Crystal internal defect detection method based on dispersion line confocal backscattering

Abstract

The application discloses a crystal internal defect detection method based on dispersion line scanning confocal backscattering. A point light source emits visible light which is focused, filtered through a small hole and collimated to enter a beam splitter prism. The beam splitter prism reflects the collimated visible light beam to a line array scanning device. The line array scanning device scans the incident collimated visible light beam into a periodic fan-shaped visible light beam. A telecentric field lens makes the effective visible light beam parallel and focused. Then, a double-telecentric dispersion objective lens focuses light of different wavelengths at different depths of a measured sample. The internal defects of the sample scatter light of corresponding depth wavelengths. The scattered light is sequentially incident on a dispersion element through the double-telecentric dispersion objective lens, the telecentric field lens, the line array scanning device and the beam splitter prism. Finally, a spectral line is formed on the plane of a detection device. The depth of the defect is determined by extracting the peak value of the spectral line. The application realizes the rapid detection of the internal defects of the crystal by using the dispersion line scanning confocal backscattering technology, and has high detection efficiency and detection sensitivity.

Description

Technical Field

[0001] This invention belongs to the field of optical detection technology, and more specifically relates to a method for detecting internal defects in crystals based on dispersive linear scanning confocal backscattering technology. Background Technology

[0002] Inertial confinement fusion (ICF) is a crucial direction in humanity's search for clean energy. A core technology in ICF is the laser driver. The United States was one of the first countries to develop this technology, and its latest generation driver—the National Ignition Facility (NIF)—is currently the most advanced and highest-energy large-scale laser driver. It contains over 7,000 large-aperture optical elements and over 20,000 small-aperture optical elements. If these optical elements have defects, when exposed to strong laser radiation, the incident laser will encounter these defects and scatter, causing beam deflection. This ultimately leads to a decrease in the quality and energy of the output beam, and in severe cases, damage to the optical elements themselves and downstream optical components, significantly reducing the load capacity of the optical elements. To address the impact of defects in optical elements on the entire optical system, defect screening must be performed during the manufacturing process to determine the size and location of defects within the optical crystal. This avoids using defective optical crystals in the fabrication of optical elements, which is a necessary measure to improve the quality of the optical system.

[0003] Existing methods for detecting crystal defects include digital holographic imaging, parallel confocal microscopy, and laser scattering tomography, each with its own advantages and disadvantages.

[0004] Digital holographic imaging utilizes the interference and diffraction of light to acquire and reproduce a three-dimensional image of an object. An object illuminated by a laser produces scattered light beams, a portion of which is recorded as a hologram by a CCD (Computer-Controlled Optical Disc Recognition) sensor. The hologram is then stored in a computer, which simulates the optical diffraction process to achieve holographic reconstruction and processing of the recorded object. Finally, the internal defects of the optical crystal are detected by observing the hologram. While digital holographic imaging boasts high accuracy, its accuracy is highly dependent on the holographic inversion algorithm. The large amount of interference data redundancy and the extremely time-consuming calculations result in extremely slow detection speeds in industrial production, failing to meet the online inspection requirements of crystal manufacturing.

[0005] Parallel confocal microscopy uses a beam-splitting element to split the light source into an array of point sources, allowing each beam to be imaged onto the object being inspected, forming a spot. Each spot, in turn, is imaged onto the detector, creating a point-to-point correspondence that enables the acquisition and recording of two-dimensional information about the object. After recording each two-dimensional surface, three-dimensional reconstruction of the object can be achieved, thus enabling the detection of internal defects in optical crystals. However, the biggest problem with parallel confocal microscopy is the contradiction between the numerical aperture (NA) and the field of view (FOV). A large numerical aperture limits the linear field of view to the millimeter level, making rapid detection of internal crystal defects impossible. Therefore, this technology cannot meet the rapid inspection requirements of crystal production.

[0006] Laser scattering tomography (LST) uses a Powell lens to uniformly project a collimated laser beam at a specific angle, allowing the fan-shaped laser beam to penetrate the two-dimensional plane of the object being tested. When the laser beam encounters a defect, it scatters. A CCD collects the scattered light from each layer, enabling the detection of internal defects in the object. While LST is very fast, it lacks confocal focusing capabilities, and CCD imaging suffers from depth-of-field limitations. This leads to severe crosstalk from scattered light, high background noise, and a low signal-to-noise ratio, making it difficult to achieve efficient detection of internal crystal defects. It is prone to errors, resulting in a lower yield rate in crystal production.

[0007] Therefore, there is an urgent need for a method that can detect the size and location of defects inside crystals both quickly and accurately, so as to provide effective protection for the fabrication of optical components. Summary of the Invention

[0008] To overcome the shortcomings of the prior art, this invention provides a crystal internal defect detection method based on dispersive linear scanning confocal backscattering, which can detect the size and location of crystal internal defects both quickly and accurately, thus providing effective protection for the fabrication of optical components.

[0009] To achieve its objectives, the present invention employs the following technical solution:

[0010] The characteristics of this invention's crystal internal defect detection method based on dispersive linear scanning confocal backscattering are:

[0011] Visible light emitted from a point source is focused by the first lens group, filtered by a pinhole, and then collimated by the second lens group to obtain a collimated visible beam A0.

[0012] The visible light beam A0 is reflected by a beam splitter to a linear array scanning device. The rotating linear array scanning device scans the visible light beam into a periodic fan-shaped visible light beam A1. A portion of the single-period fan-shaped visible light beam is used to scan the trigger element, and the remaining effective visible light beam A2 is incident on the telecentric field lens. The telecentric field lens parallels and focuses the effective visible light beam A2 onto the back focal plane of the telecentric field lens. Then, through the double telecentric dispersive objective lens, light of different wavelengths in the visible light is focused into a spot at the corresponding depth along the axis of the sample being tested, forming a cylindrical light column with a diameter of 2ω, which is the cylindrical region being tested, where ω is the radius of a scanning spot.

[0013] For a defect located at depth z of the sample being tested, the defect causes the wavelength λ at the corresponding depth to decrease. z The scanning light forms defect backscattered light B, which passes sequentially through a dual telecentric dispersive objective and a telecentric field lens, and is then reflected by the linear array scanning device to a beam splitter. The beam splitter then passes through a dispersive element and a focusing lens, focusing the light onto the calibrated detection device to form a spectrum. The wavelength λ is determined from this spectrum using a peak extraction algorithm. z This enables defect localization.

[0014] The method for detecting internal defects in crystals in this invention is characterized by: setting the back focal plane of the telecentric field lens to overlap with the front focal plane of the double telecentric dispersive objective lens to form an overlapping focal plane; so that the incident light is focused on the back focal planes at different depths of the double telecentric dispersive objective lens, and covers the entire axial depth of the sample being tested.

[0015] The method for detecting internal defects in crystals in this invention is also characterized by: fixing the sample under test on a displacement stage, and using the displacement stage to move the sample under test in the x-direction at a velocity v. x It moves to scan and detect the sample on a three-dimensional surface.

[0016] The method for detecting internal defects in crystals in this invention is also characterized by:

[0017] The velocity v x Calculated from equation (1):

[0018]

[0019] In formula (1):

[0020] ΔT is the single-sided scanning period of the linear array scanning device, ΔT = 1 / (8f ps );

[0021] f ps This is the rotation frequency of the linear scanning device.

[0022] The method for detecting internal defects in crystals in this invention is also characterized by: setting a three-dimensional scanning strategy as follows: each cylindrical optical column is arranged along the Y-axis to form a cylindrical optical column array, and each cylindrical optical column array is arranged along the X-axis to achieve three-dimensional detection of the sample under test; setting the center distance between adjacent cylindrical optical columns to be no greater than Achieve three-dimensional full coverage of the sample under test by a cylindrical light beam.

[0023] The crystal internal defect detection method of the present invention is also characterized in that: the point light source is a white LED, a supercontinuum laser diode (SLD), a tunable laser source, or a halogen source; the linear array scanning device is a polyhedral prism, an acousto-optic modulator, or a digital micromirror scanner; and the row triggering element is a photodiode.

[0024] The method for detecting internal defects in crystals in this invention is also characterized by:

[0025] The peak extraction algorithm first fits the spectrum acquired by the detection device into equation (2):

[0026]

[0027] In formula (2):

[0028] f(λ;A,u,σ,γ) is the light intensity at wavelength λ;

[0029] A is the amplitude, and μ is the wavelength at the peak position;

[0030] σ is the standard deviation, γ is the tilt of the sample being tested, and λ is the wavelength;

[0031] The wavelength λ is calculated using equation (2). z .

[0032] The characteristic of the crystal internal defect detection method of this invention is that: the axial position calibration refers to the relationship between the depth z of the sample under test and the wavelength λ obtained by extracting the spectral peak. z During the matching process, the axial position calibration is performed as follows:

[0033] A reflector is placed at different axial depths z' of the scanning light column. The detection device detects the reflected light at different axial depths z' multiple times and records the spectrum of each measurement. The wavelength μ at the peak of each spectrum is obtained through a peak extraction algorithm. n The wavelength λ is obtained by taking its average value. z ;

[0034] In actual measurements, the scanning light undergoes refraction during its incidence on the sample, which differs from the wavelength λ. z The corresponding calibration depth z is calculated using equation (3):

[0035] z = z' + Δz (3)

[0036] In formula (3):

[0037] Δz is the difference between the depth of the reflector due to refraction and the actual depth reached by the scanning light within the sample being measured. This difference Δz is calculated using equation (4):

[0038]

[0039] In equation (4):

[0040] n is the refractive index of the sample being tested;

[0041] 2θ is the convergence angle of the scanning beam when it is incident on the sample being tested.

[0042] The characteristic of the crystal internal defect detection method of this invention also lies in: setting the detection process as follows:

[0043] First, the sample to be tested is positioned and fixed in the set position; then the displacement stage is moved at the set speed v. x Movement; when a row trigger signal is detected, the detection device begins sampling at a set sampling rate until the next row trigger signal is detected;

[0044] The spectrum of the defect backscattered light obtained by the detection device is processed, and the processing result is matched point by point with the sample under test to complete the detection of one row of axial depth. The same process is repeated for each row of axial depth until the entire three-dimensional defect detection is completed, forming a defect point cloud map of the sample under test.

[0045] Compared with existing technologies, the beneficial effects of this invention are reflected in:

[0046] This invention addresses the problem of low detection efficiency in existing detection technologies by employing a dispersive linear scanning confocal backscattering detection technique. This technique focuses light of different wavelengths in the visible spectrum at various depths along the axis of the sample being tested. The linear scanning device can scan in the y-direction, and the sample only needs to be moved in one direction to achieve a three-dimensional scan. This greatly reduces the time and steps required for detection and provides a technical basis for the high-efficiency detection of crystal defects. Attached Figure Description

[0047] Figure 1 This is a schematic diagram of the optical path structure of the detection device used in this invention;

[0048] Figure 2 This is a schematic diagram of the scanning principle in the detection device used in this invention;

[0049] Figure 3 This is a schematic diagram of the axial scanning method inside the crystal according to the present invention;

[0050] Figure 4 This is a schematic diagram of the three-dimensional scanning strategy of the present invention;

[0051] Figure 5 This is a schematic diagram of the axial position calibration of the present invention;

[0052] Figure 6 This is a flowchart of the detection method of the present invention.

[0053] The following are the labels in the diagram: 1. Point light source, 2. First lens group, 3. Pinhole, 4. Second lens group, 5. Beam splitter, 6. Linear array scanning device, 7. Telecentric field lens, 8. Double telecentric dispersive objective lens, 9. Sample under test, 10. Line trigger element, 11. Dispersive element, 12. Focusing lens, 13. Detection device, 14. Overlapping focal plane, A0. Visible beam, A1. Periodic fan-shaped visible beam, A2. Effective visible beam, B. Defect backscattered light. Detailed Implementation

[0054] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0055] This embodiment uses a crystal internal defect detection method based on dispersive linear scanning confocal backscattering:

[0056] like Figure 1 and Figure 2 As shown, visible light emitted from point light source 1 is focused by first lens group 2, filtered by pinhole 3, and then collimated by second lens group 4 to obtain collimated visible light beam A0.

[0057] The visible beam A0 is reflected by the beam splitter prism 5 to the linear array scanning device 6. The rotating linear array scanning device 6 scans the visible beam A0 into a periodic fan-shaped visible beam A1. A portion of the single-period fan-shaped visible beam A1 is used to scan the trigger element 10, and the remaining effective visible beam A2 is incident on the telecentric field mirror 7. The telecentric field mirror 7 parallels and focuses the effective visible beam A2 onto the back focal plane of the telecentric field mirror 7. Then, through the double telecentric dispersive objective lens 8, the light of different wavelengths in the visible light is focused into a spot at the corresponding depth along the axis of the sample 9 being measured, forming a cylindrical light column with a diameter of 2ω, which is the cylindrical region being measured, where ω is the radius of a scanning spot.

[0058] The defect at depth z of the tested sample 9 causes the wavelength λ at the corresponding depth to decrease. zThe scanning light forms defect backscattered light B, which passes sequentially through the dual telecentric dispersive objective lens 8 and the telecentric field lens 7, and is then reflected by the linear array scanning device 6 to the beam splitter prism 5. The beam splitter prism 5 then passes sequentially through the dispersive element 11 and the focusing lens 12, focusing the light onto the calibrated detection device 13 to form a spectrum. The wavelength λ is determined based on the spectrum using a peak extraction algorithm. z This enables defect localization.

[0059] In practice, the corresponding technical measures also include:

[0060] The rear focal plane of the telecentric field lens 7 is set to overlap with the front focal plane of the dual telecentric dispersive objective lens 8 to form an overlapping focal plane; so that the incident light is focused on the rear focal planes at different depths of the dual telecentric dispersive objective lens 8 and covers the entire axial depth of the sample 9 being tested. Figure 2 As shown, the effective visible light beam A2 entering the telecentric field lens 7 is focused on the rear focal plane of the telecentric field lens 7. Since the front focal plane of the dual telecentric dispersive objective lens 8 overlaps with the rear focal plane of the telecentric field lens 7, the focused visible light is re-intruded into the dual telecentric dispersive objective lens 8 through the overlapping focal plane 14. The visible light beam is a continuous wavelength of light, and different wavelengths of light are focused by the dual telecentric dispersive objective lens 8 at different depths of the sample 9 being measured.

[0061] The sample 9 to be tested is fixedly mounted on the displacement stage, and the displacement stage is used to move the sample 9 in the x direction at a velocity v. x It moves to scan and detect the sample 9 on a three-dimensional surface.

[0062] speed v x Calculated from equation (1):

[0063]

[0064] In formula (1):

[0065] ΔT is the single-sided scanning period of the linear array scanning device 6, ΔT = 1 / (8f ps );

[0066] f ps This is the rotation frequency of the linear scanning device 6.

[0067] The three-dimensional scanning strategy is set as follows: each cylindrical light column is arranged into a cylindrical light column array along the Y direction, and each cylindrical light column array is arranged along the X direction to realize the three-dimensional detection of the sample 9 under test.

[0068] The center distance between adjacent cylindrical light columns is set to be no greater than [value missing]. Achieve three-dimensional full coverage of the sample 9 by the cylindrical light beam.

[0069] In a specific implementation, the point light source 1 can be a white LED, a supercontinuous laser diode (SLD), a tunable laser source, or a halogen source; the linear array scanning device 6 can be a polyhedral prism, an acousto-optic modulator, or a digital micromirror scanner; and the row trigger element 10 is a photodiode.

[0070] The peak extraction algorithm first fits the spectrum acquired by the detection device 13 into equation (2):

[0071]

[0072] In formula (2):

[0073] f(λ;A,u,v,γ) is the light intensity at wavelength λ;

[0074] A is the amplitude, and μ is the wavelength at the peak position;

[0075] σ is the standard deviation, γ is the tilt of the sample 9 being tested, and λ is the wavelength;

[0076] The wavelength λ is calculated using equation (2). z .

[0077] Figure 5 This embodiment illustrates the axial position calibration, which takes into account the refraction of the scanning light during its incidence on the sample in actual measurements. Axial position calibration refers to the calibration of the depth z of the sample 9 being measured and the wavelength λ obtained after extracting the spectral peak. z The matching process; axial position calibration is performed as follows:

[0078] The reflector is placed at each axial depth z' of the scanning light column. The detection device 13 detects the reflected light at different axial depths z' multiple times and records the spectrum of each time. The wavelength μ at the peak of each spectrum is obtained by the peak extraction algorithm. n The wavelength λ is obtained by taking its average value. z ;

[0079] In actual measurements, the scanning light undergoes refraction during its incidence on the sample 9, which differs from the wavelength λ. z The corresponding calibration depth z is calculated using equation (3):

[0080] z = z' + Δz (3)

[0081] In formula (3):

[0082] Δz is the difference between the depth of the reflector due to refraction and the actual depth reached by the scanning light in the sample (9) under test. The difference Δz is calculated by equation (4):

[0083]

[0084] In equation (4):

[0085] n is the refractive index of the sample 9 being tested;

[0086] 2θ is the convergence angle of the scanning beam when it is incident on the sample 9.

[0087] Figure 6 The detection method flow of this embodiment is shown, and the detection process is set as follows:

[0088] First, the sample 9 to be tested is positioned and fixed in the set position; then the displacement stage is moved at the set speed v. x Movement; When the row trigger element 10 detects the row trigger signal, the detection device 13 starts sampling according to the set sampling rate, and ends the sampling of the current cycle when the next row trigger signal is detected, and starts the sampling of the next cycle.

[0089] The spectrum of the defect backscattered light obtained by the detection device 13 is processed, and the processing result is matched point by point with the sample 9 to complete the detection of one row of axial depth. The same process is followed to detect the axial depth of each row until the entire three-dimensional defect detection is completed, forming the defect point cloud map of the sample 9.

[0090] Figure 3 This embodiment illustrates the axial scanning method inside the crystal: The linear array scanning device 6 is used so that only one cylindrical region with a diameter of 2ω is scanned at any given moment in the sample being tested. The cylindrical region has different depths λ. z Each wavelength corresponds to a different wavelength of a visible light source.

[0091] Figure 4 This embodiment illustrates the 3D scanning strategy, showing two columns of light beams for illustrative purposes: To ensure the focused light beam array completely covers all positions of the sample 9 under test, the distance between adjacent focused light beams must be less than [a certain value]. Where ω is the radius of each focused beam; each cylinder is the actual focused beam array during synchronous scanning, and obviously the array is tilted. The corresponding design is as follows: the axes of the first and last beams of any two linear focused beam arrays in any adjacent scanning period ΔT are on the same plane, thus obtaining the moving speed v of the sample 9 in the X-axis direction driven by the moving device. x , is obtained by calculation from equation (1).

[0092] In this invention, unless explicitly specified and limited, features are intertwined and do not necessarily exist independently. The above description includes the basic principles, main features, and advantages of this invention. Those skilled in the art should understand that this invention is not limited to the above embodiments; the embodiments and description are merely preferred examples and not intended to limit the invention or be the only option. Within the spirit and scope of the invention, it can be further modified and optimized. All improvements and optimizations to this invention fall within the scope of the claims, which are defined by the appended claims and their equivalents.

Claims

1. A crystal internal defect detection method based on dispersive linear scanning confocal backscattering, characterized by: Visible light emitted from a point light source (1) is focused by the first lens group (2), filtered by the pinhole (3), and then collimated by the second lens group (4) to obtain a collimated visible light beam A0. The visible beam A0 is reflected by the beam splitter prism (5) to the linear array scanning device (6). The rotating linear array scanning device (6) scans the visible beam into a periodic fan-shaped visible beam A1. A portion of the single-period fan-shaped visible beam is used to scan the trigger element (10), and the remaining effective visible beam A2 is incident on the telecentric field mirror (7). The telecentric field mirror (7) parallels and focuses the effective visible beam A2 on the back focal plane of the telecentric field mirror (7). Then, through the double telecentric dispersive objective (8), the light of different wavelengths in the visible light is focused into a spot at the corresponding depth of the axial direction of the sample (9) to form a cylindrical light column with a diameter of 2ω, which is the cylindrical region to be measured. ω is the radius of a scanning spot. The defect at depth z of the sample (9) being tested causes the wavelength λ at the corresponding depth to be reduced. z The scanning light forms defect backscattered light B, which passes sequentially through a double telecentric dispersive objective (8) and a telecentric field lens (7), and is then reflected by a linear array scanning device (6) to a beam splitter (5). The beam splitter (5) then passes sequentially through a dispersive element (11) and a focusing lens (12) and is focused onto a calibrated detection device (13) to form a spectrum. The wavelength λ is determined based on this spectrum using a peak extraction algorithm. z This enables defect localization.

2. The method for detecting internal defects in a crystal according to claim 1, characterized in that: The back focal plane of the telecentric field lens (7) is set to overlap with the front focal plane of the double telecentric dispersive objective (8) to form an overlapping focal plane; so that the incident light is focused on the back focal planes of different depths of the double telecentric dispersive objective (8) and covers the entire axial depth of the sample (9) being tested.

3. The method for detecting internal defects in a crystal according to claim 2, characterized in that: The sample to be tested (9) is fixedly mounted on the displacement stage, and the displacement stage is used to move the sample to be tested (9) in the x direction at a velocity v. x The sample (9) is moved to scan and detect the sample on a three-dimensional surface.

4. The method for detecting internal defects in a crystal according to claim 3, characterized in that: The velocity v x Calculated from equation (1): In formula (1): ΔT is the single-sided scanning period of the linear array scanning device (6), ΔT=1 / (8f ps ); f ps is the rotation frequency of the linear scanning device (6).

5. The method for detecting internal defects in a crystal according to claim 3, characterized in that: The three-dimensional scanning strategy is set as follows: each cylindrical light column is arranged in a cylindrical light column array along the Y direction, and each cylindrical light column array is arranged along the X direction to realize the three-dimensional detection of the sample (9) to be tested. The center distance between adjacent cylindrical light columns is set to be no greater than [value missing]. Achieve three-dimensional full coverage of the sample (9) by the cylindrical light beam.

6. The method for detecting internal defects in a crystal according to claim 1, characterized in that: The point light source (1) is a white LED, a supercontinuum laser diode (SLD), a tunable laser source, or a halogen source. The linear array scanning device (6) is a polyhedral prism, an acousto-optic modulator, or a digital micromirror scanner; The row triggering element (10) is a photodiode.

7. The method for detecting internal defects in a crystal according to claim 1, characterized in that: The peak extraction algorithm first fits the spectrum acquired by the detection device (13) into equation (2): In formula (2): f(λ;A,u,σ,γ) is the light intensity at wavelength λ; A is the amplitude, and μ is the wavelength at the peak position; σ is the standard deviation, γ is the tilt of the sample (9) being tested, and λ is the wavelength; The wavelength λ is calculated using equation (2). z .

8. The method for detecting internal defects in a crystal according to claim 1, characterized in that: The axial position calibration refers to the relationship between the depth z of the sample (9) being measured and the wavelength λ obtained after extracting the spectral peak. z During the matching process, the axial position calibration is performed as follows: The reflector is placed at different axial depths z' of the scanning light column. The detection device (13) detects the reflected light at different axial depths z' multiple times and records the spectrum of each time. The wavelength μ at the peak of each spectrum is obtained by the peak extraction algorithm. n The wavelength λ is obtained by taking its average value. z ; In actual measurements, the scanning light is refracted during its incidence on the sample (9), and its wavelength λ is different from that of the light source. z The corresponding calibration depth z is calculated using equation (3): z = z' + Δz (3) In formula (3): Δz is the difference between the depth of the reflector due to refraction and the actual depth reached by the scanning light within the sample (9) being measured. This difference Δz is calculated using equation (4): In equation (4): n is the refractive index of the sample (9) being tested; 2θ is the convergence angle of the scanning beam when it is incident on the sample (9).

9. The method for detecting internal defects in a crystal according to claim 3, characterized in that: The detection process is set as follows: First, the sample (9) to be tested is positioned so that it is fixed in the set position; then the displacement stage is moved at the set speed v. x Movement; when a row trigger signal is detected, the detection device (13) starts sampling at the set sampling rate until the next row trigger signal is detected; The spectrum of the defect backscattered light obtained by the detection device (13) is processed, and the processing result is matched point by point with the sample (9) to complete the detection of one row of axial depth; the same process is followed to detect each row of axial depth until the entire three-dimensional defect detection is completed, forming the defect point cloud map of the sample (9).

Images