Device and method for eliminating speckles in deep ultraviolet online wavefront detection

Abstract

The invention discloses a device and method for eliminating speckles in deep ultraviolet online wavefront detection, and belongs to the technical field of optical imaging system measurement, the device comprises a motor, a worm, a worm gear, a bearing, a base, a diffuser, a pressing ring and a cover plate; wherein the motor provides continuous power for the device and is fixed on the base; the worm is mounted on a rotor of the motor; the worm gear is meshed with the worm; the worm gear is mounted on the bearing; the bearing is mounted on the base; the cover plate is fixed on the base; and the diffuser is fixed on the worm gear through the pressing ring, so that the motor drives the diffuser to rotate quickly. According to the invention, ideal dodging and decoherence effects can be realized, precise regulation and control of rotation parameters of the diffuser can be realized, and the rotation acceleration, speed and direction of the diffuser can be regulated and controlled.

Description

Technical Field

[0001] The present invention belongs to the technical field of optical imaging system measurement, and in particular relates to a device and method for eliminating speckles in deep ultraviolet online wavefront detection. Background Art

[0002] In modern semiconductor manufacturing, the ability to achieve high-precision wavefront detection in deep ultraviolet (DUV) systems directly impacts chip manufacturing accuracy and yield. The light source is a crucial factor in DUV wavefront measurement, as its quality directly impacts the stability of the calibration process and the quality of the light spot received by the wavefront sensor. When calibrating the wavefront aberration of the objective lens under test, poor light spot quality can introduce significant systematic errors in the test results. Adding a diffuser to the optical system is a common approach to addressing light source inhomogeneity. Diffusers scatter the incident light beam, reducing the coherence of the non-interfering optical path and improving the uniformity of the light spot.

[0003] With the increasing demand for high-resolution optical systems like deep ultraviolet lithography machines, simply adding a traditional diffuser can no longer meet the current wavefront detection accuracy requirements. While advanced CGHs (computer-generated holograms) can meet the required accuracy, they are difficult and expensive to manufacture. Furthermore, existing technologies that use time-domain averaging methods, such as random vibration, to reduce the volatility of mechanical vibration shocks often require bulky devices that are difficult to install, have low integration levels, and suffer from poor adaptability. Summary of the Invention

[0004] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0005] A device for eliminating speckles in deep ultraviolet online wavefront detection includes: a motor, a worm, a worm gear, a bearing, a base, a diffuser, a pressure ring, and a cover plate; wherein the motor provides continuous power for the device and is fixed to the base; the worm is mounted on the rotor of the motor; the worm gear meshes with the worm; the worm gear is mounted on the bearing; the bearing is mounted on the base, and the cover plate is fixed to the base; the diffuser is fixed to the worm gear via the pressure ring, so that the motor drives the diffuser to rotate rapidly.

[0006] A method for eliminating speckle in deep ultraviolet online wavefront detection, used in the apparatus for eliminating speckle in deep ultraviolet online wavefront detection, comprising:

[0007] Step 1: Motor parameter setting; according to the formula , set the motor speed , rotation mode, drives the diffuser to rotate steadily, so that the randomly distributed microstructures on the diffuser surface continue to move, forming a dynamic scattering surface to prepare for subsequent beam scattering;

[0008] Step 2: Initialize the wavefront sensor; initialize the wavefront sensor and accurately set the exposure period and acquisition frame rate;

[0009] Step 3: Beam scattering: The diffuser is accurately placed at the focal point of the optical path, allowing the light beam to pass through the rotating diffuser. The diffuser's microstructure produces multi-angle diffuse reflection of the incident light.

[0010] Step 4: Image acquisition. During the exposure period of the wavefront sensor, a single frame image can capture multiple groups of light spot arrays formed by dynamic scattering. The wavefront sensor is controlled in real time to acquire images according to the exposure period and acquisition frame rate set in Step 2. During a single exposure, the captured light spot image data is promptly stored in a pre-created data container.

[0011] Step 5: Acquire the time-domain average light intensity. Due to the random distribution of the diffuser microstructure, the light intensity of the spot naturally undergoes statistical averaging in the time domain during a single exposure of the wavefront sensor. Using the single-frame image data collected by the sensor, the light intensity distribution after the time-domain statistical averaging is directly obtained.

[0012] Step 6: Data saving: Save the processed image data containing the time-domain average light intensity distribution.

[0013] The present invention has the following beneficial effects:

[0014] The present invention programs the motor controller to control the diffuser's rotation, utilizing time-domain averaging to eliminate the incident light source's speckle, achieving ideal uniform light distribution and decoherence. Closed-loop control allows precise regulation of the diffuser's rotational parameters, including acceleration, speed, and direction. Compared to static diffusers, the present invention more effectively suppresses fluctuations in the incident light source's intensity by rapidly rotating the diffuser, resolving the issue of traditional diffusers' inability to meet the wavefront detection accuracy requirements. Compared to the manufacturing difficulty and cost of CGH, the present invention's device is significantly less expensive and simpler to manufacture.

[0015] Through the mechanical design of the worm gear, the present invention can adapt to standard inch circular diffusers available on the market. At the same time, the device is compact and easy to install, with a compact structure, high integration, and high compatibility. It can quickly replace diffuser components without changing the optical path, and has good adaptability. The present invention is simple to use. It only requires installing the diffuser in the internal thread of the worm gear of the device in the existing optical path and turning on the motor drive in the host computer, effectively simplifying the experimental operation process. The present invention improves the accuracy of optical detection and is applicable to a variety of application scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1Schematic diagram of the structure of the device for eliminating speckles in deep ultraviolet online wavefront detection of the present invention, wherein 1-motor, 2-worm, 3-worm gear, 4-bearing, 5-base, 6-diffuser, 7-pressure ring, 8-cover plate;

[0017] Figure 2 Schematic diagram of the overall appearance of the device for eliminating speckles in deep ultraviolet online wavefront detection according to the present invention;

[0018] Figure 3 Schematic diagram of an experimental device for wavefront detection of an objective lens to be tested according to the present invention;

[0019] Figure 4 Schematic diagram of the uniform light effect of the device for eliminating speckles in deep ultraviolet online wavefront detection of the present invention; wherein, (a) is a schematic diagram of the uniform light effect when a diffuser is directly added, and (b) is a schematic diagram of the uniform light effect after the diffuser is rotated. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0021] This invention proposes a device for eliminating speckle in deep ultraviolet (DUV) online wavefront measurement. It employs a time-domain averaging method to eliminate light intensity fluctuations during wavefront measurement. Specifically, rotating frosted glass, a typical surface-structured diffuser, continuously changes the microstructure along the beam path through its physical motion (e.g., rotation), thereby averaging the light intensity distribution in the time domain and further reducing light intensity fluctuations. Under different pinhole sizes in the wavefront sensor's wavefront aberration calibration optical path, the diffraction spot exhibits intensity fluctuations within the wavefront sensor, indicating that the beam's uniformity is limited, affecting the wavefront aberration used to calibrate the wavefront sensor. The invention controls the diffuser 6 via a motor 1, enabling the physical motion of the diffuser 6 to effectively suppress light intensity fluctuations during wavefront measurement. Programming the motor controller allows for multi-parameter control of the diffuser 6 on a computer, enabling quantitative analysis of the experiment and laying the foundation for future automated testing. The device is compact and can be fabricated into an optical module for use in other scenarios, such as rotating glass slides.

[0022] Figure 1 Schematic diagram of the structure of the device for eliminating speckles in deep ultraviolet online wavefront detection of the present invention, Figure 2 FIG. 1 is a schematic diagram of the overall appearance of the device for eliminating speckles in deep ultraviolet online wavefront detection according to the present invention. Figure 1 、 Figure 2As shown, the device for eliminating speckle in deep ultraviolet online wavefront detection of the present invention includes a motor 1, a worm 2, a worm gear 3, a bearing 4, a base 5, a diffuser 6, a pressure ring 7, and a cover plate 8. The motor 1 provides continuous power for the device and is fixed to the base 5; the worm 2 is mounted on the rotor of the motor 1, and the two are fixed with screws; the worm gear 3 meshes with the worm 2; the worm gear 3 is mounted on the bearing 4, and the two are fixed by an interference fit; the bearing 4 is mounted on the base 5, and the cover plate 8 is fixed to the base 5 by screws; the diffuser 6 is fixed to the worm gear 3 by the pressure ring 7, so that the motor 1 can drive the diffuser 6 to rotate rapidly, achieving the purpose of eliminating optical speckle.

[0023] Motor 1 can be a DC motor. A PWM (Pulse Width Modulation) driver module and an encoder provide closed-loop speed control for motor 1, enabling precise regulation of the diffuser's rotational parameters, including acceleration, speed, and direction. Motor 1 can also be a stepper motor to meet precise control requirements in various application scenarios.

[0024] The diffuser 6 is made of frosted glass; the frosted glass is circular.

[0025] The worm gear 3 has a through-hole hollow channel with an internally threaded inner wall and a shoulder at the bottom. To eliminate speckle, the diffuser 6 is screwed into the hollow channel of the worm gear 3. The annular pressure ring 7, with a tightening thread on its inner wall, is screwed into the threaded section at the top of the worm gear 3, creating an axial preload between the lower end of the pressure ring 7 and the shoulder of the diffuser 6. Rotating the motor 1 drives the diffuser 6, eliminating speckle from the incident light source and achieving improved imaging quality.

[0026] The motor controller is programmed to control the diffuser 6 so that it rotates according to the program setting. The rotation speed in the program must be mathematically matched with the exposure time of the wavefront sensor. The mathematical matching formula is as follows:

[0027] ;

[0028] in, is the motor speed (rpm), which determines the update frequency of the microstructure on the surface of the diffuser 6, is the total number of effective microstructures on the diffuser surface, is the minimum number of valid states required within the exposure time, The exposure time (in seconds) of the wavefront sensor determines the integration time window.

[0029] The above formula can ensure that the diffuser microstructure completes a sufficient number of random state switching within the exposure time of the wavefront sensor, thereby effectively suppressing the fluctuation of light intensity and laying the foundation for realizing the time-domain averaging method to eliminate speckles. In the experimental device for wavefront detection of the objective lens to be tested, the optical path runs through the device for eliminating speckles in deep ultraviolet online wavefront detection of the present invention. Specifically, the motor 1 drives the diffuser 6 to rotate, and the randomly distributed microstructures (including pits and prism arrays) on the surface of the diffuser 6 form a dynamic scattering surface. When the light beam passes through the diffuser 6, the microstructure produces multi-angle diffuse reflection of the incident light. During the exposure cycle of the wavefront sensor, a single frame image captures multiple groups of light spot arrays formed by dynamic scattering. Due to the random distribution characteristics of the microstructure of the diffuser 6, the light spot intensity undergoes statistical averaging in the time domain during a single exposure process, and finally the wavefront sensor captures a light spot image with uniform light intensity distribution. This process realizes the engineering application of time-domain averaging method in wavefront detection through the synergistic effect of dynamic scattering and time integration, which can effectively reduce the light intensity fluctuation in wavefront calibration. By introducing dynamic scattering, the limitations of the diffuser itself can be significantly improved.

[0030] The specific steps for implementing the time-domain averaging method to eliminate speckles in deep ultraviolet online wavefront detection are as follows:

[0031] Step 1: Motor parameter setting; according to the formula , set the speed of motor 1 Parameters such as speed (rpm) and rotation mode drive the diffuser 6 to rotate stably, so that the randomly distributed microstructures (pits and prism arrays) on the surface of the diffuser 6 continue to move, forming a dynamic scattering surface to prepare for subsequent light beam scattering.

[0032] Step 2: Initialize the wavefront sensor: Initialize the wavefront sensor and accurately set key parameters such as exposure period and acquisition frame rate. Proper parameter settings ensure that the wavefront sensor can effectively capture light signals within the appropriate timescale, ensuring the accuracy and validity of the collected data.

[0033] Step 3: Light beam scattering; accurately place the diffuser 6 at the focus of the light path so that the light beam passes through the diffuser 6 in a rotating state; the microstructure of the diffuser 6 produces multi-angle diffuse reflection of the incident light.

[0034] Step 4: Image acquisition. During the wavefront sensor's exposure cycle, a single frame image can capture multiple groups of light spot arrays formed by dynamic scattering. The wavefront sensor is controlled in real time to acquire images according to the pre-set exposure cycle and acquisition frame rate. During a single exposure, the captured light spot image data is promptly stored in a pre-created data container.

[0035] Step 5: Acquire the time-domain average light intensity. Due to the random distribution characteristics of the diffuser 6 microstructure, the light intensity of the light spot naturally undergoes statistical averaging in the time domain during a single exposure of the wavefront sensor. Using the single-frame image data collected by the sensor, the light intensity distribution after the time-domain statistical averaging is directly obtained.

[0036] Step 6: Data saving: The program saves the processed image data containing the time-domain average light intensity distribution in a pre-set storage path and format.

[0037] like Figure 3 The schematic diagram of the experimental device for wavefront detection of the objective lens to be tested is shown. The experimental device for wavefront detection of the objective lens to be tested is composed of a light source, an illumination device, a diffuser, the objective lens to be tested, a pinhole and a sensor. The experiment uses a wavefront sensor to detect the light wave after the objective lens to be tested to realize wavefront detection. Starting from the light path, the light source passes through the illumination device and the device of the present invention to obtain a uniform collimated light wave, and then the collimated light wave converges through the objective lens to be tested. The light wave after the objective lens to be tested passes through the pinhole spatial filter, so that the wavefront sensor obtains a more uniform and smooth spot image. The light wave of the objective lens to be tested contains the wavefront information of the objective lens itself, such as optical aberration, etc. Therefore, the wavefront sensor can measure the wavefront aberration and other information of the light wave to realize wavefront detection for the objective lens to be tested. The advantage of the present invention is that the time domain averaging method realized by rotating the diffuser 6 controlled by a programmable motor can effectively reduce the light intensity fluctuation in wavefront calibration. By introducing dynamic scattering, the limitations of the diffuser itself can be significantly improved, providing more solutions for high-resolution optical systems.

[0038] like Figure 4 FIG. 1 is a schematic diagram showing the uniform light effect of the device for eliminating speckles in deep ultraviolet online wavefront detection according to the present invention. Figure 4 As shown in (a), if a traditional diffuser is directly added, the intensity distribution of the light spot of the image collected by the wavefront sensor will be uneven, which will affect the wavefront aberration detection of the objective lens to be tested. Figure 4 As shown in (b), after adding the device for eliminating speckles in deep ultraviolet online wavefront detection of the present invention and rotating the diffuser, the image collected by the wavefront sensor reduces the unevenness caused by the incident light source, ensuring the reliability of the incident light source. The light source is collimated ( Figure 3 The light (including the collimated light path) is incident on the objective lens and then passes through a pinhole spatial filter, making the wavefront more uniform and smooth. The resulting light spot detected by the wavefront sensor has a nearly completely uniform light field distribution, significantly improving uniformity compared to simply adding a diffuser 6. Furthermore, by using time-domain averaging to destroy the coherence of the light source, the speckle reduction device of this embodiment for online wavefront detection can effectively reduce image speckle.

[0039] As can be seen from the above description, the speckle reduction device provided by this invention for deep ultraviolet in-line wavefront detection combines a diffuser with time-domain averaging to achieve ideal light uniformity and decoherence effects while also boasting a compact structure, low cost, and high compatibility. This improves optical detection accuracy while also providing a convenient and economical solution for optimizing spot uniformity, suppressing coherent noise, and calibrating optical paths, making it suitable for a variety of application scenarios.

[0040] The above descriptions are merely embodiments of the present invention and are not intended to limit the scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention description and drawings, or directly or indirectly applied to other related system fields, are also included in the scope of protection of the present invention.

[0041] The contents not described in detail in the specification of the present invention belong to the prior art known to those skilled in the art.

Claims

1. A device for eliminating speckles in deep ultraviolet online wavefront detection, characterized in that: include: Motor, worm, worm gear, bearing, base, diffuser, pressure ring, cover plate; among them, the motor provides continuous power for the device and is fixed on the base; the worm is installed on the rotor of the motor; the worm gear engages with the worm; the worm gear is installed on the bearing; the bearing is installed on the base, and the cover plate is fixed to the base; the diffuser is fixed to the worm gear through the pressure ring, so that the motor drives the diffuser to rotate rapidly.

2. The device for eliminating speckles in deep ultraviolet online wavefront detection according to claim 1, characterized in that: The worm and the motor rotor are fixed with screws; the worm wheel and the bearing are fixed by interference fit.

3. The device for eliminating speckles in deep ultraviolet online wavefront detection according to claim 1, characterized in that: The motor is a DC motor, and closed-loop control of the motor speed is achieved through a PWM drive module and an encoder; or the motor is a stepping motor.

4. The device for eliminating speckles in deep ultraviolet online wavefront detection according to claim 1, characterized in that: The diffuser is made of frosted glass, and the frosted glass is circular.

5. The device for eliminating speckles in deep ultraviolet online wavefront detection according to claim 1, characterized in that: The worm wheel has a through hollow channel, the inner wall of the channel is processed with an internal thread structure, and a shaft shoulder is provided at the bottom of the inner wall of the channel; when it is necessary to eliminate the spots, the diffuser is screwed into the hollow channel of the worm wheel.

6. The device for eliminating speckles in deep ultraviolet online wavefront detection according to claim 5, characterized in that: The pressure ring adopts an annular structure, and a fastening thread is provided on its inner wall. By screwing into the threaded section at the top of the worm gear, an axial pre-tightening force is formed between the lower end face of the pressure ring and the diffuser shoulder. When the motor rotates, it drives the diffuser to rotate, eliminating the diffused spots of the incident light source.

7. The device for eliminating speckles in deep ultraviolet online wavefront detection according to claim 6, characterized in that: Write a program for the motor controller to control the diffuser so that the diffuser rotates according to the program settings. The mathematical matching formula between the rotation speed and the exposure time of the wavefront sensor in the program is as follows: ； in, is the motor speed, is the total number of effective microstructures on the diffuser surface, is the minimum number of valid states required within the exposure time, is the exposure time of the wavefront sensor.

8. The device for eliminating speckles in deep ultraviolet online wavefront detection according to claim 1, characterized in that: Programming the motor controller to control the diffuser to implement a time domain averaging method to eliminate the diffuse speckle of the incident light source. The time domain averaging method includes: Step 1: Motor parameter setting; according to the formula , set the motor speed , rotation mode, drives the diffuser to rotate steadily, so that the randomly distributed microstructures on the diffuser surface continue to move, forming a dynamic scattering surface to prepare for subsequent beam scattering; Among them, is the total number of effective microstructures on the diffuser surface, is the minimum number of valid states required within the exposure time, is the exposure time of the wavefront sensor; Step 2: Initialize the wavefront sensor; initialize the wavefront sensor and accurately set the exposure period and acquisition frame rate; Step 3: Beam scattering: The diffuser is accurately placed at the focal point of the optical path, allowing the light beam to pass through the rotating diffuser. The diffuser's microstructure produces multi-angle diffuse reflection of the incident light. Step 4: Image acquisition. During the exposure period of the wavefront sensor, a single frame image can capture multiple groups of light spot arrays formed by dynamic scattering. The wavefront sensor is controlled in real time to acquire images according to the exposure period and acquisition frame rate set in Step 2. During a single exposure, the captured light spot image data is promptly stored in a pre-created data container. Step 5: Acquire the time-domain average light intensity. Due to the random distribution of the diffuser microstructure, the light intensity of the spot naturally undergoes statistical averaging in the time domain during a single exposure of the wavefront sensor. Using the single-frame image data collected by the sensor, the light intensity distribution after the time-domain statistical averaging is directly obtained. Step 6: Data saving: Save the processed image data containing the time-domain average light intensity distribution.

9. The device for eliminating speckles in deep ultraviolet online wavefront detection according to claim 8, characterized in that: The motor drives the diffuser to rotate, and the randomly distributed microstructures on the diffuser surface form a dynamic scattering surface. When the light beam passes through the diffuser, the microstructure produces multi-angle diffuse reflection of the incident light. During the exposure period of the wavefront sensor, a single-frame image captures multiple groups of light spot arrays formed by dynamic scattering. Finally, the wavefront sensor collects a light spot image with uniform light intensity distribution. This process realizes the application of time-domain averaging method in wavefront detection through the synergistic effect of dynamic scattering and time integration.

10. A method for eliminating speckles in deep ultraviolet online wavefront detection, used in the device for eliminating speckles in deep ultraviolet online wavefront detection according to any one of claims 1 to 9, characterized in that: include: Step 1: Motor parameter setting; According to the formula , set the motor speed , rotation mode, drives the diffuser to rotate steadily, so that the randomly distributed microstructures on the diffuser surface continue to move, forming a dynamic scattering surface to prepare for subsequent beam scattering; Among them, is the total number of effective microstructures on the diffuser surface, is the minimum number of valid states required within the exposure time, is the exposure time of the wavefront sensor; Step 2: Initialize the wavefront sensor; initialize the wavefront sensor and accurately set the exposure period and acquisition frame rate; Step 3: Beam scattering: The diffuser is accurately placed at the focal point of the optical path, allowing the light beam to pass through the rotating diffuser. The diffuser's microstructure produces multi-angle diffuse reflection of the incident light. Step 4: Image acquisition. During the exposure period of the wavefront sensor, a single frame image can capture multiple groups of light spot arrays formed by dynamic scattering. The wavefront sensor is controlled in real time to acquire images according to the exposure period and acquisition frame rate set in Step 2. During a single exposure, the captured light spot image data is promptly stored in a pre-created data container. Step 5: Acquire the time-domain average light intensity. Due to the random distribution of the diffuser microstructure, the light intensity of the spot naturally undergoes statistical averaging in the time domain during a single exposure of the wavefront sensor. Using the single-frame image data collected by the sensor, the light intensity distribution after the time-domain statistical averaging is directly obtained. Step 6: Data saving: Save the processed image data containing the time-domain average light intensity distribution.

Images