Etching shape evaluation system and method
Patent Information
- Authority / Receiving Office
- KR · KR
- Patent Type
- Applications
- Current Assignee / Owner
- HITACHI HIGH TECH CORP
- Filing Date
- 2025-01-24
- Publication Date
- 2026-08-03
Smart Images
Figure PCT00014_ABST
Abstract
Description
Technology Field
[0001] The present invention relates to an etching shape evaluation system and method. Background Technology
[0002] In the semiconductor wafer processing process, known as the front-end process in semiconductor manufacturing, electronic circuits or integrated circuits including transistors are formed on the surface of a silicon wafer. As part of the basic process for forming these integrated circuits, there is a process in which developed photoresist, etc., is used as a mask to process into shapes such as wiring by etching processing using an etching device, etc. These are formed by repeating precise processing based on film deposition, pattern transfer, and etching.
[0003] Improvements in integrated circuit performance have historically been achieved by advancing miniaturization technology to miniaturize components and increase integration density; however, recently, the miniaturization of wiring has reached its limit, necessitating new integration technologies. Consequently, 2.5-dimensional packaging technology and 3-dimensional packaging technology are attracting attention as new integration technologies. Unlike conventional methods of improving integration in planar structures through miniaturization, these technologies improve integration density by adopting three-dimensional structures.
[0004] The etching device used in the wiring processing process is used to form a desired structure by etching (cutting) the stacked film formed on the wafer. It is a device required to form a pattern according to the designed dimensions or shape, and to precisely cut out long, thin holes (deep holes) or grooves (deep grooves) for the wiring included therein, and it is essential to the manufacturing process.
[0005] In etching processes, it is necessary to perform repetitive processing tests while changing and adjusting processing conditions to form a pattern according to the designed dimensions or shape. Since a more precise processing precision is required for the shape after processing, techniques are often used to quantitatively verify the amount of change by measuring the actual cross-sectional shape or dimensions from that cross-sectional shape as an indicator of the processed shape.
[0006] Patent Document 1 discloses a technology that captures a scanning electron microscope (SEM) image of a pattern on a substrate, performs an array pattern detection process to detect a regular arrangement of patterns within an image area, performs template matching to detect the position of the pattern within the image area, integrates the two results of array pattern detection and template matching to detect the position of individual patterns and sets a measurement cursor, and performs a measurement process of the pattern dimensions for each measurement cursor.
[0007] In addition, Patent Document 2 discloses a method for estimating the cross-sectional shape of a sample to be inspected, comprising a fitting process for fitting a plurality of shape models to cross-sectional shape data of a sample to be inspected, and a selection process for selecting at least one shape model from the plurality of shape models as an optimal model based on an error function value which is an indicator of the precision of the fitting model fitted in the fitting process. Prior art literature
[0008] Japanese Patent Publication No. 2015-21942 and Japanese Patent Publication No. 2013-73776 The problem to be solved
[0009] However, when observing semiconductor circuit patterns of deep holes or deep grooves on a Si wafer having three-dimensional structures such as fine cylinders, cones, or N-shaped prisms in a planar state from the top-view of the Si wafer using SEM, it is difficult to estimate the cross-sectional shape formed by the three-dimensional structure of the deep holes or deep grooves. For this reason, a technique in which the processing sample to be etched is cleaved and the cross-section is confirmed using SEM has been conventionally used to confirm the cross-sectional shape.
[0010] At that time, in device manufacturing, it is important to quickly start the manufacturing process and transition to a high-yield mass production system early on, so it is important to reduce the time for observing and evaluating the cross-sectional shape when evaluating the performance of the etching process through repetitive processing tests.
[0011] However, when the micro-pattern is a pattern having a deep hole in the shape of a circle (also called a hole pattern), in order to verify the cross-sectional shape, it is necessary to process the sample by cleaving it vertically (making a slight scratch along the surface with a diamond cutter, etc., and then applying force to split it to form a smooth cleavage plane). At that time, the finer the circular dimensions of the pattern, the more time is required to measure the roundness of each individual one, and it is difficult to manually cleave it while maintaining the positional relationship of the center point and aiming at the same position as the pattern position determined from the top surface using a CD-SEM (Critical Dimension-SEM), etc. Furthermore, it is even more difficult to cleave it in a plane-symmetric state by aiming at the center of the circular pattern, and to verify that it is a cross-section (symmetric plane) corresponding to the true cross-section (i.e., corresponding to the center position of the pattern at all positions from the top (surface) to the bottom (surface).
[0012] Meanwhile, observation and measurement are susceptible to the influence of the cleavage method, and in etching processing tests where processing conditions are repeatedly adjusted to achieve a desired shape, the acquisition of the cross-sectional shape, which serves as an indicator of the processing shape, is prone to affecting the precision of dimensional measurement results or shape evaluation. Furthermore, observation and measurement are also susceptible to the influence of human experience (discretion) or technology (cleavage method, judgment of structure, etc.). Although converged ion beam (FIB) processing can produce high-precision cross-sectional observation samples, it requires the acquisition of advanced processing technology and takes several hours to produce the samples, often requiring even more time depending on the material. The recognition of issues such as precision and efficiency when performing shape evaluation by cleavage and observing the cross-sectional shape of such processing samples is not sufficiently disclosed in Patent Documents 1 and 2.
[0013] Therefore, the present invention aims to provide a technology that enables improved precision or efficiency when evaluating an etched shape by observing a cross-section of a sample for processing. means of solving the problem
[0014] To solve the above problem, one of the representative etching shape evaluation systems of the present invention comprises a computer device, wherein the computer device has a processing unit and a data storage unit, wherein the data storage unit stores a base model image, wherein the processing unit acquires one or more cross-sectional observation images for creation captured by a scanning electron microscope, applies a grid shape template to the cross-sectional observation images for creation, specifies an image (material image) within a section of the grid shape template corresponding to each depth position in which the dimension width of the cross-sectional contour line takes a maximum value, extracts the material image, and creates a montage image by concatenating them, and wherein the data storage unit stores the montage image as a base model image. Effects of the invention
[0015] According to the present invention, the precision or efficiency of evaluating the etching shape by observing the cross-section of a sample for processing can be improved.
[0016] Problems, configurations, effects, etc., other than those mentioned above, are manifested in the form for implementing the invention. Brief explanation of the drawing
[0017] FIG. 1 is a flowchart showing the steps for creating a montage image in the first embodiment. FIG. 2 is a schematic diagram of a sample for processing before and after etching processing of the first embodiment. FIG. 3 is a schematic diagram illustrating several mask pattern shapes and deep hole pattern shapes. FIG. 4 is a diagram showing an example of acquiring multiple cross-sectional observation images (A) to (C). Figure 5 is a schematic diagram showing the state of applying a grid shape template to a cross-sectional observation image. Figure 6 is a schematic diagram showing the state of applying a different grid shape template to a cross-sectional observation image. FIG. 7 is a schematic diagram showing the state of applying grid shape templates A and B of different sizes to a plurality of cross-sectional observation images. FIG. 8 is a schematic diagram showing the state of applying a grid shape template to a cross-sectional observation image and measuring the cross-sectional contour in all sections. FIG. 9 is a schematic diagram illustrating the state of cleavage of a cylindrical deep hole pattern and the cross-sectional shape after cleavage. FIG. 10 is a schematic diagram illustrating the state of cleavage of a conical deep hole pattern and the cross-sectional shape after cleavage. FIG. 11 is a schematic diagram showing the misalignment of the mask pattern shape before and after etching. FIG. 12 is a schematic diagram illustrating the three-dimensional structure of an actual deep hole pattern. FIG. 13 is a schematic diagram showing the state of specifying a grid shape section that takes the maximum value. FIG. 14 is a schematic diagram showing the state of extracting an image within a grid-shaped section that takes the maximum value as a material image. FIG. 15 is a schematic diagram showing the state of synthesizing a montage image by joining cut material image data. FIG. 16 is a flowchart showing a step of performing performance evaluation of etching processing by image identification sequence processing using a base model image in the second embodiment. Figure 17 is a schematic diagram showing the state of specifically extracting a similar cross-sectional pattern image similar to a montage image. Specific details for implementing the invention
[0018] Hereinafter, embodiments of the present invention will be described with reference to the drawings. Furthermore, the present invention is not limited by these embodiments. Also, in the description of the drawings, the same parts are indicated by the same reference numerals.
[0019] [First embodiment]
[0020] (Creation of a montage image)
[0021] In the first embodiment of the present invention, the focus is on creating a montage image used as an indicator for evaluating the etching shape from an image of a cross-section obtained by cleaving a processing sample (also simply called a sample) after etching. FIG. 1 is a flowchart showing the steps for creating a montage image in the first embodiment. Steps 101 to 109 of FIG. 1 will be explained below.
[0022] (Step 101)
[0023] First, an etching test is performed on the sample for processing. FIG. 2 is a schematic diagram of the sample for processing before and after the etching process of the first embodiment. In this embodiment, a scanning electron microscope (SEM) was used as the imaging device for photographing the sample for processing. However, the imaging device may be other than an SEM.
[0024] A sample for processing (201) is composed of at least an etching layer (203) and a mask layer (202) stacked on a Si wafer. As shown in FIG. 2 (a), which is a schematic diagram (pattern top view) of the sample for processing observed from the upper surface of the mask layer, and FIG. 2 (b), which is a schematic diagram observed from a cross-section, the mask layer (202) has a circular and concave mask pattern patterned regularly arranged on its surface. When etching is performed by an etching device using the concave mask pattern, a deep hole pattern having a three-dimensional structure of a cavity, such as a cylinder, a cone, or an N-sided prism, is formed on the etching layer (203) removed by processing, depending on the shape of the mask pattern. FIG. 3 is a schematic diagram illustrating several mask pattern shapes and deep hole pattern shapes. FIG. 3 (a) is a pattern top view of a circular mask pattern and a three-dimensional view of a cylindrical deep hole pattern. FIG. 3(b) shows a top view of a circular mask pattern and a three-dimensional view of a conical deep hole pattern, FIG. 3(c) shows a top view of a triangular mask pattern and a three-dimensional view of a triangular prism-shaped deep hole pattern, and FIG. 3(d) shows a top view of a square mask pattern and a three-dimensional view of a square prism-shaped deep hole pattern.
[0025] Meanwhile, if the mask pattern is convex, the shape after etching is such that the etched layer other than the one immediately below the mask layer is removed, and a three-dimensional structure in the shape of a pillar (filler) is formed. When regularly arranged convex patterns are densely arranged, deep grooves are formed between the patterns. In this embodiment, the case of deep holes formed by a concave mask pattern is addressed, but it is also possible to apply it to cases of deep grooves.
[0026] (Step 102)
[0027] Next, the sample to be processed is cleaved, and an SEM image of the cross-section of the sample to be measured (referred to as the "cross-sectional view image") is acquired (the cross-sectional view image acquired to create a montage image is also referred to as the "cross-sectional view image for creation"). The shape of the deep hole pattern after etching is important for determining the etching shape. At that time, the ideal cleavage position A passing through the center of the circular mask pattern * -B * Ideally, the method involves observing a cross-section obtained by cleaving a sample for processing, i.e., a cross-sectional shape in which the three-dimensional structure of the deep hole pattern is in a plane-symmetric state (see FIG. 2 (a) and (b)). In this embodiment, the three-dimensional structure of the deep hole pattern is at the ideal cleavage position A * -B * The method involved creating minute scratches on the Si wafer surface using a diamond cutter or similar tool along arbitrary locations to include [the element] and ensure maximum inclusion in a plane parallel to the Y direction, and then cleaving the wafer by applying force. This method forms a relatively smooth cleavage plane, enabling cross-sectional observation. However, since the mask pattern is very fine and cleavage is performed almost entirely by hand, for the sample to be processed, an ideal cleavage position A is always... * -B *It is usually difficult to perform cleavage in this way. If cleavage at a more accurate ideal cleavage position is required, there is a method of using a dedicated scribing device equipped with a microscope, but high-precision processing such as the aforementioned converged ion beam (FIB) is difficult, and it is not possible to aim at the center of each circle of the fine circular mask pattern observed from the upper surface.
[0028] Therefore, in practice, the cross-sectional view image of the sample to be processed is observed by cleaving the sample at the actual cleaving position AB. Fig. 2(c) is a schematic diagram of the cross-sectional view image of the sample to be processed after etching. When the sample to be processed is cleaved after etching, the cross-section of the three-dimensional structure (cavity) that combines the three-dimensional structure of the mask pattern of the mask layer (202) and the three-dimensional structure of the deep hole pattern of the etched layer (203) appears by the cross-sectional outline (204). As shown in Fig. 2(c), the cross-sectional outline (204) can take various shapes, and the dimension width (CD: Critical Dimension) (205) at any depth (206) can also take various values. Thus, the shape and dimensions of the cross-sectional outline (204) can be measured, and data such as the difference or deviation from the shape and dimensions of the cross-sectional outline after processing can be obtained.
[0029] Additionally, in the following description, the direction horizontal to the upper surface of the mask layer along the actual cleavage position AB is defined as the X direction, and the direction perpendicular to the upper surface of the mask layer is defined as the Y direction.
[0030] When acquiring a cross-sectional observation image, the observation magnification is such that, with respect to the cross-sectional contour line (204), the depth (206) at any position (Y direction) from the top of the mask layer (202) to the bottom of the etching layer (203) can be measured, and the amount of change (X direction) of the dimension width (205) at each depth position can be quantitatively measured using, for example, a measuring tool on the screen of a computer device's display unit, and the observation area is adjusted so that it is included within the range of a standard image size.
[0031] In this embodiment, 200K magnification was used for the scanning electron microscope (SEM) used for observation, and multiple cross-sectional observation images were acquired from any observation position within the sample. FIG. 4 is a diagram showing an example of acquiring multiple cross-sectional observation images (A) to (C). In addition, the image data may differ in screen size, resolution, etc., depending on the basic performance of the scanning electron microscope (SEM) used or the setting of parameter values of the instrument.
[0032] In addition, when introducing multiple cross-sectional observation image data after the etching process shown in FIG. 4 into a computer device, it is sufficient to confirm that the image data consists of deep hole patterns of the same three-dimensional structure (e.g., cylindrical shape) arranged regularly, and that the image data is a fixed image with constant size, resolution, observation range, and observation magnification.
[0033] (Step 103)
[0034] Next, a grid shape template is applied to the cross-sectional view image. FIG. 5 is a schematic diagram showing the state of applying a grid shape template to a cross-sectional view image. The grid shape template (A501) is composed of grid-shaped sections (referred to as "grid shape sections" or simply "sections"), and by applying the grid shape template to the cross-sectional view image, the shape of the cross-sectional contour line (204) can be measured for each image cut into individual sections (referred to as "images within sections"). The shape of the grid shape sections is not necessarily limited to a square shape, provided that the shape of multiple cross-sectional contour lines (204) can be measured for each image within a section and a montage image can be created.
[0035] Regarding the sections arranged in the horizontal direction of the grid shape template (A501), since the mask pattern of the mask layer (202) is arranged regularly at regular intervals (see FIG. 2 (a)) and the deep hole pattern after etching is also arranged regularly according to the mask pattern, they can be arranged at the same arrangement interval as the mask pattern and the deep hole pattern, and set so that the dimension width of each cross-sectional contour line (204) fits within the size of each section (section X size) in the horizontal direction.
[0036] Regarding the horizontal size of the entire grid shape template (Template X Size), it is not specifically limited, but it should be set appropriately to a size that accommodates multiple cross-sectional contour lines that are measurement targets included in the cross-sectional observation image when applied.
[0037] Meanwhile, regarding the sections arranged in the vertical direction of the grid shape template (A501), it is possible to adjust them appropriately depending on the precision or efficiency of the evaluation of the etched shape. If the size of each section in the vertical direction (section (Y) size) is made finer, more detailed measurement data can be obtained for the cross-sectional contour and the precision of the evaluation is improved, but the time and cost required for the evaluation also increase, so adjustments can be made by considering the balance.
[0038] Regarding the vertical size (template Y size) of the entire grid shape template, it is not specifically limited, but it can be appropriately set to a size that includes multiple cross-sectional contour lines that are the measurement targets included in the cross-sectional observation image when applied. However, when measuring depth in the vertical direction, there may be cases where a reference position for the highest or lowest position that serves as the starting point for depth is specified. For example, the highest position of the starting point for depth may be placed at position C1 on the upper surface of the mask layer (202), or the highest position of the starting point for depth may be placed at position (C2) on the boundary surface between the mask layer (202) and the etching layer (203). In such cases, the template Y size can be adjusted so that the highest position (C2) or the lowest position (D) specified as the reference position is included, as shown in the example of FIG. 5.
[0039] FIG. 6 is a schematic diagram showing the state of applying a different grid shape template to a cross-sectional observation image. The grid shape template (B601) has an increased number of divisions (number of subdivisions) in the Y direction compared to the grid shape template (A501). In this way, the number of divisions in the vertical direction (Y direction) or the size of the divisions (Y) of the grid shape template can be flexibly changed according to the observation magnification, the range of depth at which observation is performed, or the precision of performance evaluation.
[0040] When the size of the grid shape section is subdivided, the number of material images in the montage composite image formed by combining images within each depth position section described later increases, thereby enabling the formation of a more detailed montage composite image. If detailed measurements are desired within a specific depth position range, the size of the grid shape section can be arbitrarily set, such as by making the size of the grid shape section fine only within that range. FIG. 7 is a schematic diagram showing the state of applying grid shape templates A and B of different sizes to a plurality of cross-sectional observation images.
[0041] (Step 104)
[0042] Next, measurements of the cross-sectional contour (204) are performed in all sections of the applied grid shape template (A501). FIG. 8 is a schematic diagram showing the state of applying a grid shape template to a cross-sectional observation image and performing measurements of the cross-sectional contour in all sections. As shown in FIG. 8, all sections of the grid shape template (A501) are specified by assigning numbers in the X direction and the Y direction. In the example of FIG. 8, numbers 1 to 10 corresponding to the hole positions of the circular mask pattern are assigned in the X direction, and numbers a to e are assigned in the Y direction, so that all sections can be specified by section numbers such as 1-a, 1-b, ~10-d, 10-e. Then, numerical data is obtained by measuring the depth (206) of the cross-sectional contour (204) in all sections and the dimension width (205) at the corresponding depth position. Additionally, the location for measuring the depth (206) within each section is not specifically limited and may be the center of gravity of the section.
[0043] However, if the three-dimensional structure of the deep hole pattern is partially significantly deviated from the ideal due to the influence of the cleavage location or the condition of the split cross-section (see within the dotted line frame of FIG. 2), and the cross-sectional contour line (204) disappears or is unclear, the measurement of the dimension width (205) and depth (206) within that section may be omitted (see × mark in FIG. 8).
[0044] When there are multiple acquired cross-sectional observation image data (see FIG. 7), the image data specifications, including the size and magnification of the images, are made identical and introduced into a computer device. In this state, a grid shape template of the same fixed size is applied to them, and numerical data is acquired by measuring the images within each section. Regarding the measurement method, an application equipped with a measurement function applicable to the computer device used may be used, but it is not limited to just one method as long as it is possible to measure a quantitative amount of change. However, it is desirable that the measurement conditions ensure that output values are obtained consistently and stably.
[0045] In one or more cross-sectional observation images, a table sheet is created by entering the measured dimensions (width (205) and depth (206) within all grid shape sections of the applied grid shape template using a table calculation application of a computer device. It is preferable to include the section number (e.g., 1-a) of the measured section along with the measurement data of the dimension width and depth in the table sheet. The measurement of the dimension width and depth can be performed by any means. It may be automatically measured by specifying both ends of the cross-sectional contour at a predetermined depth position using known image recognition technology, or the measurement may be performed by a measurement tool on the screen of the display unit of the computer device.
[0046] It is desirable to record the maximum value of the dimension width among the sections assigned section numbers (a to e) at the same depth position on the schedule sheet. In addition, by calculating and recording the minimum value, average value, or median value in addition to the maximum value, it becomes possible to verify whether the information on the obtained maximum value deviates from the desired design dimensions, which can be beneficial for improving the precision of the montage image described later. These calculations can be easily performed by using calculation functions such as table calculation applications for all obtained measurement results.
[0047] Additionally, the schedule sheet can be edited arbitrarily, and if sections within the grid shape sheet that take maximum or minimum values are highlighted using color coding or the like, it becomes easier to identify the positional relationships of the sections taking maximum values within the grid shape template later.
[0048] (Step 105)
[0049] Next, among the images within the sections of the grid shape template corresponding to each depth position, the value with the maximum dimension width of the cross-sectional contour is selected. To this end, the condition in which a discrepancy occurs between the ideal cleavage position and the actual cleavage position when cleaving the sample for processing is first explained. FIG. 9 is a schematic diagram illustrating the state of cleavage of a cylindrical deep hole pattern and the cross-sectional shape after cleavage. The interior of the deep hole pattern is an etched cavity, and the upper and lower surfaces of the three-dimensional structure of the cylindrical pattern are circular.
[0050] Figure 9 (a) is the case where cleavage is performed along the direction (Pa) passing through the center position A-A' (ideal cleavage position) of the circle, which is perpendicular to the upper plane (see (a-1)), and Qa in the cross section corresponds to the dimension width of the cross section contour (see (a-2)). In this case, Qa represents the maximum value that the dimension width of the cross section contour can take at any depth position in the vertical direction, and this state becomes the state of cleavage at the center of the circle, i.e., the plane of symmetry.
[0051] Figure 9 (b) shows the case where the cleavage is perpendicular to the upper plane and passes through the position B-B', which is slightly offset from the center position A-A' of the circle (see (b-1)), and Qb in the cross section corresponds to the dimension width of the cross section contour (see (b-2)). Since the dimension width (Qb) of the cross section contour decreases compared to the maximum value (Qa), it can be estimated that the cleavage position is in a state where the cleavage is performed on an asymmetric plane offset from the center position A-A'.
[0052] Likewise, (c) of FIG. 9 is a case where cleavage is performed along the direction Pc passing through position C-C', which is perpendicular to the upper plane and significantly offset from the center position A-A' of the circle (see (c-1)), and Qc in the cross section corresponds to the dimension width of the cross section contour (see (c-2)). Since the dimension width (Qc) of the cross section contour is significantly reduced compared to the maximum value (Qa), it can be estimated that the cleavage position is a state of cleavage in an asymmetric plane significantly offset from the center position A-A'.
[0053] Next, regarding the case where the three-dimensional structure of the deep hole pattern after etching is conical (see FIG. 3(b)), the state when the cleavage position deviates from the above position is described. In this case, the dimensional width of the cross-sectional contour of the deep hole pattern decreases at each depth position from the top surface to the bottom surface. FIG. 10 is a schematic diagram illustrating the state of cleavage of the conical deep hole pattern and the cross-sectional shape after cleavage. The interior of the deep hole pattern is an etched cavity, and the top surface of the three-dimensional structure of the conical pattern is circular.
[0054] Figure 10 (a) is a case where the cross is cleaved along the direction (Pa) passing through the center position A-A' (ideal cleavage position) of the circle, which is perpendicular to the upper plane (see (a-1)), and Qa in the cross section corresponds to the dimension width of the cross section contour (see (a-2)). In this case, Qa is the maximum value that the dimension width of the cross section contour can take at any depth position in the vertical direction, and the depth (Ra) in the vertical direction of the cross section contour is also the maximum value at any position in the horizontal direction, and this state is a state where it is cleaved at the center of the circle, that is, the plane of symmetry.
[0055] Figure 10 (b) is a case where the cleavage is perpendicular to the upper plane and passes through the position B-B' which is offset from the center position A-A' of the circle (see (b-1)), and Qb in the cross section corresponds to the dimension width of the cross section contour (see (b-2)). In addition to the dimension width (Qb) of the cross section contour decreasing compared to the maximum value (Qa), the vertical depth (Rb) also decreases compared to the maximum value (Ra), so it can be estimated that the cleavage position is a state where the cleavage is performed on an asymmetric plane offset from the center position A-A'.
[0056] In addition, the case in which the three-dimensional structure of the deep hole pattern undergoes deformation or bending in the vertical direction is described. For the regularly arranged circular mask pattern shown in FIG. 2 (a) or the N-shaped mask pattern shown in FIG. 3, as the design dimensions become finer, the dimensions of the pattern after etching process differ from the design dimensions. FIG. 11 is a schematic diagram showing the deviation of the mask pattern shape before and after etching process. Ideally, the design circular mask pattern (1101) and the actual mask patterns (1102 to 1104) observed from the top surface after the deep hole pattern after etching process would match in the same shape and dimensions, but in reality, as shown in FIG. 11, there are minute differences, and it is assumed that various deviations exist, including parts that exceed or fall below the design dimensions of the mask pattern (1101).
[0057] In addition, in the three-dimensional structure of a deep hole pattern formed by etching a circular mask pattern in a vertical direction, there is a discrepancy between the ideal shape and the actual shape. FIG. 12 is a schematic diagram illustrating the three-dimensional structure of an actual deep hole pattern. FIG. 12 (a) shows the three-dimensional structure of a deep hole pattern with an ideal shape, and also shows the upper surface (1201) and lower surface (1202) (both circular) of the deep hole pattern observed from the upper surface. In contrast, FIG. 12 (b) shows the three-dimensional structure and lower surface (1203) when the hole axis of the deep hole pattern is bent, FIG. 12 (c) shows the three-dimensional structure and lower surface (1204) when the dimension width of the deep hole pattern is reduced, FIG. 12 (d) shows the three-dimensional structure and lower surface (1205) when the deep hole pattern has a tilt and insufficient etching process, and FIG. 12 (e) shows the three-dimensional structure and lower surface (1206) when the deep hole pattern is tilted and deviates from the center line. As such, the three-dimensional structure of the actual deep hole pattern exhibits various deviations from the ideal shape.
[0058] In this way, if the three-dimensional structure of the deep hole pattern has been deformed or bent in the vertical direction, it can be assumed that a state similar to the cleavage of the conical three-dimensional structure shown in FIG. 10 will occur. Even if cleavage occurs along the direction (Pa) passing through the center position A-A' of the circle, it is expected that the shape of the cross-sectional contour will partially exhibit the dimension width of Qb or the depth of Rb due to the influence of the misalignment of the shape state in the depth direction. In this case as well, the dimension width at each depth position will decrease from the maximum value that is plane-symmetric, and it is assumed that the depth will also show a reduced measurement value.
[0059] As described above, when comparing the dimensional widths at each depth position of the cross-sectional contour lines appearing in the cross-sectional observation image obtained by cleaving the three-dimensional structure of the deep hole pattern, a larger value can be presumed to be closer to the cleavage state in the plane of symmetry (diagnostic plane) passing through the center of the mask pattern; therefore, when estimating the cleavage state in the plane of symmetry, it is desirable to select the image with the maximum value. Furthermore, the greater the depth of the cross-sectional contour line in the vertical direction, the more likely it is that the processing state is one in which the bending or deformation of the deep hole pattern is minimal.
[0060] (Step 106)
[0061] Next, the data of the selected maximum value is compared with the position information of the grid shape section of the grid shape template, and the grid shape section that takes the maximum value is identified. FIG. 13 is a schematic diagram showing the state of identifying the grid shape section that takes the maximum value. As described in Step 104, the grid shape template (A501) is applied to the cross-sectional view image (A), and data of the dimension width (205) of the cross-sectional contour line (204) in all grid shape sections (1-a to 10-e) is acquired (Fig. 13 (a)). However, measurement data for sections where the cross-sectional contour line is not clear (see × mark in FIG. 13) does not need to be acquired.
[0062] The comparison between the data of the maximum value and the position information of the grid shape section is specified by section numbers such as 1-a, 8-b, 3-c, 8-d, and 2-e in order of depth, as shown in the example in FIG. 13 (b). At this time, processing that makes selection easier may be performed, such as emphasizing the display within the section by using thick lines or changing colors to make it easier to select the image of the maximum value, or erasing or hiding images within other sections that do not correspond.
[0063] In addition, as shown in FIG. 13, when there is only one cross-sectional observation image, the section with the maximum value is identified from within that image. However, as shown in FIG. 7, when there are multiple image data, measurements are performed in all sections of all image data, and the maximum value is selected for each depth position among all images and all sections, and the section with the maximum value can be identified. However, the precision of the measurement data (dimension width or depth) may vary depending on the amount of information represented by the image data (resolution, pattern size, and observation magnification) or the size of the variable section size.
[0064] (Step 107)
[0065] Next, an image within a grid-shaped section that takes the maximum value is extracted as a material image. FIG. 14 is a schematic diagram showing the state of extracting an image within a grid-shaped section that takes the maximum value as a material image. From the position of a grid-shaped section that takes a specific maximum value in Step 106 (Fig. 14 (a)), an extraction method is adopted in which image data is cut in a fragment while maintaining the size of the section, as shown in FIG. 14 (b) (the image cut in a fragment is referred to as a "material image"). In addition, if the same maximum value exists, one can be selected by comparing, for example, the consistency with the design of the film composition or the boundary line of the laminated film, the clarity of the cross-sectional contour, and the presence or absence of deformation or foreign matter attachment.
[0066] (Step 108)
[0067] Next, the cut material image data is joined together to synthesize a montage image. FIG. 15 is a schematic diagram showing the state of joining the cut material image data to synthesize a montage image. Each material image (1-a, 8-b, 3-c, 8-d, 2-e) (Fig. 15 (a)) extracted in Step 107 is joined together to synthesize a montage image (1501) (Fig. 15 (b)). At this time, it is desirable to join the material images while making fine adjustments while maintaining an intermediate position where continuity is obtained as much as possible so that the cross-sectional contour line of the montage image does not cause a large misalignment at the seam of the adjacent material image. In addition, by using a dimension that narrows the section (Y) size of the applied grid shape template to make the seam between the upper and lower images smoother when synthesizing the montage image, it is possible to form a montage composite image with higher precision. However, if the number of vertical sections (divisions) of the grid shape template is increased and the section (Y) size is narrowed excessively, in addition to a significant increase in measurement data, it is assumed that a large amount of time will be consumed in the selection and extraction of material images and the montage image synthesis work of stitching them together; therefore, it becomes important to strike a balance with precision.
[0068] (Step 109)
[0069] Since the montage image created in this manner is an image within a section where the material image takes the maximum dimensional width at each depth position, the entire montage image can be estimated as the ideal cross-sectional contour line in the symmetry plane (diagnostic plane) separated from the center position of the mask pattern shape. Therefore, the montage image can be used as an image serving as an indicator for evaluating the etching shape (an image serving as an evaluation indicator is called a "base model image"). Thus, the montage image is stored in a data storage unit as a base model image serving as an indicator for evaluating the performance of the etching process. It is subsequently used as an indicator for evaluating the performance of the etching process.
[0070] (Hardware Configuration)
[0071] The present embodiment can be constructed as a system having a computer device as a hardware configuration, comprising a processing unit, memory, input I / F, instruction input unit, data storage unit, communication I / F, output I / F, and display unit.
[0072] The processing unit is a processor (CPU) and executes instructions according to a program stored in memory. Memory is a medium for storing data and programs, and may include random access semiconductor memory, a storage device, or a storage medium (either volatile or non-volatile). The input I / F is an interface connected to an instruction input unit, such as a keyboard or mouse, for inputting user instructions. The data storage unit stores data such as base model images including montage images or grid shape templates, and is composed of known storage such as a hard disk drive (HDD) or a solid-state drive (SSD). The communication I / F is an interface connected to a scanning electron microscope (SEM) or an etching device for transmitting and receiving data with other devices, such as captured images or etching conditions. The output I / F is an interface connected to a display unit, such as a display, for outputting various display information. The computer device may be any electronic device, such as a desktop computer or a laptop computer.
[0073] In this embodiment, for example, Step 103 (application of grid shape template), Step 104 (measurement of cross-sectional contour), Step 105 (selection of maximum value), Step 106 (specification of section taking maximum value), Step 107 (extraction of section taking maximum value), and Step 108 (creation of montage image) can be automatically processed by the processing unit of the computer device.
[0074] As such, the montage image can be estimated as an ideal cross-sectional contour and serves as an indicator that contributes to improving the precision of shape evaluation. In addition, there is the advantage of being able to be efficiently generated by a computer device based on cross-sectional observation images.
[0075] [Second embodiment]
[0076] (Performance evaluation of etching process)
[0077] The second embodiment of the present invention performs a performance evaluation of an etching process using a base model image including a montage image created in the first embodiment. FIG. 16 is a flowchart illustrating the steps for performing a performance evaluation of an etching process by image identification sequence processing using the base model image in the second embodiment. The following description follows steps 1601 to 1609 of FIG. 16. Since the hardware configuration is the same as that of the first embodiment, the description will focus on the differences.
[0078] (Step 1601)
[0079] An etching processing test is performed on a sample for processing under etching conditions different from those of the first embodiment. However, the sample for processing used is the same as the one used when creating the montage image in the first embodiment, and has the same mask pattern arrangement. However, it is acceptable to perform the processing test under the same etching conditions as the first embodiment.
[0080] (Step 1602)
[0081] As in the first embodiment, a cross-sectional observation image to be measured is acquired (the cross-sectional observation image acquired to evaluate the etching shape is also referred to as an "evaluation cross-sectional observation image"). The observation environment, such as the observation range and observation magnification, is assumed to be the same as that of the first embodiment.
[0082] (Step 1603)
[0083] A cross-sectional observation image consisting of any number of images acquired in Step 1602 is stored in the data memory of a computer device.
[0084] (Step 1604)
[0085] Image identification sequence processing is initiated in the processing unit of a computer device.
[0086] (Step 1605)
[0087] First, the montage image created in the first embodiment and saved as a base model image (step 109) is read from the data storage unit. Likewise, the cross-sectional view image saved in step 1603 is also read.
[0088] (Step 1606)
[0089] Next, the similarity between the base model image and the cross-sectional observation image is determined. The technique for determining similarity is not specifically limited, and it is also possible to employ image identification technology based on AI (artificial intelligence).
[0090] If it is determined that there is no similarity, that is, that an image of a cross-sectional contour with equivalent features cannot be detected, return to Step 1602 and perform cleavage again to acquire a cross-sectional observation image. Alternatively, return to Step 1601 and perform a new processing test again. The following factors may be considered as reasons for determining that there is no similarity.
[0091] (1) The desired design dimensions could not be achieved under etching conditions.
[0092] (2) A cross-sectional view was taken near the center position (plane of symmetry) of the mask pattern, but a photograph was taken that was significantly off-center or did not capture a deep hole pattern.
[0093] (3) The cross-section area was contaminated or damaged, making the cross-section outline unclear.
[0094] (4) The deep hole was blocked by sediment, etc.
[0095] When it is determined that an image of a cross-sectional contour line with similarity, that is, equivalent features, has been detected, an image consisting of a section containing the similar cross-sectional contour line (referred to as a "similar cross-sectional pattern image") is identified and extracted along with its location. FIG. 17 is a schematic diagram showing the state of identifying and extracting a similar cross-sectional pattern image similar to a montage image. There may be multiple similar cross-sectional pattern images to be extracted.
[0096] (Step 1608)
[0097] For the similar cross-sectional pattern image extracted in Step 1607, the dimension width and depth of the cross-sectional contour in all grid shape sections are measured, just as in Step 104 of the first embodiment.
[0098] (Step 1609)
[0099] The pseudo-cross-sectional pattern image can be used as a new base model image because, like the montage image, it can be estimated as an ideal cross-sectional contour in the plane of symmetry (diagnostic plane) cleaved from the center position of the mask pattern shape. Therefore, the pseudo-cross-sectional pattern image is stored in the data memory along with the measurement data acquired in Step 1608. The stored pseudo-cross-sectional pattern image is then read as a base model image during the subsequent image identification sequence processing (Step 1605).
[0100] In this way, by using image identification technology utilizing AI, etc., to determine similarity for cross-sectional observation images obtained from etching processing tests using montage images as base model images and performing performance evaluation of etching processing, it is possible to realize efficient performance evaluation in a short time and reduce the influence of differences in experience values depending on the person.
[0101] In addition, since similar cross-sectional pattern images determined to have similarity to montage images or existing base model images can be used as new base model images to update AI training data, an improvement in the precision of similarity determination in image identification sequence processing using AI can also be expected.
[0102] Although embodiments of the present invention have been described above, the present invention is not limited to the embodiments described above, and various modifications are possible within the scope that does not deviate from the gist of the present invention.
[0103] The embodiments that may constitute the content of the present invention are described below, but are not limited thereto.
[0104] (Sun 1)
[0105] As an etching shape evaluation system equipped with a computer device,
[0106] The above computer device has a processing unit and a data storage unit, and
[0107] The above data memory unit stores a base model image, and
[0108] The above processing unit is,
[0109] Acquire one or more cross-sectional observation images for creation captured by a scanning electron microscope, and
[0110] Apply a grid shape template to the cross-sectional observation image for the above creation, and
[0111] Among the sections of the grid shape template corresponding to each depth position, specify an image (material image) within the section where the dimension width of the cross-sectional contour takes the maximum value, and
[0112] Extract the above material images, stitch them together to create a montage image, and
[0113] The above data storage unit preserves the above montage image as a base model image,
[0114] Etching shape evaluation system.
[0115] (Sun 2)
[0116] As an etching shape evaluation system described in Solar 1,
[0117] An etching shape evaluation system in which the vertical size of the section of the above grid shape template can be changed according to the precision of the performance evaluation.
[0118] (Sun 3)
[0119] As an etching shape evaluation system described in Solar 1 or 2,
[0120] An etching shape evaluation system in which a mask layer of a processing sample to be evaluated for etching shape has a circular mask pattern formed thereon.
[0121] (Sun 4)
[0122] As an etching shape evaluation system described in any one of Sun 1 to 3,
[0123] The above processing unit is,
[0124] Acquire one or more cross-sectional observation images for evaluation captured by a scanning electron microscope, and
[0125] An etching shape evaluation system that determines the similarity between a base model image and a cross-sectional observation image for evaluation.
[0126] (Sun 5)
[0127] As an etching shape evaluation system described in Solar 4,
[0128] An etching shape evaluation system comprising a base model image including a similar cross-sectional pattern image extracted from an evaluation cross-sectional observation image that is determined to have similarity in the above determination.
[0129] (Sun 6)
[0130] As an etching shape evaluation system described in Solar 4 or 5,
[0131] An etching shape evaluation system characterized by the above determination being performed by image identification using artificial intelligence (AI).
[0132] (Sun 7)
[0133] As a method for evaluating etching shape,
[0134] In the processing unit of a computer device,
[0135] Acquire one or more cross-sectional observation images for creation captured by a scanning electron microscope, and
[0136] Apply a grid shape template to the cross-sectional observation image for the above creation, and
[0137] Among the sections of the grid shape template corresponding to each depth position, specify an image (material image) within the section where the dimension width of the cross-sectional contour takes the maximum value, and
[0138] Extract the above material images, stitch them together to create a montage image, and
[0139] In the data storage unit of the above computer device, the montage image is stored as a base model image.
[0140] Etching shape evaluation method.
[0141] (Sun 8)
[0142] As an etching shape evaluation method described in Sun 7,
[0143] In the above processing unit,
[0144] Acquire one or more cross-sectional observation images for evaluation captured by a scanning electron microscope, and
[0145] An etching shape evaluation method for determining the similarity between a base model image and a cross-sectional observation image for evaluation. Explanation of the symbols
[0146] 201… Sample for processing 202… mask layer 203… Etched layer 204… Sectional contour 205… Dimension Width (CD) 206… depth 501… Grid Shape Template A 601… Grid Shape Template B 1101~1104… Mask pattern 1201… Top surface 1202~1206… if 1501… Montage image
Claims
Claim 1 An etching shape evaluation system equipped with a computer device, wherein the computer device has a processing unit and a data storage unit, wherein the data storage unit stores a base model image, the processing unit acquires one or more cross-sectional observation images for creation captured by a scanning electron microscope, applies a grid shape template to the cross-sectional observation images for creation, identifies a material image within a section of the grid shape template corresponding to each depth position in which the dimension width of the cross-sectional contour line takes a maximum value, extracts the material image, and concatenates it to create a montage image, and the data storage unit stores the montage image as the base model image. Claim 2 An etching shape evaluation system according to claim 1, wherein the size in the vertical direction of the section of the grid shape template is changeable according to the precision of the etching shape evaluation. Claim 3 In claim 1, the etching shape to be evaluated is an etching shape evaluation system in which the etching shape is a shape etched using a hole mask pattern. Claim 4 An etching shape evaluation system according to claim 1, wherein the processing unit acquires one or more evaluation cross-sectional observation images captured by a scanning electron microscope and determines the similarity between a base model image and the evaluation cross-sectional observation images. Claim 5 An etching shape evaluation system according to claim 4, wherein the base model image includes a similar cross-sectional pattern image extracted from an evaluation cross-sectional observation image that is determined to have similarity in the above determination. Claim 6 An etching shape evaluation system according to claim 4, characterized in that the above determination is performed by image identification using artificial intelligence (AI). Claim 7 A method for evaluating an etching shape, wherein, in a processing unit of a computer device, one or more cross-sectional observation images for creation captured by a scanning electron microscope are acquired, a grid shape template is applied to the cross-sectional observation images for creation, a material image is identified as an image within a section of the grid shape template corresponding to each depth position in which the dimension width of the cross-sectional contour line takes a maximum value, the material image is extracted and joined together to create a montage image, and the montage image is stored as a base model image in a data storage unit of the computer device. Claim 8 An etching shape evaluation method according to claim 7, wherein the processing unit acquires one or more evaluation cross-sectional observation images captured by a scanning electron microscope and determines the similarity between a base model image and the evaluation cross-sectional observation images.