JPH11243047A - Positioning method in charged-beam exposure and line-width measuring method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To measure connecting accuracy directly at the circuit pattern on a substrate, by measuring the connecting accuracy between transferred patterns by the image comparison of a plurality of the transferred patterns with the transferred pattern, which is formed by the transfer based on a plurality of the transferred pattern and the pattern, wherein the connecting part is deviated beforehand. SOLUTION: The connecting accuracy between the transferred patterns as a plurality of patterns at the different times is measured by the image comparison of a plurality of the transferred patterns, the transferred pattern which is obtained by transferring the designed pattern wherein the connecting part is deviated beforhand in the design stage by exposure, and development or the transferred pattern which is estimated by simulation. For example, the design pattern of line and base, which are deviated by, e.g. 20 nm each as the specified amount in the Y direction, is formed. The design pattern is exposed to photosensitive resist film formed on a substrate by exposure and development. The design pattern is exposed. Furthermore, the development of the exposed resist film is performed. Thus, the transfer pattern when the pattern forming is actually performed is obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a positioning method and a line width measuring method in charged beam exposure, and more particularly, to a charged beam for transferring a continuous pattern as a plurality of patterns at different times on a substrate at a design stage. The present invention relates to a positioning method and a line width measuring method in exposure.

[0002]

2. Description of the Related Art As a technique for forming a desired fine pattern on a sample surface such as a semiconductor substrate, a charged beam exposure technique has attracted attention. A high throughput (processing capacity per unit time) is required for forming a large-scale circuit pattern. Therefore, at present, the circuit pattern is divided into rectangles and trapezoids with different dimensions for exposure (variable molding exposure),
Exposure collectively of these rectangles and trapezoids (batch exposure)
The method of doing is becoming mainstream. There is also a charged beam exposure apparatus having both types of exposure. In such an exposure method, a series of figures is drawn in a plurality of shots in the design stage while being divided into a plurality of shots in the design stage, so that connections are made between the patterns transferred in each shot.

In the charged beam exposure technique, a desired pattern is transferred to an arbitrary position on a substrate by beam deflection and stage movement. However, if the area that can be drawn by beam deflection is smaller than one circuit area, the beam is deflected. Connections between the deflection areas occur. As described above, when a series of figures are drawn in a plurality of shots in the exposure stage in the design stage, a pattern for measuring connection accuracy has conventionally been required to be drawn simultaneously.

[0004]

However, it is necessary to measure the connection accuracy. However, in the conventional method, the position is detected by scanning the pattern for measuring the connection accuracy with an electron beam. There are problems such as the need for a dedicated pattern, the need for a dedicated pattern with a large line width, and the need for a high step in the dedicated pattern (in other words, the need to increase the film thickness of the dedicated pattern). In the case of the problem described in the above section, it is necessary to insert an extra measurement pattern in the circuit pattern in order to measure the connection accuracy of the circuit pattern. descend. In addition, the circuit area is increased, which adversely affects the yield. In the case of the above problems, if the design rules for the connection accuracy measurement pattern and the circuit pattern are significantly different, there is a possibility that the connection accuracy differs between the circuit pattern and the measurement pattern, making accurate measurement difficult. In the case of the above problems, it is necessary to make the thickness of the connection accuracy measurement pattern different from the circuit pattern, and it is not possible to draw at the same time. Furthermore, inserting an extra measurement pattern in the circuit pattern increases the number of rectangles that need to be drawn and decreases the throughput.

[0005]

SUMMARY OF THE INVENTION The present invention is a positioning method in charged beam exposure which has been made to solve the above problems. A first aspect of the present invention is a positioning method in a charged beam exposure for transferring a continuous pattern as a plurality of patterns at different times on a substrate at a design stage, wherein the connection accuracy between the patterns transferred as the plurality of patterns is transferred. For example, based on the plurality of patterns thus determined and the
The measurement is performed by comparing an image with a transfer pattern formed by transfer by development or a transfer pattern predicted by simulation.

[0006] In the positioning method in the charged beam exposure according to the first invention, the connection accuracy between the patterns transferred as a plurality of patterns at different times is determined in advance by connecting the transferred portions with the transferred patterns at the design stage. Since the shifted pattern is measured by comparing an image with a transfer pattern formed by transferring by exposure and development or a transfer pattern predicted by simulation, there is no need to provide a dedicated pattern for measuring connection accuracy. Moreover, the connection accuracy is measured with the circuit pattern formed on the substrate without considering the difference in the design rules.

According to a second aspect of the present invention, there is provided a positioning method in a charged beam exposure in which a continuous pattern is transferred as a plurality of patterns on a substrate at different times in a design stage, and a connection accuracy between the patterns transferred as a plurality of patterns is provided. Based on a plurality of transferred patterns and a pattern in which connection portions are shifted in advance in a design stage, for example,
The measurement is performed by comparing dimensions with a transfer pattern formed by transfer by development or a transfer pattern predicted by simulation.

[0008] In the positioning method in the charged beam exposure according to the second aspect of the present invention, the connection accuracy between the patterns transferred as a plurality of patterns at different times is determined by setting the connection between the transferred plurality of patterns and the connecting portion in advance at the design stage. Since the shifted pattern is measured by comparing dimensions with a transfer pattern formed by transferring by exposure and development or a transfer pattern predicted by simulation, there is no need to provide a dedicated pattern for measuring connection accuracy. Moreover, the connection accuracy is measured with the circuit pattern formed on the substrate without considering the difference in the design rules.

A third aspect of the present invention is a line width measuring method for measuring the width of a pattern using an electron beam, wherein an image obtained by scanning the pattern to be measured with an electron beam in a certain direction. The dimension of the length pattern to be measured is measured based on the distribution of the sum of outputs.

In the line width measuring method according to the third aspect of the invention, the distribution of the sum of image outputs in a certain direction of an image obtained by scanning the pattern to be measured with an electron beam is determined by the following method. Because the scattering of the electron beam changes at the edge,
The image output also changes. Therefore, it is possible to measure the dimension of the length pattern to be measured based on the change. In such a measurement method, measurement using a circuit pattern is possible, so that measurement accuracy is improved.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An example of an embodiment according to the first invention will be described with reference to a flowchart of FIG. 1, an explanatory diagram of a pattern in which connection portions are shifted in a design stage shown in FIG.
This will be described with reference to the explanatory view of the pattern after exposure and development shown in (2). In the flowchart of FIG. 1, a circuit pattern that is continuous in a design stage is divided into a plurality of shots (divided into a plurality of drawing fields) at different times and is drawn on a substrate to be transferred as a plurality of patterns. This is shown below. In (1) and (2) of FIG. 2, a direction perpendicular to the connection surface of the pattern is defined as a Y direction, and a direction parallel to the connection surface is defined as an X direction.

First, according to the "creation of hinaga pattern data in which the connection position is shifted little by little" (see FIG. 1), for example, as shown in FIG.
A line-and-space design pattern 11 shifted by nm is created. This design pattern 11 is a pattern in which the connection portions are shifted in advance in the design stage referred to in the claims. Then, by "exposure" and "development" (see FIG. 1), the photosensitive resist film (not shown) formed on the substrate is exposed to the design pattern 11, and the exposed resist film is developed. . When the pattern is actually formed in this way, the transfer pattern 21 as shown in (2) of FIG. 2 is obtained, thereby completing “the completion of the hinaga pattern” (see FIG. 1). This transfer pattern 21 is a transfer pattern created based on a pattern in which connection portions are shifted in advance in a design stage referred to in the claims.

[0013] Alternatively, according to "Creation of hinaga pattern data in which connection positions are slightly shifted from each other" (see FIG. 1), for example, as shown in (1) of FIG. An AND space design pattern 11 is created. By performing “exposure simulation” and subsequently “development simulation” (see FIG. 1) based on the design pattern 11, a transfer pattern 31 as shown in FIG. 3 is obtained. "Completion of the pattern" (see FIG. 1) may be performed in this manner. The transfer pattern 31 is a transfer pattern created based on a pattern in which connection portions are shifted in advance in a design stage referred to in the claims.

Next, the image of the connection portion in the transfer pattern 21 as shown in (2) of FIG. 2 is stored together with the shift amount by "image storage of the hinaga pattern for each connection shift amount" (see FIG. 1). . Alternatively, the image of the connection portion in the transfer pattern 31 as shown in FIG. 3 obtained by the exposure simulation and the development simulation is stored together with the shift amount. Thus, the creation of the pattern data is completed.

Next, by "searching for pattern suitable for connection accuracy measurement from pattern data and storing coordinates" (see FIG. 1), a figure suitable for connection accuracy measurement, that is, hinaga, is obtained from the stored hinaga pattern data. The pattern is searched, and the coordinates of the searched pattern are stored. Subsequently, an image of a connection portion of a transfer pattern for measuring the amount of deviation (however, there is no deviation at the design stage) is captured by "image capture of connection portion" (see FIG. 1). The transfer patterns for measuring the shift amount are a plurality of transferred patterns referred to in the claims.

Next, the displacement of the connecting portion is measured by comparing the images. This image comparison employs a so-called template matching method. That is, pattern matching is performed by “pattern matching” (see FIG. 1) between an image in which a pattern of hinata patterns formed by shifting connection portions in various ways and a connection portion of a transfer pattern for measuring a shift amount. Then, the image of the hinaga pattern that best matches the connection portion of the transfer pattern is searched for and selected. Then, the shift amount set in the pattern or the shift amount calculated by fitting is set as the connection accuracy.

At this time, the amount of connection deviation may be estimated based on the degree of pattern matching by “calculation of connection accuracy from matching rate”. For example, the shift amount is +20 nm
When the matching ratio is 40% and the matching ratio is +40 nm and the matching ratio is 80%, the shift amount is [40 / (40 + 8)
0) × 20] + [80 / (40 + 80) × 40] = 3
It becomes 3.3 nm.

The pattern may be a transfer pattern transferred to an actual resist film or a transfer pattern obtained by exposure simulation and development simulation.

In the positioning method in the charged beam exposure according to the first invention, the connection accuracy between the patterns transferred as a plurality of patterns at different times is determined in advance by connecting the transferred portions to the transferred patterns at the design stage. The shifted design pattern 11 is measured by comparing an image with a transfer pattern 21 obtained by transferring by exposure and development or a transfer pattern 31 predicted by simulation. Therefore, the connection accuracy can be obtained from the pattern in the circuit pattern. Therefore, it is not necessary to provide a dedicated pattern for measuring the connection accuracy, and it is not necessary to consider differences in design rules. Since the connection accuracy can be measured using the circuit pattern in this manner, there is no increase in the area for forming a pattern dedicated to measurement. for example,
Even if a dedicated connection accuracy measurement pattern is added, a very small area increase is sufficient. This is because, for example, measurement can be performed using only one line pattern. Also in this case, it is possible to design with the same design rule as that of the circuit pattern, so that the influence of the difference in the design rule can be eliminated. Further, since a film thickness similar to that of the circuit pattern is sufficient, it is possible to draw simultaneously with the circuit pattern.

Next, an example of the embodiment according to the second invention will be described with reference to the flowchart of FIG. In the flowchart of FIG. 1, a circuit pattern that is continuous in a design stage is divided into a plurality of shots (divided into a plurality of drawing fields) at different times and is drawn on a substrate, and is transferred as a plurality of patterns. Show the case.

In the same manner as described with reference to the flow chart of FIG. 1, as shown in FIG. 4, "creation of hinaga pattern data in which connection positions are slightly shifted", "exposure",
By "development", a transfer pattern 21 as shown in FIG. 2 (2) is obtained. In this way, the “finish pattern” is completed. Alternatively, the transfer pattern 31 as shown in FIG. 3 is obtained by performing “exposure simulation” and “development simulation” after “creation of the pattern data in which the connection position is slightly shifted”. "Completion of the hinata pattern" may be performed in this manner.

Thereafter, when the transfer pattern 21 obtained by actual exposure and development is used as the pattern, the line width of the connection portion of the transfer pattern 21 is measured by "measurement of the connection portion". On the other hand, exposure simulation,
When the transfer pattern 31 obtained by the development simulation is used, the length of the line width of the connection part of the transfer pattern 31 is measured by “measurement of the connection part”. At this time, the thickest or thinnest part is the connection part,
Measure there. The measurement of the line width of the connection portion will be described in detail in an embodiment according to the third invention.

The relation between the line width change amount of the connection part (vertical axis) and the connection deviation amount set in the design stage (horizontal axis) is obtained by "creation of a relational graph of connection part line width and connection part deviation amount". Is obtained, and a graph showing the above relationship as shown in FIG. 5 is created. Also, in consideration of a case where the transfer pattern is separated due to an excessively large connection shift amount, that is, a case where the length measurement result is 0, the connection portion separation amount (vertical axis) as shown in FIG.
And the connection deviation amount (horizontal axis) set at the design stage is determined in advance. Thus, the creation of the pattern data is completed.

Next, in the same manner as described with reference to FIG. 1, "search for pattern suitable for connection accuracy measurement and storage of coordinates from pattern data" and "image capture of connection portion" are performed.

Next, the displacement of the connecting portion is measured by comparing the dimensions. In this dimensional comparison, based on the measurement results of the thickness due to the overlap of the pattern of the connection portion or the thinness due to the separation of the pattern, the measurement is performed from the transfer pattern 21 or the transfer pattern 31 in which the connection portion is variously shifted in advance. A pattern that best matches the long result is selected, and a set shift amount in the selected pattern or a shift amount calculated by fitting is set as connection accuracy.

That is, the dimensions of the connection portion of the transfer pattern for measuring the amount of displacement taken in the image are measured by "image processing". Further, the amount of displacement in the X direction is measured by “calculation of the amount of displacement in the X direction”. The displacement in the X direction is the first
Like a pattern having a shift amount in the Y direction as described in the embodiment of the present invention, a design pattern 41 shifted in the design direction in the X direction as shown in FIG. A transfer pattern 51 as shown in FIG. 7B obtained by performing exposure and development based on the design pattern 41.
8 is used to determine the connection accuracy by comparing the images, and as shown in FIG.
There is a method of directly measuring the length d of the deviation B in the X direction. And "calculation of connection accuracy from the relationship between line width and shift amount"
Thus, the relationship between the connection line width and the connection portion shift amount is obtained, and the connection accuracy is calculated from the relationship.

In the positioning method in the charged beam exposure according to the second aspect of the present invention, the connection accuracy between patterns transferred as a plurality of patterns at different times is determined by connecting the transferred portions with the transferred patterns in advance in the design stage. The shifted design pattern 11 is measured by comparing dimensions with a transfer pattern 21 obtained by transferring by exposure and development or a transfer pattern 31 predicted by simulation. Therefore, the connection accuracy can be obtained from the pattern in the circuit pattern. Therefore, it is not necessary to provide a dedicated pattern for measuring the connection accuracy, and it is not necessary to consider differences in design rules. Since the connection accuracy can be measured using the circuit pattern in this manner, there is no increase in the area for forming a pattern dedicated to measurement. for example,
Even if a dedicated connection accuracy measurement pattern is added, a very small area increase is sufficient. For example, it is possible to perform measurement using only one line pattern. Also in this case, it is possible to design with the same design rule as that of the circuit pattern, so that the influence of the difference in the design rule can be eliminated. Furthermore, since a film thickness similar to the circuit pattern is sufficient,
It becomes possible to draw simultaneously with the circuit pattern.

Next, an example of an embodiment according to the third invention will be described with reference to FIGS. In FIG. 9, a scanning electron microscope (SEM) is shown.
2 shows a schematic diagram of an image of a measured length pattern captured by the method described above, and a graph of data obtained as a result of image processing for adding the luminance of pixels in the vertical direction. FIG. 10 shows connection examples of various patterns, and FIGS. 11 to 14 show graphs obtained by performing image processing on each pattern.

As shown in FIG.
In the design stage, a continuous circuit pattern is divided into a plurality of shots (divided into a plurality of drawing fields) at different times, drawn and transferred onto a substrate as a plurality of patterns. This measured pattern 111
Image obtained by scanning an electron beam (for example, SEM
In the image, the edge portion (step portion) E of the measured length pattern 111 is generally bright, and the flat surface S is darkly displayed. Therefore, the distribution I of the sum of the image outputs in a fixed direction of the measured pattern 111, for example, the length direction of the measured pattern 111 (Y direction in the drawing) is obtained, and based on the obtained distribution I, the size and pattern of the measured pattern 111 are determined. The amount of connection deviation etc. is measured. More specifically, if the luminance of the pixels in the direction parallel to the length direction of the measured pattern 111 (the Y direction in the drawing) is added to make a graph, the line width of the connection portion 112 and the line width of the connection portion 112 can be determined without specifying the connection position. It becomes possible to measure the displacement in the X direction of the drawing. Details will be described in an embodiment according to the third invention.

In order to facilitate understanding, FIG. 10 shows the pattern shape (portion indicated by a two-dot chain line) and the pattern shape (portion indicated by a solid line) shown in the SEM image at the stage of designing various measured patterns. 11 to FIG. 14 and FIG.
7 shows a graph obtained by adding the luminance in the vertical direction (Y direction in the drawing) of each measured pattern indicated by. 11 to 14, the vertical axis indicates luminance, and the horizontal axis indicates position.

As shown in FIG. 10A, the measured length pattern 111A which overlaps in the Y direction and is displaced in the X direction.
Then, the luminance distribution is as shown in FIG. As shown in FIG. 10B, the measured length pattern 111B overlapping in the Y direction and having no displacement in the X direction has a luminance distribution as shown in FIG. As shown in FIG.
Length pattern 1 that is separated in the X direction and has no deviation in the X direction
In 11C, the luminance distribution is as shown in FIG. FIG.
As shown in (D) of FIG. 0, in the measured pattern 111D separated in the Y direction and shifted in the X direction, the luminance distribution is as shown in FIG.

From the measured length pattern 111B as shown in FIG. 10B, the line width Ws of the connecting portion 112B and
The line width W of the normal part 113B other than the connection part 112B can be obtained. That is, as shown in FIG. 12, a portion indicating a bright pixel portion exists outside the peak Pb of the normal portion due to the thickness of the connection portion 112B, and the line width Ws of the connection portion is obtained from the waveform Lb of that portion. In the normal case, when obtaining the line width from the waveform data (detecting the edge of the edge), the position where the waveform is flat (the first differential value is 0) or the inflection point position of the waveform (the second differential value) Uses 0).

In the embodiment of the present invention, as shown in FIGS. 10A and 10D, even if the measured length patterns 111A and 111D are shifted in the X direction, the line width can be stably reduced. For measurement, the line widths W of the normal portions 113A and 113D are measured using the distance of the waveform peak Pa1-Pa1 or the distance of Pa2-Pa2 shown in FIG. 11 for the measured pattern 111A, and for the measured pattern 111D. The distance between the waveform peaks Pd1-Pd1 or Pd2-P shown in FIG.
It measures using the distance of d2. In addition, the inflection point of the waveform is used when obtaining the line width Ws of the connection portion. However, since the waveform depends on the image state of the SEM, any condition may be used as long as the edge of the pattern to be measured 111 can be detected.

The amount of change in the line width of the connecting portion as described in the embodiment of the second invention can be obtained from the difference in the line width of the connecting portion. That is, as shown in FIG. 10C, the measured length pattern 111C is thinned, and the normal portion 113C has a bright pixel inside the peak Pc (the edge of the normal portion 113C) as shown in FIG. Since there is a portion indicating the portion (the edge of the connection portion 112C), the line width Ws of the connection portion 112C is obtained from the waveform Lc of the portion.

As shown in FIGS. 11 and 14, the shift amount in the X direction is obtained from the two peaks obtained from the measured patterns 111A and 111D (see FIGS. 10A and 10D). Measured length pattern 111A (see FIG. 10A)
When the shift occurs in the X direction as shown in FIG. 10, two edges of the normal portion 113A (see FIG. 10A) appear close to each other, that is, two peaks Pa1 and Pa2 are formed. The distance between the peaks Pa1 and Pa2 of the two peaks is the shift amount d in the X direction. On the other hand, if the measured length pattern 111D (see FIG. 10D) also shifts in the X direction, similarly, the normal portion 113
Two edges of [see FIG. 10D] appear close to each other, that is, two peaks Pd1 and Pd2 are formed. The distance between the peaks Pa1 and Pa2 of the two peaks is the shift amount d in the X direction. The line width variation described in the embodiment according to the second invention is a value obtained by subtracting the line width of the normal part and the shift amount in the X direction from the line width of the connection part.

As can be seen from the length-measured patterns 111A and 111B shown in FIGS. 10A and 10B, when a shift occurs in the X direction, the overlapping area of the patterns constituting the length-measured pattern changes. (The larger the displacement in the X direction is, the smaller the overlapping area is). Therefore, depending on the displacement in the X direction, Y
The amount of line width change (thickening) due to the direction shift changes. Therefore, the displacement in the X direction is, for example, 10 nm, 20 nm, 3
In the case of 0 nm, 40 nm, 50 nm, and 60 nm, for each shift in the X direction, the line width change amount (vertical axis) of the connection portion as shown in FIG. 6), and the relationship between the amount of connection separation (vertical axis) and the amount of connection deviation set at the design stage (horizontal axis) as shown in FIG. It is desirable from the point of view.

When an actual circuit pattern is used as the measured pattern 111, to improve the measurement accuracy,
It is preferable to select and use a pattern that is easy to measure.

By combining the above-described methods, the present invention can be applied not only to patterns such as line and space, but also to measurements using circuit patterns of actual circuit devices. However,
In consideration of the easiness of the measurement, it is desirable to search for a pattern such as an isolated line suitable for the connection accuracy measurement by the method of the present invention from the circuit pattern data. Also,
If a dedicated connection accuracy measurement pattern can be inserted in the circuit pattern, the method of the present invention may be applied to the dedicated pattern.

[0039]

As described above, according to the first aspect of the present invention, the connection accuracy between the patterns transferred as a plurality of patterns at different times is determined by comparing the transferred patterns with the connected patterns at the design stage. Is measured by comparing an image with a transfer pattern obtained by transferring the pattern shifted by exposure and development or a transfer pattern predicted by simulation, so that there is no need to provide a dedicated pattern for measuring connection accuracy, and the design rules It is possible to directly measure the connection accuracy with the circuit pattern formed on the substrate without considering the difference. Therefore, measurement accuracy can be improved. Further, since it is not necessary to provide a dedicated measurement pattern, the throughput and the yield can be improved as compared with the case where a dedicated measurement pattern is formed.

According to the second aspect of the present invention, the connection accuracy between the patterns transferred as a plurality of patterns at different times is determined by transferring the transferred plurality of patterns and the pattern in which the connection portion is shifted in advance in the design stage. Since measurement is performed by comparing the size of the obtained transfer pattern or the transfer pattern predicted by simulation, it is formed on the substrate without providing a dedicated pattern for measuring the connection accuracy and without considering the difference in design rules. The connection accuracy can be measured directly with the circuit pattern used. Therefore, measurement accuracy can be improved. Further, since it is not necessary to provide a dedicated measurement pattern, the throughput and the yield can be improved as compared with the case where a dedicated measurement pattern is formed.

According to the third aspect, the distribution of the sum of the image outputs in a fixed direction of the image obtained by scanning the pattern to be measured with an electron beam changes at the edge of the pattern to be measured. Therefore, it is possible to measure the dimension of the pattern to be measured based on the change. According to such a measuring method, it is possible to easily measure the displacement amount of the connection portion of the measured pattern, the line width of the measured pattern, and the like. In addition, since measurement using a circuit pattern is enabled, measurement accuracy can be improved.

[Brief description of the drawings]

FIG. 1 is a flowchart illustrating an example of an embodiment according to a first invention.

FIG. 2 is an explanatory view of a pattern in which a connecting portion is shifted in a design stage and an explanatory view of a pattern after exposure and development.

FIG. 3 is an explanatory diagram of a transfer pattern obtained by a simulation.

FIG. 4 is a flowchart illustrating an example of an embodiment according to the second invention.

FIG. 5 is a diagram illustrating a relationship between a line width change amount of a connection portion and a set connection deviation amount.

FIG. 6 is a diagram illustrating a relationship between a separation amount of a connection unit and a set connection deviation amount. It is.

7A and 7B are an explanatory diagram of a pattern in which a connecting portion is shifted in a design stage and an explanatory diagram of a pattern after exposure and development.

FIG. 8 is an explanatory diagram of a method of obtaining an X-direction shift amount.

FIG. 9 is a schematic diagram of an SEM image of a measured length pattern and a graph of image processing data.

FIG. 10 is an explanatory diagram of connection examples of various patterns.

FIG. 11 is a luminance distribution diagram obtained by performing image processing on the measured length pattern shown in FIG.

12 is a luminance distribution diagram obtained by performing image processing on the measured length pattern shown in FIG.

FIG. 13 is a luminance distribution diagram obtained by performing image processing on the measured length pattern shown in FIG.

FIG. 14 is a luminance distribution diagram obtained by performing image processing on the measured length pattern shown in FIG.

[Explanation of symbols]

 11 Design pattern, 21, 31 Transfer pattern

Claims (20)

[Claims] 1. A positioning method in a charged beam exposure for transferring a continuous pattern as a plurality of patterns on a substrate at different times in a design stage, wherein the connection accuracy between the patterns transferred as the plurality of patterns is A positioning method in charged beam exposure, wherein the measurement is performed by comparing images of a plurality of transferred patterns and a transfer pattern formed based on a pattern in which connection portions are shifted in advance in a design stage. 2. The positioning method according to claim 1, wherein the transfer pattern is obtained by exposure based on a pattern in which a connection portion is shifted in advance in a design stage, and development performed thereafter. Positioning method in charged beam exposure. 3. The charged beam exposure positioning method according to claim 1, wherein the transfer pattern is obtained by a simulation prediction based on a pattern in which connection portions are shifted in advance in a design stage. Positioning method. 4. The positioning method in charged beam exposure according to claim 3, wherein the simulation includes an exposure simulation and a development simulation. 5. The positioning method in charged beam exposure according to claim 1, wherein, in comparing the images, connection accuracy is calculated in accordance with a matching ratio. 6. The positioning method in charged beam exposure according to claim 2, wherein, when comparing the images, the connection accuracy is calculated in accordance with a matching ratio. 7. The positioning method in charged beam exposure according to claim 3, wherein, when comparing the images, the connection accuracy is calculated in accordance with a matching ratio. 8. The positioning method in charged beam exposure according to claim 4, wherein, when comparing the images, the connection accuracy is calculated in accordance with a matching ratio. 9. A positioning method in charged beam exposure for transferring a continuous pattern as a plurality of patterns on a substrate at different times in a design stage, wherein the connection accuracy between the patterns transferred as the plurality of patterns is A positioning method in charged beam exposure, wherein the measurement is performed by comparing dimensions of a plurality of transferred patterns and a transferred pattern formed based on a pattern in which connection portions are previously shifted in a design stage. 10. The positioning method in charged beam exposure according to claim 9, wherein the transfer pattern is obtained by exposure based on a pattern in which connection portions are shifted in advance in a design stage, and development performed thereafter. Positioning method in charged beam exposure. 11. The charged beam exposure method according to claim 9, wherein the transfer pattern is obtained by predicting a simulation based on a pattern in which connection portions are shifted in advance in a design stage. Positioning method. 12. The positioning method in charged beam exposure according to claim 11, wherein the simulation includes an exposure simulation and a development simulation. 13. The charged beam exposure method according to claim 9, wherein the dimensional comparison calculates a connection accuracy based on a relationship between a connection line width and a connection deviation amount. Positioning method for exposure. 14. The charged beam exposure method according to claim 10, wherein in the dimension comparison, a connection accuracy is calculated based on a relationship between a connection line width and a connection deviation amount. Positioning method for exposure. 15. The charged beam exposure method according to claim 11, wherein the dimensional comparison calculates connection accuracy based on a relationship between a connection line width and a connection deviation amount. Positioning method for exposure. 16. The charged beam exposure method according to claim 12, wherein the dimensional comparison calculates connection accuracy based on a relationship between a connection line width and a connection deviation amount. Positioning method for exposure. 17. A line width measuring method for measuring a width of a pattern to be measured using an electron beam, wherein a sum of image outputs in a certain direction of an image obtained by scanning the pattern to be measured with an electron beam. A dimension of the length pattern to be measured based on the distribution of the line width. 18. The line width measuring method according to claim 13, wherein the predetermined direction of the image is a length direction of the length pattern to be measured. 19. The line width measuring method according to claim 13, wherein a continuous pattern is transferred to the substrate as a plurality of patterns at different times in the design stage based on the distribution of the sum of the image outputs. A line width measuring method characterized by measuring a line width of a connection portion of a measured pattern. 20. The line width measuring method according to claim 13, wherein a continuous pattern is transferred to the substrate as a plurality of patterns at different times in a design stage based on a distribution of a sum of the image outputs. A line width measuring method characterized by measuring a connection shift amount in a direction parallel to a connection surface of a pattern to be measured.