JPH07273160A - Pattern evaluation method and pattern transfer method - Google Patents

Abstract

PURPOSE:To provide a method for evaluating GO/NO-GO of a pattern accurately through a relatively simple processing by providing a coordinate value measuring step, a relative position determining step, and a shift evaluation step, specifically. CONSTITUTION:GO/NO-GO is determined for a plurality of patterns formed on one translucent substrate based on a design coordinate data representative of each position of the pattern. In such pattern evaluation method, a step for measuring the coordinate values at more than one position of each pattern formed on the translucent substrate in a common coordinate system to determine a coordinate measurement data is provided. A step for aligning the coordinate measurement data with a design coordinate data by superposing both coordinates on each other so that the total shift is minimized statistically is also provided. Furthermore, a step for deciding whether the shift between both data is within an allowable range or not is provided.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION The present invention relates to LSI manufacturing and LCD
A photomask used as a mask for exposing a fine pattern in the production of a flat panel for (liquid crystal display), etc., and a pattern evaluation method for evaluating the quality of a fine pattern formed on a flat panel for an LCD, in particular, one substrate The present invention relates to a so-called multi-chambered board evaluation method having a plurality of patterns formed thereon, and a pattern transfer method utilizing the pattern evaluation method.

[0002]

2. Description of the Related Art Photomasks used as fine pattern exposure masks in the production of LSIs, flat panels for LCDs, etc., can be obtained by exposing a transferred plate once.
For example, there is a so-called multi-faced photomask for transferring integrated circuit patterns of a plurality of semiconductor devices, or transferring a black matrix or RGB patterns of a plurality of color filters. FIG. 5 is a plan view of a two-chamfered photomask 1 in which two patterns 11 and 12 are formed on a transparent substrate 10 among the photomasks of this type. In this photomask, a resist film is usually formed on a photomask blank in which a light-shielding film is formed on the surface of a transparent substrate, and an exposure means such as electron beam exposure is used to draw a fine pattern on the resist film with an electron beam. After exposure, it is formed by developing and etching. In this electron beam drawing, a photomask blank is held on a moving stage, a pattern to be drawn is stored in a computer as design coordinate data, and the drawing position of the electron beam and the position of the moving stage are controlled based on this design coordinate data. Do it.

Here, exposure is performed by this electron beam drawing,
It is necessary to ensure that the fine pattern actually formed on the photomask by development and etching and the fine pattern indicated by the design coordinate data match exactly. If pattern transfer is performed using photomasks that do not match each other with an accuracy of a predetermined level or more, a large number of defective products will be produced in the transferred pattern. Therefore, in the manufactured photomask, it is necessary to examine and evaluate whether or not the actually formed fine pattern and the fine pattern indicated by the design coordinate data match with each other with a precision higher than a predetermined accuracy.

As the evaluation method, there are the following methods. That is, as shown in FIG. 6, the widths X1 and Y in the X and Y directions of the patterns 11 and 12, respectively.
Width Y including 1, X2, Y2 and two patterns
3 is measured, the pattern positional accuracy in each plane is evaluated from the values of X1, Y1, X2, and Y2, and the relative positional accuracy of the pattern is evaluated by Y3.

[0005]

However, the above-mentioned method is, so to speak, one-dimensionally looking at the deviation of the distance individually, so that it is not always the best from the viewpoint of observing the overall pattern arrangement deviation. It is hard to say that it is a method. In addition, since the distance may be slightly different depending on the place on the substrate to be measured, there is a problem that the evaluation result is different depending on where the measurement is made, resulting in lack of accuracy and reliability.

The present invention has been made under the above-mentioned background, and a pattern evaluation method capable of accurately evaluating the quality of a pattern by a relatively simple process, and an accurate pattern transfer using this method. It is an object of the present invention to provide a possible pattern transfer method.

[0007]

In order to solve the above-mentioned problems, a pattern evaluation method according to the present invention comprises: (Structure 1) Based on design coordinate data indicating each position of a pattern, on the same transparent substrate. A pattern evaluation method for evaluating the quality of a plurality of patterns formed on a substrate, comprising: measuring coordinate values in a coordinate system common to two or more positions in each pattern formed on the transparent substrate. The coordinate value measurement process for obtaining the measured coordinate data and the coordinate systems of both data are overlapped so that the total deviation between the measured coordinate data obtained in this coordinate value measurement process and the design coordinate data is statistically minimized. Mutual position determining step of both data to be aligned, and deviation amount evaluating step of evaluating whether or not the deviation amount of both data at the position determined by the mutual position determining step is within an allowable range. Or (Structure 2) a transfer pattern formed on a photomask used in an exposure apparatus equipped with a transfer pattern correction function, wherein design coordinate data indicating each position of the pattern is provided. In a pattern evaluation method of a photomask for evaluating the quality of a plurality of transfer patterns formed on the same transparent substrate based on the above, a pattern common to two or more positions in each pattern group formed on the photomask The coordinate value measuring step of measuring coordinate values in the coordinate system to obtain the measured coordinate data, and the measured coordinate data determined in the coordinate value measuring step and the design coordinate data are compared, and the measured coordinate data is transferred to the exposure apparatus. The same correction as the transfer pattern correction function installed in the The mutual position determination step of the both data in which the coordinate systems of the both data are superposed and aligned so that the displacement is statistically minimized, and the displacement amount of both data at the position determined by the mutual position determination step is allowed. And a deviation amount evaluating step for evaluating whether or not it is within the range. As an aspect of these configurations 1 or 2, (configuration 3), in the pattern evaluation method of configuration 1 or 2, In the data mutual position determination process, one or both of the two data are coordinated in one coordinate system so that the sum of the squares of the deviation amounts of the points in the actual measurement coordinate data and the design coordinate data, which have a corresponding relationship with each other, becomes the minimum. In the above, the configuration is characterized by including an operation of performing either or both of rotational movement and parallel movement around an arbitrary point.

The pattern transfer method according to the present invention is (Structure 4) a pattern transfer method for transferring a pattern formed on a photomask by using an exposure device having a transfer pattern correction function, As the exposure apparatus, a photomask evaluated by the mutual position determining step in the photomask pattern evaluation method according to claim 2 as being within an allowable range of the deviation amount of both data is used as the exposure apparatus. In the photomask pattern evaluation method described in any one of the above, the configuration is characterized by using the one having a function of performing the same correction as the correction performed in the mutual position determining step.

[0009]

According to the above-mentioned configurations 1 to 3, since the displacement of the overall pattern arrangement is statistically observed, as in the conventional method for determining the displacement of the distance between the patterns,
The result does not vary depending on the measurement location, and extremely accurate pattern evaluation is possible.

Further, according to the structure 4, it is possible to perform the pattern transfer in which the deviation amount is always within the allowable range, and it is possible to prevent the occurrence of the transfer failure.

[0011]

1 is a diagram showing the outline of the procedure of a pattern evaluation method and a pattern transfer method according to an embodiment of the present invention, FIG.
FIG. 3 is a diagram showing a configuration of a pattern evaluation apparatus used when implementing the method of the embodiment of the present invention, and FIG. 3 shows measured coordinate data and design coordinate data whose mutual positions are determined by the first embodiment of the pattern evaluation method. FIG. 4 is a diagram schematically showing a state in which they are overlapped, and FIG. 4 is a diagram schematically showing a state in which the mutual position is determined by the second embodiment of the pattern evaluation method and the actual measurement coordinate data and the design coordinate data are overlapped. Hereinafter, the pattern evaluation method and the pattern transfer method according to the embodiment will be described in detail with reference to these drawings. In the following description, the photomask to be evaluated will be described first, then the configuration of the pattern evaluation apparatus will be described, then the first embodiment of the pattern evaluation method will be described, and then the pattern evaluation will be described. A second example of the method will be described, and then an example of the pattern transfer method will be described.

Photomask to be evaluated The photomask to be evaluated is obtained as follows.

First, a chromium light-shielding film having a thickness of 980 angstroms is formed on the surface (main surface) of a low-expansion glass substrate of 400 mm × 400 mm × 5 mm by a sputtering method, and a resist (AZ-1350: a product of Hoechst Co., Ltd.) is formed thereon. Name) for 10,000 angstroms.

Next, the resist thus formed is subjected to laser drawing according to the design coordinate data. In this embodiment, a photomask having two patterns (first pattern 11 and second pattern 12) arranged in the Y direction as shown in FIG. 5 is formed, and 3 pieces in the X direction are used as the design coordinate data. A row having a total of 18 reference points, 6 rows in total, 3 rows for each pattern in the Y direction, was used. In this case, the distance between the reference points in the X direction is 96520.
The reference point within the chip is 72
It was 390.00 μm and the distance between chips was 18000.00 μm.

Next, the resist is developed and baked to form a resist pattern, and the chrome film is etched with a predetermined etching solution using the resist pattern as a mask to form a chrome light-shielding pattern. After that, the remaining resist film is peeled off, and a washing / drying process is performed to obtain a double chamfered photomask as shown in FIG.

Pattern Evaluation Device In FIG. 1, reference numeral 1 is a photomask which is an object of pattern evaluation, reference numeral 2 is a coordinate value measuring device, and reference numeral 100 is a computer.

The photomask 1 has 18 reference points 1 described above.
It is a photomask in which a chrome light-shielding pattern having a is formed.

The coordinate value measuring device 2 mounts the photomask 1 and moves it in the XY directions, and at the same time, accurately obtains the amount of movement and measures the XY coordinates, and an XY stage 2a.
It is composed of a position detection optical system 2b for optically detecting the reference point 1a on the surface of the photomask 1 and a coordinate value measurement / control function of the computer 100. That is, in this coordinate value measuring device 2, the XY stage 2a and the position detecting optical system 2b are controlled by the coordinate value measuring / control function 2c of the computer 100, and the reference point 1a of the light shielding pattern formed on the photomask 1 is shown. Is detected, the coordinate value is measured, and the measured value is taken into the computer 100 as actually measured coordinate data.

The computer 100 fetches the design coordinate data in addition to the fetched actual measurement coordinate data, and also carries out a certain statistical processing, coordinate conversion processing and arithmetic processing so that the deviation between the two data is statistically smallest. The function 3 for determining the mutual positions of the coordinate systems of both data is provided so that At the same time, in the mutual positional relationship determined in this way, a function 4 is provided for obtaining the deviation amount of each point (reference point) of both data and evaluating whether or not the deviation amount is within a predetermined allowable range. It is what

First Embodiment of Pattern Evaluation Method The outline of the procedure of the pattern evaluation method of this embodiment is as shown in FIG. 1, and this procedure is executed by using the above-described pattern evaluation apparatus. Hereinafter, steps (a) to (e)
Will be described in detail.

(A) Coordinate value measuring step The coordinate value measuring device 2 is used to measure the coordinate value of each reference point of the photomask 1, and this is measured coordinate data (Pxij, P).
yij) (where i = 1 to 3 and j = 1 to 6) is taken into the computer 100.

(B) Step of importing design coordinate data In the computer 100, reference points (Xi
j, Yij) (where i = 1 to 4, j = 1 to 7) are taken in. Note that this step may be performed before the coordinate value measurement step.

(C) Step of superposing the coordinate systems of both data The origins of the coordinate systems of both data taken in the steps a and b are made coincident and both data are superposed.

(D) Mutual Position Determination Process of Both Coordinate Systems The mutual position of both data is determined so that the overall deviation between the measured coordinate data obtained in the coordinate value measurement process and the design coordinate data is statistically minimized. . Specifically, the procedure is as follows.

Now, actually measured coordinate data (Pxij, Pyij)
The coordinates obtained by adding the rotational movement (θ) about the point (cx, cy) and the parallel movement (ax, ay) to (P′xij, P ′
yij), (P'xij, P'yij) becomes (1)
It is represented by a formula.

[0026]

[Equation 1] Then, if the total sum D of the squares of the differences between the X and Y coordinates of each point pair corresponding to this (P'xij, P'yij) and the design coordinate data (Xij, Yij) is the smallest, It can be said that the overall deviation of the coordinates is statistically minimized.
Here, D is represented by the following equation (2).

[0027]

[Equation 2] Therefore, if the rotation amount θ and the parallel movement amounts ax and ay of both coordinate systems required to minimize D in the equation (2) are obtained,
The mutual position of both coordinate systems can be determined. Specifically, a simultaneous equation that holds when the value obtained by partially differentiating D in the equation (1) with each parameter (ax, ay, θ) becomes 0 may be solved. This simultaneous equation is expressed by the following equations (3) to (5).

[0028]

[Equation 3] By solving the simultaneous equations (3) to (5), θ, ax, ay can be obtained. The obtained θ, ax, and ay are expressed by the following equations (6) to (8).

[0029]

[Equation 4] However, in the above equations (6) to (8), n is the total number of reference points.

By substituting θ, ax, and ay thus obtained in the equation (1), coordinate values (P) are coordinate-converted so that the overall deviation between the measured coordinate data and the design coordinate data is statistically minimized. ′ Xij, P′yij) is obtained. FIG. 3 is a diagram schematically showing the measured coordinate data whose mutual positions have been determined for the photomask 1 and the design coordinate data. Note that, in FIG. 3, each point indicated by ● is the coordinate value (Xij, Yij) of the reference point of the design coordinate data, and each point indicated by × is the reference point after the coordinate conversion of the actually measured coordinate data. Coordinate values (P'xij, P'yij) of the. Further, in FIG. 3, the scale showing the amount of deviation is enlarged to 1.5 × 10 ⁴ times the scale showing the interval between adjacent reference points.

(E) Deviation amount evaluation process Coordinate values (P'xij, P'yij) of reference points after coordinate conversion of design coordinate data (Xij, Yij) and measured coordinate data.
The amount of deviation of each corresponding point is determined, whether or not the amount of deviation is within the allowable range is evaluated, and whether the photomask 1 is a good product or a defective product is determined.

At this time, the in-plane positional accuracy of each pattern is evaluated by the first pattern 11 and the second pattern 1 respectively.
2 design coordinate data (Xij, Yij) (X = 1 to 3, Y =
1 to 3), and (Xij, Yij) (X = 1 to 3, Y = 4 to
6) and the coordinate values (P′xi) of the reference points of the first pattern 11 and the second pattern 12 after the coordinate conversion of the actually measured coordinate data.
j, P'yij) (x = 1 to 3, y = 1 to 3), and (P
′ Xij, P′yij) (x = 1 to 3, y = 4 to 6) is calculated for the corresponding deviation amount, and the deviation amount is, for example, three times the standard deviation (3σ). Then, it is evaluated whether or not each of these variations is within the allowable range. As for the relative positional accuracy between patterns, all design coordinate data (Xij, Yij) (X
= 1 to 3, Y = 1 to 6) and the coordinate value (P'xij, P'yij) of the reference point after the coordinate conversion of the actually measured coordinate data (x = 1
~ 3, y = 1-6)
For each of them, for example, the variation of the deviation amount is obtained in the form of three times the standard deviation (3σ), and whether or not each variation is within the allowable range is evaluated. Then, these evaluations are combined to determine whether the photomask 1 is a good product or a defective product.

Second Embodiment of Pattern Evaluation Method This embodiment is an example in which a transfer pattern formed on a photomask used in an exposure apparatus equipped with a transfer pattern correction function is evaluated. The pattern evaluation of the photomask used in the exposure apparatus equipped with such a correction function is
Based on this correction, it is necessary to evaluate whether or not it is within the allowable range when it is assumed that this correction has been performed. Therefore, in this embodiment, it is necessary to add the processing in the case where the above correction is performed in advance in the "(d) Mutual position determining step of both coordinate systems" in the above-mentioned first embodiment. The steps of are the same as those in the first embodiment. Therefore, in the following, only "(d) mutual position determining step of both coordinate systems" unique to this embodiment will be described, and description of other steps will be omitted.

(D) Mutual Position Determination Process of Both Coordinate Systems In the second embodiment, the overall deviation between the measured coordinate data obtained in the coordinate value measuring process and the design coordinate data is statistically minimized. The measured coordinate data is subjected to orthogonality correction and magnification correction equivalent to the correction by the correction function provided in the exposure apparatus that transfers the pattern using this photomask to determine the mutual position of both data.

Therefore, first, for each of the design coordinate data and the actually measured coordinate data, the average value of the inclinations of the line segments connecting the adjacent reference points in the X direction and the Y direction is determined, and it is necessary for these to be equal. Is the orthogonality correction coefficient a
Ask as. That is, first, the rotational movement is performed so that the average value of the inclination of the line segment in the X direction becomes equal in both data.
Next, the X component is converted so that the average value of the inclination of the line segment in the Y direction becomes equal in both data. The X component to be converted at this time depends on the Y component, the degree of dependence on Y is the orthogonality correction value a, and a is obtained from the condition that the average value of the inclination of the line segment in the Y direction is the same for both data. You can Specifically, it is obtained as follows.

Let Ipx be the inclination of the measured data (Pxij, Pyij) with respect to the X-axis, and Irx be the inclination of the design coordinate data (Xij, Yij) with respect to the X-axis.
It can be regarded as the average value of the inclination of the line segment connecting the reference points adjacent to each other in the direction, and can be expressed by the following equation.

[0037]

[Equation 5] A rotation angle θ is calculated so that Ipx and Irx in the equations (9) and (10) are equal.

Now, actual measurement coordinate data (Pxij, Pyij)
When the rotational movement of θ obtained above is performed around the point (cx, cy), the measured coordinate data (Pxij, Pyij) becomes
Move as shown in equation (11).

[0039]

[Equation 6] In the equation (11), the center of rotation may be used as the center of gravity of the actually measured coordinate data (Pxij, Pyij) for convenience.

Measured coordinate data (P'xij, P'after rotation)
yij) is further transformed by the orthogonality correction operation expressed by the following equation (12).

[0041]

[Equation 7] The rotated measured coordinate data (P'xij, P'yij)
Let Ip be the inclination with respect to the Y axis after the conversion of the orthogonality correction operation of the equation (12), and Ir be the inclination with respect to the Y axis of the design coordinate data (Xij, Yij). Then, Ip and Ir are given by the following (13). ，
It is expressed by equation (14).

[0042]

[Equation 8] Assuming that Ip and Ir in the above equations (13) and (14) are equal, a determined at that time is represented by the following equation (15).

[0043]

[Equation 9] Using the rotation angle θ and the orthogonality correction a thus obtained, the actually measured coordinate data (Pxij, Pyij) are rotated and the orthogonality correction is performed.

Next, with respect to the actually measured coordinate data on which the above-mentioned orthogonality correction has been performed, a magnification correction including a magnification change and a parallel movement is further performed so that the deviation from the design coordinate data is minimized. Specifically, it is performed as follows.

That is, certain measured coordinate data (Pxij,
Pyij) for some design coordinate data (Xij, Yij),

[Equation 10] When the magnification is changed and the parallel movement (ax, ay) is performed, it is assumed that the distance is closest. This operation is expressed by the following equation (16).

[0046]

[Equation 11] The condition that the above two are closest to each other is D expressed by the following equation (17).
Is to take the minimum value.

[0047]

[Equation 12] Therefore, the following simultaneous equations (18) to (21) are solved so that the value obtained by partially differentiating D in the above equation (17) with each parameter becomes 0. Magnification changes and parallel movements that satisfy these equations are obtained.

[0048]

[Equation 13] In the above equation, the X component and the Y component can be solved independently, and since they are objects, only the X component will be solved. When the above equations (18) and (19) are rewritten and transformed into the following equations (22) and (23), kx can be obtained from the following equation (24).

[0049]

[Equation 14] Thereby, ax can be calculated by the following equation (25).

[0050]

[Equation 15] Similarly, the Y component can be obtained.

Using the magnification change and the parallel movement thus obtained, the magnification of the actually measured coordinate data (Pxij, Pyij) is changed, and then the parallel movement is performed, whereby the magnification correction can be executed.

FIG. 4 is a diagram schematically showing the measured coordinate data and the design coordinate data in which the mutual positions are determined at the stage when the magnification correction is performed after the orthogonality correction is performed.
The display in FIG. 4 is the same as that in FIG.

In the above embodiment, in the calculation of the orthogonality correction, the average value of the inclinations of the line segments connecting the adjacent reference points in the X direction and the Y direction is used, but the present invention is not limited to this. Draw a line segment from the reference point of by the method of least squares,
The inclination of the line segment may be used.

Example of Pattern Transfer Method In this example, a photomask evaluated as a non-defective product by the photomask pattern evaluation method according to the above-described example is selected (non-defective product selection step) and used, and a transfer pattern correction function is provided. A pattern was transferred to an object to be transferred (photomask blank with resist) by an exposure device (MPA-2000 manufactured by Canon Inc.) (transfer step).

In this exposure apparatus, the light from the light source passes through the slit and reaches the transfer target through the photomask to transfer the pattern. At that time, the orthogonality correction described above is performed by shifting the scanning direction of the photomask and the transfer target with respect to the scanning direction of the slit by a predetermined angle, and the magnification correction is performed by adjusting the magnification of the optical system. Done.

The transfer pattern formed by using such an exposure apparatus had an accuracy within a predetermined allowable range and was extremely close to the design coordinate data.

In each of the above-described embodiments, the rotation and the parallel movement in the mutual position determining process of both coordinate systems are performed around an arbitrary point (cx, cy), but this is, for example, Alternatively, the center of gravity of one of the coordinate systems may be used as the center.

Further, in the above-mentioned embodiment, as a concrete method for minimizing the overall deviation between the measured coordinate data and the design coordinate data statistically, the coordinate value (P ′ Xij, P′yij) and the design coordinate data (Xij, Yij), the condition that the sum D of the sums of the squares of the differences between the X and Y coordinates of the corresponding point pairs is calculated is as follows.
This may be obtained, for example, under the condition that the standard deviation of the difference between corresponding points of both data is minimized, or another statistical method may be used.

Further, in each of the embodiments, the coordinates of the reference point on the light-shielding pattern of the photomask are used as the design coordinate data and the actually-measured coordinate data, but this may also be achieved by using a mark provided outside the light-shielding pattern area. Good. Further, the accuracy increases as the number of points used increases, but in principle, it can be implemented if there are two or more points.

Further, in each of the above-described embodiments, an example in which the photomask pattern patterned by laser drawing is evaluated is shown, but the present invention is not limited to this.
Of course, it is also possible to evaluate a patterned photomask pattern or a pattern of an LCD flat panel by exposing through a reticle. In this case, the coordinates of the reference point of the reticle may be measured in advance and the measurement data of the reticle may be used as the design coordinate data. Furthermore, it is not limited to two chamfers,
It is also applicable to multi-chamfering such as 3-chamfering and 4-chamfering.

[0061]

As described in detail above, according to the pattern evaluation method of the present invention, since the deviation of the overall pattern arrangement is statistically observed, the deviation of the distance between the patterns is Unlike the conventional method, the result does not vary depending on the measurement location, and extremely accurate pattern evaluation is possible.

Further, according to the pattern transfer method of the present invention, it is possible to always perform the pattern transfer in which the displacement amount is within the allowable range, and it is possible to prevent the transfer failure from occurring.

[Brief description of drawings]

FIG. 1 is a diagram showing a summary of a procedure of a pattern evaluation method according to an embodiment of the present invention.

FIG. 2 is a diagram showing a configuration of a pattern evaluation device.

FIG. 3 is a diagram schematically showing a state in which actual measurement coordinate data whose mutual positions are determined by the first embodiment of the pattern transfer method and design coordinate data are superimposed.

FIG. 4 is a diagram schematically showing a state in which actual measurement coordinate data whose mutual positions are determined by the second embodiment of the pattern transfer method and design coordinate data are superimposed.

FIG. 5 is an explanatory diagram of a double-faced photomask.

FIG. 6 is an explanatory diagram of a conventional example of a pattern evaluation method for a double-chamfered photomask.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Photomask, 2 ... Coordinate value measurement device, 3 ... Mutual position determination function of design coordinate data and measurement coordinate data, 4 ... Deviation amount evaluation function, 100 ... Computer.

Claims (4)

[Claims] 1. A pattern evaluation method for evaluating the quality of a plurality of patterns formed on the same transparent substrate on the basis of design coordinate data indicating each position of the pattern, wherein the pattern is formed on the transparent substrate. A coordinate value measuring step of measuring coordinate values of two or more positions in each pattern formed in the coordinate system common to each pattern to obtain measured coordinate data; and the measured coordinate data obtained in this coordinate value measuring step and the design Mutual position determining step of overlapping the coordinate systems of both data so that the total deviation from the coordinate data is statistically minimized, and both data at the position determined by the mutual position determining step And a deviation amount evaluating step of evaluating whether or not the deviation amount is within an allowable range. 2. A transfer pattern formed on a photomask used in an exposure apparatus equipped with a transfer pattern correction function, the transfer pattern being formed on the same transparent substrate based on design coordinate data indicating each position of the pattern. A pattern evaluation method for a photomask for evaluating the quality of a plurality of transfer patterns, wherein 2 in each pattern group formed on the photomask
The coordinate value measuring step of measuring coordinate values in the coordinate system common to each pattern at the above positions to obtain the measured coordinate data, and the measured coordinate data determined in the coordinate value measuring step are compared with the design coordinate data The measured coordinate data is subjected to a correction equivalent to the correction by the transfer pattern correction function provided in the exposure apparatus, and the overall deviation between the corrected measured coordinate data and the design coordinate data is statistically minimized. First, it is evaluated whether or not the mutual position determining step of the both data in which the coordinate systems of the both data are superposed and aligned, and the deviation amount of the both data at the position determined by the mutual position determining step is within the allowable range. A pattern evaluation method comprising a shift amount evaluation step. 3. The pattern evaluation method according to claim 1 or 2, wherein the data mutual position determination step is a sum of squares of deviation amounts of respective points in the actual measurement coordinate data and the design coordinate data which have a corresponding relationship to each other. So that either one or both of the two data may be rotated or translated about an arbitrary point in one coordinate system, or both of them may be minimized. Pattern evaluation method. 4. A pattern transfer method for transferring a pattern formed on a photomask using an exposure device having a transfer pattern correction function, wherein the photomask pattern evaluation according to claim 2 is used as the photomask. In the photomask pattern evaluation method according to any one of claims 1 to 3, wherein a photomask evaluated by the mutual position determining step in the method as having a deviation amount of both data within an allowable range is used. A pattern transfer method characterized by using a device equipped with a function of performing the same correction as that performed in the mutual position determining step.