System and method for determining target feature focus in image-based overlay measurements

Abstract

To realize an overlay measurement based on a single image capture feature on a plurality of layers that requires a reference measurement on the basis of an external tool or a full-wafer measurement in order to provide a desired measurement accuracy.SOLUTION: A measurement system contains a through focus imaging measurement sub-system that is communicably coupled to a controller having a processor constructed so as to receive a plurality of training images captured at one or more focal point position. The processor generates a mechanical learning classification device on the basis of a plurality of training images, receives a target feature selection for a target overlay measurement corresponded to a target feature, determines the target focal point position on the basis of the target feature selection by using the mechanical learning classification device, receives a target image captured at the target focal point position, and determines an overlay.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure generally relates to overlay measurement, and more particularly to machine learning for target feature focus.

Background Art

[0002] Image-based overlay measurement can typically include determining the relative offset between two or more layers on a sample based on the relative imaged positions of overlay target features in different layers of interest. The accuracy of overlay measurement can thus be sensitive to the image quality associated with the imaged features on each sample layer, which can vary based on factors such as the depth of field or position (e.g., focus position) of the plane relative to the sample. Thus, overlay measurement procedures typically involve a trade-off between image quality and throughput in a particular sample layer. For example, overlay measurements based on separate images of each sample layer may provide the highest quality images of overlay target features. However, capturing multiple images per target can reduce throughput. As another example, overlay measurements based on single image capture features on multiple layers may provide relatively high throughput, but may require external tools or reference measurements based on full wafer measurements to provide the desired measurement accuracy.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0004] Therefore, it is desirable to provide a system and method for hardening the defects described above. [Means for solving the problem]

[0005] Measurement systems according to one or more embodiments of the present disclosure are disclosed. In one embodiment, the measurement system includes a controller communicatively coupled to one or more through-focus imaging measurement subsystems, the controller includes one or more processors configured to execute a set of program instructions stored in memory, the set of program instructions configured to cause one or more processors to: receive a plurality of training images captured at one or more focal positions, each containing one or more training features of a training sample; generate a machine learning classifier based on the plurality of training images captured at one or more focal positions; receive one or more target feature selections for one or more target overlay measurements corresponding to one or more target features of a target sample; determine one or more target focal positions based on one or more target feature selections using the machine learning classifier; receive one or more target images captured at one or more target focal positions, each containing one or more target features of a target sample; and determine one or more overlay measurements based on the one or more target images.

[0006] Measurement systems according to one or more embodiments of the present disclosure are disclosed. In one embodiment, the measurement system includes one or more through-focus imaging measurement subsystems. In another embodiment, the measurement system includes a controller communicatively coupled to one or more measurement subsystems, the controller includes one or more processors configured to execute a set of program instructions stored in memory, the set of program instructions configured to cause one or more processors to: receive a plurality of training images captured at one or more focal positions, each containing one or more training features of a training sample; generate a machine learning classifier based on the plurality of training images captured at one or more focal positions; receive one or more target feature selections for one or more target overlay measurements corresponding to one or more target features of a target sample; determine one or more target focal positions based on one or more target feature selections using the machine learning classifier; receive one or more target images captured at one or more target focal positions, each containing one or more target features of a target sample; and determine one or more overlay measurements based on the one or more target images.

[0007] Disclosed are methods for measuring overlays using one or more through-focus imaging measurement subsystems, according to one or more embodiments of the present disclosure. In one embodiment, the method includes receiving a plurality of training images captured at one or more focal positions, the plurality of training images containing one or more training features of a training sample. In another embodiment, the method includes generating a machine learning classifier based on the plurality of training images captured at one or more focal positions. In yet another embodiment, the method includes receiving one or more target feature selections for one or more target overlay measurements corresponding to one or more target features of a target sample. In yet another embodiment, the method includes determining one or more target focal positions based on one or more target feature selections using the machine learning classifier. In yet another embodiment, the method includes receiving one or more target images captured at one or more target focal positions, the one or more target images containing one or more target features of a target sample. In yet another embodiment, the method includes determining one or more overlay measurements based on one or more target images.

[0008] It should be understood that both the above summary and the following detailed description are illustrative and descriptive only and do not necessarily limit the claimed invention. The accompanying drawings incorporated into and constituting part of the specification illustrate embodiments of the invention and, together with the general description, help to illustrate the principles of the invention. [Brief explanation of the drawing]

[0009] Many of the advantages of this disclosure can be better understood by those skilled in the art by referring to the accompanying drawings. [Figure 1] This is a conceptual diagram showing a measurement system according to one or more embodiments of the present disclosure. [Figure 2] This is a simplified schematic diagram illustrating a measurement system according to one or more embodiments of the present disclosure. [Figure 3]This flowchart illustrates the steps performed in a method for measuring an overlay according to one or more embodiments of the present disclosure. [Figure 4] This flowchart illustrates the steps performed in a method for measuring an overlay according to one or more embodiments of the present disclosure. [Modes for carrying out the invention]

[0010] Hereinafter, we refer in detail to the disclosed subject matter shown in the accompanying drawings. This disclosure has been specifically shown and described with respect to particular embodiments and their particular features. The embodiments described herein are to be construed as illustrative rather than restrictive. It should be readily apparent to those skilled in the art that various changes and modifications in form and detail can be made without departing from the spirit and scope of this disclosure. Hereinafter, we refer in detail to the disclosed subject matter shown in the accompanying drawings.

[0011] Embodiments of this disclosure relate to through-focus imaging systems and methods for overlay targets on a sample to provide self-reference overlay measurement recipes for additional overlay targets on a sample, as well as inter-wafer process monitoring.

[0012] Semiconductor devices are typically formed as multiple patterned layers of patterned material on a substrate. Each patterned layer may be fabricated through a series of process steps, including, but not limited to, one or more material deposition steps, one or more lithography steps, or one or more etching steps. Furthermore, the features within each patterned layer must typically be fabricated within selected tolerances to properly construct the final device. For example, overlay errors related to the relative misalignment of features on different sample layers must be well characterized and controlled within each layer and with respect to previously fabricated layers.

[0013] Therefore, overlay targets can be fabricated on one or more sample layers to enable efficient characterization of interlayer feature overlays. For example, an overlay target may include features fabricated on multiple layers arranged to facilitate accurate overlay measurements. In this regard, overlay measurements on one or more overlay targets distributed across a sample can be used to determine the overlays of corresponding device features associated with the semiconductor device being fabricated.

[0014] Image-based overlay measurement tools typically capture one or more images of an overlay target and determine the overlay between sample layers based on the relative positions of the captured features of the overlay target on the layer of interest. For example, the features of an overlay target suitable for image-based overlays located on different sample layers (e.g., box-in-box targets, advanced imaging measurement (AIM) targets, etc.) may, but do not necessarily, be positioned so that features on all layers of interest are visible simultaneously. In this regard, the overlay can be determined based on the relative positions of features on the layer of interest within one or more images of the overlay target. Furthermore, the overlay target may be designed to facilitate overlay measurement between any number of sample layers in either a single measurement step or multiple measurement steps. For example, features in any number of sample layers may be visible simultaneously for single-measure overlay determination between all sample layers. In another example, the overlay target may have different sections (e.g., cells) to facilitate overlay measurement between selected layers. In this regard, the overlay between all layers of interest may be determined based on measurements of multiple parts of the overlay target.

[0015] The accuracy of an image-based overlay can depend on several factors related to image quality, including but not limited to resolution or aberrations. For example, system resolution can affect the accuracy with which feature locations can be determined (e.g., edge locations, centers of symmetry, etc.). As another example, aberrations in the imaging system can distort the size, shape, and spacing of features, so that image-based position measurements may not accurately represent the physical sample. Furthermore, image quality can vary as a function of focal position. For example, features outside the focal volume of the imaging system may appear blurred and / or have lower contrast between the overlay target feature and background space than features within the focal volume, which may affect the accuracy of position measurements (e.g., edge measurements, etc.).

[0016] Therefore, capturing separate images of features on different layers of a sample (e.g., located at different depths within the sample) may provide accurate overlay measurement. For example, the focal position of an image-based overlay measurement system (e.g., the object plane) may be adjusted to correspond to the depth of the captured feature on each layer of interest. In this regard, features on each layer of interest may be captured under conditions designed to mitigate focal position-dependent effects.

[0017] However, it is recognized herein that capturing multiple images of an overlay target at different depths can negatively impact the throughput of the system, which can offset the accuracy gain associated with the multiple images. Embodiments of the present disclosure are directed to overlay measurements at multiple focus positions, where the multiple focus positions are determined in real time by a measurement system and can correspond to one or more changing depths at which one or more portions of one or more measurement targets are located. For example, overlay measurements can be generated for an overlay target based on multiple images captured at multiple focus positions (e.g., including the depth of focus corresponding to the position of an overlay target feature), and the measurement system determines the multiple focus positions in real time (e.g., by use of a machine learning classifier).

[0018] Additional embodiments of the present disclosure relate to translating one or more portions of a measurement system along one or more adjustment axes. For example, the optimal coordinate position at which a particular overlay measurement is taken may be different from the optimal coordinate position for subsequent overlay measurements.

[0019] Further embodiments of the present disclosure relate to generating a control signal based on overlay measurements across samples provided as feedback and / or feedforward data to processing tools (e.g., lithography tools, metrology tools, etc.).

[0020] FIG. 1 is a conceptual diagram showing an overlay measurement system 100 according to one or more embodiments of the present disclosure. The system 100 can include, without limitation, one or more measurement subsystems 102. The system 100 can further include, without limitation, a controller 104, which includes one or more processors 106, a memory 108, and a user interface 110.

[0021] One or more measurement subsystems 102 can include any measurement subsystems known in the art, including but not limited to an optical measurement subsystem. For example, measurement subsystem 102 may include, but is not limited to, an optically based measurement system, a broadband measurement system (e.g., a broadband plasma measurement system), or a narrowband inspection system (e.g., a laser-based measurement system). In another example, measurement subsystem 102 can include a scatterometry-based measurement system. As another example, one or more measurement subsystems 102 can include any through-focus imaging measurement subsystem (e.g., an imaging measurement subsystem configured to construct one or more images of a sample, the one or more images being of a desired focus and constructed using a plurality of images of the sample captured at different focus positions).

[0022] In one embodiment, controller 104 is communicatively coupled to one or more measurement subsystems 102. In this regard, one or more processors 106 of controller 104 can be configured to generate and provide one or more control signals configured to perform one or more adjustments to one or more portions of one or more measurement subsystems 102.

[0023] In another embodiment, controller 104 is configured to receive a plurality of training images captured at one or more focus positions, the plurality of training images including one or more training features of a training sample. For example, controller 104 may be configured to receive the plurality of training images from one or more measurement subsystems 102.

[0024] In another embodiment, controller 104 can be configured to generate a machine learning classifier based on the plurality of training images. For example, controller 104 may be configured to use the plurality of training images as input to the machine learning classifier.

[0025] In another embodiment, the controller 104 is configured to receive one or more target feature selections for one or more target overlay measurements, where the one or more target feature selections correspond to one or more target features of a target sample. For example, the controller 104 may be configured to receive one or more target feature selections from a user via the user interface 110.

[0026] In another embodiment, the controller 104 may be configured to determine one or more target focus locations based on the selection of one or more target features. For example, the controller 104 may be configured to determine one or more target focus locations using a machine learning classifier.

[0027] In another embodiment, the controller 104 may be configured to receive one or more target images captured at one or more target focal positions. For example, the controller 104 may be configured to receive one or more target images containing one or more target features at one or more target focal positions.

[0028] In another embodiment, the controller 104 may be configured to determine one or more overlay measurements based on one or more target images. For example, the controller 104 is configured to determine the overlay between the first layer of the target sample and the second layer of the target sample based on one or more target features formed on the first and second layers, respectively.

[0029] Figure 2 shows a simplified schematic diagram of System 100 according to one or more embodiments of the present disclosure. In particular, System 100 shown in Figure 2 includes an optical measurement subsystem 102 such that System 100 operates as an optical inspection system.

[0030] The optical inspection subsystem 102 may include any optical-based inspection known in the art. The measurement subsystem 102 may include, but is not limited to, an illumination source 112, an illumination arm 111, an acquisition arm 113, and a detector assembly 126.

[0031] In one embodiment, the measurement subsystem 102 is configured to inspect and / or measure a sample 120 placed on a stage assembly 122. The illumination source 112 may include any illumination source known in the art for generating illumination 101, including but not limited to illumination sources configured to provide wavelengths of light including but not limited to vacuum ultraviolet (VUV), deep ultraviolet (DUV), ultraviolet (UV), visible, or infrared (IR) radiation. In another embodiment, the measurement subsystem 102 may include an illumination arm 111 configured to direct the illumination 101 towards the sample 120. It should be noted that the illumination source 112 of the measurement subsystem 102 may be configured in any orientation known in the art, including but not limited to dark-field orientation, bright-field orientation, etc. For example, one or more optical elements 114, 124 may be selectively tuned to configure the measurement subsystem 102 in dark-field orientation, bright-field orientation, etc.

[0032] Sample 120 may include any sample known in the art, including but not limited to wafers, reticles, and photomasks. Sample 120 may include any sample having one or more overlay measurement targets known in the art to be suitable for image-based overlay measurement. For example, Sample 120 may include an overlay measurement target having target features of one or more layers printed in one or more lithographically separate exposures. The target and / or target features may possess various symmetries, such as two- or four-fold rotational symmetry, reflectional symmetry, etc.

[0033] In one embodiment, the sample 120 is placed on a stage assembly 122, which is configured to facilitate movement of the sample 120 (e.g., movement along one or more of the x, y, or z directions). In another embodiment, the stage assembly 122 is a movable stage. For example, the stage assembly 122 may include, but is not limited to, one or more translation stages suitable for selectively translating the sample 120 along one or more linear directions (e.g., the x, y, and / or z directions). As another example, the stage assembly 122 may include, but is not limited to, one or more rotation stages suitable for selectively rotating the sample 120 along a rotational direction. As yet another example, the stage assembly 122 may include, but is not limited to, rotation and translation stages suitable for selectively translating the sample 120 along a linear direction and / or rotating the sample 120 along a rotational direction. It should be noted herein that the system 100 may operate in any measurement mode known in the art.

[0034] The illumination arm 111 may include any number and type of optical components known in the art. In one embodiment, the illumination arm 111 includes one or more optical elements 114, a set of one or more optical elements 115, a beam splitter 116, and an objective lens 118. In this regard, the illumination arm 111 may be configured to focus illumination 101 from the illumination source 112 onto the surface of the sample 120. The one or more optical elements 114 may include, but are not limited to, one or more mirrors, one or more lenses, one or more polarizers, one or more beam splitters, waveplates, etc., and may include any optical elements known in the art.

[0035] In another embodiment, the measurement subsystem 102 includes a collection arm 113 configured to collect illumination reflected or scattered from the sample 120. In another embodiment, the collection arm 113 may direct and / or focus the reflected and scattered light to one or more sensors of a detector assembly 126 via one or more optical elements 124. The one or more optical elements 124 include, but are not limited to, one or more mirrors, one or more lenses, one or more polarizers, one or more beam splitters, waveplates, etc., and may include any optical elements known in the art. It should be noted that the detector assembly 126 may include any sensor and detector assembly known in the art for detecting illumination reflected or scattered from the sample 120.

[0036] In another embodiment, the detector assembly 126 of the measurement subsystem 102 is configured to collect inspection data of sample 120 based on illumination reflected or scattered from sample 120. In yet another embodiment, the detector assembly 126 is configured to transmit the collected / acquired images and / or measurement data to the controller 104.

[0037] The measurement system 100 can be configured to image the sample 120 on any selected measurement plane (e.g., any position along the z-direction). For example, the position of the object plane associated with the image generated on the detector assembly 126 relative to the sample 120 can be adjusted using any combination of components of the measurement system 100. For example, the position of the object plane associated with the image generated on the detector assembly 126 relative to the sample 120 can be adjusted by controlling the position of the stage assembly 122 relative to the objective lens 118. As another example, the position of the object plane associated with the image generated on the detector assembly 126 relative to the sample 120 can be adjusted by controlling the position of the objective lens 118 relative to the sample 120. For example, the objective lens 118 may be mounted on a translational stage configured to adjust the position of the objective lens 118 along one or more adjustment axes (e.g., the x-direction, y-direction, or z-direction). As yet another example, the position of the object plane associated with the image generated on the detector assembly 126 relative to the sample 120 can be adjusted by controlling the position of the detector assembly 126. For example, the detector assembly 126 may be mounted on a translational stage configured to adjust the position of the detector assembly 126 along one or more adjustment axes. As another example, the position of the object plane related to the image generated on the detector assembly 126 relative to the sample 120 can be adjusted by controlling the position of one or more optical elements 124. For example, one or more optical elements 124 may be mounted on a translational stage configured to adjust the position of one or more optical elements 124 along one or more adjustment axes. It should be noted in this specification that the controller 104 may be configured to perform any of the above adjustments by providing one or more control signals to one or more parts of the measurement subsystem 102.

[0038] As previously stated herein, the controller 104 of system 100 may include one or more processors 106 and memory 108. Memory 108 may include program instructions configured to cause one or more processors 106 to perform various process steps described throughout this disclosure. For example, program instructions may be configured to cause one or more processors 106 to adjust one or more characteristics of the measurement subsystem 102 in order to perform one or more of the process steps of this disclosure. Furthermore, the controller 104 may be configured to receive data from the detector assembly 126, including but not limited to image data related to sample 120.

[0039] One or more processors 106 of the controller 104 may include any processor or processing element known in the art. For the purposes of this disclosure, the terms “processor” or “processing element” may be broadly defined to include any device having one or more processing or logic elements (e.g., one or more microprocessor devices, one or more application-specific integrated circuit (ASIC) devices, one or more field-programmable gate arrays (FPGAs), or one or more digital signal processors (DSPs)). In this sense, one or more processors 106 may include any device configured to execute algorithms and / or instructions (e.g., program instructions stored in memory). In one embodiment, one or more processors 106 may be embodied as a desktop computer, a mainframe computer system, a workstation, an image computer, a parallel processor, a networked computer, or any other computer system configured to execute a program that operates with or is configured to operate with the measurement system 100, as described throughout this disclosure.

[0040] Furthermore, different components of system 100 may include processors or logic elements suitable for performing at least some of the steps described herein. Therefore, the above description should not be construed as an limitation to embodiments of the disclosure, but rather as an example. Furthermore, the steps described throughout the disclosure may be performed by a single controller 104, or alternatively by multiple controllers. Furthermore, controller 104 may include one or more controllers housed in a common housing or multiple housings. In this way, any controller or combination of controllers can be separately packaged as modules suitable for integration into the measurement system 100. Furthermore, controller 104 can analyze data received from detector assembly 126 and supply the data to additional components within or outside the measurement system 100.

[0041] Memory 108 may include any storage medium known in the art that is suitable for storing program instructions executable by one or more associated processors 106. For example, memory 108 may include non-temporary storage media. Another example of memory 108 may include, but is not limited to, read-only memory (ROM), random access memory (RAM), magnetic or optical memory devices (e.g., disks), magnetic tape, solid-state drives, etc. Furthermore, it should be noted that memory 108 may be housed in a common controller housing together with one or more processors 106. In one embodiment, memory 108 may be located remotely from the physical locations of one or more processors 106 and controller 104. For example, one or more processors 106 of controller 104 may have access to remote memory (e.g., a server) accessible via a network (e.g., the internet, an intranet, etc.).

[0042] In one embodiment, the user interface 110 is communicatively coupled to the controller 104. The user interface 110 may include, but is not limited to, one or more desktops, laptops, tablets, etc. In another embodiment, the user interface 110 includes a display used to display data from the system 100 to the user. The display of the user interface 110 may include any display known in the art. For example, the display may include, but is not limited to, a liquid crystal display (LCD), an organic light-emitting diode (OLED) based display, or a CRT display. Those skilled in the art will recognize that any display device that can be integrated with the user interface 110 is suitable for implementation in this disclosure. In another embodiment, the user may input selections and / or commands in response to data displayed to the user via the user input device of the user interface 110.

[0043] In another embodiment, the controller 104 is communicatively coupled to one or more elements of the measurement system 100. In this regard, the controller 104 can transmit and / or receive data from any component of the measurement system 100. Furthermore, the controller 104 can direct or otherwise control any component of the measurement system 100 by generating one or more control signals for the relevant component. For example, the controller 104 can be communicatively coupled to the detector assembly 126 to receive one or more images from the detector assembly 126.

[0044] Figure 3 shows a method 300 for measuring an overlay according to one or more embodiments of the present disclosure.

[0045] In step 302, multiple training images captured at one or more focal positions are received. For example, multiple training images 125 may be received by the controller 104 from the measurement subsystem 102. In this regard, the multiple training images 125 may include optical training images. In additional and / or alternative embodiments, the controller 104 may be configured to receive one or more training images 125 from one or more sources other than the measurement subsystem 102. For example, the controller 104 may be configured to receive one or more training images 125 of features of sample 120 from an external storage device and / or memory 108. In another embodiment, the controller 104 may be further configured to store the received training images 125 in memory 108.

[0046] The multiple training images 125 may include one or more training features of a training sample. For example, the multiple training images 125 may include images captured at multiple depths of the training sample. In this regard, one or more training features of a training sample may include one or more training target features formed on different layers of the training sample. The measurement subsystem 102 may be configured to capture the multiple training images 125 at one or more focal positions corresponding to the depth (e.g., position along the z-direction) of a particular training feature. In one embodiment, the measurement subsystem 102 may be configured to capture the multiple training images 125 within a focal training range, where the focal training range includes multiple focal positions corresponding to multiple depths of one or more training features. The focal training range may be provided by the user via the user interface 110.

[0047] In step 304, a machine learning classifier is generated based on multiple training images. For example, the controller 104 may be configured to generate a machine learning classifier based on multiple training images. The controller 104 may be configured to generate a machine learning classifier via any one or more techniques known in the art, including, but not limited to, supervised learning and unsupervised learning.

[0048] For example, in the context of supervised learning, the multiple training images 125 may have varying degrees of focus based on the focal position from which each of the multiple training images was captured (for example, the multiple training images may include multiple through-focus images of one or more features of a sample). In this regard, the controller 104 may receive one or more optimal focus tolerances so that the controller 104 can determine one or more of the multiple training images that fall within one or more optimal focus tolerances. Thus, the multiple training images 125 and one or more optimal focus tolerances can be used as input for training a machine learning classifier. The controller 104 may be further configured to store the multiple training images 125, the optimal focus tolerances, and the generated machine learning classifier in memory 108.

[0049] One or more optimal focus tolerances may be configured such that a machine learning classifier determines one or more target focal positions for one or more target overlay measurements. In this regard, one or more optimal focus tolerances may be configured to ensure that one or more target images that can be subsequently captured at one or more target focal positions are of sufficient quality for overlay measurements by the controller 104. One or more optimal focus tolerances may be provided by the user via the user interface 110. In another embodiment, the controller 104 may be configured to determine one or more optimal focus tolerances using any technique known in the art. For example, the controller 104 may be configured to determine one or more optimal focus tolerances based on a contrast accuracy function (e.g., a function configured to determine the focus with the least noise based on multiple images captured at multiple focal positions). In an alternative embodiment, the controller 104 may be configured to determine one or more optimal focus tolerances using a Linnik interferometer integrated into or generated by one or more parts of the measurement subsystem 102. As another example, one or more parts of the measurement subsystem 102 (e.g., one or more parts of the illumination source 112 and / or illumination arm 111) may be configured to illuminate a sample, and the controller 104 may be configured to generate a Linnik interferometer (e.g., a low-coherence interferometer) based on the illumination collected by the detector assembly 126. In this regard, the controller 104 may also be configured to determine the interferometer peak (e.g., the point with the highest contrast in all collected images) and may associate the peak with the through-focus position along the z-axis of the sample.

[0050] It should be noted that embodiments of the present disclosure are not limited to a controller 104 that generates or references a Linnik interferometer to determine one or more optimal focus tolerances. For example, the controller 104 may be configured to generate a machine learning classifier configured to determine the best focus and / or best position based on a plurality of training images generated using one or more signals indicating illumination (e.g., illumination generated by a bright-field and / or dark-field microscope) emanating from various focal positions and / or positions (e.g., coordinate positions on the x and / or y axes, and the plurality of training images may be captured at various coordinate positions along one or both of such axes via a translatable stage, etc.). In at least the embodiments described above, the controller 104 may be configured to determine the best focus and / or best position based on the image contrast and / or contrast accuracy of the plurality of training images. In this regard, the machine learning classifier may be configured to determine one or more optimal focus tolerances based on a plurality of training images, the plurality of training images comprising focal slice images generated at various focal positions along the z-axis of the sample. The controller 104 may be configured to classify and / or label each focal slice image with its corresponding best focus based on the image contrast and / or contrast accuracy of the focal slice images. In this way, the controller 104 may be configured to determine one or more optimal focus tolerances by interpolation based on multiple training images.

[0051] In another embodiment, the machine learning classifier may be configured to determine one or more optimal focus tolerances based on a plurality of training images, the plurality of training images including images captured at various focal positions along the z-axis of a sample, and the focus of the plurality of training images is modified using a coarse focusing mechanism. Specifically, it should be noted that a coarse focusing mechanism (or any other common focusing mechanism known in the art that is suitable for the purposes contemplated by this disclosure) may enable the machine learning classifier to determine one or more optimal focus tolerances in a more precise manner (e.g., more precisely than other focusing methods described herein or known in the art). For example, as in step 308 (described below), the coarse focusing mechanism may be configured to enable the controller 104 to determine and / or set a coarse focus (e.g., the overall focal position can be fine-tuned later (see step 308 below)). It should be noted that the coarse focusing system includes, but is not limited to, a lens triangulation mechanism, a bicell detector, and / or any distance measuring system, any coarse focusing mechanism known in the art to be suitable for the purposes intended by this disclosure.

[0052] It should be noted herein that the machine learning classifier generated in step 304 may include, but is not limited to, any type of machine learning algorithm / classifier and / or deep learning technique or classifier known in the art, including, but not limited to, random forest classifiers, support vector machine (SVM) classifiers, ensemble learning classifiers, and artificial neural networks (ANNs). As another example, the machine learning classifier may include a deep convolutional neural network (CNN). For example, in some embodiments, the machine learning classifier may include ALEXNET and / or GOOGLENET. In this regard, the machine learning classifier may include any algorithm, classifier, or predictive model, but is not limited to, any algorithm, classifier, or predictive model configured to generate a Linnik interferometer and use the Linnik interferometer to determine one or more target focal positions for one or more target overlay measurements. In some embodiments, the machine learning classifier may comprise a neural network having multiple layers and receptors. For example, the machine learning classifier may comprise a neural network having about five layers and about fifteen receptors.

[0053] In step 306, one or more target feature selections for one or more target overlay measurements corresponding to one or more target features of the target sample are received. For example, the controller 104 may be configured to receive one or more target feature selections for one or more target overlay measurements from a user via the user interface 110. One or more target feature selections may include one or more signals configured to instruct the system 100 to capture one or more target images containing one or more target features of the target sample. Upon receiving one or more target feature selections, the controller 104 may be configured to determine one or more expected depths of one or more target features in the target sample. For example, the controller 104 may be provided by a user via the user interface 110 with one or more expected depths of one or more target features for overlay measurements. In other embodiments, the controller 104 may determine one or more expected depths of one or more target features for overlay measurements by referring to one or more design files or other data corresponding to the target sample stored in memory 108. In other embodiments, the controller 104 may determine one or more expected depths of one or more target features for overlay measurements based on a Linnik interferometer. For example, the controller 104 may determine one or more expected depths by referring to Linnik interferometer peaks associated with the through-focus position along the z-axis of the sample.

[0054] In step 308, one or more target focal positions based on one or more target feature selections are determined using a machine learning classifier. For example, the controller 104 may be configured to use a machine learning classifier to determine one or more target focal positions based on one or more target feature selections. As another example, the controller 104 may provide one or more expected depths of one or more target features of a target sample as input to the machine learning classifier. In this regard, the machine learning classifier may be configured to provide one or more target focal positions based on multiple training images 125 and an optimal focus tolerance. Note that the machine learning classifier may be configured to determine one or more target focal positions for one or more target overlay measurements by determining one or more focal positions within one micron of focal positions provided by a contrast accuracy function, a Linnink interferometer function, or any other method described herein.

[0055] In some embodiments, once one or more target focal positions are determined based on one or more target feature selections, the controller 104 may be configured to determine and / or provide one or more control signals to one or more parts of one or more measurement subsystems 102. The one or more control signals are configured to cause one or more parts of one or more measurement subsystems 102 to be translated along one or more adjustment axes (e.g., x, y, and / or z directions). For example, the controller 104 may provide one or more control signals to the stage assembly 122 and / or the detector assembly 126 so that the target sample is positioned at one or more of the determined target focal positions. In another embodiment, the controller 104 may be configured to provide one or more control signals to at least one of the optical elements 114, 115, the beam splitter 116, the objective lens 118, or the optical element 124 so that the measurement subsystem 102 can capture one or more target images at one or more target focal positions.

[0056] In the case of a machine learning classifier configured to determine one or more optimal focus tolerances based on multiple training images using a coarse focusing mechanism, the machine learning classifier may be configured to determine one or more control signals to the coarse focusing system (and the controller 104 may be configured to provide one or more control signals to the coarse focusing system). Here, one or more control signals may be configured to cause the coarse focusing system to adjust the focus to a focal position within ±2 micrometers of the target focal position. In this regard, the machine learning classifier may be configured to determine one or more optimal focus tolerances by determining one or more fine focus adjustments to correct the focal position determined by the coarse focusing system.

[0057] In step 310, one or more target images of one or more target features captured at one or more target focal positions are received. For example, the controller 104 may be configured to receive one or more target images 135 from the measurement subsystem 102. As used herein, the term “target image” may refer to an image of one or more target features captured at one or more target focal positions and on which one or more overlay measurements are determined. Thus, the term “target image” can be distinguished from “training image,” which may be considered an image of training features that can be used as input for training a machine learning classifier.

[0058] It should be noted that, in this specification, any discussion relating to the acquisition of training images 125 may be considered applicable to the acquisition of target images 135 unless otherwise specified herein. In additional and / or alternative embodiments, the controller 104 may be configured to receive one or more target images 135 from one or more sources other than the measurement subsystems 102. For example, the controller 104 may be configured to receive one or more target images 135 of sample 120 from an external storage device and / or memory 108.

[0059] In step 312, one or more overlay measurements are determined based on one or more target images. For example, the controller 104 is configured to determine the overlay between the first layer and the second layer of the target sample based on a first overlay measurement corresponding to one or more target features formed on a first layer of the target sample and a second overlay measurement corresponding to one or more target features formed on a second layer of the target sample. In this regard, the controller 104 may be configured to determine an offset (e.g., PPE) between the first and second layers. The one or more overlay measurements may include any overlay measurements known in the art that are suitable for the purposes contemplated by this disclosure, including overlay measurements configured to be used with specific target features of the sample. In this regard, the controller 104 may be configured to utilize one or more overlay algorithms stored in memory 108 or provided to the controller 104 to determine the one or more overlay measurements.

[0060] In some embodiments, method 300 may include step 314, in which one or more control signals are provided. For example, one or more control signals are provided for adjusting one or more process tools (e.g., lithography tools). As an additional example, controller 104 may provide one or more control signals (or corrections to control signals) to one or more parts of one or more process tools to adjust one or more parameters of one or more process tools (e.g., manufacturing settings, configuration, etc.) so that one or more parameters of one or more process tools are adjusted. Controller 104 may determine one or more control signals based on one or more overlay measurements of a sample. Control signals (or corrections to control signals) may be provided by controller 104 as part of a feedback and / or feedforward control loop. Controller 104 may cause one or more process tools to perform one or more adjustments to one or more parameters of the process tools based on the control signals, or controller 104 may warn the user to perform one or more adjustments to one or more parameters. In this sense, one or more control signals can compensate for errors in one or more manufacturing processes of one or more process tools, and thus enable one or more process tools to maintain an overlay within a selected tolerance range across multiple exposures on subsequent samples in the same or different lots.

[0061] Figure 4 shows a method 400 for measuring an overlay according to one or more embodiments of the present disclosure.

[0062] In step 402, one or more target feature selections for one or more target overlay measurements corresponding to one or more target features of the target sample are received. For example, the controller 104 may be configured to receive one or more target feature selections for one or more target overlay measurements from the user via the user interface 110. One or more target feature selections may include one or more signals configured to instruct the system 100 to capture one or more target images containing one or more target features of the target sample. Upon receiving one or more target feature selections, the controller 104 may be configured to determine one or more expected positions (e.g., positions along the x, y, and / or z directions) of one or more target features in the target sample. For example, the controller 104 may be provided by the user via the user interface 110 with one or more expected positions of one or more target features for overlay measurements. In other embodiments, the controller 104 may determine one or more expected depths of one or more target features for overlay measurements by referring to one or more design files or other data corresponding to the target sample stored in memory 108.

[0063] In step 404, one or more target focal positions are determined based on one or more target feature selections. For example, the controller 104 may be configured to determine one or more target focal positions using a machine learning classifier based on one or more target feature selections, where one or more target feature selections correspond to one or more target features for one or more overlay measurements. As another example, the controller 104 may provide one or more predicted depths of one or more target features of a target sample as input to the machine learning classifier. In this regard, the machine learning classifier may be configured to provide one or more target focal positions based on multiple training images 125 and an optimal focus tolerance.

[0064] In some embodiments, once one or more target focal positions are determined based on the selection of one or more target features, as shown in step 406, the controller 104 may be configured to determine one or more control signals and / or provide one or more control signals to one or more parts of one or more measurement subsystems 102. The one or more control signals are configured to translate one or more parts of one or more measurement subsystems 102 along one or more adjustment axes (e.g., x, y, and / or z directions). For example, in one embodiment, the controller 104 may be configured to provide one or more control signals to the stage assembly 122 and / or detector assembly so that one or more target features for overlay measurement are centered within the field of view of one or more components of the measurement subsystem 102. In this regard, the controller 104 may be configured to provide one or more control signals to one or more parts of the measurement subsystem 102 so that the target image 135 includes one or more target features at the center of the target image 135. In another embodiment, the controller 104 may be configured to supply one or more control signals to the stage assembly 122 so that the target sample is located at one or more of the determined target focal positions. In another embodiment, the controller 104 may be configured to provide one or more control signals to at least one of the optical elements 114, 115, beam splitter 116, objective lens 118, or optical element 124, in order to enable the measurement subsystem 102 to capture one or more target images at one or more target focal positions.

[0065] It should be noted that the controller 104 may be configured to simultaneously provide one or more control signals to one or more parts of one or more measurement subsystems 102. For example, the controller 104 may be configured to provide one or more control signals configured to cause simultaneous adjustments to the stage assembly 122 along the x and / or y directions and one or more other parts of the measurement subsystem 102 along the z direction (e.g., the detector assembly 126 and / or the objective lens 118).

[0066] Furthermore, it should be noted that the controller 104 may be configured to use a machine learning classifier to determine one or more control signals configured to cause transformations in one or more parts of one or more measurement subsystems 102. For example, the machine learning classifier may be configured to associate one or more locations (e.g., locations along the x and / or y axes) with one or more desired target features. In this regard, the machine learning classifier may be configured to determine locations (e.g., coordinate locations on the x and / or y axes) that automatically center one or more target features in the field of view.

[0067] In step 408, one or more target images of one or more target features captured at one or more target focal positions are received. For example, the controller 104 may be configured to receive one or more target images 135 from the measurement subsystem 102.

[0068] All methods described herein may include storing the results of one or more steps of an embodiment of the method in memory. The results may include any of the results described herein and may be stored in any form known in the art. The memory may include any memory described herein or any other suitable storage medium known in the art. After the results are stored, they may be accessed from memory, used by any of the method or system embodiments described herein, formatted for display to a user, used by another software module, method, or system, etc. Furthermore, the results may be stored “permanently,” “semi-permanently,” “temporarily,” or over a period of time. For example, the memory may be random-access memory (RAM), and the results do not necessarily persist in memory indefinitely.

[0069] Each of the embodiments of the above-described methods is further intended to include any other steps of any other methods described herein. In addition, each of the embodiments of the above-described methods can be carried out by any of the systems described herein.

[0070] Those skilled in the art will recognize that the constituent actions, devices, objects, and associated discussions described herein are used as examples for conceptual clarity, and that various configuration modifications are considered. Therefore, as used herein, the specific examples and associated discussions described are intended to be representative of their more general class. In general, the use of any particular example is intended to represent its class, and the exclusion of specific constituent actions, devices, and objects should not be interpreted as limitation.

[0071] When used herein, terms indicating direction, such as “up,” “down,” “up,” “down,” “up,” “upward,” “downward,” and “downward,” are intended to provide relative positions for illustrative purposes and not to indicate an absolute reference frame. Various modifications to the embodiments described will be apparent to those skilled in the art, and the general principles defined herein may be applicable to other embodiments.

[0072] With regard to the use of substantially any plural and / or singular terms herein, those skilled in the art can convert from plural to singular and / or singular to plural as appropriate to the context and / or use. Various singular / plural substitutions are not explicitly stated herein for the sake of clarity.

[0073] The subject matter described herein illustrates different components that, in some cases, are contained within or connected to other components. It should be understood that such depicted architectures are merely illustrative, and in practice, many other architectures can be implemented to achieve the same functionality. Conceptually, any arrangement of components to achieve the same function is effectively “associated” in such a way that the desired function is achieved. Thus, any two components in this specification combined to achieve a particular function, whether in architecture or as intermediate components, can be considered “associated” with each other in such a way that the desired function is achieved. Similarly, any two such associated components can also be considered “connected” or “joined” with each other in such a way that the desired functionality is achieved, and any two components that can be associated in such a way can also be considered “joinable” with each other in such a way that the desired functionality is achieved. Specific examples of joinable components include, but are not limited to, physically joinable and / or physically interacting components and / or wirelessly interactable and / or wirelessly interacting components and / or logically interacting and / or logically interactable components.

[0074] Furthermore, it should be understood that the present invention is defined by the appended claims. Generally, it will be understood by those skilled in the art that the terms used herein and in particular in the appended claims (e.g., the body of the appended claims) are generally intended to be “open” terms (e.g., the term “including” should be interpreted as “including but not limited to,” the term “having” should be interpreted as “at least having,” and the term “includes” should be interpreted as “includes but not limited to,” etc.). It will further be understood by those skilled in the art that if a specific number of claims to be introduced is intended, such intention will be explicitly stated in that claim, and if such statement is not made, such intention does not exist. For example, for the sake of understanding, the following appended claims may include introducing the claims using the introductory phrases “at least one” and “one or more.” However, the use of such phrases should not be interpreted as meaning that the introduction of a claim description with the indefinite article "a" or "an" limits any particular claim containing such introduced description to an invention containing only one such description. The same applies to the use of clear articles used to introduce a claim description, even if the same claim contains an introductory phrase such as "one or more" or "at least one" and an indefinite article such as "a" or "an" (for example, "a" and / or "an" should typically be interpreted as meaning "at least one" or "one or more"). Furthermore, even if a specific number of claims being introduced is explicitly listed, it will be recognized that such descriptions should typically be interpreted as meaning at least the number listed (for example, a bare list of "two lists" without other modifying factors typically means at least two lists, or two or more lists).Furthermore, in cases where a conventional expression similar to "at least one of A, B, and C" is used, such a configuration is generally intended to be understood by a person skilled in the art as including, but not limited to, a system having only A, only B, only C, both A and B, both A and C, both B and C, and / or a system having both A, B, and C. It will be further understood by those skilled in the art that virtually any separate words and / or phrases presenting two or more alternative terms, wherever they may appear in a description, claim, or drawing, should be understood as construing the possibility of including one of the terms, either of the terms, or both terms. For example, the phrase "A or B" will be understood to include the possibilities of "A" or "B" or "A and B".

[0075] Many of the present disclosure and its associated advantages will be understood from the foregoing description, and it will become clear that various modifications can be made to the form, structure, and arrangement of the components without departing from the disclosed subject matter or sacrificing any of its material advantages. The described forms are for illustrative purposes only, and it is the intent of the following claims to encompass and include such modifications. Furthermore, it should be understood that the present invention is defined by the appended claims.

Claims

1. It is a system, A controller communicatively coupled to one or more through-focus imaging measurement subsystems, the controller comprising one or more processors configured to execute a set of program instructions stored in memory, the set of program instructions is configured to execute the one or more processors The steps include receiving one or more training images, which are multiple training images captured at one or more focal positions and which include one or more training features of a training sample, The steps include generating a machine learning classifier based on multiple training images captured at one or more focal positions, The steps include receiving one or more target feature selections, which are user-selected target features, for one or more target overlay measurements corresponding to one or more target features of a target sample, The steps include: determining one or more target focus locations based on the selection of one or more target features using the machine learning classifier; Steps include receiving one or more target images captured at one or more target focal positions, wherein the one or more target images include one or more target features of a target sample, The steps include determining one or more overlays based on the one or more target images, A system characterized by having a controller configured to perform the following actions.

2. The system according to claim 1, wherein the one or more through-focus imaging measurement subsystems comprises at least one of an optical-based measurement subsystem or a scattering-based measurement subsystem.

3. The system according to claim 1, characterized in that the one or more focal positions include a plurality of focal positions within the focus training range.

4. The system according to claim 1, characterized in that the plurality of training images are captured at one or more focal positions by translating one or more parts of the measurement subsystem along one or more adjustment axes.

5. The set of program instructions is provided to one or more processors. One or more control signals are provided to one or more parts of one or more of the measurement subsystems, and the one or more control signals cause one or more parts of the measurement subsystems to move in parallel along one or more adjustment axes. The system according to claim 4, further characterized by being configured as follows.

6. The system according to claim 5, characterized in that the one or more control signals are configured to center one or more portions of the one or more through-focus imaging measurement subsystems.

7. The step of determining one or more overlays based on one or more target images of one or more target features is: The system according to claim 1, comprising the step of determining the overlay between the first layer of the target sample and the second layer of the target sample based on a first overlay measurement corresponding to one or more target features formed on the first layer of the target sample and a second overlay measurement corresponding to one or more target features formed on the second layer of the target sample.

8. The system according to claim 1, characterized in that the selection of one or more target features for one or more overlay measurements is provided by the user via a user interface.

9. The system according to claim 1, wherein the set of program instructions is further configured to provide one or more control signals to one or more process tools for one or more processors.

10. The system according to claim 1, characterized in that the machine learning classifier comprises at least one of a deep learning classifier, a convolutional neural network, an ensemble learning classifier, a random forest classifier, or an artificial neural network.

11. It is a system, One or more through-focus imaging and measurement subsystems, A controller communicatively coupled to one or more through-focus imaging measurement subsystems, comprising one or more processors configured to execute a set of program instructions stored in memory, wherein the set of program instructions is configured to execute the one or more processors The steps include receiving one or more training images, which are multiple training images captured at one or more focal positions and which include one or more training features of a training sample, The steps include generating a machine learning classifier based on multiple training images captured at one or more focal positions, A step of receiving one or more target feature selections, which are user-selected target features, for one or more target overlay measurements corresponding to one or more target features of a target sample; A step of determining one or more target focus locations based on the selection of one or more target features using a machine learning classifier, Steps include receiving one or more target images captured at one or more target focal positions, wherein one or more target images include one or more target features of a target sample; A step of determining one or more overlays based on one or more target images, A system characterized by being configured to perform the following action.

12. An overlay measurement method using one or more through-focus imaging measurement subsystems, A step of receiving multiple training images captured at one or more focal positions, wherein one or more training images include one or more training features of a training sample, The steps include generating a machine learning classifier based on the plurality of training images captured at one or more focal positions, The steps include receiving one or more target feature selections, which are user-selected target features, for one or more target overlay measurements corresponding to one or more target features of a target sample, The steps include: determining one or more target focus locations based on the selection of one or more target features using the machine learning classifier; A step of receiving one or more target images captured at one or more target focal positions, wherein the one or more target images include one or more target features of a target sample. The steps include determining one or more overlays based on the one or more target images, A method that includes this.

13. The method according to 12, characterized in that the plurality of training images and the one or more target images are captured by one or more through-focus imaging measurement subsystems, each comprising at least one of an optical-based measurement subsystem or a scattering measurement-based measurement subsystem.

14. The method according to 12, characterized in that the one or more focal positions include a plurality of focal positions within the focus training range.

15. The method according to 13, characterized in that the plurality of training images are captured at one or more focal positions by translating one or more parts of the measurement subsystem along one or more adjustment axes.

16. The method according to claim 15, further comprising providing one or more control signals to one or more parts of the one or more through-focus imaging measurement subsystems, wherein the one or more control signals are configured to move one or more parts of the one or more through-focus imaging measurement subsystems in parallel along one or more adjustment axes.

17. The method according to 16, characterized in that the one or more control signals are configured to center the one or more portions of the one or more through-focus imaging measurement subsystems.

18. The step of determining one or more overlays based on one or more target images of one or more target features is: The method according to 12, comprising the step of determining the overlay between the first layer of the target sample and the second layer of the target sample based on a first overlay measurement corresponding to one or more target features formed on a first layer of the target sample and a second overlay measurement corresponding to one or more target features formed on a second layer of the target sample.

19. The method according to 12, characterized in that the selection of one or more target features for one or more overlay measurements is provided by the user via a user interface.

20. The method according to claim 12, further comprising the step of providing one or more control signals to one or more process tools.

21. The method according to 12, characterized in that the machine learning classifier comprises at least one of a deep learning classifier, a convolutional neural network, an ensemble learning classifier, a random forest classifier, or an artificial neural network.

Images