Electron beam lithography pattern data structure and pattern exposure control method

Abstract

The invention discloses an electron beam lithography pattern data structure and a pattern exposure control method, and the data structure comprises a file header which is used for file recognition; the dose table is used for adjusting the pattern exposure dose; the index block is used for providing a field position and a graphic index; the module library is used for providing a graph multiplexing function; the graph segment is used for providing a subfield graph specific definition; and the file tail is used for identifying file ending. The control method is divided into three-stage pipeline parallel processing, the first-stage processing P1 is responsible for processing graphic data files, and meanwhile graphic data is provided for the second-stage processing P2; the second-stage processing P2 is responsible for analyzing and instructing sub-field graphic data, providing a scanning control instruction for the third-stage processing P3, and feeding back current state information to the first-stage processing P1 at the same time; and the third-stage processor P3 is responsible for scanning a control command and feeding back state information to the second-stage processor P2. The method has the advantages of complete information, high data compression rate, easiness in implementation and the like.

Description

Technical Field

[0001] This invention relates to the field of electron beam technology, specifically to an electron beam lithography pattern data structure and a pattern exposure control method. Background Technology

[0002] Electron beam lithography is a precision pattern transfer technology that uses electromagnetic fields to confine high-energy electrons to create patterns on materials. Due to its extremely high resolution and flexibility, it plays a crucial role in integrated circuit manufacturing. Integrated circuit design layouts often use standard GDSII or OASIS data formats, which frequently have very complex pattern structures and large layout sizes. However, electron beam lithography equipment can only sequentially realize a limited number of predetermined basic patterns within a finite size range (millimeter level). This necessitates the appropriate segmentation, organization, and description of the design layout (i.e., file conversion). The converted data file is usually many times larger than the original file, and additional data besides the pattern is needed to control correct machine exposure; therefore, the aforementioned design layout formats are no longer suitable. In fact, various foreign electron beam equipment manufacturers have their own machine data formats for their equipment, but these are not publicly disclosed. There are currently no mature electron beam lithography machines in China. Only the Institute of Electrical Engineering publicly disclosed an electron beam exposure data format, EDF, in the early 21st century. Its data structure is too simple (only three types of nodes: ID, TX, and LB), and the data volume is too large (each basic pattern occupies about 128 bytes). It does not have an efficient reuse mechanism other than referencing for the large number of repetitive structures in the design layout, and cannot meet the increasingly large layout exposure requirements. Summary of the Invention

[0003] To address the technical problems existing in the prior art, this invention provides an electron beam lithography pattern data structure and pattern exposure control method that is information-complete, has a high data compression rate, and is easy to implement.

[0004] To solve the above-mentioned technical problems, the technical solution proposed by this invention is as follows: An electron beam lithography pattern data structure includes: The file header is used for file identification and provides overview information; Dosimeter, used for adjusting graphic exposure dosage; Index blocks are used to provide field locations and graphical indexes; A module library for providing graphics reuse functionality; The graphics section is used to provide specific definitions for the subfield graphics; The end of a file is used to indicate the end of the file.

[0005] Preferably, the file header consists of a file header identifier, basic file information, graphics setting parameters, graphics overview information, and a data block address table; The file header identifier includes a specific ID, a data structure version number, and vendor identification information; Basic file information includes the timestamp of data file creation, the total number of data files, and the sequence number of the current file; Graphical settings parameters include control flags, size units, layout size, main field size, subfield size, main field resolution, subfield resolution, and accelerating voltage; The graphical overview information includes the number of dose levels, the number of modules in the module library, the number of main fields and subfields contained in the layout, the number of subfield data blocks in the graphical segment, and the number of various basic graphics; The data block address table contains, in order, the starting address of the dose table, the starting address of the index block, the starting address of the module library, the starting address of the graphics segment, and the starting address of the file end; where the starting addresses are all offsets relative to the file header.

[0006] Preferably, the dosing table consists of a dosing table identifier, a number of dose levels, a dose accuracy, and a dose data list; the dose data list sequentially arranges multiple dose data, and uses their sequence number in the list as an index for reference by dose adjustment instructions.

[0007] Preferably, the index block consists of an index block identifier, a number of home games, a home game data block address table, and a home game data list; The home field data block address table provides the starting address of each home field data block in sequence, and the home field order in the address table is the exposure order of the home fields. The home field data list contains multiple home field data blocks in sequence. Each home field data block consists of a home field data block identifier, the number of subfields within the home field, the home field location, and a subfield index list. The home field location refers to the position of the home field center in the map coordinate system. The subfield index list contains multiple subfield index data. Each subfield index data consists of the subfield location, the starting address of the subfield data block in the graphic segment, and the data length. The subfield location indicates the position of the subfield area in the home field coordinate system, which is represented by the vertex or center of the subfield.

[0008] Preferably, the module library consists of a module library identifier, a number of modules, a module data block address table, and a module data list; The module data block address table provides the starting address of each module data block in sequence, and uses its sequence number in the list as an index for library reference instructions to call module data. The module data list contains multiple module data blocks in sequence. Each module data block is a single basic graphics data defined by basic graphics instructions, or a structure data block containing multiple basic graphics defined by structure instructions, or an array data block containing multiple basic graphics and / or multiple structure data defined by array instructions.

[0009] Preferably, the graphic segment consists of a graphic segment identifier, a number of subfield data blocks, and a list of subfield data blocks; The subfield data block list contains multiple subfield data blocks in sequence. Each subfield data block consists of a subfield data block identifier, the number of graphic data, and a list of graphic data. The list of graphic data contains multiple graphic data in sequence. The graphic data can be basic graphic data defined by basic graphic instructions, library reference data defined by library reference instructions, or dose adjustment data defined by dose adjustment instructions. The number of graphic data refers to the sum of the number of basic graphic instructions, library reference instructions, and dose adjustment instructions. That is, both library reference data and dose adjustment data are counted as graphic data.

[0010] Preferably, the file end consists of a file end identifier and file integrity verification data; the file integrity verification data consists of a verification method, verification data length, and verification result data; the verification result data is the result calculated from all data from the file header identifier to the verification data length.

[0011] The present invention also discloses a pattern exposure control method for electron beam lithography based on the electron beam lithography pattern data structure described above. It is divided into three-stage pipelined parallel processing. The first-stage processing P1 is responsible for reading, verifying, parsing, decompressing and reconstructing the pattern data file, controlling the main field positioning, and providing complete and flattened pattern data to the second-stage processing P2 in units of subfields. The second-level processor P2 is responsible for parsing and instructing the subfield graphic data, compiling the graphic data into a limited number of machine instructions and data, providing continuous scan control instruction data for the third-level processor P3, and feeding back the current status information to the first-level processor P1. The third-level processor P3 is responsible for implementing the scanning control commands. Based on the commands and data, it controls the main field deflector, sub-field deflector, brake, and data update clock, and feeds back the status information to the second-level processor P2.

[0012] Preferably, the first-level processing P1 includes the following steps: S1: Read and verify file data; Read the file, locate the end of the file according to the end-of-file address in the file header data block address table, perform verification calculations on the file data according to the integrity verification method in the end-of-file address, and compare the verification results; If it does not meet the requirements, prompt and exit; if it does meet the requirements, proceed to the next step. S2: Check the graphic file information; check and print the output file header information, including data format version number, main field size, subfield size, main field resolution, subfield resolution, and accelerating voltage; if it does not match the current device settings, prompt and exit; if it matches, proceed to the next step; S3: Initialize and start the main loop; read the addresses of each data block in the data block address table into the corresponding data block pointers, and locate the corresponding data blocks; set the number of main loops according to the index block data; locate the first main data block according to the main data block address table; S4: Position the main field and start the sub-field loop; set the number of sub-field loops based on the main field data block data, and position the first sub-field index data; control the stage to position itself to the current main field based on the main field position data, and after the stage position is stable, send the start exposure command to the second-level processing P2; S5: Subfield graphic data decompression and reconstruction; create a new subfield graphic data block (SFB), write the subfield position of the current subfield index data into the SFB, and locate the subfield data through the graphic segment pointer according to the subfield data start address in the index data, read the data volume of the subfield data length, flatten it, and write it into the SFB; S6: Wait or write to the buffer; if the first buffer is full, wait; otherwise, write the subfield graphic data block SFB to the buffer, locate the next subfield index data, and return to S5 for loop execution until the set number of loops is reached, then proceed to the next step. S7: Wait for the current primary field exposure to complete; wait for the second-level processing P2 to be idle, then send it a stop exposure command; locate the next primary field data block, and return to S4 to execute the loop until the set number of loops is reached; S8: Determine if the current file number is equal to the total number of files; if yes, end the image exposure; otherwise, open the next file and jump to S1 to execute the loop.

[0013] Preferably, in S5, the flattening process targets dose adjustment data and library reference data. If it is dose adjustment data, the actual dose data is obtained through the dose index number and dose table pointer of the data, and written into the graphic data block SFB along with the dose adjustment instruction. If it is library reference data, the module data block is located through the module reference number and module pointer of the data, the library reference data is replaced with the corresponding module data, and written into the graphic data block SFB. Specifically, if the library references basic graphic data, the graphic data is copied. If the library references structural data, the graphic data within the structure is read sequentially, and the structural data or array data is further expanded until only basic graphic data remains. If the library references array data, the array unit graphics are first flattened, and then a set number of basic graphic data blocks are generated according to the array definition, and the generated graphic data blocks are sorted according to predetermined rules.

[0014] Compared with the prior art, the advantages of the present invention are as follows: The graphic data structure and graphic exposure control method for electron beam lithography equipment provided by this invention use index blocks as the core and subfields as the basic units for storing, organizing, and controlling graphic data. Its advantages are as follows: 1) Complete Information. The graphic data structure of this invention contains a complete hierarchical structure, providing a four-level hierarchical structure and positional relationships of layout, main field, subfield, and graphic; it includes basic graphic definitions and descriptions, providing limited basic graphic definition and description methods; it includes graphic dosage, providing adjustment information for graphic exposure dosage; it includes exposure sequence and path, providing main field / subfield and graphic exposure sequence information to reduce positioning and movement; it includes condition settings for graphic data generation, providing graphic setting information such as field size, resolution, and coordinate system; it also includes control information for graphic implementation, such as step size and fill method, which can provide complete graphic and control data for electron beam lithography equipment.

[0015] 2) High data compression ratio. Addressing the widespread repetitive structures in integrated circuit layouts, this invention provides array control, library referencing, and subfield referencing functions to compress the data volume of graphics files. Index data from different main fields, or different subfields within the same main field, can reference the same subfield data block to reuse the subfield graphics data. The index block design allows subfield graphics data blocks to be reorganized into new main fields without adding extra graphics data, facilitating multiple exposures. Basic graphics can be defined with attached array structure definitions to reduce data volume. Repetitive basic graphics, graphics structure blocks, or graphics array blocks can be defined as library modules for referencing. When referencing library modules, array structure definitions can be attached, or operations such as rotation and scaling can be performed, significantly reducing the graphics data volume. Furthermore, for different hardware (such as systems with a sub-deflection DAC bit width equal to or less than 16 bits), a short address mode can be used to further reduce the graphics data volume.

[0016] 3) Easy to implement. The graphic data structure and graphic exposure control method provided by this invention can be implemented using a three-stage pipeline and a two-stage buffer parallel processing, with each stage of processing relying on different hardware platforms. By parsing and decompressing the graphic data in the first stage of processing and reconstructing it into flattened graphic data in units of subfields, the amount of data and complexity of subsequent processing are significantly reduced. This makes subsequent graphic processing easier to implement on embedded platforms with limited resources but high real-time requirements, thereby reducing the overall technical difficulty of graphic exposure control. Attached Figure Description

[0017] Figure 1 This is a structural diagram of the electron beam lithography pattern data structure of the present invention in an embodiment.

[0018] Figure 2 This is a flowchart of an embodiment of the graphic exposure control method of the present invention.

[0019] Figure 3 This is a flowchart of the first-level processing P1 of the present invention in an embodiment. Detailed Implementation

[0020] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0021] like Figure 1 As shown, the electron beam lithography pattern data structure provided in this embodiment of the invention includes, in order: The file header is used for file identification and to provide overview information, including basic file information, graphical settings parameters, graphical overview information, and internal data block address table. The dose table is used for adjusting the exposure dose in graphics and contains all the dose data used in the graphics. The index block provides field location and graph index, including the main field location, the location of subfields within the main field, and the starting address of the graph data; The module library provides graphics reuse functionality and includes referenceable basic graphics, graphics structure blocks, array structure blocks, etc. The graphics section provides specific definitions for subfield graphics, including basic graphics data, dose adjustment data, and library reference data. The end of the file is used to indicate the end of the file and contains file integrity verification data.

[0022] The graphics data structure is stored as a binary data file, and can optionally be stored as one or more files. When stored as multiple files, the files are named in ascending order of positive integers. Each file contains the complete structure, and all necessary information is defined within the file itself, without referencing data from other files. The graphics data file uses a 32-bit basic addressing unit and is aligned to four bytes; all addresses within the data structure are in four-byte units.

[0023] Specifically, the file header consists of a file header identifier, basic file information, graphic setting parameters, graphic overview information, and a data block address table, with fixed data order and length. The file header identifier includes, but is not limited to: a specific ID identifier, data structure version number, and vendor identifier; the basic file information includes, but is not limited to: the timestamp of data file creation, the total number of data files, and the current file's sequence number; the graphic setting parameters include, but are not limited to: control flags, size units, layout size, main field size, subfield size, main field resolution, subfield resolution, and accelerating voltage; the graphic overview information includes, but is not limited to: the number of dose levels, the number of modules in the module library, the number of main fields and subfields included in the layout, the number of subfield data blocks in the graphic segment, and the number of various basic graphics; the data block address table sequentially includes the starting address of the dose table, the starting address of the index block, the starting address of the module library, the starting address of the graphic segment, and the starting address of the file tail; all starting addresses are offsets relative to the file header (address 0). The file header structure of this embodiment is shown in Table 1: Table 1

[0024] Specifically, the dose table consists of a dose table identifier (a specific ID), a number of dose levels, a dose precision, and a dose data list. The data order is fixed, but the length is variable. The dose data list sequentially arranges multiple dose data points, indexed by their order number in the list (e.g., the index of the first dose data point is 1), for use by dose adjustment instructions. The dose table structure of this embodiment is shown in Table 2: Table 2

[0025] Specifically, an index block consists of an index block identifier (a specific ID identifier), the number of home venues, a home venue data block address table, and a home venue data list. The data order is fixed, but the length is variable.

[0026] The home data block address table provides the starting address (offset relative to the starting address of the index block) of each home data block in sequence. The home order in the address table is the exposure order of the home data blocks. The home field data list sequentially contains multiple home field data blocks. Each home field data block consists of a home field data block identifier (a specific ID), the number of sub-fields within the home field, the home field location, and a sub-field index list. The home field location refers to the position of the home field center in the map coordinate system. The sub-field index list contains multiple sub-field index data. Each sub-field index data consists of the sub-field location, the starting address of the sub-field data block in the graphic segment (offset relative to the starting address of the graphic segment), and the data length. The sub-field location indicates the position of the sub-field region in the home field coordinate system, which can be represented by the vertex or center of the sub-field. The index header structure of this embodiment is shown in Table 3. Table 3

[0027] Specifically, the module library consists of a module library identifier (a specific ID identifier), the number of modules, a module data block address table, and a module data list, with the data order being fixed and the length being variable.

[0028] The module data block address table provides the starting address (offset relative to the starting address of the module library) of each module data block in sequence, and uses its sequential number in the list as an index (e.g., the index of the first module data block is 1), for library reference instructions to call module data; the module data list contains multiple module data blocks in sequence. Each module data block can be a single basic graphics data defined by basic graphics instructions, or a structure data block containing multiple basic graphics defined by structure instructions, or an array data block containing multiple basic graphics and / or multiple structure data defined by array instructions. The module library structure of this embodiment is shown in Table 4: Table 4

[0029] Specifically, a graphic segment consists of a graphic segment identifier (a specific ID), the number of subfield data blocks, and a list of subfield data blocks, with a fixed data order and variable length. The subfield data block list sequentially contains multiple subfield data blocks, each consisting of a subfield data block identifier (a specific ID), the number of graphic data blocks, and a list of graphic data blocks. The graphic data list sequentially contains multiple graphic data blocks, which can be basic graphic data defined by basic graphic instructions, library reference data defined by library reference instructions, or dose adjustment data defined by dose adjustment instructions. The number of graphic data blocks refers to the sum of the number of basic graphic instructions, library reference instructions, and dose adjustment instructions; that is, both library reference data and dose adjustment data are counted as graphic data. The graphic segment structure of this embodiment is shown in Table 5. Table 5

[0030] Specifically, the file tail consists of a file tail identifier (a specific ID) and file integrity verification data. The file integrity verification data consists of a verification method (such as MD5, SHA256, etc.), the verification data length, and the verification result data; the verification result data is the result calculated using a specified verification algorithm on all data from the file head identifier to the verification data length. The file tail structure of this embodiment is shown in Table 6: Table 6

[0031] Specifically, the basic graphic data in the module library begins with a graphic instruction (a specific ID identifier) ​​and includes a graphic code, control flags, graphic position, and graphic attribute data.

[0032] The graphic code is used to distinguish different basic graphics. Defined basic graphics include, but are not limited to, rectangles, arbitrary trapezoids, 45-degree trapezoids, polylines, and circles. Control flags include, but are not limited to, enabling repeatability, enabling X / Y axis mirroring, and enabling long address mode. Graphic position refers to the graphic's location in the subfield coordinate system; depending on the graphic code, it can use vertex, center point, etc. Position data uses a short address mode (e.g., 16 bits) by default to reduce data size; a long address mode (e.g., 24 bits or 32 bits) can be used depending on the control flag settings. Graphic attribute data describes the specific outline of the graphic and can be used flexibly, with different numbers of data allowed depending on the graphic code.

[0033] The basic graphic data of this invention embodiment are shown in Table 7: Table 7

[0034] The structure data block begins with a structure instruction (a specific ID identifier), and its data structure is consistent with that of the subfield data block. It can reference other defined structure data and can also adjust the dosage.

[0035] The array data block begins with an array instruction (a specific ID identifier) ​​and includes the number of cell graphic data, layout control data, and a list of graphic data. The number of cell graphic data refers to the number of graphic data in each repeating cell; the layout control data includes the repeating dimension (one-dimensional or two-dimensional), the number of repetitions in the first repeating direction, and the position offset vector. Δ X1, Δ Y1), and the optional number of repetitions and position offset vector in the second repetition direction ( Δ X2, Δ Y2); the graphic data list has the same structure as the graphic data list in the subfield data block.

[0036] The array data block structure of this invention embodiment is shown in Table 8: Table 8

[0037] Library reference data begins with a library reference instruction (a specific ID identifier) ​​and includes reference control data and a library module index number. Reference control data includes, but is not limited to, flags for enabling repeat, enabling X / Y axis mirroring, enabling scaling / rotation, enabling long address mode, and optional scaling factors and rotation angles; the library module index number refers to the sequence number of the module data block's starting address in the module data block address table.

[0038] Dosage adjustment data begins with a dose adjustment instruction (a specific ID flag) and includes control data and a dose index number. Control data includes, but is not limited to, the scope of the dose data (which can be global or local); the dose index number refers to the sequence number of the dose data in the dose data table.

[0039] like Figure 2 As shown, this embodiment of the invention also provides a pattern exposure control method for electron beam lithography, which is divided into three-stage pipelined parallel processing. The first-stage processing P1 is responsible for reading, verifying, parsing, decompressing and reconstructing the pattern data file, controlling the main field positioning, and providing complete and flattened pattern data to the second-stage processing P2 in units of subfields. The second-level processor P2 is responsible for parsing and instructing the subfield graphic data, compiling the graphic data into a limited number of machine instructions and data, providing continuous scan control instruction data for the third-level processor P3, and feeding back the current status information to the first-level processor P1. The third-level processor P3 is responsible for implementing the scanning control commands. Based on the commands and data, it controls the main field deflector, sub-field deflector, brake, and data update clock, and feeds back the status information to the second-level processor P2.

[0040] The above control method is divided into three-level pipelined processing. The first-level processing is responsible for the verification, parsing, data reconstruction, main field positioning and logic control of the graphic data file, and outputs complete and flat graphic data in units of subfields. The second-level processing is responsible for the parsing and instruction conversion of the subfield graphic data, and outputs continuous scan control instruction data. The third-level processing is responsible for the implementation and conversion of scan control instruction data, and outputs control-related hardware.

[0041] In the first-level processing, the index block data serves as the core, employing a double-layer nested loop control between the main field and sub-fields. Sub-field graphic data is read based on the sub-field index data and flattened to reconstruct complete, reference-free sub-field graphic data blocks. Within this first-level processing, subsequent graphic files can be identified and opened based on the number and file number information for continuous and complete graphic exposure.

[0042] A first data buffer B1 with first-in-first-out (FIFO) is set between the first-level processing and the second-level processing, and is managed by the first-level processing P1; a second data buffer B2 with FIFO is set between the second-level processing and the third-level processing, and is managed by the second-level processing P2.

[0043] like Figure 3 As shown, the first-level processing P1 processes the graphic data file according to the following procedure: S1: Read and verify file data. Read the file, locate the end of the file using the end-of-file address in the header data block address table, and use the corresponding verification algorithm to perform verification calculations on the file data according to the integrity verification method in the end-of-file table. Compare the verification results. If a discrepancy is found, prompt and exit; otherwise, proceed to the next step.

[0044] S2: Check graphic file information. Check and print the output file header information, including but not limited to data format version number, main field size, subfield size, main field resolution, subfield resolution, accelerating voltage, etc. If it does not match the current device settings, prompt and exit; if it matches, proceed to the next step.

[0045] S3: Initialize and start the main loop. Read the addresses of each data block in the data block address table into the corresponding data block pointers, and locate the corresponding data block (dose table, index block, module library, graphics segment). Set the number of main loops based on the index block data; locate the first main data block based on the main data block address table.

[0046] S4: Position the main field and start the subfield loop. Based on the main field data block data, set the number of subfield loops and position the first subfield index data; based on the main field position data, control the stage to position to the current main field, and after the stage position is stable, send the start exposure command to the second-level processing P2.

[0047] S5: Subfield graphic data decompression and reconstruction. Create a new subfield graphic data block (SFB), write the subfield position and subfield number of the current subfield index data into the SFB, locate the subfield data through the graphic segment pointer according to the subfield data start address in the index data, read the data volume of the subfield data length, flatten it, and write it into the SFB.

[0048] S6: Wait or write to the buffer. If the first buffer is full, wait; otherwise, write the subfield graphic data block SFB to the buffer, locate the next subfield index data, and return to S5 for loop execution until the set number of loops is reached, then proceed to the next step.

[0049] S7: Wait for the current master exposure to complete. After the second-level processor P2 is idle, send it a stop exposure command. Locate the next master data block and return to S4 to execute the loop until the set number of loops is reached.

[0050] S8: Determine if the current file number is equal to the total number of files. If yes, the image exposure ends; otherwise, open the next file and jump to S1 to execute the loop.

[0051] In S5, flattening is performed on dose adjustment data and library reference data. If it's dose adjustment data, the actual dose data is obtained using the dose index number and dose table pointer, and written to the graphic data block SFB along with the dose adjustment command. If it's library reference data, the module data block is located using the module reference number and module pointer, the library reference data is replaced with the corresponding module data, and written to the graphic data block SFB. Specifically, if the library references basic graphic data, the graphic data is copied; if the library references structural data, the graphic data within the structure is read sequentially, and the structural data or array data is further expanded until only basic graphic data remains; if the library references array data, the array unit graphics are first flattened, then a set number of basic graphic data blocks are generated according to the array definition, and the generated graphic data blocks are sorted according to predetermined rules.

[0052] The second-level processor P2 retrieves the subfield graphic data block (SFB) from the first data buffer B1, parses the data, and generates control commands, including but not limited to: parsing subfield position data and generating main field positioning commands; parsing basic graphic position data and generating subfield positioning commands; parsing dose adjustment data and generating dose control commands; parsing graphic codes and graphic attribute data and generating multiple consecutive line scan commands according to the set scan direction; generating brake switch commands before / after graphic line scan commands; and generating delay commands after main / subfield positioning commands and brake switch commands according to set parameters. P2 stores the control commands and data in the second data buffer B2 in the order they are generated and feeds back status information to P1, including but not limited to: status information fed back by P3, the command length in B2 buffer, the subfield number currently being processed, the total number of graphics in the current subfield, and the number of graphics already processed. P2 is preferably implemented in a digital signal processor (DSP) or an ARM processor containing a DSP.

[0053] The third-level processor, P3, retrieves instructions and data from the second data buffer, B2, and outputs corresponding hardware actions. The main field positioning instruction outputs the main field coordinates to control the deflection of the main deflector; the subfield positioning instruction outputs the subfield coordinates to control the deflection of the sub-deflector; the clamping gate switch instruction outputs the clamping gate opening or closing; the graphic line scan instruction controls the subfield coordinates in the corresponding direction to change from the initial value to the end value according to the set step size. The data update cycle of the subfield coordinates is adjusted by the dose control instruction; the delay instruction does not take any action within the set cycle. P3 feeds back status information to P2, including but not limited to the main deflection DAC value, sub-deflection DAC value, clamping gate status, and the current subfield data update cycle. P3 is preferably implemented in a field-programmable gate array (FPGA).

[0054] Compared to existing technologies, the graphic data structure and graphic exposure control method for electron beam lithography equipment provided by this invention, which uses index blocks as the core and subfields as the basic units for storing, organizing, and controlling graphic data, has the following advantages: 1) Complete Information. The graphic data structure of this invention contains a complete hierarchical structure, providing a four-level hierarchical structure and positional relationships of layout, main field, subfield, and graphic; it includes basic graphic definitions and descriptions, providing limited basic graphic definition and description methods; it includes graphic dosage, providing adjustment information for graphic exposure dosage; it includes exposure sequence and path, providing main field / subfield and graphic exposure sequence information to reduce positioning and movement; it includes condition settings for graphic data generation, providing graphic setting information such as field size, resolution, and coordinate system; it also includes control information for graphic implementation, such as step size and fill method, which can provide complete graphic and control data for electron beam lithography equipment.

[0055] 2) High data compression ratio. Addressing the widespread repetitive structures in integrated circuit layouts, this invention provides array control, library referencing, and subfield referencing functions to compress the data volume of graphics files. Index data from different main fields, or different subfields within the same main field, can reference the same subfield data block to reuse the subfield graphics data. The index block design allows subfield graphics data blocks to be reorganized into new main fields without adding extra graphics data, facilitating multiple exposures. Basic graphics can be defined with attached array structure definitions to reduce data volume. Repetitive basic graphics, graphics structure blocks, or graphics array blocks can be defined as library modules for referencing. When referencing library modules, array structure definitions can be attached, or operations such as rotation and scaling can be performed, significantly reducing the graphics data volume. Furthermore, for different hardware (such as systems with a sub-deflection DAC bit width equal to or less than 16 bits), a short address mode can be used to further reduce the graphics data volume.

[0056] 3) Easy to implement. The graphic data structure and graphic exposure control method provided by this invention can be implemented using a three-stage pipeline and a two-stage buffer parallel processing, with each stage of processing relying on different hardware platforms. By parsing and decompressing the graphic data in the first stage of processing and reconstructing it into flattened graphic data in units of subfields, the amount of data and complexity of subsequent processing are significantly reduced. This makes subsequent graphic processing easier to implement on embedded platforms with limited resources but high real-time requirements, thereby reducing the overall technical difficulty of graphic exposure control.

[0057] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should be considered within the scope of protection of the present invention.

Claims

1. An electron beam lithography pattern data structure, characterized in that, include: The file header is used for file identification and provides overview information; Dosimeter, used for adjusting graphic exposure dosage; Index blocks are used to provide field locations and graphical indexes; A module library for providing graphics reuse functionality; The graphics section is used to provide specific definitions for the subfield graphics; The end of a file is used to indicate the end of the file.

2. The electron beam lithography pattern data structure according to claim 1, characterized in that, The file header consists of a file header identifier, basic file information, graphics setting parameters, graphics overview information, and a data block address table; The file header identifier includes a specific ID, a data structure version number, and vendor identification information; Basic file information includes the timestamp of data file creation, the total number of data files, and the sequence number of the current file; Graphical settings parameters include control flags, size units, layout size, main field size, subfield size, main field resolution, subfield resolution, and accelerating voltage; The graphical overview information includes the number of dose levels, the number of modules in the module library, the number of main fields and subfields contained in the layout, the number of subfield data blocks in the graphical segment, and the number of various basic graphics; The data block address table contains, in order, the starting address of the dose table, the starting address of the index block, the starting address of the module library, the starting address of the graphics segment, and the starting address of the file end; where the starting addresses are all offsets relative to the file header.

3. The electron beam lithography pattern data structure according to claim 1, characterized in that, The dosing table consists of a dosing table identifier, a number of dose levels, a dose accuracy, and a dose data list. The dose data list sequentially arranges multiple dose data points, with their order number in the list serving as an index for dose adjustment instructions.

4. The electron beam lithography pattern data structure according to claim 1, 2, or 3, characterized in that, The index block consists of an index block identifier, a number of home games, a home game data block address table, and a home game data list; The home field data block address table provides the starting address of each home field data block in sequence, and the home field order in the address table is the exposure order of the home fields. The home field data list contains multiple home field data blocks in sequence. Each home field data block consists of a home field data block identifier, the number of subfields within the home field, the home field location, and a subfield index list. The home field location refers to the position of the home field center in the map coordinate system. The subfield index list contains multiple subfield index data. Each subfield index data consists of the subfield location, the starting address of the subfield data block in the graphic segment, and the data length. The subfield location indicates the position of the subfield area in the home field coordinate system, which is represented by the vertex or center of the subfield.

5. The electron beam lithography pattern data structure according to claim 1, 2, or 3, characterized in that, The module library consists of a module library identifier, a number of modules, a module data block address table, and a module data list; The module data block address table provides the starting address of each module data block in sequence, and uses its sequence number in the list as an index for library reference instructions to call module data. The module data list contains multiple module data blocks in sequence. Each module data block is a single basic graphics data defined by basic graphics instructions, or a structure data block containing multiple basic graphics defined by structure instructions, or an array data block containing multiple basic graphics and / or multiple structure data defined by array instructions.

6. The electron beam lithography pattern data structure according to claim 1, 2, or 3, characterized in that, The graphic segment consists of a graphic segment identifier, the number of subfield data blocks, and a list of subfield data blocks; The subfield data block list contains multiple subfield data blocks in sequence. Each subfield data block consists of a subfield data block identifier, the number of graphic data, and a list of graphic data. The list of graphic data contains multiple graphic data in sequence. The graphic data can be basic graphic data defined by basic graphic instructions, library reference data defined by library reference instructions, or dose adjustment data defined by dose adjustment instructions. The number of graphic data refers to the sum of the number of basic graphic instructions, library reference instructions, and dose adjustment instructions. That is, both library reference data and dose adjustment data are counted as graphic data.

7. The electron beam lithography pattern data structure according to claim 1, 2, or 3, characterized in that, The file tail consists of a file tail identifier and file integrity verification data; the file integrity verification data consists of a verification method, verification data length, and verification result data; the verification result data is the result calculated from all data from the file head identifier to the verification data length.

8. A pattern exposure control method for electron beam lithography based on the electron beam lithography pattern data structure according to any one of claims 1-7, characterized in that, It is divided into three levels of pipelined parallel processing. The first level of processing, P1, is responsible for reading, verifying, parsing, decompressing and reconstructing the graphics data file, controlling the main field positioning, and providing complete and flattened graphics data to the second level of processing, P2, in units of sub-fields. The second-level processor P2 is responsible for parsing and instructing the subfield graphic data, compiling the graphic data into a limited number of machine instructions and data, providing continuous scan control instruction data for the third-level processor P3, and feeding back the current status information to the first-level processor P1. The third-level processor P3 is responsible for implementing the scanning control commands. Based on the commands and data, it controls the main field deflector, sub-field deflector, brake, and data update clock, and feeds back the status information to the second-level processor P2.

9. The pattern exposure control method for electron beam lithography according to claim 8, characterized in that, The first-level processing P1 includes the following steps: S1: Read and verify file data; Read the file, locate the end of the file based on the end-of-file address in the file header data block address table, perform integrity verification calculations on the file data according to the integrity verification method in the end-of-file address, and compare the verification results; if they do not match, prompt and exit; otherwise, proceed to the next step. S2: Check the graphic file information; check and print the output file header information, including data format version number, main field size, subfield size, main field resolution, subfield resolution, and accelerating voltage; if it does not match the current device settings, prompt and exit; if it matches, proceed to the next step; S3: Initialize and start the main loop; read the addresses of each data block in the data block address table into the corresponding data block pointers, locate the corresponding data blocks; set the number of main loop iterations based on the index block data; Based on the home data block address table, locate the first home data block; S4: Position the main field and start the sub-field loop; set the number of sub-field loops based on the main field data block data, and position the first sub-field index data; control the stage to position itself to the current main field based on the main field position data, and after the stage position is stable, send the start exposure command to the second-level processing P2; S5: Subfield graphic data decompression and reconstruction; create a new subfield graphic data block (SFB), write the subfield position of the current subfield index data into the SFB, and locate the subfield data through the graphic segment pointer according to the subfield data start address in the index data, read the data volume of the subfield data length, flatten it, and write it into the SFB; S6: Wait or write to the buffer; if the first buffer is full, wait; otherwise, write the subfield graphic data block SFB to the buffer, locate the next subfield index data, and return to S5 for loop execution until the set number of loops is reached, then proceed to the next step. S7: Wait for the current primary field exposure to complete; wait for the second-level processing P2 to be idle, then send it a stop exposure command; locate the next primary field data block, and return to S4 to execute the loop until the set number of loops is reached; S8: Determine if the current file number is equal to the total number of files; if yes, end the image exposure; otherwise, open the next file and jump to S1 to execute the loop.

10. The pattern exposure control method for electron beam lithography according to claim 9, characterized in that, In S5, flattening is performed on dose adjustment data and library reference data. If it is dose adjustment data, the actual dose data is obtained through the dose index number and dose table pointer of the data, and written into the graphic data block SFB along with the dose adjustment instruction. If it is library reference data, the module data block is located through the module reference number and module pointer of the data, the library reference data is replaced with the corresponding module data, and written into the graphic data block SFB. Specifically, if the library references basic graphic data, the graphic data is copied. If the library references structural data, the graphic data within the structure is read sequentially, and the structural data or array data is further expanded until only basic graphic data remains. If the library references array data, the array unit graphics are flattened first, and then a set number of basic graphic data blocks are generated according to the array definition, and the generated graphic data blocks are sorted according to predetermined rules.

Images