Automated portable raman microscope

Abstract

An automated portable Raman microscope combines the simplicity, ruggedness, and selectivity of Raman spectroscopy with the sensitivity of single-particle identification. It includes a visible camera, controllable illumination, and a moving optical head that together acquire images of the sample at different focal planes and under different lighting conditions. An internal processor automatically identifies particles or other targets in these images for Raman analysis, e.g., shifted excitation difference Raman spectroscopy. The moving optical head scans the Raman pump beam across the sample to different targets. A built-in user interface displays the results of the Raman analysis superimposed on visual images of the sample. The Raman microscope's simplicity, reliability, and low maintenance make it easy to use for a wide variety of applications, including analyzing and identifying particles and residues, linking chemical and biometric evidence, and forensics analysis.

Description

Attorney Docket No. PEND-014WO01Automated Portable Raman MicroscopeCROSS-REFERENCE TO RELATED APPLICATION(S)

[0001] This application claims the priority benefit, under 35 U.S.C. 119(e), of U.S. Application No. 63 / 689,761, filed September 1, 2024, which is incorporated herein by reference in its entirety for all purposes.BACKGROUND

[0002] Raman microscopes, first developed in the mid-1970s, enable chemical analysis of microscopic samples. Traditionally, these systems are benchtop instruments designed for laboratory use by expert users, primarily scientists or specialized technicians. Commercial Raman microscopes are often generalist instruments intended to cover a wide range of applications, offering various customization options such as different laser sources, spectrometers, gratings, and charge-coupled devices (CCDs). These features allow customization for specific applications, such as using a shorter laser wavelength for low signal detection or a longer wavelength to mitigate high laser-induced fluorescence. Higher-end instruments may feature specialized objectives, vibration isolation tables, and regular alignment of optical components. The size, complexity, and cost of these instruments generally restrict their use to laboratory environments, handled by well-trained professionals.

[0003] However, there are numerous applications where chemical analysis at the microscopic level would be beneficial outside the laboratory. These include scenarios where trace chemicals should be analyzed in the field to identify detailed chemical compositions or detect possible contamination. In forensic science, for example, identifying trace materials at the singleparticle level, directly on surfaces or objects, or collected on swabs, could aid in detecting explosives, illicit substances, or other hazardous chemicals.

[0004] Field chemical analysis is often performed using mass spectrometry (MS), but field MS systems are typically bulky and power-intensive, requiring heating of internal elements (e.g., sample desorption, gas chromatography (GC) column). They are susceptible to poisoning or saturation from excessive sample input, can be damaged by corrosive samples, and are affected by ambient chemicals. These systems also demand extensive training, frequent maintenance, and regular replacement of consumables and internal parts. While MS-based systems are the gold standard for laboratory chemical analysis, fieldable MS systems suffer from significantAttorney Docket No. PEND-014WO01 performance trade-offs, including limited selectivity, frequent false positives, and restricted chemical library sizes. (As the term is used herein, a “fieldable” instrument is an instrument that is both compact and rugged enough to be transported outside a laboratory environment and simple enough to be used by personnel collecting samples in their line of duty, rather than specially trained chemical analysis technicians or scientists.)

[0005] Conversely, handheld Raman spectroscopy systems have proven effective for chemical identification in the field. These instruments are compact, reliable, low-maintenance, and usable by non-experts with minimal training. Recent advancements, such as the Pendar XI 0 commercialized by Pendar Technologies, Inc., address issues related to laser safety and the risk of igniting dark or sensitive materials and mitigate the effects of laser-induced fluorescence using shifted excitation difference Raman spectroscopy (SERDS). However, these systems are currently limited to analyzing visible amounts of material.SUMMARY

[0006] An inventive automated portable Raman microscope combines the benefits of handheld Raman systems with the enhanced sensitivity offered by Raman microscopes. It integrates design elements from handheld Raman devices, such as compact fixed grating spectrometers, semiconductor laser sources in telecom-style packages, rugged mechanical construction, embedded electronics, and automated measurement routines. This approach yields an automated portable Raman microscope that addresses the limitations of current laboratorybound Raman microscopes and offers significant advantages for in-field applications.

[0007] An inventive automated portable Raman microscope can be compact and portable, making it suitable for field deployment. It can be transported without needing to be recalibrated or re-aligned: the mechanical and optical components are configured to ensure that the instrument remains aligned and calibrated even after transportation, reducing or eliminating detailed re-calibration or re-alignment before use. An inventive automated portable Raman microscope may be calibrated and / or aligned from time to time using a fast, automated calibration process as described below.

[0008] An inventive automated portable Raman microscope can be easy to use. It can measure samples automatically and present high-level, directly actionable results rather than (or in addition to) raw data, streamlining the decision -making process in the field. These automationAttorney Docket No. PEND-014WO01 features enable non-expert users to operate it effectively. It can also identify a wide range of chemicals encountered in the field, supported by a large and comprehensive chemical library.

[0009] An inventive automated portable Raman microscope is safe to use, posing minimal risk of sample destruction, thereby allowing subsequent analysis by other instruments if necessary. If the sample is damaged, the damage can be detected visually or by noting the increase in background signal in the Raman spectrum from blackbody radiation. The damage is localized to the laser spot, and the laser power can be reduced to avoid damaging to the rest of the sample.

[0010] All combinations of the foregoing concepts and additional concepts discussed in greater detail below (provided such concepts are not mutually inconsistent) are contemplated as being part of the inventive subject matter disclosed herein. In particular, all combinations of claimed subject matter appearing at the end of this disclosure are contemplated as being part of the inventive subject matter disclosed herein. Terminology explicitly employed herein that also may appear in any disclosure incorporated by reference should be accorded a meaning most consistent with the particular concepts disclosed herein.BRIEF DESCRIPTIONS OF THE DRAWINGS

[0011] The skilled artisan will understand that the drawings primarily are for illustrative purposes and are not intended to limit the scope of the inventive subject matter described herein. The drawings are not necessarily to scale; in some instances, various aspects of the inventive subject matter disclosed herein may be shown exaggerated or enlarged in the drawings to facilitate an understanding of different features. In the drawings, like reference characters generally refer to like features (e.g., functionally similar and / or structurally similar elements).

[0012] FIG. 1 A illustrates a stand-alone, fieldable Raman microscope.

[0013] FIG. IB illustrates an example user interface (UI) showing results acquired with the fieldable Raman microscope of FIG. 1A.

[0014] FIG. 1C illustrates the fieldable Raman microscope system of FIG. 1A with legs extended (left), mounted on a tripod (center), and mount on a stereo-microscope mount with height adjustment (right).

[0015] FIG. ID is a block diagram of the fieldable Raman microscope of FIG. 1 A including a moving optical subassembly.Attorney Docket No. PEND-014WO01

[0016] FIG. 2A is a block diagram of a fieldable Raman microscope with a motorized sample stage.

[0017] FIG. 2B is a block diagram of a fieldable Raman microscope with a motion controller and actuator moving the entire fieldable Raman microscope.

[0018] FIG. 3 A is a perspective view of a fieldable Raman microscope a stage that moves the entire enclosure up and down with respect to a fixed sample.

[0019] FIG. 3B is a block diagram of the fieldable Raman microscope of FIG. 3 A.

[0020] FIG. 4 is a plot of intensity versus Stokes shift for a shifted excitation difference Raman spectroscopy (SERDS) measurement of acid blue 9 dye with an inventive fieldable Raman microscope.

[0021] FIG. 5A is a side view of a ring LED illuminator suitable for illuminating a sample for analysis by an inventive fieldable Raman microscope.

[0022] FIG. 5B is a plan view of the ring LED illuminator of FIG. 5 A.

[0023] FIG. 5C is a plan view of the ring LED illuminator of FIG. 5 A in a ring / 4 illumination configuration.

[0024] FIG. 5D is a side view of the ring LED illuminator of FIG. 5 A in a ring / 4 illumination configuration.

[0025] FIG. 6A is a a z-flattened image acquired by an inventive fieldable Raman microscope of dimpled acrylonitrile butadiene styrene (ABS) under full ring illumination.

[0026] FIG. 6B is a composite image acquired by imaging dimpled ABS with an inventive fieldable Raman microscope under ring / 4 illumination and flattening in the z dimension.

[0027] FIG. 7A is a z-flattened image acquired by an inventive fieldable Raman microscope flattened images of Velcro® hook-and-loop fasteners under full ring illumination.

[0028] FIG. 7B is a composite image of Velcro® hook-and-loop fasteners acquired with an inventive fieldable Raman microscope under ring / 4 illumination and flattening in the z dimension.

[0029] FIG. 8A shows full-color, color-flattened images (left) of red, green, and blue buttons on a remote control synthesized from monochromatic images of the remote control under blue (right, top), red (right, middle), and green (right, bottom) illumination.Attorney Docket No. PEND-014WO01

[0030] FIG. 8B shows a full-color, color-flattened image (left) of particles of Goody’s powder on a region of the red button of the remote control of FIG. 8 A.

[0031] FIG. 9A illustrates a camera and image processing electronics of an inventive fieldable Raman microscope.

[0032] FIG. 9B illustrates a method of z-stack image processing performed by an inventive fieldable Raman microscope.

[0033] FIG. 10 illustrates a particle identified for analysis by an inventive fieldable Raman microscope.

[0034] FIG. 11 A shows images (left) and Raman spectra (right) of a KCIO3 particle on a white polyethylene terephthalate (PET) substrate acquired with an inventive fieldable Raman microscope.

[0035] FIG. 1 IB shows an image (upper left) and Raman spectra of a KCIO3 particle together with a white PET substrate acquired with an inventive fieldable Raman microscope (upper right) and the Raman spectrum of pure KCIO3 (lower left).

[0036] FIG. 11C is a plot of signal-to-noise ratio (SNR) versus acquisition time for Raman spectra of a KCIO3 particle with the substrate (upper trace) and with the substrate removed (lower trace).

[0037] FIG. 12A shows a crosshair calibration sample for calibrating the Raman subsystem with respect to the camera subsystem in an inventive fieldable Raman microscope.

[0038] FIG. 12B shows a process for making the crosshair calibration sample of FIG. 12 A.

[0039] FIGS. 13A-13D illustrate an automatic calibration process executed by an inventive fieldable Raman microscope using the crosshair calibration sample of FIG. 12 A.

[0040] FIGS. 14A-14F show different views of a rack-and-pinion locking mechanism for securing stages in a field Raman microscope during transport.

[0041] FIGS. 15A-15C show a sampling accessory suitable for use with a fieldable Raman microscope.

[0042] FIG. 16 illustrates a heterogenous sample analysis process executed by a fieldable Raman microscope.

[0043] FIG. 17Ais an image of multiple particles of Goody’s powder identified with a fieldable Raman microscope.Attorney Docket No. PEND-014WO01

[0044] FIG. 17B shows closeups of one of the particles in FIG. 17 A.

[0045] FIG. 17C shows different spectral components of the particle in FIG. 17B identified with the fieldable Raman Microscope.

[0046] FIGS. 18A-18D show matches for different spectral components of a Raman spectrum for Goody’s powder measured with a fieldable Raman microscope.DETAILED DESCRIPTION

[0047] FIGS. 1A-1E illustrate an inventive fieldable Raman microscope 100 that can both acquire a visible image of a sample and run a Raman analysis of specific points within the field of view. The Raman microscope 100 has a working distance of between 1 cm and 20 cm. Thanks to this short working distance and a large optical aperture (e.g., between 2.5 cm and 5.0 cm), the Raman microscope 100 can collect Raman signals over a large numerical aperture, which results in high collection efficiency. The optical front-end can also be used to image the sample at shorter wavelengths (e.g., in the visible spectral range) onto an internal image sensor (e.g., a CCD or CMOS sensor).

[0048] The Raman microscope 100 includes or can be coupled to a user interface (UI) 110, such as a touchscreen or liquid-crystal display, as shown in FIGS. IB and 1C. FIG. IB also illustrates a possible result screen 111 — here, of a fingerprint on a car door handle — with a visible image of the area of interest, captured with a camera operating in the visible spectral range, and the results of Raman-based chemical identification realized at a discrete set of points across the area of interest. The locations of these points (indicated in FIG. IB as the areas corresponding to “No match”, “RDX”, “Sugar”, etc.) can be decided automatically by image analysis. For example, particles to be analyzed can be located on the image based on brightness, shape (boundaries), polarization response (if polarized light is used for illumination and / or a polarization analyzer is used in front of the camera), or other methods. Random sampling or systematic analysis of all points located on a predefined grid or set of points can also be realized.

[0049] As shown in FIGS. 1 A and 1C, the Raman microscope 100 includes a housing 130 with one or more mounting brackets / options, including independently adjustable / retractable or foldable feet 132 (shown extended at left in FIG. 1C); a tripod mounted system using a standard 14-20 threaded hole 134 or other mechanical interface (FIG. 1C, center); a microscope or stereomicroscope objective (visible / NIR optics 150, described below) with included heightAttorney Docket No. PEND-014WO01 adjustment (FIG. 1C, right). This mount can fit on the standard 3” (7.5 cm) cylinder mount of most commercial microscope mounts. The housting 130 is also rugged, with one or more shock absorbers 138 to absorb and dampen shocks that might otherwise damage the Raman microscope 100 or misalign its components.

[0050] FIG. ID illustrates the fieldable Raman microscope’s internal components, including a laser source 102 (e.g., a 785 nm dual -wavelength laser source) and spectrometer 106 coupled to a moving optical subassembly 140 via optical fibers 104 and 108, respectively. The laser source 102, spectrometer 106, and moving optical subassembly 140 are controlled by internal (and / or external) electronics 112, such as a processor, controller, and / or other circuitry. Power management circuitry 114 powers the Raman microscope 100 with electrical power from an external source, such as a wall plug, and / or an internal battery.

[0051] The moving optical subassembly 140 includes a Raman probe 142, beam steering device 144 (e.g., a microelectromechanical systems (MEMS) scanning mirror), near-infrared (NIR) optics 146, dichroic beam splitter 148, visible / NIR optics 150, visible optics 152, and (visible) camera or image detector 154. The entire optical subassembly 140 may be a rigid subsystem to maintain proper referencing between the camera image and the Raman system’s 3D point of focus. As described in greater detail below, the Raman microscope 100 tracks the Raman system’s 3D point of focus with respect to the images acquired by the camera 154 so that it can bring imaged particles into proper aim and focus for Raman measurement. In other examples, the camera 154 can be moved or adjusted with respect to the Raman system for bringing particles into proper aim and focus for imaging and / or Raman measurement.

[0052] In operation, light-emitting diodes (LEDs) 160 in a ring around the microscope’s external aperture illuminate the fixed sample 11 so that the camera 154 can image a portion of the fixed sample 11. The display / user interface 110 shows this (visible) image to the user. At the same time, the moving optical subassembly 140 directs a NIR Raman pump beam 141 emitted by the laser source 102 to a point on or within a fixed sample 11, which emits a NIR Raman signal in response. The moving optical subassembly 140 collects this NIR Raman signal and directs it to the spectrometer 106, which generates a Raman spectrum of the point on the sample illuminated by the point on the sample 11 from the collected NIR Raman signal.

[0053] The user can image and collect Raman spectra from different portions of the fixed sample 11 by entering appropriate commands via the user interface 110. The electronics 112 and a dedicated motion controller 120 respond to these commands by triggering an actuatorAttorney Docket No. PEND-014WO01122 that is mounted or coupled to the moving optical subassembly 140 and moves the entire moving optical subassembly 140 with respect to the housing 130 and the fixed sample 11, moving the Raman microscope’s field of view (FOV) with respect to the fixed sample 11. Flexible bellows 133 around the aperture facing the fixed sample allow the moving optical subassembly 140 to move without compromising the integrity of the housing 130.

[0054] The electronics 112 and / or motion controller 120 can also adjust the focus of the Raman pump beam 141 with respect to the fixed sample 11 by causing the NIR optics 146 to move back-and-forth with respect to the Raman probe 142 and steer the Raman pump beam 141 laterally within the FOV using the MEMS scanning mirror 144. The electronics 112 and / or motion controller 120 control the focus of the camera 154 independently by causing the visible optics 152 to move back-and-forth with respect to the camera 154.

[0055] In practice, it can be difficult to design an optical system with both a high object-side numerical aperture (NA) and a large field of view (FOA). The MEMS scanning mirror 144 and / or the moving optical subassembly 140 make it possible to keep the FOV of the Raman optical subsystem relatively small and to move the Raman pump beam 141 over different points of the fixed sample 11. In some cases, the camera 154 can image the sample over a FOV greater than 5 mm with an object-space NA of less than 0.2, when the optical assembly 140 has an object-space NA of at least 0.3 and a FOV of less than 2 mm.

[0056] There are several ways to move the Raman pump beam 141 with respect to the sample so the FOV can be small enough to allow a higher object-side NA. The Raman microscope 100 in FIGS. 1 A-1E features an optical head — the moving optical subassembly 140 — that moves with respect to a fixed housing 130 / instrument 100 and a fixed sample 11. Alternatively, the sample can be moved with respect to the fixed instrument and fixed optical head or the entire instrument (including optical head) can be moved with respect to a fixed sample.

[0057] FIG. 2A illustrates a fieldable Raman microscope 200 with a fixed optical subassembly 240 instead of a moving optical subassembly 140. Like the moving optical subassembly 140, this fixed optical subassembly includes a Raman probe 242, MEMS scanning mirror 244, NIR optics 246, dichroic beam splitter 248, visible / NIR optics 250, visible optics 252, and (visible) camera or image detector 254. The NIR optics 246 and visible optics 252 can be adjusted independently to provide independent focus control for the Raman pump beam 141 and camera 254, respectively. Likewise, the MEMS scanning mirror 244 can move to steer the RamanAttorney Docket No. PEND-014WO01 pump beam 141 across the FOV. The other components are fixed with respect to the housing 130. (If desired, the bellows can be omitted from the housing 130.)

[0058] In operation, the Raman microscope 200 images and analyzes a sample on a motorized sample stage 21 that can move the sample with respect to the Raman microscope’s FOV. In this example, the fieldable Raman microscope 200 includes a motion controller 220 that connects to and controls the motorized sample stage 21. In other examples, the Raman microscope 200 and motion controller 220 can be coupled to and controlled by an external computer or controller.

[0059] FIG. 2B illustrates a fieldable Raman microscope 200’ with an external motion controller 220’ and external actuator 222 that move the entire Raman microscope 200’ (or at least the housing 130 and its contents, including the optical subassembly 240) with respect to a fixed sample 21’. This moves the Raman microscope’s FOV across the fixed sample 21’. Again, the NIR optics 246 and visible optics 252 can move independently for focus control and the beam-steering MEMS mirror 244 can scan the Raman pump beam 141 across the FOV.

[0060] FIGS. 3 A and 3B shows an alternative Raman microscope 300. It includes an enclosure 330 that houses the Raman microscope’s components, including but not limited to a Raman pump laser 102, spectrometer 106, image sensor (camera 154), electronics, stages / actuators 122, and moving optical subassembly 140 as described above with respect to FIGS. 2 A and 2B. There may be a touchscreen or similar user interface 110 on the top, back, or side of the enclosure 330. The Raman microscope 300 can also connect, via a wired or wireless connection, to computer, tablet, or smartphone that controls and displays sample data acquired by the Raman microscope 300. The Raman microscope 300 acquires this sample data by illuminating a fixed sample 21’ on a sample plate 30 with a Raman pump beam 341 via an objective 350 that can move laterally with respect to the enclosure 330 and / or the sample plate 30. A (motorized) vertical stage 324 controlled by a motion controller 320 can move the entire enclosure 330, including the objective 350, vertically (i.e., in the z direction) with respect to the sample plate 30. A lateral actuator or translation stage 322 controlled by the motion controller 320 moves the moving optical subassembly 140 in the transverse (x and ) directions with respect to the sample plate 30 and the rest of the enclosure 330.

[0061] The motion can be manually driven and / or motorized, e.g., using a motion controller, in the Raman microscopes 100, 200, and 300 in FIGS. 1-3. Other Raman microscopes can use a combination of optical subassembly motion, sample motion, and whole microscope motion —Attorney Docket No. PEND-014WO01 for example, optical subassembly motion may provide fine scanning or motion in one direction, and sample motion may provide coarse scanning or motion in another direction.

[0062] Moving the FOV makes it possible to bring a high-NA Raman pump beam 141 above specific points of the sample instead of relying on an optical design with both a high NA and a large FOV for the Raman optical subsystem. Increasing the NA (e.g., to > 0.2) of the Raman optical subsystem allows for higher collection efficiency of the Raman signal and consequently a larger Raman signal. In contrast, the visible optical subsystem can have lower NA (e.g., < 0.2), since the camera can be very sensitive and produce good quality image, which makes obtaining a large FOV (e.g. 2.5 cm x 2.5 cm) for the visible optical subsystem more practical. The Raman optical sub-system and the visible optical sub-system can have independently adjustable foci to allow their respective beams (i.e., the Raman pump beam 141 and the visible beam) to come to focus at the sample surface or on a particular point of interest, which can be automatically identified by the Raman microscope’s processor / CPU. The Raman microscope can acquire a z-stack image of the sample by varying the focus of the visible image and can generate a 3D representation of the sample surface from this z-stack image. This can be used to identify and sort the location of the points to be analyzed.Laser Source

[0063] As shown in FIG. IE, the Raman microscope includes a laser source 102 that serves as the primary excitation source for inducing Raman scattering in the sample under analysis. The laser source 102 is configured to emit a highly monochromatic and coherent beam of light — the Raman pump beam 141 — typically in the visible or near-infrared (NIR) spectral range, depending on the specific application and the nature of the sample. Common wavelengths for Raman spectroscopy include 532 nm (green), 785 nm (NIR), and 1064 nm (NIR), each offering different advantages based on the sample’s characteristics.

[0064] The laser source 102 may operate at a fixed wavelength, which can be selected to increase or maximize the Raman scattering cross-section of the sample while reducing or minimizing fluorescence interference from the sample. Alternatively, the laser source 102 may also be tunable or be composed of two separate lasers for implementing Shift Excitation Difference Raman Spectroscopy (SERDS) as discussed below.

[0065] The output power of the laser source 102 is adjustable, allowing for fine-tuning of the excitation intensity to prevent sample damage while optimizing signal-to-noise ratio.Attorney Docket No. PEND-014WO01

[0066] In some configurations, the laser source 102 may be pulsed or modulated, depending on the measurement.Raman Focusing System

[0067] The optical system (e.g., NIR optics 146 and visible / NIR optics 150 in FIG. IE) focuses the laser light (Raman pump beam 141) onto the sample with high precision, ensuring optimal interaction between the excitation source and the target material for effective Raman scattering generation. The optical system can be either free-space or fiber-coupled, depending on the specific implementation.

[0068] In some configurations, the laser source emits a single spatial mode beam, which is advantageous for creating a tight focus on the sample. The optical path maintains the beam quality, reducing or minimizing aberrations and ensuring that the laser light converges to as close as possible to a diffraction-limited spot. The focusing optics, typically composed of high NA lenses or microscope objectives, can be optimized for the wavelength of the laser and the desired spot size on the sample.

[0069] When the laser light is free-space coupled, the optical system may include polarizing elements such as polarizers or waveplates to control the polarization state of the light reaching the sample. This can enhance the Raman signal by aligning the polarization with specific molecular orientations within the sample.

[0070] Alternatively, the laser light may be coupled to a single-mode optical fiber (e.g., optical fiber 104 in FIG. IE), which guides the light at least partway to the focusing optics. In this fiber-coupled configuration, the laser light may undergo polarization scrambling over the length of the fiber, resulting in a depolarized or partially polarized beam at the output. The optical system is designed to account for this potential depolarization, ensuring consistent Raman signal collection. Alternatively, a polarization-maintaining fiber may be used if it is preferable to maintain the polarization of the laser light.Motorized Motion System

[0071] Some fieldable Raman microscopes (e.g., Raman microscopes 100, 200, and 300) include a precision motorized motion system designed to manipulate the position of the sample relative to the laser beam (Raman pump beam 141), enabling accurate targeting and focusing on specific regions of interest. The motion system provides three translation degrees of freedom: two lateral (x and y) and one longitudinal (z), allowing for comprehensive spatial control of the laser spot on the sample.Attorney Docket No. PEND-014WO01

[0072] A motion system can include manual and / or motorized stages that allow for precise lateral movement of the sample along the x and axes, perpendicular to the optical axis of the Raman microscope. This enables the user to aim the laser spot onto specific portions of the sample, such as individual particles or distinct surface features. The motion system is capable of fine adjustments, ensuring accurate positioning of the laser focus within the field of view. The xy motion system also expands the effective FOV of the system relative to the single-shot FOV of the camera, allowing the Raman microscope to image a wider area by stitching together multiple camera FOVs. In one embodiment, the single-shot camera FOV is 8 mm wide, but the Raman microscope can imagine and analyze samples over an area that is 25 mm wide, using a 3 x 3 square pattern of single-shot camera FOVs.

[0073] Another manual or motorized stage provides movement along the z axis, nominally along or parallel to the Raman microscope’s optical axis. This degree of freedom allows the laser beam (Raman pump beam 141) to be focused onto the target area with high precision. By adjusting the z position, the user can ensure that the laser spot is sharply focused on the desired sample plane, optimizing the Raman signal and reducing background interference. If the z axis of motion is not be exactly parallel to the optical axis, this error can be measured and calibrated out.

[0074] The motorized motion system can be operated through a user interface (e.g., user interface 110), allowing for manual control by the operator. The user interface provides realtime feedback on the position of the sample and enables precise adjustments through intuitive controls. Additionally, the Raman microscope supports automated operation, where the motion can be governed by machine vision (described below). Using machine vision, the Raman microscope can automatically detect and target specific features on the sample, guiding the motion system to position and focus the laser spot with minimal user intervention.Raman Scattering Collection System

[0075] The Raman microscope is equipped with an optimized collection system (e.g., visible- NIR optics 150 and NIR optics 146) that efficiently gathers the weak Raman-scattered light from the sample and directs it to a spectrometer for spectral analysis. This Raman microscope may utilize the same high NA optics employed for focusing the laser light onto the sample, ensuring efficient collection of the scattered light over a broad angular range.

[0076] The high NA (relative to the NA that is used for imaging) focusing optics capture a significant portion of the Raman-scattered light, which is usually much weaker than theAttorney Docket No. PEND-014WO01 incident laser light. These optics are achromatic over the range of wavelengths expected to be collected during the Raman measurement. This achromatic design reduces or avoids strong modulations in the instrument transfer function, ensuring a consistent and accurate response across the desired Raman spectral range.

[0077] In one embodiment, the NAs are about 0.3 for the objective of the Raman subsystem and about 0.07 for the imaging subsystem.

[0078] Thin-film filters incorporated into the return optical path (e.g., in the Raman probe 142) isolate the Raman signal from the much stronger Rayleigh scattering of the laser light. These filters are precisely tuned to block the wavelength of the Raman pump beam(s) while allowing the Raman-shifted wavelengths to pass through. This ensures that only the Raman signal reaches the spectrometer, enhancing the quality of the spectral data.

[0079] The filtered Raman signal is then coupled to a spectrometer (e.g., spectrometer 106), which disperses the Raman signal into its constituent wavelengths to produce a Raman spectrum. Depending on the system configuration, this coupling can be achieved through free- space optics or via an optical fiber (e.g., optical fiber 108). Fiber coupling provides flexibility in the placement of the spectrometer within the Raman microscope, facilitating more compact or modular designs. In cases where fiber coupling is used, the fiber itself may also serve as an aperture to spatially define the input light, contributing to the spectral resolution of the system.

[0080] For polarized Raman spectroscopy, a polarizer or other polarization analyzer can be inserted into the return path (e.g., in the Raman probe 142) before the spectrometer. This analyzer allows the Raman microscope to selectively detect specific polarization states of the Raman- scattered light, enabling more detailed analysis of the sample’s molecular orientation and structure.

[0081] At the entrance of the spectrometer, a slit or pinhole is typically used to define the input light’s spatial dimensions, thereby controlling the spectral resolution and lineshape. This component ensures that the Raman spectrum obtained is both accurate and high-resolution. In fiber-coupled systems, the spatial definition may alternatively be achieved using the aperture of the optical fiber itself, simplifying the optical design.Spectrometer

[0082] The Raman spectroscopic system incorporates a spectrometer (e.g., spectrometer 106) that disperses the collected Raman-scattered light into its constituent wavelengths, enabling the generation of a detailed Raman spectrum. The spectrometer is tailored to balance spectralAttorney Docket No. PEND-014WO01 resolution, size, and portability, particularly for applications that use compact or mobile instrumentation.

[0083] The spectrometer includes a dispersive element, such as a diffraction grating or prism, to separate the Raman-shifted light into its component wavelengths. Using a diffraction grating is a reliable, effective approach for producing high-resolution spectra. For portable systems, a resolution of 8-10 cm1is generally sufficient, offering an optimal trade-off between compactness and spectral resolution. This level of resolution is adequate for many Raman applications, allowing for the identification of key molecular features while maintaining a manageable instrument size.

[0084] The spectrometer in the Raman microscope may include a single fixed diffraction grating. This fixed-grating design eliminates moving parts, which can be a source of mechanical wear and misalignment over time. By avoiding mechanical movement, the spectrometer maintains good alignment stability, reducing or eliminating recalibration, which is particularly beneficial in field applications.

[0085] The spectrometer includes a linear sensor array, such as a charge-coupled device (CCD) or complementary metal-oxide-semiconductor (CMOS) sensor, to detect the dispersed light. The linear sensor is rigidly mounted within the spectrometer housing to ensure consistent alignment and to prevent drift over time, which could otherwise degrade the accuracy and reliability of the spectral measurements.

[0086] To reduce noise, especially dark current noise, the linear sensor array may be cooled. Cooling the linear sensor array reduces thermal noise, thereby enhancing the signal -to-noise ratio, particularly during longer integration times. However, cosmic rays occasionally impact the linear sensor array, introducing anomalies in the captured data. These affected frames can be identified and corrected, ensuring that the final spectrum is free from such artifacts.

[0087] The linear sensor array integrates the Raman signal over a specified period (the frame period), depending on the intensity of the signal and the desired sensitivity. After integration, a read-out circuit digitizes the signal, converting the detected light intensity at each pixel into digital values that represent the Raman spectrum. This spectrum provides a direct correlation between light intensity and pixel position on the sensor.

[0088] The Raman microscope’s processor performs one or more calibration routines to translate the raw pixel data into meaningful spectral information. Specifically, these routines calibrate the conversion from pixel position to frequency shift (Stokes shift axis), ensuring thatAttorney Docket No. PEND-014WO01 the final output is a calibrated Raman spectrum. The calibration process accounts for any nonlinearities or imperfections in the optical system, resulting in accurate and reliable spectral data.Shifted Excitation Raman Difference Spectroscopy (SERDS)

[0089] An inventive Raman microscope can perform Shifted Excitation Raman Difference Spectroscopy (SERDS), an advanced technique designed to enhance the quality of Raman measurements, particularly in challenging scenarios where laser-induced fluorescence is present. In SERDS, the Raman microscope makes two Raman measurements in (quick) succession, using Raman pump beams at slightly different wavelengths, typically shifted by a few nanometers. This wavelength shift can be achieved by tuning a single laser or switching between lasers. For instance, the Raman microscope can employ a single laser whose wavelength can be finely tuned by adjusting that laser’s temperature. This allows for precise control of the wavelength shift without additional laser sources.

[0090] Alternatively, the laser source in the Raman microscope can include two lasers, each emitting at a slightly different wavelength. These lasers may utilize grating feedback, a distributed feedback structure directly on the semiconductor laser, a distributed Bragg grating mirror, or other wavelength-selective mechanisms to define their emission wavelengths accurately. The lasers are designed or selected to ensure that the wavelength shift between them is optimal for SERDS. The lasers may be packaged into a single laser package and beam- combined using various existing means of laser beam combining.

[0091] One benefit of using SERDS is the ability to mitigate the interference caused by laser- induced fluorescence, which can often overshadow weak Raman signals. Fluorescence typically remains constant over small wavelength shifts, while the Raman signals shift in direct proportion to the excitation wavelength. By taking two measurements with slightly different excitation wavelengths, the fluorescence background, which is nearly identical in both measurements, can be subtracted. This subtraction process highlights the Raman signal while significantly reducing or eliminating the fluorescence contribution. The result is a much clearer Raman spectrum, free from the distorting effects of fluorescence, allowing for more accurate and reliable identification of the sample’s molecular components.

[0092] FIG. 4 illustrates a SERDS measurement of acid blue 9 dye. It shows raw Raman spectra acquired with Raman pump beams (Laserl and Laser2; upper traces) at slightly different wavelengths and the difference between these spectra (lower trace). Taking theAttorney Docket No. PEND-014WO01 difference of the raw Raman spectra eliminates or reduces the fluorescence contribution and highlights the Raman spectrum of the acid blue 9 day. It also illustrates the advantages of integrating SERDS into a fieldable Raman microscope for samples prone to fluorescence, such as biological materials, complex mixtures, or substances with intrinsic chromophores. SERDS enhances the Raman microscope’s ability to deliver high-quality spectral data even in the presence of strong fluorescence, expanding the range of samples that can be effectively analyzed.Imaging and Illumination System

[0093] A fieldable Raman microscope includes an imaging and illumination subsystem (e.g., camera 154, visible optics 152, and ring LED illuminator 160) for visual observation and documentation of the sample. This imaging capability is integrated into part of the same optical path used for focusing the Raman pump beam and collecting the Raman- scattered signal, necessitating a carefully optimized objective design. The imaging system uses the same objective lens (e.g., visible-NIR optics 150) that focuses the Raman laser beam onto the sample and collects the Raman-scattered light. This shared use places specific demands on the objective’s design, including good performance across a broad range of wavelengths. The objective should have a high NA and excellent optical performance near the Raman microscope’s optical axis to ensure precise focusing of the Raman laser beam. Simultaneously, the objective should have a lower NA with a wide FOV and minimal imaging aberrations for the camera system, enabling accurate and clear (visible) imaging of the sample. The use of a single shared objective allows the Raman microscope to be more compact, but it does present trade-offs in term of imaging performance, signal collection, and spatial resolution.

[0094] Alternatively, a Raman microscope can have multiple objectives: a wide FOV objective allowing to image the sample and locate targets for analysis, and a higher magnification and narrower FOV objective allowing detailed imaging of small targets. A turret or translation mechanism, possibly motorized, may be used to swap the objectives.

[0095] The camera-based imaging subsystem includes an illuminator (e.g., ring LED illuminator 160). The illuminator, or illumination system, illuminates the sample with visible light, which scatters and / or is reflected by the sample and collected by the imaging system. The properties of the illumination system (including location with respect to the imaging focal plane, directionality, and wavelength) affect the properties of the image, in particular theAttorney Docket No. PEND-014WO01 contrast between particles of interest and the substrate. Moreover, the optimal illumination configuration may be sample-dependent.

[0096] FIGS. 5A-5D show that the illuminator 160 includes many visible light sources (e.g., visible LEDs), each with independently controllable brightness, at different positions with respect to the imaging focal plane to provide different illumination configurations. This example includes forty -two LEDs total: forty ring LEDs 162 arranged around the Raman microscope’s external a 5-centimeter-diameter ring or circle for top-illumination and two wing LEDs 164 for high-angle side illumination. Other versions of the illuminator may have more or fewer LEDs and / or LEDs arranged or mounted differently with respect to the sample. As explained in greater detail below, each LED is a co-packaged red-green-blue (RGB) LED that can emit red, green, blue, or white light. The illuminator 160 can also include infrared (IR) and / or ultraviolet (UV) light sources in addition to or instead of visible light sources. UV light sources are especially useful for fluorescent imaging (meaning that the illuminator 160 illuminates the sample with UV light and the camera acquires a visible image of the sample, from possible fluorescence of the sample). This can increase the contrast between different materials or between particles and a substrate and may help highlight portions of the sample of biological origin. The LEDs are individually addressable and can be synchronized to camera exposure by the onboard electronics. This control enables the Raman microscope to acquire successive frames under different illumination conditions.

[0097] The LEDs are positioned at different angles of incidence to the focal plane to provide different illumination configurations for different samples. To discern particles from the background surface, on some surfaces it can be helpful to combine the information from multiple images taken under different illumination conditions. A flat sample (such as a piece of paper) is often best imaged under grazing-angle illumination, whereas a sample with surface topology (rough surface, concave or convex regions, or a porous sample such as Velcro) may benefit from illumination at an angle-of-incidence selected such that the illumination can reach all points on the sample.

[0098] In some cases, the imaging and illumination system can acquire multiple images of the same scene (sample area) with illumination from different orientations using the ring LEDs 162 and / or wing LEDs 164. For instance, different illumination conditions may be obtained by turning on LEDs in different quadrants of the ring, or turning on specific individual LEDs, or turning on every second, third, fourth, fifth, etc. LED in the ring — FIGS. 5C and 5D show the ring LEDs in a ring / 4 configuration, which uses every fourth LED (i.e., ten of the forty LEDsAttorney Docket No. PEND-014WO01 on and the other thirty off) for each frame, then rotates the lighting from one frame to the next. The illuminator 160 illuminates the sample with each of these illumination configurations in succession, while the camera 154 acquires corresponding images. The Raman microscope’s onboard processor / electronics 112 can use these images to create a high-contrast composite image, also called a flattened image, for example, highlighting small particles as bright spots, while the background substrates the particles are deposited on may be comparatively attenuated. This can be done in conjunction with movement of the z-axis to acquire a stack of composite high contrast images, in order to locate particles of interest on uneven, tilted, or curved substrates.

[0099] Regions of samples with irregular surface topology may appear bright due to the interaction of the topology with the incident illumination. For example, a bump or divot can specularly reflect the incident illumination into the imaging system, causing a bright spot in the acquired image. This bright spot may be caused by a single illumination source (LED), whose position happens to be such that its emitted light interacts with the local topology to yield a bright spot. Particles of interest, on the other hand, may be smaller (e.g., < 100 pm) than the characteristic length of the topology features, which tends to cause the light scattered from a particle and captured by the imaging system to depend less on the precise position of the illumination source. This difference in scattering behavior between small particles and substrate topology can be exploited to improve imaging contrast: multiple images of the sample, each taken under a different illumination configuration, can be combined using appropriate image processing to reduce the bright artifacts caused by the sample topology. This increases the contrast of the particles with respect to the substrate.

[0100] FIGS. 6A, 6B, 7A, and 7B illustrate how different illumination configurations can be used to reduce or suppress bright spots in images of samples with irregular surface topology. They show flattened images of dimpled black acrylonitrile butadiene styrene (ABS) (FIGS. 6A and 6B) and Velco® hook-and-loop fasteners (FIGS. 7A and 7B) illuminated with full ring illumination (FIGS. 6A and 7A) and ring / 4 illumination (FIGS. 6B and 7B). The ring / 4 lighting removes bright artifacts caused by the dimples in the ABS. It also removes most of the Velcro® fiber edges.

[0101] Each LED in the illuminator 160 can emit light at different wavelengths to improve contrast and / or create color images with a monochrome camera / image sensor. For example, illuminating a sample with light at three different wavelengths (e.g., red, green, and blue) in successive images captured by a monochrome camera sensor can be used to constructAttorney Docket No. PEND-014WO01 a color image of the sample. As another example, a white particle sitting on a red substrate has better contrast under blue illumination than white illumination because the red substrate scatters less blue light than red (or green) light.

[0102] FIGS. 8A and 8B illustrate how monochromatic red, green, and blue illumination can be used to generate color-flattened images, which is especially useful for finding particles. FIG. 8A shows full-color, color-flattened images (left) of red, green, and blue buttons on a remote control synthesized from monochromatic images of the remote control under blue (right, top), red (right, middle), and green (right, bottom) illumination. The Raman microscope’s user interface can display the full-color image to the user while using the monochromatic images for particle finding or other tasks.

[0103] FIG. 8B shows a full-color, color-flattened image (left) of particles of Goody’s powder on a region of the red button of the remote control of FIG. 8 A. FIG. 8B also shows the monochromatic images, or channels, used to create the color-flattened image. (The contrast ratios in the blue and green images are inverted for clarity.) The contrast is much higher in the blue and green channels than in the red channel. The Raman microscope can use this contrast difference to find particles by performing brightness thresholding separately on each channel (monochromatic image), then combining the thresholded images for particle finding. This yields several benefits, including (1) improved ability to identify a particle with good contrast in only one or two of the three color channels and (2) the ability to assign each particle separate red, green, and blue values. In addition, there is no need to know a priori the best choice of illumination color for a given substrate.

[0104] The electronics 112 (FIG. IE) may include or be configured to act as an illumination controller that controls the location / illumination angle, brightness, and wavelength of the light source(s) (LED(s)) used to illuminate the sample. Some of the schemes described above for increasing imaging contrast use multiple images of the sample captured under different illumination configurations. To achieve this, there is electrical communication between the camera and the illumination controller, such that the illumination controller receives notice when the camera finishes exposing one image and can then modify the illumination configuration prior to the exposure of the subsequent image.

[0105] The imaging system operates in a wavelength range that is typically close to, but distinct from, the Raman excitation and scattering wavelengths. The dichroic beam splitter 148 / 248 or another dichroic filter in the optical path splits the light between the RamanAttorney Docket No. PEND-014WO01 detection path and the imaging path. Depending on the specific Raman pump wavelength(s), either a short-pass or long-pass dichroic filter can be used to direct the appropriate (visible) imaging wavelengths to the camera / image sensor, while the remaining Raman signal wavelengths continue to the Raman spectrometer. The imaging wavelength range can be selected to match the sensitivity of the image sensor (e.g., a silicon CCD or CMOS sensor), ensuring high-quality images with good contrast and detail, for example, in visible to nearinfrared spectrum where CCD or CMOS sensors perform well.

[0106] If desired, the aperture of the optical system on the imaging path can be adjusted to enhance the spatial resolution and expand the field of view of the imaging system. Closing the aperture increases the depth of field and reduces aberrations, allowing for clearer images across a wider area of the sample. This aperture control is particularly beneficial when imaging larger sample areas or when detailed structural information is desired. The optical system on the imaging path may also include a fixed iris to define the numerical aperture and maintain good image quality.

[0107] The objective 150 / 250 is configured to balance the Raman and imaging systems. It provides a high NA and good focusing quality for Raman spectroscopy and a lower NA, wide FOV, and low aberrations for imaging. The high NA and good focusing quality for Raman spectroscopy ensures that the Raman excitation light (Raman pump beam) is sharply focused on the sample, reducing the focal spot size for high-resolution Raman analysis and increasing signal collection. The lower NA, wide FOV, and low aberrations for imaging allow the camera 154 / 254 to capture high-quality images with minimal distortion or aberrations across a larger sample area. This dual functionality maintains a wide FOV for imaging and a narrow laser spot for the Raman pump beam.

[0108] The Raman microscope’s optical train, including the objective 150 / 250, is configured to keep the Raman pump beam and the camera imaging plane in focus simultaneously. This ensures that when the sample is in the desired position for Raman analysis, the camera captures sharp images of the region being analyzed. This is particularly useful for correlating visual and spectroscopic data, aiding in the precise localization of Raman measurements on specific sample features.

[0109] In practice, the simultaneous focus may not be strictly perfect and may depend on the color of the illumination used since there is likely to be some chromatic aberration to the optical system. The camera lens (visible optics 152 / 252) can be adjusted such that whenAttorney Docket No. PEND-014WO01 the image is in focus under white light illumination, the Raman signal is increased or maximized. However, the Raman microscope can also illuminate the sample with a single color (e.g., red, green, or blue). The Raman microscope can then be calibrated so that the Raman pump beam and visible RED image (e.g., the red image) have the same focal plane. For the other images (e.g., the green and blue images), the Raman microscope uses a calibration parameter to account for the wavelength differences (e.g., z green - z_IR and z blue - z_IR).

[0110] An inventive Raman microscope can capture images of a sample at different focal planes, e.g., by moving the camera 154 / 254 with respect to the sample along the z axis and / or adjusting the focus of the camera 154 / 254 and combine the images into a z-stack, or 3D representation of the sample. Generating a z-stack is useful when sample’s surface is not in focus at a single depth, e.g., because the sample is tilted, curved, undulating, or otherwise nonuniform. The z-stack can also be flattened into a single composite 2D image.

[0111] An inventive Raman microscope can generate z-stacks or flattened images using any of several different image acquisition and processing modes. In a first mode, the camera acquires images at different z-axis positions of the optical system while illumination is constant. In a second mode, both the optical system’s z-axis position and the illumination color change from image to image (e.g., the camera acquires separate images of the sample under monochromatic red, green and blue illumination at each z position). In a third mode, both the optical system’s z-axis position and the illumination direction (subset the activated LEDs) change from image to image. And in a fourth mode, the optical system’s z-axis position, the illumination direction, and the illumination color(s) change from image to image. Changing the optical system’s z-axis position allows the camera to obtain a composite image that is in focus across the entire field of view even when the sample is curved or tilted. Changing the illumination color and / or direction enhance the contrast between the substrate and particles on the substrate.

[0112] FIGS. 9A and 9B illustrate how a camera 154 / 254 in a Raman microscope can use a laptop 900 attached to the Raman microscope to acquire and process a z-stack of a sample. The laptop 900 includes multiple CPU cores and an optional graphics processing unit (GPU) 910, along with CPU random access memory (RAM) 902, and CPU shared RAM 906. The laptop 900 uses these components to perform pre-processing and other data conditioning 904 and focus processing 908, which involves determining a focus metric that is used to identify the sharpest image within a stack of sub-images. In more detail, when performing a z-stack, the camera acquires images at different distances from the substrate. The microscope splitsAttorney Docket No. PEND-014WO01 these images into smaller tiles, with a z-stack for each tile. Using the focus metric calculated for each tile and at each height gives the height with the sharpest image for this tile. Different tiles may be sharpest at different height. By taking the sharpest image for each tile, the laptop 900 can create a composite image that is sharp across the full field of view. Knowing the focus metric, or height at which the image is sharpest, for each tile makes it possible to rapidly focus the Raman probe on a particle within a given tile.

[0113] (Alternative Raman microscopes may include some or all of these components instead of being coupled to a separate laptop 900.) The camera 154 / 254 has a frame period, frame rate, and image size / resolution that may be fixed or adjustable within a certain range. For a better user experience, the laptop 900 generates each z-stack fast enough to keep up with the camera frame rate, which makes the speed of the GPU 910 especially useful. Suppose, for example, that the camera 154 / 254 acquires 8 MP frames at a frame period of 23 ms and the CPU and GPU 910 can calculate the image sharpness in about 500 ms and about 13 ms, respectively. Using the GPU 910 to calculate the image sharpness enables the laptop 900 and / or the Raman microscope to display the flattened image stops moving at the end of the z-stack image acquisition.

[0114] FIG. 9B illustrates the operations performed by the Raman microscope and laptop 900 to generate a flattened z-stack image of a sample and use that flattened z-stack image to locate and acquire Raman spectra of particles on or regions of the sample. First, the Raman microscope performs a rough focus (952) and locates one or more regions of interest (e.g., each up to 2.5 cm x 2.5 cm) of the sample (954). Next, the Raman microscope selects an illumination pattern for the visible images to be acquired by the camera (956). The camera acquires visible images of the sample at different focal planes (a z-stack), and the Raman microscope’s onboard processor and / or an auxiliary processor (e.g., laptop 900) flattens the visible images into a flattened z-stack image that brings the entire sample into focus (958). The Raman microscope can automatically locate particles of interest in the flattened image and records metrics such as the length, width, shape, brightness, and color of each particle (960). It uses these metrics to identify a subset of the particles to be targets for Raman measurement (962), for example, by filtering the particles based on size, shape, brightness, or color or by sampling the full space of particle metrics. The user can also locate and select particles for analysis using the Raman microscope’s user interface. The Raman microscope moves the optical head to each particle (or each particle to the optical head) and performs Raman measurements of the particles, followed by chemical identification (964).Attorney Docket No. PEND-014WO01Machine Vision Subsystem

[0115] The Raman microscope includes a sophisticated machine vision subsystem that leverages the imaging capabilities of the Raman microscope’s camera to enhance the efficiency, accuracy, and automation of Raman analysis. The machine vision subsystem processes the images collected by the camera to perform several functions, aiding in both manual and automated target identification, as well as optimizing the system’s focus for precise Raman measurements.

[0116] FIG. 10 shows a visual image (with inverted contrast) of a particle identified for Raman analysis by the machine vision subsystem. The crosshairs represent the position of the focused Raman pump beam. Targets for analysis can be identified in manual target selection mode or automated target selection mode. In manual target selection mode, the Raman microscope’s user interface or the display of a laptop coupled to the Raman microscope displays the visible image to the user. The user can manually select points or regions of interest on the image for analysis using Raman spectroscopy. This interactive selection process allows the user to focus on specific features or particles on the sample that may be of particular importance.

[0117] In automated target identification mode, the Raman microscope automatically detects and identifies potential targets, such as particles deposited on a surface, based on the image data. The Raman microscope’s processor or an auxiliary processor analyzes the image to detect features that meet predefined criteria, such as size, shape, or contrast. The Raman microscope’s processor or an auxiliary processor ranks or prioritizes these targets for analysis based on factors like size, brightness, or other relevant metrics. The highe st-priori ty targets can then be automatically selected for Raman analysis, streamlining the process and reducing manual intervention.Substrate Detection and Raman Measurements

[0118] FIGS. 11A andl lB illustrate another feature of an inventive fieldable Raman microscope: substrate detection and removal / compensation. In addition to locating particles, the Raman microscope can locate the sample substrate (i.e., sample locations without particles). The Raman microscope can measure the sample substrate’s Raman spectrum and subtract it from the raw Raman spectrum of each target or particle to yield the Raman spectrum of the particle by itself. If desired, the Raman microscope can measure the substrate’s Raman spectrum at each of multiple locations, average the measured substrate Raman spectra, andAttorney Docket No. PEND-014WO01 subtract the average. The Raman microscope can also measure the Raman spectrum of the sample close to each target / particle and subtract that measurement from the raw Raman spectrum of that target / particle. In each of these cases, subtracting the substrate Raman spectrum from the raw target / particle Raman spectrum yields a Raman spectrum that more faithfully represents the material that makes up the target / particle.

[0119] FIG. 11C illustrates how subtracting the substrate Raman spectrum improves the Raman microscope’s ability to identify different compounds. It shows a plot of the signal - to-noise ratio (SNR) calculated on the full / raw Raman spectrum of a KCIO3 particle measured with a Raman microscope (upper trace) and on the difference between the full / raw Raman spectrum and the Raman spectrum of the substrate supporting the KCIO3 particle (lower trace) versus measurement acquisition time. In both cases, the SNR increases with measurement acquisition time. Often, however, measurement acquisition time is limited — that is, faster measurements are preferred. Under these conditions, the Raman microscope may halt the measurement once the SNR reaches a threshold (e.g., represented by the dashed horizontal line in FIG. 11C) high enough to detect a chemical. In this example, however, if SNR is calculated on the full spectrum, the threshold is exceeded and spectral acquisition would halt. Only the substrate would be identified. Subtracting the substrate Raman spectrum before calculating SNR gives a better measure of when the Raman microscope will identify the KCIO3 particle successfully. In this case, the Raman microscope acquires signal for a longer period before halting spectrum acquisition so that it identifies the KCIO3 particle in addition to the substrate.Camera-Stage Calibration

[0120] To ensure precise targeting for Raman analysis, the machine vision subsystem converts the pixel coordinates of identified targets into corresponding stage positions (e.g., for the optical head, the sample, or the entire Raman microscope) for the X, Y, and Z axes. This conversion is achieved through two calibration routines, one for the X and Y axes and one for the Z axis. The calibration ensures that the machine vision system can accurately drive the actuators or motors, positioning the Raman probe to focus on the selected targets of interest. Following the movement of the stage to the desired coordinates, further refinement of the stage position (for example, based on “template matching” the local environment to the desired target, or a subsequent autofocus routine) can be performed for even higher positional accuracy. Calibration brings the stage as close as possible to the desired target, which reduces and, in some cases, obviates further refinement. This seamless integration between the imaging data and motor control enhances the accuracy and efficiency of the Raman measurements.Attorney Docket No. PEND-014WO01

[0121] XY calibration generates a mapping function from an arbitrary pixel coordinate (Xcam, ycam ) within the camera’s field of view (FOV) to a stage coordinate (xstage, ystage), which is the position of the stage that would bring the feature at (xCam, ycam) to the center of the camera’s FOV. This mapping function is affected by distortion in the imaging system and the relative orientation of the camera xy axes and the stage xy axes to within manufacturing tolerances. The mapping function may be calibrated on each individual Raman microscope. A reference sample contains features across the full FOV whose spatial locations on the sample (xsampie, ysampie) are known (for example, grid points arrayed at a pitch of 500 pm.) The reference sample is first imaged by the camera system, and image processing by the Raman microscope’s processor yields a mapping function from (xCam, ycam) to (xsampie, ysampie). Next, the stage is moved between several different points on the sample, which establishes a mapping function from (xsampie, ysampie) to (xstage, ystage). Taken together, the composite function provides the desired mapping from (xCam, ycam) to (xstage, ystage).

[0122] Z calibration generates a mapping function dzstage(xcam, ycam). Using the focus metric stack procedure described below, a value for the stage z position zstage is assigned to each pixel (xcam, ycam) of a flattened image. When the optical head is moved to bring the object located at (xCam, ycam) to the optical axis, the z position that optimizes the focus is zstage + dZstage(Xcam, ycam), where dzstage is a correction term that accounts for curvature and / or tilt of the focal surface with respect to the xy plane of the stage axes. These effects may be caused by field curvature in the imaging system, deviations between the stage z axis and the optical axis, or image sensor tilt with respect to the stage xy plane.

[0123] This correction function dzstage(xcam, ycam) can be determined empirically using a reference or calibration sample with a discrete feature located in each cell of an equispaced grid across the full FOV. The Raman microscope acquires a flattened z-stack image of the reference sample and assigns values of Zstage, Xcam, and ycam to each feature. The optical head then moves to one of the features and obtains a new focus metric stack to determine the optimal stage z position when the feature is at the center of the FOV, Zstage, center-FOV. The correction value dzstage is then given by Zstage, center-FOV—Zstage. The procedure is repeated for the features on the reference sample, which provides many samples of the function dzstage(xcam, ycam) distributed over the full FOV. Fitting a second-order polynomial fit yields the continuous, smooth function dZstage(Xcam, ycam).Raman-Camera CalibrationAttorney Docket No. PEND-014WO01

[0124] For a fieldable or portable Raman microscope, the alignment between the Raman system and the camera-based imaging system can shift over time, for example, due to mechanical shock during transport. Even with a shared objective (described above), there are elements that are not shared between the Raman and imaging systems, including the Raman probe 142 / 242, MEMS mirror 144 / 244, NIR optics 146 / 246, and visible optics 152 / 252 (FIGS. IE, 2, and 3). Calibration compensates for the changing relative alignment of these elements. Since calibration addresses the relative alignment, the camera focal plane can be treated as fixed, and the focused laser spot (the center of the beam waist) can drift in x, y, and z with respect to the camera focal plane.

[0125] Experimentally, the drift in z appears to be lower in magnitude than the drift in x and y, such that a calibration procedure compensating for laser beam waist drifts only in x and y may be sufficient.

[0126] Raman-camera XY calibration determines the coordinates (xCam, ycam) of the camera pixel that collects the maximum Raman signal from a feature on a calibration sample. This feature includes a Raman-active region whose spatial extent is on the order of the laser beam waist size surrounded by a Raman-inactive material. The camera can resolve the Raman- active feature well enough to determine the Raman-active feature’s location. If the Raman- active feature is too small to be easily seen on the camera image, alignment markers on the sample can be used as visual indicators for determining the location of the Raman-active feature.

[0127] FIG. 12A shows an example crosshair calibration sample 1200 for performing Raman-camera XY calibration of a fieldable Raman microscope. This crosshair calibration sample 1200 includes Raman-active features 1202 arranged in a square array and each surrounded by a Raman-inactive (e.g., gold) coating 1206. Each Raman-active feature 1202 is a 5-micron-diameter circle formed in the gold coating 1206 to expose an underlying Raman- active silicon substrate. The Raman-active features 1202 are surrounded by respective registration or fiducial markers 1204, which in this case are 125-micron-diameter ring-shaped or annular trenches formed in the gold coating to expose the underlying Raman-active silicon substrate. Each annular trench is concentric with the corresponding circle (Raman-active feature 1202).

[0128] FIG. 12B illustrates a method of making the crosshair calibration sample 1200 from a silicon wafer 1212. First (upper left), the silicon wafer 1212 is coated with a firstAttorney Docket No. PEND-014WO01 chromium layer 1214, a gold layer 1216, and a second chromium layer 1218. Photoresist 1220 is selectively deposited on the second chromium layer 1218 to define the Raman-active features and reference markers — in this case, 5-micron diameter circles and concentric 125-micron- diameter rings. Then (lower left) the second chromium layer 1218 is etched away except in the regions protected by the photoresist 1220 to expose the gold layer 1216. Additional gold is electroplated onto the exposed gold layer 1216 to form the Raman -inactive coating 1206 (upper right). Next, the photoresist 1220 is removed, and the chromium and gold layers 1214, 1216, 1218 that had been protected by the photoresist 1220 are etched away to expose portions of the Raman-active silicon substrate 1212. This fabrication process can be carried out using standard semiconductor processes to pattern Raman-inactive metal (here, gold) onto Raman-active silicon in such a way to expose only small areas of silicon as the Raman-active features.

[0129] FIGS. 13A-13D illustrates an automated alignment routine using the crosshair calibration target carried out by an inventive fieldable Raman microscope. In FIG. 13A, the user focuses the Raman pump beam and the camera roughly one of the crosshair calibration target’s Raman-active features using the crosshairs for aiming the Raman pump beam. Next, in FIG. 13B, the Raman microscope performs an automatic grid search followed by hill climbing to find maximum Raman signal. Each dot in FIG. 13B is a different point on the grid, with lighter dots representing higher intensity Raman signals. The lightest dot indicates the grid point with the peak Raman signal. Next, because the 5 pm Raman-active feature 1202 is occupies only a few pixels in the visible image, the Raman microscope’s processor identifies the registration mark — here, the 125 pm diameter annulus 1204, which is easier to locate by image analysis and has the 5 pm Raman-active feature 1202 at its center — as shown in FIGS. 13C and 13D.

[0130] In short, the Raman microscope adjusts the position of the optical subassembly 140 / 240 relative to the Raman-active feature so as to maximize the collected Raman signal. With the optical subassembly 140 / 240 positioned to maximize the collected Raman signal, the camera 154 / 254 acquires an image of the crosshair calibration sample 1200. The Raman microscope’s processor determines the location (Xcam, ycam ) of the Raman-active feature in the camera image and records this location as the position where Raman signal is increased or maximized. This calibration procedure is fast (less than 3 minutes) and automated. It can be performed by the end user in the field as part of regularly scheduled maintenance, or anytime the Raman microscope is subjected to mechanical shock.Focus Metric Calculation and Z-Axis OptimizationAttorney Docket No. PEND-014WO01

[0131] The machine vision subsystem also optimizes the focus of the Raman system by calculating focus metrics based on image sharpness. This optimization process involves image sharpness analysis, focus metric stack generation, focus determination, and alignment.Image Sharpness Analysis

[0132] For image sharpness analysis, the machine vision subsystem divides the captured images are divided into sub-tiles and calculates focus metrics for each tile based on the sharpness of the image. Various methods can be used to assess sharpness, such as gradientbased methods, contrast analysis, or edge detection techniques. These metrics indicate how well the image (or specific regions of it) is in focus.Focus Metric Stack (z-Stack) Generation

[0133] The camera 154 / 254 captures a stack of images as its focus moves incrementally along the z axis. The machine vision subsystem calculates focus metrics for each image in the stack, creating a set of focus values for each sub-tile across the different z axis positions.Optimal Focus Determination

[0134] The machine vision subsystem identifies the z axis position that corresponds to the maximum focus metric for each sub-tile, indicating the optimal focus point for that region of the image. The machine vision subsystem uses this information to adjust the z position of the sample relative to the optical subassembly 140 / 240 (e.g., by actuating the motorized sample stage 31), ensuring that the Raman microscope is focused precisely on the target located within that sub-tile. This also allows the Raman microscope to create a flattened image, which comprises the stitched-together maximally focused sub-tiles across the whole image stack.Alignment Considerations

[0135] The imaging and Raman systems are designed to be co-aligned, so the focus determined by the imaging system is generally applicable to the Raman system. Any small misalignment or systematic error between the two systems can be calibrated out using the Raman / camera calibration process described above, ensuring accurate focus for Raman measurements.System Electronics

[0136] An inventive fieldable Raman microscope is equipped with a comprehensive embedded electronics subsystem (electronics 112 in FIGS. IE, 2, and 3) configured to manage and control the various components of the system. This electronics subsystem ensures seamlessAttorney Docket No. PEND-014WO01 operation, precise control, and reliable data acquisition across the entire Raman setup. Its functions include laser drive control, temperature control, spectrometer readout, motion system control, power conversion and distribution, camera control and image acquisition, and interfacing external electronics, such as an auxiliary laptop or computer network.Laser Drive Control

[0137] The electronics 112 drive the laser source 102, ensuring stable and precise operation. It manages the laser source’s power output, enabling fine adjustments to the excitation intensity as determined by the measurement conditions. The electronics 112 also handle any modulation or pulsing of the laser source, if applicable, ensuring that the laser source 102 operates consistently within the desired parameters.Temperature Control

[0138] The Raman microscope may include one or several thermoelectric coolers (TEC) for temperature control, which can be used for maintaining the stability and performance of the laser source 102, as well as other temperature-sensitive components. The electronics 112 regulate the TECs, ensuring that the operating temperature remains within optimal ranges.Spectrometer Readout

[0139] The electronics 112 read out the signal from the spectrometer 106. This involves digitizing the analog signals received from the spectrometer’s sensor, typically a CCD or CMOS array. The electronics 112 ensure that the digitized data is accurate, with appropriate resolution and minimal noise, ready for further processing and analysis.Motion System Control

[0140] The electronics 112 controls the motorized motion system (e.g., motion controller 120 / 220 / 320, actuator 122 / 322, motorized sample stage 21), which includes the control of the x, y, and z axes. This control includes positioning the sample accurately under the Raman pump beam spot, focusing the Raman probe, and / or executing movements directed by the machine vision subsystem or user commands. The embedded electronics 112 and / or motion controller 120 / 220 / 320 provide drive signals to the actuators 122 / 322 and stage 21 and monitor their feedback to ensure precise and reliable movement.Power Conversion and Distribution

[0141] The Raman microscope includes power management electronics 114 that convert the input power to the various voltages used by the different subsystems. This involvesAttorney Docket No. PEND-014WO01 generating stable and isolated power rails for the laser source 102, TEC(s), spectrometer 106, motion controller 120 / 220 / 320, actuators 122 / 322, camera 154 / 254, and other components. The power conversion circuitry 114 reduces or minimizes noise and interference, ensuring that all subsystems operate reliably without cross-coupling or electrical noise affecting sensitive measurements.Camera Control and Image Acquisition

[0142] The embedded electronics 112 control the imaging camera 154 / 254, managing its operation, including exposure time, frame rate, and image acquisition. The electronics 112 handle the timing and synchronization of image capture, particularly in relation to other system operations such as laser excitation or motor movement. Acquired images are processed or transmitted to the system’s computer interface for display or further analysis by the machine vision subsystem.Computer Interface

[0143] The electronics 112 provide an interface to a computer, such as an embedded processor or an external computer (e.g., laptop 900) connected via a suitable communication protocol (e.g., USB, Ethernet, or wireless communication). This interface allows the computer to control the Raman microscope, send commands, receive data, and provide a user interface for system operation. The computer interface is also used for data storage, processing, and external communication, enabling integration with other software tools or databases.User Interface

[0144] An inventive fieldable Raman microscope is equipped with a versatile and intuitive user interface 110 (FIGS. IB, IE, 2, and 3) designed to provide users with comprehensive control and real-time feedback throughout the operation of the Raman microscope. The interface caters to both novice and expert users, offering a range of functionalities that enhance the usability, precision, and diagnostic capabilities of the Raman microscope. For instance, the user interface 110 can display the measured Raman spectra (either the Raman spectra obtained at each wavelength, difference Raman spectra, or both), a list of particles identified with morphology or color metrics, a list of targeted particles, measurement results (identified chemicals), and controls for the motion stages, image display (brightness, contrast, etc.), lightning (light sources, type, color, etc.), and other system parameters (integration time, laser power, etc.).

[0145] Attorney Docket No. PEND-014WO01Live Image Display

[0146] The user interface 110 can provide a live display of the image captured by the camera 154 / 254. This real-time visualization allows users to visualize the sample, perform rough positioning and focusing, and identify the target. More specifically, thanks to the user interface 110, a user can see the sample in real-time, which is particularly helpful for understanding the sample’s structure, texture, and features. A user can manually adjust the sample’s position using the x and y stages, aligning it roughly with the Raman probe by observing the live image on the user interface 110. The live image display assists a user in manually focusing the Raman microscope on the sample surface using the z-axis control, ensuring that the sample is properly aligned before more precise focusing routines are employed. A users can also explore the sample’s features and manually select points or regions of interest for Raman analysis by clicking on the image, allowing for intuitive and direct interaction with the sample.Simplified User Mode

[0147] In addition to the standard and expert modes, the user interface 110 provides a Simplified User Mode designed for minimally trained users. This mode provides a streamlined experience, allowing users to run pre-defined, application-specific routines targeted at specific types of analysis on particular samples. This mode simplifies operation by offering pre-defined routines and / or a guided workflow. For instance, a user can select from a set of pre-configured routines designed for specific applications, such as counterfeit drug pill analysis, particle analysis on swabs, and fingerprint imaging and particle identification. Counterfeit drug pill analysis is useful for identifying counterfeit medications by comparing Raman spectra to a database of known drugs. Particle analysis on swabs is tailored for analyzing particles collected on swabs, useful in forensic or environmental applications. Fingerprint imaging and particle identification involves automatically imaging fingerprints and identifying particles embedded within the fingerprints, enhancing forensic investigations. And with guided workflow, the user interface 110 guides a user through each step of the analysis process, reducing or minimizing manual adjustments or parameter selection, and ensuring consistent results for different operators.Expert Mode

[0148] For more advanced users, the user interface 110 provides includes an expert mode that provides granular control over the Raman microscope’s parameters. Expert modeAttorney Docket No. PEND-014WO01 offers flexibility and precision for those who desire detailed customization of the system's operation. Features of expert mode may include but are not limited to:• Manual Laser Power Adjustment: Users can manually set the laser power according to their sample;• Raman Frame Time Control: Users can adjust the integration time for the Raman signal, optimizing the signal-to-noise ratio based on the sample’s characteristics;• Camera Exposure Settings: Manual control of camera exposure settings allows for finetuning of the image capture process, ensuring clear and detailed images under various lighting conditions;• Manual Control of XYZ Stages: Users can directly control the motorized stages, adjusting the sample’s position and focus with high precision;• Laser On / Off Control: The user interface provides controls to turn the laser source on or off as desired, ensuring safe and controlled operation;• Live Raman Signal Acquisition and Display: Users can initiate live Raman data acquisition and view the Raman spectrum in real-time, making it easier to adjust parameters and monitor the measurement process dynamically; and• Other Controls: Additional expert controls may include settings for temperature control, filter adjustments, and system calibration routines, offering comprehensive management of all system subsystems.Target Selection Management

[0149] The user interface 110 includes features that allow users to manage target selection efficiently. After the machine vision subsystem identifies potential targets within the sample, the user interface 110 can automatically present a list of possible targets to the user, including relevant details such as size, location, and preliminary classification. Auser can select or unselect targets from this list, refining the set of points or regions to be analyzed. This feature provides users with the flexibility to focus on specific areas of interest while ignoring irrelevant or less critical regions.Measurement Control

[0150] The user interface 110 includes clear and accessible controls for starting and aborting Raman measurements. These controls ensure that users can easily initiate or halt a measurement sequence, providing flexibility and safety during operation. The user interfaceAttorney Docket No. PEND-014WO01110 can provide feedback on the measurement status, including progress indicators and any alerts or warnings related to the measurement process.Review and Reporting of Past Measurements

[0151] The user interface 110 enables users to review past measurements, offering a comprehensive view of the data collected during previous analyses. For instance, the user interface 110 can show images of the targets that were measured, allowing for visual confirmation and comparison. The user interface 110 can display the Raman spectra obtained from each target, with the ability to overlay multiple spectra for comparison. The user interface 110 can also display the results of chemical identification based on the Raman spectra, giving users detailed information about the compounds present in the analyzed targets. In additional, a user can generate a report from the user interface 110. For instance, a user can select specific measurements and compile them into a report summarizing the findings. The report may include images, spectra, and chemical identification data, formatted for easy interpretation and sharing. This feature is particularly useful for generating documentation for research, quality control, or forensic analysis.Self -Test Feature

[0152] To enhance reliability and user confidence, the user interface 110 includes a self-test feature. This feature causes the Raman microscope to perform a series of diagnostic checks on its subsystems, including the laser source, spectrometer, TEC(s), motion system, camera, and power supply. The self-test can identify potential issues, such as misalignments, power fluctuations, or sensor malfunctions, and provide users with detailed reports on the system's status. This proactive diagnostic tool helps prevent measurement errors, reduces downtime, and assists in maintenance by pinpointing specific components that may benefit from attention.Automated Spectral Processing and Chemical Identification

[0153] An inventive fieldable Raman microscope can automatically process and analyze the raw spectra collected during measurements. This ability transforms the Raman microscope from a simple data collection tool into a powerful analytical instrument capable of providing direct, actionable information, particularly for field use by non-scientists.

[0154] Upon acquiring a Raman spectrum, the Raman microscope’s processor immediately begins processing the raw spectral data. The processor uses sophisticated chemometric processes to compare the measured Raman spectrum with an extensive library ofAttorney Docket No. PEND-014WO01 reference spectra. The processor identifies the chemicals present in the sample. As a result, the output (e.g., presented via the user interface) is not just the raw spectral data but also the identities of the detected chemicals, providing users with clear and actionable information.

[0155] This automated chemical identification capability is particularly valuable for field applications where the Raman microscope may be operated by non-scientists. By converting complex spectral data into easily interpretable results, the Raman microscope ensures that users can make informed decisions without specialized knowledge in Raman spectroscopy or chemical analysis.

[0156] The Raman microscope’s ability to directly output the identities of the detected chemicals means that users can take immediate action based on the analysis. For example, in a security setting, the identification of an explosive compound can trigger a rapid response, while in pharmaceutical quality control, the detection of counterfeit drugs can lead to immediate intervention.

[0157] As mentioned above, the Raman microscope’s user interface 110 presents results in a way that highlights the identified chemicals along with relevant details such as concentration estimates or confidence levels in the identification. This makes the Raman microscope an invaluable tool in various field applications where rapid and accurate chemical identification is desired.

[0158] The Raman microscope’s electronics 112 include memory that stores customizable chemical libraries, enabling the Raman microscope to be tailored to specific applications. Users can update the Raman microscope’s library with new chemicals or modify existing entries, e.g., via the user interface 110 or an auxiliary computer (e.g., laptop 900), ensuring that the Raman microscope remains up-to-date with the latest chemical threats or analysis. The Raman microscope can also learn from new data, improving its spectral matching accuracy over time as it encounters more samples. This continuous learning capability enhances the Raman microscope’s reliability and effectiveness in real-world scenarios.

[0159] The software that powers the Raman microscope’s advanced processing can run on different types of computing hardware, including an embedded processor or an external computer, offering flexibility depending on the application and operational desires. A fully integrated, compact, and portable Raman microscope may include an embedded processor that handles all processing and analysis tasks itself. This embedded solution is especially useful where space and power efficiency are desired. Alternatively, the Raman microscope can beAttorney Docket No. PEND-014WO01 connected to an external computer (e.g., laptop 900), which may offer more processing power and flexibility. This setup is useful for applications that use more extensive data processing, storage, or additional software integration. The external computer can be connected via standard interfaces such as USB, Ethernet, or wireless connections, allowing seamless communication between the instrument and the computing platform.Motion System Architecture

[0160] The Raman microscope’s motion system is configured with flexibility and versatility in mind. Atypical conventional microscope has a fixed optical head for looking at a sample that can be moved. However, the portable Raman microscope 100 in FIG. IE has an optical head (optical subassembly 140) — comprising the objective and associated optical elements — mounted on an X-Y-Z stage (actuator(s) 122), while the sample sample 11 remains fixed. This innovative approach offers several advantages, particularly in scenarios where the sample 11 is large, delicate, or otherwise impractical to move.

[0161] This motion system design is particularly beneficial when working with large, bulky, or immovable samples. By keeping the sample stationary and moving the optical head (optical subassembly 140) instead, the Raman microscope can analyze samples that would be impractical or impossible to reposition. The Raman microscope can also be placed directly over large objects, with the optical head moving across the object’s surface to perform Raman analysis. This capability is useful for applications such as scanning surfaces for trace chemical contamination or analyzing large artifacts in situ.Optical Head Mounted on X-Y-Z Stage

[0162] The moving optical subassembly 140 in FIG. IE can move independently along each of three axes. The actuator(s) 122 can move the optical subassembly 140 laterally across the sample surface in the X and Y directions. This lateral motion allows the system to scan across different areas of the sample 11 without repositioning the sample itself. The actuator(s) 122 can also move the optical subassembly 140 vertically, i.e., in the Z direction, allowing for precise focusing of the optical head subassembly 140 the sample 11. This motion of the optical subassembly 140 is in addition to independent (fine) focus control of the Raman pump beam and camera 154 provided by the NIR optics 146 and visible optics 152, respectively, and by the visible / NIR optics 150. The actuator(s) 122 ensures that the Raman probe 142 can be accurately aligned with the sample surface or specific features of interest, maintaining optimal focus during analysis.Attorney Docket No. PEND-014WO01Contained Motion System

[0163] The entire motion system — optical subassembly 140, motion controller 120, and actuator(s) 122 — can be fully contained within the housing 130 of the Raman microscope 100. This is particularly advantageous for maintaining a controlled environment inside the housing 130 and protecting sensitive components from external contaminants.Baffles and Sliding Plates

[0164] Flexible bellows 133, baffles, and / or sliding plates accommodate the motion of the optical subassembly 140 while keeping the housing 130 sealed. These bellows 133 allow the optical subassembly 140 to move freely within the housing 130 while preventing dust, debris, or other contaminants from entering the housing 130. The bellows 133 help to maintain the instrument's performance and reliability, especially in field environments.Flexible Mounting

[0165] For even more versatility, the Raman microscope can be mounted on an articulated arm. This arm can be motorized or manually adjustable, allowing the Raman microscope to be positioned with precision over various parts of an object. The articulated arm provides additional flexibility in reaching difficult-to-access areas, making the Raman microscope adaptable to a wide range of sample geometries.

[0166] FIGS. 1A and 1C show several other mounting options for fieldable Raman microscopes. For example, a Raman microscope 100 may have feet or legs 132 or mounts 134 or brackets to secure it in place during analysis or to enhance its stability. The legs or feet 132 extend from the housing 130 and allow the Raman microscope to be placed securely on top of a flat surface, such as a table or workbench. This is particularly useful when scanning large surfaces for contamination, as it keeps the Raman microscope remains stable and aligned during analysis.

[0167] An L-bracket accessory may be used to hold both the Raman microscope and the sample in a rigid configuration, e.g., as shown at right in FIG. 1C. This setup is beneficial for ensuring consistent alignment and reducing or minimizing vibrations or movement during sensitive measurements. The Raman microscope 100 can also fit in or be coupled to any of a variety of sample holders designed to accommodate different sample shapes and sizes. These holders secure the sample in place, preventing movement during analysis and ensuring consistent and reliable results. Whether dealing with flat surfaces, cylindrical objects, orAttorney Docket No. PEND-014WO01 irregular shapes, these custom holders enhance the Raman microscope’s adaptability and ease of use.

[0168] Versions of the Raman microscope with this flexible motion system architecture, with a mobile optical head and stationary sample, are highly versatile and well- suited for a wide range of applications, especially those involving large or immovable samples. The ability to position the Raman microscope directly over a sample and move the optical head with precision, combined with the protective design of the instrument’s interior, ensures reliable, high-quality Raman analysis in diverse environments.Portable Design Features and Strategies

[0169] Inventive Raman microscopes are portable, enabling their use in field applications and various on-site environments. Several design strategies and features ensure that the Raman microscope is compact, rugged, and easy to transport while maintaining high performance and reliability. These features include a miniature grating spectrometer with a fixed grating, which provides 8-10 cm1resolution with small size, balancing performance with portability. Small, diced thin-film filters reject reflected and / or scattered light at the Raman pump beam wavelength(s), reducing or minimizing the footprint of the optical system, in turn contributing to the overall compactness of the instrument. The laser source can include a semiconductor laser housed in a compact telecom-style package, which is both space-efficient and reliable. And the Raman microscope’s electronics, including laser drive, spectrometer readout, TEC controllers, and other system-level controls, are compact. This integration reduces the overall size of the electronics housing and eliminates bulky external components.

[0170] The Raman microscope has few heat sources and uses effective heat management strategies to avoid large heat dissipation structures. Compact thermal management solutions, such as small heat sinks and efficient TECs, keep the Raman microscope at the desired temperature without increasing size.

[0171] The Raman microscope can be transported easily, e.g., in a rugged storm case for protection during travel. The Raman microscope can be easily pulled out of its case and made ready for use with minimal to no setup. This ensures that users can quickly deploy the Raman microscope in the field, saving time and effort.

[0172] The Raman microscope is built to withstand the rigors of field use, with a focus on durability and stability. Most of the Raman microscope’s optical elements are securely fixed in place and glued to prevent long-term drift and misalignment. This ensures that the RamanAttorney Docket No. PEND-014WO01 microscope maintains its calibration and alignment even after extended use and transport. The use of a spectrometer with a fixed grating and a linear sensor eliminates moving parts, reducing the risk of position drift and mechanical failure over time. Optical components, such as the laser and spectrometer, are mounted on rigid substrates. This enhances stability and reduces the likelihood of misalignment due to mechanical stresses or shocks. Optical subsystems are shielded from shocks using shock damping mounting structures, such as grommets, springs, or flexure frames. These structures absorb impacts and vibrations, protecting the delicate optical components during transport.Stage Locking Mechanism

[0173] As described above, a Raman microscope can include motorized stages for automation. These stages can be protected during transport by a locking mechanism that holds the stages in place. This locking mechanism unwanted motion and strain on the motion system components, such as rails and bearings, ensuring they remain in optimal condition. During Raman microscope shutdown, the stages move to predefined location(s). One or more screws can be turned by hand to securely attach the moving optical head to the instrument enclosure (housing 130). The Raman microscope may include a motorized latch or an electronically actuated locking mechanism, such as an electromagnet, instead of or in addition to the screws to automatically secure the optical head to the housing. This simplifies the locking procedure and reduces the risk of user error during the locking procedure.

[0174] FIGS. 14A-14E show different views of a rack and pinion that can lock an inventive fieldable Raman microscope in three axes (x,y, and z). The rack and pinion is locked in FIG. 14E and unlocked in FIGS. 14A-14D. This rack and pinion includes a pinion 1406 that is attached to a small, geared stepper motor 1422 to drive two perpendicular gear racks 1402 and 1404 simultaneously. A locking pin 1408 at the end of each gear rack 1402, 1404 slides into a corresponding hole or slot. One locking pin 1408 locks the x and j’ axes; the other locking pin 1408 locks the z axis.

[0175] The Raman microscope may also include a small reserve power source to ensure that the stages can be locked even in the event of a sudden power loss. This reserve power source can be an internal battery or supercapacitor that allows the system to home to the storage position and lock the stages. This feature adds an extra layer of safety, ensuring the Raman microscope is ready for transport whenever it is turned off.MEMS Mirror for Rapid Raman Spot AdjustmentAttorney Docket No. PEND-014WO01

[0176] As shown in FIGS. IE 2, and 3, the Raman microscope 100 / 200 / 300 may include a micro-electro-mechanical systems (MEMS) mirror 144 / 244 integrated into the Raman optical path to enable precise and rapid adjustment of the position of the Raman pump beam spot to target a specific point on the sample. The MEMS mirror 144 / 244, which can tilt along multiple axes, allows the Raman microscope to dynamically control the direction of the Raman pump beam, effectively steering the Raman pump beam spot across the sample surface with high precision.

[0177] The MEMS mirror 144 / 244 can be used to swiftly adjust the position of the Raman pump beam spot on the sample, allowing for fine-tuning of the Raman pump beam’s aim before or during a measurement. This capability is particularly valuable for ensuring that the Raman pump beam is accurately targeted on a specific point of interest, especially in scenarios where the initial alignment based on large motion of the xyz stage may be slightly off.

[0178] The Raman microscope can actively use the MEMS mirror 144 / 244 to compensate for aiming errors caused by vibrations or other external disturbances. During the measurement process, the MEMS mirror 144 / 244 can continuously adjust the Raman pump beam spot position in real-time, ensuring that the Raman pump beam spot remains fixed on the target area despite any movements of the Raman microscope or sample. This feature enhances the robustness of the system in field applications where vibrations and environmental instability may otherwise affect measurement accuracy.

[0179] By integrating feedback mechanisms, such as real-time image analysis or laser spot tracking, the Raman microscope can detect any misalignment or drift of the Raman pump beam spot relative to the target. The MEMS mirror 144 / 244 can automatically correct these errors, maintaining the Raman pump beam’s focus on the desired location and improving the reliability and precision of the Raman measurements.Fine Aiming of the Raman Laser

[0180] Achieving precise aiming and focusing of the Raman pump beam onto a specific target, such as a particle deposited on a surface, enables accurate and reliable Raman measurements. Given the potential for small positional errors during the movement of the XYZ stage, it is useful to implement strategies that correct these errors without significantly increasing the Raman microscope’s complexity or cost.Attorney Docket No. PEND-014WO01

[0181] After the XYZ stage (e.g., for the moving optical subassembly 140, the sample, or the entire Raman microscope) moves the Raman pump beam into the approximate position and focus relative to the target on the sample, there may still be small positional errors due to mechanical tolerances or stage inaccuracies. While encoders can be employed on the XYZ stages to enhance accuracy, this approach can add significant complexity and cost to the system. An alternative, cost-effective solution is to correct these small errors with a feedback loop involving image acquisition and analysis.

[0182] Once the initial motion of the XYZ stage is complete, the camera 154 / 254 acquires an image of the target area. The Raman microscope’s processor uses pattern recognition (template matching) to find the precise location of the target, such as a particle of interest, in this image. By comparing the actual position of the target in the newly acquired image to its expected position, the Raman microscope’s processor can calculate any positional error. The Raman microscope can make fine adjustments to the position of the XYZ stages based on the calculated positional error. These adjustments ensure that the Raman pump beam is accurately focused on the target, compensating for any initial misalignment. This feedback mechanism allows for high-precision targeting without more expensive or complex hardware solutions.

[0183] Mechanical backlash — small but significant delays or deviations that occur when reversing the direction of a stage movement — can introduce further errors in positioning. The Raman microscope can mitigate the effects of backlash using a consistent directional approach and / or image-based corrections to ensure consistent and accurate stage movements. The consistent directional approach reduces or minimizes the impact of backlash by approaching the target position from the same direction for every measurement. By controlling the direction of the final movement, the system can reduce variability in positioning, as the backlash is more predictable and can be compensated for more effectively. Combining backlash mitigation with image-based error correction provides a robust method for fine-tuning the position of the Raman pump beam. By systematically approaching the target from the same direction and making real-time corrections based on image analysis, the Raman microscope achieves a high level of precision in laser targeting, even in the presence of mechanical imperfections.Applications of a Fieldable Raman MicroscopeAttorney Docket No. PEND-014WO01

[0184] Inventive fieldable Raman microscopes described here are versatile tools with a wide range of potential applications across various fields, leveraging its advanced imaging, particle analysis, and chemical identification capabilities. These applications include automated swab reading, narcotics analysis, Surface-Enhanced Raman Scattering (SERS) coupon reading, assay reading, fingerprint analysis, and trace chemical analysis.

[0185] For instance, a Raman microscope can be used as an automated swab reader, similar to the type of swabs used by Ion Mobility Spectrometry (IMS) instruments. The swab material, texture, and color can be customized to increase or maximize imaging contrast, enhancing the Raman microscope’s ability to resolve and identify particles collected on the swab. The term “swab” encompasses a broad range of sampling techniques, including “Q-tip” type samplers, tape, or other adhesives, and can vary in form, being smooth / flat or woven.

[0186] In this application, the swab sample (e.g., swab coupon) is inserted into the Raman microscope, possibly through a sampling accessory, compartment, or chamber with an interlocked door that shields ambient light and ensures that no laser light escapes, maintaining a Class 1 laser safety rating. The microscope connects mechanically and electronically to this accessory, ensuring that the laser source cannot be activated unless the door is securely locked. The Raman microscope automatically images the swab, locates particles on the swab, sorts and triages the particles for analysis, and then successively analyzes the particles. The Raman microscope ultimately reports whether a threat chemical has been identified, providing actionable information in security or forensic contexts.

[0187] The Raman microscope can also function as a narcotics analyzer, suitable for testing seized illicit substances in powder or pill form. As with the swab reader application, the Raman microscope can include or be coupled to a sample compartment that ensures Class 1 laser safety. A user inserts a sample of the seized substance into the sample compartment, potentially while still in a transparent bag to preserve evidence. The Raman microscope may discern different spots on the pill or powder based on color contrast or measure multiple spots in a predefined pattern (e.g., grid).

[0188] For more detailed analysis, the sample can be crushed and the resulting powder dispersed on a flat, mirror-like substrate. This approach allows the Raman microscope to analyze individual particles, increasing the likelihood of detecting chemicals present in low concentrations. The Raman microscope can use a smart particle triage process to cluster particles based on visible features (color, shape, size, reflectivity, etc.) and selectively analyzeAttorney Docket No. PEND-014WO01 a few particles within each cluster. This can help increase the likelihood of measuring a particle whose composition is dominated by an ingredient that is otherwise in low concentration in the bulk sample. This method is particularly useful for identifying synthetic drugs, such as fentanyl analogs, often found in low concentrations (<1%) in illicit drug samples. This application can be generalized to any analysis of solid mixtures where identifying low-concentration ingredients is desirable, such as detecting contaminants in food.

[0189] A Raman microscope can be used to analyze SERS coupons. It can perform automated measurements at multiple locations on a SERS coupon, averaging the responses to account for spatial variations in intensity and spectrum. This ensures a more accurate and representative analysis of the sample's chemical composition.

[0190] A Raman microscope can serve as an assay reader, providing spatial discrimination through Raman measurements at different locations on a sample. Each location may have functionalization or other mechanisms that favor the binding of specific chemicals, enhancing selectivity. This application can be paired with SERS to amplify the Raman signal. Combining functionalization or other molecular selectivity with Raman spectroscopy can help reduce the risk of false alarms and improve the overall accuracy of chemical identification.

[0191] A Raman microscope can be paired with a spatial separation technique, such as chromatography, to allow automated Raman measurements along the direction of spatial separation. This setup can analyze the chemical composition of separated components as they move along the chromatographic path, providing detailed information about the sample’s chemical makeup.

[0192] A Raman microscope can be used to obtain combined biometric and chemical information from fingerprints. By imaging a fingerprint and analyzing the chemical composition of particles embedded within it, e.g., as shown in FIG. IB, the instrument can associate identity with activity. For example, it can link the fingerprint of a bomb maker with trace explosives found on the fingerprinted surface, providing valuable forensic evidence.

[0193] A Raman microscope can identify trace chemical compositions on surfaces, particularly when presented as individual particles. This application is especially useful for detecting and analyzing minute quantities of hazardous materials, contaminants, or residues on various surfaces, making it valuable for security, environmental monitoring, and industrial quality control.Sampling Accessory for Class 1 Laser SafetyAttorney Docket No. PEND-014WO01

[0194] FIGS. 15A-15C show a specialized sampling accessory 1500 that be securely attached to a Raman microscope 300 to ensure the highest standards of laser safety and enhance versatility. This sampling accessory 1500 is engineered to fully screen ambient light and prevent any laser light from escaping the sample compartment, thereby rendering the Raman microscope compliant with Class 1 laser safety standards.

[0195] The sampling accessory 1500 creates a controlled environment within the sample compartment, preventing external light from interfering with the Raman measurements and preventing laser light from escaping and posing a safety hazard. The sampling accessory 1500 is designed with materials and structures that effectively block ambient light and contain laser radiation within the compartment. The sampling accessory 1500 is constructed with lightabsorbing materials and seals that prevent any external light from entering the sample compartment. The sampling accessory 1500 completely contains the Raman pump beam within the sample compartment, ensuring that no laser radiation escapes into the surrounding environment. This containment provides Class 1 laser safety certification, which is the safest classification, indicating that the laser source is not hazardous under normal operating conditions.

[0196] The sampling accessory 1500 can feature a sample compartment where samples can be securely placed for analysis. This sample compartment is designed with user convenience and safety in mind. The sample compartment is accessed through a door 1502 or another type of opening that allows the user to insert the sample easily. The door 1502 can close tightly, maintaining the light-tight and laser-safe environment inside the sample compartment. The sampling accessory 1500 may include gaskets or other sealing mechanisms to ensure a secure closure that blocks light and laser radiation effectively.

[0197] To further ensure user safety, the sampling accessory 1500 can be electronically connected to the Raman microscope 300, allowing for the implementation of an interlock. This interlock provides an additional layer of protection by preventing the Raman laser from being activated unless the sample compartment door is properly closed. The interlock can be integrated into the Raman microscope’s control electronics (electronics 112) and directly linked to the sampling accessory. When the sample compartment door 1502 is open or not closed securely, the interlock prevents the laser (laser source 102) from being turned on, ensuring that no laser light can escape. Only when the door 1502 is fully closed and the sample compartment is sealed does the interlock allow the laser to be activated. This interlock helps maintain Class 1 laser safety, as it ensures that the laser can operate only when all safety conditions are met.Attorney Docket No. PEND-014WO01This interlock provides peace of mind for operators, particularly in environments where laser safety is a priority, such as in public or industrial settings.

[0198] The sampling accessory 1500 is easily attachable and detachable from the Raman microscope 300, allowing for quick setup and changeover between different types of samples or analytical tasks. Despite its robust safety features, the sampling accessory 1500 is user-friendly, ensuring that even users with minimal training can operate the Raman microscope 300 and sampling accessory safely. The sampling accessory 1500 integrates seamlessly with the Raman microscope 300, both mechanically and electronically, providing a cohesive and coherent user experience. The sampling accessory 1500 does not impede the Raman microscope’s functionality or ease of use while enhancing safety and measurement integrity. The sampling accessory 1500 can be adapted to different types of samples and analytical desires, making it a versatile addition to the Raman microscope 300. Whether analyzing solid samples, powders, or other materials, the sampling accessory 1500 ensures that the Raman microscope 300 can be used safely and effectively across a wide range of applications.User-Friendly Design Features

[0199] An inventive fieldable Raman microscope may include one or more of several user-friendly features that enhance its ease of use, making it accessible to both novice and experienced operators. These features include the integration of a camera, motorized stages, and auto-calibration capabilities, all of which contribute to the Raman microscope’s automation and precision.

[0200] The Raman microscope’s high-resolution camera (camera 154 / 254) images the sample, enabling a range of automated machine vision routines that simplify and streamline the analysis process. The camera can measure image sharpness, providing data for an autofocus system that ensures the Raman laser is accurately focused on the sample, improving the quality and consistency of measurements. The camera can be used to automatically locate particles or features of interest on the sample. This automation reduces manual targeting, making the Raman microscope easier and faster to use while improving accuracy. The camera also allows for the acquisition and storage of images of the sample, providing a visual record of the analysis that can be reviewed or included in reports, further enhancing the instrument’s utility.Attorney Docket No. PEND-014WO01

[0201] The Raman microscope’s motorized X-Y-Z stages facilitate precise and automated motion control, contributing to both the ease of use and the versatility of the Raman microscope. The motorized stages enable automated motion, such as auto-aiming on individual particles and auto-focusing on specific areas of the sample. This automation reduces operator intervention, making the Raman microscope more efficient and user-friendly.

[0202] To protect the motorized stages during transport, the Raman microscope includes locks for keeping the stages in place, preventing movement and potential damage. This ensures that the Raman microscope remains ready for use upon arrival at a new location. The motorized stages also allow for the automation of the calibration procedures described above. By automating these processes, the instrument maintains accuracy and reliability with minimal user input, further enhancing its ease of use. The Raman microscope’s auto-calibration features ensure that the Raman microscope remains calibrated and ready for use, reducing manual adjustments and the potential for user error.Analyzing Complex Powder Mixtures

[0203] Raman spectroscopy is a powerful tool for chemical identification and is widely used in laboratories to analyze solid, liquid, or powdered substances. In conventional fieldbased systems, chemical identification is typically performed by acquiring a single bulk- averaged Raman spectrum from a heterogeneous sample. This spectrum is then compared against a library of known reference spectra using regression or pattern-matching techniques. While these approaches are often adequate for relatively simple mixtures containing known compounds, they suffer significant limitations when confronted with complex, poorly characterized materials, especially those containing unknown components (not represented in the library of reference spectra).

[0204] An inventive fieldable Raman microscope is particularly well-suited for identifying chemical components within a heterogeneous powder mixture using spatially resolved particle analysis, automated sampling, and unsupervised spectral interpretation. It can detect unknown components, disaggregate the complexity of mixtures by targeting individual particles within the sample, and produce chemically meaningful outputs without an expert user. The system can perform trace-level analysis in field applications such as forensic drug detection and hazardous material screening.

[0205] Heterogenous sample analysis with a fieldable Raman microscope can include three stages: (1) sample preparation and spatial separation; (2) automated Raman spectraAttorney Docket No. PEND-014WO01 acquisition; and (3) (automated) spectral computation and interpretation. Sample preparation and spatial separation begins with physically processing a small quantity of a heterogeneous sample to improve analyzability. If the sample is solid, it is crushed into a fine powder (1602). The powder is dispersed on a flat substrate (1604), such as aluminum foil, to reduce or avoid clumping and ensure individual particles are spatially separated. The resulting dispersed powder should include particles with diameters ranging from 5 pm to a few hundred micros. This size range ensures that each particle is more likely to contain a limited number of chemical species at relatively high concentration, compared to the complex average of the bulk sample. This sample preparation facilitates spatially resolved measurements, where each target is assumed to be simpler than the whole.

[0206] FIG. 16 illustrates a heterogenous sample analysis process executed by a fieldable Raman microscope. The Raman microscope uses a machine vision system (camera and processor) equipped with one or more contrast mechanisms, including adjustable illumination (including adjustable color, direction, and incidence angle), to locate and measure Raman spectra of particles within its field of view (1610). To obtain focus across a curved or tilted substrate, the system performs z-stack imaging by acquiring images of the sample at multiple positions along the optical axis and creating a sharp composite image from the most in-focus subregion of each depth slice (1612). The microscope calculates a corresponding z- height for each image subregion and uses the z-heights to generate a 3D profile of the sample surface (and thus 3D coordinates of each particle)(l 614). Using this 3D map, the microscope automatically navigates the Raman probe beam across the sample surface to measure a specified number of particles (e.g., 100 particles each with a diameter of 20-50 pm)(1616). The microscope measures the Raman spectrum of each particle (1618) in a consistent and reproducible manner without human intervention.

[0207] After collecting the Raman spectra from the particles, the microscope or a processor coupled to the microscope corrects the backgrounds of the Raman spectra (1620), optionally filters noise from the Raman spectra (1622), and normalizes the Raman spectra (1624). The microscope or processor can estimate how many distinct component spectra are present in the Raman spectra based on variations across the Raman spectra for the different particles (1626). The number of distinct component spectra serves as the basis for decomposition.

[0208] In addition to or instead of averaging the data and / or matching to known spectra directly, the microscope or processor can use internal variability across particles to deriveAttorney Docket No. PEND-014WO01 independent component spectra from the Raman spectra (1628). These independent component spectra represent the likely pure chemical constituents within the sample. The microscope or processor can compare the measured Raman spectra and / or the derived component spectra to a reference spectral library to identify known substances and report unknowns in the absence of matching library entries (1630). The microscope or processor outputs a set of recovered spectral components and a mapping of their presence across measured particles (1632). This offers a more granular and informative picture of sample composition than bulk analysis.

[0209] Analyzing heterogeneous samples using several spatially separated measurements instead of a single, bulk-averaged measurement offers several benefits for field applications, including but not limited to: improved specificity by isolating individual particles rather than averaging mixtures; ability to detect unknowns (components not in a pre-existing spectral database); automation, enabling non-scientific personnel to perform complex chemical analysis; high throughput, with the ability to analyze dozens to hundreds of particles in minutes; and adaptability to field-use with portable instrumentation. This is especially useful for analyzing seized drugs in forensic or law enforcement settings, identifying unknown powders in hazmat or military environments, and environmental or pharmaceutical screening involving compositional heterogeneity.

[0210] FIGS. 17A-17C and 18A-18D illustrate measurements and the result of unsupervised decomposition by a fieldable Raman microscope of one hundred particles of a Goody’s powder, which an off-the-shelf drug containing acetaminophen, aspirin, caffeine, and lactose, among other main components. FIG. 17A shows an image of the sample obtained by the Raman microscope with the particles marked by X’s. The Raman microscope identified and located the particles using the techniques described above. FIG. 17B shows more detail of one particle, including a template matched to the particle (left), a closeup of the particle (center), and the particle marked by an X in an image of the entire sample (right). FIG. 17C shows the particle’s Raman spectrum (bottom trace) decomposed into Raman spectra of the particle’s principle components, acetylsalicylic acid (top trace), acetaminophen (upper middle trace), and alpha-lactose (lower middle trace). The Raman microscope generated this trace and the results in the tables below by analyzing the measurements from each particle independent of one another by doing a regression against its library of reference Raman spectra.

[0211] FIGS. 18A-18D show Raman spectra of principal components from another sample of Goody’s powder obtained with a fieldable Raman microscope. Here, the Raman microscope measured the Raman spectra of one hundred particles, and analyzed the entireAttorney Docket No. PEND-014WO01 dataset to extract pure spectral components indicative of the composition of the sample. The microscope identified the spectral components against a library of reference Raman spectra. In this example, each spectral component was in the Raman microscope’s library, but the Raman microscope didn’t use its library for decomposition and would have returned independent pure spectral components even if the spectral components were unknown.TABLE 1 : Result Detection Rate (*)Attorney Docket No. PEND-014WO01Particle SummaryTABLE 2: Particle SummaryAttorney Docket No. PEND-014WO01Conclusion

[0212] While various inventive embodiments have been described and illustrated herein, those of ordinary skill in the art will readily envision a variety of other means and / or structures for performing the function and / or obtaining the results and / or one or more of the advantages described herein, and each of such variations and / or modifications is deemed to be within the scope of the inventive embodiments described herein. More generally, those skilled in the art will readily appreciate that all parameters, dimensions, materials, and configurations described herein are meant to be exemplary and that the actual parameters, dimensions, materials, and / or configurations will depend upon the specific application or applications for which the inventive teachings is / are used. Those skilled in the art will recognize or be able to ascertain, using no more than routine experimentation, many equivalents to the specific inventive embodiments described herein. It is, therefore, to be understood that the foregoing embodiments are presented by way of example only and that inventive embodiments may be practiced otherwise than as specifically described. Inventive embodiments of the present disclosure are directed to each individual feature, system, article, material, kit, and / or method described herein. In addition, any combination of two or more such features, systems, articles, materials, kits, and / or methods, if such features, systems, articles, materials, kits, and / or methods are not mutually inconsistent, is included within the inventive scope of the present disclosure.

[0213] Also, various inventive concepts may be embodied as one or more methods, of which an example has been provided. The acts performed as part of the method may be ordered in any suitable way. Accordingly, embodiments may be constructed in which acts are performed in an order different than illustrated, which may include performing some acts simultaneously, even though shown as sequential acts in illustrative embodiments.

[0214] All definitions, as defined and used herein, should be understood to control over dictionary definitions, definitions in documents incorporated by reference, and / or ordinary meanings of the defined terms.

[0215] The indefinite articles “a” and “an,” as used herein, unless clearly indicated to the contrary, should be understood to mean “at least one.”

[0216] The phrase “and / or,” as used herein, should be understood to mean “either or both” of the elements so conjoined, i.e., elements that are conjunctively present in some cases and disjunctively present in other cases. Multiple elements listed with “and / or” should beAttorney Docket No. PEND-014WO01 construed in the same fashion, i.e., “one or more” of the elements so conjoined. Other elements may optionally be present other than the elements specifically identified by the “and / or” clause, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example, a reference to “A and / or B”, when used in conjunction with open-ended language such as “comprising” can refer, in one embodiment, to A only (optionally including elements other than B); in another embodiment, to B only (optionally including elements other than A); in yet another embodiment, to both A and B (optionally including other elements); etc.

[0217] As used herein, “or” should be understood to have the same meaning as “and / or” as defined above. For example, when separating items in a list, “or” or “and / or” shall be interpreted as being inclusive, i.e., the inclusion of at least one, but also including more than one, of a number or list of elements, and, optionally, additional unlisted items. Only terms clearly indicated to the contrary, such as “only one of’ or “exactly one of,” “consisting of,” will refer to the inclusion of exactly one element of a number or list of elements. In general, the term “or” as used herein shall only be interpreted as indicating exclusive alternatives (i.e., “one or the other but not both”) when preceded by terms of exclusivity, such as “either,” “one of,” “only one of,” or “exactly one of.” “Consisting essentially of’ shall have its ordinary meaning as used in the field of patent law.

[0218] As used herein, the phrase “at least one,” in reference to a list of one or more elements, should be understood to mean at least one element selected from any one or more of the elements in the list of elements, but not necessarily including at least one of each and every element specifically listed within the list of elements and not excluding any combinations of elements in the list of elements. This definition also allows that elements may optionally be present other than the elements specifically identified within the list of elements to which the phrase “at least one” refers, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example, “at least one of A and B” (or, equivalently, “at least one of A or B,” or, equivalently “at least one of A and / or B”) can refer, in one embodiment, to at least one, optionally including more than one, A, with no B present (and optionally including elements other than B); in another embodiment, to at least one, optionally including more than one, B, with no A present (and optionally including elements other than A); in yet another embodiment, to at least one, optionally including more than one, A, and at least one, optionally including more than one, B (and optionally including other elements); etc.

[0219] As used herein, all transitional phrases such as “comprising,” “including,” “carrying,” “having,” “containing,” “involving,” “holding,” “composed of,” and the like are toAttorney Docket No. PEND-014WO01 be understood to be open-ended, i.e., to mean including but not limited to. Only the transitional phrases “consisting of’ and “consisting essentially of’ shall be closed or semi-closed transitional phrases, respectively, as set forth in the United States Patent Office Manual of Patent Examining Procedures, Section 2111.03.

Claims

Attorney Docket No. PEND-014WO01CLAIMS1. A portable Raman microscope for inspecting a sample, the portable Raman microscope comprising: a housing; an array of light sources, mounted to the housing, to illuminate the sample with light at each of a plurality of different illumination conditions; a camera, contained within the housing, to acquire images of the sample under different illumination conditions; a processor, operably coupled to the array of light sources and the camera, to automatically identify a target for inspection from the images of the sample; a laser source, contained within the housing, to illuminate the target with at least one Raman pump beam; and a spectrometer, contained within the housing, to measure a Raman spectrum of light scattered by and / or emitted from the target in response to the at least one Raman pump beam.

2. The portable Raman microscope of claim 1, wherein the array of light sources comprises light sources arranged around an aperture in the housing through which the camera images the sample, the laser source illuminates the target, and the spectrometer receive the light scattered by and / or emitted from the target.

3. The portable Raman microscope of claim 2, wherein the processor is configured to actuate different light sources in the array of light sources to produce the plurality of different illumination conditions.

4. The portable Raman microscope of claim 1, wherein the array of light sources comprises a light source arranged to illuminate the sample at a glancing angle.

5. The portable Raman microscope of claim 1, wherein the array of light sources comprises at least one light source configured to illuminate the sample with red light, at least one light source configured to illuminate the sample with green light, and at least one light source configured to illuminate the sample with blue light and the images of the sample include a monochromatic red image, a monochromatic green image, and a monochromatic blue image.Attorney Docket No. PEND-014WO016. The portable Raman microscope of claim 5, wherein the processor is configured to automatically identify the target by: thresholding the monochromatic red image, the monochromatic green image, and the monochromatic blue image; flattening the monochromatic red image, the monochromatic green image, and the monochromatic blue image into a color-flattened image; and locating at least one particle in the color-flattened image.

7. The portable Raman microscope of claim 1, wherein the camera is configured to acquire the images at different focal planes and the processor is configured to generate a composite image from the images at the different focal planes.

8. The portable Raman microscope of claim 1, further comprising: an optical head, contained in the housing and in optical communication with the camera and the laser source, to position a conjugate plane of the camera and to focus the at least one Raman pump beam on the sample; and an actuator, contained in the housing, to move the optical head in up to three dimensions with respect to the sample.

9. The portable Raman microscope of claim 8, wherein the optical head has a first numerical aperture for the at least one Raman pump beam and a second numerical aperture smaller than the first numerical aperture for the camera.

10. The portable Raman microscope of claim 8, wherein the actuator is configured to move the optical head so as to shift a focal plane of the camera and / or a focal plane of the at least one Raman pump beam.

11. The portable Raman microscope of claim 1, wherein the laser source and the spectrometer are configured to perform a shifted excitation difference Raman spectroscopy (SERDS) measurement of the target.

12. A method of inspecting a sample, the method comprising: illuminating the sample with light from an array of light sources mounted to a housing at each of a plurality of different illumination conditions; acquiring images of the sample under different illumination conditions with a camera contained within the housing;Attorney Docket No. PEND-014WO01 identifying a target for inspection from the images of the sample; illuminating the target with at least one Raman pump beam emitted by a laser source contained within the housing; and measuring, with a spectrometer contained within the housing, a Raman spectrum of light scattered by and / or emitted from the target in response to the at least one Raman pump beam.

13. The method of claim 12, wherein the array of light sources comprises light sources arranged around an aperture in the housing through which the camera images the sample, the laser source illuminates the target, and the spectrometer receive the light scattered by and / or emitted from the target.

14. The method of claim 12, wherein illuminating the sample with light from the array of light sources comprises actuating different light sources in the array of light sources to produce the plurality of different illumination conditions.

15. The method of claim 12, wherein illuminating the sample with light from the array of light sources comprises illuminating the sample at a glancing angle.

16. The method of claim 12, wherein illuminating the sample with light from the array of light sources comprises illuminating the sample with red light, green light, and blue light and acquiring the images of the sample comprises acquiring a monochromatic red image, a monochromatic green image, and a monochromatic blue image.

17. The method of claim 16, wherein identifying the target comprises: thresholding the monochromatic red image, the monochromatic green image, and the monochromatic blue image; flattening the monochromatic red image, the monochromatic green image, and the monochromatic blue image into a color-flattened image; and locating at least one particle in the color-flattened image.

18. The method of claim 12, wherein acquiring the images comprises acquiring the images at different focal planes and further comprising: generating a composite image from the images at the different focal planes.

19. The method of claim 12, further comprising:Attorney Docket No. PEND-014WO01 positioning a conjugate plane of the camera and focusing the at least one Raman pump beam on the sample.

20. The method of claim 19, further comprising: shifting a focal plane of the camera and / or a focal plane of the at least one Raman pump beam.

21. The method of claim 12, wherein measuring the Raman spectrum of light comprises performing a shifted excitation difference Raman spectroscopy (SERDS) measurement of the target.

22. A method of measuring a Raman spectrum of a particle on a substrate, the method comprising: acquiring images of the particle and the substrate under each of a plurality of different illumination conditions; identifying, based on the images, a first region containing the particle and the substrate and a second region containing only the substrate; measuring a Raman spectrum of the first region; measuring a Raman spectrum of the second region; and determining the Raman spectrum of the particle based on a difference between the Raman spectrum of the first region and the Raman spectrum of the second region.

23. The method of claim 22, wherein acquiring the images of the particle and the substrate comprises adjusting illumination provided by an array of light sources.

24. The method of claim 22, wherein acquiring the images of the particle and the substrate comprises moving an optical head within a housing and coupled to a camera and a Raman laser source within the housing with respect to the particle and the substrate and the housing.

25. The method of claim 22, wherein measuring the Raman spectrum of the second region comprises acquiring a Raman signal from the second region for a measurement acquisition period based on a signal-to-noise ratio of the difference between the Raman spectrum of the first region and the Raman spectrum of the second region.Attorney Docket No. PEND-014WO0126. A system for measuring a Raman spectrum of a particle on a substrate, the system comprising: an array of light sources to illuminate the particle and the substrate with light at each of a plurality of different illumination conditions; an image sensor to acquire images of the particle and the substrate under each of the plurality of different illumination conditions; a processor, operably coupled to the image sensor, to identify, based on the images, a first region containing the particle and the substrate and a second region containing only the substrate; and a Raman laser source and a spectrometer, operably coupled to the processor, to measure a Raman spectrum of the first region, to measure a Raman spectrum of the second region, and to determine the Raman spectrum of the particle based on a difference between the Raman spectrum of the first region and the Raman spectrum of the second region.

27. The system of claim 26, further comprising: a movable optical head, in optical communication with the image sensor and the Raman laser source, to focus the image sensor and the Raman laser source with respect to the particle and the substrate.

28. The system of claim 26, wherein the Raman laser source and the spectrometer are configured to measure the Raman spectrum of the second region by acquiring a Raman signal from the second region for a measurement acquisition period based on a signal -to-noise ratio of the difference between the Raman spectrum of the first region and the Raman spectrum of the second region.

29. A method of calibrating a portable Raman microscope comprising a Raman pump laser and a spectrometer aligned with respect to a camera with a calibration sample comprising a Raman target located at a known position with respect to a reference mark and formed of Raman-active material, the method comprising: acquiring an image of the Raman target with the camera; measuring a Raman spectrum of the Raman target with the Raman pump laser and the spectrometer; estimating, based on the Raman spectrum, a location of the Raman target; and determining an offset between the camera and the Raman pump laser based on the image of the Raman target and the location of the Raman target.Attorney Docket No. PEND-014WO0130. The method of claim 29, wherein the Raman target has a maximum lateral dimension on the order of a resolvable spot size of the camera, the calibration sample comprises a reference mark, the image of the Raman target is of the reference mark and the Raman target, and determining the offset is based further on the location of the reference mark in the image.

31. The method of claim 29, wherein measuring the Raman spectrum of the Raman target comprises measuring Raman spectra from different points on the calibration sample and estimating the location of the Raman target comprises: determining which of the different points on the calibration sample has the Raman spectrum with the highest amplitude.

32. The method of claim 29, further comprising: acquiring an image of a target with the camera; and aiming a Raman pump beam emitted by the Raman pump laser at the target based on the image of the target and the offset between the camera and the Raman pump laser.

33. A sample for calibrating a Raman microscope, the sample comprising: a Raman-active feature having a maximum lateral dimension less than a resolvable spot size of the Raman microscope; a Raman-inactive material surrounding the Raman-active feature; and a fiducial marker at a known position with respect to the Raman-active feature and large enough to be resolved by the Raman microscope.

34. The sample of claim 33, wherein the Raman-active feature is in an array of Raman- active features.

35. The sample of claim 33, wherein Raman-inactive material comprises a gold layer and the Raman-active feature comprises silicon exposed through a hole in the gold layer.

36. A method of analyzing a heterogeneous sample comprising particles with a Raman microscope, the method comprising: locating the particles in a field of view of the Raman microscope; measuring a Raman spectrum of each of the particles with the Raman microscope; estimating spectral components of each Raman spectrum; for each of the particles, estimating a composition of that particle based on the spectral components of the Raman spectrum of that particle; andAttorney Docket No. PEND-014WO01 displaying the compositions of the particles.

37. The method of claim 36, wherein locating the particles comprises generating a z-stack image of the heterogeneous sample.

38. The method of claim 36, wherein locating the particles comprises generating a surface profile of the heterogeneous sample and measuring the Raman spectrum of each particle comprises focusing a Raman pump beam to a height of that particle based on surface profile.

39. The method of claim 36, wherein estimating the composition of one of the particles comprises comparing the spectral components for that particle to a library of Raman spectra.

40. A Raman microscope for analyzing a heterogeneous sample comprising particles with a Raman microscope, the Raman microscope comprising: an image sensor to acquire an image of the particles in a field of view of the Raman microscope; a Raman pump source and a spectrometer to measure a Raman spectrum of each of the particles; a processor, operably coupled to the image sensor, the Raman pump source, and the spectrometer, to estimate (i) spectral components of each Raman spectrum and, for each of the particles, (ii) a composition of that particle based on the spectral components of the Raman spectrum of that particle; and a display, operably coupled to the processor, to display the compositions of the particles.

41. The Raman microscope of claim 40, wherein the image sensor and the processor are configured to generate a z-stack image of the heterogeneous sample.

42. The Raman microscope of claim 40, wherein the processor is configured to locate the particles by generating a surface profile of the heterogeneous sample and further comprising: an optical head, in optical communication with the Raman pump source, to focus a Raman pump beam to a height of each particle based on surface profile.

43. The Raman microscope of claim 40, wherein the processor is configured to estimate the composition of one of the particles by comparing the spectral components for that particle to a library of Raman spectra.

Images