Crystal quality detection method and system

Abstract

The invention provides a crystal quality detection method and system, and the method comprises the steps: positioning a wafer sample to a position below an objective lens through a sample conveying platform, and enabling the focus of the objective lens to be at the center of the wafer sample; when the objective lens linearly moves within the radius range relative to the wafer sample, carrying out microscopic imaging to obtain an image; when the intensity fluctuation of the image is greater than a set threshold value, performing Raman imaging to obtain a three-dimensional hyperspectrum; processing each Raman spectrum in the three-dimensional hyperspectrum to obtain a spectrum peak curve, a peak width curve and a peak position curve; when a first-order derivative tends to zero after the spectral peak curve, the peak width curve and the peak position curve are fitted, longitudinal scanning is ended; converting the spectrum peak intensity, the peak width and the peak position generated by the three-dimensional hyperspectrum into characteristic information, and calculating a wafer quality coefficient; determining the quality of the wafer sample according to the wafer quality coefficient; the measurement of different depths can be carried out on the crystal (including a wafer sample); and nondestructive, rapid and accurate quality analysis at normal temperature and normal pressure is realized.

Description

Technical Field

[0001] The present invention relates to the field of crystal detection technology, and in particular to a crystal quality detection method and system. Background Art

[0002] The core goal of crystal quality assessment is to optimize material functions and application effects by detecting the structure, defects and physical properties of crystals. Currently, the main technologies used for crystal quality assessment are X-ray diffraction and X-ray tomography.

[0003] X-ray diffraction is a method of analyzing crystal structure, orientation, stress, and defects (such as polytypes and dislocations) by generating diffraction patterns through the scattering of X-rays by periodically arranged atoms in the crystal.

[0004] X-ray tomography utilizes the differences in X-ray absorption inside the crystal and reconstructs crystal projections taken at different angles through an algorithm to obtain three-dimensional imaging of the internal structure of the crystal and reveal the two-dimensional or three-dimensional distribution of defects inside the crystal.

[0005] X-ray diffraction lacks sensitivity and is insensitive to micro-scale defects. The focused X-ray spot is typically a line spot, typically on the order of millimeters, making it impossible to measure at different depths within the crystal surface.

[0006] X-ray tomography can achieve high-resolution tomographic imaging of the surface of crystals, but the contrast differences caused by the X-ray projection are difficult for the detector to distinguish. At the same time, X-ray tomography often requires 360-degree rotation of the sample. High-resolution imaging can only be obtained for smaller crystals, but cannot be achieved at the wafer level. Summary of the Invention

[0007] In view of this, the purpose of the present invention is to provide a crystal quality detection method and system, which can measure crystals (including wafer samples) at different depths and realize non-destructive, rapid and accurate quality analysis at room temperature and pressure.

[0008] In a first aspect, an embodiment of the present invention provides a crystal quality detection method, the method comprising:

[0009] Positioning the wafer sample under the objective lens by using the sample transport stage so that the focus of the objective lens is at the center of the wafer sample;

[0010] When the objective lens moves linearly within a radius relative to the wafer sample, microscopic imaging is performed to obtain an image;

[0011] When the intensity fluctuation of the image is greater than a set threshold, Raman imaging is performed to obtain a three-dimensional hyperspectrum;

[0012] Processing each Raman spectrum in the three-dimensional hyperspectral spectrum to obtain a spectrum peak curve, a peak width curve and a peak position curve;

[0013] When the first-order derivative of the spectrum peak curve, the peak width curve and the peak position curve after fitting tends to zero, the longitudinal scan ends;

[0014] Converting the peak intensity, peak width, and peak position generated by the three-dimensional hyperspectral spectrum into characteristic information and calculating the wafer quality factor;

[0015] The quality of the wafer sample is determined according to the wafer quality coefficient.

[0016] Furthermore, each Raman spectrum in the three-dimensional hyperspectral spectrum is processed to obtain a spectrum peak curve, a peak width curve and a peak position curve, including:

[0017] Inputting each Raman spectrum in the three-dimensional hyperspectral spectrum into a Lorentz fitting algorithm to obtain a three-dimensional graph of spectral peak intensity, a three-dimensional graph of peak width, and a three-dimensional graph of peak position;

[0018] The three-dimensional graph of the spectrum peak intensity, the three-dimensional graph of the peak width and the three-dimensional graph of the peak position are respectively scanned with the Z axis to obtain the spectrum peak curve, the peak width curve and the peak position curve.

[0019] Furthermore, the characteristic information includes stress, and converting the peak intensity, peak width, and peak position generated by the three-dimensional hyperspectral spectrum into characteristic information includes:

[0020] The stress is calculated according to the following formula:

[0021] Δv(cm -1 )=K×σ(Gpa)

[0022] Wherein, Δv is the Raman peak shift, which represents the shift of the peak position relative to the stress-free state, K is the stress sensitivity coefficient, and σ is the stress.

[0023] Furthermore, the characteristic information includes doping concentration, and converting the peak intensity, peak width and peak position generated by the three-dimensional hyperspectral spectrum into characteristic information includes:

[0024] The doping concentration is calculated according to the following formula:

[0025]

[0026] Among them, ω + is the peak position, ω l is the uncoupled LO longitudinal phonon frequency, ω p is the plasma frequency, which is related to the carrier concentration n, ω T is the uncoupled LO transverse phonon frequency.

[0027] Furthermore, the plasma frequency is achieved by:

[0028]

[0029] Where n is the carrier concentration, e is the electron charge, ∈ ∞ is the high frequency dielectric constant, m * is the effective mass.

[0030] Furthermore, the wafer quality factor is calculated, including:

[0031] Get the number of wafers that passed the test and the total number of wafers;

[0032] Calculating a yield rate based on the number of wafers that passed the test and the total number of wafers;

[0033] Obtain the total number of wafer surface defects, wafer area, and maximum allowable defect density;

[0034] Calculating defect density based on the total number of defects on the wafer surface and the wafer area;

[0035] The wafer quality factor is calculated according to the yield, the defect density and the maximum allowable defect density.

[0036] In a second aspect, an embodiment of the present invention provides a crystal quality detection system, the system comprising a sample transport stage, a microscopic imager, a Raman spectrometer, and a controller:

[0037] The sample transport stage is used to position the wafer sample below the objective lens so that the focus of the objective lens is at the center of the wafer sample;

[0038] The microscopic imager is used to perform microscopic imaging and obtain an image when the objective lens moves linearly within a radius relative to the wafer sample;

[0039] The Raman spectrometer is used to perform Raman imaging to obtain a three-dimensional hyperspectrum when the intensity fluctuation of the image is greater than a set threshold;

[0040] The controller is used to process each Raman spectrum in the three-dimensional hyperspectrum to obtain a spectrum peak curve, a peak width curve and a peak position curve; when the first-order derivatives of the spectrum peak curve, the peak width curve and the peak position curve after fitting tend to zero, the longitudinal scan ends; the spectrum peak intensity, peak width and peak position generated by the three-dimensional hyperspectrum are converted into characteristic information, and a wafer quality coefficient is calculated; and the quality of the wafer sample is determined according to the wafer quality coefficient.

[0041] Furthermore, the system further comprises a laser, a ring illumination excitation module, a first dichroic mirror and a second dichroic mirror;

[0042] The laser is used to excite the spot size of the laser inside the wafer sample;

[0043] The annular illumination excitation module is used to make the energy of the laser incident on the surface of the wafer sample at a high angle;

[0044] The second dichroic mirror is used to achieve high reflection of white light with a wavelength shorter than the Raman excitation light and high transmission of light with a wavelength longer than the Raman excitation light;

[0045] The first dichroic mirror is used for highly reflecting the excitation light wavelength band and highly transmitting the Raman scattered light wavelength band longer than the excitation light wavelength.

[0046] In a third aspect, an embodiment of the present invention provides an electronic device, including a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and the processor implements the above-mentioned method when executing the computer program.

[0047] In a fourth aspect, an embodiment of the present invention provides a computer-readable medium having a non-volatile program code executable by a processor, wherein the program code enables the processor to execute the method as described above.

[0048] An embodiment of the present invention provides a crystal quality detection method and system, including: positioning a wafer sample under an objective lens through a sample transport stage so that the focus of the objective lens is at the center of the wafer sample; performing microscopic imaging to obtain an image when the objective lens moves linearly within a radius relative to the wafer sample; performing Raman imaging to obtain a three-dimensional hyperspectrum when the intensity fluctuation of the image is greater than a set threshold; processing each Raman spectrum in the three-dimensional hyperspectrum to obtain a spectral peak curve, a peak width curve, and a peak position curve; ending the longitudinal scan when the first-order derivative of the spectral peak curve, the peak width curve, and the peak position curve tends to zero after fitting; converting the spectral peak intensity, peak width, and peak position generated by the three-dimensional hyperspectrum into characteristic information, and calculating a wafer quality coefficient; determining the quality of the wafer sample based on the wafer quality coefficient; being able to measure crystals (including wafer samples) at different depths; and achieving non-destructive, rapid, and accurate quality analysis at room temperature and pressure.

[0049] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or understood by practicing the present invention. The purposes and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0050] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0052] Figure 1 A flow chart of the crystal quality detection method provided in Example 1 of the present invention;

[0053] Figure 2 This is a schematic diagram of a crystal quality detection system provided in Example 2 of the present invention. DETAILED DESCRIPTION

[0054] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0055] To facilitate understanding of this embodiment, the embodiment of the present invention is described in detail below.

[0056] Example 1:

[0057] Figure 1 This is a flow chart of the crystal quality detection method provided in Example 1 of the present invention.

[0058] In order to fully meet the needs of front-line personnel, the detection system is based on an artificial intelligence model. On the one hand, it provides real-time guidance on the instrument operation process to determine that the instrument is always in the best working state. On the other hand, it accelerates the process from instrument data acquisition to crystal quality assessment, reducing the complexity of instrument use. Figure 1 , the method comprises the following steps:

[0059] Step S101, positioning a wafer sample under an objective lens via a sample transport stage so that the focus of the objective lens is at the center of the wafer sample;

[0060] Step S102, when the objective lens moves linearly within a radius relative to the wafer sample, microscopic imaging is performed to obtain an image;

[0061] Step S103, when the intensity fluctuation of the image is greater than a set threshold, Raman imaging is performed to obtain a three-dimensional hyperspectral spectrum;

[0062] Specifically, rapid microscopic imaging scans are performed, and algorithm feedback locates the target inspection area and plans the scanning path. To increase inspection speed, a rotating sample and linear positioning method are used to move the objective lens linearly within a radius relative to the wafer. Rapid microscopic imaging is combined with image intensity fluctuations. When intensity fluctuations exceed a set threshold, Raman imaging of the area is performed to generate a three-dimensional hyperspectral image.

[0063] Step S104, processing each Raman spectrum in the three-dimensional hyperspectral spectrum to obtain a spectrum peak curve, a peak width curve, and a peak position curve;

[0064] Step S105, when the first-order derivative of the peak curve, peak width curve and peak position curve after fitting approaches zero, the longitudinal scan ends;

[0065] Step S106, converting the peak intensity, peak width, and peak position generated by the three-dimensional hyperspectral spectrum into characteristic information, and calculating the wafer quality factor;

[0066] Step S107 , determining the quality of the wafer sample according to the wafer quality coefficient.

[0067] Specifically, the Raman microscope automatically adjusts experimental parameters to optimize the spectral signal-to-noise ratio to 5. It also automatically optimizes the laser power to achieve a crystal Raman scattering spectrum peak signal-to-noise ratio of 5. Using a Lorentz fitting algorithm, it obtains real-time information on the peak intensity, peak width, and peak position. A real-time curve of peak intensity, peak position, and peak width versus Z-axis scanning depth is plotted. The longitudinal scan ends when the first-order derivative approaches zero after the curve fitting. The wafer quality factor includes process-related parameters such as defect density, overall doping concentration, and the horizontal and vertical distribution of defects. Finally, a report is generated to switch samples.

[0068] This example utilizes high-speed scanning, a low-latency system, and online data processing to improve detection speed. It utilizes the correlation of spectral peak position, half-width, and peak intensity with three-dimensional spatial location, along with simultaneous dark-field imaging, to acquire multidimensional data and enhance assessment accuracy. Furthermore, a standardized measurement-analysis process, combined with rapid analysis using a neural network model, enables rapid and accurate assessment of crystal quality. This system addresses the challenges of traditional detection, including slow speed and low accuracy of single-dimensional data, providing efficient guidance for crystal manufacturing and processing.

[0069] Furthermore, step S104 includes the following steps:

[0070] Step S201, inputting each Raman spectrum in the three-dimensional hyperspectral spectrum into a Lorentz fitting algorithm to obtain a three-dimensional graph of peak intensity, a three-dimensional graph of peak width, and a three-dimensional graph of peak position;

[0071] Step S202 , the three-dimensional graph of peak intensity, the three-dimensional graph of peak width and the three-dimensional graph of peak position are respectively scanned with the Z axis to obtain a spectrum peak curve, a peak width curve and a peak position curve.

[0072] Furthermore, the characteristic information includes stress. In step S106, the peak intensity, peak width, and peak position generated by the three-dimensional hyperspectral spectrum are converted into characteristic information, including:

[0073] Calculate the stress according to formula (1):

[0074] Δv(cm -1 )=K×σ(Gpa) (1)

[0075] Where Δv is the Raman peak shift, representing the shift in peak position relative to the stress-free state, K is the stress sensitivity coefficient (which can be determined through calibration experiments), and σ is the stress. Compressive stress causes the peak position to shift toward higher frequencies, while tensile stress causes the opposite.

[0076] Furthermore, the characteristic information includes doping concentration (based on the plasma-phonon coupling model). In step S106, the peak intensity, peak width and peak position generated by the three-dimensional hyperspectral spectrum are converted into characteristic information, including:

[0077] The doping concentration is calculated according to formula (2):

[0078]

[0079] Among them, ω + is the peak position, i.e. the A(LO) phonon peak position observed in the experiment (cm -1 ),ω l is the uncoupled LO longitudinal phonon frequency (e.g. 4H-SiC, ω l ≈964.5cm -1 ),ω p is the plasma frequency, which is related to the carrier concentration n, ω T is the uncoupled LO transverse phonon frequency.

[0080] Furthermore, the plasma frequency is achieved by:

[0081]

[0082] Where n is the carrier concentration (cm -3 ), e is the electron charge (1.6×10 -19 C),∈ ∞ is the high frequency dielectric constant (such as 6.52 for 4H-SiC), m * is the effective mass (for electrons, m * ≈0.3m e ).

[0083] For example, if ω + =970cm -1 , substitute into the formula to infer ω p , and then calculate n.

[0084] In addition, the nitrogen concentration in 4H-SiC increases from 2.1×10 18 to 1.2×10 19 cm -3 The change of will lead to a huge change in the shape (half-maximum width) of the A1(LO) phonon.

[0085] For a doping concentration of 2.1×10 18 cm -3 4H-SiC, phonon damping Γ = 4.43 cm -1 , corresponding to a half-peak width of about 8.86 cm -1 .

[0086] For a doping concentration of 1.2×10 19 cm -3 4H-SiC, phonon damping Γ = 25.1 cm -1 , corresponding to a half-peak width of approximately 50.2 cm -1 .

[0087] Furthermore, in step S106, calculating the wafer quality coefficient includes the following steps:

[0088] Step S301, obtaining the number of wafers that passed the test and the total number of wafers;

[0089] Step S302, calculating the yield rate based on the number of wafers that pass the test and the total number of wafers;

[0090] Step S303, obtaining the total number of wafer surface defects, wafer area, and maximum allowable defect density;

[0091] Step S304, calculating the defect density based on the total number of defects on the wafer surface and the wafer area;

[0092] Step S305 , calculating the wafer quality factor according to the yield, defect density and maximum allowable defect density.

[0093] Here, referring to formulas (4), (5) and (6):

[0094] Quality factor = yield × (1 - Defect density / maximum allowable defect density) (4)

[0095] Yield = Number of wafers that passed the test / Total number of wafers × 100% (5)

[0096] Defect density = total number of defects on wafer surface / wafer area (6)

[0097] Specifically, data collection and definition: Wafer number (N_total): the total number of wafers produced in the same batch or under the same process conditions; the number of wafers that passed the test (N_pass); the total number of wafer surface defects (D_total); wafer area (A): based on standard wafer size (e.g., 8-inch wafer area ≈ 200cm 2 ); maximum allowable defect density (D_max).

[0098] Count the number of wafers that pass the test in a batch and calculate the yield; calculate the defect density; and finally calculate the quality factor.

[0099] Quality judgment and optimization suggestions: quality coefficient threshold setting;

[0100] Analysis and Improvement Measures: If defect density is high or yield is low: Check process stability (such as doping uniformity). Optimize the cleanroom environment or improve the process. Finally, conduct dynamic monitoring.

[0101] The following analysis takes the three-dimensional stress distribution of the wafer surface layer as an example:

[0102] Application objectives: Detect stress distribution inside wafers and evaluate lattice distortion and defects caused by manufacturing processes (such as grinding, polishing, and annealing).

[0103] Sample preparation: Clean the wafer surface to avoid contaminants. Mark the area to be measured (such as edge, center, specific structure area).

[0104] Data acquisition: Use confocal Raman microscopy and select appropriate laser wavelength.

[0105] 1) Set the Z-axis stepping (e.g., 0.1 μm step) and collect Raman spectra layer by layer to cover the target depth (e.g., 10 μm).

[0106] 2) Perform raster scanning in the XY plane (e.g., with a step size of 5 μm) to cover the entire region of interest.

[0107] Data processing: Fit the peak shape and extract the Raman peak shift of each point (e.g. the base of 4H-SiC is 777cm -1 Peak), the stress is calculated according to formula (1) (stress coefficient K is -1.96). A tomographic reconstruction algorithm (such as filtered back projection) is used to integrate the data at each depth to generate a three-dimensional stress distribution map.

[0108] Result analysis:

[0109] Visualize stress gradients and identify high-stress areas (such as interfaces or pattern edges). Compare stress changes before and after processing to optimize process parameters.

[0110] The following analysis takes the wafer and epitaxial layer doping concentration and doping uniformity as an example:

[0111] Application Objectives: Evaluate doping concentration and uniformity after ion implantation and annealing. Doping Concentration Quantification: Determine the concentration distribution of doping elements such as nitrogen (N) and aluminum (Al) in SiC. Uniformity Assessment: Measure the uniformity of doping element distribution in the horizontal (XY plane) and vertical (Z axis) directions.

[0112] Sample preparation:

[0113] 1) Using semi-insulating (undoped) SiC as the A(LO) phonon peak position (e.g. 964.5 cm -1 ) benchmark.

[0114] 2) Prepare gradient samples with known doping concentrations (e.g., by ion implantation and annealing at different doses) to calibrate the relationship between Raman signal and concentration.

[0115] 3) Establish a calibration curve of Raman peak parameters (peak position shift and half-peak width) and doping concentration for the reference standard.

[0116] 4) Clean the surface of the wafer to be tested to avoid interference from contaminants.

[0117] Data collection:

[0118] 1) Calibrate the instrument and select appropriate parameters.

[0119] 2) Collect Raman spectra at multiple depths (e.g., 0-3 μm, every 0.5 μm) combined with lateral scans.

[0120] Data processing:

[0121] 1) Fit peak positions and extract information such as peak position and full width at half maximum (FWHM).

[0122] 2) Construct a concentration calibration curve based on the changes in peak position shift and full width at half maximum (FWHM).

[0123] 3) Integrate the data of each depth layer to generate a three-dimensional distribution map of the doping concentration.

[0124] 4) Calculate the standard deviation (σ) and uniformity index of the horizontal / vertical concentration distribution.

[0125] Result analysis:

[0126] 1) Verify doping uniformity, detect abnormal concentration areas (such as edge accumulation), and optimize annealing processes (such as annealing parameters).

[0127] 2) Detect doping dead zones (such as lattice collapse areas caused by high concentrations) and adjust ion implantation energy or annealing conditions.

[0128] This application addresses the challenges of surface layer quality assessment in wafer manufacturing by proposing a nondestructive testing system based on Raman spectroscopy. High-speed scanning, a low-latency system, and online data processing are employed to improve detection speed. Multidimensional data is acquired by correlating spectral peak position, half-width, and peak intensity with three-dimensional spatial position, along with simultaneous dark-field imaging, to enhance assessment accuracy. Combined with rapid analysis using a neural network model, a standardized measurement-analysis process is established to enable rapid and accurate assessment of crystal quality. This system addresses the issues of slow traditional testing and low accuracy of single-dimensional data, providing efficient guidance for crystal manufacturing and processing.

[0129] Example 2:

[0130] Figure 2 This is a schematic diagram of a crystal quality detection system provided in Example 2 of the present invention.

[0131] Reference Figure 2 The system includes a sample transport stage, a microscopic imager, a Raman spectrometer, and a controller:

[0132] The sample transport stage is used to position the wafer sample under the objective lens so that the focus of the objective lens is at the center of the wafer sample;

[0133] A microscopic imager is used to perform microscopic imaging and obtain an image when the objective lens moves linearly within a radius relative to the wafer sample;

[0134] A Raman spectrometer is used to perform Raman imaging to obtain a three-dimensional hyperspectral spectrum when the intensity fluctuation of the image is greater than a set threshold;

[0135] The controller is used to process each Raman spectrum in the three-dimensional hyperspectrum to obtain a spectrum peak curve, a peak width curve and a peak position curve; when the first-order derivative of the spectrum peak curve, the peak width curve and the peak position curve after fitting tends to zero, the longitudinal scan ends; the spectrum peak intensity, peak width and peak position generated by the three-dimensional hyperspectrum are converted into characteristic information, and the wafer quality coefficient is calculated; the quality of the wafer sample is determined based on the wafer quality coefficient.

[0136] Furthermore, the system also includes a laser, a ring illumination excitation module, a first dichroic mirror, and a second dichroic mirror;

[0137] Laser, used to excite the laser spot size inside the wafer sample;

[0138] A ring illumination excitation module is used to make the laser energy incident at a high angle on the surface of the wafer sample;

[0139] A second dichroic mirror (dichroic mirror 2) is used to achieve high reflection of white light below the wavelength of the Raman excitation light and high transmission of light above the wavelength of the Raman excitation light;

[0140] The first dichroic mirror (dichroic mirror 1) is used for highly reflecting the excitation light band and highly transmitting the Raman scattered light band that is longer than the wavelength of the excitation light.

[0141] To ensure the system's rapid detection capabilities in crystal quality assessment, the system also includes:

[0142] A high numerical aperture laser (objective lens focusing) is used to excite the laser spot size inside the sample, ensuring the energy density of the excitation light and the collection capacity of the Raman scattered light signal, so that the system can shorten the exposure time as much as possible while maintaining a certain signal-to-noise ratio.

[0143] The ring illumination excitation module is used to make the laser energy incident on the sample surface at a high angle, improving the focusing characteristics of the light spot and increasing the energy density.

[0144] The galvanometer scanning technology is used within the sample surface to reduce the movement of mechanical parts and increase the scanning speed of the light spot in the crystal surface layer;

[0145] The Raman spectrometer uses an electron-multiplying charge-coupled device (EMCCD) to reduce the number of horizontal pixels in the EMCCD midline array, enabling it to cover a range of 150 wavenumbers before and after the peak of the crystal Raman characteristic peak, reducing signal redundancy at the data acquisition end and improving spectrum transmission efficiency.

[0146] Two dichroic mirrors are used in the light beam transmission path to avoid mechanical switching of the optical path between microscopic imaging and Raman spectrum collection. At the same time, dichroic mirror 2 is used to achieve high reflection of white light below the wavelength of Raman excitation light and high transmission of light above the wavelength of Raman excitation light, meeting the microscopic imaging capability; dichroic mirror 1 is used to achieve high reflection of the excitation light band and high transmission of the Raman scattered light band above the excitation light wavelength. In conjunction with the back-end Raman filter, it ensures the synchronous Raman spectrum collection capability.

[0147] The data acquisition and processing in the controller uses online accelerated data processing chips such as FPGA or ASIC to improve the online processing speed of the spectrum.

[0148] An embodiment of the present invention further provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the crystal quality detection method provided in the above embodiment are implemented.

[0149] An embodiment of the present invention further provides a computer-readable medium having non-volatile program code executable by a processor. The computer-readable medium stores a computer program. When the computer program is executed by the processor, the steps of the crystal quality detection method of the above embodiment are executed.

[0150] The computer program product provided in the embodiments of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods described in the previous method embodiments. For specific implementation, please refer to the method embodiments and will not be repeated here.

[0151] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems and devices can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0152] In addition, in the description of the embodiments of the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; they may refer to mechanical connections or electrical connections; they may refer to direct connections or indirect connections through an intermediate medium; and they may refer to internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0153] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0154] In the description of the present invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They are not intended to indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation, and therefore should not be construed as limitations on the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0155] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the above-described embodiments, those skilled in the art should understand that any person skilled in the art can modify or easily conceive of changes to the technical solutions described in the above-described embodiments within the technical scope disclosed by the present invention, or replace some of the technical features therein with equivalents. Such modifications, changes, or replacements do not deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A crystal quality detection method, characterized in that: The method comprises: Positioning the wafer sample under the objective lens by using the sample transport stage so that the focus of the objective lens is at the center of the wafer sample; When the objective lens moves linearly within a radius relative to the wafer sample, microscopic imaging is performed to obtain an image; When the intensity fluctuation of the image is greater than a set threshold, Raman imaging is performed to obtain a three-dimensional hyperspectrum; Processing each Raman spectrum in the three-dimensional hyperspectral spectrum to obtain a spectrum peak curve, a peak width curve and a peak position curve; When the first-order derivative of the spectrum peak curve, the peak width curve and the peak position curve after fitting tends to zero, the longitudinal scan ends; Converting the peak intensity, peak width, and peak position generated by the three-dimensional hyperspectral spectrum into characteristic information and calculating the wafer quality factor; The quality of the wafer sample is determined according to the wafer quality coefficient.

2. The crystal quality detection method according to claim 1, characterized in that: Each Raman spectrum in the three-dimensional hyperspectral spectrum is processed to obtain a spectrum peak curve, a peak width curve and a peak position curve, including: Inputting each Raman spectrum in the three-dimensional hyperspectral spectrum into a Lorentz fitting algorithm to obtain a three-dimensional graph of spectral peak intensity, a three-dimensional graph of peak width, and a three-dimensional graph of peak position; The three-dimensional graph of the spectrum peak intensity, the three-dimensional graph of the peak width and the three-dimensional graph of the peak position are respectively scanned with the Z axis to obtain the spectrum peak curve, the peak width curve and the peak position curve.

3. The crystal quality detection method according to claim 1, characterized in that: The characteristic information includes stress, and converting the peak intensity, peak width and peak position generated by the three-dimensional hyperspectral spectrum into characteristic information includes: The stress is calculated according to the following formula: Δv(cm -1 )=K×σ(Gpa) Wherein, Δv is the Raman peak shift, which represents the shift of the peak position relative to the stress-free state, K is the stress sensitivity coefficient, and σ is the stress.

4. The crystal quality detection method according to claim 1, characterized in that: The characteristic information includes doping concentration, and converting the peak intensity, peak width and peak position generated by the three-dimensional hyperspectral spectrum into characteristic information includes: The doping concentration is calculated according to the following formula: Among them, ω + is the peak position, ω l is the uncoupled LO longitudinal phonon frequency, ω p is the plasma frequency, which is related to the carrier concentration n, ω T is the uncoupled LO transverse phonon frequency.

5. The crystal quality detection method according to claim 4, characterized in that: The plasma frequency is achieved by: Where n is the carrier concentration, e is the electron charge, ∈ ∞ is the high frequency dielectric constant, m * is the effective mass.

6. The crystal quality detection method according to claim 1, characterized in that: Calculate wafer quality factor, including: Get the number of wafers that passed the test and the total number of wafers; Calculating a yield rate based on the number of wafers that passed the test and the total number of wafers; Obtain the total number of wafer surface defects, wafer area, and maximum allowable defect density; Calculating defect density based on the total number of defects on the wafer surface and the wafer area; The wafer quality factor is calculated according to the yield, the defect density and the maximum allowable defect density.

7. A crystal quality detection system, characterized in that: The system includes a sample transport stage, a microscopic imager, a Raman spectrometer, and a controller: The sample transport stage is used to position the wafer sample below the objective lens so that the focus of the objective lens is at the center of the wafer sample; The microscopic imager is used to perform microscopic imaging and obtain an image when the objective lens moves linearly within a radius relative to the wafer sample; The Raman spectrometer is used to perform Raman imaging to obtain a three-dimensional hyperspectrum when the intensity fluctuation of the image is greater than a set threshold; The controller is used to process each Raman spectrum in the three-dimensional hyperspectrum to obtain a spectrum peak curve, a peak width curve and a peak position curve; when the first-order derivatives of the spectrum peak curve, the peak width curve and the peak position curve after fitting tend to zero, the longitudinal scan ends; the spectrum peak intensity, peak width and peak position generated by the three-dimensional hyperspectrum are converted into characteristic information, and a wafer quality coefficient is calculated; and the quality of the wafer sample is determined according to the wafer quality coefficient.

8. The crystal quality detection system according to claim 7, characterized in that: The system also includes a laser, a ring illumination excitation module, a first dichroic mirror and a second dichroic mirror; The laser is used to excite the spot size of the laser inside the wafer sample; The annular illumination excitation module is used to make the energy of the laser incident on the surface of the wafer sample at a high angle; The second dichroic mirror is used to achieve high reflection of white light with a wavelength shorter than the Raman excitation light and high transmission of light with a wavelength longer than the Raman excitation light; The first dichroic mirror is used for highly reflecting the excitation light wavelength band and highly transmitting the Raman scattered light wavelength band longer than the excitation light wavelength.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 6 is implemented.

10. A computer-readable medium having a non-volatile program code executable by a processor, characterized in that The program code causes the processor to execute the method according to any one of claims 1 to 6.

Images