JPH0786139A - Aligner - Google Patents

Abstract

PURPOSE:To provide an aligner in which correction of magnification can be executed through a simple structure. CONSTITUTION:An aligner for projecting the image of a first object onto a second object through a projection optical system while scanning them comprises means 13 for bending the first object 2 so that a curvature is provided at least in the perpendicularly intersecting snanning direction of the first and second objects. The projection optical system comprises a plurality of erecting optical systems 3a-3e for forming the erecting image of the first object.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a scanning type exposure apparatus, and more particularly to an exposure apparatus suitable for exposing a large screen.

[0002]

2. Description of the Related Art An exposure apparatus for collectively transferring a large exposure area is a slit scanning type exposure in which an image of a mask is sequentially formed on a substrate by a concentric spherical reflecting mirror while moving the mask and the substrate. The device is conventionally known as a mirror projection aligner.

[0003]

However, in the case of transferring a pattern under a high resolution in a wide exposure area, it is necessary to increase the diameter of the spherical reflecting mirror with high accuracy in the mirror projection aligner as described above. .
Therefore, there is a problem that the manufacturing of the reflecting mirror becomes difficult and the manufacturing cost is significantly increased.

Therefore, a multi-lens type exposure apparatus in which a large number of lens arrays are arranged along a one-dimensional direction and an original image (transfer pattern) and an object to be exposed are scanned in a direction orthogonal to this direction is the same applicant as the present application. Japanese Patent Application No. 5-161588. The exposure apparatus proposed in this Japanese Patent Application No. 5-161588 has a configuration in which projection optical systems for forming an equal-size erect image of a mask on a substrate are arranged in a staggered pattern. Scanning exposure is performed.
In this exposure apparatus, the number of projection optical systems is increased, so that the exposure area can be enlarged.

Recently, it has been required to change the projection magnification on the substrate for the convenience of the manufacturing process of the liquid crystal panel. Here, in the multi-lens type exposure apparatus, even if the magnification of a plurality of projection optical systems is changed, the optical axis spacing between the projection optical systems is constant, so that the magnification is changed uniformly over the entire screen. I couldn't do it. Therefore, an object of the present invention is to provide an exposure apparatus that can perform magnification correction with a simple configuration.

[0006]

In order to achieve the above object, an exposure apparatus according to the present invention has the following configuration. For example, as shown in Figure 1, the first object (2) and the second object
(6) is an exposure apparatus that projects and exposes an image of a first object onto a second object via a projection optical system while moving at least one of the first object and the second object. To bend so that it has a curvature in a direction substantially perpendicular to the direction
The projection optical system is configured to have a plurality of erecting optical systems (3a to 3e) forming an erecting image of the first object.

Further, the erecting optical system is preferably constructed so as to have optical path length varying means (4a to 4e) for varying the optical path length of the erecting optical system. The erect image in the present invention is an image in which the lateral magnification in the vertical and horizontal directions is positive.

[0008]

In the present invention having the above-described structure, at least one of the first object and the second object can be bent in a direction perpendicular to the scanning direction (scanning vertical direction), and therefore, this scanning is performed. The length of the object in the vertical direction can be made variable. Therefore, it becomes possible to substantially change the magnification in the direction orthogonal to the scanning direction.

For example, when the first object is bent, the second object
A focus error occurs in the Z direction (optical axis direction) on the object. If this focus error is within the depth of field of the projection optical system, there is no effect when projecting and transferring the first object onto the second object. Here, the depth of field of the projection optical system is a range on the object side corresponding to the depth of focus, and when the projection optical system has the same magnification, the depth of focus and the depth of field are equal.

Here, when the focus error is larger than the depth of field of the projection optical system, it is desirable to provide an optical path length varying means for varying the optical path length of the projection optical system.
Thereby, the focus error can be contained within the depth of field of the projection optical system, and the image of the first object can be formed on the second object under a favorable image formation state.

[0011]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a diagram showing the configuration of an embodiment according to the present invention. In this embodiment, the optical axis direction of the projection optical system 3 is the Z axis and the scanning direction is the Y axis. The direction orthogonal to these Z axis and Y axis is the X axis.

In FIG. 1, exposure light (eg, g-line, i-line, etc.) from the illumination optical system 1 uniformly illuminates the mask 2 placed on the mask stage 8. On the mask 2,
For example, a circuit pattern made of chrome or the like is formed, and this circuit pattern includes a plurality of projection optical systems 3a ...
It is projected and transferred onto the plate 6 placed on the plate stage 7 by 3e. Here, each of the plurality of projection optical systems 3a to 3e forms an erect normal image of the same size as the circuit pattern. Multiple projection optical systems 3a to 3e
For example, a combination of two sets of relay optical systems, a combination of inhomogeneous lenses, and the like can be applied. In FIG. 1, five sets of projection optical systems 3a-3
Although e is illustrated, the number of projection optical systems is not limited to five.

Here, the mask stage 8 and the plate stage 7 are constructed so as to be movable in the Y-axis direction (the direction perpendicular to the paper surface in the drawing), and at the time of exposure, the mask 2 and the plate 6 are moved to each other by a driving unit (not shown). Move together in the Y direction. As a result, the circuit pattern on the mask 2 is scanned and exposed on the plate 6. Now, the exposure apparatus shown in FIG. 1 is provided with a mask deforming portion 13 for bending the mask 2 in the scanning vertical direction. The mask deforming unit 13 bends the mask by applying a load to the mask 2. At this time, the mask deforming unit 13
FIG. 2 shows an example of the load applied by. In FIG. 2, an actuator that applies a load to the mask 2 is not shown.

In FIG. 2A, the mask stages 8a ₁ and 8a _{2 each} having an L-shaped cross section support the mask 2 by vacuum-adsorbing both ends of the mask 2. At this time, when the mask stages 8a ₁ and 8a ₂ are moved in the direction of the arrow (Y direction) in the drawing (applying a compressive load in the Y direction of the mask 2) so that the mask stages 8a ₁ and 8a ₂ approach each other, the mask 2 becomes
It is transformed into a shape with a convex surface facing the projection optical system.

Further, as the mask deforming portion 13, as shown in FIG.
As shown in (b), a configuration may be adopted in which a concentrated load is applied from the Z direction of the mask 2. In FIG. 2 (b), the mask 2
Are vacuum-sucked to the mask stages 8b ₁ and 8b ₂ . Then, when a load is applied to both ends of the mask 2 in the X direction from the arrow direction (Z direction) in the figure, the mask 2 is deformed into a shape with its convex surface facing the projection optical system side.
At this time, as a configuration for applying a load to the mask 2, for example, a configuration in which a rod-shaped member is brought into contact with the mask 2 and this member is driven in the Z direction by an actuator or the like can be considered.

Then, as shown in FIG. 2 (c),
Mask stage 8c with cross section₁, c ₂By mask
A configuration is conceivable in which both ends of 2 are held so as to be sandwiched. This
At this time, both ends of the mask 2 are
Along the axis of rotation about the axis (rotation about the X axis
Mask stage 8c₁, c₂When you rotate
The mask 2 is transformed into a shape with a convex surface facing the projection optical system side.
Will be. It should be noted that, in FIGS. 2 (b) and 2 (c),
If a force is applied in the direction opposite to the arrow
It will be deformed to have a concave surface facing the system side.

As the actuator that applies a force to the mask 2, for example, a laminated piezoelectric element can be applied. Although the mask 2 is flexed in this embodiment, the plate 6 may be flexed. Next, with reference to FIG. 3, a quantitative study is performed on the amount of bending of the mask 2. FIG. 3 is a cross-sectional view of the mask 2 on the YZ plane. In the following description, it is assumed that the bending of the mask 2 has an arc shape centered on a predetermined point.

In FIG. 3, the length of the bent mask 2 along the Y direction is defined by R, where R is the radius of curvature of the bending.

[0019]

## EQU1 ## L≈2Rθ (1) can be approximated. At this time, the radius of curvature R of the bending is considered on the basis of a neutral surface (a surface that does not expand or contract) in the thickness direction of the mask 2 shown by a broken line in the drawing. The one-side shrinkage amount ΔL of the lower surface of the mask 2 is defined by the thickness D of the mask 2 and a straight line extending from the curvature center point of the deflection to the center point of the mask 2 (the portion having the largest deflection amount) and the mask center from this curvature center. When the angle formed by the straight line extending to the end of 2 is θ,

[0020]

## EQU2 ## ΔL = Dθ / 2 (2) Here, the deformation amount Δ at the end of the mask 2
Z (the amount of deviation in the optical axis direction between the central portion and the end portion of the mask 2) is

[0021]

Equation 3] [Delta] Z = R represented by ^{(1-cosθ) ≒ Rθ 2} /2 ... (3). At this time, the contraction rate (elongation rate) a of the mask 2 before bending and after bending is

[0022]

## EQU00004 ## a = 2.DELTA.L / L (4) From this equation (4) and the above equation (1), the shrinkage rate (elongation rate) a is

[0023]

## EQU00005 ## a = D.theta. / L (5) From the expressions (2) to (5), the radius of curvature R of the bending is
Is erased, the deformation amount ΔZ at the end of the mask 2 is

[0024]

[Expression 6] ΔZ = a · L ² / 4D (6) For example, consider a case where the length of the bent mask 2 in the Y direction is L = 300 mm and the thickness of the mask 2 is D = 6 mm. At this time, the contraction rate of the mask 2 is a = 2 × 10 ⁻⁵ (20 pp
In order to obtain m), the mask deforming portion 13 may bend the mask 2 so that the deformation amount ΔZ of the end portion of the mask 2 due to the bending becomes ΔZ = 0.075 mm = 75 μm. Since the deformation amount ΔZ of the end portion of the mask 2 indicates the amount of deviation in the optical axis direction between the center portion and the end portion of the mask 2, when the end portion of the mask 2 is fixed, It can be regarded as the amount of deformation in the center.

The image of the circuit pattern on the bent mask 2 is formed on the plate 6 by the plurality of projection optical systems 3a to 3e. At this time, the magnification in the scanning orthogonal direction (X direction) of the circuit pattern image formed on the plate 6 is the same as the shrinkage rate a of the mask 2. in this way,
In the exposure apparatus according to the present embodiment, the scanning orthogonal direction (X direction)
It is possible to substantially change the magnification at.

In the above-described embodiment, if the deformation amount ΔZ of the mask 2 due to the bending is within the depth of field of each of the projection optical systems 3a to 3e, a good image formation state is obtained on the plate 6. Thus, the circuit pattern image of the mask 2 is formed. Hereinafter, the relationship between the deformation amount ΔZ of the mask 2 due to bending and the depth of field of the projection optical system will be specifically described. In the following description, considering the case where the length L of the mask 2 in the Y direction is divided into N (where N is an odd number) regions for exposure, the central region of the mask 2 in the Y direction is used as a reference. (1
Th).

The shift amount δ _n in the Z direction in the _nth region (n is a natural number) is the shift amount ΔZ _{n-1 in} the Z direction between the outer end portion and the center point of the n−1th region, and n. This is the difference between the Z-direction deviation amount ΔZ _n between the outer end and the center point in the th region. Therefore, the shift amount δ _n in the Z direction in the n-th region is calculated by calculating the straight line connecting the outer end of the n-th region and the center of curvature of bending, the center (center point of the center region) and the center of curvature of bending. _Let θ _n be the angle formed by the line connecting

[0028]

Equation 7] _{δ n = ΔZ n -ΔZ n-} 1 = (θ n 2 -θ n-1 2) · R / 2 ... are expressed as (7). _{However, θ n = θ / n +} 2θ / N = (2n-1) θ / n is a (n = 1,2, ‥‥ (N + 1) / 2).

Here, the regions at both ends of the mask 2 are (N +
1) / 2nd. Therefore, the shift amount δ _{(N + 1) / 2} in the Z direction in the regions at both ends of the mask ₂ is

[0030]

Represented by Equation 8] _{δ (N + 1) / 2} = (Rθ 2/2) {4 (N-1) / N 2} = ΔZ · 4 (N-1) / N 2 ... (8) . Here, Z in the regions at both ends of the mask 2
When the amount of depth of field δ _{(N + 1) / 2} in the direction is Δt (which is equal to the depth of focus in the present invention, the amount of depth of field of the projection optical system is equal to),

[0031]

## EQU9 ## It is preferable to satisfy δ _{(N + 1) / 2} ≤ Δt (9). Here, if Δt = ΔZ / b (b is a constant), the above equation (9) is

[0032]

[Equation 10] 4 (N−1) / N ² ≦ 1 / b (10) The following equation (11) can be obtained by modifying the equation (10).
It becomes like a formula.

[0033]

N ≧ 2 {b + (b ² −b) ^1/2 } (11) For example, the deflection amount ΔZ of the mask 2 (deviation in the Z direction between the central portion and the end portion of the mask 2) is 75 μm, When the depth of focus of the projection optical system is ± 12.5 μm, b = 3, so N ≧ 10.9. Therefore, in the above case, if the mask 2 is divided into more than 11 divisions, the mask 2 will be positioned within the depth of field of each projection optical system.

In the above example, the focus error due to the bending of the mask 2 is corrected by dividing the exposure area. Here, for example, when the deformation amount ΔZ of the mask 2 due to bending becomes larger than the depth of focus of each of the projection optical systems 3a to 3e, an optical path length correction unit is arranged in the optical path of each of the projection optical systems 3a to 3e. The focus positions of the projection optical systems 3a to 3e may be corrected.

Specifically, as shown in FIG. 1, the optical path length correction members 4a to 4e are inserted in the optical paths of the projection optical systems 3a to 3e. As a result, the focus error caused by the bending of the mask 2 can be corrected by making the optical path lengths of the projection optical systems 3a to 3e different from each other. Hereinafter, a quantitative examination will be made on the configuration of the optical path length correction members 4a to 4e. In the following description, the length L of the mask 2 in the Y direction is the same as in the above example.
Consider a case in which exposure is divided into N (where N is an odd number) areas for exposure.

The shift amount δ _m in the Z direction in the m-th region (m is a natural number) is the shift amount ΔZ _{m-1 in} the Z direction between the outer end and the center point in the m-1 th region and the m-th region. This is the difference between the outer edge and the center point in the region of Z in the shift amount ΔZ _{m in} the Z direction. Here, the average of the deformation amount in the m-th region (the average value of the deformation amount with respect to the center point) ΔZ _ma is

[0037]

ΔZ _ma = (ΔZ _m + ΔZ _m-1 ) / 2 = (R / 2) (θ _m ² + θ _m-1 ² ) / 2 = (4 m ² -8 m + 5) R θ ² / 2N ² (12) ) Is represented by.

Therefore, in the optical path of the projection optical system for projecting the m-th area on the plate 6, the average Δ of the above deformation amount.
If an optical path length correction member that corrects the optical path length by Z _ma is inserted, this m-th region can be imaged on the plate 6 without a focus error. Here, the optical path length correction members 4a-
When each of 4e is formed by a plane parallel plate, if the plane parallel plate is formed so as to satisfy the following expression (13), the focus error of the mask 2 due to bending can be corrected.

[0039]

D (n-1) / n = (4m ² -8m + 5) Rθ ² / 2N ² (13) where d is the thickness of the plane-parallel plate, and n is the refractive index of the plane-parallel plate at the exposure wavelength. ,. Then, the above equation (13) can be modified as follows.

[0040]

D = n (4m ² −8m + 5) Rθ ² / {(n−1) 2N ² } (14) In the above equation (14), the center of the bent mask 2 is used as a reference. There is. That is, a plane parallel plate configured to satisfy the above expression (14) is inserted in the optical path of the projection optical system that forms an image of the m-th area from the center of the mask 2 on the plate 6.

Now, let us consider a case where the mask 2 is bent into a shape with its convex surface facing the projection optical systems 3a to 3e, as in the example of FIG. At this time, if the thickness of the plane-parallel plate in the optical path of the projection optical system for projecting the central area (first area) of the mask 2 is d ₀ , it is in the optical path of the projection optical system for projecting the m-th area. The thickness d _m of the parallel plane plates arranged is

[0042]

[Expression 15] d _m = d ₀ + d (15) Further, when the mask 2 is bent into a shape with a concave surface facing the projection optical system side, the thickness of the plane parallel plate disposed in the optical path of the projection optical system that projects the central region (first region) of the mask 2 Is d ₀ , the thickness d of the plane-parallel plate arranged in the optical path of the projection optical system that projects the m-th region.
_m satisfies the following expression (16).

[0043]

D _m = d ₀ −d (16) FIG. 4 shows an example of a specific configuration of the optical path length correction members 4a to 4e. In FIG. 4, parallel plane plates 41 to 44 having different thicknesses are provided in the opening of the frame 40 having a plurality of openings. The frame 40 is movably provided along the arrangement direction of the plane parallel plates 41 to 44, and any one of the plane parallel plates 41 to 44 is located in the optical path of the projection optical system. Alternatively, it is driven by the optical path length correction unit 14 so that none of these parallel plane plates 41 to 44 is positioned.

Next, returning to FIG. 1, the operation of magnification adjustment will be described. The exposure apparatus shown in FIG. 1 has an input unit 1 including a keyboard and the like for inputting information regarding magnification.
1 is provided. The information regarding the desired magnification input to the input unit 11 is output to the control unit 12 including a CPU and a memory unit. Further, the exposure apparatus of FIG. 1 is provided with mask deformation amount detection units 10a and 10b for detecting the deformation amount of the mask 2. The mask deformation amount detection units 10a and 10b receive light from the light projecting unit 10a that irradiates a plurality of locations on the mask 2 from a diagonal direction and light from the light projecting unit 10a reflected by the mask 2. Light receiving portion 10b that outputs information regarding the light receiving position of the light. The light receiving position information from the light receiving unit 10b is output to the control unit 12. Note that the mask deformation amount detection units 10a and 10b detect the mask deformation amount as follows.
It suffices if it is at least in the area illuminated by the illumination optical system (field of view of the projection optical system).

The control unit 12 for adjusting the magnification will be described below.
The operation of will be described. When adjusting the magnification,
First, a desired magnification is input to the input unit 11. The input information regarding the desired magnification is output to the control unit 12.
The control unit 12, based on the magnification information from the input unit 12,
The deformation amount ΔZ of the mask 2 is calculated. Then, the control unit 12
While detecting the current deformation amount of the mask 2 based on the information on the light receiving position from the light receiving unit 10b, the detected deformation amount and the calculated deformation amount ΔZ are equal to each other.
The mask deforming unit 13 is controlled to bend the mask 2.

Next, the control unit 12 calculates the deformation amount of the mask 2 for each of the divided areas. At this time, when the amount of deformation for each area is smaller than the depth of field of each of the projection optical systems 3a to 3e, the control unit 12 controls a driving unit (not shown) to cause the plate stage 7 and the mask stage 8 to operate. And are moved together to perform scanning exposure. Further, when the deformation amount for each area is larger than the depth of field of each of the projection optical systems 3a to 3e, the control unit 12 determines, based on the calculated deformation amount,
The thickness of the optical path length correction members 4a to 4e arranged in each of the projection optical systems 3a to 3e is calculated. Then, the control unit 12
The optical path length correction unit 14 is controlled to position the plane-parallel plate having the calculated thickness in the optical path of each of the projection optical systems 3a to 3e. As a result, the mask 2 with the magnification adjusted is provided on the plate 6.
Image is formed under a good imaging condition. After that, the control unit 12 controls a driving unit (not shown) to integrally move the plate stage 7 and the mask stage 8 to perform scanning exposure.

In the embodiment shown in FIG. 1, the optical path length correction members 4a to 4e are inserted on the image side of the projection optical systems 3a to 3e.
4e may be arranged on the object side of each of the projection optical systems 3a to 3e. Next, another embodiment according to the present invention will be described with reference to FIG. FIG. 5 is an optical path diagram showing an example in which the projection optical system is composed of two sets of relay optical systems.

In FIG. 5, illumination light from an illumination optical system (not shown) uniformly illuminates the mask 2. Then, the light from the mask 2 reaches the plate 6 via the three sets of projection optical systems 23a to 23c provided below the mask 2, and forms an image of the mask 2 on the plate 6. Although not shown in FIG. 5, a mask stage for supporting the mask, a plate stage for mounting the plate, and a mask deforming portion for deforming the mask are also provided in this embodiment.

Here, the projection optical systems 23a-23c are
First relay system 23a ₁ , 2 forming an intermediate image of the mask 2
3b ₁ and 23c ₁ and second relay systems 23a ₂ and 23 that form an erect image of the mask 2 based on the light from the intermediate image.
b ₂ and 23c ₂ and field diaphragms 25a, 25b and 25c provided at the intermediate image forming positions. Here, the first relay systems 23a _{1 to} 23c ₁ and the second relay systems 23a _{2 to} 23c ₂ are both side telecentric optical systems.

When the mask 2 is bent to correct the magnification by the mask deforming unit (not shown), a focus error occurs at the end of the mask 2. Here, the amount of deformation of the mask 2 in each area (the respective visual field areas of the three sets of projection optical systems 23a to 23c) is the projection optical system 23a.
When the depth of field is larger than ˜23c, in order to correct the focus error due to deformation, the optical path length correction member 24
a and 24c are arranged in the optical paths of the projection optical systems 23a and 23c. In the example of FIG. 5, the mask 2 is the projection optical system 23.
It is assumed that the convex surface faces the a to 23c side, and the circuit pattern surface in the central portion of the mask 2 and the surface on the plate 6 are conjugate.

The optical path length correcting members 24a and 24c are made of plane parallel plates, and the plane parallel plates are each made to have a thickness as shown in the above equation (15). When the projection optical system is composed of two sets of relay optical systems as in the present embodiment, the optical path length correction member 24
a and 24c are the mask 2 and the field stop 25a, 25b, 2
It is desirable to be provided in the optical path between the optical fiber 5c and 5c. Here, the optical path length correction members 24a, 24c are the field stop 25a,
When placed in the optical path between the plates 25b and 25c and the plate 6, the field diaphragms 25a and 25 projected on the plate 6 are projected.
This is not preferable because the image forming state of the images of b and 25c deteriorates.

Further, as in this embodiment, the projection optical system 2
When 3a to 23c are telecentric, it is more preferable that the optical path length correction members 24a and 24c are arranged in the optical path in which the chief ray is telecentric.
In this case, it is possible to reduce the influence of aberrations such as astigmatism and coma caused by the light rays passing through the plane-parallel plate.

Even when the projection optical system is composed of two or more sets of relay optical systems, it goes without saying that it is desirable to dispose the optical path length correcting member in the optical path between the mask and the field stop. Next, another embodiment according to the present invention will be described with reference to FIG. FIG. 6 is a schematic diagram showing an example in which the present invention is applied to an exposure apparatus according to Japanese Patent Application No. 5-161588 by the same applicant as this application.

In FIG. 6, the mask 2 is uniformly illuminated by the exposure light from the illumination optical system 10. The light from the illuminated circuit pattern on the mask 2 forms an image of the mask 2 on the plate 6 via the plurality of projection optical systems 33a to 33g. Here, the plurality of projection optical systems 33a to 33
Each g is configured to have two sets of partial optical systems. The configuration of the projection optical system will be described below with reference to FIG. FIG. 7 is a diagram showing the configuration of the projection optical system. Here, in order to simplify the explanation, the projection optical system 33
Only a will be described.

In FIG. 7, the projection optical system 33a includes a first partial optical system 33a ₁ which forms a primary image on the mask 2 and
The light from this primary image causes the mask 2 on the plate 6 to
The second partial optical system 33a ₂ for forming the next image and the field stop 25 arranged at the primary image forming position. These first and second partial optical systems 33a ₁ and 33a ₂ are each constructed by modifying a Dyson type optical system.

The light from the mask 2 is reflected by the rectangular prism P ₁ ,
The plano-convex lens component L ₁ and the meniscus lens component L ₂ are sequentially passed to reach the reflecting surface M ₁ provided on the lens component L ₂ . The light reflected by the reflecting surface M ₁ passes through the lens component L ₂ , the lens component L ₁ , and the right-angle prism P ₂ in this order, and then enters the exit side of the right-angle prism P _{2 at} the same magnification of 1 of the mask 2.
Form the next image. Here, the right-angle prisms P ₁ and P ₂ and the lens components L ₁ and L _{2 form} the first partial optical system 33a ₁ . The primary image is an inverted image (an image in which the lateral magnification in the vertical and horizontal directions is negative), and a field diaphragm 25a having an opening of a predetermined shape is provided at the primary image forming position.

The light passing through the field stop 25a passes through the right-angle prism P ₃ , the plano-convex lens component L ₃ , and the meniscus lens component L ₄ in this order, and then is reflected by the lens component L _3. Reach M ₂ . The light reflected by the reflecting surface M ₂ passes through the lens component L ₃ and the lens component L ₄ in this order, and forms a secondary image of the same size as the mask 2 on the plate 6. This secondary image is an erect image. Here, the right-angled prisms P ₃ and P ₄ and the lens components L ₃ and L _{4 form} the second partial optical system 33a ₂ .

These first and second partial optical systems 33
Reference numerals a ₁ and 33a ₂ are both-side telecentric optical systems. That is, the projection optical system 33a (33b to 33g) is a double-sided telecentric optical system. Therefore, if the projection optical systems 33a to 33g are arranged in a row along the scanning orthogonal direction (Y direction), it is not possible to expose the entire area of the mask 2 by one scanning exposure.
Therefore, in this embodiment, the projection optical systems 3a to 3g are arranged in a staggered pattern.

Here, the field of view of each projection optical system is
As shown in the plan view of FIG. 8, they are defined so as to overlap each other in the Y direction. In FIG. 8, the visual field regions 20a to 20g are defined so that the sum of the X-direction length of the non-overlapping regions and the X-direction length of the mutually overlapping regions is equal. This allows the plate 6
The exposure amount is always the same at any position along the Y direction above. The shapes of the visual field regions 20a to 20g are defined by the shapes of the openings of the visual field diaphragms of the projection optical systems 33a to 33g. Here, the opening of the field stop and the field regions 20a to 20g have similar shapes.

Now, returning to FIG. 1, these projection optical systems 33a to 33g are arranged in a staggered manner. Here, the projection optical systems 33a to 33d are arranged in a line in the scanning orthogonal direction, and the projection optical systems 33e to 33d.
g is arranged along a column in the Y direction different from the arrangement position of the projection optical systems 33a to 33d. The projection optical systems 33a to 33g form exposure regions 60a to 60d arranged on the plate 6 along the Y direction in the drawing, and the projection optical systems 33e to 33g form the exposure regions 60a to 60d.
Exposure regions 60e to 60g arranged along the Y direction are formed at positions different from a to 60d. These exposure areas 60a to 60g are erect images of the same magnification as the visual field areas 20a to 20d.

Now, also in the exposure apparatus according to this embodiment, the mask deforming portion 13 is provided for adjusting the magnification on the plate 6. As a result, the mask 2 is bent so as to have a curvature in the scanning orthogonal direction,
The magnification on the plate 6 can be changed. Then, as shown in FIG. 7, the field stop 2 in the projection optical system 33a of the present embodiment.
An optical path length correction member 24 for correcting the focus error due to the bending is provided in the optical path between the optical path 5a and the first partial optical system 33a _1.
a is provided. As a result, on the plate 6,
An image of the mask 2 whose magnification has been adjusted is formed under a favorable image formation state.

The mask 2 is placed on the mask stage 8 and the plate 6 is placed on the plate stage 7. Here, the mask stage 8 and the plate stage 7 move integrally along the X direction in the drawing. As a result, the images of the mask 2 illuminated by the illumination optical system 10 are sequentially transferred onto the plate 6, and so-called scanning exposure is performed. By moving the mask 2, the visual field 2
When the scanning of the entire surface of the mask 2 with 0a to 20g is completed, the image of the mask 2 is transferred over the entire surface of the plate 6.

In the projection optical systems 33a to 33d, the thickness of the optical path length correction member may be considered with reference to the center line of the mask 2 in the Y direction. As described above, according to the present invention, the magnification can be adjusted even when there are two or more projection optical systems arranged in the scanning orthogonal direction.

[0064]

As described above, according to the present invention, it is possible to provide an exposure apparatus capable of executing magnification correction with a simple structure. Further, by disposing optical path length varying means for varying the optical path length in the plurality of projection optical systems, it is possible to excellently correct a focus error that occurs during magnification correction due to bending.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of an exemplary embodiment according to the present invention.

FIG. 2 is a diagram showing an example of a load applied by a mask deforming unit.

FIG. 3 is a diagram illustrating the relationship between the amount of bending and the amount of contraction of a mask.

FIG. 4 is a diagram showing a configuration of an optical path length correction member.

FIG. 5 is a diagram showing the configuration of another embodiment according to the present invention.

FIG. 6 is a perspective view showing the configuration of still another embodiment according to the present invention.

FIG. 7 is a diagram showing an example of a configuration of a projection optical system.

FIG. 8 is a plan view showing a field of view by a projection optical system.

[Explanation of symbols]

 2 ... Mask, 3a-3e ... Projection optical system, 4a-4e ... Optical path length correction member, 6 ... Plate, 13 ... Mask deformation part,

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Office reference number FI technical display location 7352-4M 515 D 7352-4M 518 7352-4M 526 B

Claims (2)

[Claims] Claim: What is claimed is: 1. While moving a first object and a second object, an image of the first object is transferred to the second object via a projection optical system.
In the exposure apparatus for performing projection exposure on the object, the apparatus includes means for bending at least one of the first object and the second object so as to have a curvature in a direction substantially orthogonal to the moving direction. The exposure apparatus is characterized in that the projection optical system has a plurality of erecting optical systems that form an erecting image of the first object. 2. The exposure apparatus according to claim 1, wherein the erecting optical system has an optical path length varying means for varying an optical path length of the erecting optical system.