Evaluation device, information processing device, program, film formation system, and method for manufacturing articles

Abstract

To provide a technique advantageous in obtaining, in detail, the position and the shape of an abnormality of a composition on a substrate which is obtained by a film forming process.SOLUTION: An evaluation apparatus includes an obtaining unit configured to obtain an image of an evaluation region including a film forming region on which a film is formed by a film forming process, and a processor configured to process the image for evaluation. The processor is configured to output a feature concerning an abnormality in the image in accordance with a learned model. The image and design information representing a geometrical feature of the film forming region are input to the learned model.SELECTED DRAWING: Figure 10

Description

Technical Field

[0001] The present invention relates to an evaluation apparatus, an information processing apparatus, a program, a film forming system, and an article manufacturing method.

Background Art

[0002] The practical application of imprint technology, which is a technology for forming fine patterns, is progressing. One of the imprint technologies is the photo-curing method. An imprint apparatus employing the photo-curing method irradiates light to cure an imprint material while bringing a master (mold) into contact with a photo-curable moldable material (imprint material) supplied onto a substrate. Thereafter, the mold is separated from the cured imprint material, thereby forming a pattern on the substrate. For example, for manufacturing a semiconductor device or the like, an apparatus applying step-and-flash imprint lithography is effective (Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] When forming a pattern on a substrate using imprint technology, which is one of the film forming technologies, the imprint material may overflow outside the pattern area because the supply amount of the imprint material is too large (bleeding). Conversely, there are cases where the imprint material does not spread evenly and a pattern cannot be formed partially because the supply amount of the imprint material is small (unfilled). When bleeding occurs, not only does the bleeding part become a pattern formation defect, but it also causes the pattern of the mold that has come into contact with that part to be damaged. Also, when unfilling occurs, a pattern is not formed in that part, resulting in a defect as a semiconductor device.

[0005] Therefore, after the imprinting process, it is necessary to detect the presence or absence of such seepage and incomplete filling, and to prevent defects by adjusting the supply amount and position of the imprint material according to the detection results. However, since seepage and incomplete filling occur in minute areas, it is necessary to check a huge number of observation images obtained with a high-magnification microscope with a narrow detection range, which is difficult for humans to do. Therefore, there is a need for a technology that can inspect for seepage and incomplete filling from observation images without human intervention and determine pattern formation defects caused by seepage and incomplete filling. In the following, pattern formation defects caused by seepage and incomplete filling will also be referred to as "abnormalities."

[0006] The way this anomaly manifests varies depending on the imprint conditions. Adjusting the imprint conditions, including the amount of imprint material supplied, requires detailed detection of not only the presence or absence of an anomaly, but also its location, shape, and other information.

[0007] The present invention provides a technique advantageous for precisely determining the location and shape of abnormalities in a composition on a substrate obtained by a film formation process. [Means for solving the problem]

[0008] According to one aspect of the present invention, an evaluation apparatus for evaluating a film on a substrate that has undergone a film formation process in which a film of a composition is formed on a film formation region of the substrate using a mold, comprising: an acquisition unit that acquires an image of an evaluation region including the film formation region in which the film has been formed by the film formation process; and a processing unit that processes the image for the evaluation, wherein the processing unit is configured to output features related to abnormalities in the image according to a learning model, and the image and design information indicating the geometric features of the film formation region are input to the learning model. The learning model then calculates a confidence score representing the reliability of the detected anomaly. An evaluation device characterized by the above is provided. [Effects of the Invention]

[0009] According to the present invention, it is possible to provide a technique that is advantageous for precisely determining the location and shape of abnormalities in a composition on a substrate obtained by a film formation process. [Brief explanation of the drawing]

[0010] [Figure 1] A diagram showing the configuration of an imprinting device. [Figure 2] A diagram showing the configuration of a wide-area alignment measuring instrument. [Figure 3] A diagram showing the configuration of a product manufacturing system. [Figure 4] A flowchart illustrating the operation of an imprinting device. [Figure 5] A diagram illustrating seepage and incomplete filling. [Figure 6] A diagram illustrating images including seepage and incomplete filling. [Figure 7] A diagram showing an example of unfilled areas at the marked locations. [Figure 8] This figure shows an example of an image where the boundaries of the shot area and the original shape of the marks are unknown. [Figure 9] A diagram illustrating design information corresponding to an image obtained through imaging. [Figure 10] A diagram illustrating examples of information input to and output from a machine learning model. [Figure 11] A flowchart illustrating a method for detecting anomalies from images. [Figure 12] A schematic diagram of the methods used during learning and testing. [Figure 13] A flowchart showing how to create a learning model. [Figure 14] A diagram showing the configuration of the evaluation device. [Figure 15] A diagram illustrating the method for manufacturing articles in an embodiment. [Modes for carrying out the invention]

[0011] Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. Note that the following embodiments do not limit the invention according to the claims. Although a plurality of features are described in the embodiments, not all of these plurality of features are essential for the invention, and the plurality of features may be arbitrarily combined. Further, in the accompanying drawings, the same or similar configurations are denoted by the same reference numerals, and redundant descriptions are omitted.

[0012] The following embodiments relate to a film forming system including a film forming apparatus. The film forming apparatus is used for manufacturing devices such as semiconductor devices as articles, arranges an uncured composition on a substrate, forms the arranged composition with a mold, and forms a film of the composition on the substrate. The film forming apparatus may be called a forming apparatus, and similarly, the film forming process may be called a forming process.

[0013] The film forming process includes a contact step of bringing a formable material supplied onto a substrate into contact with a mold (master, template). By this contact, the formable material is formed. The film forming process may further include a curing step of curing the formable material while the formable material and the mold are in contact with each other. Thereby, a composition made of a cured product of the formable material is formed on the substrate. The film forming process may further include a separation step of separating the composition made of the cured product of the formable material from the mold.

[0014] The film forming apparatus can be used as an imprint apparatus that transfers the pattern of the mold to the imprint material, which is a formable material supplied onto the shot area where a pattern is to be formed on the substrate, by bringing the imprint material into contact with the pattern portion of the mold. The pattern can be, for example, a pattern (device pattern) of a semiconductor device. In the imprint apparatus, imprinting can be performed for each of a plurality of shot areas formed on the substrate. Alternatively, the imprint apparatus can be configured to perform imprinting (i.e., contact) collectively for a plurality of shot areas of the substrate (in units of the entire substrate surface or one or more columns of shot areas).

[0015] Alternatively, the film forming apparatus can also be used as a planarizing apparatus that performs a planarizing process to form a planarized film on a substrate by bringing the formable material on the substrate into contact with a member having a flat surface (the flat surface of a mold).

[0016] In the following section, to illustrate a specific example, we will describe a system that includes an imprinting apparatus, which is an example of a film formation apparatus.

[0017] Figure 1(a) schematically shows the configuration of the imprint apparatus IMP in the embodiment. The imprint apparatus IMP performs an imprint process in which the imprint material IM on the substrate S is brought into contact with the pattern area MP of the mold M, the imprint material IM is cured, and the cured imprint material IM is separated from the mold M. Through this imprint process, a pattern consisting of the cured imprint material IM is formed on the substrate S.

[0018] The imprint material used is a curable composition (sometimes called an uncured resin) that hardens when curing energy is applied. The curing energy can be electromagnetic waves, heat, etc. Electromagnetic waves may be light selected from a wavelength range of 10 nm to 1 mm, such as infrared rays, visible light, ultraviolet rays, etc. The curable composition may be a composition that hardens by irradiation with light or by heating. Of these, the photocurable composition that hardens by irradiation with light contains at least a polymerizable compound and a photopolymerization initiator, and may further contain a non-polymerizable compound or a solvent as needed. The non-polymerizable compound is at least one selected from the group of sensitizers, hydrogen donors, internal release agents, surfactants, antioxidants, polymer components, etc. The imprint material may be arranged on the substrate in the form of droplets, islands formed by multiple droplets being connected, or as a film. The viscosity of the imprint material (viscosity at 25°C) may be, for example, 1 mPa·s to 100 mPa·s. The substrate material can be, for example, glass, ceramics, metal, semiconductor, or resin. If necessary, a component made of a different material from the substrate may be provided on the surface of the substrate. Examples of substrates include silicon wafers, compound semiconductor wafers, and quartz glass.

[0019] In this specification and the accompanying drawings, directions are indicated in an XYZ coordinate system where the direction parallel to the surface of the substrate S is the XY plane. The directions parallel to the X, Y, and Z axes in the XYZ coordinate system are denoted as the X direction, Y direction, and Z direction, respectively, and the rotations around the X, Y, and Z axes are denoted as θX, θY, and θZ, respectively. Control or driving with respect to the X, Y, and Z axes means control or driving with respect to the direction parallel to the X, Y, and Z axes, respectively. Control or driving with respect to the θX, θY, and θZ axes means control or driving with respect to rotation around the axis parallel to the X, Y, and Z axes, respectively. Position is information that can be determined based on the coordinates of the X, Y, and Z axes, and orientation is information that can be determined by the values ​​of the θX, θY, and θZ axes. Positioning means controlling the position and / or orientation. Alignment may include controlling the position and / or orientation of at least one of the substrate and the mold.

[0020] The imprint apparatus (IMP) may include a substrate holding section 102 for holding a substrate S, a substrate driving mechanism 105 for driving the substrate S by driving the substrate holding section 102, a base 104 for supporting the substrate holding section 102, and a position measuring section 103 for measuring the position of the substrate holding section 102. The substrate driving mechanism 105 may include, for example, a motor such as a linear motor.

[0021] The imprinting apparatus (IMP) may include a mold holding section 121 for holding a mold M, a mold driving mechanism 122 for driving the mold M by driving the mold holding section 121, and a support structure 130 for supporting the mold driving mechanism 122. The mold driving mechanism 122 may include, for example, a motor such as a voice coil motor.

[0022] The substrate drive mechanism 105 and the mold drive mechanism 122 constitute a drive mechanism that adjusts the relative position and relative orientation of the substrate S and the mold M. The adjustment of the relative position of the substrate S and the mold M by the drive mechanism includes driving for contact between the mold and the imprint material on the substrate S, and for separating the mold from the cured imprint material (cured pattern). The substrate drive mechanism 105 may be configured to drive the substrate S along multiple axes (e.g., three axes: X, Y, and θZ; preferably six axes: X, Y, Z, θX, θY, and θZ). The mold drive mechanism 122 may be configured to drive the mold M along multiple axes (e.g., three axes: Z, θX, and θY; preferably six axes: X, Y, Z, θX, θY, and θZ).

[0023] The imprint apparatus IMP may include a mold transport mechanism 140 for transporting mold M and a mold cleaner 150. The mold transport mechanism 140 may be configured, for example, to transport mold M to a mold holding section 121, or to transport mold M from the mold holding section 121 to a master plate stocker (not shown) or a mold cleaner 150, etc. The mold cleaner 150 cleans mold M with ultraviolet light, chemicals, etc.

[0024] The mold holding portion 121 may include a window member 125 that forms a pressure control space CS on the back side of the mold M (the side opposite to the pattern region MP on which the pattern to be transferred to the substrate S is formed). The imprint apparatus IMP may include a deformation mechanism 123 that deforms the pattern region MP of the mold M into a convex shape toward the substrate S by controlling the pressure in the pressure control space CS (hereinafter referred to as cavity pressure), as schematically shown in Figure 1(b).

[0025] The imprint apparatus (IMP) may include an alignment measuring instrument 106, a wide-angle alignment measuring instrument 151, a hardening unit 107, an imaging unit 112, and an optical member 111. The alignment measuring instrument 106 measures the relative position between the alignment marks by illuminating the alignment marks on the substrate S and the alignment marks on the mold M and capturing their images. The alignment measuring instrument 106 may be positioned by a drive mechanism (not shown) according to the position of the alignment marks to be observed. The wide-angle alignment measuring instrument 151 is a measuring instrument with a wider field of view than the alignment measuring instrument 106, and measures the position of the substrate S by illuminating the alignment marks on the substrate S and capturing their images. By measuring the position of the substrate S with the wide-angle alignment measuring instrument, the alignment marks on the substrate S can be moved within the field of view of the alignment measuring instrument 106.

[0026] The curing unit 107 irradiates the imprint material IM with energy (for example, light such as ultraviolet light) to cure the imprint material IM via the optical member 111, thereby curing the imprint material IM. The imaging unit 112 images the substrate S, mold M, and imprint material IM via the optical member 111 and the window member 125.

[0027] The wide-angle alignment measuring instrument 151 may have a mechanism for switching the wavelength of illumination light. For example, the wide-angle alignment measuring instrument 151 may have wavelength filters arranged on the optical path and a mechanism for switching the wavelength filters. Alternatively, the wide-angle alignment measuring instrument 151 may have a configuration that allows simultaneous imaging of multiple wavelengths, as shown in Figure 2. The measuring instrument in Figure 2 includes a light source LS, multiple half-mirrors 162 branching from the optical path, multiple wavelength filters 163 each transmitting different wavelengths, and multiple image sensors 164, allowing simultaneous imaging of images of different wavelengths. The wide-angle alignment measuring instrument 151 may also have a mechanism for switching the light intensity of illumination light. For example, the wide-angle alignment measuring instrument 151 may have a mechanism for switching ND filters arranged on the optical path. Furthermore, the wide-angle alignment measuring instrument 151 may have multiple optical systems, such as bright-field optical systems and dark-field optical systems, and a mechanism for switching the optical system through which the image to be captured passes. Furthermore, the wide-angle alignment measuring instrument 151 may have a mechanism for switching the polarization of the illumination light or the received light. For example, the wide-angle alignment measuring instrument 151 may also have a mechanism for switching the polarization filters placed in the optical path.

[0028] The imprint apparatus IMP may include a dispenser 108 for placing imprint material IM on a substrate S. The dispenser 108 dispenses the imprint material IM so that it is placed on the substrate S according to a drop recipe indicating the placement of the imprint material IM. The imprint apparatus IMP may include a control unit 110 for controlling a substrate drive mechanism 105, a mold drive mechanism 122, a deformation mechanism 123, a mold transport mechanism 140, a mold cleaner 150, an alignment measuring instrument 106, a curing unit 107, an imaging unit 112, the dispenser 108, etc. The control unit 110 may be composed of, for example, a PLD (Programmable Logic Device) such as an FPGA (Field Programmable Gate Array), or an ASIC (Application Specific Integrated Circuit), or a general-purpose computer with a program installed, or a combination of all or part of these.

[0029] Figure 3 illustrates the configuration of an article manufacturing system 401 for manufacturing articles such as semiconductor devices. The article manufacturing system 401 may include, for example, one or more imprint devices (IMPs) and one or more inspection devices 405 (e.g., an overlay inspection device, a CD inspection device, a defect inspection device, an electrical characteristic inspection device). The article manufacturing system 401 may also include one or more substrate processing devices 406 (etching devices, film deposition devices). Furthermore, the article manufacturing system 401 may also include an evaluation device 407, which will be described later. These devices may be connected via a network 402 to a control device 403, which is an external device separate from the imprint device (IMP), and controlled by the control device 403.

[0030] The evaluation device 407 may be composed of an information processing device. This information processing device may be composed of, for example, a PLD (Programmable Logic Device) such as an FPGA (Field Programmable Gate Array), or an ASIC (Application Specific Integrated Circuit), or a general-purpose computer with a program installed, or a combination of all or part of these.

[0031] Figure 14 shows an example configuration of the evaluation device 407. The evaluation device 407 may include a control unit 4071, a RAM 4072 that stores temporary data and provides a work area to the control unit 4071, and a ROM 4073 that stores fixed data and programs. The evaluation device 407 may further include a storage device 4072, a display device 4076, and an input device 4075. The storage device 4072 stores a program 4074a for executing the evaluation method according to this embodiment. The network I / F 4077 is an interface for connecting to the network 402. In this embodiment, the network I / F 4077 may function as an acquisition unit that acquires an image including a composition formed on a substrate by a film formation process. The control unit 4071 may also function as a processing unit that processes the acquired image for evaluation. Furthermore, the control unit 4071 may also function as a display control unit that controls the display of the display unit 4076.

[0032] The functions of the evaluation device 407 may be realized by the control unit 110 of the imprint device IMP, the control unit 403, or the control unit of the inspection device 405, or a combination thereof. In this embodiment, the system including the imprint device IMP and the evaluation device 407 may be understood as a film formation system or a lithography system.

[0033] The lithography method according to this embodiment will be described below. In this embodiment, after imprint processing, an image of the shot area (film formation area), which is the area where the pattern should be formed, and the surrounding area (evaluation area) is acquired by imaging. Leakage and unfilling are detected using this image. Machine learning is used to detect leakage and unfilling. Furthermore, anomaly detection can be achieved by using an object detection algorithm in machine learning.

[0034] The operation of the imprint device (IMP) will be explained with reference to the flowchart in Figure 4. The operation shown in Figure 4 can be controlled by the control unit 110.

[0035] In step S101, the substrate S is transported from the transport source (for example, the relay section between the preprocessing device and the imprinting device IMP) onto the substrate holding section 102 by a substrate transport mechanism (not shown). The position of the transported substrate S on the substrate holding section 102 is measured by observing the marks on the substrate S using a wide-angle alignment measuring instrument 151. The control unit 110 positions the substrate S based on the position obtained from the measurement.

[0036] In steps S102 to S106, an imprint process (pattern formation) is performed on a selected shot area from among multiple shot areas of the substrate S.

[0037] In step S102, the imprint material IM is placed on the selected shot area by the dispenser 108. This process can be performed by dispensing the imprint material IM from the dispenser 108 while driving the substrate S with the substrate driving mechanism 105.

[0038] In step S103, the substrate S and the mold M are driven relative to each other by at least one of the mold driving mechanism 122 and the substrate driving mechanism 105 so that the pattern area MP of the mold M contacts the imprint material IM on the shot area. In one example, the mold M is driven by the mold driving mechanism 122 so that the pattern area MP of the mold M contacts the imprint material IM on the shot area. In the process of bringing the pattern area MP of the mold M into contact with the imprint material IM, the pattern area MP of the mold M may be deformed into a convex shape toward the substrate S by the deformation mechanism 123.

[0039] In step S104, alignment can be performed between the shot area and the pattern area MP of the mold M. Alignment can be performed by measuring the relative position between the alignment mark of the shot area to be imprinted and the alignment mark of the mold M using an alignment measuring instrument 106, so that the relative position falls within the tolerance range of the target relative position. During alignment, the substrate S and the mold M are driven relative to each other by at least one of the mold driving mechanism 122 and the substrate driving mechanism 105. The target relative position between the alignment mark of the shot area to be imprinted and the alignment mark of the mold M can be determined by a correction value determined from the results of past overlay inspection devices, etc.

[0040] In step S105, energy for curing the imprint material IM is irradiated onto the imprint material IM between the substrate S and the pattern region MP of the mold M by the curing unit 107. As a result, the imprint material IM hardens, and a cured product of the imprint material IM is formed.

[0041] In step S106, the substrate S and the mold M are driven relative to each other by at least one of the mold driving mechanism 122 and the substrate driving mechanism 105 so that the cured imprint material IM and the pattern region MP of the mold M are separated. In one example, the mold M is driven by the mold driving mechanism 122 so that the cured imprint material IM and the pattern region MP of the mold M are separated. Even when the cured imprint material IM and the pattern region MP of the mold M are separated, the pattern region MP of the mold M may be deformed into a convex shape toward the substrate S. Furthermore, imaging is performed by the imaging unit 112, and the state of separation between the imprint material IM and the mold M is observed based on the captured image.

[0042] In step S107, the control unit 110 determines whether the imprint processing steps S102 to S106 have been performed on all shot areas of the substrate S. If the imprint processing steps S102 to S106 have been performed on all shot areas of the substrate S, the process proceeds to step S108. If there are unprocessed shot areas, the process returns to step S102. In this case, the imprint processing steps S102 to S106 are performed on selected shot areas from among the unprocessed shot areas.

[0043] In step S108, an image of the region including the imprinted shot region (film formation region) (evaluation region) is acquired in order to detect anomalies. For example, the control unit 110 uses a wide-angle alignment measuring instrument 151 to capture an image of the shot region and its surroundings. If the field of view of the wide-angle alignment measuring instrument 151 is narrow relative to the shot region, the position of the substrate S may be changed by driving the substrate driving mechanism 105, and multiple images may be captured to obtain an image of the desired region. The image obtained in step S108 can be used as a training image, as described later. The image obtained in step S108 can also be used as an image for anomaly detection. Here, an example of capturing an image with a wide-angle alignment measuring instrument 151 is described, but it is not limited to this. For example, imaging may be performed using an alignment measuring instrument 106 or an imaging unit 112, etc.

[0044] Furthermore, in the above procedure, step S108 is executed after imprint processing has been performed on all of the multiple shot areas, but this is not the only option. For example, after forming a pattern in each shot area (after step S106), the imprinted pattern may be imaged for each shot area. Also, as will be described later, the pattern of the substrate removed from the imprint device may be imaged using a device other than the imprint device in the same manner as in step S108.

[0045] In step S109, the substrate S is transported from the substrate holding unit 102 to the transport destination (for example, the relay unit between the imprinting device IMP and the post-processing device) by a substrate transport mechanism (not shown). The operation shown in Figure 4 is performed for each of the multiple substrates when a lot consisting of multiple substrates is being processed.

[0046] Next, we will explain examples of abnormalities that occur during imprinting. Figure 5 is a side view of the state in which the mold M and the imprint material IM on the substrate S are in contact (after the completion of process S103, for example, in processes S104 and S105). Leakage refers to the state in which the imprint material IM spills out from the contact area between the mold M and the imprint material IM, as shown in Figure 5(a). Incomplete filling refers to the state in which there is a portion between the mold M and the substrate S that is not filled with imprint material IM, as shown in Figure 5(b).

[0047] Figure 6 shows examples of images obtained by imaging in step S108 when seepage and incomplete filling occur. Figure 6 shows an image of the state in which a pattern of imprint material IM has been formed in the shot area by the imprint process. Normally, as shown in Figure 6(a), the imprint material IM fills up to the boundary 601 of the shot area and forms a pattern. On the other hand, in the case of incomplete filling, as shown in Figure 6(b), the imprint material IM does not reach the boundary 601 of the shot area, and the unfilled area is imaged as white (or black). Also, in the case of seepage, as shown in Figure 6(c), the imprint material IM extends beyond the boundary 601 of the shot area and is imaged as black (or white) beyond the boundary 601.

[0048] Figure 7 shows an example of unfilled areas at the location of marks used for alignment and inspection. Figures 7(a) and (b) show images (IMG) of the shot area, respectively. The gray areas in each image represent the shot area of ​​the inspection target, which is filled with imprint material (IM). The shot area contains two marks, a first mark 701 and a second mark 702, which have different shapes. In Figure 7(a), the first mark 701 and the second mark 702 are shown in black, representing a normal state where the imprint material (IM) has filled the inside of the marks. On the other hand, in Figure 7(b), the first mark 701 and the second mark 702 are shown in white in some areas, representing a state where the imprint material has not sufficiently filled the inside of the marks (unfilled area N).

[0049] If mold M comes into contact with a shot area where bleed has occurred, the pattern formed on mold M may be destroyed. Furthermore, if unfilled areas occur, no pattern will be formed there, resulting in a defective semiconductor device. Therefore, it is necessary to detect the presence or absence of bleed or unfilled areas after the imprint process and adjust the imprint conditions to prevent these defects.

[0050] One example of an adjustment method is to change the amount of imprint material supplied based on the size of any seepage or unfilled areas. To perform this adjustment, it is necessary to obtain information on the location, size, and shape of areas where the amount of imprint material is insufficient or excessive. In this embodiment, this information is obtained (detected) from images using machine learning. Detection methods include using models such as those exemplified below. • Model of Convolutional Neural Network structure • Models with an AutoEncoder mechanism such as U-net, • A model based on Region-Convolutional Neural Network (R-CNN). Using these models, it is possible to determine whether a target object exists for each pixel in an image, and then label them to obtain the detailed shape of the object.

[0051] On the other hand, there are the following challenges in detecting the location, size, and shape of anomalies from captured images. Figures 6(a) to 6(c) show lines representing the boundary 601 of the shot area, but such lines do not necessarily exist in reality. Therefore, if an image like Figure 8(a) is obtained through imaging, the information from this image alone does not reveal the boundary position of the shot area, making it impossible to determine the size of the unfilled or seeped area 801 of the imprint material. Also, if images of marked areas 802 and 803 like Figure 8(b) are obtained through imaging, it is impossible to determine whether these are abnormal or not from these images alone, as the original shape is unknown.

[0052] To address this problem, this embodiment uses design information that shows the geometric features of the shot area to detect anomalies. Geometric features of the shot area may include information that identifies the boundary position of the shot area, the position and shape of marks within the shot area, etc., as will be described later. Hereafter, this information will be referred to as "design information". Figure 9(a) is an image showing the design information corresponding to the position where the image shown in Figure 6(a) was captured. The image in Figure 9(a) shows how far the imprint material IM in Figure 6(a) should be filled. In this embodiment, as shown in Figure 10(a), two images, the image obtained by capturing Figure 6(a) and the image showing the design information in Figure 9(a), are input to the machine learning model as features. As a result, the machine learning model can recognize the boundary of the shot area and correctly detect the location and size of unfilled or seepage areas.

[0053] Furthermore, the image in Figure 9(b) is an image showing the design information (position and shape) of the mark corresponding to the position where the image shown in Figure 7(a) was captured. The image in Figure 9(b) shows the precise position and shape of the mark in Figure 7(a). In this embodiment, as shown in Figure 10(b), the image obtained by capturing in Figure 6(b) and the design information in Figure 9(b) hmm Two images, one representing information and the other, are input into the machine learning model as features. machine The learning model can correctly detect unfilled areas in the marked sections.

[0054] Furthermore, semiconductor devices typically form circuits by layering multiple different patterns. Therefore, patterns are already formed on the substrate, and the image obtained through imaging may include patterns other than the one to be detected. In such cases, efficient and accurate anomaly detection can be performed by excluding the non-detectable patterns from the design information of the already formed non-detectable patterns.

[0055] Furthermore, the above example described a method in which design information is represented in image format, and the image representing that design information is input into the machine learning model. Alternatively, the design information could be represented as information about vertices, line segments, or polygons, such as the shape of shot boundaries or marks, and this information could be input into the machine learning model.

[0056] Design information, specifically the geometric characteristics of the shot area, such as the boundary position of the shot area and the position and shape of marks within the shot area, can be obtained, for example, from recipe information pre-entered for pattern formation. Alternatively, design information may be obtained by taking measurements using an imaging device or measuring device on a substrate that has been correctly imprinted (without any abnormalities occurring) under the same conditions as the object being inspected.

[0057] In this embodiment, abnormalities in the shot edge region and marks have been described, but it is also possible to similarly detect abnormalities in patterns other than marks formed on the shot region.

[0058] Referring to the flowchart in Figure 11, the image evaluation method performed by the evaluation device 407 will be explained. The evaluation method involves detecting abnormalities contained in the image acquired in process S108 and determining the type of abnormality (e.g., seepage / unfilled). 11 The evaluation method program corresponding to the flowchart is stored, for example, in the storage device 4074, loaded into the RAM 4072, and then executed by the control unit 4071 (processing unit).

[0059] In S201, the control unit 4071 loads a machine learning model (inference model; hereinafter simply referred to as "model") that outputs one or more anomaly features in the image. The model is a model created in advance using the conditions of the imprint material to be inspected and images acquired under conditions similar to the measurement conditions of the image acquired in S108. The procedure for creating the model will be described later.

[0060] Subsequently, the inspection results for each image are obtained by repeating steps S202 to S205. In S202, the control unit 4071 reads the image acquired in S108 as an image for inspection. In S203, the control unit 4071 acquires the design information of the pattern corresponding to the image read in S202. In S204, the control unit 4071 provides the image read in S202 and the design information acquired in S203 as input to the model read in S201, and obtains the characteristics of anomalies on the input image as output. Anomaly characteristics are obtained for each anomaly on the image. The obtained anomaly characteristics may include the location, size, and shape of the anomaly, as well as the type of anomaly (seepage / unfilled), the coordinates of the vertices of a rectangle surrounding the anomaly area, and the confidence level of the detected anomaly. Here, the confidence level of the detected anomaly is a value that represents the reliability of the detection result and is automatically calculated by the model for each detected anomaly. The types of anomalies that can be detected include seepage and unfilled as described above, but other types of anomalies can also be detected by training the model.

[0061] In S205, the control unit 4071 performs post-processing on the output from the model obtained in S204. For example, the post-processing may include classifying the detected anomalies by comparing the confidence level of the anomaly with a predetermined threshold. For example, when the confidence level is expressed as a value between 0 and 1, anomalies with a confidence level of 0.5 or less can be classified as anomalies that are not subject to detection. The classification conditions, such as the threshold used when classifying based on the confidence level, can be changed by various data such as the imprint material to be inspected, recipe information, lighting conditions during imaging, and imaging mode.

[0062] Next, with reference to Figures 12 and 13, we will describe how the evaluation device 407 calculates (learns) an inference model (learning model) for anomaly detection. Figure 12 is a schematic diagram of the method performed during learning and inspection. Figure 13 is a flowchart of the learning process for anomaly detection. The evaluation device 407 may be equipped with a machine learning unit that generates the inference model by machine learning. The machine learning unit performs machine learning on the relationship between the image of the evaluation region, which is a region including the film formation region, and the design information of the film formation region, as inputs, and the features related to anomalies, as outputs. This will be explained in detail below.

[0063] In step S301, the control unit 4071 acquires training images 801 in the same manner as in step S108. Specifically, the control unit 4071 collects images 801 similar to the images taken during inspection, based on the conditions of the imprint material used when capturing images during inspection and the measurement conditions. The control unit 4071 also collects training images 801 from multiple shot regions on multiple substrates. It is desirable to have a large number of images used for training, and that they contain a sufficient amount of samples of the abnormalities to be detected. In addition, the control unit 4071 also acquires design information 802 corresponding to the training images 801, in the same manner as in S203.

[0064] In process S302, feature information is created that indicates the characteristics of anomalies corresponding to each image acquired in process S301. Specifically, each acquired image is visually inspected, and feature information 803 is created for all anomalies present in each image, including information such as the category, size, and location of the anomaly.

[0065] In step S303, the control unit 4071 performs machine learning on the relationship between the image and design information acquired in S301 and the anomaly feature information 803 created in S302, and creates a learning model. Here, for example, optimization is performed by using the image 801 and design information 802 as input data for a pre-created neural network, and the anomaly feature information 803 as the output (training). Through this optimization, a learning model (neural network) 804 is created.

[0066] In step S304, the control unit 4071 saves the created learning model 804 to the storage unit 805. In the example described above, the evaluation device 407 is described as an information processing device that performs both learning and inspection. However, the information processing device that performs learning and the information processing device that performs inspection may be configured as separate devices. In that case, the first information processing device creates a learning model and transfers that learning model to the second information processing device that performs inspection. The second information processing device uses the learning model transferred from the first information processing device to inspect the input image.

[0067] According to the embodiments described above, it is possible to automate the detection of anomalies in the peripheral area of ​​the shot region and to classify anomalies in more detail. In the above-described embodiment, an imprint apparatus was explained. As mentioned above, in the case of an imprint apparatus, the "evaluation area" is the area including the shot area and its surroundings. In contrast, when the present invention is applied to a planarization apparatus, it is assumed that the "evaluation area" will be the area including the entire substrate.

[0068] <Embodiment of Article Manufacturing Method> The article manufacturing method according to an embodiment of the present invention is suitable for manufacturing articles such as microdevices, semiconductor devices, and elements having microstructures. The article manufacturing method of this embodiment may include the steps of forming a layer of composition on a substrate using a molding apparatus in a molding system, and processing the substrate on which the layer has been formed. Furthermore, such a manufacturing method may include other well-known steps (oxidation, film formation, vapor deposition, doping, planarization, etching, resist stripping, dicing, bonding, packaging, etc.). The article manufacturing method of this embodiment is advantageous over conventional methods in at least one of the performance, quality, productivity, and production cost of the article.

[0069] The patterns of cured materials formed using an imprint apparatus are used permanently on at least a portion of various articles, or temporarily during the manufacturing of various articles. Articles include electrical circuit elements, optical elements, MEMS, recording elements, sensors, or molds. Examples of electrical circuit elements include volatile or non-volatile semiconductor memories such as DRAM, SRAM, flash memory, and MRAM, as well as semiconductor elements such as LSI, CCD, image sensors, and FPGAs. Examples of molds include molds for imprinting.

[0070] The pattern of the cured material is either used as is as a component of at least a part of the above-mentioned article, or temporarily used as a resist mask. After etching or ion implantation is performed during the substrate processing process, the resist mask is removed.

[0071] Next, the manufacturing method for the article will be explained. In step SA shown in Figure 15, a substrate 1z such as a silicon substrate with a workpiece 2z such as an insulator formed on its surface is prepared, and then an imprint material 3z is applied to the surface of the workpiece 2z by an inkjet method or the like. Here, multiple droplet-shaped imprint material 3z are shown applied to the substrate.

[0072] In step SB of Figure 15, the mold 4z for imprinting is positioned opposite the imprint material 3z on the substrate, with the side where the uneven pattern is formed facing it. In step SC of Figure 15, the substrate 1z to which the imprint material 3z has been applied is brought into contact with the mold 4z, and pressure is applied. The imprint material 3z fills the gap between the mold 4z and the workpiece 2z. In this state, when light is irradiated through the mold 4z as curing energy, the imprint material 3z hardens.

[0073] In process SD shown in Figure 15, after the imprint material 3z is cured, the mold 4z and the substrate 1z are separated, forming a pattern of the cured imprint material 3z on the substrate 1z. In this cured pattern, the recesses of the mold correspond to the protrusions of the cured material, and the protrusions of the mold correspond to the recesses of the cured material. In other words, the uneven pattern of the mold 4z is transferred to the imprint material 3z.

[0074] In process SE of Figure 15, etching is performed using the cured material pattern as an etching-resistant mask. This removes the portion of the workpiece 2z surface where there is no cured material or where only a thin layer remains, creating grooves 5z. In process SF of Figure 15, removing the cured material pattern yields an article with grooves 5z formed on the surface of the workpiece 2z. Although the cured material pattern was removed here, it may also be used without removal after processing, for example, as an interlayer insulating film included in semiconductor devices, i.e., as a component of the article.

[0075] (Other embodiments) The present invention can also be realized by supplying a program that implements one or more of the functions of the above-described embodiments to a system or device via a network or storage medium, and by a process in which one or more processors in the computer of that system or device read and execute the program. It can also be realized by a circuit (e.g., an ASIC) that implements one or more functions.

[0076] The disclosures herein include at least the following evaluation apparatus, information processing apparatus, program, film formation system, and article manufacturing method. (Item 1) An evaluation apparatus for evaluating a film on a substrate that has undergone a film formation process in which a film of a composition is formed on a film formation region of the substrate using a mold, An acquisition unit that acquires an image of an evaluation region including a film formation region where a film has been formed by the aforementioned film formation process, A processing unit that processes the aforementioned image for the aforementioned evaluation, It has, The processing unit is configured to output features related to anomalies in the image according to a learned model. The aforementioned image and design information showing the geometric features of the film formation region are input to the learning model. An evaluation device characterized by the following features. (Item 2) The evaluation apparatus according to item 1, characterized in that the learning model is a model obtained by machine learning of the relationship between an image of an evaluation region including a film-forming region and design information of the film-forming region as inputs, and features related to anomalies as outputs. (Item 3) The evaluation device according to item 1, further comprising a machine learning unit that generates the aforementioned learning model by machine learning. (Item 4) The evaluation apparatus according to item 3, characterized in that the machine learning unit performs machine learning on the relationship between an image of an evaluation region including a film formation region and design information of the film formation region as inputs, and features related to anomalies as outputs. (Item 5) The evaluation device according to any one of items 1 to 4, characterized in that the learning model calculates a confidence score representing the reliability of the detected anomaly. (Item 6) The evaluation device according to item 5, characterized in that the features relating to the abnormality include information on the type, location, size, and confidence level of the abnormality in the image. (Item 7) The evaluation apparatus according to item 6, characterized in that the types of abnormalities include the overflow of the composition from the film-forming region and the incomplete filling of the composition in the film-forming region. (Item 8) The evaluation apparatus according to any one of items 1 to 7, characterized in that the design information includes information on the boundary position of the film formation region. (Item 9) The evaluation apparatus according to any one of items 1 to 7, characterized in that the design information includes information on the position and shape of the mark portion in the film formation region. (Item 10) The evaluation apparatus according to any one of items 1 to 9, characterized in that the design information is represented in the form of an image. (Item 11) The evaluation apparatus according to any one of items 1 to 10, characterized in that the film formation process is an imprint process in which the pattern of the mold is transferred to the imprint material by bringing into contact the imprint material, which is the composition supplied on the film formation region. (Item 12) The evaluation apparatus according to any one of items 1 to 10, characterized in that the film formation process is a planarization process in which a planarization film made of the composition is formed on the substrate by bringing the composition supplied onto the film formation region into contact with the flat surface of the mold. (Item 13) A program to cause a computer to function as one of the components of an evaluation apparatus described in any one of items 1 through 12. (Item 14) For a substrate on which a film formation process has been performed to form a film of a composition on a film formation region of the substrate using a mold, an acquisition unit acquires an image of an evaluation region including the film formation region and design information showing the geometric characteristics of the film formation region. A learning unit performs machine learning on the relationship between the aforementioned image and design information and the features related to anomalies in the image, and creates a learning model. An information processing device characterized by having the following features. (Item 15) A program that causes a computer to function as one of the components of the information processing device described in item 14. (Item 16) A film forming apparatus that performs a film forming process by forming a film of a composition on a substrate using a mold, An evaluation device described in any one of items 1 to 12, A film formation system characterized by including the following: (Item 17) The process of forming a film on a substrate using the film forming apparatus in the film forming system described in item 16, A step of processing the substrate on which the film is formed, A method for manufacturing an article, characterized by having a substrate and manufacturing an article from the processed substrate.

[0077] The invention is not limited to the embodiments described above, and various modifications and variations are possible without departing from the spirit and scope of the invention. Accordingly, claims are attached to disclose the scope of the invention. [Explanation of Symbols]

[0078] IMP: Imprint machine, S: Substrate, M: Mold, 110: Control unit, 102: Substrate holder, 105: Substrate drive mechanism, 121: Mold holder, 122: Mold drive mechanism, 407: Evaluation device

Claims

1. An evaluation apparatus for evaluating a film on a substrate that has undergone a film formation process in which a film of a composition is formed on a film formation region of the substrate using a mold, An acquisition unit that acquires an image of an evaluation region including a film formation region where a film has been formed by the aforementioned film formation process, A processing unit that processes the aforementioned image for the aforementioned evaluation, It has, The processing unit is configured to output features related to anomalies in the image according to a learned model. The aforementioned image and design information showing the geometric features of the film formation region are input to the learning model. The aforementioned learning model calculates a confidence score representing the reliability of the detected anomaly. An evaluation device characterized by the following features.

2. The evaluation apparatus according to claim 1, characterized in that the learning model is a model obtained by machine learning of the relationship between an image of an evaluation region including a film formation region and design information of the film formation region as inputs, and features related to anomalies as outputs.

3. The evaluation device according to claim 1, further comprising a machine learning unit that generates the aforementioned learning model by machine learning.

4. The evaluation apparatus according to claim 3, characterized in that the machine learning unit performs machine learning on the relationship between an image of an evaluation region including a film formation region and design information of the film formation region as inputs, and features related to anomalies as outputs.

5. The evaluation apparatus according to claim 1, characterized in that the features relating to the abnormality include information on the type, location, size, and confidence level of the abnormality in the image.

6. The evaluation apparatus according to claim 5, characterized in that the types of abnormalities include the overflow of the composition from the film-forming region and the incomplete filling of the composition in the film-forming region.

7. The evaluation apparatus according to claim 1, characterized in that the design information includes information on the boundary position of the film formation region.

8. The evaluation apparatus according to claim 1, characterized in that the design information includes information on the position and shape of the mark portion in the film formation region.

9. The evaluation apparatus according to claim 1, characterized in that the design information is represented in the form of an image.

10. The evaluation apparatus according to claim 1, characterized in that the film formation process is an imprint process in which the imprint material, which is the composition supplied on the film formation region, is brought into contact with the pattern portion of the mold to transfer the pattern of the mold to the imprint material.

11. The evaluation apparatus according to claim 1, characterized in that the film formation process is a planarization process in which a planarized film made of the composition is formed on the substrate by bringing the composition supplied onto the film formation region into contact with the flat surface of the mold.

12. An evaluation apparatus for evaluating a film on a substrate that has undergone a film formation process in which a film of a composition is formed on a film formation region of the substrate using a mold, An acquisition unit that acquires an image of an evaluation region including a film formation region where a film has been formed by the aforementioned film formation process, A processing unit that processes the aforementioned image for the aforementioned evaluation, It has, The processing unit is configured to output features related to anomalies in the image according to a learned model. The aforementioned features include the abnormal size and shape, The aforementioned image and design information showing the geometric features of the film formation region are input to the learning model. The evaluation apparatus is characterized in that the design information includes information on the boundary position of the film formation region and information on the position and shape of the mark portion in the film formation region.

13. The evaluation apparatus according to claim 12, characterized in that the film formation process is an imprint process in which the pattern of the mold is transferred to the imprint material by bringing into contact the imprint material, which is the composition supplied on the film formation region.

14. The evaluation apparatus according to claim 12, characterized in that the film formation process is a planarization process in which a planarized film made of the composition is formed on the substrate by bringing the composition supplied onto the film formation region into contact with the flat surface of the mold.

15. A program for causing a computer to function as a component of the evaluation apparatus described in any one of claims 1 to 14.

16. For a substrate on which a film formation process has been performed to form a film of a composition on a film formation region of the substrate using a mold, an acquisition unit acquires an image of an evaluation region including the film formation region and design information showing the geometric characteristics of the film formation region. The system includes a learning unit that performs machine learning on the relationship between the aforementioned image and design information and the features related to anomalies in the image, and creates a learning model. The aforementioned features include the size and shape of the abnormality, and the design information includes information on the boundary position of the film formation region, and information on the position and shape of the mark portion in the film formation region.

17. A program for causing a computer to function as each part of the information processing apparatus described in claim 16.

18. A film forming apparatus that performs a film forming process by forming a film of a composition on a substrate using a mold, An evaluation apparatus according to any one of claims 1 to 14, A film formation system characterized by including the following:

19. A step of forming a film on a substrate using the film forming apparatus in the film forming system according to claim 18, A step of processing the substrate on which the film is formed, A method for manufacturing an article, characterized by having a substrate and manufacturing an article from the processed substrate.

Images