Intelligent detection method and device for quartz sand material inclusions
By combining laser confocal microscopy and Raman spectroscopy, the problem of simultaneous analysis of the composition and spatial location of inclusions in quartz sand materials was solved, achieving efficient and comprehensive detection and providing information on the type, content, and spatial distribution of inclusions.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ADVANCED QUARTZ MATERIAL (HANGZHOU) CO LTD
- Filing Date
- 2026-03-04
- Publication Date
- 2026-06-12
AI Technical Summary
Traditional testing methods struggle to synchronize and correlate the compositional analysis and spatial location of inclusions in quartz sand materials, resulting in low evaluation efficiency and incomplete results.
Three-dimensional morphological data were acquired using laser confocal microscopy, and component spectra were collected using Raman spectroscopy. The type, content, and spatial distribution information of inclusions were output through a morphology-component correlation model.
It enables non-destructive, in-situ, and continuous detection of inclusions, ensuring a strict one-to-one correspondence between three-dimensional morphology and chemical composition data in spatial coordinates, thereby improving the comprehensiveness and accuracy of the evaluation results.
Smart Images
Figure CN122193186A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of materials testing technology, specifically relating to an intelligent detection method and device for inclusions in quartz sand materials. Background Technology
[0002] Quartz sand is the core raw material for preparing high-purity quartz crucibles. As a key consumable in the process of pulling semiconductor single-crystal silicon, its purity and internal defects directly determine the quality of the final silicon crystal and the single-crystal formation rate.
[0003] Quartz sand and the quartz crucibles it prepares often contain various inclusions, including bubbles, unmelted quartz grains, metallic impurity particles, and carbides. These inclusions are the main intrinsic defects affecting its performance. The presence of these defects not only weakens the mechanical strength and high-temperature stability of the quartz crucible, but may also cause crucible breakage or contamination of the molten silicon during crystal pulling, leading to a series of serious quality problems such as crystal dislocations and resistivity inhomogeneity, resulting in significant economic losses.
[0004] To accurately assess the grade of quartz sand raw materials, effectively trace the source of defects, scientifically predict product risks, and provide reliable data support for optimizing quartz sand beneficiation processes, precise detection of internal inclusions is necessary, especially obtaining information on the chemical composition and three-dimensional spatial distribution characteristics of the inclusions. However, traditional detection methods often struggle to achieve simultaneous and correlated compositional analysis and spatial location, resulting in low assessment efficiency and incomplete results. Summary of the Invention
[0005] In view of this, the present invention provides an intelligent detection method and device for quartz sand material inclusions, in order to solve the technical problem that traditional detection methods in the prior art are difficult to achieve the synchronization and correlation of compositional analysis and spatial positioning of quartz sand material inclusions, resulting in low evaluation efficiency and incomplete results.
[0006] To achieve the above objectives, this application adopts the following approach:
[0007] A method for intelligent detection of inclusions in quartz sand materials includes the following steps:
[0008] S10. Pre-treat the material to be tested and fix the pre-treated material to be tested in place;
[0009] S20. Obtain the three-dimensional morphological data of the fixed internal inclusions of the material to be tested;
[0010] S30. Based on the three-dimensional morphology data, perform compositional spectral acquisition on the inclusion region corresponding to the three-dimensional morphology data to obtain the compositional characteristic spectrum of the inclusion;
[0011] S40. Input the three-dimensional morphology data and the component spectral data into the "morphology-component" association model of the inclusion, and output the type, content and spatial distribution information of the inclusion through the "morphology-component" association model of the inclusion.
[0012] Preferably, in step S20, laser confocal technology is used to scan the material to be tested in order to obtain three-dimensional morphological data of the inclusions inside the material to be tested.
[0013] Preferably, in step S30, based on the three-dimensional morphology data, Raman spectroscopy is triggered to acquire the compositional spectrum of the inclusion region corresponding to the three-dimensional morphology data, so as to obtain the compositional characteristic spectrum of the inclusion.
[0014] Preferably, in step S20, the three-dimensional morphology data includes the size, spatial coordinates, distribution density, gas-liquid two-phase, pure gas phase, and solid phase of the inclusion.
[0015] Preferably, step S40 includes acquiring historical data and training a "morphology-composition" association model of the inclusion based on the historical data. The historical data includes historical three-dimensional morphology data, historical component feature spectra, and information on the type, content, and spatial distribution of the inclusions corresponding to the historical three-dimensional morphology data and the historical component feature spectra.
[0016] Preferably, the material to be tested is quartz sand or a quartz crucible.
[0017] An apparatus for implementing the above-mentioned intelligent detection method for inclusions in quartz sand materials includes a detection platform, a three-dimensional morphology data acquisition module, a compositional characteristic spectrum acquisition module, a synchronization control module, and an intelligent data processing module. The three-dimensional morphology data acquisition module and the compositional characteristic spectrum acquisition module are both electrically connected to the synchronization control module, and the three-dimensional morphology data acquisition module and the compositional characteristic spectrum acquisition module are both electrically connected to the intelligent data processing module.
[0018] The detection platform is used to fix the material to be detected. The three-dimensional morphology data acquisition module is used to acquire the three-dimensional morphology data of the inclusions inside the material to be detected. The composition feature spectrum acquisition module is used to acquire the composition spectrum of the inclusion region corresponding to the three-dimensional morphology data to obtain the composition feature spectrum of the inclusion. The synchronization control module is used to control the timing synchronization of the three-dimensional morphology data acquisition module and the composition feature spectrum acquisition module. The intelligent data processing module is used to execute intelligent algorithms and output the detection results.
[0019] Preferably, the three-dimensional morphology data acquisition module is a laser confocal unit, and the component characteristic spectrum acquisition module is a Raman spectroscopy unit.
[0020] Preferably, the laser confocal unit includes a laser emitter, a scanning galvanometer, and an imaging block, wherein the laser emitter is used to emit a probe laser; the scanning galvanometer is disposed in the optical path of the probe laser and is used to deflect the probe laser to scan the material to be detected; the imaging block is used to receive the reflected signal from the material to be detected and generate a corresponding three-dimensional morphology data signal to be transmitted to the intelligent data processing module.
[0021] Preferably, the Raman spectroscopy unit includes a Raman excitation source and a spectrometer, wherein the Raman excitation source is used to emit excitation light to irradiate the inclusion region corresponding to the three-dimensional morphology data; the spectrometer is used to collect and analyze the Raman scattered light signal from the inclusion region, generate compositional characteristic spectrum data, and transmit it to the intelligent data processing module.
[0022] The aforementioned intelligent detection method and device for inclusions in quartz sand materials combines two key steps that are traditionally performed separately: microscopic morphology observation (e.g., microscopy) and chemical composition analysis (e.g., spectrometry). Three-dimensional morphology data is not only used for observation but also serves as a "spatial map" guiding compositional analysis; while compositional spectral data imbues morphological characteristics with chemical meaning. The final output includes not only the chemical type of the inclusion but also its three-dimensional spatial distribution and content information, making the final output results quantifiable, statistically significant, and more comprehensive, thus improving evaluation efficiency and the comprehensiveness of the results. Through non-destructive, in-situ, and continuous detection of the same inclusion sample, a strict one-to-one correspondence between the three-dimensional morphology and chemical composition data in spatial coordinates is ensured, fundamentally solving the problems of data misalignment and lack of correlation caused by sample movement and target loss in traditional step-by-step detection. Attached Figure Description
[0023] Figure 1 This is a flowchart of this application.
[0024] Figure 2 This is a functional block diagram of this application.
[0025] The figure shows a detection platform 100, a three-dimensional morphology data acquisition module 200, a laser emitter 210, a scanning galvanometer 220, an imaging block 230, a compositional characteristic spectrum acquisition module 300, a Raman excitation source 310, a spectrometer 320, a synchronization control module 400, and an intelligent data processing module 500. Detailed Implementation
[0026] To facilitate understanding of this application, a more comprehensive description will be provided below with reference to the accompanying drawings. Preferred embodiments of this application are also given. However, this application can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to enable a more thorough and complete understanding of the disclosure of this application.
[0027] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein in the specification of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.
[0028] Please refer to Figure 1 In one specific embodiment, a smart detection method for inclusions in quartz sand material includes the following steps:
[0029] S10. Pre-treat the material to be tested and fix the pre-treated material to be tested in place;
[0030] S20. Obtain the three-dimensional morphological data of the fixed internal inclusions of the material to be tested;
[0031] S30. Based on the three-dimensional morphology data, perform compositional spectral acquisition on the inclusion region corresponding to the three-dimensional morphology data to obtain the compositional characteristic spectrum of the inclusion;
[0032] S40. Input the three-dimensional morphology data and the component spectral data into the "morphology-component" association model of the inclusion, and output the type, content and spatial distribution information of the inclusion through the "morphology-component" association model of the inclusion.
[0033] First, the material to be tested (quartz sand or quartz crucible) undergoes necessary pretreatment, such as cleaning the surface of the quartz crucible to remove dirt and then fixing it in place to provide a stable sample for subsequent testing. Second, three-dimensional morphological data of inclusions within the material to be tested are acquired to achieve stereoscopic imaging of the microstructure. Subsequently, based on the obtained three-dimensional morphological data, targeted chemical composition analysis is performed on the located inclusion regions to obtain the compositional characteristic spectra of the inclusions. The inclusion "morphology-composition" correlation model deeply fuses and analyzes the three-dimensional morphological data (geometric information) and compositional spectral data (chemical information) from the same inclusion. Using this model, the specific type, relative content, and three-dimensional spatial distribution information of the inclusions within the material can be automatically and quickly resolved and output.
[0034] The method provided in this application combines two key steps that are traditionally performed separately: microscopic morphology observation (e.g., microscopy) and chemical composition analysis (e.g., spectrometry). Three-dimensional morphology data is not only used for observation but also serves as a "spatial map" guiding compositional analysis; while compositional spectral data imbues morphological characteristics with chemical meaning. The final output includes not only the chemical type of the inclusions but also their three-dimensional spatial distribution and content information, making the final output results quantifiable, statistically significant, and more comprehensive, thus improving evaluation efficiency and the comprehensiveness of the results. Through non-destructive, in-situ, and continuous detection of the same inclusion sample, a strict one-to-one correspondence between three-dimensional morphology and chemical composition data in spatial coordinates is ensured, fundamentally solving the problems of data misalignment and lack of correlation caused by sample movement and target loss in traditional stepwise detection.
[0035] This upgrades the assessment results from the traditional "what is there" to "how much, where, and how it is distributed," providing unprecedented comprehensive data support for a deeper understanding of the source, formation process, and impact on the properties of quartz sand materials (such as high-temperature stability and chemical purity), and greatly enhancing the scientific and application value of the test results.
[0036] Specifically, in step S20 above, laser confocal technology is used to scan the material to be tested in order to obtain three-dimensional morphological data of the inclusions inside the material to be tested.
[0037] In this embodiment, the laser confocal microscopy technique possesses excellent axial resolution, enabling layer-by-layer optical slicing of inclusions and reconstruction of their precise three-dimensional coordinates and morphology. This provides a foundation for subsequent guided spectral analysis. Furthermore, laser confocal scanning does not cause substantial damage or alteration to the sample, ensuring that after morphological scanning, the same inclusion region can be used intact and unchanged for subsequent component spectral acquisition, guaranteeing the consistency of the data source and the effectiveness of the correlation. Moreover, this technique provides high-contrast, high-resolution microscopic images, and the acquired three-dimensional morphological data, including inclusion size, shape, and surface texture, are of high quality, laying a reliable data foundation for subsequent intelligent algorithms to effectively extract features and build models.
[0038] Specifically, in step S30 above, based on the three-dimensional morphology data, Raman spectroscopy is triggered to acquire the compositional spectrum of the inclusion region corresponding to the three-dimensional morphology data, so as to obtain the compositional characteristic spectrum of the inclusion.
[0039] In this embodiment, the Raman spectroscopy technique can provide unique spectral features based on molecular vibrations, like a chemical "fingerprint," which can accurately identify the phases (such as minerals, gases, melts, etc.) and specific chemical components of inclusions. This makes the component feature spectrum a decisive data source for the chemical qualitative and quantitative interpretation of three-dimensional morphology.
[0040] Specifically, in step S20 above, the three-dimensional morphology data includes the size, spatial coordinates, distribution density, gas-liquid two-phase, pure gas phase, and solid phase of the inclusion.
[0041] Spatial coordinates (such as X, Y, and Z) are the sole key to achieving a one-to-one correspondence and automatic matching between morphological and spectral data. They serve not only as a "navigation map" guiding spectral probes (such as Raman spectroscopy) to perform precise in-situ detection, but also as the logical basis for data alignment and correlation in subsequent fusion models, fundamentally solving the problems of ambiguous spatial positioning and difficulty in data correspondence in traditional methods. Size and distribution density are the foundation for quantitative and statistical evaluation: correlation positioning and component type identification alone are insufficient. The size information of inclusions (such as volume and equivalent diameter) is an important physical quantity affecting material properties; while distribution density (the number or volume percentage of inclusions per unit volume) is a macroscopic statistical indicator for evaluating the overall purity and homogeneity of the material. The introduction of these quantitative data allows the method's output to leap from qualitative description to quantitative analysis, providing a key dimension for comprehensively evaluating material quality.
[0042] Specifically, step S40 includes acquiring historical data and training a "morphology-composition" association model of the inclusion based on the historical data. The historical data includes historical three-dimensional morphology data, historical component feature spectra, and information on the type, content, and spatial distribution of the inclusions corresponding to the historical three-dimensional morphology data and the historical component feature spectra.
[0043] Furthermore, the material to be tested is quartz sand or a quartz crucible.
[0044] Please refer to Figure 2 This application also provides an apparatus for implementing the above-mentioned intelligent detection method for inclusions in quartz sand materials, comprising a detection platform 100, a three-dimensional morphology data acquisition module 200, a compositional characteristic spectrum acquisition module 300, a synchronization control module 400, and an intelligent data processing module 500. The three-dimensional morphology data acquisition module 200 and the compositional characteristic spectrum acquisition module 300 are both electrically connected to the synchronization control module 400, and the three-dimensional morphology data acquisition module 200 and the compositional characteristic spectrum acquisition module 300 are both electrically connected to the intelligent data processing module 500.
[0045] The detection platform 100 is used to fix the material to be detected. The three-dimensional morphology data acquisition module 200 is used to acquire the three-dimensional morphology data of the inclusions inside the material to be detected. The composition feature spectrum acquisition module 300 is used to acquire the composition spectrum of the inclusion region corresponding to the three-dimensional morphology data to obtain the composition feature spectrum of the inclusion. The synchronization control module 400 is used to control the timing synchronization between the three-dimensional morphology data acquisition module 200 and the composition feature spectrum acquisition module 300. The intelligent data processing module 500 is used to execute intelligent algorithms and output detection results.
[0046] In the device provided in this application, the material to be tested, such as quartz sand or a quartz crucible, is fixed on a detection platform 100. Then, a three-dimensional morphology data acquisition module 200 scans the material fixed on the detection platform 100 to obtain three-dimensional morphology data of inclusions within the material. A compositional characteristic spectrum acquisition module 300, based on the coordinates determined by the acquired three-dimensional morphology data, performs compositional spectrum acquisition on the target inclusion region to obtain the compositional characteristic spectrum of the inclusion. A synchronous control module 400 coordinates and controls the working sequence and action logic of the two acquisition modules to ensure precise and orderly coordination between morphology scanning and spectral acquisition in space and time. After acquiring the data, both acquisition modules transmit the raw data (three-dimensional morphology data and compositional spectrum data) to an intelligent data processing module 500 in real time. The intelligent data processing module 500 receives and fuses the heterogeneous data from the two acquisition modules, executes multimodal fusion algorithms such as deep learning, automatically establishes a "morphology-composition" correlation model, and finally outputs comprehensive detection results such as the type, content, and spatial distribution of the inclusions.
[0047] The device provided in this application ensures a strict one-to-one correspondence between three-dimensional morphology and chemical composition data in spatial coordinates through non-destructive, in-situ, and continuous detection of the same inclusion sample. This fundamentally solves the problems of data misalignment and lack of correlation caused by sample movement and target loss in traditional stepwise detection. The final output not only includes the chemical type of the inclusion but also its three-dimensional spatial distribution and content information. This upgrades the evaluation results from the traditional "what is there" to "how much, where, and how it is distributed," providing unprecedented comprehensive data support for a deeper understanding of the origin, formation process, and impact on the properties of quartz sand materials (such as high-temperature stability and chemical purity), greatly enhancing the scientific and application value of the detection results.
[0048] Furthermore, the three-dimensional morphology data acquisition module 200 is a laser confocal unit, and the component characteristic spectrum acquisition module 300 is a Raman spectroscopy unit.
[0049] Furthermore, the laser confocal unit includes a laser emitter 210, a scanning galvanometer 220, and an imaging block 230. The laser emitter 210 emits a probe laser; the scanning galvanometer 220 is disposed in the optical path of the probe laser and is used to deflect the probe laser to scan the material to be detected; the imaging block 230 receives the reflected signal from the material to be detected and generates a corresponding three-dimensional topographic data signal, which is transmitted to the intelligent data processing module 500.
[0050] In this embodiment, the scanning galvanometer 220 realizes programmed and gridded deflection control of the probe laser, thereby performing a systematic scan of the sample, which is a prerequisite for generating accurate spatial coordinate data; the imaging block 230 is used to receive the reflected signal from the material to be detected and generate the corresponding three-dimensional morphology data signal, ensuring that the generated three-dimensional morphology data has high resolution and high contrast.
[0051] Furthermore, the Raman spectroscopy unit includes a Raman excitation source 310 and a spectrometer 320. The Raman excitation source 310 is used to emit excitation light to irradiate the inclusion region corresponding to the three-dimensional morphology data. The spectrometer 320 is used to collect and analyze the Raman scattered light signal from the inclusion region, generate compositional characteristic spectrum data, and transmit it to the intelligent data processing module 500.
[0052] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and such modifications or substitutions should all be covered within the scope of protection of the present invention.
Claims
1. A method for intelligent detection of inclusions in quartz sand materials, characterized in that, Includes the following steps: S10. Pre-treat the material to be tested and fix the pre-treated material to be tested in place; S20. Obtain the three-dimensional morphological data of the fixed internal inclusions of the material to be tested; S30. Based on the three-dimensional morphology data, perform compositional spectral acquisition on the inclusion region corresponding to the three-dimensional morphology data to obtain the compositional characteristic spectrum of the inclusion; S40. Input the three-dimensional morphology data and the component spectral data into the "morphology-component" association model of the inclusions, and output the type, content and spatial distribution information of the inclusions through the "morphology-component" association model of the inclusions.
2. The intelligent detection method for inclusions in quartz sand materials according to claim 1, characterized in that, In step S20, laser confocal technology is used to scan the material to be tested in order to obtain three-dimensional morphological data of the inclusions inside the material to be tested.
3. The intelligent detection method for inclusions in quartz sand materials according to claim 1, characterized in that, In step S30, based on the three-dimensional morphology data, Raman spectroscopy is triggered to acquire the compositional spectrum of the inclusion region corresponding to the three-dimensional morphology data, so as to obtain the compositional characteristic spectrum of the inclusion.
4. The intelligent detection method for inclusions in quartz sand materials according to claim 1, characterized in that, In step S20, the three-dimensional morphology data includes the size, spatial coordinates, distribution density, gas-liquid two-phase, pure gas phase, and solid phase of the inclusion.
5. The intelligent detection method for inclusions in quartz sand materials according to claim 1, characterized in that, Step S40 includes acquiring historical data and training a "morphology-composition" association model of the inclusions based on the historical data. The historical data includes historical three-dimensional morphology data, historical component feature spectra, and information on the type, content, and spatial distribution of the inclusions corresponding to the historical three-dimensional morphology data and the historical component feature spectra.
6. The intelligent detection method for inclusions in quartz sand materials according to claim 5, characterized in that, The material to be tested is quartz sand or a quartz crucible.
7. An apparatus for implementing the intelligent detection method for inclusions in quartz sand material according to any one of claims 1 to 6, characterized in that, It includes a detection platform, a three-dimensional morphology data acquisition module, a component feature spectrum acquisition module, a synchronization control module, and an intelligent data processing module. The three-dimensional morphology data acquisition module and the component feature spectrum acquisition module are both electrically connected to the synchronization control module, and the three-dimensional morphology data acquisition module and the component feature spectrum acquisition module are both electrically connected to the intelligent data processing module. The detection platform is used to fix the material to be detected. The three-dimensional morphology data acquisition module is used to acquire the three-dimensional morphology data of the inclusions inside the material to be detected. The composition feature spectrum acquisition module is used to acquire the composition spectrum of the inclusion region corresponding to the three-dimensional morphology data to obtain the composition feature spectrum of the inclusion. The synchronization control module is used to control the timing synchronization of the three-dimensional morphology data acquisition module and the composition feature spectrum acquisition module. The intelligent data processing module is used to execute intelligent algorithms and output the detection results.
8. The intelligent detection device for inclusions in quartz sand materials according to claim 7, characterized in that, The three-dimensional morphology data acquisition module is a laser confocal unit, and the component characteristic spectrum acquisition module is a Raman spectroscopy unit.
9. The intelligent detection device for inclusions in quartz sand materials according to claim 8, characterized in that, The laser confocal unit includes a laser emitter, a scanning galvanometer, and an imaging block. The laser emitter emits a probe laser. The scanning galvanometer is positioned in the optical path of the probe laser to deflect the probe laser and scan the material to be tested. The imaging block receives reflected signals from the material to be tested and generates corresponding three-dimensional topographic data signals, which are then transmitted to the intelligent data processing module.
10. The intelligent detection device for inclusions in quartz sand materials according to claim 8, characterized in that, The Raman spectroscopy unit includes a Raman excitation source and a spectrometer. The Raman excitation source emits excitation light to irradiate the inclusion region corresponding to the three-dimensional morphology data. The spectrometer collects and analyzes the Raman scattered light signal from the inclusion region, generates compositional characteristic spectrum data, and transmits it to the intelligent data processing module.