Method for assisting in cutting infrared crystal by utilizing three-dimensional X-ray microscopic imaging technology

Abstract

The invention discloses a method for assisting in cutting an infrared crystal by utilizing a three-dimensional X-ray microscopic imaging technology, which comprises the following steps of: S1, detecting and distinguishing different types of defects in the infrared crystal by adopting a dual-energy CT (Computed Tomography) scanning technology; s2, aiming at a low atomic number material in the infrared crystal, detecting the defect of small density difference in the infrared crystal; s3, reconstructing three-dimensional orientation distribution of the whole infrared crystal through CT scanning of the infrared crystal according to a crystallographic orientation algorithm, completing three-dimensional modeling of the infrared crystal, then superposing the detected defect distribution with a crystallographic orientation graph in the model, and analyzing relevance between the defect distribution and crystallographic orientation; and S4, the diamond loop wire cutting machine cuts the defect-free infrared crystal away from the defect area. The method has the advantages that the defect recognition precision is high, the variety is complete, the cutting path is optimized, and the crystal utilization rate is high.

Description

Technical Field

[0001] This invention belongs to the field of infrared crystal cutting technology, specifically relating to a method for cutting infrared crystals using three-dimensional X-ray microscopy technology. Background Technology

[0002] Traditional infrared crystal quality assessment mainly adopts the following methods: (1) Manual visual inspection combined with directional etching: Operators observe the surface features of the crystal through an optical microscope and use chemical etching (such as hot phosphoric acid etching of yttrium iron garnet crystal) to expose defects such as dislocations. This method is highly subjective and can only detect surface or near-surface defects. It is powerless to detect internal defects, resulting in a blind cutting process. Moreover, the sample cannot be used after inspection and can only be sampled for inspection. (2) Infrared transmission imaging technology: InGaAs camera is used to image in the 900-1700nm band. The quality of the material is indirectly judged by analyzing the infrared transmission characteristics. Although this method has a certain penetration ability, high resolution cannot penetrate the crystal, so the resolution is limited (usually 10μm-100μm) and it cannot accurately distinguish the defect type and three-dimensional distribution. Summary of the Invention

[0003] This invention aims to provide a method for cutting infrared crystals using three-dimensional X-ray microscopy to accurately identify defects, improve crystal utilization, and solve the problems of current manual visual inspection combined with directional etching, which can only detect surface or near-surface defects and is a destructive inspection, as well as the limited resolution of infrared transmission imaging technology, which makes it impossible to accurately distinguish defect types and three-dimensional distribution.

[0004] Therefore, the technical solution adopted in this invention is as follows: a method for cutting infrared crystals using three-dimensional X-ray microscopy, comprising the following steps:

[0005] Step S1: Using dual-energy CT scanning technology, the sample is scanned with X-rays of two different energies to obtain the attenuation coefficient distribution under different energies. By utilizing the differences in attenuation characteristics of materials under different energy X-rays due to differences in composition and density, different types of defects in infrared crystals can be detected and distinguished.

[0006] Step S2: For low atomic number materials in infrared crystals, phase contrast imaging technology is used to enhance the image contrast of the object by detecting the phase change of X-rays after passing through the object, thereby detecting defects with small density differences in infrared crystals.

[0007] Step S3: Perform overall orientation for each crystal plane on the outer surface of the infrared crystal, and then perform micro-area orientation for the micron-level region to determine the orientation of the outer surface of the infrared crystal in both macroscopic and microscopic dimensions. Scan the infrared crystal with CT and reconstruct the three-dimensional orientation distribution of the entire infrared crystal based on the above crystallographic orientation results to complete the three-dimensional modeling of the infrared crystal. Then, overlay the detected defect distribution with the crystallographic orientation map in the model to analyze the correlation between defect distribution and crystallographic orientation.

[0008] Step S4: After determining the crystal plane orientation and defect location, the diamond ring cutter cuts out a defect-free infrared crystal, avoiding the defect area.

[0009] As a preferred embodiment of the above scheme, in step S1, X-rays with two energies, 80kV and 140kV, are used for scanning. The sample is one of the infrared crystals such as yttrium iron garnet (YIG), zinc germanium phosphorus (ZnGeP2), zinc selenide (ZnSe), silver gallium sulfide (AgGaS2), silver gallium selenide (AgGaSe2), gallium selenide (GaSe), zinc telluride (ZnTe), indium antimonide (InSb), and germanium single crystal (Ge). The infrared crystal has different types of defects, namely volume defects and surface defects, such as flux inclusions and microcracks. Flux inclusions are regions of non-uniform composition generated during crystal growth. The chemical composition of flux inclusions can even be identified, and the scanning identification accuracy is high.

[0010] Further preferably, in step S2, the detected defect is a line defect, such as a cluster of tiny dislocations, which has high identification accuracy.

[0011] More preferably, in step S3, each crystal plane is oriented as a whole using an X-ray diffractometer, and the micro-area is oriented at the micrometer level using an X-ray micro-diffractometer. The three-dimensional modeling of the infrared crystal adopts a 3D U-Net architecture. The 3D U-Net architecture integrates all tomographic scan data to achieve three-dimensional reconstruction and defect localization. By analyzing the 3D image data obtained from X-ray scanning, processing image slices, and then performing three-dimensional reconstruction, 3D U-Net can achieve high-precision defect segmentation.

[0012] More preferably, in step S4, the multi-objective optimization cutting planning algorithm based on the genetic algorithm considers maximizing the volume of the defect-free crystal, maximizing the area that meets the required crystal orientation, minimizing the cutting difficulty and processing time, and meeting multiple wafer size requirements. It outputs the Pareto optimal solution set, allowing operators to select the most suitable cutting scheme according to actual needs, providing the optimal cutting scheme, ensuring the utilization rate of the infrared crystal, and effectively avoiding the removal of the excellent area of ​​the crystal.

[0013] The beneficial effects of this invention are:

[0014] (1) Compared with the current manual visual inspection combined with directional etching and the limited resolution of infrared transmission imaging technology, this application uses X-ray microscopy (micro-CT) imaging technology to the field of infrared crystal cutting. It utilizes the difference in X-ray absorption coefficient of materials with different densities to visualize the internal structure, thereby breaking through the physical limitation that the human eye and traditional optical means cannot see through infrared crystals, realizing non-destructive, high-resolution three-dimensional imaging of defects inside and outside infrared crystals, and identifying defects with high accuracy and variety.

[0015] (2) Establish a three-dimensional model relating defect distribution and crystallographic orientation to achieve precise spatial positioning of defects, optimize cutting paths, avoid defect areas to the greatest extent, improve crystal utilization, replace the traditional blind cutting method that relies on experience, and form a standardized and replicable precision processing flow.

[0016] In summary, this invention has the advantages of high accuracy in identifying defects, comprehensive range of defects, optimized cutting path, and high crystal utilization. Attached Figure Description

[0017] Figure 1 A schematic diagram showing the defects of flux inclusions within crystals;

[0018] (a) Flux inclusion defects within crystals Figure 1 ;

[0019] (b) Flux inclusion defects within crystals Figure 2 ;

[0020] (c) Complete and defect-free crystal diagram.

[0021] Figure 2 This is a schematic diagram showing the distribution and size of two-dimensional internal defects in a yttrium iron garnet crystal.

[0022] Figure 3 This is an X-ray microscopy scan of a yttrium iron garnet crystal.

[0023] Figure 4 This is a three-dimensional distribution diagram of internal defects in a yttrium iron garnet crystal. Detailed Implementation

[0024] The present invention will be further described below with reference to the embodiments and accompanying drawings:

[0025] Combination Figure 1 — Figure 4 As shown, a method for cutting infrared crystals using three-dimensional X-ray microscopy is described, with the following specific steps:

[0026] Step S1: Use dual-energy X-ray microscopy (DCT) to scan the sample with two different energies of X-rays to obtain the attenuation coefficient distribution at different energies. Utilize the differences in attenuation characteristics of the material under different energy X-rays due to differences in composition and density to detect and distinguish different types of defects in the infrared crystal.

[0027] In step S1, X-rays with energies of 80kV and 140kV were used for scanning. The sample was selected from one of the following infrared crystals: yttrium iron garnet (YIG), zinc germanium phosphorus (ZnGeP2), zinc selenide (ZnSe), silver gallium sulfide (AgGaS2), silver gallium selenide (AgGaSe2), gallium selenide (GaSe), zinc telluride (ZnTe), indium antimonide (InSb), and germanium single crystal (Ge). Different types of defects in infrared crystals are volume defects and surface defects, such as flux inclusions and microcracks. A comparison image of flux inclusion defects in yttrium iron garnet crystals is shown below. Figure 1 As shown.

[0028] Step S2: For low atomic number materials in infrared crystals, phase contrast imaging technology is used to enhance the image contrast of the object by detecting the phase change of X-rays after passing through the object, thereby detecting defects with small density differences in infrared crystals.

[0029] In step S2, the detected defect is a line defect, such as a cluster of tiny dislocations.

[0030] Step S3: Perform overall orientation for each crystal plane on the outer surface of the infrared crystal, and then perform micro-area orientation for the micron-level region to determine the orientation of the outer surface of the infrared crystal in both macroscopic and microscopic dimensions. Scan the infrared crystal with CT and reconstruct the three-dimensional orientation distribution of the entire infrared crystal based on the above crystallographic orientation results to complete the three-dimensional modeling of the infrared crystal. Then, overlay the detected defect distribution with the crystallographic orientation map in the model to analyze the correlation between defect distribution and crystallographic orientation.

[0031] In step S3, each crystal plane is oriented as a whole using an X-ray diffractometer, and the micro-area is oriented at the micrometer level using an X-ray micro-diffractometer. The three-dimensional modeling of the infrared crystal adopts a 3D U-Net architecture, which integrates all tomographic scan data to achieve three-dimensional reconstruction and defect localization.

[0032] Step S4: After determining the crystal plane orientation and defect location, the diamond ring cutter cuts out a defect-free infrared crystal, avoiding the defect area.

[0033] In step S4, the multi-objective optimization cutting planning algorithm based on genetic algorithm considers maximizing the volume of defect-free crystals, maximizing the area that meets the required crystal orientation, minimizing the cutting difficulty and processing time, and meeting multiple wafer size requirements. It outputs the Pareto optimal solution set, which allows operators to select the most suitable cutting scheme according to actual needs.

Claims

1. A method of cutting an infrared crystal assisted by three-dimensional X-ray microscopy, characterized by, The method comprises the following steps: Step S1, X-ray microscopic dual-energy CT scanning is adopted, the sample is scanned by two kinds of energy X-rays respectively, the attenuation coefficient distribution under different energies is obtained, and different types of defects in the infrared crystal are detected and distinguished by using the attenuation characteristic difference of the material under different energy X-rays due to the component difference and density difference; Step S2, for the low atomic number material in the infrared crystal, the phase contrast imaging technology is adopted to enhance the image contrast of the object imaging by detecting the phase change of the X-ray after passing through the object, so as to detect the defects with small density difference in the infrared crystal; Step S3, the overall orientation is carried out on each crystal surface on the outer surface of the infrared crystal, and then the micro-area orientation is carried out on the micron-level area, so as to determine the orientation of the outer surface of the infrared crystal in macro and micro, the three-dimensional orientation distribution of the entire infrared crystal is reconstructed by CT scanning the infrared crystal and according to the above crystallographic orientation results, the three-dimensional modeling of the infrared crystal is completed, then the detected defect distribution is superimposed with the crystallographic orientation map in the model, and the correlation between the defect distribution and the crystallographic orientation is analyzed; Step S4, after the crystal surface direction and the defect position are determined, the diamond ring line cutting machine avoids the defect area to cut out the defect-free infrared crystal.

2. The method for cutting infrared crystal assisted by three-dimensional X-ray microscopic imaging technology according to claim 1, characterized in that: In the step S1, the three-dimensional X-ray microscope is adopted to scan by 80kV and 140kV two kinds of energy X-rays, and the different types of defects in the infrared crystal are bulk defects and surface defects.

3. The method for cutting infrared crystals using three-dimensional X-ray microscopy as described in claim 1, characterized in that: In the step S2, the detected defect is a line defect.

4. The method for cutting infrared crystals using three-dimensional X-ray microscopy as described in claim 1, characterized in that: In the step S3, the overall orientation is carried out on each crystal surface by the X-ray diffractometer, the micro-area orientation is carried out on the micron-level area by the X-ray micro-diffraction instrument, the three-dimensional modeling of the infrared crystal adopts the 3D U-Net architecture, and the 3D U-Net architecture realizes three-dimensional reconstruction and defect positioning by integrating all tomographic data.

5. The method for cutting infrared crystals using three-dimensional X-ray microscopy as described in claim 1, characterized in that: In the step S4, the multi-objective optimization cutting planning algorithm based on genetic algorithm considers the maximization of the volume of defect-free crystal, the maximization of the area meeting the required crystal direction, the minimization of cutting difficulty and processing time, and the satisfaction of multiple wafer size requirements, and outputs a Pareto optimal solution set for the operator to select the most suitable cutting scheme according to the actual demand.

Images