Method and apparatus for layout pattern selection

Abstract

A method for determining training patterns in a layout patterning process is described herein. The method includes: generating a plurality of features from patterns in a pattern set; grouping the patterns in the pattern set into separate groups based on similarities of the plurality of generated features; and selecting representative patterns from the separate groups to determine the training patterns. In some embodiments, the method is a method for training a machine learning model in a layout patterning process. For example, the method may include providing representative patterns from the separate groups to the machine learning model to train the machine learning model to predict a continuous transmission mask (CTM) mapping for optical proximity correction (OPC) in the layout patterning process.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to Chinese application No. PCT / CN2019 / 073714 filed on January 29, 2019, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The description herein generally relates to mask manufacturing and patterning processes. More particularly, the description relates to methods and apparatus for layout pattern selection for training machine learning models. Background Art

[0004] Lithographic projection apparatus may be used, for example, in the manufacture of integrated circuits (ICs). In this case, a patterning device (e.g., a mask) may include or provide a pattern corresponding to a single layer of the IC (a "design layout"), and this pattern may be transferred to a target portion (e.g., comprising one or more dies) on a substrate (e.g., a silicon wafer) that has been coated with a layer of radiation-sensitive material ("resist") by methods such as irradiating the target portion through the pattern on the patterning device. Typically, a single substrate comprises a plurality of adjacent target portions to which the pattern is transferred successively, one target portion at a time, by the lithographic projection apparatus. In one type of lithographic projection apparatus, the pattern on the entire patterning device is transferred to a target portion in one operation; such an apparatus is typically referred to as a stepper. In an alternative apparatus, typically referred to as a stepper-scan apparatus, the projection beam is scanned over the patterning device in a given reference direction (the "scanning" direction) while synchronously moving the substrate in a direction parallel or antiparallel to the reference direction. Different portions of the pattern on the patterning device are gradually transferred to a target portion. Since typically the lithographic projection apparatus will have a reduction ratio M (e.g. 4), the rate F at which the substrate is moved will be 1 / M times the rate at which the projection beam scans the patterning device. Further information on lithographic apparatus as described herein can be gleaned, for example, from US 6,046,792, which is incorporated herein by reference.

[0005] Before the pattern is transferred from the pattern forming device to the substrate, the substrate may undergo various processes, such as primer coating, resist coating and soft baking. After exposure, the substrate may undergo other processes ("post-exposure processes"), such as post-exposure baking (PEB), development, hard baking and measurement / inspection of the transferred pattern. This series of processes is used as the basis for manufacturing a single layer of a device (e.g., IC). Afterwards, the substrate may undergo various processes, such as etching, ion implantation (doping), metallization, oxidation, chemical mechanical polishing, etc., all of which are intended to finally complete a single layer of the device. If the device requires multiple layers, the entire process or its variation is repeated for each layer. Ultimately, the device will be set in each target portion on the substrate. Afterwards, these devices are separated from each other by techniques such as slicing or cutting, so that individual devices can be mounted on a carrier, connected to pins, etc.

[0006] Manufacturing devices (such as semiconductor devices) typically involves using multiple manufacturing processes to process a substrate (e.g., a semiconductor wafer) to form various features and multiple layers of the device. These layers and features are typically manufactured and processed using, for example, deposition, lithography, etching, chemical mechanical polishing, and ion implantation. Multiple devices can be manufactured on multiple dies on a substrate and then separated into separate devices. This device manufacturing process can be considered a patterning process. The patterning process involves a patterning step, such as optical and / or nanoimprint lithography using a patterning device in a lithographic apparatus to transfer the pattern on the patterning device to the substrate, and typically but optionally involves one or more related pattern processing steps, such as resist development by a developing apparatus, substrate baking using a baking tool, etching using a pattern using an etching apparatus, etc. In addition, one or more measurement processes are typically involved in the patterning process.

[0007] As mentioned, photolithography is a core step in the manufacture of devices such as ICs, where patterns formed on a substrate define the functional elements of the device, such as microprocessors, memory chips, etc. Similar photolithography techniques are also used to form flat panel displays, microelectromechanical systems (MEMS), and other devices.

[0008] As semiconductor manufacturing processes continue to advance, over the decades, the size of functional elements has continued to decrease while the number of functional elements (such as transistors) per device has steadily increased, following a trend commonly referred to as "Moore's Law." In the current state of the art, multiple layers of a device are fabricated using a lithographic projection apparatus that projects a design layout onto a substrate using illumination from a deep ultraviolet illumination source, thereby forming a single functional element having a size well below 100 nm (i.e., less than half the wavelength of the radiation from the illumination source (e.g., a 193 nm illumination source)).

[0009] This process in which features with dimensions smaller than the classical resolution limit of the lithographic projection apparatus are printed is generally referred to as low-k1 lithography, and is based on the resolution formula CD=k1×λ / NA, where λ is the wavelength of the radiation employed (currently 248 nm or 193 nm in most cases), NA is the numerical aperture of the projection optics in the lithographic projection apparatus, CD is the "critical dimension" (usually the smallest feature size printed), and k1 is an empirical resolution factor. In general, the smaller k1 is, the more difficult it becomes to reproduce a pattern on a substrate that resembles the shape and dimensions planned by the designer to achieve specific electrical functionality and performance. To overcome these difficulties, complex fine-tuning steps are applied to the lithographic projection apparatus, the design layout, or the patterning device. These steps include, for example, but are not limited to: optimization of NA and optical coherence settings, customized illumination schemes, use of phase-shifting patterning devices, optical proximity correction (OPC, sometimes also called "optical and process correction") in the design layout, or other methods generally defined as "resolution enhancement techniques" (RET). The term "projection optical element" as used herein should be broadly interpreted to cover various types of optical systems, including, for example, refractive optics, reflective optics, apertures, and reflective-refractive optics. The term "projection optical element" may also include components that operate according to any of these design types for guiding, shaping, or controlling a projection radiation beam, either collectively or individually. The term "projection optical element" may include any optical component in a lithographic projection apparatus, regardless of where the optical component is located on the optical path of the lithographic projection apparatus. Projection optical elements may include optical components for shaping, adjusting, and / or projecting radiation from a source before the radiation passes through a pattern forming device, or optical components for shaping, adjusting, and / or projecting the radiation after the radiation passes through a pattern forming device. Projection optical elements typically do not include a source and a pattern forming device. Summary of the invention

[0010] According to an embodiment, a method for training a machine learning model for a wafer patterning process is provided. The method includes: generating a plurality of features for each pattern in a pattern set; grouping the patterns in the pattern set into a plurality of separate groups based on the similarity of the plurality of generated features; and providing representative patterns from the plurality of separate groups to a computational lithography application of the wafer patterning process. The application may include source mask optimization (SMO), optical proximity correction (OPC), lithography manufacturability check (LMC), etc.

[0011] In an embodiment, the plurality of features generated from the patterns in the pattern set are information in addition to geometric information and / or vertex information already included in the pattern set. In an embodiment, SMO includes source and mask co-optimization for a full-chip layout of a wafer during the wafer patterning process. In an embodiment, the OPC includes a full-chip OPC for a wafer during the layout patterning process. In an embodiment, LMC includes lithography manufacturability and lithography performance inspection for a full-chip layout of a wafer during the wafer patterning process. In an embodiment, the plurality of generated features include geometric features and lithography-aware features. In an embodiment, grouping the patterns in the pattern set into a plurality of separate groups based on the similarity of a plurality of generated features includes using a machine learning clustering method to cluster unique patterns in the pattern set into a plurality of separate groups based on the similarity of a plurality of generated features.

[0012] According to another embodiment, a method for determining training patterns for a wafer patterning process is provided. The method includes: generating a plurality of features from patterns in a pattern set; grouping the patterns in the pattern set into a plurality of separate groups based on similarities of a plurality of generated features; and selecting representative patterns from the plurality of separate groups to determine the training patterns.

[0013] In an embodiment, the plurality of generated features include geometric features and lithography-aware features. The geometric features include one or more of: a target mask image, a frequency map, a pattern density map, or a pattern occurrence rate of a unique pattern in the pattern set. The lithography-aware features include one or more of: a sub-resolution auxiliary feature guide map (SGM), a diffraction order, or a diffraction pattern of a unique pattern in the pattern set. In an embodiment, the plurality of features generated from the patterns in the pattern set are information in addition to the geometric information and / or vertex information already included in the pattern set.

[0014] In an embodiment, grouping the patterns in the pattern set into a plurality of groups based on a plurality of generated features is performed using unsupervised machine learning. In an embodiment, grouping the patterns in the pattern set into a plurality of separate groups based on similarity of a plurality of generated features comprises clustering unique patterns in the pattern set into a plurality of separate groups based on the similarity of a plurality of generated features. The clustering comprises a continuous series of clustering steps performed using different features of a plurality of generated features for different clustering steps, the continuous series of clustering steps forming a subset of the patterns in the pattern set such that the representative pattern is selected from the subset to determine the training pattern. In an embodiment, the clustering comprises a machine learning clustering method (e.g., k-means clustering).

[0015] In an embodiment, the consecutive series of clustering steps comprises a cross-validation step performed using given features for a given step. The cross-validation step comprises adjusting which patterns are included in a given subset.

[0016] In an embodiment, selecting representative patterns from the separate group to determine the training pattern comprises selecting a target number of representative patterns. The target number of representative patterns is determined based on a stopping criterion. The stopping criterion is configured to cause a change in the training pattern. In an embodiment, the method further comprises determining an amount of change in the training pattern. In an embodiment, the stopping criterion is further configured to ensure that the amount of change in the training pattern exceeds a change amount threshold.

[0017] In an embodiment, the target number of representative patterns is randomly selected from the plurality of separate groups. In an embodiment, the target number of representative patterns is randomly reselected in response to the variation of the training pattern not exceeding the variation threshold.

[0018] In an embodiment, selecting representative patterns from the plurality of separate groups to determine the training patterns comprises selecting a center-most pattern from each separate group. The center-most pattern is closest to the centroid of a specified feature space for the separate group relative to other patterns in the plurality of separate groups. In an embodiment, the specified feature space is a target mask image feature space, a frequency mapping feature space, a pattern density mapping feature space, a pattern occurrence feature space, an SGM feature space, a diffraction order feature space, or a diffraction pattern feature space.

[0019] In an embodiment, the method further comprises providing the training pattern to a deep convolutional neural network to train the deep convolutional neural network. In an embodiment, the method further comprises performing optical proximity effect correction as part of a wafer patterning process using the trained deep convolutional neural network.

[0020] According to another embodiment, a computer program product is provided. The computer program product includes a non-transitory computer-readable medium having instructions recorded thereon. The instructions implement the method described above when executed by a computer. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate one or more embodiments and together with the specification explain these embodiments. Embodiments of the present invention will now be described, by way of example only, with reference to the accompanying schematic drawings, in which corresponding reference numerals indicate corresponding parts, and in which:

[0022] Figure 1A block diagram showing various subsystems of a lithography system according to an embodiment is shown.

[0023] Figure 2 The present invention is a flowchart of a method for determining a pattern forming device pattern (or mask pattern) from an image (e.g., a continuous transmission mask image, a binary mask image, a curved mask image, etc.) according to an embodiment, wherein the image corresponds to a target pattern to be printed on a substrate via a patterning process involving a photolithography process.

[0024] Figure 3 An exemplary flow chart for simulating lithography in a lithographic projection apparatus according to an embodiment is illustrated.

[0025] Figure 4 An overview of the operation of the present method for determining training patterns during wafer patterning is illustrated in accordance with an embodiment.

[0026] Figure 5 The illustrations show aspects resulting from features of the embodiments.

[0027] Figure 6 The diagram illustrates clustering patterns in a set of patterns into separate groups based on similarity of features generated for unique patterns, according to an embodiment.

[0028] Figure 7 An example graph illustrating sum of squared errors versus k value as part of a cluster validation operation according to an embodiment is illustrated.

[0029] Figure 8 Illustrate k-fold cross validation according to an embodiment.

[0030] Fig. 9 Illustrated is a selection of patterns used to form a representative example set of patterns for use as training patterns, according to an embodiment.

[0031] Fig.10 is a block diagram of an example computer system according to an embodiment.

[0032] Fig.11 is a schematic diagram of a lithographic projection apparatus according to an embodiment.

[0033] Fig.12 is a schematic diagram of another lithography projection apparatus according to an embodiment.

[0034] Fig.13 According to the embodiment Fig.12 A more detailed view of the devices in .

[0035] Fig.14 According to the embodiment Fig.12 and Fig.13 A more detailed view of the source collector module SO of the device. DETAILED DESCRIPTION

[0036] Performing pattern selection from a full-chip Graphic Database System (GDS) file (e.g., a GDSII file) is a challenging task. Training patterns are used to train deep convolutional neural networks (DCNNs) and / or other machine learning models (which are part of full-chip optical proximity correction (OPC) applications, source mask optimization (SMO) applications, lithography manufacturability inspection (LMC) applications), and / or for other purposes. If a user generates a training pattern set for a layout pattern with patterns that are not completely representative or otherwise insufficient, and such training data is provided to a machine learning model (e.g., a DCNN) for training, an accurate CTM mapping will not be predicted by a machine learning model (e.g., a DCNN) for a full-chip OPC application. Inaccurate CTM mappings lead to lithography hot spots and process window limitations, and / or cause difficulties during subsequent mask correction operations when trying to meet lithography performance specifications. Currently, users manually select training patterns from a full-chip GDS. Manual selection requires significant effort from the user, and layout pattern selection (e.g., representative coverage of various patterns) depends on the user's experience and prior knowledge about the full-chip GDS design. Advantageously, the present method and apparatus systematically analyzes the full-chip GDS patterns and selects representative patterns to construct a machine learning model training set, wherein the pattern coverage adequately represents the target layout patterns across the full-chip GDS file.

[0037] As a brief introduction, although specific reference may be made herein to the manufacture of ICs, it should be clearly understood that the description herein has many other possible applications. For example, the description herein may be used to manufacture integrated optical systems, guidance and detection patterns for magnetic domain memories, liquid crystal display panels, thin film magnetic heads, etc. In these alternative applications, 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 should be considered interchangeable with the more general terms "mask", "substrate" and "target portion", respectively. In addition, it should be noted that the methods described herein may have many other possible applications in diverse fields, such as language processing systems, self-driving cars, medical imaging and diagnosis, semantic segmentation, denoising, chip design, electronic design automation, etc. The present method may be applied to any field where quantifying uncertainty in machine learning model predictions is advantageous.

[0038] In this document, the terms "radiation" and "beam" are used to cover all types of electromagnetic radiation, including ultraviolet radiation (e.g., having a wavelength of 365, 248, 193, 157 or 126 nm) and EUV (i.e., extreme ultraviolet radiation, e.g., having a wavelength in the range of about 5 to 100 nm).

[0039] The pattern forming device may include or may form one or more design layouts. The design layout may be generated using a CAD (i.e., computer-aided design) process. This process is often referred to as EDA (i.e., electronic design automation). Most CAD programs follow a set of predetermined design rules to generate a functional design layout / pattern forming device. These rules are set based on processing and design constraints. For example, the design rules regulate the spatial tolerance between devices (such as gates, capacitors, etc.) or interconnects to ensure that the devices or lines do not interact with each other in an undesirable manner. One or more of the restrictions in the design rule restrictions may be referred to as a "critical dimension" (CD). The critical dimension of a device may be defined as the minimum width of a line or hole, or the minimum space / gap between two lines or two holes. Thus, CD determines the total size and density of the designed device. One of the goals in device manufacturing is to faithfully reproduce the original design intent (via a pattern forming device) on a substrate.

[0040] The term "mask" or "patterning device" as used herein may be broadly interpreted as referring to a general patterning device that can be used to impart a patterned cross-section to an incident radiation beam, the patterned cross-section corresponding to the pattern to be produced in the target portion of the substrate. The term "light valve" may also be used in this context. In addition to classical masks (transmissive or reflective, binary, phase-shifted, hybrid, etc.), other examples of such patterning devices include programmable mirror arrays. An example of such a device may be a matrix-addressable surface having a viscoelastic control layer and a reflective surface. The basic principle on which such an apparatus is based is that (for example) addressed areas of the reflective surface reflect the incident radiation into diffracted radiation, while unaddressed areas reflect the incident radiation into non-diffracted radiation. Using a suitable filter, the non-diffracted radiation can be filtered out from the reflected beam, leaving only the diffracted radiation; in this way, the beam is patterned according to the addressing pattern of the matrix-addressable surface. The required matrix addressing can be performed using suitable electronic devices. Other examples of such patterning devices also include programmable LCD arrays. An example of such a configuration is given in US Pat. No. 5,229,872, which is incorporated herein by reference.

[0041] Figure 1An exemplary lithographic projection apparatus 10A is shown. The main components are a radiation source 12A, which may be a deep ultraviolet (DUV) excimer laser source or other type of source, including an extreme ultraviolet (EUV) source (as discussed above, the lithographic projection apparatus need not have a radiation source itself); illumination optics, which, for example, define a partial coherence (expressed as a standard deviation) and may include transmission optics 14A, 16Aa, and 16Ab that shape the radiation from source 12A; a patterning device 18A; and transmission optics 16Ac that projects an image of the patterning device pattern onto a substrate plane 22A. An adjustable filter or aperture 20A at a pupil plane of the projection optics may constrain the range of beam angles impinging on the substrate plane 22A, with the maximum possible angle defining the numerical aperture NA=n sin(Θ) of the projection optics. max ), where n is the refractive index of the medium between the substrate and the last element of the projection optics, and θ max is the maximum angle of the beam emerging from the projection optics that can still impinge on the substrate plane 22A.

[0042] In a lithographic projection apparatus, a source providing irradiation (i.e., radiation) to a pattern forming device and projection optics directs and shapes the irradiation onto a substrate via the pattern forming device. The projection optics may include at least some of the transmission optics 14A, 16Aa, 16Ab, and 16Ac. An aerial image (AI) is a distribution of radiation intensity at the substrate level. A resist model may be used to calculate a resist image based on the aerial image, and an example of this situation may be found in U.S. Patent Application Publication No. US2009-0157630, the entire disclosure of which is hereby incorporated by reference. The resist model is related to the properties of the resist layer (e.g., the effects of chemical processes occurring during exposure, post-exposure baking (PEB), and development). The optical properties of the lithographic projection apparatus (e.g., the properties of irradiation, pattern forming device, and projection optics) indicate the aerial image and may be defined in an optical model. Since the patterning device used in a lithographic projection apparatus can be varied, 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 source and the projection optics. Techniques and models for transforming a design layout into various lithographic images (e.g., aerial images, resist images, etc.), using which optical proximity correction (OPC) is applied and performance (e.g., in terms of process windows) is evaluated, are described in U.S. Patent Application Publication Nos. US2008-0301620, 2007-0050749, 2007-0031745, 2008-0309897, 2010-0162197, and 2010-0180251, the disclosures of each of which are hereby incorporated by reference in their entirety.

[0043] Optical Proximity Correction (OPC) enhances the integrated circuit patterning process by compensating for deformations that occur during processing. The deformations occur during processing because the features printed on the wafer are smaller than the wavelength of light used in the patterning and printing process. OPC verification identifies OPC errors or weaknesses in the post-OPC wafer design that could potentially lead to patterning defects on the wafer. For example, ASML Tachyon Lithography Manufacturability Check (LMC) is an OPC verification product.

[0044] OPC addresses the fact that the final size and placement of the image of the design layout projected onto the substrate will be different from or only depend on the size and placement of the design layout on the patterning device. In the context of resolution enhancement techniques (RET) such as OPC, a physical patterning device does not have to be used, but the design layout can be used to represent the physical patterning device. For the small feature sizes and high feature densities present on some design layouts, the location of a particular edge of a given feature will be affected to some extent by the presence or absence of other adjacent features. These proximity effects arise from tiny amounts of radiation coupled from one feature to another and / or non-geometric optical effects such as diffraction and interference. Similarly, proximity effects can arise from diffusion and other chemical effects during post-exposure baking (PEB), resist development, and etching, typically following lithography.

[0045] In order to increase the chance that the projected image of the design layout is in accordance with the requirements of a given target circuit design, complex numerical models, corrections or pre-distortions of the design layout can be used to predict and compensate for proximity effects. The paper "Full-ChipLithography Simulation and Design Analysis-how OPC Is Changing IC Design" (C. Spence, Proc. SPIE, Vol. 5751, pp. 1-14 (2005)) provides an overview of the current "model-based" optical proximity correction process. In a typical high-end design, almost every feature of the design layout has some modification in order to achieve high fidelity of the projected image to the target design. These OPC modifications can include offsets or biases of edge positions or line widths and / or the application of "auxiliary" features that are expected to assist the projection of other features.

[0046] One of the simplest forms of OPC is selective biasing. Given a CD versus pitch curve, all different pitches can be forced to produce the same CD by changing the CD at the patterning device level, at least under optimal focus and exposure conditions. Thus, if a feature is printed too small at the substrate level, the patterning device level feature will be biased to be slightly larger than the nominal feature, and vice versa. Since the pattern transfer process from the patterning device level to the substrate level is nonlinear, the bias is not simply the CD error measured under optimal focus and exposure conditions multiplied by the reduction ratio, but the appropriate bias can be determined using modeling and experiments. Selective biasing is an incomplete solution to the problem of proximity effects, especially when it is applied only under nominal process conditions. Although in principle such a bias can be applied to give a uniform CD versus pitch curve under optimal focus and exposure conditions, once the exposure process changes from nominal conditions, each bias pitch curve will respond differently, resulting in different process windows for different features. The process window is the range of values ​​of two or more process parameters (e.g., focus and radiation dose in a lithographic apparatus) within which features are sufficiently properly produced (e.g., features with CD within a certain range, such as ±10% or ±5%). Thus, an "optimal" bias for giving the same CD versus pitch may even have a negative impact on the overall process window, thereby narrowing rather than enlarging the focus and exposure range within which all target features are printed on a substrate within the desired process tolerances.

[0047] Other more complex OPC techniques have been developed for applications beyond the one-dimensional biasing example above. A two-dimensional proximity effect is line-end shortening. Line ends have a tendency to "pull back" from their desired endpoint locations depending on exposure and focus. In many cases, the degree of shortening of the ends of long line ends can be several times greater than the corresponding line narrowing. This type of line-end pullback can cause catastrophic failure of the manufactured device if the line end does not completely cross over the underlying layer it is expected to cover (such as the polysilicon gate layer above the source-drain region). Since this type of pattern is extremely sensitive to focus and exposure, it is not enough to simply bias the line end to be longer than the designed length, because the line will be too long under optimal focus and exposure conditions or under underexposure conditions, resulting in short circuits when the extended line end touches adjacent structures, or unnecessarily large circuit sizes if more space is added between individual features in the circuit. Since one of the goals of integrated circuit design and manufacturing is to minimize the area required per chip while maximizing the number of functional elements, adding excess spacing is an undesirable solution.

[0048] Two-dimensional OPC methods can help solve the problem of line end pullback. Additional structures such as "hammer heads" or "wiring" (also referred to as "auxiliary features") can be added to the line ends to effectively anchor the line ends in place and provide reduced pullback throughout the entire process window. Even under optimal focus and exposure conditions, these additional structures are still not resolved, but they change the appearance of the main features without being fully resolved by each alone. "Main features" as used herein mean features that are expected to be printed on the substrate under some or all conditions in the process window. Auxiliary features can take a much more radical form than a simple hammer head added to the line end, so as to achieve a degree to which the pattern on the pattern forming device is no longer just the desired substrate pattern whose size increases according to the reduction ratio. Auxiliary features such as wiring can be applied to more situations than just reducing line end pullback. Internal wiring or external wiring can be applied to any edge, especially a two-dimensional edge, to reduce corner rounding (i.e., chamfering) or edge extrusion. With sufficient selective biasing and assist features of all sizes and polarities, the features on the patterning device bear less and less resemblance to the desired final pattern at substrate level. Typically, the patterning device pattern becomes a pre-distorted version of the substrate level pattern, where the distortion is intended to offset or reverse the pattern deformation that will occur during the manufacturing process to produce a pattern on the substrate that is as close as possible to what the designer intended.

[0049] Another OPC technique also involves the use of completely independent and indistinguishable auxiliary features instead of or in addition to those auxiliary features (e.g., wiring) that are connected to the main features. The term "independent" here means that the edges of these auxiliary features are not connected to the edges of the main features. These independent auxiliary features are not expected or required to be printed on the substrate as features, but are intended to modify the aerial image of nearby main features to enhance the printability and process tolerance of the main features. These auxiliary features (often referred to as "scattering bars" or "SBARs") can include: sub-resolution auxiliary features (SRAFs), which are features outside the edges of the main features; and sub-resolution inverse features (SRIFs), which are features dug out from the inside of the edges of the main features. The presence of SBARs adds another layer of complexity to the pattern forming device pattern. A simple example of the use of scattering bars is where a regular array of indistinguishable scattering bars is drawn on both sides of an isolated line feature, which has the effect of making the isolated line appear from an aerial image point of view as a single line within an array of dense lines more representative, resulting in a process window that is closer in focus and exposure tolerances to those of the dense pattern. The common process window between such a decorated isolated feature and the dense pattern will have a greater common tolerance to focus and exposure variations than would be the case if the feature were drawn as isolated at patterning device level.

[0050] Assist features can be considered the difference between features on the patterning device and features in the design layout.The terms "main feature" and "assist feature" do not imply that a particular feature on the patterning device must be labeled as a main feature or an assist feature.

[0051] Another aspect of understanding the lithographic process is understanding the interaction of the radiation with the patterning device. The electromagnetic field of the radiation after the radiation passes through the patterning device can be determined based on the electromagnetic field of the radiation before it reaches the patterning device and a function that characterizes the interaction. Such a function can be called a mask transmission function (which can be used to describe the interaction by the transmission patterning device and / or the reflection patterning device).

[0052] The mask transmission function can have a variety of different forms. One form is binary. A binary mask transmission function has any of two values ​​(e.g., zero and a positive constant) at any given location on the pattern forming device. A mask transmission function in binary form can be referred to as a binary mask. Another form is continuous. That is, the modulus of the transmittance (or reflectance) of the pattern forming device is a continuous function of the location on the pattern forming device. The phase of the transmittance (or reflectance) can also be a continuous function of the location on the pattern forming device. A mask transmission function in continuous form can be referred to as a continuous tone mask or a continuous transmission mask (CTM). For example, a CTM can be represented as a pixelated image, in which each pixel can be assigned, that is, assigned a value between 0 and 1 (e.g., 0.1, 0.2, 0.3, etc.) instead of a binary value of 0 or 1. In an embodiment, the CTM may be a pixelated grayscale image where each pixel has multiple values ​​(eg, in the range [-255, 255], normalized values ​​in the range [0, 1] or [-1, 1] or other suitable range).

[0053] The thin mask approximation, also known as the Kirchhoff boundary condition, is widely used to simplify the determination of the interaction of the radiation with the patterning device. The thin mask approximation assumes that the thickness of the structures on the patterning device is very small compared to the wavelength, and the width of the structures on the mask is very large compared to the wavelength. Therefore, the thin mask approximation assumes that the electromagnetic field after the patterning device is the multiplication of the incident electromagnetic field and the mask transmission function. However, as lithography processes use radiation with shorter and shorter wavelengths, and the structures on the patterning device become smaller and smaller, the assumption or assumption of the thin mask approximation breaks down. For example, the interaction of radiation with structures (such as the edge between the top surface and the sidewalls) can become significant due to the finite thickness of the structures ("mask 3D effect" or "M3D"). Including this scattering in the mask transmission function may enable the mask transmission function to better capture the interaction of the radiation with the patterning device. The mask transmission function under the thin mask approximation may be referred to as a thin mask transmission function. A mask transmission function covering M3D may be referred to as an M3D mask transmission function.

[0054] Figure 2 1 is a flow chart of a method 200 for determining a patterning device pattern (or mask pattern hereinafter) from an image (e.g., a continuous transmission mask image, a binary mask image, a curvilinear mask image, etc.) corresponding to a target pattern to be printed on a substrate via a patterning process involving a photolithography process. In embodiments, the design layout or the target pattern may be a binary design layout, a continuous tone design layout, or another suitable form of design layout.

[0055] Method 200 is an iterative process in which an initial image (e.g., an enhanced image, mask variables initialized from a CTM image, etc.) is progressively modified to produce different types of images according to different processes of the present disclosure to ultimately produce information including a mask pattern or image (e.g., mask variables corresponding to a final curved mask) that is further used to make / manufacture a mask. The iterative modification of the initial image may be based on a cost function, wherein during the iteration, the initial image may be modified so that the cost function is reduced, and in an embodiment minimized. In an embodiment, method 200 may also be referred to as a binarized CTM process, wherein the initial image is an optimized CTM image, and the optimized CTM image is further processed according to the present disclosure to produce a curved mask pattern (e.g., a geometric shape or polygonal representation of a curved mask or curved pattern). In an embodiment, the initial image may be an enhanced image of a CTM image. The curved mask pattern may be in the form of a vector, a table, a mathematical equation, or other form representing a geometric / polygonal shape.

[0056] In an embodiment, process P201 may involve obtaining an initial image (e.g., a CTM image or an optimized CTM image, or a binary mask image). In an embodiment, the initial image 201 may be a CTM image generated by a CTM generation process based on a target pattern to be printed on a substrate. The CTM image may then be received by the process P201. In an embodiment, the process P201 may be configured to generate a CTM image. For example, in CTM generation technology, the inverse lithography problem is formulated as an optimization problem. The variables are related to the values ​​of pixels in the mask image, and lithography indicators such as EPE or sidelobe printing are used as cost functions. In an iteration of the optimization, the mask image is constructed from the variables and then a process model (e.g., a Tachyon model) is applied to obtain an optical or resist image and a cost function is calculated. The cost calculation then gives a gradient value, which is used in an optimization solver to update variables (e.g., pixel intensity). After several iterations during optimization, a final mask image is generated, which is additionally used as a guide map for pattern extraction (e.g., as implemented in Tachyon SMO software). Such an initial image (e.g., a CTM image) may include one or more features corresponding to the target pattern to be printed on the substrate via the patterning process (e.g., features of the target pattern, SRAF, SRIF, etc.).

[0057] In an embodiment, a CTM image (or an enhanced version of a CTM image) may be used to initialize mask variables that may be used as the initial image 201, which is iteratively modified as discussed below.

[0058] Process P201 may involve generating an enhanced image 202 based on the initial image 201. The enhanced image 202 may be an image in which certain selected pixels within the initial image 201 are amplified. The selected pixels may be pixels within the initial image 201 having relatively low values ​​(or weak signals). In an embodiment, the selected pixels are pixels having signal values ​​lower than, for example, the average intensity of pixels throughout the initial image or a given threshold. In other words, pixels within the initial image 201 having weaker signals are amplified, thereby enhancing one or more features within the initial image 201. For example, a second-order SRAF around a target feature may have a weak signal that may be amplified. Thus, the enhanced image 202 may highlight or identify additional features (or structures) that may be included in a mask image (generated later in the method). In conventional methods (e.g., CTM methods) for determining a mask image, weak signals within the initial image may be ignored, and as such, the mask image may not include features that may be formed by weak signals in the initial image 201.

[0059] The generation of the enhanced image 202 involves applying an image processing operation such as a filter (e.g., an edge detection filter) to amplify weak signals within the initial image 201. Alternatively or in addition, the image processing operation may be deblurring, averaging, and / or feature extraction or other similar operations. Examples of the edge detection filter include Prewitt operators, Laplacian operators, Laplacian of Gaussian (LoG) filters, etc. The generation step may also involve combining the amplified signal of the initial image 201 with the original signal of the initial image 201 with or without modifying the original strong signal of the initial image 201. For example, in an embodiment, for one or more pixel values ​​at one or more locations (e.g., at a contact hole) across the initial image 201, the original signal may be relatively strong (e.g., above a threshold value such as 150 or below -50), then the original signal at the one or more locations (e.g., at the contact hole) may not be modified or combined with the amplified signal for the location.

[0060] In an embodiment, the noise (e.g., random variations in brightness or color or pixel values) in the initial image 201 may also be amplified. Therefore, alternatively or additionally, a smoothing process may be applied to reduce the noise (e.g., random variations in brightness or color or pixel values) in the combined image. Examples of image smoothing methods include Gaussian blur, moving average, low-pass filter, etc.

[0061] In an embodiment, an edge detection filter may be used to generate the enhanced image 202. For example, an edge detection filter may be applied to the initial image 201 to generate a filtered image that highlights the edges of one or more features within the initial image 201. The resulting filtered image may be further combined with the original image (i.e., the initial image 201) to generate the enhanced image 202. In an embodiment, the combination of the initial image 201 and the image obtained after edge filtering may involve modifying only those parts of the initial image 201 with weak signals without modifying areas with strong signals, and the combination process may be weighted based on signal strength. In an embodiment, amplifying the weak signals may also amplify the noise within the filtered image. Therefore, according to an embodiment, a smoothing process may be performed on the combined image. Smoothing of an image may refer to an approximation function that attempts to capture important patterns (e.g., target patterns, SRAF) in an image while leaving out noise or other fine-scale structures / fast phenomena. In smoothing, the data points of the signal may be modified so that individual points (roughly due to noise) may be reduced, and points that may be lower than neighboring points may be increased, resulting in a smoother signal or a smoother image. Thus, after the smoothing operation, a further smoothed version of the enhanced image 202 with reduced noise may be obtained according to an embodiment of the present disclosure.

[0062] In process P203, the method may involve generating mask variables 203 based on the enhanced image 202. In a first iteration, the enhanced image 202 may be used to initialize the mask variables 203. In later iterations, the mask variables 203 may be iteratively updated.

[0063] The contour extraction of a real-valued function f of n real variables is a collection of the following form:

[0064] L c (f) = {(x1, x2, ... x n )|f(x1, x2, ...x n )=c}

[0065] In two-dimensional space, the set defines the points on the surface where the function f is equal to a given value c. In two-dimensional space, the function f is able to extract closed contours to be rendered to the mask image.

[0066] In the above equation, x1,x2,…x n Refers to a mask variable, such as the intensity of an individual pixel, that determines where a curved mask edge exists at a given constant value c (e.g., at a threshold plane as discussed below in process P205).

[0067] In an embodiment, at an iteration, the generation of the mask variable 203 may involve modifying one or more values ​​of a variable within the enhanced image 202 (e.g., pixel values ​​at one or more locations) based on, for example, initialization conditions or a gradient map (which may be generated later in the method). For example, the one or more pixel values ​​may be increased or decreased. In other words, the amplitude of one or more signals within the enhanced image 202 may be increased or decreased. Such modified amplitudes of signals enable different curve patterns to be generated depending on the amount of change in the amplitude of the signal. Thus, the curve pattern gradually evolves until the cost function is reduced (minimized in an embodiment). In an embodiment, further smoothing may be performed on the horizontal mask variable 203.

[0068] In addition, process P205 involves generating a curved mask pattern 205 (e.g., having a polygonal shape represented in vector form) based on the mask variables 203. The generation of the curved mask pattern 205 may involve thresholding of the mask variables 203 to trace or generate a curved (or curved) pattern from the mask variables 203. For example, thresholding may be performed using a threshold plane (e.g., an xy plane) having a fixed value that intersects the signal of the mask variables 203. The intersection of the threshold plane with the signal of the mask variables 203 generates a trace or contour (i.e., a curved polygonal shape), which forms a polygonal shape used as a curved pattern for the curved mask pattern 205. For example, the mask variables 203 may intersect a zero plane parallel to the (x, y) plane. Thus, the curved mask pattern 205 may be any curved pattern generated as above. In an embodiment, the curved pattern traced or generated from the mask variables 203 depends on the signal of the enhanced image 202. As such, the image enhancement process P203 results in an improvement in the pattern produced for the final curvilinear mask pattern.The final curvilinear mask pattern may be further used by a mask manufacturer to manufacture masks for use in a lithography process.

[0069] Process P207 may involve rendering the curved mask pattern 205 to produce a mask image 207. Rendering is an operation performed on the curved mask pattern, which is a process similar to converting a rectangular mask polygon into a discrete grayscale image representation. This process can generally be understood as sampling a box function of continuous coordinates (polygons) into values ​​at each point of an image pixel.

[0070] The method further involves performing a forward simulation of the patterning process using a process model, the process model generating or predicting a pattern that can be printed on the substrate based on the mask image 207 as a process model image 209. For example, process P209 may involve executing and / or simulating the process model using the mask image 207 as an input, and generating a process model image 209 (e.g., an aerial image, a resist image, an etch image, etc.) on the substrate. In an embodiment, the process model may include a mask transmission model coupled to an optics model, which is further coupled to a resist model and / or an etch model (e.g., as described below). The output of the process model may be a process model image 209 that factors in different process variations during the simulation process. The process image may be further used to determine parameters of the patterning process (e.g., EPE, CD, overlap, side lobes, etc.) by, for example, tracing the contours of the pattern within the process image. The parameters may also be used to define a cost function, which is further used to optimize the mask image 207 so that the cost function is reduced, or minimized in an embodiment.

[0071] In process P211, a cost function may be evaluated based on the process model image 209 (also referred to as a simulated substrate image or substrate image or wafer image). Thus, the cost function may be considered process-aware in the case of variations in the patterning process, thereby enabling the generation of a curved mask pattern that takes into account variations in the patterning process. For example, the cost function may be an edge placement error (EPE), a side lobe, a mean square error (MSE), a pattern placement error (PPE), a normalized image logarithm, or other suitable variables defined based on the pattern contour in the process image. The EPE may be an edge placement error associated with one or more patterns, and / or the sum of all edge placement errors associated with all patterns of the process model image 209 and the corresponding target pattern. In an embodiment, the cost function may include more than one condition that may be reduced or minimized simultaneously. For example, in addition to the probability of MRC violation, the number of defects, EPE, overlap, CD or other parameters may also be included, and all conditions may be reduced (or minimized) simultaneously.

[0072] In addition, one or more gradient maps may be generated based on the cost function (e.g., EPE), and mask variables may be modified based on such gradient maps. Therefore, the gradient calculation can be expressed as And the gradient values ​​are updated by capturing the inverse mathematical relationship from the mask image (MI) to the curvilinear mask polygons to the mask variables. Thus, a chain of derivatives of the cost function with respect to the mask image can be calculated from the mask image to the curvilinear mask polygons and from the curvilinear mask polygons to the mask variables, which allows to modify the values ​​of the mask variables at the mask variables.

[0073] In an embodiment, image regularization can be added to reduce the complexity of the mask pattern that can be produced. Such image regularization can be a mask rule check (MRC). MRC refers to the constraints of the mask manufacturing process or equipment. Thus, the cost function can include different components, for example, based on EPE and MRC violation penalties. The penalty can be an item of the cost function, which depends on the amount of violation, such as the difference between the mask measurement value and a given MRC or mask parameter (e.g., the mask pattern width and the allowed (e.g., minimum or maximum) mask pattern width). Thus, according to an embodiment of the present disclosure, a mask pattern can be designed, and a corresponding mask can be made not only based on the forward simulation of the patterning process, but also based on the manufacturing constraints of the mask manufacturing equipment / process. Thus, a manufacturable curve mask with high yield (i.e., minimum defects) and high accuracy can be obtained based on, for example, the overlap on the EPE or printed pattern.

[0074] The pattern corresponding to the process image should be exactly the same as the target pattern, however, such an exact target pattern may not be feasible (e.g., typically sharp corners), and some conflicts are introduced due to variations in the patterning process itself and / or approximations in the model of the patterning process. In the first iteration of the method, the mask image 207 may not produce a pattern (in the resist image) that is similar to the target pattern. The determination of the accuracy or acceptance of the printed pattern in the resist image (or etching image) can be based on a cost function such as EPE. For example, if the EPE of the resist pattern is high, it indicates that the printed pattern using the mask image 207 is unacceptable and the pattern in the mask variable 203 must be modified.

[0075] To determine whether the mask image 207 is acceptable, process P213 may involve determining whether the cost function is reduced or minimized, or whether a given number of iterations has been reached. For example, the EPE value of the previous iteration may be compared with the EPE value of the current iteration to determine whether the EPE has been reduced, minimized, or converged (i.e., no significant improvement in the printed pattern is observed). When the cost function is minimized, the method may stop, and the resulting curve mask pattern information is considered to be an optimization result.

[0076] However, if the cost function is not reduced or minimized, the mask-related variables or enhanced image-related variables (e.g., pixel values) may be updated. In an embodiment, the update may be based on a gradient-based approach. For example, if the cost function is not reduced, the method 200 continues to the next iteration of generating the mask image after executing processes P215 and P217 indicating how to further modify the mask variables 203.

[0077] Process P215 may involve generating a gradient map (labeled as reference numeral 215) based on the cost function. The gradient map may be a derivative and / or partial derivative of the cost function. In an embodiment, the partial derivatives of the cost function may be determined with respect to the pixels of the mask image, and the derivatives may be further linked to determine the partial derivatives with respect to the mask variables 203. Such gradient calculations may involve determining an inverse relationship between the mask image 207 and the mask variables 203. Furthermore, the inverse relationship of any smoothing operations (or functions) performed in processes P205 and P203 must be considered.

[0078] The gradient map 215 may provide suggestions for increasing or decreasing the value of the mask variable in a manner that reduces (in an embodiment minimizes) the value of the cost function. In an embodiment, an optimization algorithm may be applied to the gradient map 215 to determine the mask variable value. In an embodiment, an optimization solution process may be used to perform gradient-based calculations (in process P217).

[0079] In an embodiment, for an iteration, the mask variables may be changed, while the threshold plane may remain fixed or unchanged in order to gradually reduce or minimize the cost function. Thus, the resulting curve pattern may gradually evolve during the iteration so that the cost function is reduced, or in an embodiment, minimized. In another embodiment, both the mask variables and the threshold plane may be changed to achieve faster convergence of the optimization process. A final set of binarized CTM results (i.e., a modified version of an enhanced image, mask image, or curve mask) may be generated after several iterations and / or minimizations of the cost function.

[0080] In an embodiment of the present disclosure, the transition from CTM optimization performed with a grayscale image to a binarized CTM optimization performed with a curved mask can be simplified by replacing the threshold setting process (i.e., P203 and P205) by a different process, where a sigmoid transform is applied to the enhanced image 202 and a corresponding change in the gradient calculation is performed. The sigmoid transform of the enhanced image 202 produces a transformed image that gradually evolves into a curved pattern during the optimization process (e.g., minimizing the cost function). During the iteration or optimization step, variables associated with the sigmoid function (e.g., steepness and / or threshold) can be modified based on the gradient calculation. Since the sigmoid transform becomes steeper during successive iterations (e.g., the steepness of the slope of the sigmoid transform increases), a gradual transition from the CTM image to the final binarized CTM image can be achieved, thereby allowing improved results for the final binarized CTM optimization performed with a curved mask pattern.

[0081] In embodiments of the present disclosure, additional steps / processes may be inserted into the optimized iterative loop to enhance the result to have selected or desired properties. For example, smoothness may be ensured by adding a smoothing step, or other filters may be used to enhance the image to favor horizontal / vertical structure.

[0082] The present method has several features or aspects. For example, a CTM mask image optimized using an image enhancement method is used to improve the signal, which can also be used as a seeding in the optimization process. In another aspect, a threshold setting method (called binarized CTM) using CTM technology is used to enable the generation of a curved mask pattern. In another aspect, a complete formulation of gradient calculation (i.e., closed loop formulation) also allows the use of a gradient-based solution process for mask variable optimization. The binarized CTM result can be used as a local solution (as a hotspot repair) or as a full chip solution. The binarized CTM result can be used as an input together with machine learning. This can allow the use of machine learning to accelerate the binarized CTM. In another aspect, the method includes an image regularization method to improve the result. In another aspect, the method involves a continuous optimization stage to achieve a smoother transition from a grayscale image CTM to a binary curve mask binarized CTM. The method allows tuning of the optimized threshold to improve the result. The method includes additional transformations to the optimized iteration to strengthen the good properties of the result (requiring smoothness in the binarized CTM image).

[0083] As lithography nodes continue to shrink, more and more complex masks are needed. The method can be used in key layers using DUV scanners, EUV scanners, and / or other scanners. The method according to the present disclosure can be included in different aspects of the mask optimization process (including source mask optimization (SMO), mask optimization, and / or OPC).

[0084] As described above, it is often desirable to be able to computationally determine how a patterning process will produce a desired pattern on a substrate. Thus, a simulation can be provided to simulate one or more portions of the process. For example, it is desirable to be able to simulate the photolithography process of transferring a patterning device pattern to a resist layer of a substrate and the resulting pattern in the resist layer after development of the resist.

[0085] Figure 3 An exemplary flow chart for simulating lithography in a lithographic projection apparatus is illustrated in FIG. 331. An illumination model 331 represents the optical properties of the illumination (including radiation intensity distribution and / or phase distribution). A projection optics model 332 represents the optical properties of the projection optics (including changes in radiation intensity distribution and / or phase distribution caused by the projection optics). A design layout model 335 represents the optical properties of a design layout (including changes in radiation intensity distribution and / or phase distribution caused by a given design layout), the design layout being a representation of the arrangement of features formed on or by a pattern forming device. An aerial image 336 may be simulated using the illumination model 331, the projection optics model 332, and the design layout model 335. A resist image 338 may be simulated from the aerial image 336 using a resist model 337. The simulation of lithography may, for example, predict the profile and / or CD in the resist image.

[0086] More specifically, the illumination model 331 can represent the optical properties of the illumination, including but not limited to the NA-mean square deviation (σ) setting, and any specific illumination shape (e.g., off-axis illumination, such as annular, quadrupole, dipole, etc.). The projection optics model 332 can represent the optical properties of the projection optics, including, for example, aberrations, deformations, refractive index, physical size or dimensions, etc. The design layout model 335 can also represent one or more physical properties of a physical pattern forming device, such as described in, for example, U.S. Patent No. 7,587,704, which is incorporated herein by reference in its entirety. The optical properties associated with the lithographic projection apparatus (e.g., the properties of the illumination, the pattern forming device, and the projection optics) specify the aerial image. Since the pattern forming device used in the lithographic projection apparatus can be changed, it is desirable to separate the optical properties of the pattern forming device from the optical properties of the rest of the lithographic projection apparatus (hence the design layout model 335), including at least the illumination and the projection optics.

[0087] The resist model 337 can be used to calculate the resist image from the aerial image, an example of which can be found in U.S. Pat. No. 8,200,468, which is hereby incorporated by reference in its entirety. The resist model is typically related to the properties of the resist layer (e.g., the effects of chemical processes occurring during exposure, post-exposure baking, and / or development).

[0088] The goal of the simulation is to accurately predict, for example, edge placement, aerial image intensity slope, and / or CD, which can then be compared to an intended design, which is typically defined as a pre-OPC design layout that can be provided in a standardized digital file format such as GDSII, OASIS, or another file format.

[0089] From the design layout, one or more portions referred to as "clips or chips" may be identified. In an embodiment, a set of clips is extracted, which represent complex patterns (e.g., pattern sets) in the design layout (typically about 50 to 1000 clips, but any number of clips may be used). As will be appreciated by those skilled in the art, these patterns or clips represent small portions of the design (e.g., circuits, cells, etc.) for which specific attention and / or verification (e.g., pattern sets) is required. In other words, a clip may be a portion of the design layout, or may be similar, or have similar behavior of a portion of the design layout whose critical features are identified empirically (including clips provided by a customer), identified by trial and error, or identified by performing full chip simulations. A clip typically contains one or more test patterns or gauge patterns. An initial larger set of clips may be provided a priori by a customer based on known critical feature areas in the design layout that require specific image optimization. Alternatively, in another embodiment, an initial larger set of clips may be extracted from the entire design layout by using the process of identifying the critical feature areas described below.

[0090] In some examples, the simulation and modeling can be used to configure one or more features of the pattern forming device pattern (e.g., to perform optical proximity effect correction), one or more features of the illumination (e.g., to change one or more characteristics of the spatial / angular intensity distribution of the illumination, such as changing the shape), and / or one or more features of the projection optical device (e.g., numerical aperture, etc.). Such configurations can be generally referred to as mask optimization, source optimization, and projection optimization, respectively. Such optimizations can be performed independently or combined in different combinations. One such example is source-mask optimization (SMO), which involves configuring one or more features of the pattern forming device pattern together with one or more features of the illumination. Optimization techniques can focus on one or more clips among a plurality of clips. The optimization can use the machine learning models described herein to predict the values ​​of various parameters (including images, etc.).

[0091] In some embodiments, illumination model 331, projection optics model 332, design layout model 335, resist model 337, SMO model, and / or other models associated with and / or included in the integrated circuit manufacturing process may be empirical models for performing the operations of the methods described herein. The empirical models may predict outputs based on correlations between various inputs (e.g., one or more characteristics of a mask or wafer image, one or more characteristics of a design layout, one or more characteristics of the patterning device, one or more characteristics of the illumination used in the lithography process (such as wavelength), etc.).

[0092] As an example, the empirical model can be a machine learning model. In some embodiments, the machine learning model can be and / or include a mathematical equation, an algorithm, a plotted graph, a chart, a network (e.g., a neural network), and / or other tools and machine learning model components. For example, the machine learning model can be and / or include one or more neural networks having an input layer, an output layer, and one or more intermediate or hidden layers. In some embodiments, the one or more neural networks can be and / or include a deep neural network (e.g., a neural network having one or more intermediate or hidden layers between the input layer and the output layer).

[0093] As an example, the one or more neural networks may be based on a large collection of neural units (or artificial neurons). The one or more neural networks may loosely, i.e., not strictly, mimic the way a biological brain works (e.g., via large clusters of biological neurons connected by axons). Each neural unit of a neural network may be connected to many other neural units of the neural network. Such connections may strengthen or inhibit their influence on the activation state of the connected neural units. In some embodiments, each individual neural unit may have a summation function that combines the values ​​of all its inputs. In some embodiments, each connection (or neural unit itself) may have a threshold function such that the signal must exceed the threshold before it is allowed to propagate to other neural units. These neural network systems may be self-learning and trained, rather than explicitly programmed, and may perform significantly better in certain problem-solving areas compared to traditional computer programs. In some embodiments, one or more neural networks may include multiple layers (e.g., where the signal path traverses from the front layer to the back layer). In some embodiments, backpropagation techniques may be utilized by a neural network, where forward stimulation is used to reset weights for the "front end" neural units. In some embodiments, stimulation and inhibition for one or more neural networks may flow more freely, where connections interact in a more chaotic and complex manner. In some embodiments, one or more intermediate layers of the neural network include one or more convolutional layers, one or more recurrent layers, and / or other layers.

[0094] The one or more neural networks may be trained (i.e., their parameters determined) using a set of training data. The training data may include 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). 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. For example, given a model of {(x1, y1), (x2, y2), ..., (xN ,y N )} is a set of N training samples such that x i is the eigenvector of the i-th example and y i In the case of a signal for which it is managed, the training algorithm finds 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 that represents a numerical feature of a certain target (e.g., a chip design in the example above). The vector space associated with these vectors is often called a feature space. After training, the neural network can be used to make predictions using new samples.

[0095] For example, as described above, in order to train DCNN and / or other machine learning models to predict CTM mappings, users are required to generate CTM images as training data (e.g., via ASML's Tachyon product). However, it is difficult to select an appropriate representative portion of the pattern in the full-chip GDS as a training pattern set to train the DCNN to predict the CTM mapping. Manual selection requires the user to have significant prior knowledge, i.e., preliminary knowledge, about the full-chip GDS design and to invest several hours in the selection process. Although random selection is not time-consuming relative to manual selection, the unclear pattern coverage and low stability associated with random selection make random selection infeasible or unsustainable in real-world applications.

[0096] To address these and other disadvantages of previous systems, the present method and apparatus provide users with an efficient tool to automatically perform training pattern selection from a full chip GDS file.The present method and apparatus are configured to automatically select representative patterns from a pattern set in less time than prior art systems.

[0097] Figure 4 An overview of the operations of the present method 400 for determining training patterns during wafer patterning is illustrated. Figure 4 The method shown in is a method or part of a method for training a machine learning model in a wafer patterning process (e.g., as described herein). The operations include: generating 441 (e.g., using a feature generation engine) multiple features from unique patterns in a pattern set 443 to form an enhanced pattern set 445 with additional features; grouping 442 (e.g., using a pattern clustering engine) the patterns in the enhanced pattern set 445 into multiple separate groups 447 based on the similarity of the multiple generated features; and selecting 444 (e.g., using a pattern selection engine) representative patterns 448 from the multiple separate groups 447 to determine the training patterns 449. The selected representative patterns 448 from the multiple separate groups 447 (e.g., which form the training patterns 449) can be provided to a machine learning model (e.g., a DCNN, Figure 4The machine learning model may be used to train the machine learning model to predict a continuous transmission mask (CTM) map for optical proximity correction (OPC) in the wafer patterning process, and / or may be provided for other applications. For example, the patterns in the pattern set 443 may be, for example, a unique pattern library 446 generated by, for example, a collect patterns from layout (PCL) function 451 in a Tachyon PRO (Pattern Recognition and Optimization) product (and / or other similar products) based on a full chip layout GDS file 453, and / or may be part of the unique pattern library 446.

[0098] In operation 441, the plurality of generated features include geometric features, lithography-perceptual features, and / or other features. The geometric features include one or more of the following: a target mask image, a frequency map, a pattern density map, a pattern occurrence rate of a unique pattern in a pattern set, and / or other geometric features. The lithography-perceptual features include one or more of the following: a sub-resolution auxiliary feature guide map (SGM), a diffraction order, a diffraction pattern of a unique pattern in a pattern set, and / or other lithography-perceptual features. Geometric features and lithography-perceptual features are generated for each of a plurality of patterns in a pattern set. In an embodiment, a feature set (including geometric and / or lithography-perceptual features) is generated for each pattern (in set 443). The plurality of features generated from a plurality of patterns in a pattern set are features in addition to the geometric information and / or vertex information already included in the pattern set.

[0099] For example, a pattern set 446 (e.g., a pattern library (PLIB)) generated by PCL 451 may include geometric information related to multiple individual unique patterns. However, only vertex information for an individual pattern may be stored as part of pattern set 446 to represent a specific pattern design. For example, vertex information is easy to use for precise pattern matching, but vertex information is insufficient to guide the pattern grouping method to fully group similar patterns (they are not exactly the same) because such a process is a fuzzy matching process. To improve the robustness of pattern grouping in operation 442, operation 441 includes generating additional features for multiple individual patterns (e.g., in addition to vertex geometric information), including both geometric and lithography-perceptual features. As described above, the geometric features include one or more of the following: a target mask image, a frequency map, a pattern density map, a pattern occurrence rate of a unique pattern in a pattern set, and / or other geometric features. The lithography-perceptual features include one or more of the following: a sub-resolution auxiliary feature guide map (SGM), a diffraction order, a diffraction pattern of a unique pattern in a pattern set, and / or other lithography-perceptual features. However, this specification is not intended to be limiting. Operation 441 may be customized by the user to extend to additional and / or different features generated according to the user's specifications.

[0100] Figure 5 Picture shows Figure 4 The features shown in FIG. 4 are aspects of operation 441. Figure 5 As shown, operation 441 includes generating geometric features 502 and lithography-sensing features 504. Figure 4 ) generates a geometric feature 502 and a lithographically-perceptible feature 504 for each of the patterns 501 in the image processing engine 441 such that a set of features 502, 504 is generated for each pattern 501. Examples of geometric features 502 generated as part of operation 441 (e.g., by a feature generation engine) include a target mask image 506, a frequency map 508, a pattern density map 510, and a pattern occurrence rate (count) 512 of the pattern in the set of patterns. Examples of lithographically-perceptible features 504 generated as part of operation 441 (e.g., by a feature generation engine) include an SGM map 514, a diffraction order 516, and a diffraction pattern 518. These features are examples only and are not intended to be limiting.

[0101] Back to Figure 4 In operation 442, the patterns in the enhanced pattern set 445 are grouped into a plurality of separate groups based on similarities of a plurality of generated features (e.g., generated at operation 441). Operation 442 is performed using unsupervised machine learning. In an embodiment, the grouping includes clustering and / or other grouping operations. The clustering includes a continuous series of clustering steps 455, 457, 459, which are performed using different features of the plurality of generated features for different clustering steps 455, 457, 459. Figure 4 In the example shown in FIG. 4 , clustering step 455 may be performed based on the first feature, clustering step 457 may be performed based on the second feature, and clustering step 459 (and / or any other intermediate clustering steps) may be performed based on the nth feature. The continuous series of clustering steps 455, 457, 459 form corresponding subsets 461, 463, 465 of unique patterns in the pattern set, such that the representative pattern 448 (operation 444) is selected from the subsets 461, 463, 465 to determine the training pattern 449.

[0102] Figure 6 The diagram shows Figure 4 442. For example, Figure 6 The diagram shows a method based on multiple generated features (e.g., Figure 4The patterns in the enhanced pattern set 445 are clustered into a plurality of separate groups based on similarity (generated by, for example, a feature generation engine) at operation 441 shown in FIG. 4. Because the patterns exhibit multi-dimensional features, clustering can be performed based on a plurality of separate features. Unsupervised machine learning can be used to perform clustering on each pattern. Figure 6 The clusters shown in FIG. 1 consider multiple features for each pattern in a stepwise manner to group together patterns that share similarities in multidimensional features (e.g., Figure 6 ) For example, the user can define which features should be used to cluster the patterns, and the sequence of features used for clustering. Figure 6 In the example shown in , clustering includes a continuous series of clustering steps 455, 457, 459, which are performed using different features among the multiple features generated by different clustering steps 455, 457, 459 (for example, the identities and order of these clustering steps are defined by the user). For example, clustering step 455 can be performed based on a first feature, clustering step 457 can be performed based on a second feature, and clustering step 459 (and / or any other intermediate clustering steps) can be performed based on the nth feature. The continuous series of clustering steps 455, 457, 459 form corresponding subgroups 461 (for example, groups A and B), 463 (for example, groups Aa, Ab, Ba, Bb, Bc), and 465 (for example, Aan, Abn, Ban, Bbn, Bcn, etc.) of unique patterns. For example, Figure 6 The patterns in the groups in the nth level of the hierarchy shown in (clustering step 459) share similarities of n features.

[0103] In an embodiment, the continuous series of clustering steps 455, 457, 459 includes a cross-validation step performed using a given feature for a given step. The cross-validation step includes: adjusting the number of groups clustered at multiple individual clustering steps 455, 457, 459, determining and / or adjusting which unique patterns are included in a given subgroup, and / or other cross-validation operations. In an embodiment, the cross-validation step includes: optimizing the number of groups clustered at multiple individual clustering steps 455, 457, 459, and optimizing which unique patterns are included in a given subgroup, and / or other cross-validation operations.

[0104] In an embodiment, clustering and / or cross validation comprises a machine learning clustering method, such as k-means clustering and / or other clustering methods. K-means clustering is configured to partition n observations into k clusters, where each observation belongs to the cluster with the closest mean, which serves as a representative example of the cluster. This results in the partitioning of the data space into Voronoi cells. Given a set of patterns (x1, x2, ..., xn ) and one of its corresponding features, the feature can be regarded as a d-dimensional real vector, so that the k-means cluster is configured to partition the n observations into k (≤n) sets S = {S1, S2, ..., S k} to minimize the within-cluster sum of squares. Formally, the goal is to find:

[0105]

[0106] where μ i YesS i The mean of the points in .

[0107] Although k-means clustering is an efficient clustering method, one of its disadvantages is that it requires the number of clusters k to be specified before clustering occurs. In order to address the need for pre-selecting a certain number of clusters for each k-means clustering process, the present apparatus and method are configured to perform a k-fold cross validation test using an elbow method to determine the value of k (e.g., an optimal value). The elbow method includes running k-means clustering on a data set for a range of k values ​​(e.g., k is from 1 to 14, which is not intended to be limiting), and for each k value, determining the sum of squared errors (SSE). After determining the SSE, a line graph of the SSE is drawn for each k value.

[0108] Figure 7 An example line graph (labeled as reference numeral 700) of SSE 702 versus k value 704 is illustrated. If the line graph 700 looks similar to an arm, then the "elbow" 706 on the arm corresponds to the optimal value of k. In principle, a relatively small SSE is preferred, but the SSE tends to decrease toward zero as k increases (the SSE is 0 when k is equal to the number of data points in the data set because each data point is its own cluster and there is no error between a data point and the center of its cluster). The present method and apparatus are configured to determine smaller values ​​of k that still have a low SSE, and the "elbow" 706 in the line graph 700 corresponds to the location in the line graph 700 where increasing k produces diminishing returns. For example, at Figure 7 , the "elbow" 706 is located at k=5, indicating that for this example data set, the optimal k is 5. As k increases above 5, the SSE 702 changes less and less as it approaches zero.

[0109] The present method and apparatus are configured such that a k-fold cross validation test is performed for the determination of the sum of squared errors for each value of k. In the k-fold cross validation, the original sample is randomly divided into p sub-samples of equal size. A single sub-sample of the p sub-samples is retained as validation data for determining the sum of squared errors, and the remaining p-1 sub-samples are used as training data for the k-means cluster. The cross validation process is then repeated p times, with each of the p sub-samples being used exactly once as validation data. The p results can then be averaged to produce a single estimate of the sum of squared errors. The advantage of this method over repeated random sub-sampling is that multiple separate observations are used for both training and validation, and multiple separate observations are used for validation exactly once.

[0110] By way of non-limiting example, Figure 8 The figure shows k-fold cross validation, where k = 4. Figure 8 As shown, the original sample 800 is randomly divided into p sub-samples 802 of equal size (e.g., five points in each sub-sample 802). Figure 8 , there are p=4 subsamples in sample 800. A single subsample 804 of the p=4 subsamples is retained as validation data 806 for determining the sum of squared errors, and the remaining p-1 subsamples 802 are used as training data 808 for k-means clustering. The cross-validation process is then repeated p times, with each of the p subsamples used exactly once as validation data (e.g., 810, 812, 814). The p results can then be averaged to produce a single estimate of the sum of squared errors.

[0111] Back to Figure 4 , selecting 444 representative patterns 448 from a plurality of individual groups 447 to determine training patterns 449 includes selecting a target number 467 (“M”) of representative patterns 448 from a corresponding number of groups 447 (e.g., groups G(1), G(2), …, G(M)). In an embodiment, selecting representative patterns 448 from a plurality of individual groups 447 of a target number 467 to determine training patterns 449 includes selecting a center-most pattern 469 (e.g., from a plurality of individual groups 447 of a target number 467 (G(1), G(2), …, G(M)) Figure 4 CP(1), CP(2), ..., CP(M)) shown in.

[0112] In an embodiment, the most central pattern 469 from a given group (CP(1), CP(2), ..., CP(M)) is the pattern that is closest to the centroid of a specified feature space for the multiple individual groupings relative to other patterns in the multiple individual groupings. For example, the specified feature space can be a target mask image feature space, a frequency mapping feature space, a pattern density mapping feature space, a pattern occurrence feature space, an SGM feature space, a diffraction order feature space, a diffraction pattern feature space, and / or other feature space. In an embodiment, the method and apparatus can be configured such that by default the target mask image feature space is used to select the center sample. However, because the enhanced pattern set (e.g., Figure 4 Each sample in 445) includes multi-dimensional features (e.g., frequency mapping, pattern density mapping, etc.), and the method and device can be configured so that the user can specify the feature space and / or select the center sample based on the target feature.

[0113] The present method and apparatus are configured such that: assuming that there are a total number (N) of multiple individual groups 447, and in response to N being greater than a target number M 467 of representative patterns 448, M patterns are selected from the N groups as representative example patterns set to be used as training patterns 449. The M selected patterns are configured to maximize the variation of the training patterns 449 in a given (e.g., user-selected) feature space (e.g., the default target mask image feature space).

[0114] Fig. 9 The diagram illustrates a method for forming a representative set of example patterns for use as training patterns (eg, Figure 4 449) from n groups 900 (e.g., Fig. 9 The "n" in is similar to the pattern selection of "N" described above. Group 900 is formed by images 901 (eg, clips) C1 902, C2 904, ..., Cn 906. Fig. 9 As shown, for a pattern set including n samples (e.g., n images (or clips) of different patterns), with a sample image size of H pixels by W pixels (e.g., H×W), C j P i indicates the i-th pixel in the j-th image (clip), where j∈[1,n], i∈[1,H*W]. For pixel P i , P i The variation of along different images is given as follows:

[0115]

[0116] For each item in the above formula (P ij -P i ) 2, the change of the j-th pixel among n images of the same pixel position is determined, and the pixel changes of all H×W pixels are summed to represent a criterion for selecting M patterns from N groups as a representative example pattern set, wherein the selection criterion is configured to maximize the change in the training patterns.

[0117] Back to Figure 4 As described above, the method and apparatus are configured to select M samples (M<N) from N candidates to maximize the variation of the training pattern 449 (e.g., in the target mask image feature space). In an embodiment, determining the training pattern 449 may include performing a global search method across all N candidates so that the training pattern 449 has the maximum amount of variation. The global search method includes traversing all permutations of multiple candidate patterns and multiple groups, because this is a gradient-free search. The total number of permutations is:

[0118]

[0119] Typically, both M and N are typically greater than hundreds or even thousands. Thus, the total number of permutations is an extremely large number. This makes it difficult to traverse all permutations within a limited and feasible amount of time.

[0120] In an embodiment, different methods may be used to select M samples from N candidates. In an embodiment, a target number 467 of representative patterns 448 is determined based on a stopping criterion and / or other information. The stopping criterion is configured to promote changes in the training pattern 449. The stopping criterion may be determined based on information from previous training patterns, entered and / or selected by a user, determined at the manufacturing device, and / or determined in other ways. In an embodiment, method 400 includes determining an amount of change in the training pattern 449. In an embodiment, the stopping criterion is configured to ensure that the amount of change in the training pattern 449 exceeds an amount of change threshold. The amount of change threshold may be determined based on information from previous training patterns, entered and / or selected by a user, determined at the manufacturing device, and / or determined in other ways.

[0121] In an embodiment, a target number 467 of representative patterns 448 are randomly selected from a total number (N) of individual groups 447 (e.g., patterns are randomly selected from M groups). In an embodiment, the target number of representative patterns are randomly reselected in response to the amount of change in the training pattern not breaking through a threshold amount of change, in response to an increase in the amount of change in a subsequently selected training pattern 449 relative to an immediately previous iteration in the training pattern 449, and / or for other reasons. For example, the amount of change in the training pattern 449 may be determined after an iteration in which patterns are randomly selected from M groups from a total number N of groups. If the change in the current iteration is greater than the previous change (e.g., because an increased change is required), the training pattern 449 may be updated (e.g., the randomly selected patterns constituting the training pattern 449 may be randomly reselected). The stopping criteria may be and / or include a maximum number of iterations (e.g., N_iter>max_iter), an amount of change that breaks through a threshold (e.g., change>=threshold), a break through of a maximum iteration time, and / or other stopping criteria. This search method is time-efficient and can be used to exploit (eg, good enough) variation to drive the training set.

[0122] In an embodiment, method 400 further includes: Figure 4 (not shown) is provided to a deep convolutional neural network and / or other machine learning model to train the deep convolutional neural network. In an embodiment, method 400 also includes using the trained deep convolutional neural network to perform ( Figure 4 ) optical proximity effect correction as part of the wafer patterning process. For example, training patterns 449 (e.g., generated based on a full-chip GDS as described herein) can be provided to train a deep convolutional neural network and / or other machine learning model to predict a CTM mapping. For example, training patterns 449 (e.g., continuous transmission mask (CTM) images) are a set of representative patterns that are provided to a machine learning model (e.g., DCNN) for training so that an accurate CTM mapping will be predicted by a machine learning model (e.g., DCNN) for a full-chip OPC application. This more accurate CTM mapping results in fewer lithography hotspots and fewer process window limitations, and / or reduces difficulties during subsequent mask correction operations when attempting to meet lithography performance specifications (e.g., related to aerial images, resist images, etc.). Other applications of method 400 are contemplated.

[0123] Fig.101 is a block diagram illustrating a computer system 100 that may 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 may 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.

[0124] The computer system 100 can be coupled to a display 112, such as a cathode ray tube (CRT) or a flat panel display or a touch panel display, via the bus 102 for displaying information to a computer user. An input device 114 including alphanumeric keys and other keys is coupled to the bus 102 for communicating information and command selections to the 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 the processor 104 and for controlling cursor movement on the display 112. Such an input device typically has two degrees of freedom on 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.

[0125] According to one embodiment, according to one embodiment, portions of one or more methods described herein 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. These 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, hard-wired 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.

[0126] As used herein, the term "computer-readable medium" 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 disks 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, floppy 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, carriers as described below, or any other media that can be read by a computer.

[0127] Various forms of computer readable media are involved in 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 over a telephone line using a modem. A modem local to computer system 100 may receive the 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 the instructions. The instructions received by main memory 106 may optionally be stored on storage device 110 before or after execution by processor 104.

[0128] The computer system 100 may also include a communication interface 118 coupled to the bus 102. The communication interface 118 provides a bidirectional data communication coupling to a network link 120 connected to a local area network 122. For example, the communication interface 118 may 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 may be a local area network (LAN) card to provide a data communication connection to a compatible LAN. A wireless link may 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.

[0129] The network link 120 typically provides data communication to other data devices through one or more networks. For example, the network link 120 may provide a connection to a host computer 124 or to data equipment operated by an Internet Service Provider (ISP) 126 through a local area network 122. The ISP 126, in turn, provides data communication services through 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 example forms of carrier waves that transport the information.

[0130] The computer system 100 can send messages and receive data including process code via the network, network link 120, and communication interface 118. In the Internet example, server 130 can transmit the requested process code for an application via Internet 128, ISP 126, local area network 122, and communication interface 118. For example, one such downloaded application can provide all or part of the methods described herein. The received process code can be executed by processor 104 when received, and / or stored in storage device 110 or other non-volatile storage for later execution. In this way, computer system 100 can obtain application code in the form of a carrier wave.

[0131] Fig.11 An exemplary lithographic projection apparatus that can be utilized in conjunction with the techniques described herein is schematically depicted. The apparatus comprises:

[0132] - 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;

[0133] a first object table (e.g. patterning device table) MT having a patterning device holder for holding a patterning device MA (e.g. mask table) and connected to a first positioner for accurately positioning the patterning device relative to the item PS;

[0134] a second object table (substrate table) WT having 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 item PS; and

[0135] - a projection system ("lens") PS (eg a refractive, reflective or catadioptric optical system) that can image the irradiated portion of the patterning device MA onto a target portion C of the substrate W (eg comprising one or more dies).

[0136] As described herein, the apparatus may be of the transmissive type (e.g., employing a transmissive patterning device). However, in general, it may be of the reflective type (e.g., employing a reflective patterning device). The apparatus may employ a different kind of patterning device than a classical mask; examples include a programmable mirror array or an LCD matrix.

[0137] A source SO (e.g. a mercury lamp or an excimer laser, 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 comprise 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 comprise 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.

[0138] about Fig.10 It should be noted that the source SO can be within the housing of the lithographic projection apparatus (this is often the case when the source SO is, for example, a mercury lamp), but it can also be remote from the lithographic projection apparatus, into which the radiation beam it generates is directed (for example, with the aid of appropriate directional mirrors); the latter situation is often the case when the source SO is an excimer laser (for example, based on KrF, ArF or F2 laser action).

[0139] 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 PB can pass 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 PW2 (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 can be accurately positioned with the aid of a second positioning device PW2 (and the interferometry device IF). Fig.11 Movement of the object tables 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.

[0140] The depicted tools can be used in two different modes:

[0141] - in step mode, the patterning device table MT is kept substantially stationary and the entire patterning device image is projected at once (i.e. a single "flash") onto a target portion C. The substrate table WT is then shifted in the x and / or y direction so that a different target portion C can be illuminated by the beam PB;

[0142] - In scan mode, essentially the same applies, except that a given target portion C is not exposed in a single "flash". Alternatively, the patterning device table MT can be moved at a velocity v in a given direction (the so-called "scanning direction", e.g. the y-direction) so that the projection beam B scans across the patterning device image; while the substrate table WT is simultaneously moved in the same or opposite direction at a velocity 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.

[0143] Fig.12 Another exemplary lithographic projection apparatus 1000 that may be utilized in conjunction with the techniques described herein is schematically depicted.

[0144] The lithography projection device 1000 comprises:

[0145] - Source collector module SO;

[0146] - an illumination system (illuminator) IL configured to condition a radiation beam B (e.g. EUV radiation);

[0147] a support structure (e.g. patterning device table) MT configured to support a patterning device (e.g. mask or reticle) MA and connected to a first positioning device PM configured to accurately position the patterning device;

[0148] 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 positioning device PW configured to accurately position the substrate; and

[0149] - 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).

[0150] like Fig.12As depicted in FIG. 1 , the apparatus 1000 is of the reflective type (e.g., using a reflective patterning device). It should be noted that because most materials are absorptive in the EUV wavelength range, the patterning device can have a multilayer reflector including, for example, multiple stacks of molybdenum and silicon. In one example, the multilayer 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 sheet 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 where features will not be printed (negative resist).

[0151] The illuminator IL receives an extreme ultraviolet (EUV) radiation beam from a source collector module SO. Methods for generating EUV radiation include, but are not necessarily limited to, utilizing one or more emission lines of the EUV range to convert a material into a plasma state having at least one element, such as xenon, lithium, or tin. In one such method, often referred to as laser produced plasma ("LPP"), a plasma may be generated by irradiating a fuel (such as a droplet, stream, or cluster of material having a line emitting element) with a laser beam. The source collector module SO may be a laser ( Fig.12 The plasma is a part of an EUV radiation system (not shown) for providing 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.

[0152] In these 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, such as when the radiation source is a discharge produced plasma EUV generator, often referred to as a DPP source, the source may be an integral part of the source collector module. In embodiments, a DUV laser source may be used.

[0153] 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. Furthermore, the illuminator IL may include various other components, 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.

[0154] The radiation beam B is incident on a patterning device (e.g. a mask) MA held on a support structure (e.g. a patterning device 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 positioning device 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 in order to position different target portions C in the path of the radiation beam B. Similarly, a first positioning device 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.

[0155] The depicted device 1000 may be used in at least one of the following modes:

[0156] In step mode, the support structure (e.g. patterning device table) MT and the substrate table WT are kept substantially stationary while an entire pattern imparted to the radiation beam is projected at once (i.e. a single static exposure). Subsequently, the substrate table WT is shifted in the X and / or Y direction so that a different target portion C can be exposed.

[0157] In scan mode, the support structure (e.g. patterning device table) MT and the substrate table WT are scanned synchronously while a pattern imparted to the radiation beam is projected onto a target portion C (i.e. a single dynamic exposure). The speed and direction of the substrate table WT relative to the support structure (e.g. patterning device table) MT may be determined by the (de-)magnification and image reversal characteristics of the projection system PS.

[0158] In another mode, the support structure (e.g. patterning device 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 can be readily applied to maskless lithography utilizing a programmable patterning device, such as a programmable mirror array of the type mentioned above.

[0159] Fig.13The apparatus 1000 is shown in more detail, and includes 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. A plasma source can be formed by a discharge to generate an EUV radiation emitting plasma 210. 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. The very hot plasma 210 is generated by, for example, a discharge that causes at least a partially ionized plasma. For efficient generation of radiation, a partial pressure of, for example, 10Pa 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.

[0160] The radiation emitted by the hot plasma 210 is transferred from the source chamber 211 to the collector chamber 212 through 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 known in the art, the contaminant trap or contaminant barrier 230 further indicated herein includes at least a channel structure.

[0161] The collector chamber 212 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 in 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.

[0162] 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.

[0163] Typically more elements than shown may 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. Additionally, more mirrors than shown may be present, for example more than 200 mirrors may be present in the projection system PS. Fig.13 The reflective elements shown have from 1 to 6 additional reflective elements.

[0164] like Fig.14 The illustrated collector optics CO 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 can be used in combination with a discharge produced plasma source, often referred to as a DPP source.

[0165] Alternatively, the source collector module SO may be as follows Fig.14 . The laser LA 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 eV. 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.

[0166] The embodiments may be further described using the following aspects:

[0167] 1. A method for training a machine learning model for a layout patterning process, the method comprising:

[0168] generating a plurality of features from patterns in a pattern collection;

[0169] grouping the patterns in the set of patterns into a plurality of separate groups based on similarities of a plurality of generated features; and

[0170] Representative patterns from the plurality of individual groups are provided to the machine learning model to train the machine learning model to predict a continuous transmission mask (CTM) mapping for optical proximity correction (OPC) for the layout patterning process.

[0171] 2. The method according to clause 1, wherein the plurality of features generated from the pattern in the pattern set is information in addition to geometric information and / or vertex information already included in the pattern set.

[0172] 3. The method according to clause 1 or 2, wherein the OPC comprises full-chip OPC for a wafer during the layout patterning process.

[0173] 4. The method of any one of clauses 1 to 3, wherein the plurality of generated features comprises geometric features and lithography-aware features.

[0174] 5. A method according to any one of aspects 1 to 4, wherein grouping the patterns in the pattern set into multiple separate groups based on the similarity of multiple generated features includes using a machine learning clustering method to cluster unique patterns in the pattern set into multiple separate groups based on the similarity of multiple generated features.

[0175] 6. A method for determining a training pattern for a layout patterning process, the method comprising:

[0176] generating a plurality of features from patterns in a pattern collection;

[0177] grouping the patterns in the set of patterns into a plurality of separate groups based on similarities of a plurality of generated features; and

[0178] Representative patterns are selected from the plurality of individual groups to determine the training patterns.

[0179] 7. The method of clause 6, wherein the plurality of generated features include geometric features and lithography-aware features.

[0180] 8. The method of clause 7, wherein the geometric feature comprises one or more of: a target mask image, a frequency map, a pattern density map, or a pattern occurrence rate of a unique pattern in the pattern set.

[0181] 9. A method according to any one of aspects 7 or 8, wherein the lithography-aware features include one or more of: a sub-resolution assist feature guide map (SGM), a diffraction order, or a diffraction pattern of a unique pattern in the pattern set.

[0182] 10. A method according to any one of clauses 6 to 9, wherein the plurality of features generated from the pattern in the pattern set is in addition to geometric information and / or vertex information already included in the pattern set.

[0183] 11. The method according to any one of aspects 6 to 10, wherein grouping the patterns in the set of patterns into a plurality of groups based on a plurality of generated features is performed using unsupervised machine learning.

[0184] 12. A method according to any one of aspects 6 to 11, wherein grouping the patterns in the pattern set into multiple separate groups based on the similarity of multiple generated features includes clustering the patterns in the pattern set into multiple separate groups based on the similarity of multiple generated features.

[0185] 13. A method according to aspect 12, wherein the clustering includes a continuous series of clustering steps performed using different features from a plurality of generated features for different clustering steps, the continuous series of clustering steps forming a subset of unique patterns in the pattern set so that the representative pattern is selected from the subset to determine the training pattern.

[0186] 14. A method according to aspect 12 or 13, wherein the clustering includes a machine learning clustering method.

[0187] 15. A method according to clause 13 or 14, wherein the consecutive series of clustering steps comprises a cross-validation step performed using given features for a given step, the cross-validation comprising adjusting which patterns are included in a given subset.

[0188] 16. The method of any one of aspects 6 to 15, wherein selecting representative patterns from the separate set to determine the training patterns comprises selecting a target number of representative patterns.

[0189] 17. A method according to clause 16, wherein the target number of representative patterns is determined based on a stopping criterion, the stopping criterion being configured to induce a variation of the training pattern.

[0190] 18. The method of clause 17, further comprising: determining an amount of change in the training pattern.

[0191] 19. The method of clause 18, wherein the stopping criterion is further configured to ensure that the variation of the training pattern breaks through a variation threshold.

[0192] 20. The method of clause 19, wherein the target number of representative patterns are randomly selected from the plurality of separate groups.

[0193] 21. The method of clause 20, wherein the target number of representative patterns is randomly reselected in response to the variation of the training pattern not exceeding the variation threshold.

[0194] 22. A method according to any one of aspects 6 to 21, wherein selecting representative patterns from the multiple individual groups to determine the training patterns includes selecting a most central pattern from each individual group, wherein the most central pattern is closest to the centroid of the specified feature space for the individual group relative to other patterns in the multiple individual groups.

[0195] 23. The method of clause 22, wherein the specified feature space is a target mask image feature space, a frequency map feature space, a pattern density map feature space, a pattern occurrence feature space, an SGM feature space, a diffraction order feature space, or a diffraction pattern feature space.

[0196] 24. The method according to any one of aspects 6 to 23 further comprises: providing the training pattern to a deep convolutional neural network to train the deep convolutional neural network.

[0197] 25. The method of clause 24, further comprising: performing optical proximity effect correction as part of a wafer patterning process using a trained deep convolutional neural network.

[0198] 26. A computer program product comprising a non-transitory computer readable medium having recorded thereon instructions, the instructions when executed by a computer implementing the method according to any one of aspects 1 to 25.

[0199] The concepts disclosed herein can simulate or mathematically model any general imaging system for imaging sub-wavelength features, and can be used in particular with emerging imaging technologies that can produce shorter and shorter wavelengths. Emerging technologies already in use include extreme ultraviolet (EUV), DUV lithography that can produce 193nm wavelengths by using ArF lasers and even 157nm wavelengths by using fluorine lasers. In addition, EUV lithography can produce wavelengths in the range of 5nm to 20nm by using synchrotrons or by using high energy electrons to bombard materials (solid or plasma) to produce photons in this range.

[0200] 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 applicable with any type of lithography imaging system, for example, a lithography imaging system for imaging on substrates other than silicon wafers.

[0201] The above description is intended to be illustrative rather than limiting. Accordingly, those skilled in the art will appreciate that modifications may be made to the described invention without departing from the scope of the claims set forth below.

Claims

1. A method for determining a training pattern for a layout patterning process, the method comprising: generating a plurality of features from patterns in a pattern collection; grouping the patterns in the set of patterns into a plurality of separate groups based on the generated similarities of the plurality of features; as well as selecting representative patterns from the plurality of separate groups to determine the training patterns, Wherein the plurality of features include geometric features and lithographically-sensitive features, and wherein the lithographically-sensitive features include information related to lithographic transfer of a pattern to a substrate. 2 . The method of claim 1 , wherein the geometric features include one or more of: a target mask image, a frequency map, a pattern density map, or a pattern occurrence rate of a unique pattern in the pattern set. 3 . The method of claim 1 , wherein the lithographically-sensitive features comprise one or more of: a sub-resolution assist feature guide map, a diffraction order, or a diffraction pattern of a unique pattern in the set of patterns. 4 . The method of claim 1 , wherein the plurality of features generated from the patterns in the pattern set are in addition to geometric information and / or vertex information already included in the pattern set. 5 . The method of claim 1 , wherein grouping the patterns in the set of patterns into a plurality of groups based on the plurality of features is performed using unsupervised machine learning. 6 . The method of claim 1 , wherein grouping the patterns in the set of patterns into a plurality of separate groups based on similarities of the plurality of features comprises clustering the patterns in the set of patterns into a plurality of separate groups based on the similarities of the plurality of features.

7. A method according to claim 6, wherein the clustering includes a continuous series of clustering steps performed using different features of the multiple features for different clustering steps, and the continuous series of clustering steps form a subset of unique patterns in the pattern set so that the representative pattern is selected from the subset to determine the training pattern.

8. A method according to claim 7, wherein the clustering comprises a machine learning clustering method, and / or wherein the consecutive series of clustering steps comprises a cross-validation step performed using given features for a given step, the cross-validation comprising adjusting which patterns are included in a given subgroup.

9. The method of claim 1 , wherein selecting representative patterns from the separate group to determine the training patterns comprises selecting a target number of representative patterns, and / or wherein the target number of representative patterns is determined based on a stopping criterion configured to promote a change in the training patterns.

10. The method according to claim 9, further comprising: determining an amount of change of the training pattern, and / or wherein the stopping criterion is further configured to ensure that the amount of change of the training pattern exceeds an amount of change threshold, and / or wherein the target number of representative patterns are randomly selected from the plurality of separate groups, and / or wherein the target number of representative patterns are re-randomly selected in response to the amount of change of the training pattern not exceeding the amount of change threshold.

11. The method of claim 1 , wherein selecting representative patterns from the plurality of separate groups to determine the training patterns comprises selecting a center-most pattern from each separate group that is closest to a centroid of a specified feature space for the separate group relative to other patterns in the plurality of separate groups.

12. The method of claim 11, wherein the specified feature space is a target mask image feature space, a frequency mapping feature space, a pattern density mapping feature space, a pattern appearance feature space, a feature space for one or more sub-resolution auxiliary feature guide mappings, a diffraction order feature space, or a diffraction pattern feature space.

13. The method according to claim 1, further comprising: The training pattern is provided to a deep convolutional neural network to train the deep convolutional neural network.

14. A computer program product comprising a non-transitory computer-readable medium having recorded thereon instructions, the instructions implementing the method according to claim 1 when executed by a computer.

Images