WO2023137622A1 - 用于确定晶圆图形尺寸的方法、装置、设备、介质以及程序产品 - Google Patents

用于确定晶圆图形尺寸的方法、装置、设备、介质以及程序产品 Download PDF

Info

Publication number
WO2023137622A1
WO2023137622A1 PCT/CN2022/072733 CN2022072733W WO2023137622A1 WO 2023137622 A1 WO2023137622 A1 WO 2023137622A1 CN 2022072733 W CN2022072733 W CN 2022072733W WO 2023137622 A1 WO2023137622 A1 WO 2023137622A1
Authority
WO
WIPO (PCT)
Prior art keywords
pattern
test
intensity distribution
light intensity
data
Prior art date
Application number
PCT/CN2022/072733
Other languages
English (en)
French (fr)
Inventor
程垄
任谦
孙永倩
明勇
陈飞鸿
杨敏
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to CN202280006302.2A priority Critical patent/CN116802556A/zh
Priority to PCT/CN2022/072733 priority patent/WO2023137622A1/zh
Publication of WO2023137622A1 publication Critical patent/WO2023137622A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G03PHOTOGRAPHY; CINEMATOGRAPHY; ANALOGOUS TECHNIQUES USING WAVES OTHER THAN OPTICAL WAVES; ELECTROGRAPHY; HOLOGRAPHY
    • G03FPHOTOMECHANICAL PRODUCTION OF TEXTURED OR PATTERNED SURFACES, e.g. FOR PRINTING, FOR PROCESSING OF SEMICONDUCTOR DEVICES; MATERIALS THEREFOR; ORIGINALS THEREFOR; APPARATUS SPECIALLY ADAPTED THEREFOR
    • G03F1/00Originals for photomechanical production of textured or patterned surfaces, e.g., masks, photo-masks, reticles; Mask blanks or pellicles therefor; Containers specially adapted therefor; Preparation thereof
    • G03F1/36Masks having proximity correction features; Preparation thereof, e.g. optical proximity correction [OPC] design processes

Definitions

  • Embodiments of the present disclosure generally relate to the field of chip manufacturing, and more specifically relate to methods, devices, equipment, media and program products for determining wafer pattern dimensions.
  • photolithography refers to the process of transferring the pattern on the mask to the wafer by means of optical projection exposure, which is a key step in integrated circuit manufacturing.
  • Pattern fidelity is a fundamental requirement for lithography.
  • OPE optical Proximity Effect
  • the optical proximity effect means that the graphics on the mask change with the distance, size and density of the surrounding graphics, such as size shift, corner passivation, line end shrinkage, and even graphics disappear. This will lead to a large difference between the pattern on the wafer and the expected pattern, which will cause deviations in the electrical characteristics of the integrated circuit and affect the function and yield of the chip finally obtained.
  • Optical Proximity Correction is a pre-compensation technique for masks. By compensating and correcting the pattern on the mask, the pattern obtained on the wafer after exposure and development is consistent with the desired pattern.
  • OPC Optical Proximity Correction
  • the design layout becomes more and more complex, the requirements for the prediction accuracy of the OPC model are also getting higher and higher.
  • the embodiments of the present disclosure aim to provide a solution for determining the size of wafer patterns.
  • a method for determining wafer pattern dimensions includes: at least based on a test light intensity distribution data set for at least one test pattern and a wafer pattern data set, generating a development model for determining a wafer pattern size, each test light intensity distribution data in the test light intensity distribution data set indicates a light intensity distribution obtained by a light source of a lithography apparatus on a wafer through a corresponding test pattern, and each wafer pattern data in the wafer pattern data set indicates a size of a wafer pattern corresponding to the corresponding test pattern;
  • the size of the pattern, the target light intensity distribution data indicates the light intensity distribution obtained by the light source on the wafer through the target pattern.
  • the method according to the present disclosure uses the trained model structure to fit the development process that is difficult to accurately describe with formulas, so the accuracy of the development model describing the development process can be improved.
  • the light intensity distribution is used as a training sample in the process of training the model, the physical properties of the model can be guaranteed, so that the trained model is not prone to overfitting, thereby improving the prediction accuracy of the OPC model.
  • each test pattern respectively has a measurement area
  • each test light intensity distribution data in the test light intensity distribution data set indicates the light intensity distribution at the corresponding measurement area of the wafer
  • each wafer pattern data in the wafer pattern data set indicates the size of the wafer pattern at the corresponding measurement area
  • generating the development model includes: Based on the wafer pattern data set and the test pattern size data set for at least one test pattern, determine a size difference data set for at least one test pattern, each test pattern size data in the test pattern size data set indicates a corresponding test pattern
  • Each size difference data in the size difference data set indicates the size difference between the corresponding test pattern and the corresponding wafer pattern at the corresponding measurement region; and using the test light intensity distribution data set and the size difference data set to train the model structure to obtain a development model.
  • the measurement area is a one-dimensional measurement line at the measurement location, and each test light intensity distribution data in the test light intensity distribution data set indicates the one-dimensional light intensity distribution at the corresponding measurement line of the wafer.
  • each test light intensity distribution data in the test light intensity distribution data set indicates the one-dimensional light intensity distribution at the corresponding measurement line of the wafer.
  • the method further includes: based on the measurement position, using light source data and pupil data for the lithography apparatus and test mask data for the test mask to obtain a test light intensity distribution data set, the light source data indicates the distribution of the energy and phase of the light emitted by the light source with respect to the spatial position, the pupil data indicates whether light can be received at each spatial position, the test mask includes at least one test pattern, and the test mask data indicates the distribution of the light transmittance of the test mask with respect to the spatial position;
  • the target mask data is used to obtain the target light intensity distribution data, the target mask includes a target pattern, and the target mask data indicates the distribution of the light transmittance of the target mask with respect to the spatial position.
  • Obtaining the light intensity distribution in this way can ensure the physicality of the obtained test light intensity distribution data set and target light intensity distribution data, thereby ensuring the physicality of the OPC model, making the subsequent training of the development model less prone to overfitting, thereby improving the prediction accuracy of the OPC model.
  • the measurement line includes a first line segment and a second line segment, the first line segment and the second line segment are spaced apart from each other and are respectively located at the edge of the test pattern at the measurement position, at least one test pattern includes the first test pattern, the test light intensity distribution data set includes the first test light intensity distribution data for the first test pattern, and obtaining the test light intensity distribution data set includes: based on the position of the first line segment in the first test pattern, using light source data, pupil data, and test mask data to determine the first local light intensity distribution data; position in the graph, using the light source data, pupil data, and test mask data to determine the second local light intensity distribution data; and splicing the first local light intensity distribution data and the second local light intensity distribution data to obtain the first test light intensity distribution data.
  • the lengths of the first line segment and the second line segment depend on the size of the test pattern at the measurement position. In this way, the lengths of the first line segment and the second line segment can be appropriately set, so as to reduce the size of data used for subsequent model training and reduce the amount of computation when determining the light intensity distribution.
  • the target light intensity distribution data indicates the light intensity distribution at a predetermined position on the wafer
  • determining the size of the wafer pattern corresponding to the target pattern includes: based on the target light intensity distribution data, using a development model to determine the target size difference data, the target size difference data indicates the size difference between the target pattern and the corresponding wafer pattern at the predetermined position; and using the size of the target pattern at the predetermined position and the target size difference data to obtain the size of the wafer pattern corresponding to the target pattern at the predetermined position.
  • the development model can be used to predict the size of the target pattern actually obtained on the wafer after exposure and development, so as to guide the subsequent process of correcting the pattern on the mask.
  • an apparatus for determining wafer pattern dimensions includes a developing model generating module and a wafer pattern dimension determining module.
  • the development model generating module is configured to: generate a development model for determining the size of the wafer pattern at least based on the test light intensity distribution data set and the wafer pattern data set for at least one test pattern, each test light intensity distribution data in the test light intensity distribution data set indicates the light intensity distribution obtained by the light source of the lithography equipment on the wafer through the corresponding test pattern, and each wafer pattern data in the wafer pattern data set indicates the size of the wafer pattern corresponding to the corresponding test pattern.
  • the wafer pattern size determining module is configured to: use a development model to determine the size of the wafer pattern corresponding to the target pattern at least based on target light intensity distribution data for the target pattern, the target light intensity distribution data indicating the light intensity distribution obtained by the light source on the wafer through the target pattern.
  • the solution according to the present disclosure uses the trained model structure to fit the development process that is difficult to accurately describe with formulas, so the accuracy of the development model describing the development process can be improved.
  • the light intensity distribution is used as a training sample in the process of training the model, the physical properties of the model can be guaranteed, so that the trained model is not prone to overfitting, thereby improving the prediction accuracy of the OPC model.
  • each test pattern respectively has a measurement area
  • each test light intensity distribution data in the test light intensity distribution data set indicates the light intensity distribution at the corresponding measurement area of the wafer
  • each wafer pattern data in the wafer pattern data set indicates the size of the wafer pattern at the corresponding measurement area
  • the development model generation module is further configured: Based on the wafer pattern data set and the test pattern size data set for at least one test pattern, determine a size difference data set for at least one test pattern, each test pattern size data in the test pattern size data set indicates a corresponding The size of the test pattern at the corresponding measurement area, each size difference data in the size difference data set indicates the size difference between the corresponding test pattern and the corresponding wafer pattern at the corresponding measurement area; and the model structure is trained using the test light intensity distribution data set and the size difference data set to obtain a development model.
  • the measurement area is a one-dimensional measurement line at the measurement location, and each test light intensity distribution data in the test light intensity distribution data set indicates the one-dimensional light intensity distribution at the corresponding measurement line of the wafer.
  • the one-dimensional light intensity distribution at the measurement position can be calculated instead of the light intensity distribution of the entire graph, so the size of the sample data used for model training can be greatly reduced, the speed of model training can be improved, and the running speed of the scheme according to some implementations of the present disclosure can be greatly improved, thereby improving the efficiency of OPC.
  • the apparatus further includes: a test light intensity distribution data set acquisition module configured to: based on the measurement position, use light source data and pupil data for the lithography apparatus and test mask data for the test mask to acquire the test light intensity distribution data set, the light source data indicates the distribution of the energy and phase of the light emitted by the light source with respect to the spatial position, the pupil data indicates whether light can be received at each spatial position, the test mask includes at least one test pattern, the test mask data indicates the distribution of the light transmittance of the test mask with respect to the spatial position;
  • the target light intensity distribution data acquisition module is configured to: use the light source data, pupil data, and target mask data for the target mask to acquire target light intensity distribution data, the target mask includes a target pattern, and the target mask data indicates the distribution of light transmittance of the target mask with respect to spatial positions.
  • Obtaining the light intensity distribution in this way can ensure the physicality of the obtained test light intensity distribution data set and target light intensity distribution data, thereby ensuring the physicality of the OPC model, making the subsequent training of the development model less prone to overfitting, thereby improving the prediction accuracy of the OPC model.
  • the measurement line includes a first line segment and a second line segment, the first line segment and the second line segment are spaced apart from each other and are respectively located at the edge of the test pattern at the measurement position, at least one test pattern includes the first test pattern, the test light intensity distribution data set includes the first test light intensity distribution data for the first test pattern, the test light intensity distribution data set acquisition module is also configured to: use the light source data, pupil data, and test mask data based on the position of the first line segment in the first test pattern to determine the first local light intensity distribution data; The position of the line segment in the first test pattern is determined by using the light source data, pupil data, and test mask data to determine the second local light intensity distribution data; and splicing the first local light intensity distribution data and the second local light intensity distribution data to obtain the first test light intensity distribution data.
  • the lengths of the first line segment and the second line segment depend on the size of the test pattern at the measurement position. In this way, the lengths of the first line segment and the second line segment can be appropriately set, so as to reduce the size of data used for subsequent model training and reduce the amount of computation when determining the light intensity distribution.
  • the target light intensity distribution data indicates the light intensity distribution at a predetermined position on the wafer
  • the wafer pattern size determination module is further configured to: based on the target light intensity distribution data, use a development model to determine target size difference data, the target size difference data indicates the size difference between the target pattern and the corresponding wafer pattern at the predetermined position; and use the size of the target pattern at the predetermined position and the target size difference data to obtain the size of the wafer pattern corresponding to the target pattern at the predetermined position.
  • the development model can be used to predict the size of the target pattern actually obtained on the wafer after exposure and development, so as to guide the subsequent process of correcting the pattern on the mask.
  • an electronic device comprises: at least one processor; and at least one memory coupled to the at least one processor and storing instructions for execution by the at least one processor, the instructions, when executed by the at least one processor, cause the electronic device to perform a method according to the first aspect of the present disclosure.
  • the electronic device according to the present disclosure uses the trained model structure to fit the development process that is difficult to accurately describe with formulas, so the accuracy of the development model describing the development process can be improved.
  • the light intensity distribution is used as a training sample in the process of training the model, the physical properties of the model can be guaranteed, so that the trained model is not prone to overfitting, thereby improving the prediction accuracy of the OPC model.
  • a computer readable storage medium stores a computer program.
  • the computer program implements the method according to the first aspect of the present disclosure when executed by a processor.
  • the computer-readable storage medium according to the present disclosure uses the trained model structure to fit the development process that is difficult to accurately describe with formulas, thus improving the accuracy of the development model describing the development process.
  • the physical properties of the model can be guaranteed, so that the trained model is not prone to overfitting, thereby improving the prediction accuracy of the OPC model.
  • a computer program product comprises computer-executable instructions which, when executed by a processor, cause a computer to implement the method according to the first aspect.
  • the computer program product according to the present disclosure uses the trained model structure to fit the development process that is difficult to accurately describe with formulas, thus improving the accuracy of the development model describing the development process.
  • the physical properties of the model can be guaranteed, so that the trained model is not prone to overfitting, thereby improving the prediction accuracy of the OPC model.
  • FIG. 1 shows a schematic diagram of a lithographic apparatus to which some embodiments of the present disclosure can be applied;
  • Figure 2 shows a block diagram of an example environment according to some embodiments of the present disclosure
  • FIG. 3 shows a flowchart of a method for determining wafer pattern dimensions according to some embodiments of the present disclosure
  • FIG. 4 shows a schematic diagram of an exemplary test pattern according to some embodiments of the present disclosure
  • FIG. 5 shows a schematic diagram of a process for obtaining a test light intensity distribution data set according to some embodiments of the present disclosure
  • FIG. 6 illustrates a block diagram of an example apparatus for determining wafer pattern dimensions according to some embodiments of the present disclosure.
  • Fig. 7 shows a schematic block diagram of an example device that may be used to implement some embodiments according to the present disclosure.
  • the term “comprise” and its variants mean open inclusion, ie “including but not limited to”.
  • the term “or” means “and/or” unless otherwise stated.
  • the term “based on” means “based at least in part on”.
  • the terms “one example embodiment” and “one embodiment” mean “at least one example embodiment.”
  • the term “another embodiment” means “at least one further embodiment”.
  • a conventional OPC modeling scheme uses formulas to describe the optical model corresponding to the exposure process and the development model corresponding to the development process, and optimizes the optical model and the development model through an optimization algorithm.
  • the goal of optimization is to use the optical model and the development model to predict the graphics on the wafer according to the graphics on the mask. Close to the graphics actually obtained after exposure and development.
  • this conventional solution has the following problems: for advanced technology nodes, especially the Negative Tone Development (NTD) process, due to the complex performance of the photoresist during the development process, it is difficult to accurately describe the development process through formulas, which leads to poor prediction performance of the development model, thereby reducing the prediction accuracy of the OPC model.
  • NTD Negative Tone Development
  • the development model is generated by training the model by using the light intensity distribution data set for the test pattern and the actual measured wafer pattern data set, and the size of the wafer pattern is predicted by means of the generated development model.
  • FIG. 1 shows a schematic diagram of a lithographic apparatus 100 in which some embodiments of the present disclosure can be applied.
  • the lithography apparatus 100 may include a light source 110 , an illumination device 120 , a mask table 140 for carrying a mask 130 , a projection device 150 , and a support table 170 for carrying a wafer 160 . It should be understood that the lithographic apparatus 100 may also include components not shown and/or components shown may be omitted, and that the scope of the present disclosure is not limited in this respect.
  • the illuminating device 120 receives the light beam emitted from the light source 110 and can adjust the light beam so that the light beam has a desired spatial distribution and angular intensity distribution at the plane of the mask 130 .
  • the mask 130 is used to impart a patterned cross-section to the incident beam.
  • the projection device 150 projects the patterned cross section of the light beam imparted by the mask 130 onto the target portion T of the wafer 160 .
  • Diagram MA schematically shows patterns on mask 130
  • diagram WA schematically shows patterns on wafer 160 . It should be understood that in the context of the present disclosure the term "light” or "light beam” may encompass all types of electromagnetic radiation, such as ultraviolet radiation.
  • reticle or “mask” denotes any suitable general patterning component that may be used to impart a patterned cross-section to the incident beam corresponding to the pattern to be created in the target portion T of the wafer 160 .
  • the scope of the present disclosure is not limited in the above respects.
  • the pattern fidelity describes the consistency between the pattern obtained on the wafer 160 and the pattern on the mask 130 after exposure and development, and high pattern fidelity helps to ensure that the function of the finally obtained chip meets predetermined specifications.
  • the pattern on the mask 130 is compensated and corrected by means of the OPC model, so that the pattern obtained on the wafer 160 after exposure and development is consistent with the desired pattern.
  • the prediction accuracy of the OPC model plays a decisive role in the correction results, so it is expected to build a high-precision OPC model.
  • FIG. 2 shows a block diagram of an example environment 200 according to some embodiments of the present disclosure.
  • example environment 200 may generally include electronic device 240 .
  • the electronic device 240 may be a device having computing functions such as a personal computer, a workstation, a server, and the like. The scope of the present disclosure is not limited in this respect.
  • the electronic device 240 may receive as input the test light intensity distribution data set 210 and the wafer pattern data set 220 for at least one test pattern.
  • a "test pattern” refers to a pattern that is exposed and developed during the process of constructing the OPC model to obtain a pattern corresponding to the pattern that is finally actually obtained on the wafer 160 .
  • the finally actually obtained pattern on the wafer 160 is also referred to as a "wafer pattern”.
  • the test patterns may include, for example, a series of one-dimensional patterns or two-dimensional patterns.
  • the test light intensity distribution data in the test light intensity distribution data set 210 indicates the light intensity distribution on the wafer 160 obtained by the light source 110 of the lithography apparatus 100 through the corresponding test pattern.
  • the wafer pattern data in the wafer pattern data set 220 indicates the size of the wafer pattern corresponding to the corresponding test pattern, which can be obtained, for example, by measuring the size of the pattern actually obtained on the wafer 160 by means of a critical dimension scanning electron microscope (CD-SEM).
  • CD-SEM critical dimension scanning electron microscope
  • the electronic device 240 may also receive as input the target light intensity distribution data 230 for the target pattern.
  • the target light intensity distribution data 230 indicates the light intensity distribution obtained by the light source 110 on the wafer 160 via the target pattern.
  • a "target graph” means a graph that needs to be corrected by means of the OPC model.
  • the test light intensity distribution data set 210 , the wafer pattern data set 220 and the target light intensity distribution data 230 may be input into the electronic device 240 by a user.
  • the test light intensity distribution data set 210 , the wafer pattern data set 220 and the target light intensity distribution data 230 may have been pre-stored in the electronic device 240 .
  • electronic device 240 may be communicatively coupled to other devices to obtain test light intensity distribution data set 210 , wafer pattern data set 220 , and target light intensity distribution data 230 from other devices.
  • the test light intensity distribution data set 210 , the wafer pattern data set 220 and the target light intensity distribution data 230 may also be predetermined by the electronic device 240 . The scope of the present disclosure is not limited in this respect.
  • the electronic device 240 can generate a development model based on the test light intensity distribution data set 210 and the wafer pattern data set 220 , and use the generated development model to predict the size of the corresponding wafer pattern based on the target light intensity distribution data 230 . This will be described in further detail below with reference to FIGS. 3 to 5 .
  • FIG. 3 shows a flowchart of a method 300 for determining wafer pattern dimensions according to some embodiments of the present disclosure.
  • the method 300 may be executed by the electronic device 240 as shown in FIG. 2 . It should be appreciated that method 300 may also include additional blocks not shown and/or blocks shown may be omitted, and that the scope of the present disclosure is not limited in this respect.
  • the electronic device 240 acquires the test light intensity distribution data set 210 for at least one test pattern.
  • Each test light intensity distribution data in the test light intensity distribution data set 210 indicates the light intensity distribution obtained by the light source 110 of the lithography apparatus 100 on the wafer 160 via the corresponding test pattern.
  • the electronic device 240 obtains the wafer pattern data set 220 for at least one test pattern, each wafer pattern data in the wafer pattern data set 220 indicating the size of the wafer pattern corresponding to the corresponding test pattern.
  • the electronic device 240 may receive the predetermined test light intensity distribution data set 210 and the actual measured wafer pattern data set 220 from an external device.
  • the electronic device 240 may use the light source data and pupil data for the lithographic apparatus 100 and the test mask data for the test mask to obtain the test light intensity distribution data set 210 .
  • the light source data may indicate the distribution of energy and phase of light emitted by the light source 110 with respect to spatial location.
  • Pupil data is associated with the configuration of the lens groups of the lithographic apparatus 100 and may indicate whether light can be received at various spatial locations.
  • the test mask may include at least one test pattern, and the test mask data may indicate a distribution of light transmittance of the test mask with respect to a spatial position. It should be noted that a test mask may include all of the at least one test pattern, or may only include a part of the at least one test pattern, and the scope of the present disclosure is not limited in this respect.
  • the electronic device 240 can generate test light intensity distribution data based on light source data, pupil data and test mask data by means of Hopkins Euqation in optical imaging theory. Equation (1) below shows a form of the Hopkins formula:
  • the light source function J describes the distribution of the energy and phase of the light emitted by the light source 110 with respect to the spatial position
  • the pupil function P describes whether light can be received at each spatial position
  • the mask function Q describes the distribution of the light transmittance of the mask 130 with respect to the spatial position.
  • the electronic device 240 can use the acquired light source data, pupil data and test mask data to obtain the light source function J, pupil function P and mask function Q in the above formula (1), so as to calculate the test light intensity distribution data set 210.
  • the physical model described by the above formula (1) to calculate the light intensity distribution the physicality of the obtained test light intensity distribution data set can be guaranteed, so that the physicality of the OPC model can be guaranteed, so that the subsequent training of the development model is not prone to overfitting, so that the prediction accuracy of the OPC model can be improved.
  • the electronic device 240 may acquire the test light intensity distribution data set 210 in any other suitable manner, and the scope of the present disclosure is not limited in this respect.
  • FIG. 4 shows a schematic diagram of an exemplary test pattern 400 according to some embodiments of the present disclosure.
  • the test pattern 400 has five independent polygons (polygon) 410 - 1 to 410 - 5 and a measurement area 420 located at a measurement position.
  • the polygons 410-1 to 410-5 may be light-transmissive or opaque depending on the light-transmittance of the mask.
  • polygons 410-1 through 410-5 may be light transmissive in the case of dark filed.
  • polygons 410-1 to 410-5 may be opaque in case of bright filed.
  • the scope of the present disclosure is not limited in this respect.
  • the measurement area 420 is shown as a one-dimensional measurement line at the measurement location.
  • the measurement area 420 at the measurement position indicates an area where the size of the portion of the wafer pattern within the area is to be measured after the test pattern 400 is developed and exposed to obtain a corresponding wafer pattern.
  • the test pattern 400 may also have any other suitable shape, and the measurement area 420 may also have any other suitable shape, and the scope of the present disclosure is not limited in this respect.
  • the electronic device 240 can obtain the test light intensity distribution data set 210 based on the measurement position of the measurement area 420 using the light source data, the pupil data, and the test mask data.
  • the test light intensity distribution data set 210 indicates the light intensity distribution at the corresponding measurement area 420 , rather than the light intensity distribution of the entire pattern.
  • the electronic device 240 may calculate the light intensity distribution at the corresponding measurement region 420 of the wafer 160 in a manner similar to that described above with reference to Equation (1) to obtain the test light intensity distribution data set 210 .
  • the wafer pattern data may only indicate the size of the wafer pattern at the corresponding measurement area 420 , for example, only the size of the pattern at the corresponding measurement area 420 may be measured by CD-SEM.
  • the test light intensity distribution data may indicate the one-dimensional light intensity distribution at the corresponding measurement line of the wafer 160 .
  • a one-dimensional measurement line refers to a line extending in one direction.
  • FIG. 5 shows a schematic diagram of a process for obtaining a test light intensity distribution data set 210 according to some embodiments of the present disclosure.
  • Diagram 500 shows test pattern 400 shown in FIG. 4 .
  • a measurement line (not fully shown in illustration 500) includes a first line segment 520-1 and a second line segment 520-2.
  • the first line segment 520 - 1 and the second line segment 520 - 2 are spaced apart from each other and are respectively located at the edges of the test pattern 400 at the measurement position.
  • the center of the first line segment 520 - 1 and the center of the second line segment 520 - 2 may respectively be located at the edges of the test pattern 400 at the measurement position.
  • first line segment 520-1 is centered at the left edge of polygon 410-3
  • second line segment 520-2 is centered at the right edge of polygon 410-3.
  • the lengths of the first line segment 520 - 1 and the second line segment 520 - 2 depend on the size of the test pattern 400 at the measurement position, for example, the size in the X direction in FIG. 4 and diagram 500 .
  • the length of the first line segment 520 - 1 and the second line segment 520 - 2 may be a quarter of the width of the test pattern 400 at the measurement position.
  • the ratio of the length of the first line segment 520-1 and the second line segment 520-2 to the width of the test pattern 400 at the measurement position may also be any other suitable value, and the scope of the present disclosure is not limited in this respect. In this way, the lengths of the first line segment 520-1 and the second line segment 520-2 can be appropriately set, so as to reduce the size of data used for subsequent model training and reduce the amount of computation when determining the light intensity distribution.
  • the electronic device 240 may use the light source data, pupil data, and test mask data to determine the first local light intensity distribution data based on the position of the first line segment 520-1 in the test pattern 400.
  • the first local light intensity distribution data indicates the light intensity distribution of the light source 110 of the lithographic apparatus 100 via the test pattern 400 at the position corresponding to the first line segment 520 - 1 on the wafer 160 .
  • the graph 502 in FIG. 5 exemplarily shows the first local light intensity distribution data for the first line segment 520 - 1 , wherein the axis L of abscissa indicates position, and the axis I of ordinate indicates light intensity.
  • the electronic device 240 may use the light source data, pupil data, and test mask data based on the position of the second line segment 520-2 in the test pattern 400 to determine the second local light intensity distribution data, the second local light intensity distribution data indicating the light intensity distribution of the light source 110 of the lithography apparatus 100 at the position corresponding to the second line segment 520-2 on the wafer 160 via the test pattern 400.
  • the graph 504 in FIG. 5 exemplarily shows the second local light intensity distribution data for the second line segment 520 - 2 , wherein the axis L of abscissa indicates the position, and the axis I of ordinate indicates the light intensity.
  • the electronic device 240 may splice the first local light intensity distribution data and the second local light intensity distribution data to obtain the first test light intensity distribution data for the test pattern 400 .
  • the diagram 506 in FIG. 5 shows the first test light intensity distribution data obtained by splicing the first local light intensity distribution data shown in the diagram 502 and the second local light intensity distribution data shown in the diagram 504, wherein the abscissa axis L indicates the position, and the ordinate axis I indicates the light intensity.
  • the electronic device 240 generates a development model for determining wafer pattern dimensions based at least on the test light intensity distribution dataset 210 and the wafer pattern dataset 220 .
  • the electronic device 240 may use the test light intensity distribution data set 210 as a sample and directly use the wafer pattern data set 220 as a label to train a model structure to obtain a development model.
  • the model structure may be a model structure such as a neural network, deep learning, etc., wherein at least a part of parameters contained in the model structure need to be trained using a data set to obtain a final usable model.
  • the electronic device 240 can also train the model structure based on the size difference between the test pattern 400 and the corresponding wafer pattern at the measurement area 420 to obtain a development model.
  • the electronic device 240 may determine a size difference data set for at least one test pattern based on the wafer pattern data set 220 and the test pattern size data set for the at least one test pattern.
  • Each test pattern size data in the test pattern size data set indicates the size of the corresponding test pattern 400 at the corresponding measurement area 420 .
  • Each size difference data in the size difference data set indicates the size difference between the corresponding test pattern 400 and the corresponding wafer pattern at the corresponding measurement area 420 .
  • the electronic device 240 may, for example, calculate the difference between the width of the polygon 410 - 3 in the measurement region 420 and its actual width on the wafer 160 as the size difference data.
  • the electronic device 240 can calculate the size difference data for each test pattern in a similar manner to construct a size difference data set. Then, the electronic device 240 can use the test light intensity distribution data set 210 and the size difference data set to train the model structure to obtain a development model. By training the model based on the size difference, the influence of the size of the test graph itself can be reduced when training the model, thereby improving the prediction accuracy of the model.
  • the model structure used to fit the test light intensity distribution data set and the size difference data set may be, for example, a convolutional neural network including 8 convolutional layers, 4 pooling layers and 2 fully connected layers, wherein a pooling layer is arranged after every 2 convolutional layers, and 2 fully connected layers are arranged at the end of the convolutional neural network.
  • the electronic device 240 can train the convolutional network network through the following configuration: the test light intensity distribution data set 210 is used as a sample, the size difference data set is used as a label, and a mean square error (Mean Square Error, MSE) loss function and an Adaptive Moment Estimation (Adaptive Moment Estimation, ADAM) optimizer are used.
  • MSE mean square error
  • ADAM Adaptive Moment Estimation
  • the accuracy of the development model describing the development process can be improved, thereby improving the prediction accuracy of the OPC model.
  • the light intensity distribution is used as a training sample in the process of training the model, the physicality of the model can be guaranteed, so that the trained model is not prone to overfitting.
  • the electronic device 240 may also generate a development model based on the test light intensity distribution data set 210 and the wafer pattern data set 220 in any other suitable manner, and the scope of the present disclosure is not limited in this regard.
  • the electronic device 240 utilizes the development model to determine the size of the wafer pattern corresponding to the target pattern based at least on the target light intensity distribution data 230 for the target pattern.
  • the electronic device 240 can obtain the target light intensity distribution data 230 by using the light source data, the pupil data, and the target mask data for the target mask 130 in a manner similar to that described above with reference to equation (1).
  • the target mask 130 includes a target pattern, and the target mask data indicates a distribution of light transmittance of the target mask 130 with respect to a spatial position. The present disclosure will not be repeated here.
  • the target light intensity distribution data 230 may only indicate the light intensity distribution at the predetermined position on the wafer 160 .
  • the electronic device 240 when the electronic device 240 obtains the development model by using the test light intensity distribution data set 210 as a sample and directly using the wafer pattern data set 220 as a label to train the model structure, the electronic device 240 can directly input the target light intensity distribution data 230 into the development model, and use the output of the development model as the size of the wafer pattern corresponding to the target pattern.
  • the electronic device 240 may use the development model to determine target size difference data based on the target light intensity distribution data 230, the target size difference data indicating the size difference between the target pattern and the corresponding wafer pattern at a predetermined position. For example, the electronic device 240 may input the target light intensity distribution data 230 indicating the light intensity distribution at a predetermined position on the wafer 160 into the development model to obtain the target size difference data.
  • the electronic device 240 may use the size of the target pattern at the predetermined position and the target size difference data to acquire the size of the wafer pattern corresponding to the target pattern at the predetermined position. For example, the electronic device 240 may add the size of the target pattern at the predetermined position to the target size difference data to predict the size of the wafer pattern at the predetermined position.
  • the electronic device 240 can use the development model to predict the size of the target pattern actually obtained on the wafer 160 after exposure and development based on the light intensity distribution, so as to guide the subsequent process of correcting the target pattern.
  • the development model is generated by using the light intensity distribution data set for the test pattern and the actually measured wafer pattern data set to train the model, and the size of the wafer pattern is predicted by means of the generated development model.
  • the accuracy of the development model describing the development process can be improved.
  • the light intensity distribution is used as a training sample in the process of training the model, the physical properties of the model can be guaranteed, so that the trained model is not prone to overfitting, thereby improving the prediction accuracy of the OPC model.
  • FIG. 6 shows a block diagram of an example apparatus 600 for determining wafer pattern dimensions according to some embodiments of the present disclosure.
  • the apparatus 600 can be used, for example, to implement the electronic device 240 as shown in FIG. 2 .
  • the apparatus 600 may include a development model generating module 602 configured to: at least based on a test light intensity distribution data set and a wafer pattern data set for at least one test pattern, generate a development model for determining the size of the wafer pattern, each test light intensity distribution data in the test light intensity distribution data set indicates the light intensity distribution obtained by the light source of the lithography equipment on the wafer through the corresponding test pattern, and each wafer pattern data in the wafer pattern data set indicates the size of the wafer pattern corresponding to the corresponding test pattern.
  • the apparatus 600 may further include a wafer pattern size determination module 604, the wafer pattern size determination module 604 is configured to: use a development model to determine the size of the wafer pattern corresponding to the target pattern at least based on the target light intensity distribution data for the target pattern, the target light intensity distribution data indicates the light intensity distribution obtained by the light source on the wafer through the target pattern.
  • At least one test pattern has a measurement area
  • each test light intensity distribution data in the test light intensity distribution data set indicates the light intensity distribution at the corresponding measurement area of the wafer
  • each wafer pattern data in the wafer pattern data set indicates the size of the wafer pattern at the corresponding measurement area
  • the development model generation module 602 is also configured to: determine a size difference data set for at least one test pattern based on the wafer pattern data set and the test pattern size data set for at least one test pattern, each test pattern size data in the test pattern size data set indicates the size of the corresponding test pattern at the corresponding measurement area, and each size difference data in the size difference data set indicates The size difference between the corresponding test pattern and the corresponding wafer pattern at the corresponding measurement area; and use the test light intensity distribution data set and the size difference data set to train the model structure to obtain the development model.
  • the measurement area is a one-dimensional measurement line at the measurement position, and each test light intensity distribution data in the test light intensity distribution data set indicates the one-dimensional light intensity distribution at the corresponding measurement line of the wafer.
  • the apparatus 600 further includes: a test light intensity distribution data set acquisition module, configured to: based on the measurement position, use the light source data and pupil data for the lithographic apparatus and the test mask data for the test mask to acquire the test light intensity distribution data set, the light source data indicates the distribution of the energy and phase of the light emitted by the light source with respect to the spatial position, the pupil data indicates whether light can be received at each spatial position, the test mask includes at least one test pattern, and the test mask data indicates the distribution of the light transmittance of the test mask with respect to the spatial position and a target light intensity distribution data acquisition module configured to: use light source data, pupil data, and target mask data for the target mask to acquire target light intensity distribution data, the target mask includes a target pattern, and the target mask data indicates the distribution of light transmittance of the target mask with respect to spatial positions.
  • a target light intensity distribution data acquisition module configured to: use light source data, pupil data, and target mask data for the target mask to acquire target light intensity distribution data, the target mask includes a target pattern, and
  • the measurement line includes a first line segment and a second line segment, the first line segment and the second line segment are spaced apart from each other and are respectively located at the edge of the test pattern at the measurement position, at least one test pattern includes the first test pattern, the test light intensity distribution data set includes the first test light intensity distribution data for the first test pattern, the test light intensity distribution data set acquisition module is also configured to: use the light source data, pupil data, and test mask data based on the position of the first line segment in the first test pattern to determine the first local light intensity distribution data; The position of the second line segment in the first test pattern is determined by using the light source data, the pupil data, and the test mask data to determine the second local light intensity distribution data; and splicing the first local light intensity distribution data and the second local light intensity distribution data to obtain the first test light intensity distribution data.
  • the lengths of the first line segment and the second line segment depend on the size of the test pattern at the measurement position.
  • the target light intensity distribution data indicates the light intensity distribution at a predetermined position on the wafer
  • the wafer pattern size determination module 604 is further configured to: based on the target light intensity distribution data, use a development model to determine the target size difference data, the target size difference data indicates the size difference between the target pattern and the corresponding wafer pattern at the predetermined position; use the size of the target pattern at the predetermined position and the target size difference data to obtain the size of the wafer pattern corresponding to the target pattern at the predetermined position.
  • the modules and/or units included in the device 600 may be implemented in various ways, including software, hardware, firmware or any combination thereof.
  • one or more units may be implemented using software and/or firmware, such as machine-executable instructions stored on a storage medium.
  • some or all of the units in apparatus 600 may be at least partially implemented by one or more hardware logic components.
  • Exemplary types of hardware logic components include, by way of example and not limitation, Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), System on Chips (SOCs), Complex Programmable Logic Devices (CPLDs), and the like.
  • modules and/or units shown in FIG. 6 may be implemented in part or in whole as hardware modules, software modules, firmware modules or any combination thereof.
  • the procedures, methods or processes described above may be implemented by hardware in the storage system or a host corresponding to the storage system or other computing devices independent of the storage system.
  • Fig. 7 shows a schematic block diagram of an example device 700 that may be used to implement some embodiments according to the present disclosure.
  • the device 700 may be used to implement the electronic device 230 as shown in FIG. 2 .
  • the device 700 includes a central processing unit (CPU) 701, which can perform various appropriate actions and processes according to computer program instructions stored in a read only memory (ROM) 702 or loaded from a storage unit 708 into a random access memory (RAM) 703.
  • ROM read only memory
  • RAM random access memory
  • various programs and data necessary for the operation of the device 700 can also be stored.
  • the CPU 701, ROM 702, and RAM 703 are connected to each other via a bus 704.
  • An input/output (I/O) interface 705 is also connected to the bus 704 .
  • I/O input/output
  • the I/O interface 705 includes: an input unit 706, such as a keyboard, a mouse, etc.; an output unit 707, such as various types of displays, speakers, etc.; a storage unit 708, such as a magnetic disk, an optical disk, etc.; and a communication unit 709, such as a network card, modem, wireless communication transceiver, etc.
  • the communication unit 709 allows the device 700 to exchange information/data with other devices over a computer network such as the Internet and/or various telecommunication networks.
  • the processing unit 701 executes various methods and processes described above, such as the method 300 .
  • method 300 may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as storage unit 708 .
  • part or all of the computer program may be loaded and/or installed on the device 700 via the ROM 702 and/or the communication unit 709.
  • the CPU 701 may be configured to execute the method 300 in any other suitable manner (for example, by means of firmware).
  • FPGAs Field Programmable Gate Arrays
  • ASICs Application Specific Integrated Circuits
  • ASSPs Application Specific Standard Products
  • SOCs System on Chips
  • CPLDs Load Programmable Logic Devices
  • Program codes for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to processors or controllers of general purpose computers, special purpose computers or other programmable data processing devices, so that the program codes cause the functions/operations specified in the flowcharts and/or block diagrams to be implemented when executed by the processors or controllers.
  • the program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.
  • a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device.
  • a machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium.
  • a machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing.
  • a machine-readable storage medium would include one or more wire-based electrical connections, a portable computer disk, a hard disk, random access memory (RAM), read only memory (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, compact disk read only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.
  • RAM random access memory
  • ROM read only memory
  • EPROM or flash memory erasable programmable read only memory
  • CD-ROM compact disk read only memory
  • magnetic storage devices or any suitable combination of the foregoing.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Exposure And Positioning Against Photoresist Photosensitive Materials (AREA)

Abstract

用于确定晶圆图形尺寸的方法(300)、装置(600)、设备(700)、介质以及程序产品,用于确定晶圆图形尺寸的方法(300)包括:至少基于针对至少一个测试图形的测试光强分布数据集和晶圆图形数据集,生成用于确定晶圆图形尺寸的显影模型(306),测试光强分布数据集(210)中的每个测试光强分布数据指示光刻设备的光源经由相应的测试图形在晶圆上得到的光强分布,晶圆图形数据集(220)中的每个晶圆图形数据指示与相应的测试图形对应的晶圆图形的尺寸;以及至少基于针对目标图形的目标光强分布数据,利用显影模型来确定与目标图形对应的晶圆图形的尺寸(308),目标光强分布数据(230)指示光源经由目标图形在晶圆上得到的光强分布。以此方式,可以提高对在晶圆上获得的实际图形尺寸的预测精确度。

Description

用于确定晶圆图形尺寸的方法、装置、设备、介质以及程序产品 技术领域
本公开的实施例总体上涉及芯片制造领域,更具体地涉及用于确定晶圆图形尺寸的方法、装置、设备、介质以及程序产品。
背景技术
在集成电路(Integrated Circuit,IC)制造中,光刻是指利用光学投影曝光的方法将掩模上的图形转移到晶圆上的过程,其是集成电路制造的关键步骤。图形保真是对光刻的基本要求。随着光刻技术发展到纳米级技术节点,光刻过程受光学临近效应(Optical Proximity Effect,OPE)的影响越来越严重。光学临近效应是指:掩模上的图形随着周围图形的距离远近、尺寸大小以及密集程度的不同而发生变化,出现诸如尺寸偏移、拐角钝化、线端缩头、甚至图形消失等现象。这会导致晶圆上的图形与所期望的图形之间存在较大差异,进而造成集成电路的电学特性出现偏差,影响最终所获得的芯片的功能和良率。
光学邻近效应修正(Optical Proximity Correction,OPC)是一种针对掩模的预补偿技术。通过对掩模上的图形进行补偿修正,使其经过曝光显影后在晶圆上得到图形与所期望的图形一致。在利用OPC对掩模进行修正补偿时,需要借助于模型建立掩模上图形与晶圆上的实际图形之间的映射关系,以便为OPC修正过程提供指导,使得修正结果能够向所期望的图形靠近。随着设计版图越来越复杂,对OPC模型的预测精确度的要求也越来越高。
发明内容
鉴于上述问题,本公开的实施例旨在提供一种用于确定晶圆图形尺寸的方案。
根据本公开的第一方面,提供了一种用于确定晶圆图形尺寸的方法。该方法包括:至少基于针对至少一个测试图形的测试光强分布数据集和晶圆图形数据集,生成用于确定晶圆图形尺寸的显影模型,测试光强分布数据集中的每个测试光强分布数据指示光刻设备的光源经由相应的测试图形在晶圆上得到的光强分布,晶圆图形数据集中的每个晶圆图形数据指示与相应的测试图形对应的晶圆图形的尺寸;以及至少基于针对目标图形的目标光强分布数据,利用显影模型来确定与目标图形对应的晶圆图形的尺寸,目标光强分布数据指示光源经由目标图形在晶圆上得到的光强分布。根据本公开的方法借助于经训练的模型结构来拟合难以用公式准确描述的显影过程,因此可以提高描述显影过程的显影模型的准确性。此外,由于在训练模型的过程中,使用光强分布作为训练样本,因此可以保证模型的物理性,使得经训练得到的模型不易出现过拟合,从而可以提高OPC模型的预测精确度。
在一些实现方式中,至少一个测试图形分别具有量测区域,测试光强分布数据集中的每个测试光强分布数据指示在晶圆的相应量测区域处的光强分布,晶圆图形数据集中的每个晶圆图形数据指示晶圆图形在相应量测区域处的尺寸,生成显影模型包括:基于晶圆图形数据集和针对至少一个测试图形的测试图形尺寸数据集,确定针对至少一个测试图形的尺寸差异数据集,测试图形尺寸数据集中的每个测试图形尺寸数据指示相应的 测试图形在对应的量测区域处的尺寸,尺寸差异数据集中的每个尺寸差异数据指示相应测试图形与对应的晶圆图形在相应量测区域处的尺寸差异;以及使用测试光强分布数据集和尺寸差异数据集对模型结构进行训练,以获得显影模型。通过基于尺寸差异来训练模型,可以在训练模型时减少测试图形本身的尺寸所带来的影响,从而可以提高模型的预测精确度。
在一些实现方式中,量测区域是在量测位置处的一维的量测线条,并且测试光强分布数据集中的每个测试光强分布数据指示在晶圆的相应量测线条处的一维光强分布。通过这种方式,可以仅计算在量测位置处的一维光强分布,而非整个图形的光强分布,因此可以大幅减小后续用于模型训练的样本数据的大小,提高模型训练的速度,并且可以大幅提升根据本公开的一些实现方式的方法的运行速度,从而可以提高OPC的效率。
在一些实现方式中,该方法还包括:基于量测位置,使用针对光刻设备的光源数据和光瞳数据、以及针对测试掩模的测试掩模数据,来获取测试光强分布数据集,光源数据指示光源发出的光的能量和相位关于空间位置的分布,光瞳数据指示在各个空间位置处是否能够接收到光,测试掩模包括至少一个测试图形,测试掩模数据指示测试掩模的透光率关于空间位置的分布;以及使用光源数据、光瞳数据、以及针对目标掩模的目标掩模数据,来获取目标光强分布数据,目标掩模包括目标图形,目标掩模数据指示目标掩模的透光率关于空间位置的分布。通过这种方式来获取光强分布,可以保证所获得的测试光强分布数据集和目标光强分布数据的物理性,从而可以保证OPC模型的物理性,使得后续对于显影模型的训练不易出现过拟合现象,从而可以提高OPC模型的预测精确度。
在一些实现方式中,量测线条包括第一线段和第二线段,第一线段和第二线段彼此间隔开并且分别位于测试图形在量测位置处的图形的边缘,至少一个测试图形包括第一测试图形,测试光强分布数据集包括针对第一测试图形的第一测试光强分布数据,获取测试光强分布数据集包括:基于第一线段在第一测试图形中的位置,使用光源数据、光瞳数据、和测试掩模数据,来确定第一局部光强分布数据;基于第二线段在第一测试图形中的位置,使用光源数据、光瞳数据、和测试掩模数据,来确定第二局部光强分布数据;以及对第一局部光强分布数据和第二局部光强分布数据进行拼接,以获得第一测试光强分布数据。通过这种方式,可以仅计算在量测线条的两个线段处的一维光强分布,而非整个量测线条处的光强分布,因此可以进一步减小后续用于模型训练的样本数据的大小,提高模型训练的速度,并且可以进一步提升根据本公开的一些实现方式的方法的运行速度,从而可以提高OPC的效率。
在一些实现方式中,第一线段和第二线段的长度取决于测试图形在量测位置处的图形的尺寸。通过这种方式,可以合适地设置第一线段和第二线段的长度,以减小用于后续模型训练的数据大小并减少确定光强分布时的运算量。
在一些实现方式中,目标光强分布数据指示在晶圆上预定位置处的光强分布,确定与目标图形对应的晶圆图形的尺寸包括:基于目标光强分布数据,利用显影模型来确定目标尺寸差异数据,目标尺寸差异数据指示目标图形与对应的晶圆图形在预定位置处的尺寸差异;以及使用目标图形在预定位置处的尺寸以及目标尺寸差异数据,来获取与目标图形对应的晶圆图形在预定位置处的尺寸。通过这种方式,可以基于光强分布利用显影模型来预测目标图形经曝光显影后在晶圆上实际获得的图形的尺寸,以用于指导后续 的修正掩模上的图形的过程。
根据本公开的第二方面,提供了一种用于确定晶圆图形尺寸的装置。该装置包括显影模型生成模块和晶圆图形尺寸确定模块。显影模型生成模块被配置为:至少基于针对至少一个测试图形的测试光强分布数据集和晶圆图形数据集,生成用于确定晶圆图形尺寸的显影模型,测试光强分布数据集中的每个测试光强分布数据指示光刻设备的光源经由相应的测试图形在晶圆上得到的光强分布,晶圆图形数据集中的每个晶圆图形数据指示与相应的测试图形对应的晶圆图形的尺寸。晶圆图形尺寸确定模块被配置为:至少基于针对目标图形的目标光强分布数据,利用显影模型来确定与目标图形对应的晶圆图形的尺寸,目标光强分布数据指示光源经由目标图形在晶圆上得到的光强分布。根据本公开的方案借助于经训练的模型结构来拟合难以用公式准确描述的显影过程,因此可以提高描述显影过程的显影模型的准确性。此外,由于在训练模型的过程中,使用光强分布作为训练样本,因此可以保证模型的物理性,使得经训练得到的模型不易出现过拟合,从而可以提高OPC模型的预测精确度。
在一些实现方式中,至少一个测试图形分别具有量测区域,测试光强分布数据集中的每个测试光强分布数据指示在晶圆的相应量测区域处的光强分布,晶圆图形数据集中的每个晶圆图形数据指示晶圆图形在相应量测区域处的尺寸,显影模型生成模块还被配置为:基于晶圆图形数据集和针对至少一个测试图形的测试图形尺寸数据集,确定针对至少一个测试图形的尺寸差异数据集,测试图形尺寸数据集中的每个测试图形尺寸数据指示相应的测试图形在对应的量测区域处的尺寸,尺寸差异数据集中的每个尺寸差异数据指示相应测试图形与对应的晶圆图形在相应量测区域处的尺寸差异;以及使用测试光强分布数据集和尺寸差异数据集对模型结构进行训练,以获得显影模型。通过基于尺寸差异来训练模型,可以在训练模型时减少测试图形本身的尺寸所带来的影响,从而可以提高模型的预测精确度。
在一些实现方式中,量测区域是在量测位置处的一维的量测线条,并且测试光强分布数据集中的每个测试光强分布数据指示在晶圆的相应量测线条处的一维光强分布。通过这种方式,可以仅计算在量测位置处的一维光强分布,而非整个图形的光强分布,因此可以大幅减小后续用于模型训练的样本数据的大小,提高模型训练的速度,并且可以大幅提升根据本公开的一些实现方式的方案的运行速度,从而可以提高OPC的效率。
在一些实现方式中,该装置还包括:测试光强分布数据集获取模块,被配置为:基于量测位置,使用针对光刻设备的光源数据和光瞳数据、以及针对测试掩模的测试掩模数据,来获取测试光强分布数据集,光源数据指示光源发出的光的能量和相位关于空间位置的分布,光瞳数据指示在各个空间位置处是否能够接收到光,测试掩模包括至少一个测试图形,测试掩模数据指示测试掩模的透光率关于空间位置的分布;以及目标光强分布数据获取模块,被配置为:使用光源数据、光瞳数据、以及针对目标掩模的目标掩模数据,来获取目标光强分布数据,目标掩模包括目标图形,目标掩模数据指示目标掩模的透光率关于空间位置的分布。通过这种方式来获取光强分布,可以保证所获得的测试光强分布数据集和目标光强分布数据的物理性,从而可以保证OPC模型的物理性,使得后续对于显影模型的训练不易出现过拟合现象,从而可以提高OPC模型的预测精确度。
在一些实现方式中,量测线条包括第一线段和第二线段,第一线段和第二线段彼此间隔开并且分别位于测试图形在量测位置处的图形的边缘,至少一个测试图形包括第一 测试图形,测试光强分布数据集包括针对第一测试图形的第一测试光强分布数据,测试光强分布数据集获取模块还被配置为:基于第一线段在第一测试图形中的位置,使用光源数据、光瞳数据、和测试掩模数据,来确定第一局部光强分布数据;基于第二线段在第一测试图形中的位置,使用光源数据、光瞳数据、和测试掩模数据,来确定第二局部光强分布数据;以及对第一局部光强分布数据和第二局部光强分布数据进行拼接,以获得第一测试光强分布数据。通过这种方式,可以仅计算在量测线条的两个线段处的一维光强分布,而非整个量测线条处的光强分布,因此可以进一步减小后续用于模型训练的样本数据的大小,提高模型训练的速度,并且可以进一步提升根据本公开的一些实现方式的方案的运行速度,从而可以提高OPC的效率。
在一些实现方式中,第一线段和第二线段的长度取决于测试图形在量测位置处的图形的尺寸。通过这种方式,可以合适地设置第一线段和第二线段的长度,以减小用于后续模型训练的数据大小并减少确定光强分布时的运算量。
在一些实现方式中,目标光强分布数据指示在晶圆上预定位置处的光强分布,晶圆图形尺寸确定模块还被配置为:基于目标光强分布数据,利用显影模型来确定目标尺寸差异数据,目标尺寸差异数据指示目标图形与对应的晶圆图形在预定位置处的尺寸差异;以及使用目标图形在预定位置处的尺寸以及目标尺寸差异数据,来获取与目标图形对应的晶圆图形在预定位置处的尺寸。通过这种方式,可以基于光强分布利用显影模型来预测目标图形经曝光显影后在晶圆上实际获得的图形的尺寸,以用于指导后续的修正掩模上的图形的过程。
根据本公开的第三方面,提供了一种电子设备。该电子设备包括:至少一个处理器;以及至少一个存储器,至少一个存储器被耦合到至少一个处理器,并且存储用于由至少一个处理器执行的指令,指令当由至少一个处理器执行时,使得电子设备执行根据本公开的第一方面的方法。根据本公开的电子设备借助于经训练的模型结构来拟合难以用公式准确描述的显影过程,因此可以提高描述显影过程的显影模型的准确性。此外,由于在训练模型的过程中,使用光强分布作为训练样本,因此可以保证模型的物理性,使得经训练得到的模型不易出现过拟合,从而可以提高OPC模型的预测精确度。
根据本公开的第四方面,提供了一种计算机可读存储介质。该计算机可读存储介质存储有计算机程序。该计算机程序被处理器执行时实现根据本公开的第一方面的方法。根据本公开的计算机可读存储介质借助于经训练的模型结构来拟合难以用公式准确描述的显影过程,因此可以提高描述显影过程的显影模型的准确性。此外,由于在训练模型的过程中,使用光强分布作为训练样本,因此可以保证模型的物理性,使得经训练得到的模型不易出现过拟合,从而可以提高OPC模型的预测精确度。
根据本公开的第五方面,提供了一种计算机程序产品。该计算机程序产品包括计算机可执行指令,计算机可执行指令在被处理器执行时,使计算机实现根据第一方面的方法。根据本公开的计算机程序产品借助于经训练的模型结构来拟合难以用公式准确描述的显影过程,因此可以提高描述显影过程的显影模型的准确性。此外,由于在训练模型的过程中,使用光强分布作为训练样本,因此可以保证模型的物理性,使得经训练得到的模型不易出现过拟合,从而可以提高OPC模型的预测精确度。
提供发明内容部分是为了简化的形式来介绍对概念的选择,它们在下文的具体实施方式中将被进一步描述。发明内容部分无意标识本公开内容的关键特征或主要特征,也 无意限制本公开内容的范围。
附图说明
通过参考附图阅读下文的详细描述,本公开的实施例的上述以及其它目的、特征和优点将变得易于理解。在附图中,以示例而非限制性的方式示出了本公开的若干实施例。
图1示出了根据本公开的一些实施例可以应用于其中的光刻设备的示意图;
图2示出了根据本公开的一些实施例的示例环境的框图;
图3示出了根据本公开的一些实施例的用于确定晶圆图形尺寸的方法的流程图;
图4示出了根据本公开的一些实施例的示例性的测试图形的示意图;
图5示出了根据本公开的一些实施例的用于获取测试光强分布数据集的过程的示意图;
图6示出了根据本公开的一些实施例的用于确定晶圆图形尺寸的示例装置的框图;以及
图7示出了可以用于实施根据本公开的一些实施例的示例设备的示意性框图。
具体实施方式
下面将参照附图更详细地描述本公开的优选实施例。虽然附图中显示了本公开的优选实施例,然而应该理解,可以以各种形式实现本公开而不应被这里阐述的实施例所限制。相反,提供这些实施例是为了使本公开更加透彻和完整,并且能够将本公开的范围完整地传达给本领域的技术人员。
在本文中使用的术语“包括”及其变形表示开放性包括,即“包括但不限于”。除非特别申明,术语“或”表示“和/或”。术语“基于”表示“至少部分地基于”。术语“一个示例实施例”和“一个实施例”表示“至少一个示例实施例”。术语“另一实施例”表示“至少一个另外的实施例”。术语“上”、“下”、“前”、“后”等指示放置或者位置关系的词汇均基于附图所示的方位或者位置关系,仅为了便于描述本公开的原理,而不是指示或者暗示所指的元件必须具有特定的方位、以特定的方位构造或操作,因此不能理解为对本公开的限制。
如上文所述,随着光刻技术发展到纳米级技术节点使得光刻过程受光学临近效应的影响越来越严重,借助于OPC来对掩模上的图形进行补偿修正已经成为保证图形保真度(Pattern Fidelity)的一种重要手段。
一种常规的OPC建模方案借助于公式来描述与曝光过程相对应的光学模型以及与显影过程相对应的显影模型,并且通过优化算法来对光学模型与显影模型进行优化,优化的目标是根据掩模上的图形借助于光学模型与显影模型对晶圆上的图形的预测结果接近经过曝光显影后实际获得的图形。然而,这种常规方案存在如下问题:针对先进技术节点,特别是负显影(Negative Tone Develop,NTD)工艺,由于显影过程中光刻胶的表现比较复杂,难以通过公式来准确地描述显影过程,因此导致显影模型预测表现不好,从而使得OPC模型的预测精确度下降。
本公开的各实施例提供了一种用于确定晶圆图形尺寸的方案。根据本公开的各实施例,通过利用针对测试图形的光强分布数据集和实际测得的晶圆图形数据集来训练模型以生成显影模型,并且借助于所生成的显影模型来预测晶圆图形的尺寸。
通过下文描述将会理解,由于根据本公开的各实施例的方案使用光强分布数据作为模型的训练样本,因此可以保证最终所获得的模型的物理性,而不易出现过拟合。与已知传统方案相比,根据本公开的各实施例的方案可以更准确地描述显影过程,从而提高OPC模型的预测精确度。
图1示出了根据本公开的一些实施例可以应用于其中的光刻设备100的示意图。光刻设备100可以包括光源110、照射装置120、用于承载掩模130的掩模台140、投影装置150、以及用于承载晶圆160的支撑台170。应当理解的是,光刻设备100还可以包括未示出的部件和/或可以省略所示出的部件,本公开的范围在此方面不受限制。
如图1所示,照射装置120接收从光源110发出的光束,并且可以调节光束,以使得光束在掩模130的平面处具有期望的空间分布和角强度分布。掩模130用于赋予入射光束图形化横截面。投影装置150将由掩模130赋予光束的图形化横截面投影到晶圆160的目标部分T上。图示MA示意性地示出了掩模130上的图形,并且图示WA示意性地示出了晶圆160上的图形。应当理解的是,在本公开的上下文中,术语“光”或“光束”可以涵盖所有类型的电磁辐射,例如紫外辐射。术语“掩模版”或“掩模”表示可以被用于赋予入射光束图形化横截面的任何合适的通用图形形成部件,该图形化横截面对应于要在晶圆160的目标部分T中被创建的图形。本公开的范围在上述方面不受限制。
图形保真度描述经曝光显影后在晶圆160上获得图形与掩模130上的图形的一致性,高的图形保真度有助于保证最终得到的芯片的功能满足预定规格。在实践中,通过借助于OPC模型对掩模130上的图形进行补偿修正,使其经过曝光显影后在晶圆160上得到图形与所期望的图形一致。OPC模型的预测精确度对修正结果起着决定性的影响,因此期望构建高精度的OPC模型。
图2示出了根据本公开的一些实施例的示例环境200的框图。如图2所示,示例环境200总体上可以包括电子设备240。在一些实施例中,电子设备240可以是诸如个人计算机、工作站、服务器等具有计算功能的设备。本公开的范围在此方面不受限制。
电子设备240可以接收针对至少一个测试图形的测试光强分布数据集210和晶圆图形数据集220作为输入。在本公开的上下文中,“测试图形”表示如下的图形,该图形在构建OPC模型过程中被曝光显影,以获得与该图形相对应的在晶圆160上最终实际得到的图形。在晶圆160上最终实际得到的图形也被称为“晶圆图形”。测试图形例如可以包括一系列一维图形或二维图形。在一些实施例中,测试光强分布数据集210中的测试光强分布数据指示光刻设备100的光源110经由相应的测试图形在晶圆160上得到的光强分布。在一些实施例中,晶圆图形数据集220中的晶圆图形数据指示与相应的测试图形对应的晶圆图形的尺寸,这例如可以借助于线宽扫描电子显微镜(Critical Dimension Scanning Electron Microscope,CD-SEM)量测在晶圆160上实际得到的图形的尺寸而被获得。
电子设备240还可以接收针对目标图形的目标光强分布数据230作为输入。目标光强分布数据230指示光源110经由目标图形在晶圆160上得到的光强分布。在本公开的上下文中,“目标图形”表示需要借助于OPC模型来被修正的图形。
在一些实施例中,测试光强分布数据集210、晶圆图形数据集220以及目标光强分布数据230可以由用户输入电子设备240。在一些实施例中,测试光强分布数据集210、晶圆图形数据集220以及目标光强分布数据230可以已经预先被存储在电子设备240中。 在一些实施例中,电子设备240可以通信地耦连到其它设备,以从其它设备获取测试光强分布数据集210、晶圆图形数据集220以及目标光强分布数据230。在一些实施例中,测试光强分布数据集210、晶圆图形数据集220以及目标光强分布数据230还可以由电子设备240预先确定。本公开的范围在此方面不受限制。
电子设备240可以基于测试光强分布数据集210和晶圆图形数据集220来生成显影模型,并借助于所生成的显影模型来基于目标光强分布数据230预测相应的晶圆图形的尺寸。这将在下文中结合图3至图5进一步详细描述。
图3示出了根据本公开的一些实施例的用于确定晶圆图形尺寸的方法300的流程图。在一些实施例中,方法300可以由如图2所示的电子设备240执行。应当理解的是,方法300还可以包括未示出的附加框和/或可以省略所示出的框,本公开的范围在此方面不受限制。
在框302,电子设备240获取针对至少一个测试图形的测试光强分布数据集210。测试光强分布数据集210中的每个测试光强分布数据指示光刻设备100的光源110经由相应的测试图形在晶圆160上得到的光强分布。在框304,电子设备240获取针对至少一个测试图形的晶圆图形数据集220,晶圆图形数据集220中的每个晶圆图形数据指示与相应的测试图形对应的晶圆图形的尺寸。在一些实施例中,电子设备240可以从外部设备接收预先确定的测试光强分布数据集210和实际测得的晶圆图形数据集220。
在一些实施例中,电子设备240可以使用针对光刻设备100的光源数据和光瞳数据、以及针对测试掩模的测试掩模数据来获取测试光强分布数据集210。光源数据可以指示光源110发出的光的能量和相位关于空间位置的分布。光瞳数据与光刻设备100的透镜组的配置相关联,并且可以指示在各个空间位置处是否能够接收到光。测试掩模可以包括至少一个测试图形,并且测试掩模数据可以指示测试掩模的透光率关于空间位置的分布。应当指出的是,一个测试掩模可以包括至少一个测试图形中的全部测试图形,或者可以仅包括至少一个测试图形中的一部分测试图形,本公开的范围在此方面不受限制。
电子设备240可以借助于光学成像理论中的霍普金斯公式(Hopkins Euqation),基于光源数据、光瞳数据和测试掩模数据来生成测试光强分布数据。下面的式子(1)示出了霍普金斯公式的一种形式:
Figure PCTCN2022072733-appb-000001
其中
Figure PCTCN2022072733-appb-000002
Figure PCTCN2022072733-appb-000003
分别表示光源函数J、光瞳函数P和掩模函数Q在频率域的傅里叶变换;
Figure PCTCN2022072733-appb-000004
Figure PCTCN2022072733-appb-000005
分别表示函数
Figure PCTCN2022072733-appb-000006
的复共轭和函数
Figure PCTCN2022072733-appb-000007
的复共轭;ξ和η是空间位置坐标;μ和ν是空间频率坐标;光源函数J描述光源110发出的光的能量和相位关于空间位置的分布;光瞳函数P描述在各个空间位置处是否能够接收到光;掩模函数Q描述掩模130的透光率关于空间位置的分布。关于霍普金斯公式可以进一步参考文章Hopkins,On the Diffraction Theory of Optical Images.Proc.Roy.Soc(Lond.)A217.1953,P408,该文章通过引用以其整体并入本文。
电子设备240可以使用所获取的光源数据、光瞳数据和测试掩模数据来获得上述式 子(1)中的光源函数J、光瞳函数P和掩模函数Q,以计算得到测试光强分布数据集210。通过借助于由上述式子(1)描述的物理模型来计算光强分布,可以保证所获得的测试光强分布数据集的物理性,从而可以保证OPC模型的物理性,使得后续对于显影模型的训练不易出现过拟合现象,从而可以提高OPC模型的预测精确度。
应当理解的是,电子设备240还可以以其它任何合适的方式来获取测试光强分布数据集210,本公开的范围在此方面不受限制。
在一些实施例中,至少一个测试图形分别具有量测区域。图4示出了根据本公开的一些实施例的示例性的测试图形400的示意图。如图4所示,测试图形400具有5个独立的多边形(polygon)410-1至410-5和位于量测位置的量测区域420。取决于掩模的透光性,多边形410-1至410-5可以是透光的或者不透光的。在一个示例中,在暗场(dark filed)的情况下,多边形410-1至410-5可以是透光的。在另一示例中,在亮场(bright filed)的情况下,多边形410-1至410-5可以是不透光的。本公开的范围在此方面不受限制。
在图4中,量测区域420被示出为在量测位置处的一维的量测线条。位于量测位置的量测区域420指示如下的区域,在测试图形400被显影曝光获得对应的晶圆图形之后,将在该量测位置处测量晶圆图形在该区域内的部分的尺寸。应当理解的是,测试图形400还可以具有其它任何合适的图形,并且量测区域420还可以具有其它任何合适的形状,本公开的范围在此方面不受限制。
在一些实施例中,电子设备240可以基于量测区域420的量测位置,使用光源数据、光瞳数据、以及测试掩模数据,来获取测试光强分布数据集210。在这种情况下,测试光强分布数据集210指示在对应的量测区域420处的光强分布,而非整个图形的光强分布。例如,电子设备240可以以与上文参考式子(1)描述类似的方式来计算在晶圆160的相应量测区域420处的光强分布,以获得测试光强分布数据集210。在这种情况下,晶圆图形数据可以仅指示晶圆图形在相应量测区域420处的尺寸,例如,可以利用CD-SEM仅量测相应量测区域420处的图形的尺寸。
特别地,在量测区域420是在量测位置处的一维的量测线条的情况下,测试光强分布数据可以指示在晶圆160的相应量测线条处的一维光强分布。在此,一维的量测线条是指在一个方向上延伸的线条。通过这种方式,可以仅计算在量测位置处的一维光强分布,即在一维的量测线条上的光强分布,而非在整个图形上的二维光强分布,因此可以大幅减小后续用于模型训练的样本数据的大小,提高模型训练的速度,并且可以大幅提升根据本公开的实施例的方法的运行速度,从而可以提高OPC的效率。
发明人经研究发现:光学邻近效应往往并不会影响掩模130上的各个图形的中央部分,而是主要对图形的边缘部分产生较大影响,使得图形边缘部分的尺寸或形状发生畸变。因此,在确定测试光强分布数据集210的过程中,可以仅关注在图形边缘部分处的光强分布,而不必再考虑在图形中央部分处的光强分布。
图5示出了根据本公开的一些实施例的用于获取测试光强分布数据集210的过程的示意图。图示500示出了图4中所示的测试图形400。在图示500中,量测线条(在图示500中未完整示出)包括第一线段520-1和第二线段520-2。第一线段520-1和第二线段520-2彼此间隔开并且分别位于测试图形400在量测位置处的图形的边缘。在一些实施例中,第一线段520-1的中心和第二线段520-2的中心可以分别位于测试图形400在量测位置处的图形的边缘。如图示500所示,第一线段520-1的中心位于多边形410-3的左 侧边缘处,并且第二线段520-2的中心位于多边形410-3的右侧边缘处。
在一些实施例中,第一线段520-1和第二线段520-2的长度取决于测试图形400在量测位置处的图形的尺寸,例如在图4和图示500中的X方向上的尺寸。示例性地,第一线段520-1和第二线段520-2的长度可以是测试图形400在量测位置处的图形的宽度的四分之一。应当理解的是,第一线段520-1和第二线段520-2的长度占测试图形400在量测位置处的图形的宽度的比例还可以是其它任何合适的值,本公开的范围在此方面不受限制。通过这种方式,可以合适地设置第一线段520-1和第二线段520-2的长度,以减小用于后续模型训练的数据大小并减少确定光强分布时的运算量。
在一些实施例中,以与上文参考式子(1)描述类似的方式,电子设备240可以基于第一线段520-1在测试图形400中的位置,使用光源数据、光瞳数据、和测试掩模数据来确定第一局部光强分布数据。该第一局部光强分布数据指示光刻设备100的光源110经由测试图形400在晶圆160上的与第一线段520-1对应的位置处的光强分布。图5中的图示502示例性地示出了针对第一线段520-1的第一局部光强分布数据,其中横坐标轴L指示位置,纵坐标轴I指示光强。
类似地,电子设备240可以基于第二线段520-2在测试图形400中的位置,使用光源数据、光瞳数据、和测试掩模数据,来确定第二局部光强分布数据,该第二局部光强分布数据指示光刻设备100的光源110经由测试图形400在晶圆160上的与第二线段520-2对应的位置处的光强分布。图5中的图示504示例性地示出了针对第二线段520-2的第二局部光强分布数据,其中横坐标轴L指示位置,纵坐标轴I指示光强。
电子设备240可以对第一局部光强分布数据和第二局部光强分布数据进行拼接,以获得针对测试图形400的第一测试光强分布数据。图5中的图示506示出了利用图示502所示的第一局部光强分布数据和图示504所示的第二局部光强分布数据拼接得到的第一测试光强分布数据,其中横坐标轴L指示位置,纵坐标轴I指示光强。通过这种方式,可以仅计算在量测线条的两个线段处的一维光强分布,而非整个量测线条处的光强分布,因此可以进一步减小后续用于模型训练的样本数据的大小,提高模型训练的速度,并且可以进一步提升根据本公开的实施例的方法的运行速度,从而可以提高OPC的效率。
返回到参考图3,在框306,电子设备240至少基于测试光强分布数据集210和晶圆图形数据集220,生成用于确定晶圆图形尺寸的显影模型。在一些实施例中,电子设备240可以使用测试光强分布数据集210作为样本并且直接将晶圆图形数据集220作为标签来训练模型结构,以获得显影模型。在本公开的上下文中,模型结构可以是诸如神经网络、深度学习等的模型结构,其中模型结构所包含的至少一部分参数需要利用数据集来训练,以获得最终可用的模型。
在一些实施例中,电子设备240还可以基于测试图形400与对应的晶圆图形在量测区域420处的尺寸差异来对模型结构进行训练,以获得显影模型。示例性地,电子设备240可以基于晶圆图形数据集220和针对至少一个测试图形的测试图形尺寸数据集,来确定针对至少一个测试图形的尺寸差异数据集。测试图形尺寸数据集中的每个测试图形尺寸数据指示相应的测试图形400在对应的量测区域420处的尺寸。尺寸差异数据集中的每个尺寸差异数据指示相应测试图形400与对应的晶圆图形在相应量测区域420处的尺寸差异。参考图4,电子设备240例如可以计算多边形410-3在量测区域420的宽度与其在晶圆160上的实际宽度之差来作为尺寸差异数据。电子设备240可以以类似的方式 计算针对每个测试图形的尺寸差异数据,以构建尺寸差异数据集。然后,电子设备240可以使用测试光强分布数据集210和尺寸差异数据集对模型结构进行训练,以获得显影模型。通过基于尺寸差异来训练模型,可以在训练模型时减少测试图形本身的尺寸所带来的影响,从而可以提高模型的预测精确度。
在一些实施例中,用于拟合测试光强分布数据集和尺寸差异数据集的模型结构例如可以是包括8个卷积层、4个池化层和2个全连接层的卷积神经网络,其中在每2个卷积层之后设置1个池化层,并且2个全连接层被设置在卷积神经网络的末尾。示例性地,电子设备240可以通过如下配置来训练该卷积网络网络:将测试光强分布数据集210作为样本,将尺寸差异数据集作为标签,并且使用均方差(Mean Square Error,MSE)损失函数和自适应矩估计(Adaptive Moment Estimation,ADAM)优化器。应当理解的是,模型结构还可以是其它任何合适的模型结构并且可以以其它任何合适的配置被训练,本公开的范围在此方面不受限制。
与已知的常规方案相比,在根据本公开的实施例中,由于借助于经训练的模型结构来拟合难以用公式准确描述的显影过程,因此可以提高描述显影过程的显影模型的准确性,从而可以提高OPC模型的预测精确度。此外,由于在训练模型的过程中,使用光强分布作为训练样本,因此可以保证模型的物理性,使得经训练得到的模型不易出现过拟合。
应当理解的是,电子设备240还可以以其它任何合适的方式来基于测试光强分布数据集210和晶圆图形数据集220而生成显影模型,本公开的范围在此方面不受限制。
在框308,电子设备240至少基于针对目标图形的目标光强分布数据230,利用显影模型来确定与目标图形对应的晶圆图形的尺寸。在一些实施例中,电子设备240可以以与上文参考式子(1)描述类似的方式,通过使用光源数据、光瞳数据、以及针对目标掩模130的目标掩模数据来获取目标光强分布数据230。目标掩模130包括目标图形,并且目标掩模数据指示目标掩模130的透光率关于空间位置的分布。本公开在此不再赘述。在一些实施例中,当期望预测晶圆160上预定位置处的图形的尺寸时,目标光强分布数据230可以仅指示在晶圆160上预定位置处的光强分布。
在一些实施例中,在电子设备240通过使用测试光强分布数据集210作为样本并且直接将晶圆图形数据集220作为标签来训练模型结构来获得显影模型的情况下,电子设备240可以直接将目标光强分布数据230输入显影模型,并且将显影模型的输出作为与目标图形对应的晶圆图形的尺寸。
在一些实施例中,在电子设备240通过使用测试光强分布数据集210作为样本并且将上述尺寸差异数据集作为标签来训练模型结构来获得显影模型的情况下,电子设备240可以基于目标光强分布数据230,利用显影模型来确定目标尺寸差异数据,该目标尺寸差异数据指示目标图形与对应的晶圆图形在预定位置处的尺寸差异。例如,电子设备240可以将指示在晶圆160上预定位置处的光强分布的目标光强分布数据230输入显影模型,以获得目标尺寸差异数据。然后,电子设备240可以使用目标图形在预定位置处的尺寸以及目标尺寸差异数据来获取与目标图形对应的晶圆图形在预定位置处的尺寸。例如,电子设备240可以将目标图形在预定位置处的尺寸与目标尺寸差异数据相加,来预测晶圆图形在预定位置处的尺寸。
通过这种方式,电子设备240可以基于光强分布利用显影模型来预测目标图形经曝 光显影后在晶圆160上实际获得的图形的尺寸,以用于指导后续的修正目标图形的过程。
通过以上结合图2至图5的描述可以看到,根据本公开的各实施例的用于确定晶圆图形尺寸的方案,通过利用针对测试图形的光强分布数据集和实际测得的晶圆图形数据集来训练模型以生成显影模型,并且借助于所生成的显影模型来预测晶圆图形的尺寸。与已知的常规方案相比,在根据本公开的实施例中,由于借助于经训练的模型结构来拟合难以用公式准确描述的显影过程,因此可以提高描述显影过程的显影模型的准确性。此外,由于在训练模型的过程中,使用光强分布作为训练样本,因此可以保证模型的物理性,使得经训练得到的模型不易出现过拟合,从而可以提高OPC模型的预测精确度。
在上文中已经参考图2至图5详细描述了根据本公开的各实施例的方法的示例实现,在下文中将描述相应的装置的实现。
图6示出了根据本公开的一些实施例的用于确定晶圆图形尺寸的示例装置600的框图。该装置600例如可以用于实现如图2中所示的电子设备240。如图6所示,装置600可以包括显影模型生成模块602,该显影模型生成模块602被配置为:至少基于针对至少一个测试图形的测试光强分布数据集和晶圆图形数据集,生成用于确定晶圆图形尺寸的显影模型,测试光强分布数据集中的每个测试光强分布数据指示光刻设备的光源经由相应的测试图形在晶圆上得到的光强分布,晶圆图形数据集中的每个晶圆图形数据指示与相应的测试图形对应的晶圆图形的尺寸。此外,装置600还可以包括晶圆图形尺寸确定模块604,该晶圆图形尺寸确定模块604被配置为:至少基于针对目标图形的目标光强分布数据,利用显影模型来确定与目标图形对应的晶圆图形的尺寸,目标光强分布数据指示光源经由目标图形在晶圆上得到的光强分布。
在一些实施例中,至少一个测试图形分别具有量测区域,测试光强分布数据集中的每个测试光强分布数据指示在晶圆的相应量测区域处的光强分布,晶圆图形数据集中的每个晶圆图形数据指示晶圆图形在相应量测区域处的尺寸。显影模型生成模块602还被配置为:基于晶圆图形数据集和针对至少一个测试图形的测试图形尺寸数据集,确定针对至少一个测试图形的尺寸差异数据集,测试图形尺寸数据集中的每个测试图形尺寸数据指示相应的测试图形在对应的量测区域处的尺寸,尺寸差异数据集中的每个尺寸差异数据指示相应测试图形与对应的晶圆图形在相应量测区域处的尺寸差异;以及使用测试光强分布数据集和尺寸差异数据集对模型结构进行训练,以获得显影模型。
在一些实施例中,量测区域是在量测位置处的一维的量测线条,并且测试光强分布数据集中的每个测试光强分布数据指示在晶圆的相应量测线条处的一维光强分布。
在一些实施例中,装置600还包括:测试光强分布数据集获取模块,被配置为:基于量测位置,使用针对光刻设备的光源数据和光瞳数据、以及针对测试掩模的测试掩模数据,来获取测试光强分布数据集,光源数据指示光源发出的光的能量和相位关于空间位置的分布,光瞳数据指示在各个空间位置处是否能够接收到光,测试掩模包括至少一个测试图形,测试掩模数据指示测试掩模的透光率关于空间位置的分布;以及目标光强分布数据获取模块,被配置为:使用光源数据、光瞳数据、以及针对目标掩模的目标掩模数据,来获取目标光强分布数据,目标掩模包括目标图形,目标掩模数据指示目标掩模的透光率关于空间位置的分布。
在一些实施例中,量测线条包括第一线段和第二线段,第一线段和第二线段彼此间隔开并且分别位于测试图形在量测位置处的图形的边缘,至少一个测试图形包括第一测 试图形,测试光强分布数据集包括针对第一测试图形的第一测试光强分布数据,测试光强分布数据集获取模块还被配置为:基于第一线段在第一测试图形中的位置,使用光源数据、光瞳数据、和测试掩模数据,来确定第一局部光强分布数据;基于第二线段在第一测试图形中的位置,使用光源数据、光瞳数据、和测试掩模数据,来确定第二局部光强分布数据;以及对第一局部光强分布数据和第二局部光强分布数据进行拼接,以获得第一测试光强分布数据。
在一些实施例中,第一线段和第二线段的长度取决于测试图形在量测位置处的图形的尺寸。
在一些实施例中,目标光强分布数据指示在晶圆上预定位置处的光强分布,晶圆图形尺寸确定模块604还被配置为:基于目标光强分布数据,利用显影模型来确定目标尺寸差异数据,目标尺寸差异数据指示目标图形与对应的晶圆图形在预定位置处的尺寸差异;使用目标图形在预定位置处的尺寸以及目标尺寸差异数据,来获取与目标图形对应的晶圆图形在预定位置处的尺寸。
装置600中所包括的模块和/或单元可以利用各种方式来实现,包括软件、硬件、固件或其任意组合。在一些实施例中,一个或多个单元可以使用软件和/或固件来实现,例如存储在存储介质上的机器可执行指令。除了机器可执行指令之外或者作为替代,装置600中的部分或者全部单元可以至少部分地由一个或多个硬件逻辑组件来实现。作为示例而非限制,可以使用的示范类型的硬件逻辑组件包括现场可编程门阵列(FPGA)、专用集成电路(ASIC)、专用标准品(ASSP)、片上系统(SOC)、复杂可编程逻辑器件(CPLD),等等。
图6中所示的这些模块和/或单元可以部分或者全部地实现为硬件模块、软件模块、固件模块或者其任意组合。特别地,在某些实施例中,上文描述的流程、方法或过程可以由存储系统或与存储系统对应的主机或独立于存储系统的其它计算设备中的硬件来实现。
图7示出了可以用于实施根据本公开的一些实施例的示例设备700的示意性框图。设备700可以用于实现实现如图2中所示的电子设备230。如图7所示,设备700包括中央处理单元(CPU)701,其可以根据存储在只读存储器(ROM)702中的计算机程序指令或者从存储单元708加载到随机访问存储器(RAM)703中的计算机程序指令,来执行各种适当的动作和处理。在RAM 703中,还可存储设备700操作所需的各种程序和数据。CPU 701、ROM 702以及RAM 703通过总线704彼此相连。输入/输出(I/O)接口705也连接至总线704。
设备700中的多个部件连接至I/O接口705,包括:输入单元706,例如键盘、鼠标等;输出单元707,例如各种类型的显示器、扬声器等;存储单元708,例如磁盘、光盘等;以及通信单元709,例如网卡、调制解调器、无线通信收发机等。通信单元709允许设备700通过诸如因特网的计算机网络和/或各种电信网络与其它设备交换信息/数据。
处理单元701执行上文所描述的各个方法和处理,例如方法300。例如,在一些实施例中,方法300可被实现为计算机软件程序,其被有形地包含于机器可读介质,例如存储单元708。在一些实施例中,计算机程序的部分或者全部可以经由ROM 702和/或通信单元709而被载入和/或安装到设备700上。当计算机程序加载到RAM 703并由CPU 701执行时,可以执行上文描述的方法300的一个或多个步骤。备选地,在其它实施例 中,CPU 701可以通过其它任何适当的方式(例如,借助于固件)而被配置为执行方法300。
本文中以上描述的功能可以至少部分地由一个或多个硬件逻辑部件来执行。例如,非限制性地,可以使用的示范类型的硬件逻辑部件包括:场可编程门阵列(FPGA)、专用集成电路(ASIC)、专用标准产品(ASSP)、芯片上系统的系统(SOC)、负载可编程逻辑设备(CPLD)等等。
用于实施本公开的方法的程序代码可以采用一个或多个编程语言的任何组合来编写。这些程序代码可以提供给通用计算机、专用计算机或其它可编程数据处理装置的处理器或控制器,使得程序代码当由处理器或控制器执行时使流程图和/或框图中所规定的功能/操作被实施。程序代码可以完全在机器上执行、部分地在机器上执行,作为独立软件包部分地在机器上执行且部分地在远程机器上执行或完全在远程机器或服务器上执行。
在本公开的上下文中,机器可读介质可以是有形的介质,其可以包含或存储以供指令执行系统、装置或设备使用或与指令执行系统、装置或设备结合地使用的程序。机器可读介质可以是机器可读信号介质或机器可读储存介质。机器可读介质可以包括但不限于电子的、磁性的、光学的、电磁的、红外的、或半导体系统、装置或设备,或者上述内容的任何合适组合。机器可读存储介质的更具体示例会包括基于一个或多个线的电气连接、便携式计算机盘、硬盘、随机存取存储器(RAM)、只读存储器(ROM)、可擦除可编程只读存储器(EPROM或快闪存储器)、光纤、便捷式紧凑盘只读存储器(CD-ROM)、光学储存设备、磁储存设备、或上述内容的任何合适组合。
此外,虽然采用特定次序描绘了各操作,但是这应当理解为要求这样操作以所示出的特定次序或以顺序次序执行,或者要求所有图示的操作应被执行以取得期望的结果。在一定环境下,多任务和并行处理可能是有利的。同样地,虽然在上面论述中包含了若干具体实现细节,但是这些不应当被解释为对本公开的范围的限制。在单独的实施例的上下文中描述的某些特征还可以组合地实现在单个实现中。相反地,在单个实现的上下文中描述的各种特征也可以单独地或以任何合适的子组合的方式实现在多个实现中。
尽管已经采用特定于结构特征和/或方法逻辑动作的语言描述了本主题,但是应当理解所附权利要求书中所限定的主题未必局限于上面描述的特定特征或动作。相反,上面所描述的特定特征和动作仅仅是实现权利要求书的示例形式。

Claims (17)

  1. 一种用于确定晶圆图形尺寸的方法,其特征在于,所述方法包括:
    至少基于针对至少一个测试图形的测试光强分布数据集和晶圆图形数据集,生成用于确定晶圆图形尺寸的显影模型,所述测试光强分布数据集中的每个测试光强分布数据指示光刻设备的光源经由相应的测试图形在晶圆上得到的光强分布,所述晶圆图形数据集中的每个晶圆图形数据指示与相应的测试图形对应的晶圆图形的尺寸;以及
    至少基于针对目标图形的目标光强分布数据,利用所述显影模型来确定与所述目标图形对应的晶圆图形的尺寸,所述目标光强分布数据指示所述光源经由所述目标图形在晶圆上得到的光强分布。
  2. 根据权利要求1所述的方法,其特征在于,所述至少一个测试图形分别具有量测区域,所述测试光强分布数据集中的每个测试光强分布数据指示在晶圆的相应量测区域处的光强分布,所述晶圆图形数据集中的每个晶圆图形数据指示晶圆图形在相应量测区域处的尺寸,生成所述显影模型包括:
    基于所述晶圆图形数据集和针对所述至少一个测试图形的测试图形尺寸数据集,确定针对所述至少一个测试图形的尺寸差异数据集,所述测试图形尺寸数据集中的每个测试图形尺寸数据指示相应的测试图形在对应的量测区域处的尺寸,所述尺寸差异数据集中的每个尺寸差异数据指示相应测试图形与对应的晶圆图形在相应量测区域处的尺寸差异;以及
    使用所述测试光强分布数据集和所述尺寸差异数据集对模型结构进行训练,以获得所述显影模型。
  3. 根据权利要求2所述的方法,其特征在于,所述量测区域是在量测位置处的一维的量测线条,并且所述测试光强分布数据集中的每个测试光强分布数据指示在晶圆的相应量测线条处的一维光强分布。
  4. 根据权利要求3所述的方法,其特征在于,所述方法还包括:
    基于所述量测位置,使用针对所述光刻设备的光源数据和光瞳数据、以及针对测试掩模的测试掩模数据,来获取所述测试光强分布数据集,所述光源数据指示所述光源发出的光的能量和相位关于空间位置的分布,所述光瞳数据指示在各个空间位置处是否能够接收到光,所述测试掩模包括所述至少一个测试图形,所述测试掩模数据指示所述测试掩模的透光率关于空间位置的分布;以及
    使用所述光源数据、所述光瞳数据、以及针对目标掩模的目标掩模数据,来获取所述目标光强分布数据,所述目标掩模包括所述目标图形,所述目标掩模数据指示所述目标掩模的透光率关于空间位置的分布。
  5. 根据权利要求4所述的方法,其特征在于,所述量测线条包括第一线段和第二线段,所述第一线段和所述第二线段彼此间隔开并且分别位于所述测试图形在所述量测位置处的图形的边缘,所述至少一个测试图形包括第一测试图形,所述测试光强分布数据集包括针对所述第一测试图形的第一测试光强分布数据,获取所述测试光强分布数据集包括:
    基于所述第一线段在所述第一测试图形中的位置,使用所述光源数据、所述光瞳数据、和所述测试掩模数据,来确定第一局部光强分布数据;
    基于所述第二线段在所述第一测试图形中的位置,使用所述光源数据、所述光瞳数据、和所述测试掩模数据,来确定第二局部光强分布数据;以及
    对所述第一局部光强分布数据和所述第二局部光强分布数据进行拼接,以获得所述第一测试光强分布数据。
  6. 根据权利要求5所述的方法,其特征在于,所述第一线段和所述第二线段的长度取决于所述测试图形在所述量测位置处的所述图形的尺寸。
  7. 根据权利要求2至6中任一项所述的方法,其特征在于,所述目标光强分布数据指示在晶圆上预定位置处的光强分布,确定与所述目标图形对应的晶圆图形的尺寸包括:
    基于所述目标光强分布数据,利用所述显影模型来确定目标尺寸差异数据,所述目标尺寸差异数据指示目标图形与对应的晶圆图形在所述预定位置处的尺寸差异;以及
    使用所述目标图形在所述预定位置处的尺寸以及所述目标尺寸差异数据,来获取与所述目标图形对应的晶圆图形在所述预定位置处的尺寸。
  8. 一种用于确定晶圆图形尺寸的装置,其特征在于,所述装置包括:
    显影模型生成模块,被配置为:至少基于针对至少一个测试图形的测试光强分布数据集和晶圆图形数据集,生成用于确定晶圆图形尺寸的显影模型,所述测试光强分布数据集中的每个测试光强分布数据指示光刻设备的光源经由相应的测试图形在晶圆上得到的光强分布,所述晶圆图形数据集中的每个晶圆图形数据指示与相应的测试图形对应的晶圆图形的尺寸;以及
    晶圆图形尺寸确定模块,被配置为:至少基于针对目标图形的目标光强分布数据,利用所述显影模型来确定与所述目标图形对应的晶圆图形的尺寸,所述目标光强分布数据指示所述光源经由所述目标图形在晶圆上得到的光强分布。
  9. 根据权利要求8所述的装置,其特征在于,所述至少一个测试图形分别具有量测区域,所述测试光强分布数据集中的每个测试光强分布数据指示在晶圆的相应量测区域处的光强分布,所述晶圆图形数据集中的每个晶圆图形数据指示晶圆图形在相应量测区域处的尺寸,所述显影模型生成模块还被配置为:
    基于所述晶圆图形数据集和针对所述至少一个测试图形的测试图形尺寸数据集,确定针对所述至少一个测试图形的尺寸差异数据集,所述测试图形尺寸数据集中的每个测试图形尺寸数据指示相应的测试图形在对应的量测区域处的尺寸,所述尺寸差异数据集中的每个尺寸差异数据指示相应测试图形与对应的晶圆图形在相应量测区域处的尺寸差异;以及
    使用所述测试光强分布数据集和所述尺寸差异数据集对模型结构进行训练,以获得所述显影模型。
  10. 根据权利要求9所述的装置,其特征在于,所述量测区域是在量测位置处的一维的量测线条,并且所述测试光强分布数据集中的每个测试光强分布数据指示在晶圆的相应量测线条处的一维光强分布。
  11. 根据权利要求10所述的装置,其特征在于,所述装置还包括:
    测试光强分布数据集获取模块,被配置为:基于所述量测位置,使用针对所述光刻设备的光源数据和光瞳数据、以及针对测试掩模的测试掩模数据,来获取所述测试光强分布数据集,所述光源数据指示所述光源发出的光的能量和相位关于空间位置的分布,所述光瞳数据指示在各个空间位置处是否能够接收到光,所述测试掩模包括所述至少一个测试图形,所述测试掩模数据指示所述测试掩模的透光率关于空间位置的分布;以及
    目标光强分布数据获取模块,被配置为:使用所述光源数据、所述光瞳数据、以及针对目标掩模的目标掩模数据,来获取所述目标光强分布数据,所述目标掩模包括所述目标图形, 所述目标掩模数据指示所述目标掩模的透光率关于空间位置的分布。
  12. 根据权利要求11所述的装置,其特征在于,所述量测线条包括第一线段和第二线段,所述第一线段和所述第二线段彼此间隔开并且分别位于所述测试图形在所述量测位置处的图形的边缘,所述至少一个测试图形包括第一测试图形,所述测试光强分布数据集包括针对所述第一测试图形的第一测试光强分布数据,所述测试光强分布数据集获取模块还被配置为:
    基于所述第一线段在所述第一测试图形中的位置,使用所述光源数据、所述光瞳数据、和所述测试掩模数据,来确定第一局部光强分布数据;
    基于所述第二线段在所述第一测试图形中的位置,使用所述光源数据、所述光瞳数据、和所述测试掩模数据,来确定第二局部光强分布数据;以及
    对所述第一局部光强分布数据和所述第二局部光强分布数据进行拼接,以获得所述第一测试光强分布数据。
  13. 根据权利要求12所述的装置,其特征在于,所述第一线段和所述第二线段的长度取决于所述测试图形在所述量测位置处的所述图形的尺寸。
  14. 根据权利要求9至13中任一项所述的装置,其特征在于,所述目标光强分布数据指示在晶圆上预定位置处的光强分布,所述晶圆图形尺寸确定模块还被配置为:
    基于所述目标光强分布数据,利用所述显影模型来确定目标尺寸差异数据,所述目标尺寸差异数据指示目标图形与对应的晶圆图形在所述预定位置处的尺寸差异;以及
    使用所述目标图形在所述预定位置处的尺寸以及所述目标尺寸差异数据,来获取与所述目标图形对应的晶圆图形在所述预定位置处的尺寸。
  15. 一种电子设备,其特征在于,包括:
    至少一个处理器;以及
    至少一个存储器,所述至少一个存储器被耦合到所述至少一个处理器,并且存储用于由所述至少一个处理器执行的指令,所述指令在由所述至少一个处理器执行时,使所述电子设备执行根据权利要求1至7中任一项所述的方法。
  16. 一种计算机可读存储介质,其特征在于,所述计算机可读存储介质存储有计算机程序,所述计算机程序在被处理器执行时实现根据权利要求1至7中任一项所述的方法。
  17. 一种计算机程序产品,其特征在于,所述计算机程序产品包括计算机可执行指令,所述计算机可执行指令在被处理器执行时,使计算机实现根据权利要求1至7中任一项所述的方法。
PCT/CN2022/072733 2022-01-19 2022-01-19 用于确定晶圆图形尺寸的方法、装置、设备、介质以及程序产品 WO2023137622A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN202280006302.2A CN116802556A (zh) 2022-01-19 2022-01-19 用于确定晶圆图形尺寸的方法、装置、设备、介质以及程序产品
PCT/CN2022/072733 WO2023137622A1 (zh) 2022-01-19 2022-01-19 用于确定晶圆图形尺寸的方法、装置、设备、介质以及程序产品

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2022/072733 WO2023137622A1 (zh) 2022-01-19 2022-01-19 用于确定晶圆图形尺寸的方法、装置、设备、介质以及程序产品

Publications (1)

Publication Number Publication Date
WO2023137622A1 true WO2023137622A1 (zh) 2023-07-27

Family

ID=87347639

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/072733 WO2023137622A1 (zh) 2022-01-19 2022-01-19 用于确定晶圆图形尺寸的方法、装置、设备、介质以及程序产品

Country Status (2)

Country Link
CN (1) CN116802556A (zh)
WO (1) WO2023137622A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117311080A (zh) * 2023-11-24 2023-12-29 全芯智造技术有限公司 用于拆分版图图案的方法、设备和介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020078430A1 (en) * 2000-12-07 2002-06-20 Ko Miyazaki Manufacturing method of semiconductor device
CN104570586A (zh) * 2013-10-23 2015-04-29 中芯国际集成电路制造(北京)有限公司 光学邻近修正模型的获取方法
CN105223773A (zh) * 2014-05-30 2016-01-06 中芯国际集成电路制造(上海)有限公司 一种光掩模图案透光强度修正的方法
CN107908071A (zh) * 2017-11-28 2018-04-13 上海集成电路研发中心有限公司 一种基于神经网络模型的光学邻近校正方法
CN109283798A (zh) * 2018-08-15 2019-01-29 上海华力集成电路制造有限公司 光阻顶部缺失的优化方法
CN110221513A (zh) * 2018-03-02 2019-09-10 中芯国际集成电路制造(上海)有限公司 光学邻近校正方法及优化光学邻近校正模型的方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020078430A1 (en) * 2000-12-07 2002-06-20 Ko Miyazaki Manufacturing method of semiconductor device
CN104570586A (zh) * 2013-10-23 2015-04-29 中芯国际集成电路制造(北京)有限公司 光学邻近修正模型的获取方法
CN105223773A (zh) * 2014-05-30 2016-01-06 中芯国际集成电路制造(上海)有限公司 一种光掩模图案透光强度修正的方法
CN107908071A (zh) * 2017-11-28 2018-04-13 上海集成电路研发中心有限公司 一种基于神经网络模型的光学邻近校正方法
CN110221513A (zh) * 2018-03-02 2019-09-10 中芯国际集成电路制造(上海)有限公司 光学邻近校正方法及优化光学邻近校正模型的方法
CN109283798A (zh) * 2018-08-15 2019-01-29 上海华力集成电路制造有限公司 光阻顶部缺失的优化方法

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117311080A (zh) * 2023-11-24 2023-12-29 全芯智造技术有限公司 用于拆分版图图案的方法、设备和介质
CN117311080B (zh) * 2023-11-24 2024-02-27 全芯智造技术有限公司 用于拆分版图图案的方法、设备和介质

Also Published As

Publication number Publication date
CN116802556A (zh) 2023-09-22

Similar Documents

Publication Publication Date Title
TWI696125B (zh) 用於計算微影之機器學習模型的訓練方法
TWI754308B (zh) 機器學習輔助光學接近誤差校正的訓練方法
US11119401B2 (en) Simulating near field image in optical lithography
JP6055436B2 (ja) 高速自由形式ソース・マスク同時最適化方法
US9091935B2 (en) Multistage extreme ultra-violet mask qualification
US8356261B1 (en) Determining the gradient and hessian of the image log slope for design rule optimization for accelerating source mask optimization (SMO)
US7921385B2 (en) Mask-pattern determination using topology types
US10025175B2 (en) Method and system to prepare, manufacture and inspect mask patterns for a semiconductor device
US8473878B2 (en) Lithographically enhanced edge determination
CN102053504A (zh) 基于衍射标记分析的对设计布局中的优化图案的选择
US11675958B2 (en) Lithography simulation method
KR102262427B1 (ko) 확률적 변동을 추산하는 모델
JP2010118655A (ja) 高速感度モデル計算のためのデルタtcc
JP2021505973A (ja) パターニングプロセスについての情報を決定する方法、測定データにおける誤差を低減する方法、メトロロジプロセスを較正する方法、メトロロジターゲットを選択する方法
Li et al. Robust pixel-based source and mask optimization for inverse lithography
US10310372B1 (en) Full-chip hierarchical inverse lithography
TW202001406A (zh) 光罩之製作方法
WO2023137622A1 (zh) 用于确定晶圆图形尺寸的方法、装置、设备、介质以及程序产品
TWI791216B (zh) 判定用於基板上之圖案的部分之度量
TWI787849B (zh) 用於判定與運算微影光罩模型相關聯之電磁場的方法
CN113219784A (zh) 获取掩膜版最优物平面方法和光学邻近修正方法及掩膜版
US20240184219A1 (en) System and method to ensure parameter measurement matching across metrology tools
EP4089484A1 (en) System and method to ensure parameter measurement matching across metrology tools
US20230066219A1 (en) Methods of preparing photo mask data and manufacturing a photo mask
US10429746B2 (en) Estimation of data in metrology

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 202280006302.2

Country of ref document: CN

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22921070

Country of ref document: EP

Kind code of ref document: A1