Extreme ultraviolet collecting mirror grating hybrid measuring system and method

Abstract

The invention provides an extreme ultraviolet collection mirror grating hybrid measurement system and method, and relates to the technical field of extreme ultraviolet (EUV) lithography and nanometer metrology, and the system comprises an atomic force microscopic module which is in communication connection with a data processing module and is used for carrying out high-resolution morphology detection on an abnormal morphology according to position information, and obtaining image data; the shared precision positioning platform module is connected with the optical scattering measurement module and the atomic force microscopy module and used for controlling movement of the optical scattering measurement module and the atomic force microscopy module so that the atomic force microscopy module can accurately move to the position where the abnormal morphology is located; wherein the data processing module is also used for generating a comprehensive detection report based on the geometric structure parameters reconstructed by the optical scattering measurement module and the image data acquired by the atomic force microscopic module. According to the invention, high-precision, high-efficiency, cross-scale and lossless comprehensive characterization of the grating on the surface of the EUV collecting mirror can be realized.

Description

Technical Field

[0001] This invention relates to the fields of extreme ultraviolet (EUV) lithography and nanometer technology, and in particular to an EUV collection mirror grating hybrid measurement system and method. Background Technology

[0002] In extreme ultraviolet (EUV) lithography systems, the collector mirror is the first critical optical component, and its performance directly affects the throughput of the light source and the overall efficiency of the lithography machine. To improve system performance, advanced EUV collector mirrors are designed with precise periodic grating structures on their surfaces. Their core function is to efficiently and selectively reflect 13.5nm wavelength EUV light while suppressing reflections from other non-working wavelengths (i.e., out-of-band light), thereby reducing aberrations and thermal load and improving lithography quality.

[0003] The performance of the surface gratings of the EUV collector mirror is strictly dependent on their physical structural parameters, including geometric features such as period, duty cycle, trench depth, and sidewall profile. Any slight deviation from the design values, whether introduced by the manufacturing process or by degradation caused by long-term exposure to extreme high temperatures and plasma radiation (such as material erosion and thermal deformation), will cause a shift in their spectral response characteristics, thereby reducing filtering efficiency, allowing more out-of-band light to enter the lithography system, and ultimately affecting exposure accuracy and yield. Therefore, high-precision, non-destructive periodic measurement of these grating structures is crucial.

[0004] Existing single measurement techniques are insufficient to meet the needs of comprehensive evaluation. While contact profilometers and atomic force microscopy (AFM) can provide surface topography or images with sub-nanometer resolution, they suffer from drawbacks such as slow measurement speed and inability to cover large-area samples. Although optical scattering measurement methods can quickly and non-destructively invert the overall geometric parameters of gratings, their measurement results are statistical averages within the spot area, which are insensitive to local and minute defects and cannot provide intuitive morphological information about defects. Therefore, the industry urgently needs a comprehensive solution for extreme ultraviolet collection mirror gratings that can simultaneously assess overall performance and detect local defects. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a hybrid measurement system and method for extreme ultraviolet (EUV) collecting mirror gratings, which can achieve high-precision, high-efficiency, cross-scale and non-destructive comprehensive characterization of the surface gratings of EUV collecting mirrors.

[0006] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: An extreme ultraviolet collection mirror grating hybrid measurement system includes: An optical scattering measurement module is used to acquire the diffraction spectrum of a grating and reconstruct the geometric parameters of the grating based on the diffraction spectrum; The data processing module is communicatively connected to the optical scattering measurement module, and is used to receive the geometric structure parameters, identify the abnormal morphology of the grating based on the geometric structure parameters, and output the location information of the abnormal morphology. An atomic force microscopy module, which is communicatively connected to the data processing module, is used to perform high-resolution morphology detection on the abnormal morphology based on the location information and acquire image data. A shared precision positioning platform module is connected to the optical scattering measurement module and the atomic force microscopy module, and is used to control the movement of the optical scattering measurement module and the atomic force microscopy module so that the atomic force microscopy module can be precisely moved to the location of the abnormal morphology; The data processing module is also used to generate a comprehensive detection report based on the geometric structure parameters reconstructed by the optical scattering measurement module and the image data acquired by the atomic force microscopy module.

[0007] Furthermore, the optical scattering measurement module includes an illumination unit, a polarization control unit, a beam splitting unit, a focusing unit, and a spectral measurement unit; The illumination unit is used to generate a probe beam; the polarization control unit is connected to the illumination unit and is used to receive the probe beam and adjust its polarization state. The beam splitting unit is connected to the polarization control unit and is used to guide the probe beam after the polarization state is adjusted to the focusing unit; the focusing unit is used to focus the probe beam from the beam splitting unit onto the surface of the grating sample under test and collect the diffracted light generated by the grating sample under test. The spectral measurement unit is connected to the beam splitting unit and is used to receive the diffracted light collected by the focusing unit and transmitted through the beam splitting unit, and to collect the light intensity signals of multiple diffraction orders to form a diffraction spectrum.

[0008] Furthermore, the data processing module identifies abnormal morphology by analyzing at least one of the following methods: analyzing the deviation between the geometric structure parameters and the design values, analyzing the local uniformity of the geometric structure parameters on the two-dimensional distribution map, or analyzing the fitting residual of the diffraction spectrum. When the deviation of the geometric structure parameters from the design value exceeds the preset tolerance range, or the distribution of the geometric structure parameters changes drastically in a local area, or the fitting residual of the diffraction spectrum is greater than a set threshold, the data processing module will mark the corresponding area as an abnormal morphology.

[0009] Furthermore, the abnormal morphology includes geometric defects and laser damage defects introduced by the processing, or surface dust particle contamination introduced by the usage environment; the atomic force microscopy module is used to perform high-resolution three-dimensional morphology imaging on the abnormal morphology marked by the data processing module, so as to directly observe and quantify the specific morphological features of the processing defects or contamination.

[0010] Furthermore, the atomic force microscopy module operates in a tapping mode, scanning the sample surface with an ultra-sharp probe; the atomic force microscopy module receives anomalous morphology location information from the data processing module, and scans the target area based on the location information to obtain three-dimensional morphology data of the anomalous morphology with sub-nanometer resolution.

[0011] Furthermore, the shared precision positioning platform module includes a clamp for horizontally holding and fixing the extreme ultraviolet collecting mirror to be tested and a high-precision multi-axis motion module. The shared precision positioning platform module receives the abnormal morphology location information sent by the data processing module and drives the high-precision multi-axis motion module to control the relative position and angle changes of the extreme ultraviolet collecting mirror and the atomic force microscopy module, so as to precisely move the ultra-sharp probe of the atomic force microscopy module to the location of the abnormal morphology.

[0012] A hybrid measurement method using an extreme ultraviolet collecting mirror and grating, for implementing the aforementioned system, includes: The optical scattering measurement module scans the grating sample fixed on the shared precision positioning platform module along a predetermined path, and collects diffraction spectra at each measurement point. The data processing module then reconstructs the geometric parameters of the grating based on the diffraction spectra. The data processing module receives and analyzes the geometric structure parameters. By judging whether the geometric structure parameters exceed the preset tolerance range, or whether the distribution of the geometric structure parameters changes drastically in a local area, or whether the fitting residual of the diffraction spectrum exceeds the set threshold, abnormal morphology in the grating is identified and the location information of the abnormal morphology is generated. The data processing module sends the location information of the abnormal morphology to the shared precision positioning platform module and the atomic force microscopy module. The shared precision positioning platform module drives the atomic force microscopy module to move, so that its ultra-sharp probe can be precisely moved to the location of the abnormal morphology and perform high-resolution scanning on the abnormal morphology to obtain its three-dimensional morphology image data. The data processing module receives the three-dimensional morphology image data acquired by the atomic force microscopy module, and correlates and analyzes it with the geometric structure parameters reconstructed in the optical scattering measurement step to generate a comprehensive inspection report that includes global parameter distribution, abnormal morphology features and classification statistics.

[0013] Furthermore, in the abnormal morphology identification step, the specific criteria used by the data processing module to identify abnormal morphologies include: The values ​​of the geometric parameters exceed the tolerance range preset according to the grating design specifications; or The geometric structure parameters exhibit localized abrupt changes across a preset threshold on the two-dimensional distribution map; or The residual obtained after fitting the measured data of the diffraction spectrum with the theoretical model is greater than the set threshold.

[0014] Furthermore, in the atomic force microscopy measurement step, the atomic force microscopy module operates in a tapping mode, with its ultra-sharp probe scanning the target area based on the location information of abnormal morphology; the acquisition of three-dimensional morphology image data specifically includes: The three-dimensional morphological features of anomalous morphology, including processing defects or surface contaminants, are directly observed and quantified by probe scanning.

[0015] Furthermore, in the comprehensive report generation step, the data processing module performs specific analyses on the 3D topography image data, including: The defect type and geometric dimensions of the abnormal morphology are identified and quantified. The precise features of the defect type are correlated and compared with the corresponding scattering measurement abnormal signals in the abnormal morphology identification step, and finally a comprehensive inspection report is generated.

[0016] The above-described solution of the present invention has at least the following beneficial effects: Leveraging the advantages of fast and non-destructive scattering measurement, it can quickly complete large-area scanning and locate suspected defect areas, solving the pain point of AFM's inability to perform rapid full-field search, without causing damage to the sample.

[0017] Leveraging the sub-nanometer resolution of AFM, precise three-dimensional morphological measurements and quantitative analysis are performed on located anomalies, solving the problems of non-unique indirect inversion results from scattering measurements and the inability to directly observe defects. By analyzing the root causes of defects through data correlation, the results can be quickly fed back to the manufacturing end, forming a closed loop of detection, analysis, and improvement, effectively improving manufacturing yield. Through a shared precision positioning platform and software algorithms, a fully automated process from anomaly detection to anomaly analysis is achieved, reducing human intervention and improving the objectivity and efficiency of detection. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of a hybrid measurement system with a grating structure on the surface of an EUV collecting mirror according to the present invention.

[0019] Figure 2 Is using Figure 1 Flowchart of the hybrid measurement method for the surface grating structure of the EUV collecting mirror implemented by the device.

[0020] The diagram is labeled as follows: 1. Sample conveyor belt; 2. Sample collection mirror grating; 3. Lens group; 4. Illumination unit; 5. Polarization control unit; 6. Semi-reflective lens; 7. Spectral measurement unit; 8. Data processing module; 9. Atomic force microscopy module; 10. Shared precision positioning platform. Detailed Implementation

[0021] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0022] like Figures 1 to 2 As shown, the present invention provides a hybrid measurement system for collecting mirror gratings in extreme ultraviolet lithography systems, which is used to achieve cross-scale, high-precision comprehensive detection of the grating structure on the surface of collecting mirrors in extreme ultraviolet lithography systems. The system mainly includes: a sample conveyor belt 1, an optical scattering measurement module, an atomic force microscopy module 9, a shared precision positioning platform 10, and a data processing module 8.

[0023] The optical scattering measurement module is used to quickly and non-destructively acquire the diffraction spectrum of the grating and reconstruct its geometric parameters. The module is specifically composed of an illumination unit 4, a polarization control unit 5, a beam splitting unit (e.g., a semi-reflective mirror 6), a focusing unit (e.g., a lens group 3), and a spectral measurement unit 7.

[0024] The data processing module 8 is communicatively connected to the optical scattering measurement module and the atomic force microscopy module 9, respectively, and is used to coordinate and control the entire measurement process; the atomic force microscopy module 9 is used to perform sub-nanometer resolution three-dimensional morphology imaging on the identified abnormal morphology; the shared precision positioning platform 10 is used to horizontally clamp and fix the sample 2 to be tested, and integrates the control of the movement of the optical scattering measurement module and the atomic force microscopy module 9 to ensure the accurate correlation and reproduction of the measurement position.

[0025] The system works as follows: At the start of the measurement, the sample conveyor belt 1 transports the sample 2 to be measured, which is then fixed by the shared precision positioning platform 10. First, the optical scattering measurement module operates: the illumination unit 4 generates a probe beam, which, after its polarization state is adjusted by the polarization control unit 5, is guided by the beam splitting unit 6 to the focusing unit 3, and finally focused onto the surface of the sample 2. The generated diffracted light is collected by the focusing unit 3, reflected by the beam splitting unit 6, and then enters the spectral measurement unit 7, where multiple diffraction orders are collected to form a diffraction spectrum.

[0026] Subsequently, the data processing module 8 receives the diffraction spectrum data acquired by the spectral measurement unit 7, and the specific implementation process of parameter reconstruction and anomaly identification is as follows: Geometric structure parameter reconstruction and 2D distribution map generation: The data processing module 8 is designed to achieve high-precision inversion from diffraction spectra to geometric parameters. Its core is to construct and apply a grating electromagnetic scattering forward model based on Rigorous Coupled-Wave Analysis (RCWA).

[0027] The specific construction process of the forward model is as follows: First, the data processing module 8 establishes a parameterized grating geometric model based on the basic design parameters of the grating sample 2 to be tested (such as the optical constants of the material and the approximate value of the grating period) and the known measurement conditions of the optical scattering measurement module (such as the wavelength range of the probe light generated by the illumination unit 4 and the incident angle range). The grating geometric model discretizes the cross-sectional structure of the grating (such as a rectangle, trapezoid, or more complex contour) into multi-layer slices and defines its key geometric parameters as variables to be inverted, mainly including period, duty cycle, groove depth, sidewall angle, etc.

[0028] Subsequently, a numerical calculation engine based on the rigorous coupled-wave theory (RCWA) is embedded into the data processing module 8. This engine can accurately simulate the complete physical process of the interaction between the probe light wave and the parameterized grating model by solving the Maxwell equations under periodic boundary conditions. The calculation process includes: expanding the incident plane wave into spatial harmonics (Fourier series), performing electromagnetic field matching in the grating region, and finally calculating the theoretical diffraction efficiency (i.e., light intensity signal) of each diffraction order (detected by the spectral measurement unit 7) generated by the grating under given geometric parameters and measurement conditions.

[0029] Thus, the forward model constructed within the data processing module 8 achieves an accurate mapping from a hypothetical set of grating geometric parameters to the corresponding theoretical diffraction spectrum. During actual inversion, the data processing module 8 uses the measured diffraction spectrum acquired by the spectral measurement unit 7 as the target and continuously adjusts the geometric parameter values ​​in the forward model through a nonlinear optimization algorithm (such as the Levenberg-Marquardt algorithm) until the difference between the theoretical spectrum and the measured spectrum output by the model (usually measured by the sum of squared residuals) is minimized. At this point, the geometric parameter values ​​used by the forward model are recognized as the actual geometric structure parameters of the grating reconstructed through inversion.

[0030] When the goodness of fit reaches the preset requirement (e.g., the residual is less than the threshold), the iteration process stops. At this time, the parameter values ​​used by the geometric model are identified as the actual grating geometric structure parameters (including period, duty cycle, groove depth, etc.) reconstructed by inversion. The data processing module 8 controls the optical scattering measurement module to scan the sample 2 point by point and repeats the above fitting and inversion process for each measurement point. Finally, the data processing module 8 generates a two-dimensional distribution map (or cloud map) of the parameter by corresponding the values ​​of the same parameter (e.g., groove depth) obtained by inversion of all measurement points with its two-dimensional coordinates on the sample surface.

[0031] Automatic identification, marking, and localization of abnormal morphologies: After obtaining the geometric structure parameters and their two-dimensional distribution across the entire field, the data processing module 8 automatically identifies abnormal morphologies through the following parallel or serial analysis process: The data processing module 8 compares the parameter values ​​obtained from the inversion of each measurement point with the pre-stored grating design specifications and their allowable tolerance range. For example, if the design groove depth is 50nm and the tolerance range is ±1nm, then if the reconstructed groove depth value of any measurement point falls outside the range of 49nm to 51nm, the point will be marked as abnormal.

[0032] The data processing module 8 performs gradient analysis or regional statistics on the two-dimensional distribution map of the parameters. It evaluates the local uniformity by calculating the difference (or standard deviation) of the parameter values ​​between each point and its surrounding neighboring points. When a sharp change in parameter values ​​is detected in a local area that crosses a preset gradient threshold (for example, the difference in trench depth between adjacent points exceeds 0.5 nm), it indicates that there may be defects or contamination in the area, and the module 8 marks this area of ​​dramatic change as an anomaly.

[0033] Based on the identification of fitting residuals, even if the inverted parameter values ​​are within the tolerance range, if the residual between the measured spectrum and the optimal fitted theoretical spectrum at that point is still greater than the set threshold, it indicates that there is a difference between the actual physical structure and the model that cannot be perfectly matched (it may be a complex defect or contamination). The data processing module 8 will also mark such points with abnormally high fitting residuals as anomalies.

[0034] The working principle is that the data processing module 8 first uses a physical model to convert the raw spectral data into geometric parameters with clear physical meaning to achieve quantification; then, by setting multiple complementary criteria, each region is screened, and any region that touches the red line will be accurately locked.

[0035] Finally, the data processing module 8 highlights the measurement points or areas that meet any one or more of the above-mentioned abnormal conditions as abnormal morphologies on the two-dimensional distribution map, and uses the world coordinate system of the shared precision positioning platform 10 to accurately record the coordinate information (such as X, Y, Z values) of each abnormal morphology, generating a list of abnormal coordinates to be precisely measured, providing navigation targets for the subsequent precise positioning of the atomic force microscopy module 9.

[0036] Next, the precision measurement stage begins: the data processing module 8 controls the shared precision positioning platform 10, driving the ultra-sharp probe of the atomic force microscopy module 9 to move to the identified coordinate points of the abnormal morphology. The atomic force microscopy module 9 performs a high-resolution scan of the target area in a tapping mode to acquire real three-dimensional morphological image data.

[0037] Finally, the data processing module 8 integrates the global parameter distribution provided by optical scattering measurement with the local high-resolution morphology data provided by atomic force microscopy, performs correlation analysis, and automatically generates a comprehensive inspection report that includes defect classification, size statistics, and overall quality assessment.

[0038] Through the above methods, the present invention achieves coordinated measurement by first conducting rapid global screening and then conducting precise local diagnosis, effectively balancing detection efficiency and accuracy.

[0039] In a preferred embodiment of the present invention, combined with Figure 1 and Figure 2 As shown, the optical scattering measurement module is composed of an illumination unit 4, a polarization control unit 5, a semi-reflective lens 6, a lens group 3, and a spectral measurement unit 7 connected in sequence to form a complete optical path detection and signal acquisition chain.

[0040] Its specific structure and working principle are as follows: The illumination unit 4 is used to generate and output a high-quality probe beam, which first enters the polarization control unit 5. The polarization control unit 5 is used to precisely control the polarization state of the probe beam (e.g., convert it to linearly polarized light or elliptically polarized light) to match the optimal incident conditions for grating measurement.

[0041] The polarization-tuned probe beam then arrives at the beam splitter (e.g., a semi-reflective mirror 6). The main function of the semi-reflective mirror 6 is to reflect the incident beam from the illumination path and guide it to the focusing unit (e.g., lens group 3). The lens group 3 converges the received collimated beam into a small spot and precisely focuses it onto the surface of the sample grating to be tested, which is fixed on the shared precision positioning platform 10.

[0042] When the probe beam illuminates the surface of a grating with a periodic structure, diffraction occurs. The lens group 3 also acts as a collecting objective, responsible for collecting the diffracted light of each order after being diffracted by the grating sample 2. This diffracted light, carrying information about the grating structure, returns along its original path, passes through the lens group 3 again, becomes collimated, and reaches the semi-reflective mirror 6 for the second time. This time, most of the diffracted light will penetrate the semi-reflective mirror 6 and enter the measurement optical path.

[0043] Finally, the diffracted light passing through the semi-reflective mirror 6 is guided into the spectral measurement unit 7. This spectral measurement unit 7 (usually a spectrometer or detector array) is responsible for collecting light intensity signals of multiple diffraction orders (such as 0th order, ±1st order, etc.) at different wavelengths or angles, thereby forming diffraction spectral data for analysis. This diffraction spectrum is the original basis for the subsequent data processing module 8 to reconstruct the grating geometric parameters.

[0044] like Figures 1 to 2 As shown, in a preferred embodiment of the present invention, the data processing module 8 is configured to execute an intelligent anomaly recognition algorithm. The data processing module 8 receives grating geometric parameters (such as period, duty cycle, and groove depth) and their two-dimensional distribution data reconstructed by the optical scattering measurement module, and automatically identifies anomalies based on at least one of the following analysis methods: First, the data processing module 8 compares the reconstructed geometric parameters with the preset design values. When the deviation of any parameter exceeds the tolerance range preset according to the process requirements, the area where the measurement point is located is determined to be abnormal.

[0045] Secondly, the data processing module 8 analyzes the local uniformity of the geometric structure parameters on the two-dimensional distribution map. By calculating the gradient change or statistical fluctuation of parameter values ​​in adjacent regions, when the parameter distribution in a local region undergoes a drastic change that crosses a set threshold, the region of such a drastic change is determined to be abnormal.

[0046] In addition, the data processing module 8 also analyzes the residual between the measured diffraction spectrum collected by the spectral measurement unit 7 and the spectrum obtained by nonlinear fitting based on the theoretical model; when the fitting residual is greater than the set threshold, it indicates that there is a significant difference between the actual structure and the ideal model, and the corresponding region is also marked as an anomaly.

[0047] Its working principle is as follows: The data processing module 8, by comprehensively applying the above criteria, quickly screens the measurement data across the entire domain. It can sensitively detect minute structural anomalies caused by factors such as processing defects, laser damage, or surface contamination. Once any anomaly condition is met, the data processing module 8 will immediately mark the corresponding area as an abnormal morphology in the world coordinate system and generate a list containing precise coordinate information. This provides a clear navigation target for the subsequent precise positioning and detailed imaging of the atomic force microscopy module 9. This process enables the automatic, objective, and efficient positioning of suspected defects from massive amounts of measurement data and is the core link in realizing the intelligence and automation of the entire hybrid measurement system.

[0048] like Figures 1 to 2 As shown, in a preferred embodiment of the present invention, the abnormal morphology mainly includes two categories: one is defects introduced during the processing, such as geometric structural defects (e.g., sidewall roughness, contour distortion) and laser damage defects (e.g., erosion pits); the other is surface contamination introduced by the usage environment, such as dust particle adhesion. The core function of the atomic force microscopy module 9 is to perform high-resolution three-dimensional morphology imaging on these abnormal morphologies marked by the data processing module 8.

[0049] Its specific working principle is as follows: Once the data processing module 8 completes the analysis of the optical scattering measurement data and outputs a list of precise coordinates for the anomalous morphology, the system initiates the precise positioning and fine detection process. The shared precision positioning platform module 10 receives the anomalous morphology location command from the data processing module 8, and its built-in high-precision multi-axis motion module begins to work collaboratively. This module controls the relative motion between the fixture holding the extreme ultraviolet collecting mirror sample 2 and the atomic force microscopy module 9, rapidly and precisely moving the ultra-sharp probe of the atomic force microscopy module 9 to the spatial location of the target anomalous morphology.

[0050] Once the localization is complete, the atomic force microscopy module 9 is activated. This module operates in a tapping mode that causes minimal damage to the sample, with its ultra-sharp probe performing high-frequency micro-amplitude vibrations and scanning on the sample surface. The interaction forces between the probe and the atoms on the anomalous morphology surface are detected in real time and converted into electrical signals. Algorithms then reconstruct three-dimensional morphology data of the region with sub-nanometer resolution. This process allows for the direct observation, clear imaging, and precise quantification of anomalous features previously indirectly indicated by scattering measurements, such as scratch depth, contaminant particle size, or the microstructure of damage pits, thus enabling the final confirmation and in-depth diagnosis of defects.

[0051] like Figures 1 to 2 As shown, this invention provides a hybrid measurement method using an extreme ultraviolet collecting mirror grating. This method is implemented through the hybrid measurement system and specifically includes the following steps: Step S1, Optical scattering measurement and parameter reconstruction: The optical scattering measurement module (including an illumination unit 4, a polarization control unit 5, a beam splitting unit 6, a focusing unit 3, and a spectral measurement unit 7) scans the sample 2 of the collection mirror grating fixed on the shared precision positioning platform module 10 along a predetermined path. At each measurement point, the beam generated by the illumination unit 4 is adjusted by the polarization control unit 5 and then illuminates the surface of the sample 2 through the beam splitting unit 6 and the focusing unit 3. The spectral measurement unit 7 simultaneously acquires the generated diffraction spectrum. Subsequently, the data processing module 8 receives the diffraction spectrum data from all measurement points and, based on a rigorous coupled-wave analysis and other electromagnetic field calculation models, inverts and reconstructs the geometric structural parameters of the grating, such as the period, duty cycle, and groove depth, as well as their two-dimensional distribution map.

[0052] Step S2: Intelligently identify and locate abnormal shapes: The data processing module 8 receives and analyzes the geometric structure parameters reconstructed in step S1. It uses a preset algorithm to determine whether the geometric structure parameter values ​​at each point exceed the tolerance range set according to the design specifications, whether the two-dimensional distribution map of the analyzed parameters shows a drastic change exceeding a set threshold in a local area, or whether the fitting residual between the measured diffraction spectrum and the theoretical model spectrum exceeds a set threshold. When any of these abnormal conditions are met, the data processing module 8 marks the corresponding area as an abnormal morphology and accurately calculates its position information in the world coordinate system of the shared precision positioning platform 10, generating an abnormal coordinate list.

[0053] Step S3, Precise positioning and fine scanning using atomic force microscopy: The data processing module 8 sends the location information of the abnormal morphology generated in step S2 to the shared precision positioning platform module 10 and the atomic force microscopy module 9; the shared precision positioning platform module 10 drives its high-precision multi-axis motion module to adjust the relative position of the sample 2 and the atomic force microscopy module 9, and precisely moves the ultra-sharp probe of the atomic force microscopy module 9 to the coordinate point of each abnormal morphology; then, the atomic force microscopy module 9 performs a high-resolution scan of the tiny area in a tapping mode, and obtains three-dimensional morphology image data of the abnormal morphology with sub-nanometer resolution by detecting the interaction force between the probe and the sample surface.

[0054] Step S4, Data Correlation Analysis and Comprehensive Report Generation: The data processing module 8 receives the three-dimensional morphology image data acquired by the atomic force microscopy module 9 in step S3. The data processing module 8 first analyzes the image, directly observing and quantifying the specific characteristics of the abnormal morphology (such as defect type, size, and depth). Subsequently, it correlates and analyzes these local high-resolution morphology data with the global geometric structure parameter distribution obtained in step S1 and the anomaly identification signal in step S2. Finally, the data processing module 8 automatically generates a comprehensive inspection report, which includes a global parameter distribution cloud map of the grating, an AFM three-dimensional morphology map of the abnormal morphology, a defect classification statistical table, and data correlation analysis conclusions.

[0055] like Figures 1 to 2 As shown, in a preferred embodiment of the present invention, the key steps of the hybrid measurement method are specifically implemented as follows: In the abnormal morphology identification step, the data processing module 8 identifies abnormal morphologies based on three criteria: Parameter tolerance criterion: The reconstructed grating geometric parameters (such as period, duty cycle, and groove depth) are compared with the tolerance range preset according to the grating design specifications; if any parameter value exceeds its corresponding upper or lower tolerance limit, it is determined that there is an anomaly in the area where the measurement point is located.

[0056] Uniformity criterion: Analyze the gradient changes of the geometric structural parameters on the two-dimensional distribution map. By calculating the differences in parameter values ​​between adjacent measurement points, when a drastic change in parameter values ​​occurs in a local area (i.e., the gradient value crosses a preset threshold), it indicates that the structural uniformity of that area has been disrupted, and it is identified as an abnormal area.

[0057] Criteria for abnormal fitting residuals: The data processing module 8 compares the measured diffraction spectrum data collected by the spectral measurement unit 7 at a specific measurement point with the optimal theoretical spectrum obtained by nonlinear fitting based on the parameterized grating forward model (specifically, a model based on the strict coupled-wave theory - RCWA) constructed internally by the data processing module 8.

[0058] The specific operation is as follows: calculate the residual between the measured spectral data points and the corresponding data points of the best theoretical spectrum obtained by fitting (usually quantified in the form of residual sum of squares or root mean square error, etc.); when this residual is greater than a preset threshold, it indicates that even if the optimal combination of geometric parameters is found, the forward model cannot simulate the measured spectrum well. This reveals that there is a significant deviation that cannot be ignored between the actual physical structure of the grating sample 2 under test and the structure that the parameterized model can describe. This deviation often stems from factors not covered by the model, such as complex local defects, surface contamination or severe structural irregularities. Therefore, the data processing module 8 marks the corresponding regions with such high fitting residuals as abnormal morphologies.

[0059] Its working principle is as follows: by setting multiple and complementary criteria, the data processing module 8 can comprehensively screen the measurement data from different dimensions (parameter values, spatial distribution, spectral signal consistency), which significantly improves the sensitivity and reliability of abnormal morphology identification and effectively avoids missed detection or false detection.

[0060] The specific implementation method in the atomic force microscopy measurement procedure is as follows: The data processing module 8 sends the precise location coordinates of the generated abnormal morphology to the shared precision positioning platform 10 and the atomic force microscopy module 9. The shared precision positioning platform 10 first performs coarse positioning, drives its high-precision multi-axis motion module, and moves the sample 2 of the collection mirror grating to be tested or the scanning head of the atomic force microscopy module 9 fixed on it over a large range, so that the ultra-sharp probe of the atomic force microscopy module 9 is initially aligned with the target area.

[0061] Subsequently, the atomic force microscopy module 9 operates in a tapping mode that minimizes sample damage. In this mode, an ultra-sharp probe mounted on a piezoelectric scanner vibrates at a high frequency with micro-amplitude near its resonant frequency; as the probe gradually approaches the surface of sample 2 to a nanoscale distance, the weak interaction forces (mainly van der Waals forces) between the probe tip and the atoms on the sample surface cause changes in the vibration amplitude or phase of the probe.

[0062] The specific scanning and signal detection process is as follows: Based on received position commands, the piezoelectric scanner drives the probe to perform precise point-by-point, line-by-line raster scanning of the target area. During the scanning process, an automatic feedback control system is activated, which detects changes in the probe's vibration state in real time (e.g., by illuminating the rear end of the probe's cantilever beam with a laser beam and detecting the displacement of the reflected light spot using a position-sensitive detector (PSD),) and converts these changes into corresponding electrical signals. This feedback system simultaneously dynamically adjusts the average distance (Z-direction) between the probe and the sample surface to maintain a constant probe vibration amplitude (constant amplitude mode). The compensation voltage signal applied to the piezoelectric scanner in the Z-direction to maintain constant amplitude directly corresponds to the height of the undulations on the sample surface.

[0063] The process of acquiring and reconstructing three-dimensional topographic image data specifically includes: The data acquisition system of the atomic force microscopy module 9 records the Z-direction compensation voltage value corresponding to each point (X, Y coordinates) within the scanning area. Using the known piezoelectric ceramic sensitivity coefficient, these voltage values ​​are converted into true height values. By combining the height information of all points in the entire scanning area, a high-resolution true three-dimensional topography map (topology map) of the abnormal morphology area can be directly reconstructed.

[0064] This process provides nanometer-level and even sub-nanometer-level longitudinal resolution, enabling intuitive observation and precise quantification of the microscopic features of abnormal morphology. For example, for processing defects (such as scratches), their width, depth, and cross-sectional shape can be measured; for structural fractures, the steepness of the fracture surface can be assessed; and for surface contaminants (such as dust particles), their particle size, height, and distribution can be measured. This precise morphological data provides the most direct evidence for the diagnosis and classification of defects.

[0065] In the comprehensive report generation step, the data processing module 8 performs the following detailed analysis on the three-dimensional morphology image data acquired by the atomic force microscopy module 9: Defect identification and quantification: The data processing module 8 first preprocesses the acquired three-dimensional topography data matrix. For example, the data processing module 8 first performs background tilt correction. Because the sample 2 may have a slight tilt angle when mounted on the shared precision positioning platform 10, the acquired three-dimensional topography data contains a tilted reference plane. The data processing module 8 uses a plane fitting algorithm based on the least squares method to correct the background plane tilt caused by the sample mounting tilt. The core objective of this process is to find a mathematical plane that best represents the macroscopic tilt trend of the entire scanning area and remove it from the original data. The specific implementation steps are as follows: The data processing module 8 extracts the coordinates (X, Y) and corresponding height values ​​(Z) of all pixels from the three-dimensional topographic data of the entire scanned area acquired by the atomic force microscopy module 9. Module 8 assumes that these data points are distributed on an ideal inclined plane, whose mathematical equation can be expressed as Z = Ax + By + C, where A, B, and C are the plane parameters to be determined, representing the slopes of the plane in the X and Y directions and the intercept in the Z direction, respectively.

[0066] Next, the data processing module 8 uses the least squares method to solve for the unknown parameters A, B, and C in the aforementioned plane equation. The goal of the least squares method is to find a set of values ​​for A, B, and C that minimizes the sum of squares of the vertical distances (i.e., residuals) between the ideal plane and all actual data points. By solving a normal system of equations constructed from all data points, the data processing module 8 calculates the numerical solutions for parameters A, B, and C that optimally satisfy this condition. Thus, module 8 obtains a fitted plane equation that accurately describes the macroscopic tilted background.

[0067] Finally, the data processing module 8 performs background subtraction. It iterates through each pixel in the original three-dimensional topography data and calculates the background tilt height value Zfitted corresponding to the point based on the coordinates (X, Y) of the point using the obtained fitting plane equation Zfitted = AX + BY + C. Then, it subtracts the calculated background value Zfitted from the original height value Zoriginal to obtain the corrected new height value Zcorrected = Zoriginal - Zfitted.

[0068] Through the above process, the data processing module 8 effectively removes the macroscopic systematic tilt from the original morphology data and generates a three-dimensional morphology dataset with the reference surface flattened. This allows subsequent analysis to accurately focus on the characteristics of the real microscopic undulations, periodic structures, and abnormal morphology on the surface of the grating sample 2.

[0069] After eliminating the tilted background, the data processing module 8 further applies a digital filter to suppress random noise in the image. This noise may originate from minute vibrations of the probe in the atomic force microscope module 9, thermal noise from the electronic system, or environmental interference. The data processing module 8 typically employs a nonlinear filter, such as a median filter, to perform convolution operations on the processed data. This filter uses a sliding window of a specific size (e.g., 3x3 pixels), traversing each pixel in the image and sorting the height values ​​of all pixels within the window. The median value is then taken as the new height value of the current center pixel. This method can effectively filter out isolated, high-frequency noise points with abnormal amplitudes. (Such as spike-like noise), while better preserving the true edge information of defects, thus significantly improving the signal-to-noise ratio of the image and highlighting the true morphological features, preparing for subsequent accurate threshold segmentation and defect quantization; subsequently, the data processing module 8 uses a threshold segmentation algorithm to automatically separate the abnormal morphological region from the normal raster background surface based on the significant difference between the pixel height value and the surrounding background region, generating a binary image to mark the potential defect region; then, after obtaining the preliminary binary image through threshold segmentation, the data processing module 8 sequentially applies two morphological operations, closing and opening operations, to optimize the morphological features of the defect region. This process targets the white connected regions representing the abnormal morphology.

[0070] Closing operations are used to fill holes and connected areas: Step 1: Dilation Operation. The data processing module 8 first performs a convolution calculation on a predefined shape called a structuring element (typically a small circular or square array of pixels). This structuring element acts as a probe, sliding across the binary image. For each background pixel in the image, if at least one foreground (white) pixel exists in its neighborhood, the data processing module 8 also converts that background pixel into a foreground pixel. This operation causes all white areas to expand outward, effectively bridging narrow gaps between white areas and filling black holes within the areas that are smaller than those created by the structuring element.

[0071] The second step is the erosion operation. Immediately afterward, the data processing module 8 performs an erosion operation on the dilated image. It also uses structuring elements for sliding calculations, but for each foreground (white) pixel in the image, if there is at least one background (black) pixel in its neighborhood, the foreground pixel is transformed into a background pixel. This operation causes all white areas to shrink inward, which removes the boundary burrs introduced by the dilation operation, making the area boundaries more regular, while retaining the area connection and hole filling effects achieved in the previous step. After this dilation followed by erosion operation, defective areas that may have small breaks or internal holes are connected and filled, forming a more complete and continuous connected area.

[0072] Opening operations are used to smooth boundaries and filter out noise. Step 1: Erosion operation. The data processing module 8 performs an opening operation on the image after the closing operation. First, the erosion operation is performed. This step will uniformly shrink all white areas. Its key function is to eliminate those very small, isolated white pixel clumps. These clumps are usually pseudo-defect points caused by random noise.

[0073] The second step is the dilation operation. Then, the data processing module 8 performs a dilation operation on the eroded image. This step causes the remaining white areas to expand again, restoring them to near their pre-erosion size. However, the small noise points that were previously completely eroded cannot be recovered. The effect is to smooth the boundary contours of larger defect areas while completely eliminating tiny isolated noise.

[0074] By performing a continuous process of first closing the operation and then opening the operation, the data processing module 8 effectively purifies the preliminary segmentation results, so that the final connected regions representing abnormal morphology have complete shapes, smooth boundaries, and minimize noise interference.

[0075] After completing the morphological optimization, the data processing module 8 calculates the geometric feature descriptors for each independent connected region. These descriptors include: For area calculation, the data processing module 8 obtains the pixel area by counting the total number of pixels contained in the connected region; then, the module 8 multiplies the pixel area by the actual physical area corresponding to a single pixel (this conversion factor is pre-calibrated by the scanning scale and image resolution of the atomic force microscopy module 9) to obtain the actual physical projection area of ​​the abnormal morphology.

[0076] For perimeter measurement, the data processing module 8 uses a boundary tracking algorithm to accurately extract the contour pixel sequence of the connected region. By calculating and accumulating the straight-line distance between the center points of these adjacent contour pixels, the pixel length of the perimeter of the region is obtained, and it is also converted into the actual physical length according to the calibration coefficient.

[0077] The aspect ratio is calculated by the data processing module 8, which calculates the minimum bounding rectangle of the connected region. By traversing the set of boundary points of the region, the rectangle that can completely enclose the region and has the smallest area is found, and the length of the major axis and the length of the minor axis of the rectangle are determined. The aspect ratio is the ratio of the length of the major axis to the length of the minor axis.

[0078] The circularity assessment is performed by the data processing module 8, which evaluates the degree to which the shape of the region is close to a circle by the relationship between the area and the perimeter. The closer the numerical result is to 1, the closer the shape of the region is to an ideal circle.

[0079] For surface roughness calculation, the data processing module 8 extracts the height values ​​of all pixels within the original three-dimensional topography data (after background correction) corresponding to the connected region; then, it calculates the statistical distribution dispersion of these height values ​​relative to their average height, usually using an algorithm that calculates the standard deviation of all height values. The resulting value is the surface roughness characterizing the degree of surface undulation of the abnormal topography.

[0080] For profile contour curve extraction and analysis, the data processing module 8 defines a straight path through the centroid of the connected region or its longest axis direction. Along this path, the height value of each point is extracted from the three-dimensional topography data to obtain a profile contour curve. By analyzing this curve, the data processing module 8 can directly read the depth of the abnormal topography (the height difference between the highest and lowest points), evaluate the inclination angle of its sidewalls (by calculating the slope of the side segment of the curve), and determine its overall contour shape (such as V-shape, U-shape, or trapezoid).

[0081] The data processing module 8 compares and performs pattern recognition on these calculated feature values ​​with various defect feature templates pre-stored in the internal database. For example, a region with a high aspect ratio, low roundness, and a continuous concave profile is identified as a scratch; a region with a low aspect ratio, high roundness, and a convex profile is identified as particulate matter; and a region with a specific contour (such as an approximately trapezoidal concavity) may be identified as a structural distortion.

[0082] After completing defect classification and basic size measurement, the data processing module 8 accurately estimates the volume of the abnormal morphology using a volume integration algorithm. This process treats the defect as a three-dimensional entity and quantifies the volume by calculating the size of the space enclosed by the defect and the ideal reference plane. The specific steps are as follows: To define a reference plane, the data processing module 8 first selects a flat area around the abnormal morphology connected region, which is considered to represent the original surface of the undamaged grating sample 2. It then calculates the average value of the height of all pixels in this area and defines this average value as the height of the reference plane for volume calculation. For concave defects, such as scratches or pits, the volume of material missing below this reference plane needs to be calculated. For convex defects, such as contaminant particles, the volume of the object above this reference plane needs to be calculated.

[0083] Calculating the volume contribution of a single pixel: The data processing module 8 then traverses every pixel within the connected region of the anomalous morphology. For each pixel, the module performs the following operations: First, it obtains the true height value of the pixel after correcting the background tilt; then, it calculates the difference between the pixel height and the height of the reference plane, which represents the defect height deviation at the pixel location. Next, it multiplies this height deviation value by the actual physical area corresponding to the single pixel (this area value is determined by the scanning scale calibration of the atomic force microscopy module 9) to obtain the unit volume contributed by this specific pixel.

[0084] The total volume is obtained by integration and summation: Finally, the data processing module 8 sums up the unit volumes calculated for all pixels in the defect area. This summation process is to integrate the volume of the entire defect area and finally obtain the estimated total volume of the abnormal morphology.

[0085] Through the aforementioned volume integration algorithm, the data processing module 8 can transform two-dimensional topographic image data into accurate three-dimensional volume quantification data. This provides key quantitative indicators for a more in-depth assessment of the severity of defects, the quantity of contaminants or material loss, and their potential impact on the optical performance of the grating.

[0086] Data correlation and root cause analysis: Subsequently, the data processing module 8 performs data correlation analysis. It correlates and compares the precise defect features directly observed and quantified by the atomic force microscopy module 9 (e.g., a defect identified as a scratch with a depth measurement of 5 nanometers and a V-shaped profile) with the anomalous signals corresponding to the same location measured by optical scattering in the anomalous morphology identification step (e.g., the location shows a sharp change in duty cycle parameter in the scattering measurement parameter distribution map, and the diffraction spectrum fitting residual at this point is also significantly high). This correlation can confirm the specific physical origin of the anomalous signals detected by optical scattering measurement, thus providing key and intuitive evidence for in-depth judgment of the cause of defects (e.g., mechanical scratches during processing, laser damage leading to erosion, or particulate contamination in the usage environment).

[0087] Finally, the data processing module 8 systematically integrates all the above analysis results, including the two-dimensional distribution cloud map of global geometric parameters provided by optical scattering measurement, the high-resolution three-dimensional topographic map of abnormal morphology provided by atomic force microscopy module 9, the defect classification and quantitative statistics table (including type, quantity, and size distribution), and key data correlation analysis conclusions (e.g., the duty cycle mutation and high-fit residual at coordinates (X, Y), which were confirmed by AFM to be caused by a V-shaped scratch with a depth of 5 nm), and automatically generates a comprehensive inspection report that integrates global statistics, local fine observation, qualitative judgment, and quantitative data.

[0088] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A hybrid measurement system of extreme ultraviolet collecting mirror and grating, characterized in that, include: An optical scattering measurement module is used to acquire the diffraction spectrum of a grating and reconstruct the geometric parameters of the grating based on the diffraction spectrum; The data processing module is communicatively connected to the optical scattering measurement module, and is used to receive the geometric structure parameters, identify the abnormal morphology of the grating based on the geometric structure parameters, and output the location information of the abnormal morphology. An atomic force microscopy module, which is communicatively connected to the data processing module, is used to perform high-resolution morphology detection on the abnormal morphology based on the location information and acquire image data. A shared precision positioning platform module is connected to the optical scattering measurement module and the atomic force microscopy module, and is used to control the movement of the optical scattering measurement module and the atomic force microscopy module so that the atomic force microscopy module can be precisely moved to the location of the abnormal morphology; The data processing module is also used to generate a comprehensive detection report based on the geometric structure parameters reconstructed by the optical scattering measurement module and the image data acquired by the atomic force microscopy module.

2. The extreme ultraviolet collecting mirror grating hybrid measurement system according to claim 1, characterized in that, The optical scattering measurement module includes an illumination unit, a polarization control unit, a beam splitting unit, a focusing unit, and a spectral measurement unit; The illumination unit is used to generate a probe beam; the polarization control unit is connected to the illumination unit and is used to receive the probe beam and adjust its polarization state. The beam splitting unit is connected to the polarization control unit and is used to guide the probe beam after the polarization state is adjusted to the focusing unit; the focusing unit is used to focus the probe beam from the beam splitting unit onto the surface of the grating sample under test and collect the diffracted light generated by the grating sample under test. The spectral measurement unit is connected to the beam splitting unit and is used to receive the diffracted light collected by the focusing unit and transmitted through the beam splitting unit, and to collect the light intensity signals of multiple diffraction orders to form a diffraction spectrum.

3. The extreme ultraviolet collecting mirror grating hybrid measurement system according to claim 2, characterized in that, The data processing module identifies abnormal morphology by analyzing at least one of the following methods: the deviation between the geometric structure parameters and the design values, the local uniformity of the geometric structure parameters on the two-dimensional distribution map, or the fitting residual of the diffraction spectrum. When the deviation of the geometric structure parameters from the design value exceeds the preset tolerance range, or the distribution of the geometric structure parameters changes drastically in a local area, or the fitting residual of the diffraction spectrum is greater than a set threshold, the data processing module will mark the corresponding area as an abnormal morphology.

4. The extreme ultraviolet collecting mirror grating hybrid measurement system according to claim 3, characterized in that, The abnormal morphology includes geometric defects and laser damage defects introduced by the processing, or surface dust particle contamination introduced by the usage environment; the atomic force microscopy module is used to perform high-resolution three-dimensional morphology imaging on the abnormal morphology marked by the data processing module, so as to directly observe and quantify the specific morphological features of the processing defects or contamination.

5. The extreme ultraviolet collecting mirror grating hybrid measurement system according to claim 4, characterized in that, The atomic force microscopy module operates in tapping mode, scanning the sample surface with an ultra-sharp probe; The atomic force microscopy module receives the location information of the abnormal morphology from the data processing module, and scans the target area based on the location information to obtain three-dimensional morphology data of the abnormal morphology with sub-nanometer resolution.

6. The extreme ultraviolet collecting mirror grating hybrid measurement system according to claim 5, characterized in that, The shared precision positioning platform module includes a clamp for horizontally holding and fixing the extreme ultraviolet collecting mirror under test and a high-precision multi-axis motion module. The shared precision positioning platform module receives the abnormal morphology location information sent by the data processing module and drives the high-precision multi-axis motion module to control the relative position and angle changes of the extreme ultraviolet collecting mirror and the atomic force microscopy module, so as to precisely move the ultra-sharp probe of the atomic force microscopy module to the location of the abnormal morphology.

7. A hybrid measurement method using an extreme ultraviolet collecting mirror and grating, for implementing the system as described in any one of claims 1 to 6, characterized in that, include: The optical scattering measurement module scans the grating sample fixed on the shared precision positioning platform module along a predetermined path, and collects diffraction spectra at each measurement point. The data processing module then reconstructs the geometric parameters of the grating based on the diffraction spectra. The data processing module receives and analyzes the geometric structure parameters. By judging whether the geometric structure parameters exceed the preset tolerance range, or whether the distribution of the geometric structure parameters changes drastically in a local area, or whether the fitting residual of the diffraction spectrum exceeds the set threshold, abnormal morphology in the grating is identified and the location information of the abnormal morphology is generated. The data processing module sends the location information of the abnormal morphology to the shared precision positioning platform module and the atomic force microscopy module. The shared precision positioning platform module drives the atomic force microscopy module to move, so that its ultra-sharp probe can be precisely moved to the location of the abnormal morphology and perform high-resolution scanning on the abnormal morphology to obtain its three-dimensional morphology image data. The data processing module receives the three-dimensional morphology image data acquired by the atomic force microscopy module, and correlates and analyzes it with the geometric structure parameters reconstructed in the optical scattering measurement step to generate a comprehensive inspection report that includes global parameter distribution, abnormal morphology features and classification statistics.

8. The extreme ultraviolet collecting mirror grating hybrid measurement method according to claim 7, characterized in that, In the abnormal morphology identification step, the specific criteria used by the data processing module to identify abnormal morphologies include: The values ​​of the geometric parameters exceed the tolerance range preset according to the grating design specifications; or The geometric structure parameters exhibit localized abrupt changes across a preset threshold on the two-dimensional distribution map; or The residual obtained after fitting the measured data of the diffraction spectrum with the theoretical model is greater than the set threshold.

9. The extreme ultraviolet collecting mirror grating hybrid measurement method according to claim 8, characterized in that, In the atomic force microscopy (AFM) measurement step, the AFM module operates in tapping mode, with its ultra-sharp probe scanning the target area based on the location information of abnormal morphology. The acquisition of three-dimensional morphology image data specifically includes: The three-dimensional morphological features of anomalous morphology, including processing defects or surface contaminants, are directly observed and quantified by probe scanning.

10. The extreme ultraviolet collecting mirror grating hybrid measurement method according to claim 9, characterized in that, In the comprehensive report generation step, the data processing module performs the following specific analyses on the 3D topography image data: The defect type and geometric dimensions of the abnormal morphology are identified and quantified. The precise features of the defect type are correlated and compared with the corresponding scattering measurement abnormal signals in the abnormal morphology identification step, and finally a comprehensive inspection report is generated.

Images