Observation system, observation method, and program

Abstract

To provide an observation system capable of observing a formation position of a target shape that cannot be directly irradiated with an electron beam.SOLUTION: An observation system 1 includes an electron microscope 100 and a calculator 200. Here, the electron microscope 100 radiates an electron beam 103 to a first surface position on a sample 108, which is different from a formation position of a target shape on the sample 108, scatters the inside of the sample 108 from the first surface position, detects a predetermined electron emitted from the forming position of the target shape to the outside of the sample 108, and performs output as a detection signal, and the calculator 200 outputs one or more values related to the target shape on the basis of the detection signal.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an observation system, an observation method, and a program, and more particularly to an observation system, an observation method, and a program for performing observation based on an image captured by an electron microscope (electron microscope image), for example. [Background technology]

[0002] A technique for observing a sample using an electron microscope is described, for example, in Patent Document 1. That is, Patent Document 1 describes a technique for estimating the depth of a pattern on a sample irradiated with primary electrons by a primary electron beam irradiation means provided in the electron microscope. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2019-185972 Summary of the Invention [Problem to be solved by the invention]

[0004] In Patent Document 1, in order to observe the depth of a pattern formed on a sample, an electron beam is directly irradiated onto a location to be observed, and the depth of the pattern is observed from a signal detected by the irradiation of the electron beam. In other words, the technique described in Patent Document 1 has a problem in that when a location to which an electron beam cannot be directly irradiated is set as a location of interest, the location of interest cannot be observed.

[0005] An object of the present invention is to provide an observation system, an observation method, and a program that are capable of observing a target area that cannot be directly irradiated with an electron beam.

[0006] Other objects and novel features of the present invention will become apparent from the description of this specification and the accompanying drawings. [Means for solving the problem]

[0007] A brief outline of a representative embodiment of the present invention will be described below.

[0008] That is, an observation system according to one embodiment includes an electron microscope and a computer. Here, the electron microscope irradiates electrons to a first surface position on the sample that is different from the formation position (point of interest) of the shape of interest on the sample, detects predetermined electrons that are scattered from the first surface position inside the sample and exit from the formation position of the shape of interest to the outside of the sample, and outputs the detected electrons as a detection signal, and the computer outputs one or more values ​​related to the shape of the shape of interest based on the detection signal.

[0009] In another embodiment, an observation method is provided, and in yet another embodiment, a program executed in a program is provided to determine a shape change of a feature of interest on a sample. Effect of the Invention

[0010] To briefly explain the effect obtained by a representative embodiment of the invention disclosed in this application, it is possible to provide an observation system that can observe the formation position of a target shape that cannot be directly irradiated with an electron beam. [Brief description of the drawings]

[0011] [Figure 1] 1 is a block diagram showing a configuration of an observation system according to a first embodiment. [Diagram 2] 1 is a block diagram showing a configuration of a computer according to a first embodiment. [Diagram 3] 1A to 1D are diagrams showing the shape of a pattern formed on a sample and its electron microscope images. [Figure 4] FIG. 1 is an explanatory diagram for explaining the principle of an observation system according to the first embodiment. [Diagram 5] 4 is a flowchart showing the operation of the observation system according to the first embodiment. [Figure 6]13 is a flowchart showing the operation of the observation system according to the first modification of the first embodiment. [Figure 7] FIG. 2 is a diagram showing a configuration of a GUI according to the first embodiment. [Figure 8] 13(A) to 13(D) are diagrams illustrating a shape of interest according to Modification 2 of Embodiment 1. [Figure 9] FIG. 11 is a block diagram showing the configuration of a computer according to a second embodiment. [Figure 10] FIG. 11 is a diagram for explaining an observation system according to the second embodiment. [Figure 11] FIG. 11 is a diagram for explaining an observation system according to the second embodiment. [Figure 12] 11 is a flowchart showing the operation of the observation system according to the second embodiment. [Figure 13] FIG. 11 is a diagram showing a configuration of a GUI according to the second embodiment. [Figure 14] 13 is a flowchart showing the operation of the observation system according to the third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] The following embodiments will be described with reference to the drawings. Note that the following embodiments do not limit the scope of the invention, and not all of the elements and combinations thereof described in the embodiments are necessarily essential to the solution of the invention.

[0013] According to the present invention, the shape of a target location (shape of interest) that is the object of observation is observed, and in the embodiment described below, the shape is observed as the amount of change of the shape of interest relative to a reference shape that serves as a reference. That is, the shape of interest is observed as a relative amount relative to the reference shape. Of course, the reference shape may be fixed and the amount of change may be observed as an absolute amount. An example of the reference shape will be described later with reference to FIG. 7 etc.

[0014] (Embodiment 1) <Overall configuration of the observation system> Fig. 1 is a block diagram showing the configuration of an observation system according to embodiment 1. In Fig. 1, reference numeral 1 indicates the observation system. The observation system 1 includes an electron microscope 100, a control device 120, power supply devices 121 and 122, and a computer 200.

[0015] In the electron microscope 100, an electron beam (electron beam) 103 is irradiated onto a sample. The electron microscope 100 outputs a detection signal obtained based on the irradiation of the electron beam 103. The observation system 1 includes components necessary for forming a signal waveform and an image based on the detection signal from the electron microscope 100. First, an example of the electron microscope 100 will be specifically described with reference to FIG. 1.

[0016] An electron beam 103 extracted from an electron source 101 by an extraction electrode 102 is accelerated by an acceleration electrode (not shown). The accelerated electron beam 103 is focused by a condenser lens 104, which is a type of converging lens. The focused electron beam 103 is scanned one-dimensionally or two-dimensionally on a sample 108 by a scanning electrode 105. The electron beam 103 is decelerated by a negative voltage applied to an electrode built into a sample stage 109, and is focused by the lens action of an objective lens 106, and is irradiated onto the sample 108.

[0017] When the electron beam 103 is irradiated onto the sample 108, it scatters from the irradiated location (irradiation location) and / or from the irradiation location into the sample 108, and is emitted as electrons 110 such as secondary electrons and backscattered electrons from a location different from the irradiation location. The emitted electrons 110 are accelerated toward the electron source 101 by an acceleration action based on a negative voltage applied to the sample 108, and collide with a conversion electrode 112 to generate secondary electrons 111. The secondary electrons 111 emitted from the conversion electrode 112 are captured by a detector 113, and a detection signal, which is an output of the detector 113, changes depending on the amount of the captured secondary electrons 111.

[0018] The detection signal output from the detector 113 is supplied to the computer 200 by the control device 120. The computer 200 is equipped with a display device (not shown). The brightness of an image displayed on this display device changes according to the detection signal. That is, the amount of electrons captured by the detector 113 (electron amount) is displayed as brightness on the display device.

[0019] For example, when displaying a two-dimensional image on a display device, the deflection signal supplied to the scanning electrode 105 and the detection signal output from the detector 113 are synchronized, so that the brightness of the image in the scanning area scanned by the deflection signal is displayed on the display device.

[0020] 1 is provided with a deflector (not shown) for moving the scanning area of ​​the electron beam 103, although this is not particularly limited. This deflector is used to display images of patterns of the same shape that exist at different positions on a display device. This deflector is also called an image shift deflector, and can move the viewing position of the electron microscope 100 without moving the sample 108 using a sample stage (e.g., sample stage 109) that moves the sample 108. The image shift deflector and the scanning electrode 105 may be a common deflector, and a signal for image shifting and a deflection signal may be superimposed and supplied to the deflector.

[0021] Detection signals (image, brightness profile, brightness, etc.) from the electron microscope 100 are supplied to the computer 200 via the control device 120. The computer 200 calculates values ​​related to changes in the shape of the target shape of the observation target based on the supplied detection signals, and outputs one or more of these calculated values. Note that the computer 200 may be integrated with the electron microscope 100.

[0022] The control device 120 controls the power supply devices 121 and 122 according to instructions from the computer 200. By controlling the power supply device 122, the voltage applied to the extraction electrode 102 and the acceleration electrode (not shown) changes. Similarly, by controlling the power supply device 121, the voltage applied to the sample 108 changes. Furthermore, the control device 120 controls the deflection signal supplied to the scanning electrode 105 and the signal supplied to the objective lens 106 according to instructions from the computer 200. Furthermore, the control device 120 supplies the detection signal output from the detector 113 to the computer 200, as described above.

[0023] <Computer configuration> Next, the computer 200 according to the first embodiment will be described with reference to the drawings. Fig. 2 is a block diagram showing the configuration of the computer according to the first embodiment.

[0024] The computer 200 has a plurality of functional blocks, but only the functional blocks necessary for the explanation are shown in Fig. 2. In Fig. 2, 210 indicates a computer core (hereinafter also simply referred to as a computer), 201 indicates an electron scattering range database (Data Base: hereinafter also referred to as an electron scattering range DB), and 208 indicates a feature database (hereinafter also referred to as a feature DB). The electron scattering range DB 201 and the feature DB 208 are databases stored in a storage device (not shown) connected to the computer 210.

[0025] In the electron scattering range DB 201, the accelerating voltage for accelerating the electron beam 103 (see FIG. 1) and the electron scattering range for each material of the sample 108 (see FIG. 1) are registered. Here, the electron scattering range indicates the range in which electrons incident on the sample 108 made of, for example, a single material are scattered. The electron scattering range can be obtained, for example, by performing a simulation using the accelerating voltage and the material of the sample as parameters, that is, an electron scattering simulation. One example of the electron scattering simulation is a Monte Carlo simulation. Of course, the parameters used in the electron scattering simulation are not limited to the accelerating voltage and the material of the sample.

[0026] A group of feature amounts of the image of the sample 108 is registered in the feature amount DB 208. The feature amounts are, for example, the brightness of the pattern top (hereinafter also referred to as pattern top brightness or top brightness), the brightness of the pattern edge (hereinafter also referred to as pattern edge brightness or edge brightness), the brightness of the region between the pattern top and the pattern edge, and their differential values. In the first embodiment, these feature amounts are used to represent the characteristics of the pattern shape at a position (point of interest) where the electron beam cannot be directly irradiated. The feature amounts do not need to be registered in the database as described above. For example, the user may design the feature amounts based on the image of the sample 108 and input them to the computer core 210. An example of the pattern top brightness and the pattern edge brightness related to the feature amounts will be described later using FIG. 3, so a detailed description will be omitted here.

[0027] The computer 210 includes a processor 209 that executes a program, and an input / output device 205, a display device 206, and a memory 207 that are coupled to the processor 209. The input / output device 205 is, for example, a mouse, a keyboard, etc., and is used by a user to input data, instructions, etc. to the processor 209. The display device 206 is used to display data, etc. calculated by the processor 209. The memory 207 is used to store data, etc. when the processor 209 executes a program.

[0028] The processor 209 reads and executes a program stored in, for example, a storage medium (not shown). Functional units realized in the processor 209 by executing the program are shown in Fig. 2 as a feature amount sensitivity calculation functional unit 202, a feature amount determination functional unit 203, and a dimension calculation functional unit 204 that determines (calculates) and outputs one or more values ​​related to a change in the target shape.

[0029] The feature amount sensitivity calculation function unit 202 calculates the image feature amount and sensitivity to the shape change of the shape of interest at the point of interest from the pattern of the sample 108 input by the user using the input / output device 205, information on the point of interest of the observation target, information on the electron scattering range registered in the electron scattering range DB 201, and the feature amount registered in the feature amount DB 208. From the obtained image feature amount and sensitivity, the feature amount determination function unit 203 determines the feature amount and sensitivity to be used when observing the shape change of the shape of interest, taking into account the sensitivity, variation, etc.

[0030] The dimension calculation functional unit 204 calculates a value related to a shape change of the target shape based on a detection signal actually supplied from the electron microscope 100 (FIG. 1) via the control device 120 (FIG. 1), using the feature amount and sensitivity determined by the feature amount determination functional unit 203 or the feature amount and sensitivity determined by the feature amount sensitivity calculation functional unit 202. One or more values ​​determined by this dimension calculation functional unit 204 are displayed by the display device 206.

[0031] 2 shows an example in which the feature amount determination function unit 203 is used to obtain a value related to a shape change of the target shape, but the feature amount determination function unit 203 is not essential. For example, instead of the feature amount determination function unit 203, the feature amount and sensitivity obtained by the feature amount sensitivity calculation function unit 202 may be displayed on the display device 206, the user may determine the feature amount and sensitivity to be used, and the determined feature amount and sensitivity may be input using the input / output device 205. Also, a plurality of feature amounts and sensitivities obtained by the feature amount sensitivity calculation function unit 202 may be displayed on the display device 206, and the user may select an appropriate one from the plurality of feature amounts and sensitivities. In this case, the dimension calculation function unit 204 uses the feature amount and sensitivity input or selected by the user to obtain and output one or more values ​​related to a shape change of the target shape.

[0032] In the first embodiment, the electron scattering range DB 201 is created in advance for each material and acceleration voltage by, for example, Monte Carlo simulation, as described above. Also, in Fig. 2, the electron scattering range DB 201 is provided in the computer 200, but this is not limiting. For example, the electron scattering range DB may be provided in a server installed outside the computer 200, and the electron scattering range may be provided to the computer 200 by a communication means connecting the server and the computer 200.

[0033] Furthermore, the electron scattering range may be calculated in the computer 200 without using the electron scattering range DB 201. For example, a program for simulating the electron scattering range (scattering simulation) may be prepared, the program may be executed in the processor 209, and an electron scattering range calculation function unit (not shown) for calculating the electron scattering range by the processor 209 may be realized on the computer 200. In this case, the user uses the input / output device 205 to supply, for example, the acceleration voltage and the material of the sample 108 to the electron scattering range calculation function unit. The electron scattering range calculated by the electron scattering range calculation function unit is used instead of the electron scattering range registered in the electron scattering range DB 201. In this case, the electron scattering range DB 201 is no longer essential, and the electron scattering range is generated according to the input of the acceleration voltage, the material of the sample 108, and the like by the user.

[0034] <Example of pattern shape formed on sample> Next, an example of a pattern shape formed on the sample 108 and an example of an electron microscope image obtained when the pattern shape is imaged (observed) by the electron microscope 100 will be described with reference to the drawings. FIG. 3 shows a pattern shape formed on the sample and its electron microscope image. Here, FIG. 3(A) shows a perspective view of the pattern shape, and FIG. 3(B) shows the electron microscope image. FIG. 3(C) shows a cross-sectional view of the pattern shape shown in FIG. 3(A), and FIG. 3(D) shows a brightness profile (change in brightness due to change in coordinates) in the electron microscope image shown in FIG. 3(B).

[0035] Here, a semiconductor wafer used in the manufacture of semiconductor devices will be taken as an example of sample 108. Being a semiconductor wafer, the material of sample 108 is silicon (Si). By etching sample 108, a pattern shape 300 as shown in FIG. 3(A) is formed. Here, an L / S (Line and Space) having a bow shape will be taken as an example of the pattern of pattern shape 300.

[0036] 3A, UP indicates a main surface (first surface) of a line L of the pattern shape 300, and SD indicates a sidewall (second surface) of the line L. Also, DW indicates a back surface of the sample 108 opposite to the main surface UP.

[0037] 3B is an image captured by the electron microscope 100 from vertically above the sample 108. That is, the image captured by irradiating the sample 108 with the electron beam 103 from vertically above the main surface UP is the electron microscope image 310. Note that when the electron beam 103 is irradiated onto the main surface UP, the side wall SD and the back surface DW are not directly irradiated with the electron beam 103.

[0038] Fig. 3(C) shows a cross section 302 of the pattern shape 300 at a cross-sectional position 301 in the pattern shape 300 shown in Fig. 3(A). Also, Fig. 3(D) shows a brightness profile 312 at a cross-sectional position 311 in the electron microscope image 310 shown in Fig. 3(B).

[0039] Here, the bow shape will be explained. The bow shape refers to a shape in which, as shown in FIG. 3(C), in the cross section 302, the width dimension w at a position below the pattern top (for example, at a position of depth h) is smaller than the width dimension wt of the pattern top, which is the main surface UP of the line L, and the side wall SD cannot be seen directly from vertically above the sample 108 (main surface UP). In FIG. 3(B), the width dimension increases again from the depth h downward, but when viewed in cross section, the width of the line L, which is the pattern, decreases from the pattern top (main surface UP) toward the bottom (back surface DW), a so-called reverse taper shape, is also considered to be part of the bow shape. The bow shape is not particularly limited, but is generated, for example, by overetching of a semiconductor wafer.

[0040] In the following description, the pattern top refers to the central position of the pattern top (main surface UP) of the line L indicated by the arrow 303 in FIG. 3C. The image brightness of this pattern top in the electron microscope image is the pattern top brightness at the position indicated by the arrow 313 in FIG. 3D. The pattern edge refers to the position near the pattern edge of the line L indicated by the arrow 304 in FIG. 3C. The image brightness of the pattern edge in the electron microscope image is the pattern edge brightness at the position indicated by the arrow 314 in FIG. 3D. As shown in FIG. 3D, the image brightness of the pattern edge brightness 314 is a local maximum value. In other words, the local maximum value of the image brightness is the pattern edge brightness 314.

[0041] In the electron microscope 100, as described above, the electron beam 103 is irradiated toward the sample 108 from vertically above the main surface UP. In the bowing shape, the width dimension of the line L below the pattern top is smaller than that of the pattern top, so the electron beam 103 is not directly irradiated onto the sidewall SD of the line L. Therefore, it is difficult to observe the sidewall SD with the electron microscope 100.

[0042] <Explanation of the principle> In the first embodiment, the sidewall of a pattern shape having a bow-like shape when viewed in cross section is observed using the above-mentioned pattern top luminance and / or pattern edge luminance as a feature amount. For example, in the first embodiment, the pattern top luminance is used as a feature amount to observe the dimensional change of the width dimension w (hereinafter also referred to as middle width dimension) of a position below the pattern top including the bow shape where the electron beam is not directly irradiated. Next, the principle of observing the dimensional change of the middle width dimension using the feature amount will be described with reference to the drawings.

[0043] FIG. 4 is an explanatory diagram for explaining the principle of the observation system according to the first embodiment. In FIG. 4, 401 indicates a pattern shape (cross-sectional pattern shape) of the outer shape of a line L that does not have a bowing shape, and 402 and 403 indicate pattern shapes (cross-sectional pattern shapes) of the outer shape of a line L that has a bowing shape. As shown in FIG. 4, the middle width dimensions w of the pattern shapes 401 to 403 are different from each other, and the middle width dimension w increases from the pattern shape 403 to 401. In addition, the shape of the dashed line 410 drawn overlapping the pattern shapes 401 to 403 indicates the scattering range of electrons that are scattered inside and outside the sample when an electron beam is irradiated onto the pattern top.

[0044] In FIG. 4, 421 to 423 show cross-sectional views in which the pattern shapes 401 to 403 and the electron scattering range 410 are superimposed. As can be seen from the cross-sectional views 421 to 423, at a position of depth h from the main surface UP, that is, at the measurement position of the middle width dimension w, the number of electrons (electron amount) that fly out from the side wall SD of the line L to the outside of the sample changes depending on the change in the bow shape. That is, by observing the amount of electrons that fly out from the side wall SD, the change in the bow shape can be observed. Some of the electrons that fly out from the side wall SD are reflected, for example, by the side wall of the adjacent line, and are attracted toward the electron source 101 by the electric field, becoming electrons 110 that move toward the electron source 101 shown in FIG. 1. The electron microscope 100 outputs a detection signal based on the secondary electrons 111 caused by the electrons 110, so that the brightness represented by the detection signal changes depending on the amount of electrons that fly out from the side wall SD.

[0045] Here, the electrons emitted from the side wall SD have been described, but the electrons 110 also include electrons that are reflected by the side wall SD inside the sample, scattered, and move from the main surface UP toward the electron source 101. Therefore, the brightness represented by the detection signal also changes depending on the amount of electrons reflected by the side wall SD inside the sample.

[0046] A specific explanation will be given below with reference to cross-sectional views 421 to 423.

[0047] As shown in cross-sectional view 421, in a pattern shape 401 having a large middle width dimension w, the amount of electrons that escape to the outside of the pattern shape 401 at the middle width observation position of depth h is relatively small, so the pattern top luminance 313 (FIG. 3(D)) is relatively small. In contrast, as shown in cross-sectional view 423, in a pattern shape 403 having a small middle width dimension w, the amount of electrons that escape to the outside of the pattern shape 403 at the middle width observation position of depth h is relatively large, so the pattern top luminance 313 (FIG. 3(D)) is relatively large.

[0048] In addition, in the case of pattern shape 402 in which the middle width dimension w at the middle width observation position at depth h is a value between that of pattern shapes 401 and 403, as shown in cross-sectional view 422, the amount of electrons escaping to the outside of pattern shape 402 at the middle width observation position is a value between that of pattern shapes 401 and 403, and therefore pattern top luminance 313 is also a value between that of pattern shapes 401 and 403. The above-mentioned pattern top luminance 313 is represented by a detection signal output from electron microscope 100.

[0049] That is, the pattern top luminance 313 changes depending on the middle width dimension w. In Fig. 4, 430 shows a correlation graph showing the correlation between the pattern top luminance 313 (written as "top luminance" in Fig. 4) and the middle width dimension. As shown in the correlation graph 430, as the middle width dimension increases, the pattern top luminance decreases. By using this correlation graph 430, the middle width dimension can be obtained based on the pattern top luminance even when a bowing shape exists.

[0050] In the first embodiment, the bowing shape is the shape of interest to be observed. The sidewall SD having the bowing shape cannot be directly irradiated with an electron beam. However, according to the first embodiment, a value related to the shape of interest can be obtained based on a detection signal obtained by directly irradiating the surface (main surface UP) of the sample, which is a position different from the location of interest (sidewall SD), where the shape of interest is to be formed, with electrons (electron beam).

[0051] The value related to the shape of interest can be calculated from the pattern top brightness represented by the detection signal by quantifying the correlation graph 430. For example, in the correlation graph 430, the change in the pattern top brightness accompanying the change in the middle width dimension can be fitted with a linear function to obtain a regression line 430_1, and the middle width dimension w can be calculated from the pattern top brightness using the regression line 430_1. In this case, the slope of the regression line 430_1 can be regarded as the sensitivity indicating the amount by which the pattern top brightness changes due to the change in the middle width dimension.

[0052] <Observation system operation> 5 is a flowchart showing the operation of the observation system according to Embodiment 1. The operation of the observation system according to Embodiment 1 will be described with reference to FIGS.

[0053] In step S501 in Fig. 5, information 501 on the material of the sample 108 and the dimension of interest of the location of interest is input to the processor 209 (Fig. 2) through the user's operation of the input / output device 205 (Fig. 2). In Fig. 5, the middle width dimension w at the position of depth h from the main surface UP of the line L is shown as information 501. Although not shown in Fig. 5, silicon is also input as the material of the sample.

[0054] The processor 209 of the calculator 210 executes a program corresponding to the processing of steps S502 to S506. Here, the program corresponding to step S202 including steps S502 to S504 is executed by the processor 209, whereby the feature amount sensitivity calculation function unit 202 shown in Fig. 2 is realized on the processor 209. Also, the program corresponding to step S204 including steps S505 and S506 is executed by the processor 209, whereby the dimension calculation function unit 204 shown in Fig. 2 is realized on the processor 209.

[0055] In step S502, the value of the target dimension (e.g., middle width dimension) input in step S501 is set in the processor 209. In step S503, the processor 209 retrieves an electron scattering range 502 from the electron scattering range DB 201 (FIG. 2), superimposes information 501 of the set target dimension on the electron scattering range 502, and counts the number of electrons (electron amount) that fly out from the sidewall SD, which is the target shape, to the outside of the line L, i.e., to the outside of the pattern shape of the line L, as indicated by reference numeral 503 in FIG.

[0056] When the processor 209 finishes counting the number of electrons protruding outside the pattern shape, it executes step S502 again. In this case, the processor 209 updates the value of the dimension of interest (middle width dimension) in step S502. Then, the processor 209 executes step S503 again. Since the value of the dimension of interest has been updated, step S503 is executed again, whereby the processor 209 counts the number of electrons corresponding to the updated value of the dimension of interest.

[0057] Processor 209 repeats steps S502 and S503 until the value of the dimension of interest reaches a predetermined value. By repeating steps S502 and S503, a plurality of amounts of electrons corresponding to a plurality of dimensions of interest (middle width dimensions) are obtained. That is, by processor 209 repeating steps S502 and S503, a first process is executed to obtain a change in the amount of electrons accompanying a change in the dimension of interest.

[0058] Next, the processor 209 calculates the sensitivity in step S504 (second process) using the multiple target dimensions and the multiple corresponding electron amounts obtained in step S503. For example, the processor 209 sets the multiple target dimensions as the horizontal axis of the correlation graph 430 in Fig. 4, sets the top brightness corresponding to the multiple electron amounts as the vertical axis of the correlation graph 430, and obtains the slope of the regression line 430_1 as the sensitivity. Also, although not particularly limited, the processor 209 displays the obtained sensitivity on the display device 206 of the computer 210 in step S504.

[0059] After the above-mentioned steps S501 to S504 have been executed in advance, observation is performed using the detection signal from the electron microscope 100. That is, in the computer 210, the processor 209 executes the above-mentioned step S202 in advance to obtain the sensitivity. Thereafter, the processor 209 uses the previously obtained sensitivity and the detection signal from the electron microscope 100 to observe the shape change of the shape of interest.

[0060] That is, in the next step S505, the processor 209 acquires a detection signal output from the electron microscope 100 via the control device 120. Then, in step S506, the processor 209 uses the sensitivity and the detection signal to output one or more values ​​related to a shape change of the shape of interest. The values ​​to be output include a change in middle width dimension, a change in pattern top luminance, a change in pattern edge luminance, etc.

[0061] The processor 209 also retrieves and uses information on features (pattern top, pattern edge) from the feature DB 208 (FIG. 2) when executing steps S503, S505, etc. For example, when the electron scattering range and the pattern shape are superimposed, the processor 209 uses the information retrieved from the feature DB 208 to determine the position of the pattern top.

[0062] According to the first embodiment, by irradiating the main surface UP of the sample with an electron beam, it is possible to observe the change in the shape of interest at a location of interest (side wall SD) that cannot be directly irradiated with the electron beam, based on the value output in step S506.

[0063] <Variation 1> Next, an observation system using the feature amount determining functional unit 203 shown in FIG.

[0064] In the first modification, an observation system is provided in which the feature amount is optimized and the optimized feature amount is used to output a value related to a shape change of a shape of interest.

[0065] Fig. 6 is a flowchart showing the operation of the observation system according to the first modification of the first embodiment. Fig. 6 is similar to Fig. 5, so the differences will be mainly described. The differences are that steps S601 and S602 are added to the flowchart of Fig. 5 in Fig. 6. Note that the dashed line indicating the separation between steps S202 and S204 has been omitted in Fig. 6 to avoid complicating the drawing.

[0066] 2 is realized on the processor 209 by the processor 209 executing a program corresponding to steps S601 and S602 added in FIG.

[0067] 4, the pattern top luminance 313, the pattern edge luminance 314, the luminance at any position between the pattern top and the pattern edge, and a value combining these luminances (for example, the amount of luminance change between the luminance of the pattern top and the luminance of the pattern edge), but is not limited to these and may be information related to luminance other than these. Variation 1 will be described taking the pattern top luminance 313 and the pattern edge luminance 314 as examples of the feature amounts.

[0068] In step S601, the processor 209 changes the feature amount used to calculate the sensitivity. In the first modification, the processor 209 selects the pattern top luminance 313 to be used as the feature amount, and then changes the feature amount to the pattern edge luminance 314.

[0069] In step S601, the processor 209 selects the pattern top luminance 313 as a feature amount, and repeatedly executes steps S502 to S504. By executing such processing, the sensitivity is calculated as described in Fig. 5. That is, a correlation graph 430 (Fig. 4) showing the correlation between the middle width dimension w and the pattern top luminance 313 is generated, and the slope of the regression line 430_1 is obtained as the sensitivity regarding the pattern top luminance.

[0070] When the calculation of the sensitivity related to the middle width dimension w and the pattern top luminance 313 is completed in step S504, in step S601, the processor 209 selects the pattern edge luminance 314 as a feature amount. Thereafter, the processor 209 calculates the sensitivity related to the middle width dimension w and the pattern edge luminance 314 by repeatedly executing steps S502 to S504. That is, a correlation graph showing the correlation between the middle width dimension w and the pattern edge luminance 314 is generated, and the slope of the regression line in the correlation graph is obtained as the sensitivity related to the pattern edge.

[0071] When the calculation of the sensitivity for the pattern edge luminance is completed in step S504, the processor 209 executes step S602. In step S602, the processor 209 determines the feature amount and sensitivity to be used based on the previously obtained sensitivity for the pattern top luminance and the sensitivity for the pattern edge luminance. For example, the processor 209 compares the sensitivity for the pattern top luminance and the sensitivity for the pattern edge luminance, selects the highest sensitivity as the optimal sensitivity, and determines the optimal sensitivity and the corresponding optimal feature amount (pattern top luminance or pattern edge luminance) to be used in step S506.

[0072] Steps S505 and S506 are similar to those in FIG. 5 and therefore will not be described. However, the difference from FIG. 5 is that the sensitivity used in step S506 is the optimum sensitivity determined in step S602.

[0073] 6, in step S602, the processor 209 selects the highest sensitivity, but this is not limiting. For example, in step S602, the user may select the sensitivity related to the pattern top luminance or the sensitivity related to the pattern edge luminance as the optimal sensitivity.

[0074] According to the first modification, it is possible to output one or more values ​​related to the shape change of the shape of interest using optimal sensitivity, so that it is possible to output a more accurate value.

[0075] <Example of display on a display device> In the observation system 1 of embodiment 1, the processor 209 displays, on the display device 206, the features and sensitivities (including the optimal features and optimal sensitivity), and one or more values ​​related to the shape change of the shape of interest determined using the features and sensitivities.

[0076] Next, a description will be given of an example of a GUI (Graphical User Interface) displayed on the display device 206. Fig. 7 is a diagram showing the configuration of the GUI according to the first embodiment.

[0077] The displayed GUI is composed of an input screen area 701 and an output screen area 751. The user inputs a model of a pattern to be observed (pattern shape, material of a sample, etc.) in the input screen area 701. According to the input model, the processor 209 (FIG. 2) displays the shape and material (silicon: Si) of the model in the input screen area 701. The user specifies a portion (a target dimension in FIG. 7) 702 to be observed with a target shape in the displayed model, and specifies a value of the specified portion 702 in target dimension information 703 as a numerical value. The user also sets imaging conditions 704 of the electron microscope 100 (FIG. 1). The imaging conditions 704 include an acceleration voltage, a position of the detector 113 (FIG. 1), etc. For example, the value set as the acceleration voltage is notified to the control device 120 shown in FIG. 1, and the control device 120 controls the power supply devices 121 and 122 according to the notified acceleration voltage.

[0078] Thereafter, the user presses the calculation start button 705. When the calculation start button 705 is pressed, the processor 209 executes, for example, steps S502 to S504 and S601 shown in FIG.

[0079] When the calculation of the sensitivity etc. is completed, the processor 209 displays the progress as "calculation completed" in the output screen area 751 to indicate that the calculation is completed, and further displays the sensitivity (correlation graph) 752 for each feature amount, a list 753 of the feature amount and the sensitivity, and an optimal feature amount (optimum feature amount) 754. As described in the first modification, FIG. 7 shows a case where the pattern top luminance (top luminance) and the pattern edge luminance (edge ​​luminance) are used as the feature amount, and the pattern top luminance is selected as the optimal feature amount in step S602 of FIG. 6. In the example shown in FIG. 7, the sensitivity for the pattern top luminance is 0.7, and the sensitivity for the pattern edge luminance is 0.2.

[0080] Next, when a measurement button 710 displayed in the input screen area 701 is pressed, the processor 209 acquires an image captured by the electron microscope 100 under the conditions set in the imaging conditions 704, and displays the acquired image 755 in the output screen area 751. The processor 209 also executes step S506 shown in Fig. 6, and displays the execution result (a value related to the shape change of the shape of interest) in the output screen area 751. In Fig. 7, a change amount 756 of the shape of interest from the reference is displayed as a value related to the shape change of the shape of interest.

[0081] Here, the reference shape that is the basis for the shape of interest will be explained. Reference numeral 757 displayed in output screen area 751 indicates a plan view of a semiconductor wafer that is sample 108 (FIG. 1). A plurality of semiconductor chips having the same configuration are formed on semiconductor wafer 757. In the first embodiment, although not particularly limited, a pattern shape (e.g., line L in FIG. 3) at a predetermined position of semiconductor chip (reference semiconductor chip) 758 formed at the center position of semiconductor wafer 757 is used as the reference shape.

[0082] That is, for example, the middle width dimension of the reference shape formed in the reference semiconductor chip 758 is used as a reference for the amount of change in the middle width dimension of the shape of interest. Using the middle width dimension of the reference shape in the reference semiconductor chip 758 as a reference, for example, the amount of change in the middle width dimension of the pattern shape in the semiconductor chips (target semiconductor chips) 759 and 760 is displayed as the amount of change 756 from the reference of the shape of interest. In this case, the pattern shape serving as the reference shape and the pattern shape to be measured are arranged at the same positions in the reference semiconductor chip 758 and the target semiconductor chips 759 and 760. Although FIG. 7 shows an example in which the reference semiconductor chip 758 and the target semiconductor chips 759 and 760 are formed on the same semiconductor wafer 757, for example, the reference semiconductor chip and the target semiconductor chip may be formed on different semiconductor wafers.

[0083] By outputting the amount of change in the shape of interest from the reference, it is also possible to obtain the variation in the semiconductor wafer.

[0084] When the amount of change 756 of the shape of interest from the reference is displayed, the processor 209 displays the progress as "measurement completed" to indicate that the measurement has been completed in the output screen area 751. Although not shown, a plurality of chips may be measured within a semiconductor wafer, and a distribution (wafer map) of dimensions of interest within the semiconductor wafer may be created and displayed.

[0085] Fig. 7 shows the configuration of a GUI in the case of an observation system according to Modification 1. In the case of an observation system that executes the flowchart shown in Fig. 5, one sensitivity (correlation graph) is displayed in the output screen area 751 instead of the sensitivity 752 for each feature amount, and one feature amount and sensitivity are displayed in the list 753 of feature amounts and sensitivities. Also, the optimal feature amount 754 is not displayed in the output screen area 751.

[0086] <Variation 2> Although the L / S pattern has been described as an example of the shape of interest, the shape of interest is not limited to this. In Modification 2, an example other than the L / S pattern will be described.

[0087] FIG. 8 is a diagram for explaining a shape of interest according to the second modification of the first embodiment. In FIG.

[0088] In FIG. 8(A), 810 denotes a semiconductor layer (material: silicon) which is a sample. A plurality of hole patterns 800 extending from the main surface UP to the back surface DW are formed in the semiconductor layer 810. FIG. 8(B) shows a cross-sectional view 803 of the semiconductor layer 810 in which the hole patterns 800 are formed, seen at a cross section 801. As shown in FIG. 8(B), the black hole patterns are formed so as to extend from the main surface UP to the back surface DW. Therefore, even if an electron beam is irradiated to the main surface UP, the electron beam is not directly irradiated to the sidewalls SD of the hole patterns.

[0089] According to the first embodiment, a predetermined or arbitrary processor repeatedly executes steps S502 to S504 illustrated in FIG. 5 to obtain sensitivity in advance, and the width w of the hole pattern 800 at the position of depth h can be obtained by using the detection signal and sensitivity obtained from an electron microscope obtained by irradiating an electron beam onto the position 804 on the main surface UP, for example.

[0090] Fig. 8(C) shows a sample on which a pattern shape 850 of a laminated structure called GAA (Gate All Around) is formed. Also, a cross-sectional view 853 seen from the cross section 851 is shown in Fig. 8(D). As shown in Fig. 8(D), the pattern shape (GAA) 850 has a recessed region formed with respect to a virtual line 855 connecting two layers protruding in the lateral direction. Even if an electron beam is irradiated to the main surface UP, the electrons are not directly irradiated to the recessed region, and it is difficult to observe the recess amount (indentation amount) Δw.

[0091] According to the first embodiment, a predetermined or arbitrary processor can repeatedly execute steps S502 to S504 illustrated in FIG. 5 to obtain the sensitivity in advance, and the indentation amount Δw can be obtained, for example, by using a detection signal from an electron microscope obtained by irradiating an electron beam onto position 854 of the main surface UP and the sensitivity.

[0092] According to the first embodiment, one or more values ​​relating to the shape change of the location of interest can be output using a detection signal obtained based on electrons irradiated to a surface position on the sample other than the location of interest.

[0093] For example, by obtaining a distribution of values ​​related to a change in the shape of a target shape within the surface of a semiconductor wafer, the obtained distribution can be used to improve and control the manufacturing process.

[0094] Although an example using one feature quantity (e.g., pattern top brightness or pattern edge brightness) represented by a detection signal output from an electron microscope has been described, the present invention is not limited to this. For example, a combination of a plurality of feature quantities represented by a detection signal may be used. As an example, the feature quantities of pattern top brightness and pattern edge brightness may be combined. In this case, the sensitivity related to the pattern top brightness previously determined and the sensitivity related to the pattern edge brightness previously determined are weighted and added, the result of the addition is divided by 2 (1 / 2), and the result is used as the combined sensitivity. In this case, one or more values ​​related to the shape change of the target shape are output based on the pattern top brightness and pattern edge brightness represented by the detection signal and the combined sensitivity previously determined.

[0095] Furthermore, a cross-sectional image prepared by performing a simulation while changing the dimension of interest or a cross-sectional image obtained by a cross-sectional SEM (Scanning Electron Microscope) may be used as the correct value of the shape of interest to calibrate the value related to the shape change of the shape of interest determined by the observation system 1. Calibration improves the accuracy of the value output from the observation system 1, so that the value output from the observation system 1 can be used to grasp the absolute value of the shape of interest of the pattern shape or to evaluate and inspect the pattern shape. Of course, the method of obtaining the correct value of the dimension of interest is not limited to simulation or cross-sectional SEM.

[0096] (Embodiment 2) In the second embodiment, a technique for optimizing the imaging conditions of an electron microscope when observing a shape change of a shape of interest in the observation system described in the first embodiment will be described. Here, the imaging conditions will be described taking the acceleration voltage for accelerating electrons in the electron microscope as an example. The acceleration voltage, which is an imaging condition, is instructed to the control device 120 (FIG. 1) from the processor 209 shown in FIG. 2 as an output of the computer 200 (FIG. 1). The control device 120 controls the electron microscope so as to accelerate electrons at the instructed acceleration voltage.

[0097] Also in the second embodiment, the feature of interest will be described by taking as an example the bow-shaped side wall SD described with reference to FIG.

[0098] Fig. 9 is a block diagram showing the configuration of a computer according to the second embodiment. The electron microscope and control device according to the second embodiment are similar to those shown in Fig. 1, and therefore their description will be omitted unless necessary. Fig. 9 is similar to Fig. 2, and therefore the differences will be mainly described. The differences are that in Fig. 9, an imaging condition determination function unit 901 is added to the processor 209, and the electron scattering range DB 201 has been changed.

[0099] The imaging condition determination functional unit 901, like other functional units (eg, the feature amount sensitivity calculation functional unit 202), is realized on the processor 209 by the processor 209 executing a corresponding program.

[0100] The change in the electron scattering range DB 201 will be described later with reference to Fig. 10 and Fig. 11, and will not be described here. Note that in Fig. 9, the feature amount DB 208 shown in Fig. 2 is omitted.

[0101] <Accelerating voltage and electron scattering range> First, the relationship between the acceleration voltage and the scattering range of electrons scattered within a sample will be described. Fig. 10 and Fig. 11 are diagrams for explaining an observation system according to the second embodiment. Fig. 10 shows a schematic diagram of the relationship between the acceleration voltage and the scattering range of electrons. In Fig. 10, 1001 denotes a sample to which an electron beam is irradiated. Here, it is assumed that the sample 1001 is made of a uniform material (silicon).

[0102] The electron beam is irradiated from the main surface UP of the sample 1001 toward the back surface DW. When the electron beam is irradiated, the penetration depth and scattering range of the electrons incident on the sample 1001 change according to the acceleration voltage. In FIG. 10, the scattering range of the electrons in the sample 1001 is indicated by dashed lines 1002 to 1004. That is, when the acceleration voltage is low, the acceleration of the electrons is low, so the penetration depth is shallow and the scattering range is narrow, as shown in the scattering range 1002. On the other hand, when the acceleration voltage is high, the acceleration of the electrons is high, so the penetration depth is deep and the scattering range is wide, as shown in the scattering range 1004. When the acceleration voltage is a voltage (intermediate voltage) between the low acceleration voltage and the high acceleration voltage, the acceleration of the electrons is intermediate, so the penetration depth and scattering range are values ​​between the low acceleration voltage and the high acceleration voltage, as shown in the scattering range 1003.

[0103] It should be noted that when the material of the sample 1001 is changed, the shapes of the scattering regions 1002 to 1004 also change.

[0104] FIG. 11 shows a schematic diagram of the relationship between the electron scattering range and an L / S pattern having a bow shape.

[0105] The relationship between the electron scattering range and the L / S pattern when the accelerating voltage is low is shown as 1111 in Fig. 11, and the relationship between the electron scattering range and the L / S pattern when the accelerating voltage is high is shown as 1131 in Fig. 11. Similarly, the relationship between the electron scattering range and the L / S pattern when the accelerating voltage is intermediate is shown as 1121 in Fig. 11.

[0106] As shown in the relationship 1111, when the acceleration voltage is low, the electrons are accelerated at a low speed, and thus the scattering range is narrow, as shown in the scattering range 1002. Because the scattering range of the electrons is narrow, the electrons are not scattered to the side wall SD having a bowing shape, and few electrons fly out of the sample from the side wall SD. In contrast, as shown in the relationship 1131, when the acceleration voltage is high, the electrons are accelerated at a high speed, and thus the scattering range is like the scattering range 1004, and the position where the electrons are scattered is lowered toward the back surface DW ( FIG. 10 ) side opposite to the main surface UP. In other words, the position of the scattering range 1004 of the electrons is lower than the position of the bowing shape, and few electrons fly out of the sample from the side wall SD having a bowing shape.

[0107] On the other hand, when the acceleration voltage is an intermediate voltage, as shown in the relationship 1121, the position of the electron scattering range 1003 almost coincides with the position of the bowing shape, so that more electrons fly out of the sample from the bowing-shaped side wall SD.

[0108] As described in the first embodiment, the observation system observes the amount of electrons ejected from the bowing shape, thereby observing the change in the bowing shape based on the detection signal. In this case, the more electrons that are ejected, the more information related to the bowing shape is included in the detection signal. In other words, the more electrons that are ejected, the more information related to the bowing shape represented by the detection signal. Therefore, it is appropriate and desirable for more electrons to be ejected from the bowing shape. That is, there is an appropriate acceleration voltage for observing the shape of interest. In the example shown in FIG. 10 and FIG. 11, the intermediate voltage is the appropriate voltage for observing the shape of interest.

[0109] In the second embodiment, a simulation is performed in advance to obtain the electron scattering range while changing the material and the acceleration voltage. By this simulation, a plurality of electron scattering ranges corresponding to a plurality of acceleration voltages are obtained for each material, and the obtained electron scattering ranges corresponding to each acceleration voltage are registered in advance in the electron scattering range DB201.

[0110] As will be described later, an optimal electron scattering range suited to the shape of interest is selected from the electron scattering range DB 201.

[0111] <Observation system operation> Fig. 12 is a flowchart showing the operation of the observation system according to the second embodiment. Fig. 12 is similar to Fig. 6, so the differences will be mainly described. The differences are that in Fig. 12, steps S1201 to S1204 are added to the steps shown in Fig. 6. Also, in Fig. 12, step S602 shown in Fig. 6 is omitted. The processor 209 shown in Fig. 9 executes a program corresponding to steps S1201 to S1204, thereby realizing the imaging condition determination function unit 901 shown in Fig. 9.

[0112] The operations in steps S501 to S506 and step S601 have already been explained with reference to FIG. 5 and FIG. 6, and therefore will not be repeated here.

[0113] The processor 209 executes steps S502 to S504 and S601 to calculate the sensitivity for each feature amount under a predetermined imaging condition (for example, a condition in which the acceleration voltage is low). That is, the processor 209 reads out the electron scattering range (for example, 1002 in FIG. 10) when the acceleration voltage is low from the electron scattering range DB 201, and calculates the sensitivity for each feature amount (for example, pattern top luminance and pattern edge luminance) using the read electron scattering range.

[0114] In step S1201, the processor 209 determines whether or not the change of the imaging conditions is completed. For example, when there are three imaging conditions (three conditions in which the acceleration voltage is a low voltage, a medium voltage, and a high voltage), the processor 209 determines in step S1201 whether or not all of the three imaging conditions are completed.

[0115] If it is determined in step S1201 that the conditions for the medium acceleration voltage and the high acceleration voltage have not yet been completed, the processor 209 changes the imaging conditions in the next step S1202. That is, the processor 209 reads, for example, the electron scattering range (for example, 1003 in FIG. 10) when the acceleration voltage is the medium voltage from the electron scattering range DB 201. Thereafter, the processor 209 executes steps S502 to S504 and S601. By executing steps S502 to S504 and S601, the sensitivity for each feature amount is calculated using the electron scattering range when the acceleration voltage is the medium voltage.

[0116] In step S1201, the processor 209 repeatedly executes steps S502 to S504, S601, S1201, and S1202 until the processor 209 determines that the changes to all the imaging conditions have been completed. When the changes to all the imaging conditions have been completed, the sensitivity is calculated for each imaging condition and for each feature amount.

[0117] Thereafter, in step S1203, the processor 209 calculates the correlation between the acceleration voltage and the sensitivity. A correlation graph obtained by this calculation is shown as 1201 in Fig. 12. The correlation graph 1201 shows an example in which the pattern top luminance is used as the feature amount, with the horizontal axis indicating the acceleration voltage (acceleration) and the vertical axis indicating the sensitivity of the pattern top luminance (top luminance). According to the correlation graph 1201, the sensitivity is highest when the acceleration voltage is around 30 keV.

[0118] Next, in step S1204, the processor 209 determines the imaging conditions. For example, the most sensitive acceleration voltage (30 keV) is determined as the imaging condition. The processor 209 notifies the control device 120 (FIG. 1) of the determined imaging condition, and the control device 120 sets the acceleration voltage of the electron microscope 100 (FIG. 1) according to the imaging condition. Following step S1204, the processor 209 acquires a detection signal from the electron microscope 100 in step S505, and outputs one or more values ​​related to the shape change of the shape of interest in step S506. Steps S505 and S506 have already been described, so they will be omitted, but in step S505, the acceleration voltage, which is a setting condition of the electron microscope 100, is the one determined in step S1204.

[0119] 12, an example in which the imaging conditions are determined by the imaging condition determination function unit 901 has been described, but the present invention is not limited to this. For example, the processor 209 may display a correlation graph for each feature amount on a display device, and the user may select an appropriate imaging condition.

[0120] <Example of display on a display device> Fig. 13 is a diagram showing the configuration of a GUI according to embodiment 2. Fig. 13 is similar to Fig. 7, so differences will be mainly described. The differences are that in Fig. 13, input screen area 701 in Fig. 7 has been changed to input screen area 1301, and output screen area 751 in Fig. 7 has been changed to output screen area 1351.

[0121] In the input image area 1301, the imaging condition 704 is changed to the imaging condition 1302. In the imaging condition 1302, an acceleration voltage range is provided so that the acceleration voltage to be set in the electron microscope 100 can be input as a voltage range, instead of inputting it as a predetermined value as in FIG. 7. The user inputs the acceleration voltage range in the acceleration voltage range. The acceleration voltage value set in the acceleration voltage range is used in step S1201 in FIG. 12. For example, the lower limit voltage (0.2 keV in FIG. 13) set in the acceleration voltage range corresponds to the predetermined imaging condition described in FIG. 12 (for example, the condition in which the acceleration voltage is a low voltage), and the upper limit voltage (60 keV) corresponds to the high voltage described in FIG. 12. Therefore, the processor 209 executes steps S502 to S506, S601, and S1201 to S1204 shown in FIG. 12 while changing the acceleration voltage from the lower limit voltage to the upper limit voltage shown in FIG. 13.

[0122] In the output screen area 1351, the list 753 of features and sensitivities and the optimal feature (optimum feature) 754 shown in FIG. 7 are changed to a list 1353 of optimal features and optimal acceleration conditions, and the sensitivity for each feature (correlation graph) 752 shown in FIG. 7 is changed to acceleration dependency of sensitivity for each feature (correlation graph) 1352.

[0123] The processor 209 displays the acceleration voltage, which is an imaging condition determined by executing step S1204 in Fig. 12, as an optimal acceleration voltage (optimum acceleration in Fig. 13) in a list table 1353 of optimal feature amounts and optimal acceleration conditions. The processor 209 also displays the feature amount determined by executing step S602 (omitted in Fig. 12) shown in Fig. 6 in a list table 1353 of optimal feature amounts and optimal acceleration conditions. The processor 209 also displays a correlation graph of the sensitivity for each feature amount obtained by executing step S1203 in Fig. 12 as acceleration dependency (correlation graph) 1352 of the sensitivity for each feature amount.

[0124] In response to a user pressing a measurement button 710, the processor 209 acquires an image 755 of the sample from the electron microscope 100, calculates a feature amount from the acquired image, and outputs a change amount 756 of the shape of interest from a reference.

[0125] Although the imaging condition is described by taking the acceleration voltage of the electrons irradiated to the sample as an example, the imaging condition is not limited to this. For example, the electron scattering range DB201 (FIG. 9) may store not only the electron scattering range but also the electron trajectory information including the electron energy, the electron position, and / or the electron traveling direction. In this case, in step S503 (FIG. 12), the number of electrons is counted taking into consideration not only the number of electrons but also the direction in which the electrons are emitted and the energy at that time, and the arrangement of the detector 113 (FIG. 1) of the electron microscope 100 and the sensitivity for each band of the detection energy filter (not shown) arranged in front of the detector are calculated in step S504, and the correlation between the acceleration and the sensitivity is calculated for each arrangement of the detector 113 and each detection energy filter in step S1203 (FIG. 12), and the optimal arrangement of the detector 113 and the value of the detection energy filter can be derived.

[0126] According to the second embodiment, it is possible to optimize the imaging conditions of the electron microscope when observing a change in the shape of a target shape that cannot be directly irradiated with an electron beam.

[0127] (Embodiment 3) In the first and second embodiments, the electron scattering range obtained by the electron scattering simulation is used to observe the shape change of the target shape. In the third embodiment, an example will be described in which an image actually captured by an electron microscope is used instead of the electron scattering simulation.

[0128] FIG. 14 is a flowchart showing the operation of the observation system according to the third embodiment.

[0129] First, in step S1401, a plurality of samples having shapes of interest with different dimensions are prepared.

[0130] In step S1402, the processor 209 (FIG. 2) acquires images or brightness of the multiple specimens prepared in step S1401 as samples using the electron microscope 100 (FIG. 1) to create one image group. In step S1402, multiple image groups are acquired by acquiring images of the samples while changing the imaging conditions (e.g., acceleration voltage) of the electron microscope 100.

[0131] In step S1403, the processor 209 analyzes the feature amounts (e.g., pattern top luminance, pattern edge luminance, etc.) of the image group (measurement image group) and calculates the relationship (sensitivity) between the change in the shape of interest and the change in the feature amount. In step S1404, the processor 209 determines whether the calculation in step S1403 has been completed for all imaging conditions. If the calculation in step S1403 has not been completed for all imaging conditions, the processor 209 repeats steps S1403 and S1404 to perform the calculation in step S1403 on the measurement image group for the imaging conditions that have not been completed.

[0132] When the processor 209 determines in step S1404 that the calculation is completed for all imaging conditions, the processor 209 executes step S1405. In this step S1405, the processor 209 calculates the relationship between the imaging condition and the sensitivity for each feature amount, and displays it on the display device 206 (FIG. 2). In FIG. 14, an example of a correlation graph displayed on the display device 206 is indicated by reference numeral 1401. What is displayed on the display device 206 is not limited to the correlation graph 1401 in FIG. 14. For example, the relationship between the feature amount, the imaging condition, and the sensitivity may be displayed in a three-dimensional (3D) map with the horizontal axis representing the acceleration voltage as the imaging condition, the vertical axis representing the feature amount, and the depth direction representing the magnitude of the sensitivity.

[0133] In step S1406, the processor 209 determines the imaging conditions and the feature amounts using the feature amount determination function unit 203 (FIG. 2) and the imaging condition determination function unit 901 (FIG. 9). Of course, the user may determine the imaging conditions and the feature amounts based on the sensitivity and the like shown in the correlation graph 1401.

[0134] Thereafter, in step S505, the processor 209 acquires a detection signal from the electron microscope 100, and in step S506, outputs one or more values ​​related to the shape change of the shape of interest.

[0135] As described above, multiple specimens are prepared in step S1401, but for example, patterns of different shapes of interest formed at different positions in a single semiconductor wafer may be prepared as the shapes of interest. Of course, multiple specimens may be prepared using multiple semiconductor wafers instead of a single semiconductor wafer.

[0136] According to the third embodiment, since the sensitivity is obtained using an actual sample, a value related to the shape change of the shape of interest can be output using a more accurate sensitivity. Moreover, the third embodiment may be combined with the first or second embodiment. For example, the sensitivity obtained in the third embodiment may be used to calibrate the sensitivity obtained in the first or second embodiment, and the calibrated sensitivity may be used to output a value related to the shape change of the shape of interest. By using the calibrated sensitivity, a value related to the shape change can be output more accurately.

[0137] In the above-described embodiment, information such as a program for implementing each functional unit can be stored in a recording device such as a memory, a hard disk, or an SSD (Solid State Drive), or in a recording medium such as an IC card, an SD card, a DVD, etc. Of course, each functional unit may be implemented by combining logic circuits, etc.

[0138] In embodiments 1 to 3, the shape change of a shape of interest that cannot be directly irradiated with an electron beam can be quantified using a detection signal obtained by irradiating an electron beam to a predetermined location (e.g., the position of the first surface: the first surface position) different from the location of interest where the shape of interest is formed (the formation position where the shape of interest is formed).

[0139] The invention made by the inventor has been specifically described above based on an embodiment, but it goes without saying that the present invention is not limited to the above embodiment and can be modified in various ways without departing from the gist of the invention. [Explanation of symbols]

[0140] 1 Observation System 100 Electron Microscope 103 Electron Beam 108 Samples 120 Control device 200 calculator DW Back SD side wall (second surface) UP Main surface (1st surface)

Claims

1. An observation system having an electron microscope and a computer, The electron microscope is An electron source that generates an electron beam; a detector for detecting secondary electrons generated in a sample when the sample is irradiated with the electron beam accelerated by an accelerating voltage; Equipped with the sample has a bow-shaped pattern having a main surface and a side wall, the main surface is irradiated with the electron beam, and the secondary electrons from the main surface are detected by the detector; the computer includes a processor to which a detection signal of brightness corresponding to an amount of the secondary electrons detected by the detector is supplied; the processor superimposes a predetermined electron scattering range and the bowing shape while changing the bowing shape, obtains an amount of electrons flying out from a side wall of the bowing shape, defines the obtained amount of electrons as luminance, and obtains a change in luminance accompanying the change in the bowing shape as sensitivity in advance; the processor outputs a value related to a shape change of the bow shape based on the brightness detection signal output from the electron microscope and the previously obtained sensitivity. Observation system.

2. 2. The observation system according to claim 1, the computer obtains the scattering range of electrons while changing the acceleration voltage, and includes a database in which a plurality of acceleration voltages and a plurality of corresponding scattering ranges of electrons are registered; the processor selects an electron scattering range from the plurality of electron scattering ranges registered in the database, superimposes the selected electron scattering range with the bowing shape while changing the bowing shape, determines the amount of electrons escaping from a sidewall of the bowing shape, determines a change in brightness accompanying the change in the bowing shape as sensitivity for each of the selected electron scattering ranges, determines a correlation between the acceleration voltage corresponding to the selected electron scattering range and the sensitivity, and controls the electron microscope so that the electron beam is accelerated at an acceleration voltage corresponding to a high sensitivity. Observation system.

3. An observation method for observing a sample based on a detection signal output from an electron microscope, comprising: the electron microscope includes a detector that detects secondary electrons generated in the sample by irradiating the sample with an electron beam accelerated by an accelerating voltage; the sample has a bow-shaped pattern having a main surface and a side wall, the main surface is irradiated with the electron beam, and secondary electrons from the main surface are detected by the detector, a computer to which a detection signal of brightness corresponding to the amount of secondary electrons detected by the detector is supplied, superimposing a predetermined electron scattering range and the bowing shape while changing the bowing shape, determining the amount of electrons flying out from the side wall of the bowing shape, determining the determined amount of electrons as brightness and a change in brightness accompanying the change in the bowing shape as sensitivity in advance; the computer outputs a value related to the change in the bowing shape based on the brightness detection signal output from the electron microscope and the previously obtained sensitivity. Observation method.

4. In the observation method according to claim 3, The computer includes a processor and a database in which a scattering range of electrons is calculated while changing the acceleration voltage, and a plurality of acceleration voltages and a corresponding plurality of scattering ranges of electrons are registered; the processor selects an electron scattering range from the plurality of electron scattering ranges registered in the database, superimposes the selected electron scattering range with the bowing shape while changing the bowing shape, determines the amount of electrons escaping from a sidewall of the bowing shape, determines a change in brightness accompanying the change in the bowing shape as sensitivity for each of the selected electron scattering ranges, determines a correlation between the acceleration voltage corresponding to the selected electron scattering range and the sensitivity, and controls the electron microscope so that the electron beam is accelerated at an acceleration voltage corresponding to a high sensitivity. Observation method.

5. A program for causing a processor to execute a process for observing a sample using a detection signal output from an electron microscope, comprising: The sample has a bow-shaped pattern with a main surface and a sidewall, the electron microscope irradiates the principal surface of the sample with an electron beam accelerated by an accelerating voltage, and supplies secondary electrons from the principal surface to the processor as the detection signal; The processor, a first process for superimposing a predetermined electron scattering range and the bowing shape while changing the bowing shape, determining an amount of electrons escaping from a sidewall of the bowing shape, defining the determined amount of electrons as luminance, and determining a change in luminance accompanying the change in the bowing shape as sensitivity; a second process for outputting a value related to a change in the bowing shape based on the detection signal output from the electron microscope and the sensitivity obtained in the first process; Execute the program.

6. In the program according to claim 5, the computer obtains the scattering range of electrons while changing the acceleration voltage, and includes a database in which a plurality of acceleration voltages and a plurality of corresponding scattering ranges of electrons are registered; In the first process, the processor selects an electron scattering range from the plurality of electron scattering ranges registered in the database, superimposes the selected electron scattering range and the bowing shape while changing the bowing shape, calculates an amount of electrons flying out from a side wall of the bowing shape, calculates a change in brightness accompanying the change in the bowing shape as a sensitivity for each of the selected electron scattering ranges, and calculates a correlation between the acceleration voltage corresponding to the selected electron scattering range and the sensitivity. determining an acceleration voltage corresponding to high sensitivity based on the correlation between the acceleration voltage and the sensitivity determined in the first process, and controlling the electron microscope in the second process so that the electron beam is accelerated at the determined acceleration voltage. program.

Images