Automated spectrometric analysis of microside microplastic

Abstract

The detection of the micro-plastic needs to be combined with various technologies. For example, position detection techniques such as cross polarization detection, autofluorescence detection or photo-thermal infrared imaging are utilized to determine the position of the microplastics in a sample. The microplastics are then characterized by detecting the infrared absorption at these locations. Compared with the prior art, the method has the advantage that the content of the micro-plastics can be positioned and represented more quickly and more accurately.

Description

Cross-references to related applications

[0001] This application claims priority to U.S. Patent Application No. 18 / 350,690, filed July 11, 2023, the entire disclosure of which is incorporated herein by reference. Technical Field

[0002] This disclosure relates to the detection or analysis of materials using optical methods (i.e., infrared, visible, or ultraviolet light). Specifically, this disclosure relates to the use of optical photothermal detection technology to perform infrared spectroscopy and imaging with submicron spatial resolution, thereby identifying the chemical composition of micron-sized microplastic particle clusters. Background Technology

[0003] Optical photothermal technology has been described in, for example, U.S. Patents 9,091,594 and 9,841,324. These documents typically use different names and abbreviations to refer to the technology. For the purposes of this application, these technologies are collectively referred to as Optical Photothermal Infrared (OPTIR).

[0004] Multiple research teams have been working on the broad field of OPTIR, including researchers from the U.S. Naval Research Laboratory, Purdue University, the University of Notre Dame, Boston University, and MIT. The instruments developed by these laboratories use visible light beams to detect the photothermal response of samples after absorbing infrared radiation. Possibly relevant background publications and patents include: (1) R. Furstenberg, CA Kendziora, MR Papantonakis, V. Nguyen and RA McGill, “Chemical Imaging using Infrared Photo-thermal Microspectroscopy” Proc. of SPIE Vol. 8374, 837411 (2012); (2) R. Furstenberg, C. Kendziora, NDBassim, RA McGill, and VK Nguyen, US Patent 9,091,594 B2 (2015); (3) C. Li, D. Zhang, MN Slipchenko, and J.-X. Cheng, Anal. Chem., 89, 9, 4863-4867 (2017); (4) D. Zhang, C. Li, C. Zhang, MN Slipchenko, G. Eakins, and J.-X. Cheng, Science Advances, 2, 9, e1600521 (2016); (5) Z. Li, K. Aleshire, M. Kuno, and GV Hartland, The Journal of Physical Chemistry B, 121, 37,8838-8846 (2017); (6) Z. Li, M. Kuno, and G. Hartland, "Super-resolution imaging with mid-IR photothermal microscopy on the single particle level", inSPIE Nanoscience+ Engineering (International Society for Optics and Photonics, 2015), p. 954912-954912-954918; (7) Z. Li, M. Kuno, and G.Hartland, "Super-resolution Mid-infrared Imaging using Photothermal Microscopy", inConference on Lasers and Electro-Optics (Optical Society of America, SanJose, California, 2016), p. ATu3J.7.; (8) A. Mërtiri, A. Totachawattana, H.Liu, MK Hong, T. Gardner, MY Sander, and S. Erramilli, "Label free mid-IR photothermal imaging of bird brain with quantum cascade laser", in CLEO:Applications and Technology (Optical Society of America, 2014), p. AF1B. 4; (9) MY Sander, "Mid-infrared photothermal imaging", in Laser Science (Optical Society of America, 2015), p. LM1I. 2. .

[0005] The working principle of OPTIR technology lies in measuring the minute changes that occur after a probe beam interacts with the infrared absorption region of a sample. The infrared absorption region of the sample converts absorbed infrared radiation into heat, causing a localized temperature increase. This temperature increase alters the morphology, size, surface position, and / or refractive index of the infrared absorption region. One or more of these changes alter the intensity, angle, and / or phase of the probe beam after its interaction with the sample. Light reflected, scattered, and / or transmitted from the sample can be collected and analyzed. Detecting and measuring changes in the collected probe light under infrared illumination indicates the infrared absorption of the sample region illuminated by the probe light. Detecting changes in the collected probe light with infrared wavelength / wavenumber yields a signal indicating the infrared absorption spectrum of the sample material. By measuring changes in the collected probe light with sample position, an image indicating the infrared absorption of the sample can be generated, thus providing an image of the distribution of different chemical substances within the sample.

[0006] microplastics

[0007] Microplastics (MPs), which are plastic waste particles with a diameter of less than about 5 millimeters, such as particles ranging in size from 1 micrometer to 5 millimeters, have been recognized as a global problem (see, for example, Andrady, AL, The plastic in microplastics: A review. Marine pollution bulletin, 2017. 119(1): p. 12-22; Hale, RC, et al., A Global Perspective on Microplastics. Journal of Geophysical Research: Oceans, 2020. 125(1): p. e2018JC014719; and Ivleva, NP, Chemical analysis of microplastics and nanoplastics: challenges, advanced methods, and perspectives. Chemical Reviews, 2021. 121(19): p.11886-11936). Since the 1970s, plastic production has increased more than tenfold, with 335 million tons of plastic waste generated in 2017 alone. It is projected that plastic waste entering the environment will triple in the next 20 years. Microplastics, primarily derived from the decomposition of plastic waste, are now widely present in the environment, found in large quantities in water, air, and soil. Marine organisms frequently ingest microplastics, thus they have infiltrated the food chain and are frequently detected in seafood. A recent review by Dąbrowska states that "marine microplastics are considered one of the most critical issues requiring research and resolution in the coming years" (Dąbrowska, A., Raman Spectroscopy of Marine Microplastics - A shortcomprehensive compendium for the environmental scientists. MarineEnvironmental Research, 2021. 168: p. 105313.). Microplastics can enter the human body through drinking contaminated water, ingesting contaminated food, and / or inhaling airborne particles. Microplastics may have a profound impact on human health, especially micron-sized particles, which can penetrate the intestinal wall and accumulate in tissues, thereby affecting organ function and causing other damage.A recent review detailed the results of human cell culture assays and in vivo assays in mouse models, which showed that microplastics have adverse effects on organisms, including particulate toxicity, oxidative stress, cytokine secretion, cell damage, inflammation and immune response, DNA damage, neurotoxicity and metabolic effects (Yong, CQY, S. Valiyaveettil, and B.L. Tang, Toxicity of microplastics and nanoplastics in mammalian systems. International Journal of Environmental Research and Public Health, 2020. 17(05): p. 1509).

[0008] To understand the impact of microplastics on human health and to provide insights into possible remedies, the microplastics research community widely recognizes the need to characterize the distribution of microplastic particles in the environment, biological tissues, and food. As a recent opinion piece published in *Science* states, “There is an urgent need for suitable analytical tools to sample, separate, detect, quantify, and characterize small microplastics (<10 µm)” (Vethaak, AD and J. Legler, *Microplastics and humanhealth. Science, 2021. 371(6530): p. 672-674). At the same time, the industry clearly recognizes that existing particle analysis techniques have significant limitations when dealing with tiny particles (i.e., <10 µm), performing poorly on one or more key metrics, including minimum particle size, chemical identification accuracy, ability to analyze colored / stained microplastics, and measurement / analysis speed and automation.

[0009] For example, Raman spectroscopy often fails when samples have high levels of autofluorescence, as the autofluorescence drowns out the Raman signal. Raman spectroscopy also struggles with dark and / or colored samples, which may suffer photodamage under Raman-excited lasers. In a recent multilaboratory study, Raman spectroscopy achieved only 5% accuracy for some colored plastics, 5% accuracy for dyed cellulose, and only 18% accuracy for natural particles, many of which were misidentified as plastics. Furthermore, in the same study, Raman spectroscopy users successfully identified only 24 microplastic particles in the 1–20 µm size range, compared to over 1700 particles in a larger size range (cited in: DeFrond, H., et al., Monitoring microplastics in drinking water: Aninterlaboratory study to inform effective methods for quantifying and characterizing microplastics. Chemosphere, 2022. 298: p. 134282).

[0010] There are three main limitations to the application of Raman spectroscopy in the analysis of microplastics: (1) Raman spectroscopy is extremely inefficient, with only about 1 / 10 of the photons incident on the sample being used. 6 (1) Raman scattering can occur, which leads to the need for high power and / or excessively long integration time; (2) Many plastics, especially colored plastics, have a strong fluorescent background that can mask the Raman signal; (3) Colored / dark particles often suffer destructive photodamage under typical Raman laser irradiation. As the DeFrond literature mentioned above shows, the combined effect of these factors can make chemical identification significantly difficult.

[0011] Traditional infrared spectroscopy techniques, including laser direct infrared (LDIR) methods, often fail when analyzing small microplastic particles due to inherent resolution limitations, sensitivity constraints, and scattering artifacts related to particle size and shape. Therefore, a recent protocol published by the Southern California Coastal Water Research Project (SCCWRP) recommends using conventional infrared and Raman spectroscopy only for microplastic particles larger than 20 μm. For example, see the following literature: (1) Xu, J.-L., et al., FT-IR and Raman imaging for microplastics analysis: State of theart, challenges and prospects. TrAC Trends in Analytical Chemistry, 2019.119: p. 115629; (2) De Frond, H., et al., Monitoring microplastics in drinkingwater: An interlaboratory study to inform effective methods for quantifying and characterizing microplastics. Chemosphere, 2022. 298: p. 134282; (3) Wong, CS and S. Coffin, Standard Operating Procedures for Extraction and Measurement by Raman Spectroscopy of Microplastic Particles in DrinkingWater. 2021, Southern California Coastal Water Research Project Authority andCalifornia State Water Resources Control Board; (4) Wong, CSC, Scott, Standard Operating Procedures for Extraction and Measurement by InfraredSpectroscopy of Microplastic Particles in Drinking Water.2021, SouthernCalifornia Coastal Water Research Project Authority and California StateWater Resources Control Board;(5)Primpke, S., M. Godejohann, and G. Gerdts,Rapid identification and quantification of microplastics in the environmentby quantum cascade laser-based hyperspectral infrared chemical imaging.Environmental Science & Technology, 2020. 54(24): p. 15893-15903;(6)Cabernard, L., et al., Comparison of Raman and Fourier transform infraredspectroscopy for the quantification of microplastics in the aquaticenvironment. Environmental science & technology, 2018. 52(22): p. 13279-13288。.

[0012] OPTIR has been used for the artificial measurement of small amounts of microplastic particles, as described in the following references, which are hereby incorporated herein by reference: (1) Böke, JS, J. Popp, and C. Krafft, Optical photothermalinfrared spectroscopy with simultaneously acquired Raman spectroscopy for two-dimensional microplastic identification. Scientific reports, 2022. 12(1):p. 1-13; (2) Shi, Y., et al., Visual characterization of microplastics in cornflour by near field molecular spectral imaging and data mining. Science of The Total Environment, 2023. 862: p. 160714; (3) Barrett, J., et al., Microplastic Pollution in Deep-Sea Sediments From the Great Australian Bight. Frontiers in Marine Science, 2020. 7(808); (4) Chen, CK, et al., A portable purification system for the rapid removal of microplastics from environmentalsamples. Chemical Engineering Journal, 2022. 428: p. 132614; (5) Krafft, C., Optical photothermal infrared spectroscopic applications in microplastics—comparison with Fourier transform infrared and Raman spectroscopy, inMolecular and Laser Spectroscopy. 2022, Elsevier. p. 305-336; (6) Yan, F., et al., Development of a binary digestion system for extraction microplastics infish and detection method by Optical Photothermal Infrared (O-PTIR).Frontiers in Marine Science, 2022: p. 99;(7)Anderson, J., et al. SubmicronSimultaneous IR and Raman Spectroscopy (IR+ Raman): Breakthrough Developmentsin Optical Photothermal IR (O-PTIR) Combined for Enhanced Failure Analysis.in ISTFA 2019. 2019. ASM International;(8)Su, Y., et al., Steam disinfectionreleases micro (nano) plastics from silicone-rubber baby teats as examined byoptical photothermal infrared microspectroscopy. Nature Nanotechnology, 2022.17(1): p. 76-85;(9)Huang, W., et al., Comparison of Lead Adsorption on theAged Conventional Microplastics, Biodegradable Microplastics andEnvironmentally-Relevant Tire Wear Particles: Effects of Aquatic Chemistryand Desorption Behavior. Biodegradable Microplastics and Environmentally-Relevant Tire Wear Particles: Effects of Aquatic Chemistry and DesorptionBehavior。.

[0013] However, none of the literature listed above describes a fully automated OPTIR analysis, including automated particle identification, automated OPTIR spectral analysis, and automated chemical identification of microplastic particles. Therefore, a highly reliable spectroscopic identification method for microplastic particles with a diameter less than 20 μm has long been lacking. This paper proposes corresponding methods and apparatus to address this previously unmet technical need. Summary of the Invention

[0014] According to the embodiments described herein, the microscopic analysis of samples utilizes optical imaging, image particle analysis, and automated spectral analysis to rapidly identify the chemical composition of micron-sized microplastic particles while covering a wide range of chemical compositions, achieving a high success rate in chemical identification.

[0015] According to a first embodiment, the present invention discloses a method for automatically characterizing a sample containing a group of microplastic particles using photothermal infrared spectroscopy. The method includes: acquiring a polarization optical image of the sample; analyzing the polarization optical image to automatically identify the positions of multiple microplastic particles in the sample; automatically positioning the microplastic particles under the detection beam of a photothermal infrared spectroscopy system based on the positions of the multiple microplastic particles; and irradiating the microplastic particles with multiple infrared wavelengths. The method further includes: collecting at least one detection beam light from the microplastic particles at a detector, selected from reflected light, scattered light, or transmitted light; and measuring changes in the collected detection light from the microplastic particles, the changes corresponding to the infrared absorption of the microplastic particles.

[0016] According to another embodiment, the present invention discloses an optical system. This optical system can be used to automatically characterize samples containing microplastic particle clusters. The system includes a cross-polarization microscopy subsystem configured to acquire polarization optical images of the sample and identify the locations of multiple microplastic particles in the sample. The system also includes a photothermal infrared spectroscopy system configured to illuminate the locations of the multiple microplastic particles with multiple infrared wavelengths to detect the photothermal infrared absorption corresponding to each location of the multiple microplastic particles.

[0017] According to another embodiment, the present invention discloses a method for automatically characterizing a sample containing microplastic particle clusters using photothermal infrared spectroscopy. The method includes: acquiring at least one photothermal infrared image of the sample at multiple infrared wavelengths; analyzing at least one optical image and / or photothermal infrared image to identify multiple particle locations; analyzing at least one photothermal infrared image of the sample to select a subset of the multiple particle locations as possible microplastic particle locations; automatically positioning the particles under the detection beam of a photothermal infrared spectroscopy system based on the multiple possible microplastic particle locations; irradiating the particle with multiple infrared wavelengths; and detecting changes in the detection light collected from the particle, the changes corresponding to the infrared absorption of the particle.

[0018] The above overview is not intended to describe every exemplary embodiment or every implementation of the subject matter herein. The accompanying drawings and detailed descriptions that follow provide more specific illustrative examples of various embodiments. Attached Figure Description

[0019] The subject matter of this invention can be more fully understood through a detailed description of the following embodiments taken in conjunction with the accompanying drawings, wherein:

[0020] Figure 1A and Figure 1B A simplified schematic diagram of an OPTIR-based method for the automated detection and analysis of microplastic particles and the automated analysis of their chemical composition is shown.

[0021] Figure 2 A simplified schematic diagram of a polarization optical microscope used for detecting microplastic particles is shown.

[0022] Figure 3 A simplified schematic diagram of an apparatus for the automated detection and analysis of microplastic particles is shown.

[0023] Figures 4A-4F An example OPTIR spectrum of microplastic particles, automatically measured after detection by polarization microscopy, is shown, including microplastic particles composed of materials that do not typically exhibit bulk birefringence.

[0024] Figure 5 Examples of optical microscope images taken using bright-field, autofluorescence, and cross-polarization microscopy are shown, along with corresponding particle analysis diagrams.

[0025] Figures 6A-6B The illustration shows a method for correcting optical microscope image distortion to improve the positioning accuracy of microplastic particles in OTPIR measurements.

[0026] Figures 7A-7B The illustration shows a method for calculating one or more position correction vectors using OPTIR DC images and optical microscope images to improve the positioning accuracy of microplastic particles in OPTIR measurements.

[0027] Figure 8 The average infrared absorption spectra of plastics and organic materials are shown.

[0028] While various embodiments can be modified and alternatively implemented, the specific details shown in the accompanying drawings are merely illustrative examples. However, it should be understood that the invention of the claims is not intended to be limited to the specific embodiments described. Rather, it is intended to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the subject matter defined in the claims. Detailed Implementation

[0029] This specification describes methods and apparatus for performing optical photothermal infrared (OPTIR) imaging and spectral analysis, which have improved sensitivity, better signal-to-noise ratio, and reduced background signal.

[0030] For the purposes of this specification, the following terms are specifically defined as follows:

[0031] "Analyzer / controller" refers to a system used to assist in data acquisition and control of an OPTIR system. This controller can be a single integrated electronic device or comprise multiple distributed components. The control components can control the positioning and / or scanning of the probe tip and / or sample. The control components can also collect data on probe deflection, motion, or other responses, and control the power, polarization, pointing, focusing, and / or other functions of the radiation source. The aforementioned control components can include computer program methods or digital logic methods and can be implemented using any combination of various computing devices (computers, personal electronic devices), analog and / or digital discrete circuit elements (transistors, resistors, capacitors, inductors, diodes, etc.), programmable logic devices, microprocessors, microcontrollers, application-specific integrated circuits, or other circuit elements. A memory configured to store a computer program can be used in conjunction with discrete circuit elements to execute one or more of the processes described herein.

[0032] A beam combiner is an optical element that combines two beams of light into a single optical path. In one configuration, the beam combiner can be a beam splitter used in reverse, combining a beam of light reflected at the beam splitter interface with another beam of light passing through that interface. For example, a beam splitter cube can be used as both a beam splitter and a beam combiner. Commercially available optical elements labeled as beam splitters can also be used as beam combiners, even if they are not used to split light into two paths. For example, a Mach-Zehnder interferometer uses a beam splitter to split the incident light into two paths, and then uses a second beam splitter to recombin the two beams. In this case, the second beam splitter is used as a beam combiner. In a Michelson interferometer, a single beam splitter is used both to split the incident light and to recombin it. Therefore, the beam splitter in a Michelson interferometer functions as both a beam splitter and a beam combiner. Beam combiners can also be fiber-based devices, such as combining light from two input fibers into a single output fiber; such devices include 1×2 fiber couplers. A single 1×2 fiber coupler can be used as both a beam splitter and a beam combiner.

[0033] A beam splitter is an optical element capable of splitting a beam of light into at least two optical paths. Beam splitters can include plate, cube, and / or prism types, or other shapes / structures capable of dividing a beam. Beam splitters can include a thin film with partial reflectivity at the wavelength of interest, such that a portion of the incident beam is reflected and another portion is transmitted. Beam splitters can be polarized, meaning they essentially transmit light of one polarization state and reflect light of another polarization state orthogonal to that polarization state. Beam splitters can also distribute light to two transmission paths based on polarization, for example, when the beam splitter is a Nomaski prism or a Wollaston prism. Beam splitters can also be non-polarized, where light is split into two optical paths substantially independent of the polarization state of the incident light. Beam splitters can also be fiber-based devices, such as splitting light from one input fiber into light from at least two output fibers, for example, a 1×2 fiber coupler. Beam splitters can be 50:50 beam splitters, where approximately equal proportions of light are directed onto two different optical paths. Beam splitters can also be unbalanced, such as 90:10 or 70:30 or similar ratios, so that 90% of the light is directed onto one optical path and 10% onto another, or 70% onto one optical path and 30% onto another.

[0034] A "camera" refers to an array of photodetectors containing multiple photosensitive pixels. A camera may employ one or more technologies, including but not limited to CCD, EM-CCD, CMOS, s-CMOS, and / or other photosensitive array technologies. The camera can support frame rates ranging from a few frames per second, hundreds of frames per second, or even thousands of frames per second or higher.

[0035] "Collecting probe light" and "collecting probe radiation" refer to collecting the radiation from the probe beam that interacts with the sample. The probe light can be collected after reflection, scattering, transmission, evanescent wave coupling, and / or transmission through an aperture probe.

[0036] "Confocal microscopy" refers to an optical microscopy technique in which the detector collects only light that passes through a small volume region within the three-dimensional focusing volume of the optical objective on the sample. Confocal microscopy is typically achieved by placing a "confocal stop" on a focal plane equivalent to the sample's focal plane, thereby blocking stray light that has not passed through the sample's focusing volume.

[0037] A "detector" is a device capable of generating a signal indicating the power, intensity, and / or energy of light / radiation incident on the surface of the detector. This signal is typically an electrical signal, such as voltage, current, and / or charge. A detector can be a photodiode, phototransistor, or charge-coupled device (CCD). In some cases, a detector can be a semiconductor detector, such as a silicon PIN photodiode. A detector can also be an avalanche photodiode, photomultiplier tube, or any other device capable of generating changes in current, voltage, charge, conductivity, or similar parameters upon light incidence. Detectors can include single elements, multiple detector elements (e.g., dual-cell or quad-cell detectors), linear arrays, or two-dimensional arrays of detector elements, including camera-based detectors.

[0038] The "diffraction limit" of a light beam refers to the minimum distance at which a detector can distinguish two light sources. For a microscope with a specific numerical aperture (NA) operating at wavelength λ, the Abbe diffraction limit d is defined as d = λ / (2 · NA). Due to the physical limitations of the microscope's numerical aperture, which prevent it from being very large, the diffraction limit of a microscope largely depends on the operating wavelength used for detection. The longer the wavelength, the lower the corresponding resolution, while the shorter the wavelength corresponds to higher resolution.

[0039] "Demodulation" refers to extracting the information-carrying signal from the overall signal, typically (but not necessarily) at a specific frequency. For example, in this application, the probe light collected by the photodetector represents the overall signal. The demodulation process extracts the portion of the signal that is disturbed due to the absorption of infrared light by the sample. Demodulation can be achieved using a lock-in amplifier, a fast Fourier transform (FFT), calculation of the discrete Fourier components at the desired frequency, a resonant amplifier, a narrowband bandpass filter, or any other technique that can significantly enhance the target signal while suppressing background and noise signals that are out of sync with the modulation.

[0040] A demodulator is a device or system that performs demodulation.

[0041] A "feedback loop" is a control loop designed to keep parameters substantially near a target value ("setpoint") by adjusting control elements. A "phase feedback loop" is a feedback loop designed to maintain the optical phase difference of an interferometer at a target phase value. It's important to note that despite the presence of a feedback loop, the optical phase difference typically does not remain truly constant or fixed because disturbances in the optical phase difference can occur too quickly for the phase feedback loop to compensate for. Therefore, an error, especially transient errors, usually exists between the target phase value and the actual phase value.

[0042] "Performance metrics" refers to any measure or indicator that measures the relative quality of a signal or measurement. Performance metrics can be, for example, measurement sensitivity, signal strength, noise level, signal-to-noise ratio, background level, signal-to-background ratio, or any combination of these metrics, or other indicators that can rank the relative quality of signals and / or measurements.

[0043] "Fluorescence" refers to the phenomenon that a sample emits light at another wavelength through a fluorescence excitation and emission process when excited at one wavelength.

[0044] "Irradiation," "illumination," and "irradiation" refer to directing radiation onto an object, such as a sample surface, probe tip, and / or the area where the probe interacts with the sample. Irradiation can include the infrared wavelength range, visible wavelengths, and other radiation from ultraviolet to millimeter-scale or longer wavelengths. Irradiation can include any configuration of radiation sources, reflective elements, focusing elements, and any other elements used for beam pointing control or adjustment.

[0045] "Infrared absorption spectrum" refers to a spectrum in which the relationship between the infrared absorption coefficient, absorbance, or similar physical quantities characterizing the infrared absorption properties of a sample and wavelength is directly proportional. An example of infrared absorption spectrum is the absorption measurement produced by a Fourier transform infrared (FTIR) spectrometer, i.e., an FTIR absorption spectrum. Typically, infrared light is absorbed (i.e., a portion of the infrared absorption spectrum), transmitted (i.e., a portion of the infrared transmission spectrum), or reflected. The intensity of the reflected or transmitted spectrum of the collected probe light may differ at various wavelengths compared to the intensity of the probe light source at each wavelength. It is worth noting that infrared measurements are typically plotted as transmitted light rather than absorbed light. For the purposes of this definition, infrared transmission and infrared absorption spectra are considered equivalent sets of data because a simple relationship exists between the two measurements.

[0046] "Infrared source" and "infrared radiation source" refer to one or more light sources that generate or emit radiation in the infrared wavelength range (typically 2-25 micrometers). The radiation source can be one of a variety of light sources, including heat sources or Globar sources, supercontinuum laser sources, frequency combs, difference frequency generators, sum frequency generators, harmonic generators, optical parametric oscillators (OPO), optical parametric generators (OPG), quantum cascade lasers (QCL), inter-band cavity lasers (ICL), synchrotron infrared sources, nanosecond, picosecond, femtosecond, and attosecond laser systems, carbon dioxide lasers, microheaters, electric sparks or chemical sparks, and / or any other light source capable of generating infrared radiation emission. In a preferred embodiment, the light source emits infrared radiation, but it can also emit radiation in other wavelength ranges, such as from ultraviolet to terahertz (THz). The light source can be narrowband, for example, with a spectral width of less than 10 cm⁻¹. -1 or less than 1 cm-1 It can also be broadband, for example, with a spectral width greater than 10 cm. -1 Greater than 100 cm -1 or greater than 500 cm -1 Broadband light sources can be converted into narrowband light sources using filters, monochromators, and other devices. Infrared light sources can also consist of discrete emission spectra, such as those tuned to specific absorption bands of the target substance.

[0047] In the context of interaction with a sample, “interaction” refers to light that is irradiated onto a sample and undergoes at least one of the following: being scattered by the sample, being refracted by the sample, being absorbed by the sample, being distorted by the sample, being deflected by the sample, being diffracted through the sample, being transmitted through the sample, and being reflected by the sample.

[0048] "Interference" refers to the interaction of at least two beams of light, including the coherent superposition of these beams.

[0049] An "interferometer" is an optical device that combines light from at least two paths to produce interference between at least two beams of light. In the context of this application, these two paths are sometimes referred to as a "sample arm" (in which light interacts with a sample) and a "reference arm" (in which light interacts with a reference reflector).

[0050] "Asymmetric interferometer" refers to an interferometer configuration in which there is interference between light from the sample arm and light from the reference arm, wherein the intensity of the light from the reference arm is greater than that from the sample arm. It is important to note that this term differs from other terms that may refer to different interferometer arm lengths. The asymmetric interferometer described herein may have equal or unequal interferometer arms, but "asymmetric" here refers to the difference in optical power between the reference arm and the sample arm.

[0051] A "common-path interferometer" refers to an interferometer in which at least most of the optical paths of the sample arm and the reference arm are shared, with only a few optical paths being different.

[0052] An "orthogonal interferometer" is an interferometer having at least two optical paths that interfere and at least two detectors for detecting the interference radiation on these two optical paths. An orthogonal interferometer can be configured such that there is an optical phase delay of approximately 90° between the two interfering optical paths and between the corresponding two detectors. An orthogonal interferometer can also have more detectors, such as four detectors.

[0053] A "heterodyne interferometer" is an optical interferometer in which the optical frequency and / or phase of at least one arm of the interferometer is modulated relative to the other arm. A heterodyne interferometer may include a "phase modulator," which is an optical element capable of modulating the phase and / or frequency of light transmitted through or reflected from it. Phase modulators may include acousto-optic modulators, electro-optic modulators, liquid crystal modulators, or other devices capable of changing the phase / frequency of light in response to an input signal or stimulus.

[0054] A "lock-in amplifier" is an example of a "demodulator" (as defined above), a device, system, and / or algorithm that demodulates the response of a system at one or more reference frequencies. A lock-in amplifier can be an electronic component comprising analog electronic elements, digital electronic elements, or a combination of both. It can also be a computational algorithm implemented on digital electronic devices such as microprocessors, field-programmable gate arrays (FPGAs), digital signal processors, and personal computers. A lock-in amplifier can generate signals indicative of various parameters of an oscillating system, including amplitude, phase, in-phase (X) components, and quadrature (Y) components, or any combination thereof. In this case, the lock-in amplifier can also produce such measurements at the reference frequency, higher harmonics of the reference frequency, and / or sideband frequencies of the reference frequency.

[0055] A mass spectrometer is an instrument used to analyze the molecular weight distribution of analytes.

[0056] When referring to radiation incident on a sample, "modulation" refers to the periodic alteration of the intensity of infrared laser light at a specific location. Modulation of beam intensity can be achieved, for example, by mechanically chopping the beam, controlled laser pulse emission, and / or deflecting the laser beam. Beam deflection can be achieved using tilting mirrors, which can be tilted or deformed by electrostatic, electromagnetic, piezoelectric actuators, or other methods, or by using high-speed rotating mirror devices. Modulation can also be achieved by providing time-varying transmittance, such as acousto-optic modulators, electro-optic modulators, photoelastic modulators, Pockels cells, etc. Modulation can also be achieved using diffraction effects, for example, through MEMS-based diffraction modulators, or through high-speed shutters, attenuators, or other devices capable of altering the intensity, angle, and / or phase of the laser light incident on the sample.

[0057] "Near-infrared light" usually refers to infrared (IR) light with a wavelength range of 0.75 µm–2 µm.

[0058] "Optical properties" refers to the optical properties of a sample, including but not limited to refractive index, absorption coefficient, reflectivity, absorptivity, the real and / or imaginary part of the refractive index, the real and / or imaginary part of the sample's dielectric function, and / or any property that can be mathematically derived from one or more of these optical properties.

[0059] "Optical response" refers to the result of the interaction between radiation and a sample. Optical response is related to one or more of the aforementioned optical properties. It can be radiation absorption, temperature rise, thermal expansion, photodynamic activity, light reflection and / or scattering, or other responses of a material due to its interaction with the irradiating radiation.

[0060] "Narrowband light source" refers to a light source with a narrow bandwidth or linewidth, such as a linewidth of less than 8 cm. -1 The light source can be narrow enough that it does not cover the spectral range of interest of the sample.

[0061] "Polarization optical microscopy" refers to one or more techniques used to generate microscopic images characterizing polarization rotation, optical activity, and / or chirality in a sample. Examples include cross-polarization microscopy, microellipsometers, Stokes polarimeters, polarization modulation microscopy, polarization-sensitive optical tomography, second harmonic generation, and circular dichroism microscopy. "Polarization optical image" refers to an image obtained using one or more polarization optical microscopy techniques.

[0062] A polarizer is an optical device that preferentially polarizes a beam of light in a desired direction, typically by allowing one type of polarized light to pass through while significantly attenuating, blocking, and / or redirecting the direction of another orthogonally polarized light. Examples include wire-grating polarizers, dichroic thin-film polarizers, nanoparticle linear thin-film polarizers, liquid crystal polarizers, polarizing beam splitters, Wollaston prisms or other prism polarizers, reflective polarizers utilizing Brewster's angle reflection, and birefringent crystals. Alternatively, polarizers can also produce circularly polarized light with a preferred rotation direction, i.e., right-handed or left-handed circularly polarized light.

[0063] "Photothermal distortion" refers to the change in properties of a sample caused by the absorption of light energy (such as the absorption of infrared radiation). Photothermal distortion may manifest as changes in refractive index, reflectivity, thermal expansion, surface distortion, or other effects detectable by a probe beam.

[0064] "Detector source," "detector light source," or "detector radiation source" refers to a radiation source that can be used to sense the optical properties of a sample. A detector light source can be used to sense the sample's response to incident light from an infrared light source. This radiation source can include, for example, a gas laser, a laser diode, a superluminescent diode (SLD), a near-infrared laser, or an ultraviolet and / or visible laser beam generated by sum-frequency or difference-frequency generation. The radiation source can also include any other near-infrared, ultraviolet, and / or visible light sources that can be focused onto a spot smaller than 2.5 micrometers, even smaller than 1 micrometer, or possibly smaller than 0.5 micrometers. In some embodiments, the operating wavelength of the detector light source can be outside the tuning or emission range of the infrared light source, but the detector light source can also be a fixed-wavelength light source operating at a selected wavelength overlapping with the tuning range of the infrared light source. "Detector beam" or "sensing beam" refers to the beam initially emitted by the detector light source.

[0065] "Probe beam" refers to a beam of light or radiation that is directed onto a sample to detect photothermal distortion or other optical changes caused by the interaction of infrared radiation with the sample, such as detecting the absorption of infrared radiation by the sample.

[0066] A "retarder" is an optical element that introduces a relative phase delay in an optical path. Examples of retarders include waveplates, such as half-wave plates, quarter-wave plates, and eighth-wave plates. One or more retarders / waveplates can be used to introduce an optical phase difference between two polarized lights, for example, between the two optical paths of an orthogonal interferometer.

[0067] A "signal indicating..." refers to a signal that has a mathematical relationship with the property of interest. This signal can be an analog signal, a digital signal, and / or one or more values ​​stored in a computer or other digital electronic device. The signal can be voltage, current, or any other signal that is easily converted and recorded. Mathematically, the signal can be identical to the property being measured, for example, a definite absolute phase signal or absorption coefficient. It can also be a signal that has a mathematical relationship with one or more properties of interest, for example, involving linear or other scaling, offsetting, inversion, or even more complex mathematical operations.

[0068] A signal processor is an analog and / or digital device used to process one or more signals, such as filtering, level shifting, squaring, summing, RMS summing, and / or any other calculation or transformation of analog and / or digital signals. A signal processor can be an analog circuit where all filtering / conversion / calculation is performed by discrete analog components, or it can be one or more digital processors, such as a central processing unit (CPU), computer, field-programmable gate array (FPGA), digital signal processor (DSP), or any other suitable digital computing device. A signal processor can also be a hybrid of analog and digital components.

[0069] "Spectrum" refers to a measurement of one or more properties of a sample as a function of wavelength, or equivalently (and more commonly) as a function of wavenumber.

[0070] The terms “approximately” or “about” and similar terms are synonyms used to indicate that the modified numerical value has a generally accepted range of values, which may be ±20%, ±15%, ±10%, ±5%, or ±1%.

[0071] The term “substantially” is used to indicate that a result (such as a measurement) is close to a target value, where “close” can be understood, for example, as the result being within 80%, 90%, 95%, or 99% of the target value.

[0072] Automated Spectroscopic Analysis of Micron-Sized Microplastics Using OPTIR

[0073] Figure 1 shows a simplified flow chart of the process described herein for automated spectroscopic analysis of micron-sized microplastic particle clusters. The first step 100 involves preparing the microplastic particle clusters on a substrate by filtration or other means. In one embodiment, a liquid sample 103 containing microplastic particle clusters is filtered using a filter bottle device 101. One or more filters 105 are provided at the bottom of the filter bottle, with pore sizes arranged to capture particles within a target size range. In one embodiment, the filters may include two sequentially arranged filters: a first filter with a pore size of approximately 20 micrometers and a second filter with a pore size of approximately 0.8 micrometers. The first filter captures larger particles larger than 20 micrometers, and the second filter captures micron-sized particles ranging from approximately 1 micrometer to 20 micrometers in size. A vacuum bottle 101 includes a vacuum interface 107 for drawing liquid through the filters into a receiving bottle 109. The filtrate in the receiving bottle may be further filtered as needed to capture smaller particles that have passed through the first filter. After any filtration step, the filter can be removed from the filter bottle and the particles on its surface can be analyzed using OPTIR. Suitable filters include, for example, gold-plated polycarbonate filters and silicon membrane filters. Many filters consist of thin plastic membranes, which, if uncoated, generate their own OPTIR spectral signals, interfering with measurements of the spectra of the tiny plastic particles on the filter surface. Coating plastic filters with a thin metallic coating blocks the absorption of infrared light, thus avoiding this problem. Filters made of silicon, silicon nitride, silicon dioxide, thin metal films, and / or inorganic materials can also be used, and in some cases, no additional coating is required.

[0074] In step 110, an optical image of the sample is acquired, optionally and preferably using an optical technique that provides high contrast between the particles and the filter surface, particularly between the microplastic particles and the pores of the filter surface. Suitable techniques include dark-field imaging, fluorescence microscopy imaging by adding a fluorescent dye (e.g., Nile Red), and autofluorescence imaging. Methods for highlighting microplastics using fluorescent dyes have been discussed, for example, in Kang et al.'s paper "Modification of a Nile Red Staining Method for Microplastics Analysis—ANile Red Plate Method" (Water 2020, 12, 3251; doi:10.3390 / w12113251).

[0075] One particularly effective method is cross-polarization microscopy imaging, whose reference... Figure 2 The following description is provided. Cross-polarization microscopy uses two substantially orthogonal polarizers in an optical microscope, one in the illumination path and the other in the camera path. This effectively blocks light from the sample substrate, especially from the filter pores, allowing light to pass through and enter the camera only when the sample causes a polarization rotation in the light reflected / scattered from it. The end result is an approximately binary image 111, where the filter material (or other substrate) 115 appears substantially black, while the microplastic particles (e.g., 113) appear bright. Contrast enhancement techniques can be used to adjust the data range of the cross-polarized image to maximize the coverage of the displayed color levels. Thresholding can also be applied to create a binary image, forcing brighter areas to be set to maximum values ​​and darker areas to be set to minimum values.

[0076] In step 120, the high-contrast optical image can be analyzed to determine the location of the microplastic particle cluster and, optionally, acquire various size parameters. Many suitable algorithms can be used to determine the particle location and size parameters. For example, OpenCV provides library functions for image thresholding, blob detection, edge detection, and other related functions. Particle identification can optionally include any number of pre-filtering steps (e.g., field flattening, low-pass filtering, median filtering, and thresholding, such as the Otsu method) to create a binary image that separates the image into a filtered background and a particle foreground. Particle boundaries can then be defined using watershed segmentation (Beucher, S. and C. Mathmatique, The Watershed Transformation Applied To Image Segmentation. Scanning. Microsc., 2000. 6) or similar algorithms. Depending on the particle density, other algorithms can be employed to erode and segment overlapping particles. Once the particles are identified, algorithms can be used to analyze their size, area, perimeter, roundness, and any other morphological features. If needed, a color image of the microplastic particle cluster can also be acquired to enhance the color representation of each particle. The location and other properties of the particles can be summarized into a "particle table," which can then be used to initiate automated spectral analysis of microplastic particles.

[0077] In step 130, one or more positional offset corrections can be determined between the particle position coordinates measured from the optical image in steps 110–120. The XY position coordinates can then be applied to the positioning platform to position the sample relative to the OPTIR probe beam and / or to position the probe beam relative to the sample. The positions determined in steps 110–120 are typically the centroids of the microplastic particle cluster, given in pixel coordinates of the optical image. These pixel coordinates then need to be converted to positioning platform coordinates, taking into account nonlinear scaling errors, origin offsets, and rotations that may exist between the camera image and the positioning platform's translation axes. Nonlinear distortions in the optical image may include barrel / pincushion distortion and / or higher-order distortions. These issues can have significant effects if not carefully considered and compensated for. For example, consider a low-magnification optical image with a field of view of 1.0 mm. Even a distortion level of only 0.5% can result in a positional error of up to 5 μm, which is larger than the size of many target particles. In practice, it is desirable to achieve an overall positioning error of less than 1 μm, preferably less than 0.5 μm. This can be achieved in several ways. First, image distortion / rotation can be corrected by acquiring an image of a known reference sample, and then the measured optical position of the object with a known XY position can be determined. For example, Thorlabs and other optical suppliers sell calibration / distortion targets for mapping image distortion. Once the distortion field is determined, it can be inverted to convert the measured coordinates in image space to the corrected coordinates in real space. The positioning platform can also be calibrated similarly using calibration standards, interferometry, or other techniques. Furthermore, the position correction can be calibrated using the probe beam of an OPTIR instrument. For example, an image similar to bright-field or transmission imaging can be generated by scanning the sample under the probe beam, or by scanning the probe beam across the sample while recording the DC intensity of the scattered / transmitted light. This image can be cross-correlated with the wide-field optical image obtained in step 110, or even with the polarization image, to generate a corresponding mapping or function that converts the image coordinates to real-space coordinates, more specifically, to platform coordinates. If imaging a swarm of microplastic particles, this correction calibration can be performed on a single point or multiple particles. One semi-automatic method involves automatically acquiring an OPTIR DC image of a single particle and then prompting the user to click on the center of the particle image. The software can then calculate the offset between the user's click location and the nominal desired center location. This offset (Δx, Δy) can be applied to each subsequent sample location before performing OPTIR particle analysis. Alternatively, multiple offsets (Δx, Δy) can be determined for multiple different particles. i , Δy iThis is especially true when the optimal offset varies with the image position. Regardless of the method used, each particle to be measured needs to be calibrated to ensure that even very small particles are properly centered on the OPTIR probe beam before measurement.

[0078] In step 140, OPTR spectral measurements are acquired at multiple XY positions corresponding to the particle positions in the particle table. The resulting measurements are a series of spectra 141, each corresponding to a different microplastic particle.

[0079] In step 150, the chemical composition of various microplastic particles is determined by performing a spectral search on a known reference spectral database. Typically, the measured spectra are compared with those in the reference library, and a matching quality index (HQI) is calculated for each searched reference spectrum, as shown in Figure 151, which compares the sample data (dashed line) with a known spectrum of polymethyl methacrylate (PMMA) 153, which has an HQI of 0.903. The matching quality index can be calculated using various metrics, including but not limited to: cross-correlation, Euclidean distance, cosine similarity, and machine learning methods such as support vector machines and artificial neural networks. Before performing the correlation search, some preprocessing steps may be required, such as normalization, smoothing, baseline correction, mean centering, wavelength correction, first or second derivative processing, ATR correction, and / or other forms of filtering and preprocessing. Once the matching quality index for each reference spectrum is calculated, one or more candidate “matches” can be displayed, including the corresponding material identification and associated matching quality index or “score.” This process is repeated for each measured OPTIR spectrum (or at least for OPTIR spectra with signal-to-noise ratio).

[0080] In step 160, the results are summarized into a particle summary report 161. This report may include the particle's location, size, area, aspect ratio, roundness, and / or any other desired morphological or color parameters. Morphological factors may also be used to indicate the particle type, such as spherical, irregular shape, fibrous, fragmented, etc.

[0081] A key improvement of the method described above with reference to Figure 1 is the generation of high-contrast images that highlight the particles relative to the background of the filter or other substrate. We have implemented two methods for obtaining high-contrast images of microplastic particles with high success rates: (1) cross-polarization; and (2) autofluorescence. These two methods will be described below.

[0082] Detection of microplastic particles smaller than 20 micrometers in amorphous materials based on cross-polarization

[0083] Figure 2Cross-polarization imaging of microplastic particles is illustrated. An illumination beam 200 (typically with random incident polarization, as shown by arrow 202) passes through a polarizer 204 (whose first polarization angle is shown by arrow 206), resulting in a beam polarization of the transmitted beam as shown by arrow 208. The illumination beam is then incident on a beam splitter 210, which reflects beam 212 onto the back surface of objective lens 214. In the illustrated example, the polarization direction of the back surface of objective lens 214 is the X-direction, as shown by arrow 216. The initial polarization direction can be changed to any angle, but for ease of illustration, it is chosen to be along the X-direction in this example.

[0084] Objective lens 214 focuses the illumination light onto a spot 218 on sample 220, with a polarization direction as shown by arrow 222, which is approximately the same as the incident polarization direction shown by arrow 216. Sample 220 contains multiple particles 224, which are illuminated by the spot 218. Subsequently, the scattered / reflected light is collected again by objective lens 214 and passed through beam splitter 210. The collected light typically contains light with multiple polarization angles, a portion of which has the original polarization direction (as shown by arrow 226), and another portion of which has a rotated polarization direction (as shown by arrow 228). The rotation angles are for illustrative purposes only and are not drawn to scale.

[0085] A second polarizer 230 is positioned in the beam path, its polarization axis typically being substantially perpendicular to the polarization direction of the incident illumination light. In the example shown, the polarization axis of polarizer 230, indicated by arrow 232, is along the Y-axis and substantially orthogonal to the polarization direction of the illumination light at the sample along the X-axis. Polarizer 232 essentially blocks unrotated illumination light from the sample and primarily allows light from regions of the sample where polarization rotation has occurred to pass through. As described below, polarization rotation is an efficient mechanism for distinguishing microplastic particles from the filter substrate and filter pores. It should be noted that if the incident beam 200 is already polarized (e.g., using an intrinsically polarized light source, such as a laser), the first polarizer 204 can be omitted. It should be noted that in this context, "cross-polarization microscopy" refers to the use of polarizers in the camera optical path to block unrotated polarized light, but does not specifically mean that two polarizers with substantially orthogonal directions must be used. More generally, polarization microscopy does not necessarily require a cross polarizer, but rather refers to any imaging technique that is highly sensitive to polarization rotation, such as polarization modulation microscopy, microellipsometers, polarization-sensitive optical coherence tomography, and Stokes polarization measurements.

[0086] Sierra et al. used polarized light microscopy to study microplastic particles in wastewater (see "Identification of microplastics in wastewater samples by means of polarized light optical microscopy," Environmental Science and Pollution Research (2020) 27:7409–7419, https: / / doi.org / 10.1007 / s11356-019-07011-y.). Sierra used polarized microscopy to detect larger microplastic particles ranging in size from 70 to 600 μm in wastewater samples and identified categories of microplastic materials with sufficient optical activity (i.e., the ability to rotate optical polarization) to be detected by cross-polarized microscopy.

[0087] Previous studies have shown that cross-polarization microscopy is not suitable for certain plastics.

[0088] Sierra's paper also points out an inherent limitation of their method: "a significant limitation in identifying microplastics from non-birefringent polymers such as polyvinyl chloride (PVC)." The paper actually classifies microplastic particles as "optically active" or "optically inactive" based on whether they exhibit "optical activity" (i.e., the ability to rotate light polarization). Materials identified as having very low or no optical activity include polyvinyl chloride (PVC), polymethyl methacrylate (PMMA), and some polyethylene (PE) and polyamide (PA) samples. The paper further explains that amorphous polymers such as PVC and PMMA exhibit extremely low optical activity and are therefore undetectable by polarization microscopy. This previous study concluded that polarization microscopy is suitable for detecting certain microplastic materials but not others. In fact, many common polymers are amorphous (i.e., non-crystalline), such as atactic polystyrene (PS), polymethyl methacrylate (PMMA), high-density polyethylene (HDPE), polyvinyl chloride (PVC), polycarbonate (PC), certain types of polyethylene terephthalate (PET), polyvinyl acetate (PVAc), and polytetrafluoroethylene (PTFE, such as Teflon). Therefore, it can be expected that many important plastics may not be easily detected by cross-polarization microscopy.

[0089] Unexpected optical activity in small-sized microplastic particles of bulk amorphous polymers

[0090] However, the inventors have discovered that, in practice, cross-polarization microscopy can very effectively detect small-sized microplastic particles, even those composed of materials that are nominally amorphous in the bulk phase and therefore expected to be non-birefringent. Specifically, the inventors have used cross-polarization microscopy to detect particles smaller than 20 micrometers in PMMA, PS, HDPE, and PC, materials that are nominally amorphous in the bulk samples and are not expected to be birefringent. Figure 3 The device shown has also been used to detect and measure microplastic particles of polyethylene (PE) and polyamide (PA), which, according to Sierra’s literature, are not optically active in certain cases.

[0091] Several mechanisms could explain the unexpected success in detecting microplastic particles smaller than 20 micrometers using cross-polarization microscopy.

[0092] First, the successful detection of microplastic particles smaller than 20 micrometers can likely be attributed to the mechanical stress during their fragmentation process. Microplastic particles are typically the fragmentation product of larger particles and / or bulk plastics. For larger plastic particles to break down into micrometer-sized microplastic fragments, they may need to withstand stresses exceeding their yield strength. This stress can lead to molecular orientation effects, resulting in optical activity.

[0093] Inhomogeneities at minute scales can also make particles smaller than 20 micrometers detectable. Polymers that are considered amorphous or non-crystalline in bulk may exhibit molecular order at even smaller length scales. Therefore, micrometer-sized particles may contain local anisotropy, thus endowing them with optical activity and polarization rotation properties.

[0094] Scattering effects can also lead to detectability. When the particle size is close to the wavelength of the incident light (e.g., on the micrometer scale), Mie scattering has a complex dependence on the scattering angle and can alter the polarization state.

[0095] The reflection effect of tiny particles may also be one of the reasons for their detectability. The reflection coefficient depends on the polarization state and the angle of incidence. For tiny particles with large curvature, the incident light will encounter various different surface angles, resulting in a reflection distribution and thus changing the polarization state of the reflected light.

[0096] Multiple scattering on small particles may also be one of the reasons for their detectability. Light may be scattered on multiple surfaces of the particle, resulting in polarization rotation.

[0097] Experiments show that, contrary to expectations of detectability based on existing published literature, one or more of these mechanisms can enhance the detection of the vast majority of microplastic particles, including polymers that are typically considered amorphous / non-crystalline and / or expected to be optically inactive.

[0098] Figure 3 This is a simplified schematic diagram of an apparatus for rapid automated analysis of microplastic particles using OPTIR. An illumination source 340 emits a beam 342, which is selectively collected and / or substantially collimated by a collecting optics element 344, and then passes through a first polarizer 346. The polarized illumination beam 348 can be selectively focused by one or more additional optics elements 350 (including a beam splitter 352). If Kohler illumination is used, the light from the illuminator can be focused onto the back focal plane 354 of the objective lens 308. In any case, this arrangement allows a region of the sample to be illuminated by a spot 311 of polarized light.

[0099] Combination Figure 2 As described, the microplastic particles in sample 312 cause a rotation in the polarization of reflected / scattered / transmitted light. In the illustrated configuration, after the illumination light interacts with the sample, the backscattered / reflected light is collected by objective 308 and passes through beam splitter 352 to reach second polarizer 358. Second polarizer 358 substantially blocks the light in its original illumination polarization state and allows the scattered / reflected light, whose polarization has been rotated due to interaction with the microplastic particles, to pass through. In a transmission configuration, an additional objective is placed on the other side of the sample, and a second polarizer is arranged after the second objective.

[0100] exist Figure 3 In the configuration shown, light from the sample passes through polarizer 358 and is then focused onto the surface of camera 362 by one or more focusing optical elements 360 (e.g., tubular lenses). Camera 362 then generates a polarization image 364 of the sample, which includes a generally darker region 363 associated with the background filter / substrate and a brighter region 365 corresponding to the microplastic particles.

[0101] It should be noted that this configuration can be easily converted into a fluorescence microscope as an alternative to or supplement to a cross-polarization microscope. For example, polarizer 346 can be replaced with a fluorescence excitation filter, beam splitter 352 can be replaced with a suitable fluorescence dichroic mirror, and polarizer 358 can be replaced with a fluorescence emission filter.

[0102] Next, particle analysis software 366 processes one or more polarization images 364 (and / or fluorescence images) of the sample, calculates particle parameters in the images 364, such as particle size, area, and coordinate position (X,Y), and uses these parameters to generate a particle summary table 368. An optional position correction algorithm 370 can be applied before or after generating the particle summary table to create positional offsets to be applied to the particle positions calculated by the particle analysis software 366. Using the optionally position-corrected particle positions in the particle summary table, the sample can be sequentially positioned for OPTIR spectral analysis of multiple microplastic particles. In the illustrated configuration, sample 312 is mounted on an XY translation stage that allows for rapid movement of different microplastic particles, bringing them below the measurement beam of the OPTIR system.

[0103] Alternatively, the measurement beam of the OPTIR system can be moved relative to the sample, for example, by scanning the objective above the sample and / or scanning the OPTIR measurement beam within the objective's field of view. This is similar to that in laser scanning confocal microscopy and has been specifically described in the literature "Video-rate mid-infrared photothermal imaging by single-pulse photothermal detection per pixel" (Science Advances, June 14, 2023, Vol. 9, No. 24, DOI: 10.1126 / sciadv.adg8814). Any combination of translation of the sample, objective, and / or OPTIR measurement beam can be used to measure multiple microplastic particles. In some embodiments, OPTIR measurements can be performed in a wide-field configuration, for example using camera 362 (or an alternative camera not shown) as an OPTIR detector, as described in the applicant’s previously filed and published applications US 2023 / 0063843 and Cheng et al.’s US 2022 / 0381684, the contents of which are incorporated herein by reference in their entirety.

[0104] OPTIR measurement of microplastic particles

[0105] The following section describes an embodiment of an OPTIR system for infrared spectroscopy measurements of micron-sized microplastic particles, for example, after particle identification and localization have been performed using polarization microscopy and image particle analysis. (Referring to...) Figure 3An infrared light source 300 emits a beam of infrared radiation 302, which passes through an optional dichroic mirror 304 and is then guided by a reflector 306 onto an objective lens 308. The objective lens 308 focuses the infrared radiation into a spot 310, illuminating a portion of a sample 312 containing a cluster of microplastic particles distributed on a filter or other substrate. The infrared light source 300 is tunable and / or broadband, thus it can excite the microplastic particles in the sample 312 at multiple infrared wavelengths, which, for example, correspond to different absorption bands common in polymer materials. In one embodiment, the infrared light source 300 can scan at least a portion of the infrared fingerprint region, for example, at least 400-2000 cm⁻¹. -1 Sub-regions within the range. It can also support other tuning ranges, such as covering approximately 2700-3600 cm. -1 The CH bond / single bond region. The sample can also be illuminated with a probe beam to measure the photothermal changes in the sample due to the absorption of infrared light. For example, a probe beam source 314 emits a probe radiation 316, which can selectively pass through a polarizer and / or waveplate 318 before entering a polarizing beamsplitter 320. A portion of the probe radiation is reflected from the polarizing beamsplitter 320 and passes through an optional waveplate 322. In one embodiment, waveplate 318 is a half-wave plate and waveplate 322 is a quarter-wave plate. In this configuration, the polarization of the probe beam after passing through waveplate 322 is elliptically polarized. The dichroic mirror 304 in the configuration shown is configured to be reflective of the probe beam wavelength and transmissive of the infrared beam wavelength, thereby causing the infrared beam and the probe beam to collinearly combine. The opposite configuration can also be used, where the dichroic mirror is reflective of infrared light and transmissive of the probe light.

[0106] Collinear infrared and probe beams are reflected by mirror 306 to objective 308, which also focuses the probe beam into spot 322, illuminating sample 312. This spot at least partially overlaps with the infrared illumination region 310 of the sample. Absorption of infrared radiation by microplastic particles in the sample causes a local temperature increase in the microplastic particles, resulting in subtle changes in their size and reflection / scattering characteristics. These photothermal changes in the sample, in turn, cause changes in the intensity and distribution of probe light reflected, scattered, and / or transmitted from the sample. Probe light leaving the sample can be recollected by objective 308 in a coaxial reflection (epi) configuration, or as an alternative / supplement, by an additional objective or collecting optics (not shown) located on the same or opposite side of the sample. Collecting light on the opposite side constitutes a transmission detection configuration.

[0107] In the illustrated coaxial reflection detection configuration, the probe light collected from the sample is reflected by mirrors 306 and 304, returns, and is reflected again by waveplate 304 before entering polarization beamsplitter 322. Waveplate 322 can be used to generate a 90° polarization rotation, allowing the probe light returning from the sample to pass through polarization beamsplitter 320. Subsequently, the collected probe light can be selectively focused through optical element 324 onto confocal pinhole 326, and then selectively refocused through one or more additional optical elements 327 onto one or more detectors 328. The signal from the detectors is analyzed by demodulator / data acquisition (demod / DAQ) system 330 to generate a signal indicating the infrared absorption of the microplastic particle sample. More specifically, demodulator / data acquisition system 330 can be used alone or in conjunction with additional computing devices (e.g., CPU or other numerical processors) to determine variations in the collected probe radiation, which are coordinated with the pulse and / or energy modulation of the infrared beam 302. The signal indicating infrared absorption is measured over multiple infrared wavelength / wavenumber ranges, thereby generating multispectral measurements of particle infrared absorption. The infrared light source can perform continuous frequency sweeps within the mid-infrared wavelength range, for example, to construct measurements indicating the infrared absorption spectrum 336 of microplastic particles. Alternatively, the system can perform multiple measurements of the signal indicating infrared absorption at several discrete wavenumbers (i.e., corresponding to common spectral peaks in polymer materials). For example, measuring infrared absorption at a discrete number (e.g., 5-10) of infrared wavelengths can also be used to distinguish common polymer categories. In either case, the infrared absorption measurements of microplastic particles obtained at multiple infrared wavelengths are then compared with a spectral database 338 to determine whether the measured spectrum 336 matches well with any spectrum in the database.

[0108] In one embodiment, the measured spectrum 336 is compared with at least a portion of reference spectra in a spectral database, and a matching quality index (HQI) is calculated for each analyzed reference spectrum. If the highest HQI is higher than a user-configurable threshold, the chemical composition of the reference spectrum with the highest HQI is returned as the chemical identification (CHEM ID, 339) of the target microplastic particle. After all particle spectra have been compared and analyzed with the reference database, the HQI and chemical identification can be added to a particle summary table 341. The confocal pinhole 326 is particularly useful for the spectral analysis of microplastics for several reasons. First, microplastic particles may have many surfaces that scatter light, resulting in complex patterns in the collected probe light. The confocal aperture can purify the beam, enabling more robust detection of the sample's infrared absorption. The confocal pinhole 326 also provides improved and faster autofocus to optimize the focusing height of the objective lens 308 relative to the microplastic particle. Conventional OPTIR autofocus focuses on the OPTIR signal itself at a specific infrared wavelength. However, for unknown particles, it is difficult to determine the optical wavelength used for autofocus. Using a confocal stop allows for autofocus that is independent of the infrared wavelength, utilizing the intensity of the collected probe beam (i.e., the so-called DC signal). Since the DC signal does not need to be demodulated synchronously with the infrared pulse, DC autofocus is extremely fast, thus having minimal impact on the overall measurement throughput.

[0109] Figures 4A-4F Figure 1 to 10 show the use of... Figure 3 The method and apparatus described herein automatically measure the OPTIR spectra of various microplastic particles after identifying them using a cross-polarization microscope. Each image is labeled with the polymer abbreviation corresponding to the best-matching chemical identification. Figure 4A PMMA = Polymethyl methacrylate; Figure 4B PS = Polystyrene; Figure 4C PE = Polyethylene; Figure 4D PA = Polyamide; Figure 4E PI = Polyimide; Figure 4F PP = Polypropylene. In each figure, the solid line represents the measured spectrum, and the dashed line represents the best-matched reference spectrum. It should be noted that PMMA and PS are typically amorphous materials and therefore not expected to have birefringence, but they are easily detectable by cross-polarization microscopy in the size range of less than 20 µm in diameter.

[0110] Figure 5 This demonstrates the advantages of using cross-polarization microscopy to detect and characterize microplastic particles. Figure 5 A to Figure 5 C shows three images of the same region of a microplastic particle sample located on a filter membrane, obtained using different optical microscopy techniques.Figure 5 A is a bright field (BF) image of 500. Figure 5 B is the autofluorescence (AF) image 502. Figure 5 C represents cross-polarized (XPOL) image 504. Each image has a field of view of 296 µm and depicts a group of microplastic particles on a gold-plated polycarbonate filter membrane with a pore size of 0.8 µm. Figure 5 D to Figure 5 F shows the pair Figure 5 A to Figure 5 The results of thresholded particle analysis of the original images 500, 502, and 504 in C. Crosshairs indicate the locations of suspected detected particles.

[0111] for Figure 5 D in the middle Figure 5 Particle analysis of the BF image 500 in image A revealed 247 matching points in the thresholded image 506. However, a large proportion of these "particle" matching points were false positives related to the pore structure of the filter membrane. Automated OPTIR spectral analysis of the crosshair marker locations in image 506 yielded lower-than-expected success rates because many marker locations did not actually contain microplastic particles, but rather two or more closely spaced pores or other contrast deviations unrelated to microplastic particles. Figure 5 Image 508 in E is... Figure 5 Image 502 in B, representing autofluorescence (AF), is a thresholded particle analysis result. This analysis identified 83 particles with very high confidence; the locations marked by the cursors correspond to actual microplastic particles because background filters and their pores are very effectively suppressed in the AF image. It is worth noting that measuring particles with high AF levels using Raman spectroscopy can be challenging because AF can mask Raman signals. AF can serve as an effective method for highlighting certain types of microplastic particles. Figure 5 In image 502 of B, autofluorescence excitation was performed using ultraviolet light with a central excitation wavelength of 355–375 nm, and detection was performed at wavelengths of 425 nm and above. This excitation / emission wavelength range can detect a large number of microplastic particles.

[0112] The main drawback of using autofluorescence for particle detection is that while many plastics exhibit autofluorescence, many others do not. Therefore, using autofluorescence can bias the analysis of microplastic particles towards detecting those with autofluorescence, while suppressing the measurement of plastics without autofluorescence. This can potentially bias the analytical results regarding the distribution of plastics detected in environmental samples. In contrast, Figure 6A Particle analysis image 510 in F (based on) Figure 6BAnalysis of XPOL image 504 in C identified 124 particles, 50% more than the number of microplastic particles detected by autofluorescence image 508. The XPOL particle image demonstrates a very high microplastic particle detection rate, thus enabling successful particle identification using OPTIR. It is important to note that in some cases, two or more images can be combined, such as autofluorescence images and cross-polarized images, to detect and locate more microplastic particles than a single technique can detect.

[0113] High-speed, high-reliability spectroscopic analysis of 2-20 µm microplastic particles

[0114] Key metrics for this process include the number of particles analyzed per hour and the success rate of chemical identification of microplastic particles. Using the methods and apparatus described herein, developers have achieved high-quality measurements of microplastic particles smaller than 20 µm at rates exceeding 100 particles per hour, and in some embodiments even up to 160 particles per hour. The systems and methods described herein also achieve chemical identification success rates exceeding 50%, and in some embodiments even up to 80%.

[0115] To the best of our knowledge, no other technology currently enables such high-speed and highly reliable spectroscopic analysis of microplastics with diverse chemical compositions, encompassing most common plastics (including dark and colored plastic particles).

[0116] The following sections will discuss some key aspects of current methods and equipment that enable high throughput and high chemical identification accuracy in spectroscopic analysis:

[0117] Low optical power infrared beam and probe beam

[0118] A key factor in reliable OPTIR spectral measurements of various microplastic particles is the use of extremely low infrared and probe light power, for example, approximately 200 microwatts each. Microplastic particles smaller than 20 μm have very small thermal mass, and using high infrared / probe light power can easily lead to thermal / optical damage. Even if the infrared and probe light power are below the damage threshold, sufficient photothermal distortion may occur, inducing displacement or instability in the thin plastic membrane filter used to filter and support the microplastic sample for measurement. Researchers have found that controlling the infrared and probe light power to the sub-milliwatt level (e.g., approximately 200 μW) can significantly improve measurement stability while minimizing thermal / optical damage. These microwatt-level powers are typically lower than the power range used in conventional Raman spectroscopy, which generally employs milliwatts to tens of milliwatts or even higher. To support OPTIR measurements with such low probe light power, the detector 328 is preferably a high-sensitivity / low-illuminance detector, such as an avalanche photodiode (APD).

[0119] Position correction is performed on the particle positions calculated from the optical images.

[0120] Figure 6 illustrates a method for correcting image distortion within the field of view of an optical microscope. First, a camera image is acquired of a known reference sample using an objective lens for particle analysis. This known reference sample preferably has a pattern whose feature placement accuracy is higher than the accuracy required for locating microplastic particles within the desired size range. For example, if it is necessary to study particles as small as 2 μm, ideally, the feature spacing of the reference sample should be known to be 0.2 μm or higher. Figure 7A An optical image is shown, consisting of an array of metallic circles on a regular grid pattern with known spacing. Particle analysis is performed on this optical image to determine the centroid positions of the circles in the image, and these positions are compared with the known positions of the circles. Through this comparison, a correction table can be calculated, for example, as shown below. Figure 7B The vector field diagram 602 is shown in the image. The correction value, as a function of the XY position in the optical image, can be fitted by, for example, a polynomial or other function to correct the coordinates of the microplastic particles determined from subsequent optical microscopic images.

[0121] Calibration / correction of inter-objective offset

[0122] Often, it is desirable to use an objective lens, such as a low-power objective, to perform an initial survey of particle positions, followed by the acquisition of OPTIR measurement data for the particle array using a high-power objective lens and / or an infrared-compatible objective lens (e.g., a Schwarzschild objective). There may be offsets between the field centers of these two or more objectives, and these offsets can change over time due to thermal drift and / or errors in the mechanical mechanisms used to change different objectives. Therefore, it is necessary to measure and correct for any changes in the center positions of the scanning and measuring objectives. This can be achieved semi-automatically or fully automatically. One approach is to navigate to a reference feature (e.g., a reference mark on the sample or stage, or a particle, defect, or other identifiable feature). The position of a point on the reference feature is then marked with a cursor, and / or that point is moved to a known position in the optical image, such as the field center marked by crosshairs. Afterward, the objective lens is switched to the measuring objective lens, and the process is repeated. The current offset between the scanning and measuring objectives can be calculated by measuring the displacement of the XY platform required to center the reference feature in the field of view of the measuring objective, and / or by measuring the pixel position of the feature in the field of view of the measuring objective. Once this offset is known, it can be applied to all subsequent particle coordinate positions calculated in the particle analysis step. This correction can be used in conjunction with any of the corrections described below, or it can be used alone, depending on the relative proportion of the calculated correction value.

[0123] Another potential source of positioning error is offset, which can be caused by changes in the position of the OPTIR probe beam relative to the optical image. This problem can be addressed by acquiring an OPTIR image using a measurement objective and correlating the positions of features in the OPTIR image with the positions of features in the optical image obtained using a scanning objective and / or a measurement objective. A particularly effective method is to use a so-called “DC image,” which plots the intensity of the probe light collected from the sample as a function of the XY sample position. The advantages of a DC image are mainly twofold. First, it is unaffected by the infrared absorption characteristics of the sample; second, it allows for very rapid image acquisition because it does not require demodulation of minute photothermal intensity changes as in conventional OPTIR chemical imaging. Cross-correlation can be performed on the DC image and the optical image to calculate the offset vector, thereby correcting for changes in the position of the probe beam relative to the optical image. A single offset value can be calculated and applied, or a correction vector field similar to Figure 602 in Figure 6 can be calculated as needed. This offset field can be calculated automatically or with user assistance, such as manually moving the OPTIR DC image to achieve optimal superposition with the optical measurement image. This image overlay / correlation analysis can also be used to correct any rotation between the scanning axes of optical and OPTIR images.

[0124] Figure 7 illustrates how the offset between the probe beam and the center of the optical image is calculated using the correlation between the OPTIR DC image and the optical scan image. Figure 8 A magnified region 700 in the optical scan image is shown, containing various particles on a substrate. Box 702 marks the extent of the OPTIR DC image acquired over an area that at least partially overlaps with the optical image 700. The OPTIR DC image is converted to a binary image using an image thresholding technique and then superimposed on the optical image 700. In the OPTIR DC superimposed image, the particles are colored white, and the background is set to transparent. In this way, the amount of offset required to superimpose the image of particle 704 in the OPTIR DC image with the image of the same particle 706 in the optical scan image can be determined. Vector 707 shows an example of the magnitude and direction of the image offset required to align the OPTIR DC image 702 with the optical image 700. Measuring this offset can generate one or more correction vectors that can be applied to subsequent OPTIR measurements of microplastic particles. ​ Image 708 shows the result of repeating the process associated with image 700 after successful application of alignment and correction vectors. In this case, it is noteworthy that the white dots in the OPTIR DC image achieve good overlap with the particles in the optical image. Therefore, after this step, there is a good correlation between the position measured in the optical scan image and the commanded position of the sample relative to the probe laser (or equivalently, the position of the probe laser relative to the sample).

[0125] OPTIR imaging for microplastic pre-screening

[0126] Microplastic particles often coexist with non-plastic materials of similar size, where the concentration of non-plastic materials may be much higher than that of plastic materials. One study found that 66% of the Raman spectra were uninterpretable, “indicating the presence of a large amount of biological material…and salt precipitates originating from the laboratory preparation process itself” (Cabernard, L., et al., Comparison of Raman and Fourier transform infrared spectroscopy for the quantification of microplastics in the aquatic environment. Environmental science & technology, 2018. 52(22): p. 13279-13288.).

[0127] As a complement or alternative to polarization microscopy, OPTIR chemical imaging can be used to rapidly pre-screen sample regions, thereby excluding non-plastic particles as much as possible before investing significant time in OPTIR and / or Raman spectroscopy analysis of the particles. This can be achieved by rapidly evaluating infrared absorption at a few infrared bands, which are typically strong absorption bands of plastics, and optionally evaluating strong absorption bands of one or more non-plastic materials that are often mixed with microplastics (MPs).

[0128] ​ Graph 800 is shown, containing two curves: the first curve 802 represents the average infrared spectrum of hundreds of plastics, and the second curve 804 represents the average spectrum of many common non-plastic natural materials that often coexist with microplastics. Key absorption bands are labeled with their peak wavenumbers. From this graph, specific absorption bands that highly indicate the presence of polymers can be identified, such as 1720 cm⁻¹. -1 1464 cm -1 and 1242 cm -1 The absorption band at 1526 cm⁻¹; conversely, some absorption bands closely associated with organic matter can also be identified, such as at 1526 cm⁻¹. -1 and 1630 cm -1 The peaks at these or other indicative absorption bands. Therefore, acquiring OPTIR images at these or other indicative absorption bands can effectively predict whether particles are plastic and determine whether a more comprehensive OPTIR spectral analysis is warranted.

[0129] This pre-screening is particularly important when microplastic particles coexist with other "interference" (i.e., non-plastic particles of similar size). For example, in seawater samples, common interferences might include particles composed of sand, plant tissue, proteins, bone / teeth, shells, various salt compounds, and other substances. These interferences may contain silica, silicates and other substances, cellulose, calcium phosphate / hydroxyapatite, calcium carbonate, etc. Common interferences can be classified, and their spectral characteristics can be used to select the infrared wavelengths for performing pre-screening measurements, thus determining the likelihood that a given particle is microplastic or interference. These screening measurements can employ relatively "lenient" thresholds, for example, allowing a certain number of interferences to be included to increase the probability of selecting statistically significant groups of microplastic particles for measurement. For example, even if the pre-screening process excludes 75% of the interferences, it can still significantly improve measurement throughput without requiring the complete removal of all interferences. It should be noted that pre-screening can be performed by OPTIR imaging of the sample area and / or by measuring the OPTIR signal intensity in a few discrete bands (rather than the full infrared spectrum). Machine learning, deep learning, artificial neural networks, and / or artificial intelligence can be used to distinguish between microplastics and interfering substances. For example, an experienced spectroscopic analyst can manually train an artificial neural network using a dataset to classify the spectra of different particles as either microplastics or interfering substances. Then, based on the training dataset, the relative absorption of a small number of infrared wavelengths can be used to predict whether a particle is a microplastic or an interfering substance. The same method can be used to distinguish microplastics of different compositions, for example, by taking measurements at multiple wavelengths corresponding to common infrared absorption bands of plastic materials, particularly those wavelengths that can distinguish different plastic materials.

[0130] The pre-screening described herein may employ optical detection, photothermal imaging, autofluorescence imaging, cross-polarization imaging, or any combination thereof. For certain samples, using one or more of these techniques may achieve more accurate or convenient screening compared to other techniques. In some embodiments, the aforementioned machine learning techniques may include pre-screening using techniques determined by algorithms to be superior to or most likely to be superior to other techniques.

[0131] The embodiments described herein are merely examples. Those skilled in the art will understand that modifications, rearrangements, alternative processes, and alternative elements may be made to these embodiments without departing from the technical concepts set forth herein, and such modifications should still fall within the scope of the technical teachings disclosed herein. One or more steps, processes, or methods described herein may be performed by one or more appropriately programmed processing devices and / or digital devices.

[0132] According to embodiments, some actions, events, or functions in the method steps described herein may be performed in a different order, or may be added, combined, or even omitted entirely (e.g., not all described actions or events are necessary for the implementation of the algorithm). Furthermore, in some embodiments, actions or events may be performed in parallel rather than sequentially.

[0133] The various exemplary logic modules, optical elements, control elements, and method steps described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the substitutability between hardware and software, the exemplary components, modules, and steps have generally been described from a functional perspective. Whether a hardware or software implementation is adopted depends on the specific application and the design constraints imposed on the overall system. The functionality may be implemented in different ways for different applications, but such implementation choices should not be construed as departing from the scope of this disclosure.

[0134] The various exemplary logic and functional modules described in conjunction with the embodiments disclosed herein can be implemented or executed by a machine, such as a processor configured with specific instructions, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, discrete gate circuits or transistor logic, discrete hardware components, or any combination thereof, to perform the functions described herein. The processor may be a microprocessor; alternatively, the processor may be a controller, a microcontroller, a state machine, or a combination thereof. The processor may also be implemented by a combination of multiple computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, a configuration of one or more microprocessors combined with a DSP core, or any other similar configuration.

[0135] The elements of the methods, processes, or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly in hardware, as software modules executed by a processor, or a combination of both. Software modules can be stored in RAM, flash memory, ROM, EPROM, EEPROM, registers, hard disks, removable storage disks, CD-ROMs, or any other computer-readable storage medium known in the art. Exemplary storage media can be coupled to a processor, enabling the processor to read information from and write information to the storage medium; alternatively, the storage medium can be integrated with the processor. The processor and storage medium can be co-located in an ASIC. Software modules can include computer-executable instructions for causing a hardware processor to execute corresponding computer-executable instructions.

[0136] The conditional language used herein, such as “may,” “possibly,” “perhaps,” “for example,” etc., unless otherwise expressly stated or clearly understood in context to have a different meaning, is generally used to indicate that certain embodiments include corresponding features, elements, and / or states, while other embodiments do not. Therefore, such conditional language should not generally be construed as implying that any feature, element, or state is necessary for one or more embodiments, nor should it be construed as implying that any embodiment necessarily includes logic for determining whether to include or perform such features, elements, or states.

[0137] The terms “including,” “comprising,” “having,” and “involving” have the same meaning and are used in an open-ended, inclusive manner, not excluding additional elements, features, actions, or operations. Furthermore, the term “or” is used in an inclusive sense (rather than an exclusive one), for example, when connecting a list of multiple elements, indicating one, several, or all of the listed elements.

[0138] Unless otherwise expressly stated, optional expressions such as "at least one X, Y, or Z" should be understood to mean that the object can be any one of X, Y, or Z, or any combination thereof (e.g., X, Y, and / or Z). Therefore, such expressions should not be construed as requiring that at least one X, one Y, and one Z must be included in some embodiments.

[0139] The terms “approximately,” “about,” and similar expressions have the same meaning and are used to indicate that the numerical value they modify has an understandable range, such as ±20%, ±15%, ±10%, ±5%, or ±1%. The term “substantially” is used to indicate that a result (e.g., a measurement) is close to a target value, where “close” can, for example, mean that the result is within 80%, 90%, 95%, or 99% of the target value.

[0140] Unless otherwise expressly stated, articles such as “a” or “an” should generally be interpreted as including one or more of the said objects. Therefore, the expression “a device configured to…” should be understood to include one or more of the said devices, which may also be configured cooperatively to perform the said function. For example, “a processor configured to perform A, B, and C” could include a combination of a first processor for performing A and a second processor for performing B and C.

[0141] While the foregoing detailed description has shown, described and pointed out novel features in conjunction with exemplary embodiments, it should be understood that various omissions, substitutions and modifications may be made to the form and details of the illustrated apparatus or method without departing from the spirit of this disclosure. As those skilled in the art will understand, some embodiments described herein may be implemented in a form that does not provide all the features and effects described herein, as some features may be used or implemented independently of the others. All modifications falling within the meaning and equivalents of the claims should be included within their scope of protection.

[0142] This document has described various embodiments of systems, apparatuses, and methods. These embodiments are merely examples and are not intended to limit the scope of the claimed invention. Furthermore, it should be understood that different features in the described embodiments can be combined in various ways to form a large number of other embodiments. Moreover, although specific materials, sizes, shapes, structures, and locations, etc., that can be used in the disclosed embodiments have been described herein, other solutions besides those disclosed herein may be employed without departing from the scope of the claims.

[0143] Those skilled in the art will understand that the subject matter herein may include fewer features than those shown in any single embodiment. The embodiments described herein are not an exhaustive list of possible combinations of features. The embodiments are not mutually exclusive combinations of features, but may include combinations of different individual features selected from different embodiments. Furthermore, unless otherwise stated, elements described in connection with one embodiment may also be implemented in other embodiments, even if not explicitly described in those other embodiments.

[0144] While dependent claims may refer to a specific combination with one or more other claims in the claims, other embodiments may also include combinations of the subject matter of the dependent claim with any other dependent claim, or combinations of one or more features with other dependent or independent claims. Unless explicitly stated otherwise, all such combinations are considered as embodiments proposed herein.

[0145] Any inclusion of the foregoing references shall be limited to the inclusion of subject matter that contradicts the express disclosure herein. Any inclusion of the foregoing references is also limited to the exclusion of any claims thereof. Furthermore, any inclusion of the foregoing references is limited to the exclusion of any definitions provided therein unless expressly incorporated herein.

[0146] In interpreting the claims, it is intended that the provisions of 35 USC §112(f) should not apply unless the specific expressions “means for” or “step for” are expressly used in the claims.

Claims

1. A method for automatically characterizing a sample containing a population of microplastic particles using photothermal infrared spectroscopy, the method comprising: a) acquiring a polarized optical image of the sample; b) analyzing the polarized optical image to automatically identify a plurality of microplastic particle locations in the sample; c) automatically positioning a microplastic particle under a probe beam of a photothermal infrared spectroscopy system according to the plurality of microplastic particle locations; d) illuminating the microplastic particle with a plurality of infrared wavelengths; e) collecting at least one of reflected, scattered, or transmitted probe beam light from the microplastic particle at a detector; f) measuring a change in the collected probe light from the microplastic particle, the change corresponding to an infrared absorption of the microplastic particle.

2. The method of claim 1, further comprising determining a chemical identity of the microplastic particle using a signal indicative of the infrared absorption of the microplastic particle.

3. The method of claim 1, wherein the probe beam is generated by a fixed probe beam source, and positioning a microplastic particle under the probe beam comprises translating the sample relative to the probe beam source while the probe beam source remains fixed.

4. The method of claim 1, wherein the sample remains in a fixed position, and positioning a microplastic particle under the probe beam comprises translating the probe beam source relative to the sample.

5. The method of claim 1, wherein the probe beam is generated by a probe beam source that remains in a fixed position relative to the sample, and positioning a microplastic particle under the probe beam comprises translating the probe beam relative to a field of view of a focusing optical element prior to the probe beam interacting with the sample.

6. The method of any preceding claim, further comprising repeating steps c) and d) to illuminate a plurality of microplastic particles corresponding to at least some of the plurality of microplastic particle locations.

7. The method of any of claims 1-5, wherein the microplastic particles have an effective size in at least one dimension of less than 20 microns.

8. The method of claim 7, wherein the microplastic particles have an effective size in at least one dimension of less than 5 microns.

9. The method of claim 1, wherein the polarized optical image is created using cross-polarized microscopy.

10. The method of claim 1, wherein the polarized optical image includes images of microplastic particles of a polymer that do not exhibit bulk birefringence.

11. The method of claim 1, wherein the polarized optical image includes images of microplastic particles of a polymer that are substantially amorphous in the bulk.

12. The method of any of claims 1-5 or 8-10, wherein measuring comprises analyzing a distribution of image pixel intensities in the polarized optical image to create a list of microplastic particle locations.

13. The method of any of claims 1-5 or 8-10, wherein analyzing comprises identifying regions in the polarized optical image that are above a threshold value.

14. The method of any one of claims 1-5 or 8-10, wherein the analyzing comprises identifying regions in the polarized optical image that are above a local average of a region in the polarized optical image.

15. The method of any one of claims 1-5 or 8-10, wherein the positioning step comprises at least one position correction to compensate for a positional error between a first microscope objective used to create the polarized optical image and a second microscope objective used to focus the probe light beam onto the microplastic particles.

16. The method of claim 15, wherein the position correction is applied automatically without user intervention.

17. The method of claim 11, wherein the position correction is applied automatically after a user performs a manual calibration step.

18. The method of claim 15, wherein the position correction comprises a correction for image non-uniformity caused by the first microscope objective.

19. The method of any one of claims 1-5 or 8-10, wherein the probe light beam is provided at a sub-milliwatt power level, and the illumination at multiple infrared wavelengths is also provided at a sub-milliwatt power level.

20. The method of any one of claims 1-5 or 8-10, wherein collecting the probe light comprises passing the probe light through a pinhole before the probe light interacts with the detector.

21. The method of claim 20, wherein the pinhole is a confocal pinhole, and the method further comprises performing a direct current autofocus to rapidly adjust the focus of the objective on the plurality of microplastic particles.

22. An optical system for automatically characterizing a sample containing a population of microplastic particles, the optical system comprising: a cross-polarization microscope subsystem configured to acquire a polarized optical image of the sample and identify locations of a plurality of microplastic particles in the sample; a photothermal infrared spectroscopy subsystem configured to illuminate the locations of the plurality of microplastic particles with multiple infrared wavelengths to detect respective optical photothermal infrared absorption of a microplastic particle at each of the locations of the plurality of microplastic particles.

23. The optical system of claim 22, further comprising an objective lens, wherein: the cross-polarization microscope subsystem comprises an illumination light source configured to direct an illumination light beam through the objective lens toward the sample; the photothermal infrared spectroscopy subsystem comprises an infrared light source configured to direct an infrared light beam having the multiple infrared wavelengths through the objective lens onto the sample; and the photothermal infrared spectroscopy subsystem comprises a probe light source configured to direct a probe light beam through the objective lens toward the sample.

24. The optical system of claim 23, wherein the illumination light source is configured to operate at a sub-milliwatt power level, and the infrared light source is configured to operate at a sub-milliwatt power level.

25. The optical system of claim 23, further comprising a confocal pinhole disposed along a path of the probe light between the objective lens and the detector.

26. The optical system of claim 23, further comprising a movable reflector configured to selectively direct the illumination light beam toward the objective lens or to selectively direct the infrared light beam and the probe light beam into the objective lens.

27. The optical system of claim 22, wherein the optical system is configured to characterize at least 100 particles per hour.

28. The optical system of claim 22, wherein the optical system is configured to characterize a plurality of particles with a success rate of over 50%.

29. A method for automated characterization of a sample containing a population of microplastic particles using photothermal infrared spectroscopy, the method comprising: a) acquiring at least one photothermal infrared image of the sample at a plurality of infrared wavelengths; b) analyzing at least one optical image and / or photothermal infrared image to identify a plurality of particle locations; c) analyzing at least one photothermal infrared image of the sample to select a subset of locations from the plurality of particle locations as possible microplastic particle locations, d) automatically positioning a particle under a probe beam of a photothermal infrared spectroscopy system according to the plurality of possible microplastic particle locations; d) illuminating the particle with a plurality of infrared wavelengths; and e) detecting a change in probe light collected from the particle, the change corresponding to an infrared absorption of the particle.

30. The method of claim 29, further comprising determining a chemical identity of the population of microplastic particles in the sample.

Images