JPH10149980A - Method for deciding exposing area of scanning aligner and projection optical system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method for deciding the shape of the exposed area of a scanning aligner by which the distortion of an exposed picture can be reduced. SOLUTION: In a method by which the shape of the exposed area 4 of a scanning aligner which transfers a pattern 1 on a reticle 1 to a wafer 2 in the longitudinal direction (x) of the area 4 perpendicular to the scanning direction (y) of the aligner is decided by forming the image of the pattern 1a in the area 4 on the wafer 2 through a projection optical system 3 and synchronously scanning the wafer 2 and reticle 1 at the speed ratio corresponding to the magnification of the optical system 3, and then, evenly forming the width of the area 4 in the scanning direction (y), the shape of the area 4 in the longitudinal direction (x) is decided so that the phase shift Φbetween the image in the area 4 and the pattern 1a caused by the optical system 3 can become the minimum based on the distortion aberration D (x, y) of the optical system and the arranging direction ϕ of the pattern 1a on the reticle 1.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a shape of an exposure area of a scanning type exposure apparatus, and more particularly to a method for determining the shape of an exposure area in a longitudinal direction orthogonal to a scanning direction.

[0002]

2. Description of the Related Art In recent years, the integration ratio per unit area of integrated circuits has become increasingly finer, and the overall exposure area has been expanding more and more. As a result, the design and manufacture of projection optical systems have become more and more difficult. ing. In the collective exposure method in which the pattern on the reticle is collectively transferred to a wafer, the number of pixels that can be transferred is clearly approaching the physical limit. Therefore, if the number of pixels per chip is further increased, it is necessary to further shorten the exposure wavelength and increase the chip area. On the other hand, the scanning exposure method moves a reticle placed on an object plane and a wafer placed on an image plane synchronously at a speed ratio corresponding to the magnification of the projection optical system. . According to this method, in the longitudinal direction (x direction) orthogonal to the scanning direction, there is a limitation due to the lens diameter,
In the scanning direction (y direction), an infinite length can be secured in principle. Therefore, the scanning exposure method is a technique that is perfect for increasing the chip area, and has attracted attention in recent years.

[0003] In the normal reduced projection batch exposure method, the distortion of the optical image to be exposed has a spatial one-to-one correspondence with the distortion of the projection optical system. Therefore, it is necessary to minimize distortion of the projection optical system over the entire illumination field. On the other hand, in the scanning exposure method, the exposure area is limited by arranging an opening in the projection optical system. Therefore, the distortion required for the projection optical system only needs to be focused on this exposure area portion, and the design and manufacture of the projection optical system are somewhat easier.

Here, the shape of the opening arranged in the projection optical system, that is, the shape of the exposure area has been conventionally determined empirically or qualitatively. The following is an example of such a conventional method. That is, in the design and manufacturing stages of the projection optical system, the distortion of the projection optical system in the longitudinal direction (x direction) orthogonal to the scanning direction is reduced as much as possible,
In this method, a rectangular opening having a longitudinal direction in the x direction is provided (US Pat. No. 5,281,996).
issue).

[0005]

In the scanning exposure system,
Since the exposure amount must be uniform in the longitudinal direction (x direction) of the exposure area, the width of the exposure area in the scanning direction (y direction) needs to be uniform in the longitudinal direction (x direction). However, it is not always necessary that the shape of the exposure area be rectangular. In other words, the above-described prior art has a problem that the theoretical basis for making the shape of the exposure area rectangular is poor and lacks a quantitative evaluation on the distortion of the image to be exposed. Accordingly, an object of the present invention is to provide a method of determining the shape of an exposure area of an exposure area of a scanning exposure apparatus, which can reduce distortion of an image to be exposed.

[0006]

In the scanning exposure method, since the exposure is performed by scanning the reticle and the wafer, the relationship between the distortion aberration of the projection optical system and the distortion of the figure printed on the wafer is complicated. Become. Some research has been done on this problem, but recently, the literature, J. Braat and
P. Rennspies: Appl Opt 35 No. 4 690-700 (1996) clarified the relationship between the distortion of the projection optics and the features on the wafer being printed. The present invention has been made by developing this knowledge, that is, an image in an exposure area of an image of a pattern on a reticle is formed on a wafer by a projection optical system, and the wafer and the reticle are projected optically. An exposure area of a scanning exposure apparatus that transfers a pattern on a reticle onto a wafer by scanning synchronously at a speed ratio according to the magnification of the system and uniformly forming the width of the exposure area in the scanning direction. A method of determining a shape in a longitudinal direction orthogonal to the scanning direction of the projection optical system, based on the distortion aberration of the projection optical system and the arrangement direction of the pattern on the reticle, the image and pattern in the exposure area by the projection optical system A method for determining an exposure area of a scanning exposure apparatus, characterized in that the shape of the exposure area in the longitudinal direction is determined so that the phase shift between the exposure areas is minimized.

According to the present invention, an image in an exposure area of a pattern image on a reticle is formed on a wafer by a projection optical system, and the wafer and the reticle are adjusted to a speed ratio corresponding to a magnification of the projection optical system. The projection optical system of a scanning type exposure apparatus for transferring a pattern on a reticle onto a wafer by forming a uniform width of an exposure area in the scanning direction by scanning in synchronization with the scanning direction. A component in the scanning direction of the aberration;
A projection optical system of a scanning type exposure apparatus, characterized in that a ratio of a component in a longitudinal direction orthogonal to a scanning direction is formed to be substantially constant over the entire exposure area.

According to the present invention, an image in an exposure area of an image of a pattern on a reticle is formed on a wafer by a projection optical system, and the wafer and the reticle are adjusted to a speed ratio corresponding to a magnification of the projection optical system. The scanning type exposure apparatus that transfers the pattern on the reticle onto the wafer by forming the width of the exposure region in the scanning direction uniformly by scanning in synchronization with the scanning direction of the exposure region in the longitudinal direction orthogonal to the scanning direction of the exposure region. A method of determining a shape, wherein a projection optical system is formed such that a ratio of a distortion component of a projection optical system in a scanning direction to a component in a longitudinal direction orthogonal to the scanning direction is substantially constant. Then, based on the distortion of the projection optical system, the longitudinal shape of the exposure region is determined so that the phase shift between the image and the pattern in the exposure region by the projection optical system is minimized. , Exposure area of scanning exposure equipment A determination method.

[0009]

Embodiments of the present invention will be described.
FIG. 1 shows an example of a scanning exposure apparatus to which the present invention is applied,
The pattern on the reticle 1 is projected and exposed on the wafer 2 by the projection optical system 3. An opening 4 is provided in the projection optical system 3. Therefore, only a portion of the pattern on the reticle 1 corresponding to the shape of the opening 4 is projected on the wafer 4. The reticle 1 and the wafer 2 are scanned synchronously at a speed ratio corresponding to the magnification of the projection optical system 2, and thus all the patterns on the reticle 1 are projected and exposed on the wafer 2. Here, among the directions orthogonal to the optical axis z of the projection optical system 2, the scanning direction is defined as the y direction, and the direction orthogonal to the y direction is defined as the x direction.

FIG. 2 shows the shape of the opening 4, that is, the shape of the exposure area 4. Since the exposure amount in the x direction is uniform, the width 2c _{1 in} the y direction of the exposure region 4 is uniform at any position in the x direction, that is, the half width c ₁ is a constant independent of x. is there. On the other hand, the midpoint c ₂ of width 2c ₁ in the y direction of the exposure area 4 is a variable in the present invention. That is, the object of the present invention is to provide the shape c ₂
It is to provide a method for defining (x). However, since the width 2c ₁ of the exposure region 4 in the y direction is uniform, the object of the present invention is to make the shape c ₂ (x) of the exposure region 4 on the tip side in the scanning direction.
+ C ₁ , or the shape c ₂ (x) −c ₁ on the rear end side.

In FIG. 1, the opening 4 is arranged immediately after the reticle 1. However, since the opening 4 defines an exposure area, it can be arranged immediately before the wafer 2.
When the projection optical system is performing intermediate imaging, it can be arranged at the position of the intermediate imaging. Although FIG. 1 shows that the scanning directions of the reticle 1 and the wafer 2 are the same, scanning may be performed in parallel and opposite directions. Whether the scanning is performed in the same direction in parallel or in the opposite direction depends on the presence or absence of the intermediate image formation or the number of times of image formation. It also depends on whether a figure similar to the pattern projected on the wafer or a mirror-symmetric figure is used as the reticle. Further, when a reflection optical system is used as the projection optical system, the scanning directions of the reticle 1 and the wafer 2 are not even parallel.

FIG. 3 is a schematic diagram showing a pattern 1 a on the reticle 1. The interval between the patterns 1a is represented by Λ, the arrangement direction of the patterns 1a is represented by S, and the azimuth of the arrangement direction S from the x-axis is represented by φ. FIG. 4 shows the arrangement direction S of the pattern 1a.
FIG. 4 is a schematic diagram showing a light intensity distribution on a reticle surface along with a light intensity distribution on a wafer surface along a line. As shown in the figure, due to the aberration of the projection optical system 3, the modulation depth of the light intensity distribution on the wafer surface decreases by the ratio | m _D |, and at the same time, a phase shift occurs by Φ in the S direction. . An object of the present invention is to provide a method for determining the shape c ₂ (x) of the middle point of the exposure region 4 so that the phase shift Φ in the S direction is minimized.

Now, as shown in the above document, 1
When the two-dimensional diffraction grating object 1a is placed on the reticle surface and exposed by an exposure device that scans in the y direction, the modulation depth decreases | m _D
And the phase shift Φ, the exposure distribution on the wafer surface is as follows. E (x, y): Exposure amount distribution on wafer surface m _D (x): Normalized modulation function (normalized modulation f)
unction) Φ (x): phase shift in array direction S ν _x : x component of spatial frequency of image of one-dimensional diffraction grating 1a ν _y : y component of spatial frequency of image of one-dimensional diffraction grating 1a here, M: magnification of the projection lens Λ: pitch of the one-dimensional diffraction grating 1a φ: angle formed with the x-axis in the arrangement direction S of the one-dimensional diffraction grating 1a.

Further, the normalized modulation function m _D (x) is connected to the distortion D (x, y) of the projection lens in the following relationship. _{However, η = y + v w t} α: the position of the scanning direction one end side of the exposure area beta: Position D _x in the scanning direction other end side of the exposure region (x, y): the distortion aberration of the projection lens x component D _y (x , Y): y component of distortion aberration of the projection lens v _w : scanning speed of the wafer in the y direction t: time where: c ₁ : half width of exposure area c ₂ : center position of width of exposure area

In a semiconductor exposure apparatus that requires extremely faithful pattern transfer, a decrease in modulation depth and a phase shift are fatal defects. In particular, the phase shift Φ is a major cause of disturbing the similarity of the image, and must be carefully removed. When the distortion aberration D (x, y) of the projection lens is sufficiently smaller than the pitch of the one-dimensional diffraction grating image projected on the wafer surface, the phase shift Φ (x) of the one-dimensional diffraction image is as follows. Is represented by Here, α ₀ and β ₀ are 0th order expansion coefficients obtained by expanding the distortion aberration D _x (x, y) in the x direction and the distortion aberration D _y (x, y) in the y direction of the projection lens by Legendre polynomials, respectively. And It is.

Since the phase shift Φ (x) may be in the positive direction or in the negative direction, the phase shift Φ (x) is required to reduce the image distortion.
The absolute value of (x) needs to be minimized. From equation (5), minimizing the absolute value of the phase shift Φ (x) Is equivalent to minimizing the absolute value of Further, as is apparent from the equation (6), α ₀ and β ₀ are functions of the half width c ₁ and the center position c ₂ of the exposure region 4 in the scanning direction in addition to the functions of x. Therefore, g is a function of the orientation φ of the one-dimensional diffraction grating 1a as well as a function of x, c ₁ , and c ₂ . In the first embodiment described in detail below, for a given φ and c ₁ , c ₂ is determined as a function of x so as to minimize the absolute value of the value of g. In the optimal exposure shape determined in this way, the phase shift Φ becomes minimum with respect to the one-dimensional diffraction grating having the azimuth φ.

Now, the distortion D in the x and y directions of the projection lens
_X (x, y) and D _y (x, y) are Zernike polynomials Z
It is expanded as follows using _{n and m} . To further convert to polar coordinates, Then, the Zernike polynomial Z up to the third order of ρ
_{n and m} are as shown in Table 1.

[0018]

[Table 1]

On the other hand, the distortion D in the x and y directions of the projection lens
Zero-order expansion coefficients α ₀ , β ₀ expanded by Legendre polynomials of _X (x, y) and D _y (x, y), and expansion coefficients A _{n, m} , B _{n of the} Zernike polynomials up to the third order _{, m} have the following relationship. In equation (9a), α ₀ is replaced by β ₀ , and A _{n, m}
Is replaced by B _{n, m} is equation (9b).

Thus, first, the distortions D _x (x, y) and D _y (x, y) of the projection lens in the x and y directions are obtained, and the expansion coefficients A _{n, m} and B _{n, m} are _{calculated based} on these. Then, α ₀ and β ₀ are formulated using the equations (9a) and (9b), and further, (7)
By formulating g (x, c ₁ , c ₂ , φ) in the equation, and then obtaining c ₂ as a function of x so that the absolute value of g is minimized, the given projection lens and pattern arrangement direction the azimuth φ and the half-width c ₁ of the exposure area, it is possible to define the shape c ₂ of the exposure region image distortion is minimized.

FIG. 5 is a flowchart of the first embodiment according to the present invention based on the above steps. First, measurement of the distortion aberrations D _x (x, y) and D _y (x, y) of the projection lens in the x and y directions (a in FIG. 5) and calculation of expansion coefficients A _{n, m} and B _{n, m} (B). Next, after the value of x is initialized to the minimum value (c), the process enters the x loop. Then as long as x is not beyond the maximum value (the d), the value of c ₂ is initialized to a minimum value of the search range (the same e), after initializing the maximum value the minimum value g _min of the absolute value of g entering (at f), c ₂ loop. Next, as long as the value of c ₂ does not exceed the maximum value of the search range (the same g), g
(X, c ₁ , c ₂ , φ) is calculated (h), and when the absolute value of g is smaller than the minimum value g _min (i), the minimum value g
updates the _min, and updates the value of c ₂ at that time as the best value (same j). Thereafter, the value of c ₂ is increased by an appropriate difference Δc ₂ (k), and the c ₂ loop is repeated. c
_When the value of ₂ exceeds the maximum value of the search range (g), the c ₂ loop is terminated, and the best value of c ₂ at x at that time is stored (l). Thereafter, the value of x is increased by an appropriate difference Δx (the same m), and the x loop is repeated.
When the value of x exceeds the maximum value (d), the x loop is terminated and the program is terminated (n). In FIG. 5, P (x) represents the center position of the optimum width of the exposure area at the position x.

[0022] Then as an example, the distortion aberration D _x (x, y) in the x direction of the projection lens expansion coefficient expand in the Zernike Polynomials A _n, of _m, only A _{2, 0} is A _{2, 0} ≠ 0, all other expansion coefficients are 0, ie, D _x (x, y) = A _2,0 Z _2,0 , and similarly D _y (x, y) = B _2,2 Z a _2,2, and shows the calculation result of the shape of the optimum exposure region when a a _{2, 0} = B _2,2 in FIGS. 6-9. The arrow in each figure represents the distortion D (x, y). FIG. 6 shows that the azimuth φ in the pattern arrangement direction S is φ = 0.
7, 8 and 9 show the optimum exposure region shapes when φ = 45 °, φ = 90 °, and φ = 135 °, respectively. It can be seen that the optimum shape of the exposure area changes depending on the azimuth φ in the arrangement direction of the diffraction grating. As described above, in the scanning type semiconductor exposure apparatus, if the shape of the opening for limiting the exposure area is determined by the above-described means, the figure transferred onto the wafer will be similar to the figure on the reticle with less image distortion. It becomes a figure with high character.

When the distortion of the projection lens has a higher order distortion than the third order, the equations (9a) and (9b) can be expressed as follows. In this case, (10a) at each position x with respect to the opening half-width c ₁ given, by substituting the (10b) Equation (7), (7) the center position c of the opening type is minimized By obtaining ₂ , the aperture that minimizes the phase shift can be determined.

Next, a second embodiment will be described. FIG.
In the flowchart of the first embodiment shown in FIG. 7, a single azimuth is designated in advance as the azimuth φ in the pattern arrangement direction S, and the aperture is set so that the absolute value of g in Expression (7) is minimized in that azimuth φ The shape was determined. As a result, FIGS.
As shown in the figure, the optimum aperture shape was different for each of the specified directions φ. Therefore, in the second embodiment,
As shown in FIG. 10, φ is placed between the x loop and the c ₂ loop.
A loop is provided, and the minimum value g _min of the absolute value of g and the value of c ₂ at that time are obtained for each of the plurality of directions φ, and then the direction in which the minimum value g _min of the absolute value of g is the maximum The value of c ₂ at φ is adopted.

That is, the method of the second embodiment is designed to minimize the image distortion in the azimuth where the image distortion is the worst.
This determines the opening shape. As a result, the center position c _{2 of the} opening obtained for each position x is a certain direction φ
May not necessarily be the best position, but the image distortion in the direction φ is never worse than the image distortion in the direction in which the image distortion is the worst. As a plurality of directions φ repeated in the φ loop, for example, φ = 0
, 45 °, 90 ° and 135 °.

Next, a third embodiment will be described. As is clear from the above description, the optimum opening shape differs for each direction φ in the pattern arrangement direction S. That (7) and a set of x, c _1, c ₂ of the alpha ₀ to a minimum in the formula, x is a beta ₀ to a minimum, because generally different from the set of c _1, c _2, the equation (7) The set of x, c ₁ , and c ₂ that minimizes the whole changes depending on the direction φ in the pattern arrangement direction. An ordinary reticle for IC is composed of a one-dimensional diffraction grating pattern having various directions φ, and a set of x, c ₁ , and c ₂ that minimizes α ₀ in equation (7). , Β ₀ , there is no aperture shape that minimizes image distortion for any orientation φ, as long as the combination of x, c ₁ , and c ₂ that minimizes β.

Thus, if the distortion D _x (x,
y) and the distortion aberration D _y (x, y) in the y direction, When there is a correlation of: It is expressed as That is, as shown on the right side of the equation (12), the quantity (ν _x + kν _y ) related to the orientation φ of the diffraction grating,
The variable α ₀ (x, c ₁ , c ₂ ) related to distortion can be separated into variables. Therefore (11) by fabricating the projection optical system so as to satisfy the equation, x where g is minimum set of c _1, c ₂ is, x the alpha ₀ is minimized, c _1, c ₂
And as a result c
₂ can be obtained independently of the orientation φ of the diffraction grating.

The relation of the equation (11) can be expressed as follows using expansion coefficients A _{n, m} and B _{n, m} developed by Zernike polynomials. That is, first, the distortion D
X component D _x (x, y) and y component D _y (x, y) of (x, y)
y) means that they are composed of the same type of aberration,
That is, a set of (n _i , m _i ) where A _{n, m} ≠ 0 (i = m
1 to j) and (n _i , m _i ) where B _{n, m} ≠ 0 ( _i )
= 1 to j) and that the match, the second, the match (n _i, A _n of the set of _{m i) (i = 1~j)} , m (≠ 0) and B _{n, m} (≠ 0) To satisfy the relationship.

As an example, in the basic aberrations shown in Table 1, when only (3,3) aberrations among (n, m) aberrations exist, that is, At this time, α ₀ (x, c ₁ , c ₂ ) becomes as follows by setting expansion coefficients other than A _{3 and 3 in} equation (9) to 0.

The optimum aperture for minimizing the phase shift is (1
The opening is determined from the parentheses on the right side of Equation 4), and the shape of the center line of the opening is as follows. FIG. 11 shows the aperture shape given by the equation (15) together with the distortion figure. Note that this distortion aberration figure satisfies the expression (11), that is, the size and the direction of the arrow differ depending on each position (x, y), but the arrows are parallel to each other including the opposite direction.

Similarly, (3, −) of the set (n, m)
When only the set of aberrations of 1) exists, that is, In this case, α ₀ (x, c ₁ , c ₂ ) becomes as follows by setting the expansion coefficients other than A _{3, -1 in} equation (9) to 0.

In this case, there are two optimum apertures for minimizing the phase shift. The first optimal aperture has a centerline of x
A rectangular aperture that coincides with the axis and is given by: The second optimum opening is an opening determined from the parentheses on the right side of the equation (16), and the shape of the center line of the opening is as follows. FIG. 12 shows the second opening shape given by Expression (18) as follows:
This is shown together with the distortion figure.

As described above, when the distortion is a single low-order aberration, the optimum aperture shape can be easily obtained analytically. When it is difficult to obtain the value analytically, the value can be obtained, for example, by the method shown in FIG. The method of FIG. 13 is a flowchart showing a general solution of the third embodiment, which is substantially the same as the method of FIG. 5 except that g is replaced by α ₀ .

In order to apply the method of the third embodiment, it is necessary that the projection optical system substantially satisfies the expression (11) as described above. By the way, in order to apply the method of the third embodiment, it is necessary that Expression (11) is satisfied in the search range. However, after the aperture shape is finally determined, distortion in portions other than the aperture is determined. Aberration is no longer a problem. Thus: That is, the present invention
Although the invention of the method as a method of determining an exposure area and the invention of a product as a projection optical system are included, the invention of the method includes (11) a range including an opening shape that will be finally determined. The expression must be satisfied. On the other hand, the invention of the object only needs to be a projection optical system that satisfies the expression (11) only in the finally determined aperture.

It is not always easy to manufacture a projection optical system so as to satisfy the expression (11). However, if the condition is close to satisfying the expression (11), in other words,
If the arrows of the distortion aberration shown in FIG. 11 or FIG. 12 are close to a state parallel to each other, the set of x, c ₁ , c ₂ that minimizes α ₀ and the β ₀ in equation (7) are minimized. X, c ₁ , c ₂
Approaching the pair. In this state, when determining the aperture shape on the basis of the method of the second embodiment, the resulting center position c ₂ openings take in any orientation phi, so that the best position and not inferior. Therefore, it is preferable to manufacture the projection optical system so as to satisfy Expression (11) as much as possible.

[0036]

As described above, according to the present invention, the optimum aperture shape is determined by using the function g uniquely determined from the distortion of the projection lens and the azimuth φ of the figure on the reticle. is there. The present invention also discloses a method of determining an optimum aperture shape using a function α ₀ uniquely determined only from the distortion of the projection lens. Therefore, by these methods, an aperture region that minimizes distortion (phase shift) of an exposed image can be obtained. In a semiconductor exposure apparatus, a decrease in modulation depth and a phase shift are fatal defects, and in particular, a phase shift is a major cause of disturbing the similarity of figures. Therefore, the method for determining the aperture shape according to the present invention that minimizes the phase shift can be said to be a reasonable method.

[Brief description of the drawings]

FIG. 1 is a schematic longitudinal sectional view showing a scanning type exposure apparatus to which the present invention is applied.

FIG. 2 is a plan view showing an opening shape of a projection optical system of the scanning exposure apparatus.

FIG. 3 is a plan view showing a reticle pattern.

FIG. 4 is a schematic diagram showing a phase shift between a reticle pattern and its image.

FIG. 5 is a flowchart showing a method of determining an exposure area according to the first embodiment.

FIG. 6 shows a first example when the reticle pattern orientation is 0 °.
FIG. 5 is a diagram illustrating an example of an exposure area determined by an example and distortion aberration used for the determination.

FIG. 7 is a diagram illustrating an example of an exposure area determined by the first embodiment and distortion aberration used for the determination when the reticle pattern orientation is 45 °.

FIG. 8 is a diagram showing an example of an exposure area determined by the first embodiment and distortion aberration used for the determination when the reticle pattern orientation is 90 °.

FIG. 9 shows a case where the orientation of the reticle pattern is 135 °.
FIG. 7 is a diagram illustrating an example of an exposure area determined by the first embodiment and distortion aberration used for the determination.

FIG. 10 is a flowchart showing a method of determining an exposure area according to a second embodiment.

FIG. 11 is a diagram showing an example of an exposure area determined by the third embodiment and distortion aberration used for the determination.

FIG. 12 is a diagram showing another example of the exposure area determined by the third embodiment and the distortion used for the determination.

FIG. 13 is a flowchart showing a general method for determining an exposure area according to a third embodiment.

[Explanation of symbols]

1 ... reticle 1a ... pattern (1
2D wafer 3 Projection optical system 4 Aperture (exposure area)

Claims (5)

[Claims] An image in an exposure area of an image of a pattern on a reticle is formed on a wafer by a projection optical system, and the wafer and the reticle are adjusted to a speed ratio according to a magnification of the projection optical system. Scanning in synchronization with each other and uniformly forming the width of the exposure region in the scanning direction, thereby transferring the pattern on the reticle onto the wafer. And a method of determining the shape in the longitudinal direction perpendicular to the, based on the distortion aberration of the projection optical system and the arrangement direction of the pattern on the reticle, the image and the pattern in the exposure area by the projection optical system And determining a shape of the exposure region in the longitudinal direction so that a phase shift between the exposure regions is minimized. 2. A method according to claim 1, wherein a single direction is designated as an arrangement direction of the pattern, and the longitudinal shape of the exposure area is determined.
A method for determining an exposure area of the scanning exposure apparatus according to claim 1. 3. A plurality of directions are designated as an arrangement direction of the pattern, and in any one of the plurality of arrangement directions, the longitudinal direction of the exposure region is minimized so that the phase shift is minimized. 2. The method according to claim 1, wherein the shape is determined. 4. An image in an exposure area of a pattern image on a reticle is formed on a wafer by a projection optical system, and the wafer and the reticle are adjusted to a speed ratio according to a magnification of the projection optical system. The projection optical system of a scanning exposure apparatus for transferring a pattern on the reticle onto the wafer by forming a uniform width of the exposure region in the scanning direction. The ratio of the component in the scanning direction of the distortion of the projection optical system to the component in the longitudinal direction orthogonal to the scanning direction is formed so as to be substantially constant over the entire exposure area. A projection optical system of a scanning exposure apparatus, which is characterized by the following. 5. An image in an exposure area of a pattern image on a reticle is formed on a wafer by a projection optical system, and the wafer and the reticle are adjusted to a speed ratio according to a magnification of the projection optical system. Scanning in synchronization with each other and uniformly forming the width of the exposure region in the scanning direction, thereby transferring the pattern on the reticle onto the wafer. And determining a shape in the longitudinal direction orthogonal to the scanning direction, wherein the ratio of the component of the distortion of the projection optical system in the scanning direction and the component in the longitudinal direction orthogonal to the scanning direction is substantially constant. The projection optical system is formed such that a phase shift between the image in the exposure area and the pattern by the projection optical system based on the distortion of the projection optical system is minimized. Define the longitudinal shape of the area It characterized Rukoto method of determining the exposure area of the scanning type exposure apparatus.