Polarization aberration compensation parameter acquisition method and system and photoetching machine

Abstract

The invention provides a polarization aberration compensation parameter acquisition method, a polarization aberration compensation parameter acquisition system and a photoetching machine, and relates to the technical field of polarization illumination. The method comprises the following steps: acquiring a Mueller matrix sequence and an environmental parameter sequence corresponding to a photoetching machine system, wherein the time sequences of the Mueller matrix sequence and the environmental parameter sequence are aligned; performing feature extraction on the Mueller matrix sequence and the environmental parameter sequence, and combining features of the same time sequence to obtain a feature tensor sequence; and determining a target compensation parameter according to the feature tensor sequence, wherein the target compensation parameter is used for compensating the polarization aberration of the photoetching machine system. The method can improve the accuracy of obtaining the polarization aberration compensation parameter, thereby further improving the overall compensation accuracy and imaging precision of the system.

Description

Technical Field

[0001] The present invention relates to the technical field of polarization illumination, and particularly to a method and system for obtaining polarization aberration compensation parameters and a lithography machine. Background Art

[0002] In advanced lithography technology, improving pattern resolution and overlay accuracy are the core driving forces for continuously shrinking the integrated circuit process nodes. Currently, high numerical aperture (NA) and polarization illumination technologies are the key technologies for improving the resolution of lithography machines. When the numerical aperture reaches 0.85 or above, polarization aberration cannot be ignored. Polarization aberration refers to the change in the polarization state of light waves on the exit pupil surface after polarized light passes through an optical system, which is mainly caused by the different transmission characteristics of light waves with different polarization states by each component inside the optical system, such as the intrinsic birefringence of optical materials, stress birefringence, and mask absorption effects. When the polarization aberration is large, it will seriously affect the reduction of the contrast and resolution of lithography patterns, especially the overlay accuracy. Therefore, polarization compensation for the optical system is required. Currently, lithography equipment usually models the measured Mueller matrix through Fourier expansion, Zernike polynomial fitting, etc., and extracts polarization aberration parameters to achieve subsequent system compensation.

[0003] Under actual working conditions, optical components may be affected by factors such as thermal drift and stress evolution, showing significant non-linear or time-varying behaviors. However, the Mueller matrix is essentially a linear system response model, and the existing methods for obtaining compensation parameters are difficult to accurately characterize these dynamic changes. Therefore, the accuracy of the obtained polarization aberration compensation parameters is not high. Summary of the Invention

[0004] The present invention provides a method and system for obtaining polarization aberration compensation parameters and a lithography machine, which are used to solve the defect of low accuracy in polarization aberration compensation in the prior art. The multi-source data composed of the Mueller matrix and environmental parameters are fused and features are extracted to obtain a sequence of feature tensors, and target compensation parameters are determined according to the sequence of feature tensors. It can improve the accuracy of obtaining polarization aberration compensation parameters, thereby further improving the overall system compensation accuracy and imaging accuracy.

[0005] The present invention provides a method for obtaining polarization aberration compensation parameters, including the following steps: Obtain a sequence of Mueller matrices and a sequence of environmental parameters corresponding to the lithography machine system, the time series of the sequence of Mueller matrices and the sequence of environmental parameters are aligned, the sequence of Mueller matrices is used to characterize the change of polarized light during transmission in the lithography machine system, and the sequence of environmental parameters is used to characterize the change of environmental parameters that affect the stability of the lithography machine system; Extract features from the sequence of Mueller matrices and the sequence of environmental parameters respectively, and combine the features at the same time series to obtain a sequence of feature tensors; Determine a target compensation parameter according to the sequence of characteristic tensors, where the target compensation parameter is used to compensate for the polarization aberration of the lithography machine system.

[0006] According to the acquisition of a polarization aberration compensation parameter provided by the present invention, the sequence of environmental parameters includes at least one of a temperature sequence, a pressure sequence, and a vibration sequence. The temperature sequence is used to characterize the temperature change of the lithography machine system, the pressure sequence is used to characterize the pressure change of the lithography machine system, and the vibration sequence is used to characterize the vibration acceleration change of the moving parts and / or optical parts of the lithography machine system.

[0007] According to the acquisition of a polarization aberration compensation parameter provided by the present invention, the sequence of environmental parameters includes the temperature sequence, the pressure sequence, and the vibration sequence. Feature extraction is respectively performed on the Mueller matrix sequence and the environmental parameter sequence, and the features at the same time series are combined to obtain a sequence of characteristic tensors, including: Perform principal component analysis dimensionality reduction processing on the Mueller matrix sequence to obtain a sequence of Mueller matrix eigenvectors after dimensionality reduction; Perform low-pass filtering processing on the temperature sequence and the pressure sequence respectively to obtain a filtered temperature sequence and a filtered pressure sequence; Perform sliding window aggregation on the vibration sequence to obtain an aggregated acceleration sequence; Normalize the sequence of Mueller matrix eigenvectors after dimensionality reduction, the filtered temperature sequence, the filtered pressure sequence, and the aggregated acceleration sequence, and then splice the features at the same time series to obtain the sequence of characteristic tensors.

[0008] According to the acquisition of a polarization aberration compensation parameter provided by the present invention, the Mueller matrix sequence includes at least one 4×4 Mueller matrix. Performing principal component analysis dimensionality reduction processing on the Mueller matrix sequence to obtain a sequence of Mueller matrix eigenvectors after dimensionality reduction includes: Expand each 4×4 Mueller matrix in the at least one 4×4 Mueller matrix into a 16-dimensional row vector by row priority to obtain at least one 16-dimensional row vector; Stack the at least one 16-dimensional row vector to form a data matrix, and extract the first K principal components from the data matrix to obtain the sequence of Mueller matrix eigenvectors after dimensionality reduction.

[0009] According to the acquisition of a polarization aberration compensation parameter provided by the present invention, the sampling periods corresponding to the temperature sequence and the pressure sequence are both the first sampling period. Performing low-pass filtering processing on the temperature sequence and the pressure sequence respectively to obtain a filtered temperature sequence and a filtered pressure sequence includes: Perform moving average filtering on the temperature sequence through a first sliding window to obtain the filtered temperature sequence; Perform moving average filtering on the air pressure sequence through a second sliding window to obtain the filtered air pressure sequence, where the time lengths of the first sliding window and the second sliding window are greater than or equal to two first sampling periods, and the moving average filtering includes taking the mean value of the values of at least two sampling points within the window as the value of the current sampling point; The second sampling period corresponding to the vibration sequence is less than the first sampling period, and the third sliding window is a non-overlapping sliding window. The sliding window aggregation of the vibration sequence to obtain the aggregated acceleration sequence includes: Determine that the time length of the third sliding window is equal to the first sampling period; Perform sliding aggregation on the vibration sequence through the third sliding window to obtain the processed vibration acceleration data. The sliding aggregation includes calculating the mean value and variance of the acceleration data within the window as the value of the current sampling point; Perform time series alignment on the processed vibration acceleration data according to the first sampling period to obtain the aggregated acceleration sequence.

[0010] According to the acquisition of a polarization aberration compensation parameter provided by the present invention, determining the corresponding target compensation parameter according to the feature tensor sequence includes: Input the feature tensor sequence into a trained compensation parameter prediction model, and output the target compensation parameter. The trained compensation parameter prediction model is obtained by inputting the measured Mueller matrix sequence and the environmental parameter sequence into an initial compensation parameter prediction model and optimizing the model parameters according to the mean square error between the measured imaging parameter corresponding to the predicted compensation parameter output by the model and the target imaging parameter.

[0011] According to the acquisition of a polarization aberration compensation parameter provided by the present invention, the trained compensation parameter prediction model includes a long short-term memory neural network model and a fully connected layer. The fully connected layer is used to fuse the time series features output by the long short-term memory neural network model and generate the target compensation parameter. The target compensation parameter includes at least one of the polarization plate angle compensation amount and the fast axis azimuth angle compensation amount of the quarter-wave plate.

[0012] According to the acquisition of a polarization aberration compensation parameter provided by the present invention, the target compensation parameter includes the polarization plate angle compensation amount and the fast axis azimuth angle compensation amount of the quarter-wave plate. The method further includes: Adjust the polarization plate angle and the fast axis azimuth angle of the quarter-wave plate in the polarization state generator according to the target compensation parameter.

[0013] Obtaining polarization aberration compensation parameters according to the present invention, the obtaining of the Mueller matrix sequence and the environmental parameter sequence corresponding to the lithography machine system includes: Obtain multiple sets of Stokes vectors at each sampling point according to the first sampling period, and perform least-squares fitting on the multiple sets of Stokes vectors to obtain the Mueller matrix corresponding to each sampling point. The Mueller matrices corresponding to multiple sampling points constitute the Mueller matrix sequence; Synchronously obtain the temperature parameter and the air pressure parameter corresponding to each sampling point according to the first sampling period. The temperature parameters and the air pressure parameters corresponding to multiple sampling points constitute the temperature sequence and the air pressure sequence; Obtain multiple vibration acceleration parameters according to the second sampling period. The multiple vibration acceleration parameters constitute the vibration sequence, and the second sampling period is less than the first sampling period.

[0014] The present invention also provides a system for obtaining polarization aberration compensation parameters, including the following modules: A data acquisition module for obtaining the Mueller matrix sequence and the environmental parameter sequence corresponding to the lithography machine system. The time sequences of the Mueller matrix sequence and the environmental parameter sequence are aligned. The Mueller matrix sequence is used to characterize the change of polarized light during transmission in the lithography machine system, and the environmental parameter sequence is used to characterize the change of environmental parameters that affect the stability of the lithography machine system; A feature acquisition module for respectively extracting features from the Mueller matrix sequence and the environmental parameter sequence and combining the features at the same time sequence to obtain a feature tensor sequence; A parameter prediction module for determining target compensation parameters according to the feature tensor sequence. The target compensation parameters are used to compensate for the polarization aberration of the lithography machine system.

[0015] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and running on the processor. When the processor executes the computer program, the method for obtaining polarization aberration compensation parameters as described in any one of the above is implemented.

[0016] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the method for obtaining polarization aberration compensation parameters as described in any one of the above is implemented.

[0017] The present invention also provides a computer program product, including a computer program. When the computer program is executed by a processor, the method for obtaining polarization aberration compensation parameters as described in any one of the above is implemented.

[0018] The method, system and lithography machine for obtaining polarization aberration compensation parameters provided by the present invention fuse multi-source data composed of Mueller matrices and environmental parameters and extract features to obtain a sequence of feature tensors, and determine target compensation parameters according to the sequence of feature tensors. It can improve the accuracy of obtaining polarization aberration compensation parameters, thereby further improving the overall compensation accuracy and imaging accuracy of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0020] Figure 1 It is a schematic flow chart of the method for obtaining polarization aberration compensation parameters provided by the present invention.

[0021] Figure 2 It is a schematic flow chart of the method for obtaining a sequence of Mueller matrices and a sequence of environmental parameters corresponding to a lithography machine system provided by the present invention.

[0022] Figure 3 It is a schematic structural diagram of the lithography machine system provided by the present invention.

[0023] Figure 4 It is a schematic flow chart of the method for obtaining a sequence of feature tensors provided by the present invention.

[0024] Figure 5 It is a schematic diagram of feature extraction for a sequence of Mueller matrices and a sequence of environmental parameters provided by the present invention.

[0025] Figure 6 It is a schematic overall flow chart of the method for obtaining polarization aberration compensation parameters provided by the present invention.

[0026] Figure 7 It is a schematic structural diagram of the polarization aberration compensation system provided by the present invention.

[0027] Figure 8 It is a schematic physical structure diagram of an electronic device provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0028] To make the objectives, technical solutions and advantages of the present invention clearer, the following will clearly and completely describe the technical solutions in the present invention in conjunction with the drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0029] In advanced lithography technology, improving pattern resolution and overlay accuracy are the core driving forces for continuously shrinking the integrated circuit process nodes. Currently, high numerical aperture (NA) and polarization illumination technologies are the key technologies to improve the resolution of lithography machines. When the numerical aperture reaches 0.85 or above, polarization aberration cannot be ignored. Polarization aberration refers to the change in the polarization state of the light wave on the exit pupil plane after polarized light passes through the optical system. It is mainly caused by the different transmission characteristics of the optical elements inside the optical system for light waves with different polarization states, such as the intrinsic birefringence of optical materials, stress birefringence, and mask absorption effects. When the polarization aberration is large, it will seriously affect and reduce the contrast and resolution of the lithography pattern, especially affecting the overlay accuracy. Therefore, polarization compensation for the optical system is required. Currently, lithography machines mainly use the Mueller Matrix modeling method to characterize and compensate the polarization transmission behavior of the optical system. The Mueller Matrix is a mathematical tool that describes the propagation and change of polarized light in an optical system. It is a 4×4 real number matrix used to represent the action of an optical element or system on a beam of light with an arbitrary polarization state. Since the Mueller Matrix describes the linear change from the incident polarized light to the outgoing polarized light, polarization compensation for the lithography machine illumination system can be achieved based on Mueller Matrix modeling. In current practical applications, lithography equipment usually models the measured Mueller Matrix through methods such as Fourier expansion and Zernike polynomial fitting, and extracts polarization aberration parameters to achieve subsequent system compensation.

[0030] However, under actual working conditions, optical elements may be affected by factors such as thermal drift and stress evolution, showing significant non-linear or time-varying behaviors. The Mueller Matrix is essentially a linear system response model, and the existing methods for obtaining compensation parameters are difficult to accurately characterize these dynamic changes. Therefore, the accuracy of the obtained polarization aberration compensation parameters is not high.

[0031] In view of this, an embodiment of the present invention provides a method for obtaining polarization aberration compensation parameters, including obtaining a sequence of Mueller Matrices and a sequence of environmental parameters corresponding to the lithography machine system, where the time series of the sequence of Mueller Matrices and the sequence of environmental parameters are aligned; respectively extracting features from the sequence of Mueller Matrices and the sequence of environmental parameters and combining the features at the same time series to obtain a sequence of feature tensors; determining target compensation parameters according to the sequence of feature tensors, where the target compensation parameters are used to compensate for the polarization aberration of the lithography machine system. This method can improve the accuracy of obtaining polarization aberration compensation parameters, thereby further improving the overall system compensation accuracy and imaging accuracy.

[0032] Next, the technical solutions in the embodiments of the present invention will be described in conjunction with the accompanying drawings in the embodiments of the present invention.

[0033] Figure 1It is a schematic flowchart of a method for obtaining polarization aberration compensation parameters provided by the present invention. The method for obtaining polarization aberration compensation parameters can be applied to a lithography machine system. As Figure 1 shown, the method may include the following steps 101 to 103: Step 101: Obtain a sequence of Mueller matrices and a sequence of environmental parameters corresponding to the lithography machine system. The timings of the sequence of Mueller matrices and the sequence of environmental parameters are aligned. The sequence of Mueller matrices is used to characterize the change of polarized light during transmission in the lithography machine system, and the sequence of environmental parameters is used to characterize the change of environmental parameters that affect the stability of the lithography machine system.

[0034] It should be noted that there are many methods for obtaining the sequence of Mueller matrices and the sequence of environmental parameters corresponding to the lithography machine system. For example, they can be obtained by collecting Mueller matrices and environmental parameters at multiple sampling points according to a fixed sampling period, or by receiving information transmitted from other devices. The present invention does not limit the method for obtaining the sequence of Mueller matrices and the sequence of environmental parameters corresponding to the lithography machine system.

[0035] Among them, the Mueller matrix can be collected by the lithography machine system, and the environmental parameters can be collected by a data acquisition module. The data acquisition module can include various types of sensors, and the type of sensor is determined according to the environmental parameters to be collected. Exemplarily, the environmental references can include temperature parameters, pressure parameters, etc.

[0036] Step 102: Respectively perform feature extraction on the sequence of Mueller matrices and the sequence of environmental parameters, and combine the features at the same timing to obtain a sequence of feature tensors.

[0037] It should be noted that there are many ways to respectively perform feature extraction on the sequence of Mueller matrices and the sequence of environmental parameters. For example, filtering methods, aggregation methods, dimensionality reduction methods, etc. can be used. The way to combine the features at the same timing for the extracted features can be feature splicing, fusion, etc. The present invention does not limit the way to respectively perform feature extraction on the sequence of Mueller matrices and the sequence of environmental parameters and combine the features at the same timing.

[0038] Step 103: Determine target compensation parameters according to the sequence of feature tensors. The target compensation parameters are used to compensate for the polarization aberration of the lithography machine system.

[0039] It should be noted that there are many ways to determine the target compensation parameters according to the sequence of feature tensors. For example, they can be obtained by inputting the sequence of feature tensors into a trained prediction model, or calculated according to the sequence of feature tensors and a preset algorithm, etc. The present invention does not limit the method for determining the target compensation parameters according to the sequence of feature tensors.

[0040] It is understandable that the existing methods for obtaining polarization aberration compensation parameters are difficult to capture the non-linear response mechanism because the Mueller matrix is essentially a linear system response model, which is applicable to the modeling of an optical system under steady-state conditions. However, under actual working conditions, optical components may be affected by factors such as thermal drift and stress evolution, showing significant non-linear or time-varying behavior, and it is difficult for existing modeling methods to accurately characterize these dynamic changes. The present invention provides a method for synchronously collecting a sequence of Mueller matrices and a sequence of environmental parameters according to a fixed sampling period, then performing feature extraction and combination to obtain a sequence of feature tensors, and then determining target compensation parameters according to the sequence of feature tensors. This method incorporates non-linear response into the Mueller matrix to determine target compensation parameters, which can improve the accuracy of obtaining polarization aberration compensation parameters, thereby further improving the overall compensation accuracy rate and imaging accuracy of the system.

[0041] In some embodiments, the sequence of environmental parameters may include at least one of a temperature sequence, a pressure sequence, and a vibration sequence. The temperature sequence is used to characterize the temperature change of the lithography machine system, the pressure sequence is used to characterize the pressure change of the lithography machine system, and the vibration sequence is used to characterize the vibration acceleration change of the moving parts and / or optical components of the lithography machine system.

[0042] It should be noted that determining the temperature change, pressure change, and vibration acceleration change of the lithography machine system has an obvious impact on the stability of the lithography machine system, that is, it has a clear impact on polarization aberration. Therefore, at least one of the temperature sequence, pressure sequence, and vibration sequence can be used as the sequence of environmental parameters to capture non-linear response characteristics. Among them, the vibration acceleration monitoring of the lithography machine system mainly targets the moving parts (wafer stage, mask stage) and / or the optical core components (projection objective, illumination system), and it is necessary to eliminate its impact on polarization aberration through high-frequency sampling and time-series modeling. Therefore, the vibration acceleration may include vibration data of at least one component of the wafer stage, mask stage, projection objective, or illumination system. It can improve the accuracy of obtaining polarization aberration compensation parameters, thereby further improving the overall compensation accuracy rate and imaging accuracy of the system.

[0043] Figure 2 is a schematic flowchart of the method for obtaining the sequence of Mueller matrices and the sequence of environmental parameters corresponding to the lithography machine system provided by the present invention. As Figure 2 shown, the obtaining of the sequence of Mueller matrices and the sequence of environmental parameters corresponding to the lithography machine system may include: Step 201: Obtain multiple groups of Stokes vectors at each sampling point according to the first sampling period, and perform least squares fitting on the multiple groups of Stokes vectors to obtain the Mueller matrix corresponding to each sampling point. The Mueller matrices corresponding to multiple sampling points constitute the sequence of Mueller matrices.

[0044] It should be noted that the Mueller matrix sequence can be obtained through the original lithography system. That is, at each sampling point corresponding to the fixed sampling period, i.e., the first sampling period, multiple sets of Stokes vectors are obtained, and the multiple sets of Stokes vectors are subjected to least squares fitting to obtain the Mueller matrix corresponding to each sampling point.

[0045] Exemplarily, the lithography system may include devices such as an illumination device and a polarization state generator.

[0046] Step 202: Synchronously obtain the temperature parameter and the pressure parameter corresponding to each sampling point according to the first sampling period. The temperature parameters and the pressure parameters corresponding to multiple sampling points constitute the temperature sequence and the pressure sequence.

[0047] Step 203: Obtain multiple vibration acceleration parameters according to the second sampling period. The multiple vibration acceleration parameters constitute the vibration sequence, and the second sampling period is less than the first sampling period.

[0048] It should be noted that the environmental parameter sequence includes a temperature sequence, a pressure sequence, and a vibration sequence. To obtain the environmental parameter sequence, corresponding sensor modules can be set in the lithography system, such as setting a temperature sensor, a pressure sensor, and an acceleration sensor, etc. While collecting the Mueller matrix, the environmental parameters are collected to obtain the environmental parameter sequence. The sampling periods of the temperature sequence and the pressure sequence are the same as the sampling period of the Mueller matrix, while the sampling frequency corresponding to the vibration acceleration is higher, so the sampling period is shorter than the sampling period of the Mueller matrix.

[0049] Figure 3 is a schematic structural diagram of the lithography system provided by the present invention. As Figure 3 shown, the lithography system is also a collection system for the Mueller matrix sequence and the environmental parameter sequence, and mainly includes the following modules: an illumination and lithography module, a data collection module, a polarization state analyzer (PSA), and a charge-coupled device (CCD) detector.

[0050] Among them, the illumination and lithography module includes: an illumination device: generating excimer laser; a polarization state generator (PSG): composed of a polarizer and a quarter-wave plate, used to generate a linearly polarized light beam; a projection objective: used to reduce and transfer the circuit pattern on the mask to the surface of the wafer; a collimating mirror: used to adjust the light beam output by the projection objective into a parallel light to ensure uniform energy distribution and consistent angle in the optical path.

[0051] The data collection module includes: temperature, pressure, and acceleration sensors, respectively used to measure information such as the temperature, pressure, and vibration acceleration of the lithography system; A polarization state analyzer and a charge-coupled device detector are used to detect the parameters of the Stokes vector. By measuring multiple sets of Stokes vectors and through fitting calculations, the Mueller matrix at the current moment can be obtained.

[0052] It can be understood that obtaining the Mueller matrix and environmental parameters at each sampling point according to a fixed sampling period lays a foundation for subsequent feature extraction and combination, and also provides conditions for obtaining accurate target compensation parameters.

[0053] Figure 4 It is a schematic flowchart of the method for obtaining a feature tensor sequence provided by the present invention. As Figure 4 shown, the environmental parameter sequence includes the temperature sequence, the air pressure sequence, and the vibration sequence. The feature extraction of the Mueller matrix sequence and the environmental parameter sequence respectively and the combination of the features at the same time series to obtain a feature tensor sequence may include: Step 301: Perform principal component analysis dimensionality reduction processing on the Mueller matrix sequence to obtain a reduced-dimensional Mueller matrix eigenvector sequence.

[0054] It should be noted that the principal component analysis (PCA) dimensionality reduction method can be used. That is, the Mueller matrix is a 4×4 real matrix (16 elements), but there are strong physical constraints and correlations among the actual elements. Therefore, dimensionality reduction can be first performed through PCA (principal component analysis method) to remove redundant information.

[0055] Step 302: Perform low-pass filtering processing on the temperature sequence and the air pressure sequence respectively to obtain a filtered temperature sequence and a filtered air pressure sequence.

[0056] It should be noted that the low-pass filtering method can be used. That is, low-pass filtering is performed on parameters with a relatively large change time scale, such as temperature and air pressure, etc., to smooth out some short-term fluctuations and noises and reduce the interference of minute fluctuations.

[0057] Step 303: Perform sliding window aggregation on the vibration sequence to obtain an aggregated acceleration sequence.

[0058] It should be noted that the sliding aggregation method can be used. That is, window aggregation is performed on the high-frequency vibration acceleration information with a relatively small change time scale for time scale alignment.

[0059] Step 304: Normalize the reduced-dimensional Mueller matrix eigenvector sequence, the filtered temperature sequence, the filtered air pressure sequence, and the aggregated acceleration sequence, and then splice the features at the same time series to obtain the feature tensor sequence.

[0060] It should be noted that after obtaining the dimension-reduced Mueller matrix eigenvector sequence, the filtered temperature sequence, the filtered air pressure sequence, and the aggregated acceleration sequence, the processed features can be normalized, and the features at the same time series are combined into a three-dimensional input tensor: ; Among them, the feature tensor sequence at each time point includes: ; In the above formula, is the dimension-reduced Mueller matrix eigenvector, , are the temperature value and air pressure value after low-pass smoothing filtering, and are the mean and variance values of the acceleration after aggregation.

[0061] It can be understood that the existing methods ignore the component coupling effect, that is, the optical components are independently adjusted for different aberration components (such as depolarization, retardation), ignoring the coupling effect of different components and lacking the ability to adaptively process complex situations. The method for obtaining the feature tensor sequence provided by the present invention combines multiple features to form a feature tensor sequence, which can improve the accuracy of the obtained polarization aberration compensation parameters.

[0062] Furthermore, the Mueller matrix sequence includes at least one 4×4 Mueller matrix. The principal component analysis dimension reduction process for the Mueller matrix sequence to obtain the dimension-reduced Mueller matrix eigenvector sequence may include: expanding each 4×4 Mueller matrix in the at least one 4×4 Mueller matrix into a 16-dimensional row vector by row priority to obtain at least one 16-dimensional row vector; stacking the at least one 16-dimensional row vector to form a data matrix, and extracting the first K principal components from the data matrix to obtain the dimension-reduced Mueller matrix eigenvector sequence.

[0063] It should be noted that the dimension reduction process of the Mueller matrix may include the following steps: (1) Each 4×4 Mueller matrix M(t n ) is expanded into a 16-dimensional row vector m(t n ), m(t n ) = [M 11 (t n ), M 12 (t n ), …, M 14 (t n ), M 21 (t n ), …, M 44 (t n )] ∈ R 1×16; (2) The data matrix X is vertically stacked by the unfolded vectors at N moments: , where represents the real number field, represents a real matrix with N rows and 16 columns. Multiply the centralized data matrix with the projection matrix V K ∈ R 16×K to obtain the sequence of Mueller matrix eigenvectors after dimensionality reduction, as shown in the following formula. It can be seen that the original 16-dimensional eigenvectors are reduced to K dimensions.

[0064] .

[0065] It can be understood that the method of dimensionality reduction processing of the Mueller matrix can extract key features and improve the accuracy of the obtained polarization aberration compensation parameters.

[0066] Exemplarily, in the model training stage, the adopted process can be as follows: 1. Data centralization. Let the original training data set be X ∈ R N×16 , where N is the number of samples, and each row is a 16-dimensional Mueller matrix unfolded vector. Perform centralization processing on each column (feature dimension): The centralized data , where μ = [μ1, μ2,..., μ 16 T is the mean vector of each feature dimension.

[0067] , where 1 N is an N×1 column vector with all elements being 1, represents the element in the i-th row and j-th column of the matrix.

[0068] 2. Calculate the covariance matrix of the centralized data, C ∈ R 16×16 .

[0069] ; where, represents the matrix obtained after centralizing the matrix, represents the transpose of the matrix obtained after centralizing the matrix.

[0070] Perform eigenvalue decomposition on C to obtain the eigenvalues λ1 ≥ λ2 ≥ … ≥ λ 16 and the corresponding eigenvectors v1, v2, …, v 16 ; 3. Select the eigenvectors corresponding to the first K largest eigenvalues to construct the projection matrix VK ∈R 16×K : V K = [v1, v2, …, v k 。

[0071] 4. Project the centralized data for dimensionality reduction to obtain the feature matrix Y after dimensionality reduction train ∈R N×K 。

[0072] 。

[0073] Use the data matrix after dimensionality reduction for model training to obtain a trained compensation parameter prediction model

[0074] In the prediction stage, perform dimensionality reduction on new samples. For a new sample x ∈ R 1×16 , process it according to the following steps: 1. Centralization: Use the mean μ in the training stage for centralization: The new sample data after centralization , where represents the sequence of Mueller matrix eigenvectors obtained in the prediction stage; 2. Projection: Through the obtained by training for dimensionality reduction: The new sample data after dimensionality reduction ; 3. Output y new ∈R 1×K is the eigenvector after dimensionality reduction

[0075] The method for dimensionality reduction processing of the Mueller matrix can extract key features and improve the accuracy of the obtained polarization aberration compensation parameters

[0076] Furthermore, the sampling periods corresponding to the temperature sequence and the air pressure sequence are both the first sampling period. The method of respectively performing low-pass filtering on the temperature sequence and the air pressure sequence to obtain the filtered temperature sequence and the filtered air pressure sequence may include: performing moving average filtering on the temperature sequence through a first sliding window to obtain the filtered temperature sequence; performing moving average filtering on the air pressure sequence through a second sliding window to obtain the filtered air pressure sequence, where the time lengths of the first sliding window and the second sliding window are greater than or equal to two first sampling periods, and the moving average filtering includes taking the mean value of the values of at least two sampling points within the window as the value of the current sampling point

[0077] Exemplarily, a moving average filter with sliding window sizes of W1 and W2 is applied to the temperature sequence and the air pressure sequence, that is, by calculating the average value of the data corresponding to the W1 sampling points before each current sampling point, a filtered temperature sequence is obtained, and by calculating the average value of the data corresponding to the W2 sampling points before each current sampling point, a filtered air pressure sequence is obtained, where W1 and W2 are different.

[0078] Exemplarily, W1 and W2 can be the same W, and a moving average filtering process is performed on the measured temperature and air pressure sequences by sliding. A suitable window size W is set, and the average value of the W data within the sliding window corresponding to the current sampling point is calculated as the filtered value at the current moment. By the above method, short-term fluctuations can be smoothed out and the influence of high-frequency noise can be reduced.

[0079] It can be understood that the moving average filtering process on the measured temperature and air pressure can extract key features and improve the accuracy of the obtained polarization aberration compensation parameters.

[0080] Further, the sliding window aggregation of the vibration sequence to obtain the aggregated acceleration sequence may include: determining that the time length of the third sliding window is equal to the first sampling period; performing sliding aggregation on the vibration sequence through the third sliding window, that is, calculating the mean and variance of the acceleration data within each sliding window to obtain the processed vibration acceleration data; and then synchronizing the processed vibration acceleration data in time according to the first sampling period to obtain the aggregated acceleration sequence.

[0081] It should be noted that sliding window aggregation can be performed on the original signal of the acceleration sensor to calculate the mean value of the acceleration within the time window of the first sampling period ΔT and variance .

[0082] It can be understood that the sliding aggregation process on the measured vibration acceleration data can extract key features and improve the accuracy of the obtained polarization aberration compensation parameters.

[0083] Figure 5 This is a schematic diagram of feature extraction for the Mueller matrix sequence and the environmental parameter sequence provided by the present invention. As Figure 5 shown, the sampling periods corresponding to the Mueller matrix sequence, the temperature sequence, and the air pressure sequence are all T, the sampling period corresponding to the vibration sequence is much smaller than T, the sliding window W1 corresponding to the temperature sequence is equal to three sampling periods, the sliding window W2 corresponding to the air pressure sequence is equal to two sampling periods, the aggregation window corresponding to the vibration sequence is W3, and the result of sliding aggregation is to aggregate multiple values within the W3 window into one mean value and variance. The time length of W3 is equal to the sampling period T, and aggregation processing is performed on all points within the period T to obtain one sampling value.

[0084] In some embodiments, determining the corresponding target compensation parameter according to the sequence of feature tensors may include: inputting the sequence of feature tensors into a trained compensation parameter prediction model to output the target compensation parameter, where the trained compensation parameter prediction model is obtained by inputting a measured sequence of Mueller matrices and a sequence of environmental parameters into an initial compensation parameter prediction model and optimizing the model parameters according to the mean square error between the measured imaging parameters corresponding to the predicted compensation parameters output by the model and the target imaging parameters.

[0085] It should be noted that the existing methods for obtaining polarization aberration compensation parameters are highly dependent on models and have low automation. That is, they usually rely on predefined mode expansions (such as Fourier or Zernike decomposition) to characterize the properties of Mueller matrices. These methods rely on parameter structures selected manually, the modeling process depends on domain experience, lack the ability of adaptive learning, and it is difficult to cover complex polarization distortion patterns.

[0086] In the present invention, a measured sequence of Mueller matrices and a sequence of environmental parameters are input into an initial compensation parameter prediction model, and the model parameters are optimized according to the mean square error between the measured imaging parameters corresponding to the predicted compensation parameters output by the model and the target imaging parameters to obtain a trained compensation parameter prediction model. Inputting the sequence of feature tensors into the trained compensation parameter prediction model can output the target compensation parameter.

[0087] It can be understood that during the training process, joint modeling and collaborative optimization of multiple adjustment parameters affecting polarization performance are realized, thereby avoiding the system coupling effect caused by independent adjustment of each compensation unit in the traditional scheme, improving the overall compensation accuracy and imaging accuracy, and having good system integration ability and engineering feasibility.

[0088] Further, the trained compensation parameter prediction model includes a long short-term memory neural network model and a fully connected layer. The fully connected layer is used to fuse the temporal features output by the initial compensation parameter prediction model and generate the target compensation parameter, where the target compensation parameter includes at least one of the polarization angle compensation amount of the polarizer and the fast axis azimuth angle compensation amount of the quarter-wave plate.

[0089] It should be noted that the long short-term memory network (LSTM) is a special recurrent neural network (RNN). With its powerful self-learning ability, it can automatically deeply extract key features from data, thus effectively overcoming the dependence on manual experience in existing methods and realizing efficient characterization of complex polarization characteristics. Features are extracted through the LSTM model, and the features are fused by the fully connected layer, and finally the compensation parameters of the polarization controller are output. The compensation parameters can be at least one of the polarization angle compensation amount of the polarizer and the fast axis azimuth angle compensation amount of the quarter-wave plate.

[0090] In addition, a feedback verification module can be adopted, that is, a polarization controller is used to compensate the polarization parameters of the illumination system according to the output of the machine learning model. And the compensation effect is fed back to the long short-term memory neural network model to optimize the model parameters.

[0091] It can be understood that the prior art generally adopts an end-to-end global optimization method, while the invention adopts a long short-term memory neural network model (Long Short-Term Memory, LSTM) as a data-driven modeling framework, which can realize the joint modeling and collaborative optimization of multiple adjustment parameters affecting polarization performance (such as polarizer angle, wave plate phase delay, etc.) during the training process, thereby avoiding the system coupling effect caused by the independent adjustment of each compensation unit in the traditional scheme, improving the overall compensation accuracy and imaging accuracy, and having good system integration ability and engineering feasibility.

[0092] In some embodiments, the target compensation parameters include the polarizer angle compensation amount and the fast axis azimuth angle compensation amount of the quarter-wave plate, and the method may further include: adjusting the polarizer angle and the fast axis azimuth angle of the quarter-wave plate in the polarization state generator according to the target compensation parameters.

[0093] It should be noted that after obtaining the polarizer angle compensation amount and the fast axis azimuth angle compensation amount of the quarter-wave plate, the polarizer angle and the fast axis azimuth angle of the quarter-wave plate in the polarization state generator can be adjusted through the polarizer angle compensation amount and the fast axis azimuth angle compensation amount of the quarter-wave plate to achieve polarization aberration compensation.

[0094] It can be understood that to overcome the limitations of the existing polarization aberration compensation method based on the Mueller matrix in adaptive feature extraction, non-linear response modeling, and coupled effect collaborative optimization, the present invention proposes a polarization aberration compensation method based on a long short-term memory neural network (Long Short-Term Memory, LSTM). This method obtains time series data sets such as Mueller matrix parameters and lithography machine system environment state parameters (such as temperature, air pressure, vibration acceleration), and inputs them into a trained LSTM model. The trained LSTM model has learned the change law of polarized light in the optical system, and performs polarization compensation on the result obtained by computational simulation through a feedback system, which can improve the accuracy of obtaining polarization aberration compensation parameters, and thus reduce polarization aberration and improve imaging quality.

[0095] In addition, the present invention adopts a polarization aberration dynamic compensation system architecture based on LSTM and Mueller matrix, and protects the multi-source data fusion method of "Mueller matrix - environmental parameters" and the sliding window aggregation technology through a complete technical chain composed of Mueller matrix sequence acquisition, environmental sensing data fusion, PCA dimensionality reduction preprocessing, LSTM time series modeling, and closed-loop compensation control. It can improve the accuracy of obtaining polarization aberration compensation parameters, thereby further improving the overall compensation accuracy and imaging accuracy of the system.

[0096] The exemplary application of the embodiments of the present invention in an actual application scenario will be described below.

[0097] In the lithography process of a lithography machine, it can be briefly divided into several processes such as a simulation design stage, an exposure preparation stage, and an exposure stage. Among them, in the simulation design stage, the lithography process is simulated through simulation software, and the imaging effect can be predicted and parameters (including illumination mode, mask design, polarization configuration, etc.) can be optimized. In particular, although the parameters obtained by computational simulation take into account the factor of polarization aberration, generally the results cannot completely avoid the influence of polarization phase difference, and polarization compensation usually needs to be carried out on this basis. Polarization compensation is carried out in the exposure preparation stage. In this stage, according to the latest measured Mueller matrix or a preset model, the polarization controller (such as a rotating wave plate, a micro mirror array) is adjusted to compensate for the aberration. In the exposure stage, exposure is usually carried out according to the parameters of the computational simulation design and polarization aberration compensation, and the polarization state is generally not adjusted. The present invention is applied in the exposure preparation stage, extracts the features of the Mueller matrix sequence and the temperature, air pressure, and vibration sequences of the lithography system through an LSTM model, predicts the parameter compensation amount of the polarization generator PSG, and fine-tunes the polarization parameters through a feedback system for feed-forward compensation.

[0098] Specifically, in the simulation design stage, the ideal parameters of the design simulation stage are obtained, including: the polarization compensation angle sequence of the polarizer {θ0(t0), θ0(t1),..., θ0(t n )}, and the fast axis azimuth angle sequence of the quarter-wave plate {φ0(t0), φ0(t1),..., φ0(t n )}; the imaging parameter error sequence, that is, the target critical dimension (CD) error sequence {CD0(t0), CD0(t1),..., CD0(t n )}, (where t n represents the nth measurement moment, and ΔT = t n - t n-1 is a fixed sampling interval). Exemplarily, the imaging parameter error sequence can be the mean square error sequence between the predicted imaging line width and the measured imaging line width.

[0099] Figure 6This is a schematic diagram of the overall process for obtaining polarization aberration compensation parameters provided by the present invention. As Figure 6 shown, obtain the measurement data sequence in the exposure preparation stage, including: Polarization state analysis unit (polarization state analyzer PSA and charge-coupled device CCD detector): Measure the Stokes vector parameters at a fixed period ΔT, and calculate the measured Mueller matrix sequence {M(t0), M(t1),..., M(t n )}.

[0100] Environmental monitoring unit: Measure the temperature sequence {T(t0), T(t1),..., T(t n )} and the air pressure sequence {P(t0), P(t1),..., P(t n )} of the lithography machine at a synchronous period. Since the vibration frequency is relatively high, the original acceleration sequence {a(τ0),..., a(τ m )} is output at a higher frequency Δτ (Δτ = ΔT / k).

[0101] Then, perform low-pass filtering on the obtained temperature sequence, perform low-pass filtering on the obtained air pressure sequence, perform sliding aggregation on the obtained vibration sequence, and perform PCA dimensionality reduction on the obtained Mueller matrix sequence to obtain the feature tensor sequence.

[0102] Then, input the obtained feature tensor sequence into the LSTM model, and output the target compensation parameters through the fully connected layer of the LSTM model. The target compensation parameters include the polarization angle compensation amount and the fast axis azimuth angle compensation amount of the quarter-wave plate .

[0103] Among them, input the processed parameters into the trained LSTM neural network for prediction, and output them as the compensation amount sequence through the fully connected layer network, .

[0104] Closed-loop timing compensation control includes: For each moment t_{n+1}, perform parameter compensation as follows, and then adjust the polarization angle and the fast axis azimuth angle of the quarter-wave plate through the polarization controller.

[0105]

[0106] Among them, and are the ideal parameters in the design and simulation stage, and are the polarization aberration compensation parameters output by the LSTM neural network.

[0107] In addition, Figure 6The steps within the dashed-line box are used to simulate the transformation of the exposure system, and are used to backpropagate the error between the measured critical dimension (CD) and the simulated ideal CD to and for training the LSTM model. The loss function is the mean square sum of the simulated target CD and the actually measured CD.

[0108] The system also includes a machine learning prediction module: mainly including extracting features by the LSTM model, and fusing the features by the fully connected layer, and finally outputting the compensation parameters of the polarization controller, mainly the angle of the polarizer and the fast axis azimuth angle of the quarter-wave plate.

[0109] The system may also include a feedback verification module: mainly the polarization controller, which compensates the polarization parameters of the illumination system according to the output of the machine learning model.

[0110] It can be understood that, in order to overcome the limitations of the above-mentioned existing polarization aberration compensation method based on the Mueller matrix in adaptive feature extraction, non-linear response modeling, and collaborative optimization of coupling effects, the present invention proposes a polarization aberration compensation method based on a long short-term memory neural network (LSTM). This method constructs a time series data set of Mueller matrix parameters, lithography system state parameters (such as temperature, air pressure, vibration acceleration), etc., trains the LSTM model to learn the variation law of polarized light in the optical system, and performs polarization compensation on the results obtained by computational simulation through a feedback system, thereby reducing polarization aberration and improving imaging quality.

[0111] It has the following advantages: Adaptive feature learning: The Long Short-Term Memory (LSTM) network is a special type of Recurrent Neural Network (RNN). With its powerful self-learning ability, it can automatically deeply mine key features from data, thus effectively overcoming the dependence on artificial experience in existing methods and achieving efficient characterization of complex polarization characteristics; Having the ability of non-linear mapping and time-series modeling: Although the Mueller matrix itself is a model of linear transformation, due to the non-linear response of components in the lithography system in certain states, these non-linear responses will be manifested in the measured Mueller matrix. Therefore, by feeding the sequence of Mueller matrices at different times into the LSTM model for training, the learning of the non-linear mapping of the lithography system can be achieved; End-to-end global optimization: As a data-driven modeling framework, the LSTM model can jointly model and co-optimize multiple adjustment parameters (such as polarizer angle, waveplate phase delay, etc.) that affect polarization performance during the training process, thereby avoiding the system coupling effect caused by independent adjustment of each compensation unit in the traditional scheme, improving the overall compensation accuracy and imaging accuracy, and having good system integration ability and engineering feasibility.

[0112] Based on the foregoing embodiments, an embodiment of the present invention provides a system for obtaining polarization aberration compensation parameters. Each module included in the system, as well as each unit included in each module, can be implemented by a processor; of course, it can also be implemented by specific logic circuits; during implementation, the processor can be a central processing unit (CPU), a microprocessor (MPU), a digital signal processor (DSP), or a field programmable gate array (FPGA), etc.

[0113] The polarization aberration compensation parameter system provided by the present invention will be described below. The system for obtaining polarization aberration compensation parameters described below can be mutually corresponding and referred to with the method for obtaining polarization aberration compensation parameters described above.

[0114] Figure 7 is a schematic structural diagram of the polarization aberration compensation system provided by the present invention. As Figure 7 shown, the system 400 includes a data acquisition module 401, a feature acquisition module 402, and a parameter prediction module 403, wherein: The data acquisition module 401 is used to obtain a sequence of Mueller matrices corresponding to the lithography system and a sequence of environmental parameters. The time series of the sequence of Mueller matrices and the sequence of environmental parameters are aligned. The sequence of Mueller matrices is used to characterize the change of polarized light during transmission in the lithography system, and the sequence of environmental parameters is used to characterize the change of environmental parameters that affect the stability of the lithography system; A feature acquisition module 402, configured to perform feature extraction on the Mueller matrix sequence and the environmental parameter sequence respectively, and combine the features at the same time series to obtain a feature tensor sequence; A parameter prediction module 403, configured to determine a target compensation parameter according to the feature tensor sequence, where the target compensation parameter is used to compensate for the polarization aberration of the lithography machine system.

[0115] In some embodiments, the environmental parameter sequence includes at least one of a temperature sequence, a pressure sequence, and a vibration sequence. The temperature sequence is used to characterize the temperature change of the lithography machine system, the pressure sequence is used to characterize the pressure change of the lithography machine system, and the vibration sequence is used to characterize the vibration acceleration change of the moving parts and / or optical parts of the lithography machine system.

[0116] In some embodiments, the environmental parameter sequence includes the temperature sequence, the pressure sequence, and the vibration sequence. The feature acquisition module 402 includes: A dimensionality reduction processing unit, configured to perform principal component analysis dimensionality reduction processing on the Mueller matrix sequence to obtain a dimensionality-reduced Mueller matrix feature vector sequence; A filtering processing unit, configured to perform low-pass filtering processing on the temperature sequence and the pressure sequence respectively to obtain a filtered temperature sequence and a filtered pressure sequence; A window aggregation unit, configured to perform sliding window aggregation on the vibration sequence to obtain an aggregated acceleration sequence; A feature combination unit, configured to normalize the dimensionality-reduced Mueller matrix feature vector sequence, the filtered temperature sequence, the filtered pressure sequence, and the aggregated acceleration sequence, and then splice the features at the same time series to obtain the feature tensor sequence.

[0117] In some embodiments, the Mueller matrix sequence includes at least one 4×4 Mueller matrix. The dimensionality reduction processing unit is specifically configured to: expand each 4×4 Mueller matrix in the at least one 4×4 Mueller matrix into a 16-dimensional row vector by row priority to obtain at least one 16-dimensional row vector; stack the at least one 16-dimensional row vector to form a data matrix, and extract the first K principal components from the data matrix to obtain the dimensionality-reduced Mueller matrix feature vector sequence.

[0118] In some embodiments, the sampling periods corresponding to the temperature sequence and the air pressure sequence are both the first sampling period. The filtering processing unit is specifically configured to: perform moving average filtering on the temperature sequence through a first sliding window to obtain the filtered temperature sequence; perform moving average filtering on the air pressure sequence through a second sliding window to obtain the filtered air pressure sequence, where the time lengths of the first sliding window and the second sliding window are greater than or equal to two first sampling periods, and the moving average filtering includes taking the mean value of the values of at least two sampling points within the window as the value of the current sampling point.

[0119] In some embodiments, the second sampling period corresponding to the vibration sequence is less than the first sampling period, and the third sliding window is a non-overlapping sliding window. The window aggregation unit is specifically configured to: determine that the time length of the third sliding window is equal to the first sampling period; perform sliding aggregation on the vibration sequence through the third sliding window to obtain the processed vibration acceleration data, where the sliding aggregation includes calculating the mean value and variance of the acceleration data within the window as the value of the current sampling point; perform time series alignment on the processed vibration acceleration data according to the first sampling period to obtain the aggregated acceleration sequence.

[0120] In some embodiments, the parameter prediction module 403 is specifically configured to: input the feature tensor sequence into a trained compensation parameter prediction model and output the target compensation parameter. The trained compensation parameter prediction model is obtained by inputting the measured Mueller matrix sequence and the environmental parameter sequence into an initial compensation parameter prediction model and optimizing the model parameters according to the mean square error between the measured imaging parameter corresponding to the predicted compensation parameter output by the model and the target imaging parameter.

[0121] In some embodiments, the trained compensation parameter prediction model includes a long short-term memory neural network model and a fully connected layer. The fully connected layer is used to fuse the time series features output by the long short-term memory neural network model and generate the target compensation parameter, and the target compensation parameter includes at least one of a polarizer angle compensation amount and a fast axis azimuth angle compensation amount of a quarter-wave plate.

[0122] In some embodiments, the target compensation parameter includes a polarizer angle compensation amount and a fast axis azimuth angle compensation amount of a quarter-wave plate. The device further includes a compensation implementation module for adjusting the polarizer angle compensation amount and the fast axis azimuth angle of the quarter-wave plate in the polarization state generator according to the target compensation parameter.

[0123] In some embodiments, the data acquisition module 401 includes: The first acquisition unit is configured to obtain multiple sets of Stokes vectors at each sampling point according to the first sampling period, and perform least-squares fitting on the multiple sets of Stokes vectors to obtain a Mueller matrix corresponding to each sampling point. The Mueller matrices corresponding to multiple sampling points form the Mueller matrix sequence; The second acquisition unit is configured to synchronously obtain the temperature parameter and the pressure parameter corresponding to each sampling point according to the first sampling period. The temperature parameters and the pressure parameters corresponding to multiple sampling points form the temperature sequence and the pressure sequence; The third acquisition unit is configured to obtain multiple vibration acceleration parameters according to the second sampling period. The multiple vibration acceleration parameters form the vibration sequence, and the second sampling period is less than the first sampling period.

[0124] In the embodiment of the present invention, multi-source data composed of a Mueller matrix and environmental parameters is fused and features are extracted to obtain a feature tensor sequence, and target compensation parameters are determined according to the feature tensor sequence. The accuracy of obtaining polarization aberration compensation parameters can be improved, thereby further improving the overall compensation accuracy and imaging accuracy of the system.

[0125] Figure 8 It is a schematic physical structure diagram of an electronic device provided by the present invention. As Figure 8 shown, the electronic device may include: a processor 510, a communication interface 520, a memory 530, and a communication bus 540. Among them, the processor 510, the communication interface 520, and the memory 530 communicate with each other through the communication bus 540. The processor 510 can call the logical instructions in the memory 530 to execute a method for obtaining polarization aberration compensation parameters. The method includes: obtaining a Mueller matrix sequence and an environmental parameter sequence corresponding to a lithography machine system, where the time sequences of the Mueller matrix sequence and the environmental parameter sequence are aligned. The Mueller matrix sequence is used to characterize the change of polarized light during transmission in the lithography machine system, and the environmental parameter sequence is used to characterize the change of environmental parameters that affect the stability of the lithography machine system; performing feature extraction on the Mueller matrix sequence and the environmental parameter sequence respectively, and combining the features at the same time sequence to obtain a feature tensor sequence; determining target compensation parameters according to the feature tensor sequence, where the target compensation parameters are used to compensate for the polarization aberration of the lithography machine system.

[0126] In addition, when the logical instructions in the aforementioned memory 530 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs.

[0127] On the other hand, the present invention also provides a computer program product. The computer program product includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the method for obtaining polarization aberration compensation parameters provided by the above-mentioned various methods. The method includes: obtaining a sequence of Mueller matrices and a sequence of environmental parameters corresponding to a lithography system, where the timing of the sequence of Mueller matrices and the sequence of environmental parameters is aligned. The sequence of Mueller matrices is used to characterize the change of polarized light during transmission in the lithography system, and the sequence of environmental parameters is used to characterize the change of environmental parameters that affect the stability of the lithography system; respectively performing feature extraction on the sequence of Mueller matrices and the sequence of environmental parameters and combining the features at the same timing to obtain a sequence of feature tensors; determining a target compensation parameter according to the sequence of feature tensors, and the target compensation parameter is used to compensate for the polarization aberration of the lithography system.

[0128] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, they generate all or part of the processes or functions described in the embodiments of the present invention. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a server or data center that includes one or more integrated available media. The available medium may be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid-state disk (SSD)).

[0129] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it implements a method for obtaining polarization aberration compensation parameters provided by the above methods. The method includes: obtaining a sequence of Mueller matrices and a sequence of environmental parameters corresponding to a lithography system, wherein the timings of the sequence of Mueller matrices and the sequence of environmental parameters are aligned. The sequence of Mueller matrices is used to characterize the change of polarized light during transmission in the lithography system, and the sequence of environmental parameters is used to characterize the change of environmental parameters that affect the stability of the lithography system; respectively performing feature extraction on the sequence of Mueller matrices and the sequence of environmental parameters and combining the features at the same timing to obtain a sequence of feature tensors; determining a target compensation parameter according to the sequence of feature tensors, and the target compensation parameter is used to compensate for the polarization aberration of the lithography system.

[0130] The above computer-readable storage medium may adopt any combination of one or more computer-readable media. The computer-readable media may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the computer-readable storage medium include: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM) or a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0131] The computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries the computer-readable program code. Such a propagated data signal may take various forms, including - but not limited to - an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer-readable signal medium may also be any computer-readable medium other than the computer-readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0132] The program code contained on the computer-readable medium may be transmitted by any appropriate medium, including - but not limited to - wireless, wire, optical cable, radio frequency (RF), etc., or any suitable combination of the above.

[0133] Computer program code for performing the operations of this specification can be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (e.g., by connecting through the Internet using an Internet service provider).

[0134] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without creative work.

[0135] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0136] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for obtaining polarization aberration compensation parameters, characterized in that, Including: Obtain a Mueller matrix sequence and an environmental parameter sequence corresponding to a lithography machine system, where the time sequences of the Mueller matrix sequence and the environmental parameter sequence are aligned. The Mueller matrix sequence is used to characterize the change of polarized light during transmission in the lithography machine system, and the environmental parameter sequence is used to characterize the change of environmental parameters that affect the stability of the lithography machine system; Perform feature extraction on the Mueller matrix sequence and the environmental parameter sequence respectively, and combine the features at the same time sequence to obtain a feature tensor sequence; Determine target compensation parameters according to the feature tensor sequence, and the target compensation parameters are used to compensate for the polarization aberration of the lithography machine system.

2. The method for obtaining polarization aberration compensation parameters according to claim 1, characterized in that The environmental parameter sequence includes at least one of a temperature sequence, a pressure sequence, and a vibration sequence. The temperature sequence is used to characterize the temperature change of the lithography machine system, the pressure sequence is used to characterize the pressure change of the lithography machine system, and the vibration sequence is used to characterize the vibration acceleration change of the moving parts and / or optical parts of the lithography machine system.

3. The method for obtaining polarization aberration compensation parameters according to claim 2, wherein The environmental parameter sequence includes the temperature sequence, the pressure sequence, and the vibration sequence. The performing feature extraction on the Mueller matrix sequence and the environmental parameter sequence respectively, and combining the features at the same time sequence to obtain a feature tensor sequence includes: Perform principal component analysis and dimensionality reduction processing on the Mueller matrix sequence to obtain a dimensionality-reduced Mueller matrix eigenvector sequence; Perform low-pass filtering on the temperature sequence and the pressure sequence respectively to obtain a filtered temperature sequence and a filtered pressure sequence; Perform sliding window aggregation on the vibration sequence to obtain an aggregated acceleration sequence; Normalize the dimensionality-reduced Mueller matrix eigenvector sequence, the filtered temperature sequence, the filtered pressure sequence, and the aggregated acceleration sequence, and then splice the features at the same time sequence to obtain the feature tensor sequence.

4. The method for obtaining polarization aberration compensation parameters according to claim 3, wherein The Mueller matrix sequence includes at least one 4×4 Mueller matrix. The performing principal component analysis and dimensionality reduction processing on the Mueller matrix sequence to obtain a dimensionality-reduced Mueller matrix eigenvector sequence includes: Expand each 4×4 Mueller matrix in the at least one 4×4 Mueller matrix into a 16-dimensional row vector by row priority to obtain at least one 16-dimensional row vector; Stack the at least one 16-dimensional row vector to form a data matrix, and extract the first K principal components from the data matrix to obtain the dimensionality-reduced Mueller matrix eigenvector sequence.

5. The method for obtaining polarization aberration compensation parameters according to claim 3, characterized in that, The sampling periods corresponding to the temperature sequence and the pressure sequence are both the first sampling period. The performing low-pass filtering on the temperature sequence and the pressure sequence respectively to obtain a filtered temperature sequence and a filtered pressure sequence includes: Perform moving average filtering on the temperature sequence through a first sliding window to obtain the filtered temperature sequence; Performing moving average filtering on the air pressure sequence through a second sliding window to obtain the filtered air pressure sequence, where the time lengths of the first sliding window and the second sliding window are greater than or equal to two first sampling periods, and the moving average filtering includes taking the mean value of the values of at least two sampling points within the window as the value of the current sampling point; The second sampling period corresponding to the vibration sequence is less than the first sampling period, and the third sliding window is a non-overlapping sliding window. The performing sliding window aggregation on the vibration sequence to obtain the aggregated acceleration sequence includes: Determining that the time length of the third sliding window is equal to the first sampling period; Performing sliding aggregation on the vibration sequence through the third sliding window to obtain the processed vibration acceleration data, where the sliding aggregation includes calculating the mean value and variance of the acceleration data within the window as the value of the current sampling point; Performing time series alignment on the processed vibration acceleration data according to the first sampling period to obtain the aggregated acceleration sequence.

6. The method for obtaining polarization aberration compensation parameters according to claim 1, wherein The determining the corresponding target compensation parameter according to the feature tensor sequence includes: Inputting the feature tensor sequence into a trained compensation parameter prediction model to output the target compensation parameter. The trained compensation parameter prediction model is obtained by inputting the measured Mueller matrix sequence and the environmental parameter sequence into an initial compensation parameter prediction model and optimizing the model parameters according to the mean square error between the measured imaging parameters corresponding to the predicted compensation parameters output by the model and the target imaging parameters.

7. The method for obtaining polarization aberration compensation parameters according to claim 6, characterized in that The trained compensation parameter prediction model includes a long short-term memory neural network model and a fully connected layer. The fully connected layer is used to fuse the time series features output by the long short-term memory neural network model and generate the target compensation parameter, and the target compensation parameter includes at least one of the polarization angle compensation amount of the polarizer and the fast axis azimuth angle compensation amount of the quarter-wave plate.

8. The method for obtaining polarization aberration compensation parameters according to claim 2, wherein, The environmental parameter sequence includes the temperature sequence, the air pressure sequence, and the vibration sequence. The obtaining the Mueller matrix sequence and the environmental parameter sequence corresponding to the lithography machine system includes: Obtaining multiple sets of Stokes vectors at each sampling point according to the first sampling period, and performing least squares fitting on the multiple sets of Stokes vectors to obtain the Mueller matrix corresponding to each sampling point. The Mueller matrices corresponding to multiple sampling points constitute the Mueller matrix sequence; Synchronously obtaining the temperature parameter and the air pressure parameter corresponding to each sampling point according to the first sampling period. The temperature parameters and the air pressure parameters corresponding to multiple sampling points constitute the temperature sequence and the air pressure sequence; Obtaining multiple vibration acceleration parameters according to the second sampling period. The multiple vibration acceleration parameters constitute the vibration sequence, and the second sampling period is less than the first sampling period.

9. An acquisition system for polarization aberration compensation parameters, characterized in that, Including: A data acquisition module, configured to obtain a Mueller matrix sequence and an environmental parameter sequence corresponding to a lithography machine system, wherein the time sequences of the Mueller matrix sequence and the environmental parameter sequence are aligned, the Mueller matrix sequence is used to characterize the change of polarized light during transmission in the lithography machine system, and the environmental parameter sequence is used to characterize the change of environmental parameters that affect the stability of the lithography machine system; A feature acquisition module, configured to perform feature extraction on the Mueller matrix sequence and the environmental parameter sequence respectively and combine the features at the same time sequence to obtain a sequence of feature tensors; A parameter prediction module, configured to determine a target compensation parameter according to the sequence of feature tensors, and the target compensation parameter is used to compensate for the polarization aberration of the lithography machine system.

10. A lithography machine, comprising a memory, a processor, and a computer program stored on the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the method for obtaining the polarization aberration compensation parameter according to any one of claims 1 to 8.

Images