JP2013200319A - Electron microscope system and pattern dimension measuring method using the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an electron microscope system for outputting contour information adapted to an actual pattern edge end of a sample.SOLUTION: A local projection waveform is generated by projecting an electron microscope image in a tangential direction to a pattern edge on each point of the pattern edge of the electron microscope image, and each local projection waveform generated on each point is applied to a library formed beforehand where a cross-sectional shape of a sample is correlated with an electron beam signal waveform. Hereby, the cross-sectional shape of a pattern transcribed onto the sample is estimated, and a position coordinate of the edge end on the basis of the cross-sectional shape is determined, and a contour of the pattern is output as a series of the position coordinate.

Description

The present invention relates to a photomask used in a lithography process of semiconductor manufacturing and a pattern image measurement and measurement system that measures a pattern image obtained by photographing a wafer after the lithography process with an electron microscope.

  With the miniaturization of semiconductor circuit patterns, the resolution of optical exposure equipment has reached its limit, and in order to form the circuit pattern as designed on the silicon wafer, a correction pattern is put in the photomask design data in advance, and the finished shape OPC (Optical Proximity Correction) processing for controlling the image is frequently performed.

  The OPC pattern is designed based on a light exposure simulation that predicts how the photomask pattern subjected to the OPC process is transferred onto the silicon wafer. In order to verify this, that is, to confirm whether the designed circuit pattern is formed on the wafer, it is necessary to observe the pattern (resist pattern) transferred to the wafer. Non-Patent Document 1 discloses a system for displaying design data (FIG. 2B) superimposed on a pattern (FIG. 2A) on a wafer imaged with an electron microscope (FIG. 2A) (FIG. 2C). Has been.

  In Patent Document 1 and Patent Document 2, an edge is detected from an electron microscope image to obtain a pattern outline, and this is overlaid with design data, and the amount of pattern deformation is numerically calculated from the positional relationship between the two at each pattern site. Is disclosed.

Japanese Patent Laying-Open No. 2007-121147 (FIG. 6) Japanese Patent Laid-Open No. 2001-338304 (FIGS. 49 and 53) Hidekazu Suzuki, Ryoichi Matsuoka, “Wafer Transfer Pattern Shape Evaluation System Based on CAD Data / Grade Scope”, LSI Testing Symposium 2002 Proceedings, pp. 31-36 (2002) J. S. Villarrubia, A. E. Vladar, J. R. Lowney, and M. T. Postek, “Scanning electron microscope analog of scatterometry,” Proc. SPIE 4689, pp. 304-312 (2002)

  In order to verify the OPC process more accurately, it is important that the pattern contour line obtained from the electron microscope image conforms to the actual pattern edge end of the sample. However, in Patent Document 1 and Patent Document 2, no consideration is given to correctly capturing the edge of the pattern edge.

  FIG. 3 shows an edge detection method disclosed in Patent Document 1. On the secondary electron microscope image (SEM image), the edge portion appears as a bright band and a white band, but according to the method of FIG. 3, the center line of the white band (see FIG. 604) is generally set as the edge point. .

  FIG. 4 shows an edge detection method disclosed in Patent Document 2. A threshold value T is given to the profile acquired in the profile acquisition section of FIG. 4A as shown in FIG. 4B, and the intersection of the profile and the threshold value T (indicated by Q in the figure) is the edge point. Is done.

  On the other hand, as shown in Non-Patent Document 2, the SEM signal waveform changes depending on the cross-sectional shape of the target pattern. FIG. 5A is a part of the SEM image shown in FIG. The white band width differs depending on the part, which is an indication that the cross-sectional shape of the sample differs depending on the part. The part A where the white band (white part in the figure) has a narrow width is estimated to have a sharp edge. , It is presumed that the portion B where the white band is wide has a taper (the inclination of the edge is relatively large). FIG. 5B is a schematic diagram of the cross-sectional shape of the pattern corresponding to the part A and the part B, and the secondary electron signal waveform. In general, what is desired to be regarded as the pattern edge end is the bottom end of the pattern (the portion indicated as the edge end that is desired to be detected in the figure), and corresponds to the black circled position on the secondary electron signal waveform.

  However, according to the pattern edge detection method (FIG. 3) described in Patent Document 1, since the detected edge point is the center line of the white band, the position is indicated by the black triangle in FIG. 5 (c). Also, according to the pattern edge detection method described in Patent Document 2 (FIG. 4), the detected edge point depends on the threshold value T, but for example, the midpoint between max and min is set as the threshold value. Then, the black square in FIG.

  In FIGS. 5C and 5D, it is apparent that the bottom end of the pattern as shown by the black circle in FIG. 5B is not detected. A further problem is that in FIGS. 5C and 5D, the distance between the detected edge point and the bottom end of the pattern differs depending on the cross-sectional shape of the pattern (part A and part B in the figure). .

  As apparent from the fact that the white band width varies depending on the part in FIG. 5A, the cross-sectional shape of the pattern is not uniform. In the prior art, there is a situation in which there is a deviation between the detected edge point and the actual pattern edge end, and the deviation amount varies depending on the location.

  The conventional technique is that no attention is paid to the change in the detected signal waveform depending on the cross-sectional shape of the target pattern. The detected edge point is either the method using the center line of the white band shown in Patent Document 1 as the edge point or the method using the intersection of the threshold value and the profile shown in Patent Document 2 as the edge point. The physical meaning of is ambiguous.

  The purpose of verifying the OPC process is to check whether a circuit pattern as designed is formed on the wafer, and to change the size of the correction pattern if it is found that the pattern does not conform to the design value. As described above, in the situation where the pattern edge edge is not correctly detected, it is clear that the adjustment amount of the correction pattern size may be misread and high-precision OPC verification cannot be realized.

  An object of the present invention is to provide an electron microscope system capable of extracting dimensional information corresponding to the actual pattern edge of a sample and performing highly reliable dimension measurement.

  In order to achieve the above object, in the present invention, an electron microscope system includes an electron beam image acquisition means for acquiring an image of a desired portion of a sample having a pattern formed on the surface using a scanning electron microscope, and a pattern of the sample. Storage means for storing a library for associating the cross-sectional shape of the sample and electron beam signal waveform information corresponding to the cross-sectional shape of the pattern, and information obtained from the image of the sample pattern acquired by the electron beam image acquiring means is stored in the storage means. Dimension information extracting means for extracting pattern dimension information by processing with reference to the library information, and output means for displaying on the screen information related to the dimension of the pattern extracted by the dimension information extracting means. Configured.

  In order to achieve the above object, in the present invention, in a pattern dimension measuring method using an electron microscope system, an image of a desired portion of a sample having a pattern formed on the surface is obtained using a scanning electron microscope. A library that correlates the cross-sectional shape of the sample pattern and the electron beam signal waveform information corresponding to the cross-sectional shape of the pattern is stored, and information obtained from the acquired image of the pattern of the sample is stored in the storage means. Pattern size information is extracted by referring to the processing, and information related to the extracted pattern size is displayed on the screen.

According to the present invention, the projection waveform at each point of the pattern edge of the electron microscope image is applied to a library that correlates the cross-sectional shape of the sample and the electron beam signal waveform, thereby estimating the cross-sectional shape of the pattern. Since an edge end (for example, the bottom end) of a desired definition is used as an edge point and the series is used as a pattern contour line, a contour line corresponding to the actual pattern edge end of the sample is output.
The case where the cross-sectional shape of the pattern changes depending on the part is particularly effective. As described in “Problems to be Solved by the Invention”, in the conventional technique, the deviation amount between the actual pattern edge edge and the detected edge edge differs depending on the part. This problem is particularly serious when trying to perform highly accurate OPC verification. According to the present invention, since the cross-sectional shape is estimated for each part, and the edge point is determined according to the estimated cross-sectional shape, the problem that the deviation amount from the actual pattern edge end varies depending on the part.

  The present invention can be applied to various types of charged particle beam apparatuses (SEM, FIB, etc.), but in the following examples, a description will be given of a SEM (Scanning Electron Microscope) as a representative.

  A flow of shape measurement in the semiconductor wafer shape measurement apparatus according to the present invention is shown in FIG.

  First, a method for acquiring an SEM image using the SEM apparatus 10 will be described with reference to FIG. An electron beam 1002 emitted from an electron gun 1001 of the SEM 10 is accelerated by an acceleration electrode 1003, converged by a focusing lens 1004, and deflected by a deflector 1005 in the X direction and Y direction (in a plane perpendicular to the drawing in FIG. 1A). The surface of the sample 1007 is scanned and irradiated with the objective lens 1006 so that the focus of the electron beam is aligned with the surface of the sample 1007 on which the measurement target pattern is formed. Although not shown in FIG. 1A, the sample 1007 is placed on a table and can be moved in a plane, and a desired region on the surface of the sample 1007 is positioned in the irradiation region of the electron beam 1002. To be controlled.

  A part of the secondary electrons generated from the surface of the sample 1007 irradiated with the electron beam 1002 is detected by the detector 1008, converted into an electric signal, sent to the overall control / image processing unit 1009, and an SEM image is created. Then, the SEM image is processed using the information of the SEM waveform library stored in the storage unit 1011 by the calculation unit 1010 to calculate the dimension of the pattern, and the result is displayed on the screen of the output unit 1012. The pattern dimension calculation result is stored in the server 1013 via the communication line. The overall control / image processing unit 1009 also controls the entire SEM 10 including a table on which a sample 1007 (not shown) is placed.

  As shown in FIG. 1B, the calculation unit 1010 includes a provisional contour extraction unit 10101, a local projection waveform forming part 10102, a library matching unit 10103, an edge position determination unit 10104, a coordinate conversion / contour line extraction unit 10105, a dimension measurement. Part 1016.

  The steps of S100 to S600 will be described with reference to the flow shown in FIG.

  (S100): The temporary contour extraction unit 10101 extracts a temporary contour line from the SEM image 20 acquired by the SEM 10. As a provisional outline extraction method, the conventional technique shown in FIG. 3 can be applied. This will be described below with reference to FIG. The input image 105 is subjected to smoothing processing for removing noise (601), edges are detected by edge enhancement filter processing such as a Sobel filter (602), and binarization processing is performed to thereby obtain an edge portion. A binary image of “1” and “0” except for the edge portion is obtained (603). Subsequently, the center line is detected by thinning processing such as contraction until the pattern width becomes one pixel (604). The obtained line drawing (112) becomes a temporary outline.

  (S200): In this step, in the local projection waveform forming portion 10102, an image is projected in the tangential direction of the contour line in the local region along the contour line obtained in S100 to generate a local projection waveform. This step is supplemented in FIG. FIG. 6A is an enlarged view of the vicinity of the part B in FIG. A dotted line 112 is the temporary contour line extracted in S100 (as described above, when the method shown in FIG. 3 is used, the temporary contour line is approximately the center line of the white band). An image is projected along the provisional contour 112 by a fixed length p1 in the tangential direction. That is, a local projection waveform 210 as shown in FIG. 6B is obtained by projecting an image in the direction of the arrow for each of the local rectangular regions 202, 203, 204. The projection length p1 and the pitch p2 for obtaining the projection waveform may be fixed values, and according to the curvature of the contour line, p1 and p2 are small at a portion with a large curvature, and p1 and p2 are large at a portion with a small curvature. Thus, it may be variable according to the pattern shape.

  (S300): In this step, the library matching unit 10103 performs matching between the local projection waveform obtained in S200 and the SEM waveform library 30 stored in advance in the storage unit 1011. The SEM waveform library 30 associates the cross-sectional shape of the sample with the SEM waveform. For example, SEM in various cross-sectional shapes by the method shown in Non-Patent Document 2, that is, electron beam simulation (Monte Carlo simulation). It can be created by calculating the waveform.

  FIG. 7 is a conceptual diagram of the SEM waveform library. The horizontal axis indicates the taper angle of the sample, and the vertical axis indicates the degree of rounding of the top corner. The SEM waveform 702 in each cross-sectional shape 701 of the pattern calculated by Monte Carlo simulation is stored as a pair with the cross-sectional shape. By selecting the waveform that best matches the local projection waveform obtained in S200 from among the SEM waveforms stored in the library by waveform matching, the cross-sectional shape of the pattern at the site where the local projection waveform was acquired is estimated. Will be. In FIG. 7, for the sake of simplicity, the cross-sectional shape of the pattern is expressed by two parameters, ie, the taper angle and the rounding of the top corner. However, in actuality, the above-described two cross-sectional shapes are used to express the cross-sectional shape according to the measurement target pattern. It is necessary to have necessary parameters in addition to one parameter. For example, the height of the pattern and the rounding of the bottom corner can be considered. It is also possible to use the light exposure simulation result described in the section “Background Technology” in determining the parameter type for expressing the cross-sectional shape of the pattern and the variation range of each parameter.

  (S400): In this step, the edge position determination unit 10104 determines the edge position on the local projection waveform based on the cross-sectional shape of the pattern estimated in S300. On the SEM waveform library 30, the correspondence between the cross-sectional shape of the pattern and the SEM waveform of the portion is known. Therefore, by defining the edge end (for example, the bottom end) of the desired pattern with respect to the cross-sectional shape of the pattern, The position is propagated to the SEM waveform and the position corresponding to the local projection waveform is further propagated.

  FIG. 8 shows an example of a GUI screen for the user to specify the edge of the pattern cross section. The edge edge at a predetermined percentage is detected as shown in FIG. 8B (or 0% to be the bottom edge), or the edge edge at a predetermined height is detected as shown in FIG. 8C. (Specify 0nm as the bottom edge).

  (S500): In this step, the coordinate conversion / contour extraction unit 10105 converts the edge position on the local projection waveform determined in S400 to a position on the SEM image.

  (S600): Further, the coordinate transformation / contour extraction unit 10105 obtains a final contour as a series of coordinates on the SEM image obtained in S500.

  (S700): The dimension measuring unit 10106 measures the dimension of the desired position of the pattern or the interval between adjacent patterns using the contour line obtained in S600.

  The result measured in S700 is displayed on the screen of the output unit 1012 and stored in the server 1013 via the communication line.

  As an example displayed on the screen of the output unit 1012, FIG. 9 shows a case where the final contour line obtained in S600 corresponding to the same part as FIG. 6 is displayed. In FIG. 9, a dotted line 112 displayed superimposed on the SEM image 901 is a provisional contour line obtained in S100, and a solid line 113 is a final contour line.

  The obtained outline 113 can be used, for example, for comparison with the design data 114 shown in the thick practice. Thus, by comparing with the design data, for example, an evaluation of how much the edge placement error amount E or the design line width L1 is different from the actual line width L2 is obtained by the conventional technique. Compared with the case where the contour 112 to be used is used, it is possible to perform the test with higher reliability and to perform OPC verification with higher accuracy.

  In addition, the present invention is not limited to OPC verification, and can be manually applied to a use for monitoring process variations. In addition to the shape measurement of the pattern transferred onto the silicon wafer, the same method can be applied to the shape measurement of the photomask.

  FIG. 10 is a flow of shape measurement according to the second embodiment of the present invention. The configuration of the SEM system in the present embodiment is basically the same as the configuration shown in FIG. 1 described in the first embodiment, but the configuration of the arithmetic unit 1010 is slightly different.

  FIG. 10A shows the configuration of the calculation unit 1020 in the present embodiment. The calculation unit 1020 includes a provisional contour extraction unit 10201, a local projection waveform formation 10202, a feature amount calculation unit 10203, a representative point extraction unit 10204, a library matching unit 10205, an edge position determination unit 10206, a coordinate conversion / contour line extraction unit 10207, A dimension measuring unit 10208 is provided.

  Hereinafter, each step will be described along the flow shown in FIG.

  (S110): Similar to S100 in the first embodiment, the provisional contour extraction unit 10201 generates a provisional contour line from the SEM image 20 imaged by the SEM 10 and output from the overall control / image processing unit 1009. Extract.

  (S210): Similar to S200 in the first embodiment, in the local projection waveform forming portion 10202, an image is projected in the tangential direction of the contour line in the local region along the contour line obtained in S100, A projection waveform is generated.

  (S211): In this step, the feature quantity calculation unit 10203 calculates various feature quantities for the pattern cross-sectional shape 1100 as shown in FIG. 11A from the local projection waveform generated in S210. In FIG. 11B, the feature quantity f1 is the white band width. The white band width is a feature amount that reflects the expected width of the edge portion when viewed from above in the vertical direction. The feature amount f2 is a feature amount that is an average width outside the peak position in the white band portion and reflects the magnitude of the curvature of the bottom portion 1102. The feature amount f3 is an average width in the white band portion, which is an average width inside the peak position, and is a feature amount that reflects the magnitude of the curvature of the top portion 1101. The feature quantity f4 is the magnitude of the signal intensity, and is a feature quantity that reflects the taper angle as shown in FIG.

  (S212): In this step, the representative point extraction unit 10204 determines a representative point for performing library matching based on the calculation result of the feature amount at each point on the temporary outline in S211. FIG. 12B shows how the feature quantity f1 shown in FIG. 11B, that is, the white band width changes along the temporary contour 122 on the SEM image 121 shown in FIG. It is shown whether or not. The white band width is almost constant from the start point to point P1 in FIG. 12 (a), then gradually increases to point P2, then decreases sharply to point P3, and remains almost constant from P3 to the end point. . It can be said that the transition of the feature amount is an appearance of the transition of the sectional shape. If the various feature amounts are constant, it means that the cross-sectional shapes are substantially equal, and therefore it is not necessary to estimate the cross-sectional shapes at all points in the next step.

  Therefore, in this step, a representative point on which the cross-sectional shape should be estimated is determined based on the transition of the feature amount. In FIG. 12B, the points with black circles are representative points. As shown in the figure, the representative point interval is wide in the region where the change rate of the feature amount is small and narrow in the region where the change rate of the feature amount is large. When a plurality of feature amounts are used as shown in FIG. 11B, the representative point interval is obtained for each feature amount, and the average value or the minimum representative point interval is adopted.

  (S310): In this step, the library matching unit 10205 matches the representative point determined in S210 with the SEM waveform library 30 created in advance and stored in the storage unit 1011 to estimate the cross-sectional shape. I do. The matching method is the same as the method described in S300 of the first embodiment.

  (S410): In this step, the edge position determination unit 10206 determines the edge position on the local projection waveform based on the cross-sectional shape estimated in S301. The method is the same as the method described in S400 of the first embodiment.

  (S411): In this step, the edge position determination unit 10206 calculates the position on the local projection waveform of the temporary contour calculated in S110 at each representative point (black triangle point in FIG. 13A) and the calculation in S410. The distance d from the edge position (black circle point in FIG. 13B) is calculated. The result of plotting the calculation result of d is shown as a white circle on FIG.

  (S412): In this step, the edge position determination unit 10206 interpolates the value d (white circle point in FIG. 13B) calculated at the representative point (the broken line in FIG. 13B is interpolated). Result) The d values of all points are obtained, and the positions of the temporary contour lines are corrected based on the d values, thereby determining the edge positions on all the projected waveforms.

  (S510) As in the first embodiment, the coordinate conversion / contour extraction unit 10207 converts the edge position on the local projection waveform determined in S412 to a position on the SEM image.

  (S610) As in the first embodiment, the coordinate transformation / contour extraction unit 10208 obtains the final contour as a series of coordinates on the SEM image obtained in S510.

  (S710): The dimension measuring unit 10208 measures the dimension of the desired position of the pattern or the interval between adjacent patterns using the contour line obtained in S610.

  The result measured in S710 is displayed on the screen of the output unit 1012 and stored in the server 1013 via the communication line.

  The display on the screen of the output unit 1012 is the same as that described with reference to FIG. 9 in the first embodiment.

  According to the second embodiment, library matching processing that requires computation time is performed only for representative points, so higher speed processing is possible.

  FIG. 14 is a flow of shape measurement according to the third embodiment of the present invention. The flow of shape measurement according to the present embodiment is basically the same as that of the first embodiment described with reference to FIG. 1, but in the first and second embodiments, the local projection waveform is used. In order to estimate the cross-sectional shape, the library 30 in which the cross-sectional shape 701 of the pattern stored in the storage unit 1011 is associated with the SEM waveform 702 is used. In the present embodiment, the cross-sectional shape of the pattern is used. The difference is that the SEM waveform feature quantity library 31 that associates the feature quantities calculated from 701 ′ and the SEM waveform 702 ′ is stored in the storage unit 1011 and is used. That is, in the processing flow shown in FIG. 14, S221 and S320 are different from the processing flow described in FIG.

  FIG. 15B shows a conceptual diagram of the SEM waveform feature amount library 31. The SEM waveform feature amount library 31 creates the SEM waveform library 30 as shown in FIG. 15A by the electron beam simulation shown in FIG. 7, and then the waveform shown in FIG. 11 from the obtained SEM waveform 702 ′. It is created by calculating feature amounts f1 to f4. As shown in FIG. 15A, the SEM waveform feature quantity library 31 is a table describing changes in feature quantities with respect to cross-sectional shape parameters (taper angle, rounding of top corner, etc.) for each feature quantity.

  In FIG. 14, in the step of matching with the SEM waveform feature quantity library 31 (S320), the feature quantity groups (f1 to fn) calculated from the local projection waveforms by applying the maximum likelihood method or the like most closely match. Find the cross-sectional shape parameter combination (taper angle, rounded top corner, etc.).

  Note that this embodiment can be combined with the second embodiment.

FIG. 16 is a flow of shape measurement according to the fourth embodiment of the present invention.
The flow of shape measurement according to the present embodiment is basically the same as that of the first embodiment described with reference to FIG. 1 except that a temporary contour line is extracted in this embodiment. After (S130), matching with the design data is performed, and by referring to the design data, the distance from the neighboring edge is calculated for each point on the contour line (S131).

  FIG. 17A shows an image diagram of the design data. In this figure, for example, the point of interest 111 is given the information “large distance” because there is no other edge in the vicinity, and the point of interest Reference numeral 112 gives the information “distance w” because the neighboring edge exists at the position of the distance w.

  The step of creating a local projection waveform (S230) is the same as S200 of the first embodiment. A conceptual diagram of the SEM waveform library 32 used in the present embodiment is shown in FIG. When creating the library, electron beam simulation is performed by changing the distance to the neighboring edge to obtain the relationship between the cross-sectional shape 701 ″ of the pattern and the SEM waveform 702 ″, and all the results are stored as a library. When matching the local projection waveform with the SEM waveform library 32 (S330), information on the distance to the neighboring edge assigned in S131 is used, and a library corresponding to the distance is used.

  Since the distance to the neighboring edge is not always the same as the design data, when performing matching (S330), not only the distance information given in S131 but also the given distance ± α [nm] library and You may make it match.

  The reason why the SEM waveform library 32 corresponding to the distance to the neighboring edge is used is that the secondary electron signal waveform changes depending on the distance to the neighboring edge. For example, when the edge interval is narrow, the secondary electron signal intensity from the part located between the edges is affected. When the pattern width or pattern interval becomes less than several tens of nanometers with the miniaturization of the circuit pattern, it is necessary to apply this embodiment.

FIG. 18 is a flow of shape measurement according to the fifth embodiment of the present invention.
The flow of shape measurement according to the present embodiment is basically the same as that of the first embodiment described with reference to FIG. 1 except that a temporary contour line is extracted in this embodiment. After (S140), a partial image is cut out in a local region along the temporary contour line (S240), and matching with the SEM image library 33 created by performing electron beam simulation on the three-dimensional cross-sectional shape is performed. (S340) point. The steps after S440 are basically the same as the steps after S400 described in FIG.

  This embodiment is effective in a portion where there is no linear pattern edge, such as a hole pattern or a line end. As in the fourth embodiment, the design data may be referred to switch between using a local projection waveform or a partial image for each part.

  FIG. 19A shows a flow of shape measurement according to the sixth embodiment of the present invention. Steps S150, S250, and S251 are the same as steps S110, S210, and S211 of the second embodiment. In the second embodiment, the waveform feature value calculated in S210 is used for the three choices of representative points. However, in this embodiment, the calculation result of the waveform feature value calculated in S251 is presented to the user. (S252) The process ends.

  FIG. 19B is an example of a waveform feature amount display screen (the parts are the same as those in FIG. 9). The figure shows the feature value f1 (white band width) calculated from the local projection waveform generated along the provisional contour line 112 extracted in S150, in a color-coded manner according to the value.

  As described above, since the waveform feature value reflects the cross-sectional shape of the sample, it is effective for the purpose of capturing how the cross-sectional shape changes depending on the part. The data that is the basis of the color-coded display is the data shown in the lower part of FIG. 12, but this data can be stored and used for process monitoring.

  In the first to fifth embodiments, a single contour line is output. For example, the height is 0% (2001), 25% (2002), 50% (2003), 75 on the SEM image 2000. % (2004) or the like, a plurality of edge ends may be output (see FIG. 20). Alternatively, since the cross-sectional shape at each point has been estimated, it may be output as a contour line.

  In resist patterns, it is important to understand the situation of so-called film reduction (resist height reduction). In particular, the results of light exposure simulations described in the section of “Background Technology” are used as targets for comparing output results. In the case of comparison, this embodiment is effective.

It is a flowchart explaining the 1st Embodiment of this invention. It is a figure which shows the example of the superimposition display of a SEM image and design data. It is a figure which shows the conventional technique for extracting an outline from a SEM image. It is a figure which shows another example of the prior art for extracting an outline from a SEM image. It is a figure explaining the problem of the prior art. It is explanatory drawing of the method of producing | generating a local projection waveform from a SEM image. It is a conceptual diagram of a SEM waveform library. It is a figure which shows an example of the GUI screen which a user defines an edge end. It is explanatory drawing of the comparison evaluation with a final output result and design data. It is a flowchart explaining the 2nd Embodiment of this invention. It is a figure which shows the example of a waveform feature-value. It is supplementary explanatory drawing of the 2nd Embodiment of this invention. It is supplementary explanatory drawing of the 2nd Embodiment of this invention. It is a flowchart explaining the 3rd Embodiment of this invention. It is supplementary explanatory drawing of the 3rd Embodiment of this invention. It is a flowchart explaining the 4th Embodiment of this invention. It is supplementary explanatory drawing of the 4th Embodiment of this invention. It is a flowchart explaining the 5th Embodiment of this invention. It is a flowchart explaining the 6th Embodiment of this invention. It is supplementary explanatory drawing of the 7th Embodiment of this invention.

DESCRIPTION OF SYMBOLS 10 ... Electron microscope 20 ... SEM image 30-33 ... Library 112 ... Temporary outline 113 ... Final output outline 114 ... Design data 202-20 ... Local rectangular area 210 ... Local projection waveform 212 ... Waveform feature quantity displayed in different colors S100 to S700 ... Shape measurement step

Claims (14)

An electron beam image acquisition means for acquiring an image of a desired portion of a sample having a pattern formed on the surface using a scanning electron microscope;
Storage means for storing a library associating the cross-sectional shape of the pattern of the sample with the electron beam signal waveform information corresponding to the cross-sectional shape of the pattern;
Dimension information extraction means for extracting dimension information of the pattern by processing information obtained from the image of the pattern of the sample acquired by the electron beam image acquisition means with reference to library information stored in the storage means; ,
An electron microscope system comprising: output means for displaying information related to the dimension of the pattern extracted by the dimension information extracting means on a screen.   2. The electron microscope system according to claim 1, wherein the dimension information extraction unit processes information obtained from the image of the sample with reference to library information stored in the storage unit. An electron microscope system characterized in that contour information of a pattern formed on the sample is obtained.   3. The electron microscope system according to claim 2, wherein the information obtained from the sample image in the dimension information extracting means is tangential to the pattern edge at each point or representative point of the pattern edge in the sample image. An electron microscope system characterized by a local projection waveform generated by projecting the image onto   2. The electron microscope system according to claim 1, wherein the dimension information extracting means extracts pattern width information and pattern height information as the dimension information of the pattern.   2. The electron microscope system according to claim 1, wherein the dimension information extraction unit extracts information on a plurality of pattern intervals as dimension information of the pattern, and the output unit extracts information on the intervals of the plurality of extracted patterns. Is displayed on a screen together with the design data of the plurality of patterns.   2. The electron microscope system according to claim 1, wherein the library stored in the storage means is obtained by computer simulation of electron beam signal waveforms of samples having various cross-sectional shapes. system.   2. The electron microscope system according to claim 1, wherein in the library stored in the storage unit, electron beam signal waveform information corresponding to the cross-sectional shape of the pattern associated with the cross-sectional shape of the pattern of the sample is at an edge of the pattern. An electron microscope system characterized by an electron beam signal waveform characteristic amount obtained from a local projection waveform generated by projecting an image of the pattern in a tangential direction. Obtain an image of the desired part of the sample with a pattern formed on the surface using a scanning electron microscope,
Storing a library associating the cross-sectional shape of the pattern of the sample with the electron beam signal waveform information corresponding to the cross-sectional shape of the pattern;
Extracting the pattern dimension information by processing the information obtained from the acquired pattern image of the sample with reference to the library information stored in the storage means,
A pattern dimension measuring method using an electron microscope system, wherein information relating to the extracted dimension of the pattern is displayed on a screen.   9. The pattern dimension measuring method according to claim 8, wherein extracting the dimension information of the pattern is performed by processing information obtained from the image of the sample with reference to the stored library information. A pattern dimension measuring method using an electron microscope system, wherein contour information of a pattern formed on the sample is obtained as dimension information.   10. The pattern dimension measuring method according to claim 9, wherein in the step of extracting dimension information of the pattern, information obtained from the sample image is a pattern at each point or representative point of a pattern edge in the sample image. A pattern dimension measurement method using an electron microscope system, wherein the pattern is a local projection waveform generated by projecting the image in a tangential direction with respect to an edge.   9. The pattern dimension measuring method according to claim 8, wherein, in the step of extracting dimension information of the pattern, pattern width information and pattern height information are extracted as the dimension information of the pattern. Pattern dimension measurement method using an electron microscope system.   The pattern dimension measuring method according to claim 8, wherein in the step of extracting dimension information of the pattern, information on a plurality of pattern intervals is extracted as the dimension information of the pattern and displayed on the screen. A pattern dimension measuring method using an electron microscope system, characterized in that information on intervals between the extracted patterns is displayed on a screen together with design data of the patterns.   9. The pattern dimension measuring method according to claim 8, wherein the library to be stored is obtained by computer simulation of electron beam signal waveforms of samples having various cross-sectional shapes. The pattern dimension measurement method used.   9. The pattern dimension measuring method according to claim 8, wherein in the library to be stored, electron beam signal waveform information corresponding to a cross-sectional shape of the pattern associated with the cross-sectional shape of the pattern of the sample is obtained with respect to an edge of the pattern. A pattern dimension measuring method using an electron microscope system, which is a feature amount of an electron beam signal waveform obtained from a local projection waveform generated by projecting an image of the pattern in a tangential direction.

Images