Measurement system and method

Abstract

Provided is a technology capable of more accurately acquiring / measuring specific pattern positions and / or the like from an image of a semiconductor device captured by a charged particle beam apparatus, even in cases where there is point irradiation of a specific pattern. The present invention provides a measurement system comprising a computer system that takes measurements of a pattern formed on a semiconductor sample by performing image processing on the basis of data pertaining to images obtained by a charged particle beam apparatus (e.g., an SEM). The computer system acquires a first image captured before point irradiation with respect to a target pattern is performed using a charged particle beam and a second image captured after point irradiation is performed (steps S1, S4), calculates an amount of positional shift in the target pattern between the first image and the second image (steps S5, S7), and determines a measurement position for the target pattern on the basis of the calculated amount of positional shift (step S8).

Description

Measurement system and method

[0001] The present disclosure relates to techniques for measuring, analyzing, observing, inspecting, evaluating, etc. semiconductor devices. In this specification, measurement, analysis, observation, inspection, evaluation, etc. may be collectively referred to as measurement.

[0002] A semiconductor device measurement / inspection system irradiates an irradiation point (in other words, a target point) on a semiconductor device, which is a sample, with an electron beam, which is a charged particle beam, detects particles such as secondary charged particles that arise from the irradiation point, and performs inspections such as measuring the dimensions of the semiconductor device and inspecting its electrical characteristics (electrical defect inspection) based on the detected signals.

[0003] A scanning electron microscope (SEM) or the like is used as a charged particle beam device in a semiconductor device measurement / inspection system. The SEM irradiates a sample with a charged particle beam, detects particles such as secondary charged particles generated from the sample at the irradiation point as signals, and captures an image (in other words, an SEM image).

[0004] One method of semiconductor device inspection using an SEM is to perform an electrical characteristic inspection by irradiating a target point on a sample with a pulsed electron beam and analyzing a voltage contrast image obtained by detecting secondary electrons, etc. The electrical characteristic inspection is an inspection to determine whether a specific pattern such as a plug satisfies predetermined electrical characteristics in the design and whether there are any defects.

[0005] An example of the prior art is Japanese Patent Laid-Open No. 2021-22440 (Patent Document 1). Patent Document 1 states that "a system capable of evaluating the electrical characteristics of an element formed on a sample is provided." Patent Document 1 also states that "electrical characteristics are derived by referring to information relating to two or more features extracted from at least two or more image data obtained under at least two image acquisition conditions and information relating to the electrical characteristics of the element formed on the sample."

[0006] Japanese Patent Application Laid-Open No. 2021-22440

[0007] In order to derive the electrical characteristics of an element formed on a sample using a technique such as that described in Patent Document 1, it is necessary to accurately acquire the specific pattern position from each image acquired under different image acquisition conditions. However, if the image acquisition conditions are different, the specific pattern position in the image may differ from image to image. In other words, even if the specific pattern position is acquired in an image under certain image acquisition conditions, the specific pattern position may differ in another image acquired under different image acquisition conditions.

[0008] One of the reasons why the position of the specific pattern differs from image to image is that the conditions for capturing the images, such as the positional relationship between the sample on the stage and the field of view, are different. The amount of positional deviation of the specific pattern caused by differences in the image capturing positions can be calculated by aligning images captured under different image capturing conditions. However, such alignment alone cannot accurately obtain the amount of positional deviation of the specific pattern between images.

[0009] This is because distortions can occur in images depending on the image acquisition conditions. For example, in an SEM image taken after irradiating an electron beam onto a specific pattern position (i.e., an irradiation point) on a sample, local distortions occur due to the non-uniform charge state on the sample surface. If local distortions occur in an SEM image, even if the same location on the sample is imaged with an SEM, the position of the pattern in the SEM image will be shifted in accordance with the local distortion of the SEM image.

[0010] Even if a user and an SEM for measuring / inspecting the semiconductor device set the same image acquisition conditions to capture an SEM image, the actual charging state of the sample surface will differ depending on whether or not a specific pattern is point-irradiated (in other words, point pre-dose, etc.). The SEM image may contain distortions depending on the charging state. The position of the specific pattern may be shifted depending on the distortion. Even when the specific pattern is point-irradiated as described above, it is desirable to more accurately acquire and measure the position of the specific pattern, taking into account the effect of the point irradiation, i.e., the distortion of the image depending on the charging state.

[0011] An object of the present disclosure is to provide a technology for the above-mentioned semiconductor device measurement and the like that can more accurately acquire and measure the position of a specific pattern, etc., from an image of a semiconductor device captured by a charged particle beam device such as an SEM, even when point irradiation of the specific pattern is performed.

[0012] The present disclosure proposes a method for more accurately acquiring and measuring the position of a specific pattern in an image with different image acquisition conditions, even when local distortion occurs in the SEM image due to point irradiation, etc., by calculating the amount of positional deviation corresponding to the local distortion in the SEM image, and acquiring information regarding the feature quantities of the specific pattern.

[0013] A representative embodiment of the present disclosure has the following configuration: This embodiment is a measurement system including a computer system that measures a pattern formed on a semiconductor sample by performing image processing based on image data obtained by a charged particle beam device that irradiates a semiconductor sample with a charged particle beam and captures an image, wherein the computer system acquires a first image captured before point-irradiating the pattern to be measured with the charged particle beam and a second image captured after point-irradiating the pattern with the charged particle beam, performs processing to calculate a positional deviation amount related to the position of the pattern to be measured between the first image and the second image, and determines the position of the pattern to be measured based on the calculated positional deviation amount.

[0014] According to a representative embodiment of the present disclosure, in relation to a technique for measuring a semiconductor device, it is possible to more accurately acquire and measure the position of a specific pattern from an image of a semiconductor device captured by a charged particle beam device such as an SEM, even when a specific pattern is irradiated with a point. Problems, configurations, effects, etc. other than those described above are described in the description of the embodiment of the invention.

[0015] FIG. 1 is a schematic explanatory diagram illustrating a case where a plug is irradiated with an electron beam for point irradiation as an example of a specific pattern on a sample surface. FIG. 2 is a schematic explanatory diagram illustrating an example of normal / defective plugs on a sample surface. FIG. 3 is a schematic explanatory diagram illustrating a pulsed electron beam for point irradiation. FIG. 4 is a schematic explanatory diagram illustrating changes occurring in an SEM image due to irradiation of an electron beam on a plug. FIG. 5 is a diagram illustrating the configuration of a measurement system of Example 1. FIG. 6 is a diagram illustrating a basic flow of the measurement system of Example 1. FIG. 7 is a diagram illustrating an example of a GUI screen in the measurement system of Example 1. FIG. 8 is a schematic explanatory diagram illustrating a specific example of processing content in Example 1. FIG. 9 is a diagram illustrating a specific example of processing content in Example 1. FIG. 10 is a diagram illustrating a flow of a specific processing example of step S7 in Example 1. FIG. 11 is a diagram illustrating an example of an image crop size in Example 1. FIG. 12 is a schematic explanatory diagram illustrating a specific example of processing content in step S7 in Example 1. FIG. 13 is a schematic explanatory diagram illustrating a specific example of processing content in step S9 in Example 1. FIG. 14 is a schematic explanatory diagram illustrating a specific example of processing content in step S10 in Example 1. FIG. 15 is a diagram illustrating a basic flow of a measurement system of Example 2. FIG. 16 is a diagram illustrating an example of a plug and a design drawing image in Example 2. FIG. 10 is a schematic explanatory diagram showing a specific example of processing content in Example 2. FIG. 11 is a diagram showing a basic flow of a measurement system in Example 3. FIG. 12 is a diagram showing an example of contour line extraction in Example 3. FIG. 13 is a schematic explanatory diagram showing a specific example of non-rigid registration and the like in Example 3. FIG. 14 is a diagram showing a basic flow of a measurement system in Example 4. FIG. 15 is a schematic explanatory diagram showing a specific example of processing content in Example 4. FIG. 16 is a schematic explanatory diagram showing a specific example of SEM images in a modified example of Example 1. FIG. 17 is a schematic explanatory diagram showing a specific example of SEM images in a modified example of Example 1.

[0016] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. In the drawings, identical parts are generally designated by the same reference numerals, and repeated explanations will be omitted. In the drawings, the representation of components may not represent their actual positions, sizes, shapes, ranges, etc., in order to facilitate understanding of the invention.

[0017] For the purpose of explanation, when describing processing by a program, the program, functions, processing units, etc. may be described as the main components, but the main hardware components are the processor, or a controller, device, computer, system, etc. that is configured with the processor, etc. The computer executes processing according to the program read into memory using resources such as memory and communication interfaces as appropriate through the processor. This realizes predetermined functions, processing units, etc. The processor is configured, for example, with semiconductor devices such as a CPU / MPU or GPU. Processing is not limited to software program processing, and can also be implemented using dedicated circuits. Dedicated circuits such as FPGAs, ASICs, and CPLDs can be used.

[0018] The program may be pre-installed as data on the target computer, or may be distributed as data from a program source to the target computer. The program source may be a program distribution server on a communication network, or a non-transitory computer-readable storage medium, such as a memory card or disk. The program may be composed of multiple modules. The computer system may be composed of multiple devices. The computer system may be composed of a client-server system, a cloud computing system, an IoT system, etc. Various data and information may be composed of structures such as tables and lists, for example, but are not limited to these. Expressions such as identification information, identifiers, IDs, names, and numbers are interchangeable.

[0019] [Solution, etc.] The measurement system of this embodiment includes a charged particle beam device (in other words, an imaging unit) that irradiates an electron beam onto a semiconductor device serving as a sample and detects secondary electrons, etc. emitted from the sample to acquire an image; a position calculation unit (in other words, a pattern measurement unit) that calculates / measures the position of a specific pattern (e.g., a plug, etc.) to be measured / inspected from the image acquired from the imaging unit, taking into account point irradiation, etc. of the specific pattern (i.e., the resulting effect of image distortion depending on the charged state); a feature calculation unit that analyzes and extracts feature amounts (e.g., brightness) of the specific pattern based on the position of the specific pattern in the image; and an inspection unit that performs electrical defect inspection of the specific pattern based on the extracted feature amounts.

[0020] The position calculation unit calculates / measures the position of the target pattern after correction from an image (e.g., an SEM image) so as to eliminate positional deviations corresponding to the distortion, taking into account the influence of image distortion due to the charging state caused by point irradiation, etc., on the target specific pattern (e.g., a plug).

[0021] According to this embodiment, the position of a specific pattern of interest can be more accurately acquired and measured from images with different image acquisition conditions (for example, images with or without point irradiation on the specific pattern), feature values ​​can be acquired based on the acquired position, and electrical defect inspection can be performed more accurately (in other words, with higher precision) based on the acquired feature values.

[0022] In the following examples, an example will be described in which measurement and electrical defect inspection are performed by irradiating an electron beam (particularly point irradiation) onto a semiconductor device and a specific pattern, which is a circuit in which plugs, which are circular circuit patterns that constitute a semiconductor memory, are arranged in a regular pattern.

[0023] First Embodiment A semiconductor device measuring system and the like according to a first embodiment will be described with reference to FIG. 1 and subsequent figures.

[0024] [Example of Specific Pattern: Plug] First, FIG. 1 is a schematic explanatory diagram of a case where a plug is included as an example of a specific pattern in a semiconductor device sample, and shows a schematic diagram of the plug being irradiated with an electron beam b1 using an SEM. In state A, multiple plugs 301 are arranged, for example, vertically and horizontally in a 10x10 pattern on the sample surface (here, assumed to be the X-Y plane). The plugs 301 are wiring formed by embedding a conductor in a hole shape. The plug 302 is an example of a plug to be measured and inspected, and is a plug to be irradiated with the electron beam b1 for point irradiation. Point p1 is the target point / irradiation point of point irradiation and represents the position of the target plug 302, which in this case is the center point as the center of gravity of the circular plug region.

[0025] State B is a cross-sectional view (here, assumed to be the X-Z plane) of the vicinity of the sample surface corresponding to state A. During point irradiation, an electron beam b1 is irradiated from the vertical direction (Z direction) to a point p1 of the target plug 302. The electron beam b1 for point irradiation is, for example, a pulsed electron beam, and the pulse irradiation time, shut-off time, etc. are controlled (see FIG. 3, described below). The target plug 302 becomes charged by point irradiation. Factors that affect the charged state include, for example, the material and wiring state of the pattern. The brightness in the SEM image varies depending on the charged state. Electrical characteristics (e.g., electrical resistance and capacitance) can be measured and inspected depending on the brightness in the SEM image.

[0026] Regarding the target plug (target pattern) of the sample for point irradiation, a recipe may be set by specifying the target plug or a target region including the target plug from an SEM image in advance (see FIG. 7 described later). Also, the acceleration voltage and the like can be set as optical conditions for point irradiation.

[0027] 1 illustrates a case where the electron beam b1 is irradiated perpendicularly to one target plug 302 during point irradiation (in other words, point pre-dose), but the point irradiation technique is not limited to this. Other examples include irradiating multiple target plugs with the electron beam simultaneously, irradiating multiple target plugs sequentially while scanning them with the electron beam, irradiating the target plug with the electron beam from an oblique direction (tilt direction), etc.

[0028] A charged particle beam device (e.g., SEM) irradiates a specific target plug 302 with an electron beam b1 at a point. This causes the target plug 302 to become charged. The charged particle beam device (e.g., SEM) captures an image (SEM image) of the charged sample. This changes the brightness of the SEM image due to changes in the surface potential of the sample. Brightness is a value that indicates the brightness of the image acquired by the charged particle beam device (e.g., SEM). The amount of charge reflects the electrical properties of the sample, such as its electrical resistance and capacitance.

[0029] This measurement system can perform electrical characteristic inspection (electrical defect inspection) of a sample by acquiring and analyzing the brightness of an SEM image of a specific target plug after point irradiation. For example, by comparing the brightness of the target plug with a threshold value, it is possible to estimate and determine the presence or absence of defects in the target plug and the type of defect. Patent Document 1 describes an example of inspection using point irradiation.

[0030] A brief explanation of point irradiation (point pre-dose) is provided. The voltage contrast method, a method for observing SEM images, utilizes contrast reflecting differences in surface potential resulting from charging caused by irradiating a specific pattern on the sample surface with an electron beam. This method uses differences in brightness of specific patterns in SEM images to identify and detect defective areas (defects) in order to inspect for defects in the electrical characteristics of specific patterns on elements. Brightness is, for example, inversely proportional to the electrical resistance of the element and proportional to its capacitance. Based on the brightness differences, defective areas on the element can be identified.

[0031] However, point irradiation of the electron beam charges the sample, especially the specific pattern, causing distortion in the SEM image depending on the charged state. Therefore, the target pattern position specified before point irradiation may differ from the target pattern position in the SEM image after point irradiation. Therefore, when point irradiation is used, it is necessary to accurately identify the target pattern position.

[0032] In addition, during normal imaging, an SEM irradiates a sample with an electron beam while scanning it, but the effect of charging differs between such irradiation during normal imaging and point irradiation. This embodiment is a technology that attempts to correct the positional deviation of a specific pattern on an image, taking into account the effect of point irradiation.

[0033] [Example of Normal / Defective State of Plug] FIG. 2 is a cross-sectional view showing an example of the normal / defective state of a plug such as that shown in FIG. 1. This figure shows a variety of plug states collectively as a schematic diagram. A conductor is embedded in each plug region (hole shape). The plug is surrounded by an insulating layer. The plug PL1 is in contact (electrically connected) with the wiring WL1 that is electrically floating above the silicon substrate SS1. The plug PL1 is an example of a plug that is in a normal state as a result of the manufacturing process. Here, the state in which the bottom end of the plug is in contact with the specified wiring is considered to be the normal state. In an electrical characteristics test, it is desirable to determine such a plug PL1 as normal (in other words, without defects).

[0034] The plug PL2 is in a state where it is not in complete contact with the wiring WL1, and is an example of a defective state. In an electrical characteristic inspection, it is desirable to determine such a plug PL2 as defective (in other words, abnormal, defective). Furthermore, the plug PL3 is in a state where it is coupled to the wiring WL1 with electrical resistance, and is an example of a defective state. In an electrical characteristic inspection, it is desirable to determine such a plug PL3 as defective. A defect such as the plug PL2 has a high electrical resistance but a low electrostatic capacitance. A defect such as the plug PL3 has a high electrostatic capacitance but a low electrical resistance.

[0035] The plug PL4 is in contact with the wiring WL2 located in a layer below the wiring WL1, which is an example of a normal state. The plug PL5 should have been in contact with the wiring WL1 located in the upper layer, but instead penetrated the wiring WL1 and bonded to the wiring WL2 located in the lower layer, which is an example of a defective state.

[0036] Furthermore, plug PL6 is an example of a plug that is in normal contact with impurity diffusion layer ID1 of silicon substrate SS1. Plug PL7 is an example of a defective state in which plug PL7 is not in normal contact with impurity diffusion layer ID2. Furthermore, plugs PL8 and PL9 are examples of a defective state in which conduction occurs between two adjacent plugs.

[0037] This measurement system can determine whether a plug such as the above example is normal or defective in the electrical characteristic test described below.

[0038] [Point Irradiation: Pulsed Electron Beam] FIG. 3 is an explanatory diagram of the irradiation of a pulsed electron beam during point irradiation (point pre-dose). The time (total time), voltage, etc. can be set for one point irradiation for one pattern. Furthermore, the pulse irradiation time (on time) and shut-off time (off time) for the pulsed electron beam during point irradiation can be set. State A shown in the figure is a setting under condition 1, where the pulse irradiation time is 0.1 μs and the shut-off time is 0.1 μs. State B shown in the figure is a setting under condition 2, where the pulse irradiation time is 0.1 μs and the shut-off time is 0.5 μs. Even depending on these settings, the effect of point irradiation can change.

[0039] 4 is a schematic diagram illustrating the concept of changes that occur in an SEM image before and after point irradiation of a specific pattern, such as a plug, with an electron beam, particularly point irradiation. These changes include distortion according to the charging state and positional deviation corresponding to the distortion.

[0040] State A shows an SEM image (first image) A2 before point irradiation. The image shows the sample surface (assumed to be the X-Y plane). Plug A1 is an example of a target plug for point irradiation. In this example, the plug area is relatively large and is therefore represented in white, while the other areas are relatively small and are therefore represented in black.

[0041] State B shows an SEM image (second image) A3 after point irradiation, which was taken of the same location (plug A1) as image A2 in state A after point irradiation. The plug A1 in SEM images A2 and A3 shows different brightness levels of the plug A1 (A1b) that was point irradiated before and after point irradiation. The plug A1 in state B (A1b for distinction) has a lower brightness than the plug A1 in state A, and is represented by a diagonal hatching pattern.

[0042] State C is a conceptual diagram showing SEM image A2 of state A and SEM image A3 of state B superimposed to explain local distortion (corresponding positional misalignment) of the SEM images. A4 indicated by a solid circle is the plug contour (i.e., plug region) corresponding to plug A1 in SEM image A2 before point irradiation. A5 indicated by a dashed circle is the plug contour (plug region) corresponding to plug A1b in SEM image A3 after point irradiation.

[0043] State D is a conceptual diagram corresponding to state C, in which the amount of in-plane positional deviation caused by point irradiation is expressed by the direction and length of the arrow. Arrow A6 indicates the amount of positional deviation of the image (particularly plugs A1 and A1b) caused by point irradiation. Note that the amount of positional deviation A6 is an arrow representing the positional deviation of each plug before and after point irradiation shown in state C, with the longer the arrow, the greater the amount of positional deviation. In the example of state D, the amount of positional deviation increases from the bottom left (-X, -Y) to the top right (+X, +Y) in the image of the sample surface.

[0044] In the illustrated example, the size and shape of each plug are almost unchanged before and after point irradiation, but this is not limiting. The size and shape of the pattern may change depending on the charge.

[0045] Each of the present examples (Examples 1 to 4 described below) describes a method for more accurately identifying the position of a specific pattern to be measured and inspected in an SEM image after point irradiation and acquiring feature information (particularly brightness information) based on that position. The specific electrical characteristic inspection method using the acquired feature information (particularly brightness information) is not particularly limited. Furthermore, each of the present examples illustrates the case where brightness is extracted as a feature of a specific pattern to be measured and inspected, but the feature is not limited to brightness, and other feature amounts such as pattern dimensions and circularity of the pattern shape can also be used.

[0046] [Measurement System] Fig. 5 shows the configuration of the measurement system of Example 1. The system of Fig. 5 is a measurement system for semiconductor devices, and roughly comprises a computer system 1 and a charged particle beam device 2 which is an SEM 2. The computer system 1 and the SEM 2 are connected to each other via a communication interface or the like.

[0047] In other words, the computer system 1 is an image processing unit and a measurement processing unit. The computer system 1 includes a processor 101, a memory 102, an input / output unit 103, etc., which are interconnected via a bus or the like. The computer system 1 acquires an SEM image 121, etc. from the SEM 2 via the input / output unit 103, and stores the image in the memory 102. The memory 102 stores a control program and various data and information. For example, the memory 102 stores the SEM image 121, measurement information 122, etc.

[0048] The processor 101 executes processing related to measurement of semiconductor devices in accordance with a control program stored in the memory 102. Functional blocks realized by the processing of the processor 101 include a position calculation unit 111, a feature calculation unit 112, and an inspection unit 113.

[0049] The position calculation unit 111 measures and calculates the position of a specific pattern (e.g., a plug) to be measured and inspected from the SEM image 121. The feature calculation unit 112 extracts brightness information as a feature based on the position of the specific pattern from the SEM image 121. The inspection unit 113 performs an electrical characteristic inspection (electrical defect inspection) based on the brightness information, which is the extracted feature. The processor 101 appropriately stores data and information being processed by the position calculation unit 111, feature calculation unit 112, inspection unit 113, etc., and processing result information, as measurement information 122, in the memory 102. The computer system 1 can also output the SEM image 121 and measurement information 122 to an external device (e.g., a user terminal or a server).

[0050] The input / output unit 103 is equipped with an input / output interface and a communication interface, and has input / output devices 104 externally connected or built-in. The input / output devices 104 may be input devices or output devices operated by the user U1, or may be a user terminal such as a client PC of the user U1. The user U1 is an operator who performs work related to measurement. A display device is also connected to the input / output unit 103 as one of the input / output devices 104. The screen of the display device can display a setting screen, measurement / inspection results, etc. based on the SEM image 121 and measurement information 122.

[0051] 5 , the computer system 1 may be connected to an external server or system (e.g., a manufacturing execution system) via the input / output unit 103 and a communication network. For example, the computer system 1 may refer to and acquire design data, etc., related to the semiconductor device 13, which is the sample 13, from the external server or system. Alternatively, the user U1 may input and set such data and information as appropriate via the input / output device 104.

[0052] The SEM 2 irradiates a semiconductor device 13, which is a sample 13, with a charged particle beam (electron beam 8) and detects the signal to construct and capture an SEM image 120. The SEM 2 includes an overall control unit 21, an electron optical system control unit 20, an A / D converter 16, a detection signal processing unit 17, a memory unit 18, a stage controller 19, etc. The SEM 2 includes, in a housing (column) 6, an electron gun 7, a condenser lens 9, a deflector 10, an objective lens 11, a stage 12, a detector 15, etc.

[0053] The SEM 2 controls the movement of the stage 12 (at least in the X and Y directions shown) based on the stage controller 19, and sets the imaging position and the like for the sample 13 on the stage 12. The stage controller 19 receives a control signal from the overall control unit 21 and controls the stage 12 on which the sample 13 is placed.

[0054] The SEM 2 generates and irradiates an electron beam 8 from an electron gun 7 based on an electron optical system control unit 20. The electron optical system control unit 20 receives control signals from an overall control unit 21 and controls the electron gun 7, condenser lens 9, deflector 10, objective lens 11, etc. The electron beam 8 is converged and irradiated onto a sample 13 on a stage 12 by the condenser lens 9, deflector 10, objective lens 11, etc. When the electron beam 8 irradiates the sample 13, secondary electrons 14 and the like are generated. The generated secondary electrons 14 and the like are detected by a secondary electron detector 15. An analog signal, which is a detection signal from the secondary electron detector 15, is converted into a digital signal by an A / D converter 16. A detection signal processing unit 17 constructs an SEM image 120 from the digital signal and stores it in a memory unit 18.

[0055] The overall control unit 21 also has an input / output interface and a communication interface (or is connected to a part that has them), and transmits the SEM image 120 and related information in the storage unit 18 to the computer system 1 via communication or the like. Note that the SEM 2 may transmit the SEM image 120 as a response in response to a request or instruction from the computer system 1.

[0056] In this embodiment, the inspection unit 113 also performs an electrical defect inspection, but the present invention is not limited to this. At least, the position calculation unit 111 calculates the position of the specific pattern, that is, corrects the positional deviation of the specific pattern taking into account the influence of point irradiation, thereby obtaining a minimum specific effect.

[0057] [Basic Flow] Fig. 6 shows the basic flow of Example 1. The measurement system of Fig. 5 operates and processes in accordance with this basic flow. One feature of the processing of the flow of Fig. 6 is that it performs two-stage alignment (calculation of the corresponding positional deviation amount): alignment of the entire SEM image in step S5, and local alignment of the periphery of the target pattern in step S7.

[0058] In the first step S1, an SEM image (first image) is taken by the SEM 2 of a specific pattern (e.g., a plug as shown in FIG. 1) of the sample 13 to be measured and inspected before point irradiation. The computer system 1 acquires this SEM image (first image) from the SEM 2. The SEM image taken before point irradiation may be referred to as the first image. The SEM image 121 in FIG. 5 includes data and information of various acquired SEM images corresponding to the SEM image 120 of the SEM 2.

[0059] In the following step S2, the computer system 1 determines and sets, through the input / output unit 103, a target pattern (plug) to be measured and inspected and a position ("point irradiation position") to be point-irradiated with the electron beam 8 on the target pattern in the SEM image (first image) captured in step S1 based on the operation of the user U1. The user U1 may set these on a screen or the like through the input / output device 104 (see FIG. 7, described below). This determines the position coordinates of the target pattern (plug) to be measured and inspected in the SEM image (first image) before point irradiation. The position coordinates of this target pattern are assumed to be (X0, Y0).

[0060] FIG. 7 shows an example of step S2 in which user U1 specifies the position coordinates of a target pattern on a screen with a GUI (graphical user interface). On this screen, user U1 may view an SEM image (first image) before point irradiation and specify a point on the target pattern with a cursor or the like, and set this specified point as the position of the target pattern and the point irradiation position. Alternatively, a position definition (e.g., a center of gravity point) may be provided in advance for each pattern, such as a plug, and the definition may be selected and used. In the example screen shown in FIG. 7 , an SEM image file can be specified in field 701, and the SEM image can be displayed in field 702. User U1 can specify the target pattern in field 702 with a cursor or the like. Furthermore, point irradiation conditions can be set in field 703.

[0061] The point irradiation position may be the same as or different from the position of the target pattern (plug). For example, the center point of the plug's circle is typically set as the position of the target plug, and the point irradiation position is also set as the same position. Furthermore, multiple target plugs may be processed in one processing run. In other words, multiple target plugs can be processed together.

[0062] In step S3, the SEM 2 performs point irradiation on the determined point irradiation position of the target pattern, thereby charging the sample 13 (particularly the target pattern).

[0063] In step S4, an SEM image of the charged sample 13 after point irradiation is taken by the SEM 2. The SEM image taken after point irradiation may be referred to as a second image. The charged state of the sample surface due to point irradiation causes distortion in the SEM image. Therefore, the position coordinates of the target pattern in the SEM image (second image) may deviate from the position coordinates (X0, Y0) determined in steps S1 and S2.

[0064] Therefore, in this first embodiment, in order to obtain a more accurate position of the target pattern after point irradiation, in the subsequent step S5, the computer system 1 aligns the entire two SEM images (the first image in step S1 and the second image in step S4) before and after point irradiation by pattern matching. This calculates the amount of misalignment in the entire SEM image caused by point irradiation. The amount of misalignment in the entire SEM image caused by point irradiation is defined as (ΔX0, ΔY0). The entire SEM image here refers to, for example, an SEM image containing multiple 5 × 5 plugs as shown in FIG. 4. Naturally, the number of specific patterns (e.g., plugs) captured in a single SEM image varies depending on the imaging conditions of the sample 13 and the SEM 2.

[0065] In step S6, the computer system 1 checks and determines whether it is necessary to calculate the amount of misalignment corresponding to the local distortion around the target pattern. This step S6 may be performed by checking whether the function (the function related to step S7) is set to ON (enabled) in the user settings of the measurement system.

[0066] For example, if the amount of charge caused by point irradiation is extremely small, it is assumed that the local distortion of the SEM image is sufficiently small. In such a case, it is not necessary to calculate the amount of misalignment corresponding to the local distortion (S6-NO), and the process proceeds to step S8.

[0067] In this case, in step S8, the computer system 1 adds the amount of misalignment for the entire image calculated in step S5 (first amount of misalignment) to the position coordinates of the target pattern set in step S2. The added position coordinates are (X0 + ΔX0, Y0 + ΔY0). This position is considered to be appropriate as the target pattern position on the SEM image (second image) after point irradiation. Therefore, the computer system 1 determines this position as the measurement / inspection position of the target pattern. In other words, the position where the misalignment has been corrected taking into account the influence of point irradiation is the measurement / inspection position of the target pattern. Note that this position corresponds to, for example, the center of gravity position of the plug region.

[0068] Furthermore, for example, if user U1 has set in advance via input / output unit 103 that precise measurement and inspection results are required (i.e., the function is turned on), it is determined that calculation of the amount of misalignment corresponding to the local distortion around the target pattern is required (S6-YES), and the process proceeds to step S7.

[0069] Alternatively, the confirmation and determination in step S6 can be performed by comparing the feature quantities of the SEM images before and after point irradiation in the vicinity of the target pattern position coordinates on the SEM image (first image) before point irradiation. For example, the computer system 1 sets a square region centered on the target pattern position coordinates on the SEM image before point irradiation and calculates the brightness difference of the SEM images before and after point irradiation within this region. If a brightness difference of a certain level or more occurs, the computer system 1 determines that the image distortion (corresponding positional deviation) is large and determines that it is necessary to calculate the positional deviation corresponding to the local distortion around the target pattern (S6-YES). The feature quantity used in this case is not limited to the brightness of the SEM image, but various feature quantities such as brightness variance and gradient can also be applied.

[0070] In step S7, the computer system 1 cuts out a pre-point irradiation SEM image (first image) around the target pattern (plug). In other words, the computer system 1 cuts out and references a partial image including the periphery of the target pattern (plug) from the first image captured in step S1. The computer system 1 calculates a local misalignment amount by pattern matching between the local image (cut-out image) around the target pattern before point irradiation and the corresponding post-point irradiation image (the entire SEM image or the cut-out image) including the periphery of the target pattern. The local misalignment amount at this time is defined as (ΔX1, ΔY1). The computer system 1 calculates the misalignment amount (ΔX1, ΔY1) by pattern matching the image cut out from the pre-point irradiation SEM image as a template with the post-point irradiation SEM image (e.g., the entire SEM image).

[0071] In this alignment by pattern matching stage of step S7, the misalignment amount of the target pattern has been calculated in advance as a first stage (whole image) by the alignment by pattern matching of the entire SEM image stage of step S5. That is, the first misalignment amount (ΔX0, ΔY0) described above has already been acquired. In step S7, the misalignment amount of the target pattern is additionally calculated in a second stage (local). That is, the second misalignment amount (ΔX1, ΔY1) is obtained. In other words, in this flow, correction of the target pattern position and calculation of the misalignment amount are performed by two-stage alignment, the alignment of step S5 and the alignment of step S7. This enables efficient alignment and correction. If it is determined that calculation of the misalignment amount corresponding to the local distortion around the target pattern is unnecessary (S6-NO), the process including pattern matching of step S7 is not executed.

[0072] 8A and 8B are supplementary explanatory diagrams related to the flow of FIG. 1, showing specific examples of SEM images, misalignment, etc. State A in FIG. 8A shows the concept of plug positions before and after point irradiation, similar to state C in FIG. 4, and the position coordinates of the target plug A7 in the SEM image before point irradiation are shown as (X0, Y0). A8 shows the outline of the target plug in the SEM image after point irradiation. A9 shows the outline of the SEM image before alignment between the SEM images before and after point irradiation.

[0073] State B shows the plug position and image contour after alignment across the entire SEM image in step S5. A10 shows the contour of the SEM image before point irradiation after alignment across the entire image. A11 shows the amount of misalignment across the entire image as (ΔX0, ΔY0).

[0074] 8B shows the plug position and image contour after image cropping in step S7. A12 shows the contour of a cropped image before point irradiation, without alignment around the target plug. This cropping example is for a case where the image size is set to include the first adjacent plug in both the X and Y directions, with the target plug at the center.

[0075] State D in Fig. 8B shows the plug position and image contour after alignment around the target plug in step S7. A13 shows the contour of the cut-out image before point irradiation after alignment around the target plug. A14 shows the amount of local misalignment around the target plug as (ΔX1, ΔY1). A15 shows the position coordinates of the corrected target pattern obtained by adding the amounts of misalignment due to the two-stage alignment in steps S5 and S7 as (X0 + ΔX0 + ΔX1, Y0 + ΔY0 + ΔY1).

[0076] In this example, the entire SEM image (including, for example, a 5x5 plug) is used as the image after point irradiation that is compared with the template (the cut-out image before point irradiation). However, in order to reduce calculation costs, it is also possible to cut out a portion of the SEM image and use it.

[0077] 6 , in step S8, the computer system 1 determines the measurement and inspection position of the target pattern (plug) in the SEM image (second image) after point irradiation, as described above, using the results of steps S5 and S7 (i.e., the amount of misalignment). In other words, the position of the target pattern whose misalignment has been corrected taking into account the influence of point irradiation is measured and calculated.

[0078] In one embodiment, the results up to step S8, i.e., the measurement result information of the position coordinates of the target pattern, may be saved and the process may end. This measurement result can be obtained as more accurate position information when point irradiation is used.

[0079] In step S9, the computer system 1 (particularly the feature calculation unit 112 in FIG. 5 ) sets a measurement / inspection area at the coordinates of the measurement / inspection position of the target pattern, which is the result of step S8. In other words, this measurement / inspection area is the area used to extract features in step S10. Specifically, the computer system 1 places a figure representing the measurement / inspection area at the coordinates of the measurement / inspection position. Note that this setting and placement of the measurement / inspection area does not erase brightness information and the like contained in the SEM image, and does not result in loss of information. The figure representing the measurement / inspection area can be, for example, a figure representing the shape of the target pattern that can be extracted from the SEM image (first image) before point irradiation, such as a circle representing a plug, but is not limited to this.

[0080] In step S10, the computer system 1 (feature calculation unit 112) extracts brightness as a feature from the measurement / inspection area set in step S9. In one embodiment, the results up to this point, i.e., the feature (brightness) information of the target pattern, may be saved and the process may end. In another embodiment, the brightness information may be used to measure the dimensions, circularity, etc. of the target pattern. Dimensions 1301 and 1302 in FIG. 13 are example measurements of the length and width of the diameter of the plug.

[0081] In step S11, the computer system 1 (particularly the inspection unit 113 in FIG. 5 ) performs an electrical defect inspection using the brightness information, which is the feature extracted in step S10. For example, the presence or absence of a defect in the target plug is determined based on a comparison between the brightness and a threshold value. In other embodiments, the brightness information may be used to estimate or calculate electrical resistance or capacitance.

[0082] [Processing in Step S7] FIG. 9 shows details of the processing in step S7, and a flow of a specific processing example.

[0083] The computer system 1 uses a SEM image (first image) cut out around the target pattern (plug) before point irradiation as a template and performs pattern matching with the corresponding SEM image (second image) after point irradiation. In this case, it is necessary to determine and set an appropriate cutout size for the SEM image. From the perspective of the computational cost of pattern matching, a smaller cutout size is preferable. However, if the cutout size is too small, the possibility of pattern matching failure increases, for example, when the brightness of the target pattern changes due to point irradiation.

[0084] Therefore, in the flow of FIG. 9 of the first embodiment, the cutout size is gradually increased from the initial size as needed, thereby setting the minimum necessary cutout size.

[0085] In the first step S71, the computer system 1 cuts out a local image around the target pattern from the SEM image (first image) before point irradiation, using an initial cutout size (first size). The initial cutout size (first size) may be determined and set in advance by the user U1 via the input / output unit 103.

[0086] Alternatively, the computer system 1 may measure the spacing between patterns and automatically determine and set the size, for example, to include up to the Nth adjacent pattern for the target pattern.

[0087] A supplementary explanation of the cutout size will be provided using FIG. 10 . Here, the target plug is assumed to be plug PL0. In the direction in which the patterns are adjacent to plug PL0, for example, on the positive side (+X) of the X direction, the first adjacent plug is plug PL1 and the second adjacent plug is plug PL2. The same can be considered for each direction. Examples of cutout sizes (in other words, cutout image areas) are size 1000, size 1001, size 1002, size 1003, etc. Size 1000 is an image area that includes only a single target plug and its adjacent background area (insulator). Size 1001 is an image area that includes up to the first adjacent plug PL1. Size 1002 is an image area that includes up to the second adjacent plug PL2. Size 1003 is an image area that includes up to the third adjacent plug PL3. Spacing T11 is the spacing between plug PL0 and the first adjacent plug PL1, and has a fixed set value in design.

[0088] In step S72, the computer system 1 uses the extracted SEM image as a template and performs pattern matching with the corresponding SEM image after point irradiation (here, the entire SEM image).

[0089] In step S72 (or more broadly, step S7), the second image to be compared with the template (a cut-out image of the first image) is, for example, the entire SEM image. However, as a modified example, an image portion cut out from the SEM image after point irradiation, having a size approximately equivalent to the periphery of the target pattern, may be used as the second image to be compared, similar to the cut-out image of the first image.

[0090] In the next step S73, the computer system 1 calculates the local positional deviation amount (ΔX1, ΔY1) based on the result of the pattern matching.

[0091] In step S74, the computer system 1 determines whether the pattern matching in step S72 was successful. If it was successful (S74-YES), this flow ends, and the positional deviation amount obtained by this pattern matching is adopted as the local positional deviation amount (ΔX1, ΔY1) around the target pattern.

[0092] On the other hand, if the result is unsuccessful (S74-NO), in step S75 the computer system 1 increases the cutout size. The specific extent to which the cutout size should be increased (cutout size increase amount) may be set in advance by the user U1. Alternatively, a method in which the computer system 1 determines the cutout size using the spacing between patterns or the like so that the next adjacent pattern is also included is also applicable.

[0093] An example of expanding or changing the cutout size is as follows, using Fig. 10. Assume that the initial cutout size is, for example, size 1001 in Fig. 10, which includes up to the first adjacent pattern. If the first pattern matching fails, the cutout size for the next process is expanded to size 1002, which includes up to the second adjacent pattern.

[0094] The steps from step S71 onward are repeated using the expanded cutout size in step S75, gradually expanding the cutout size until pattern matching is successful. Note that an upper limit may be set for the number of attempts to expand the cutout size. If pattern matching fails even after the maximum number of attempts, the computer system 1 ends this flow with an error.

[0095] In step S7, when using a local image portion (cut-out image) around the target pattern, either an image containing only the target pattern (e.g., an image portion of size 1000 in FIG. 10 ) or an image containing up to the Nth pattern adjacent to the target pattern in the direction (e.g., the X and Y directions) where other patterns are adjacent to the target pattern can be used. However, if the difference between the images before and after point irradiation is too large, alignment by pattern matching may become difficult and fail if the size of the cut-out image is small. Therefore, in this embodiment 1 ( FIG. 9 ), the cut-out size is varied and trialed, for example, by gradually enlarging it from the initial size. This allows for a good balance between the success and stability of alignment and the computational cost.

[0096] Furthermore, in the first embodiment, the cutout size of the template image is initially set to a minimum and gradually increased with each trial, giving priority to the viewpoint of calculation cost. However, this is not limited to this. In a modified example, the cutout size can be set to a maximum initially and gradually reduced with each trial. For example, in the case of FIG. 10, the initial size is set to size 1003, and in the next trial, it is reduced to size 1002. Compared to the embodiment (FIG. 9), the modified example requires a slightly higher calculation cost, but the larger template image improves the stability of pattern matching.

[0097] [Specific Example] Figure 11 is a supplementary explanatory diagram regarding the flow of Figure 9, and particularly shows an example of cutting out. State A shows an example of a method for cutting out a local image from an SEM image before point irradiation. A21 is a case where the cutout size includes only a single target plug (plug A1). A22 is a case where the cutout size includes the target plug and a first adjacent plug in the X and Y directions. A23 is a case where the cutout size includes the target plug, a first adjacent plug, and a second adjacent plug in the X and Y directions.

[0098] In addition, if the area corresponding to the cut-out size exceeds the outline of the original SEM image depending on the position of the target plug within the SEM image, the cut-out area will be within the outline of the original SEM image, as shown in the figure.

[0099] State B shows a case where pattern matching is performed using a cut-out image A21 having a size that includes only the target plug.

[0100] Returning to FIG. 6 , after calculating the local misalignment amount (ΔX1, ΔY1) in step S7, the computer system 1 calculates and determines the position coordinates of the target pattern on the SEM image (second image) after point irradiation in the following step S8. The computer system 1 uses the misalignment amount (ΔX0, ΔY0) obtained by pattern matching of the entire SEM image before and after point irradiation in step S5 and the local misalignment amount (ΔX1, ΔY1) around the target pattern obtained in step S7. In step S8, the computer system 1 adds the misalignment amount (ΔX0, ΔY0) of the entire image and the local misalignment amount (ΔX1, ΔY1) to the position coordinates (X0, Y0) of the target pattern on the SEM image (first image) before point irradiation. This results in the measurement / inspection position (X0 + ΔX0 + ΔX1, Y0 + ΔY0 + ΔY1) of the target pattern on the SEM image (second image) after point irradiation.

[0101] As mentioned above, if the calculation of the local positional deviation amount in step S7 is not performed, in step S8, only the positional deviation amount (ΔX0, ΔY0) of the entire image obtained in step S5 is added together to determine the measurement / inspection position (X0+ΔX0, Y0+ΔY0) of the target pattern on the SEM image after point irradiation.

[0102] In this first embodiment, a case has been shown in which two-stage alignment is performed, in which pattern matching is performed on the entire SEM image (step S5) and then pattern matching is performed around the target plug (step S7), and the corresponding amount of misalignment is calculated. However, the present invention is not limited to this, and a configuration in which alignment and misalignment calculation are performed in more stages by varying the size of the image to be processed, etc., is also possible (modifications described later).

[0103] In step S9, a circle is placed as a graphic representing the measurement / inspection area at the coordinates of the measurement / inspection position of the target pattern (plug) determined in step S8. In this embodiment 1, a graphic such as a circle is placed instead of the outline of the target plug as a graphic representing the measurement / inspection area because there is a possibility that the outline of the target plug will disappear due to point irradiation. For example, if the brightness of the target plug at the point irradiation position shown by A1 in state B of Figure 4 becomes the same brightness as the surrounding area (black in the figure) after point irradiation, the outline of the plug cannot be extracted.

[0104] In this first embodiment, since the shape of the plug to be measured and inspected is a circle when viewed from the sample surface, a circle is selected and adopted as the figure representing this measurement and inspection region. This is not limiting, and other figures, such as a polygon or a ring, may be adopted depending on the target pattern. Furthermore, the size of the figure (e.g., a circle) of the measurement and inspection region may be a value that is preset based on design data, or may be the measured value of the dimensions of other plugs in the SEM image or a statistical value of measurements of multiple plugs.

[0105] FIG. 12 is a supplementary explanatory diagram regarding step S9. FIG. 12 shows an example of the arrangement of circles representing distortion of the target plug and the measurement / inspection region. In the example of FIG. 12, plug 1201 indicates the original shape of the target plug region before point irradiation. Plug 1202 indicates the shape of the plug region in which distortion has occurred due to charging after point irradiation. Point p11 indicates the plug position (e.g., center of gravity) before point irradiation corresponding to the distorted plug 1202. Point p12 indicates the plug position after the positional deviation corresponding to the distortion has been corrected. Circle C0 is a circle representing the measurement / inspection region arranged at the corrected plug position p12.

[0106] In step S10, the computer system 1 acquires brightness information of the SEM image (second image) after point irradiation within the measurement / inspection region (for example, circle C0) set in step S9.

[0107] 13 is a supplementary explanatory diagram regarding step S10, and corresponds to the example of FIG. 12. Brightness is acquired from each pixel included in the circle C0 of the measurement / inspection area. By using the circle C0 of the measurement / inspection area at the corrected position (point p12) where the influence of distortion has been eliminated, brightness information can be obtained more appropriately.

[0108] The results of the above measurements are displayed in the measurement result field 704 on the screen of FIG. 7, for example, and can be confirmed by the user U1.

[0109] In this first embodiment, by using the above-described series of steps, it is possible to accurately set the position and area to be measured and inspected, and extract and acquire the feature value (brightness) even from an SEM image of a sample that has been point-irradiated with a specific pattern. By using this brightness information, it is possible to perform electrical defect inspection more accurately.

[0110] In this Example 1 (FIG. 6), the case where the SEM image before point irradiation is cut out and used as a template is shown, but it is also possible to cut out the SEM image after point irradiation and calculate the positional deviation amount of the target plug.

[0111] [Effects (1)] According to the first embodiment, the position of a specific pattern can be more accurately acquired and measured from an image of a semiconductor device captured by a charged particle beam device such as an SEM, even when the specific pattern is point-irradiated. In the first embodiment, the target pattern is subjected to point irradiation for electrical characteristic inspection of the specific pattern (e.g., a plug), and the electrical characteristics and presence or absence of defects of the target pattern can be determined from the brightness of the SEM image after point irradiation. However, local distortion may occur in the SEM image after point irradiation due to the influence of charging. In this case, the first embodiment calculates the amount of positional deviation of the target pattern between the SEM images before and after point irradiation. This makes it possible to obtain a more accurate position as the measurement / inspection position of the target pattern in the SEM image after point irradiation. In particular, in the first embodiment, as shown in FIG. 6 , the position of the target pattern can be efficiently obtained after correcting the influence of distortion due to point irradiation by performing two-stage alignment (steps S5 and S7) and controlling the crop size (step S7).

[0112] In this embodiment 1, a case where an SEM 2 is applied as a charged particle beam device (in other words, an imaging device, an image acquisition tool) is shown, but this is not limitative and a focused ion beam (FIB) device or the like can also be applied in the same way.

[0113] [Modifications] The following modifications of the first embodiment are also possible.

[0114] In the first embodiment, the flow ends at step S11 in FIG. 6 . After the flow ends, the measurement system does not need to move the SEM 2 to capture another SEM image. That is, the SEM images used for pattern measurement and inspection (steps S10 and S11) have already been captured as the first image in step S1 and the second image in step S4, and there is no need to capture them again. This is not a limitation. In a modified example, after the flow ends, the measurement system may move the SEM 2 to capture another SEM image of the same target pattern and perform measurement and inspection using the captured SEM image. In this case, the measurement system uses the accurate measurement and inspection position, etc. obtained in step S8, etc., to capture one or more SEM images again using the SEM 2 (point irradiation is not necessarily performed at this time), and then performs measurement and inspection (e.g., dimension measurement or electrical characteristic inspection) using the SEM image.

[0115] As another modification, the process of calculating the amount of misregistration due to alignment of the entire image in step S5 in FIG. 6 may be omitted, and the process of calculating the amount of misregistration due to local alignment in step S7 may always be performed.

[0116] As another modification, the process of calculating the amount of misregistration due to alignment of the entire image in step S5 of FIG. 6 may be always performed, and the process of calculating the amount of misregistration due to local alignment in step S7 may be omitted.

[0117] 9, instead of gradually increasing the image clipping size from the initial size as pattern matching is attempted, the image clipping size may be gradually reduced from the initial size as pattern matching is attempted, which may change the number of attempts required until pattern matching is successful.

[0118] As a modified example of performing alignment and misregistration calculation in multiple stages, the following is possible. The above-mentioned flow ( FIG. 6 ) is an example of a configuration in which alignment and misregistration calculation are performed in two stages, where the comparison image region in the first stage (step S5) is the entire SEM image, and the comparison image region in the second stage (step S7) is a local region (variable as a cut-out image) around the target plug. This is not limiting, and various configurations are possible regarding the number of stages and the size of the comparison image region.

[0119] FIG. 22 shows a specific example of an SEM image in this modified example. SEM image 2200 shows, for example, a 10×10 plug. Plug PL0 is an example of a target plug. In a first step, the measurement system sets a region 2201 (e.g., a size that includes up to the third adjacent plug) as a first region within the entire SEM image 2200, and uses region 2201 to align images (the first image and the second image) before and after point irradiation and calculate a misalignment amount (first misalignment amount). In a second step, the measurement system sets a region 2202 (e.g., a size that includes up to the first adjacent plug) smaller than region 2201 as a second region within the entire SEM image 2200, and uses region 2202 to align images (the first image and the second image) before and after point irradiation and calculate a misalignment amount (second misalignment amount). The measurement system determines the measurement and inspection position of target plug PL0 by adding the first misalignment amount and the second misalignment amount. The first region may be set to the entire SEM image, in which case it will be the same as in Fig. 6. When processing the first region and the second region, the crop size variable control as in Fig. 9 may be similarly applied.

[0120] In another example, in a first stage, the measurement system performs alignment using the entire SEM image 2200 as a first region and calculates a first misalignment amount. In a second stage, the measurement system performs alignment using a region 2201 of the entire SEM image 2200 as a second region and calculates a second misalignment amount. In a third stage, the measurement system performs alignment using a region 2202 as a third region and calculates a third misalignment amount. The measurement system determines the measurement / inspection position of the target plug PL0 by adding the first misalignment amount, the second misalignment amount, and the third misalignment amount.

[0121] 23 shows a specific example of an SEM image in another modified example. When an image area is cut out at a predetermined size centered on the target pattern, as in the state A of FIG. 11 described above, the image area may extend beyond the entire range of the SEM image. In the case of the state A of FIG. 11, if the cutout size is set to A23 for the target plug A1, the image area will extend beyond the entire range of the SEM image in the right or upward direction. Even in the case of the state A of FIG. 11, for example, there is an image area that includes a predetermined number of adjacent plugs in the left or downward direction of the target plug A1, so alignment is possible using the image area (the area within the entire range of the SEM image).

[0122] On the other hand, as in the modified example of FIG. 23 , when an image region is cut out from an SEM image, it may be cut out to a predetermined size so as to fit within the entire SEM image, rather than centered on the target pattern. In other words, the predetermined size may be set by extending the image region in at least one direction from the target pattern so as to ensure coverage up to the Nth adjacent plug. In this example, the target plug PL0 is located in the upper right corner of the entire SEM image 2300 (a 6×6 plug array). For example, if an image region is cut out with a size centered on the target plug in the X and Y directions and extending up to the first adjacent plug in each direction, it will result in a region 2301, which exceeds the range of the SEM image 2300. If an image region is cut out with a size that includes up to the second adjacent plug, it will result in a region 2302, which exceeds the range of the SEM image 2300.

[0123] On the other hand, when it is desired to take an image region of a predetermined size within the range of the SEM image 2300, it may be set, for example, as region 2303 or region 2304 shown at the bottom. Region 2303 is set to be the same size as region 2301, including from target plug PL0 to the second adjacent plug in the X-left direction and Y-downward direction. Region 2304 is set to be the same size as region 2302, including from target plug PL0 to the fourth adjacent plug in the X-left direction and Y-downward direction. Even when such image regions are used, alignment is possible.

[0124] <Example 2> Example 2 will be described. The basic configuration of Example 2 etc. is the same as or common to Example 1, and the following mainly describes the components of Example 2 etc. that are different from Example 1. Example 2 describes an example of a measurement sequence using a design drawing (in other words, design data) of a semiconductor circuit, which is a sample.

[0125] 14 shows a basic flow of the measurement system of Example 2. In the first step S21, similar to step S1 of Example 1, an SEM image (first image) of the sample 13 before point irradiation is taken by the SEM 2.

[0126] In step S22, the computer system 1 aligns the SEM image (first image) captured in step S21 with design data (design drawing) of the target sample 13. The design data (design drawing) is, for example, CAD data in a known GDS format or the like, and is data that represents the semiconductor circuit (including a specific pattern) of the sample 13 as a set of shapes such as rectangles. The method for acquiring the design data is not limited. For example, as shown in FIG. 5 , the computer system 1 may store design data related to the specific pattern of the target sample 13 in memory 102 in advance. Alternatively, the computer system 1 may acquire the design data as needed from an external server or system via communication.

[0127] In step S23, the computer system 1 selects and specifies a target pattern (e.g., a plug) and a point irradiation position for measurement and inspection on the design data in step S22 through the input / output unit 103 based on the operation of the user U1. For example, a design drawing including a specific pattern on the sample surface or an image in which the design drawing and the SEM image before point irradiation are superimposed may be displayed on the screen of the input / output device 104, and the user U1 may select and specify the target pattern from the design drawing. In step S2 of the first embodiment, the target pattern was selected on the SEM image before point irradiation (first image) ( FIG. 7 ). In the second embodiment, however, a design figure (e.g., a rectangle representing a plug) corresponding to the target pattern is selected on the design drawing. The computer system 1 obtains, for example, a centroid position as a representative point in the design figure of the target pattern selected in step S23. That is, in the second embodiment, the centroid position as a representative point of this design figure is used as the target pattern position. The centroid position as a representative point, which is the target pattern position, is set to (X0, Y0).

[0128] The point selected as the representative point of the design figure as the target pattern position may be any coordinate that can correspond one-to-one with the design figure corresponding to the target pattern, and is not limited to the center of gravity. Also, the order of steps S22 and S23 may be reversed.

[0129] FIG. 15 is a supplementary explanatory diagram of step S23, etc. In state A, a design drawing image 1500 of the design data (design drawing) of the sample 13 has a rectangle D0, which is a design figure representing a plug. For the sake of explanation, the actual plug shape is illustrated superimposed with a dashed line. If the target plug is, for example, PL0, the position of the target plug PL0 is the center of gravity (X0, Y0) of the rectangle D0, with the representative point p0. The center of gravity (X0, Y0) corresponds to the center point of the rectangle D0. In the design drawing, the rectangle D0 has a horizontal width h1, a vertical width h2, etc. Furthermore, the arrangement of multiple rectangles has intervals T1, T2, etc. In another example, the position of the target plug PL0 may be the upper left point of the rectangle D0 of the design figure.

[0130] State B is an example of another pattern. A design drawing image 1501 has a rectangle E0 as a design figure representing a plug. A representative point p0 of the target plug PL0 is, for example, the center of gravity position (X0, Y0) of the rectangle E0.

[0131] In step S24, the target pattern selected in step S23 is subjected to point irradiation by the SEM 2. The point irradiation position is the same as the center of gravity position as a representative point set inside or near the outside of the design figure corresponding to the target pattern selected in step S23. As in the first embodiment, the point irradiation position may be the same as or different from the center of gravity position, which is the target pattern position.

[0132] In step S25, similarly to step S4 in the first embodiment, an SEM image (second image) after point irradiation is taken by the SEM 2.

[0133] In step S26, the computer system 1 cuts out the design data (design drawing) around the target pattern position (here, the cut-out design drawing image is referred to as the first image), and aligns the cut-out design drawing image (first image) with the SEM image (second image) taken in step S25 after point irradiation by pattern matching. In step S5 of the first embodiment, alignment was performed across the entire image between the SEM images before and after point irradiation, but the flow of this second embodiment shows a case where such alignment across the entire image is not performed.

[0134] Furthermore, in Example 1, local alignment in step S7 was performed only when it was determined that calculation of the amount of local misalignment around the target pattern was necessary (step S6-YES), but in the flow of Example 2, local alignment around the target pattern in step S26 is always performed.

[0135] In addition, in step S7 of the first embodiment, a portion of the SEM image (first image) before point irradiation was cut out and aligned, but in the flow of the second embodiment, a design drawing related to the target pattern is cut out and aligned. In this step S26, the amount of positional deviation of the target pattern due to point irradiation is calculated. This amount of positional deviation is defined as (ΔX, ΔY).

[0136] In step S27, the computer system 1 adds the positional deviation (ΔX, ΔY) calculated in the previous step S26 to the position (X0, Y0) of the target pattern before point irradiation, thereby calculating the measurement / inspection position (X0+ΔX, Y0+ΔY) of the target pattern in the SEM image (second image) after point irradiation.

[0137] In step S28, the computer system 1 sets a measurement / inspection area for the measurement / inspection position obtained in step S27. Specifically, the computer system 1 places a figure representing the measurement / inspection area at the measurement / inspection position (X0 + ΔX, Y0 + ΔY). As with step S9 in the first embodiment, the figure to be placed is not limited. The figure to be placed may be a design figure corresponding to the target pattern based on the design data, or a figure obtained by modifying the design figure.

[0138] In step S29, the computer system 1 extracts brightness as a feature amount using the measurement / inspection area of ​​the target pattern. In step S30, the computer system 1 performs an electrical characteristic inspection using the brightness information of the target pattern.

[0139] [Specific Example] Figure 16 is a supplementary explanatory diagram regarding the flow of Figure 14. State A shows the SEM image (first image) before point irradiation in step S22, and design data (design drawing image) aligned with the SEM image (first image) before point irradiation. A9 is the outline of the SEM image before point irradiation. A1 is an example of a target plug for point irradiation. A4 is the outline of the plug in the SEM image before point irradiation. A7 is the target plug in the SEM image before point irradiation. A18 is a design figure (e.g., a rectangle) representing the plug in the design drawing aligned with the SEM image before point irradiation. A19 is the design figure (e.g., a rectangle) of the target plug.

[0140] State B shows the SEM image (second image) after point irradiation in step S25 and the corresponding design data (design drawing image). The plug position in the SEM image (second image) after point irradiation is shifted from the position before point irradiation. A5 is the outline of the plug in the SEM image (second image) after point irradiation. A8 is the outline of the target plug in the SEM image (second image) after point irradiation.

[0141] State C shows the SEM image (second image) after point irradiation and the design image cut out around the target plug position in step S26. A20 is the outline of the design image cut out to include the target plug.

[0142] State D shows the SEM image (second image) after point irradiation and the design image after alignment around the target plug position in step S26. A21 is the design figure (rectangle) of the target plug aligned with the SEM image (second image) after point irradiation. A22 is the design figure (rectangle) of the target plug not aligned with the SEM image (second image) after point irradiation.

[0143] [Effects (2)] According to the second embodiment, it is possible to obtain the same effects as those of the first embodiment. In the second embodiment, it is possible to efficiently perform processing by utilizing the design data.

[0144] The following modifications of the second embodiment are also possible. In the flow of FIG. 14 of the second embodiment, only the misregistration amount is calculated by local alignment around the target pattern in step S26, but this is not limiting. In a modification, the calculation of the misregistration amount by alignment in two stages (steps S5 and S7) may be applied, as in the first embodiment. Alternatively, a configuration in which alignment and misregistration calculation are performed in more stages is also possible. Furthermore, in a modification, the processing may be branched depending on the confirmation such as in step S6, as in the first embodiment.

[0145] Furthermore, as a modification of the second embodiment, a flow may be adopted in which the capturing of the SEM image (first image) before point irradiation in step S21 and the alignment of the SEM image before point irradiation with the design data in step S22 are omitted. Even if there is no SEM image (first image) before point irradiation, as long as there are design data and an SEM image (second image) after point irradiation, the position coordinates (X0, Y0) of the pattern to be measured and inspected (e.g., a plug) and the point irradiation position can be obtained through steps S26 and S27.

[0146] In Example 2, by performing steps S21 and S22, a more detailed electrical characteristic inspection can be performed in step S30. In this case, for example, information on the change in brightness between the SEM image (first image) before point irradiation and the SEM image (second image) after point irradiation can be obtained. Knowing such a change in brightness is important for electrical characteristic inspection. Therefore, in Example 2, step S21 is always performed. However, depending on the inspection content, there may be cases where it is sufficient to capture only the image after point irradiation, and therefore the above-mentioned modified example can be applied.

[0147] Third Embodiment A third embodiment will be described below. In the third embodiment, non-rigid registration is used as a registration method for calculating the amount of misregistration.

[0148] 17 shows the basic flow of the measurement system of Example 3. The initial steps S31 and S32 are similar to steps S1 and S2 of Example 1, and the position coordinates (X0, Y0) of the pattern to be measured and inspected (e.g., a plug) and the point irradiation position are obtained.

[0149] In step S33, the computer system 1 extracts the contour of the target pattern from the SEM image (first image) before point irradiation. As a specific processing example, the contour can be extracted by applying an image processing filter, such as a differential filter that calculates the brightness difference between adjacent pixels, to the SEM image.

[0150] 18 is a schematic explanatory diagram of an example of extracting the contour of a target pattern from the SEM image (first image) before point irradiation in step S33. As a result of applying a filter to the target plug PL0 in the image, a contour 1801 is extracted.

[0151] Steps S34 and S35 are similar to steps S4 and S5 in the first embodiment, and an SEM image (second image) after point irradiation is obtained.

[0152] In step S36, the computer system 1 performs non-rigid alignment processing between the SEM images (first image and second image) before and after point irradiation in order to calculate the amount of local positional deviation of the SEM image (second image) caused by point irradiation.

[0153] Non-rigid registration is a well-known technique that involves image deformation, and refers to a registration method using affine transformation or non-linear transformation. The non-rigid registration process includes a first process of estimating the amount of misalignment between a first image and a second image, and a second process of calculating the validity of the estimated amount of misalignment.

[0154] As a result of the non-rigid alignment process in step S36, a misregistration amount corresponding to the local distortion around the target pattern position is calculated between the SEM images before and after point irradiation. The calculated misregistration amount is represented as (ΔX, ΔY).

[0155] In step S37, the computer system 1 calculates the measurement / inspection position (X0+ΔX, Y0+ΔY) of the target pattern based on the local positional deviation amount calculated in the previous step S36, as in the second embodiment.

[0156] In step S38, the computer system 1 places the contour line of the target plug extracted in step S33 as a figure representing the measurement / inspection area at the measurement / inspection position (X0+ΔX, Y0+ΔY) determined in step S37.

[0157] When an arbitrary figure is set as a measurement / inspection area at a measurement / inspection position of a target pattern as in the first and second embodiments, the measurement / inspection area will have a different shape and size from the actual target pattern. That is, if the area is smaller than the actual target pattern, it will be impossible to measure and acquire all of the brightness inside the target plug area. Also, if the area is larger than the actual target plug, the brightness outside the target plug area will also be measured and acquired.

[0158] In Example 3, the contour line of the target pattern extracted from the SEM image (first image) before point irradiation is used as the measurement / inspection area, making it possible to measure and obtain all brightness within the actual target pattern area, excluding the outside.

[0159] In this Example 3, the contour lines of the target pattern are extracted from the SEM image (first image) before point irradiation and used, but the contour lines of the target pattern may be extracted from the SEM image (second image) after point irradiation and used. Furthermore, the extracted contour lines may be enlarged or reduced as appropriate, and the resulting processed figure may be used in the measurement / inspection area.

[0160] Furthermore, if the contour line cannot be extracted in the processing of step S33, another method may be used as an alternative, for example, a figure representing the measurement / inspection area may be placed as in Example 1, or a design figure may be placed as in Example 2.

[0161] In step S39, the computer system 1 measures and acquires brightness as a feature amount from the measurement / inspection area of ​​the target pattern, similarly to Example 1. In step S40, the computer system 1 performs an electrical characteristic inspection using the acquired brightness information.

[0162] [Specific Example] Figure 19 is a supplementary explanatory diagram regarding non-rigid alignment and the like in Example 3. State A is an SEM image (first image) before point irradiation. State B is an SEM image (second image) after point irradiation. A1 (A1b) is an example of a target plug. Point irradiation causes non-uniform distortion within the surface of the SEM image. State C shows the plug positions before and after point irradiation superimposed. The grid lines in the figure represent the arrangement of multiple plugs (plug positions) before point irradiation. In State C, the grid lines intersect perpendicularly to each other. State D shows the plug positions in the SEM images before and after point irradiation after non-rigid alignment in step S36. In State D, the grid lines are curved due to non-rigid alignment.

[0163] [Effects (3)] According to the third embodiment, it is possible to obtain the same effects as those of the first and second embodiments. In the third embodiment, it is possible to preferably realize measurement and inspection by using the pattern contour line extracted from the SEM image before point irradiation and non-rigid alignment.

[0164] <Fourth Embodiment> A fourth embodiment will be described below. The fourth embodiment is characterized by a method for calculating the amount of local positional deviation around the target pattern.

[0165] [Basic Flow] Figure 20 shows the basic flow of the measurement system of Example 4. Steps S41 to S45 are the same as steps S1 to S5 of Example 1. Up to step S45, SEM images (first image, second image) before and after point irradiation, position coordinates (X0, Y0) of the pattern to be measured / inspected, and positional deviation amounts (ΔX0, ΔY0) before and after point irradiation are obtained. Step S45 is an example of alignment of the entire SEM image.

[0166] In step S46, the computer system 1 divides the SEM image (first image) before point irradiation into images of a certain size. The division size may be set in advance by the user U1 via the input / output unit 103. Alternatively, the computer system 1 may automatically set the division size by measuring the intervals between specific patterns or by acquiring the intervals from the design data and using the intervals to set the cutout size such that each cutout image (in other words, divided image) contains one specific pattern.

[0167] In step S47, the computer system 1 aligns, by pattern matching, the images around the position coordinates of the target pattern among the images (divided images) cut out by division in step S46 with the SEM image (second image) after point irradiation, thereby calculating the amount of positional deviation corresponding to the local distortion around the position coordinates of the target pattern.

[0168] In step S48, the local positional deviation amount (ΔX1, ΔY1) in the position coordinates of the target pattern is calculated by interpolating the positional deviation amounts of the surrounding (e.g., adjacent) patterns, thereby determining the measurement and inspection position of the target pattern in the SEM image (second image) after point irradiation.

[0169] Steps S49, S50, and S51 are the same as steps S9, S10, and S11 in the first embodiment.

[0170] In this fourth embodiment, the SEM image (first image) before point irradiation is divided in step S46, but this is not limited to this, and it is also possible to divide the SEM image (second image) after point irradiation and calculate the amount of local positional deviation.

[0171] 21A and 21B are supplementary explanatory diagrams regarding the flow of Example 4. State A in FIG. 21A shows the state in which the SEM image (first image) before point irradiation in step S46 is divided. In this example, the SEM image (first image) is divided vertically and horizontally into nine equal parts of 3 x 3. A41 is an example of a division size. A1 is an example of a target plug.

[0172] State B shows the coordinates of the center of each divided image obtained by dividing, and the coordinates of the position of the pattern to be measured and inspected. A42 is an example of a divided image (in other words, a divided area), and each divided image is divided into g ｉｊ A43 is an example of the coordinates of the center of the divided image, and each coordinate of the center is represented by point P ｉｊ The position coordinate point of the pattern (A1) to be measured and inspected is represented as Pc (Xc, Yc).

[0173] State C in FIG. 21B is a conceptual diagram of a case where the amount of local positional deviation in the target pattern position coordinates in step S48 is calculated using a bilinear interpolation method. Bilinear interpolation is a method of performing linear interpolation in two directions, vertically and horizontally (in this example, the X and Y directions shown in the figure). The position of each divided image (point P ｉｊ ) is the amount of positional deviation at (ΔX ｉｊ , ΔY ｉｊ ) is indicated.

[0174] In states A, B, and C, images before equally divided point irradiation are used, but it is also possible to use cut-out images showing plugs near the target plug (in other words, adjacent plugs), as in state D. A44a, A44b, A44c, and A44d are examples of cut-out images of the periphery of the plugs, for four plugs that are vertically and horizontally near the target plug A1. Interpolation is possible using these cut-out images.

[0175] In state D, the image after division by cutting out shows one target plug and one nearby plug, but a method of cutting out an image of a larger size is also applicable. Also, a method of repeatedly trying pattern matching by changing the cutout size, as in steps S71 to S75 of Fig. 9 in Example 1, is also applicable.

[0176] In this Example 4, the SEM image (first image) before point irradiation was divided, but this is not limited to this, and it is also possible to divide the SEM image (second image) after point irradiation to calculate the amount of local positional deviation.

[0177] [Effects (4)] According to the fourth embodiment, it is possible to obtain the same effects as those of the first, second, and third embodiments. In the fourth embodiment, measurement and inspection can be suitably performed using the divided images and the interpolation method.

[0178] Although the embodiments of the present disclosure have been specifically described above, they are not limited to the above-described embodiments and can be modified in various ways without departing from the spirit of the present disclosure. Except for essential components, components can be added, deleted, or replaced in each embodiment. Unless otherwise specified, each component may be singular or plural. A combination of each embodiment and its variations is also possible.

[0179] 1...computer system, 2...SEM, 13...sample, 101...processor, 102...memory, 111...position measurement unit, 112...feature amount calculation unit, 113...inspection unit

Claims

1. A measurement system comprising a computer system that measures a pattern formed on a semiconductor sample by performing image processing based on image data obtained by a charged particle beam device that irradiates a semiconductor sample with a charged particle beam and captures an image, wherein the computer system acquires a first image captured before point-irradiating the pattern to be measured with the charged particle beam and a second image captured after point-irradiating the pattern with the charged particle beam, performs processing to calculate a positional deviation amount regarding the position of the pattern to be measured between the first image and the second image, and determines the position of the pattern to be measured based on the calculated positional deviation amount.

2. A measurement system according to claim 1, wherein a measurement area is set at the determined position of the pattern to be measured, and the pattern is measured based on the measurement area.

3. A measurement system according to claim 2, wherein brightness information is acquired as a feature amount from the measurement area in measuring the pattern.

4. A measurement system according to claim 2, wherein the pattern measurement comprises calculating dimensions of the pattern from the measurement area.

5. A measurement system according to claim 3, wherein the brightness information acquired from the measurement area is used to measure the pattern, and an electrical characteristic inspection of the pattern is performed.

6. A measurement system according to claim 1, wherein the process of calculating the amount of misalignment comprises: a first process of calculating a first amount of misalignment across the entire image between the first image and the second image; a second process of calculating a second amount of misalignment in a local image region around the pattern within the image between the first image and the second image; and a process of adding together the first amount of misalignment and the second amount of misalignment.

7. A measurement system according to claim 1, wherein the process of calculating the amount of positional deviation is performed in multiple stages, and includes at least a first process of setting a first region including the pattern to be measured within the entire range of the first image and the second image, and calculating a first amount of positional deviation in the first region; a second process of setting a second region smaller than the first region including the pattern to be measured, and calculating a second amount of positional deviation in the second region, between the first image and the second image; and a process of adding together the first amount of positional deviation and the second amount of positional deviation.

8. A measurement system according to claim 1, wherein the process of calculating the amount of positional deviation is performed using an image area in the first image and the second image that includes at least the pattern that is the object of the measurement and is closest to the Nth pattern in the direction in which other patterns are adjacent.

9. A measurement system according to claim 1, wherein the process of calculating the amount of positional deviation includes a process of calculating the amount of positional deviation in a local image region around the pattern in the image between the first image and the second image, and the process of calculating the amount of positional deviation in the local image region is performed by aligning the first image with the second image using an image region in the first image and the second image that includes up to the Nth closest pattern in the direction in which other patterns are adjacent to the pattern that is the target of the measurement, and the size of the image region is gradually enlarged from an initial size depending on failure of the alignment.

10. A measurement system according to claim 1, wherein the process of calculating the amount of positional deviation includes a process of calculating the amount of positional deviation in a local image region surrounding the pattern in the image between the first image and the second image, and the process of calculating the amount of positional deviation in the local image region is performed by aligning the first image with the second image using an image region in the first image and the second image that includes up to the Nth closest pattern in the direction in which other patterns are adjacent to the pattern to be measured, and the size of the image region is gradually reduced from its initial size depending on failure of the alignment.

11. A measurement system according to claim 1, wherein design data of the pattern is used in place of the first image in the process of calculating the amount of positional deviation.

12. A measurement system according to claim 2, wherein the measurement area is set based on a design figure in design data of the pattern.

13. A measurement system according to claim 2, wherein the measurement area is set based on the contour line of the pattern obtained from the first image or the second image.

14. A measurement system according to claim 1, wherein the process of calculating the amount of positional deviation is performed based on the results of a non-rigid registration process between the first image and the second image.

15. A measurement system according to claim 1, wherein when determining the position of the pattern to be measured, the position of the pattern to be measured is calculated by an interpolation method based on the amount of positional deviation of other patterns in the vicinity of the pattern in question.

16. A measurement method executed by a computer system that measures a pattern formed on a semiconductor sample by performing image processing based on image data obtained by a charged particle beam device that irradiates a semiconductor sample with a charged particle beam and captures an image, the measurement method comprising the steps of: the computer system acquiring a first image captured before point-irradiating the pattern to be measured with the charged particle beam, and a second image captured after point-irradiating the pattern with the charged particle beam; performing processing to calculate a positional deviation amount related to the position of the pattern to be measured between the first image and the second image; and determining the position of the pattern to be measured based on the calculated positional deviation amount.

Images