A wafer detection signal variable sampling collection method based on a synchronous control of an acousto-optic modulator and a wafer defect detection system

Abstract

This invention relates to a wafer inspection signal variable sampling acquisition method and a wafer defect detection system based on synchronous control of an acousto-optic modulator (AOM). The method includes: planning the full-wafer scan sampling rate based on wafer defect detection system parameters; configuring scan motion parameters; generating an actual sampling rate signal based on the scan motion parameters and the full-wafer scan sampling rate, and sending it to a signal acquisition card; the signal acquisition card receiving the actual sampling rate signal, synchronously controlling the AOM and photodetector to acquire the photoelectric signal corresponding to the scattered light from the wafer surface, obtaining raw sampling data; the signal acquisition card interpolating the raw sampling data to ensure that the sampling point density is the same under the same rotational scan stroke; and performing defect algorithm analysis on the interpolated image data to identify and detect defects on the wafer surface. This invention resolves the contradiction between incident light power and wafer surface damage threshold, achieving synchronous adjustment between laser light power and sampling rate.

Description

Technical Field

[0001] This invention relates to the field of semiconductor wafer defect detection technology, and to a wafer defect detection system and a wafer inspection signal variable sampling acquisition method based on synchronous control of an acousto-optic modulator (AOM). In particular, it relates to a wafer inspection signal variable sampling acquisition method and wafer defect detection system adapted to laser rotating point scanning mode and based on synchronous control of an acousto-optic modulator. Background Technology

[0002] In semiconductor manufacturing, wafer defect detection is a crucial step in ensuring chip yield. Defect detection on patternless wafers primarily relies on laser scattering rotational point scanning imaging technology. This technology uses a focused Gaussian beam incident on the wafer surface, collecting the scattering signal from defects. This signal is then received by a photodetector, enabling defect identification and location, offering advantages such as high sensitivity and high yield. Its defect detection sensitivity is closely related to the incident light wavelength and power density. As the defect size targeted in wafer inspection decreases, the incident laser wavelength has gradually transitioned from the ultraviolet band to the deep ultraviolet band, and the incident light power has also increased. However, high-power laser incident light can lead to potential wafer damage, especially for wafers with surface coatings. The optical power threshold of the coating is much lower than that of the wafer substrate, and prolonged exposure can easily cause coating ablation and other problems.

[0003] Furthermore, the motion mode of laser rotating point scanning has inherent drawbacks. When the laser scans the inner ring of the wafer, the radius is small, the scanning linear velocity is low, and the laser irradiation time per unit area is much longer than that on the outer ring. Conversely, the outer ring has a high scanning speed and a short laser irradiation time per unit area. This leads to a contradiction between motion characteristics and laser power parameters. If a high-power laser is used for uniform rotational speed scanning across the entire area, the short irradiation time on the outer ring will not damage the wafer and will meet the detection sensitivity requirements. However, when stepping into the inner ring for scanning, the film layer may be burned. If the incident light power parameters are limited, the detection sensitivity may not meet the requirements. At the same time, under the condition of low linear velocity on the inner ring, a fixed sampling method will cause signal oversampling, resulting in a large amount of redundant sampled data. Summary of the Invention

[0004] To address the issues of inner layer damage, high sensitivity, and safe detection caused by fixed laser power and sampling rate in existing laser rotating point scanning wafer defect detection technologies, this invention provides a wafer inspection signal variable sampling acquisition method and wafer defect detection system based on synchronous control of an acousto-optic modulator. Specifically, it provides a wafer inspection signal variable sampling acquisition method and wafer defect detection system adapted to laser rotating point scanning mode and based on synchronous control of an acousto-optic modulator, thereby resolving the contradiction between incident light power and wafer surface damage threshold in wafer inspection systems, achieving synchronous adjustment between laser light power and sampling rate, and ensuring high sensitivity and safe detection requirements for the entire wafer.

[0005] This invention provides a method for variable sampling acquisition of wafer inspection signals based on synchronous control of an acousto-optic modulator, comprising the following steps: Step 101: Based on the parameters of the wafer defect detection system, plan the full wafer scan sampling rate; Step 102: Configure scanning motion parameters to determine the scanning range and scanning step of the wafer to be inspected, and clarify the radial range and rotation speed of this scan, so as to provide a motion reference for subsequent dynamic matching of sampling rate according to radius interval; Step 103: Generate an actual sampling rate signal based on the scanning motion parameters and the full-wafer scanning sampling rate, and send it to the signal acquisition card; Step 104: The signal acquisition card receives the actual sampling rate signal, generates the AOM control timing sequence, synchronously controls the acousto-optic modulator AOM and the photodetector, and acquires the photoelectric signal corresponding to the scattered light on the wafer surface to obtain the original sampling data. Step 105: The signal acquisition card performs interpolation processing on the original sampling data to ensure that the sampling point density is the same for the wafer under the same rotational scanning stroke; and Step 106: Perform defect algorithm analysis on the interpolated image data to identify and detect defects on the wafer surface.

[0006] Further, in step 101, the parameters of the wafer defect detection system include the laser incident power P0 on the wafer surface, the focused spot diameter d0, the wafer rotation speed RPM, and the damage radius R. D Threshold radius R T And the outer diameter R0.

[0007] Furthermore, the parameters of the wafer defect detection system also include laser source parameters, acousto-optic modulator performance parameters, photodetector parameters, scanning platform basic parameters, and wafer film layer damage threshold parameters.

[0008] Furthermore, in step 101, planning the full-wafer scan sampling rate includes: Plan the radial scan pixel size Pr and set the threshold radius R. T R T = N×Pr, where N is a power of 2 and satisfies R T >R D ; Delineate wafer scan partition boundaries: radius less than threshold radius R T The time is the wafer variable sampling rate scanning area, and the radius is greater than the threshold radius R. T This is the wafer's fixed sampling rate scanning area; The AOM laser pixel switching time t0 is planned to satisfy t0 < T, where T is the incident light spot at the damage radius R. DEquivalent exposure time at the location; The planned circumferential sampling rate ft is the number of pixels collected in one circumference at the threshold radius position, satisfying 2×pi×R. T / ft < d0; Plan the circumferential sampling rate fo of the outermost diameter R0 of the wafer, satisfying 2×pi×R0 / fo<d0, where pi is π and fo≥ft; For wafer regions according to R T / 2、R T / 4、R T / 8、…The radius range is divided into intervals by binary division at each level, corresponding to a progressive halving of the circumferential sampling rate; and… Based on the sampling period within the radius range and combined with the fixed pixel switching time t0, the corresponding pixel switching time is calculated to form a complete modulation timing sequence for each radius range.

[0009] Furthermore, the corresponding pixel's light-off time is calculated using the following formula: Pixel off time = sampling period - pixel on time t0.

[0010] Further, in step 102, the scanning motion parameters include the inner scanning diameter Rin, the outer scanning diameter Rout, and the radial scanning pixel size Pr.

[0011] Furthermore, step 102 also includes: Determine the number of acquisition cycles corresponding to the radius range interval. Number of acquisition cycles = radial width of the radius range interval / radial scan pixel size Pr.

[0012] Further, in step 103, generating the actual sampling rate signal based on the scanning motion parameters and the full-wafer scanning sampling rate includes: comparing the scanning inner diameter Rin and the scanning outer diameter Rout, determining the radius range between them, and then matching the preset circumferential sampling rate and the corresponding number of acquisition cycles within the corresponding radius range to complete the planning and generation of the variable sampling rate control signal; and / or The actual sampling rate signal includes the circumferential sampling rate corresponding to the radius interval and the number of sampling cycles corresponding to the radius range interval.

[0013] Further, in step 104, the signal acquisition card generates an AOM control timing sequence consisting of pixel on-time and pixel off-time based on the actual sampling rate signal; synchronously controlling the acousto-optic modulator (AOM) and photodetector, the AOM is used to rapidly (on the order of hundreds of nanoseconds) modulate the output beam of the laser source, based on the incident light spot within the damage radius R. D The equivalent irradiation time at the point determines the effective irradiation time of the wafer film layer; Among these measures, the effective irradiation time of the wafer film layer is controlled when the incident light spot is within the damage radius R. DWithin the equivalent irradiation time at the location, the incident light spot is within the damage radius R D The equivalent exposure time at the location is: T=60×d0 / (2×pi×R D ×RPM).

[0014] Furthermore, the switching response speed of the AOM is directly proportional to the diameter of its internal beam and inversely proportional to the speed of sound. The AOM is a quartz crystal AOM.

[0015] Further, in step 105, the original sampling data includes sampling data of the wafer fixed sampling rate scanning area and sampling data of the wafer variable sampling rate scanning area; in the interpolation processing of the original sampling data by the signal acquisition card, the sampling data of the wafer variable sampling rate scanning area is supplemented to make the circumferential sampling number of the wafer variable sampling rate scanning area equal to the circumferential sampling rate of the wafer fixed sampling rate scanning area.

[0016] The present invention also provides a wafer defect detection system, comprising: A laser source is configured to output laser light to the surface of the wafer to be inspected in order to generate scattered light; AOM is configured to rapidly switch the laser source on and off according to instructions, thereby shortening the inner ring irradiation time and avoiding film damage. The AOM control module is configured to receive instructions from the signal acquisition card and drive the AOM to switch the laser on and off. A photodetector is configured to receive the scattered light and convert the optical signal into an electrical signal for transmission to a signal acquisition card. The signal acquisition card is configured to receive instructions from the host computer, synchronously control the AOM and photodetector, convert electrical signals into digital data and perform interpolation processing. The scanning platform is configured to rotate (θ axis) and move radially (R axis) the wafer to be inspected in order to complete a full-area scan; A motion control unit, configured to control the motion of the scanning platform; and The host computer is the main control console.

[0017] The present invention has at least the following beneficial effects: 1) The present invention resolves the contradiction between incident light power and wafer surface damage threshold, and achieves synchronous adjustment between laser light power and sampling rate; 2) To solve the problem of damage in the wafer variable sampling rate scanning area, the present invention actively reduces the sampling rate of the inner circle and shortens the laser irradiation time. At the same time, it interpolates the sparse sampling data of the inner circle to make up a unified data density, so that the scattering imaging coordinates of the whole wafer are regularized, eliminating image distortion and data differences caused by partitioned sampling. In addition, after the data density is consistent, the same set of recognition logic and judgment threshold is used for the wafer variable sampling rate scanning area and the outer circle small defects. The high sensitivity advantage of the outer circle is retained, and the inner circle can also achieve the same recognition accuracy, solving the secondary problem of missed detection of small defects caused by sparse inner circle data in the traditional solution. Attached Figure Description

[0018] To further illustrate the above and other advantages and features of the various embodiments of the present invention, a more specific description of the embodiments of the invention will be presented with reference to the accompanying drawings. It is to be understood that these drawings depict only typical embodiments of the invention and are therefore not intended to limit its scope. In the drawings, identical or corresponding parts will be indicated by identical or similar reference numerals for clarity.

[0019] Figure 1 The flowchart of a wafer detection signal variable sampling acquisition method based on synchronous control of an acousto-optic modulator is shown in some embodiments of the present invention; Figure 2 Schematic diagrams of wafer defect detection systems in some embodiments of the present invention are shown; Figure 3 A schematic diagram of scanning system parameters in some embodiments of the present invention is shown; Figure 4 A schematic diagram of the modulation principle of the acousto-optic modulator in some embodiments of the present invention is shown; Figure 5 A schematic diagram of the synchronization control signal for the acousto-optic modulator and photodetector in some embodiments of the present invention is shown; Figure label: 2-Optical cavity, 201-Laser source, 202-AOM, 203-AOM control module, 204-Photodetector, 205-Signal acquisition card, 206-Scanning platform, 207-Motion control unit, 208-Host computer. Detailed Implementation

[0020] It should be noted that the components in the accompanying drawings may be shown exaggerated for illustrative purposes and may not be to scale.

[0021] In this invention, the various embodiments are merely intended to illustrate the solutions of the invention and should not be construed as limiting.

[0022] In this invention, unless otherwise specified, the quantifiers “a” and “one” do not exclude scenarios involving multiple elements.

[0023] It should also be noted that, in the embodiments of the present invention, only a portion of the parts or components may be shown for clarity and simplicity. However, those skilled in the art will understand that, under the teachings of the present invention, the required parts or components can be added as needed for specific scenarios.

[0024] It should also be noted that within the scope of this invention, the terms "same", "equal", and "equal to" do not mean that the two values ​​are absolutely equal, but allow for a certain reasonable error. In other words, the terms also cover "substantially the same", "substantially equal", and "substantially equal to".

[0025] It should also be noted that in the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. The term "host computer" refers to the main controller / computer that controls the entire detection system.

[0026] Furthermore, the embodiments of the present invention describe the process steps in a specific order. However, this is only for the convenience of distinguishing each step, and is not a limitation on the order of each step. In different embodiments of the present invention, the order of each step can be adjusted according to the process.

[0027] In the following embodiments, AOM can be RD1010-200-24-025-CA of Pulse Optoelectronics.

[0028] The following embodiment provides a method for variable sampling acquisition of wafer inspection signals based on synchronous control of an acousto-optic modulator. Figure 1 A flowchart of a wafer inspection signal variable sampling acquisition method based on acousto-optic modulator synchronous control is shown. It includes the following steps: Step 101: Based on the parameters of the wafer defect detection system, plan the full wafer scan sampling rate; Step 102: The host computer 208 configures the scanning motion parameters (scanning inner diameter Rin, scanning outer diameter Rout, and radial scanning pixel size Pr) to determine the scanning range and scanning step of the wafer to be inspected, and to clarify the radial range and rotation speed of this scan, so as to provide a motion reference for subsequent dynamic matching of sampling rate according to radius interval; Step 103: Generate the actual sampling rate signal based on the scanning motion parameters and the full wafer scanning sampling rate, and send it to the signal acquisition card 205; Step 104: Signal acquisition card 205 receives the actual sampling rate signal, generates AOM control timing, and synchronously controls the acousto-optic modulator AOM 202 and photodetector 204 to acquire the photoelectric signal corresponding to the scattered light on the wafer surface, obtaining the original sampling data (sampling data of the wafer fixed sampling rate scanning area and sampling data of the wafer variable sampling rate scanning area); AOM 202 is used to rapidly switch the output beam of laser source 201 on and off, based on the incident light spot within the damage radius R. D The equivalent irradiation time at the point determines the effective irradiation time of the wafer film layer; Step 105: The signal acquisition card 205 interpolates the original sampled data, padding the data in the wafer variable sampling rate scanning area to make the circumferential sampling number of the wafer variable sampling rate scanning area equal to the circumferential sampling rate of the wafer fixed sampling rate scanning area, so that the sampling point density is the same under the same rotation scanning stroke of the wafer; and Step 106: The host computer 208 performs defect algorithm analysis on the interpolated image data to identify and detect defects on the wafer surface.

[0029] In step 101, the parameters of the wafer defect detection system ( Figure 3 This includes the laser incident power P0 on the wafer surface, the focused spot diameter d0, the wafer rotation speed RPM, and the damage radius R. D Threshold radius R T And the outer diameter R0. It also includes parameters for the laser source 201, the acousto-optic modulator performance parameters, the photodetector 204 parameters, the scanning platform 206 basic parameters, and the wafer film damage threshold parameters. The wafer defect detection system scans samples with Cr-coated films, with the following configuration parameters: laser incident power P0 on the wafer surface is 80mW, focused spot diameter d0 is 5 micrometers, wafer rotation speed RPM is 1200 r / min, and damage radius R... D 7.5mm, threshold radius R T And the outer diameter R0.

[0030] Incident light spot at damage radius R D The equivalent exposure time at the location is: T=60×d0 / (2×pi×R D (×RPM) = 5.3μs Incident light spot at R D The equivalent irradiation energy density at the location is: U=4×P0×T / (pi×d0 2 ) The energy density U is on the same order of magnitude as the damage threshold of the metal film, which means that within the damage radius R... DAt a laser irradiation time of T=5.3μs, the energy density has reached the safety limit of the film layer, confirming that the traditional method will inevitably cause surface damage.

[0031] To avoid surface damage, it is necessary to reduce the energy density as much as possible, such as by reducing the laser power and increasing the incident spot area. However, it is noted that the sample irradiation time increases rapidly with the decrease of the inner diameter. The spot dwell time at the center position is 60 / RPM, ~0.05s. That is, the equivalent dwell time of the spot at the center position under the condition of 1200 r / min RPM on the turntable is 0.05s, which is the damage radius R. D The equivalent irradiation time T at that location is 9400 times longer, which is an inherent defect of the rotating point scanning motion mode.

[0032] To solve this problem, the power of the laser source 201 is rapidly switched and modulated using AOM 202, so that the equivalent irradiation time of the incident light spot during the full-domain scanning process is less than T. At the same time, the photodetector 204 needs to be controlled synchronously to collect effective photoelectric signals during the laser irradiation time.

[0033] A schematic diagram illustrating the principle of rapid switching modulation of laser source 201 power by AOM 202 is shown below. Figure 4 As shown. The laser beam travels at a Bragg angle (θ). Bragg When incident light is emitted into an AOM 202 crystal, the acousto-optic Bragg diffraction condition is met. Driven by radio frequency (RF), the crystal generates a periodic refractive index grating, efficiently diffracting the incident light to output 0th-order (0th) and 1st-order (1st) diffracted light. After diffraction, the 0th and 1st-order diffracted lights form a separation angle (θSep), completely separating them spatially. The 1st-order diffracted light is selected as the system's incident light path. When the RF signal is on, most of the light energy is coupled to the 1st-order diffracted light, typically achieving a diffraction efficiency greater than 85%. When the RF signal is off, almost all light exits as the 0th-order diffracted light, thus realizing an optical switch function.

[0034] The switching response speed (rise / fall time) of AOM 202 is directly proportional to its internal beam diameter and inversely proportional to the speed of sound. To achieve fast switching modulation, AOM 202 is a quartz crystal AOM 202, matched with a millimeter-scale incident beam spot, and the rise and fall time can be as low as hundreds of nanoseconds to meet the irradiation time requirements.

[0035] In step 101, planning the full-wafer scan sampling rate includes: Plan the radial scan pixel size Pr and set the threshold radius R. T R T = N×Pr, where N is a power of 2 and satisfies R T >R D ; Delineate wafer scan partition boundaries: radius less than threshold radius R TThe time is the wafer variable sampling rate scanning area, and the radius is greater than the threshold radius R. T This is the wafer's fixed sampling rate scanning area; Plan the switching time t0 of the AOM 202 laser pixel to satisfy t0 < T, where T is the incident light spot at the damage radius R. D Equivalent exposure time at the location; The planned circumferential sampling rate ft is the number of pixels collected in one circumference at the threshold radius position, satisfying 2×pi×R. T / ft < d0; Plan the circumferential sampling rate fo of the outermost diameter R0 of the wafer, satisfying 2×pi×R0 / fo<d0, where pi is π and fo≥ft; For wafer regions according to R T / 2、R T / 4、R T / 8、…The radius range is divided into intervals by binary division at each level, corresponding to a progressive halving of the circumferential sampling rate; and… Based on the sampling period of each radius range and the fixed pixel switching time t0, the corresponding pixel switching time is calculated to form a complete modulation timing sequence for each radius range. The corresponding pixel switching time is calculated according to the following formula: Pixel switching time = Sampling period - Pixel switching time t0. The full wafer scan sampling rate is shown in Table 1 (DC (Direct Current) in Table 1 refers to DC drive mode).

[0036] Table 1 Full Wafer Scan Sampling Rate Planning In step 103, generating the actual sampling rate signal based on the scanning motion parameters and the full-wafer scanning sampling rate includes: comparing the scanning inner diameter Rin and the scanning outer diameter Rout, looking up the data in Table 1, determining the radius range between them, and then matching the preset circumferential sampling rate and the corresponding number of acquisition cycles within the corresponding radius range to complete the planning and generation of the variable sampling rate control signal. For example, when the scanning inner diameter Rin is within Rout... T / 16 to R T / 8 interval, scan outer diameter Rout is in R T / 2 to R T The data acquisition plan for the specified intervals is shown in Table 2. The "Wafer Radius - Sampling Rate - Number of Acquisition Rounds" mapping table is distributed and stored in the storage unit of the lower-level computer board. The actual sampling rate signal includes the circumferential sampling rate of the corresponding radius interval and the number of acquisition rounds corresponding to the radius range interval. The number of acquisition rounds = radial width of the radius range interval / radial scan pixel size Pr.

[0037] Table 2. Sampling Rate and Number of Collection Cycles Planning Examples Figure 5 A schematic diagram of the synchronous control signals for the acousto-optic modulator and photodetector 204 is shown. The starting position for each sampling revolution is determined based on the turntable position trigger signal. The turntable is a scanning platform 206 carrying the wafer to be inspected. The turntable encoder signal pulse time Te is used to generate the current row sampling clock, corresponding to the circumferential sampling rate. The RF drive signal is consistent with the sampling clock. The AOM 202 optical signal modulation has a time delay Td. Td is related to the incident position Ld of the light spot on the crystal and the speed of sound v, where Ld = Td × v. The photodetector 204 acquisition trigger signal has a time delay Ti. Ti is related to the switching time of the AOM 202 and the Ti parameter needs to be configured to ensure that data acquisition is within the incident light illumination time range. The signal acquisition card 205 performs interpolation processing on the sampled data. For cases where the inner and outer diameter sampling rates are different, the lower-level computer uses an interpolation algorithm to interpolate the inner-circle sampling rate data to maintain consistency with the outer-circle sampling rate data, facilitating image processing and algorithm analysis by the upper-level computer 208.

[0038] This embodiment also provides a wafer defect detection system. Figure 2 A schematic diagram of a wafer defect detection system is shown, including: Optical cavity 2, which serves as both a laser emitter and a signal receiver, is provided with: Laser source 201 is configured to output laser light to the surface of the wafer to be inspected in order to generate scattered light; AOM 202 is configured to rapidly switch the laser source 201 on and off according to instructions, thereby shortening the inner ring irradiation time and avoiding film damage; and Photodetector 204 is configured to receive scattered light and convert the optical signal into an electrical signal for transmission to signal acquisition card 205; AOM control module 203 is configured to receive instructions from signal acquisition card 205 and drive AOM 202 to turn the laser on and off; The signal acquisition card 205 is configured to receive instructions from the host computer 208, synchronously control the AOM 202 and the photodetector 204, convert electrical signals into digital data and perform interpolation processing. The scanning platform 206 is configured to output encoder position signals to the signal acquisition card 205, thereby driving the wafer under test to rotate (θ axis) and move radially (R axis) to complete the full-area scanning. Motion control unit 207, configured to control the movement of scanning platform 206; and The host computer 208 is the main control console.

[0039] The wafer defect detection system fixes the wafer to be inspected on the scanning platform 206 and configures basic detection parameters through the host computer 208. The motion control unit 207 controls the movement of the scanning platform 206, including the R-axis scanning range, R-axis movement speed, and θ-axis rotation speed. The output light from the laser source 201 is modulated by the AOM 202 and finally focused onto the wafer surface after optical path transformation. The light scattering signal generated by the incident light on the wafer surface is received by the photodetector 204 and output to the signal acquisition card 205. The signal acquisition card 205 simultaneously controls the signal acquisition timing of the photodetector 204 and the trigger control signal of the AOM control module 203 to achieve synchronous adjustment between laser power and sampling rate.

[0040] While some embodiments of the present invention have been described in this application, those skilled in the art will understand that these embodiments are merely illustrative. Numerous variations, alternatives, and improvements will arise in those skilled in the art under the teachings of this invention without departing from its scope. The appended claims are intended to define the scope of the invention and thereby cover methods and structures within the scope of the claims themselves and their equivalents.

Claims

1. A method for variable sampling acquisition of wafer inspection signals based on synchronous control of an acousto-optic modulator, characterized in that, Includes the following steps: Step 101: Based on the parameters of the wafer defect detection system, plan the full wafer scan sampling rate; Step 102: Configure scanning motion parameters to determine the scanning range and scanning step of the wafer to be inspected; Step 103: Generate an actual sampling rate signal based on the scanning motion parameters and the full-wafer scanning sampling rate, and send it to the signal acquisition card; Step 104: The signal acquisition card receives the actual sampling rate signal, generates the AOM control timing sequence, synchronously controls the acousto-optic modulator AOM and the photodetector, and acquires the photoelectric signal corresponding to the scattered light on the wafer surface to obtain the original sampling data. Step 105: The signal acquisition card performs interpolation processing on the original sampling data to ensure that the sampling point density is the same for the wafer under the same rotation scanning stroke. as well as Step 106: Perform defect algorithm analysis on the interpolated image data to identify and detect defects on the wafer surface; In step 101, the parameters of the wafer defect detection system include the laser incident power P0 on the wafer surface, the focused spot diameter d0, the wafer rotation speed RPM, and the damage radius R. D Threshold radius R T And the outer diameter R0; In step 101, planning the full-wafer scan sampling rate includes: Plan the radial scan pixel size Pr and set the threshold radius R. T R T = N×Pr, where N is a power of 2 and satisfies R T >R D ; Delineate wafer scan partition boundaries: radius less than threshold radius R T The time is the wafer variable sampling rate scanning area, and the radius is greater than the threshold radius R. T This is the wafer's fixed sampling rate scanning area; The AOM laser pixel switching time t0 is planned to satisfy t0 < T, where T is the incident light spot at the damage radius R. D Equivalent exposure time at the location; The planned circumferential sampling rate ft is the number of pixels collected in one circumference at the threshold radius position, satisfying 2×pi×R. T / ft < d0; Plan the circumferential sampling rate fo of the outermost diameter R0 of the wafer, satisfying 2×pi×R0 / fo<d0 and fo≥ft; For wafer regions according to R T / 2、R T / 4、R T / 8、…The radius range is divided into intervals by binary division at each level, corresponding to a progressive halving of the circumferential sampling rate; and… Based on the sampling period of the radius range and combined with the fixed pixel light-on time t0, the corresponding pixel light-off time is calculated to form a complete modulation timing sequence for each radius range; In step 104, the signal acquisition card generates an AOM control timing sequence consisting of pixel on-time and pixel off-time based on the actual sampling rate signal; synchronously controlling the acousto-optic modulator (AOM) and photodetector, the AOM is used to rapidly switch the output beam of the laser source on and off, based on the incident light spot within the damage radius R. D The equivalent irradiation time at the point determines the effective irradiation time of the wafer film layer; Among these measures, the effective irradiation time of the wafer film layer is controlled when the incident light spot is within the damage radius R. D Within the equivalent irradiation time at the location, the incident light spot is within the damage radius R D The equivalent exposure time at the location is: T=60×d0 / (2×pi×R D ×RPM)。 2. The wafer inspection signal variable sampling acquisition method according to claim 1, characterized in that, In step 102, the scanning motion parameters include the inner scanning diameter Rin, the outer scanning diameter Rout, and the radial scanning pixel size Pr.

3. The wafer inspection signal variable sampling acquisition method according to claim 1, characterized in that, Step 102 also includes: Determine the number of acquisition cycles corresponding to the radius range interval. Number of acquisition cycles = radial width of the radius range interval / radial scan pixel size Pr.

4. The wafer inspection signal variable sampling acquisition method according to claim 3, characterized in that, In step 103, generating the actual sampling rate signal based on the scanning motion parameters and the full-wafer scanning sampling rate includes: comparing the scanning inner diameter Rin and the scanning outer diameter Rout, determining the radius range between them, and then matching the preset circumferential sampling rate and the corresponding number of acquisition cycles within the corresponding radius range to complete the planning and generation of the variable sampling rate control signal; and / or The actual sampling rate signal includes the circumferential sampling rate corresponding to the radius interval and the number of sampling cycles corresponding to the radius range interval.

5. The wafer inspection signal variable sampling acquisition method according to claim 1, characterized in that, AOM refers to quartz crystal AOM.

6. The wafer inspection signal variable sampling acquisition method according to claim 1, characterized in that, In step 105, the original sampling data includes sampling data of the wafer fixed sampling rate scanning area and sampling data of the wafer variable sampling rate scanning area; during the interpolation processing of the original sampling data by the signal acquisition card, the sampling data of the wafer variable sampling rate scanning area is supplemented to make the circumferential sampling number of the wafer variable sampling rate scanning area equal to the circumferential sampling rate of the wafer fixed sampling rate scanning area.

7. A wafer defect detection system, characterized in that, For implementing the wafer inspection signal variable sampling acquisition method according to any one of claims 1-6, the wafer defect detection system comprises: A laser source is configured to output laser light to the surface of the wafer to be inspected in order to generate scattered light; AOM, which is configured to rapidly switch the laser source; The AOM control module is configured to receive instructions from the signal acquisition card and drive the AOM to switch the laser on and off. A photodetector is configured to receive the scattered light and convert the optical signal into an electrical signal for transmission to a signal acquisition card. The signal acquisition card is configured to receive instructions from the host computer, synchronously control the AOM and photodetector, convert electrical signals into digital data and perform interpolation processing. The scanning platform is configured to rotate and radially move the wafer under inspection to complete a full-area scan. A motion control unit, configured to control the motion of the scanning platform; and The host computer is the main control console.

Images