An apparatus and method for separating errors in detecting and adjusting distortion of a lithographic objective lens

Abstract

The present invention discloses a distortion detection and adjustment error separation device and method for a lithographic objective lens. The device includes an illumination system, a mask, a mask stage, a temperature and pressure sensor, a lithographic objective lens, a Shack-Hartmann sensor, a sensor moving stage, an interferometer, and an environmental control system. Along the laser illumination direction of the illumination system, there are successively a mask mounted on the mask stage, a lithographic objective lens with a temperature and pressure sensor, a Shack-Hartmann sensor for detecting the wavefront signal, a sensor moving stage, and a measurement system and an environmental control system composed of an interferometer. After the Shack-Hartmann sensor mounted on the sensor moving stage is aligned with the test wavefront, the three-dimensional coordinates of the test wavefront are obtained by the interferometer returning an ideal spherical wave passing through 11×11 square holes of the mask and the lithographic objective lens, and the coordinate information of the test wavefront is obtained. The distortion caused by the adjustment error is separated through coordinate transformation, and the data with the adjustment error separated is added into the calculation model to obtain the actual distortion of the lithographic objective lens.

Description

Technical Field

[0001] The present invention belongs to the technical fields of optical measurement and lithography machines, and particularly relates to a distortion detection and adjustment error separation device and method for a lithography objective lens. Background Art

[0002] In the field of semiconductor manufacturing, the complexity of integrated circuits on chips increases with Moore's law, and the feature linewidth of mask patterns is shrinking. To ensure that the mask pattern can be accurately projected onto the wafer, as an important component of the lithography machine, the projection lens must keep the distortion within the tolerance range to ensure the imaging quality, and it is necessary to detect the distortion of the lithography projection lens. Distortion represents the deviation between the actual position and the ideal position of different field points on the imaging surface, which is mainly caused by level errors, lens magnification, and aberration errors.

[0003] Three common methods for detecting the distortion of a lithography projection lens are exposure measurement method, aerial image measurement method, and wavefront measurement method. Litel Corporation proposed a self-reference method for measuring the distortion of a projection lens in 2003, using the Box-in-Box exposure measurement method, which can reduce the measurement repeatability accuracy to below 1 nm. The aerial image measurement method convolves the space formed by the alignment marks on the sensor stage and the mask marks, converts the light intensity signal into an electrical signal, obtains the imaging offset through analog-to-digital conversion, and finally fits to obtain the distortion parameters to detect the distortion. Nikon and ASML have both developed this measurement technology in this way. The difference is that Nikon's Hagiwara et al. use slit scanning along the x-axis and y-axis to obtain the aerial image, while ASML's van der Laan et al. use specific transmissive image sensor (TIS) marks and marks on the wafer plane to align and scan to obtain the aerial image. The wavefront measurement method detects the wavefront information of an ideal spherical wave after passing through the mask and the projection objective lens, calculates the actual image point offset between the imaging surface and the ideal image point using the wavefront information, and then fits the distortion parameters. Tichenor et al. proposed a distortion detection method based on a point diffractometer in 2001, and ASML's Flagello et al. proposed a distortion detection method based on shear interference in 2003.

[0004] In the present invention, a method for detecting the distortion of a projection lens using wavefront measurement is proposed. The laser of the illumination system generates an ideal spherical wave after passing through the mask and the lithography objective lens. The sensor stage moves to align the Shack-Hartmann sensor with the coordinate data of 11×11 image points. By using an interferometer to record the three-dimensional coordinate information of the Shack-Hartmann sensor, an overdetermined system of equations is established and fitted using the least squares algorithm to separate the error amount of inaccurate distortion measurement caused by adjustment errors, and the actual distortion in the lithography objective lens is calculated. Summary of the Invention

[0005] In order to separate the error amount of inaccurate distortion measurement caused by adjustment errors and calculate the actual distortion in a lithographic objective lens, we propose a device and method for separating adjustment errors in lithographic objective lens distortion detection.

[0006] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0007] A device for separating adjustment errors in lithographic objective lens distortion detection, the device includes an illumination system, a mask, a mask stage, a temperature and pressure sensor, a lithographic objective lens, a Shack-Hartmann sensor, a sensor moving stage, an interferometer and an environmental control system. Among them, the illumination system is located directly above the device to provide laser illumination. The mask with a specific pattern, the mask stage carrying the mask, the lithographic objective lens, and the sensor moving stage responsible for scanning movement are on the same vertical axis but parallel to each other. The temperature and pressure sensor is placed inside the lithographic objective lens to monitor temperature and pressure. The illumination system, mask, mask stage, temperature and pressure sensor, lithographic objective lens, Shack-Hartmann sensor, sensor moving stage, and interferometer are all in the environmental control system. The Shack-Hartmann sensor is installed on the sensor moving stage for alignment, and the interferometer is installed on the sensor moving stage to detect position coordinate information. Before and after the Shack-Hartmann sensor installed on the sensor moving stage aligns the test wave, the three-dimensional coordinates of the test wavefront are obtained by the ideal spherical wave returned by the interferometer passing through 11×11 square holes on the mask and the lithographic objective lens. The distortion caused by adjustment errors is separated through coordinate transformation, and the data with the adjustment errors separated is added into the distortion calculation model to obtain the actual distortion of the lithographic objective lens.

[0008] The pattern of the mask is a 5μm-sized square hole. The mask size is 110mm×110mm, with 11×11 square holes distributed. The light passes through the square holes to form an ideal spherical wave.

[0009] The lithographic objective lens is equipped with a temperature and pressure sensor, which can real-time feedback the temperature and pressure information inside the lithographic objective lens, so as to dynamically adjust the internal temperature and pressure to keep it constant.

[0010] The Shack-Hartmann sensor measures the position of the test wavefront after the ideal spherical wave passes through the mask and the lithographic objective lens, and the measurement system composed of the sensor moving stage and the interferometer obtains the three-dimensional coordinates and wave aberration information of the test wavefront.

[0011] The sensor moving stage is a five-dimensional moving stage, which can adjust the amount of movement, which are the amounts of movement of the sensor moving stage along the x, y, and z directions and the rotation angles in the xz and yz directions respectively, so as to adjust the distortion caused by adjustment errors.

[0012] The environmental control system keeps the entire test environment in a closed environment with constant temperature, pressure, humidity, and air flow, reducing the influence of environmental factors on the measurement.

[0013] A method for separating adjustment errors in lithographic objective lens distortion detection includes the following steps:

[0014] Step 1: In the environmental control system, keep the entire test environment in a closed environment with constant temperature, pressure, humidity, and air flow. The illumination system generates a laser that passes through 11×11 square holes on the mask stage to generate an ideal spherical wave. After passing through the lithographic objective lens, the Shack-Hartmann sensor on the workpiece senses the wavefront information, and the interferometer returns the three-dimensional coordinates of the wavefront information.

[0015] Step 2: Due to mask tilt, workpiece tilt, and the surface shape of the lithographic objective lens itself, the actual image plane and the ideal image plane of the actual test are inconsistent. Therefore, adjustment errors will cause inaccurate distortion measurement. After converting the coordinates of the test wavefront information, the tilt angle can be obtained.

[0016] Step 3: The sensor moving stage drives the Shack-Hartmann sensor to move and test 11×11 field points, and records the field point coordinates and wave aberration information through interference.

[0017] Step 4: Separate the actual wavefront coordinate error caused by adjustment errors, and perform noise filtering and gross error processing on the data.

[0018] Step 5: Through the data of 11×11 field points collected after separating the adjustment errors, construct the coordinate data into an overdetermined system of equations and add it to the distortion calculation model. Using the least squares algorithm for fitting, the actual distortion of the lithographic objective lens after separating the distortion caused by adjustment errors can be obtained.

[0019] The principle of the present invention is as follows: Through the alignment of the Shack-Hartmann sensor and the measurement by the interferometer, the coordinate data of 11×11 image points are obtained. Then, an overdetermined system of equations is established and the least squares algorithm is used for fitting to separate the error amount of inaccurate distortion measurement caused by adjustment errors, and the actual distortion in the lithographic objective lens is calculated.

[0020] Compared with the prior art, the advantages of the present invention are as follows: This method and device have a simple structure and low cost. It can measure the distortion of the lithographic objective lens without exposure, separate the error amount of inaccurate distortion measurement caused by adjustment errors, and calculate the actual distortion in the lithographic objective lens. Brief Description of the Drawings

[0021] Figure 1 It is a schematic diagram of a device for separating adjustment errors in lithographic objective lens distortion detection according to the present invention;

[0022] Figure 2It is a diagram of the selected mask size and pattern distribution;

[0023] Figure 3 It is a schematic diagram of the distortion principle caused by adjustment errors;

[0024] Figure 4 They are several common distortion vector diagrams;

[0025] Figure 5 It is a test flow chart of a method for separating adjustment errors in the distortion detection of a lithographic objective according to the present invention;

[0026] In the figure: 1 is the illumination system, 2 is the mask, 3 is the mask stage, 4 is the temperature and pressure sensor, 5 is the lithographic objective, 6 is the Shack-Hartmann sensor, 7 is the sensor moving stage, 8 is the interferometer, and 9 is the environmental control system. Detailed implementation mode

[0027] In order to better illustrate the purpose, technical solution and advantages of the present invention, the following will be further described in detail with reference to the accompanying drawings, but the protection scope of the present invention should not be limited thereby.

[0028] As Figure 1 As shown, a device for separating adjustment errors in the distortion detection of a lithographic objective includes an illumination system 1, a mask 2, a mask stage 3, a temperature and pressure sensor 4, a lithographic objective 5, a Shack-Hartmann sensor 6, a sensor moving stage 7, an interferometer 8 and an environmental control system 9. Among them, the illumination system 1 is located directly above the device to provide laser illumination. The mask 2 with a specific pattern, the mask stage 3 carrying the mask 2, the lithographic objective 5, and the sensor moving stage 7 responsible for scanning movement are on the same vertical axis but parallel to each other. The temperature and pressure sensor 4 is placed inside the lithographic objective 5 to monitor temperature and pressure. The illumination system 1, the mask 2, the mask stage 3, the temperature and pressure sensor 4, the lithographic objective 5, the Shack-Hartmann sensor 6, the sensor moving stage 7, and the interferometer 8 are all in the environmental control system 9. The Shack-Hartmann sensor 6 is installed on the sensor moving stage 7 for alignment, and the interferometer 8 is installed on the sensor moving stage 7 to detect position coordinate information;

[0029] Before the Shack-Hartmann sensor 6 installed on the sensor moving stage is aligned with the test wave, the three-dimensional coordinates of the test wavefront are obtained by the ideal spherical wave returned by the interferometer passing through 11×11 square holes on the mask and the lithographic objective. The coordinate information of the test wavefront is obtained, and the distortion caused by adjustment errors is separated through coordinate transformation. The data with the adjustment errors separated is added into the distortion calculation model to obtain the actual distortion of the lithographic objective.

[0030] The lithographic objective 5 is equipped with a temperature and pressure sensor 4, which can real-time feedback the temperature and pressure information inside the lithographic objective 5 to dynamically adjust the internal temperature and pressure to keep it constant.

[0031] The Shack-Hartmann sensor 6 tests the position of the wavefront after the ideal spherical wave passes through the mask and the photolithography objective 5, and the measurement system composed of the sensor motion stage 7 and the interferometer 8 obtains the three-dimensional coordinates and wave aberration information of the test wavefront.

[0032] The sensor motion platform 7 is a five-dimensional motion platform that can adjust The amount of exercise, They are respectively the movement amount of the sensor motion stage along the x, y, and z directions and the rotation angle amount in the xz and yz directions, thereby adjusting the distortion caused by the adjustment error.

[0033] The environmental control system 9 places the entire test environment in a closed environment where the temperature, pressure, humidity and airflow are kept constant, thereby reducing the impact of environmental factors on the measurement.

[0034] like Figure 2 As shown, the pattern of the mask is 5μm square holes with a spacing of 10mm. The mask size is 110mm×110mm, with 11×11 square holes distributed. Light passes through the square holes to form an ideal spherical wave. The sensor motion platform carries the Shack-Hartmann sensor to align the actual image points formed by the 11×11 square holes of the mask in an I-shaped scan on the imaging surface.

[0035] like Figure 3 As shown in the error adjustment principle diagram, the tilt will cause the actual test image points to be not on the same plane, which will have a certain impact on the calculated distortion and lead to inaccurate distortion measurement of the lithography objective. The three-dimensional coordinates of the actual image point can be expressed as:

[0036] (1)

[0037] in, , , are the three-dimensional coordinates of the test wavefront, is the distance between the image point and the zero point (the zero point is the center of the mask), is the aperture angle, is the elevation angle.

[0038] (2)

[0039] in, , , are the coordinates of the actual image point, is the change in aperture angle due to adjustment error, is the elevation angle change caused by the adjustment error, , It is the offset between the actual image point and the ideal image point caused by the adjustment error.

[0040] Figure 4 Several common distortions are given, such as Figure 4 the translation shown in (a), Figure 4 the rotation shown in (b), Figure 4 the magnification shown in (c), Figure 4 (d), Figure 4 the trapezoidal distortion shown in (e), Figure 4 the wedge distortion shown in (f), Figure 4 the radial distortion shown in (g). The adjustment error mainly affects the trapezoidal distortion and the wedge distortion. Specifically, it is reflected in the offset of the actual image point and the ideal image point in the x and y directions due to the tilt. The distortion calculation model is as follows:

[0041] (3)

[0042] Where:

[0043] , are the offsets of the ideal image point and the actual image point in the x and y directions after removing the offsets caused by the adjustment error, respectively;

[0044] , are the x and y coordinates of the actual image point after removing the offsets caused by the adjustment error, respectively;

[0045] , are the offsets in the x and y directions caused by the translation, respectively;

[0046] , are the offsets in the x and y directions caused by the magnification error, respectively;

[0047] , are the offsets in the x and y directions caused by the rotation, respectively;

[0048] , is the offset in the x direction caused by the trapezoidal distortion;

[0049] , is the offset in the y direction caused by the trapezoidal distortion;

[0050] , are the offsets in the x and y directions caused by the wedge distortion, respectively;

[0051] , The offsets in the x and y directions caused by radial distortion, respectively;

[0052] , The residuals in the x and y directions, respectively.

[0053] The test process of the method for separating adjustment errors in the distortion detection of a lithographic objective in the present invention is as Figure 5 shown:

[0054] Step 1: In the environmental control system, the entire test environment is placed in a closed environment where the temperature, pressure, humidity, and air flow are kept constant. The illumination system generates a laser, and an ideal spherical wave is generated through 11×11 square holes installed on the mask stage. After passing through the lithographic objective, the Shack-Hartmann sensor on the workpiece senses the wavefront information, and the interferometer returns the three-dimensional coordinates of the wavefront information;

[0055] Step 2: Due to the tilt of the mask, the tilt of the workpiece, and the surface shape of the lithographic objective itself, the actual image plane and the ideal image plane in the actual test are inconsistent. Therefore, the adjustment error will cause inaccurate distortion measurement. After converting the coordinates of the test wavefront information, the tilt angle can be obtained;

[0056] Step 3: The sensor moving stage drives the Shack-Hartmann sensor to move to test 11×11 field points and records the field point coordinates and wave aberration information through interference;

[0057] Step 4: Separate the actual wavefront coordinate error caused by the adjustment error, and perform noise filtering and gross error processing on the data;

[0058] Step 5: Through the data of 11×11 field points collected after separating the adjustment error, construct the coordinate data into an overdetermined system of equations and add it to the distortion calculation model. Using the least squares algorithm for fitting, the actual distortion of the lithographic objective after separating the distortion caused by the adjustment error can be obtained.

Claims

1. A method for separating adjustment errors in lithographic objective lens distortion detection, using a device for separating adjustment errors in lithographic objective lens distortion detection, characterized in that: The device includes an illumination system (1), a mask (2), a mask stage (3), a temperature and pressure sensor (4), a lithography objective lens (5), a Shack-Hartmann sensor (6), a sensor moving stage (7), an interferometer (8), and an environmental control system (9). Among them, the illumination system (1) is located directly above the device to provide laser illumination. The mask (2) with a pattern, the mask stage (3) carrying the mask (2), the lithography objective lens (5), and the sensor moving stage (7) responsible for scanning movement are on the same vertical axis but parallel to each other. The temperature and pressure sensor (4) is placed inside the lithography objective lens (5) to monitor temperature and pressure. The illumination system (1), the mask (2), the mask stage (3), the temperature and pressure sensor (4), the lithography objective lens (5), the Shack-Hartmann sensor (6), the sensor moving stage (7), and the interferometer (8) are all within the environmental control system (9). The Shack-Hartmann sensor (6) is installed on the sensor moving stage (7) for alignment, and the interferometer (8) is installed on the sensor moving stage (7) to detect position coordinate information. The method includes the following steps: Step 1: In the environmental control system (9), make the entire test environment in a closed environment where temperature, pressure, humidity, and air flow are kept constant. The illumination system (1) generates laser light, and an ideal spherical wave is generated through 11×11 square holes of the mask (2) installed on the mask stage (3). After passing through the lithography objective lens (5), the wavefront information is tested by the Shack-Hartmann sensor (6) on the sensor moving stage (7), and the three-dimensional coordinates of the wavefront information are returned by the interferometer (8). Step 2: Due to the tilt of the mask (2), the tilt of the sensor moving stage (7), and the surface shape of the lithography objective lens (5) itself, the actual image plane and the ideal image plane in the test are inconsistent. Therefore, adjustment errors will cause inaccurate distortion measurement. After converting the coordinates of the tested wavefront information, the tilt angle can be obtained. Step 3: The sensor moving stage (7) drives the Shack-Hartmann sensor (6) to move to test 11×11 field points and records the field point coordinates and wave aberration information through the interferometer (8). Step 4: Separate the actual wavefront coordinate error caused by adjustment errors, and perform noise filtering and gross error processing on the data. Step 5: Through the data of the 11×11 field points collected after separating the adjustment errors, construct an overdetermined system of equations with the coordinate data and add it to the distortion calculation model. Using the least squares algorithm for fitting, the actual distortion of the lithography objective lens after separating the distortion caused by adjustment errors can be obtained.

2. The method for separating adjustment errors in lithographic objective lens distortion detection according to claim 1, characterized in that: The pattern of the mask (2) is square holes with a size of 5 μm. The mask size is 110 mm×110 mm, and there are 11×11 square holes distributed. Light passes through the square holes to form an ideal spherical wave.

3. The method for separating adjustment errors in lithographic objective lens distortion detection according to claim 1, characterized in that: The lithography objective lens (5) is equipped with a temperature and pressure sensor (4), which can real-time feedback the temperature and pressure information inside the lithography objective lens (5) to dynamically adjust the internal temperature and pressure to keep it constant.

4. The method for separating adjustment errors in lithographic objective lens distortion detection according to claim 1, characterized in that: The Shack-Hartmann sensor (6) measures the positions of the test wavefront after an ideal spherical wave passes through the mask (2) and the lithographic objective lens (5). The measurement system composed of the sensor moving stage (7) and the interferometer (8) obtains the three-dimensional coordinates and wave aberration information of the test wavefront.

5. The method for separating adjustment errors in lithographic objective lens distortion detection according to claim 1, characterized in that: The sensor moving stage (7) is a five-dimensional moving stage capable of adjusting the amount of movement, which are respectively the amount of movement of the sensor moving stage in the x, y, and z directions and the rotation angles in the xz and yz directions, so as to adjust the distortion caused by the adjustment error.

6. The method for separating adjustment errors in lithographic objective lens distortion detection according to claim 1, characterized in that: The environmental control system (9) keeps the entire test environment in a sealed environment with constant temperature, pressure, humidity, and air flow, reducing the influence of environmental factors on the measurement.

7. The method for separating adjustment errors in lithographic objective lens distortion detection according to claim 1, characterized in that: The coordinate data of 11×11 image points are obtained through the alignment of the Shack-Hartmann sensor and the measurement of the interferometer. Then, an overdetermined system of equations and the least squares algorithm are established for fitting, separating the error amount of inaccurate distortion measurement caused by adjustment errors, and calculating the actual distortion in the lithographic objective lens.

Images