Method for improving process-based contour information of structure in image
By combining edge placement gauges and critical dimension gauges with machine learning models, the simulated contours are modified to match the measurement data, solving the problem of overfitting in the lithography model and achieving a more accurate lithography and etching model, thus improving the efficiency and accuracy of the patterning process.
Patent Information
- Application Number
- CN202411022511.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2020-04-20
- Filing Date
- 2020-08-01
- Publication Date
- 2024-11-15
AI Technical Summary
In existing photolithography technology, the calibration method of photolithography model has the problem of overfitting, which leads to inaccurate prediction of the printed contour on the substrate. In particular, it is difficult to accurately predict the contour of the solid structure beyond the measurement area when modeling complex patterns and high-order terms.
By using edge placement gauges and critical dimension gauges in conjunction with machine learning models, the simulated contours are modified to match measurement data, generating more accurate measurement gauges for training lithography and etching models, thereby improving the accuracy and efficiency of the patterning process.
It improves the accuracy of photolithography and etching models, reduces overfitting problems, provides more accurate contour shape information, and enhances the yield and process window of the patterning process.
Smart Images

Figure CN118963068A_ABST
Abstract
Description
[0001] This application is a divisional application of the patent application with application number 202080058841.1, whose invention name is “Method for improving process-based contour information of structures in images” and whose international application date is 2020-08-01 and international application number is PCT / EP2020 / 071742, and whose national phase entry date in China is February 18, 2022. Technical Field
[0002] The present disclosure relates to techniques for improving the performance of metrology tools and device manufacturing processes. The techniques can be used in conjunction with lithography equipment metrology or profile-based manufacturing processes associated with device manufacturing. Background Art
[0003] A lithographic apparatus is a machine that applies a desired pattern to a target portion of a substrate. For example, a lithographic apparatus can be used in the manufacture of integrated circuits (ICs). In this case, a pattern forming device (alternatively referred to as a mask or reticle) can be used to produce a circuit pattern corresponding to a single layer of the IC, and the pattern can be imaged onto a target portion (e.g., a portion including a die, one or several dies) on a substrate (e.g., a silicon wafer) having a layer of radiation-sensitive material (resist). Typically, a single substrate will include a network of adjacent target portions that are exposed in succession. Known lithographic apparatus include so-called steppers, in which each target portion is irradiated by exposing the entire pattern to the target portion at once, and so-called scanners, in which each target portion is irradiated by scanning the pattern along a given direction (the "scanning" direction) with the beam while simultaneously scanning the substrate in a direction parallel or anti-parallel to the direction. Summary of the invention
[0004] In an embodiment, a method of generating a metrology gauge for measuring a physical property of a structure on a substrate is provided. The method includes: obtaining (i) measurement data associated with the physical property of the structure printed on the substrate, and (ii) at least a portion of a simulated profile of the structure, the portion of the simulated profile associated with the measurement data; modifying the portion of the simulated profile of the structure based on the measurement data; and generating the metrology gauge on or near the modified portion of the simulated profile, the metrology gauge being placed to measure the physical property of the simulated profile of the structure.
[0005] In addition, in an embodiment, a method for determining a hot spot location associated with a substrate is provided. The method includes: obtaining (i) a simulated profile associated with one or more patterns, the simulated profile being associated with measurement data of a physical property of the one or more patterns printed on the substrate, and (ii) a metrology gauge associated with the simulated profile; determining a value of the physical property associated with the one or more patterns based on the metrology gauge; and determining the hot spot location on the substrate based on the physical property value, wherein the hot spot location is a location on the substrate where the physical property value is less than a hot spot threshold associated with the one or more patterns.
[0006] In addition, in an embodiment, a method for training a model associated with a patterning process is provided. The method includes: obtaining (i) measurement data associated with a physical property of a structure printed on the substrate, and (ii) a metrology gauge associated with a simulated profile of the structure to be printed on the substrate, the simulated profile being associated with a defined location on the substrate where the physical property is measured; and using the measurement data and the metrology gauge to train the model so that a performance indicator of the patterning process is improved around the defined location on the substrate, the performance indicator being a function of the metrology gauge and the physical property.
[0007] In addition, in an embodiment, a method for generating a metrology gauge for measuring physical properties of a structure on a substrate is provided, the method comprising: obtaining (i) measurement data associated with the physical properties of the structure printed on the substrate, and (ii) at least a portion of a simulated contour of the structure, the portion of the simulated contour being associated with the measurement data; generating a modified contour of the portion of the simulated contour of the structure based on the measurement data; and providing the modified contour to a model of the patterning process to determine parameters of the patterning process.
[0008] Furthermore, in an embodiment, a computer program product is provided, including a non-transitory computer-readable medium having instructions recorded thereon, the instructions implementing the aforementioned method when executed by a computer.
[0009] In addition, in an embodiment, a method for training a machine learning model associated with a patterning process is provided. The method includes: obtaining (i) contour data of an after-development image (ADI) pattern on a substrate, (ii) measurement data of an after-etching image (AEI) pattern printed on the substrate, and (iii) obtaining a reference bias value based on the contour data of the ADI pattern and the measurement data of the AEI pattern; and using the measurement data and the contour data as training data to train the machine learning model to determine a bias value to be applied to the ADI contour.
[0010] In addition, in an embodiment, a method for determining a bias vector associated with an after-development image (ADI) pattern is provided. The method includes: obtaining (i) a probability distribution function (PDF) corresponding to particles within the ADI pattern deposited on a substrate, and (ii) a profile function characterizing an ADI profile associated with the ADI pattern; determining a deposition rate of the particles at a specified location on the ADI profile based on a combination of the PDF of the particles over an area of the ADI profile and the profile function; and determining a bias vector associated with the ADI pattern based on the deposition rate, the bias vector generating an after-etch image (AEI) profile when applied to the ADI profile of the ADI pattern.
[0011] In addition, in an embodiment, a method for determining a bias vector for a contour is provided. The method includes: obtaining (i) a probability distribution function (PDF) corresponding to a process to be performed on the contour, and (ii) a contour function characterizing the shape of the contour; convolving the contour function with the PDF over an area of the contour to determine a process rate at a specified location on the contour; and determining a bias vector to be applied to the contour based on the process rate to generate a biased contour indicating the effect of the process applied to the contour.
[0012] Furthermore, in an embodiment, a non-transitory computer-readable medium is provided, comprising instructions that, when executed by one or more processors, result in the operation of the method steps discussed herein. BRIEF DESCRIPTION OF THE DRAWINGS
[0013] Embodiments will now be described, by way of example only, with reference to the accompanying drawings, in which:
[0014] Figure 1 A block diagram showing various subsystems of a lithography system according to an embodiment;
[0015] Figure 2 depicts an example flow chart for modeling and / or simulating at least a portion of a patterning process according to an embodiment;
[0016] Figure 3A is a flow chart of a method of generating a measurement gauge (eg, an edge placement gauge, a CD gauge, etc.) for measuring a physical property of a structure on a substrate according to an embodiment;
[0017] Figure 3B According to the embodiment Figure 3A A flowchart of an example implementation of the steps used in modifying a simulation profile in a method of;
[0018] Figure 4Aillustrates an example of simulated profiles and measurement data at a location (e.g., within the FOV of a SEM tool) according to an embodiment;
[0019] Figure 4B The embodiment according to the present invention is shown Figure 4A An example of a modified profile associated with a simulated profile;
[0020] Figure 5 shows examples of signals associated with a simulated profile and thresholds used to produce a modified profile according to an embodiment;
[0021] Figure 6 is a flow chart of a method for determining a hot spot location associated with a substrate according to an embodiment;
[0022] Figure 7 is a flow chart of a method for training a model associated with a patterning process according to an embodiment;
[0023] Figure 8 illustrates an example model such as a convolutional neural network (CNN) including a plurality of layers, each layer being associated with model parameters such as weights and biases, in accordance with an embodiment;
[0024] Fig. 9 is a flow chart of a method for training a model associated with a patterning process according to an embodiment;
[0025] FIG. 10A to FIG. 10C is an example of etching biasing a resist profile and problems caused by etching biasing according to an embodiment;
[0026] Fig.11 is a flow chart of a method for determining a bias vector associated with an after developed image (ADI) pattern;
[0027] Fig.12 is an illustration of particles in a resist trench according to an embodiment;
[0028] Fig.13 is an example bias in the normal direction according to an embodiment;
[0029] Fig.14A and Fig. 14B According to the embodiment Fig.11 Example bias in the direction determined in ;
[0030] Fig.15 is a flow chart of a method for determining a bias vector associated with a process according to an embodiment;
[0031] Fig.16A and Fig. 16B illustrates an example application of an offset profile according to an embodiment;
[0032] Fig.17 schematically depicts an embodiment of a scanning electron microscope (SEM) according to an embodiment;
[0033] Fig.18 schematically depicts an embodiment of an electron beam inspection apparatus according to an embodiment;
[0034] Fig.19 is a block diagram of an example computer system according to an embodiment;
[0035] Fig. 20 is a schematic diagram of a lithographic projection apparatus according to an embodiment;
[0036] Fig.21 is a schematic diagram of an extreme ultraviolet (EUV) lithography projection apparatus according to an embodiment;
[0037] Fig. 22 According to the embodiment Fig.21 A more detailed view of the devices in
[0038] Fig.23 According to the embodiment Fig.21 and Fig. 22 A more detailed view of the source collector module of a device. DETAILED DESCRIPTION
[0039] Before describing the embodiments in detail, it is instructive to provide an exemplary environment in which the embodiments may be implemented.
[0040] Figure 1 An exemplary lithographic projection apparatus 10A is shown. The main components are: a radiation source 12A, which may be a deep ultraviolet excimer laser source or other types of sources including extreme ultraviolet (EUV) sources; a source (which, as discussed above, is not required by the lithographic projection apparatus itself); an illumination optical device, which, for example, defines a partial coherence (labeled σ) and may include optical devices 14A, 16Aa and 16Ab, which shapes the radiation from source 12A; a patterning device 18A; and a transmission optical device 16Ac, which projects an image of the patterning device pattern onto a substrate plane 22A. An adjustable filter or aperture 20A at the pupil plane of the projection optics may limit the range of beam angles impinging on the substrate plane 22A, wherein the largest possible angle defines the numerical aperture NA=nsin(Θmax) of the projection optics, where n is the refractive index of the medium between the substrate and the last element of the projection optics, and Θmax is the maximum angle of the beam emerging from the projection optics that may still impinge on the substrate plane 22A.
[0041] In a lithographic projection apparatus, a source provides illumination (i.e., radiation) to a pattern forming device, and a projection optical element directs the illumination onto a substrate via the pattern forming device and shapes the illumination. The projection optical element may include at least some of components 14A, 16Aa, 16Ab, and 16Ac. An aerial image (AI) is a distribution of radiation intensity at the substrate level. A resist layer on the substrate is exposed, and the aerial image is transferred to the resist layer as a latent "resist image" (RI) therein. A resist image (RI) may be defined as a spatial distribution of the solubility of the resist in the resist layer. A resist model may be used to calculate a resist image from an aerial image, an example of which may be found in commonly assigned U.S. Patent Application Publication No. US 2009-0157360, the disclosure of which is incorporated herein by reference in its entirety. The resist model is only related to the properties of the resist layer (e.g., the effects of chemical processes occurring during exposure, PEB, and development). The optical properties of a lithographic projection apparatus (e.g., properties of the source, patterning device, and projection optics) define the aerial image. Because the patterning device used in a lithographic projection apparatus can be varied, it may be desirable to separate the optical properties of the patterning device from the optical properties of the rest of the lithographic projection apparatus, including at least the source and projection optics.
[0042] In an embodiment, assist features (sub-resolution assist features and / or printable resolution assist features) may be placed in the design layout based on how the design layout is optimized according to the methods of the present disclosure. For example, in an embodiment, the method employs a machine learning based model to determine the patterning device pattern. The machine learning model may be a neural network (such as a convolutional neural network) that may be trained in a manner (e.g., as discussed in FIG. 3 ) to obtain accurate predictions at a faster rate, thereby enabling full chip simulation of the patterning process.
[0043] A neural network can be trained (i.e., parameters of the neural network are determined) using a set of training data. The training data may include or consist of a set of training samples. Each sample may be a pair including an input object (usually a vector, which may be referred to as a feature vector) and a desired output value (also referred to as a management signal), or a pair consisting of the input object and the desired output value. A training algorithm analyzes the training data and adjusts the behavior of the neural network by adjusting the parameters of the neural network (e.g., the weights of one or more layers) based on the training data. After training, the neural network can be used to map new samples.
[0044] In the context of determining a patterning device pattern, the feature vector may include one or more characteristics of the design layout included or formed by the patterning device (e.g., shape, arrangement, size, etc.), one or more characteristics of the patterning device (e.g., one or more physical properties, such as size, refractive index, material composition, etc.), and one or more characteristics of the illumination used in the lithography process (e.g., wavelength). The management signal may include one or more characteristics of the patterning device pattern (e.g., CD, profile, etc. of the patterning device pattern).
[0045] In the given form {(x1,y1),(x2,y2),…,(x N ,y N )} a set of N training samples such that x i is the feature vector of the i-th example and y i In the case of a signal for which it is managed, the training algorithm seeks a neural network g:X→Y, where X is the input space and Y is the output space. A feature vector is an n-dimensional vector representing a numerical feature of an object. The vector space associated with these vectors is often called the feature space. Sometimes the following operation is convenient: Using the scoring function to represent g such that g is defined as returning the value that gives the highest score Assume that F represents the space of scoring functions.
[0046] The neural network may be probabilistic, where g takes the form of a conditional probability model g(x)=P(y|x), or f takes the form of a joint probability model f(x,y)=P(x,y).
[0047] There are two basic approaches to choosing f or g: empirical risk minimization and structural risk minimization. Empirical risk minimization seeks to best fit the neural network to the training data. Structural risk minimization includes a penalty function that controls the bias / variance tradeoff. For example, in an embodiment, the penalty function may be based on a cost function, which may be squared error, number of defects, EPE, etc. The function (or weights within the function) may be modified so that the variance is reduced or minimized.
[0048] In both cases, the training set is assumed to consist of independent and identically distributed pairs (x i ,y i ) or is composed of one or more samples. In an embodiment, in order to measure how well the function fits the training data, a loss function is defined: For the training sample (x i ,y i, predict the value The loss is
[0049] Define the risk R(g) of a function g as the expected loss of g. This can be estimated from the training data as
[0050] In an embodiment, the machine learning model of the patterning process can be trained to predict, for example, the contour, pattern, CD of the mask pattern, and / or the contour, CD, edge placement (e.g., edge placement error) in the resist and / or etched image on the wafer. The goal of the training is to achieve accurate prediction of, for example, the contour, spatial image intensity slope, and / or CD of the printed pattern on the wafer. The contour refers to the pattern to be printed on the substrate or the outline of the printed pattern on the substrate. For example, the contour can be obtained via an image processing algorithm such as edge detection or other customized algorithm. The intended design (e.g., a wafer target layout to be printed on a wafer) is typically defined as a pre-OPC design layout that can be provided in a standardized digital file format such as GDSII or OASIS or other file formats.
[0051] Fig. 22 An exemplary flow chart for modeling and / or simulating portions of a patterning process is illustrated. As will be appreciated, the models may represent different patterning processes and need not include all of the models described below. The source model 1200 represents the optical properties of the illumination of the patterning device (including the radiation intensity distribution, bandwidth, and / or phase distribution). The source model 1200 may represent the optical properties of the illumination, including, but not limited to, numerical aperture settings, illumination standard deviation (σ) settings, and any specific illumination shape (e.g., off-axis radiation shapes such as toroidal, quadrupole, dipole, etc.), where the standard deviation (or σ) is the outer radial extent of the illuminator.
[0052] The projection optics model 1210 represents the optical properties of the projection optics (including changes in the radiation intensity distribution and / or phase distribution caused by the projection optics). The projection optics model 1210 may represent the optical properties of the projection optics, including aberrations, deformations, one or more refractive indices, one or more physical sizes, one or more physical dimensions, etc.
[0053] The patterning device / design layout model module 1220 captures how the design features are arranged in the pattern of the patterning device, and the patterning device / design layout model module 1220 may include a representation of the detailed physical properties of the patterning device, as described, for example, in U.S. Pat. No. 7,587,704, which is incorporated herein by reference in its entirety. In an embodiment, the patterning device / design layout model module 1220 represents the optical properties (including the changes in radiation intensity distribution and / or phase distribution caused by a given design layout) of a design layout (e.g., a device design layout corresponding to features of an integrated circuit, memory, electronic device, etc.), which is a representation of the arrangement of features on or formed by the patterning device. Because the patterning device used in the lithographic projection apparatus can be changed, it is desirable to separate the optical properties of the patterning device from the optical properties of the rest of the lithographic projection apparatus, including at least the illumination and projection optics. The goal of the simulation is often to accurately predict, for example, edge placement and CD, which can then be compared to the device design. The device design is typically defined as a pre-OPC patterning device layout and will be provided in a standardized digital file format such as GDSII or OASIS.
[0054] An aerial image 1230 may be simulated from the source model 1200, the projection optics model 1210, and the patterning device / design layout model 1220. The aerial image (AI) is the radiation intensity distribution at substrate level. The optical properties of the lithographic projection apparatus (e.g., the properties of the illumination, the patterning device, and the projection optics) dictate the aerial image.
[0055] A resist layer on a substrate is exposed by the aerial image, and the aerial image is transferred to the resist layer as a latent "resist image" (RI) therein. A resist s image (RI) can be defined as the spatial distribution of the solubility of the resist in the resist layer. A resist image 1250 can be simulated from the aerial image 1230 using a resist model 1240. The resist model can be used to calculate the resist image from the aerial image, and an example of this situation can be found in U.S. Patent Application Publication No. US2009-0157360, the entire disclosure of which is hereby incorporated by reference. The resist model typically describes the effects of chemical processes occurring during resist exposure, post-exposure baking (PEB), and development, in order to predict, for example, the profile of resist features formed on the substrate, and therefore it is typically only related to these properties of the resist layer (e.g., the effects of chemical processes occurring during exposure, post-exposure baking, and development). In an embodiment, the optical properties of the resist layer, such as refractive index, film thickness, propagation and polarization effects, may be captured as part of the projection optics model 1210 .
[0056] Therefore, in general, the connection / relationship between the optical model and the resist model is the simulated spatial image intensity within the resist layer, which originates from the projection of radiation onto the substrate, the refraction at the resist interface, and multiple reflections in the resist film stack. The radiation intensity distribution (spatial image intensity) is converted into a latent "resist image" by the absorption of the incident energy, which is further modified by diffusion processes and various loading effects. An efficient simulation method that is fast enough for full-chip applications approximates the actual 3D intensity distribution in the resist stack through a 2D spatial (and resist) image.
[0057] In an embodiment, the resist image may be used as input to a post-pattern transfer process model module 1260. The post-pattern transfer process model 1260 defines the performance of one or more post-resist development processes (eg, etching, developing, etc.).
[0058] The simulation of the patterning process can, for example, predict the profile, CD, edge placement (e.g., edge placement error), etc. in the resist and / or etched image. Thus, the goal of the simulation is to accurately predict, for example, edge placement of the printed pattern, and / or aerial image intensity slope, and / or CD, etc. These values can be compared to the intended design to, for example, correct the patterning process, identify where defects are predicted to occur, etc. The intended design is typically defined as a pre-OPC design layout, which can be provided in a standardized digital file format such as GDSII or OASIS or other file formats.
[0059] Thus, the model formulation describes most, if not all, of the known physical properties and chemistry of the overall process, and each of the model parameters desirably corresponds to a different physical or chemical effect. The model formulation thus sets an upper limit on how well the model can be used to simulate the overall manufacturing process.
[0060] In an example, computational analysis of the lithography or etching process employs a predictive model (e.g., as described above with respect to Figure 2 As discussed above, the predictive model, when properly calibrated, can produce accurate predictions of the dimensions output from the lithography and / or etching process. The model of the lithography or etching process is usually calibrated based on empirical measurements. This calibration includes running a test wafer with different process parameters, measuring the critical dimensions obtained after the etching process, and calibrating the model to the measured results. In practice, fast and accurate models are used to improve device performance or yield, enhance process windows, or increase design options. It will be appreciated by those skilled in the art that the methods described herein are not limited to the specific model of the lithography. In order to calibrate the desired model, an image can be obtained after any semiconductor manufacturing step. For example, an aerial image, a resist image, an etching image, an image after chemical mechanical polishing, or other images related to the process of the patterning process.
[0061] In computational lithography models, critical dimension (CD) gauges measured by scanning electron microscopes (CD-SEM) are often used as input data to calibrate the model. The goal of lithography modeling is to predict accurate resist profiles for each site on the substrate. However, when using radical model formalisms or deep convolutional neutral networks, the calibration produces models that suffer from overfitting. When such an overfitted model is used to predict, for example, the resist profile, it may deviate from the printed profile on the substrate, especially for those patterns that do not have an available CD gauge.
[0062] To alleviate this overfitting problem, the present disclosure provides a method for extracting metrology gauges such as edge placement (EP) gauges based on CD SEM raw images to provide much better pattern coverage. The EP gauge can help cover complex 2D patterns (such as holes). Complex 2D patterns are defined by at least 2 dimensions (such as width and length), and it may not be easy to place CD tangents, and they may not have reliable CD measurement options. In addition, existing measurement tools require several days of additional data processing time, which may be difficult to fit into a tight production schedule. Even more challenging is that it is sometimes very difficult to extract accurate 2D profiles from SEM images due to scanning direction, shadow effects, and / or charging effects.
[0063] As such, there are several limitations to methods that utilize only CD gauges from CD SEM metrology to generate computational lithography models. The limitations stem from the fact that lithography and plasma etching processes consist of complex physical and chemical reactions that are so complex that linear terms can only model pattern-dependent etch biases to a certain extent. However, more complex higher-order terms or deep convolutional neutral networks are prone to overfitting, which fails to predict the contours of physical structures beyond the metrology measurement site. In order to prevent overfitting using CD SEM metrology data, methods are needed to expand the CD metrology data to provide better data coverage and prevent overfitting.
[0064] The method of the present disclosure provides a way to generate a metrology gauge such as an EP gauge based on a CD gauge and a model to alleviate the problem of model overfitting. In addition, a method is provided for modifying the simulated model profile to match the measured CD data of, for example, a printed substrate. Thus, a model calibrated using the metrology gauge of the present disclosure can provide a better model that can also provide accurate profile shape information.
[0065] In an embodiment, a method for using a CD gauge associated with a printed substrate and an EP gauge associated with a model simulation to train a DCNN lithography and / or etching model is provided.
[0066] In an embodiment, CD metrology data (e.g., from a CD-SEM) and a solid model are used to generate a modified simulated profile that matches the metrology data. Additionally, simulated metrology data (e.g., an EP gauge) is generated based on the modified profile. This simulated metrology provides more metrology information than CD gauges obtained, for example, from CD-SEM alone.
[0067] Figure 3A Flow chart of a method for generating a metrology gauge (e.g., edge placement gauge, CD gauge, etc.) for measuring physical properties of a structure on a substrate. The method 300 generates a metrology gauge for measuring physical properties of a structure. In an embodiment, a metrology tool may be used to perform the measurement. In an embodiment, the metrology gauge (e.g., in a GDS file format) may be exported to a model (e.g., OPC, etch model, resist model, etc.) for improving the patterning process. In addition, in an embodiment, the method 300 may also be used to generate a modified simulation profile and export such a modified profile (e.g., in a GDS file format) to a model (e.g., etch model) for improving the patterning process.
[0068] In an embodiment, the term "gauge" or "gauge gauge" refers to a structure for measuring dimensions (e.g., size, shape) associated with physical characteristics of a structure on a substrate (e.g., a memory pattern, or other circuit pattern). In an embodiment, the gauge may be, for example, a visual marker or visual display of such information. In an embodiment, the gauge used to measure edge placement (e.g., a point at the outline of the structure) is referred to as an edge placement (EP) gauge. Similarly, a gauge used to measure the critical dimension (CD) of a structure may be referred to as a CD gauge. The gauge is also associated with a location on the substrate. The location may be a defined location (e.g., user-defined) or other location of interest, such as a location with a minimum or maximum size associated with the structure. For example, the location may be associated with a minimum CD value for a line or grid bar forming structure. EP and CD gauges are examples used to explain the concept. However, the present disclosure is not limited to gauges used to measure the physical characteristics associated with the structure of the substrate.
[0069] Step P301 comprises obtaining (i) measurement data 301 associated with the physical property of the structure printed on the substrate, and (ii) at least a portion 302 of a simulated profile of the structure, the portion of the simulated profile being associated with the measurement data 301. In an embodiment, the portion of the simulated profile is the portion of the simulated profile within a defined area around the measurement data 301 associated with the structure. In an embodiment, the portion may be the entire simulated profile.
[0070] In an embodiment, obtaining the portion 302 of the simulated profile includes defining an area of the substrate around a defined location associated with the measurement data 301; and simulating a patterning process within the defined area of the substrate to obtain the portion 302 of the simulated profile of the structure. For example, the defined location can be a field of view (FOV) of a metrology tool or a user-selected area around the portion 302 of the structure. In an embodiment, the FOV is a limited area on the substrate captured for observation or measurement purposes. For example, the FOV is an area around the structure printed on the substrate, a location where the CD value of the structure is measured, or other given location. In an embodiment, the size of the defined location (i.e., local area) can be selected so that within the area, the profile shape has optimal physical fidelity. When two CD gauges are very close to each other, the area can be selected so that they do not overlap each other.
[0071] In an embodiment, the measurement data 301 is obtained via a metrology tool. In an embodiment, the metrology tool is a scanning electron microscope (SEM) and the measurement data 301 is obtained from a SEM image. In an embodiment, the SEM tool captures an image of the structure printed on the substrate. The image can be acquired at a given location using a FOV.
[0072] The simulated profile is the outline of the structure to be printed on the substrate. Figure 2 ) and obtain the simulated profile. In an embodiment, the simulation process can be configured to be performed only on a specific part (e.g. Figure 2 Rather than simulating the entire substrate, a process model of the substrate is used. Simulating only a portion of the substrate allows for faster execution and reduced computing resources compared to simulating the entire substrate.
[0073] Figure 4A Examples of simulated profiles 401a and 401b (collectively referred to as 401) and measured data 410 at a site (e.g., within the FOV of a SEM tool) are shown. In an embodiment, the simulated profiles 401a and 401b (collectively referred to as 401) and measured data 410 are shown. Figure 2 In an embodiment, the measurement data 410 is a physical property (e.g., CD, EPE, etc.) associated with the structure and the patterning process is simulated to obtain the simulation profile 401. In an embodiment, the measurement data 410 is a physical property (e.g., CD, EPE, etc.) associated with the structure. The value associated with the physical property can also be obtained from the simulation profile 401. However, the simulated value of the physical property can be significantly different from the actual measured value of the physical property. Therefore, if the measurement is based on such a simulation profile, the measurement will eventually be inaccurate and may affect the yield of the patterning process. The present disclosure provides a method for modifying the simulation profile and further generating the measurement gauge (e.g., EP gauge, CD gauge) based on the modified profile. For example, process P303 is a method (as an example) for modifying the simulation profile. Figure 4B Examples of modified profiles (eg, 411a and 411b) of the simulation profile 401 are illustrated.
[0074] In an embodiment, the measured data 410 is a CD value associated with the structure at the given location on the substrate. In an embodiment, the CD value is the distance between two contours at the given location. In an embodiment, the measured CD value is substantially different from the CD value obtained from the simulated contour 401. In an embodiment, the simulated contour 401 is modified so that the measured CD value and the simulated CD value are similar.
[0075] Step P303 comprises modifying a portion 302 of the simulated profile of the structure based on the measurement data 301, thereby generating a modified profile 304 of the simulated profile. Figure 3B An example implementation of the step for modifying the simulation profile is discussed below.
[0076] Process P311 includes determining simulated data 312 associated with the physical property of the simulated profile of the structure based on the portion 302 of the simulated profile. Process P313 includes determining a difference between the measured data 301 associated with the physical property of the structure and the simulated data 312. Process P315 includes modifying the portion 302 of the simulated profile based on the difference 314 so that the difference 314 between the measured data 301 and the simulated data 312 is reduced. The resulting modified profile 304 can also be used in various applications related to the patterning process (e.g., improving patterns, determining process parameters, OPC, etc.).
[0077] As mentioned earlier, the measured data are CD values at defined locations associated with the structure.Next, the portion 302 of the simulated profile is modified based on the difference 314 between the simulated CD value associated with the structure and the measured CD value.
[0078] Figure 4B An example of a modified profile 411 and measurement data 410 associated with the simulated profile 401 at a given location (e.g., within the FOV of a SEM tool) is shown. The modified profile 411 can be obtained using procedures P311, P312, and P315 (or P317) as discussed herein. For example, the simulated profile 401 can be modified based on measurement data 410 such as CD values. In an embodiment, the simulated profile is used to measure the CD value at the same location as the measurement data. For example, the simulated CD can be measured between the simulated profiles 401a and 401b. Next, the difference between the simulated CD value and the measured CD value is calculated. Based on the CD difference, the simulated profile is modified within the FOV so as to minimize the CD difference. In an embodiment, the difference is such that the size of the simulated profile is increased to the modified profiles 411a and 411b, so as to reduce (in an embodiment, minimize) the CD difference. In addition, based on the modified profile 411, a measurement gauge is generated. The resulting metrology gauge, such as an EP gauge, may also be used to accurately measure properties of the structures on the substrate.
[0079] In another example, the modification of the portion 302 of the simulated profile includes adjusting a threshold associated with obtaining the simulated profile (e.g., for obtaining the simulated profile in a level setting method). For example, in an embodiment, processes P311, P313, and P315 may be employed. Process P311 includes determining simulated data 312 associated with the physical properties of the simulated profile of the structure based on the portion 302 of the simulated profile. Process P313 includes determining a difference 314 between the simulated data 312 associated with the physical properties of the structure and the measured data. Process P317 includes adjusting a threshold used to generate the simulated profile based on the difference 314, so that the difference 314 between the measured data 301 and the simulated data 312 is reduced, wherein the adjusted threshold modifies the portion 302 of the simulated profile. Thus, a modified profile 304' is generated and it can also be used in different applications (e.g., OPC) related to the patterning process, as mentioned earlier.
[0080] In an embodiment, the measured data is a CD of a feature. In such a case, in an example, the modification of the portion 302 of the simulated profile comprises using the portion 302 of the simulated profile to determine a simulated CD value at a defined location on the substrate where the measured CD value is obtained; determining a difference 314 between the simulated CD value and the measured CD value; and adjusting a threshold value based on the difference 314 so as to reduce the difference 314 between the CD values, the adjusted threshold value modifying the portion 302 of the simulated profile.
[0081] Figure 5 An example of a signal 501 associated with a simulated profile and a threshold value for producing a modified profile is shown. The signal can be imagined as a mountain-shaped profile in 3 dimensions (e.g., x, y, and z). For example, a patterning process simulation can involve a level setting method for receiving a signal 501 (e.g., an image intensity associated with a simulated pattern). In addition, the level setting method uses a threshold 510, such as in the form of a plane that cuts across the signal. Then, the intersection of the plane and the signal produces the simulated profile. Depending on the threshold, different simulated profiles can be produced. Thus, according to the present disclosure, the difference between the measured data and the simulated data from the simulated profile can be used to adjust the threshold 510 to different thresholds 520. The adjusted threshold 520 makes it produce the simulated profile, and the simulated profile makes it reduce or minimize the difference between the measured data and the simulated data associated with the physical property. For example, the threshold 510 can be modified with respect to the difference between the simulated data and the measured data.
[0082] Step P305 includes generating the measurement gauge (e.g., edge placement gauge) on or near the modified portion of the simulated profile, the measurement gauge being placed to measure the physical property of the simulated profile of the structure. In an embodiment, generating the measurement gauge includes specifying a mark such as a point on (or near) the modified portion of the simulated profile; and exporting the location of the point as the measurement gauge (e.g., the edge placement gauge). In an embodiment, the location can be exported or output as a text file, a GDS file, or other format for processing by a computer. Figure 4B Illustrated are example edge placement gauges EP1, ..., EP10, ..., EPn generated along a modified profile 411. In an embodiment, the edge placement gauges are points at or around the modified profile. In an embodiment, the edge placement gauges may be generated by dragging a line from the simulated profile to the modified profile in a direction perpendicular to the modified profile.
[0083] In an embodiment, the method 300 may be modified to generate a modified profile from the simulated profile, the modified profile being used to improve the patterning process. In an embodiment, the improvement of the patterning process includes performing a simulation based on the patterning process simulation (e.g., see Figure 2 ) determine the parameters of the patterning process.
[0084] In an embodiment, the method 300 may be modified as follows. As explained in process P301, the method includes obtaining (i) measurement data 301 associated with the physical property of the structure printed on the substrate, and (ii) at least a portion 302 of the simulated profile of the structure, the portion 302 of the simulated profile associated with the measurement data. In addition, as explained with respect to process P303, the method includes generating a modified profile of the portion 302 of the simulated profile of the structure based on the measurement data 301. In an embodiment, the modified profile (discussed with respect to P303) may be generated by shifting the simulated profile based on a difference 314 between the measurement data 301 and the simulated data 312. In an embodiment, the simulated profile is shifted to reduce the CD difference between the measured CD and simulated CD values, for example, at a given location.
[0085] In addition, the method includes providing the modified profile to a model of the patterning process to determine parameters of the patterning process. Figure 2 The etch model or resist model is used to further improve the accuracy of the simulated etch profile or the simulated resist profile.
[0086] Figure 66 is a flow chart of a method 600 for determining the location of a hot spot on a substrate. The method 600 may be an application of a metrology gauge such as an EP gauge or a CD gauge. For example, the EP gauge generated by P305 may be used to determine the location of a hot spot. The hot spot detection algorithm may use an EP gauge (e.g., EP1, ..., EPn) to determine the pattern and location of the hot spot. In an embodiment, a hot spot is a process window limiting pattern or a pattern that is most likely to fail after imaging on the substrate. An example method for determining a hot spot is explained using steps P601, P603, and P605. However, the metrology gauge may be used in any other hot spot detection algorithm that is configured to determine a hot spot based on a metrology gauge and a simulated profile.
[0087] Step P601 includes obtaining (i) a simulated profile 601 associated with one or more patterns, wherein the simulated profile 601 is associated with measurement data of physical properties of the one or more patterns printed on the substrate, and (ii) a measurement gauge 602 (e.g., an edge placement gauge and / or a CD gauge) associated with the simulated profile 601.
[0088] In an embodiment, obtaining the metrology gauge 602 includes: determining the simulated profile 601 associated with one or more patterns by simulating a patterning process using the measurement data; modifying at least a portion of the simulated profile 601 based on the measurement data associated with the one or more patterns; and generating the metrology gauge 602 on or at the modified portion of the simulated profile 601. For example, the method 300 can be used to modify the simulated profile 601 and further generate the metrology gauge 602 such as an EP gauge.
[0089] Step P603 includes determining 604 a value of the physical property associated with the one or more patterns based on the metrology gauges 602. In an embodiment, determining 604 the value of the physical property includes measuring 604 the value of the physical property at one or more metrology gauges in the metrology gauges 602. In an embodiment, the metrology gauges 602 can be used to measure an edge placement error (EPE) of a simulated profile relative to a reference pattern (e.g., a target pattern), a CD gauge, or other physical property.
[0090] Step P605 includes determining a hot spot 606 or a hot spot location 606 on the substrate based on the physical property value 604, wherein a hot spot or a hot spot location refers to a pattern or location on the substrate whose physical property value is less than a hot spot threshold associated with the one or more patterns.
[0091] In an embodiment, determining the hot spot location 606 includes: determining whether the value of the physical property associated with the one or more patterns exceeds the hot spot threshold; and in response to exceeding the threshold, identifying the location of the metrology gauge 602 associated with exceeding the threshold. For example, the hot spot threshold can be a minimum CD or EPE value of the features to be printed on the substrate.
[0092] Figure 7 700 is a flow chart of a method for training a model associated with a patterning process. The method 700 is an example application of a metrology gauge 702 produced using the method 300 herein. Because the metrology gauge 702 is more accurate, a process model associated with the patterning process trained based on the metrology gauge 702 will provide more accurate results (e.g., closely matching the measurement data). The results of the model can also be used to determine improved parameters of the patterning process, thereby resulting in a higher yield from the actual patterning process. The example processes involved in the method 700 are discussed in detail below.
[0093] Step P701 comprises obtaining (i) measurement data 701 associated with the physical properties of the structure printed on the substrate, and (ii) a measuring gauge 702 (e.g., an EP gauge or a CD gauge) associated with a simulated contour of the structure to be printed on the substrate, wherein the simulated contour is associated with a defined location on the substrate where the physical properties are measured.
[0094] Step P703 includes using the measurement data 701 and the metrology gauge 702 to train the model 704 so that the performance index of the patterning process is improved around the defined location on the substrate, and the performance index is a function of the metrology gauge 702 and the physical property. After completing the training process, the model is called a trained model 704.
[0095] In an embodiment, the training of the model is an iterative process. Iteration includes: determining a simulated profile of the structure to be printed on the substrate and simulated data associated with the physical properties of the simulated profile of the structure by executing the model; determining a first difference between the simulated data and the measured data 701, and a second difference between a point along the simulated profile and a measurement gauge 702; and determining model parameters based on a gradient of a performance indicator with parameters of a patterning process so as to minimize the performance indicator, the performance indicator being a function of the first difference and the second difference.
[0096] Figure 8An example model such as a convolutional neural network (CNN) comprising multiple layers is illustrated, each layer being associated with model parameters such as weights and biases. When an input (e.g., a feature vector) is passed through the layers, the input is weighted and biased according to the values assigned to each layer, and an output (e.g., an output vector of simulated contours and patterning process parameters) is generated.
[0097] As mentioned earlier, the training of the machine learning model such as CNN 800 is an iterative process. Iterations include: initializing model parameters of CNN 800; predicting values of the physical properties associated with the substrate; and adjusting the model parameter values of CNN 800 so as to reduce the cost function.
[0098] In an embodiment, adjusting the model parameter values is based on a gradient descent of the cost function. In an embodiment, the cost function is minimized. In an embodiment, adjusting the model parameter values of CNN 800 includes determining a gradient map or gradient graph of a first cost function that varies according to the model parameters. Then, based on the gradient map, the model parameter values are determined so as to minimize the cost function.
[0099] In an embodiment, adjusting the model parameter values comprises adjusting the values of: one or more weights of a layer of a convolutional neural network, one or more biases or offsets of a layer of a convolutional neural network, a hyperparameter of a CNN, and / or the number of layers of a CNN. In an embodiment, the number of layers is a hyperparameter of a CNN that may be preselected and may not be changed during the training process. In an embodiment, a series of training processes may be performed where the number of layers may be modified.
[0100] In an embodiment, the cost function is the difference between measured data and simulated data (e.g., predicted by CNN 800). The difference is reduced by modifying the values of CNN model parameters (e.g., weights, biases, strides, etc.). In an embodiment, the gradient corresponding to the difference may be dcost / dparameter, where the cnn_parameters value may be updated based on an equation (e.g., parameter=parameter-learning_rate*gradient). In an embodiment, the parameters may be weights and / or biases, and learning_rate, i.e., the learning rate, may be a hyperparameter used to tune the training process and may be selected by a user or computer to improve convergence (e.g., faster convergence) of the training process.
[0101] In an embodiment, the model is at least one of the process models, such as an etch model configured to predict an etch image; or a resist model configured to predict a resist image.
[0102] The computational analysis of the etching process employs a calibrated prediction model that can predict the size of the etched structure produced by the etching process. As mentioned herein, the model associated with the etching process can be calibrated based on empirical measurements. The calibration process includes patterning a test wafer with different process parameters, measuring the critical dimension (CD) of the pattern on the test wafer after the etching process, and calibrating the model based on the measured CD. In practice, a fast and accurate model can be used to improve the performance of the patterning equipment, the patterning yield, the process window of the patterning process, or to increase design options associated with, for example, determining a mask pattern.
[0103] After the etching process, the etch profile of the etched pattern deviates from the corresponding resist profile of the resist pattern on the substrate. The deviation is pattern dependent. A constant deviation may not be applied to the resist profile to produce the etch profile. In etching modeling, the resist profile can be used as input, and the goal is to predict the etching bias values to be applied to different points on the resist profile. In existing modeling methods, the pattern-dependent etching bias values are modeled by linear equations that use multiple linear terms that describe pattern characteristics.
[0104] There are several limitations on using the linear equation to model the bias values that depend on the pattern. The limitations arise from the fact that the etching process (e.g., using dry etching) includes complex chemical reactions and physical particle bombardment, which are so complex that linear terms can only model the etching bias values that depend on the pattern within a limited range. Thus, etching effects that cannot be accurately modeled by linear terms should be considered to produce a more accurate etching model. In an embodiment, the etching model can also be used in various applications related to lithography. For example, the etching model can be used to determine, for example, an OPC associated with a mask pattern in order to improve patterning performance or yield.
[0105] Currently, an etch profile is generated by applying bias values (e.g., determined by the etch model) at different points of the resist profile. The bias values are applied to the resist profile in the local normal direction. However, this approach tends to result in over-calculation of bias values at points of high curvature, and the resulting etch profile can exhibit non-physical behavior (e.g., FIG. 10A to FIG. 10C The present disclosure describes methods for determining an etch profile and a bias direction to address the aforementioned problems associated with the etch profile.
[0106] Fig. 9900 is an exemplary process for training a machine learning model associated with a patterning process according to an embodiment of the present disclosure. The training is based on measurement data related to an after-development image (ADI) and an after-etch image (AEI). After training, the trained model can determine bias values that can be applied to the ADI profile to generate an etch profile. The exemplary process 900 includes different processes discussed in detail below.
[0107] Step P901 includes: obtaining (i) contour data 901 of an ADI pattern on a substrate, (ii) measurement data 902 of an AEI pattern printed on the substrate, and (iii) a reference bias value 903 based on both the contour data 901 of the ADI pattern and the measurement data 902 of the AEI pattern. For example, the reference bias value 903 is determined based on the difference between the measurement values of the ADI pattern and the AEI pattern.
[0108] In an embodiment, the contour data 901 may be represented in the form of an image of a contour associated with one or more features of the ADI pattern. In an embodiment, the image is generated from a simulated contour of a simulated ADI pattern. In an embodiment, obtaining the contour data 901 involves executing one or more process models associated with the patterning process using a design pattern to be printed on the substrate as input to generate a simulated ADI pattern. The patterning process includes a resist process or a resist model for simulating the resist process. From the simulated ADI pattern, an ADI contour may be extracted. Each contour is a contour of a feature within the simulated ADI pattern. In an embodiment, one or more process models include at least one of: an optical device model configured to determine an aerial image, and a resist model configured to determine a resist image. About Figure 2 Example simulation processes employing different models related to the patterning process are discussed.
[0109] In an embodiment, the image may be obtained from a metrology device (e.g., a SEM) configured to capture an image of the substrate after a resist process on the substrate. In an example, the profile may be a resist profile that may be extracted from a resist image (e.g., a SEM image of the resist pattern printed on the substrate).
[0110] In an embodiment, measurement data 902 is obtained at a specified metrology gauge. As mentioned earlier, the metrology gauge can be an edge placement gauge, a critical dimension (CD) gauge associated with the AEI pattern, or both. For example, the measurement data 902 at the metrology gauge includes a location of an edge placement gauge associated with a contour of the AEI pattern printed on the substrate; and / or a CD value associated with the AEI pattern printed on the substrate.
[0111] In an embodiment, when the measuring gauge is a CD gauge, the reference offset value 903 is obtained via a calibration process configured to determine an offset value associated with a given CD gauge. The offset value indicates the amount of CD reduction to be applied to the ADI pattern to produce the AEI pattern. In an embodiment, the offset value is provided at the end of a given CD gauge. The offset values at the two ends may not be equal. In other words, the offset value may be asymmetric about the center of the CD gauge.
[0112] In an embodiment, the calibration process includes determining a bias model as a linear combination of multiple terms characterizing the pattern. The bias model can determine a bias value at a specific resist contour point. An example bias model is given by the following linear model.
[0113]
[0114] In the above process, Term_bias i is the model term associated with point i of the ADI profile, and c i is the coefficient associated with Term_bias at point i. In embodiments, the model terms may be linear expressions, or physical terms related to an aspect of the patterning process (e.g., CD, dose, focus, MSD, resist thickness). In embodiments, the bias model may be combined with a lithography simulation process (e.g., Figure 2 In an embodiment, the resist profile is then biased in the normal direction using the bias value predicted by the model to obtain a corresponding etching profile.
[0115] In an embodiment, the terms in the bias model may be represented by CD, and point i refers to the first end or the second end of a CD gauge (e.g., a horizontal or vertical line drawn across a profile to measure the CD of the profile). Thus, in an embodiment, the bias model may determine bias values at the ends of a CD gauge. When working with a CD gauge, the bias values are split into the two ends of the CD gauge, since the bias is not always symmetrical with respect to the center of the gauge. The method of splitting the bias for a CD gauge uses the calibrated bias model described above, which may produce asymmetric bias values at a given CD gauge. In an embodiment, the gauge center is used as a reference, and the bias values are equally split into the two CD gauge ends. The split CD bias values are then used to train a CNN model. In an embodiment, when an edge placement (EP) gauge is used, there is no asymmetric splitting of the bias values. The bias values are determined for each EP gauge and such bias values may be used directly to train the CNN model.
[0116] In an embodiment, the ADI pattern or the ADI profile extracted therefrom may first be transformed into a different image format, and then the ADI pattern or the ADI profile extracted therefrom may be used to train the model. For example, the image format may include a filtered downsampled resist image, i.e., a filtered downsampled resist image (FDRI). For example, the FDRI may be a low pass filter image generated by applying a low pass filter to the profile extracted from the ADI pattern. In an embodiment, the profile may be a binary image, and if it is used directly to train the model, the training process may be very slow compared to using the FDRI. In addition, the FDRI is a grayscale image, which provides greater flexibility in modifying each pixel value during the training process, so that the model output converges to the desired result at a faster rate. In an embodiment, the image may be generated by transforming the ADI profile or other mathematical transformation of the ADI profile with a bias model term. The transformation may result in a better correlation between the bias model term and the etching process.
[0117] Step P903 includes training the machine learning model using the measurement data 902 and the profile data 901 as training data to determine the bias value to be applied to the ADI profile. After the training process, a trained model 905 is generated. The trained model 905 can also be applied to one or more aspects of the patterning process to improve, for example, lithography performance, patterning yield, adjustment parameters of the patterning process, etc.
[0118] In an embodiment, training the model includes adjusting model parameters of the machine learning model so that the bias value will be within a specified range determined based on the reference bias value 903. For example, the values of the weights and biases of a model (e.g., a convolutional neural network (CNN)) may be adjusted to cause the model to generate bias values within the specified range. In an embodiment, the specified range indicates that the bias values generated by the model converge to the reference bias value 903. For example, the specified range may be defined as a value that deviates within 0 to 5% of each reference bias value at a given location of the ADI pattern (e.g., a reference bias value ±0.1 nm). In an embodiment, the specified range may be defined as a value that deviates within 0 to 5% of each reference bias value.
[0119] In an embodiment, the training of the machine learning model is an iterative process. Iteration includes: (a) executing the machine learning model using the measurement data 902, the profile data 901, and given values of the model parameters to generate a bias map associated with the profile data 901, the bias map including a bias value; (b) adjusting the model parameters of the machine learning model based on the gradient of the difference between the model-based bias value and the reference bias value 903 so as to reduce the difference; and (c) performing steps (a) to (b) until the difference is minimized.
[0120] In an embodiment, the model parameters are weights and biases of the model. Adjusting the weights and biases of one or more layers of the model causes the model to produce a bias value approximately the same as the reference bias value 903. In an embodiment, the gradient of the difference between the bias value produced by the model and the reference bias value 903 guides the adjustment of the value of the model parameter. For example, the gradient can be a mapping of the derivative of the difference relative to the model parameter. The mapping includes peaks and troughs, wherein the trough indicates a minimized point. In an embodiment, the training process includes adjusting the value of the model parameter so as to minimize the difference. This minimization can be associated with the trough of the gradient mapping. For example, the minimization is achieved by changing the model parameter value in the direction of the valley value of the trough.
[0121] In an embodiment, the machine learning model is configured to generate a representation of a bias map for the ADI profile. In an embodiment, the bias map may be represented as a pixelated image, with each pixel indicating a bias value. Additionally, the pixel locations may be associated with coordinates of a target layout, or coordinates of the ADI pattern. In an embodiment, the bias value may be positive, negative, or zero. A positive bias value indicates that the ADI profile should be reduced and a negative bias value may indicate that the ADI profile should be increased, or vice versa.
[0122] In an embodiment, the bias map generated via the trained machine learning model includes an etch bias value to be applied to the resist profile to determine the etch profile to be printed on the substrate. In an embodiment, the bias map includes coordinates associated with the entire wafer or die. Each coordinate system is associated with a bias value. In an embodiment, the etch bias value is applied to the resist profile in a local normal direction. The local normal direction is the direction perpendicular to the resist profile at a given point on the resist profile. Thus, each point on the resist profile will have a different normal direction. In an embodiment, the bias map is a pixelated image, each pixel having an intensity value indicating a bias value.
[0123] In an embodiment, as mentioned earlier, applying the bias value to the ADI profile in the local normal direction may result in an unrealistic etch profile. FIG. 10A to FIG. 10C Examples of existing biasing methods and associated problems are illustrated.
[0124] exist Fig. 10A , bias values b1, b2, b3, b4, and b5 may be applied to different locations of the resist profile 1001. The bias values b1 to b5 are applied in the normal direction to produce an etch profile 1020. In the case where the bias values b1 to b5 are large enough, these bias values may cause a fish-mouth-like irregular shape 1021 in the etch profile 1020. This fish-mouth shape 1021 is an unrealistic representation of the etch pattern.
[0125] like Fig. 10B , the bias values intersect at the curvature region 1030. This intersection of the biases results in the fish mouth 1021. In embodiments, large bias values that may not intersect may produce sharp line ends (e.g., as Fig. 10C ). Fig. 10C A resist profile 1050 is shown to which the bias values b10, b11, and b12 may be applied to produce an etch profile 1060. The bias values b10 and b11 are large enough to produce a sharp line end like a knife tip. Therefore, shifting the ADI profile in the local normal direction by the bias value calculated by the calibrated bias model may not result in an accurate AEI profile. Fig.11 A method is provided in which a bias vector is determined that can be applied to, for example, a resist profile.
[0126] Fig.111 is an exemplary process 1100 for determining a bias vector associated with an after-development image (ADI) pattern according to an embodiment of the present disclosure. In an embodiment, the bias vector includes a bias direction that directs the bias value in a direction that does not cause a profile curvature intersection when biased. In an embodiment, the method 1100 includes the following procedures discussed in detail below. In an embodiment, the bias value can be obtained according to a trained model (e.g., 905) configured to generate a bias value for any given pattern, the bias vector of the method 1100, a user-defined bias value, or other bias determination algorithm or method.
[0127] Step P1101 includes obtaining (i) a probability distribution function 1101 (PDF) of particle deposition within an ADI pattern on a substrate and (ii) a profile function 1102 characterizing an ADI profile associated with the ADI pattern.
[0128] In an embodiment, a PDF 1101 of particle deposition is determined or calibrated based on measured substrate data. The measured substrate data may include deposition data of the particle, and a measured etching pattern. In an embodiment, the PDF 1101 of the particle characterizes the net deposition effect or net etching effect of the particle in contact with the ADI profile. Embodiments are described in detail herein using the terms "deposition" or "deposition rate," wherein the resulting profile is derived by applying a bias inward from an original profile. However, it should be understood that the disclosed mechanism for determining the bias direction may also be extended to applications where the resulting profile may be derived by applying a bias outward from the original profile and by using a negative deposition rate. In an embodiment, the PDF 1101 may be a Gaussian distribution. However, this is merely exemplary; any other suitable form of function may be used without departing from the scope of the present disclosure. In an embodiment, obtaining the PDF 1101 includes determining the variance or standard deviation (σ) of the Gaussian distribution to which the measured data is fitted. This will be discussed later in this article with respect to Fig.13 and FIG. 14A to FIG. 14B An example of how the variance of the Gaussian distribution affects the bias direction and the etching profile is discussed below.
[0129] Step P1103 includes determining a deposition rate 1103 of the particle at a specified location on the ADI profile based on a combination of the PDF 1101 of the particle over the area of the ADI profile and the profile function 1102. In an embodiment, the deposition rate 1103 may be positive (e.g., corresponding to a contraction of the profile) or negative (e.g., corresponding to an expansion of the profile). In an embodiment, determining the deposition rate 1103 of the particle includes convolving the profile function 1102 with the PDF 1101 of the particle and integrating over the area of the ADI profile.
[0130] Fig.12 The example influence of particles on the resist profile represented by the profile function R (x, y) is illustrated. As shown, at the point P on the resist profile, the bias direction points to the particle position (marked by a star). In an embodiment, the particle position is characterized by the concentration of the particle. In an embodiment, the particle will be deposited on the resist wall, so the resist profile will decrease in the direction toward the particle. In an embodiment, the resist groove will include etching particles whose dispersion is characterized by, for example, a Gaussian distribution G (r). In an embodiment, the resist profile R (x, y) is integrated with all particles on the region of the resist profile to find the final etching profile E (x, y). In other words, the etching profile is not determined by only one particle, but by all particles in the resist groove.
[0131] In an embodiment, the deposition rate 1103, such as D(x,y), may be determined based on the following equation:
[0132] D(x,y)=k∫∫R(u,v)G(xu,yv)dudv
[0133] In the above process, R(u,v) is a profile function used to characterize the geometry of the profile (e.g., resist profile) in the ADI; and G(xu,yv) is a deposition rate function of particles in a groove at a distance r from the resist wall. In an embodiment, the deposition rate function is a Gaussian function characterized by a mean and a variance. In an embodiment, the variance of the Gaussian function can be determined based on measured data (e.g., an etching profile on a printed substrate). In an embodiment, G(xu,yv) serves as a guide to the direction of the bias value. For example, Fig.14A and Fig. 14B The graph shows that changing the variance of the Gaussian function affects the bias direction and the final etch profile.
[0134] Step P1105 includes determining a bias vector 1105 associated with the ADI pattern based on the deposition rate 1103. The bias vector 1105 generates an after-etch image (AEI) profile when applied to the ADI profile of the ADI pattern. In an embodiment, the bias vector 1105 includes a bias direction at a specific location of the ADI profile. In an embodiment, the method may also include the step of applying a bias value along the bias direction to generate the AEI profile. For example, the bias vector includes a bias direction along which a bias value may be applied at a specific location on a resist profile, as discussed herein (e.g., see FIG. 14A to FIG. 14B ).
[0135] In an embodiment, determining the bias vector 1105 includes determining a gradient of the deposition rate 1103 relative to a first direction and a second direction of the ADI pattern. For example, the first direction (eg, along the x-axis) and the second direction (eg, along the y-axis) are perpendicular to each other.
[0136] In an embodiment, for the above deposition rate D(,), the gradient of the deposition rate 1103 is determined based on the following equation:
[0137]
[0138] In the above process, the gradient of the deposition rate Expressed as a combination of the x- and y-components of the deposition rate in a given direction.
[0139] In an embodiment, the bias direction at each designated location on the ADI profile is associated with a bias value. When the bias values at different locations are applied to the ADI profile, the bias vectors 1105 at different locations do not intersect each other. In an embodiment, the bias direction of the bias vector 1105 includes a direction that is not perpendicular to the ADI profile. In an embodiment, the variance of the Gaussian distribution of the particles causes the bias vector 1105 to change. Thus, in an embodiment, the variance can be adjusted to generate the bias vector 1105 that does not cause the ADI profiles to intersect when the bias values are applied.
[0140] In an embodiment, when the ADI pattern includes multiple contours, a set of bias vectors 1105 are determined separately for each ADI contour. For example, the ADI pattern may include features on a first layer and on a second layer on top of the first layer. In an example, a feature may be surrounded by adjacent features of the ADI pattern. In an example, the density or closeness of adjacent features may be combined to calculate the bias value. However, regardless of the density of the adjacent features, after applying the bias value, the bias vectors will not cause the ADI contours to intersect.
[0141] Fig.13 An example of applying bias values to a resist profile RC1 in the normal direction at different points on the resist profile to produce a bias profile EC1 (also referred to as an etch profile EC1) is illustrated. Note that in the case of the curvature of the resist profile RC1, the bias vectors intersect each other in region R1. As mentioned earlier, this intersection causes irregular or non-physical behavior of the etch profile EC1. For example, the resist profile RC1 is moved by the bias value to cause the bias profile EC1 to have a fish mouth or sharp line end in the region R1.
[0142] In an embodiment, the bias profile EC1 may be similar to the profile generated by applying the method 1100 discussed above. For example, the bias profile EC1 may be generated by setting the variance of the Gaussian function to substantially zero. Fig.14A and Fig. 14B The effect of changing the variance of the Gaussian function is further illustrated in FIG.
[0143] Fig.14A and Fig. 14B are example results of applying method 1100 using the Gaussian function with variances of, for example, 30 and 60, respectively. In an embodiment, the method 1100 determines a bias vector based on a Gaussian function with a first variance and determines another bias vector based on a Gaussian function with a second variance relatively higher than the first variance. When the bias vector is applied to the resist profile RC1, it does not cause the bias values to intersect, and bias profiles EC2 and EC3 are generated.
[0144] As the variance of the Gaussian function increases, the intersection point of the bias value (associated with the resist profile) moves to the left. For example, the intersection point in region R3 is on the relatively left side of the intersection point in region R2. In an embodiment, the intersection point indicates a relatively high concentration of particles within the resist trench. Therefore, the bias value points to the intersection point.
[0145] Comparing the offset profiles EC2 and EC3 shows that the profile portions within R2 and R3 do not have sharp edges or fish mouth shapes. In addition, the portion of the offset profile EC2 (within R2) is relatively sharper (pointier) than the portion of the offset profile EC3 (within R3).
[0146] In an embodiment, the variance value of the Gaussian function may be calibrated based on measurement data (e.g., etch profile data of a printed substrate), as discussed earlier. Using the calibrated Gaussian function, the method 1100 may be used to determine the bias direction. For example, the gradient is determined In addition, the bias value determined, for example, using a trained model 905 (e.g., CNN), and the bias direction at each point of the resist profile are used. The etch profile may be produced.
[0147] In an embodiment, methods 900 and 1100 can be used for various applications related to the patterning process. Example applications include but are not limited to SMO, OPC, hot spot detection, defect detection, adjusting parameters of lithography equipment during the manufacturing process, adjusting parameters of post-lithography processes, and other related applications.
[0148] For example, in OPC applications, a mask pattern can be used to generate a resist profile. Using the resist profile as an input to the trained model 905, the bias value can be determined. The bias value can be applied to the resist profile to determine the etching profile. In an embodiment, the bias value can be applied in a normal direction or in a bias direction determined by method 1100. In addition, depending on the difference between the etching profile and the target profile to be printed on the substrate, an optical proximity effect correction for the mask pattern can be determined. In an embodiment, the aforementioned steps can be repeated until the difference between the etching profile and the target profile is minimized.
[0149] In an embodiment, the method 1100 is not limited to patterning processes. The method 1100 can be extended to determine bias profiles for other applications. In an example, modifications of the method 1100 are discussed as follows.
[0150] In an embodiment, Fig.15 is a flow chart of an exemplary process 1500 for determining an offset vector for a contour. The method 1500 includes the following steps.
[0151] Procedure P1501 includes obtaining (i) a probability distribution function 1501 (PDF) corresponding to a process to be performed on the profile, and (ii) a profile function 1502 that characterizes the shape of the profile. For example, the PDF 1501 can represent the behavior of a processing process via a machining tool, a measurement process via a metrology tool, a lithography-related process as discussed herein, guiding a robotic device along a profile, or other processes involving profile-based operations. In an example, the profile can be a geometric shape associated with a part to be machined. In another example, the profile can characterize a tool travel path during a machining process, a tool travel path during a measurement process, a robot motion path, or respective limits of other properties associated with the profile. In an embodiment, the PDF 1501 can represent properties of a tool used in the process. For example, the PDF 1501 can be specified for a specific tool having a specified size used during a machining operation, etching, a robotic part size, or other properties that affect the profile when the process is performed on the profile.
[0152] Step P1503 includes convolving the profile function 1502 with the PDF 1501 over the area of the profile to determine a process rate 1503 at a specified location on the profile. In an embodiment, the process results in the addition or removal of material forming the profile, which addition or removal results in a change in the shape of the profile. In an embodiment, the process rate characterizes the behavior of the addition or removal of material forming the profile. For example, the addition or removal of material during a machining process, or the addition or removal of material during an etching process associated with lithography. The PDF 1501 of the process can be a Gaussian function fitted based on measurement data related to the process performed on the profile.
[0153] Process P1505 includes determining a bias vector 1505 to be applied to the profile based on the process rate 1503 for producing a bias profile indicating the effect of the process applied to the profile. For example, the bias vector 1505 includes a bias value applied inwardly or outwardly relative to the profile to produce the bias profile. For example, in a removal process, the bias value can be applied in an inward direction. In an addition process, the bias value can be applied in an outward direction. The processes discussed herein, such as machining, etching, robotic movement, etc., are exemplary for explaining the concepts and do not limit the scope of the invention.
[0154] Fig.16A and Fig. 16B An example of a contour-based process is shown. For example, Fig.16AThe machining operation performed on the die via a machining tool (e.g., a grinding tool) is illustrated. The component comprises a profile 1601 before the machining process is performed. After machining, a machined profile 1602 is obtained. Such a machined profile 1602 represents an offset profile determined via a PDF characterizing the machining process performed using a tool of a specified size.
[0155] Fig. 16B Another example of a process based on a profile is illustrated. For example, profile 1611 represents the initial profile of a component to be machined (or scanned) via tool MT1. After machining, the offset profile 1612 is obtained. In an embodiment, based on the profile 1611 and the offset profile 1612, a tool path (represented by the horizontal and dotted lines inside the offset profile 1612) can be determined. As shown, tool MT1 is a circle with a specified radius and machining speed for generating or tracking the offset profile 1612. It will be appreciated that the present disclosure is not limited to specific tools. The tools used in the process may be machining tools, etching tools, scanning tools, or other tools associated with the photolithography process for generating or tracking the offset profile.
[0156] In an embodiment, one or more processes of methods 300, 600, 700, 900, 1100 and 1500 may be implemented on one or more processors of a computer system. In an embodiment, a computer program product is provided, including a non-transitory computer-readable medium having instructions recorded thereon, which, when implemented by a computer, execute one or more processes of the above methods.
[0157] For example, in an embodiment, a non-transitory computer-readable medium includes instructions that, when executed by one or more processors, result in operations including: obtaining (i) a probability distribution function (PDF) corresponding to particles deposited within an after-development image (ADI) pattern on a substrate, and (ii) a profile function characterizing an ADI profile associated with the ADI pattern; determining a deposition rate of the particle at a specified location on the ADI profile based on a combination of the profile function over an area of the ADI profile and the PDF of the particle; and determining a bias vector associated with the ADI pattern based on the deposition rate, the bias vector producing an after-etch image (AEI) profile when applied to the ADI profile of the ADI pattern.
[0158] In an embodiment, in the non-transitory computer readable medium, obtaining the probability distribution function (PDF) of the particles is based on measured substrate data, the measured substrate data comprising deposition data of the particles, and the measured etching pattern. In an embodiment, obtaining the PDF comprises determining the variance of a Gaussian distribution fitted to the measured data.
[0159] In an embodiment, in the non-transitory computer readable medium, the determining the deposition rate of the particle comprises instructions for: convolving the PDF of the particle with the profile function; and integrating over the area of the ADI profile. In the non-transitory computer readable medium, the determining the bias vector comprises determining a gradient of the deposition rate relative to a first direction and a second direction of the ADI pattern, the first direction and the second direction being perpendicular to each other.
[0160] In an embodiment, in the non-transitory computer readable medium, the bias vector includes: a bias direction at a location of the ADI profile, and also includes applying the bias value to generate the AEI profile. In an embodiment, in the non-transitory computer readable medium, the bias direction is determined so that when the bias values at different locations are applied to the ADI profile, the bias vectors at different locations do not intersect each other. In an embodiment, in the non-transitory computer readable medium, the bias direction includes: a direction that is not perpendicular to the ADI profile.
[0161] In an embodiment, in the non-transitory computer readable medium, the PDF of the particle represents a deposition or etching process of the particle on the ADI profile, and wherein the deposition rate is positive or negative. In an embodiment, in the non-transitory computer readable medium, the bias value is obtained from a trained machine learning model configured to generate a bias map for a given resist pattern. In an embodiment, in the non-transitory computer readable medium, when the ADI pattern includes multiple profiles, a set of bias vectors is determined separately for each ADI profile.
[0162] In an embodiment, a non-transitory computer-readable medium is provided, comprising instructions that, when executed by one or more processors, result in operations including: obtaining (i) a probability distribution function (PDF) corresponding to a process to be performed on a contour, and (ii) a contour function characterizing the shape of the contour; convolving the contour function with the PDF over an area of the contour to determine a process rate at a specified location on the contour; and determining a bias vector to be applied to the contour based on the process rate to generate a biased contour indicating the effect of the process applied to the contour.
[0163] In an embodiment, in the non-transitory computer readable medium, the process results in the addition or removal of material forming the contour, the addition or the removal resulting in a change in the shape of the contour. In an embodiment, in the non-transitory computer readable medium, the process rate characterizes the behavior of the addition or the removal of the material forming the contour.
[0164] In an embodiment, the trained machine learning model can be used for various applications related to the patterning process to improve the yield of the patterning process. For example, method 300 also includes: predicting a substrate image for a design layout via a trained machine learning model; determining a mask layout to be used to manufacture a mask for a patterning process via an OPC simulation using the design layout and the predicted substrate image. In an embodiment, the OPC simulation includes simulating a patterning process model using the geometry of the design layout and corrections associated with multiple fragments to determine a simulated pattern to be printed on a substrate; and determining an optical proximity effect correction for the design layout so as to reduce the difference between the simulated pattern and the design layout. In an embodiment, determining the optical proximity correction is an iterative process. Iteration includes adjusting the shape and / or size of the geometry of the main features and / or one or more auxiliary features of the design layout so as to reduce the performance index of the patterning process. In an embodiment, one or more auxiliary features are extracted from the predicted OPC image of the machine learning model.
[0165] In some embodiments, the inspection device may be a scanning electron microscope (SEM) that produces images of structures exposed or transferred onto the substrate (eg, some or all structures of the device). Fig.17 An embodiment of a SEM tool is depicted.A primary electron beam EBP emitted from an electron source ESO is converged by a condenser lens CL and then passes through a beam deflector EBD1 , an E×B deflector EBD2 , and an objective lens OL to illuminate a substrate PSub on a substrate stage ST at a focal length.
[0166] When the substrate PSub is irradiated with the electron beam EBP, secondary electrons are generated from the substrate PSub. The secondary electrons are deflected by the E×B deflector EBD2 and detected by the secondary electron detector SED. A two-dimensional electron beam image can be obtained by detecting the electrons generated from the sample in synchronization with, for example, two-dimensional scanning of the electron beam by the beam deflector EBD1 in the X or Y direction or with repeated scanning of the electron beam EBP by the beam deflector EBD1, and continuously moving the substrate PSub in the other direction of the X or Y direction by the substrate stage ST.
[0167] The signal detected by the secondary electron detector SED is converted into a digital signal by an analog / digital (A / D) converter ADC, and the digital signal is sent to the image processing system IPU. In an embodiment, the image processing system IPU may have a memory MEM to store all or part of the digital image for processing by the processing unit PU. The processing unit PU (e.g., specially designed hardware, or a combination of hardware and software) is configured to convert or process the digital image into a data set representing the digital image. In addition, the image processing system IPU may have a storage medium STOR configured to store the digital image and the corresponding data set in a reference database. The display device DIS may be connected to the image processing system IPU so that the operator can perform the necessary operations of the equipment with the aid of a graphical user interface.
[0168] As mentioned above, SEM images can be processed to extract contours in the image that describe the edges of objects representing device structures. These contours are then quantified via indicators such as CD. Thus, images of device structures are usually compared and quantified via simplistic indicators such as distance between edges (CD) or simple pixel differences between images. Typical contour models that detect the edges of objects in images in order to measure CD use image gradients. In practice, those models rely on strong image gradients. But in practice, images are usually noisy and have discontinuous boundaries. Techniques such as smoothing, adaptive thresholding, edge detection, erosion and dilation can be used to process the results of image gradient contour models to address noisy and discontinuous images, but will ultimately result in low-resolution quantization of high-resolution images. Thus, in most instances, mathematical processing of images of device structures to reduce noise and automated edge detection will result in a loss of resolution of the image, thereby resulting in a loss of information. Therefore, the result is a low-resolution quantization equivalent to an oversimplified representation of a complex high-resolution structure.
[0169] Thus, it is desirable to have a mathematical representation that can preserve resolution and can also describe the general shape of a structure (e.g., a circuit feature, an alignment mark, or a portion of a measurement target (e.g., a grating feature), etc.) produced or expected to be produced using a patterning process, whether, for example, the structure is in a latent resist image, in a developed resist image, or in a layer transferred to a substrate, such as by etching. In the context of a lithography or other patterning process, the structure can be a device being manufactured or a portion thereof, and the image can be a SEM image of the structure. In some cases, the structure can be a feature of a semiconductor device (e.g., an integrated circuit). In this case, the structure can be referred to as a pattern or expected pattern that includes multiple features of the semiconductor device. In some cases, the structure can be an alignment mark or a portion thereof (e.g., a grating of an alignment mark) used in an alignment measurement process to determine the alignment of an object (e.g., a substrate) with another object (e.g., a patterning device), or a measurement target or a portion thereof (e.g., a grating of a measurement target) used to measure parameters of the patterning process (e.g., overlay, focus, dose, etc.). In an embodiment, the measurement target is a diffraction grating used to measure (e.g.,) overlay.
[0170] Fig.18 Another embodiment of the inspection device is schematically illustrated. The system is used to inspect a sample 90 (such as a substrate) on a sample platform 88 and includes a charged particle beam generator 81, a condenser lens module 82, a probe forming objective lens module 83, a charged particle beam deflection module 84, a secondary charged particle detector module 85, and an image forming module 86.
[0171] The charged particle beam generator 81 generates a primary charged particle beam 91. The condenser lens module 82 condenses the generated primary charged particle beam 91. The probe forming objective lens module 83 focuses the converged primary charged particle beam into a charged particle beam probe 92. The charged particle beam deflection module 84 scans the formed charged particle beam probe 92 across the surface of the area of interest on the sample 90 fastened on the sample platform 88. In an embodiment, the charged particle beam generator 81, the condenser lens module 82 and the probe forming objective lens module 83 or their equivalent designs, alternatives or any combination thereof together form a charged particle beam probe generator that generates a scanning charged particle beam probe 92.
[0172] The secondary charged particle detector module 85 detects secondary charged particles 93 emitted from the sample surface after being bombarded by the charged particle beam probe 92 (and possibly together with other reflected or scattered charged particles from the sample surface) to generate a secondary charged particle detection signal 94. An image forming module 86 (e.g., a computing device) is coupled to the secondary charged particle detector module 85 to receive the secondary charged particle detection signal 94 from the secondary charged particle detector module 85 and to form at least one scan image accordingly. In an embodiment, the secondary charged particle detector module 85 and the image forming module 86 or their equivalent designs, alternatives, or any combination thereof together form an image forming device that forms a scan image from the detected secondary charged particles emitted by the sample 90 bombarded by the charged particle beam probe 92.
[0173] In an embodiment, the monitoring module 87 is coupled to the image forming module 86 of the image forming device to monitor, control, etc. the patterning process and / or use the scanned image of the sample 90 received from the image forming module 86 to derive parameters for patterning process design, control, monitoring, etc. Therefore, in an embodiment, the monitoring module 87 is configured or programmed to cause the method described herein to be performed. In an embodiment, the monitoring module 87 includes a computing device. In an embodiment, the monitoring module 87 includes a computer program encoded on a computer-readable medium forming or disposed within the above-mentioned monitoring module 87 to provide the functionality herein.
[0174] In an embodiment, similar to using a probe to inspect a substrate Fig.17 Electron beam inspection tools, Fig.18 The electron current in a system such as Fig.17 The inspection device discussed above may be a single beam device or a multi-beam device, which is significantly larger than the CD SEM depicted in FIG, so that the detection spot is large enough so that the inspection speed can be faster. However, due to the large detection spot, the resolution may not be as high as compared to the CD SEM. In an embodiment, without limiting the scope of the present disclosure, the inspection device discussed above may be a single beam device or a multi-beam device.
[0175] Can process from e.g. Fig.17 and / or Fig.18 The SEM images of the system are extracted to extract contours describing the edges of the objects (representing the device structures) in the image. These contours are then quantified via a metric such as CD, typically at a user-defined tangent line. Thus, the images of the device structures are typically compared and quantified via a metric such as the edge-to-edge distance (CD) measured for the extracted contours or a simple pixel difference between the images.
[0176] Fig.191 is a block diagram illustrating a computer system 100 that can assist in implementing the methods, processes, or apparatus disclosed herein. The computer system 100 includes a bus 102 or other communication mechanism for communicating information, and a processor 104 (or multiple processors 104 and 105) coupled to the bus 102 for processing information. The computer system 100 also includes a main memory 106, such as a random access memory (RAM) or other dynamic storage device, coupled to the bus 102 for storing information and instructions to be executed by the processor 104. The main memory 106 can also be used to store temporary variables or other intermediate information during the execution of instructions to be executed by the processor 104. The computer system 100 also includes a read-only memory (ROM) 108 or other static storage device coupled to the bus 102 for storing static information and instructions for the processor 104. A storage device 110, such as a magnetic disk or optical disk, is provided and coupled to the bus 102 for storing information and instructions.
[0177] The computer system 100 can be coupled by bus 102 to a display 112 for displaying information to a computer user, such as a cathode ray tube (CRT) or a flat panel display or a touch panel display. An input device 114 including alphanumeric keys and other keys is coupled to bus 102 for communicating information and command selections to processor 104. Another type of user input device is a cursor control 116, such as a mouse, trackball, or cursor direction keys, for communicating directional information and command selections to processor 104 and for controlling cursor movement on display 112. Such an input device typically has two degrees of freedom in two axes: a first axis (e.g., x) and a second axis (e.g., y), which allows the device to specify a position in a plane. A touch panel (screen) display can also be used as an input device.
[0178] According to one embodiment, portions of the process may be performed by computer system 100 in response to processor 104 executing one or more sequences of one or more instructions contained in main memory 106. Such instructions may be read into main memory 106 from another computer-readable medium, such as storage device 110. Execution of the sequence of instructions contained in main memory 106 causes processor 104 to perform the process steps described herein. One or more processors in a multi-processing arrangement may also be used to execute the sequence of instructions contained in main memory 106. In alternative embodiments, hardwired circuitry may be used instead of or in combination with software instructions. Therefore, the description herein is not limited to any specific combination of hardware circuitry and software.
[0179] The term "computer-readable medium" as used herein refers to any medium that participates in providing instructions to the processor 104 for execution. Such media can take many forms, including but not limited to non-volatile media, volatile media, and transmission media. Non-volatile media include, for example, optical or magnetic disks, such as storage device 110. Volatile media include volatile memory, such as main memory 106. Transmission media include coaxial cables, copper wires, and optical fibers, including wires that contain bus 102. Transmission media can also take the form of sound waves or light waves, such as sound waves or light waves generated during radio frequency (RF) and infrared (IR) data communications. Common forms of computer-readable media include, for example, floppy disks, flexible disks, hard disks, magnetic tapes, any other magnetic media, CD-ROMs, DVDs, any other optical media, punch cards, paper tapes, any other physical media with hole patterns, RAM, PROMs and EPROMs, FLASH-EPROMs, any other memory chips or cartridges, carrier waves as described below, or any other media that can be read by a computer.
[0180] Various forms of computer-readable media may be involved when carrying one or more sequences of one or more instructions to processor 104 for execution. For example, the instructions may initially be carried on a disk of a remote computer. The remote computer may load the instructions into its volatile memory and send the instructions via a telephone line using a modem. A modem local to computer system 100 may receive data on the telephone line and use an infrared transmitter to convert the data into an infrared signal. An infrared detector coupled to bus 102 may receive the data carried in the infrared signal and place the data on bus 102. Bus 102 carries the data to main memory 106, from which processor 104 retrieves and executes instructions. The instructions received by main memory 106 may optionally be stored on storage device 110 before or after being executed by processor 104.
[0181] The computer system 100 also desirably includes a communication interface 118 coupled to the bus 102. The communication interface 118 provides a bidirectional data communication coupling to a network link 120, which is connected to a local area network 122. For example, the communication interface 118 can be an integrated services digital network (ISDN) card or a modem to provide a data communication connection to a corresponding type of telephone line. As another example, the communication interface 118 can be a local area network (LAN) card to provide a data communication connection to a compatible LAN. A wireless link can also be implemented. In any such implementation, the communication interface 118 sends and receives electrical, electromagnetic, or optical signals that carry digital data streams representing various types of information.
[0182] The network link 120 typically provides data communication to other data devices via one or more networks. For example, the network link 120 may provide a connection from a local area network 122 to a host computer 124 or to data equipment operated by an Internet Service Provider (ISP) 126. The ISP 126 in turn provides data communication services via a global packet data communication network (now commonly referred to as the "Internet") 128. Both the local area network 122 and the Internet 128 use electrical, electromagnetic or optical signals that carry digital data streams. The signals through the various networks and the signals on the network link 120 and through the communication interface 118, which carry the digital data to and from the computer system 100, are exemplary forms of carrier waves that transport the information.
[0183] The computer system 100 can send messages and receive data, including program code, via the network, network link 120, and communication interface 118. In the Internet example, server 130 may transmit requested program code for an application via Internet 128, ISP 126, local area network 122, and communication interface 118. One such downloaded application may provide, for example, irradiation optimization of an embodiment. The received code may be executed by processor 104 as it is received, and / or stored in storage device 110 or other non-volatile storage for later execution. In this manner, computer system 100 may obtain application code in the form of a carrier wave.
[0184] Fig. 20 An exemplary lithographic projection apparatus that can be utilized in conjunction with the techniques described herein is schematically depicted. The apparatus comprises:
[0185] - an illumination system IL, said illumination system being used to condition the radiation beam B. In such a particular case, the illumination system also comprises a radiation source SO;
[0186] a first stage (e.g. patterning device table) MT provided with a patterning device holder for holding a patterning device MA (e.g. a reticle) and connected to a first positioner for accurately positioning the patterning device relative to the object PS;
[0187] a second stage (substrate table) WT provided with a substrate holder for holding a substrate W (e.g. a silicon wafer coated with resist) and connected to a second positioner for accurately positioning the substrate relative to the item PS;
[0188] - a projection system ("lens") PS (eg a refractive, reflective or catadioptric optical system) for imaging the irradiated portion of the patterning device MA onto a target portion C of the substrate W (eg comprising one or more dies).
[0189] As depicted herein, the device is of the transmissive type (i.e., has a transmissive patterning device). However, in general, the device may also be of, for example, the reflective type (with a reflective patterning device). The device may use a different kind of patterning device than a classical mask; examples include a programmable mirror array or an LCD matrix.
[0190] A source SO (e.g. a mercury lamp or an excimer laser, a laser produced plasma (LPP) EUV source) generates a radiation beam. This beam is fed into an illumination system (illuminator) IL, for example directly or after having traversed a conditioning device such as a beam expander Ex. The illuminator IL may include an adjustment device AD for setting the outer radial extent and / or the inner radial extent (commonly referred to as σ-outer and σ-inner, respectively) of the intensity distribution in the beam. In addition, the illuminator IL will typically include various other components, such as an integrator IN and a condenser CO. In this way, the beam B impinging on the patterning device MA has a desired uniformity and intensity distribution in its cross section.
[0191] about Fig. 20 It should be noted that the source SO can be within the housing of the lithographic projection device (this is often the case when the source SO is, for example, a mercury lamp), but the source can also be remote from the lithographic projection device, the radiation beam generated by the source being guided into the device (for example with the aid of suitable directional reflectors); this latter situation is often the case when the source SO is an excimer laser (for example based on KrF, ArF or F2 laser action).
[0192] The beam PB then intercepts the patterning device MA which is held on the patterning device table MT. Having traversed the patterning device MA, the beam B passes through a lens PL which focuses the beam B onto a target portion C of the substrate W. With the aid of the second positioning device (and the interferometry device IF), the substrate table WT can be accurately moved, for example in order to position a different target portion C in the path of the beam PB. Similarly, the first positioning device can be used to accurately position the patterning device MA relative to the path of the beam B, for example after mechanical retrieval of the patterning device MA from a patterning device library or during scanning. Typically, the patterning device MA will be accurately positioned with the aid of the second positioning device (and the interferometry device IF) which is not in the path of the beam PB. Fig. 20 Movement of the stage MT, WT is achieved by a long-stroke module (coarse positioning) and a short-stroke module (fine positioning) explicitly depicted in FIG. 2. However, in the case of a stepper (as opposed to a step-and-scan tool), the patterning device table MT may be connected to the short-stroke actuator only, or may be fixed.
[0193] The depicted tools can be used in two different modes:
[0194] - in step mode, the patterning device table MT is kept substantially stationary and the entire patterning device image is projected onto a target portion C at one time (i.e. a single "flash"). The substrate table WT is then shifted in the x-direction and / or y-direction so that a different target portion C can be irradiated by the beam PB;
[0195] - In scan mode, essentially the same situation applies, except that a given target portion C is not exposed in a single "flash". Instead, the patterning device table MT can be moved in a given direction (the so-called "scanning direction", e.g. the y-direction) at a speed v, so that the projection beam B is scanned over the entire patterning device image; in parallel, the substrate table WT is simultaneously moved in the same direction or in the opposite direction at a speed V=Mv, where M is the magnification of the lens PL (typically M=1 / 4 or 1 / 5). In this way, a relatively large target portion C can be exposed without having to compromise resolution.
[0196] Fig.21 Another exemplary lithographic projection apparatus 1000 is schematically depicted, comprising:
[0197] - a source collector module SO, said source collector module being adapted to provide radiation;
[0198] - an illumination system (illuminator) IL configured to condition a radiation beam B (e.g. EUV radiation) coming from the source collector module SO;
[0199] a support structure (e.g., mask table) MT configured to support a patterning device (e.g., mask or reticle) MA and connected to a first positioner PM configured to accurately position the patterning device;
[0200] a substrate table (e.g. wafer stage) WT configured to hold a substrate (e.g. a resist-coated wafer) W and connected to a second positioner PW configured to accurately position the substrate; and
[0201] - a projection system (eg, a reflective projection system) PS configured to project the pattern imparted to the radiation beam B by the patterning device MA onto a target portion C of the substrate W (eg, comprising one or more dies).
[0202] As depicted here, the apparatus 1000 is of the reflective type (e.g., using a reflective mask). It should be noted that because most materials are absorptive in the EUV wavelength range, the patterning device can have a multilayer reflector comprising, for example, multiple stacked layers of molybdenum and silicon. In one example, the multi-stacked layer reflector has 40 layer pairs of molybdenum and silicon, where each layer is a quarter wavelength thick. X-ray lithography can be used to produce smaller wavelengths. Since most materials are absorptive at EUV and x-ray wavelengths, a thin piece of patterned absorbing material on the patterning device topography (e.g., a TaN absorber on top of a multilayer reflector) defines where features will be printed (positive resist) or not (negative resist).
[0203] refer to Fig.21 , the illuminator IL receives a beam of extreme ultraviolet radiation from a source collector module SO. Methods for generating EUV radiation include, but are not necessarily limited to, converting a material having at least one element (e.g., xenon, lithium, or tin) into a plasma state using one or more emission lines in the EUV range. In one such method, often referred to as laser produced plasma "LPP", a plasma may be generated by irradiating a fuel (e.g., a droplet, stream, or cluster of material having the line emitting element) with a laser beam. The source collector module SO may be a laser ( Fig.21 The laser is used to provide a laser beam for exciting the fuel. The resulting plasma emits output radiation, such as EUV radiation, which is collected using a radiation collector disposed in a source collector module. For example, when a CO2 laser is used to provide a laser beam for fuel excitation, the laser and the source collector module may be separate entities.
[0204] In such cases, the laser is not considered to form part of the lithographic apparatus, and the radiation beam is delivered from the laser to the source collector module by means of a beam delivery system comprising, for example, suitable directing mirrors and / or a beam expander. In other cases, for example, when the radiation source is a discharge produced plasma EUV generator (often referred to as a DPP radiation source), the radiation source may be an integral part of the source collector module.
[0205] The illuminator IL may include an adjuster for adjusting the angular intensity distribution of the radiation beam. Typically, at least the outer radial extent and / or the inner radial extent (commonly referred to as σ-outer and σ-inner, respectively) of the intensity distribution in a pupil plane of the illuminator may be adjusted. In addition, the illuminator IL may include various other elements, such as a faceted field mirror arrangement and a faceted pupil mirror arrangement. The illuminator may be used to adjust the radiation beam to have a desired uniformity and intensity distribution in its cross-section.
[0206] The radiation beam B is incident on a patterning device (e.g. a mask) MA held on a support structure (e.g. a mask table) MT and is patterned by the patterning device. After reflection from the patterning device (e.g. a mask) MA, the radiation beam B passes through a projection system PS which focuses the beam onto a target portion C of the substrate W. With the aid of a second positioner PW and a position sensor PS2 (e.g. an interferometric device, a linear encoder or a capacitive sensor), the substrate table WT can be accurately moved, for example so that different target portions C are positioned in the path of the radiation beam B. Similarly, a first positioner PM and a further position sensor PS1 can be used to accurately position the patterning device (e.g. a mask) MA relative to the path of the radiation beam B. The patterning device (e.g. a mask) MA and the substrate W can be aligned using patterning device alignment marks M1, M2 and substrate alignment marks P1, P2.
[0207] The depicted device 1000 may be used in at least one of the following modes:
[0208] 1. In step mode, the support structure (e.g. mask table) MT and substrate table WT are held substantially stationary (i.e. a single static exposure) while the entire pattern imparted to the radiation beam is projected at one time onto a target portion C. Next, the substrate table WT is shifted in the X and / or Y direction so that a different target portion C can be exposed.
[0209] 2. In scan mode, the support structure (e.g. mask table) MT and substrate table WT are scanned synchronously (i.e. single dynamic exposure) as a pattern imparted to the radiation beam is projected onto a target portion C. The speed and direction of the substrate table WT relative to the support structure (e.g. mask table) MT may be determined by the (minimum) magnification and image reversal characteristics of the projection system PS.
[0210] 3. In another mode, the support structure (e.g. mask table) MT is kept substantially stationary, thereby holding the programmable patterning device, and the substrate table WT is moved or scanned, while a pattern imparted to the radiation beam is projected onto a target portion C. In this mode, a pulsed radiation source is typically used, and the programmable patterning device is updated as required after each movement of the substrate table WT or between successive radiation pulses during a scan. This mode of operation may be readily applicable to maskless lithography using a programmable patterning device, such as a programmable mirror array of the type mentioned above.
[0211] Fig. 22The apparatus 1000 is shown in more detail, including a source collector module SO, an illumination system IL and a projection system PS. The source collector module SO is constructed and arranged so that a vacuum environment can be maintained in an enclosure 220 of the source collector module SO. An EUV radiation emitting plasma 210 can be formed by a discharge-generated plasma radiation source. EUV radiation can be generated by a gas or vapor (e.g., Xe gas, Li vapor or Sn vapor), wherein a very hot plasma 210 is generated to emit radiation in the EUV range of the electromagnetic spectrum. For example, a very hot plasma 210 is generated by a discharge that causes at least a partially ionized plasma. For efficient generation of radiation, a partial pressure of, for example, 10 Pa of Xe, Li, Sn vapor or any other suitable gas or vapor may be required. In an embodiment, an excited tin (Sn) plasma is provided to generate EUV radiation.
[0212] The radiation emitted by the hot plasma 210 is transferred from the source chamber 211 to the collector chamber 212 via an optional gas barrier or contaminant trap 230 (also referred to as a contaminant barrier or foil trap in some cases) positioned in or behind an opening in the source chamber 211. The contaminant trap 230 may include a channel structure. The contaminant trap 230 may also include a gas barrier, or a combination of a gas barrier and a channel structure. As is known in the art, the contaminant trap or contaminant barrier 230 further indicated herein includes at least a channel structure.
[0213] The collector chamber 211 may include a radiation collector CO which may be a so-called grazing incidence collector. The radiation collector CO has an upstream radiation collector side 251 and a downstream radiation collector side 252. Radiation traversing the collector CO may be reflected from the grating spectral filter 240 to be focused into a virtual source point IF along the optical axis indicated by the dotted line "O". The virtual source point IF is often referred to as an intermediate focus, and the source collector module is arranged so that the intermediate focus IF is located at or near an opening 221 in the enclosure structure 220. The virtual source point IF is an image of the radiation emitting plasma 210.
[0214] The radiation then traverses an illumination system IL which may include a faceted field mirror arrangement 22 and a faceted pupil mirror arrangement 24 arranged to provide a desired angular distribution of the radiation beam 21 at the patterning device MA, and a desired uniformity of radiation intensity at the patterning device MA. After reflection of the radiation beam 21 at the patterning device MA held by the support structure MT, a patterned beam 26 is formed and is imaged by the projection system PS via reflective elements 28, 30 onto a substrate W held by a substrate table WT.
[0215] More elements than shown may typically be present in the illumination optics unit IL and the projection system PS. Depending on the type of lithographic apparatus, a grating spectral filter 240 may optionally be present. In addition, more mirrors may be present than shown in the figures, for example, more than 100 mirrors may be present in the projection system PS. Fig. 22 The reflective elements shown have from 1 to 6 additional reflective elements.
[0216] like Fig. 22 The collector optics CO illustrated in FIG. 1 is depicted as a nested collector with grazing incidence reflectors 253, 254 and 255, merely as an example of a collector (or collector mirror). The grazing incidence reflectors 253, 254 and 255 are arranged axially symmetrically around the optical axis O, and this type of collector optics CO is desirably used in combination with a discharge produced plasma radiation source.
[0217] Alternatively, the source collector module SO may be as follows Fig.23 Components of the LPP radiation system shown. The laser LAS is arranged to deposit laser energy into a fuel such as xenon (Xe), tin (Sn) or lithium (Li), thereby producing a highly ionized plasma 210 with an electron temperature of tens of electron volts. The high-energy radiation produced during the deexcitation and recombination of these ions is emitted from the plasma, collected by the near normal incidence collector optics CO, and focused onto an opening 221 in the enclosure structure 220.
[0218] Embodiments of the present disclosure are further described in the following aspects.
[0219] 1. A method of producing a metrology gauge for measuring a physical property of a structure printed on a substrate, the method comprising:
[0220] obtaining (i) measurement data associated with the physical property of the structure printed on the substrate, and (ii) at least a portion of a simulated profile of the structure, the portion of the simulated profile associated with the measurement data;
[0221] modifying the portion of the simulated profile of the structure based on the measured data; and
[0222] The metrology gauge is generated on or near the modified portion of the simulated contour, the metrology gauge being positioned to measure the physical property of the simulated contour of the structure.
[0223] 2. A method according to aspect 1, wherein the portion of the simulated contour is the portion of the simulated contour that is within a defined area around the measurement data associated with the structure.
[0224] 3. The method according to aspect 1, wherein obtaining the portion of the simulated profile comprises:
[0225] confining an area of the substrate around a defined location associated with the measurement data; and
[0226] A patterning process is simulated within the defined area of the substrate to obtain the portion of the simulated outline of the structure.
[0227] 4. The method according to any one of aspects 1 to 3, wherein modifying the portion of the simulation profile comprises:
[0228] determining simulation data associated with the physical property of the simulated profile of the structure based on the portion of the simulated profile;
[0229] determining a difference between the measured data and the simulated data associated with the physical property of the structure; and
[0230] The portion of the simulated profile is modified based on the difference such that the difference between the measured data and the simulated data is reduced.
[0231] 5. A method according to any one of aspects 1 to 4, wherein the measurement data is a CD value at a defined location associated with the structure.
[0232] 6. A method according to any one of clauses 5, wherein the modification of the portion of the simulated profile is based on a difference between a simulated CD value and a measured CD value associated with the structure.
[0233] 7. The method according to any one of aspects 1 to 6, wherein modifying the portion of the simulation profile comprises:
[0234] determining simulation data associated with the physical property of the simulated profile of the structure based on the portion of the simulated profile;
[0235] determining a difference between the measured data and the simulated data associated with the physical property of the structure; and
[0236] A threshold used to generate the simulated profile is adjusted based on the difference such that the difference between the measured data and the simulated data is reduced, wherein the adjusted threshold modifies the portion of the simulated profile.
[0237] 8. The method according to any one of aspects 1 to 7, wherein modifying the portion of the simulation profile comprises:
[0238] determining a simulated CD value at a defined location associated with a measured CD value using the portion of the simulated profile;
[0239] determining a difference between the simulated CD value and the measured CD value; and
[0240] The threshold is adjusted based on the difference such that the difference between the simulated CD value and the measured CD value decreases, the adjusted threshold modifying the portion of the simulated profile.
[0241] 9. The method according to any one of aspects 1 to 8, wherein generating the measurement gauge comprises:
[0242] specifying points along the modified portion of the simulated profile; and
[0243] The location of the point is exported as the measuring gauge.
[0244] 10. The method according to any one of aspects 1 to 9, wherein the measurement data is obtained via a metrology tool.
[0245] 11. The method of clause 9, wherein the metrology tool is a scanning electron microscope (SEM) and the measurement data is obtained from a SEM image.
[0246] 12. The method according to any one of aspects 1 to 11, wherein the measuring gauge is an edge placement gauge and / or a CD gauge.
[0247] 13. A method for determining a hot spot location associated with a substrate, the method comprising:
[0248] obtaining (i) a simulated profile associated with one or more patterns, the simulated profile associated with measured data of physical properties of the one or more patterns printed on the substrate, and (ii) a metrology gauge associated with the simulated profile;
[0249] determining a value of the physical property associated with the one or more patterns based on the metrology gauge; and
[0250] The hot spot locations on the substrate are determined based on physical property values, wherein the hot spot locations are locations on the substrate where the physical property values are less than a hot spot threshold associated with the one or more patterns.
[0251] 14. The method according to aspect 13, wherein obtaining the measurement gauge comprises:
[0252] determining a simulated profile associated with the one or more patterns by simulating a patterning process using the measured data;
[0253] modifying at least a portion of the simulated profile based on the measurement data associated with the one or more patterns; and
[0254] The metrology gauge is generated along the modified portion of the simulated profile.
[0255] 15. The method according to any one of aspects 13 to 14, wherein determining the value of the physical property comprises:
[0256] The value of the physical property is measured at one or more of the gages.
[0257] 16. The method according to aspect 15, wherein determining the hot spot location comprises:
[0258] determining whether a value of the physical property associated with the one or more patterns exceeds the hot spot threshold;
[0259] In response to breaching the threshold, a location of the measurement gauge associated with breaching the threshold is identified.
[0260] 17. A method for training a model associated with a patterning process, the method comprising:
[0261] obtaining (i) measurement data associated with a physical property of a structure printed on the substrate, and (ii) a metrology gauge associated with a simulated profile of the structure to be printed on the substrate, the simulated profile associated with a defined location on the substrate at which the physical property is measured; and
[0262] Training the model using the measurement data and the metrology gauge improves a performance metric of the patterning process around the defined location on the substrate, the performance metric being a function of the metrology gauge and the physical property.
[0263] 18. The method according to aspect 17, wherein said training said model is an iterative process, the iteration comprising:
[0264] determining, via executing the model, a simulated profile of the structure to be printed on the substrate and simulation data associated with the physical properties of the simulated profile of the structure;
[0265] determining a first difference between the simulated data and the measured data, and a second difference between a point along the simulated profile and the measurement gauge; and
[0266] Model parameters are determined based on a gradient of the performance indicator with parameters of the patterning process so as to minimize the performance indicator, the performance indicator being a function of the first difference and the second difference.
[0267] 19. The method of clause 18, wherein the model is at least one of:
[0268] an etching model configured to predict an etching image; or
[0269] A resist model configured to predict a resist image.
[0270] 20. A method of producing a metrology gauge for measuring a physical property of a structure on a substrate, the method comprising:
[0271] obtaining (i) measurement data associated with the physical property of the structure printed on the substrate, and (ii) at least a portion of a simulated profile of the structure, the portion of the simulated profile associated with the measurement data;
[0272] generating a modified profile of the portion of the simulated profile of the structure based on the measured data; and
[0273] The modified profile is provided to a model of the patterning process to determine parameters of the patterning process.
[0274] 21. A method according to aspect 20, wherein said generating a modified profile of said portion of said simulated profile comprises:
[0275] determining simulation data associated with the physical property of the simulated profile of the structure based on the portion of the simulated profile;
[0276] determining a difference between the measured data and the simulated data associated with the physical property of the structure; and
[0277] The portion of the simulated profile is modified based on the difference such that the difference between the measured data and the simulated data is reduced.
[0278] 22. A computer program product, comprising a non-transitory computer-readable medium having instructions recorded thereon, the instructions implementing the method according to any one of the above aspects when executed by a computer.
[0279] 23. A method of training a machine learning model associated with a patterning process, the method comprising:
[0280] obtaining (i) profile data of an after-development image (ADI) pattern on a substrate, (ii) measurement data of an after-etching image (AEI) pattern printed on the substrate, and (iii) obtaining a reference bias value based on the profile data of the ADI pattern and the measurement data of the AEI pattern; and
[0281] The machine learning model is trained using the measurement data and the profile data as training data to determine bias values to be applied to the ADI profile.
[0282] 24. The method according to aspect 23, wherein the training comprises:
[0283] A model parameter of the machine learning model is adjusted so that the bias value is within a specified range determined based on the reference bias value.
[0284] 25. A method according to aspect 23, wherein the machine learning model is configured to generate a representation of a bias map for the ADI profile.
[0285] 26. The method of clause 23, wherein the contour data represents an image of a contour associated with one or more features of the ADI pattern.
[0286] 27. A method according to clause 26, wherein the image is generated from a simulated profile of a simulated ADI pattern and / or obtained from a metrology device configured to capture an image of the substrate after a process of preforming a resist on the substrate.
[0287] 28. The method according to any one of aspects 23 to 27, wherein obtaining the profile data comprises:
[0288] executing one or more process models associated with the patterning process using a design pattern to be printed on the substrate as input to generate the simulated ADI pattern, the patterning process including a resist process; and
[0289] Contours are extracted from the simulated ADI pattern, each contour being a contour of a feature within the simulated ADI pattern.
[0290] 29. The method of clause 28, wherein the one or more process models include at least one of the following:
[0291] an optics model configured to determine an aerial image; and
[0292] A resist model is configured to determine a resist image.
[0293] 30. The method of any one of clauses 23 to 29, wherein the measurement data is obtained at a metrology gauge, the metrology gauge being an edge placement gauge, and / or a critical dimension (CD) gauge associated with the AEI pattern.
[0294] 31. The method of aspect 28, wherein the measurement data at the metrology gauge comprises:
[0295] a location of the edge placement gauge associated with an outline of the AEI pattern printed on the substrate; and / or
[0296] A CD value associated with the AEI pattern printed on the substrate.
[0297] 32. A method according to aspect 28, wherein when the measuring gauge is a CD gauge, the reference offset value is obtained via a calibration process, the calibration process being configured to determine an offset value associated with a given CD gauge, the offset value indicating an amount of CD reduction to be applied to the ADI pattern to produce the AEI pattern.
[0298] 33. A method according to aspect 32, wherein the offset values are provided at the ends of the given CD gauge, the offset values being unequal or asymmetric about the center of the CD gauge.
[0299] 34. A method according to any one of aspects 23 to 33, wherein the training of the machine learning model is an iterative process, the iteration comprising:
[0300] (a) executing the machine learning model using the measurement data, the profile data and given values of the model parameters to produce a bias map associated with the profile data, the bias map comprising bias values;
[0301] (b) based on the gradient of the difference between the model-based bias value and the reference bias value, adjusting the model parameters of the machine learning model so as to reduce the difference; and
[0302] (c) performing steps (a) to (b) until the difference is minimized.
[0303] 35. A method according to aspects 23 to 34, wherein the bias map generated via a trained machine learning model includes etch bias values to be applied to a resist profile to determine an etch profile to be printed on the substrate.
[0304] 36. The method of clause 35, wherein the etch bias value is applied to the resist profile in a local normal direction.
[0305] 37. A method according to any one of clauses 23 to 36, wherein the bias map is a pixelated image, each pixel having an intensity value indicating a bias value.
[0306] 38. A method for determining a bias vector associated with an after developed image (ADI) pattern, the method comprising:
[0307] obtaining (i) a probability distribution function (PDF) corresponding to particles within the ADI pattern deposited on a substrate, and (ii) a profile function characterizing an ADI profile associated with the ADI pattern;
[0308] determining a deposition rate of the particle at a specified location on the ADI profile based on a combination of the PDF of the particle over a region of the ADI profile and the profile function; and
[0309] A bias vector associated with the ADI pattern is determined based on the deposition rate, the bias vector producing an after-etch image (AEI) profile when applied to the ADI profile of the ADI pattern.
[0310] 39. The method of clause 38, wherein the obtaining the probability distribution function (PDF) of particles is based on measured substrate data, the measured substrate data comprising deposition data of particles and a measured etch pattern.
[0311] 40. The method of clause 39, wherein obtaining the PDF comprises determining the variance of a Gaussian distribution fitted to the measurement data.
[0312] 41. The method of any one of aspects 38 to 40, wherein said determining the deposition rate of said particles comprises:
[0313] convolving the PDF of the particle with the profile function; and
[0314] An integration is performed over the area of the ADI profile.
[0315] 42. The method of any one of aspects 38 to 41, wherein determining the bias vector comprises:
[0316] A gradient of the deposition rate with respect to a first direction and a second direction of the ADI pattern is determined, the first direction and the second direction being perpendicular to each other.
[0317] 43. The method of clause 38, wherein the bias vector comprises: a bias direction at a location of the ADI profile, and the method further comprises applying a bias value along to generate the AEI profile.
[0318] 44. A method according to aspect 43, wherein the bias direction is determined such that when the bias values at different locations are applied to the ADI profile, the bias vectors at different locations do not intersect each other.
[0319] 45. The method of clause 44, wherein the bias direction comprises a direction that is not perpendicular to the ADI profile.
[0320] 46. The method of any one of aspects 38 to 45, wherein the PDF of the particle represents a deposition or etching process of the particle on the ADI profile, and wherein the deposition rate is positive or negative.
[0321] 47. A method according to any one of aspects 43 to 46, wherein the bias value is obtained from a trained machine learning model configured to generate a bias map for a given resist pattern.
[0322] 48. A method according to any one of clauses 38 to 47, wherein when the ADI pattern comprises a plurality of profiles, a set of bias vectors is determined separately for each ADI profile.
[0323] 49. A method for determining an offset vector for a contour, the method comprising:
[0324] obtaining (i) a probability distribution function (PDF) corresponding to a process to be performed on the contour, and (ii) a contour function characterizing a shape of the contour;
[0325] convolving the profile function with the PDF over an area of the profile to determine a process rate at a specified location on the profile; and
[0326] A bias vector is determined to be applied to the profile based on the process rate to produce a bias profile indicative of an effect of the process applied to the profile.
[0327] 50. The method of aspect 49, wherein the process results in the addition or removal of material forming the contour, the addition or the removal resulting in a change in the shape of the contour.
[0328] 51. The method of clause 50, wherein the process rate characterizes the behavior of the addition or the removal of the material forming the profile.
[0329] 52. A non-transitory computer readable medium comprising instructions that, when executed by one or more processors, result in operations comprising:
[0330] obtaining (i) profile data of an after-development image (ADI) pattern on a substrate, (ii) measurement data of an after-etching image (AEI) pattern printed on the substrate, and (iii) obtaining a reference bias value based on the profile data of the ADI pattern and the measurement data of the AEI pattern; and
[0331] The machine learning model is trained using the measurement data and the profile data as training data to determine bias values to be applied to the ADI profile.
[0332] 53. The non-transitory computer-readable medium of clause 52, wherein the training comprises:
[0333] A model parameter of the machine learning model is adjusted so that the bias value is within a specified range determined based on the reference bias value.
[0334] 54. A non-transitory computer-readable medium according to aspect 52, wherein the machine learning model is configured to generate a representation of a bias map for the ADI profile.
[0335] 55. The non-transitory computer-readable medium of aspect 52, wherein the contour data represents an image of a contour associated with one or more features of the ADI pattern.
[0336] 56. A non-transitory computer-readable medium according to aspect 55, wherein the image is generated from a simulated profile of a simulated ADI pattern and / or obtained from a metrology device configured to capture an image of the substrate after a pre-forming resist process on the substrate.
[0337] 57. The non-transitory computer-readable medium of any one of aspects 52 to 56, wherein obtaining the profile data comprises:
[0338] executing one or more process models associated with the patterning process using a design pattern to be printed on the substrate as input to generate the simulated ADI pattern, the patterning process including a resist process; and
[0339] Contours are extracted from the simulated ADI pattern, each contour being a contour of a feature within the simulated ADI pattern.
[0340] 58. The non-transitory computer-readable medium of aspect 57, wherein the one or more process models include at least one of:
[0341] an optics model configured to determine an aerial image; and
[0342] A resist model is configured to determine a resist image.
[0343] 59. The non-transitory computer-readable medium of any one of aspects 52 to 58, wherein the measurement data is obtained at a metrology gauge, the metrology gauge being an edge placement gauge, and / or a critical dimension (CD) gauge associated with the AEI pattern.
[0344] 60. The non-transitory computer-readable medium of aspect 57, wherein the measurement data at the metrology gauge comprises:
[0345] a location of the edge placement gauge associated with an outline of the AEI pattern printed on the substrate; and / or
[0346] A CD value associated with the AEI pattern printed on the substrate.
[0347] 61. A non-transitory computer-readable medium according to aspect 57, wherein when the measuring gauge is a CD gauge, the reference offset value is obtained via a calibration process, the calibration process being configured to determine an offset value associated with a given CD gauge, the offset value indicating an amount of CD reduction to be applied to the ADI pattern to produce the AEI pattern.
[0348] 62. The non-transitory computer-readable medium of aspect 61, wherein the offset values are provided at ends of the given CD gauge, the offset values being unequal or asymmetric about a center of the CD gauge.
[0349] 63. The non-transitory computer-readable medium of any one of clauses 52 to 62, wherein said training said machine learning model is an iterative process, the iterations comprising:
[0350] (a) executing the machine learning model using the measurement data, the profile data and given values of the model parameters to produce a bias map associated with the profile data, the bias map comprising bias values;
[0351] (b) based on the gradient of the difference between the model-based bias value and the reference bias value, adjusting the model parameters of the machine learning model so as to reduce the difference; and
[0352] (c) performing steps (a) to (b) until the difference is minimized.
[0353] 64. A non-transitory computer-readable medium according to aspects 52 to 63, wherein the bias map generated via a trained machine learning model includes etch bias values to be applied to a resist profile to determine an etch profile to be printed on the substrate.
[0354] 65. The non-transitory computer-readable medium of clause 64, wherein the etch bias value is applied to the resist profile in a local normal direction.
[0355] 66. The non-transitory computer-readable medium of any one of clauses 52 to 66, wherein the bias map is a pixelated image, each pixel having an intensity value indicative of a bias value.
[0356] 67. A non-transitory computer readable medium comprising instructions that, when executed by one or more processors, result in operations comprising:
[0357] Obtaining (i) a probability distribution function (PDF) corresponding to particles within an after-development image (ADI) pattern deposited on a substrate, and (ii) a profile function characterizing an ADI profile associated with the ADI pattern;
[0358] determining a deposition rate of the particle at a specified location on the ADI profile based on a combination of the PDF of the particle over a region of the ADI profile and the profile function; and
[0359] A bias vector associated with the ADI pattern is determined based on the deposition rate, the bias vector producing an after-etch image (AEI) profile when applied to the ADI profile of the ADI pattern.
[0360] 68. The non-transitory computer-readable medium of aspect 67, wherein the obtaining the probability distribution function (PDF) of the particles is based on measured substrate data, the measured substrate data comprising deposition data of the particles and a measured etch pattern.
[0361] 69. The non-transitory computer-readable medium of aspect 68, wherein obtaining the PDF comprises determining a variance of a Gaussian distribution that fits the measurement data.
[0362] 70. The non-transitory computer-readable medium of any one of aspects 67 to 69, wherein the determining the deposition rate of the particles comprises:
[0363] convolving the PDF of the particle with the profile function; and
[0364] An integration is performed over the area of the ADI profile.
[0365] 71. The non-transitory computer-readable medium of any one of aspects 67 to 70, wherein determining the bias vector comprises:
[0366] A gradient of the deposition rate with respect to a first direction and a second direction of the ADI pattern is determined, the first direction and the second direction being perpendicular to each other.
[0367] 72. The non-transitory computer-readable medium of aspect 67, wherein the bias vector comprises: a bias direction at a location of the ADI profile, and the method further comprises applying a bias value along to generate the AEI profile.
[0368] 73. The non-transitory computer-readable medium of aspect 72, wherein the bias direction is determined such that when the bias values at different locations are applied to the ADI profile, the bias vectors at different locations do not intersect each other.
[0369] 74. The non-transitory computer-readable medium of aspect 73, wherein the bias direction comprises a direction that is not perpendicular to the ADI profile.
[0370] 75. The non-transitory computer-readable medium of any one of aspects 67 to 74, wherein the PDF of the particle represents a deposition or etching process of the particle on the ADI profile, and wherein the deposition rate is positive or negative.
[0371] 76. A non-transitory computer-readable medium according to any one of aspects 67 to 75, wherein the bias value is obtained from a trained machine learning model configured to generate a bias map for a given resist pattern.
[0372] 77. The non-transitory computer-readable medium of any one of aspects 67 to 76, wherein when the ADI pattern comprises a plurality of profiles, a set of bias vectors is determined separately for each ADI profile.
[0373] 78. A non-transitory computer readable medium comprising instructions that, when executed by one or more processors, result in operations comprising:
[0374] obtaining (i) a probability distribution function (PDF) corresponding to a process to be performed on the contour, and (ii) a contour function characterizing a shape of the contour;
[0375] convolving the profile function with the PDF over an area of the profile to determine a process rate at a specified location on the profile; and
[0376] A bias vector is determined to be applied to the profile based on the process rate to produce a bias profile indicative of an effect of the process applied to the profile.
[0377] 79. The non-transitory computer-readable medium of aspect 78, wherein the process results in the addition or removal of material forming the contour, the addition or the removal resulting in a change in the shape of the contour.
[0378] 80. The non-transitory computer-readable medium of aspect 79, wherein the process rate characterizes an act of the addition or the removal of the material forming the profile.
[0379] The concepts disclosed herein can simulate or mathematically model any general imaging system for imaging sub-wavelength features, and can be particularly useful for emerging imaging technologies that can produce wavelengths with increasingly smaller sizes. Emerging technologies already in use include extreme ultraviolet (EUV) lithography, which can produce wavelengths of 193 nm using ArF lasers and even 157 nm using fluorine lasers. In addition, EUV lithography can produce wavelengths in the range of 20 nm to 5 nm using a synchrotron or by using high energy electrons to impinge on a material (solid or plasma) to produce photons in this range.
[0380] Although the concepts disclosed herein may be used for imaging on substrates such as silicon wafers, it should be understood that the disclosed concepts may be used with any type of lithography imaging system, for example, a lithography imaging system for imaging on substrates other than silicon wafers.
[0381] Although specific reference may be made herein to the use of embodiments in IC manufacturing, it should be understood that the embodiments herein may have many other possible applications. For example, it may be used to manufacture integrated optical systems, guidance and detection patterns for magnetic domain memories, liquid crystal displays (LCDs), thin film magnetic heads, micromechanical systems (MEMs), etc. Those skilled in the art will appreciate that in the context of such alternative applications, any use of the terms "reticle", "wafer" or "die" herein may be considered synonymous with or interchangeable with the more general terms "patterning device", "substrate" or "target portion", respectively. The substrates mentioned herein may be processed in, for example, a track (a tool that typically applies a resist layer to a substrate and develops the exposed resist) or a measurement or detection tool before or after exposure. When applicable, the disclosure herein may be applied to such and other substrate processing tools. In addition, the substrate may be processed more than once (for example) to produce a multi-layer IC, so that the term substrate used herein may also refer to a substrate that already contains multiple processed layers.
[0382] In this document, the terms "radiation" and "beam" as used herein encompass all types of electromagnetic radiation, including ultraviolet radiation (e.g., having a wavelength of about 365 nm, about 248 nm, about 193 nm, about 157 nm, or about 126 nm) and extreme ultraviolet (EUV) radiation (e.g., having a wavelength in the range of 5 nm to 20 nm), as well as particle beams, such as ion beams or electron beams.
[0383] As used herein, the term "optimizing (optimizing / optimization)" refers to or means adjusting a patterning device (e.g., a lithographic device), a patterning process, etc. so that the result and / or process has more desired characteristics, such as a higher projection accuracy of the designed pattern on the substrate, a larger process window, etc. Therefore, as used herein, the term "optimizing (optimizing / optimization)" refers to or means a process of identifying one or more values for one or more parameters, which provide an improvement in at least one related indicator, such as a local optimum, compared to an initial set of one or more values for those one or more parameters. "Optimal" and other related terms should be interpreted accordingly. In an embodiment, the optimization step can be applied iteratively to provide further improvement of one or more indicators.
[0384] Aspects of the present invention may be implemented in any convenient form. For example, an embodiment may be implemented by one or more appropriate computer programs, which may be performed on an appropriate carrier medium that may be a tangible carrier medium (e.g., a disk) or an intangible carrier medium (e.g., a communication signal). An embodiment of the present invention may be implemented using a suitable device that may specifically take the form of a programmable computer that executes a computer program configured to implement a method as described herein. Therefore, embodiments of the present disclosure may be implemented in hardware, firmware, software, or any combination thereof. Embodiments of the present disclosure may also be implemented as instructions stored on a machine-readable medium that may be read and executed by one or more processors. A machine-readable medium may include any mechanism for storing or transmitting information in a form that can be read by a machine (e.g., a computing device). For example, a machine-readable medium may include: a read-only memory (ROM); a random access memory (RAM); a disk storage medium; an optical storage medium; a flash memory device; a propagation signal in electrical, optical, acoustic, or other form (e.g., a carrier wave, an infrared signal, a digital signal, etc.), etc. In addition, firmware, software, routines, instructions may be described herein as performing certain actions. However, it should be understood that these descriptions are merely for convenience and that these actions are actually caused by a computing device, processor, controller, or other device executing firmware, software, routines, instructions, etc.
[0385] In the block diagrams, the illustrated components are depicted as discrete functional blocks, but embodiments are not limited to systems in which the functionality described herein is organized as illustrated. The functionality provided by each of the components may be provided by software or hardware modules that are organized in a manner different from that currently depicted, for example, may be mixed, combined, replicated, decomposed, distributed (e.g., within a data center or by region), or otherwise organize such software or hardware in a different manner. The functionality described herein may be provided by one or more processors of one or more computers executing program code stored on a tangible, non-transitory machine-readable medium. In some cases, a third-party content distribution network may host some or all of the information communicated via the network, in which case, in the case of purportedly supplying or otherwise providing information (e.g., content), the information may be provided by sending instructions to obtain the information from the content distribution network.
[0386] Unless otherwise specifically stated, as is clear from the discussion, it should be understood that throughout this specification, discussions utilizing terms such as "processing," "computing," "calculating," "determining," etc. refer to actions or processes of a specific apparatus such as a special purpose computer or similar special purpose electronic processing / computing device.
[0387] The reader should understand that this application describes several inventions. These inventions have been grouped into a single file, rather than separating those inventions into multiple separate patent applications, because the related subject matter of the inventions contributes to economic development in the application. However, the different advantages and aspects of these inventions should not be combined. In some cases, the embodiments solve all the defects mentioned herein, but it should be understood that the inventions are independently useful, and some embodiments only solve a subset of these problems or provide other unmentioned benefits, which will be understood by those skilled in the art who review this disclosure. Due to cost constraints, some of the inventions disclosed herein may not be claimed at present, and the inventions may be claimed in a later application (such as a continuation application or by amending the present technical solution). Similarly, due to space constraints, the abstract and content of the invention sections of this document should not be regarded as a comprehensive list containing all these inventions or all aspects of these inventions.
[0388] It should be understood that the specification and drawings are not intended to limit the invention to the particular forms disclosed, but on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the invention as defined by the appended claims.
[0389] In view of this specification, those skilled in the art will understand the modification and alternative embodiments of various aspects of the present invention. Therefore, this specification and the accompanying drawings should be understood to be illustrative only and for the purpose of teaching those skilled in the art to carry out the general mode of the present invention. It should be understood that the form of the present invention shown and described herein should be regarded as an example of an embodiment. Elements and materials can replace the elements and materials illustrated and described herein, parts and processes can be reversed or omitted, certain features can be utilized independently, and the features of embodiments or embodiments can be combined, as will be understood by those skilled in the art after obtaining the benefit of this specification. Changes can be made to the elements described herein without departing from the spirit and scope of the present invention as described in the appended claims. The titles used herein are only to achieve organizational purposes and are not intended to limit the scope of this specification.
[0390] As used throughout this application, the words "may" or "could" are used in a permissive sense (i.e., meaning possibly) rather than a mandatory sense (i.e., meaning must). The words "include / including / includes" and the like mean including, but not limited to. As used throughout this application, the singular forms "a / an / the" include plural references unless the context clearly indicates otherwise. Thus, for example, reference to "an element / a element" includes a combination of two or more elements, although other terms and phrases, such as "one or more", may be used with respect to one or more elements. Unless otherwise indicated, the term "or" is non-exclusive, i.e., covers both "and" and "or". Terms describing conditional relationships, such as "in response to X, and Y", "after X, then Y", "if X, then Y", "when X, Y", etc., cover causal relationships where the antecedent is a necessary causal condition, the antecedent is a sufficient causal condition, or the antecedent is a contributing causal condition to the result, for example, "after condition Y is obtained, state X occurs" is common to "only after Y, then X occurs" and "after Y and Z, then X occurs". These conditional relationships are not limited to results obtained by immediately following the antecedent, because some results can be delayed, and in the conditional statement, the antecedent is connected to its result, for example, the antecedent is related to the probability of the result occurring. Unless otherwise indicated, a statement that multiple properties or functions are mapped to multiple objects (e.g., one or more processors that perform steps A, B, C, and D) covers both all of these properties or functions being mapped to all of these objects and subsets of properties or functions being mapped to subsets of properties or functions (e.g., all processors each perform steps A to D, and where processor 1 performs step A, processor 2 performs step B and part of step C, and processor 3 performs part of step C and step D). In addition, unless otherwise indicated, a statement that a value or action is "based on" another condition or value encompasses both the case where the condition or value is a separate factor and the case where the condition or value is a factor of multiple factors. Unless otherwise indicated, a statement that "each" instance of a set has a property should not be read to exclude the case where some otherwise identical or similar members of the larger set do not have the property (i.e., each does not necessarily mean every one). References to selections from a range include the endpoints of the range.
[0391] In the above description, any process, description or block in the flowchart should be understood to represent a module, segment or portion of a program code, which includes one or more executable instructions for implementing specific logical functions or steps in the process, and alternative implementations are included in the scope of the exemplary embodiments of the present invention, in which functions may be performed out of the order shown or discussed, including substantially simultaneously or in reverse order, depending on the functionality involved, as should be understood by those skilled in the art.
[0392] To the extent that certain U.S. patents, U.S. patent applications, or other materials (e.g., articles) have been incorporated by reference, the text of such U.S. patents, U.S. patent applications, and other materials is incorporated only to the extent that there is no conflict between such materials and the statements and drawings set forth herein. In the event of such a conflict, any such conflicting text in such incorporated by reference U.S. patents, U.S. patent applications, and other materials is not specifically incorporated by reference herein.
[0393] Although certain embodiments have been described, these embodiments are presented only as examples and are not intended to limit the scope of the present disclosure. Indeed, the novel methods, devices, and systems described herein may be embodied in a variety of other forms; furthermore, various omissions, substitutions, and changes in the form of the methods, devices, and systems described herein may be made without departing from the spirit of the present disclosure. The accompanying claims and their equivalents are intended to cover such forms or modifications as would be within the scope and spirit of the present disclosure.
Claims
1. A method for training a machine learning model associated with a patterning process, the method comprising: Obtaining (i) profile data of an after-development image (ADI) pattern on a substrate, (ii) measurement data of an after-etching image (AEI) pattern printed on the substrate, and (iii) obtaining a reference bias value based on the profile data of the ADI pattern and the measurement data of the AEI pattern; and The machine learning model is trained using the measurement data and the profile data as training data to determine bias values to be applied to the ADI profile.
2. The method according to claim 1, wherein the training comprises: A model parameter of the machine learning model is adjusted so that the bias value is within a specified range determined based on the reference bias value.
3. The method of claim 1 , wherein the machine learning model is configured to generate a representation of a bias map for the ADI profile. 4 . The method of claim 1 , wherein the contour data represents an image of a contour associated with one or more features of the ADI pattern.
5. The method of claim 4, wherein the image is generated from a simulated profile of a simulated ADI pattern and / or obtained from a metrology device configured to capture an image of the substrate after a process of preforming a resist on the substrate.
6. The method according to any one of claims 1 to 5, wherein obtaining the profile data comprises: executing one or more process models associated with the patterning process using a design pattern to be printed on the substrate as input to generate the simulated ADI pattern, the patterning process including a resist process; and Contours are extracted from the simulated ADI pattern, each contour being a contour of a feature within the simulated ADI pattern.
7. The method of claim 6, wherein the one or more process models include at least one of: an optics model configured to determine an aerial image; and A resist model is configured to determine a resist image.
8. The method according to any one of claims 1 to 7, wherein the measurement data is obtained at a metrology gauge, the metrology gauge being an edge placement gauge, and / or a critical dimension (CD) gauge associated with the AEI pattern.
9. The method of claim 6, wherein the measurement data at the measurement gauge comprises: a location of the edge placement gauge associated with an outline of the AEI pattern printed on the substrate; and / or A CD value associated with the AEI pattern printed on the substrate.
10. The method of claim 6, wherein when the measuring gauge is a CD gauge, the reference offset value is obtained via a calibration process, the calibration process being configured to determine an offset value associated with a given CD gauge, the offset value indicating an amount of CD reduction to be applied to the ADI pattern to produce the AEI pattern.
11. The method of claim 10, wherein the offset values are provided at the ends of the given CD gauge, the offset values being unequal or asymmetric about the center of the CD gauge.
12. The method according to any one of claims 1 to 11, wherein the training of the machine learning model is an iterative process, the iteration comprising: (a) executing the machine learning model using the measurement data, the profile data and given values of the model parameters to produce a bias map associated with the profile data, the bias map comprising bias values; (b) based on the gradient of the difference between the model-based bias value and the reference bias value, adjusting the model parameters of the machine learning model so as to reduce the difference; and (c) performing steps (a) to (b) until the difference is minimized.
13. The method of claims 1 to 12, wherein the bias map generated via a trained machine learning model comprises etch bias values to be applied to a resist profile to determine an etch profile to be printed on the substrate. The method of claim 13 , wherein the etch bias value is applied to the resist profile in a local normal direction.
15. A method according to any one of claims 1 to 14, wherein the bias map is a pixelated image, each pixel having an intensity value indicative of a bias value.
16. A non-transitory computer-readable medium comprising instructions that, when executed by one or more processors, result in operations comprising: Obtaining (i) a probability distribution function (PDF) corresponding to particles within an after-development image (ADI) pattern deposited on a substrate, and (ii) a profile function characterizing an ADI profile associated with the ADI pattern; determining a deposition rate of the particle at a specified location on the ADI profile based on a combination of the PDF of the particle over a region of the ADI profile and the profile function; and A bias vector associated with the ADI pattern is determined based on the deposition rate, the bias vector producing an after-etch image (AEI) profile when applied to the ADI profile of the ADI pattern.
17. The non-transitory computer readable medium of claim 16, wherein the obtaining the probability distribution function (PDF) of particles is based on measured substrate data, the measured substrate data comprising deposition data of particles and a measured etch pattern.
18. The non-transitory computer-readable medium of claim 17, wherein obtaining the PDF comprises determining a variance of a Gaussian distribution that fits the measurement data.
19. The non-transitory computer readable medium of any one of claims 16 to 18, wherein said determining the deposition rate of the particles comprises: convolving the PDF of the particle with the profile function; and An integration is performed over the area of the ADI profile.
20. The non-transitory computer readable medium of any one of claims 16 to 19, wherein determining the bias vector comprises: A gradient of the deposition rate with respect to a first direction and a second direction of the ADI pattern is determined, the first direction and the second direction being perpendicular to each other.
21. The non-transitory computer readable medium of claim 16, wherein the bias vector comprises: The method further comprises applying a bias value along the position of the ADI profile to generate the AEI profile.
22. The non-transitory computer-readable medium of claim 21, wherein the bias direction is determined such that when the bias values at different locations are applied to the ADI profile, the bias vectors at different locations do not intersect each other.
23. The non-transitory computer readable medium of claim 22, wherein the bias direction comprises: A direction that is not perpendicular to the ADI profile.
24. The non-transitory computer-readable medium of any one of claims 16 to 23, wherein the PDF of the particle represents a deposition or etching process of the particle on the ADI profile, and wherein the deposition rate is positive or negative.
25. The non-transitory computer-readable medium of any one of claims 16 to 24, wherein the bias value is obtained from a trained machine learning model configured to generate a bias map for a given resist pattern.
26. The non-transitory computer readable medium of any one of claims 16 to 25, wherein when the ADI pattern includes a plurality of profiles, a set of bias vectors is determined separately for each ADI profile.
27. A non-transitory computer-readable medium comprising instructions that, when executed by one or more processors, result in operations comprising: obtaining (i) a probability distribution function (PDF) corresponding to a process to be performed on the contour, and (ii) a contour function characterizing a shape of the contour; convolving the profile function with the PDF over an area of the profile to determine a process rate at a specified location on the profile; and A bias vector is determined to be applied to the profile based on the process rate to produce a bias profile indicative of an effect of the process applied to the profile.
28. The non-transitory computer readable medium of claim 27, wherein the process results in the addition or removal of material forming the contour, the addition or the removal resulting in a change in the shape of the contour.
29. The non-transitory computer readable medium of claim 28, wherein the process rate characterizes behavior of the addition or the removal of the material forming the profile.
