Wafer detection method and device

Abstract

The invention discloses a wafer detection method and device, and relates to the technical field of wafer detection. The method comprises the steps of detecting whether an illumination light spot emitted to a movable workpiece table by an illumination system meets a preset optical constraint condition or not before wafer detection operation is executed by utilizing the illumination system; under the condition that the illumination light spot meets a preset optical constraint condition, wafer detection is carried out on the to-be-detected wafer, and a detection image containing speckle interference is obtained; based on an optical memory effect, carrying out self-correlation processing and Fourier transform on a detection image containing speckle interference, and extracting frequency domain information of the detection image; and processing the frequency domain information by using a phase recovery algorithm to obtain a wafer structure image of the to-be-detected wafer. According to the scheme, a real wafer structure image is reversely deduced from an actual image containing speckle interference, and the limitation that a front speckle elimination mode mainly depends on light source transformation or external modulation is broken through.

Description

Technical Field

[0001] The present application belongs to the field of wafer detection technology, and in particular relates to a wafer detection method and device. Background Art

[0002] In the field of wafer defect detection, laser illumination is widely used due to its high brightness and directionality, but its high coherence also easily produces speckle noise, which interferes with image quality and thus affects the accuracy and efficiency of detection results.

[0003] Currently, several speckle elimination methods used in practical applications have problems such as limited effectiveness, high cost, or increased system complexity, making it difficult to strike a balance between performance and practicality. Summary of the Invention

[0004] The embodiments of the present application provide a wafer inspection method and apparatus that can significantly improve the accuracy of defect detection.

[0005] A first aspect of an embodiment of the present application provides a wafer inspection method, the method comprising:

[0006] Before performing a wafer inspection operation using the illumination system, detecting whether the illumination spot emitted by the illumination system onto the movable workpiece stage meets a preset optical constraint condition, where the movable workpiece stage is used to carry the wafer to be inspected;

[0007] When the illumination spot meets the preset optical constraint conditions, wafer inspection is performed on the wafer to be inspected to obtain an inspection image containing speckle interference;

[0008] Based on the optical memory effect, the detection image containing speckle interference is subjected to autocorrelation processing and Fourier transform to extract the frequency domain information corresponding to the detection image containing speckle interference;

[0009] The frequency domain information is processed using a phase recovery algorithm to obtain a wafer structure image of the wafer to be inspected.

[0010] According to a second aspect of an embodiment of the present application, a wafer inspection device is provided, the device comprising:

[0011] A pre-detection module is used to detect whether the illumination spot emitted by the illumination system onto the movable workpiece stage meets the preset optical constraint conditions before performing the wafer detection operation using the illumination system. The movable workpiece stage is used to carry the wafer to be detected;

[0012] The detection module is used to perform wafer detection on the wafer to be detected when the illumination spot meets the preset optical constraint conditions, and obtain a detection image containing speckle interference;

[0013] a processing module for performing autocorrelation processing and Fourier transform on the detection image containing speckle interference based on the optical memory effect, and extracting frequency domain information corresponding to the detection image containing speckle interference;

[0014] The conversion module is used to process the frequency domain information using a phase recovery algorithm to obtain a wafer structure image of the wafer to be inspected.

[0015] According to a third aspect of an embodiment of the present application, an electronic device is provided, which includes: a memory and a program or instruction stored in the memory and runnable on a processor, wherein when the program or instruction is executed by the processor, a wafer detection method as provided in any one of the above-mentioned embodiments of the present application is implemented.

[0016] According to a fourth aspect of the embodiments of the present application, a readable storage medium is provided, on which a program or instruction is stored. When the program or instruction is executed by a processor, a wafer detection method as provided in any one of the above-mentioned embodiments of the present application is implemented.

[0017] According to a fifth aspect of the embodiments of the present application, a computer program product is provided. When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes the wafer detection method provided in any one of the above-mentioned embodiments of the present application.

[0018] The wafer inspection method provided in this application optically evaluates the light spot formed by the illumination system before inspection to determine whether it meets the preset optical constraints, thereby ensuring that the illumination system has the foundation to achieve the optical memory effect. Under this premise, wafer inspection is performed to obtain the actual image. Considering that the image may be superimposed with speckle interference, the image is further subjected to autocorrelation analysis and Fourier transform based on the optical memory effect to extract stable frequency domain amplitude information. The phase distribution of the image is estimated using a phase recovery algorithm to construct a true, speckle-free wafer structure image. This solution significantly improves image clarity and defect presentation accuracy without compromising the advantages of high laser coherence. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0020] Figure 1 1 is a flow chart of a wafer inspection method provided in one embodiment of the present application;

[0021] Figure 2 is a schematic diagram of speckle imaging provided by an embodiment of the present application;

[0022] Figure 3 This is a schematic diagram of the effect of compressing the short axis of the illumination spot using a cylindrical mirror provided in one embodiment of the present application;

[0023] Figure 4 (a) and (b) are schematic diagrams showing the effect of a homogenizer providing a method of homogenizing the intensity of the long-axis beam of an illumination spot according to an embodiment of the present application;

[0024] Figure 5 (a) and (b) are schematic diagrams of the speckle elimination effect provided by an embodiment of the present application;

[0025] Figure 6 This is a schematic structural diagram of a wafer inspection device provided by one embodiment of the present application;

[0026] Figure 7 This is a schematic diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0027] The features and exemplary embodiments of various aspects of the present application will be described in detail below. In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, rather than to limit the present application. For those skilled in the art, the present application can be implemented without the need for some of these specific details. The following description of the embodiments is merely to provide a better understanding of the present application by illustrating the examples of the present application.

[0028] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, the elements defined by the phrase "comprising..." do not exclude the presence of other identical elements in the process, method, article, or device comprising the elements.

[0029] It should be noted that the acquisition, storage, use, and processing of data in the technical solution of this application are in compliance with the relevant provisions of national laws and regulations. In the embodiments of this application, certain software, components, models, and other existing solutions in the industry may be mentioned. These should be considered as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of this application, but it does not mean that the applicant has or will necessarily use such solutions.

[0030] First, the terms involved in one or more embodiments of the present application are explained.

[0031] Wafers, silicon slices processed by specific processing techniques, are the basic materials for semiconductor manufacturing.

[0032] Speckle noise, a common noise phenomenon in optical imaging, is generated when coherent light (such as laser light) illuminates a rough surface or passes through a scattering medium. It appears as randomly distributed bright and dark spots in the image, reducing image resolution and contrast, and interfering with detection and analysis.

[0033] The optical memory effect is a physical phenomenon that occurs when waves propagate in a scattering medium. It refers to the situation where when the angle or position of the incident light changes slightly, the scattering pattern after transmission or reflection will undergo a corresponding linear or coherent translation, but the overall structure remains highly correlated.

[0034] In other words, by changing the direction or position of the incident light within a certain range, the resulting speckle image does not change completely randomly, but rather has a "memory"—there is a predictable correlation between the new image and the original image.

[0035] Lasers or high-coherence light sources are usually used when performing defect detection on wafers. When light is irradiated on the surface of the wafer, the defects on it will cause the incident light to scatter, and defect detection can be achieved by collecting the scattered light. However, in areas with high-density repetitive patterns on the wafer surface, due to the regular and strong periodicity of the surface structure, it is easier to form a stable interference pattern between the scattered light fields in different areas, making the speckle noise more obvious, resulting in drastic fluctuations in the intensity of the background signal, reducing the signal-to-noise ratio, and affecting the accuracy of subsequent defect identification. In addition, the presence of speckle requires the detection system to process more noise information, which increases the complexity and computational complexity of image processing, thereby extending the detection time and reducing detection efficiency.

[0036] In order to overcome the impact of speckle on wafer inspection, some current methods generally reduce the coherence of the laser to reduce speckle interference. The following are some common methods:

[0037] Breaking the spatial coherence means using devices such as scattering sheets to change the angle of the incident light, and coordinating with the rotation of the motor so that the laser irradiation angle changes over time, thereby breaking its spatial coherence and reducing the consistency of the speckle. However, this method has problems such as large energy loss and high system complexity, which seriously affect the efficiency of the production line. In addition, in some niche areas (such as optical inspection of patterned wafers), changes in the lighting angle may cause certain tiny defects to not be effectively illuminated, so a single, optimized lighting angle must usually be selected. In addition, if the motor speed is insufficient, resulting in insignificant changes in the speckle pattern during the camera imaging cycle, the suppression effect is also unsatisfactory.

[0038] Temporal coherence is broken by introducing a phase-modulation device and applying a voltage to introduce a time-varying optical path difference, thereby reducing the temporal coherence of the laser. This approach is effective in some imaging applications, but is limited by the device's frame rate (typically around 20kHz), making it difficult to match the high frame rate requirements of linear array cameras in this field, resulting in limited speckle reduction effectiveness.

[0039] Using multi-wavelength or incoherent light sources, such as multi-wavelength lasers or directly using incoherent light sources such as LEDs (Light Emitting Diodes), can weaken coherence at the source. Although this approach is feasible in some systems, in wafer inspection, due to the large differences in the surface material's response to different wavelengths, significant chromatic aberration can be introduced, posing a significant challenge to subsequent optical design. Chromatic aberration compensation requires adding lens groups, which not only increases system complexity but may not completely eliminate chromatic aberration, ultimately affecting image quality.

[0040] Introducing multiple polarization states, that is, superimposing multiple polarization states, can reduce speckle contrast. However, any polarization state can be decomposed into two basic components, S and P. Therefore, the theoretical suppression limit of this method is that the speckle contrast is reduced to 1 / √2 of the original value, which has a clear performance bottleneck.

[0041] In summary, existing speckle suppression or elimination schemes mainly avoid speckle generation by reducing the coherence of the laser. However, in wafer defect detection scenarios, due to the high requirements for imaging accuracy, lighting stability and system complexity, such methods have great limitations in terms of performance, implementation cost and applicability. It is difficult to achieve effective speckle suppression without weakening the inherent advantages of laser lighting (such as high brightness and high directionality).

[0042] To address the above-mentioned technical issues, the inventors of this application have taken a different approach, proposing a wafer inspection method and apparatus based on the perspective of inferring the true image structure from speckle images. This wafer inspection method does not rely on modifications to the light source itself or external modulation devices. Instead, it uses the acquired actual inspection image containing speckle interference to construct a true and accurate image of the wafer structure through optical memory effects, autocorrelation analysis, frequency domain processing, and a phase recovery algorithm. Compared to the original inspection image, this image has reduced or eliminated speckle interference. While retaining the advantages of laser illumination imaging, it effectively reduces the interference of speckle on inspection accuracy, improving image quality and the reliability of defect identification.

[0043] For example, the wafer detection method provided in the embodiment of the present application can be applied to the production line of a semiconductor manufacturing enterprise to perform defect detection on semiconductor devices generated during the production process. In practical applications, an image of the wafer to be inspected is first obtained, and the image may contain speckle interference generated by laser illumination. Subsequently, an autocorrelation operation is performed on the image and a Fourier transform is performed to extract the frequency domain features of the image. Based on the frequency domain information, a phase recovery algorithm is used to construct a wafer structure image without speckle interference, thereby obtaining a high-quality imaging result that truly reflects the wafer structure. Based on this image, defect identification, classification and quantification processing can be further carried out to improve the accuracy and reliability of the detection results, and provide effective support for subsequent process optimization and quality control.

[0044] It should be noted that the application scenarios described in the above embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Persons skilled in the art will appreciate that, as new application scenarios emerge, the technical solutions provided by the embodiments of the present application are equally applicable to similar technical problems. The wafer inspection method provided in the embodiments of the present application can be applied to various application scenarios requiring noise removal from wafer images.

[0045] The wafer inspection method provided by the embodiment of the present application is introduced below. In practical applications, the execution subject of the wafer inspection method of the embodiment of the present application can be an electronic device.

[0046] First, the core design idea of ​​this application method is briefly described.

[0047] As previously mentioned, the optical memory effect refers to the phenomenon where, under certain conditions, the speckle pattern recorded by a scattering medium changes in its specific form when the incident angle of the illumination beam changes. However, the overall structure of its intensity distribution remains highly correlated, meaning that the speckle pattern exhibits a stable "memory." This effect demonstrates that, while the speckle image appears chaotic and disorganized, it actually has a traceable relationship to the illumination direction and the structural characteristics of the illuminated object.

[0048] Figure 2 is an example of speckle imaging. Figure 2 (a) with Figure 2 The top of (b) is the corresponding speckle intensity distribution curve, the middle is the speckle image, and the bottom is the schematic line of the illumination direction. Figure 2 (a) is lighting from right to left, Figure 2 (b) Illumination is from left to right. It can be observed that although the specific speckle pattern varies slightly due to different illumination directions, the overall intensity distribution curve trend remains highly consistent, indicating that the image contains structural invariance. In other words, for the same periodic structure on the wafer, the speckle distribution exhibits a regular and repeatable response.

[0049] Based on the above rules, this application proposes a reverse reconstruction method: by establishing a mathematical mapping relationship between the detection image containing speckle interference and the actual wafer structure image, and combining the characteristics of the optical memory effect, without changing the illumination system structure, the wafer structure image without speckle interference can be inferred.

[0050] Figure 1 FIG. 1 shows a flow chart of a wafer inspection method provided by an embodiment of the present application. Figure 1 As shown, the method includes steps S110 to S140.

[0051] S110 , before performing a wafer inspection operation using the illumination system, inspect whether the illumination spot emitted by the illumination system onto the movable workpiece stage meets a preset optical constraint condition.

[0052] The illumination system is an optical device used to provide illumination light source for wafer surface imaging.

[0053] Typically, an illumination system includes a light source and its associated emission control components, which are used to control the intensity, divergence angle, and illumination range of the light beam and may include, for example, a collimator, beam expander, and lens assembly.

[0054] Wafers to be inspected refer to semiconductor wafers that are in the manufacturing process and require surface defect inspection.

[0055] The movable workpiece stage is used to carry the wafer to be inspected and move to the inspection area according to a preset path or scanning trajectory to cooperate with the image acquisition process.

[0056] During actual inspection, the illumination beam emitted by the lighting system is modulated through an optical path and then illuminates a predetermined position on the movable workpiece stage, where the wafer to be inspected is located, forming an illumination spot. To ensure the accuracy and stability of subsequent image processing algorithms, it is necessary to determine whether this illumination spot meets the preset optical constraints before executing the inspection operation.

[0057] The preset optical constraint condition refers to the parameter range restrictions set on the spatial geometric parameters and energy distribution characteristics of the illumination spot in order to make the image reconstruction method based on the optical memory effect valid in the calculation.

[0058] It should be understood that only when the illumination spot meets the aforementioned optical constraints can the speckle characteristics in subsequent images be stable and controllable, thus ensuring the applicability of frequency domain analysis and phase recovery algorithms based on the optical memory effect. If the illumination spot deviates from the constraints, it may lead to unstable spot morphology, thus affecting the consistency of the image spectrum, undermining the basis for autocorrelation processing, and reducing the effectiveness and accuracy of speckle reduction.

[0059] S120 : When the illumination spot meets the preset optical constraint condition, perform wafer inspection on the wafer to be inspected, and obtain an inspection image containing speckle interference.

[0060] The detection image refers to the image data formed by the imaging system collecting the reflected or scattered light signals on the wafer surface under the illumination of the illumination beam.

[0061] It should be understood that the detection image is the original, unprocessed direct imaging result, which contains not only the structural information and optical properties of the wafer surface, but also speckle interference.

[0062] Once the illumination spot is confirmed to meet the preset optical constraints, the wafer inspection operation begins. In one implementation, the illumination system continuously projects an illumination beam onto the wafer surface, while the imaging system simultaneously collects the reflected light signal within the illumination area to generate image data of the wafer to be inspected. This image data is the inspection image.

[0063] S130 , based on the optical memory effect, perform autocorrelation processing and Fourier transform on the detection image to extract frequency domain information corresponding to the detection image containing speckle interference.

[0064] After completing wafer inspection and obtaining the inspection image containing speckle interference, in order to extract effective feature information that can be used to reconstruct the real wafer structure, the optical memory effect is used to carry out frequency domain analysis operations.

[0065] In this step, autocorrelation processing is first performed on the inspection image. Autocorrelation is a fundamental operation in signal processing, used to measure the similarity between a signal (or image) and itself at different position offsets (or delays). The autocorrelation operation effectively enhances periodic structures in the image and weakens highly random speckle noise, thereby highlighting the spatial structural characteristics of the wafer pattern.

[0066] Then, a Fourier transform is performed. The Fourier transform is used to convert the signal from the spatial domain to the frequency domain, so that various types of information in the image are displayed in the form of frequency components.

[0067] Frequency domain information refers to the frequency domain amplitude information extracted from the detection image through Fourier transform.

[0068] Spectral analysis reveals characteristic frequency regions corresponding to the true structure, clearly distinguishable from the speckle background. This frequency domain information not only reveals the periodicity and symmetry of the wafer pattern but also guides subsequent phase recovery and image reconstruction steps, enabling the reverse restoration of the speckle-free image.

[0069] S140 , using a phase recovery algorithm to process the frequency domain information to obtain a wafer structure image of the wafer to be inspected.

[0070] Since speckle interference in the direct imaging process mainly affects the phase information, and the frequency domain amplitude can often more stably preserve the spatial frequency characteristics of the detected object, the true structure of the original image can be inferred by restoring the phase information.

[0071] Wafer structure image refers to the image result that can truly reflect the pattern structure, defect characteristics and other information on the wafer surface after reducing or removing speckle interference.

[0072] Phase recovery algorithm refers to a technical method that estimates the phase information of the original image through certain mathematical optimization or iterative calculation methods when the Fourier amplitude of the image is known but its phase is unknown, thereby achieving image reconstruction.

[0073] Exemplarily, the phase recovery algorithm may be a common phase reconstruction method based on the Gerchberg-Saxton (GS) algorithm, the Hybrid Input-Output (HIO) algorithm, Wirtinger Flow, or the Ptychographic Iterative Engine (PIE).

[0074] In one implementation, step S140 may include the following steps.

[0075] S141. Estimate the phase distribution of the wafer structure image using a phase recovery algorithm based on the frequency domain amplitude information.

[0076] In step S130, after autocorrelation and Fourier transform processing, the frequency domain amplitude information of the detected image is obtained. However, the spectral information obtained by Fourier transform consists of two key components: the amplitude spectrum and the phase spectrum. The phase spectrum has a decisive influence on the structure and details of the image. However, because image acquisition equipment generally only records light intensity information, phase information is lost during detection and cannot be directly obtained.

[0077] To achieve image reconstruction, the missing phase information must be estimated through a phase recovery algorithm. Specifically, based on the frequency domain amplitude information and spatial domain constraints, an iterative optimization method is used to gradually converge to a phase distribution that matches the actual structure.

[0078] S142. Combine the frequency domain amplitude information and phase distribution, perform inverse Fourier transform, and obtain a wafer structure image.

[0079] After successfully recovering the phase information, it is combined with the known frequency-domain amplitude information to construct a complete complex spectrum. Subsequently, an inverse Fourier transform is performed to remap the frequency-domain data back to the spatial domain, resulting in a restored image of the wafer structure.

[0080] The two steps in this implementation method work together to complete the transition from "frequency domain information" to "visual image", solving the problem of being unable to directly observe the true structure of the wafer under speckle interference. By using phase recovery and inverse Fourier transform, the periodicity, symmetry and detailed features of the wafer pattern can be accurately restored even in the lack of traditional imaging clarity, significantly improving detection accuracy and analysis reliability.

[0081] The wafer inspection method provided in this application introduces a light spot quality assessment mechanism before inspection to detect whether the illumination spot emitted by the illumination system to the movable workpiece table meets the preset optical constraints, ensuring that the illumination system operates under the conditions of the optical memory effect, thereby improving the feasibility and stability of subsequent image reconstruction from the source. On this basis, the actual inspection image containing speckle interference is collected, and the image is subjected to autocorrelation processing and Fourier transform based on the principle of optical memory effect to extract stable frequency domain amplitude information. The phase distribution of the image is estimated by combining the phase recovery algorithm, and finally the wafer structure image without speckle interference is restored by inverse Fourier transform. This solution effectively removes speckle interference without relying on complex hardware modulation or reducing laser coherence, while retaining the high contrast and high sensitivity advantages of laser illumination imaging, significantly improving image quality and the accuracy of wafer defect recognition, and is suitable for high-precision and high-reliability semiconductor manufacturing inspection scenarios.

[0082] In one embodiment, step S130 may include the following steps.

[0083] S131. Based on the optical memory effect, a convolution imaging model of a detection image including speckle interference is constructed.

[0084] based on Figure 2 The memory characteristics shown indicate that the generation process of the speckle image satisfies the linear system theory, and can therefore be represented by mathematical modeling as a convolution imaging process. The formula of the convolution imaging model is as follows:

[0085] I=O·h; (1)

[0086] Wherein, I is the actual acquired detection image;

[0087] O is the wafer structure image to be restored;

[0088] h is the point spread function (PSF), which is used to describe the response characteristics of the optical system under given lighting conditions.

[0089] The model shows that the inspection image can be obtained by the convolution between the wafer structure image and the point spread function.

[0090] Considering that during actual wafer inspection, the wafer being inspected is scanned on a two-dimensional movable worktable in the XY plane, and the image data changes over time t, and the spatial characteristics of the illumination spot significantly influence the speckle distribution, the illumination spot can be simplified as an ellipse in modeling, with its minor axis dx along the x-direction and major axis dy along the y-direction. Based on this, the convolution imaging model can be further expressed as:

[0091] I(x(t),y(t),dx,dy)=O(x(t),y(t))*h(dx,dy); (2)

[0092] Among them, x(t) and y(t) are used to describe the periodic motion of the wafer in the x and y directions with time t;

[0093] dx and dy are the sizes of the elliptical spot in the X and Y directions, respectively, which will affect the spatial diffusion degree of the system point spread function h;

[0094] h(dx,dy) is the point spread function related to the illumination spot size, which determines the intensity distribution and diffusion range of the speckle during the imaging process.

[0095] It should be noted that the larger the spot size (i.e., the larger the dx and dy values), the more scattered wavelet sources are excited in the illuminated area, and the more complex the speckle pattern formed by interference. Within the millimeter range, increasing the spot area significantly increases the density of the speckle distribution in the image.

[0096] It should be noted that, when the illumination spot satisfies the preset optical constraints, the point spread function h(dx,dy) can be further simplified into an equivalent constant related to time.

[0097] Introduce preset optical constraints, including but not limited to any of the following:

[0098] The short axis width of the illumination spot is smaller than a preset threshold;

[0099] The beam intensity in the long axis direction of the illumination spot is a flat top distribution.

[0100] The preset threshold refers to a specific standard or limit value set when designing an optical system based on system performance requirements and imaging accuracy, which is used to control and limit the maximum allowable range of the short-axis width of the illumination spot.

[0101] For example, the preset threshold may be 100 micrometers, which means that the short axis width of the illumination spot cannot exceed 100 micrometers.

[0102] Flat-top distribution means that within a specific range, the intensity distribution of a light beam or signal presents a nearly constant "flat-top" shape, that is, the light intensity changes very little or remains consistent within this range, similar to a horizontal rectangular waveform.

[0103] It should be understood that, on the one hand, compressing the short-axis width of the illumination spot effectively reduces the interference path difference, enhancing the spatial consistency and stability of the speckle image. On the other hand, by homogenizing the intensity distribution along the long axis of the spot into a flat-top shape, local nonlinear interference caused by energy unevenness can be significantly reduced, improving the uniformity and reconfigurability of the imaging. Taken together, these two constraints help simplify the behavior of the system's point spread function, making the convolution model more stable and providing high-quality input for subsequent image processing.

[0104] When the above constraints are met, the geometry and energy distribution of the illumination spot tend to be stable, and the speckle formation mechanism is also simplified, so that the point spread function h can be approximately regarded as an equivalent constant h(t) that changes only with time. At this point, the convolution model can be further simplified as:

[0105] I(x(t),y(t),dx,dy)=O(x(t),y(t))*h; (3)

[0106] S132. Perform autocorrelation processing on the convolution imaging model to obtain an autocorrelation result.

[0107] Based on the convolution imaging model constructed in S131, an autocorrelation operation is performed on the parameter-containing detection image I(x(t), y(t), dx, dy). Since the detection image can be represented as the convolution relationship between the wafer structure image and the point spread function, according to the mathematical relationship between convolution and autocorrelation, we have:

[0108]

[0109] To further simplify the model, system constraints are introduced. It should be understood that in the actual application of wafer defect detection, the inspection sample is usually a wafer surface pattern with a periodic structure. The movement of its image in two-dimensional space (x, y) over time t is repetitive and symmetrical, showing the characteristic of "periodic movement" of the image position over time. However, the periodic movement of the image position over time will be offset by each other in the autocorrelation operation, which can be simplified as follows:

[0110]

[0111] Furthermore, in one implementation, considering that the laser beam itself has a narrow linewidth characteristic, the constructed point spread function has a stable spatial distribution when meeting the preset optical constraints, and the autocorrelation result has a sharp concentrated feature. It is approximately the unit shock function δ, that is:

[0112]

[0113] Based on this, the formula can be further simplified as:

[0114]

[0115] In other words, under these conditions, the autocorrelation results of the inspection image can be directly regarded as the autocorrelation results of the wafer structure image. This simplification significantly reduces the dependence on the original system modeling and lays a solid foundation for subsequent frequency domain processing and phase recovery operations.

[0116] S133. Perform Fourier transform on the autocorrelation result to extract frequency domain information.

[0117] This embodiment effectively improves the ability to suppress speckle interference and image recovery accuracy during wafer inspection by constructing a convolution imaging model and combining it with autocorrelation processing and frequency domain analysis. First, a simplified convolution imaging model is established based on the optical memory effect, reducing the complexity of the optical system. Second, autocorrelation processing is used to eliminate the impact of position changes during the wafer's periodic scanning process, making the speckle characteristics in the image more stable. Finally, frequency domain information is extracted through Fourier transform, further enhancing the image's spectral characteristics and providing precise frequency domain support, laying a solid foundation for subsequent defect detection and phase recovery. Overall, this process significantly improves image quality and reconstructability, effectively reducing noise and nonlinear interference.

[0118] In some embodiments, the preset optical constraints may include at least the following:

[0119] The minor axis width of the illumination spot in the first direction is smaller than a preset threshold; and

[0120] The intensity of the light beam along the long axis direction of the illumination spot in the second direction is a flat top distribution.

[0121] The first direction and the second direction are perpendicular to each other.

[0122] This embodiment defines preset optical constraints, requiring the illumination system's light spot to meet different conditions in different directions. The spot size requirement in the first direction is used to reduce path differences in the speckle image, thereby improving speckle stability and image clarity. The requirement for a flat-top intensity distribution in the second direction improves illumination uniformity, avoids localized overexposure or underexposure, and reduces the risk of nonlinear interference. Overall, these optical constraints are designed to provide more stable and accurate baseline data for subsequent speckle reduction.

[0123] In one implementation, when the illumination spot is an elliptical spot, the first direction is the direction of the line where the short axis of the elliptical spot is located, and the second direction is the direction of the line where the long axis of the elliptical spot is located.

[0124] In one embodiment, the method of the present application further includes: optimizing the lighting system when the lighting spot does not meet the preset optical constraint conditions.

[0125] In one implementation, the method for optimizing the lighting system may include at least one of the following:

[0126] At least one cylindrical mirror is provided in the illumination system to compress the size of the illumination spot along a first direction.

[0127] In one example, when the illumination spot is an elliptical spot, the method is: setting at least one cylindrical mirror in the illumination system to compress the short-axis width of the illumination spot.

[0128] The cylindrical mirror changes the degree of refraction of the light beam in one direction, making the light spot more compact in the short axis direction, thereby reducing the interference path difference and improving the imaging stability and clarity of the speckle image.

[0129] Figure 3 A schematic diagram shows the effect of a cylindrical mirror on the short-axis compression of the illumination spot, showing that after passing through the cylindrical mirror, the short-axis size of the spot is significantly reduced, the focusing effect is more precise, and the contrast and stability of the image are enhanced.

[0130] A beam homogenizer is provided in the illumination system to homogenize the beam intensity of the illumination spot along the second direction.

[0131] Illustratively, the beam homogenizer may be a lens array, an optical shaper, or an optical waveguide, etc., and this application does not limit its type.

[0132] In one example, when the illumination spot is an elliptical spot, the method is: providing a beam homogenizer in the illumination system to homogenize the beam intensity in the long axis direction of the illumination spot.

[0133] Figure 4 The schematic diagram shows the effect of the homogenizer on the intensity of the long axis beam of the illumination spot. The homogenizer adjusts the light intensity distribution so that the light intensity of the light spot in the long axis direction changes from the original Gaussian distribution (such as Figure 4 (a)) into a flat-top distribution (as shown in Figure 4 (as shown in (b)).

[0134] In this embodiment, the lighting system is optimized when the illumination spot does not meet the preset optical constraints, which can ensure the stability of the geometric shape and energy distribution of the light spot, thereby improving the controllability and consistency of the imaging process.

[0135] Specifically, through the above two optimization measures, the performance of the illumination system has been significantly improved. First, the cylindrical mirror compresses the short-axis width of the illumination spot, reducing the path difference in the speckle image, thereby improving the stability of the speckle and imaging clarity. Second, the beam homogenizer adjusts the light intensity distribution along the long axis of the spot to a flat-top type, avoiding the local overexposure and underexposure problems caused by the Gaussian distribution, ensuring the uniformity of the light intensity of the entire image, and reducing the nonlinear interference that may occur during the imaging process. In summary, these optimizations not only improve image quality and consistency, but also provide more stable and accurate data input for subsequent image restoration and defect detection, thereby enhancing the accuracy and robustness of wafer inspection.

[0136] Figure 5 A schematic diagram of the speckle elimination effect after applying the method of the present application is shown. Figure 5 (a) is the image before speckle removal (i.e., the detection image), Figure 5 (b) is the image after speckle removal (i.e., wafer structure image). As can be seen, after processing by this method, the speckle noise in the image is significantly reduced and the image quality is significantly improved.

[0137] Based on the wafer detection method, the present application also provides a specific embodiment of the image processing device.

[0138] Figure 6 A schematic structural diagram of a wafer inspection device provided in an embodiment of the present application is shown.

[0139] like Figure 6 As shown, the wafer inspection apparatus 1000 may include the following modules.

[0140] The pre-detection module 1001 is used to detect whether the illumination spot emitted by the illumination system onto the movable workpiece stage meets the preset optical constraint conditions before performing the wafer detection operation using the illumination system.

[0141] The movable workpiece stage is used to carry the wafer to be inspected.

[0142] The detection module 1002 is configured to perform wafer detection on the wafer to be detected and obtain a detection image containing speckle interference when the illumination spot meets preset optical constraints.

[0143] The detection image contains speckle interference.

[0144] The processing module 1003 is configured to perform autocorrelation processing and Fourier transform on the detection image containing speckle interference based on the optical memory effect, and extract frequency domain information corresponding to the detection image containing speckle interference.

[0145] The conversion module is used to process the frequency domain information using a phase recovery algorithm to obtain a wafer structure image of the wafer to be inspected.

[0146] Based on the wafer detection method, the present application also provides specific embodiments of the electronic device.

[0147] Figure 7 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application is shown.

[0148] The electronic device may include a processor 7001 and a memory 7002 storing computer program instructions.

[0149] Specifically, the processor 7001 may include a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0150] The memory 7002 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 7002 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 7002 may include removable or non-removable (or fixed) media. Where appropriate, the memory 7002 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 7002 is a non-volatile solid-state memory.

[0151] The processor 7001 implements any one of the wafer detection methods in the above embodiments by reading and executing computer program instructions stored in the memory 7002.

[0152] In one example, the electronic device may further include a communication interface 7003 and a bus 7004. Figure 5 As shown, the processor 7001, the memory 7002, and the communication interface 7003 are connected via a bus 7004 and communicate with each other.

[0153] The communication interface 7003 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.

[0154] Bus 7004 comprises hardware, software or both, and the parts of this electronic device are coupled to each other.For example, and not limitation, bus can comprise accelerated graphics port (AGP) or other graphics bus, enhanced industry standard architecture (EISA) bus, front side bus (FSB), hypertransport (HT) interconnection, industry standard architecture (ISA) bus, infinite bandwidth interconnection, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnection (PCI) bus, PCI-Express (PCI-X) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus or other suitable bus or two or more of these combinations.In suitable cases, bus 7004 can comprise one or more buses.Although the present application embodiment describes and shows specific bus, the application considers any suitable bus or interconnection.

[0155] In addition, in conjunction with the wafer inspection method in the above embodiments, the present application can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when the computer program instructions are executed by a processor, any one of the wafer inspection methods in the above embodiments is implemented.

[0156] In addition, in combination with the wafer detection method in the above-mentioned embodiments, the embodiments of the present application may provide a computer program product for implementation. When the instructions in the computer program product are executed by the processor of an electronic device, the electronic device executes the wafer detection method provided by any aspect of the above-mentioned embodiments of the present application.

[0157] It should be understood that the present application is not limited to the specific configurations and processes described above and illustrated in the figures. For the sake of brevity, a detailed description of known methods is omitted here. In the above embodiments, several specific steps are described and illustrated as examples. However, the method process of the present application is not limited to the specific steps described and illustrated. Those skilled in the art can make various changes, modifications, and additions, or change the order of the steps after understanding the spirit of the present application.

[0158] The functional blocks shown in the above-described block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of the present application are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier wave. "Machine-readable medium" can include any medium that can store or transmit information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROMs, flash memories, erasable ROMs (EROMs), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.

[0159] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps. In other words, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0160] Aspects of the present disclosure have been described above with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable data processing device to produce a machine so that these instructions executed by the processor of the computer or other programmable data processing device enable the implementation of the function / action specified in one or more boxes of the flowchart and / or block diagram. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor or a field programmable logic circuit. It is also understood that each box in the block diagram and / or flowchart and the combination of the boxes in the block diagram and / or flowchart can also be implemented by dedicated hardware that performs the specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0161] The above description is only a specific embodiment of the present application. Those skilled in the art will clearly understand that for the convenience and brevity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the scope of protection of the present application.

Claims

1. A wafer detection method, characterized in that: include: Before performing a wafer inspection operation using an illumination system, detecting whether an illumination spot emitted by the illumination system onto a movable workpiece stage satisfies a preset optical constraint condition, wherein the movable workpiece stage is used to carry the wafer to be inspected; When the illumination spot satisfies the preset optical constraint condition, performing wafer inspection on the wafer to be inspected to obtain an inspection image containing speckle interference; Based on the optical memory effect, performing autocorrelation processing and Fourier transform on the detection image containing speckle interference to extract frequency domain information corresponding to the detection image containing speckle interference; The frequency domain information is processed using a phase recovery algorithm to obtain a wafer structure image of the wafer to be inspected.

2. The method according to claim 1, characterized in that The preset optical constraints include: The size of the illumination spot in the first direction is smaller than a preset threshold; and The beam intensity of the illumination light spot in the second direction is a flat-top distribution, and the first direction and the second direction are perpendicular to each other.

3. The method according to claim 2, characterized in that The method further comprises: When the illumination spot does not satisfy the preset optical constraint condition, performing at least one of the following operations: At least one cylindrical mirror is provided in the illumination system to compress the size of the illumination spot along the first direction; A beam homogenizer is provided in the illumination system to homogenize the beam intensity of the illumination spot along the second direction.

4. The method according to claim 2, characterized in that In the case that the illumination spot is an elliptical spot, the first direction is the direction of the line where the short axis of the elliptical spot is located, and the second direction is the direction of the line where the long axis of the elliptical spot is located.

5. The method according to any one of claims 1 to 4, characterized in that The step of performing autocorrelation processing and Fourier transform on the detection image containing speckle interference based on the optical memory effect to extract frequency domain information corresponding to the detection image containing speckle interference includes: Based on the optical memory effect, a convolution imaging model of the detection image containing speckle interference is constructed, wherein the wafer structure image and a point spread function are convolved in the convolution imaging model to generate the detection image containing speckle interference; Performing autocorrelation processing on the convolution imaging model to obtain an autocorrelation result; Performing Fourier transform on the autocorrelation result to extract the frequency domain information.

6. The method according to claim 5, characterized in that The preset optical constraint condition includes that the illumination beam in the illumination system is a laser beam; The performing autocorrelation processing on the convolution imaging model to obtain an autocorrelation result includes: Based on the narrow linewidth characteristic of the laser beam, the autocorrelation result of the point spread function is simplified to a unit impulse function, so that the autocorrelation result of the detection image containing speckle interference is equivalent to the autocorrelation result of the wafer structure image.

7. The method according to any one of claims 1 to 4, characterized in that The frequency domain information is frequency domain amplitude information; The method of processing the frequency domain information by using a phase recovery algorithm to obtain a wafer structure image of the wafer to be inspected includes: estimating the phase distribution of the wafer structure image using the phase recovery algorithm according to the frequency domain amplitude information; Combining the frequency domain amplitude information and the phase distribution, an inverse Fourier transform is performed to obtain the wafer structure image.

8. A wafer inspection device, characterized in that: include: A pre-detection module is used to detect whether the illumination spot emitted by the illumination system onto the movable workpiece stage for carrying the wafer to be inspected meets preset optical constraints before performing a wafer inspection operation using the illumination system; a detection module, configured to perform wafer detection on the wafer to be detected and obtain a detection image containing speckle interference when the illumination spot satisfies the preset optical constraint condition; a processing module, configured to perform autocorrelation processing and Fourier transform on the detection image containing speckle interference based on an optical memory effect, and extract frequency domain information corresponding to the detection image containing speckle interference; The conversion module is used to process the frequency domain information using a phase recovery algorithm to obtain a wafer structure image of the wafer to be inspected.

9. An electronic device, characterized in that: The device comprises: a processor and a memory storing computer program instructions; the processor reads and executes the computer program instructions to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer storage medium stores computer program instructions, which, when executed by a processor, implement the method according to any one of claims 1 to 7.

Images