Mask near-field simulation operation method and device under partial coherence illumination and electronic equipment

Abstract

The application discloses a mask near-field simulation operation method and device under partial coherent illumination and electronic equipment, and relates to the field of photolithography systems. The method comprises the following steps: dividing a pupil surface into a preset number of light source points with different incident directions; determining four diffraction near-field matrices of a plurality of input training layout patterns through strict electromagnetic field simulation for each light source point; determining light source point near-field distribution training data corresponding to a preset pattern category based on a plurality of groups of four diffraction near-field matrices corresponding to the plurality of training layout patterns; determining diffraction transmission matrices of all preset pattern categories based on a plurality of light source point near-field distribution training data through a least square method; in the case that a target incident light source is received, determining a light source point corresponding to the target incident light source; and performing operation on a near-field distribution of a target mask pattern corresponding to the target incident light source based on a diffraction transmission matrix corresponding to the light source point, so as to determine a near-field result.

Description

Technical Field

[0001] This application relates to the field of photolithography systems, and more particularly to a method, apparatus, and electronic device for near-field simulation of masks under partially coherent illumination. Background Technology

[0002] With the continuous evolution of integrated circuit technology, the size of circuit patterns in chips is constantly decreasing, and transistor density is significantly improved, thereby enhancing chip performance. However, this also brings enormous challenges to the chip manufacturing process. In current advanced chips, the minimum spacing between patterns is only tens or even hundreds of nanometers, while mainstream lithography technologies, such as ArF deep ultraviolet lithography, use light sources with a wavelength of 193 nanometers. At such tiny dimensions, the diffraction effect of light becomes particularly severe.

[0003] Existing fast computation methods are based on convolution calculations. First, rigorous electromagnetic field calculations are used to calculate the accurate near-field distribution corresponding to a subset of patterns, which is then used as a training set. Next, using the training set as a standard, a convolution kernel is calculated to reflect the relationship between the pattern and its corresponding near-field distribution. Finally, the entire pattern is convolved using this kernel to obtain the near-field distribution corresponding to the target pattern. The results of fast computation methods have smaller errors compared to the accurate results of rigorous electromagnetic field calculations, while offering a significant speed improvement, making it possible to simulate the near-field distribution of large-scale patterns with high accuracy. However, on the one hand, there is still room for improvement in both the accuracy and speed of fast computation methods; on the other hand, for DUV lithography, because the spacing between the fabricated patterns is much smaller than the wavelength of the light source, the edge interference effect in the near field is significant, such as… Figure 2 As shown in (a), a "wavy" field strength distribution will appear inside the graph. Figure 2 As shown in (b), the convolution method cannot restore the "wavy" field intensity distribution caused by diffraction. In our previous research, the thick mask model based on decomposition machine learning was proposed, using a "Diffraction Transfer Matrix" (DTM) instead of a traditional single convolution kernel. Figure 2 As shown in (c), this method can effectively restore the "wavy" field intensity distribution caused by diffraction, improve the accuracy of the calculation results, and reduce the calculation time.

[0004] However, this method is based on coherent illumination systems for calculations, while actual lithography systems use partially coherent illumination as the light source. Therefore, this method cannot simulate the imaging model of partially coherent illumination lithography systems, thus hindering the application of subsequent resolution enhancement techniques (such as OPC, SMO, etc.).

[0005] Figure 3A schematic diagram illustrating an example of photolithography imaging to demonstrate the limitations of using coherent illumination in the near field. Figure 3 'a' shows the pupil plane. The ring L represents the ring illumination used in this example, with an inner coherence factor of 0.4 and an outer coherence factor of 0.6. In the lithography system, we use the Abbe model to calculate the lithographic spatial image. Figure 3 In the rigorous Abbe model shown in b, four light source points are used on the pupil surface to represent annular illumination. These four light source points are... Figure 3 K is used as the denotor in a. The near fields of the four light sources are calculated independently, and the spatial image is obtained through the Abbe imaging model; in Figure 3 In the approximate Abbe model shown in c, the near field of coherent illumination is used to approximate the near field of ring illumination, and the coherent illumination source point is at... Figure 3 The denoting element M is used in a. After obtaining the near field of coherent illumination, the spatial image is calculated using an imaging model with four light sources. Figure 3 d shows the difference between the spatial images of the strict Abbe model and the approximate Abbe model; Figure 3 e shows the spatial image cross-sections of the strict Abbe model and the approximate Abbe model. Using a threshold function to approximate the photoresist, it can be seen that the exposure cross-section length at the cross-section is 27 nm for the strict Abbe model and 38 nm for the approximate Abbe model. The critical dimension (CD) of the layout at this location is 54 nm.

[0006] As can be seen, the spatial image calculated by the approximate Abbe model has a large error. Therefore, for the near-field model, calculating only the near field of coherent illumination is limited and cannot meet the requirements of a high-precision imaging calculation model. Summary of the Invention

[0007] The purpose of this application is to provide a method, apparatus, and electronic device for near-field simulation of masks under partially coherent illumination, so as to solve the problem that existing near-field simulation methods cannot meet the requirements of high-precision imaging calculation models.

[0008] In a first aspect, this application provides a method for simulating the near-field of a mask under partially coherent illumination, the method comprising:

[0009] The pupil plane is divided into a preset number of light source points with different incident directions;

[0010] For each of the aforementioned light source points, four diffraction near-field matrices of the multiple input training pattern graphics are determined through rigorous electromagnetic field simulation.

[0011] Based on the multiple sets of the four diffraction near-field matrices corresponding to the multiple training pattern graphics, the near-field distribution training data of the light source points corresponding to the preset pattern types are determined respectively.

[0012] Based on the near-field distribution training data of multiple light source points, the diffraction transfer matrix of each of the preset pattern types is determined by the least squares method.

[0013] Upon receiving a target incident light source, the light source point corresponding to the target incident light source is determined;

[0014] The near-field distribution of the target mask pattern corresponding to the incident light source is calculated based on the diffraction transmission matrix corresponding to the light source point to determine the near-field result.

[0015] With the above technical solution, the mask near-field simulation calculation method under partially coherent illumination described in this application divides the pupil plane into a preset number of light source points with different incident directions; for each light source point, through rigorous electromagnetic field simulation, four diffraction near-field matrices of multiple input training pattern graphics are determined; based on multiple sets of the four diffraction near-field matrices corresponding to the multiple training pattern graphics, training data for the near-field distribution of light source points corresponding to the preset pattern types are determined; based on the training data for the near-field distribution of light source points, the diffraction transfer matrix of all preset pattern types is determined using the least squares method; when a target incident light source is received, the light source point corresponding to the target incident light source is determined; based on the diffraction transfer matrix corresponding to the light source point, the near-field distribution of the target mask pattern corresponding to the target incident light source is calculated to determine the near-field result. This method can achieve fast and accurate calculation of the near-field generated by any light source point on the pupil plane, thereby providing a computational basis for subsequent resolution enhancement technology under partially coherent illumination, expanding the application scenarios of near-field calculation, and ensuring that the speed and accuracy of near-field calculation meet the requirements of high-precision imaging calculation models.

[0016] In one possible implementation, determining the near-field distribution training data of light source points corresponding to the preset pattern types based on multiple sets of the four diffraction near-field matrices corresponding to multiple training pattern patterns includes:

[0017] Each training pattern is divided into a convex corner portion, a concave corner portion, a vertical edge portion, and a horizontal edge portion according to preset division conditions.

[0018] For each part of the image, the training data of the near-field distribution of the light source point corresponding to each corresponding light source point is determined by multiple sets of the four diffraction near-field matrices.

[0019] In one possible implementation, determining the diffraction transfer matrix for each of the preset pattern types using the least squares method based on near-field distribution training data of multiple light source points includes:

[0020] Obtain the training data set for each of the partial graphics corresponding to the near-field distribution training data of multiple light source points;

[0021] For each of the said partial patterns, the diffraction transfer matrix of each partial pattern is determined using the least squares method based on the corresponding training data set.

[0022] In one possible implementation, the step of calculating the near-field distribution of the target mask pattern corresponding to the target incident light source based on the diffraction transfer matrix corresponding to the light source point, and determining the near-field result, includes:

[0023] Based on the preset division conditions, the target mask pattern is divided into corresponding target convex corner part pattern, target concave corner part pattern, target vertical edge part pattern and target horizontal edge part pattern;

[0024] Based on the vectorized mask pattern corresponding to each partial pattern, near-field distribution calculation is performed in combination with the diffraction transfer matrix of each partial pattern corresponding to the light source point to obtain the near-field distribution corresponding to each partial pattern.

[0025] In one possible implementation, dividing the pupil plane into a predetermined number of light source points with different incident directions includes:

[0026] The pupil surface is divided into 13 light source points with different incident directions by symmetrical arrangement at equal intervals.

[0027] Secondly, this application also provides a mask near-field simulation computing device under partially coherent illumination, the device comprising:

[0028] The partitioning module is used to divide the pupil plane into a preset number of light source points with different incident directions;

[0029] The first determining module is used to determine the four diffraction near-field matrices of the multiple input training pattern graphics for each of the light source points through rigorous electromagnetic field simulation.

[0030] The second determining module is used to determine the near-field distribution training data of light source points corresponding to the preset graphic types based on the multiple sets of the four diffraction near-field matrices corresponding to the multiple training pattern graphics.

[0031] The third determining module is used to determine the diffraction transfer matrix of all the preset graphic types respectively by using the least squares method based on the near-field distribution training data of multiple light source points.

[0032] The fourth determining module is used to determine the light source point corresponding to the target incident light source when the target incident light source is received;

[0033] The fifth determining module is used to calculate the near-field distribution of the target mask pattern corresponding to the target incident light source based on the diffraction transmission matrix corresponding to the light source point, and determine the near-field result.

[0034] In one possible implementation, the second determining module includes:

[0035] The first division submodule is used to divide each training pattern into a convex corner part, a concave corner part, a vertical edge part, and a horizontal edge part according to a preset division condition.

[0036] The first determining submodule is used to determine, for each part of the graphic, the near-field distribution training data of the light source point corresponding to each corresponding light source point through multiple sets of four diffraction near-field matrices.

[0037] In one possible implementation, the third determining module includes:

[0038] The acquisition submodule is used to acquire the training data set of each of the partial graphics corresponding to the near-field distribution training data of multiple light source points;

[0039] The second determining submodule is used to determine the diffraction transfer matrix of each of the partial graphics using the least squares method based on the corresponding training data set.

[0040] In one possible implementation, the fifth determining module includes:

[0041] The second division module is used to divide the target mask pattern into corresponding target convex corner part pattern, target concave corner part pattern, target vertical edge part pattern and target horizontal edge part pattern based on the preset division conditions;

[0042] The third determining submodule is used to perform near-field distribution calculation based on the vectorized mask pattern corresponding to each part of the pattern and the diffraction transfer matrix of each part of the pattern corresponding to the light source point, so as to obtain the near-field distribution corresponding to each part of the pattern.

[0043] The beneficial effects of the mask near-field simulation computing device under partially coherent illumination provided in the second aspect are the same as those of the mask near-field simulation computing method under partially coherent illumination described in the first aspect or any possible implementation of the first aspect, and will not be repeated here.

[0044] Thirdly, this application also provides an electronic device, comprising: one or more processors; and one or more machine-readable media having instructions stored thereon, which, when executed by the one or more processors, cause the mask near-field simulation computation method under partially coherent illumination as described in any of the first aspects to be performed.

[0045] The beneficial effects of the electronic device provided in the third aspect are the same as those of the mask near-field simulation calculation method under partially coherent illumination described in the first aspect or any possible implementation of the first aspect, and will not be repeated here. Attached Figure Description

[0046] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0047] Figure 1 A schematic diagram of a photolithography model provided in an embodiment of this application is shown;

[0048] Figure 2 This illustration shows a schematic diagram of an electric field distribution provided in an embodiment of this application;

[0049] Figure 3 A schematic diagram illustrating an example of photolithographic imaging to demonstrate the limitations of using coherent illumination in the near field;

[0050] Figure 4 A flowchart illustrating a mask near-field simulation calculation method under partially coherent illumination provided in an embodiment of this application is shown.

[0051] Figure 5 A flowchart illustrating another method for simulating near-field mask operation under partially coherent illumination, as provided in an embodiment of this application, is shown.

[0052] Figure 6 This illustration shows a pixelated pupil surface and a process for establishing a training set according to an embodiment of this application; Figure 7 A schematic diagram showing the near-field calculation results of the mask near-field simulation method under partially coherent illumination at different light source points; Figure 7 b shows the positions of the three light source points on the pupil plane in the example, a scene diagram labeled with P;

[0053] Figure 8 This illustration shows a schematic diagram of a mask near-field simulation computing device under partially coherent illumination provided in an embodiment of this application;

[0054] Figure 9 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of this application is shown;

[0055] Figure 10 This is a schematic diagram of the chip structure provided in the embodiments of this application.

[0056] Figure Labels

[0057] 400 - Electronic equipment; 410 - Processor; 440 - Communication line; 420 - Communication interface; 430 - Memory; 500 - Chip; 540 - Bus system. Detailed Implementation

[0058] To facilitate a clear description of the technical solutions in the embodiments of this application, the terms "first" and "second" are used in the embodiments of this application to distinguish identical or similar items with essentially the same function and effect. For example, the first threshold and the second threshold are only used to distinguish different thresholds and do not limit their order. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and the terms "first" and "second" are not necessarily different.

[0059] It should be noted that, in this application, the terms "exemplary" or "for example" are used to indicate that something is being described as an example, illustration, or illustration. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or design solutions. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0060] In this application, "at least one" means one or more, and "more than one" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can mean: a, b, c, a combination of a and b, a combination of a and c, a combination of b and c, or a, b, and c, where a, b, and c can be single or multiple.

[0061] Figure 1 A schematic diagram of a photolithography model provided in an embodiment of this application is shown, such as... Figure 1 As shown, the main function of photolithography is to use optical projection to accurately transfer the circuit device pattern on the photomask to the photoresist coated on the silicon wafer surface. In photolithography systems, partial coherent illumination is widely used as the light source type, and tilted illumination is used to improve the resolution of the photolithography system.

[0062] With the continuous evolution of integrated circuit technology, the size of circuit patterns in chips is constantly decreasing, and transistor density is significantly improved, thereby enhancing chip performance. However, this also brings enormous challenges to the chip manufacturing process. In current advanced chips, the minimum spacing between patterns is only tens or even hundreds of nanometers, while mainstream lithography technologies, such as ArF deep ultraviolet lithography, use light sources with a wavelength of 193 nanometers. At such tiny dimensions, the diffraction effect of light becomes particularly severe.

[0063] To improve the quality of lithographic patterns, resolution enhancement techniques are necessary, such as optical proximity correction and light source mask optimization. Most lithography systems use partially coherent illumination as the light source, leveraging its characteristics in conjunction with tilted illumination to improve the system's resolution. In resolution enhancement techniques, calculating the near-field diffraction distribution of the target mask pattern under partially coherent illumination is a crucial step.

[0064] Existing methods for calculating the near-field distribution of photolithographic diffraction mainly include the Kirchhoff approximation method, the rigorous electromagnetic field calculation method, and the fast calculation method. The Kirchhoff approximation method uses a scalar model and assumes that the mask satisfies Kirchhoff boundary conditions: in the transparent region of the mask, the field distribution and its partial derivatives are completely different from those without a mask; in the blocked region of the mask, the field distribution and its partial derivatives are both zero. Therefore, the near-field distribution of the mask can be expressed as a two-dimensional amplitude transmittance function of the mask pattern. Although the Kirchhoff approximation method is simple and easy to implement, it does not consider the three-dimensional effects of the mask and the influence of the boundaries on the light field, and it cannot reflect the influence of light sources with different incident angles on the near-field distribution. Therefore, the accuracy of the near-field distribution results obtained by the Kirchhoff approximation method is relatively low.

[0065] In contrast, rigorous electromagnetic field calculation methods compute the diffraction near-field of thick masks by solving Maxwell's equations. The near-field results obtained by this method are theoretically completely accurate. However, their computation time is too long, making them unsuitable for calculating near-field distributions of large-scale masks. Therefore, to address the limitations of these existing methods, it is necessary to investigate more efficient and accurate methods for calculating lithographic near-field distributions.

[0066] Among fast computation methods, convolution is a typical approach. Based on convolution calculations, it first uses rigorous electromagnetic field calculations to determine the accurate near-field distribution of a portion of a graphic under illumination from a specific light source, and uses this as a training set. Next, using the training set as a benchmark, a convolution kernel is calculated to reflect the relationship between the graphic and its corresponding near-field distribution. Finally, the graphic is convolved using this kernel, and the mask graphic is reconstructed point-by-point to obtain the near-field distribution corresponding to the target graphic.

[0067] Compared to the accurate results obtained by the rigorous electromagnetic field calculation method, the convolution method has smaller errors but significantly improved calculation speed. Although the fast calculation method achieves a good balance between accuracy and calculation speed, there is still room for further optimization to meet the ever-increasing requirements of photolithography technology.

[0068] In previous research, we proposed a post-mask model based on a decomposition-based machine learning method, which achieved fast and high-precision near-field diffraction simulations under strong edge interference effects. Compared to the convolution method, it achieved a significant improvement in computational accuracy within an acceptable computation time. However, previous studies mainly focused on coherent lithography systems, therefore this thick mask model is insufficient to simulate the imaging model in partially coherent lithography systems.

[0069] Existing methods for calculating near-field distribution in photolithography can be categorized into Kirchhoff's approximation, rigorous electromagnetic field calculation, and fast calculation methods. Kirchhoff's approximation can quickly obtain the near-field distribution corresponding to the mask pattern. However, as chip manufacturing processes shrink year by year, the pattern spacing in photolithography is now much smaller than the wavelength of the light source, rendering Kirchhoff's approximation invalid and resulting in low accuracy of the near-field distribution obtained using this method. Rigorous electromagnetic field calculation can yield accurate near-field distributions, but the computational load is too enormous, making the simulation of large-scale near-field distributions prohibitively time-consuming.

[0070] Existing fast computation methods are based on convolution calculations. First, rigorous electromagnetic field calculations are used to calculate the accurate near-field distribution corresponding to a subset of patterns, which is then used as a training set. Next, using the training set as a standard, a convolution kernel is calculated to reflect the relationship between the pattern and its corresponding near-field distribution. Finally, the entire pattern is convolved using this kernel to obtain the near-field distribution corresponding to the target pattern. The results of fast computation methods have smaller errors compared to the accurate results of rigorous electromagnetic field calculations, while offering a significant speed improvement, making it possible to simulate the near-field distribution of large-scale patterns with high accuracy. However, on the one hand, there is still room for improvement in both the accuracy and speed of fast computation methods; on the other hand, for direct UV lithography, because the spacing between the fabricated patterns is much smaller than the wavelength of the light source, the edge interference effect in the near field is significant, such as… Figure 2 As shown in (a), a "wavy" field strength distribution will appear inside the graph. Figure 2 As shown in (b), the convolution method cannot restore the "wavy" field intensity distribution caused by diffraction. In our previous research, the thick mask model based on decomposition machine learning was proposed, using a "Diffraction Transfer Matrix" (DTM) instead of a traditional single convolution kernel. Figure 2As shown in (c), this method can effectively restore the "wavy" field intensity distribution caused by diffraction, improve the accuracy of the calculation results, and reduce the calculation time.

[0071] However, this method is based on coherent illumination systems for calculations, while actual lithography systems use partially coherent illumination as the light source. Therefore, this method cannot simulate the imaging model of partially coherent illumination lithography systems, thus hindering the application of subsequent resolution enhancement techniques (such as OPC, SMO, etc.).

[0072] Figure 3 A schematic diagram illustrating an example of photolithography imaging to demonstrate the limitations of using coherent illumination in the near field. Figure 3 'a' shows the pupil plane. The ring L represents the ring illumination used in this example, with an inner coherence factor of 0.4 and an outer coherence factor of 0.6. In the lithography system, we use the Abbe model to calculate the lithographic spatial image. Figure 3 In the rigorous Abbe model shown in b, four light source points are used on the pupil surface to represent annular illumination. These four light source points are... Figure 3 K is used as the denotor in a. The near fields of the four light sources are calculated independently, and the spatial image is obtained through the Abbe imaging model; in Figure 3 In the approximate Abbe model shown in c, the near field of coherent illumination is used to approximate the near field of ring illumination, and the coherent illumination source point is at... Figure 3 The denoting element M is used in a. After obtaining the near field of coherent illumination, the spatial image is calculated using an imaging model with four light sources. Figure 3 d shows the difference between the spatial images of the strict Abbe model and the approximate Abbe model; Figure 3 e shows the spatial image cross-sections of the strict Abbe model and the approximate Abbe model. Using a threshold function to approximate the photoresist, it can be seen that the exposure cross-section length at the cross-section is 27 nm for the strict Abbe model and 38 nm for the approximate Abbe model. The critical dimension (CD) of the layout at this location is 54 nm.

[0073] As can be seen, the spatial image calculated by the approximate Abbe model has a large error. Therefore, for near-field models, calculating only the near field of coherent illumination is limited and cannot meet the requirements of high-precision imaging calculation models. Therefore, this application proposes a method, apparatus, and electronic device for simulating the near field of a mask under partially coherent illumination. In a DUV lithography scene with partially coherent illumination, a corresponding diffraction near-field dataset is established. The least squares method (a machine learning method) is used to fit the DTM corresponding to the light source point. After obtaining the fitted DTM, matrix multiplication is performed between the layout and the DTM to achieve fast and high-accuracy simulation calculation of the near-field distribution of the target mask pattern, thus meeting the requirements of high-precision imaging calculation models. Specific embodiments are as follows:

[0074] Figure 4 This document illustrates a flowchart of a mask near-field simulation method under partially coherent illumination, as provided in an embodiment of this application. Figure 4 As shown, the near-field simulation method for masks under partially coherent illumination includes:

[0075] Step 101: Divide the pupil plane into a preset number of light source points with different incident directions.

[0076] In this application, the pupil surface can be divided into 13 light source points with different incident directions by symmetrical arrangement at equal intervals.

[0077] The embodiments of this application do not limit the specific incident direction, and can be adjusted according to the actual application scenario.

[0078] Step 102: For each of the light source points, determine the four diffraction near-field matrices of the multiple input training pattern graphics through rigorous electromagnetic field simulation.

[0079] Step 103: Based on the multiple sets of the four diffraction near-field matrices corresponding to the multiple training pattern graphics, determine the near-field distribution training data of the light source points corresponding to the preset pattern types.

[0080] In this application, each training pattern can be divided into a convex corner part, a concave corner part, a vertical edge part, and a horizontal edge part according to preset division conditions. For each part, the near-field distribution training data of the light source point corresponding to each corresponding light source point is determined by multiple sets of four diffraction near-field matrices.

[0081] Step 104: Based on the near-field distribution training data of multiple light source points, determine the diffraction transfer matrix of each of the preset pattern types using the least squares method.

[0082] In this application, a training data set for each of the partial patterns corresponding to the near-field distribution training data of multiple light source points can be obtained; for each partial pattern, the diffraction transfer matrix of each partial pattern is determined by the least squares method based on the corresponding training data set.

[0083] Step 105: Upon receiving the target incident light source, determine the light source point corresponding to the target incident light source.

[0084] Step 106: Calculate the near-field distribution of the target mask pattern corresponding to the incident light source based on the diffraction transfer matrix corresponding to the light source point, and determine the near-field result.

[0085] In this application, the target mask pattern can be divided into corresponding target convex corner part pattern, target concave corner part pattern, target vertical edge part pattern and target horizontal edge part pattern based on the preset division conditions; based on the vectorized mask pattern corresponding to each part pattern, near-field distribution calculation is performed in combination with the diffraction transfer matrix of each part pattern corresponding to the light source point to obtain the near-field distribution corresponding to each part pattern.

[0086] In summary, the near-field simulation method for masks under partially coherent illumination described in this application divides the pupil plane into a predetermined number of light source points with different incident directions. For each light source point, four diffraction near-field matrices of multiple input training pattern graphics are determined through rigorous electromagnetic field simulation. Based on multiple sets of the four diffraction near-field matrices corresponding to the multiple training pattern graphics, training data for the near-field distribution of light source points corresponding to the corresponding predetermined pattern types are determined. Based on the training data for the near-field distribution of light source points, the diffraction transfer matrix of all the corresponding predetermined pattern types is determined using the least squares method. When a target incident light source is received, the light source point corresponding to the target incident light source is determined. Based on the diffraction transfer matrix corresponding to the light source point, the near-field distribution of the target mask pattern corresponding to the target incident light source is calculated to determine the near-field result. This method enables fast and accurate calculation of the near-field generated by any light source point on the pupil plane, thereby providing a computational basis for subsequent resolution enhancement technology under partially coherent illumination, expanding the application scenarios of near-field calculation, and ensuring that the speed and accuracy of near-field calculation meet the requirements of high-precision imaging calculation models.

[0087] Figure 5 This document illustrates a flowchart of another mask near-field simulation method under partially coherent illumination provided in an embodiment of this application. Figure 5 As shown, the method includes:

[0088] Step 201: Divide the pupil plane into a preset number of light source points with different incident directions.

[0089] In this application, the pupil surface can be divided into 13 light source points with different incident directions by symmetrical arrangement at equal intervals.

[0090] Figure 6 The illustration shows a pixelated pupil surface and a training set establishment process provided in an embodiment of this application, as shown in the diagram. Figure 6 As shown in (a), each light source point represents an incident direction of illumination. In this application, the pupil surface can be divided into 13 light source points.

[0091] Step 202: For each of the light source points, determine the four diffraction near-field matrices of the multiple input training pattern graphics through rigorous electromagnetic field simulation.

[0092] In this application, Figure 6 (b) shows the process of establishing the training set. For each light source point, through rigorous electromagnetic field simulation, four diffraction near-field matrices of each input training pattern are calculated. These diffraction near-field matrices can be used as training data for that light source point to perform the next step of fitting the diffraction transfer matrix (DTM) of the corresponding light source point.

[0093] In this application, DTM represents the transformation relationship between a vectorized mask pattern and the corresponding vectorized diffraction near field. "T" is used in equations (1) and (2) of this application.

[0094] Step 203: Divide each training pattern into a convex corner portion, a concave corner portion, a vertical edge portion, and a horizontal edge portion according to the preset division conditions.

[0095] In this application, since the patterns on the photomask are all Manhattan patterns, that is, patterns composed of vertical and horizontal rectangles, the convex and concave corners of the patterns can be found, and points can be selected at certain intervals along the horizontal and vertical sides. Based on these points, the entire pattern can be decomposed and summarized into four types of parts: convex corner parts, concave corner parts, vertical side parts, and horizontal side parts.

[0096] Step 204: For each partial image, determine the training data of the near-field distribution of the light source point corresponding to each corresponding light source point by using multiple sets of the four diffraction near-field matrices.

[0097] The training data for the near-field distribution of light source points may include... In this application, for each partial graphic element, the following relationship exists:

[0098]

[0099] in, The mask pattern representing the selected region is transformed into a vector representation. If the size of the selected mask pattern region is (N... m ×N m After transformation For (N) m 2 A vector of size ×1, whose elements are 1 and 0, representing the light transmission or obstruction at the corresponding position of the mask pattern. Representing the near-field distribution at the same location, and also transformed into a vector representation. If the selected mask pattern area size is (N d ×N d After transformation For (N)d 2 A vector of size (×1), whose elements are the near-field amplitudes at the corresponding positions. T is the DTM corresponding to the type of shape (convex corner, concave corner, vertical edge, horizontal edge) under the corresponding light source point, and is of size (N). d 2 ×N m 2 The matrix represents the transformation from the mask pattern to the near-field distribution of the light source point within this pattern type.

[0100] Step 205: Based on the near-field distribution training data of multiple light source points, determine the diffraction transfer matrix of each of the preset pattern types using the least squares method.

[0101] In this application, a training data set for each of the partial patterns corresponding to the near-field distribution training data of multiple light source points can be obtained; for each partial pattern, the diffraction transfer matrix of each partial pattern is determined by the least squares method based on the corresponding training data set.

[0102] Specifically, after obtaining the training data on the near-field distribution of the corresponding light source points, for a specific type of graphic, that is, for a selected portion of the graphic (training graphic corresponding to convex corners, concave corners, vertical edges, or horizontal edges), the mask graphic region and its corresponding near-field distribution region of this type of partial graphic can be found in the training set. and All are known. Calculate the DTM for this graphic type using the least squares method:

[0103]

[0104] In this context, the subscript "x" represents a type of shape (convex corner, concave corner, vertical edge, horizontal edge); the superscript "T" represents the matrix transpose; and the superscript "-1" represents the matrix inverse.

[0105] Step 206: Upon receiving the target incident light source, determine the light source point corresponding to the target incident light source.

[0106] In this application, when a target incident light source is received, a light source point that matches the incident angle of the target incident light source can be determined.

[0107] Step 207: Calculate the near-field distribution of the target mask pattern corresponding to the incident light source based on the diffraction transfer matrix corresponding to the light source point, and determine the near-field result.

[0108] In this application, the target mask pattern can be divided into corresponding target convex corner part pattern, target concave corner part pattern, target vertical edge part pattern and target horizontal edge part pattern based on the preset division conditions; based on the vectorized mask pattern corresponding to each part pattern, near-field distribution calculation is performed in combination with the diffraction transfer matrix of each part pattern corresponding to the light source point to obtain the near-field distribution corresponding to each part pattern.

[0109] Specifically, after determining the kernel matrix (i.e., the diffraction transfer matrix) corresponding to the four pattern types, the near-field distribution of the target mask pattern can be quickly calculated under the corresponding light source. First, the target mask pattern can be divided into multiple regions using the same segmentation method as the training set pattern, and then categorized into different pattern types. According to formula (1), the corresponding DTM is used to perform matrix multiplication with the vectorized mask pattern to obtain the near-field distribution corresponding to each region. In existing methods for quickly calculating near-field distribution, the calculation is limited to mask diffraction near-field under coherent illumination conditions and cannot be applied to imaging calculations of some coherent illumination lithography systems.

[0110] In this application, a training set can be established for each of the light source points on the pupil surface, and the corresponding DTM can be fitted on this set. This allows for independent, fast, and high-precision near-field calculations for each light source point. The obtained near-field results can be further applied to partially coherent illumination imaging models, providing a foundation for subsequent photolithography imaging simulations and resolution enhancement technologies.

[0111] This application proposes a fast mask near-field calculation method based on a partially coherent illumination lithography system, which can efficiently calculate the diffraction near-field of the target pattern under partially coherent illumination conditions while ensuring high calculation accuracy. Figure 7 b shows the positions of the three light source points on the pupil plane in the example, labeled by P; Figure 7 Image a shows the near-field calculation results using this method at different light source points. The first, second, and third rows of images show the diffraction near-fields at light source points 3, 4, and 8, respectively. The first, second, third, and fourth columns show the E(XX) diffraction near-field obtained from rigorous electromagnetic field simulation, the E(XX) diffraction near-field calculated using this method, the E(YY) diffraction near-field obtained from rigorous electromagnetic field simulation, and the E(YY) diffraction near-field calculated using this method, respectively. Since the overall amplitude of the E(XY) and E(YX) diffraction near-fields is much smaller than that of the E(XX) and E(YY) diffraction near-fields (approximately 1%), their impact on the results of partially coherent illumination lithography imaging models is relatively small and is therefore not shown in this paper.

[0112] The diffraction near-field (DNF) represents the electric field distribution beneath the mask during photolithography. DNF has four components, representing: the electric field generating the X-polarized component when a unit of X-polarized light is incident; the electric field generating the Y-polarized component when a unit of X-polarized light is incident; the electric field generating the X-polarized component when a unit of Y-polarized light is incident; and the electric field generating the X-polarized component when a unit of Y-polarized light is incident. Due to the metal shielding layer structure and tiny pattern size of the mask, the diffraction near-field exhibits severe diffraction and edge interference effects, leading to a discrepancy between its distribution and the designed mask pattern. The method proposed in this application rapidly calculates the diffraction near-field generated at a given light source point from a given mask pattern.

[0113] Table 1 shows the diffraction near-field errors calculated using this method. We use diffraction near-field data obtained from rigorous electromagnetic field simulations as the standard, and demonstrate the accuracy of the calculation results using root-mean-square (RMS) error and signal-to-noise ratio (SNR). Figure 5 As shown in Table 1, this method can achieve high-precision simulation of the diffraction near field generated by different light source points. This near-field model can be further applied to partially coherent lithography imaging models to improve the computational efficiency of imaging models and resolution enhancement techniques (OPC, SMO, etc.).

[0114] Table 1

[0115]

[0116] In this application, after pixelating the entire pupil, a near-field dataset for each light source point is established using rigorous electromagnetic field simulation, and the DTM under the corresponding conditions is fitted using the least squares method. For a given new layout, the near-field distribution under that light source point can be quickly calculated using the DTM of the corresponding light source point. In this way, the near-field generated by any light source point on the pupil surface can be calculated quickly and accurately, thus providing a computational basis for subsequent resolution enhancement techniques under partially coherent illumination.

[0117] The near-field simulation method for masks under partially coherent illumination described in this application divides the pupil plane into a predetermined number of light source points with different incident directions. For each light source point, rigorous electromagnetic field simulation is used to determine four diffraction near-field matrices for multiple input training pattern graphics. Based on multiple sets of the four diffraction near-field matrices corresponding to the multiple training pattern graphics, near-field distribution training data for light source points corresponding to the corresponding predetermined pattern types are determined. Based on the near-field distribution training data for multiple light source points, the least squares method is used to determine the diffraction transfer matrices for all the corresponding predetermined pattern types. When a target incident light source is received, the light source point corresponding to the target incident light source is determined. Based on the diffraction transfer matrix corresponding to the light source point, the near-field distribution of the target mask graphic corresponding to the target incident light source is calculated to determine the near-field result. This method enables fast and accurate calculation of the near-field generated by any light source point on the pupil plane, thus providing a computational foundation for subsequent resolution enhancement techniques under partially coherent illumination, expanding the application scenarios of near-field calculation, and ensuring that the speed and accuracy of near-field calculation meet the requirements of high-precision imaging calculation models.

[0118] Figure 8 This application provides a schematic diagram of the structure of a mask near-field simulation computing device under partially coherent illumination, as illustrated in an embodiment of the present application. Figure 8 As shown, the mask near-field simulation computing device 300 under partially coherent illumination includes:

[0119] The dividing module 301 is used to divide the pupil surface into a preset number of light source points with different incident directions;

[0120] The first determining module 302 is used to determine the four diffraction near-field matrices of the multiple input training pattern graphics for each of the light source points through rigorous electromagnetic field simulation.

[0121] The second determining module 303 is used to determine the near-field distribution training data of the light source points corresponding to the preset graphic type based on the multiple sets of the four diffraction near-field matrices corresponding to the multiple training pattern graphics.

[0122] The third determining module 304 is used to determine the diffraction transfer matrix of all the preset graphic types respectively by using the least squares method based on the near-field distribution training data of multiple light source points.

[0123] The fourth determining module 305 is used to determine the light source point corresponding to the target incident light source when the target incident light source is received;

[0124] The fifth determining module 306 is used to calculate the near-field distribution of the target mask pattern corresponding to the target incident light source based on the diffraction transmission matrix corresponding to the light source point, and determine the near-field result.

[0125] Optionally, the second determining module includes:

[0126] The first division submodule is used to divide each training pattern into a convex corner part, a concave corner part, a vertical edge part, and a horizontal edge part according to a preset division condition.

[0127] The first determining submodule is used to determine, for each part of the graphic, the near-field distribution training data of the light source point corresponding to each corresponding light source point through multiple sets of four diffraction near-field matrices.

[0128] Optionally, the third determining module includes:

[0129] The acquisition submodule is used to acquire the training data set of each of the partial graphics corresponding to the near-field distribution training data of multiple light source points;

[0130] The second determining submodule is used to determine the diffraction transfer matrix of each of the partial graphics using the least squares method based on the corresponding training data set.

[0131] Optionally, the fifth determining module includes:

[0132] The second division module is used to divide the target mask pattern into corresponding target convex corner part pattern, target concave corner part pattern, target vertical edge part pattern and target horizontal edge part pattern based on the preset division conditions;

[0133] The third determining submodule is used to perform near-field distribution calculation based on the vectorized mask pattern corresponding to each part of the pattern and the diffraction transfer matrix of each part of the pattern corresponding to the light source point, so as to obtain the near-field distribution corresponding to each part of the pattern.

[0134] Optionally, the partitioning module includes:

[0135] The third dividing submodule is used to divide the pupil surface into 13 light source points with different incident directions in a symmetrical arrangement with equal spacing.

[0136] The near-field simulation computing device for masks under partially coherent illumination described in this application divides the pupil plane into a preset number of light source points with different incident directions. For each light source point, through rigorous electromagnetic field simulation, four diffraction near-field matrices for multiple input training pattern graphics are determined. Based on multiple sets of the four diffraction near-field matrices corresponding to the multiple training pattern graphics, training data for the near-field distribution of light source points corresponding to preset pattern types are determined. Based on the training data for the near-field distribution of light source points, the diffraction transfer matrix for all preset pattern types is determined using the least squares method. When a target incident light source is received, the light source point corresponding to the target incident light source is determined. Based on the diffraction transfer matrix corresponding to the light source point, the near-field distribution of the target mask pattern corresponding to the target incident light source is calculated to determine the near-field result. This allows for fast and accurate calculation of the near-field generated by any light source point on the pupil plane, thus providing a computational foundation for subsequent resolution enhancement techniques under partially coherent illumination, expanding the application scenarios of near-field calculation, and ensuring that the speed and accuracy of near-field calculation meet the requirements of high-precision imaging calculation models.

[0137] This application provides a near-field simulation computing device for masks under partially coherent illumination, used to achieve, for example... Figures 4 to 7 To avoid repetition, the near-field simulation method for masks under any of the partially coherent illumination conditions shown will not be described again here.

[0138] The electronic device in this application embodiment can be a device, or a component, integrated circuit, or chip in a terminal. The device can be a mobile electronic device or a non-mobile electronic device. For example, a mobile electronic device can be a mobile phone, tablet computer, laptop computer, PDA, in-vehicle electronic device, wearable device, ultra-mobile personal computer (UMPC), netbook, or personal digital assistant (PDA), etc., while a non-mobile electronic device can be a server, network-attached storage (NAS), personal computer (PC), television (TV), ATM, or self-service machine, etc., and this application embodiment does not specifically limit the scope.

[0139] The electronic device in this application embodiment can be a device with an operating system. This operating system can be Android, iOS, or other possible operating systems; this application embodiment does not specifically limit the specific operating system used.

[0140] Figure 9A schematic diagram of the hardware structure of an electronic device according to an embodiment of this application is shown. Figure 9 As shown, the electronic device 400 includes a processor 410.

[0141] like Figure 9 As shown, the processor 410 described above can be a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits used to control the execution of the program of the present application.

[0142] like Figure 9 As shown, the electronic device 400 may further include a communication line 440. The communication line 440 may include a path for transmitting information between the components.

[0143] Optional, such as Figure 9 As shown, the above-described electronic device may further include a communication interface 420. There may be one or more communication interfaces 420. The communication interface 420 may use any transceiver-like device for communicating with other devices or communication networks.

[0144] Optional, such as Figure 9 As shown, the electronic device may further include a memory 430. The memory 430 stores computer execution instructions for implementing the scheme of this application, and its execution is controlled by a processor. The processor executes the computer execution instructions stored in the memory to implement the method provided in the embodiments of this application.

[0145] like Figure 9As shown, memory 430 can be read-only memory (ROM) or other types of static storage devices capable of storing static information and instructions, random access memory (RAM) or other types of dynamic storage devices capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. Memory 430 can exist independently and be connected to processor 410 via communication line 440. Memory 430 can also be integrated with processor 410.

[0146] Optionally, the computer execution instructions in the embodiments of this application may also be referred to as application code, and the embodiments of this application do not specifically limit this.

[0147] In a specific implementation, as one example, such as Figure 9 As shown, processor 410 may include one or more CPUs, such as Figure 9 CPU0 and CPU1 in the CPU.

[0148] In a specific implementation, as one example, such as Figure 9 As shown, the terminal device may include multiple processors, such as Figure 9 The processors in the system. Each of these processors can be a single-core processor or a multi-core processor.

[0149] Figure 10 This is a schematic diagram of the chip structure provided in an embodiment of this application. Figure 10 As shown, the chip 500 includes one or more processors 410.

[0150] Optional, such as Figure 10 As shown, the chip also includes a communication interface 420 and a memory 430. The memory 430 may include read-only memory and random access memory, and provides operation instructions and data to the processor. A portion of the memory may also include non-volatile random access memory (NVRAM).

[0151] In some implementations, such as Figure 10 As shown, memory 430 stores the following elements: execution modules or data structures, or subsets thereof, or extended sets thereof.

[0152] In the embodiments of this application, such as Figure 10 As shown, the corresponding operation is executed by calling the operation instructions stored in the memory (which can be stored in the operating system).

[0153] like Figure 10 As shown, the processor 410 controls the processing operations of any one of the terminal devices. The processor 410 can also be called a central processing unit (CPU).

[0154] like Figure 10 As shown, memory 430 may include read-only memory and random access memory, providing instructions and data to the processor. A portion of memory 430 may also include NVRAM. For example, in an application, memory, communication interfaces, and memory are coupled together via a bus system, which may include, in addition to a data bus, a power bus, a control bus, and a status signal bus, etc. However, for clarity, in... Figure 10 The general labeled all buses as Bus System 540.

[0155] like Figure 10 As shown, the methods disclosed in the embodiments of this application can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by integrated logic circuits in the processor's hardware or by instructions in software form. The processor can be a general-purpose processor, a digital signal processor (DSP), an ASIC, a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly embodied as being executed by a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules can be located in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. This storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method.

[0156] On the one hand, a computer-readable storage medium is provided, which stores instructions that, when executed, implement the functions performed by the terminal device in the above embodiments.

[0157] On the one hand, a chip is provided that is used in a terminal device. The chip includes at least one processor and a communication interface. The communication interface and at least one processor are coupled together. The processor is used to run instructions to implement the functions performed by the mask near-field simulation calculation method under partially coherent illumination in the above embodiments.

[0158] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed on a computer, the processes or functions described in the embodiments of this application are performed entirely or partially. The computer can be a general-purpose computer, a special-purpose computer, a computer network, a terminal, a user equipment, or other programmable device. The computer program or instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, the computer program or instructions can be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; it can also be an optical medium, such as a digital video disc (DVD); or it can be a semiconductor medium, such as a solid-state drive (SSD).

[0159] Although this application has been described herein in conjunction with various embodiments, those skilled in the art, by reviewing the accompanying drawings, the disclosure, and the appended claims, will understand and implement other variations of the disclosed embodiments in carrying out the claimed application. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude multiple instances. A single processor or other unit can implement several functions listed in the claims. While different dependent claims may recite certain measures, this does not mean that these measures cannot be combined to produce good results.

[0160] Although this application has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made thereto without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely exemplary illustrations of this application as defined by the appended claims, and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from the spirit and scope of this application. Thus, if such modifications and modifications of this application fall within the scope of the claims of this application and their equivalents, this application is also intended to include such modifications and modifications.

Claims

1. A method for simulating the near-field of a mask under partially coherent illumination, characterized in that, The method includes: The pupil plane is divided into a preset number of light source points with different incident directions; For each of the aforementioned light source points, four diffraction near-field matrices of the multiple input training pattern graphics are determined through rigorous electromagnetic field simulation. Based on the multiple sets of the four diffraction near-field matrices corresponding to the multiple training pattern graphics, the near-field distribution training data of the light source points corresponding to the preset pattern types are determined respectively. Based on the near-field distribution training data of multiple light source points, the diffraction transfer matrix of each of the preset pattern types is determined by the least squares method. Upon receiving a target incident light source, the light source point corresponding to the target incident light source is determined; The near-field distribution of the target mask pattern corresponding to the incident light source is calculated based on the diffraction transmission matrix corresponding to the light source point to determine the near-field result.

2. The method for near-field simulation of a mask under partially coherent illumination according to claim 1, characterized in that, The step of determining the near-field distribution training data of light source points corresponding to the preset pattern types based on multiple sets of the four diffraction near-field matrices corresponding to multiple training pattern patterns includes: Each training pattern is divided into a convex corner portion, a concave corner portion, a vertical edge portion, and a horizontal edge portion according to preset division conditions. For each part of the image, the training data of the near-field distribution of the light source point corresponding to each corresponding light source point is determined by multiple sets of the four diffraction near-field matrices.

3. The method for near-field simulation of a mask under partially coherent illumination according to claim 2, characterized in that, The step of determining the diffraction transfer matrix for all preset pattern types based on near-field distribution training data of multiple light source points using the least squares method includes: Obtain the training data set for each of the partial graphics corresponding to the near-field distribution training data of multiple light source points; For each of the said partial patterns, the diffraction transfer matrix of each partial pattern is determined using the least squares method based on the corresponding training data set.

4. The method for near-field simulation of a mask under partially coherent illumination according to claim 2, characterized in that, The near-field distribution of the target mask pattern corresponding to the incident light source is calculated based on the diffraction transfer matrix corresponding to the light source point to determine the near-field result, including: Based on the preset division conditions, the target mask pattern is divided into corresponding target convex corner part pattern, target concave corner part pattern, target vertical edge part pattern and target horizontal edge part pattern; Based on the vectorized mask pattern corresponding to each partial pattern, near-field distribution calculation is performed in combination with the diffraction transfer matrix of each partial pattern corresponding to the light source point to obtain the near-field distribution corresponding to each partial pattern.

5. The method for near-field simulation of a mask under partially coherent illumination according to claim 1, characterized in that, The method of dividing the pupil plane into a predetermined number of light source points with different incident directions includes: The pupil surface is divided into 13 light source points with different incident directions by symmetrical arrangement at equal intervals.

6. A near-field simulation computing device for a mask under partially coherent illumination, characterized in that, The device includes: The partitioning module is used to divide the pupil plane into a preset number of light source points with different incident directions; The first determining module is used to determine the four diffraction near-field matrices of the multiple input training pattern graphics for each of the light source points through rigorous electromagnetic field simulation. The second determining module is used to determine the near-field distribution training data of light source points corresponding to the preset graphic types based on the multiple sets of the four diffraction near-field matrices corresponding to the multiple training pattern graphics. The third determining module is used to determine the diffraction transfer matrix of all the preset graphic types respectively by using the least squares method based on the near-field distribution training data of multiple light source points. The fourth determining module is used to determine the light source point corresponding to the target incident light source when the target incident light source is received; The fifth determining module is used to calculate the near-field distribution of the target mask pattern corresponding to the target incident light source based on the diffraction transmission matrix corresponding to the light source point, and determine the near-field result.

7. The near-field simulation computing device for a mask under partially coherent illumination according to claim 6, characterized in that, The second determining module includes: The first division submodule is used to divide each training pattern into a convex corner part, a concave corner part, a vertical edge part, and a horizontal edge part according to a preset division condition. The first determining submodule is used to determine, for each part of the graphic, the near-field distribution training data of the light source point corresponding to each corresponding light source point through multiple sets of four diffraction near-field matrices.

8. The near-field simulation computing device for a mask under partially coherent illumination according to claim 7, characterized in that, The third determining module includes: The acquisition submodule is used to acquire the training data set of each of the partial graphics corresponding to the near-field distribution training data of multiple light source points; The second determining submodule is used to determine the diffraction transfer matrix of each of the partial graphics using the least squares method based on the corresponding training data set.

9. The near-field simulation computing device for a mask under partially coherent illumination according to claim 7, characterized in that, The fifth determining module includes: The second division module is used to divide the target mask pattern into corresponding target convex corner part pattern, target concave corner part pattern, target vertical edge part pattern and target horizontal edge part pattern based on the preset division conditions; The third determining submodule is used to perform near-field distribution calculation based on the vectorized mask pattern corresponding to each part of the pattern and the diffraction transfer matrix of each part of the pattern corresponding to the light source point, so as to obtain the near-field distribution corresponding to each part of the pattern.

10. An electronic device, characterized in that, include: One or more processors; And one or more machine-readable media thereon storing instructions, which, when executed by the one or more processors, cause to perform the mask near-field simulation computation method under partially coherent illumination as described in any one of claims 1 to 5.

Images