Method and apparatus for multi-modal polarization holographic spectroscopy for material analysis

Abstract

A two-dimensional spectrum recording apparatus has a collimated laser light source that irradiates a sample plate on which a sample is placed. Light passing through the sample forms object light, and light not passing through the sample forms reference light that interferes with the object light, thereby forming a holographic image at the sample plate. A beam splitter receives the collimated laser after the collimated laser passes through the sample and forms separated first and second beams. A linear polarizer receives the first light beam and passes the first light beam to a polarization camera having a full Stokes mask at its input. The polarization camera records light that passes through the Stokes mask. A slit aperture receives the second light beam and transforms the second light beam into slit light applied to a diffraction grating. A diode array receives the second light beam after the second light beam passes through the diffraction grating and forms a recorded spectrum.

Description

Cross-references to related patent applications This application claims priority under 35 U.S.C. Section 119(e) to U.S. Application No. 63 / 516,228, filed on July 28, 2023, which is incorporated herein by reference in its entirety. Technical Field

[0001] The invention relates to a multi-modal polarization holographic spectrometer which can simultaneously examine the biochemical, physical and morphological characteristics of a detected sample. Background Art

[0002] For biomedical, environmental and chemical research, non-invasive, identifiable sample and material analysis without labeling or extensive sample preparation is highly desirable. Material analysis and characterization by spectroscopy is accurate but slow. Spectral signals need to be recorded pixel by pixel, which is time consuming. Temporal redundancy does not guarantee high-throughput material identification.

[0003] The morphological and structural characteristics are related to the composition of the sample and to the weathering and transfer processes of the sample object. The morphological characteristics can be precisely observed by microscopy, i.e. scanning electron microscopy (SEM). However, these characteristics, such as roundness, are not discernible and vary with the production process.

[0004] Polarization images recorded by current polarization imaging (PI) systems are poor and cumbersome. Manual rotation and adjustment of the polarizer is required, which limits the use of real-time and automatic imaging processes. Chinese Patent No. CN104198040A discloses a system that requires two laser sources with different wavelengths. Polarization image recording requires modulation of the beams of the two lasers. The method disclosed in this patent is unable to record spectra and capture spectral information related to materials. Moreover, there is no feature analysis method based on automatic imaging processes or machine learning in this patent.

[0005] U.S. Published Application 2016 / 0011050A1 discloses the use of a polarizer and an analyzer to identify differences introduced by a sample. Image recording requires manual rotation, which is cumbersome and reduces the speed of image recording. In addition, this patent requires a specially designed retarder element to implement its function. This retarder element needs to be precisely manufactured, which limits the industrial production of the system and method. Compared to recording rich information on holograms and spectra, U.S. Patent No. 7,649,660 attempts to achieve high-throughput spectral recording through holographic multiplexing for material identification through high-precision and multimodal imaging. The system and method of this patent cannot record the polarization information of the sample. Summary of the invention

[0006] In order to solve the problems of the prior art, the present invention proposes a device and method for realizing high-throughput two-dimensional (2D) spectrum recording. Compared with the widely used Raman spectroscopy and Fourier transform infrared spectroscopy (FT-IR), the present invention records 2D spectra in both the x-axis direction and the y-axis direction. This strategy significantly improves time efficiency.

[0007] The present invention uses polarization holography to encode holographic patterns through spatial coherence with polarized light. By separating and filtering overlapping holograms in the frequency domain, the morphological surface and three-dimensional structure of the sample can be retrieved. In addition, polarization holography can be used to analyze the birefringence properties of the sample through the optical polarization axis azimuth and phase delay. The absorption spectrum reveals the structure and composition of the material. The position and intensity of the absorption peak in the spectrum indicate the presence of a specific molecular structure or functional group. In general, the system and method of the present invention provide an advanced and discernible multimodal material analysis tool.

[0008] Polarization features (such as birefringence and anisotropy) are extracted and calculated for complementary morphological features. In addition, refractive index, transmittance and fringe contrast are obtained by holographic imaging with holograms. Rich physical and structural information is collected by the system and provides discernible features for sample material identification.

[0009] In addition, synthetic polarization imaging recording is achieved in system hardware with a multi-directional polarization recording mask. Advanced image configuration and registration methods are used to illustrate images with specific polarization states. No additional optical polarization analyzer is required, which eliminates troublesome manual hardware adjustments and allows automatic imaging. Moreover, the apparatus and method of the present invention eliminate the field of view and light source intensity limitations of existing spectral analysis systems, and add phase and molecular structure characteristics through PI and digital holography (DH). The system is integrated into advanced electrical imaging sensor hardware and implemented through advanced image processing methods, thereby providing multi-dimensional fingerprints for real-time material analysis.

[0010] The invention has a wide range of applications, including biomedical research, materials science, environmental assessment and biophysical analysis, for example, microplastic detection for marine pollution monitoring, biological tissue diagnosis for cancer research, chemical and biological agent detection, semiconductor material inspection with defect-oriented detection, etc.

[0011] The system hardware setup of the present invention is relatively simple and compact. A specially manufactured polarization mask is installed inside the imaging sensor and is not easily observable from the outside. This novel design and specific innovations improve the compactness of the system and reduce the system cost, making it suitable for commercial and industrial use. In addition, advanced image registration, feature extraction and processing methods and algorithms are integrated into the present invention. Therefore, the present invention is a composite system in the form of an intelligent tool that exceeds the limitations of a single imaging and spectral analysis system. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The above and other objects and advantages of the present invention will become more apparent when taken in conjunction with the following detailed description and the accompanying drawings, in which like reference numerals represent like elements, and in which:

[0013] Figure 1A is a schematic diagram of a system according to the present invention, and Figure 1B Showing its embedded holographic part;

[0014] Figure 2 is a schematic diagram of the working principle for forming an absorption spectrum;

[0015] Figure 3A is an image of the recorded chromatogram according to the invention, and Figure 3B is a graph of the extracted spectral features;

[0016] Figure 4 A diagram showing the working principle of the polarization imaging device parts with a Stokes mask of the device of the present invention;

[0017] Figure 5 Visualization of holographic patterns of test samples;

[0018] Figure 6 is a pattern for the holographic pattern and the spatial difference between the two fringes;

[0019] Figure 7 are correlation matrices for polarization, holographic, textual, spectral, and morphological features; and

[0020] Figure 8 Classification confusion matrix for a pilot study of the materials analysis method and apparatus of the present invention. DETAILED DESCRIPTION

[0021] A schematic diagram of the apparatus of the system of the present invention is shown in Figure 1AThe system can innovatively record polarization-sensitive holographic patterns and spectra of a sample simultaneously. In an exemplary embodiment, the laser light 10 from the laser source 11 has a wavelength of 532 nm and is first collimated by a collimating lens 13. After the laser light passes through the sample object on the sample plate 15, the object information is encoded with the laser wavefront. The object light wave is then separated into two optical paths by a beam splitter 17.

[0022] exist Figure 1A In the setup in , the holographic imaging system is implemented by an embedded setup. The reference light wave and the object light wave use the same optical path, such as Figure 1B As shown. A portion of the light having a plane wavefront from the light source passes through the sample and forms an object light wave. The other portion of the plane wave is a plane reference light wave. A holographic image is formed at the sample plate 15 in FIG. 1 and is recorded at the sensor 27. One path of the object light interferes with the plane reference light and forms a holographic image. Specifically, a collimated laser source is irradiated on a sample plate on which sample particles are placed. The light that passes through the sample forms the object light, and the light that does not pass through the sample forms the reference light that interferes with the object light, thereby forming a holographic image at the sample plate.

[0023] The first light path passes from the beam splitter 17 through a linear polarizer 21 and then through a quarter wave plate 23 to a specific polarization camera 25. This camera is used to replace the prior art charge-coupled device (CCD) or complementary metal oxide semiconductor (CMOS) camera for full Stokes holographic image recording. The other light path from the beam splitter passes in turn through an entrance or aperture slit 31 to form light, which is dispersed by a diffraction grating 35 after being reflected by a mirror 33. The diffraction grating forms a 2D spectrum captured on a diode array 37. Mirrors are placed in the optical path to increase the compactness of the system by folding the light.

[0024] like Figure 2 As seen in FIG. 1 , the light beam 10 passes through a sample plate 15 or cuvette, which is a straight-sided, optically transparent container used to hold a liquid sample in a spectrophotometer. A portion of the light beam is absorbed and scattered by the molecules of the liquid sample. Thereafter, the light beam is filtered through a slit 31 and is also Figure 1A The light is dispersed by a diffraction grating 35 as shown in FIG. A diode array 37 receives the dispersed light from the grating 35 and is configured to record the spectrum of the light after it has passed through the sample. Figure 2 The difference from part of FIG. 1 is that it does not show the use of mirror 33 to make the layout more compact.

[0025] A spectrum is a graphical representation of the response of light as a function of wavelength, which provides insight into the structure and properties of a sample material. Absorption spectroscopy involves shining UV or visible light onto a sample and observing its response to this stimulus. When an atom or molecule absorbs a photon, it gains energy from the absorbed particle. This can induce a transition of the particle to a higher energy level or excited state, which requires that the energy of the absorbed photon matches the energy difference between the ground state and the excited state of the absorbing particle. Therefore, only photons with a specific energy or wavelength can be absorbed.

[0026] Absorption spectroscopy is obtained by measuring the amount of light absorbed by a sample as a function of the wavelength or frequency of the light. This provides valuable insight into the electronic structure and energy levels of the absorbing particles. Specifically, absorption spectroscopy is a valuable tool for identifying specific atoms or molecules in a sample and determining their energy levels and transition probabilities. Absorption spectroscopy is widely used in the fields of chemistry, physics, and biology to study material properties and identify unknown substances.

[0027] Figure 3A Displays the recorded spectrum, where different colors represent different frequencies or wavelengths of light waves. Figure 3B A graph showing the wavelength position of the peak.

[0028] By utilizing a spectrophotometer or imaging sensor, the intensity of light transmitted through the sample is measured as it varies with wavelength. An absorption spectrum can be obtained by subtracting the intensity of the transmitted light from the intensity of the incident light and plotting this result as a function of wavelength.

[0029] The Beer-Lambert law is the basic principle used to calculate the absorbance of a sample. The absorbance is defined as the negative logarithm to the base 10 of the sample's transmission. Mathematically, Among them I 原始 and I 样品 is the intensity of light before and after passing through the sample. The Beer-Lambert law describes the absorbance (A) of a sample, its ability to absorb light of a specified wavelength, and its absorption rate (ε A ), the degree of energy absorption is defined as: A=ε A ·L·c, (3) Where L is the optical path distance between samples and c is the sample concentration. A In L·mol -1 cm -1 The unit of c is mol·L -1 .

[0030] The working principle of a specific polarization camera with a Stokes mask 26 is Figure 4The Stokes mask 26 is placed inside the polarization camera 25 and the imaging sensor 27 ( Figure 1B ) before. Figure 4 The image of the Stokes mask shown in front of the camera in the image but actually inside it is a magnified image in order to explain how the Stokes mask is made and works. The yellow box gives a magnified image of the four cells in the Stokes mask, each cell in the Stokes mask has a corresponding polarization state of 0 degrees, 45 degrees, 90 degrees and 135 degrees. Therefore, the light from the sample image will be passed to the cell of the sensor according to its polarization.

[0031] By using the system of the present invention, the polarization-rich features of the sample under examination can be captured and facilitate birefringence and material analysis. The working principle is described as follows. The Jones matrix of a non-depolarized sample is described as Where φ is the phase delay and θ defines the optical axis orientation. In the system of the present invention, information of four polarization states is recorded, which are 0°, 45°, 90° and 135° states. The original light intensity recorded in each polarization state is expressed as

[0032] The two dummy variables that simplify the expression are defined as ε 1 = sinφsin 2Θ (9) ε 2 = sinφcos 2Θ (10) The phase retardation and optical axis orientation of the sample are rewritten as

[0033] The holographic pattern contains information about the structure and refractive index of the sample being examined. Figure 5 The recorded holographic sample images are shown. MP1 represents microplastic sample 1, MP2 represents microplastic sample 2, and BG represents a background area without a sample. The spatial and phase differences between the interference of MP1 and MP2 exist in the intensity difference, holographic fringe contrast, and holographic fringe pattern. The spatial and phase differences between the two interference fringes reveal the refractive index, structure, and surface information of the sample. Figure 5 Middle, scale bar: 0.5 mm.

[0034] Figure 6 The light intensity distribution of the holographic fringe pattern is shown in Figure 2. where x′ is the horizontal position of the fringe in the imaging plane in Cartesian coordinates. cis the light intensity of the central bright spot. k defines the wave number. D is the distance between the light source and the imaging plane. d is the distance between the two light waves. In the imaging plane, Figure 6 As shown in , the distance between two stripes is expressed as

[0035] For the optical system, a light source with one wavelength is used. For an imaging plane at a certain distance, the displacement between the two holographic fringe patterns depends only on the sample refractive index n.

[0036] The recorded polarization holographic images and spectra are processed by machine learning and deep learning methods for automatic and in-depth feature extraction and material analysis. Figure 7 The correlation matrix for the features is shown in Figure 7 , the vertical axis corresponds to the correlation between features, where -1 indicates negative correlation and 1 indicates positive correlation. Figure 7 The color and size of the circles in represent the value of the correlation, where the darker the color and the larger the circle, the stronger the correlation.

[0037] Figure 7 The features in show high independence and provide rich information for material identification. Specifically, Figure 7 The correlation matrix shown in shows the part of the features extracted from the original image data. These features include polarization-related (phase delay average, phase delay standard deviation, etc.), holographic (contrast, etc.) and spectral-related characteristics. Relatively high independence is shown among the feature categories. The extracted image features are used for further category classification.

[0038] Figure 8 The confusion matrix shows the results of a pilot study of the material analysis method and device of the present invention. It indicates the classification results among different microplastic samples and categories. The selected research sample materials are polycarbonate (PC), polyethylene terephthalate (PET), polymethyl methacrylate (PMMA), polypropylene (PP) and polyvinyl chloride (PVC). Figure 8 In the figure, the color bar indicates the classification category. -2 to 2 gives 5 categories, indicating 5 microplastic categories (PC, PET, PMMA, PP, PVC). The last five rows of the matrix also have group relationships. The color and intensity show their relationship. Figure 8 The classification confusion matrix demonstrates the salient features for material analysis utilizing the system of the present invention.

[0039] The above is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modification or substitution obvious to a person skilled in the art will fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be subject to the scope of protection of the claims.

[0040] Although the present invention has been described with respect to certain embodiments, it should be understood that various modifications of the description will become apparent to those skilled in the art after reading the description. Therefore, it should be understood that the invention disclosed herein is intended to cover such modifications that fall within the scope of the appended claims.

Claims

1. A two-dimensional spectrum recording device, characterized in that: include: A collimated laser source, wherein the collimated laser source is irradiated onto a sample plate on which a sample is placed, light passing through the sample forms object light, and light not passing through the sample forms reference light, and the reference light interferes with the object light, thereby forming a holographic image at the sample plate; a beam splitter that receives the collimated laser light after the collimated laser light passes through the sample and forms a separated first light beam and a second light beam; a linear polarizer, the linear polarizer receiving the first light beam; a polarization camera having a full Stokes mask at its input and receiving the first beam after it passes through the linear polarizer, the polarization camera recording the light that passes through the Stokes mask; a slit aperture that receives the second light beam and transforms the second light beam into a slit light; a diffraction grating that receives the second light beam after the second light beam passes through the slit aperture; as well as A diode array receives the second light beam after it passes through the diffraction grating and forms the recorded spectrum.

2. The two-dimensional spectrum recording device according to claim 1, characterized in that: The collimated laser source is a laser diode and a collimating lens. The laser diode generates a laser beam, and the collimating lens receives the laser and converts the laser into a collimated laser beam.

3. The two-dimensional spectrum recording device according to claim 1, characterized in that: Further included is a mirror located after the slit aperture to bend the second light beam to be parallel to the first light beam to make the structure more compact.

4. The two-dimensional spectrum recording device according to claim 1, characterized in that: The Stokes mask is a series of four-element groups, each of which has polarizations of 0°, 45°, 90°, and 135°, respectively.

5. The two-dimensional spectrum recording device according to claim 4, characterized in that: The second light beam passes through a linear polarizer after passing through the beam splitter and passes through a quarter wave plate after passing through the linear polarizer and before passing through the polarization camera.

6. The two-dimensional spectrum recording device according to claim 1, characterized in that: The recorded holographic microplastic sample 1 (MP1) and holographic microplastic sample 2 (MP2) are formed, and the spatial and phase difference between the interference of MP1 and MP2 exists in the intensity difference, the contrast of the holographic fringes and the pattern of the holographic fringes, and the spatial and phase difference between the two interference fringes discloses the refractive index, structure and surface information of the sample.

Images