCTF device for Raman imaging or other chemical imaging applications in which thickness variations occur in the stripes, horizontal bands or depicted fringes extending across the X-Y field of the filter element. Fig. 7 shows a three-stage Lyot-type filter according to an embodiment of the invention, and is similar to the previous embodiments in that each stage may comprise a liquid crystal cell and an optional retarder element such as a birefringent crystal. Each stage is shown interposed between two polarizers (at the entrance and exit points of the stage). The birefringent crystal and liquid crystal have optical axes aligned at 45 to the orientation of the input polarizer (the initial polarizer does not)As shown in fig. 7). A conductive layer, typically indium tin oxide, may be provided on at least one glass substrate of the liquid crystal cell patterned with rows (or columns) of a given width. Furthermore, each row or column can be supplied with a respective control voltage source V₁～V_nConnected to the same or different control voltages. In one embodiment, the columns (or rows) of each stage are aligned with similarly positioned columns (or rows) of other stages. Referring to fig. 7, stage 710 is receiving a voltage source V₁，V_i…V_n. The filter may be used as a conventional tunable filter by setting the voltage to each column the same in the same stage, and the filter may be arranged to tune each column or row to a different voltage for providing a different wavelength bandpass. The successive stages 710, 720, 730 cooperate to pass the same wavelength through each aligned column (or row) in the stacked stage.

[0059]By applying different voltages across multiple columns (or rows) of liquid crystal cells of stage 710 and the associated columns (or rows) of the following stages 720, 730, such that substantially independent tunable filter structures are formed throughout the columns (e.g., Lyot structures having thicknesses R, 2R, 4R), different bandpass (λ) s₁-λ_n) May be determined in different columns (which is shown in fig. 8 and discussed further below). This transforms the imaging LCTF into a dispersive spectrometer. Although the first stage 710 is shown connected to a voltage source V₁～V_nThe principles of the disclosure are not so limited and stage 710 may be designed to receive only one voltage source. Also, the voltage sources may be applied along multiple columns of liquid crystal segments of stage 710. The application voltage source is not limited to only one stage of the filter, but may be designed such that each of the plurality of stages is driven by the voltage source. Multiple columns (or rows) may also be bound together such that a portion of the image received from the CCD is at wavelength 1, for example, and another portion of the image received from the CCD is at wavelength 2. Stage 710 may also include uniform retarders, stepped or wedge retarders, and complementary wedges as described above. A black matrix mask may be placed over the exit polarizer of the last stage to prevent light leakage at the inter-pixel or inter-column regions when the filter is not tunable.

[0060]Fig. 8 is a schematic diagram of a column LCTF device for Raman or other chemical imaging applications. Referring to fig. 8, the first stage 810 is connected to several voltage sources V₁-V_n. Voltage source V₁-V_nCan be operated independently of the others to power the optical electro-plating tuned liquid crystal cell of stage 810. This results in a dynamic LCTF with high spatial resolution and filters that can operate to differentiate according to image or wavelength. By providing different voltages across a column (or row) of a liquid crystal segment and the associated column (or row) in subsequent stages, the tunable filter can be reconfigured along the X-axis to provide a 1D spectrum (Y, λ) as a function of the excited state of the LCTF at a certain location X. The liquid crystals used in the column system of fig. 8 may be nematic or smectic. Although the embodiment of fig. 8 depicts only one of the three stages as receiving an independent voltage, the disclosure is not so limited and stages 820 and 830 may also be configured for column tuning.

[0061] The filter of fig. 8 can be coupled to a photon emission source, a photon detection source, an optical lens, and a processor to form a system that obtains spatially accurate wavelength resolved images of a sample having first and second dimensions. The photon detector may include a charge coupled device, a complementary metal oxide semiconductor, a charge injection device, an enhanced charge injection device, an electron multiplying charge coupled device, a silicon photodiode, a silicon avalanche diode, and a focal plane array. The photon emitting source may be a laser, a light emitting device or a fluorescent device.

[0062] Fig. 9 is a schematic diagram of an unshaped Raman imaging or other chemical imaging application system in accordance with another embodiment of the disclosure. Referring to fig. 9, an optical filter 900 receives scattered photons from an object (optionally a Raman object) 905 and produces a spectrum 910. The wedge shaped liquid crystal filter 900 may be connected to a voltage source 907. Applying voltage 907 to filter 900 changes the crystal delay at locations along filter 900. Filter 900 may produce a variable delay by applying voltages to filter 900 at various locations along the lateral length of the filter. In alternative embodiments, each of rows l through N may receive a different voltage.

[0063]In fig. 9, the wedge design can create a series of bandpass regions. Each bandpass region may allow a different wavelength (λ)₁-λ_n) In fig. 9 different positions of the filter 900 are passed. Similar to the example embodiments of fig. 4 and 5, the bandpass regions may be arranged horizontally in the X-direction. The different band pass regions allow filter 900 to function as a dispersive spectrometer. The resolution of the filter may be a function of the pixel size of the CCD camera and the wedge angle of each filter stage (if a multi-stage filter is used). The additional liquid crystal cells in each stage may add a tunable feature to the filter so that each band pass region of the filter can be tuned independently of the other regions. In contrast to the embodiments of fig. 4C and 5, where voltages are applied to the entire liquid crystal cell such that the retardation of each stage satisfies the Lyot filter structure (i.e., R, 2R, 4R, etc.), the embodiment of fig. 9 may be configured to have the filter at a different location along the X-direction (e.g., at location X, which is not shown)₁-X_n) Producing a series of band pass regions (λ)₁-λ_n). The retardation of the liquid crystal cell can be changed by changing the applied voltage, so that the liquid crystal cell functions as a variable retarder. Thus, at different positions X_i(not shown), the bandpass wavelength can be changed by changing the voltage to the liquid crystal cell. Specific wavelength lambda_iMay be combined, e.g., by the same wavelength λ_iThe pass-through region is constructed by a computer.

[0064] FIG. 10 is a schematic diagram of a compact imaging filter for a hand-held system using a tunable Fabry-Perot micro-electro-mechanical system. The sample 1005, shown here as numeral 12, is an opaque surface that is illuminated by light and reflectively scatters incident light or absorbs light energy and re-emits energy in all directions at a characteristic wavelength. The rays diverge from any point (x, y) on the sample 1005. An optical system, generally represented by lens 1009 in the figure, directs radiation from point (x, y) through one or more small 2D Raman imaging filter elements 1010, two of which are shown in the example, in direction 1012. Receiving optics, generally represented by lens 1011 in the drawing, are used to obtain a spatially accurate wavelength resolved image of the sample on the surface of detector 1020.

[0065] In fig. 10, the imaging filter comprises a series of filter elements having pairs of partially transparent parallel thin Si plates separated by air so that the pairs each define an optical cavity. The plates may be made by micromachining and may be moved by a microelectromechanical positioner (not shown) to adjust the thickness of the cavities between the plates to tune the resonant wavelengths of the cavities. Preferably, two or more pairs of substantially reflective (but less than 100% reflective) plates 1010 form the cavity of one or more Fabry-Perot interferometers. The plurality of cavities are arranged at the same cavity pitch. Rays oriented perpendicular to the plate at the resonant wavelength pass through, while other wavelengths are reflected back along the optical path.

[0066] Optical plate 1010 can be made from a planar Si wafer, and the wafer can be configured to form a support frame by micromachining steps, such as etching using chemical or ion beams. By means of further processing steps known to the person skilled in the art, connection elements are formed between the support frame and the substantially reflective thin Si plate (e.g. along the sides and corners of each Si filter element) in order to provide an electronically controlled positioning actuator for setting the cavity thickness. These actuators (not shown) move one or both optical plates 1010 defining the Fabry-Perot interferometer in pairs. The plate 1010 can be moved uniformly into and out of the plane of the respective support frame (i.e., in a direction parallel to the optical axis and perpendicular to the parallel plane of the optical plate) for setting the cavity thickness as indicated by the arrows.

[0067] In one embodiment, the actuator may be energized by a voltage source. Depending on the desired output, the actuator may be shifted between discrete tuning positions (cavity thicknesses) or can be adjusted to a desired point within the actuator displacement range. The set of planar structures containing the Si filter elements and the on-board actuators are stacked along the optical axis to form a Fabry-Perot imaging filter. In this stack, each parallel thin Si plate is parallel to the other plates and aligned so that all points lie in a direction parallel to the central optical axis.

[0068] The final number of plates may determine the Fabry-Perot etalon between pairs or may be odd if successive pairs of plates are also used as resonant cavities to define another pair. In this case, the pitch of the plates and the pitch between such successive pairs must each be controlled to use the same resonant pitch. Different combinations of MEMS actuator displacement between pairs of filter elements and different stack distances between sets of plates (to form stages) will allow narrow-passband transmission of light of selected wavelengths within the wavelength range to which the device is controllably tuned.

[0069] The filter wavelength λ is selected by varying the cavity or distance between each of the Fabry-Perot filter elements in a predetermined manner. The actuator displacement, the spacing between the plates, and the number of plates determine the wavelength range over which transmission is achieved when the actuator is changed (i.e., tuned), and the wavelength window (i.e., bandpass) over which light of wavelength λ is transmitted. The bandpass of such devices can be as narrow as 0.25nm (high resolution) or as high as 10nm (lower resolution). The wavelength range over which the device operates (i.e. filters light of different wavelengths) may be, for example, 400nm to 1800 nm. Design tradeoffs to achieve this performance are chosen to optimize the number of elements (cost and simplicity) and overall transfer function (optical efficiency) for any particular measurement requirement, such as Raman, fluorescence, VIS or NIR chemical imaging.

[0070] For Raman, fluorescent, visible or NIR operation, a specific arrangement of plate stacks may be used, with a predetermined set of actuator voltages known to provide the required wavelength filtering characteristics. For a particular mode of operation, the required set of voltages is then recalled by the computer and applied to the actuator to image each wavelength. Scanning a set of actuators and acquiring data over the entire image 1020 produces a wavelength-resolved spatially accurate image.

[0071] Preferably, a 2D image (X, Y) of the sample is generated on detector 1020. The tunable optical cavity produces an image at one wavelength at a time and is controllable to tune to two or more wavelengths and, optionally, selected wavelengths within a range. The tuning speed to change between cavity spacings to select an image at a new wavelength may be less than 1 second.

[0072] Optics 1009 and 1011 before and after the Fabry-Perot plate may be used to allow light scattered or emitted from sample 1005 to accurately and faithfully reproduce each (x, y) position of sample 1005 onto detector plate 1020 at (x ', y'). A particular detector pixel forms an image or spatially accurate representation corresponding to the location of a point in the sample image. It is possible that the transmitted light is discriminated without using the optical structures of the optical systems 1009 and 1011 including the lenses. For example, a stacked Fabry-Perot array, each etalon discriminating wavelengths due to the thickness of the cavity in the normal direction, has the property of selectively passing only normally directed rays, effectively collimating the light from the image and providing a spatially resolved image that is applied to the detector.

[0073] The embodiment of fig. 10 enables a subminiature high resolution Raman or fluorescence imaging device that can be selectively fine tuned to obtain Raman or fluorescence spectra corresponding to each spatial element of the sample. The individually addressable Fabry-Perot filter element 1010 may allow the acquisition of Raman or fluorescence spectra corresponding to spatial elements of the sample 1005. When detector 1020 is a CMOS detector, individual pixels can be individually sensed by tuning the wavelength applied to a particular pixel element, which is not possible in typical CCD detector devices. The use of such a Fabry-Perot imaging filter is novel because previous MOEMS-based Fabry-Perot filters could not perform imaging-only wavelength selection of the light source. A further advantage of the present invention is that the CMOS detector in this preferred embodiment can perform pixel selection to focus on only important pixels in the sample-thereby speeding up and simplifying the selection of the most important data from a particular range of samples.

[0074] The Fabry-Perot filter element not only transmits, but in alternative embodiments can reflect light to individual pixels of the CMOS sensor to form a single wavelength Raman or fluorescence imaging object 1020 of spatially accurate spectrally resolved pixels of the sample. That is, in alternative embodiments, one or more Fabry-Perot filter elements may be used as a reflective wavelength filter rather than a transmissive wavelength filter.

[0075] The wedge or stepped birefringent interference filters, MOEMS devices and dispersive spectrometers disclosed herein can be made very small and are particularly well suited for use in hand-held imaging systems. Also, the filter may be configured to operate in two modes: an imaging mode and a spectral mode. The tuning method may include line scanning in the 1D spectrum. Thus, at a certain position X, the sample can be scanned as a function of Y and λ. The tuning speed may be as low as about 20ms per wavelength or less than about 1 second per scan line.

[0076]In this regard, a "small" or "hand-held" or "portable" version should be considered to include a hand-held calculator, cell phone, PDA, etc. of similar size, i.e., a self-powered unit of a size that can be conveniently carried in a pocket or used entirely by hand. Preferably, the handheld device is approximately 36in³Or less (3 × 6 × 2in) and may be as small as 9in³(3 x 6 x 0.5in) or less and the optical path measured from the detector to the sample is about 2-4 inches.

[0077] In a handheld system using the tunable filter structure disclosed herein, a controller for controlling the tunable filter and the CCD may be included. The controller may be in the form of a processor programmed by software to communicate with an operator via a keyboard or display unit. The handheld system may also include a photon emitting source, a polarizing beam splitter, and a power source. The power source may be a battery. The photon emitting source may include a laser (for Raman scattering), an LED (for white light reflection applications or fluorescent emission), a near infrared source, a fluorescent source, or combinations thereof. The handheld device may also include one or more rejection filters for preventing the transmission source from interfering with the LCTF and the detector.

[0078] FIG. 11A is a schematic diagram of a hand-held chemical imaging threat assessor apparatus, according to one embodiment of the invention. Referring to the example embodiment of FIG. 11A, the CHITA device includes a miniaturized assembly effectively packaged in a portable compact form for handheld operation. The unit includes an illumination/excitation source (laser source 1110 and light emitting diode source 1105), a lens and reflective surface 1114, a scatter/emission/reflection light conditioning filter 1107, a polarizing beam splitter 1120, a combination of a filter 1124 (e.g., a fluorescent liquid crystal tunable Filter (FLC)) and a detection system 1126 (e.g., CMOS or CCD) for sample area screening and selective broadband detection, a filter 1130 (e.g., Raman liquid crystal tunable filters (RLC1 and RLC2)) and a second combination of detection subsystems 1132 (e.g., CMOS or CCD), control electronics 1134, a processor unit 1136 and a battery 1140. Imaging filter 1124(FLC) is a fluorescent LCTF. The imaging filter 1130(RLC1, RLC2) may be a Raman LCTF. An adjusting filter 1107, which may be a "notch filter," is positioned in front of polarizing beamsplitter 1120 to prevent laser light from overwhelming beamsplitter 1120 and subsequent detection systems 1124, 1130, and lenses 1122 and 1128 are positioned between beamsplitter 1120 and liquid crystals 1124 and 1130, respectively.

[0079] In the example device of fig. 11A, a laser 1110 provides a photon beam 1112 that can be used for narrow band excitation and analysis, including Raman analysis. A first filter/detection subsystem comprising Light Emitting Diodes (LEDs) may be used for screening purposes.

[0080] FIG. 12 is an auxiliary illumination source according to one embodiment of the disclosure. In fig. 12, the LEDs are configured in concentric rings 1200 perpendicular to the axis intersecting the sample. In other words, they form a ring 1200 and are positioned around the sample (e.g., 1105 of FIG. 11A) to illuminate the sample with photons. The embodiment of fig. 12 shows a diode illumination ring 1200 with LEDs 1205 and 1210 that can illuminate the sample using different wavelengths. In other words, the ring 1200 may include a similar series of different light emitting diodes for illuminating the sample to allow sample detection and identification. Additionally, sample illumination may enhance laser detection and identification. The LEDs 1205 and 1210 may be selected to operate in the visible, NIR or UV bands depending on the desired application. They may also be selected to include a mixture of different diodes selected for screening combinations of different chemical stations or biological agents. The illumination source of fig. 12 may also be configured as a fluorescent light source.

[0081] Referring again to FIG. 11A, the sample can be illuminated and analyzed using laser light from source 1110 and emitted photons from LED source 1105 substantially simultaneously or sequentially. Next, the photon beam 1118 scattered and emitted from the sample is collected by lens 1116 and reflected by surface 1114 into the wavelength imaging filter. Reflective face 1114 may be configured to allow laser light to pass through while reflecting scattered and emitted photon beam 1118. Illumination filter 1107 removes wavelengths of Raleigh scattered illumination light that can overwhelm the detector. Polarizing beam splitter 1120 splits one polarization of the transmitted and scattered light and allows it to be analyzed by two different filter/detector subsystems shown to the left and below beam splitter 1120. A CMOS (shown) or CCD detector (1132, 1126) may record the selected wavelength/filter signal read and analyzed by control module 1134 and analyzer. The analyzer may be software stored in the CPU that uses data stored in the device memory. Alternatively, the processor may be programmed with software to detect the chemical characteristics of the sample by comparing its spectrum to known spectra stored in the CPU database.

[0082] Fig. 11B shows a different side of the handheld device of fig. 11A. Referring to fig. 11B, the device 1100 includes a battery source 1140, a keyboard or other interface device 1152, and a screen 1150. Screen 1150 may be configured to display the resulting image, or spectrum and a bio-threat warning indication, when such a threat is identified. The visual indicator may be supplemented by an audible warning signal or other identification method. The keypad 1152 may be used for controlling and inputting data or for providing commands to the unit 1100. The device may also include one or more communication ports for electronic communication with other electronic equipment, such as servers, printers, etc.

[0083] The apparatus 1100 may be configured to illuminate or illuminate a sample and collect and analyze photons emitted or scattered by the sample. Analysis and identification can be achieved as a function of the wavelength emitted or scattered by the sample (i.e., the spectrum of the sample). Thus, a spectrum similar to that produced by a spectrometer can be obtained. In another embodiment, the principles disclosed herein are particularly advantageous because the apparatus 1100 is capable of simultaneously obtaining spatially resolved images and spectral identification of the sample. In another embodiment, the disclosure is directed to a portable system for obtaining spatially accurate wavelength resolved images of a sample having first and second spatial dimensions. The portable system may include a photon emission source that sequentially illuminates portions of the sample with a plurality of photons to generate photons scattered by the sample. The photon emission source may illuminate the sample along the first spatial dimension for each of a plurality of predetermined locations of the second spatial dimension. The system may also include an optical lens for collecting scattered photons to produce filtered photons therefrom, a dispersive spectrometer for determining a wavelength of some of the filtered photons, a photon detector for receiving the filtered photons and obtaining therefrom a plurality of spectra of the sample, and a processor for producing a two-dimensional image of the sample from the plurality of spectra.

[0084] Figure 11C is a schematic diagram of a handheld CHITA device according to another embodiment of the present invention. In particular, FIG. 11C shows that filters 1124 and 1130 are each an embodiment of a dispersive spectrometer. In another embodiment, at least one of the filters 1124 or 1130 is replaced with a dispersive spectrometer.

[0085] Fig. 13 schematically illustrates the operation of an example CHITA device. As shown, the sample is first illuminated 1305 by one or more photon sources (e.g., LED 1105 and laser 1110 in FIG. 11A). The next step 1310 is the collection and analysis of emitted and/or scattered photons from the sample. Once the wavelength and spatially resolved data/information is obtained, the data is directed to a CMOS or CCD detector in step 1315. This data is then processed by the CPU to perform various corrections on the raw data for detailed analysis. The calibration analysis may include calibration and baseline calibration 1320, separation 1325 of different spectral features (such as those disclosed in U.S. patent application 10/812,233, incorporated herein by reference as background information), searching local databases for possible spectral matches 1330 and identification 1335 of the sample. The analysis may be repeated for different spatial locations on the sample. The spectral data may be stored in CPU 1136 or compared to baseline data stored in on-board memory. Example processing steps allow detection and identification 1340 of compounds classified as biological or chemical threats.

[0086] In accordance with the foregoing embodiments, the pixilated or otherwise added portions of the tunable filter are independently tunable and may be tuned, either entirely independently or in a coordinated manner, to selectively filter data capture devices or data capture fields applied at different points in the image (e.g., the X-Y field of photosensitive devices or a row of photosensitive devices in a push-broom configuration). In a stepped thickness retarder implementation, for example, a series of pixel regions or bands may be tuned together so as to provide a series of increasing regions (at each step) tuned to successive wavelengths that differ according to the difference in retarder thickness from one step to the next. Similarly, in a wedge-shaped retarder embodiment, the tuning on the wedge selects a series of wavelengths within a range that vary continuously across the surface where the retarder thickness varies between its maximum and minimum thicknesses. In a fully pixelated filter, individual pixels may be tuned to a wavelength different from the tuned wavelengths of other possible adjacent pixels.

[0087] It is therefore an aspect of the present invention that different locations on the tunable filter region are simultaneously tuned to different wavelengths. This is a departure from the desired technique of collecting wavelength-specific light amplitude data at one wavelength over the entire filter area, and then continuing to collect the next set of data at the next wavelength until the entire spectrum is collected for each pixel location. However, the present invention provides increased speed and versatility by anticipating the need to collect a full spectrum (all wavelengths) for the entire tunable region before data analysis can be done.

[0088] It is possible to use the independent or stepped tuning capability of the present invention to collect full spectrum information for each pixel location. In that case, it is necessary to manage data collection to remember the tuning wavelength for which each light amplitude measurement is applicable.

[0089] It is also possible according to the invention to collect a series of different wavelength measurements from a sample at a time. This can be achieved, for example, by defocusing or otherwise applying the reflected light of the image diffusely over all adjustable positions in the array (e.g., wedges or series of steps of different retarder thicknesses) and accumulating wavelength data from the entire sample in a manner similar to the way light from a slit is applied to a spectrometer using a prism or grating. That is, a wedge or step or pixilated retarder scheme may operate to collect an average spectrum of the entire sample image. The same approach can also be used to collect an average spectrum over a selected portion of the image.

[0090] In one example, the tunable filter is controlled by a processor, such as processor 1136 in fig. 11A, and the processor also controls the continuous data acquisition mode. As one step, focused fluorescence imaging is used to collect images of the sample. One or more specific regions of the image may be distinguished by detection of a feature of interest, for example conversely at a certain characteristic wavelength. In the next mode, the region of interest, or alternatively scattered light from the entire sample, is optionally examined for specific wavelength relationships using very narrow bandwidth (but slow) Raman imaging. Since it is not necessary to collect full spectrum data for each pixel or other adjustable increment, sample analysis can be substantially faster than otherwise possible.

[0091] The wavelength dispersion capability of the wedge-shaped, stepped, and individually tunable retarders of the foregoing embodiments can thus be accomplished by collecting a full spectrum of pixels to accomplish spectral analysis, or by collecting an average spectrum of the image, or differential wavelength analysis where adjacent or other regions of the image are selectively tuned to different wavelengths.

[0092] Thus, in one embodiment, a hand-held detection system for threat detection or other applications may be constructed and controllably operated to use birefringent spectroscopic "agile" interference filter elements, i.e., elements that can be selectively tuned to one or more different wavelengths at a given time to accomplish fluorescence imaging, reflectance image collection, Raman image and Raman image averaging spectral collection and analysis. Any one or combination of the filter arrangements disclosed herein may be used in a handheld device. For example, with reference to fig. 11A, RLC1 and RLC2(Raman liquid crystal) may be wedge filters or tunable filters as described above. Similarly, FLC 1124 may include a dispersion filter as described above.

[0093] FIG. 14 provides packaging options according to one embodiment of the disclosure. Referring to fig. 14, device 1410 illustrates a handheld device that is applicable similar to that shown in fig. 11A and 11B. Handheld device 1410 includes a screen 1412 for displaying various information to the operator, a warning signal 1411, and a communication port 1413 for enabling data communication with other electronic devices. Device 1400 may be used for bio-threat detection because it may display images as well as text. The unit may be configured to identify a wide range of bio-threat materials determined by the size of the on-board library of bio-threat signatures contained in its memory. It may also be configured to communicate remotely with the master station using a wireless link to report important findings or update its library.

[0094] The hand-held air monitor 1420 is shown to include a port 1422. Finally, handheld surface inspection device 1430 is shown having a handle 1433, an LED source 1431 having a ring-like configuration, and a body 1432. Although not shown, the hand-held surface sensing device may also include a display, a keyboard, and one or more communication ports. The devices shown in fig. 11A, 11B and 14 illustrate that embodiments disclosed herein can be assembled and packaged in a handheld device for field use. It will be readily seen that such an apparatus is compact, although having a small optical path (measured between the sample and the detector) may be as effective as a bench-top unit. Devices 1420 and 1430 are particularly well suited for air monitoring or surface detection of bio-threats, respectively. Hand-held detector 1430 may be used to measure bio-threats on clothing or exposed body parts, which is particularly useful in battlefield or civilian environments.

[0095] While the example embodiment of fig. 14 is discussed with respect to bio-threat detection, such an apparatus is equally applicable to chemical warfare agent detection or hazardous substance monitoring. Another application of the detection device 1410 may include the detection and monitoring of chemical substances in the human body for medical purposes. A consumer device using the principles disclosed herein may also be configured for performing self-diagnostic tests that identify agents such as glucose, cholesterol, urea, hemoglobin, and alcohol.

[0096] Finally, handheld device 1440 represents an example representation of a lower cost, consumer oriented device with simplified operational controls and handset-like menu driven input. The cell may be programmed to detect a certain chemical depending on its target application.

[0097] An alternative embodiment of the hand held detector includes the use of a Raman microspectrometer as a dispersive filter for Raman scattering. To achieve the small size required for a portable hand-held detector, the micro-Raman sensor assembly can be constructed from semiconductor lithographic materials such as PMMA and x-ray lithographic processes known to those skilled in the art of semiconductor processing. Manufacture involves exposure of Polymethylmethacrylate (PMMA) provided with an X-ray mask to synchrotron radiation. Here, the exposure may be performed using an in-plane micro-optical system known to those skilled in the art and fabricated in batch mode via deep x-ray lithography. Subsequent steps may include development of the exposed PMMA and its removal, formation of the plating in the PMMA cavity, polarization of the combined materials, removal of the protected PMMA, and finally release of the resulting plated assembly. This procedure produced a PMMA grating that could be used as a small wavelength dispersive element for the CHITA handheld detector. According to the construction technique of a dispersive spectrometer, the grating is inserted into the optical path to spread the filtered wavelengths over the detector surface to detect a range of wavelengths from the sample.

[0098] The subminiature filter is characterized by the inherent optical properties of PMMA and Raman characteristics that do not substantially impair Raman detection of the bio-threat agent. Fig. 15 shows Raman spectra of PMMA collected using a high performance microscope glass optical system. FIG. 15 also shows the Raman spectrum of an anthrax (BG) collected using the same optical system. Finally, figure 15 shows the Raman spectra of BG after introducing a thin slab of PMMA into the laser illumination and collection optical path. As long as PMMA is located in the optical region where the light rays are parallel, it does not introduce significant background into the measured spectrum. That is, PMMA is an illumination optical path that does not prevent BG Raman spectral collection. The contribution to the Raman spectrum due to PMMA can also be used as an internal calibrator (intensity and wavelength) that can help with the automatic correction of instrument response and overall improved performance of the in-situ system. The instrument calibration allows compensation for instrument variations and quantitative analysis including laser line drift.

[0099] The following table shows non-proprietary example specifications for an implementation of a compact imaging spectrometer.

Specification of performance parameters

Laser excitation wavelength and bandwidth: 532 nm; less than 0.2nm

Imaging aperture: 0.5' or more

Field of view (angle incidence): +/-3 degree

Free spectral range: 500-750nm

The available Raman range: 350-3,200cm^-1

Resolution ratio: 0.25nm FWHM @500nm, < 10cm^-1

Non-peak rejection: 10,000: 1 Total energy

Transmission: the minimum is 30 percent

[00100] While the disclosure has been described using illustrative embodiments provided herein, it will be understood that the principles of the disclosure are not limited thereto and may include modifications and substitutions thereto.

Claims

1. In an optical filter comprising a plurality of filter stages, wherein a first stage of the plurality of stages comprises a retarder element and a liquid crystal cell, each of the retarder element and the liquid crystal cell having a uniform thickness, the improvement comprising replacing the first stage with a second stage having the retarder element and the liquid crystal cell, wherein the retarder element or the liquid crystal cell does not have a uniform thickness.

2. The filter of claim 1, wherein the elements and cells of the first stage have the same thickness.

3. The filter of claim 1, wherein the elements of the second stage do not have a uniform thickness.

4. The filter of claim 1, wherein the elements of the second stage are wedge-shaped.

5. The filter of claim 1, wherein the elements of the second stage have a stepped wedge shape.

6. In an optical filter comprising a plurality of filter stages, wherein a first stage of the plurality of stages comprises a retarder element and a liquid crystal cell, each of the retarder element and the liquid crystal cell having a uniform thickness, the improvement comprising replacing the first stage with a second stage having the retarder element and the liquid crystal cell, wherein said cell comprises:

a pixel-like liquid crystal structure for filtering photons;

a voltage source operatively connected to the pixelated liquid crystal structure; and

means for controlling an amount of voltage applied by the voltage source to some of the pixels;

wherein the liquid crystal is substantially uniform in thickness in the direction of photon passage.

7. In an optical filter comprising a plurality of filter stages, wherein a first stage of the plurality of stages comprises a retarder element and a liquid crystal cell, each of the retarder element and the liquid crystal cell having a uniform thickness, the improvement comprising replacing the first stage with a second stage having the retarder element and the liquid crystal cell, wherein said cell comprises:

a column-type liquid crystal structure for filtering photons;

a voltage source operatively connected to the column liquid crystal structure; and

means for controlling an amount of voltage applied by the voltage source to some of the column pillars;

8. An optical filter comprising:

a first filter stage comprising a first retarder element and a first liquid crystal cell; and

a second filtering stage for the second signal to be filtered,

wherein the first element or the first cell does not have a uniform thickness.

9. The filter of claim 8, wherein the second stage includes a second retarder element having a first uniform thickness and a second liquid crystal cell having a second uniform thickness.

10. The filter of claim 9, wherein the first uniform thickness is substantially the same as the second uniform thickness.

11. The filter of claim 8, wherein the first element does not have a uniform thickness.

12. The filter of claim 11 wherein the first element is wedge-shaped.

13. A filter according to claim 11, wherein the first element has a stepped wedge shape.

14. The filter of claim 8 wherein the first and second stages comprise birefringent interference filters.

15. The filter according to claim 8 wherein the first and second stages comprise Evans split element filters.

16. The filter of claim 8 wherein the first and second stages comprise Solc filters.

17. The filter of claim 8 wherein the first and second stages comprise Lyot filters.

18. The filter according to claim 17 comprising an Evans split element filter.

19. The filter of claim 8 wherein said first element comprises quartz.

20. The filter of claim 8 wherein said first element comprises lithium niobate.

21. The filter of claim 8, wherein the filter is a tunable filter.

22. The filter according to claim 8, wherein the filter comprises a polarizer.

23. The filter of claim 8, wherein the filter is pixellated.

24. The filter of claim 8, wherein the filter is columnar.

25. The filter of claim 8, wherein the first and second filtering stages define a hybrid filter.

26. The filter of claim 8 wherein at least one of the first and second stages is selected from the group consisting of a Lyot filter, an Evans filter, a Fabry-Perot filter and a Solc filter, and the combination of the first and second stages determines the hybrid filter.

27. A method of manufacturing an optical filter comprising the steps of:

providing a first filter stage comprising a first retarder element and a first liquid crystal cell, wherein the first element or the first cell does not have a uniform thickness;

providing a second filtering stage; and

the first filtering stage is aligned with the second filtering stage.

28. The method of claim 27, wherein the first and second stages comprise Lyot filters and comprising the steps of:

providing an Evans split element filter; and

the Evans split element filter is aligned with the Lyot filter.

29. The method of claim 27, wherein the first and second stages comprise Lyot filters and comprising the steps of:

providing a Fabry-Perot filter; and

the Fabry-Perot filter is aligned with the Lyot filter.

30. The method of claim 73 wherein the first and second stages comprise Lyot filters and comprising the steps of:

providing a Solc filter; and

the Solc filter is aligned with the Lyot filter.

31. The method of claim 27, wherein the optical filter is a hybrid filter and at least one of the first or second stages is selected from the group consisting of a Lyot filter, an Evans filter, a Solc filter, and a Fabry-Perot filter.

32. An optical filter for passing photons, the filter comprising:

a second filtering stage for the second signal to be filtered,

wherein the first element comprises an input face and an output face, an

Wherein one of the faces of the first element is oriented not substantially perpendicular to the trajectory of photons passing through the filter.

33. The filter of claim 32 wherein said second stage comprises a second retarder element having a first uniform thickness and a second liquid crystal cell having a second uniform thickness.

34. The filter of claim 33, wherein the first uniform thickness is substantially the same as the second uniform thickness.

35. The filter of claim 32 wherein the first element is wedge-shaped.

36. The filter of claim 32 wherein the first element has a stepped wedge shape.

37. The filter of claim 32 wherein the first and second stages comprise birefringent interference filters.

38. The filter according to claim 32 wherein the first and second stages comprise Evans split element filters.

39. The filter of claim 32 wherein the first and second stages comprise Solc filters.

40. The filter of claim 32 wherein the first and second stages comprise Lyot filters.

41. The filter according to claim 40 comprising an Evans split element filter.

42. The filter of claim 32 wherein said first element comprises quartz.

43. The filter of claim 32 wherein said first element comprises lithium niobate.

44. The filter of claim 32 wherein the filter is a tunable filter.

45. The filter according to claim 32 wherein the filter comprises a polarizer.

46. The filter of claim 32 wherein the filter is pixellated.

47. The filter of claim 32 wherein the filter is columnar.

48. An optical filter comprising:

a second filtering stage for the second signal to be filtered,

wherein the first element comprises an input face and an output face, an

Wherein the faces of the first elements are not oriented substantially parallel to each other.

49. The filter of claim 48 wherein said second stage comprises a second retarder element having a first uniform thickness and a second liquid crystal cell having a second uniform thickness.

50. The filter of claim 49, wherein the first uniform thickness is substantially the same as the second uniform thickness.

51. The filter of claim 48 wherein the first element is wedge-shaped.

52. A filter according to claim 48 wherein the first element has a stepped wedge shape.

53. The filter according to claim 48 wherein the first and second stages comprise birefringent interference filters.

54. The filter according to claim 48 wherein the first and second stages comprise Evans split element filters.

55. The filter of claim 48 wherein the first and second stages comprise Solc filters.

56. The filter of claim 48 wherein the first and second stages comprise Lyot filters.

57. The filter according to claim 48 comprising an Evans split element filter.

58. The filter according to claim 48 wherein said first element comprises quartz.

59. The filter of claim 48 wherein said first element comprises lithium niobate.

60. The filter of claim 48 wherein the filter is a tunable filter.

61. The filter according to claim 48 wherein the filter comprises a polarizer.

62. The filter of claim 48 wherein the filter is pixilated.

63. The filter according to claim 48 wherein the filter is columnar.

64. The filter according to claim 48 wherein the optical filter is a hybrid filter and at least one of the first or second stages is selected from the group consisting of a Lyot filter, an Evans filter, a Solc filter and a Fabry-Perot filter.

65. A system for obtaining a spectrum of a sample, comprising:

a photon emission source for illuminating the sample with a plurality of photons to generate photons scattered by the sample;

an optical lens for collecting scattered photons;

a filter for receiving the collected scattered photons and providing filtered photons therefrom; and

a photon detector for receiving the filtered photons and obtaining therefrom a spectrum of the sample,

wherein the filter comprises:

a second filtering stage for the second signal to be filtered,

wherein the first element or the first cell does not have a uniform thickness.

66. The system of claim 65, wherein the second stage comprises a second retarder element having a first uniform thickness and a second liquid crystal cell having a second uniform thickness.

67. The system of claim 66, wherein the first uniform thickness is substantially the same as the second uniform thickness.

68. The system of claim 65, wherein the first member does not have a uniform thickness.

69. The system of claim 68, wherein the first member is wedge-shaped.

70. The system of claim 68, wherein the first element has a stepped wedge shape.

71. The system of claim 65, wherein the first and second stages comprise birefringent interference filters.

72. The system according to claim 65 wherein the first and second stages comprise Evans split element filters.

73. The system of claim 65, wherein the first and second stages comprise Solc filters.

74. The system of claim 65 wherein the first and second stages comprise Lyot filters.

75. The system according to claim 74 including an Evans split element filter.

76. The system of claim 65, wherein the first element comprises quartz.

77. The system of claim 65, wherein said first element comprises lithium niobate.

78. The system of claim 65, wherein the filter is a tunable filter.

79. The system of claim 65, wherein the filter comprises a polarizer.

80. The system of claim 65, wherein the filter is pixilated.

81. The system of claim 65, wherein the filter is columnar.

82. The system of claim 65, wherein said photon detector is selected from the group consisting of a charge coupled device, a complementary metal oxide semiconductor, a charge injection device, an enhanced charge injection device, an electron multiplying charge coupled device, a silicon photodiode, a silicon avalanche diode, and a focal plane array.

83. The system of claim 65, wherein the photon emitting source is a laser.

84. The system of claim 65, wherein the photon emitting source is a light emitting diode.

85. A system as recited in claim 84, wherein said light emitting diodes are arranged in a ring.

86. The system of claim 85, wherein the light emitting diode is a plurality of light emitting diodes.

87. The system of claim 84, wherein the light emitting diode is a plurality of light emitting diodes, wherein one of the plurality of light emitting diodes emits photons at a wavelength that is different from a wavelength of photons emitted by another of the plurality of light emitting diodes.

88. The system of claim 87, wherein the photons emitted by one of the plurality of light emitting diodes have a wavelength in the ultraviolet wavelength range.

89. The system of claim 87, wherein the photons emitted by one of the plurality of light emitting diodes have a wavelength in the near infrared wavelength range.

90. A system in accordance with claim 65, wherein said spectra are obtained over a predetermined period of time and said sample is stationary during said predetermined period of time.

91. The system of claim 65, wherein the scattered photons comprise photons emitted by the sample.

92. A system according to claim 65 wherein the spectra are Raman spectra.

93. The system of claim 65, wherein the system is a portable system.

94. The system of claim 65, wherein the system is a handheld system.

95. A system according to claim 65 wherein the spectra are Raman spectra.

96. The system of claim 65, wherein the spectrum is a spatially accurate wavelength resolved image.

97. A system in accordance with claim 65, wherein said spectra are obtained over a predetermined period of time and said sample is stationary during said predetermined period of time.

98. The system of claim 65, wherein the scattered photons comprise photons emitted by the sample.

99. The system of claim 65, wherein some of the filtered photons each have a wavelength within a predetermined wavelength band.

100. The system of claim 65, wherein the first filter element comprises a liquid crystal.

101. The system of claim 65, wherein the retarder element comprises quartz.

102. The system of claim 65, wherein said retarder element comprises lithium niobate.

103. The system of claim 65, wherein the filter is a hybrid filter and at least one of the first or second filter stages is selected from the group consisting of a Lyot filter, an Evans filter, a Solc filter, and a Fabry-Perot filter.

104. A method of obtaining a spectrum of a sample, comprising:

providing a sample;

illuminating the sample with a plurality of photons to produce photons scattered by the sample;

collecting scattered photons;

receiving the collected scattered photons using a filter and providing filtered photons therefrom; and

receiving the filtered photons and obtaining therefrom a spectrum of the sample,

wherein the filter comprises:

a second filtering stage for the second signal to be filtered,

wherein the first element or the first cell does not have a uniform thickness.

105. The method of claim 104 wherein the first and second stages comprise Lyot filters and including the steps of:

providing an Evans split element filter; and

the Evans split element filter is aligned with the Lyot filter.

106. The method of claim 104 wherein the first and second stages comprise Lyot filters and including the steps of:

providing a Fabry-Perot filter; and

the Fabry-Perot filter is aligned with the Lyot filter.

107. The method of claim 104 wherein the first and second stages comprise Lyot filters and including the steps of:

providing a Solc filter; and

the Solc filter is aligned with the Lyot filter.

108. The method of claim 104, wherein the optical filter is a hybrid filter and at least one of the first or second stages is selected from the group consisting of a Lyot filter, an Evans filter, a Solc filter, and a Fabry-Perot filter.

109. An optical filter comprising:

a first liquid crystal cell including a first region having a first photon retardation value and a second region having a second photon retardation value;

a second liquid crystal cell including a first region having the first photon delay value and a second region having the second photon delay value;

wherein each of the first and second delay values is determined by a first excitation source and a second excitation source in communication with each of the first and second regions of the first and second cells, respectively.

110. The optical filter of claim 109 wherein the first region of the first stage is substantially aligned with the first region of the second stage.

111. The optical filter of claim 109 wherein the first element and the second element form one of a Lyot filter, an Evans split element filter, a Solc filter, or a hybrid filter.

112. An optical filter comprising:

a pixellated liquid crystal stage for filtering photons;

a voltage source operatively connected to the liquid crystal pixels; and

113. The optical filter of claim 112 wherein the amount of voltage applied to some of the first group of pixels is substantially the same.

114. The optical filter of claim 112 wherein the amount of voltage applied to each of the pixels is substantially the same.

115. An optical filter comprising:

a column type liquid crystal stage for filtering photons;

a voltage source operatively connected to the liquid crystal column; and

116. The optical filter of claim 115 wherein the amount of voltage applied to some of the first group of the column posts is substantially the same.

117. The optical filter of claim 115 wherein the amount of voltage applied to each of said columns is substantially the same.