A method for process parameter scanning and process window analysis in electron beam lithography EDA software

By reading the GDSII layout and calculating the ideal CD and EPE, combined with MPI parallel computing and exposure energy deposition model, the problems of low efficiency in pattern quality measurement and low efficiency in process window determination in electron beam direct writing lithography are solved, and rapid process parameter scanning and evaluation are realized.

CN123117004APending Publication Date: 2026-11-17HUNAN UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611280028.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-22
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In existing electron beam direct writing lithography processes, the efficiency of pattern quality measurement is low, the impact of process parameters on pattern quality is difficult to assess quickly, and the efficiency of determining the process window is low.

Method used

By reading the GDSII layout, extracting polygonal geometric information, calculating ideal CD and EPE, and combining MPI parallel computing, an exposure energy deposition model is established, and the dose and focal length process windows are analyzed to achieve rapid simulation and evaluation.

Benefits of technology

It improves the calculation speed and evaluation efficiency of electron beam lithography process parameters, has strong scalability, and can quickly scan and determine the optimal process parameters and process window.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123117004A_ABST
    Figure CN123117004A_ABST
Patent Text Reader

Abstract

This invention proposes a method for scanning process parameters and analyzing process windows in electron beam lithography EDA software. The method involves reading the GDSII layout, extracting the geometric information of polygon edges, determining the CD measurement pairing edges based on attitude, normal, projection overlap, and distance threshold, and calculating the ideal CD. The actual development contour is searched along the normal to calculate the EPE, thereby obtaining the actual CD, CDU, and LER. Parameters such as electron beam accelerating voltage, beam spot size, exposure dose, focusing parameters, PEC iteration count, minimum fragmentation pattern size, and development threshold are scanned. An exposure energy deposition model is established based on a double Gaussian point spread function. Based on the target CD and tolerance analysis, the pattern quality under different dose and defocusing conditions is determined, obtaining the dose process window, dose-CD curve, and focal length process window, enabling rapid simulation and evaluation of electron beam direct-write lithography process parameters.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to a method for scanning process parameters and analyzing process windows in EDA software for electron beam lithography. Based on an electron beam exposure energy deposition model, the method rapidly scans electron beam lithography process parameters through critical dimension (CD) simulation, pattern quality evaluation, and process window analysis. It features fast calculation speed, high evaluation efficiency, and strong scalability, and belongs to the field of electronic design automation. Background Technology

[0002] As integrated circuit manufacturing processes continue to advance to more advanced process nodes, the size of lithographic patterns continues to shrink, placing higher demands on pattern accuracy and processing quality. Electron Beam Direct Write Lithography (EBDW) has advantages such as high resolution, no need for masks, and flexible pattern modification, and is widely used in advanced lithographic mask manufacturing, nanodevice fabrication, small-batch chip manufacturing, and lithography process research.

[0003] During electron beam exposure, the exposure energy distribution changes due to factors such as electron forward scattering, backscattering, and proximity effect, causing deviations between the actual pattern size and the design layout. Simultaneously, equipment physical parameters such as electron beam accelerating voltage, beam spot size, exposure dose, and focusing state, as well as graphical control parameters such as the number of proximity effect correction (PEC) iterations, minimum fragmentation pattern size, and development threshold, all influence the final exposure result. Therefore, during process development, extensive simulation calculations are required for different combinations of process parameters, and a comprehensive evaluation of pattern quality indicators such as critical dimension (CD), edge placement error (EPE), critical dimension uniformity (CDU), and line edge roughness (LER) is necessary to determine the optimal process parameters and process window to meet manufacturing requirements.

[0004] This invention proposes a method for evaluating and finding optimal process parameters. It can calculate the CD and EPE of an electron beam lithography process, use the CD and EPE to calculate the CDU and LER of the layout, and use this as a basis to scan process parameters and analyze the process window. Summary of the Invention

[0005] This invention addresses the methods for scanning process parameters and analyzing process windows in electron beam lithography. It establishes a CD simulation and rapid process window evaluation method for electron beam direct-write lithography to solve the problems of low efficiency in CD measurement of layout patterns, difficulty in rapidly evaluating the impact of process parameters on pattern quality, and low efficiency in determining process windows in existing electron beam direct-write lithography processes.

[0006] The technical solution of this invention is as follows: Read the GDSII layout and extract polygonal geometric information; determine the CD measurement edge using an edge pairing method based on edge direction, normal, projection overlap, and distance threshold; calculate the ideal CD; obtain the EPE by searching the actual development contour along the edge normal; further calculate the actual CD, CDU, and LER based on EPE statistics; scan the physical parameters and graphical control parameters of different electron beam direct-write lithography equipment; and use MPI multi-node parallel computing to improve the simulation efficiency of large-scale parameter combinations; establish an exposure energy deposition model based on the double Gaussian point diffusion function; analyze the graphic changes under different exposure doses and defocusing conditions according to the target CD and its tolerance range; and determine the dose process window and focal length process window by combining CD, EPE, CDU, and LER, thereby achieving rapid simulation and evaluation of electron beam direct-write lithography process parameters and process windows.

[0007] The steps of the invention are as follows:

[0008] Step S1: Extraction and calculation of geometric information of the layout graphic;

[0009] The system reads the GDSII layout data to be analyzed, obtains the vertex set of each polygon in the layout, and converts each polygon into a set of edges according to the connection relationship of the polygon vertices. For each edge, it extracts geometric information such as its start point, end point, edge number, normal vector, horizontal or vertical attitude marker, and edge length.

[0010] Specifically, the unit normal vector pointing into the polygon is calculated based on the start and end points of the edge, and the horizontal or vertical edge is determined based on the relationship between the horizontal and vertical coordinates of the start and end points; the discrete integer grid point set corresponding to the edge is generated using the Bresenham line algorithm.

[0011] To facilitate subsequent edge pairing, CD, and EPE calculations, the geometric information of each edge is reserved and recorded with the paired edge number, EPE sampling point, ideal CD, and EPE value. Initially, the paired edge number is set to a null value. After edge pairing is completed, the EPE sampling point is determined based on the overlapping area of ​​the two paired edges, and the coordinates of the corresponding sampling points on the two paired edges are recorded. The ideal CD is obtained based on the geometric distance between the two paired edges, and the actual CD is subsequently calculated by combining the EPE values ​​at the sampling points.

[0012] Step S2: Polygon CD measurement and edge pairing;

[0013] Based on the set of polygon edges obtained in step S1, each edge is filtered by direction and the overlapping area is determined to identify the paired edges used for CD measurement.

[0014] First, the candidate pairing range is narrowed down based on the edge's orientation and normal direction: for horizontal edges, other horizontal edges are searched only in their normal direction as candidate pairs; for vertical edges, other vertical edges are searched only in their normal direction as candidate pairs, thereby reducing the search for irrelevant edges.

[0015] Subsequently, the projection interval overlap between the current edge and the candidate paired edge is determined. For two horizontal edges, it is determined whether the projection intervals of the two edges overlap in the horizontal direction; for two vertical edges, it is determined whether the projection intervals of the two edges overlap in the vertical direction. Only when the corresponding projection intervals of the two edges overlap are the candidate edges considered as valid paired candidate edges, so as to ensure that a valid CD measurement position can be established between the two edges in the future.

[0016] For candidate pairs that satisfy the directional constraints and projection overlap conditions, calculate the vertical distance between the current edge and each candidate edge, and filter them in ascending order of distance, selecting the candidate edge with the smallest distance as the pair of the current edge.

[0017] Simultaneously, a maximum allowed pairing distance threshold is set between edges. A pairing relationship between two edges is established only when the distance between the current edge and a candidate pairing edge is less than or equal to the threshold; if no candidate edge meets the distance threshold, no pairing relationship is established. Through direction constraints, projection overlap constraints, and distance threshold constraints, erroneous edge pairings are reduced, and non-target edges or edges that are too far away are avoided as CD measurement objects.

[0018] Step S3: Calculate and save the ideal CD;

[0019] Based on the pairing edge relationships obtained in step S2, calculate the ideal critical dimension CD between each pairing edge. For horizontal pairing edges, calculate the ideal CD based on the distance between the two edges in the corresponding coordinate directions; for vertical pairing edges, calculate the ideal CD based on the distance in the other coordinate direction. Save the obtained ideal CD and the corresponding pairing edge information to provide a benchmark for subsequent actual CD calculations.

[0020] Step S4: Calculate the EPE of the EPE sampling points to obtain the actual CD;

[0021] Based on the paired edges and their overlapping areas determined in step S2, EPE sampling points are determined in the discrete point set of the corresponding edges. Using the EPE sampling points as the ideal pattern boundary positions, the actual development contour is searched according to the normal direction of the corresponding edge at that position, and the actual development contour position is determined based on the exposure energy distribution and a preset development threshold. The signed distance between the ideal pattern boundary and the actual development contour in the normal direction is calculated, and this signed distance is used as the EPE of the corresponding sampling point to characterize the local positional offset of the pattern edge relative to the ideal pattern edge after exposure.

[0022] Step S5: Calculate the actual CD, as well as CDU and LER;

[0023] The actual CD after exposure is calculated based on the ideal CD obtained in step S3 and the EPE values ​​of the two EPE sampling points on the paired edge obtained in step S4.

[0024] Furthermore, the average CD and standard deviation of CD are calculated based on the actual CD obtained from multiple measurement positions in the layout, and the standard deviation of CD is used as CDU to characterize the degree of critical size change at different positions or different graphics.

[0025] Based on multiple EPE sampling data obtained along the edge direction of the graphic, the average value of EPE and its statistical dispersion are calculated. The statistical results of the edge position error are used as the LER evaluation index based on EPE to characterize the local fluctuation of the actual graphic edge relative to the ideal edge.

[0026] Step S6: Scan process parameters;

[0027] Based on a pre-defined range of process parameters, scanning step size, and parameter combination rules, a set of electron beam direct-write lithography process parameters is established, and different combinations of process parameters are scanned. The process parameters include equipment physical parameters and patterning control parameters. The equipment physical parameters include one or more of electron beam accelerating voltage, beam spot size, exposure dose, and focusing parameters. The patterning control parameters include one or more of proximity effect correction iterations, minimum fragmentation pattern size, and development threshold.

[0028] For each set of process parameters, the aforementioned layout processing, edge pairing, CD, and EPE calculation methods are applied to obtain the corresponding CD, EPE, CDU, and LER. Each combination of process parameters is then associated and stored with its corresponding graphic quality evaluation results. Multiple independent combinations of process parameters are processed using a multi-node parallel computing approach to improve the computational efficiency of large-scale process parameter scanning. After completing the calculation of all parameter combinations within the preset parameter range, the correspondence between process parameters and graphic quality evaluation indicators is output, providing a data foundation for process window analysis.

[0029] Step S7: Process window analysis;

[0030] Based on the process parameters obtained in step S6 and their corresponding CD, EPE, CDU and LER results, the dose process window and focal length process window are analyzed.

[0031] In this process, the dose process window analysis uses a double Gaussian point spread function to describe the energy deposition distribution per unit dose. The point spread function is then convolved with the pattern to be exposed to obtain the spatial energy deposition distribution under different exposure doses. The corresponding development pattern profile is obtained based on a preset development threshold, and CD, EPE, CDU, and LER are calculated for different exposure doses. The exposure dose is scanned according to a preset step size. Based on the target CD and its allowable tolerance, effective dose points that meet CD requirements are determined. Furthermore, EPE, CDU, and LER indicators are used to screen effective dose points that meet process requirements. The dose process window is determined from these effective dose points, and a correspondence between exposure dose and CD is established to obtain the dose-CD curve.

[0032] Focal length process window analysis establishes a correspondence between electron beam defocusing parameters and point spread functions, representing the beam blurring caused by defocusing as beam blurring parameters, and calculating effective forward scattering characteristic parameters under defocused conditions based on these parameters. By stepping and scanning the beam blurring parameters, the point spread function and corresponding spatial energy deposition distribution under different defocusing conditions are obtained. The corresponding development pattern profile is obtained based on the development threshold, and CD, EPE, CDU, and LER are calculated. Defocusing parameters that meet the requirements are selected based on the target CD and its allowable tolerance, and the effective defocusing range is further determined by combining EPE, CDU, and LER indicators. The focal length process window is obtained based on the correspondence between the defocusing parameters and the equipment focal length parameters. Attached Figure Description

[0033] Figure 1 This is an operation flowchart of the present invention;

[0034] Figure 2 It is a polygon CD measurement edge pairing; Detailed Implementation Plan

[0035] Step S1: Extraction and calculation of geometric information of the layout graphic;

[0036] The mask layout to be used is read in from the mask data, and the set of polygon vertices is obtained from the gdsii file. Each polygon is processed one by one, and each polygon is represented by a set of edges. Each edge has its own independent geometric information, including the edge start point, edge end point, edge number, edge normal, pose marker, length, paired edge number, line point set, EPE sampling point, ideal CD, and EPE value obtained from the polygon vertex set.

[0037] Edge start point, edge end point, edge number: Based on the extracted polygon vertex set, the points in the set are read one by one in a loop as the edge start point and the next point as the edge end point. The edge numbers are sorted from 1 to n-1 according to the order of the start points in the polygon vertex set, where n is the number of points in the polygon vertex set. Since the last point and the first point in the extracted polygon vertex set are the same point, there are actually only n vertices and n edges. In this way, the generated edges always keep the polygon vertices sorted in a counterclockwise direction.

[0038] Edge normal: Calculate a unit vector that always points inside the polygon as the basis for determining the edge normal. The normal vector x-component is calculated based on the edge's start and end points. The formula for calculating the x-component of the normal vector is:

[0039]

[0040] The formula for calculating the y-components of the normal vector is:

[0041]

[0042] In equations (1) and (2) , These are the x and y coordinates of the starting point of the edge. , These are the x and y coordinates of the endpoint of the edge;

[0043] Attitude indicator: Determine whether an edge is horizontal or vertical. If the x-coordinates of the edge's start and end points are the same, it is a vertical edge; if the y-coordinates are the same, it is a horizontal edge. Use a Boolean value to represent horizontal or vertical, with horizontal being true and vertical being false.

[0044] Length: Formula for calculating length:

[0045]

[0046] In formula (3) , These are the x and y coordinates of the starting point of the edge. , These are the x and y coordinates of the endpoint of the edge;

[0047] Paired edge number: After step S2, the number of the paired edge is stored in the structure for easy access when calculating CD. The initial value is -1.

[0048] Line point set: The set of discrete integer grid points traversed by the edge is generated using the Breseham line algorithm;

[0049] EPE sampling point: After edge pairing, select the center point of the set of straight line points overlapping with the paired edge as the EPE sampling point to sample the EPE to calculate the actual CD. Similarly, it is necessary to record the coordinates of the points on the paired edge that are opposite to it.

[0050] Ideal CD: After edge pairing, the distance between two edges is calculated as the ideal CD. When calculating the actual CD, only the EPE of two sampling points needs to be calculated to obtain the actual CD.

[0051] EPE value: Step S4 will calculate the EPE of the two sampling points and save it here;

[0052] Step S2: Polygon CD measurement and edge pairing;

[0053] To narrow down the pairing range, for horizontal edges, only horizontal edges located in their normal direction are considered; for vertical edges, only vertical edges located in their normal direction are considered. Then, it is determined whether the two edges have an overlapping area. Only if there is an overlapping area can CD be measured. Two horizontal edges can only be paired if their x-projection intervals overlap. Similarly, two vertical edges can only be paired if their y-projection intervals overlap. The formula for determining horizontal edges is as follows:

[0054]

[0055] In formula (4) It is the minimum x-coordinate of the current edge's starting and ending points. It is the minimum x-coordinate of the starting and ending points of the candidate edge. It is the maximum x-coordinate of the current edge's starting and ending points. It is the maximum x-coordinate of the starting and ending points of the candidate edge; similarly, the vertical edge is judged:

[0056]

[0057] In formula (5) It is the minimum y-coordinate of the current edge's starting and ending points. It is the minimum y-coordinate of the starting and ending points of the candidate edge. It is the maximum y-coordinate of the current edge's starting and ending points. It is the maximum y-coordinate of the starting and ending points of the candidate edge;

[0058] Next, the distance between the current edge and the candidate edge is recorded. The candidate edge with the smallest distance is selected as the pairing edge of the current edge. In order to ensure that the wrong edge is not paired and the long edge is not measured, the user will set a threshold to only pair edges with a distance less than this value.

[0059] Step S3: Calculate and save the ideal CD;

[0060] After step S2, calculate the ideal CD for each edge and its paired edge. If it is a horizontal edge, use the absolute value of the difference between the ordinates of the starting points of the two edges as the ideal CD. If it is a vertical edge, use the absolute value of the difference between the abscissas of the starting points of the two edges as the ideal CD. The calculation formula is as follows:

[0061]

[0062] In formula (6) , These are the x and y coordinates of the starting point of the current edge. , These are the x and y coordinates of the starting points of the paired edges. It is a gesture marker;

[0063] Step S4: Calculate the EPE of the EPE sampling points to obtain the actual CD;

[0064] For a single EPE measurement point, the position of the measurement point is first determined on the ideal layout boundary. This position is already stored in the EPE sampling points and can be used directly. The actual development contour is searched along the normal direction based on the normal vector to determine the position of the actual contour corresponding to the ideal measurement point. The actual contour is determined by the exposure energy distribution and the development threshold; that is, the position where the local energy reaches the development threshold is found. Finally, the signed distance between the ideal measurement point and the corresponding point of the actual contour in the normal direction is calculated, and this distance is taken as the EPE of the measurement point. The sign is determined according to the offset direction of the actual contour relative to the ideal boundary, thus characterizing the local offset of the pattern edge relative to the ideal layout edge after exposure.

[0065] Step S5: Calculate the actual CD, as well as CDU and LER;

[0066] An actual CD consists of an ideal CD for storage and two EPE values. The EPE is a signed value, and its calculation formula is as follows:

[0067]

[0068] In equation (7) and These are the EPE values ​​of the two sampling points. It is the ideal CD;

[0069] CDU is mainly used to describe the degree of CD variation at different locations or in different shapes. The average CD is calculated by measuring N CDs in the layout using the following formula:

[0070]

[0071] CDU is expressed as the standard deviation of CD, and the calculation formula is as follows:

[0072]

[0073] LER describes the degree of local fluctuation of the actual edge of a graphic relative to the ideal edge position. The edge position changes locally along the edge direction, and this change is the source of LER. The EPE that changes along the edge direction can actually be used to describe the edge roughness. In this invention, an LER evaluation method based on edge position error statistics is adopted, and the calculation formula is as follows:

[0074]

[0075] In equation (9) It is the EPE measured in the layout. This is the average of these EPEs;

[0076] Step S6: Scan process parameters;

[0077] Based on pre-set process parameter ranges, scanning step sizes, and parameter combination rules, the physical parameters of the equipment and the patterning control parameters in the electron beam direct-write lithography process are scanned to obtain the layout pattern quality evaluation results under different process parameter combinations, specifically including:

[0078] A set of process parameters is established based on the electron beam direct-write lithography process to be evaluated. These parameters include equipment physical parameters and patterning control parameters. The equipment physical parameters include one or more of electron beam accelerating voltage, beam spot size, exposure dose, and focusing parameters. The patterning control parameters include one or more of proximity effect correction iterations, minimum fragmentation pattern size, and development threshold. The scanning range and scan step size for each process parameter are set. For each parameter to be scanned, the starting value, ending value, and scan step size are set to form a corresponding discrete parameter sequence. Multiple parameter sequences are combined according to preset parameter combination rules to form a process parameter combination to be calculated. Graphical simulations of each process parameter combination are executed sequentially or in parallel. For each set of process parameters, the aforementioned methods for extracting layout geometry information, edge pairing, CD calculation, and EPE calculation are called to obtain the CD and EPE corresponding to that process parameter combination, and further, CDU and LER are obtained. Each process parameter combination is then screened according to a preset evaluation index range. When one or more of the corresponding CD, EPE, CDU, and LER indicators meet the preset process requirements, the process parameter combination is used as a candidate process parameter; when it does not meet the preset process requirements, it is marked as a non-target process parameter combination. The above steps are repeated until all parameter combinations within the preset process parameter range are scanned, and the CD, EPE, CDU, and LER results under different process parameter conditions are output, providing a data basis for subsequent process window analysis.

[0079] Step S7: Process window analysis;

[0080] Based on the different combinations of process parameters obtained in step S6 and their corresponding CD, EPE, CDU, and LER calculation results, the dose process window and focal length process window of electron beam direct writing lithography are analyzed. According to the electron beam exposure energy deposition model, the double Gaussian point diffusion function is used to describe the energy deposition distribution per unit dose.

[0081]

[0082] For the applied dose D, the unit dose energy deposition distribution is convolved with the pattern distribution of the layout to be exposed to obtain the spatial energy deposition distribution after exposure:

[0083]

[0084] Within a preset exposure dose range, the exposure dose is scanned according to a set step size to obtain the CD calculation results corresponding to different exposure doses. The CD corresponding to each exposure dose is compared with the CD allowable range, and the exposure dose that meets the CD tolerance requirement is determined as the effective dose point.

[0085] Furthermore, the EPE, CDU, and LER corresponding to each effective dose point are determined, and the exposure dose that simultaneously meets the preset EPE, CDU, and LER index requirements is determined as the effective process point. The effective process points constitute the dose process window, and the dose-CD curve is established based on the CD calculation results corresponding to different exposure doses.

[0086] Establish the correspondence between electron beam defocusing parameters and point spread function. Under ideal focusing conditions, the electron beam forward scattering characteristic parameters are: When defocusing is present, introduce a sparse blur parameter. And calculate the effective forward scattering characteristic parameters under defocused conditions:

[0087]

[0088] Will Substituting the data into a dual Gaussian point spread function model, point spread functions under different defocus conditions are obtained. These point spread functions are then convolved with the image to be exposed to obtain the spatial energy deposition distribution under the corresponding defocus conditions. Based on a preset development threshold, the developed pattern contours under different defocus conditions are obtained, and the corresponding CD, EPE, CDU, and LER are calculated. Defocus parameters that simultaneously meet the preset EPE, CDU, and LER requirements are determined as effective defocus points, and these effective defocus points constitute the allowable defocus range.

Claims

1. A method for scanning process parameters and analyzing process windows in electron beam lithography EDA software, characterized in that, Includes the following steps: Step S1: Extraction and calculation of geometric information of the layout graphic; Step S2: Polygon CD measurement and edge pairing; Step S3: Calculate and save the ideal CD; Step S4: Calculate the EPE of the EPE sampling points to obtain the actual CD; Step S5: Calculate the actual CD, as well as CDU and LER; Step S6: Scan process parameters; Step S7: Process window analysis.

2. The method for scanning process parameters and analyzing process windows in electron beam lithography EDA software as described in claim 1, characterized in that: In step S1, the geometric information of the layout graphic is extracted and calculated; the GDSII layout data is read, the vertex set of each polygon is obtained, and each polygon is converted into a set of edges composed of multiple edges; For each edge, extract the start point, end point, edge number, unit normal vector pointing to the interior of the polygon, horizontal or vertical attitude marker, and edge length to generate a set of discrete integer grid points corresponding to the edge, and reserve storage space for paired edge number, EPE sampling point, ideal CD and EPE value. The formula for calculating the x-component of the unit normal vector is: The formula for calculating the y-component of the unit normal vector is: In the formula, is , The x and y coordinates of the starting point of the edge, , Here are the x and y coordinates of the edge's endpoint; the edge's orientation is determined by whether the x and y coordinates of the edge's start and end points are the same: if the x coordinates are the same, it's a vertical edge; if the y coordinates are the same, it's a horizontal edge; the edge length is calculated using the following formula: EPE sampling points are selected from the center of the set of straight line points in the overlapping area of ​​paired edges.

3. The method for scanning process parameters and analyzing process windows in electron beam lithography EDA software as described in claim 1, characterized in that, Step S2 includes: polygon CD measurement edge pairing; narrowing the candidate pairing range based on edge attitude markers and normal directions: horizontal edges are matched only with horizontal candidate edges in their normal direction, and vertical edges are matched only with vertical candidate edges in their normal direction; determining the overlap between the current edge and the candidate paired edge in the corresponding projection direction, and retaining only candidate edges whose projection intervals overlap; calculating the vertical distance between the current edge and each valid candidate edge, and selecting the candidate edge with the smallest distance that meets the preset maximum allowed pairing distance threshold as the paired edge; the formula for determining the overlap of horizontal edge projection intervals is: The formula for determining the overlap of vertical side projection intervals is: In the formula , Let x and y be the minimum and maximum x-coordinates of the current edge's start and end points, respectively. , These are the minimum and maximum x-coordinates of the candidate edge's starting and ending points, respectively. , Let be the minimum and maximum y-coordinates of the current edge's starting and ending points, respectively. , These are the minimum and maximum ordinates of the candidate edge's starting and ending points, respectively.

4. The method for scanning process parameters and analyzing process windows in electron beam lithography EDA software as described in claim 1, characterized in that, Step S3 includes: calculating and saving the ideal CD; calculating the ideal critical dimension of each pair of paired edges based on the pairing edge relationship: the ideal CD of a horizontal paired edge is the absolute value of the difference between the ordinates of the two edges, and the ideal CD of a vertical paired edge is the absolute value of the difference between the abscissas of the two edges. The ideal CD and the corresponding paired edge information are then saved. The formula for calculating the ideal CD is: , These are the x and y coordinates of the starting point of the current edge. , These are the x and y coordinates of the starting points of the paired edges. It's a gesture marker.

5. The method for scanning process parameters and analyzing process windows in electron beam lithography EDA software as described in claim 1, characterized in that: In step S4, the EPE sampling point is determined based on the overlapping area of ​​the paired edges. The actual development contour is searched along the normal direction of the edge corresponding to the sampling point. The signed distance between the ideal pattern boundary and the actual development contour in the normal direction is calculated as the EPE of the sampling point.

6. The method for scanning process parameters and analyzing process windows in electron beam lithography EDA software as described in claim 1, characterized in that: In step S5, the actual CD, CDU, and LER are calculated. The actual CD at each measurement location is calculated using the formula: Actual CD = Sum of EPE values ​​at two EPE sampling points + Ideal CD. The average and standard deviation are calculated based on the actual CDs at multiple measurement locations, with the standard deviation used as the CDU. The average EPE and dispersion are calculated based on multiple EPE sampling values ​​along the edge direction, with 3 times the EPE standard deviation used as the LER. The formula for calculating CDU is: In the formula Measure the total number of CDs. Let CD be the actual CD at the i-th measurement position. The actual CD average value is the value of all measurement locations; the formula for calculating LER is: In the formula Let be the EPE value of the i-th EPE sampling point. This is the average EPE value for all EPE sampling points.

7. The method for scanning process parameters and analyzing process windows in electron beam lithography EDA software as described in claim 1, characterized in that: In step S6, process parameter scanning is performed; a set of process parameters including equipment physical parameters and graphical control parameters is established. The equipment physical parameters include one or more of electron beam accelerating voltage, beam spot size, exposure dose, and focusing parameters. The graphical control parameters include one or more of proximity effect correction iterations, minimum fragmentation pattern size, and development threshold. Multiple sets of process parameter combinations are generated according to a preset scanning range and step size. A multi-node parallel computing method is used to perform layout processing, edge pairing, CD and EPE calculations on each set of parameters to obtain the CD, EPE, CDU, and LER corresponding to each set of parameters. The parameter combinations are then associated with and stored with the corresponding evaluation results.

8. The method for scanning process parameters and analyzing process windows in electron beam lithography EDA software as described in claim 1, characterized in that: In step S7, process window analysis is performed; an exposure energy deposition model is established based on the double Gaussian point diffusion function, and the spatial energy deposition distribution is obtained by convolving the model with the pattern to be exposed; dose process window analysis is performed: the exposure dose is scanned by step size, and effective dose points are screened according to the target CD and tolerance range, combined with EPE, CDU, and LER indicators, to determine the dose process window and establish the dose-CD curve; Focal length process window analysis: Bead blur parameters are introduced to calculate the effective forward scattering characteristic parameters under defocused conditions. These parameters are then substituted into the point spread function model to obtain the energy deposition distribution and development profile under different defocused conditions. Combined with the target CD tolerance and EPE, CDU, and LER indices, the effective defocused range is screened to determine the focal length process window. The expression for the double Gaussian point spread function is: The formula for calculating the spatial energy deposition distribution after exposure is: In the formula For exposure dose, The ratio of backscattering to forward scattering coefficients. Forward scattering characteristic parameters, These are backscattering characteristic parameters. The distance from the exposure point, The distribution of the pattern to be exposed; the formula for calculating the effective forward scattering characteristic parameters under defocused conditions is: In the formula For speckle blur parameters, These are the forward scattering characteristic parameters under ideal focusing conditions.