CN111443569A - Method and device for establishing correction model and method and device for optimizing mask - Google Patents

Method and device for establishing correction model and method and device for optimizing mask Download PDF

Info

Publication number
CN111443569A
CN111443569A CN202010419225.6A CN202010419225A CN111443569A CN 111443569 A CN111443569 A CN 111443569A CN 202010419225 A CN202010419225 A CN 202010419225A CN 111443569 A CN111443569 A CN 111443569A
Authority
CN
China
Prior art keywords
pattern
historical
mask
exposure
mask pattern
Prior art date
Legal status (The legal status 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 status listed.)
Granted
Application number
CN202010419225.6A
Other languages
Chinese (zh)
Other versions
CN111443569B (en
Inventor
马乐
韦亚一
张利斌
陈睿
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Institute of Microelectronics of CAS
Original Assignee
Institute of Microelectronics of CAS
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 Institute of Microelectronics of CAS filed Critical Institute of Microelectronics of CAS
Priority to CN202010419225.6A priority Critical patent/CN111443569B/en
Publication of CN111443569A publication Critical patent/CN111443569A/en
Application granted granted Critical
Publication of CN111443569B publication Critical patent/CN111443569B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G03—PHOTOGRAPHY; CINEMATOGRAPHY; ANALOGOUS TECHNIQUES USING WAVES OTHER THAN OPTICAL WAVES; ELECTROGRAPHY; HOLOGRAPHY
    • G03F—PHOTOMECHANICAL PRODUCTION OF TEXTURED OR PATTERNED SURFACES, e.g. FOR PRINTING, FOR PROCESSING OF SEMICONDUCTOR DEVICES; MATERIALS THEREFOR; ORIGINALS THEREFOR; APPARATUS SPECIALLY ADAPTED THEREFOR
    • G03F1/00—Originals 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
    • G03F1/36—Masks having proximity correction features; Preparation thereof, e.g. optical proximity correction [OPC] design processes
    • G—PHYSICS
    • G03—PHOTOGRAPHY; CINEMATOGRAPHY; ANALOGOUS TECHNIQUES USING WAVES OTHER THAN OPTICAL WAVES; ELECTROGRAPHY; HOLOGRAPHY
    • G03F—PHOTOMECHANICAL PRODUCTION OF TEXTURED OR PATTERNED SURFACES, e.g. FOR PRINTING, FOR PROCESSING OF SEMICONDUCTOR DEVICES; MATERIALS THEREFOR; ORIGINALS THEREFOR; APPARATUS SPECIALLY ADAPTED THEREFOR
    • G03F7/00—Photomechanical, e.g. photolithographic, production of textured or patterned surfaces, e.g. printing surfaces; Materials therefor, e.g. comprising photoresists; Apparatus specially adapted therefor
    • G03F7/70—Microphotolithographic exposure; Apparatus therefor
    • G03F7/70425—Imaging strategies, e.g. for increasing throughput or resolution, printing product fields larger than the image field or compensating lithography- or non-lithography errors, e.g. proximity correction, mix-and-match, stitching or double patterning
    • G—PHYSICS
    • G03—PHOTOGRAPHY; CINEMATOGRAPHY; ANALOGOUS TECHNIQUES USING WAVES OTHER THAN OPTICAL WAVES; ELECTROGRAPHY; HOLOGRAPHY
    • G03F—PHOTOMECHANICAL PRODUCTION OF TEXTURED OR PATTERNED SURFACES, e.g. FOR PRINTING, FOR PROCESSING OF SEMICONDUCTOR DEVICES; MATERIALS THEREFOR; ORIGINALS THEREFOR; APPARATUS SPECIALLY ADAPTED THEREFOR
    • G03F7/00—Photomechanical, e.g. photolithographic, production of textured or patterned surfaces, e.g. printing surfaces; Materials therefor, e.g. comprising photoresists; Apparatus specially adapted therefor
    • G03F7/70—Microphotolithographic exposure; Apparatus therefor
    • G03F7/70425—Imaging strategies, e.g. for increasing throughput or resolution, printing product fields larger than the image field or compensating lithography- or non-lithography errors, e.g. proximity correction, mix-and-match, stitching or double patterning
    • G03F7/70433—Layout for increasing efficiency or for compensating imaging errors, e.g. layout of exposure fields for reducing focus errors; Use of mask features for increasing efficiency or for compensating imaging errors

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Preparing Plates And Mask In Photomechanical Process (AREA)

Abstract

The embodiment of the application discloses a method and a device for establishing a correction model and a method and a device for optimizing a mask pattern, wherein the correction model is established in advance based on a pattern parameter of a historical mask pattern, an actual parameter of the historical exposure pattern obtained by exposure of the historical mask pattern and a historical weight corresponding to the historical mask pattern, the historical weight is determined based on the pattern parameter of the historical mask pattern and/or the actual parameter of the historical exposure pattern and an initial weight of the historical mask pattern, after the mask pattern to be corrected is obtained, a prediction parameter of the exposure pattern to be corrected corresponding to the mask pattern to be corrected can be obtained by using the correction model, and the mask pattern to be corrected is corrected based on the prediction parameter and a target parameter of the exposure pattern to be corrected so as to reduce a difference value between the prediction parameter and the target parameter. The historical weight determined in the way is pertinent, so that the established correction model is more accurate, and the correction of the mask pattern to be corrected is more pertinent.

Description

Method and device for establishing correction model and method and device for optimizing mask
Technical Field
The present application relates to the field of semiconductors, and in particular, to a method and an apparatus for building a correction model, and a method and an apparatus for optimizing a mask.
Background
In the semiconductor field, photolithography is an important process in integrated circuit production, and specifically, a mask pattern on a reticle can be transferred to a photoresist layer in a certain ratio by exposure, and further transferred from the photoresist layer to an object to be processed. The pattern on the mask is determined according to the actually required pattern. However, in practical operation, as the pattern size on the wafer is smaller and smaller, the diffraction effect is more and more obvious, and in addition to other factors such as aberration, the exposure pattern obtained after exposure often deviates from the pre-designed size, for example, a line width of 200 nm is desired to be obtained by designing the mask pattern, and the line width of the exposure pattern obtained actually is 192.9 nm, that is, there is an error of 7.1 nm.
Therefore, it is necessary to correct the designed mask pattern so that the exposure pattern obtained by the corrected mask pattern approaches the size designed in advance, thereby enhancing the fidelity of the pattern. Generally, optical proximity correction can be performed on a mask pattern when the line width is smaller than the exposure wavelength, for example, for a 248 nm lithography machine, a simple correction is required when the line width of the pattern is smaller than 250 nm, and a very complex correction is required when the line width is smaller than 180 nm. Referring to fig. 1, a schematic diagram of an exposure pattern provided in an embodiment of the present application is shown, where fig. 1(a) is an exposure pattern obtained by exposing a mask pattern before correction, and fig. 1(b) is an exposure pattern obtained by exposing a mask pattern after correction, and it can be seen from the diagram that line widths of lines in the exposure pattern obtained by exposing the mask pattern before correction are not uniform, and there are some positions where there are problems of breaking (Necking) and Bridging (Bridging), etc., while these problems do not exist in the exposure pattern obtained by exposing the mask pattern after correction, and the sizes of the lines are more uniform.
In more advanced technology nodes, such as 90 nm and below, the mask pattern is usually corrected by using a model-based optical proximity correction (MB-OPC) method. The model is generally established according to a large amount of actual exposure data, and the quality of the correction effect is highly dependent on the accuracy of the model. However, the accuracy of this correction method in practical application needs to be improved.
Disclosure of Invention
In order to solve the above technical problems, embodiments of the present application provide a method and an apparatus for establishing a correction model, and a method and an apparatus for optimizing a mask, so as to improve accuracy of the model, accuracy and efficiency of mask pattern correction.
The embodiment of the application provides a method for establishing a correction model, which comprises the following steps:
acquiring historical weights corresponding to historical mask patterns; the historical weight is determined based on the graph parameters of the historical mask graph and/or the actual parameters of the historical exposure graph and the initial weight of the historical mask graph, and the initial weight of the historical mask graph corresponds to the type of the historical mask graph;
and establishing a correction model based on the graph parameters of the historical mask graph, the actual parameters of the historical exposure graph obtained by the historical mask graph exposure and the historical weight.
Optionally, the historical weight is determined based on the initial weight and a correction coefficient of the historical mask pattern, and the correction coefficient is determined based on a pattern parameter of the historical mask pattern and/or an actual parameter of the historical exposure pattern.
Optionally, the type of the historical mask pattern includes at least one of an independent line pattern, a line period pattern, an independent square block pattern, a square period array pattern, a square staggered pattern, an independent rectangular pattern, a rectangular period array pattern, a rectangular staggered pattern, an independent end-to-end pattern, an end-to-end period pattern, an independent end-to-end line pattern, an end-to-end line period pattern, an L pattern, a U pattern, a T pattern, an H pattern, an independent gap pattern, and a gap period pattern.
Optionally, the pattern parameters of the historical mask pattern include at least two kinds of critical dimensions of the historical mask pattern, and the actual parameters of the historical exposure pattern include at least one kind of critical dimensions of the historical exposure pattern; the critical dimension includes line width, period and interval.
Alternatively to this, the first and second parts may,
Figure BDA0002496289730000021
or
Figure BDA0002496289730000022
Or
Figure BDA0002496289730000023
Or
Figure BDA0002496289730000024
Wherein k is the correction coefficient, n is the limit design size of the historical mask pattern, and d1For the critical dimension of the historical mask pattern, m is the actual dimension of the historical exposure pattern obtained by exposure of the historical mask pattern with the limit design dimension, and d2Is the actual size of the historical exposure pattern.
Optionally, the historical weight is a product of the initial weight and the correction coefficient.
Optionally, the method further includes:
determining that the prediction error of the correction model is smaller than or equal to a preset error; the prediction error is the difference between the prediction parameter of the test mask pattern obtained by using the correction model and the actual parameter of the exposure pattern obtained by exposing the test mask pattern.
Optionally, the modifying the mask pattern to be modified based on the prediction parameter and the target parameter of the exposure pattern to be modified includes:
and if the difference value between the preset parameter and the target parameter of the exposure pattern to be corrected is larger than or equal to a preset value, correcting the mask pattern to be corrected based on the difference value.
The embodiment of the application also provides a mask pattern optimization method, which comprises the following steps:
obtaining a prediction parameter of the exposure pattern to be corrected corresponding to the mask pattern to be corrected by using the correction model; the correction model is obtained by the method for establishing the correction model provided by the embodiment of the application;
and correcting the mask pattern to be corrected based on the prediction parameter and the target parameter of the exposure pattern to be corrected so as to reduce the difference between the prediction parameter and the target parameter.
Optionally, the modifying the mask pattern to be modified based on the prediction parameter and the target parameter of the exposure pattern to be modified includes:
and if the difference value between the preset parameter and the target parameter of the exposure pattern to be corrected is larger than or equal to a preset value, correcting the mask pattern to be corrected based on the difference value.
The embodiment of the present application further provides an apparatus for establishing a correction model, where the apparatus includes:
the weight obtaining unit is used for obtaining the historical weight corresponding to the historical mask pattern; the historical weight is determined based on the graph parameters of the historical mask graph and/or the actual parameters of the historical exposure graph and the initial weight of the historical mask graph, and the initial weight of the historical mask graph corresponds to the type of the historical mask graph;
and the model establishing unit is used for establishing a correction model based on the graph parameters of the historical mask graph, the actual parameters of the historical exposure graph obtained by the historical mask graph exposure and the historical weight.
Optionally, the historical weight is determined based on the initial weight and a correction coefficient of the historical mask pattern, and the correction coefficient is determined based on a pattern parameter of the historical mask pattern and/or an actual parameter of the historical exposure pattern.
Optionally, the type of the historical mask pattern includes at least one of an independent line pattern, a line period pattern, an independent square block pattern, a square period array pattern, a square staggered pattern, an independent rectangular pattern, a rectangular period array pattern, a rectangular staggered pattern, an independent end-to-end pattern, an end-to-end period pattern, an independent end-to-end line pattern, an end-to-end line period pattern, an L pattern, a U pattern, a T pattern, an H pattern, an independent gap pattern, and a gap period pattern.
Optionally, the pattern parameters of the historical mask pattern include at least two kinds of critical dimensions of the historical mask pattern, and the actual parameters of the historical exposure pattern include at least one kind of critical dimensions of the historical exposure pattern; the critical dimension includes line width, period and interval.
Alternatively to this, the first and second parts may,
Figure BDA0002496289730000041
or
Figure BDA0002496289730000042
Or
Figure BDA0002496289730000043
Or
Figure BDA0002496289730000044
Wherein k is the correction coefficient, n is the limit design size of the historical mask pattern, and d1For the critical dimension of the historical mask pattern, m is the ultimate actual dimension of the historical exposure pattern obtained by exposure of the historical mask pattern with the ultimate design dimension, and d2Is the actual size of the historical exposure pattern.
Optionally, the historical weight is a product of the initial weight and the correction coefficient.
Optionally, the apparatus further comprises:
the verification unit is used for determining that the prediction error of the correction model is smaller than or equal to a preset error; the prediction error is the difference between the prediction parameter of the test mask pattern obtained by using the correction model and the actual parameter of the exposure pattern obtained by exposing the test mask pattern.
An embodiment of the present application provides a mask pattern optimization apparatus, including:
the device for establishing the correction model provided by the embodiment of the application is used for establishing the correction model;
the prediction parameter determining unit is used for obtaining the prediction parameters of the exposure pattern to be corrected corresponding to the mask pattern to be corrected by using the correction model;
and the parameter correcting unit is used for correcting the mask pattern to be corrected based on the prediction parameter and the target parameter of the exposure pattern to be corrected so as to reduce the difference between the prediction parameter and the target parameter.
Optionally, the parameter correction unit is specifically configured to:
and if the difference value between the preset parameter and the target parameter of the exposure pattern to be corrected is larger than or equal to a preset value, correcting the mask pattern to be corrected based on the difference value.
The embodiment of the application provides a method and a device for establishing a correction model, a method and a device for optimizing a mask pattern, the method and the device are used for establishing the correction model in advance based on the obtained graphic parameters of a historical mask pattern, the actual parameters of the historical exposure pattern obtained by exposure of the historical mask pattern and the historical weights corresponding to the historical mask pattern, wherein the historical weights are determined based on the graphic parameters of the historical mask pattern and/or the actual parameters of the historical exposure pattern and the initial weights of the historical mask pattern, the initial weights of the historical mask pattern correspond to the type of the historical mask pattern, after the mask pattern to be corrected is obtained, the prediction parameters of the exposure pattern to be corrected corresponding to the mask pattern to be corrected can be obtained by using the established correction model, the mask pattern to be corrected is corrected based on the prediction parameters and the target parameters of the exposure pattern to be corrected, to reduce the difference between the predicted parameter and the target parameter. The historical exposure pattern is related to the historical mask pattern, so that the obtained historical weight is also related to the pattern parameter of the historical mask pattern, the obtained historical weight is not only related to the type of the historical mask pattern, but also related to the pattern parameter of the historical mask pattern, and therefore the correction model is more pertinent.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments described in the present application, and other drawings can be obtained by those skilled in the art according to the drawings.
FIG. 1 is a schematic diagram of an exposure pattern provided in an embodiment of the present application;
FIG. 2 is a flowchart of a mask pattern optimization method according to an embodiment of the present disclosure;
FIG. 3 is a schematic illustration of a photolithography process according to an embodiment of the present application;
FIG. 4 is an exemplary graph to be modified provided by embodiments of the present application;
FIG. 5 is a schematic diagram illustrating a measurement of a historical mask pattern according to an embodiment of the present disclosure;
FIG. 6 is a schematic diagram illustrating measurement of a historical exposure pattern according to an embodiment of the present disclosure;
fig. 7 is a block diagram of a mask pattern optimization apparatus according to an embodiment of the present disclosure.
Detailed Description
In order to make the technical solutions of the present application better understood, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Currently, a mask pattern may be corrected using an optical proximity correction method based on a model, where the model may be established based on a mask pattern in actual operation and an exposure pattern that the mask pattern can form, and the mask pattern needs to have different kinds, such as an independent line pattern, a line period pattern, an end-to-end period pattern, and the like. Different types of mask patterns have different degrees of importance and contribute to model building, so different mask patterns may be weighted differently, for example, 30 for line period patterns, 60 for independent line patterns, and so on. However, the method for setting the weight basically depends on manual setting and on experience of technicians, and is often inaccurate and needs to be adjusted for many times according to modeling results. The resulting model also does not meet the practical requirements.
Based on the above technical problems, embodiments of the present application provide a method and an apparatus for building a correction model, and a method and an apparatus for optimizing a mask pattern, where a correction model is built in advance based on an obtained pattern parameter of a historical mask pattern, an actual parameter of a historical exposure pattern obtained by exposure of the historical mask pattern, and a historical weight corresponding to the historical mask pattern, where the historical weight is determined based on the pattern parameter of the historical mask pattern and/or the actual parameter of the historical exposure pattern, and an initial weight of the historical mask pattern, the initial weight of the historical mask pattern corresponds to a type of the historical mask pattern, after a mask pattern to be corrected is obtained, a prediction parameter of the exposure pattern to be corrected corresponding to the mask pattern to be corrected can be obtained by using the built correction model, the mask pattern to be corrected is corrected based on the prediction parameter and a target parameter of the exposure pattern to be corrected, to reduce the difference between the predicted parameter and the target parameter. The historical exposure pattern is related to the historical mask pattern, so that the obtained historical weight is also related to the pattern parameter of the historical mask pattern, the obtained historical weight is not only related to the type of the historical mask pattern, but also related to the pattern parameter of the historical mask pattern, and the correction model is more pertinent.
The following describes a specific implementation of a mask pattern optimization method and apparatus in the embodiments of the present application in detail by embodiments with reference to the drawings.
Referring to fig. 2, a flowchart of a mask pattern optimization method provided in an embodiment of the present application is shown, where the method may include the following steps:
s101, a correction model is established in advance.
In the photoetching process, a mask pattern can be designed according to the current process, and then the mask pattern is transferred onto a wafer by utilizing the photoetching technology, so that the wafer is etched in a targeted manner. Specifically, referring to fig. 3, a schematic diagram of a lithography process provided in an embodiment of the present application is shown, where a light source system may include a light source and a lens, the light source is configured to generate a laser beam, the lens is configured to change a direction of the laser beam to implement focusing and diverging of the laser beam, and the mask has a mask pattern, so that the laser beam can penetrate through the mask at a specific position, so that the laser beam penetrating through the mask has characteristics of the mask pattern, the laser beam penetrating through the mask is focused by a projection system, and the laser beam is irradiated on a photoresist layer on a surface of a wafer to cause a photochemical reaction on the photoresist layer, and then the photoresist layer is baked, developed, and cleaned, so that an exposure pattern is formed on the photoresist layer, and the exposure pattern and the mask pattern have a certain.
The correction model is capable of processing the mask pattern so as to simulate an exposure pattern obtained by exposing the mask pattern, wherein the exposure pattern represents an exposure pattern which can be obtained by the mask pattern in an actual exposure operation. It can be understood that if the difference between the simulated exposure pattern and the target exposure pattern is large, it means that the difference between the exposure pattern obtained by using the mask pattern and the expected exposure pattern is large, and the actual requirement cannot be met. Therefore, the correction model can predict the exposure pattern corresponding to the mask pattern, and correct the mask pattern based on the prediction result, so that a more accurate mask pattern can be obtained.
The correction model can be pre-established, specifically, the correction model can be established based on the pre-obtained graph parameters of the historical mask graph, the actual parameters of the historical exposure graph obtained by exposure of the historical mask graph and the historical weights corresponding to the historical mask graph, and different historical weights can be set for the correction model due to different contributions of different historical mask graphs to the establishment of the correction model, so that the importance of some important historical mask graphs is highlighted. Based on the correspondence between the pattern parameters of the historical mask pattern and the actual parameters of the historical exposure pattern, the correction model can be made to have the capability of predicting the pattern parameters of the exposure pattern.
In order to improve the accuracy of the correction model, in the actual operation, the historical weight can be determined according to the graph parameters of the historical mask graph and/or the actual parameters of the historical exposure graph, and because the historical exposure graph is an exposed graph obtained by exposure by using the historical mask graph, the historical exposure graph and the historical mask graph have strong correlation in parameters, the historical weight determined according to the actual parameters of the historical exposure graph also has strong correlation with the graph parameters of the historical mask graph. The historical weight which is associated with the graph parameters of the historical mask graph can better reflect the importance of the historical mask graph, the established correction model is more accurate, and the ability of predicting the graph parameters of the exposure graph is more reliable.
Specifically, the process of establishing the correction model may refer to the following description.
And S102, obtaining a prediction parameter of the exposure pattern to be corrected corresponding to the mask pattern to be corrected by using the correction model.
Based on the above description, in the embodiment of the present application, the mask pattern to be corrected is used as the mask pattern to be corrected, and the mask pattern to be corrected is a pattern used for optimization on the mask and may be a pattern selected by a photolithography engineer.
The mask pattern to be corrected may include one or more unit patterns, thereby constituting different types, wherein the unit patterns may include at least one of a horizontal independent line pattern, a vertical independent rectangular pattern, a L type pattern, a U type pattern, a T type pattern, an H type pattern, and the like.
When the mask pattern to be corrected includes a plurality of unit patterns, the mask pattern to be corrected may include at least one of the following patterns: line periodic patterns, square periodic array patterns, square staggered patterns, rectangular periodic array patterns, rectangular staggered patterns, independent end-to-end patterns, end-to-end periodic patterns, independent end-to-end line patterns, end-to-end line periodic patterns, independent gap patterns, gap periodic patterns and the like.
Specifically, the line periodic pattern may include a transverse line periodic pattern and a longitudinal line periodic pattern, the rectangular periodic array pattern may include a transverse rectangular periodic array pattern and a longitudinal rectangular periodic array pattern, the rectangular staggered pattern includes a transverse rectangular staggered pattern and a longitudinal rectangular staggered pattern, the independent end-to-end pattern includes a transverse independent end-to-end pattern and a longitudinal independent end-to-end pattern, the end-to-end periodic pattern may include a transverse end-to-end periodic pattern and a longitudinal end-to-end periodic pattern, the independent gap pattern may include a transverse independent gap pattern and a longitudinal independent gap pattern, and the gap periodic pattern may include a transverse gap periodic pattern and a longitudinal gap periodic pattern.
The transverse line periodic pattern consists of a plurality of transverse independent line patterns, and the longitudinal line periodic pattern consists of a plurality of longitudinal independent line patterns; the square periodic array graph is formed by a plurality of independent square graphs in an array mode, and the square staggered graph is formed by a plurality of independent square graphs in a staggered mode; the transverse rectangular periodic array pattern is formed by a plurality of transverse independent rectangular patterns in an array mode, and the longitudinal rectangular periodic array pattern is formed by a plurality of longitudinal independent rectangular patterns in an array mode; the transverse rectangular staggered pattern is formed by staggering a plurality of transverse independent rectangular patterns, and the longitudinal rectangular staggered pattern is formed by staggering a plurality of longitudinal independent rectangular patterns; the transverse independent end-to-end graph comprises at least two transverse independent line graphs with opposite end parts, and the longitudinal independent end-to-end graph comprises at least two longitudinal independent line graphs with opposite end parts; the transverse end-to-end graph can be formed by at least two groups of transverse line periodic graphs in the horizontal direction, and the longitudinal end-to-end graph can be formed by at least two groups of longitudinal line periodic graphs in the vertical direction; the transverse independent end-to-end line graph comprises a longitudinal independent line graph and two transverse independent line graphs arranged on two sides of the longitudinal independent line graph respectively, and the longitudinal independent end-to-end line graph comprises a transverse independent line graph and two longitudinal independent line graphs arranged on two sides of the transverse independent line graph respectively; the transverse end alignment graph can be composed of a longitudinal independent line graph and transverse line periodic graphs on two sides of the longitudinal independent line graph, and the longitudinal end alignment graph can be composed of a transverse independent line graph and longitudinal line periodic graphs on two sides of the transverse independent line graph; the transverse independent gap pattern may include a transverse gap formed by two unit patterns, and the longitudinal independent gap pattern may include a longitudinal gap formed by two unit patterns; the transverse gap periodic pattern includes more than two unit patterns, adjacent unit patterns can form a transverse gap, the longitudinal gap periodic pattern includes more than two unit patterns, and adjacent unit patterns can form a longitudinal gap.
Of course, in actual practice, the mask pattern to be corrected may be expanded according to actual circumstances, and is not limited to the above-exemplified types.
Referring to fig. 4, an exemplary graph to be modified according to an embodiment of the present application is shown, where fig. 4(a) shows a vertical independent line graph, fig. 4(b) shows a vertical line periodic graph, fig. 4(c) shows a horizontal independent line graph, fig. 4(d) shows a horizontal line periodic graph, fig. 4(e) shows an independent square block graph, fig. 4(f) shows a square block periodic array graph, fig. 4(g) shows a square block staggered graph, fig. 4(h) shows a vertical independent rectangular pattern, fig. 4(i) shows a vertical rectangular periodic array graph, fig. 4(j) shows a vertical rectangular staggered graph, fig. 4(k) shows a horizontal independent rectangular pattern, fig. 4(l) shows a horizontal rectangular periodic array graph, fig. 4(m) shows a horizontal rectangular staggered graph, fig. 4(n) shows a horizontal independent end-to end graph, fig. 4(o) shows a transverse end-to-end array pattern, fig. 4(p) shows a longitudinal independent end-to-end pattern, fig. 4(q) shows a longitudinal end-to-end array pattern, fig. 4(r) shows a transverse gap pattern, fig. 4(s) shows a longitudinal gap pattern, fig. 4(t) shows a transverse end-to-end pattern, and fig. 4(u) shows a longitudinal end-to-end pattern.
The mask pattern to be corrected may have pattern parameters for characterizing features of the mask pattern to be corrected. In this embodiment of the application, the pattern parameter of the mask pattern to be corrected may be a critical dimension (cd) of the mask pattern to be corrected, and the critical dimension may include at least two of a line width, a period, a distance, and the like, for example, a line width of an independent line pattern, a line width, a distance, a line width, a gap, a period, and the like of an independent end-to-end pattern, and a gap periodic pattern.
It should be noted that the mask pattern to be corrected may be a layout in practical application, and generally has a more complex pattern structure, including various types of patterns. In the embodiment of the present application, the pattern parameters of the mask pattern to be corrected may be obtained by the following method: and acquiring the graphic parameters of the mask graph to be corrected input by a user, or reading the graphic parameters of the mask graph to be corrected from the storage space, or identifying the mask graph to be corrected to obtain the graphic parameters of the mask graph to be corrected. For example, the boundary of the mask pattern to be corrected may be identified, so as to calculate the pattern parameters such as the line width, the pitch, the period, and the like of the mask pattern to be corrected.
In the embodiment of the application, the correction model can process the mask pattern, so that the exposure pattern corresponding to the mask pattern is predicted to obtain the prediction parameter of the exposure pattern, and the prediction parameter represents the pattern parameter of the exposure pattern which can be obtained by the mask pattern in the actual operation. It is understood that if the difference between the predicted parameter of the exposure pattern and the target parameter of the exposure pattern is large, it means that the difference between the exposure pattern obtained by using the mask pattern and the expected exposure pattern is large, and the actual requirement cannot be satisfied.
In the embodiment of the application, the exposure pattern corresponding to the mask pattern to be corrected and the pattern parameters of the exposure pattern can be obtained by utilizing the pre-established correction model, and the obtained prediction parameters of the exposure pattern have higher accuracy.
The predicted parameter of the exposure pattern may be a critical dimension, and the critical dimension may include at least one of a line width, a period, a spacing, and the like, for example, a line width of an independent line pattern, a line width, a spacing of an independent end-to-end pattern, a line width, a gap, a period, and the like of a gap period pattern.
S103, based on the prediction parameters and the target parameters, the mask pattern to be corrected is corrected so as to reduce the difference value between the prediction parameters and the target parameters.
In the embodiment of the application, for the mask pattern to be corrected, the corresponding exposure pattern to be corrected has the target parameter, the target parameter is the pattern parameter of the exposure pattern expected to be obtained after the mask pattern to be corrected is designed, and the closer the actual parameter of the exposure pattern obtained after the mask pattern to be corrected is exposed to the target parameter, the closer the design of the mask pattern to be corrected can meet the actual requirement.
The prediction parameter is a prediction result of a graphic parameter of an exposure graphic obtained after exposure of the mask graphic to be corrected based on the mask graphic to be corrected and the correction model, and on the basis that the correction model is more accurate, the closer the prediction parameter is to the target parameter, the more the mask graphic to be corrected meets the actual requirement. In the embodiment of the application, the mask pattern to be corrected can be corrected based on the prediction parameters and the target parameters, so that the difference value between the prediction parameters and the target parameters is reduced, and the design rationality and the photoetching accuracy of the mask pattern to be corrected are improved.
In specific implementation, the boundary of the mask pattern to be corrected can be moved to adjust the pattern parameters of the mask pattern to be corrected, so that the mask pattern to be corrected can be corrected.
Specifically, when the difference between the predicted parameter and the target parameter is greater than or equal to the second preset value, it may be determined that the error of the mask pattern to be corrected is not within the tolerable range, so that the mask pattern to be corrected may be corrected to reduce the difference between the preset parameter and the target parameter; when the difference value between the predicted parameter and the target parameter is smaller than the second preset value, the error of the mask pattern to be corrected is considered to be within a tolerance range, and the correction of the mask pattern to be corrected can not be carried out, so that the workload of pattern correction is reduced to a certain extent.
For example, if the predicted parameter is greater than the target parameter and the difference between the predicted parameter and the target parameter is greater than or equal to a second predetermined value, the critical dimension of the corresponding position of the mask pattern to be corrected may be reduced, so as to reduce the predicted parameter, so as to reduce the difference between the predicted parameter and the target parameter; if the predicted parameter is smaller than the target parameter and the difference between the predicted parameter and the target parameter is greater than or equal to a second preset value, the critical dimension of the corresponding position of the mask pattern to be corrected can be increased, so that the predicted parameter is increased, and the difference between the predicted parameter and the target parameter is reduced.
The embodiment of the application provides a mask pattern optimization method, which is characterized in that a correction model is established in advance based on the obtained pattern parameters of a historical mask pattern, the actual parameters of the historical exposure pattern obtained by historical mask pattern exposure and the historical weight corresponding to the historical mask pattern, wherein the historical weights are based on the pattern parameters of the historical mask pattern and/or the actual parameters of the historical exposure pattern, and determining initial weights of the historical mask patterns, the initial weights of the historical mask patterns corresponding to the types of the historical mask patterns, after the mask pattern to be corrected is obtained, the established correction model can be utilized to obtain the prediction parameters of the exposure pattern to be corrected corresponding to the mask pattern to be corrected, and the mask pattern to be corrected is corrected based on the prediction parameters and the target parameters of the exposure pattern to be corrected so as to reduce the difference between the prediction parameters and the target parameters. The historical exposure pattern is related to the historical mask pattern, so that the obtained historical weight is also related to the pattern parameter of the historical mask pattern, the obtained historical weight is not only related to the type of the historical mask pattern, but also related to the pattern parameter of the historical mask pattern, and the determined historical weight is more pertinent, so that the established correction model is more accurate, more accurate prediction parameters of the exposure pattern to be corrected can be determined, the correction of the mask pattern to be corrected is more pertinent, and the more accurate mask pattern is obtained.
The following explains a method of establishing a correction model.
S201, obtaining the historical weight of the historical mask pattern.
The correction model is capable of processing the mask pattern so as to predict an exposure pattern obtained by exposing the mask pattern, and at the same time, obtaining a prediction parameter of the exposure pattern, the prediction parameter representing a pattern parameter of the exposure pattern which can be obtained by actually operating the mask pattern.
The correction model can be established based on the graph parameters of the historical mask graph obtained in advance, the actual parameters of the historical exposure graph obtained by exposure of the historical mask graph and the historical weight corresponding to the historical mask graph.
The historical mask pattern may be a mask pattern obtained in advance, and subjected to actual exposure, and a corresponding historical exposure pattern is obtained. The historical mask pattern is similar to the mask pattern to be corrected, and can be a complete layout, or a selected part of the complete layout, or a corrected mask pattern, or an uncorrected mask pattern.
The unit patterns may include at least one of a horizontal independent line pattern and a vertical independent line pattern, and the vertical independent rectangle pattern may include a horizontal independent rectangle pattern and a vertical independent rectangle pattern.
When the history mask pattern includes a plurality of unit patterns, the history mask pattern may include at least one of the following patterns: line periodic patterns, square periodic array patterns, square staggered patterns, rectangular periodic array patterns, rectangular staggered patterns, independent end-to-end patterns, end-to-end periodic patterns, independent end-to-end line patterns, end-to-end line periodic patterns, independent gap patterns, gap periodic patterns and the like.
For an example of the historical mask pattern, reference may be made to an example of a mask pattern to be corrected, which is not described herein again.
The pattern parameter of the historical mask pattern may be a critical dimension of the mask pattern to be corrected, and the critical dimension may include at least two of a line width, a period, a spacing, and the like, for example, a line width of an independent line pattern, a line width, a spacing of an independent end-to-end pattern, a line width, a gap, a period, and the like of a gap periodic pattern.
In order to obtain a more accurate correction model, a large number of historical mask patterns and pattern parameters thereof can be obtained to cover a more comprehensive actual mask pattern, and the historical mask patterns can be of different types and can also have different critical dimensions. In practical applications, due to the limitation of the photolithography process, a minimum resolution pattern (anchor pattern) may exist in the historical mask pattern, and the critical dimension of the minimum resolution pattern is a critical process dimension, as an example, the current critical process dimension is 90 nm, and then the critical dimension of the minimum resolution pattern is 90 nm, for example, the line width of the line period pattern is 90 nm, the pitch is 90 nm, and the period is 180 nm.
Generally, the critical dimension of the historical mask pattern may be greater than or equal to the critical dimension to cover the actual lithography process, and certainly, in order to explore the actual lithography performance of the smaller-sized pattern and enable the modeling to be free, the critical dimension of the historical mask pattern may also include a range smaller than the critical dimension, and the historical mask pattern smaller than the critical dimension is often not well exposed and developed due to the process resolution limitation and becomes a sub-rule pattern, so that the corrected model can cover a wider pattern range.
Taking the line period pattern as an example, the critical dimension of the historical mask pattern may include a free combination of a plurality of line widths and a plurality of periods, for example, the historical mask pattern with a line width of 90 nm and a period of 180, 190, 200, etc., similarly, the historical mask pattern with a line width of 100 nm and a period of 80, 19, 200, etc., and the line width of 110 nm, 120 nm, etc., to cover more critical dimensions. The pattern parameters of these historical mask patterns may be stored in a layout file format, such as testpatterns.
Fig. 5 is a schematic diagram illustrating measurement of a historical mask pattern according to an embodiment of the present disclosure, wherein, referring to fig. 5(a), the schematic diagram illustrates measurement of a line period pattern according to an embodiment of the present disclosure, in which a line width is 90 nm (0.09 μm) and a period is 180 nm (0.18 μm); referring to fig. 5(b), a schematic diagram of an end-to-end period pattern measurement provided in the embodiment of the present application is shown, in which the line width is 180 nm (0.18 μm), the end-to-end spacing is 130 nm (0.13 μm), and the period is 360 nm (0.36 μm).
The historical exposure pattern obtained by historical mask pattern exposure is an exposure pattern obtained by exposing the historical mask pattern, the pattern parameters of the historical exposure pattern correspond to the pattern parameters of the historical mask pattern, and other factors in the actual photoetching process, such as diffraction effect, process characteristics, aberration and the like are included, so that the corresponding relation between the historical mask pattern and the historical exposure pattern can reflect the deviation generated in the actual photoetching process, for example, the corresponding relation is used for establishing a correction model, and the established correction model has the prediction capability of the pattern parameters of the exposure pattern.
The historical exposure patterns are also associated with exposure conditions, which may include light source shape, wavelength, Numerical Aperture (NA), illumination dose, etc., corresponding to different historical exposure patterns.
The pattern parameters of the historical exposure pattern comprise a critical dimension of the historical exposure pattern, wherein the critical dimension comprises at least one of line width, space and period, such as line width of an independent line pattern, line width and space of an independent end-to-end pattern, line width, space and period of a gap period pattern and the like.
Referring to fig. 6, a schematic diagram of a historical exposure pattern measurement provided by an embodiment of the present application is shown, in which a historical mask pattern obtained by exposure is a line period pattern shown in fig. 5(a), and a line width of the historical exposure pattern is 89.5 nm (0.895 μm). That is, the original 90 nm line width after exposure becomes 89.5 nm.
The method includes the steps of obtaining an exposure pattern on exposure, collecting graph parameters of a historical mask graph and the historical exposure graph obtained through exposure, and using the graph parameters as building data of a correction model, wherein the graph parameters are shown in table 1 and are examples of the graph parameters of the historical mask graph and the historical exposure graph obtained through exposure, wherein IS represents an independent gap graph (ISO space), DEN L represents a line period graph with a line width and space ratio of 1:1, ISO L represents an independent line graph (ISO line), L P represents a period line graph (line), e2e represents an end-to-end line graph (end to), G represents a space (gap), and the unit of each key dimension IS nanometer.
TABLE 1 Critical dimension of historical mask patterns and historical exposure patterns
Figure BDA0002496289730000141
Figure BDA0002496289730000151
The historical weight corresponding to the historical mask pattern can be determined according to the pattern parameters of the historical mask pattern and/or the actual parameters of the historical exposure pattern, so that the determined historical weight is related to the historical mask pattern and can pertinently represent the importance degree of the historical mask pattern in the aspect of modeling.
Specifically, the historical weight corresponding to the historical mask pattern may be determined based on an initial weight of the historical mask pattern corresponding to the type of the historical mask pattern and a correction coefficient determined based on a pattern parameter of the historical mask pattern and/or an actual parameter of the historical exposure pattern. In this way, the historical weights corresponding to the historical mask patterns integrate the types of the historical mask patterns and the characteristics of the pattern parameters, so that the historical mask patterns are highly correlated.
The initial weight of the historical mask graph can be determined in advance based on the type of the historical mask graph, for example, a larger weight is determined for a one-dimensional graph, a smaller weight is determined for a two-dimensional graph, and the maximum weight is set for the anchor pattern, and a line period graph has a larger weight than an independent line graph.
As an example, a weight of 60 may be determined for the independent line pattern and the independent gap pattern, a weight of 30 may be determined for the periodic line pattern and the gap periodic pattern, a weight of 1 may be determined for the two-dimensional pattern, and a weight of 100 may be determined for the anchor pattern.
The correction coefficient is determined according to the pattern parameter of the historical mask pattern and/or the actual parameter of the historical exposure pattern, generally speaking, the larger the pattern parameter of the historical mask pattern and/or the actual parameter value of the historical exposure pattern is, the less important the historical mask pattern is, the smaller the correction coefficient may be at this time, so as to reduce the historical weight corresponding to the historical mask pattern, whereas, the smaller the pattern parameter of the historical mask pattern and/or the actual parameter value of the historical exposure pattern is, the more important the historical mask pattern is, the larger the correction coefficient may be at this time, so as to increase the historical weight corresponding to the historical mask pattern. Specifically, the correction coefficient k may be determined in four ways.
(1) The method comprises the following steps: the correction coefficient k is determined by the following formula,
Figure BDA0002496289730000161
where n is the ultimate design dimension of the historical mask pattern, d1Is the critical dimension of the historical mask pattern. In the method, the ratio of the critical dimension of the historical mask pattern to the limit process dimension is considered, and the limit process dimension can be 90 nanometers.
(2) The method 2 comprises the following steps: the correction coefficient k is determined by the following formula,
Figure BDA0002496289730000171
in the method, the difference value of the critical dimension and the limit process dimension is considered, and the importance of the historical mask pattern with the critical dimension outside the rule range is further reduced.
(3) The method 3 comprises the following steps: the correction coefficient k is determined by the following formula,
Figure BDA0002496289730000172
where m is the actual size of the history exposure pattern obtained by exposing the history mask pattern having the ultimate design size, d2In the method, the ratio of the actual size of the historical exposure pattern to the limit actual size is considered.
(4) The method 4 comprises the following steps: the correction coefficient k is determined by the following formula,
Figure BDA0002496289730000173
in the method, the difference value between the critical dimension of the exposure pattern and the limit actual dimension is considered, and the importance of the historical mask pattern with the critical dimension outside the rule range is further reduced.
The correction coefficient can be determined by the four methods, and the product of the original weight w and the correction coefficient k is used as the historical weight w1, so that four historical weights can be obtained: the first weight w1, the second weight w2, the third weight w3 and the fourth weight w4 are, of course, when the correction coefficient is obtained in other manners, the historical weight may also be obtained by other calculation manners of the original weight and the correction coefficient, such as a ratio of the two. Referring to table 2, d is an example of a history weight provided in the embodiment of the present application1And d2In units of nanometers.
TABLE 2 History weights for respective History mask patterns
Figure BDA0002496289730000174
Figure BDA0002496289730000181
S202, establishing a correction model based on the graph parameters of the historical mask graph, the actual parameters of the historical exposure graph obtained by exposure of the historical mask graph and the historical weight.
The obtained data required for establishing the correction model are as follows: the pattern parameters of the historical mask pattern, the actual parameters of the exposure pattern obtained by exposure of the historical mask pattern and the historical weight corresponding to the historical mask pattern, so that the correction model can be established.
Specifically, the correction model may include an optical model considering errors generated by the light source and the lens part and a photoresist model considering errors generated by the photoresist, and the two models may be separately established.
The optical model can be established by inputting various parameters of the photoetching process, such as light source shape, wavelength, numerical aperture, light dose and the like, and simultaneously importing the graphic parameters of the historical mask graphics, the actual parameters of the exposure graphics obtained by the exposure of the historical mask graphics and the historical weight corresponding to the historical mask graphics to establish the optical model.
After the optical model is built, the optical model may be further evaluated, and evaluation indexes of the optical model are used to reflect the prediction accuracy of the optical model, for example, the evaluation indexes may include an error mean value (eMEAN), an error root mean square (etrro mean square, etrms), and the like, where the error root mean square is defined as:
Figure BDA0002496289730000191
wherein, wjAnd the historical weight corresponding to the jth historical exposure pattern is the difference value of the actual parameter and the predicted parameter of the historical exposure pattern. Examples of such applications areThe historical mask pattern is an isolated line with a line width of 120 nm, the line width of the historical exposure pattern obtained by exposure is 101.5 nm, and the value predicted by the established optical model is 108.8 nm, which is-7.3 nm.
Thus, a smaller eMEAN is better, and a smaller edrms is better. The historical weights provided by the embodiments of the present application enable the established optical model to have an eMEAN _ opt of-1.324 nm and an rms _ opt of 3.431 nm.
After the optical model is established, the photoresist model can also be established. Specifically, relevant process parameters such as the photoresist thickness, the photoresist refractive index, the anti-reflection layer thickness and the like can be input, and meanwhile, the pattern parameters of the historical mask pattern, the actual parameters of the exposure pattern obtained by exposure of the historical mask pattern, the historical weight corresponding to the historical mask pattern and the result of the optical model are imported to establish the photoresist model.
The photoresist model may have different base models, and the difference between the different base models is an increase or decrease of some factors. In actual operation, a plurality of photoresist models based on different basic models can be established, and the photoresist model which best meets the current process condition can be determined according to experience for modeling. For example, the photoresist model can be a first model based on an acid-base neutralized base model, a second model based on a gaussian convolution term base model, a third model based on a variable threshold base model, and so on.
After the photoresist model is established, the photoresist model may be evaluated, and evaluation indexes of the photoresist model are used to reflect the prediction accuracy of the photoresist model, for example, the evaluation indexes may include an error mean value (eMEAN), an error root mean square (etrms), and the like.
By using the historical weight, the evaluation indexes of the three photoresist models are respectively as follows: eMEAN _ pr1_ v-3.666 nm, eerms _ pr1_ v-12.066 nm, eMEAN _ pr2_ v-6.441 nm, eerms _ pr2_ v-20.678 nm, eMEAN _ pr3_ v-3.427 nm, and eerms _ pr3_ v-8.343 nm.
After the completion of the creation of the correction model, the correction model may be verified, and if the verification passes, the mask pattern may be corrected using the correction model, and if the verification does not pass, the creation of the correction model may be resumed. Specifically, it may be determined that the corrected model passes verification when it is determined that a predicted error of the corrected model is smaller than or equal to a preset error, where the preset error may be a difference between a predicted parameter of a test mask pattern obtained by using the corrected model and an actual parameter of an exposure pattern obtained by exposing the test mask pattern, where the test mask pattern is similar to a historical mask pattern and generally has fewer types of patterns compared to the actual mask pattern. In practice, mask patterns may be obtained in advance, wherein one part of the mask patterns is used as a history mask pattern for establishing a correction model, and the other part of the mask patterns is used as a test mask pattern for verifying the correction model.
In the embodiment of the present application, the evaluation index of the corrected model finally obtained by using the historical weight is better than the evaluation index of the corrected model obtained by using the original weight, and specifically, refer to table 3.
TABLE 3 examples of modified models created by different methods
Figure BDA0002496289730000201
Therefore, the establishment method of the correction model provided by the embodiment of the application can obtain the correction model with higher precision, so that the mask pattern can be corrected more accurately.
According to the method for establishing the correction model, the historical weight corresponding to the historical mask pattern can be obtained firstly, then the correction model is established based on the obtained graph parameters of the historical mask pattern, the actual parameters of the historical exposure pattern obtained by exposure of the historical mask pattern and the historical weight corresponding to the historical mask pattern, wherein the historical weight is determined based on the graph parameters of the historical mask pattern and/or the actual parameters of the historical exposure pattern, and the historical exposure pattern is related to the historical mask pattern, so that the obtained historical weight is also related to the graph parameters of the historical mask pattern, the obtained historical weight is not only related to the type of the historical mask pattern, but also related to the graph parameters of the historical mask pattern, and therefore the method is more pertinent, and the established correction model is also more accurate.
Based on the above method for establishing a correction model, an embodiment of the present application further provides a device 110 for establishing a correction model, and for a structural block diagram of the device for establishing a correction model provided by the embodiment of the present application, the device may include:
the weight obtaining unit is used for obtaining the historical weight corresponding to the historical mask pattern; the historical weight is determined based on the graph parameters of the historical mask graph and/or the actual parameters of the historical exposure graph and the initial weight of the historical mask graph, and the initial weight of the historical mask graph corresponds to the type of the historical mask graph;
and the model establishing unit is used for establishing a correction model based on the graph parameters of the historical mask graph, the actual parameters of the historical exposure graph obtained by the historical mask graph exposure and the historical weight.
Optionally, the historical weight is determined based on the initial weight and a correction coefficient of the historical mask pattern, and the correction coefficient is determined based on a pattern parameter of the historical mask pattern and/or an actual parameter of the historical exposure pattern.
Optionally, the type of the historical mask pattern includes at least one of an independent line pattern, a line period pattern, an independent square block pattern, a square period array pattern, a square staggered pattern, an independent rectangular pattern, a rectangular period array pattern, a rectangular staggered pattern, an independent end-to-end pattern, an end-to-end period pattern, an independent end-to-end line pattern, an end-to-end line period pattern, an L pattern, a U pattern, a T pattern, an H pattern, an independent gap pattern, and a gap period pattern.
Optionally, the pattern parameters of the historical mask pattern include at least two kinds of critical dimensions of the historical mask pattern, and the actual parameters of the historical exposure pattern include at least one kind of critical dimensions of the historical exposure pattern; the critical dimension includes line width, period and interval.
Alternatively to this, the first and second parts may,
Figure BDA0002496289730000211
or
Figure BDA0002496289730000212
Or
Figure BDA0002496289730000213
Or
Figure BDA0002496289730000214
Wherein k is the correction coefficient, n is the limit design size of the historical mask pattern, and d1For the critical dimension of the historical mask pattern, m is the ultimate actual dimension of the historical exposure pattern obtained by exposure of the historical mask pattern with the ultimate design dimension, and d2Is the actual size of the historical exposure pattern.
Optionally, the historical weight is a product of the initial weight and the correction coefficient.
Optionally, the apparatus further comprises:
the verification unit is used for determining that the prediction error of the correction model is smaller than or equal to a preset error; the prediction error is the difference between the prediction parameter of the test mask pattern obtained by using the correction model and the actual parameter of the exposure pattern obtained by exposing the test mask pattern.
In the device for establishing the correction model provided by the embodiment of the application, the historical weight corresponding to the historical mask pattern can be obtained firstly, then the correction model is established based on the obtained graph parameters of the historical mask pattern, the actual parameters of the historical exposure pattern obtained by exposure of the historical mask pattern and the historical weight corresponding to the historical mask pattern, wherein the historical weight is determined based on the graph parameters of the historical mask pattern and/or the actual parameters of the historical exposure pattern, and the historical exposure pattern is related to the historical mask pattern, so that the obtained historical weight is also related to the graph parameters of the historical mask pattern, the obtained historical weight is not only related to the type of the historical mask pattern, but also related to the graph parameters of the historical mask pattern, and therefore the correction model is more pertinent and the established correction model is more accurate.
Based on the above mask pattern optimization method, an embodiment of the present application further provides a mask pattern optimization apparatus, and referring to fig. 7, the apparatus is a block diagram of a structure of the mask pattern optimization apparatus provided in the embodiment of the present application, and the apparatus may include:
a corrected model establishing device 110 for establishing a corrected model;
a prediction parameter determining unit 120, configured to obtain, by using the correction model, a prediction parameter of the exposure pattern to be corrected, which corresponds to the mask pattern to be corrected;
a parameter modification unit 130, configured to modify the mask pattern to be modified based on the prediction parameter and a target parameter of the exposure pattern to be modified, so as to reduce a difference between the prediction parameter and the target parameter.
Optionally, the parameter correction unit is specifically configured to:
and if the difference value between the preset parameter and the target parameter of the exposure pattern to be corrected is larger than or equal to a preset value, correcting the mask pattern to be corrected based on the difference value.
The embodiment of the application provides a mask pattern optimization device, which establishes a correction model in advance based on the obtained pattern parameters of a historical mask pattern, the actual parameters of the historical exposure pattern obtained by historical mask pattern exposure and the historical weight corresponding to the historical mask pattern, wherein the historical weights are based on the pattern parameters of the historical mask pattern and/or the actual parameters of the historical exposure pattern, and determining initial weights of the historical mask patterns, the initial weights of the historical mask patterns corresponding to the types of the historical mask patterns, after the mask pattern to be corrected is obtained, the established correction model can be utilized to obtain the prediction parameters of the exposure pattern to be corrected corresponding to the mask pattern to be corrected, and the mask pattern to be corrected is corrected based on the prediction parameters and the target parameters of the exposure pattern to be corrected so as to reduce the difference between the prediction parameters and the target parameters. The historical exposure pattern is related to the historical mask pattern, so that the obtained historical weight is also related to the pattern parameter of the historical mask pattern, the obtained historical weight is not only related to the type of the historical mask pattern, but also related to the pattern parameter of the historical mask pattern, and therefore the correction model is more pertinent.
The name "first" in the names "first … …", "first … …", etc. mentioned in the embodiments of the present application is only used for name identification, and does not represent the first in sequence. The same applies to "second" etc.
As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that all or part of the steps in the above embodiment methods can be implemented by software plus a general hardware platform. Based on such understanding, the technical solution of the present application may be embodied in the form of a software product, which may be stored in a storage medium, such as a read-only memory (ROM)/RAM, a magnetic disk, an optical disk, or the like, and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network communication device such as a router) to execute the method according to the embodiments or some parts of the embodiments of the present application.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the apparatus embodiment, since it is substantially similar to the method embodiment, it is relatively simple to describe, and reference may be made to some descriptions of the method embodiment for relevant points. The above-described embodiments of the apparatus and system are merely illustrative, wherein modules described as separate parts may or may not be physically separate, and parts shown as modules may or may not be physical modules, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
The above description is only a preferred embodiment of the present application and is not intended to limit the scope of the present application. It should be noted that, for a person skilled in the art, several improvements and modifications can be made without departing from the scope of the present application, and these improvements and modifications should also be considered as the protection scope of the present application.

Claims (10)

1. A method for building a correction model, the method comprising:
acquiring historical weights corresponding to historical mask patterns; the historical weight is determined based on the graph parameters of the historical mask graph and/or the actual parameters of the historical exposure graph and the initial weight of the historical mask graph, and the initial weight of the historical mask graph corresponds to the type of the historical mask graph;
and establishing a correction model based on the graph parameters of the historical mask graph, the actual parameters of the historical exposure graph obtained by the historical mask graph exposure and the historical weight.
2. The method according to claim 1, wherein the historical weight is determined based on the initial weight and a correction factor of the historical mask pattern, the correction factor being determined based on a pattern parameter of the historical mask pattern and/or an actual parameter of the historical exposure pattern.
3. The method of claim 1, wherein the type of the historical mask pattern comprises at least one of an independent line pattern, a line period pattern, an independent square block pattern, a square period array pattern, a square staggered pattern, an independent rectangular pattern, a rectangular period array pattern, a rectangular staggered pattern, an independent end-to-end pattern, an end-to-end period pattern, an independent end-to-end line pattern, an end-to-end line period pattern, a L pattern, a U pattern, a T pattern, an H pattern, an independent gap pattern, and a gap period pattern.
4. The method according to claim 2 or 3, wherein the pattern parameters of the historical mask pattern comprise at least two kinds of critical dimensions of the historical mask pattern, and the actual parameters of the historical exposure pattern comprise at least one kind of critical dimensions of the historical exposure pattern; the critical dimension includes line width, period and interval.
5. The method of claim 4,
Figure FDA0002496289720000011
or
Figure FDA0002496289720000012
Or
Figure FDA0002496289720000013
Or
Figure FDA0002496289720000014
Wherein k is the correction coefficient, n is the limit design size of the historical mask pattern, and d1For the critical dimension of the historical mask pattern, m is the ultimate actual dimension of the historical exposure pattern obtained by exposure of the historical mask pattern with the ultimate design dimension, and d2Is the actual size of the historical exposure pattern.
6. The method of claim 1, further comprising:
determining that the prediction error of the correction model is smaller than or equal to a preset error; the prediction error is the difference between the prediction parameter of the test mask pattern obtained by using the correction model and the actual parameter of the exposure pattern obtained by exposing the test mask pattern.
7. A method for optimizing a mask pattern, the method comprising:
obtaining a prediction parameter of the exposure pattern to be corrected corresponding to the mask pattern to be corrected by using the correction model; the correction model is established by the establishing method of any one of claims 1-6;
and correcting the mask pattern to be corrected based on the prediction parameter and the target parameter of the exposure pattern to be corrected so as to reduce the difference between the prediction parameter and the target parameter.
8. The method according to claim 7, wherein the modifying the mask pattern to be modified based on the predicted parameter and the target parameter of the exposure pattern to be modified comprises:
and if the difference value between the preset parameter and the target parameter of the exposure pattern to be corrected is larger than or equal to a preset value, correcting the mask pattern to be corrected based on the difference value.
9. An apparatus for creating a correction model, the apparatus comprising:
the weight obtaining unit is used for obtaining the historical weight corresponding to the historical mask pattern; the historical weight is determined based on the graph parameters of the historical mask graph and/or the actual parameters of the historical exposure graph and the initial weight of the historical mask graph, and the initial weight of the historical mask graph corresponds to the type of the historical mask graph;
and the model establishing unit is used for establishing a correction model based on the graph parameters of the historical mask graph, the actual parameters of the historical exposure graph obtained by the historical mask graph exposure and the historical weight.
10. An apparatus for optimizing a mask pattern, the apparatus comprising:
means for establishing a correction model according to claim 9;
a prediction parameter determining unit, configured to obtain, by using the correction model established by the device for establishing a correction model, a prediction parameter of an exposure pattern to be corrected, which corresponds to the mask pattern to be corrected;
and the parameter correcting unit is used for correcting the mask pattern to be corrected based on the prediction parameter and the target parameter of the exposure pattern to be corrected so as to reduce the difference between the prediction parameter and the target parameter.
CN202010419225.6A 2020-05-18 2020-05-18 Correction model building method and device, mask optimizing method and device Active CN111443569B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010419225.6A CN111443569B (en) 2020-05-18 2020-05-18 Correction model building method and device, mask optimizing method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010419225.6A CN111443569B (en) 2020-05-18 2020-05-18 Correction model building method and device, mask optimizing method and device

Publications (2)

Publication Number Publication Date
CN111443569A true CN111443569A (en) 2020-07-24
CN111443569B CN111443569B (en) 2023-11-17

Family

ID=71655244

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010419225.6A Active CN111443569B (en) 2020-05-18 2020-05-18 Correction model building method and device, mask optimizing method and device

Country Status (1)

Country Link
CN (1) CN111443569B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113759659A (en) * 2021-09-08 2021-12-07 广东省大湾区集成电路与系统应用研究院 Light source mask optimization method and photoetching simulation device
CN115185165A (en) * 2022-09-13 2022-10-14 全芯智造技术有限公司 Construction method of assistant model, optical correction method and device and terminal
CN115587545A (en) * 2022-11-23 2023-01-10 广州粤芯半导体技术有限公司 Parameter optimization method, device and equipment for photoresist and storage medium

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7921383B1 (en) * 2006-01-11 2011-04-05 Olambda, Inc Photolithographic process simulation including efficient result computation for multiple process variation values
CN103163728A (en) * 2011-12-13 2013-06-19 上海华虹Nec电子有限公司 OPC correction method based on photoetching process window
JP2014182219A (en) * 2013-03-18 2014-09-29 Fujitsu Ltd Flawed site predictor, identification model generator, flawed site prediction program, and flawed site prediction method
US20190094680A1 (en) * 2017-09-28 2019-03-28 Taiwan Semiconductor Manufacturing Co., Ltd. Method of Mask Simulation Model for OPC and Mask Making
WO2019162346A1 (en) * 2018-02-23 2019-08-29 Asml Netherlands B.V. Methods for training machine learning model for computation lithography

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7921383B1 (en) * 2006-01-11 2011-04-05 Olambda, Inc Photolithographic process simulation including efficient result computation for multiple process variation values
CN103163728A (en) * 2011-12-13 2013-06-19 上海华虹Nec电子有限公司 OPC correction method based on photoetching process window
JP2014182219A (en) * 2013-03-18 2014-09-29 Fujitsu Ltd Flawed site predictor, identification model generator, flawed site prediction program, and flawed site prediction method
US20190094680A1 (en) * 2017-09-28 2019-03-28 Taiwan Semiconductor Manufacturing Co., Ltd. Method of Mask Simulation Model for OPC and Mask Making
WO2019162346A1 (en) * 2018-02-23 2019-08-29 Asml Netherlands B.V. Methods for training machine learning model for computation lithography

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
茅言杰 等: "基于差分进化算法的光刻机匹配方法" *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113759659A (en) * 2021-09-08 2021-12-07 广东省大湾区集成电路与系统应用研究院 Light source mask optimization method and photoetching simulation device
CN115185165A (en) * 2022-09-13 2022-10-14 全芯智造技术有限公司 Construction method of assistant model, optical correction method and device and terminal
CN115185165B (en) * 2022-09-13 2023-01-06 全芯智造技术有限公司 Construction method of assistant model, optical correction method and device and terminal
CN115587545A (en) * 2022-11-23 2023-01-10 广州粤芯半导体技术有限公司 Parameter optimization method, device and equipment for photoresist and storage medium

Also Published As

Publication number Publication date
CN111443569B (en) 2023-11-17

Similar Documents

Publication Publication Date Title
JP6735794B2 (en) Model-based process simulation method
CN102841509B (en) Integration of lithography apparatus and mask optimization process with multiple patterning process
JP4420878B2 (en) Method for performing calibration / optimization of resist process and DOE optimization to provide OPE alignment between different lithography systems
US7735053B2 (en) Correction method and correction system for design data or mask data, validation method and validation system for design data or mask data, yield estimation method for semiconductor integrated circuit, method for improving design rule, mask production method, and semiconductor integrated circuit production method
US8065636B2 (en) System and method for creating a focus-exposure model of a lithography process
JP3856197B2 (en) How to make OP mask
CN111443569A (en) Method and device for establishing correction model and method and device for optimizing mask
US8458626B1 (en) Method for calibrating an SRAF printing model
JP4989279B2 (en) Parameter value adjusting method, semiconductor device manufacturing method, and program
US8091048B2 (en) Method for predicting resist pattern shape, computer readable medium storing program for predicting resist pattern shape, and computer for predicting resist pattern shape
JP4068531B2 (en) Pattern dimension correction method and verification method using OPC, mask creation method and semiconductor device manufacturing method, and system and program for executing the correction method
JP2009302206A (en) Method of determining exposure parameter, program for determining exposure parameter, exposure method, and device manufacturing method
TW201725443A (en) Method and apparatus to correct for patterning process error
CN113050363A (en) Method for establishing optical proximity correction model and optical proximity correction method
TW201725452A (en) Method and apparatus to correct for patterning process error
TW201725442A (en) Method and apparatus to correct for patterning process error
KR102063229B1 (en) Simulation of Lithography Using Multi-Sampling of Angle Distribution of Source Radiation
JP4727288B2 (en) Method and program for optimizing a design formed on a substrate
US7966580B2 (en) Process-model generation method, computer program product, and pattern correction method
KR102642972B1 (en) Improved gauge selection for model calibration
KR100881525B1 (en) Photomask determination method, and method of manufacturing semiconductor device
JP2004163472A (en) Method for designing photomask, photomask, and semiconductor device
CN111258188B (en) Method for selecting optimum photoetching deviation of anchor point in SMO technology
CN116736626B (en) Optical proximity correction method, device, equipment and computer readable storage medium
CN110647008B (en) Method for screening SBAR rules

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant