Semiconductor graphics-free wafer dark field defect detection system and method
Through segmented variable speed spiral scanning and real-time adjustment of light intensity, spot size and TDI row frequency, the problem of mismatch between the delay integral camera and the scanning speed is solved, and the continuity and accuracy of wafer defect detection is achieved.
Patent Information
- Application Number
- CN202510934756.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-08
- Publication Date
- 2025-08-08
- Estimated Expiration
- 2045-07-08
AI Technical Summary
During the spiral scanning process, the sampling frequency of the delay integration camera does not match the scanning speed, resulting in problems with image acquisition and affecting the accuracy of wafer defect detection.
The semiconductor graphics-free wafer dark field defect detection system is adopted, including lighting light source components, beam-expanded shaping components, spiral motion control components, reflected light collection components and scattered light collection components. Through segmented variable speed spiral scanning and real-time adjustment of the light intensity, spot size and TDI row frequency, we ensure the continuity and consistency of the scanning process.
The continuity and consistency of wafer surface defect detection is achieved, the accuracy and efficiency of image processing is improved, image blur and artifact are reduced, and defect recognition capabilities are enhanced.
Smart Images

Figure CN120446150A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of semiconductor technology, and in particular to a semiconductor unpatterned wafer dark field defect detection system and method. Background Art
[0002] The semiconductor unpatterned wafer dark-field defect inspection system uses scattered light detection to detect defects on the wafer surface. During the entire wafer inspection process, the entire wafer surface must be scanned to obtain defect information. Currently, there are two mainstream scanning schemes: one uses an S-shaped scanning path in the XY coordinate system, and the other uses a spiral scanning scheme in the R and θ coordinate systems. This article discusses the second spiral scanning scheme.
[0003] During spiral scanning, a time-delayed integration (TDI) camera is often used for imaging. After capturing the image, further analysis is performed to identify the corresponding defects. However, during spiral scanning, if a fixed rotational speed is maintained, the linear velocity decreases as the scanning radius decreases. Consequently, if the TDI sampling frequency remains unchanged, a mismatch with the speed will occur, leading to image acquisition issues. Therefore, the present invention proposes a dark-field defect detection system and method for semiconductor unpatterned wafers. Summary of the Invention
[0004] The object of the present invention is to provide a semiconductor unpatterned wafer dark field defect detection system and method to solve the problems raised in the above background technology.
[0005] To achieve the above objectives, the present invention provides the following technical solution: a semiconductor unpatterned wafer dark field defect detection system, comprising: An illumination light source assembly includes a light source body, an attenuator mechanism and a controllable aperture arranged along the light propagation path, wherein the attenuator is used to control the attenuation of light intensity during wafer scanning, and the controllable aperture is used to control the size of the light spot according to the scanning process; The beam expansion and shaping component is used to expand and shape the laser beam provided by the illumination light source component to obtain a square strip-shaped light spot that is irradiated onto the wafer surface; The spiral motion control component is arranged under the wafer and is used to control the movement of the wafer to realize spiral scanning of the wafer; The spiral motion control component includes a rotary motion stage, a linear motion stage and a controller. The rotary motion stage, the linear motion stage and the controller are electrically connected. The rotary motion stage and the linear motion stage can realize spiral scanning of the wafer. The spiral scanning is divided into two sections from the outside of the wafer to the inside. The first section scans at a uniform linear speed with a uniform increase in angular speed. The second section scans at a uniform angular speed with a uniform decrease in linear speed. The light intensity attenuation and the change in spot size match the linear speed and angular speed during the scanning process. A reflected light collection component is used to collect the light path reflected from the wafer surface. The reflected light collection component is connected to a central control machine for analyzing and processing the reflected light information. The scattered light collecting component is used to collect the scattered light emitted by the wafer and transmit it to the image processor for processing the scattered light.
[0006] Furthermore, the light source body is configured as a UV light source, a DUV light source or an EUV light source.
[0007] Furthermore, a half-wave plate is arranged between the beam expansion and shaping component and the illumination light source component. The beam expansion and shaping component includes a beam expander, a collimator, a microlens array, and a focusing lens in sequence along the propagation direction of the light beam. The parallel light beam emitted by the illumination light source is diverged after passing through the beam expander mode, and is re-collimated into a parallel light source with a larger diameter after passing through the collimator. The parallel light is then incident on the microlens array to form a square strip-shaped light spot, which is finally focused by the focusing lens group and incident on the wafer surface.
[0008] Furthermore, the reflected light collection component is configured as a collector, which includes an attenuation OD plate and an energy detector PD, for collecting and analyzing the reflected light energy to ensure that the illumination laser energy obtained per unit pixel per unit time on the wafer remains consistent during the imaging process.
[0009] Furthermore, the scattered light collecting component includes an objective lens arranged above the wafer, and a first relay lens, a first beam splitter, a second beam splitter and a reflector are coaxially arranged in sequence along the light transmission direction.
[0010] Furthermore, three collection channels are set, namely the first channel, the second channel and the third channel. The light source from the first relay mirror is split into two beams of light after passing through the first beam splitter, one of which is transmitted to the first channel and the other is transmitted to the second beam splitter position. The light beam transmitted to the second beam splitter is split into two beams of light again and transmitted to the second channel and the third channel respectively.
[0011] Furthermore, a second relay mirror, a focusing lens and a detector are sequentially arranged in the first channel, the second channel and the third channel along the propagation direction of the light beam, and the detector is connected to the image processor, and the image processor is connected to the central control machine.
[0012] Furthermore, the illumination light source assembly also includes an attenuator and a controllable aperture, wherein the attenuator is configured as a linear polarizer controlled by a rotating motor, and the rotating motor is electrically connected to a motor controller. The rotating motor is started to drive the linear polarizer to rotate, thereby controlling the light power passing therethrough, thereby achieving controlled attenuation of the light intensity, so as to ensure that the illumination laser energy obtained per unit pixel per unit time on the wafer remains consistent during the imaging process.
[0013] Furthermore, the controllable aperture includes a left baffle, a right baffle, a coupling, a servo motor, a positive threaded rod and a negative threaded rod. Guide rails are provided below the left baffle and the right baffle, and are threadedly connected to the positive threaded rod and the negative threaded rod on the top, respectively. The positive threaded rod and the negative threaded rod are coaxially fixed and fixed to the output shaft of the servo motor through a coupling. When the servo motor is started, the positive threaded rod and the negative threaded rod are driven to rotate through the coupling. The threaded engagement relationship can be used to open or close the left baffle and the right baffle, thereby realizing the control of the width of the light spot.
[0014] According to a second aspect of the present invention, a method for detecting dark field defects in a semiconductor blank wafer is provided, which uses the semiconductor blank wafer dark field defect detection system described in the first aspect, and the specific steps are as follows: S1. Starting from the outer ring of the wafer, the linear motion stage and the rotary motion stage drive the wafer to begin uniform linear velocity scanning. At this time, the scanning spot size, TDI line frequency, and spot laser power remain unchanged, while the angular velocity of the rotary motion stage's rotation axis begins to gradually increase. S2. When entering the position with a radius of R1 from the rotation center, the rotation axis moves at a fixed angular velocity. At this time, the scanning spot size remains unchanged, the linear velocity begins to decrease, the TDI line frequency begins to gradually decrease, and the spot laser power also begins to decrease synchronously. S3. When entering the position with a radius of R2 from the rotation center, the radius of the scanning area from the wafer center is further reduced, and the linear velocity difference between the center of the spot and the edge of the spot is greater than 5%. At this time, the rotation axis still moves at a uniform angular velocity at the fixed angular velocity in step S2, and begins to reduce the TDI line frequency and the spot laser power. At the same time, the variable aperture is used to gradually reduce the spot size until the scan is completed.
[0015] The present invention has at least the following beneficial effects: The present invention optimizes the inner-circle scanning process of a wafer inspection system based on spiral scanning lines, adjusts the light intensity and size of the strip scanning spot in real time, and synchronously adjusts the line frequency of the delayed integration camera in real time, thereby achieving the continuity of the entire scanning process. The inner-circle scanning image is kept consistent with the outer-circle scanning image, facilitating subsequent image processing and recognition.
[0016] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 Schematic diagram of the overall structure of the detection system of the present invention; Figure 2 Schematic diagram of the structure of the beam expansion and shaping component in the present invention; Figure 3 Schematic diagram of the principle of spiral scanning of a wafer in the present invention; Figure 4 This is a schematic diagram showing the relationship between the spiral scanning time and the line frequency of the wafer in the present invention; Figure 5 Schematic diagram of external triggering of the time-delay integration camera in the present invention; Figure 6 This is a schematic diagram showing the relationship between the spiral scanning time and light intensity of the wafer in the present invention; Figure 7 Schematic diagram of the controllable aperture structure of the present invention; Figure 8 Schematic diagram of scanning of a light spot at a radius of R3 in an embodiment of the present invention; Figure 9 is a curve showing the change of the spot width with the scanning time in an embodiment of the present invention; Figure 10 Schematic diagram showing the relationship between the wafer spiral scanning time and the line frequency after correction in an embodiment of the present invention; Figure 11 This is a schematic diagram showing the relationship between the wafer spiral scanning time and the light intensity after correction in the first embodiment of the present invention; Figure 12 This is a schematic structural diagram of a collector in Example 1 of the present invention; Figure 13 This is a schematic diagram of the principle of collecting scattered light in each channel according to the first embodiment of the present invention; Figure 14 Schematic diagram of the structure of the controllable aperture in the second embodiment of the present invention; Figure 15 Schematic diagram of the detection method in Example 3 of the present invention. DETAILED DESCRIPTION
[0018] The following will be combined with the accompanying drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the embodiments described are only part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of the present disclosure.
[0019] Example 1: See also Figure 1The present invention provides a technical solution: a semiconductor unpatterned wafer dark field defect detection system, which is applied to wafer defect detection, comprising: The illumination light source assembly includes a light source body 101, an attenuator mechanism 501 and a controllable aperture 503 arranged along the light propagation path, wherein the attenuator mechanism 501 is used to control the attenuation of light intensity during the wafer scanning process, and the controllable aperture 503 is used to control the size of the light spot according to the scanning process; The beam expansion and shaping component is used to expand and shape the laser beam provided by the illumination light source component to obtain a square strip-shaped light spot that is irradiated onto the wafer surface; The spiral motion control component is arranged under the wafer and is used to control the movement of the wafer to realize spiral scanning of the wafer; The spiral motion control component includes a rotary motion stage 108, a linear motion stage and a controller 106. The rotary motion stage 108, the linear motion stage 107 and the controller 106 are electrically connected. The rotary motion stage 108 and the linear motion stage 107 can realize spiral scanning of the wafer. The spiral scanning is divided into two sections from the outside of the wafer to the inside. The first section scans at a uniform linear speed with a uniform increase in angular speed. The second section scans at a uniform angular speed with a uniform decrease in linear speed. The light intensity attenuation and the change in spot size match the linear speed and angular speed during the scanning process. A reflected light collection component is used to collect the light path reflected from the wafer surface. The reflected light collection component is connected to the central control computer 124 and is used to analyze and process the reflected light information to ensure that the illumination laser energy obtained per unit pixel per unit time on the wafer remains consistent during the imaging process; The scattered light collecting component is used to collect the scattered light emitted by the wafer and transmit it to the image processor for processing the scattered light.
[0020] The light source body 101 (not shown in the figure) is used to provide the illumination beam required for detection. The light source body 101 (not shown in the figure) can adopt various types, such as LED light source, laser light source or plasma light source, etc. Correspondingly, the wavelength range of the light source can also be selected according to specific application requirements, including wide spectrum, ultraviolet light (UV), deep ultraviolet light (DUV) or extreme ultraviolet light (EUV) and other common bands in semiconductor detection equipment. This embodiment does not make specific restrictions here, and can be selected according to actual conditions. With respect to the technical solution of this embodiment, for the convenience of explanation, the light source body 101 of this embodiment preferably adopts a DUV light source with a single wavelength continuous output, which is specifically configured as a linearly polarized The linearly polarized light emitted by the DUV light source is first incident on the half-wave plate 102. By precisely rotating the angle of the half-wave plate 102, the polarization direction of the transmitted light beam can be controlled to obtain the specific polarization angle required for irradiating the wafer surface. The polarization-modulated light beam then enters the beam expansion and shaping component 200, which performs optical processing including beam expansion, collimation, focusing, and light field distribution shaping on the light beam. In this embodiment, the output light beam processed by the beam expansion and shaping component 200 forms a specific spot shape on the wafer surface, preferably a square strip-shaped focused spot with uniform energy distribution. The wafer inspection process is completed by scanning the focused spot on the wafer.
[0021] Regarding the technical solution of this embodiment, Figure 2 As shown, the beam expansion and shaping assembly 200 includes, in sequence along the beam propagation direction: a beam expander 201, a collimator 202, a microlens array 203, and a focusing lens group 204. Its function is to shape and focus the incident beam into a focused spot with a specific shape (such as a square strip-shaped spot shape) and uniform energy distribution, and finally project it onto the wafer surface for illumination scanning; Specifically, the parallel light beam modulated by the upstream optical element half-wave plate 102 is first incident on the beam expander 201. The light beam processed by the beam expander 201 is in a divergent state. The divergent light beam is then incident on the collimator 202. After being re-collimated by the collimator 202, it is output as a parallel light beam with an increased diameter. The parallel light beam with an increased diameter is then incident on the microlens array 203. The microlens array 203 has a key structural feature: its central area is composed of densely arranged microlens units to form an effective light transmission area (or microlens working area), the shape of which is designed to match the target light spot morphology, and is preferably square in this embodiment; and the peripheral area surrounding the central effective light transmission area is The microlens array 203 is configured as a light-blocking area (or non-transparent area). Through this structural design, the microlens array 203 selectively transmits and splits the incident parallel light beam, allowing only the sub-beams that pass through the central square area to continue propagating. As a result, a square light spot (or intermediate light spot) corresponding to the shape of the effective light-transmitting area is initially formed on or near the output surface of the microlens array 203. Subsequently, the light beam emitted by the microlens array 203 (i.e., the light beam with a square outline) is incident on the focusing lens group 204. The focusing lens group 204 then performs final focusing on the square light beam, accurately adjusting its size to the specifications required for inspection, and then accurately projects the square strip-shaped focused light spot onto the surface of the wafer to be inspected. Compared with the technical solution of using a diffractive optical element (DOE) for beam shaping, the beam expansion and shaping component 200 adopted in this embodiment has significant advantages: the segmentation and superposition principle of the microlens array is conducive to achieving light field homogenization, and can form a square strip-shaped focused light spot with more uniform energy distribution. The highly uniform illumination spot directly acts on the wafer surface, which can significantly reduce image grayscale fluctuations or artifacts caused by uneven illumination; the highly uniform illumination spot is conducive to the subsequent time delay integration (TDI) camera to collect reflected light or scattered light signals. The uniform illumination ensures that the signal intensity received by different pixels in the TDI camera during the line-by-line scanning integration process is more consistent, thereby effectively improving the contrast, clarity and signal-to-noise ratio of the final collected image.
[0022] Furthermore, the reflected light collecting component is set as a collector 104. After the incident light is reflected by the wafer surface, it is collected by the collector 104. The collector 104 transmits the information to the central control computer 124 for further information analysis and processing, such as Figure 12 As shown, the collector 104 includes an attenuation OD plate 1041 and an energy detector PD1042, which are used to collect and analyze the reflected light energy to ensure that the illumination laser energy obtained per unit pixel per unit time on the wafer remains consistent during the imaging process. On the one hand, the collector 104 can prevent the reflected light from affecting the detection results after multiple reflections in the system. On the other hand, the collected light can be further analyzed, such as the spot shape, energy and other information, so that the system can monitor the incident light system in real time.
[0023] In accordance with the technical solution of this embodiment, the spiral motion control assembly is used to drive the wafer-carrying stage (not shown) to move so as to execute a specific scanning motion trajectory. In this embodiment, the spiral motion control assembly mainly includes: a linear motion stage 107, a rotary motion stage 108, and a controller 106. The controller 106 is electrically connected to the linear motion stage 107 and the rotary motion stage 108 (e.g., via a cable or bus) and coordinates the control. Specifically, the controller 106 is configured to: output a control signal to the linear motion stage 107 to drive it to perform reciprocating or unidirectional linear motion along a preset linear direction (e.g., the Z-axis direction); Output a control signal to the rotary motion stage 108 to drive it to rotate around a preset rotation axis (e.g., an axis parallel to the linear motion direction); coordinate the motion relationship between the displacement (speed) of the linear motion stage 107 and the angular displacement (angular velocity) of the rotary motion stage 108 in real time, so that the composite motion trajectory of the two motion stages forms an accurate spiral scanning curve.
[0024] With respect to the technical solution of this embodiment, the scattered light collection component includes an objective lens 109 arranged above the wafer, and a first relay lens 110, a first beam splitter 111, a second beam splitter 112 and a reflector 119 are coaxially arranged in sequence along the light transmission direction. The objective lens 109 must have the ability to capture scattered signals generated at a wide range of angles (including high angles) generated on the wafer surface, which is crucial for detecting tiny defects (such as nano-scale particles, scratches or graphic defects) because such defects often produce scattered signals with weak intensity and a wide angular distribution. In the solution of this embodiment, the use of an objective lens 109 with an NA of more than 0.9 ensures that as many scattered photons as possible can be collected, thereby significantly improving the signal intensity and signal-to-noise ratio (SNR), providing a basis for subsequent high-sensitivity detection. It should be noted that the objective lens may include various forms of objective lens designs, such as a high-NA refractive objective lens or a catadioptric objective lens with a reflective lens, which is not specifically limited in this embodiment.
[0025] Furthermore, three collection channels are provided, namely, a first channel, a second channel and a third channel. The light source from the first relay lens 110 is split into two beams of light after passing through the first beam splitter 111, one of which is transmitted to the first channel and the other is transmitted to the position of the second beam splitter 112. The light beam transmitted to the second beam splitter 112 is split into two beams of light again, which are transmitted to the second channel and the third channel respectively. The first channel, the second channel and the third channel are sequentially provided with a second relay lens 113, a focusing lens 114 and a detector 115 along the propagation direction of the light beam. The second relay lens 113 relays and optimizes the aberration of the received scattered light beam to ensure the beam quality and adapt to subsequent optical elements. The focusing lens 114 accurately focuses the relayed light beam onto the photosensitive surface of the corresponding detector. The detector 115 converts the focused scattered light signal into an electrical signal. The detector can be a high-sensitivity photomultiplier tube (PMT), an avalanche photodiode (APD) array, a scientific-grade CCD / CMOS or a sensor dedicated to a specific wavelength (such as DUV). The detector 115 is connected to the image processor 123, and the image processor 123 is connected to the central control computer 124, so as to facilitate further analysis after detecting each light beam. Figure 13 As shown, the three channels collect scattered light from three different areas respectively. The collected light is a large circular area in the entire field of view. The areas can be blocked separately by using masks, so that area one enters the first channel, area two enters the second channel, and area three enters the third channel. Specifically: The scattered light in region 1 (the central circular region) enters the first channel, mainly corresponding to the near-specular reflection direction or small-angle scattering; The scattered light from region 2 (the middle annular region) enters the second channel, which mainly corresponds to medium-angle scattering; The scattered light from region three (the outer ring region) enters the third channel, which mainly corresponds to large-angle scattering.
[0026] This spatial angle partitioning collection strategy enables different channels to specifically capture scattered light with angular distribution characteristics generated by different types of defects (such as tiny particles, shallow scratches, deep grooves, graphic defects, etc.), greatly enhancing the system's defect differentiation ability and classification accuracy.
[0027] It should be noted that the purpose of setting up the relay mirror is to lengthen the optical path length of the entire collection channel, thereby creating space for the placement of various optical devices in the optical path system, and also facilitating subsequent engineering design. Figure 1 The relay lens shown in the figure consists of only two lenses and is for illustration only. In actual situations, the design of relay lenses needs to minimize the aberration during light propagation to improve the imaging quality and facilitate better defect detection in the future.
[0028] It should be further explained that after the detector 115 collects the signal, it transmits it to the corresponding image processor 123, which processes the signal. Figure 1 The figure is only an example. In practice, the number of image processors can be matched according to the data volume of the detector. For example, the data volume of one detector can be equipped with one image processor, or the data volume of one detector can be equipped with multiple image processors 123. Finally, the central control machine 124 controls the entire system as a whole. The central control machine 124 is the carrier of the entire system software and the carrier of the human-computer interaction interface with the system users. The final defect detection results can be obtained through the central control machine 124 for further processing.
[0029] Regarding the technical solution of this embodiment, Figure 3 The specific principle of spiral scanning on the wafer is shown in the figure. The trajectory formed during the spiral scanning detection process is a scanning spiral line, and the illumination spot is a long square spot. The marks 301 and 302 in the figure are different positions of the same spot on the wafer, corresponding to the radii R1 and R2 from the center of the wafer respectively. Since the relationship between angular velocity and linear velocity obeys the following formula: (1) It can be seen from this that if the angular velocity (ω, i.e., rotational speed) is kept constant during the scanning process, the linear velocities at different radial positions (such as R1 and R2) will differ significantly (V1 / V2 = R1 / R2). However, the key sensors used in this system, such as the time delay integration camera (TDI), generally require the target's linear velocity (V) to remain constant when acquiring data along its scanning direction. This is because the charge transfer rate (line frequency) of the TDI camera is fixed. The linear velocity (V) must be precisely matched to the line frequency (i.e., V = pixel size × line frequency) to achieve blur-free cumulative integral imaging. Variations in linear velocity can cause image blur, stretching, or compression. Therefore, in order to obtain accurate and usable inspection data, it is ideal to ensure that the linear velocity (V) of the scan target relative to the detector remains constant throughout the entire scan path (constant linear velocity scanning).
[0030] In theory, achieving uniform linear velocity scanning requires that during the spiral scanning process, the angular velocity (ω) must continue to increase as the scanning radius (R) decreases (because ω = V / R). However, this requirement presents significant challenges in practical engineering implementation: Rotational mechanism limit: When the scan approaches the center of the circle (R approaches 0), the required angular velocity (ω) theoretically approaches infinity, which far exceeds the maximum speed capability of physical rotating mechanisms (such as motors and bearings); Dynamic performance and stability: Extremely high, continuously changing acceleration requirements place a significant burden on the motion control system, making it difficult to ensure smooth motion and positioning accuracy. These requirements can also cause vibrations, which in turn degrade image quality. System complexity and cost: Designing rotating shafts and drive systems to withstand extreme speeds and accelerations significantly increases engineering difficulty, complexity, and cost. In order to effectively resolve the contradiction between the above-mentioned sensor data acquisition requirements and the physical limitations of the rotating mechanism, this embodiment innovatively proposes a segmented variable speed spiral scanning control strategy. This strategy divides the entire spiral scanning trajectory into two sections with different motion characteristics. The first section is from the outer circle of the wafer to the radius R1 position. Within this section, the controller 106 dynamically adjusts the angular velocity ω of the rotating motion stage 108 so that the angle The linear velocity V gradually increases, and the linear velocity remains unchanged. Within a fixed scanning time interval, the arc length (linear velocity integral) of the scanning point movement is equal, but the rotation angle (angular displacement) gradually increases; the second section is from the radius R2 to the center of the circle. In this area, the controller 106 fixes the angular velocity ω of the rotating motion stage 108, so that the linear velocity V gradually decreases. Within a fixed scanning time interval, the rotation angle (angular displacement) of the scanning point movement is equal, but the arc length (linear velocity integral) gradually decreases.
[0031] The following uses the time-delay integration camera as an example to further discuss the detector design, focusing on the analysis of the second scan segment. In the second scan curve, the wafer rotates at a uniform angular velocity. The relationship between the speed limit, the line frequency f, and the pixel size PixelSize in the scanning direction is shown in the following formula: (2) Substituting equation (2) into equation (1), we can obtain the relationship between line frequency, angular velocity and radius: (3) The second stage starts from the radius of 25 mm and the line frequency changes linearly until the radius reaches 0.
[0032] As attached Figure 5 As shown, a time-delay integration camera can use external triggering to input the change of line frequency. Every time a trigger signal (lookup table input frequency) is received, the electron moves downward one pixel. The external signal is based on Figure 4 A lookup table (LUT) is established based on the relationship between the scanning time and line frequency, that is, the TDI line frequency change table over time. The specific line frequency relationship is linearly related to the scanning time. The TDI line frequency is input through the lookup table, so that TDI can realize variable line frequency acquisition, thereby ensuring that the line frequency matches the line speed.
[0033] During the dark field wafer inspection process, the presence and size of defects are determined by obtaining the energy of scattered photons from defects. Therefore, during the entire inspection process, it is necessary to ensure that the incident energy per unit time remains consistent to ensure that the scattered energy of the detected defects can accurately reflect the size of the defects. However, in the above-mentioned inspection process, the second stage uses uniform angular velocity motion, the linear velocity gradually decreases, and the scanning time per unit pixel becomes longer. If the incident laser energy remains unchanged during this process, the light intensity per unit pixel per unit time will increase, which makes it impossible to accurately judge the size of the scanned particles by the light intensity of the scattered signal. Figure 6 The figure shows the relationship between the change of light intensity and scanning time during the scanning process of the second stage, that is, the light intensity change with time. The figure shows that the light intensity and scanning time also have a linear change relationship. The light intensity uses normalized units, and the light intensity unit in the first stage scanning is set to 1.
[0034] like Figure 7 As shown, since the light intensity also needs to change all the time, an attenuator mechanism 501 needs to be placed in the optical path system to ensure that the illumination laser energy obtained per unit pixel per unit time on the wafer during the imaging process remains consistent. The attenuator mechanism 501 is a linear polarizer controlled by a rotating motor. The linear polarizer is installed at the output end of the rotating motor. The controller 502 is used to control the rotating motor. The incident light is a linear polarized light. By starting the rotating motor to drive the polarizer to rotate, the optical axis direction of the polarizer and the polarization direction of the incident light can be controlled. In the first scanning stage, the optical axis direction of the polarizer is made parallel to the polarization direction of the incident light. In the second scanning stage, the linear polarizer is controlled to rotate, thereby controlling the light power so that it is in accordance with the attached Figure 6 The formed scan time and light intensity lookup table is used to attenuate and control the light intensity; During the wafer inspection process, the incident light spot 301 has a certain width, such as 3mm. This is to ensure the yield of the entire system. A larger light spot will shorten the scanning time of a single wafer. However, a larger light spot will bring additional problems, such as the attached Figure 8 As shown in the figure, it is the situation when a spot is scanned at a radius of R3. As analyzed above, when in the inner circle, the system scans at a uniform angular velocity. Assuming the angular velocity is , as attached Figure 8 As shown in the figure, assuming that the linear velocities of the light spot at the far and near ends from the rotation center are V3 and V4, and the radii of the far and near ends are R3+1.5mm and R3-1.5mm respectively, the linear velocities V3 and V4 are respectively as follows according to formula (1): (4) (5) When R3=5mm, the linear velocity difference between V3 and V4 is 1.85 times according to the above formula. As described above, when scanning at variable line frequency, the TDI camera's line frequency is determined by the center position of the light spot. This will cause the linear velocity difference between the two ends relative to the center to reach more than 40%. In this case, the image scanned by TDI will appear clear in the middle and blurred at both ends.
[0035] To solve the above problem, it is necessary to adjust the spot width in the inner circle. For example, the spot width can be given a condition where the linear velocity difference between the two ends of the spot is less than 5%. The curve of spot width changing with scanning time can be obtained, as shown in the attached figure. Figure 9 shown.
[0036] Therefore, in the entire optical system, the spot size needs to be controlled according to the scanning process, as shown in the attached Figure 7 and Figure 14 As shown, the lighting source assembly also includes a controllable aperture 503. The controllable aperture 503 is driven by a servo motor 5031 to simultaneously drive two baffles to move synchronously. 504 is a controller of the controllable aperture 503. Specifically, the controllable aperture 503 includes two left and right baffles (left baffle 5037 and right baffle 5035), a coupling 5032, a servo motor 5031, a positive threaded rod 5033, and a negative threaded rod 5034. There is a guide rail 5036, which is threadedly connected to the positive thread rod 5033 and the negative thread rod 5034 on the upper side. The positive thread rod 5033 and the negative thread rod 5034 are coaxially fixed and fixed to the output shaft of the servo motor 5031 through the coupling 5032. When the servo motor 5031 is started, the positive thread rod 5033 and the negative thread rod 5034 are driven to rotate through the coupling 5032. The threaded engagement relationship can be used to open or close the left and right baffles, so that the left and right baffles can be opened or closed according to the attached Figure 9 The change curve in the image is used to form a lookup table. During the wafer inspection process, the spot width is controlled to achieve the function of spot width change scanning.
[0037] Due to the addition of the variable of spot change, if the entire wafer area still needs to be scanned, the scanning curve needs to be converted from an equidistant spiral to a non-equidistant spiral in the uniform angular velocity stage. In this case, the adjacent Figure 4 With attached Figure 6 The relationship between the TDI line frequency and the light intensity changing with the scanning time is corrected, and converted from a linear change relationship to a nonlinear change relationship, so that the matching can be guaranteed. The corrected curve of the TDI line frequency and the light intensity changing with the scanning time is shown in the attached figure. Figure 10 and Figure 11 shown.
[0038] In summary, this embodiment optimizes the inner circle scanning process of the wafer inspection system based on spiral scanning lines, adjusts the light intensity and size of the strip scanning spot in real time, and synchronously adjusts the line frequency of the delayed integration camera in real time, thereby achieving the continuity of the entire scanning process. The inner circle scanning image is consistent with the outer circle scanning image, which facilitates subsequent image processing and recognition.
[0039] Example 2: The fundamental difference between the second embodiment and the first embodiment is that the attenuator mechanism 501 is an acousto-optic deflector (AOD), which uses its first-order light to coincide with the optical axis, and the 0th-order light is attenuated using a light block. 502 is the controller of the acousto-optic deflector, which sends the required waveform signal and controls the attenuator mechanism 501 according to the attached Figure 6 The scan time and light intensity lookup table is formed and controlled in real time until the scan is completed.
[0040] Example 3: like Figure 15 As shown, this embodiment provides a semiconductor blank wafer dark field defect detection method, using a semiconductor blank wafer dark field defect detection system described in Example 1, and the specific steps are as follows: S1. The first stage of spiral scanning, the outer ring is scanned at a uniform linear speed: Starting from the outer circle of the wafer, the linear motion stage and the rotary motion stage drive the wafer to begin uniform linear speed scanning. At this time, the scanning spot size, TDI line frequency and spot laser power remain unchanged, and the angular velocity of the rotary motion stage's rotation axis begins to gradually increase. S2. The second stage of spiral scanning: the inner circle is scanned at a uniform angular velocity, and the linear velocity difference between the center and edge of the spot is less than 5%: When entering the position with a radius of R1 from the rotation center, the rotation axis moves at a fixed angular velocity. At this time, the scanning spot size remains unchanged, the linear velocity begins to gradually decrease, the TDI line frequency begins to gradually decrease, and the spot laser power also begins to gradually decrease synchronously. S3. The second stage of spiral scanning: the inner circle is scanned at a uniform angular velocity, and the difference in linear velocity between the center and edge of the spot is greater than 5%. When entering the position with a radius of R2 from the rotation center, the radius of the scanning area from the wafer center is further reduced, and the linear velocity difference between the center of the spot and the edge of the spot is greater than 5%. At this time, the rotation axis still moves at a uniform angular velocity at the fixed angular velocity in step S2, and begins to gradually reduce the TDI line frequency and the spot laser power. At the same time, the variable aperture is used to gradually reduce the spot size until the scanning is completed.
[0041] 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 variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.
[0042] For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances. When an element is referred to as being "assembled on", "installed on", "fixed on" or "set on" another element, it can be directly on the other element or there can be a central element. When an element is considered to be "connected" to another element, it can be directly connected to the other element or there can be a central element at the same time. The terms "vertical", "horizontal", "up", "down", "left", "right" and similar expressions used herein are for illustrative purposes only and are not intended to be the only embodiment.
[0043] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.
[0044] Throughout this specification, references to terms such as "one embodiment," "example," or "specific example" indicate that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present disclosure. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
Claims
1. A semiconductor unpatterned wafer dark field defect detection system, applied to wafer defect detection, characterized in that: include: An illumination light source assembly includes a light source body, an attenuator and a controllable aperture arranged along the light propagation path, wherein the attenuator is used to control the light intensity attenuation during the wafer scanning process, and the controllable aperture is used to control the light spot size according to the scanning process; The beam expansion and shaping component is used to expand and shape the laser beam provided by the illumination light source component to obtain a square strip-shaped light spot that is irradiated onto the wafer surface; The spiral motion control component is arranged under the wafer and is used to control the movement of the wafer to realize spiral scanning of the wafer; The spiral motion control component includes a rotary motion stage, a linear motion stage and a controller. The rotary motion stage, the linear motion stage and the controller are electrically connected. The rotary motion stage and the linear motion stage move in coordination to achieve spiral scanning of the wafer. The spiral scanning is divided into two sections from the outside of the wafer to the inside. The first section scans at a uniform linear speed with a uniformly increasing angular speed. The second section scans at a uniform angular speed with a uniformly decreasing linear speed. The light intensity attenuation and the change in spot size match the linear speed and angular speed during the scanning process. A reflected light collection component is used to collect the light path reflected from the wafer surface. The reflected light collection component is connected to a central control machine for analyzing and processing the reflected light information. The scattered light collecting component is used to collect the scattered light emitted by the wafer and transmit it to the image processor for processing the scattered light.
2. The semiconductor unpatterned wafer dark field defect detection system according to claim 1, characterized in that: The light source body is configured as a UV light source, a DUV light source or an EUV light source.
3. The semiconductor unpatterned wafer dark field defect detection system according to claim 2, characterized in that: A half-wave plate is arranged between the beam expansion and shaping component and the illumination light source component. The beam expansion and shaping component includes a beam expander, a collimator, a microlens array, and a focusing lens in sequence along the propagation direction of the light beam. The parallel light beam emitted by the illumination light source is diverged by the beam expander mode and re-collimated into a parallel light source with a larger diameter after passing through the collimator. The parallel light is then incident on the microlens array to form a square strip light spot. The light spot is finally focused by the focusing lens group and incident on the wafer surface.
4. The semiconductor unpatterned wafer dark field defect detection system according to claim 3, characterized in that: The reflected light collection component is configured as a collector, which includes an attenuation OD plate and an energy detector PD, and is used to collect and analyze the reflected light energy to ensure that the illumination laser energy obtained per unit pixel per unit time on the wafer remains consistent during the imaging process.
5. The semiconductor unpatterned wafer dark field defect detection system according to claim 4, characterized in that: The scattered light collecting component includes an objective lens arranged above the wafer, and a first relay lens, a first beam splitter, a second beam splitter and a reflector are coaxially arranged in sequence along the light transmission direction.
6. The semiconductor blank wafer dark field defect inspection system according to claim 5, characterized in that: The collecting channels are set to three, namely the first channel, the second channel and the third channel. The light source from the first relay mirror is split into two beams of light after passing through the first beam splitter, one of which is transmitted to the first channel and the other is transmitted to the second beam splitter position. The light beam transmitted to the second beam splitter is split into two beams of light again and transmitted to the second channel and the third channel respectively.
7. The semiconductor unpatterned wafer dark field defect detection system according to claim 6, characterized in that: A second relay mirror, a focusing lens and a detector are sequentially arranged in the first channel, the second channel and the third channel along the propagation direction of the light beam, and the detector is connected to the image processor, and the image processor is connected to the central control machine.
8. The semiconductor unpatterned wafer dark field defect detection system according to claim 7, characterized in that: The illumination light source assembly also includes an attenuator and a controllable aperture, wherein the attenuator is configured as a linear polarizer controlled by a rotating motor, and the rotating motor is electrically connected to a motor controller. The rotating motor is started to drive the linear polarizer to rotate, thereby controlling the light power passing through, and is used to achieve controlled attenuation of light intensity, so as to ensure that the illumination laser energy obtained per unit pixel per unit time on the wafer remains consistent during the imaging process.
9. The semiconductor unpatterned wafer dark field defect detection system according to claim 8, characterized in that: The controllable aperture includes a left baffle, a right baffle, a coupling, a servo motor, a positive threaded rod and a negative threaded rod. Guide rails are provided below the left baffle and the right baffle, and are threadedly connected to the positive threaded rod and the negative threaded rod on the top, respectively. The positive threaded rod and the negative threaded rod are coaxially fixed and fixed to the output shaft of the servo motor through a coupling. When the servo motor is started, the positive threaded rod and the negative threaded rod are driven to rotate through the coupling. The threaded engagement relationship can be used to open or close the left baffle and the right baffle, thereby realizing the control of the width of the light spot.
10. A semiconductor blank wafer dark field defect detection method, using a semiconductor blank wafer dark field defect detection system according to any one of claims 1 to 9, characterized in that: The specific steps are as follows: S1. Starting from the outer ring of the wafer, the linear motion stage and the rotary motion stage drive the wafer to begin uniform linear velocity scanning. At this time, the scanning spot size, TDI line frequency, and spot laser power remain unchanged, while the angular velocity of the rotary motion stage's rotation axis begins to gradually increase. S2. When entering the position with a radius of R1 from the rotation center, the rotation axis moves at a fixed angular velocity. At this time, the scanning spot size remains unchanged, the linear velocity begins to decrease, the TDI line frequency begins to gradually decrease, and the spot laser power also begins to decrease synchronously. S3. When entering the position with a radius of R2 from the rotation center, the radius of the scanning area from the wafer center is further reduced, and the linear velocity difference between the center of the spot and the edge of the spot is greater than 5%. At this time, the rotation axis still moves at a uniform angular velocity at the fixed angular velocity in step S2, and begins to reduce the TDI line frequency and the spot laser power. At the same time, the variable aperture is used to gradually reduce the spot size until the scan is completed.
Citation Information
Patent Citations
TDI-based line scanning dark field scattering wafer surface defect detection device and method
CN116359249A
Scanning method and device for wafer imaging, storage medium and electronic equipment
CN118501156A
Motion control method and system for non-graphic wafer defect detection
CN119727513A
Camera pulse triggering method and device, equipment and storage medium
CN120075427A
Laser beam reshaper
CN208713119U
Cited By
Defect detection equipment for wafer surface
CN121830717A