Pattern inspection / measuring apparatus, pattern inspection / measuring program product

Abstract

An object of the present disclosure is to provide a pattern inspection / measurement apparatus capable of accurately determining a corner point formed on a sample. A pattern inspection / measurement apparatus according to the present disclosure determines a pair of corner points on design data as a corner pair candidate, and determines a corner point on a shape pattern actually formed based on a relative relationship between the corner pair candidate on the design data and the corner pair candidate in the shape pattern (see FIG. 13).

Description

Technical Field

[0001] This disclosure relates to a pattern inspection / measuring apparatus for inspecting or measuring shapes and patterns formed on a sample. Background Technology

[0002] In semiconductor manufacturing processes, circuit patterns on semiconductor integrated circuits are typically inspected and measured. One method for inspecting and measuring circuit patterns is to use CAD (Computer-Aided Design) data as a reference pattern. In this method, the circuit pattern is evaluated by comparing an image obtained from photographing the actual formed circuit pattern (e.g., an SEM image taken with a scanning electron microscope) with the CAD data.

[0003] In semiconductor manufacturing processes, the circuit patterns formed on a wafer will not be the same as the designed shape, and shape measurement is performed to evaluate its completion.

[0004] When evaluating circuit patterns, the distance between patterns is sometimes measured. There are methods for measuring the distance between patterns that utilize design data and methods that do not. Patent Document 1 describes a method for measuring the distance between patterns that does not utilize design data.

[0005] Existing technical documents

[0006] Patent documents

[0007] Patent Document 1: JP Japanese Patent Application Publication No. 11-201919 Summary of the Invention

[0008] The problem that the invention aims to solve

[0009] In recent years, the miniaturization of semiconductor devices has led to a narrowing of the likelihood of pattern variations in circuit designs. Therefore, it is necessary to measure not only lines but also corners and distances between adjacent patterns. This is because these parameters are highly likely to deviate from design data due to the miniaturization of manufacturing processes. For example, the distance between diagonal corners of a single closed-region pattern and the distance between opposite corners of two closed-region patterns are relatively likely to deviate from design data.

[0010] Corner points are designed to be sharp (e.g., right angles) in the design data, but depending on the manufacturing process, the actual corner points may not be clearly identifiable; for example, they may sometimes be chamfered. Due to this nature of corner points, identifying the corner points themselves or the distance between corner points is difficult using existing evaluation methods such as Patent Document 1.

[0011] This disclosure is made in view of the aforementioned issues, and aims to provide a pattern inspection / measurement device capable of accurately determining corner points formed on a sample.

[0012] Methods for solving problems

[0013] The pattern inspection / measuring device disclosed herein determines a pair of corner points in the design data as corner pair candidates, and determines the corner points on the shape pattern according to the relative relationship between the corner pair candidates in the design data and the corner pair candidates in the actual formed shape pattern.

[0014] The effects of the invention

[0015] According to the pattern inspection / measuring apparatus disclosed herein, corner points formed on a sample can be determined with good accuracy. Other structures, issues, advantages, etc., will become clear from the following description of embodiments. Attached Figure Description

[0016] Figure 1 This is a structural diagram of the pattern inspection / measurement system 100.

[0017] Figure 2 This is a flowchart illustrating the process by which the pattern inspection / measurement system 100 determines corner pairs on the sample 107.

[0018] Figure 3 This is a diagram illustrating the additional information assigned to the design data in S205.

[0019] Figure 4 This is a flowchart illustrating the details of S207.

[0020] Figure 5 This is a diagram illustrating a specific example of one of the additional information assigned in S401.

[0021] Figure 6A Examples representing minute step differences.

[0022] Figure 6B An example representing a dense corner group 602.

[0023] Figure 7 An example representing polar coordinates.

[0024] Figure 8A The example shown in S402 is used to identify a corner pair candidate.

[0025] Figure 8B The example shown in S402 is used to identify a corner pair candidate.

[0026] Figure 8C The example shown in S402 is used to identify a corner pair candidate.

[0027] Figure 8D The example shown in S402 is used to identify a corner pair candidate.

[0028] Figure 8E It is a diagram that selects the combination of the orientations of the line segments surrounding the candidate corner points to form a corner.

[0029] Figure 9A This shows an example of narrowing down the candidate range for corner pairs in S403.

[0030] Figure 9B This shows an example of narrowing down the candidate range for corner pairs in S403.

[0031] Figure 9C This shows an example of narrowing down the candidate range for corner pairs in S403.

[0032] Figure 9D This shows an example of narrowing down the candidate range for corner pairs in S403.

[0033] Figure 9E Indicates to Figures 9A to 9D The results of the sorting.

[0034] Figure 10 This is a diagram representing a specific example of S404.

[0035] Figure 11 This is a flowchart illustrating the details of S208.

[0036] Figure 12 This is a diagram representing a specific example of S210.

[0037] Figure 13 This is an example of the graphical user interface (GUI) provided by computer system 116. Detailed Implementation

[0038] <Implementation Method 1>

[0039] Figure 1 This is a structural diagram of the pattern inspection / measurement system 100. The pattern inspection / measurement system 100 is a system for inspecting / measuring shape patterns formed on semiconductor samples. The pattern inspection / measurement system 100 includes a scanning electron microscope (SEM) 101, computer systems 111 and 116, a design information database 120, and input / output devices 121. They are interconnected via a network 115.

[0040] SEM 101 images the shape and pattern formed on a sample by irradiating it with an electron beam. SEM 101 includes an electron beam column 102, a vacuum sample chamber 105, and an XY stage 106. During pattern inspection, an electron beam 104 is irradiated onto a sample 107, such as a wafer on which a device has been fabricated, from an electron source 103. A multi-stage lens is used to converge the irradiated electron beam, and a scanning deflector is used for deflection scanning. Thus, the electron beam 104 scans the surface of the sample 107 in one or two dimensions. Electrons 108 (secondary electrons or backscattered electrons) emitted from the sample by the scanning of the electron beam 104 are detected by a detector and converted into digital signals by an A / D converter 109. The digital signals are input to a computer system 111 via a network 110 and stored in a storage unit 114.

[0041] Computer system 111 is connected to SEM 101 via network 110. Computer system 111 controls various modules of the apparatus, such as electron beam column 102, vacuum sample chamber 105, and XY stage 106, via control unit 113. During pattern inspection, processing unit 112 performs signal processing and image processing using digital signals stored in storage unit 114 to obtain the information to be measured (e.g., SEM image).

[0042] The processing unit 112 generates a program (recipe) to control the operation of the SEM 101 based on the design data of the semiconductor device. That is, the processing unit 112 also functions as a recipe setting unit for the SEM 101. Specifically, the processing unit 112 sets position information (such as design data, pattern outline data, desired measurement points on the simulated design data, autofocus points, auto-astigmatism correction points, addressing points, etc.) for the SEM 101 to perform the required processing, and controls the XY stage 106, deflector, etc. based on this setting.

[0043] Computer system 116 processes the measurement results (SEM images, etc.) obtained by computer system 111. Specifically, the arithmetic processing unit 117 uses the measurement results to perform processing for inspecting the shape and pattern on sample 107 (e.g., measuring the distance between corner points, as described later). Image processing unit 118 performs image processing accompanying the pattern inspection. Storage unit 119 stores the processing results.

[0044] Design information database 120 stores design information (hereinafter referred to as design data) of semiconductor circuit patterns formed on sample 107. The design data describes the shape, coordinates, etc. of the circuit patterns formed on each layer of the stacked circuit pattern. Design information database 120 can be constructed by storing the design data in a storage device.

[0045] Input / output device 121 is an operation terminal used to perform various operations of SEM 101, primarily including formula execution and formula creation operations of computer system 111. Input / output device 121 is connected to computer system 116 to instruct measurement processing and statistical processing, etc. Input / output device 121 can further access design information database 120 to compare, obtain, save, and create design data.

[0046] In the following description, the vertex where the angle between two line segments in the design data is a right angle is designated as a corner point. A corner pair refers to a combination of two corner points. The computer system 116 determines the corner points and calculates the distance between them according to the process described later, thereby inspecting the sample 107.

[0047] Figure 2 This is a flowchart illustrating the process by which the pattern inspection / measurement system 100 determines corner pairs on sample 107. The following description... Figure 2 Each step.

[0048] ( Figure 2 Step S201)

[0049] SEM101 captures an SEM image of sample 107 according to the formula created by the formula creation function of the processing unit 112. The control unit 113 stores the SEM image in the storage unit 114 and saves accompanying information such as imaging conditions in the storage unit 114.

[0050] ( Figure 2 (Steps S202-S204)

[0051] Computer system 116 obtains the SEM image and accompanying information stored in storage unit 114 of computer system 111 via the network (S202). Computer system 116 obtains the design data corresponding to the SEM image from design information database 120 (S203), and reads in the SEM image and design data (S204).

[0052] ( Figure 2 Step S205)

[0053] Computer system 116 sets additional information on the read-in design data. This additional information is assigned to distinguish whether the shape or pattern formed on the design data is a convex or concave pattern (described later). Figure 3 Let's illustrate with a specific example.

[0054] ( Figure 2 Step S206)

[0055] Computer system 116 performs alignment processing between the imported SEM image and the design data. Known methods such as template matching and pattern matching utilizing normalized correlation can be used for alignment.

[0056] ( Figure 2 Step S207)

[0057] Computer system 116 uses the alignment design data implemented in S206 with the SEM image to automatically obtain corner pairs set as candidate measurement positions. Details of this step are as follows: Figure 4 To illustrate.

[0058] ( Figure 2 Step S208)

[0059] Computer system 116 uses a SEM image obtained by photographing a circuit pattern intended for inspection to extract the outline of the circuit pattern. Figure 11 Let me explain the details of this step.

[0060] ( Figure 2 Step S209)

[0061] Computer system 116 uses the corner pairs candidate data obtained in S207 and the contour lines extracted in S208 to determine the points (or regions) on the contour lines corresponding to each corner point that forms the corner pairs candidate. Computer system 116 uses the determined positions as the estimated positions of the corner points on the actual shape pattern.

[0062] ( Figure 2 Step S209: Supplement)

[0063] In this step, as a method to estimate the position of corner points from the contour lines, the following methods can be used: estimating the position of the corner point by searching for a corresponding point in any direction starting from the corner point of the object; or, taking the intersection of a straight line with an arbitrary angle passing through the corner point of the object and the contour line as the corresponding point and setting it as the estimated position of the corner point. However, in this method, there is a possibility that the position of the corner point may not be estimated or may be falsely detected. Therefore, it is also possible to mitigate the possibility of false detection by creating a distance transformation image of the obtained contour lines and adding the weights of the gradient direction of the distance transformation image to the search direction. A distance transformation image is an image based on contour line information, where the distance to the nearest adjacent contour line is set as a brightness value, and the brightness value decreases as one approaches the contour line.

[0064] ( Figure 2 Step S210)

[0065] Computer system 116 searches for combinations of points with the shortest inter-corner distance around each presumed corner location, obtaining the combination with the shortest distance. The shortest distance obtained in this step is set as the inter-corner distance. An example of this step uses... Figure 12 To illustrate.

[0066] Figure 3 This is a schematic diagram illustrating the additional information assigned to the design data in S205. Three-dimensional shape patterns 301 and 302 are formed on sample 107. They form rectangular closed regions in a plane. Three-dimensional shape pattern 301 is a convex pattern, and three-dimensional shape pattern 302 is a concave pattern. Therefore, when sample 107 is cut through inspection surface 303, the closed region of three-dimensional shape pattern 301 is filled with material, while the closed region of three-dimensional shape pattern 302 is not filled.

[0067] In S205, to distinguish Figure 3 The illustrated convex and concave patterns are given additional information. Since the design data describes two-dimensional shapes, effort is required to characterize three-dimensional attributes. Therefore, the focus is on the line segments forming the closed regions. Specifically, in the convex pattern, it is assumed that the line segments rotate to the left, and in the concave pattern, it is assumed that they rotate to the right. In S205, the convex or concave pattern is distinguished by assigning this rotation direction attribute to each shape pattern in the design data for convenience.

[0068] Figure 4 This is a flowchart illustrating the details of S207. The following explanation... Figure 4 Each step.

[0069] ( Figure 4 Step S401)

[0070] The computer system 116 assigns the following additional information to each corner point in the design data.

[0071] ( Figure 4 Step S401: Additional Information (One of the steps)

[0072] Computer system 116 will use the corner points in the design data as a basis Figure 3 The rotation directions described are categorized into 8 types. (Use...) Figure 5 This will illustrate a specific example of the additional information. This additional information is provided for convenience in subsequent processing, acknowledging its necessity in computational operations.

[0073] ( Figure 4 Step S401: Additional Information 2)

[0074] In the actual formed shape pattern, there are cases where the formation is a tiny step difference that is not intended to form a corner point; for example, an OPC (Optical Proximity Correction) pattern is equivalent to this. The computer system 116 assigns attributes representing properties other than those of the object being inspected to corner points formed through such tiny step differences. Figure 6A To illustrate this attribute with a specific example.

[0075] ( Figure 4 Step S401: Additional Information 3)

[0076] If the step difference is not as small as that of the OPC pattern, but the distances between adjacent corners are dense, there is a possibility that the number of corner pairs for measuring the distance between corner points will increase dramatically. Therefore, the computer system 116 assigns a dense attribute to such dense corner groups. Figure 6B To illustrate a specific example of this attribute, examples of how to use this attribute will be explained in Implementation Method 2.

[0077] ( Figure 4 Step S401: Additional Information 4)

[0078] Computer system 116 assigns a normal attribute to corner points that do not belong to either object external attributes or dense attributes. A normal attribute means that the corner point is the corner point of an object that should be used to measure the distance between corner points.

[0079] ( Figure 4 Step S401: Additional Information Five)

[0080] Computer system 116 calculates polar coordinates for each corner point, with the top left corner of the design data as the origin, and assigns these polar coordinates as the position information for each corner. Figure 7 Here's an example illustrating polar coordinates.

[0081] ( Figure 4 Step S402)

[0082] Computer system 116 uses the additional information set in S301 to determine candidates for measuring the distance between corner points (hereinafter referred to as corner pair candidates). A corner pair candidate is either (a) two corner points diagonally arranged in one closed region, or (b) two corner points opposite each other in two closed regions. A specific example of this step uses... Figures 8A to 8E Please provide an explanation.

[0083] ( Figure 4 Step S402: Supplement)

[0084] Computer system 116 excludes corner points with external attributes from corner pair candidates. Furthermore, since computer system 116 only sets the two diagonally arranged corner points in (a) above as corner pair candidates, it excludes two adjacent corner points from corner pair candidates. Moreover, computer system 116, for example, treats repeated corner point pairs such as corner point 1: corner point 2 and corner point 2: corner point 1 as one pair, and uses only one arbitrary corner point pair. The same applies in S403.

[0085] ( Figure 4 Step S403)

[0086] Computer system 116 narrows down the range of corner pairs for measuring the distance between corner points from the candidate corner pairs. Specifically, regarding S402 (a), only two corner points within a threshold distance that are diagonally aligned are used as measurement objects; regarding S402 (b), only two corner points within a threshold distance that are juxtaposed are used as measurement objects. Figures 9A to 9E Let's take a specific example to illustrate this step.

[0087] ( Figure 4 Step S403: Supplement)

[0088] In this step, computer system 116 uses all corner points within the threshold distance as the measurement object. Therefore, there may be multiple candidate corner pairs within the threshold distance.

[0089] ( Figure 4 Step S404)

[0090] When a line segment connecting two corner points (potential candidates for a corner pair) intersects with a line segment forming a shape pattern, the computer system 116 excludes these two corner points from the corner pair candidates. Figure 10 Here's an example to illustrate this step.

[0091] Figure 5 This diagram illustrates a specific example of one of the additional information provided in S401. Corner points exist in four types: the lower right end, the upper right end, the lower left end, and the upper left end of a rectangle. Furthermore, for each corner point... Figure 3 The diagram illustrates left and right rotations. Therefore, it allows for the classification of corner points in the design data. Figure 5 The eight types shown.

[0092] Figure 6AThis illustrates an example of a small step difference. To exclude small step differences, computer system 116 determines whether the distance between corner points is less than a threshold (the first threshold). If it is less than the first threshold, it can be identified as a small step difference. In S401, computer system 116 assigns an external attribute to two corner points whose distance is less than the first threshold.

[0093] Figure 6B This represents an example of a dense corner group 602. The dense corner group 602 is formed by a large number of corner points clustered together in close proximity. It can be considered that, to measure the distance between corner points, measuring the distance between any two points in the dense corner group is sufficient. This is because the distances between other corner points are also approximately the same. In step S401, the computer system 116 assigns a dense attribute to two corner points where the distance between corner points is less than or equal to a first threshold.

[0094] Figure 7 An example of polar coordinates. The origin is the top left of the design data, and the coordinates of each corner point are represented by distance and angle. In the processing described later in Figures 8-9, since polar coordinates are easy to calculate, they are pre-calculated in S401 and assigned as additional information.

[0095] Figure 8A This illustrates an example of determining corner pair candidates in S402. The upper left and lower right corner points within the rectangular closed region of a convex pattern are designated as corner pair candidates (Pattern 1). The same applies to concave patterns (Pattern 2). Similarly, the upper right and lower left corner points within the rectangular region of a convex pattern are designated as corner pair candidates (Pattern 3). The same applies to concave patterns (Pattern 4). Based on the above, Figure 8A The orientation of the line segments surrounding the corner point formed in the middle is respectively limited to the corner points formed in the middle. Figure 8A The four combinations of line segment patterns shown on the right.

[0096] Figure 8B This represents an example of determining a corner pair candidate in S402. Figure 8B In Pattern 5, the first convex pattern is positioned at the upper left, and the second convex pattern is positioned at the lower right. The lower right corner of the first convex pattern and the upper left corner of the second convex pattern are positioned opposite each other. The two corner points in this configuration are also designated as corner pair candidates. However, in this case, the orientation of the line segments surrounding the corner points forming the corner pair candidates is the same as in Pattern 1 (Note: Although the orientation of the line segments is the same, the search range described later is different from that in Pattern 1, and the same applies to Pattern 6 and later). Similarly, for Pattern 6 and later, the orientation of the line segments surrounding the corner points forming the corner pair candidates is the same as in Pattern 1. Figure 8A They are all the same.

[0097] With the first convex pattern positioned at the lower left and the second convex pattern positioned at the upper right, the positional relationship is the same as that of pattern 7. Therefore, this positional relationship does not need to be considered. That is, only the positional relationship needs to be considered. Figure 8B The four positional relationships shown are sufficient.

[0098] Figure 8B The external pattern represents an example of two corner points not in an opposing configuration. An opposing configuration, as used here, means that if one corner point extends in the direction it protrudes, the other corner point is positioned accordingly. Figure 8B Except for the corner pairs in Embodiment 1, which are not opposite to each other in the external pattern of the object. Regarding the shape pattern in such a positional relationship, it can be inspected by an inspection method different from that of this disclosure.

[0099] Figure 8C This represents an example of determining a corner pair candidate in S402. Figure 8C In pattern 9, the corner point corresponding to the lower right corner point protrudes inwards towards the closed area. In this type of pattern, the upper left corner point and the lower right corner point are also designated as corner pair candidates. Pattern 10 represents a concave pattern with the same shape as pattern 9. The orientation of the line segments around the corner points forming corner pair candidates in pattern 10 is exactly the same as in pattern 9. Similarly, for patterns 11 and thereafter, the two corner points arranged diagonally are designated as corner pair candidates.

[0100] In Pattern 9, since the relationship between the upper left corner point and the lower right corner point is the same as in Pattern 1, it does not need to be considered here. The relationship between the upper left corner point and the lower right corner point in Pattern 9 is also the same as in Pattern 1. Similarly, in Pattern 10 and thereafter, repeating line segment pairs do not need to be considered.

[0101] Figure 8D This represents an example of determining a corner pair candidate in S402. Figure 8D In pattern 17, the first convex pattern is positioned at the upper left, and the second convex pattern is positioned at the lower right. The lower right corner of the first convex pattern protrudes inwards towards the enclosed area, similar to pattern 9. The lower right corner of the first convex pattern and the upper left corner of the second convex pattern are positioned opposite each other. These two corner points in this arrangement are also designated as corner pair candidates. However, in this case, the orientation of the line segments surrounding the corner points forming the corner pair candidates is the same as in pattern 9. The same applies to patterns 18 and later (equivalent to patterns 10 to 16), where the orientation of the line segments surrounding the corner points forming the corner pair candidates is the same as in pattern 9. Figure 8C They are all the same.

[0102] Figure 8E It is a diagram that selects the combination of line segments surrounding the candidate corner point that form a corner. For example... Figures 8A to 8DAs explained in the text, the orientation of the line segments surrounding the corner forming the corner point is limited to... Figure 8E The pattern shown is 8. Therefore, when the computer system 116 determines the corner pair candidates in S402, it only selects... Figure 8E The line segment combinations of the 8 patterns shown can be set as the processing objects. This simplifies the process of determining the candidates for corner pairs.

[0103] Figure 9A This illustrates an example of narrowing down the candidate range for corner pairs in S403. Figure 8A In pattern 1, since the upper left corner and lower right corner are set as potential corner pairs, the possibility of these two corner points existing is limited to... Figure 9A The area indicated by the diagonal lines. Therefore, computer system 116 searches for candidate corner pairs corresponding to the line segment combinations of pattern 1 within the diagonal line area. No search is required outside the diagonal line area. Regarding Figures 9B to 9D Similarly, when there are corner points of opposite line segments within the same area, these two corner points are used as the objects for measuring the distance between the corner points. Figures 9B to 9D Similarly, the size of the diagonal region (the radius of the sector) is pre-determined as a distance threshold. The user can also input this distance threshold. Figures 9B to 9D The same applies to Naka. Regarding... Figure 8A After pattern 2, perform the same search. Figure 9A Only an example of pattern 3 is shown.

[0104] Figure 9B This illustrates an example of narrowing down the candidate range for corner pairs in S403. Figure 8B In pattern 5, since the lower right corner of the first convex pattern and the upper left corner of the second convex pattern are designated as corner pair candidates, the possibility of these two corner points existing is limited to the area shown by the diagonal lines. Therefore, the computer system 116 searches for corner pair candidates corresponding to the line segment combination of pattern 5 within the area of ​​the diagonal lines. Regarding Figure 8B After pattern 6, the search is performed in the same way.

[0105] Figure 9C This illustrates an example of narrowing down the candidate range for corner pairs in S403. Figure 8C In pattern 9, since the upper left corner and lower right corner are set as potential corner pairs, the possibility of these two corner points existing is limited to... Figure 9C Within the area indicated by the diagonal lines. Therefore, the computer system 116 searches for candidate corner pairs within the area of ​​the diagonal lines that correspond to the line segment combinations of pattern 9. Regarding Figure 8C After pattern 10, perform the same search. Figure 9C Only an example of pattern 10 is shown.

[0106] Figure 9D This illustrates an example of narrowing down the candidate range for corner pairs in S403. Figure 8D In pattern 17, since the lower right corner of the first convex pattern and the upper left corner of the second convex pattern are designated as corner pair candidates, the possibility of these two corner points existing is limited to the area shown by the diagonal lines. Therefore, the computer system 116 searches for corner pair candidates corresponding to the line segment combinations of pattern 17 within the area shown by the diagonal lines. Regarding Figure 8D The same search is performed on pattern 17 and later.

[0107] Figure 9E Indicates to Figures 9A to 9D The results have been compiled. Figures 9A to 9D In S403, computer system 116 searches for two corner points diagonally arranged within a closed region pattern. Figure 9E The range shown by the eight combinations in the upper layer is sufficient. (b) In the case of searching for the two corner points of the opposite configuration of the two patterns, the search... Figure 9E The range shown by the eight combinations at the bottom is sufficient. This not only narrows down the range for corner pairs but also simplifies the processing used for narrowing down that range.

[0108] Figure 10 This is a diagram illustrating a specific example of S404. In Figure 10 In the middle, there are two potential corner pairs centered on the lower left corner point. When connecting the two corner points that form potential corner pairs with line segments, if they intersect with other shapes or patterns, these two corner points are excluded from the potential corner pairs. Figure 10 (NG).

[0109] Figure 11 This is a flowchart illustrating the details of S208. As a method for extracting the outline of the actually formed shape pattern, any known technique can be used. Figure 11 An example is shown. First, the edges that serve as references are detected from the SEM image obtained by photographing the pattern set as the object of measurement (S1101). The contour positions are determined based on the line contours of the reference edges obtained in S1101 (S1102). A group of contour points is generated by connecting the contour positions along the object pattern point list in the design data (S1103).

[0110] Figure 12 This is a diagram illustrating a specific example of S210. The corner points on the actual formed shape pattern are not limited to having the shortest distance from each other. Therefore, with a corner point on the shape pattern as the center, the distance between corner points is calculated for any number (e.g., a user-specified number) of corner points or all combinations of corner points. The computer system 116 uses the two corner points that have the shortest distance as the final corner pair within that given range.

[0111] <Implementation Method 1: Summary>

[0112] The pattern inspection / measurement system 100 of this embodiment 1 determines corner pairs in the design data, determines the relative positional relationship between the corner pairs and the contour lines corresponding to the corner pairs on the actual formed shape pattern, and thereby determines the corner points. Thus, corner points on the actual pattern can be determined with good accuracy.

[0113] The pattern inspection / measurement system 100 of this embodiment 1 excludes corner points formed due to minute step differences from the corner pair candidates (see reference). Figure 6A This reduces the burden of searching for candidates around corners and allows for rapid completion of checks based on the distance between corner points.

[0114] The pattern inspection / measurement system 100 of this embodiment 1, when determining candidate corner pairs, extracts line segment pairs of two diagonally arranged corner points within a closed area pattern (see reference). Figure 8A , Figure 8C Only the corner point pairs formed by the line segment pair are designated as corner pair candidates. Furthermore, the search range for corner pair candidates is limited to the area where the corner point pairs formed by the line segment pair can exist (see reference). Figure 9A , Figure 9C This reduces the burden of searching for candidates around corners and allows for rapid completion of checks based on the distance between corner points.

[0115] The pattern inspection / measurement system 100 of this embodiment 1, when determining candidate corner pairs, extracts line segment pairs of two oppositely arranged corner points that form two closed area patterns (see reference). Figure 8B , Figure 8D Only the corner point pairs formed by the line segment pair are designated as corner pair candidates. Furthermore, the search range for corner pair candidates is limited to the area where the corner point pairs formed by the line segment pair can exist (see reference). Figure 9B , Figure 9D This reduces the burden of searching for candidates around corners and allows for rapid completion of checks based on the distance between corner points.

[0116] In order to distinguish between convex and concave patterns, the pattern inspection / measurement system 100 of this embodiment 1 assigns the rotation direction of the line segments forming the corner points to the design data as an additional attribute (see reference). Figure 3 Therefore, based on the design data as two-dimensional data, it is possible to identify corner pairs while recognizing three-dimensional shape patterns.

[0117] In the pattern inspection / measurement system 100 of this embodiment 1, when the line segment connecting the corner points intersects with the shape pattern, the corner point is excluded from the corner pair candidate (see reference). Figure 10 Therefore, since only corner points that are within the shortest distance to the corner point of interest are identified as candidate corner pairs, the inspection process based on the distance between corner points can be reduced.

[0118] When multiple corner points exist within the reference threshold range to form candidate corner pairs, the pattern inspection / measurement system 100 of this embodiment 1 only selects corner point pairs whose distance between corner points is a point of difference as final candidates for corner pairs (see reference). Figure 12 This reduces the need for inspection processing based on the distance between corner points.

[0119] <Implementation Method 2>

[0120] In implementation method 1, the existing Figures 9A to 9E All corners within the range described herein are extracted. In Embodiment 2 of this disclosure, it is explained that those existing in Figures 9A to 9E The example described above illustrates an action where the range of corners within the specified area is narrowed down in advance based on the distance between corner points. Since the structure of the pattern inspection / measurement system 100 is the same as in Embodiment 1, the following mainly describes the differences related to the search range.

[0121] Following S404, computer system 116 further implements the following processing. [The remaining text appears to be incomplete and requires further context.] Figures 9A to 9E The corner points are extracted based on the typical attributes described herein. The computer system 116 uses only the two corner point pairs with the shortest distance between them as candidate corner pairs, excluding other corner point pairs from the candidate list.

[0122] Following S404, computer system 116 further implements the following processing. [The remaining text appears to be incomplete and requires further context.] Figures 9A to 9E Extraction of corner point clusters with dense attributes within the range described in the text. The extraction method includes the following two techniques.

[0123] Computer system 116 selects only the two corner point pairs with the shortest distance between them from the same dense corner point group as corner point candidates, excluding other corner point pairs (method 1). Computer system 116 selects only the central corner point from the same dense corner point group as a corner point, excluding other corner points (method 2). By using method 2, since the dense corner point group can be set as one corner point, the time for searching for the two corner point pairs with the shortest distance between them can be reduced. It is assumed that as long as the corner points belong to the same dense corner point group, the difference will be small regardless of which point is selected as the corner point. However, in this embodiment 2, the point located at the center of the dense range, which is considered to have the largest difference even within the dense range, is selected as the corner point.

[0124] By using the above methods, the computer system 116 can pre-narrow down the range of candidates for corner points. This reduces the workload of inspection processes based on the distance between corner points.

[0125] <Implementation Method 3>

[0126] Figure 13 This is an example of a graphical user interface (GUI) provided by computer system 116. Computer system 116 can display GUIs such as screens 1301 and 1302 on a display device such as a monitor included in input / output device 121.

[0127] Screen 1301 displays (1) the shape pattern 1303 in the design data, (2) the corner pair candidate 1304 determined in the design data, and (3) the line segment 1305 connecting the corner pairs.

[0128] Screen 1302 displays (1) the outline of the actual pattern 1306, (2) the estimated position of the corner point 1307, (3) the line segment connecting the estimated positions of the corners 1308, and (4) the distance between the corners 1309, etc.

[0129] Users can also switch between screens 1301 and 1302. Furthermore, they can display design data and actual patterns simultaneously.

[0130] <Variations on this disclosure>

[0131] This disclosure is not limited to the foregoing embodiments and includes various modifications. For example, the embodiments described above have been detailed for ease of understanding and explanation of this disclosure, but are not necessarily limited to having all the described structures. Furthermore, a part of the structure of a certain embodiment can be replaced with the structure of another embodiment, and the structure of another embodiment can be added to the structure of a certain embodiment. In addition, other structures can be added, deleted, or replaced to a part of the structure of each embodiment.

[0132] In the above implementation, if the calculated distance between corner points deviates from the baseline value from the design data, the computer system 116 may output an alarm indicating this. For example, it may be possible to... Figure 13 The alert is displayed on the GUI as described in the instructions, or it can be output in other forms. It can also be output along with indicators representing the quality of the pattern, such as the deviation from the design data.

[0133] In the above embodiments, the arithmetic processing unit 117 and the image processing unit 118 can also be constructed from hardware such as circuitry devices that have their functions installed, or from software that has their functions installed and executed by a processor such as a CPU (Central Processing Unit). Similarly, the arithmetic processing unit 112 and the control unit 113 can also be constructed from hardware or software executed by a processor.

[0134] In the above embodiments, SEM 101 was exemplified as a means of obtaining an image of the shape pattern formed on the sample 107, but this disclosure is not limited thereto. Other means may be used as long as the computer system 116 can process the image of the shape pattern to determine the corner points.

[0135] In the above embodiments, a semiconductor sample was exemplified as sample 107, but this disclosure is not limited thereto and can also be applied to shape patterns formed on other samples.

[0136] Explanation of reference numerals in the attached figures

[0137] 100: Pattern Inspection / Measurement System

[0138] 101: Scanning Electron Microscope

[0139] 102: Electron Beam Column

[0140] 103: Electronic Source

[0141] 104: Electron Beam

[0142] 105: Vacuum Sample Chamber

[0143] 106: XY platform

[0144] 107: Sample

[0145] 108: Secondary electrons or backscattered electrons

[0146] 109: A / D converter

[0147] 110: Internet

[0148] 111: Computer System

[0149] 112: Computation and Processing Unit

[0150] 113: Control Department

[0151] 114: Storage Department

[0152] 115: Internet

[0153] 116: Computer System

[0154] 117: Computation and Processing Unit

[0155] 118: Image Processing Department

[0156] 119: Storage Department

[0157] 120: Design Information Database

[0158] 121: Input / output device.

Claims

1. A pattern inspection / measuring apparatus for inspecting or measuring shape patterns formed on a sample, characterized in that, It includes: a computer system that detects corner patterns containing corner points within the shape pattern from an image obtained by photographing the shape pattern. The computer system obtains the coordinates and shape of the corner pattern from the design data of the shape pattern. The computer system determines two corner points within a given distance from each other as candidate corner pairs based on the coordinates and shape of the corner pattern obtained from the design data. The computer system determines the coordinates of the two corner points on the shape pattern forming the corner pair candidate based on the relative relationship between the position of the shape pattern forming the corner pair candidate and the position of the corner pair candidate on the design data. The computer system extracts two corner points (excluding adjacent corner points) from a corner pattern that forms a closed area by connecting line segments, and uses these as the internal corner point pair of the corner pattern. The computer system determines the extracted pairs of internal corner points as candidates for the corner pairs.

2. The pattern inspection / measuring device according to claim 1, characterized in that, When extracting the internal corner point pairs, the computer system searches the design data for the range of two corner points other than adjacent corner points within a closed area formed by connecting line segments.

3. The pattern inspection / measuring device according to claim 1, characterized in that, The design data is data describing a two-dimensional shape. The computer system distinguishes between protruding shape patterns formed by protrusion along the normal direction of the two-dimensional plane and concave shape patterns formed by indentation. The computer system extracts the internal corner point pairs for the protruding shape pattern and the concave shape pattern, respectively.

4. The pattern inspection / measuring device according to claim 1, characterized in that, The computer system extracts corner points whose distance to each other is less than a first threshold, and uses these as external pairs of objects. The computer system extracts corner points whose distances to each other are greater than the first threshold but less than the second threshold, and classifies them as dense corner groups. The computer system identifies two corner points, excluding the outer corner points and the dense corner group, that have the smallest mutual distance among the internal corner point pairs as candidate corner pairs. The computer system determines two corner points that are the closest to each other among the corner points contained in the internal corner point pair and the dense corner group, as candidates for the corner pair; or, the computer system extracts one point from the dense corner group containing the corner points contained in the internal corner point pair and the dense corner group, and determines the extracted corner point as a candidate for the corner pair.

5. A pattern inspection / measuring apparatus for inspecting or measuring shape patterns formed on a sample, characterized in that, It includes: a computer system that detects corner patterns containing corner points within the shape pattern from an image obtained by photographing the shape pattern. The computer system obtains the coordinates and shape of the corner pattern from the design data of the shape pattern. The computer system determines two corner points within a given distance from each other as candidate corner pairs based on the coordinates and shape of the corner pattern obtained from the design data. The computer system determines the coordinates of the two corner points on the shape pattern forming the corner pair candidate based on the relative relationship between the position of the shape pattern forming the corner pair candidate and the position of the corner pair candidate on the design data. The computer system extracts a first corner point from one of the two corner patterns forming a closed area by connecting line segments, and a second corner point from the other corner pattern, which is opposite to the first corner point, as an external corner point pair. The computer system determines the extracted external corner point pairs as candidates for the corner pairs.

6. The pattern inspection / measuring device according to claim 5, characterized in that, When extracting the external corner point pairs, the computer system searches the design data only for the range of two corner points that are opposite each other among the corner points of the two corner patterns that form a closed area by connecting line segments.

7. The pattern inspection / measuring device according to claim 5, characterized in that, The design data is data describing a two-dimensional shape. The computer system distinguishes between protruding shape patterns formed by protrusion along the normal direction of the two-dimensional plane and concave shape patterns formed by indentation. The computer system extracts the external corner point pairs for the protruding shape pattern and the concave shape pattern, respectively.

8. The pattern inspection / measuring device according to claim 5, characterized in that, The computer system extracts corner points whose distance to each other is less than a first threshold, and uses these as external pairs of objects. The computer system extracts corner points whose distances to each other are greater than the first threshold but less than the second threshold, and classifies them as dense corner groups. The computer system determines the two corner points with the smallest mutual distance among the external corner points obtained from the external corner point pair (excluding the object external pair and the dense corner group), as candidates for the corner pair. The computer system determines two corner points that are the smallest in distance from each other among the corner points contained in the external corner point pair and the dense corner group, as candidates for the corner pair; or, the computer system extracts one point from the dense corner group containing the corner points contained in the external corner point pair and the dense corner group, and determines the extracted corner point as a candidate for the corner pair.

9. The pattern inspection / measuring apparatus according to claim 1 or 5, characterized in that, The computer system extracts the first corner point and the second corner point from the corner pattern. If the distance between the first corner point and the second corner point is less than a first threshold, the computer system excludes at least one of the first corner point and the second corner point from the corner pair candidates.

10. The pattern inspection / measuring apparatus according to claim 1 or 5, characterized in that, The computer system extracts the third and fourth patterns from the corner pattern. If the line segment described in the design data intersects with the line segment connecting the third pattern and the fourth pattern, the computer system excludes at least one of the third pattern and the fourth pattern from the corner candidate.

11. The pattern inspection / measuring apparatus according to claim 1 or 5, characterized in that, The computer system calculates and outputs the distance between the two corner points that determine the coordinates, as the distance between the two corner points that form the candidate corner pair.

12. The pattern inspection / measuring apparatus according to claim 11, characterized in that, The computer system identifies multiple candidate corner pairs. If, among the plurality of corner pair candidates, the distance between the corner point forming the first corner pair candidate and the corner point forming the second corner pair candidate is within a reference threshold, and the distance between the corner point forming the other corner point forming the first corner pair candidate and the corner point forming the other corner point forming the second corner pair candidate is within the reference threshold, the computer system extracts only the combination with the shortest mutual distance among all combinations of corner points contained in the first corner pair candidate and corner points contained in the second corner pair candidate, excluding other corner points from the corner pair candidates.

13. The pattern inspection / measuring apparatus according to claim 11, characterized in that, The computer system includes a screen interface that displays the two corner points that form the corner pair candidates and the distance between the corner points.

14. A pattern inspection / measurement program product, which enables a computer to perform processing for inspecting or measuring a shape pattern formed on a sample, characterized in that, The pattern inspection / measurement program product causes the computer to perform the following steps: detect corner patterns containing corner points within the shape pattern from an image obtained by photographing the shape pattern. In the step of detecting the corner pattern, the computer performs the following steps: The coordinates and shape of the corner pattern are obtained from the design data of the shape pattern; Based on the coordinates and shape of the corner pattern obtained from the design data, two corner points within a given distance from each other are determined as candidate corner pairs; and The corner pair candidates are formed according to the relative relationship between the position of the shape pattern forming the corner pair candidate and the position of the corner pair candidate on the design data. The pattern inspection / measurement program product further enables the computer to perform the following steps: Extract the two corner points other than adjacent corner points from a corner pattern that forms a closed region by connecting line segments, and define them as the internal corner point pair of the corner pattern. The extracted pairs of internal corner points are identified as candidates for the corner pairs.

15. A pattern inspection / measurement program product that enables a computer to perform processing for inspecting or measuring shapes and patterns formed on a sample, characterized in that, The pattern inspection / measurement program product causes the computer to perform the following steps: detect corner patterns containing corner points within the shape pattern from an image obtained by photographing the shape pattern. In the step of detecting the corner pattern, the computer performs the following steps: The coordinates and shape of the corner pattern are obtained from the design data of the shape pattern; Based on the coordinates and shape of the corner pattern obtained from the design data, two corner points within a given distance from each other are determined as candidate corner pairs; and The corner pair candidates are formed according to the relative relationship between the position of the shape pattern forming the corner pair candidate and the position of the corner pair candidate on the design data. The pattern inspection / measurement program product further enables the computer to perform the following steps: Extract the first corner point of one of the two corner patterns forming a closed area by connecting line segments, and the second corner point of the other corner pattern, which is opposite to the first corner point, as the outer corner point pair. The extracted external corner point pairs are identified as candidate corner pairs.

Images