Optical system and method

Abstract

According to the optical system and method provided by the invention, the illumination numerical aperture is adjusted according to the region of interest on the edge of the to-be-detected wafer through the illumination numerical aperture adjusting device arranged in the light path where the incident light irradiates the edge of the to-be-detected wafer, so that the resolution is adjusted. An illumination numerical aperture is controlled through an aperture device, a fixed numerical aperture in a traditional optical system is changed into a dynamically optimized variable by using a physical principle that the illumination numerical aperture mainly determines a limit resolution, complex performance switching is realized through simple mechanical adjustment, high-resolution detection can be realized according to detection requirements, and the detection precision is improved. And the application range and adaptability of the optical system are enhanced.

Description

[0001] This application is a divisional application of Chinese invention patent application No. 202511599209.9, filed on November 4, 2025, entitled "An optical system and method for detecting wafer edge defects". Technical Field

[0002] This application relates to the field of semiconductor detection technology, and in particular to an optical system and method. Background Technology

[0003] In semiconductor manufacturing, wafer morphology inspection is a crucial step, directly impacting wafer quality and performance. The wafer edge is a non-planar, curved structure. A projection optical system with its optical axis parallel to the wafer surface, incorporating wafer rotation, can detect defects at different locations on the wafer's side. Furthermore, this optical system does not affect wafer loading or unloading, making it an excellent solution for detecting wafer side defects.

[0004] Currently, in wafer edge defect detection, detection accuracy (i.e., the smallest detectable defect size) and detection range are both important parameters. However, there is a trade-off between the two, making it impossible to achieve both high detection accuracy and a large detection range simultaneously. Summary of the Invention

[0005] In view of this, the present invention provides an optical system and method that can simultaneously achieve high detection accuracy and a large detection range.

[0006] To solve the above problems, this application adopts the following technical solution: One of the objectives of this application is to provide an optical system, including an illumination unit and an imaging unit; The incident light provided by the illumination unit illuminates the edge of the wafer to be inspected and generates signal light through orthographic projection. The imaging unit receives the signal light incident from the edge of the wafer and performs imaging; wherein: An illumination numerical aperture adjustment device is provided in the optical path through which the incident light illuminates the edge of the wafer to be inspected. The illumination numerical aperture adjustment device adjusts the illumination numerical aperture according to the region of interest on the edge of the wafer to be inspected, so as to adjust the resolution.

[0007] In some embodiments, an illumination numerical aperture adjustment device is provided in the optical path through which the incident light illuminates the edge of the wafer to be inspected. The illumination numerical aperture adjustment device adjusts the illumination numerical aperture according to the region of interest on the edge of the wafer to be inspected, so as to adjust the resolution.

[0008] In some embodiments, the region of interest includes at least one of a flat area, a notch, an edge bevel, an edge top, and an edge bottom. A corresponding illumination numerical aperture is pre-set according to each type of region of interest, and the illumination numerical aperture adjustment device adjusts according to the corresponding illumination numerical aperture to achieve resolution adjustment.

[0009] In some embodiments, the illumination numerical aperture adjustment device is an illumination variable aperture, the diameter of which is adjustable. By adjusting the diameter of the illumination variable aperture, the angle at which the incident light illuminates the edge of the wafer to be inspected is adjusted, thereby adjusting the illumination numerical aperture.

[0010] In some embodiments, the diameter of the illumination variable aperture varies from 1 to 6.6 mm, corresponding to an illumination numerical aperture of 0.0075 to 0.05 and a resolution of 44.7 to 6.71 μm. The relationship between the illumination numerical aperture and the resolution satisfies the following formula: d = 0.61 * λ² / NA², where d is the resolution, λ² is the wavelength of the incident light, and NA² is the illumination numerical aperture.

[0011] In some embodiments, for the flat area and the notch, the diameter of the illumination variable aperture is adjusted to make its illumination numerical aperture 0.0075~0.02, at which point the resolution is 44.7~16.78μm; For the edge slope, by adjusting the diameter of the illumination variable aperture to make its illumination numerical aperture 0.01~0.03, the resolution is 33.55~11.18μm; For the top and bottom edges, the diameter of the illumination variable aperture is adjusted to make its illumination numerical aperture 0.02~0.05, at which point the resolution is 16.78~6.71μm.

[0012] In some embodiments, the lighting unit includes an LED light source, a first lens group, a light diffuser, a second lens group, and a third lens group. The first lens group includes a first lens, a second lens, and a third lens. The second lens group includes a fourth lens, a fifth lens, and a sixth lens. The third lens group includes a seventh lens, an eighth lens, and a ninth lens. The light beam emitted from the LED light source passes sequentially through the first lens, the second lens, the third lens, the light homogenizer, the fourth lens, the fifth lens, the sixth lens, the illumination numerical aperture adjustment device, the seventh lens, the eighth lens, and the ninth lens, and the resulting incident light illuminates the edge of the wafer to be inspected.

[0013] In some embodiments, the first lens group, the second lens group, and the third lens group constitute a Kohler illumination system, wherein the incident light forms a uniform light spot at the edge of the wafer to be inspected.

[0014] The second objective of this application is to provide an optical method comprising the following steps: Incident light illuminates the edge of the wafer to be tested and generates signal light; Receive the signal light reflected from the edge of the wafer and perform imaging; wherein: An illumination numerical aperture adjustment device is installed in the optical path where the incident light illuminates the edge of the wafer to be inspected. The illumination numerical aperture is adjusted according to the region of interest on the edge of the wafer to be inspected in order to adjust the resolution.

[0015] In some embodiments, the region of interest includes at least one of a flat area, a notch, an edge bevel, an edge top, and an edge bottom. A corresponding illumination numerical aperture is pre-set according to each type of region of interest, and the illumination numerical aperture adjustment device adjusts according to the corresponding illumination numerical aperture to achieve resolution adjustment.

[0016] In some embodiments, the illumination numerical aperture adjustment device is an illumination variable aperture, the diameter of which is adjustable. By adjusting the diameter of the illumination variable aperture, the angle at which the incident light illuminates the edge of the wafer to be inspected is adjusted, thereby adjusting the illumination numerical aperture.

[0017] In some embodiments, the diameter of the illumination variable aperture varies from 1 to 6.6 mm, corresponding to an illumination numerical aperture of 0.0075 to 0.05 and a resolution of 44.7 to 6.71 μm. The relationship between the illumination numerical aperture and the resolution satisfies the following formula: d = 0.61 * λ² / NA², where d is the resolution, λ² is the wavelength of the incident light, and NA² is the illumination numerical aperture.

[0018] In some embodiments, for the flat area and the notch, the diameter of the illumination variable aperture is adjusted to make its illumination numerical aperture 0.0075~0.02, at which point the resolution is 44.7~16.78μm; For the edge slope, by adjusting the diameter of the illumination variable aperture to make its illumination numerical aperture 0.01~0.03, the resolution is 33.55~11.18μm; For the top and bottom edges, the diameter of the illumination variable aperture is adjusted to make its illumination numerical aperture 0.02~0.05, at which point the resolution is 16.78~6.71μm.

[0019] The present application adopts the above technical solution, and its beneficial effects are as follows: The optical system and method provided in this application include an illumination numerical aperture adjustment device disposed in the optical path through which incident light illuminates the edge of the wafer to be inspected. The illumination numerical aperture is adjusted according to the region of interest on the edge of the wafer to be inspected, thereby adjusting the resolution. This technical solution introduces an adjustable aperture device to control the illumination numerical aperture. By utilizing the physical principle that the illumination numerical aperture mainly determines the limiting resolution, the fixed numerical aperture in the traditional optical system is transformed into a dynamically optimized variable. Complex performance switching is achieved through simple mechanical adjustment. High-resolution detection can be achieved according to the detection requirements, enhancing the application range and adaptability of the optical system. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 The schematic diagram provided for the embodiments of this application shows a beam collimated and emitted to the entire side edge of the wafer, with horizontal opposing beams.

[0022] Figure 2 This is a schematic diagram of a beam collimated and emitted horizontally at a point on the edge of a wafer, provided in an embodiment of this application.

[0023] Figure 3 This is a schematic diagram of the depth range during wafer edge detection provided in an embodiment of the present invention.

[0024] Figure 4 This is a schematic diagram of the optical system structure for detecting wafer edge defects provided in an embodiment of the present invention.

[0025] Figure 5 This is a schematic diagram of the quality assessment of the imaging system based on the modulation transfer function index when NA=0.05, provided by an embodiment of the present invention.

[0026] Figure 6 This is a schematic diagram of the modulation transfer function index evaluation of the imaging system quality assessment when NA=0.0075, provided by an embodiment of the present invention.

[0027] Figure 7 This is a simulation effect of wafer edge through-beam imaging provided in the embodiments of the present invention.

[0028] Figure 8 This is an actual imaging schematic diagram provided by an embodiment of the present invention. Detailed Implementation

[0029] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application. In the description of this application, it should be understood that the terms "upper", "lower", "horizontal", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified. To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments.

[0030] Please see Figure 1 This is a schematic diagram of the structure of an optical system for detecting wafer edge defects provided in an embodiment of this application, including an illumination unit 10 and an imaging unit 20. The technical solution for its implementation is described in detail below.

[0031] The incident light provided by the illumination unit 10 illuminates the edge of the wafer to be inspected and generates signal light. The imaging unit 20 receives the signal light reflected from the edge of the wafer and performs imaging.

[0032] The wafer edge defect detection optical system provided in this embodiment consists of two core functional units: an illumination unit 10 and an imaging unit 20. The illumination unit 10 generates incident light and illuminates the edge of the wafer 100 to be inspected. After the incident light interacts with the edge of the wafer 100 (reflection, scattering, etc.), it generates signal light carrying sample information. The imaging unit 20 captures these signal lights and focuses them onto a sensor (such as a CCD or CMOS) to form an image, which is then used by subsequent image processing and analysis algorithms for defect identification.

[0033] Furthermore, an imaging numerical aperture adjustment device 40 is provided in the optical path through which the signal light is received to the imaging unit 20. The imaging numerical aperture adjustment device 40 adjusts the imaging numerical aperture according to the region of interest on the edge of the wafer to be detected, so as to adjust the imaging depth of field.

[0034] It is understood that in this embodiment, only the imaging end is equipped with an imaging numerical aperture adjustment device 40. That is, the imaging numerical aperture adjustment device 40 is set in the optical path where the signal light is received to the imaging unit 20. The imaging numerical aperture adjustment device 40 adjusts the imaging numerical aperture according to the region of interest on the edge of the wafer to achieve the adjustment of the imaging depth of field. The specific implementation scheme is described in detail below.

[0035] In this embodiment, the selection of the region of interest (ROI) at the wafer edge is typically based on the following strategy: First, the wafer edge typically includes the following key areas: flat areas, i.e., the edge plane area, which is prone to problems such as edge chipping and uneven milling; notches, which are V-shaped and used for wafer positioning, are areas with a high incidence of defects; edge bevels, i.e. the inclined areas of the wafer edge, which are prone to edge chipping and cracks; and the top and bottom edges, which may contain contaminants or mechanical damage.

[0036] Secondly, process feature identification: Based on the types of defects that may be caused by wafer process steps (such as CMP, etching, thin film deposition, etc.), predefine ROI, or use prior knowledge or historical inspection data to determine which areas are more prone to defects (such as uneven edge coating, particle contamination, etc.).

[0037] Re-extraction of ROI: Through initial low-resolution global scanning, regions that may contain anomalies are identified and then set as ROIs for high-resolution detection. Alternatively, machine learning algorithms (such as semantic segmentation networks) can be combined to automatically divide the wafer edge image into regions.

[0038] Furthermore, the regions of interest selected using the above strategy have the following characteristics:

[0039] Based on the above characteristics, the most critical performance metrics are determined for each type of ROI: 1. For areas with drastic changes in elevation (such as slopes, gap sidewalls) The primary metric considered is depth of field (DoF). Insufficient depth of field will result in partially blurred or unclear images of sloping or steep surfaces, leading to missed detections. In such cases, some resolution must be sacrificed for greater depth of field. This necessitates using a smaller imaging numerical aperture (low NA_img).

[0040] 2. For flat areas where minute defects need to be detected (such as the top and bottom) Resolution is the primary consideration. Since the surface is within the depth of focus, the main challenge is to distinguish minute scratches or particles. Therefore, a larger numerical aperture (high NA_img) is used. A high NA collects more diffracted light, improving the system's limiting resolution.

[0041] 3. Signal-to-noise ratio considerations. For larger apertures (high NA_img): collecting more light signals helps improve the signal-to-noise ratio, making the image brighter and cleaner; for smaller apertures (low NA_img): collecting less light signals may result in a darker image, requiring longer exposure times or stronger illumination to compensate, otherwise noise will affect defect identification.

[0042] In this embodiment, the imaging numerical aperture adjustment device 40 is an imaging variable aperture. The diameter of the imaging variable aperture is adjustable. By adjusting the diameter of the imaging variable aperture, the angle at which the signal light is received by the imaging unit is adjusted, thereby adjusting the imaging numerical aperture.

[0043] Specifically, the imaging variable aperture is located in the imaging optical path and is usually set at the back focal plane of the objective lens, i.e., the position of the aperture stop. By changing the size of this aperture, the imaging numerical aperture (NA_imaging) is directly controlled, which changes the angular range of the signal light collected by the imaging system, efficiently collecting the signal light returning from the edge of the wafer and imaging it clearly.

[0044] Furthermore, the diameter of the imaging variable aperture varies from 1 to 6.6 mm, corresponding to an imaging numerical aperture of 0.0075 to 0.05 and a depth of field range of 0.22 to 9.77 mm. The relationship between the imaging numerical aperture and the depth of field satisfies the following formula: DOF = λ1 / NA1 2 Where DOF is the depth of field, λ1 is the wavelength of the signal light, and NA1 is the numerical aperture of the imaging. The signal light here is visible light with a wavelength in the range of 0.4um-0.78um, with 0.55um visible light being preferred.

[0045] Furthermore, for flat areas and gaps, the imaging numerical aperture is adjusted to 0.0075~0.02 by adjusting the diameter of the imaging variable aperture, at which point the depth of field ranges from 9.77 to 1.38 mm.

[0046] For the edge slope, by adjusting the diameter of the imaging variable aperture to make its imaging numerical aperture 0.01~0.03, the depth of field range is 5.5~0.61mm.

[0047] For the top and bottom edges, the imaging numerical aperture is adjusted to 0.02~0.05 by adjusting the diameter of the imaging variable aperture, and the depth of field range is 1.38~0.22mm.

[0048] It's understandable that increasing the aperture of the imaging variable stop increases the imaging aperture (NA1). According to the physical formula, resolution ∝ λ1 / NA1, which directly improves the theoretical maximum resolution of the system, enabling the resolution to distinguish smaller defects. However, at the same time, the depth of field will become drastically shallower (the relationship between NA1 and depth of field satisfies the following formula: DOF=λ1 / NA1). 2 (DOF is the range of areas that can be clearly imaged), making focusing extremely demanding. Reducing the aperture of the imaging variable stop decreases the imaging NA1, resulting in a larger clear imaging range along the optical axis. Within a given imaging field of view, this allows for clear imaging of both the upper and lower surfaces of the wafer edge simultaneously, making it ideal for large-area scanning and observation. The trade-off is a decrease in system resolution. Conversely, increasing the aperture of the imaging variable stop increases the imaging NA1, resulting in a slightly smaller clear imaging range along the optical axis. This means that within a given imaging field of view, clear imaging is only possible in a small area, allowing for clear imaging of the wafer edge vertices. This is ideal for small-area scanning and observation, but the trade-off is that large areas such as flat areas at the wafer edge will be blurred.

[0049] Furthermore, an illumination numerical aperture adjustment device is provided in the optical path through which the incident light illuminates the edge of the wafer to be inspected. The illumination numerical aperture adjustment device adjusts the illumination numerical aperture according to the region of interest on the edge of the wafer to achieve resolution adjustment.

[0050] It is understood that the illumination numerical aperture adjustment device 30 is only provided at the illumination end. That is, the illumination numerical aperture adjustment device 30 is provided in the optical path where the incident light illuminates the edge of the wafer to be inspected. The illumination numerical aperture is adjusted by the illumination numerical aperture adjustment device to achieve the adjustment of resolution.

[0051] In this embodiment, the illumination numerical aperture adjustment device 30 is an illumination variable aperture, the diameter of which is adjustable. By adjusting the diameter of the illumination variable aperture, the angle at which the incident light illuminates the edge of the wafer to be inspected is adjusted, thereby adjusting the illumination numerical aperture.

[0052] Specifically, the illumination variable aperture is generally composed of multiple blades, and its opening and closing size can be adjusted by a stepper motor or manually. By changing the aperture size, the illumination numerical aperture (NA_illumination) is directly controlled, which changes the angle range of incident light illuminating the sample, thereby generating uniform, stable and appropriately bright incident light to illuminate the inspection area at the edge of the wafer.

[0053] Furthermore, the diameter of the illumination variable aperture varies from 1 to 6.6 mm, corresponding to an illumination numerical aperture of 0.0075 to 0.05 μm and a resolution of 44.7 to 6.71 μm. The relationship between the illumination numerical aperture and the resolution satisfies the following formula: d = 0.61 * λ² / NA², where d is the resolution, λ² is the wavelength of the incident light, and NA² is the illumination numerical aperture. Here, the incident light is visible light with a wavelength in the range of 0.4 μm to 0.78 μm, with 0.55 μm visible light being preferred.

[0054] Furthermore, the numerical aperture and resolution configuration strategies for illumination in different sensitive areas are as follows: 1. Flat areas and gaps The region is characterized by steep vertical walls and abrupt changes in height. The main defects are edge chipping and notch cracks, requiring maximizing defect contrast rather than limiting resolution. The interference effect generated by highly coherent illumination needs to be utilized to transform minute phase changes into significant light-dark contrast. The illumination numerical aperture NA2 is chosen to be small, ranging from 0.0075 to 0.02, for example, 0.01; corresponding to a theoretical resolution (d) of 44.7 μm to 16.78 μm. This resolution range is sufficient to clearly distinguish larger edge chipping defects, while the illumination coherence is also sufficient to enhance the defect signal.

[0055] 2. Sloping edges and steep sidewalls The region exhibits highly variable characteristics, with the main defects being microcracks and uneven coating. A balance needs to be struck between defect contrast and edge definition clarity. Sufficient coherence is required to enhance edge contrast, and adequate resolution is also needed to define the shape of the slope. The illumination numerical aperture NA2 is chosen to be small to medium, ranging from 0.01 to 0.03, for example, 0.02; corresponding to a theoretical resolution (d) of 33.55 μm to 11.18 μm. Although the theoretical resolution is low, it is sufficient for cracks on the order of tens of micrometers, prioritizing the ability to highlight the cracks from the background.

[0056] 3. Top and bottom edges The area is relatively flat. The main defects are particulate contaminants and fine scratches, requiring consideration of high resolution and low noise. It is necessary to distinguish minute particles and ensure uniform illumination for accurate assessment of defect intensity. A large illumination numerical aperture (NA2) is chosen, ranging from 0.02 to 0.05, for example, 0.04. This corresponds to a theoretical resolution (d) of 16.78 μm to 6.71 μm. At this level, low-coherence illumination provides a uniform light field, suppressing speckle noise, while the highest theoretical resolution can capture even smaller particles and finer scratches.

[0057] It is understandable that increasing the aperture of the illumination variable aperture increases the illumination NA2. According to the physical formula, the relationship between NA2 and resolution satisfies the Rayleigh criterion as follows: d = 0.61 * λ2 / NA2, where d is the smallest resolvable interval and λ2 is the wavelength. This provides large-angle incident light, which can excite scattering signals of finer sample structures, making it possible for the system to detect high-resolution defects (such as microburrs and nanoparticles). At this time, the effective resolution of the system is improved. Conversely, decreasing the aperture of the illumination variable aperture reduces the illumination NA2, providing nearly collimated small-angle incident light. Although this sacrifices some resolution, it can significantly reduce stray light, improve image contrast, and indirectly increase the imaging depth of field.

[0058] It should be noted that in the dual adjustment device setting, illumination NA2 and imaging NA1 must be used in match. Specifically, illumination NA2 can be set slightly smaller than imaging NA1 in order to better reduce stray light.

[0059] It is understandable that when the illumination numerical aperture is smaller than the imaging numerical aperture, the angle of all light rays illuminating the sample is within the range that the imaging system can receive. This means that any scattered signal excited by the illumination light (whether from a defect or a normal surface), as long as it is excited, its angle can be captured by the imaging lens; since one of the main sources of stray light is illumination light beyond the imaging NA1 range, if an illumination ray has an excessively large angle (even if it comes from the illumination system) and illuminates a defect in the sample or optical element, it may be scattered. Those rays with scattering angles beyond the imaging NA1 range will not be focused by the objective lens, but will diffusely reach the sensor as non-uniform background light (flares); when the illumination numerical aperture is larger than the imaging numerical aperture, this possibility is fundamentally eliminated, because all illumination light is within the "visible range" of the imaging system, and the scattered light they produce can theoretically be captured and correctly focused, without producing uncontrollable diffuse stray light. The background of the image is extremely "clean" and "black," with the highest signal-to-noise ratio (SNR) and contrast. Weak defect signals can be clearly highlighted from the uniform background.

[0060] It should be noted that if the illumination NA2 is greater than the imaging NA1, the resolution is still calculated based on the imaging NA1, but some stray light from the illumination may enter the imaging system. If the limiting NAs of the two are different, the overall system limiting NA is the smaller one.

[0061] When the illumination numerical aperture is larger than the imaging numerical aperture, the illumination system provides a large amount of wide-angle light that the imaging system cannot "see" (i.e., cannot collect). This "excess" wide-angle light becomes harmful light, reflecting and scattering when it strikes the surfaces of optical elements (such as lens edges and inner walls of the lens barrel). It also produces strong scattering when it strikes non-detection areas of the sample (such as the complex structure at the edge of a wafer). This scattered light from the "excess" light is mostly angularly irregular, and much of it enters the imaging system as background noise, severely reducing the signal-to-noise ratio and contrast of the image, and may even produce flares and ghosting, drowning out the true defect signals.

[0062] Therefore, in practice, the required system NA should be determined first based on the detection requirements (high resolution or large depth of field).

[0063] Next, set the imaging NA1: Adjust the imaging numerical aperture stop to a value slightly larger than the target system's NA. For example, if the target system's NA is 0.4, then set NA_imaging to 0.45. This sets the upper limit of the system's performance.

[0064] Rematch Illumination NA2: Adjust the illumination numerical aperture stop to a value slightly smaller than the imaging NA1, making it very close to but not exceeding the NA_imaging setting above. For example, set it to 0.4.

[0065] At this point, the system's effective NA = min(0.4, 0.45) = 0.4. The system operates with an NA of 0.4, achieving the required resolution and depth of field at that NA, and operating with minimal stray light and maximum contrast.

[0066] Please see Figure 2 The provided implementation diagram shows a NA adjustment device (referring to a variable aperture) installed at both the illumination and imaging ends. However, in actual implementation, only one NA adjustment device can be installed in either the incident or collection area. When the NA adjustment device is installed at the illumination end, the system's NA is controlled by adjusting the incident light angle range; when the NA adjustment device is installed at the imaging end, the system's NA is controlled by adjusting the collected signal light angle range. Figure 2 The inclusion of two adjustment devices allows for simultaneous adjustment of the incident light and signal light angle ranges to achieve NA control. In this dual-adjustment setup, the illumination NA and imaging NA must be matched. Specifically, the illumination NA2 can be set slightly smaller than the imaging NA1 to better reduce stray light. If the illumination NA2 is greater than the imaging NA1, the resolution is still calculated based on the imaging NA1, but some stray light from the illumination may enter the imaging system. If the limiting NAs for the two devices are different, the overall system limiting NA is set to the smaller one.

[0067] Please see Figure 3 As shown, in wafer edge inspection, in some cases it is desirable to detect defects over a larger area on the top and bottom surfaces, thus requiring clear imaging within the depth of field S2; in other cases, only small edge defects need to be detected, in which case clear imaging within the depth of field S1 is sufficient. Based on the relationship between NA1, resolution, and depth of field, an aperture stop is added at the imaging end to adjust NA1. Increasing NA1 allows for the clear observation of smaller burr defects on the wafer edge over a smaller area, while decreasing NA1 allows for the observation of slightly larger defects on the top and bottom surfaces of the wafer over a larger area. To meet the resolution requirements of the imaging system without introducing excessive stray light, the NA2 of the illumination unit should be matched with the NA2 of the imaging unit. Therefore, a variable aperture stop is set in the optical system to adjust imaging NA1 and illumination NA2, adjusting according to different application scenarios to meet either high-precision edge detection or low-precision detection of larger defects on the top and bottom surfaces.

[0068] Please see Figure 4 A schematic diagram of the structure of an optical system for detecting wafer edge defects provided in another embodiment of this application.

[0069] In this embodiment, the lighting unit 10 includes an LED light source 11, a first lens group G1, a light diffuser 12, a second lens group G2, and a third lens group G3. The first lens group G1 includes a first lens L1, a second lens L2, and a third lens L3. The second lens group G2 includes a fourth lens L4, a fifth lens L5, and a sixth lens L6. The third lens group G3 includes a seventh lens L7, an eighth lens L8, and a ninth lens L9. The light beam emitted from the LED light source 11 passes sequentially through the first lens L1, the second lens L2 and the third lens L3, the light homogenizer 12, the fourth lens L4, the fifth lens L5 and the sixth lens L6, the illumination numerical aperture adjustment device 30, the seventh lens L7, the eighth lens L8 and the ninth lens L9, and then forms incident light that illuminates the edge of the wafer to be tested.

[0070] It is understandable that the first lens group G1 is mainly responsible for collecting the maximum range of light emitted by the light source and forming the initial intermediate image; the second lens group G2 is responsible for transmitting the light path and further correcting aberrations, such as spherical aberration and coma, to ensure that the light spot shape is regular; the third lens group finally projects the image of the aperture onto the object side and ensures that the illumination beam has good collimation.

[0071] Furthermore, the first lens group G1, the second lens group G2, and the third lens group G3 constitute the Kohler illumination system, and the incident light forms a uniform light spot on the detection surface of the wafer to be inspected.

[0072] It should be noted that in wafer defect detection, the brightness difference (contrast) of the image captured by the camera must accurately reflect the characteristic differences of the sample surface (such as defects and morphology), and cannot be caused by the unevenness of the illumination light itself. If the illumination is uneven (bright in the center of the image and dark around the edges), the system may misclassify areas with normal edges as dark defects (false defects), or drown out bright defects in a bright background (missed detection).

[0073] Furthermore, by adjusting the illumination numerical aperture adjustment device 30 (variable aperture) located in its optical path, the illumination NA2 is controlled. The aperture is opened to achieve large-angle illumination, providing excitation for the high-resolution detection mode; the aperture is closed to achieve small-angle (collimated) illumination, providing illumination for the large depth-of-field detection mode.

[0074] This application uses Kohler illumination to form uniform illumination on the illumination surface, with a uniformity of up to 95%. This eliminates errors introduced by light source issues and ensures that any changes in brightness in the image originate from the sample itself, greatly improving the accuracy and reliability of the detection.

[0075] In this embodiment, the imaging unit 20 includes a fourth lens group G4 and a fifth lens group G5. The fourth lens group G4 includes a tenth lens L10, an eleventh lens L11, and a twelfth lens L12. The fifth lens group G5 includes a thirteenth lens L13, a fourteenth lens L14, and a fifteenth lens L15, wherein: The signal light is imaged sequentially through the tenth lens L10, the eleventh lens L11, the twelfth lens L12, the imaging numerical aperture adjustment device 40, the thirteenth lens L13, the fourteenth lens L14, and the fifteenth lens L15.

[0076] In this embodiment, the fourth lens group G4 and the fifth lens group G5 are exactly the same as the third lens group G3 and the second lens group G2 and are placed symmetrically with the wafer to be tested. The fourth lens group G4 adopts an object-side telecentric design, and the fifth lens group G5 adopts an image-side telecentric design.

[0077] It is understandable that the fourth lens group (L10-L12) employs an object-side telecentric design, meaning that the principal rays of this lens group are parallel to the optical axis on the object side (i.e., the wafer side). This implies that all principal rays emanating from the sample surface and entering the lens exit at a near-perpendicular angle, resulting in more uniform illumination and imaging, and eliminating measurement errors. However, for wafer edge morphology measurements, if the lens is not object-side telecentric, slight up-and-down movement of the sample within the depth of field will cause changes in the size of the image on the sensor (larger image at closer distances and smaller image at farther distances), leading to significant dimensional measurement errors. The object-side telecentric design completely eliminates this error, ensuring that the magnification remains constant regardless of minute changes in sample height.

[0078] Furthermore, the beam passing through the fourth lens group (L10-L12) can be adjusted to an imaging numerical aperture (NA) of 0.0075 to 0.05 under the action of the imaging numerical aperture adjustment device 40.

[0079] When the imaging NA1 = 0.05 (maximum value), the resolution is high enough to distinguish larger particles, scratches, and burrs; the depth of field is shallow, requiring precise focusing. It can be applied to high-precision defect detection modes.

[0080] When NA1 = 0.0075 (minimum), the resolution is low. At the same wavelength, the resolution decreases, and only larger defects can be seen; however, the depth of field is extremely large. It can be applied to a wide range of top and bottom surface screening modes. The huge depth of field ensures that the top, edges, and bottom surfaces of the wafer edge are clearly imaged simultaneously.

[0081] Therefore, this embodiment employs a telecentric design and a variable NA (aperture stop). Changing the NA usually affects aberrations and telecentricity. Maintaining telecentricity within such a large NA variation range demonstrates a highly complex and excellent optical design. The combination of multiple lenses effectively corrects various aberrations, ensuring geometrical accuracy of the image (telecentricity) and providing the ability to flexibly switch between resolution and depth of field (variable NA). In this embodiment, the fifth lens group G5 (L13-L15) adopts an image-side telecentric design, optimizing distortion to 0.2%. This lens group serves as a relay or rear lens group in the imaging unit, responsible for transmitting the intermediate image to the sensor and performing crucial aberration correction.

[0082] The fifth lens group G5 (L13-L15) provided in this embodiment adopts a telecentric image-side design, where the principal rays of this lens group are parallel to the optical axis on the image side (i.e., the sensor side). This means that all rays are incident on the camera sensor surface at a near-perpendicular angle, ensuring edge brightness. If the light shines obliquely onto the sensor pixels, the image edges may darken (vignetting) due to microlens offset, decreased sensitivity, etc. Distortion optimization to 0.2% means that at the outermost edge of the image field, the deviation between the actual image point and the ideal image point is less than two-thousandths of the sensor's diagonal size. For metrology applications, distortion must be strictly eliminated. Whether measuring the size of defects or the contour position of crystal edges, extremely low distortion is a prerequisite for ensuring absolute measurement accuracy. A distortion level of 0.2% indicates that the system can perform high-precision dimensional quantization without complex post-correction by software (software correction may introduce interpolation errors). This ensures the authenticity of the image and prevents the distortion of the true shape of defects due to the lens.

[0083] Therefore, this embodiment adopts a combination of object-centric telecentricity and low distortion. Object-centric telecentricity ensures magnification stability, while low distortion ensures shape accuracy. Together, they ensure that any geometric data extracted from the image is real and reliable, ensuring that the image can be transmitted to the sensor with high quality and high fidelity, laying a solid foundation for subsequent accurate measurement and analysis.

[0084] In summary, this embodiment balances the trade-off between accuracy and range through object-side telecentricity and variable NA, while ensuring the reliability of measurement and fidelity through image-side telecentricity and extremely low distortion, fully meeting the stringent requirements of the semiconductor industry for accuracy and flexibility.

[0085] Please refer to the following: Figure 4 The diagram below shows a schematic of a wafer edge defect detection optical system according to a specific embodiment of this application, including the following: 1. Lighting unit 10, including: Light source 11: 550nm LED (monochromatic light, to avoid color difference interference), emitting surface diameter 2mm.

[0086] Lens Group 1 G1 (L1-L3): Collimates the LED beam.

[0087] Lens group 2 (L4-L6): works with frosted glass to achieve uniform light distribution.

[0088] Lens group 3 (L7-L9): Image aperture 1 onto the edge of the wafer to form Kohler illumination.

[0089] Light homogenizer 12: Eliminates LED speckle and further achieves uniform light.

[0090] 2. Imaging unit 20, comprising: Lens Group 4 (L10-L12): Object-side telecentric design, NA=0.0075~0.05 adjustable.

[0091] Lens Group 5 (L13-L15): Image-side telecentric design, distortion optimized to 0.2%.

[0092] 3. Aperture adjustment device: The illumination numerical aperture adjustment device 30 is an illumination variable aperture that controls the illumination NA2 to achieve a resolution (adjustable from 6.71 to 44.7 μm). The relationship between NA2 and resolution is satisfied by the Rayleigh criterion as follows: d = 0.61 * λ2 / NA2, where d is the smallest resolvable interval and λ2 is the wavelength, preferably 550 nm in this example.

[0093] The imaging numerical aperture adjustment device 40 is an imaging variable aperture that controls the imaging NA1 to achieve dynamic adjustment of the depth of field (adjustable from 0.22 to 9.77 mm). The relationship between NA1 and the depth of field satisfies the following formula: DOF = λ1 / NA1^2, where DOF is the range in which the image can be clearly captured.

[0094] In this embodiment, the lens diameter is <17mm, the total optical length of the imaging system is <190mm, and the total length of the entire optical system is <450mm; a segmented lens group design is adopted, the lens groups are reused, and the lens materials are all selected from commonly used materials.

[0095] To further reduce costs and design complexity, G1 and G2 are placed symmetrically with identical lenses. The illumination modules of G2 / G3 are also placed symmetrically with the imaging modules of G4 / G5, with identical lenses. Only 6 different types of lenses were used for the 15 lenses.

[0096] G1 consists of a negative focal length lens L1 and positive focal length lenses L2 / L3, with a combined positive focal length of 22mm and a collimated LED light source.

[0097] G2 is exactly the same as G1, placed symmetrically, and used with a frosted glass homogenizer for light homogenization.

[0098] G3 consists of a negative focal length lens L7 and positive focal length lenses L8 / L9, with a combined focal length of 66mm, which images the aperture 1 onto the edge of the wafer to form Kohler illumination.

[0099] G4 and G5 are exactly the same as G3 and G2, and are placed symmetrically to image the wafer edge onto the camera's detection surface.

[0100] The illumination units G1 / G2 / G3 form the Kohler illumination system, which forms a uniform light spot on the wafer detection surface with a uniformity of >95%. The illumination modules G2 / G3 and the imaging modules G4 / G5 are projection imaging lenses with a projection magnification of 3X. The sampling object has a dual telecentric design with a telecentricity of <0.01°.

[0101] The variable aperture diameter ranges from 1 to 6.6 mm, corresponding to an NA of 0.0075 to 0.05, a resolution of 44.7 to 6.71 μm, and a depth of field range of 0.22 to 9.77 mm. It can achieve high-resolution detection or large depth-of-field scanning according to the detection requirements.

[0102] Please see Figure 5 This is a schematic diagram of the modulation transfer function (MTF) index for evaluating the imaging system quality when NA=0.05, as provided in this embodiment.

[0103] Please see Figure 6This is a schematic diagram of the modulation transfer function (MTF) index for evaluating the imaging system quality when NA=0.0075, as provided in this embodiment.

[0104] Depend on Figure 5 and Figure 6 It can be seen that the imaging quality reaches the diffraction limit when NA=0.05 and NA=0.0075.

[0105] Please see Figure 7 This is a simulation effect of wafer edge through-beam imaging provided in this embodiment. Figure 8 This is a schematic diagram of the actual imaging process. Figure 7 and 8 As can be seen, the actual imaging effect is close to the simulation effect.

[0106] The wafer edge defect detection optical system provided in the above embodiments of this application includes an imaging numerical aperture adjustment device disposed in the optical path where the signal light is received to the imaging unit. This device adjusts the imaging numerical aperture according to the region of interest on the wafer edge to achieve adjustment of the imaging depth of field. This technical solution introduces an adjustable aperture device and, based on the physical principle that the imaging numerical aperture mainly determines the limiting resolution and depth of field, transforms the fixed numerical aperture in the traditional optical system into a dynamically optimized variable. Complex performance switching is achieved through simple mechanical adjustment. It can achieve high-resolution detection or large depth-of-field scanning according to the detection requirements, thereby enhancing the application range and adaptability of the optical system.

[0107] This application also provides an optical method for detecting wafer edge defects, applied to the wafer edge defect detection optical system provided in any of the above embodiments. The wafer edge defect detection optical method includes the following steps: Step S10: Incident light illuminates the edge of the wafer to be tested and generates signal light.

[0108] Step S20: Receive signal light reflected from the edge of the wafer and perform imaging; wherein: the imaging numerical aperture adjustment device, which is set in the optical path where the signal light is received to the imaging unit, adjusts the imaging numerical aperture according to the region of interest on the edge of the wafer to achieve the adjustment of the imaging depth of field.

[0109] In this embodiment, the imaging numerical aperture adjustment device 40 is an imaging variable aperture. The diameter of the imaging variable aperture is adjustable. By adjusting the diameter of the imaging variable aperture, the angle at which the signal light is received by the imaging unit is adjusted, thereby adjusting the imaging numerical aperture.

[0110] Specifically, the imaging variable aperture is located in the imaging optical path and is usually set at the back focal plane of the objective lens, i.e., the position of the aperture stop. By changing the size of this aperture, the imaging numerical aperture (NA_imaging) is directly controlled, which changes the angular range of the signal light collected by the imaging system, efficiently collecting the signal light returning from the edge of the wafer and imaging it clearly.

[0111] Furthermore, the diameter of the imaging variable aperture varies from 1 to 6.6 mm, corresponding to an imaging numerical aperture of 0.0075 to 0.05 and a depth of field range of 0.22 to 9.77 mm. The relationship between the imaging numerical aperture and the depth of field satisfies the following formula: DOF = λ1 / NA1 2 Where: DOF is the depth of field, λ1 is the wavelength of the signal light, and NA1 is the numerical aperture of the imaging.

[0112] Furthermore, for flat areas and gaps, the imaging numerical aperture is adjusted to 0.0075~0.02 by adjusting the diameter of the imaging variable aperture, at which point the depth of field ranges from 9.77 to 1.38 mm.

[0113] For the edge slope, by adjusting the diameter of the imaging variable aperture to make its imaging numerical aperture 0.01~0.03, the depth of field range is 5.5~0.61mm.

[0114] For the top and bottom edges, the imaging numerical aperture is adjusted to 0.02~0.05 by adjusting the diameter of the imaging variable aperture, and the depth of field range is 1.38~0.22mm.

[0115] It's understandable that increasing the aperture of the imaging variable stop increases the imaging NA1. According to the physical formula, resolution ∝ λ / NA1, which directly improves the theoretical maximum resolution of the system, enabling the resolution to distinguish smaller defects. However, at the same time, the depth of field becomes drastically shallower (the relationship between NA1 and depth of field satisfies the following formula: DOF=λ1 / NA1^2, where DOF is the range that can be clearly imaged), making focusing extremely demanding. Conversely, decreasing the aperture of the imaging variable stop reduces the imaging NA1. When the aperture is reduced, the system can have a larger range of clear imaging along the optical axis, allowing the upper and lower surfaces of the wafer edge to be clearly imaged simultaneously. This is very suitable for large-scale scanning and observation, but the trade-off is a decrease in system resolution.

[0116] Furthermore, an illumination numerical aperture adjustment device is provided in the optical path through which the incident light illuminates the edge of the wafer to be inspected. The illumination numerical aperture adjustment device adjusts the illumination numerical aperture according to the region of interest on the edge of the wafer to achieve resolution adjustment.

[0117] It is understood that the illumination numerical aperture adjustment device 30 is only provided at the illumination end. That is, the illumination numerical aperture adjustment device 30 is provided in the optical path where the incident light illuminates the edge of the wafer to be inspected. The illumination numerical aperture is adjusted by the illumination numerical aperture adjustment device to achieve the adjustment of resolution.

[0118] In this embodiment, the illumination numerical aperture adjustment device 30 is an illumination variable aperture, the diameter of which is adjustable. By adjusting the diameter of the illumination variable aperture, the angle at which the incident light illuminates the edge of the wafer to be inspected is adjusted, thereby adjusting the illumination numerical aperture.

[0119] Specifically, the illumination variable aperture is generally composed of multiple blades, and its opening and closing size can be adjusted by a stepper motor or manually. By changing the aperture size, the illumination numerical aperture (NA_illumination) is directly controlled, which changes the angle range of incident light illuminating the sample, thereby generating uniform, stable and appropriately bright incident light to illuminate the inspection area at the edge of the wafer.

[0120] Furthermore, the diameter of the illumination variable aperture varies from 1 to 6.6 mm, corresponding to an illumination numerical aperture of 0.0075 to 0.05 and a resolution of 44.7 to 6.71 μm. The relationship between the illumination numerical aperture and the resolution satisfies the following formula: d = 0.61 * λ² / NA², where d is the resolution, λ² is the wavelength of the incident light, and NA² is the illumination numerical aperture.

[0121] Furthermore, the numerical aperture and resolution configuration strategies for illumination in different sensitive areas are as follows: 1. Flat areas and gaps The region is characterized by steep vertical walls and abrupt changes in height. The main defects are edge chipping and notch cracks, requiring maximizing defect contrast rather than limiting resolution. The interference effect generated by highly coherent illumination needs to be utilized to transform minute phase changes into significant light-dark contrast. The illumination numerical aperture NA2 is chosen to be small, ranging from 0.0075 to 0.02, for example, 0.01; corresponding to a theoretical resolution (d) of 44.7 μm to 16.78 μm. This resolution range is sufficient to clearly distinguish larger edge chipping defects, while the illumination coherence is also sufficient to enhance the defect signal.

[0122] 2. Sloping edges and steep sidewalls The region exhibits highly variable characteristics, with the main defects being microcracks and uneven coating. A balance needs to be struck between defect contrast and edge definition clarity. Sufficient coherence is required to enhance edge contrast, and adequate resolution is also needed to define the shape of the slope. The illumination numerical aperture NA2 is chosen to be small to medium, ranging from 0.01 to 0.03, for example, 0.02; corresponding to a theoretical resolution (d) of 33.55 μm to 11.18 μm. Although the theoretical resolution is low, it is sufficient for cracks on the order of tens of micrometers, prioritizing the ability to highlight the cracks from the background.

[0123] 3. Top and bottom edges The area is relatively flat. The main defects are particulate contaminants and fine scratches, requiring consideration of high resolution and low noise. It is necessary to distinguish minute particles and ensure uniform illumination for accurate assessment of defect intensity. A large illumination numerical aperture (NA2) is chosen, ranging from 0.02 to 0.05, for example, 0.04. This corresponds to a theoretical resolution (d) of 16.78 μm to 6.71 μm. At this level, low-coherence illumination provides a uniform light field, suppressing speckle noise, while the highest theoretical resolution can capture even smaller particles and finer scratches.

[0124] It is understandable that increasing the aperture of the illumination variable aperture increases the illumination NA2. According to the physical formula, the relationship between NA2 and resolution satisfies the Rayleigh criterion as follows: d = 0.61 * λ2 / NA2, where d refers to the smallest resolvable interval and λ2 is the wavelength. This provides large-angle incident light, which can excite scattering signals of finer structures in the sample, making it possible for the system to detect high-resolution defects (such as microburrs and nanoparticles). At this time, the effective resolution of the system is improved. Conversely, decreasing the aperture of the illumination variable aperture reduces the illumination NA, providing nearly collimated small-angle incident light. Although this sacrifices some resolution, it can significantly reduce stray light, improve image contrast, and indirectly increase the imaging depth of field.

[0125] It should be noted that in the dual adjustment device setting, illumination NA2 and imaging NA1 must be used in match. Specifically, illumination NA2 can be set slightly smaller than imaging NA1 in order to better reduce stray light.

[0126] It is understandable that when the illumination numerical aperture is smaller than the imaging numerical aperture, the angle of all light rays illuminating the sample is within the range that the imaging system can receive. This means that any scattered signal excited by the illumination light (whether from a defect or a normal surface), as long as it is excited, its angle can be captured by the imaging lens; since one of the main sources of stray light is illumination light beyond the imaging NA1 range, if an illumination ray has an excessively large angle (even if it comes from the illumination system) and illuminates a defect in the sample or optical element, it may be scattered. Those rays with scattering angles beyond the imaging NA1 range will not be focused by the objective lens, but will diffusely reach the sensor as non-uniform background light (flares); when the illumination numerical aperture is larger than the imaging numerical aperture, this possibility is fundamentally eliminated, because all illumination light is within the "visible range" of the imaging system, and the scattered light they produce can theoretically be captured and correctly focused, without producing uncontrollable diffuse stray light. The background of the image is extremely "clean" and "black," with the highest signal-to-noise ratio (SNR) and contrast. Weak defect signals can be clearly highlighted from the uniform background.

[0127] It should be noted that if the illumination NA2 is greater than the imaging NA1, the resolution is still calculated based on the imaging NA1, but some stray light from the illumination may enter the imaging system. If the limiting NAs of the two are different, the overall system limiting NA is the smaller one.

[0128] When the illumination numerical aperture is larger than the imaging numerical aperture, the illumination system provides a large amount of wide-angle light that the imaging system cannot "see" (i.e., cannot collect). This "excess" wide-angle light becomes harmful light, reflecting and scattering when it strikes the surfaces of optical elements (such as lens edges and inner walls of the lens barrel). It also produces strong scattering when it strikes non-detection areas of the sample (such as the complex structure at the edge of a wafer). This scattered light from the "excess" light is mostly angularly irregular, and much of it enters the imaging system as background noise, severely reducing the signal-to-noise ratio and contrast of the image, and may even produce flares and ghosting, drowning out the true defect signals.

[0129] Therefore, in practice, the required system NA should be determined first based on the detection requirements (high resolution or large depth of field).

[0130] Next, set the imaging NA1: Adjust the imaging numerical aperture stop to a value slightly larger than the target system's NA. For example, if the target system's NA is 0.4, then set NA_imaging to 0.45. This sets the upper limit of the system's performance.

[0131] Rematch Illumination NA2: Adjust the illumination numerical aperture stop to a value slightly smaller than the imaging NA1, making it very close to but not exceeding the NA_imaging setting above. For example, set it to 0.4.

[0132] At this point, the system's effective NA = min(0.4, 0.45) = 0.4. The system operates with an NA of 0.4, achieving the required resolution and depth of field at that NA, and operating with minimal stray light and maximum contrast.

[0133] Please see Figure 2 The provided implementation diagram shows a NA adjustment device (referring to a variable aperture) installed at both the illumination and imaging ends. However, in actual implementation, only one NA adjustment device can be installed in either the incident or collection area. When the NA adjustment device is installed at the illumination end, the system's NA is controlled by adjusting the incident light angle range; when the NA adjustment device is installed at the imaging end, the system's NA is controlled by adjusting the collected signal light angle range. Figure 2 The inclusion of two adjustment devices allows for simultaneous adjustment of the incident light and signal light angle ranges to achieve NA control. In this dual-adjustment setup, the illumination NA and imaging NA must be matched. Specifically, the illumination NA can be set slightly smaller than the imaging NA to better reduce stray light. If the illumination NA is greater than the imaging NA, the resolution is still calculated based on the imaging NA, but some stray light from the illumination may enter the imaging system. If the limiting NAs for the two devices are different, the overall system limiting NA is set to the smaller one.

[0134] Please see Figure 3 As shown, in wafer edge detection, in some cases it is desirable to detect defects over a larger area on the top and bottom surfaces, thus requiring clear imaging within the depth of field S2; in other cases, only small edge defects need to be detected, in which case clear imaging within the depth of field S1 is sufficient. Based on the relationship between NA1, resolution, and depth of field, an aperture stop is added at the imaging end to adjust NA1. Increasing NA1 allows for the clear observation of smaller burr defects on the wafer edge over a smaller area, while decreasing NA1 allows for the observation of slightly larger defects on the top and bottom surfaces of the wafer over a larger area. To meet the resolution requirements of the imaging system without introducing excessive stray light, the NA2 of the illumination unit should be matched with the NA1 of the imaging unit. Therefore, a variable aperture stop is set in the optical system to adjust the imaging NA1 and illumination NA2, adjusting according to different application scenarios to meet either high-precision edge detection or low-precision detection of larger defects on the top and bottom surfaces.

[0135] The optical method for wafer edge defect detection provided in the above embodiments of this application adjusts the imaging numerical aperture according to the region of interest on the edge of the wafer to be tested by using an imaging numerical aperture adjustment device disposed in the optical path where the signal light is received to the imaging unit, thereby adjusting the imaging depth of field. This technical solution introduces an adjustable aperture device to control the imaging numerical aperture. By utilizing the physical principle that the imaging numerical aperture mainly determines the limiting resolution and depth of field, the fixed numerical aperture in the traditional optical system is transformed into a dynamically optimized variable. Complex performance switching is achieved through simple mechanical adjustment. High-resolution detection or large depth-of-field scanning can be achieved according to the detection requirements, thereby enhancing the application range and adaptability of the optical system.

[0136] The above are merely preferred embodiments of this application, and only specifically describe the technical principles of this application. These descriptions are only for explaining the principles of this application and should not be construed as limiting the scope of protection of this application in any way. Based on this explanation, any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application, as well as other specific embodiments of this application that can be conceived by those skilled in the art without creative effort, should be included within the scope of protection of this application.

Claims

1. An optical system characterized by comprising: The optical system comprises an illumination unit and an imaging unit; The illumination unit provides incident light to irradiate the edge of a wafer to be detected and generates signal light through orthogonal projection; The imaging unit receives the signal light from the edge of the wafer to be detected and forms an image; wherein: The light path of the incident light irradiating the edge of the wafer to be detected is provided with an illumination numerical aperture adjusting device, which adjusts the illumination numerical aperture according to a region of interest on the edge of the wafer to be detected to realize adjustment of resolution.

2. The optical system of claim 1, wherein The region of interest comprises at least one of a flat area, a notch, an edge bevel, an edge top and an edge bottom, and a corresponding illumination numerical aperture is set in advance for each type of region of interest, and the illumination numerical aperture adjusting device adjusts according to the corresponding illumination numerical aperture to realize adjustment of resolution.

3. The optical system of claim 2, wherein, The illumination numerical aperture adjusting device is an illumination variable diaphragm, the diameter of the illumination variable diaphragm is adjustable, and the angle of the incident light irradiating the edge of the wafer to be detected is adjusted by adjusting the diameter of the illumination variable diaphragm to adjust the illumination numerical aperture.

4. The optical system of claim 3, wherein, The diameter of the illumination variable diaphragm changes in the range of 1-6.6 mm, the corresponding illumination numerical aperture is 0.0075-0.05, the resolution is 44.7-6.71 μm, and the relationship between the illumination numerical aperture and the resolution satisfies the formula d=0.61*λ2 / NA2, wherein d is the resolution, λ2 is the wavelength of the incident light, and NA2 is the illumination numerical aperture.

5. The optical system of claim 4, wherein For the flat area and the notch, the diameter of the illumination variable diaphragm is adjusted so that the illumination numerical aperture is 0.0075-0.02, and at this time the resolution is 44.7-16.78 μm; For the edge bevel, the diameter of the illumination variable diaphragm is adjusted so that the illumination numerical aperture is 0.01-0.03, and at this time the resolution is 33.55-11.18 μm; For the edge top and the edge bottom, the diameter of the illumination variable diaphragm is adjusted so that the illumination numerical aperture is 0.02-0.05, and at this time the resolution is 16.78-6.71 μm.

6. The optical system of claim 1, wherein, The illumination unit comprises an LED light source, a first lens group, a light homogenizing sheet, a second lens group and a third lens group, the first lens group comprises a first lens, a second lens and a third lens, the second lens group comprises a fourth lens, a fifth lens and a sixth lens, and the third lens group comprises a seventh lens, an eighth lens and a ninth lens, wherein: The light beam emitted by the LED light source passes through the first lens, the second lens, the third lens, the light homogenizing sheet, the fourth lens, the fifth lens, the sixth lens, the illumination numerical aperture adjusting device, the seventh lens, the eighth lens and the ninth lens in sequence to form incident light irradiating the edge of the wafer to be detected.

7. The optical system of claim 6, wherein, The first lens group, the second lens group and the third lens group form a Kohler illumination system, and the incident light forms a uniform light spot at the edge of the wafer to be detected.

8. An optical method, characterized by, The method comprises the following steps: incident light is irradiated to the edge of a wafer to be detected and generates signal light; the signal light reflected from the edge of the wafer is received and imaged; wherein: an illumination numerical aperture adjusting device is arranged in the optical path of the incident light irradiated to the edge of the wafer to be detected, and the illumination numerical aperture is adjusted according to the region of interest on the edge of the wafer to be detected, so as to realize the adjustment of the resolution.

9. The optical method of claim 8, wherein the region of interest includes at least one of a flat area, a notch, an edge bevel, an edge top and an edge bottom, and a corresponding illumination numerical aperture is set in advance according to each type of region of interest, and the illumination numerical aperture adjusting device adjusts according to the corresponding illumination numerical aperture to realize the adjustment of the resolution.

10. The optical method of claim 9, wherein the illumination numerical aperture adjusting device is an illumination variable diaphragm, the diameter of the illumination variable diaphragm can be adjusted, and the angle of the incident light irradiated to the edge of the wafer to be detected is adjusted by adjusting the diameter of the illumination variable diaphragm, so as to adjust the illumination numerical aperture.

11. The optical method of claim 10, wherein, The diameter of the illumination variable diaphragm varies in the range of 1-6.6 mm, the corresponding illumination numerical aperture is 0.0075-0.05, the resolution is 44.7-6.71 μm, and the relationship between the illumination numerical aperture and the resolution satisfies the formula d=0.61*λ2 / NA2, wherein d is the resolution, λ2 is the wavelength of the incident light, and NA2 is the illumination numerical aperture.

12. The optical method of claim 11, wherein for the flat area and the notch, the diameter of the illumination variable diaphragm is adjusted so that the illumination numerical aperture is 0.0075-0.02, and at this time the resolution is 44.7-16.78 μm; for the edge bevel, the diameter of the illumination variable diaphragm is adjusted so that the illumination numerical aperture is 0.01-0.03, and at this time the resolution is 33.55-11.18 μm; for the edge top and the edge bottom, the diameter of the illumination variable diaphragm is adjusted so that the illumination numerical aperture is 0.02-0.05, and at this time the resolution is 16.78-6.71 μm.

Images