JP2013096711A - Method, apparatus and program for measuring semiconductor pattern, and storage medium storing semiconductor pattern measuring program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To solve such problems that a method for measuring a semiconductor pattern corresponding to various types of patterns have not been disclosed, especially, a method for measuring an object pattern with both sides thereof being constituted of straight or curved lines and being nonparallel have not been disclosed.SOLUTION: The method for measuring a semiconductor pattern is a method in which capturing an image of a pattern formed on a semiconductor wafer and measuring a pattern dimension on the basis of the captured image. The method includes the steps of: identifying edge positions on both sides of the pattern on the image; performing a measurement on the basis of a line segment connecting a first point and a second point, where the first point is located on a first edge of the pattern, and the second point is located on a second edge opposite to the first edge, a first tangential line of the first edge at the first point, and a second tangential line of the second edge at the second point; and measuring a distance of the line segment.

Description

  The present invention relates to a semiconductor pattern measurement method, a semiconductor pattern measurement device, a semiconductor pattern measurement program, and a storage medium storing a semiconductor pattern measurement program.

  In the dimension measurement of a semiconductor pattern, the pattern dimension is measured using a high-resolution image captured by a scanning electron microscope. If the pattern is a straight line and parallel to the vertical direction of the image, the position of both sides of the pattern is detected from the signal waveform obtained by vertically projecting the image signal of a certain section in the vertical direction of the image, and the pattern dimensions are determined. Can be sought. When the pattern is a straight line and parallel to the horizontal direction of the image, the same applies to the horizontal direction. On the other hand, with the increase in the density of semiconductor circuits, a curved pattern has been seen. As an edge detection method for a curved pattern, the method of Patent Document 1 is disclosed.

Patent No. 4040809

  Patent Document 1 is a method for obtaining a pattern edge of a curved pattern, and after projecting an image signal waveform projection section for obtaining an edge, the signal waveform is projected in the tangential direction of the pattern in the sectioned section. , Edge. However, a measurement method corresponding to various patterns is not disclosed, and in particular, there is no disclosure about a measurement method of a pattern in which both sides of a measurement target pattern are configured by straight lines or curves, and both sides are non-parallel.

  Semiconductor memory devices have been increased in capacity due to the miniaturization of patterns and the reduction of cell area by devising the pattern layout. The conventional pattern layout has been put into practical use in which bit lines and word lines are arranged in a linear lattice pattern to form memory cells, whereas bit lines are devised in a curved line shape. In the future, the appearance of various shape patterns is expected, and for the purpose of measuring these patterns by a comprehensive method, the object of the present invention is a pattern formed on a semiconductor wafer by an electron microscope image, Provided are a pattern dimension measuring method, a pattern dimension measuring apparatus, a program for causing a computer to execute a pattern dimension measuring method, and a storage medium for the pattern dimension measuring method, a pattern dimension measuring apparatus for measuring a pattern in which both sides are straight lines or curved lines and both sides are non-parallel with high accuracy. There is to do.

The following is a brief description of an outline of typical inventions disclosed in the present application.
(1) A method of capturing an image of a pattern formed on a semiconductor wafer and measuring a pattern dimension based on the captured image, the step of specifying edge positions on both sides of the pattern on the image, and a first pattern A line connecting the first point and the second point at a first point on the edge and a second point on the second edge that is the opposite edge; and the first point A step of measuring based on a first tangent of the first edge at the point, a second tangent of the second edge at the second point, and a step of measuring the distance of the line segment A semiconductor pattern measuring method characterized by the following.

  According to the present invention, it is possible to measure with high accuracy a pattern formed on a semiconductor wafer, the pattern of which both sides are constituted by straight lines or curves and whose both sides are non-parallel.

It is a block diagram of the Example of this application. It is an internal block diagram of an image processing part. It is a figure which shows the example of a captured image. It is a conceptual diagram which shows the conventional measuring method. It is a conceptual diagram which shows the measuring method by this application. It is a figure which shows the measuring method of the maximum width of a pattern width. It is a figure which shows the measuring method of the maximum width of a pattern width. It is a measurement processing flowchart of the maximum width. It is a figure which shows the example of a captured image. It is a figure of the signal waveform of a measurement object pattern. It is a conceptual diagram which shows the determination method of a definition corresponding | compatible point. It is a processing flow figure for determining a definition corresponding point. It is a measurement processing flowchart of the minimum width. It is a measurement processing flowchart of a designated position. It is a measurement processing flowchart of the maximum width. It is a measurement processing flowchart of the maximum width. It is a display example of a measurement result. It is a display example of a measurement result. It is a display example of a measurement result. It is a display example of a measurement result.

  FIG. 1 shows an overall configuration diagram of a fine pattern measuring apparatus according to the present invention. A length measuring SEM 100 according to this embodiment detects a stage 106 on which a measurement wafer 107 is placed, an irradiation optical system that controls an electron beam 101 emitted from an electron gun 102, and secondary electrons emitted from a sample. It comprises a detector 108 and a signal processing system for detection signals. The irradiation optical system includes an electron gun 102, a condenser lens 103, a deflection coil 104, and an objective lens 105 on the path of the electron beam 101. The electron beam 101 is condensed by this optical system in a predetermined area where the pattern to be measured on the wafer 107 is present. Secondary electrons detected by the detector 108 are converted into digital signals by the A / D converter 109. The converted digital signal is sent to the image processing unit 110. The image processing unit 110 extracts the digital signal stored in the memory as necessary, performs image processing, and performs pattern dimension measurement and the like. Reference numeral 111 denotes a stage controller, 112 denotes an electron optical system control unit, 113 denotes a control unit for the entire apparatus, and 114 denotes a control terminal connected to the control unit. A recording medium (not shown) can be connected to the image processing unit 110, the overall control unit 113, and the control terminal 114. A program executed by the image processing unit 110 is read from the recording medium, and an image is displayed. The processing unit 110 can be loaded.

  FIG. 2 shows a configuration diagram of the image processing unit 110. The secondary electron signal converted into a digital signal by the A / D converter 109 is sent to the memory 203 via the data input I / F 205 and stored so as to be readable as image data in the memory 203. The image processing program is read from the memory 203 or the storage medium by the image processing control unit 201. The image processing control unit 201 controls the arithmetic unit 202 according to a program, processes image data stored in the memory 203 or intermediate processing data obtained as a result of processing the image data, and measures a pattern. The pattern measurement result is sent to the overall control unit 113 via the input / output I / F 200, and the measurement result is displayed on the control terminal shown in FIG. An operation command for the image processing unit 110 is input from the overall control unit 113 to the image processing control unit via the input / output I / F 200. Data transmission / reception in the image processing unit 110 is performed via the bus 204.

  FIG. 3 is a schematic diagram of the captured image 301 of the measurement target pattern 302. A feature of the measurement target pattern in this embodiment is that there is a portion that is not parallel to the left edge 303 and the right edge 304. In FIG. 3, both edges 303 and 304 are drawn with curves, but one edge or both edges may be straight. As for the coordinates on the image, the horizontal right direction is defined as the x direction and the vertical downward direction is defined as the y direction. The measurement range 305 of the measurement target pattern 302 is a range between the upper limit 306 (y = ys) and the lower limit 307 (y = ye). In FIG. 3, the measurement range 305 is different in ys and ye and has a certain width, but ys and ye are not necessarily different.

  FIG. 4 shows a case where a conventional measurement method is applied to the pattern shown in FIG. In the conventional method, the pattern is scanned in the horizontal direction of the image, and the left and right edges are detected and measured. Specifically, the point L1 on the left edge 303 and the point R1 on the right edge 304 are detected at y = y1 in FIG. 4, and the distance between L1 and R1 is measured. Similarly, at other y positions, for example, y2 and y3, L2 and R2 were detected when y = y2, and L3 and R3 were detected when y = y3, and the distances between L2 and R2 and L3 and R3 were measured. When considering the minimum width of the pattern, as can be seen from FIG. 4, the width of the pattern can be appropriately measured with the pattern width obtained by scanning in the horizontal direction, and the minimum width cannot be found. FIG. 5 shows a conceptual diagram of a pattern width measuring method using the method of this embodiment. In this method, an angle α4 formed by a line segment L4-R4 connecting a point L4 on the left edge 303 and a point R4 on the right edge 304 and a tangent line (not shown) of the left edge 303 at L4, and a line segment L4 The points L4 and R4 at which the angles β4 formed by the tangent (not shown) of the right edge 304 at −R4 and R4 are equal are detected, and the distance between L4 and R4 is measured as the pattern width. Based on this definition, the pattern widths L5-R5 and L6-R6 can be determined at the same positions as y = y2 and y = y3 shown in FIG. 4, and the minimum pattern width can be obtained at y = y3. it can.

  FIG. 6 shows an example of measuring the maximum pattern width as an example in which a more remarkable effect can be obtained by the pattern width measuring method according to the method of the present embodiment. FIG. 6 is a schematic diagram of the pattern 606. Points 601 and 602 are points determined by this method. A broken line 605 is the circumference of a circle centered on the point 601 and having the distance between the points 601 and 602 as a radius. As shown in FIG. 6, the distance between the points 604 and 601 in the horizontal direction from the point 601 is larger than the distance between the points 601 and 602. That is, the points 604 and 601 give the maximum width when the edges on both sides are determined in the horizontal direction. FIG. 7 is obtained by rotating the pattern shown in FIG. 6 around a point 601. A broken line 701 is the circumference of a circle centered on the point 601 and having a distance between the points 601 and 604 as a radius. The distance between the point 702 and the point 601 in the horizontal direction from the point 601 is larger than the distance between the point 601 and the point 604, and the point 601 and the point 702 give the maximum width when the edges of both sides are determined in the horizontal direction. In this way, even with the same pattern, different maximum widths can be obtained with the conventional measurement method. On the other hand, according to the left and right edge points 601 and 602 according to the method of the present embodiment, the same point can be obtained even if the pattern is rotated, and the same maximum width can be obtained in FIGS.

  As described above, as shown in FIG. 5, the line segment lr connecting the point l on the left edge 303 (for example, L4 in FIG. 5) and the point r on the right edge 304 (for example, R4 in FIG. 5), The point l such that the angle formed by the tangent line of the left edge 303 at (for example, α4 in FIG. 5) and the angle formed by the tangent line of the right edge 304 at the line segment lr (for example, β4 in FIG. 5) are equal. By defining and r, it is possible to uniquely determine the minimum width or the maximum width with the same definition. Also, it is clear that the width can be uniquely determined at any specified location by this definition. This definition is hereinafter referred to as a corresponding point definition, and a pair of points determined by the corresponding point definition is referred to as a definition corresponding point pair. The conventional measurement of parallel straight lines corresponds to the case where the angle between the line segment between the definition corresponding points and the tangent on both sides is 90 degrees. Although the minimum width measurement shown in FIG. 5 can be realized by the shortest distance search, the measurement according to this definition can include such an object or measurement location with the same definition.

  FIG. 8 shows a flow of pattern measurement processing based on the above-described definition. The flow in FIG. 8 measures the maximum width or the average width in the vicinity of the position giving the maximum width. First, the edges 303 and 304 of the pattern in the image obtained by imaging the measurement target pattern are detected (S800). A method for detecting a pattern edge point will be described with reference to FIGS. FIG. 9 schematically shows a grayscale image obtained when the same pattern as in FIG. 3 is captured as an SEM image, and shows how the edges of the pattern become bright due to the edge effect. FIG. 10 shows an image signal waveform on the line segment AB shown in FIG. The horizontal axis of the graph in FIG. 10 is the x direction of the image, and the vertical axis is the signal value. Two peaks 1001 and 1002 of the waveform correspond to signal peaks at the pattern edge. By applying an arbitrary threshold value 1003, the left edge point is searched from the smaller x to the larger, and at the position xl where the signal value exceeds the threshold, the right edge point is from the larger x. The search is made in the smaller direction, and detection is possible at a position xr where the signal value exceeds the threshold value. By sequentially performing the same detection in the y direction, the position of the left edge point and the right edge point of the pattern at each ancestor position in the y direction can be detected, and the pattern edges 303 and 304 of the measurement target pattern are detected. be able to. Since each pixel position is in the y direction, y of the detected edge point is an integer value, but x can be a real value by using interpolation. The edges of the measurement target pattern may not be sufficiently smooth. For example, the roughness of the resist causes the edges to become rough. Without edge smoothness, the accuracy of the calculated tangent is reduced. Therefore, smoothing processing is performed on the obtained edge data as necessary. As the smoothing process, a Gaussian filter, an average value process, an intermediate value process, and the like can be considered, but the smoothing technique is not limited to this. As a result, the accuracy for obtaining the tangent is improved, and the reproducibility of the minimum width, the maximum width, or the measurement width at an arbitrary designated place is stabilized.

Next, the variable max is initialized (S801). Thereafter, S803 to S806 between S802a and S802b are repeated from Y = ys to Y = ye with a step width step. The value of step is arbitrary.
In step S803, the definition corresponding point R [Y] of the point L [Y] of y = Y on the left edge 303 is searched. The definition corresponding point search method will be described with reference to FIG. FIG. 11 is an enlarged view of a part of the pattern shown in FIG. For the point L [Y] at y = Y on the left edge 303, a corresponding point R [Y] of L [Y] defined by the corresponding point definition is searched. α1, α2, and α3 are the line segments L [Y] R1, L [Y] R2, L [Y] R3 connecting the point L [Y] with the points R1, R2, and R3, and the point L [Y]. Each angle formed by the tangent line 1100 of the left edge 303. β1 is an angle formed by the line segment L [Y] R1 and the tangent line 1101 of the right edge 302 at the point R1, β2 is an angle formed by the line segment L [Y] R2 and the tangent line 1102 of the right edge 302 at the point R2, β3 is an angle formed by the line segment L [Y] R3 and the tangent 1103 of the right edge 302 at the point R3. In FIG. 11, R2 is the definition corresponding point R [Y] of L [Y], and α1 = β2. As is clear from FIG. 11, α1 <β1 and α3> β3 at the points before and after R2, and the magnitude relationship is reversed across R2 which is the definition corresponding point corresponding to L [Y]. A definition corresponding point is searched using this reversing property.

FIG. 12 shows a detailed flow of S804 in FIG. First, a tangent L [Y] -tan at L [Y] is derived (S1200). The tangent line L [Y] -tan can be obtained by functional approximation of 2n + 1 points from L [Yn] to L [Y + n] adjacent to L [Y]. n is an arbitrary number. S1201a and S1201b represent loop calculations, and S1202 to S1206 in the meantime are repeated from yy = YS to Y = YE with a step width step. The value of step is arbitrary.
The tangent line R [yy] -tan of the right edge 304 at the point R [yy] of y = yy of the right edge 304 is derived (S1202). Next, an angle α formed between the line segment L [Y] R [yy] and the tangent line L [Y] -tan, and an angle β formed between the line segment L [Y] R [yy] and the tangent line R [yy] -tan are calculated. (S1203). If yy is not YS. It is confirmed whether (α−β) × (α1−β1) <0 (S1205). Since α1 and β1 are assigned in S1206 in the loop shown by S1201a and S1201b, α1 and β1 are substituted in S1206. This is the judgment of the reversal of the magnitude relationship of β. If (α−β) × (α1−β1) ≧ 0, α is substituted into α1 and β is substituted into β1 (S1206), and the loop is continued. If (α−β) × (α1−β1) <0, the process goes out of the loop, and R [Y] is determined based on, for example, the internal ratio as shown in S1207. Α and β given by L [Y] and R [Y] determined in this way are not exactly the same because they include a calculation error due to the calculation of a finite number of digits, but there is no problem in measurement.

  It is confirmed whether the midpoint of the searched definition corresponding points L [Y], R [Y] is within the measurement range (S804). If within the range, the distance between L [Y] and R [Y] is max. It is determined whether it is larger (S805). If larger, max is updated with the distance between L [Y] and R [Y], and the value of Y is recorded in Ymax (S806). After completion of the loop between S802a and S802b, variables D and count are cleared (S807)), the distance between L [Y] and R [Y] is added to D, and count is incremented (S809), Y = Ymax It repeats from −a / 2 to Y = Ymax + a / 2 with a step width step (S808a, S808b). The value of step is the same as S802a. a is the size of the interval in which the length measurement values are averaged, and a value of 0 or more is given. After the loop processing defined in S808a and S808b, the length measurement value is obtained by dividing D by count (S810).

  When there is a midpoint of the definition corresponding point in the measurement region in FIG. 13, the minimum value of the distance between the definition corresponding points or the average value of the distance between the definition corresponding points in the vicinity of the position that gives the minimum value is measured. Indicates the method of value. First, the edges 303 and 304 of the pattern in the image obtained by capturing the measurement target pattern are detected (S1300). The method for detecting the edge point of the pattern is as described with reference to FIGS. Next, a value equal to or larger than the diagonal distance of the image is substituted for the variable min (S1201). Thereafter, S1303 to S1306 between S1302a and S1302b are repeated from Y = ys to Y = ye with a step width step. The value of step is arbitrary. In step S1303, the definition corresponding point R [Y] of the point L [Y] of y = Y on the left edge 303 is searched. It is confirmed whether the midpoint of the searched definition corresponding points L [Y] and R [Y] is within the measurement range (S1304). If within the range, the distance between L [Y] and R [Y] is min. It is determined whether it is smaller (S1305). If smaller, min is updated with the distance between L [Y] and R [Y], and the value of Y is recorded in Ymin (S1306). After completion of the loop between S1302a and S1302b, the variables D and count are cleared (S1307), the addition of the distance of L [Y] and R [Y] to D, and the increment of count (S1309) are set as Y = Ymax− Repeat from a to Y = Ymax + a with a step width step (S1308a, S1308b). The value of step is the same as S1302a. a is the size of the interval in which the measurement values are averaged, and a value of 0 or more is given. After the loop processing defined in S1308a and S1308b, the length measurement value is obtained by dividing D by count (S1310).

  FIG. 14 shows a method of using the average value of the distance between the points corresponding to the definition corresponding points at the position designated in advance and the distance between the points corresponding to the definition corresponding points in the vicinity of the position designated in advance. First, the edges 303 and 304 of the pattern in the image obtained by imaging the measurement target pattern are detected (S1400). Next, the variables D and count are cleared (S1401). Thereafter, S1403 to S1405 between S1402a and S1402b are repeated from Y = Yc−a / 2−m to Y = Yc + a / 2 + m with a step width step. Yc is a measurement designated position. The value of step is arbitrary. a is the size of the interval in which the measurement values are averaged, and a value of 0 or more is given. m is a margin for the detection range of the definition corresponding point. Next, the definition corresponding point R [Y] of the point L [Y] of y = Y on the left edge 303 is searched (S1403), and the Y coordinate of the midpoint of L [Y] and R [Y] is the section [Yc. -A / 2, Yc + a / 2] is determined (S1304), and if within the interval, the distance between L [Y] and R [Y] is added to D, and count is incremented (S1405). . After completing the loop processing of S1402a and S1402b, a length measurement value is obtained by dividing D by count (S1406). Although not shown in FIG. 14, when S1304 does not pass through S1304 in the loop of S1402a and S1402b, for example, considering the case of a = 0, the midpoint between L [Y] and R [Y] closest to Yc The distance between L [Y] and R [Y] having the coordinates is stored separately, and if it never passes S1304, this separately stored value is taken as the measurement value.

  In the measurement flow of FIG. 8, when the midpoint of the definition corresponding point detected in the measurement range is the measurement region, the maximum value of the distance between the definition corresponding points or the definition corresponding point in the vicinity of the position where the maximum value is given. The method of using the average value of the distance between points as the measured value was shown. 15 and 16 show measurement processing flows by providing two different measurement areas.

  FIG. 15 is a flow of a method for measuring the maximum width of the pattern or the average width of the vicinity of the position giving the maximum width on condition that L [Y] enters the measurement range. First, the edges 303 and 304 of the pattern in the image obtained by imaging the measurement target pattern are detected (S1500). The method for detecting the edge point of the pattern is as described with reference to FIGS. In step S1501, the variable max is cleared. Thereafter, S1503 to S1505 between S1502a and S1502b are repeated from Y = ys to Y = ye with a step width step. The value of step is arbitrary. In 1503, the definition corresponding point R [Y] corresponding to the point L [Y] of y = Y on the left edge 303 is searched. It is determined whether the distance between L [Y] and point R [Y] is greater than max (S1504). If greater, max is updated with the distance between L [Y] and point R [Y], and the value of Y is set to Ymax. 130505). After completion of the loop between S1502a and S1502b, the variables D and count are cleared (S1506), and the addition of the distance of L [Y] and R [Y] to D and the increment of count (S1508) are set as Y = Ymax− Repeat from a to Y = Ymax + a with a step width step (1507a, S1507b). The value of step is the same as S1502a. a is the size of the interval in which the length measurement values are averaged, and a value of 0 or more is given. After the loop processing defined in S1507a and S1507b, the length measurement value is obtained by dividing D by count (S1509).

  FIG. 16 is a flow of a method for measuring the maximum width of the pattern or the average width of the width in the vicinity of the position that gives the maximum width on condition that R [Y] enters the measurement range. First, the edges 303 and 304 of the pattern in the image obtained by imaging the measurement target pattern are detected (S1600). The method for detecting the edge point of the pattern is as described with reference to FIGS. In step S1601, the variable max is cleared. Thereafter, S1603 to S1605 between S1602a and S1602b are repeated from Y = ys to Y = ye with a step width step. The value of step is arbitrary. In 1603, the definition corresponding point L [Y] corresponding to the point R [Y] of y = Y on the right edge 304 is searched. It is determined whether the distance between L [Y] and point R [Y] is larger than max (S1604). If larger, max is updated with the distance between L [Y] and point R [Y], and the value of Y is changed to Ymax. Store (S1605). After completion of the loop between S1602a and S1602b, the variables D and count are cleared (S1606), and the addition of the distance of L [Y] and R [Y] to D and the increment of count (S1608) are set as Y = Ymax− Repeat from a to Y = Ymax + a with a step width step (S1607a, S1607b). The value of step is the same as S1602a. a is the size of the interval in which the length measurement values are averaged, and a value of 0 or more is given. After the loop processing defined in S1607a and S1607b, the length measurement value is obtained by dividing D by count (S1609).

  In order to apply the processing flow shown in FIG. 15 and FIG. 16 to the minimum width measurement, the flow for measuring the maximum width in FIG. 8 is changed and the same change as the flow in FIG. 13 is made. Just do it. Also, in order to measure the processing flow shown in FIGS. 15 and 16 at a specified position, the determination in S1304 in FIG. 8 is performed, and in the case of FIG. 15, the Y coordinate of L [Y] is [Yc−a / 2, Yc + a / 2], in the case of FIG. 16, in the case of determining whether the Y coordinate of R [Y] is in the range [Yc-a / 2, Yc + a / 2]. Replace it.

  The above-described measurement processing method can be processed by a program executed by a computer, and this program is recorded in a storage medium such as a memory or an external storage medium, and is executed by being read from the storage medium.

  The processing procedure of the image data for measurement and the data obtained from the image data described with reference to FIGS. 8 to 12 is performed on the length measurement SEM shown in FIG. 1 and the image processing unit shown in FIG. The method of implementation is described below. A wafer 107 having a measurement target pattern is loaded onto the length measurement SEM 100 shown in FIG. After placing the wafer, the stage 106 is controlled from the overall control unit 113 through the stage controller 111, and the stage 106 is moved so that the measurement pattern on the wafer enters the field of view of the electron optical system. Next, the electron optical system control unit 112 controls the deflection coil 104, and the electron beam 101 scans the measurement target pattern or a pattern whose position relative to the measurement target pattern is known. Secondary electrons obtained from the measurement target are converted into digital signals by the A / D converter 109, and a digital image of the pattern is stored in the memory 203 via the data input I / F 205 in the image processing unit 110. The stored digital image is processed by the image processing control unit 201 using the calculation unit 202, the position of the pattern is obtained, detection position information is output from the image processing unit 110 to the overall control unit 113, and the overall control unit 113 receives The detected position information is transferred to the electron optical system processing unit 112. The electron optical system processing unit 112 converts the received detection position information into control information of the electron optical system, and controls the deflection coil 104 based on this to capture the pattern at the center of the visual field of the electron optical system. As a result, it is possible to accurately position and view the imaging target pattern.

  After positioning the measurement target pattern, the measurement target pattern is scanned with the electron beam 101 and the digital image is taken into the memory 203 in the image processing unit 110 in the same manner. The image data captured in the memory 203 is processed by a computer program for executing the above-described measurement processing procedure using the image processing control unit 201 and the calculation unit 202, and the measurement processing result is stored in the memory 203. The The program is input via an external storage medium (not shown) connected to the overall control unit 113 or the image processing unit, or a local area network connected to the overall control unit 113 or the image processing unit, and the memory 203 Alternatively, the program is stored in a non-volatile memory (not shown) in the memory 203 or the storage device 206, and the program is stored in the non-volatile memory (not shown) in the memory 203 or the storage device 206 from the next time. The program is read into the memory 203 and executed. The overall control unit reads out the measurement processing result stored in the memory via the input / output I / F 200 and outputs the measurement processing result to the display 114 of the control terminal, or the measurement processing result is connected to the overall control unit 113. Output to an external storage medium (not shown), or output the measurement processing result to a host server (not shown) via a local area network (not shown) connected to the overall control unit 113. Or In addition, the range on the image serving as the measurement area is input in advance via the control terminal 114 or a local area network connected to the overall control unit 113.

  FIG. 17 shows a display example of the measurement result of the pattern shown in FIG. FIG. 17 shows an example in which the minimum width is measured by the measuring method shown in FIG. Reference numeral 1700 denotes a display screen. The measurement value of the minimum width in the measurement range indicated by 1707 and 1708 is indicated by 1706, the definition corresponding points of the measurement points are indicated by 1704 and 1705, and the line segment connecting the definition corresponding points is indicated by 1701. Reference numeral 1702 denotes a line segment connecting the definition corresponding points at the upper end of the measurement range, and 1703 indicates a line segment connecting the definition corresponding points at the lower end of the measurement range.

  FIG. 18 is also a display example of the measurement result of the pattern shown in FIG. 3 and shows an example in which the minimum width is measured by the measurement method shown in FIG. Reference numeral 1801 denotes a line segment connecting a plurality of definition corresponding points detected in S803 in the loop of FIG. 8 excluding 1701, 1702, and 1703 shown in FIG. Reference numeral 1801 may be all detected line segments, or may be displayed with appropriate thinning. Reference numeral 1803 denotes a center line connecting the midpoints of 1801. It is a tangent of the 1802 pattern edge, and the position for displaying the tangent is arbitrary. Although the tangent line at the point 1804 is displayed in FIG. 18, the tangent line at the point designated by the user may be used. Moreover, although this example is the minimum width, you may display both the points of the definition corresponding point of the measurement location, or the tangent of one point. 17 and 18 are not displayed separately, and part or all of the display contents shown in FIGS. 17 and 18 may be displayed on one display screen. Note that the display of the line segment determined by the definition corresponding point and the two angles indicated by the display of both tangent lines are basically the same, but an error of the display resolution may be included.

  FIG. 19 is a display example of a result obtained by measuring the pattern shown in FIG. 3 with the distance between the points corresponding to the definition over the y direction. FIG. 20 is a schematic diagram in which patterns of the same process or the same product measured at different times are superimposed and displayed in the format shown in FIG. 19. By recording such display or data for display, it is possible to capture the change in pattern shape in more detail, and to perform more detailed process management. The display shown in FIGS. 19 and 20 may be displayed on the screen simultaneously with the display shown in FIG. 17 or FIG.

As described above, the curve pattern has been described. However, it is clear that the single width method based on the definition corresponding points can be applied to a pattern that is non-parallel on both sides and is configured with straight lines on one side or both sides. . The location to be measured may be a space that is a gap between patterns.
According to the present invention, in a pattern formed on a semiconductor wafer, a pattern in which both sides are constituted by straight lines or curves and a pattern in which both sides are non-parallel can be measured with a uniquely determined width, so that stable measurement is possible. As a result, highly accurate measurement with high measurement reproducibility is possible.

  The semiconductor pattern dimension measuring method of the present invention is a method of imaging a pattern formed on a semiconductor wafer and measuring the pattern dimension based on the captured image, and specifying the edge positions on both sides of the pattern on the image And connecting the first point and the second point at the first point on the first edge of the pattern and the second point on the second edge which is the opposite edge. Measuring based on a line segment, a first tangent of the first edge at the first point, and a second tangent of the second edge at the second point; and a distance of the line segment A semiconductor pattern dimension measuring method having a step of measuring.

Further, the semiconductor pattern dimension measuring method of the present invention is a pattern in which the edges on both sides are in a non-parallel relationship, and both sides are curved, or both sides are straight, or one side is straight and the opposite side is a curve. It was set as the semiconductor pattern dimension measuring method to do.
Further, the semiconductor pattern dimension measuring method of the present invention is a semiconductor pattern dimension measuring method in which the pattern dimension is the dimension of the pattern itself or the dimension of the space that is the gap between the patterns.
Further, the semiconductor pattern dimension measuring method of the present invention compares the angle formed by the line segment and the first tangent line with the angle formed by the line segment and the second tangent line, the first point, The semiconductor pattern dimension measuring method includes a step of measuring the distance of the line segment defined by the second point.

In the semiconductor pattern dimension measuring method of the present invention, the first angle formed by the line segment and the first tangent line is equal to the second angle formed by the line segment and the second tangent line. The semiconductor pattern dimension measuring method includes a step of measuring a distance between a line defined by the first point and the second point.
In the semiconductor pattern dimension measuring method of the present invention, the equal angle may be calculated as a difference between the first angle and the second angle, and the line segment displayed superimposed on the captured image. A semiconductor pattern dimension measuring method which is in a state including one or both of an angle represented by the first tangent line and a display error represented as a difference between a line segment and an angle represented by the second tangent line; did.
Further, the semiconductor pattern dimension measuring method of the present invention is a semiconductor pattern dimension measuring method for measuring the distances of a plurality of line segments within a predetermined range on an image.
The semiconductor pattern dimension measuring method of the present invention is a semiconductor pattern dimension measuring method in which midpoints of the plurality of line segments are within a range on the predetermined image.
Further, the semiconductor pattern dimension measuring method of the present invention is a semiconductor pattern dimension measuring method in which end points in a predetermined direction of a plurality of line segments are within a range on the predetermined image.
Further, in the semiconductor pattern dimension measuring method of the present invention, the minimum value, or the average value of the measured values at nearby positions that give the minimum value, or the maximum value, from the plurality of measurement values that are the measurement results of the distances of the plurality of line segments. Value or the average value of the measured values of the neighboring positions that give the maximum value, the measured value based on the predetermined position, or the average of the measured values of the neighboring positions that give the measured value based on the predetermined position The semiconductor pattern dimension measuring method using the value as the measurement result.
Moreover, the semiconductor pattern dimension measuring method of the present invention is a semiconductor pattern dimension measuring method for displaying the line segment superimposed on the captured image.
Further, the semiconductor pattern dimension measuring method of the present invention is a semiconductor pattern dimension measuring method for displaying the plurality of detected line segments superimposed on the captured image on the basis of the predetermined range on the image. .
Further, in the semiconductor pattern dimension measuring method of the present invention, the first tangent used for measurement, the second tangent, or the first tangent and the second tangent are superimposed and displayed on the captured image. It was set as the semiconductor pattern dimension measuring method to do.
Further, the semiconductor pattern dimension measuring method of the present invention is a semiconductor pattern dimension measuring method in which a midpoint of the line segment or a line passing through the midpoint of the line segment is displayed superimposed on the captured image.
In the semiconductor pattern size measuring method of the present invention, the distances of the detected plurality of line segments are displayed in a superimposed manner on the captured image based on the predetermined range on the image, or a plurality of lines A semiconductor pattern dimension measuring method for recording a minute distance was adopted.
Further, the semiconductor pattern dimension measuring method of the present invention superimposes a plurality of line segment distances for each pattern detected from the measurement target pattern measured at different times on the basis of the predetermined range on the image. The semiconductor pattern size measurement method records the result that can be output or superimposeable.

In the semiconductor pattern dimension measuring method of the present invention, the semiconductor pattern dimension measuring method includes a step of smoothing the detected edge point sequence.
Moreover, the semiconductor pattern dimension measuring method of the present invention is a semiconductor pattern dimension measuring method having an imaging step with an electron beam microscope.
In order to achieve the above object, a semiconductor pattern dimension measuring apparatus of the present invention is an apparatus that images a pattern formed on a semiconductor wafer and measures the pattern dimension based on the captured image, and images the pattern. An electron optical system; an electron receiving unit obtained from the pattern; a converting unit that converts the received signal into a digital signal; a storage unit that stores the digital signal as a digital image; and the storage unit that stores the digital signal. Based on the information of the pattern, identification of edge positions on both sides of the pattern, and a first point on the first edge of the pattern and a second on the second edge that is the opposite edge A line segment connecting the first point and the second point, a tangent line of the first edge at the first point, and the line segment and the second point. Based on the tangent of the second edge 1 point and to obtain a semiconductor pattern dimension measuring hand device having a calculation unit for calculating the distance of a point of the second.
Further, the calculation unit in the semiconductor pattern dimension measuring apparatus of the present invention compares the angle formed between the line segment and the first tangent line with the angle formed between the line segment and the second tangent line. A semiconductor pattern dimension measuring device having a function of calculating a distance between a point and a line segment defined by the second point is provided.
The semiconductor pattern dimension measuring apparatus according to the present invention further includes an input unit for inputting a range on an image for defining a range for detecting a plurality of the point pairs, and a point having a minimum distance among the plurality of detected point pairs. A line segment connecting pairs, a line segment connecting point pairs having the maximum distance, a line segment connecting point pairs at a predetermined position, or between all or some of the plurality of detected point pairs. A semiconductor pattern dimension measuring apparatus having a display unit that displays the connecting line segment superimposed on the image.

In order to achieve the above object, a semiconductor pattern measurement program of the present invention is a semiconductor pattern measurement program capable of executing the above semiconductor pattern measurement method by a computer.
In order to achieve the above object, a storage medium storing a semiconductor pattern measurement program of the present invention stores a computer-readable semiconductor pattern measurement program recording a program for causing a computer to execute the semiconductor pattern measurement method. It was.

DESCRIPTION OF SYMBOLS 100 ... Measuring SEM, 101 ... Electron beam, 102 ... Electron gun, 103 ... Condenser lens, 104 ... Deflection coil, 105 ... Objective lens, 106 ... Stage, 107 ... Wafer, 108 ... Detector, 109 ... A / D converter , 110 ... Image processing unit, 111 ... Stage controller, 112 ... Electro-optical system control unit, 113 ... Overall control unit, 114 ... Control terminal, 200 ... Input / output I / F, 201 ... Image processing control unit, 202 ... Calculation unit , 203 ... memory, 204 ... bus, 205 ... data input I / F, 301 ... captured image, 302 ... measurement target pattern, 303 ... left edge, 304 ... right edge, 305 ... measurement range, 306 ... upper limit, 307 ... lower limit , 601, 602 ... definition corresponding point, 603 ... line segment, 604 ... point, 605 ... circumferential line, 606 ... pattern, 701 ... circumferential line, 70 ..., points 1001, 1002 ... waveform peak, 1003 ... threshold, 1100, 1101, 1102, 1103 ... tangent, 1700 ... display screen, 1701, 1702, 1703 ... line segment display, 1704, 1705 ... definition corresponding point display, 1706 ... Measured value display, 1707 ... Measurement range upper end display, 1708 ... Measurement range lower end display, 1801 ... Definition corresponding point display, 1802 ... Tangential display, 1803 ... Definition corresponding point center line display, 1804 ... Definition corresponding point

Claims (23)

A method of imaging a pattern formed on a semiconductor wafer and measuring a pattern dimension based on the captured image,
Identifying edge positions on both sides of the pattern on the image;
A line segment connecting the first point and the second point at the first point on the first edge of the pattern and the second point on the second edge which is the opposite edge. Measuring based on a first tangent of the first edge at the first point and a second tangent of the second edge at the second point;
A method of measuring a semiconductor pattern, comprising the step of measuring a distance of the line segment.   2. The semiconductor pattern according to claim 1, wherein the two side edges are in a non-parallel relationship, and are curved on both sides, straight on both sides, or straight on one side and curved on the opposite side. Measurement method.   2. The semiconductor pattern measuring method according to claim 1, wherein the pattern dimension is a dimension of a pattern itself or a dimension of a space which is a gap between patterns.   The pattern measurement method according to claim 1, wherein an angle formed between the line segment and the first tangent line is compared with an angle formed between the line segment and the second tangent line, the first point, A method for measuring a semiconductor pattern comprising measuring a distance of a line segment defined by a second point.   The pattern measurement method according to claim 1, wherein a first angle formed by the line segment and the first tangent line is equal to a second angle formed by the line segment and the second tangent line. A method for measuring a semiconductor pattern, comprising: measuring a distance between a line defined by the first point and the second point.   The equal angle includes a calculation error expressed as a difference between the first angle and the second angle, an angle represented by the line segment displayed on the captured image and the first tangent line, and 6. The semiconductor pattern measurement method according to claim 5, wherein one or both of display errors appearing as a difference in display of the angle represented by the line segment and the second tangent line are included.   The semiconductor pattern measuring method according to claim 1, wherein the distances of the plurality of line segments are measured within a predetermined range on the image.   8. The semiconductor pattern measuring method according to claim 7, wherein the midpoints of the plurality of line segments are within a range on the predetermined image.   8. The semiconductor pattern measuring method according to claim 7, wherein end points in a predetermined direction of the plurality of line segments are within a range on the predetermined image.   From the plurality of measurement values that are the measurement results of the distances of the plurality of line segments, the minimum value, or the average value of the measurement values of the neighboring positions that give the minimum value, or the measurement of the neighboring position that gives the maximum value 10. The measurement result is an average value of values, a measurement value based on a predetermined position, or an average value of measurement values at nearby positions that give a measurement value based on a predetermined position. The semiconductor pattern measuring method according to any one of the above.   The semiconductor pattern measurement method according to claim 1, wherein the line segment is displayed so as to be superimposed on the captured image.   The semiconductor pattern measuring method according to claim 7, wherein the plurality of detected line segments are superimposed on the captured image with reference to the predetermined range on the image. .   2. The semiconductor according to claim 1, wherein the first tangent used for measurement, the second tangent, or the first tangent and the second tangent are superimposed and displayed on the captured image. Pattern measurement method.   The semiconductor pattern measuring method according to claim 1, wherein a midpoint of the line segment or a line passing through the midpoint of the line segment is displayed superimposed on the captured image.   The distances of the plurality of detected line segments are displayed superimposed on the captured image based on the predetermined range on the image, or the distances of the plurality of line segments are recorded. Item 10. A semiconductor pattern measurement method according to any one of Items 7 to 9. Based on the predetermined range on the image, the distance of a plurality of line segments for each pattern detected from the measurement target pattern measured at different times is superimposed and output, or the superimposable result is recorded. The semiconductor pattern measuring method according to claim 7, wherein the semiconductor pattern measuring method is a semiconductor pattern measuring method.   2. The semiconductor pattern measuring method according to claim 1, wherein the step of specifying the edge position further comprises a step of smoothing the detected edge point sequence.   The semiconductor pattern measurement method according to claim 1, wherein the imaging is imaging with an electron beam microscope. An apparatus for imaging a pattern formed on a semiconductor wafer and measuring a pattern dimension based on the captured image,
An electron optical system for imaging the pattern;
An electron receiver obtained from the pattern;
A converter for converting the received signal into a digital signal;
A storage unit for storing the digital signal as a digital image;
Based on the information of the pattern stored in the storage unit, the edge positions on both sides of the pattern are identified, the first point on the first edge of the pattern, and the first edge that is the opposite edge. A pair of second points on two edges, a line segment connecting the first point and the second point, a tangent line of the first edge at the first point, and the line segment And a computing unit that computes the distance between the first point and the second point based on the tangent of the second edge at the second point. The pattern measuring device according to claim 19,
The calculation unit is defined by the first point and the second point by comparing the angle formed by the line segment and the first tangent line with the angle formed by the line segment and the second tangent line. A semiconductor pattern measuring device, characterized in that it is a calculation unit having a function of calculating a distance of a line segment to be measured. The pattern measuring device according to claim 19,
Furthermore, an input unit for inputting a range on the image that defines a range for detecting a plurality of pairs of points;
Among the plurality of detected point pairs, a line segment connecting the point pairs having the minimum distance, a line segment connecting the point pairs having the maximum distance, a line segment connecting the point pairs at predetermined positions, or A semiconductor pattern measuring apparatus, comprising: a display unit configured to superimpose and display a line segment connecting all or a part of the plurality of detected point pairs on the image.   A semiconductor pattern measurement program for causing a computer to execute the semiconductor pattern measurement method according to claim 1.   A storage medium storing a computer-readable semiconductor pattern measurement program recording a program for causing a computer to execute the semiconductor pattern measurement method according to claim 1.

Images