JPH09244254A - Exposure device for liquid crystal - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an exposure device for liquid crystals characterized by a small size, lightweightness, low cost, high resolution and wide field. SOLUTION: A front stage microlens array 10 is arranged behind a mask 1 recorded with patterns to be transferred to a wafer 2, by which inverted and unmagnified pattern images 3 are formed at every area rotated by 180 deg.C with each arranging area of this front stage microlens array 10. A rear stage microlens array 15 consisting of the same arrangement as the arrangement of the front stage microlens array 10 is arranged behind these inverted and unmagnified pattern images 3 of every area, by which erecting unmagnified images 4 of partial patterns are formed on the wafer 2. The mask 1 and the wafer 2 as well as the front stage and rear stage microlens arrays 10, 15 are scanned relatively toward the row direction x of the patterns 1a and the scanning is repeated by relatively shifting the scan in the column direction y of the patterns, by which the patterns 1a are transferred onto the wafer 2.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a liquid crystal exposure apparatus.

[0002]

2. Description of the Related Art As a prior art of a liquid crystal exposure apparatus which requires a wide exposure area, a proximity type has been commercialized. According to this technique, as shown in FIG. 11, the mask 1 and the wafer 2 are brought into close contact with each other at a minute interval, and the pattern recorded on the mask 1 is magnified to the resist 2a on the wafer 2.
Is transferred to According to this technique, a wide exposure area having the same size as the mask 1 can be secured, but the adhesion between the mask 1 and the wafer 2 is limited.

In addition to the above-mentioned conventional example, there is also an apparatus of a type in which a pattern on a mask is projected on a wafer by an image forming system of the same size and the exposure area is expanded by scanning. For example,
There are Offner type, Dyson type, and Double Dyson type. The Offner type uses a catadioptric imaging system, and the Dyson type and the double Dyson type use a catadioptric imaging system. For example, FIG. 12 shows an Offner type device, in which the mask 1 is irradiated with linear illumination light, the light from the mask 1 is bent by a mirror 50, reflected by a concave mirror 51 and a convex mirror 52, and then bent again by a mirror 53. An image is formed on the wafer 2 and the mask 1 and the wafer 2 and the image forming optical system are moved relatively to each other to perform scanning so as to cover the entire area of the pattern on the mask 1. The effective diameter of the optical element of these imaging systems was as large as several hundreds of millimeters. Also, the optical system had only one optical axis.

[0004]

In the proximity type apparatus, the adhesion between the mask 1 and the wafer 2 is low, and the line width is widened by diffraction, so that a high resolution cannot be obtained. Further, in the unity-magnification imaging scan type apparatus, although a high resolution can be obtained by the imaging system, the imaging system is large and expensive. Further, due to the aberration of the image forming system, there is a limit to widening the exposure area. That is, with the conventional technology, it was not possible to construct a small-sized, lightweight, low-cost, high-resolution, wide-field liquid crystal exposure apparatus. Therefore, the present invention is
An object is to provide a light-weight, low-cost, high-resolution, wide-field liquid crystal exposure apparatus.

[0005]

The present invention solves the above problems by using a micro lens array (hereinafter referred to as MLA) as an imaging element. MLA is an imaging element in which a large number of element lenses are two-dimensionally arranged. That is, according to the present invention, by arranging the front-stage micro lens array behind the mask on which the pattern to be transferred onto the wafer is recorded, the front-stage micro lens array is rotated by 180 ° for each arrayed region. An inverted equal-magnification partial pattern image is formed.
By arranging a post-stage micro lens array having the same arrangement as that of the lens array, an erecting equal-magnification image of the partial pattern is formed on the wafer, and the mask and the wafer and the pre-stage and post-stage micro lenses are formed. The array is relatively scanned in the row direction of the pattern, and the mask and the wafer and the front and rear micro lens arrays are relatively shifted in the column direction of the pattern to perform scanning. This is an exposure apparatus for liquid crystal that repeats to transfer a pattern onto a wafer.

The principle of the present invention will be described with reference to FIG. In the same figure, an erecting mask image of equal magnification is made by the synthetic image forming system by MLA, and the case where there are five MLA element lenses is shown. First, an inverted 1 × image of the mask 1 is created by the front stage MLA 10. This inverted equal-magnification image is xy that is orthogonal to the optical axis z for each array area of the front stage MLA 10.
Inverted equal-magnification image 3 for each area rotated 180 ° in the plane. The inverted equal-magnification image 3 for each area is converted into an equal-magnification erect image 4 for each area by the rear stage MLA 15 having the same arrangement as the arrangement of the preceding stage MLA 10.
The wafer 2 is placed at the position.

As shown in the figure, in the image formation by both MLAs 10 and 15, an optical axis z exists for each of the element lenses 10a and 15a constituting each MLA. Therefore, the regions of the inverted equal-magnification image 3 and the erecting equal-magnification image 4 for each region, that is, the fields of view are different for each element lens 10a and 15a, but by forming the erect image 4 by both MLAs 10 and 15, The image areas of the respective element lenses 10a and 15a are combined in a matched manner to obtain a wide field as a whole. In order to limit the areas of the inverted equal-magnification image 3 and the erecting equal-magnification image 4 for each area, it is preferable to arrange aperture arrays 5a and 5b, respectively, as shown in FIG.

According to the present invention, both MLAs 10 and 1
5 also uses zigzag scanning, so that a wider exposure area can be obtained. 2 shows the entire exposure area,
That is, M having an exposure area narrower than the pattern area 1a
A method of exposing the entire pattern area 1a by zigzag scanning using LAs 10 and 15 is shown. In the arrangement shown in the figure, the mask 1 and the wafer 2 are simultaneously scanned in the x-minus direction, and this scan is shifted in the y-direction and repeated, so that the partial pattern image 4a of the pattern region 1a on the mask 1 is sequentially transferred to the wafer. Then, the entire area 1a can be finally transferred onto the wafer 2.

However, the numerical aperture of the MLA can be manufactured up to about 0.5, so that high resolution imaging can be performed. Further, the image formation of the MLA is performed for each element lens, and the field in charge of each element lens is as small as the size of the element lens (about several hundreds μm). Therefore, the aberration generated in the element lens is extremely small. Therefore, as long as the number of element lenses is increased, a wide exposure field can be easily obtained without causing deterioration of aberration. Further, by using a scanning optical system, a wider exposure area can be obtained.

As the method for manufacturing MLA, a mask is provided on a glass substrate to form a distributed refractive lens array by ion exchange, a method of coating a glass substrate with a photoresist or the like for etching or heat melting, and a binary optical element. Fresnel lenses and the like are known, and these are lower in cost than the conventional image forming system. Therefore, by using such an MLA, a small-sized, lightweight, and inexpensive liquid crystal exposure apparatus can be configured.

Even if there is a gap between the image areas, the gap in the row direction is naturally filled by the scan in the row direction, and the gap in the column direction can also be filled by the scan. The image areas therefore do not necessarily have to be arranged in a continuous plane, that is to say they can be arranged discretely in a plane. As a configuration in which the image areas are discretely arranged in a plane, the image areas viewed from the row direction may be completely separated in the column direction and may simply be in line contact in the column direction. However, it is more preferable to arrange the image areas viewed from the row direction so as to overlap in the column direction.
Further, it is preferable that the integrated exposure amount by scanning is uniform in the column direction.

If the image areas are arranged discretely, it is also possible to carry out Koehler illumination on the area covering all of the conjugate areas on the mask with respect to the image areas. However, it is preferable to illuminate only the discrete conjugate areas on the mask for the discrete image areas discretely and uniformly within the discrete conjugate areas because stray light is prevented. Further, when illuminating only the discrete conjugate area, the discrete illumination light diaphragm array can be arranged in close contact with the mask, but it is more preferable to illuminate the mask by imaging the discrete illumination light diaphragm array. preferable.

[0013]

FIG. 3 shows an embodiment of a zigzag scan type liquid crystal exposure apparatus using an MLA. Light source 6,
The illumination optical system 7 and the imaging optical system 8 by MLA are relatively aligned and fixed. The mask 1 and the wafer 2 are held by movable stages 1b and 2b which are movable in x, y and z directions, respectively. An interferometer (not shown) is installed on each moving axis of each movable stage 1b, 2b so that the positions of the mask 1 and the wafer 2 can be measured. The light source 6 is a high pressure mercury lamp,
The light from the light source 6 is shaped by the illumination optical system 7 to illuminate the mask 1. The wavelength used is the g-line (436 nm). The mask pattern 1a is formed on the wafer 2 by the image forming optical system 8 as an equal-size erect image. The pattern area 1a of the mask 1 is 300 mm × 200 mm, and
The side of 00 mm is the x-axis and the side of 200 mm is the y-axis.

The distance between the mask 1 and the image forming optical system 8 and the distance between the wafer 2 and the image forming optical system 8 are both 70 at the image plane position.
It is set to 0 μm. A plurality of gap sensors 1c and 2c are arranged around the image forming optical system 8, and the gap between the mask 1 and the image forming optical system 8 during scanning and the gap between the wafer 2 and the image forming optical system 8 are set. Feedback is provided to a z-axis movement motor (not shown) so as to be kept accurate. In this arrangement, the mask 1 and the wafer 2 are interlocked with each other to scan at a constant speed in the x direction, and after the constant speed scan, the pattern 1 on the mask 1 is repeated by zigzag scanning.
The whole a can be transferred to the wafer 2. Details of scanning and shifting will be described later.

Next, details of the image forming optical system 8 will be described. FIG. 4 shows the structure of the imaging optical system 8 viewed from the z-axis direction. The imaging optical system 8 is supported by a metal holder 24 of 20 mm × 20 mm, and the center thereof is 10 mm × 10 m.
It is an opening of m. FIG. 5 shows the structure of the imaging optical system 8 viewed from the x direction. As shown in the figure, the imaging optical system 8
Is composed of four flat plates MLA 11, 12, 16, and 17, two aperture stop arrays 20 and 21, and one field stop array 22. These elements are the first stage flat plate MLA11, the front stage aperture stop array 20, the front stage second flat plate MLA12, the field stop array 22, the rear stage first flat plate MLA16, and the rear stage aperture stop array 21 from the mask 1 side to the wafer 2 side. , And the second-stage second flat plate MLA17 in this order. In addition, the first flat plate MLA11 and the rear first plate MLA11
The flat plate MLA16 is arranged with the substrate surface facing the wafer 1, and the front stage second flat plate MLA12 and the rear stage second flat plate MLA17 are arranged.
Are arranged with the back surface of the substrate facing the wafer 1.

The thickness t of each flat plate MLA 11, 12, 16, 17 is t = 0.9 mm, the back focus f ₁ from the substrate surface to the front side focal position is f ₁ = 700 μm, and the back focus from the substrate back surface to the rear surface side focal position. Back focus f ₂ is f ₂ =
200 μm. The thicknesses of the aperture stop arrays 20 and 21 and the field stop array 22 are both 20 μm. Mask 1, front aperture stop array 20, field stop array 2
2. The rear aperture stop array 21 and the wafer 2 are positioned by the spacer 23 so as to be positioned at the front side back focus f ₁ or the back side back focus f ₂ of the corresponding flat plate MLA 11, 12, 16, 17, respectively. And is incorporated in the metal holder 24. Further, the openings of the respective elements are aligned with each other also in the xy directions.

In this arrangement, an inverted intermediate image is formed for each area by the front stage MLA10 composed of the front stage first flat plate MLA11 and the front stage second flat plate MLA12, which is composed of the rear stage first flat plate MLA16 and the rear stage second flat plate MLA17. By forming an image again in the latter stage MLA15, an erect image of the same size is obtained. This arrangement is a telecentric imaging system.

FIG. 6 is a plan view of each flat plate MLA11,
The structure of 12, 16, 17 is shown. However, the number of element lenses 11a, 12a, 16a, 17a of each flat plate MLA is not actual as described below. When h = 70 μm, the diameter of the element lens is 6 h (= 420 μm).
The arrangement of the element lenses is determined such that the image areas of the respective element lenses viewed from the scanning direction (x direction) overlap in the shift direction (y direction). Here, 2
An element lens array in which two element lenses are arranged in the y direction with no clearance is shifted in the x direction by an element lens diameter of 6 h (= 420 μm), and the element lens radius is 3 h (= 21
22 rows were arranged by alternately shifting by 0 μm) in the y direction. At this time, each flat plate MLA11, 12,
The openings of 16 and 17 are 9.24 mm × 9.45 mm.

The flat plates MLA11, 12, 16, 17
Was manufactured by masking a glass substrate having a thickness of 1.7 mm with a circular aperture array having the same shape as in FIG. 6, and forming the distributed index lens array by ion exchange at the aperture.
The substrate has a thickness of 0.9 m after forming the distributed index lens array.
It was polished so as to be m. In addition, the wavelength of 4
An antireflection film for 36 nm was formed.

FIG. 7 shows each aperture stop array 2 viewed from the z direction.
The structure of 0,21 is shown. One aperture stop element 20a, 2
The diameter of 1a is 140 μm, and the arrangement in the xy directions is adjusted to the flat plate MLA. Here, considering that the front-side back focus of the flat plate MLA is f ₁ = 700 μm, it can be said that the numerical aperture of the imaging system is 0.1. The aperture stop arrays 20 and 21 were manufactured by etching stainless steel having a thickness of 20 μm into the shape shown in FIG. 7 and applying black alumite plating or the like having a low reflectance.

FIG. 8 shows the field stop array 22 viewed from the z direction.
The structure of is shown. The cycle in the xy direction is the same as the aperture stop array.
It is fitted to the flat plate MLA. The field stop elements 22a of the field stop array are regular hexagons inscribed in a circle having a diameter of 4h (= 280 μm), and are arranged so that one side is parallel to the y direction, as shown in FIG. This field stop array 22
Was manufactured by etching stainless steel having a thickness of 20 μm into the shape shown in FIG. 8 and applying black alumite plating or the like having a low reflectance.

The field stop element 22a has a diameter of 4h.
Since it is a regular hexagon inscribed in a circle of (= 280 μm) and one side is parallel to the y direction, the central rectangular portion A
Has a height of 2h (= 140μm), and the upper and lower isosceles 3
The height of each of the corners B _U and B _L is h (= 70 μm). On the other hand, a field stop element 22b adjacent to a certain field stop element 22a in the x direction is 3h (= 210) in the y direction.
Since the two field stop elements 22a and 22b are viewed from the x direction, the central rectangular portion A does not overlap in the y direction, but the upper and lower isosceles triangles B _U and B are arranged. _{For L} , the upper 2 isosceles 3 of one element 22a
The corner B _U and the lower isosceles three corner B _{L of} the other element 22b will just overlap. Moreover, the sum of the lengths of both isosceles triangles B _U and B _{L in} the x direction when viewed at the same height y is exactly the width of the rectangular portion A in the x direction at any height y. Match.

By adopting such a structure, the rectangular portion A having no overlap and the overlapped portion 2
The line integration apertures in the x direction of the equilateral triangular portion B become equal, and therefore the fields (the field of view has the same shape as the field diaphragm array 22) of the regions A and B when scanned at a constant speed in the x direction. The integrated exposure dose can be made equal. That is, by using such a field stop array 22, the entire image can be smoothly joined also in the y direction orthogonal to the scanning direction.

The size K of the visual field in the y direction of the field stop array 22 smoothly joined in the y direction as described above.
_y is 9.17 mm, and the visual field width K _{x in the} x direction is 9.17 mm.
It is 06 mm. Here Synthesis field K in the y-direction _y is the area of the rectangular portion A, 3 corners a pair of isosceles B _U, the total length of the region B consisting of B _L, i.e., an imaging optical system 8a y It is defined as a region where the exposure amount is equal even when used without shifting in the direction. As actually shown in Figure 8, there is an upper junction region C _U of height h decreases the exposure amount is linearly on the upper side of the synthetic field K _y (= 70μm), the bottom side of the composite field K _y Also, the height h (= 7
0 μm) lower junction region C _L.

That is, the upper junction region C _U and the lower junction region C _L are located at the upper and lower ends of the field stop array 22, and when they are not shifted in the y direction, there is no overlap and the exposure dose is directed upward and downward. It is a region that decreases linearly. In this embodiment, zigzag scanning is performed while shifting in the y direction so that the upper junction region C _U and the lower junction region C _L sequentially overlap to cover the entire exposure region. Details will be described later.

Next, the illumination optical system will be described. FIG. 9 shows the structure of the illumination optical system. The illumination optical system 7 corresponds to the arrangement of the imaging optical system 8 of FIG. 4, and has a structure that illuminates only a substantial exposure area of the imaging optical system 8. ing. In FIG. 9, after shaping the light from the light source 6 by the high-pressure mercury lamp, the light is expanded by the beam expander 30 and the fly eye
The light enters the lens array 31, and the condenser lens 33 illuminates the illumination light diaphragm array 34. A spatial coherence diaphragm 32 is arranged at the focal position of the fly-eye lens array 31, and the spatial coherence of the illumination light can be adjusted by adjusting the size thereof. This is a general Koehler illumination system used in semiconductor manufacturing equipment and the like, and the light intensity on the illumination light diaphragm array 34 has high uniformity. Where the fly eye
The lens array 31 has 10 × 10 elements, and the size of the element lens is 4 mm square and the focal length is 8 mm. The focal length of the condenser lens 33 is 18.6 mm. Therefore, about 9.3 mm on the illumination light diaphragm array 34.
An illumination area of × 9.3 mm is obtained.

The illumination light diaphragm array 34 is an MLA of the image forming system.
The same structure as the field stop array 22 used in FIG. The direction of the illumination light diaphragm array 34 is the MLA of the imaging system.
The field stop array 22 is aligned in the same direction. The illumination light diaphragm array 34 includes a first flat plate ML for illumination.
The mask 1 is formed by the A35 and the second flat plate MLA36 for illumination.
It is imaged on the surface. The first illumination flat plate MLA35 and the second illumination flat plate MLA36 are the first stage flat plate ML of the imaging optical system 8.
It is the same as that used as A11 and the former second flat plate MLA12, or the latter first flat plate MLA16 and the latter second flat plate MLA17. With the above optical system, an illumination light array having the same shape as the field stop array 22 of the imaging optical system 8 is generated on the mask 1. That is, the illumination optical system 7 has a structure for selectively illuminating the field of view of the MLA of the imaging optical system 8. The illumination flat plates MLA35, 3
It is desirable that the image forming system according to 6 be capable of finely adjusting the magnification so that the size of the region of the elements of the illumination light array can be adjusted to an appropriate value. The fine adjustment of the magnification can be performed, for example, by providing the flat plates MLA 35 and 36 with a mechanism capable of moving in parallel in the z direction while keeping the distance therebetween.

After all, the image forming optical system 8 by the above-mentioned MLA.
The field of view and the illumination light array of the illumination optical system 7 are both shaped as shown in FIG. 8, and both are accurately aligned, and in that state, the mask 1 and the wafer 2 are interlocked to perform zigzag scanning. By this, mask pattern 1
The whole a can be transferred to the wafer 2. Zigzag scanning refers to repetition of constant speed scanning and constant width shift. In this embodiment, constant speed scanning is performed in the x direction and constant width shift is performed in the y direction. FIG. 10 shows this state. However, in FIG. 10, the mask 1 and the wafer 2 are
Is fixed and the illumination optical system 7 and the imaging optical system 8 are drawn as if they were moved. However, since the movements of the mask 1 and the wafer 2 are relative movements with respect to the illumination optical system 7 and the imaging optical system 8, the illumination optical system 7 and the imaging optical system 8 are fixed in FIG. Is equivalent to the exposure area when is moved.

In FIG. 10, starting from P ₁ , a constant velocity scan of 310 mm is performed in the x direction up to Q ₁ , and then a constant width shift of 9.24 mm is performed in the y direction from Q ₁ to Q ₂ .
Then again subjected to constant-speed scanning in the x direction from Q ₂ to P _2, in the order of ..., total up to P ₂₂ 2
300 mm by performing two constant speed scans
The pattern area 1a of × 200 mm is covered. At this time, each constant speed scan is performed in the upper junction region C _U of FIG.
And the lower junction region C _L are overlapped with each other. This is the constant width shift amount of each time, y of MLA
In the combined visual field K _y (= 9.17 mm) in the direction, the height h (= 70 μm) of the upper bonding region C _U or the lower bonding region C _L.
This can be done by adding 9.24 mm.
By doing so, each point in the total combined visual field after scanning has an equal exposure amount.

At this time, the lower junction area C _L in the first constant speed scan and the upper junction area C _U in the 22nd (final) constant speed scan are underexposed areas that never overlap. Therefore, this area needs to be a non-use area during the entire exposure. Eventually, the maximum total combined visual field K ′ _y in the y direction that can be obtained by 22 constant-speed scans is 22 times K _y and the junction region C is
_U, a 203.21mm obtained by adding 21 times the height h of the C _L. Further, the maximum combined visual field K ′ _x in the x direction is the visual field width K _{x in} the x direction of the MLA from the constant speed scan distance of 310 mm.
Is calculated to be 300.94 mm. In this way, it was possible to cover the pattern area of 300 mm × 200 mm.

In the above, the scanning method assuming that the wafer 2 does not expand or contract has been described. Hereinafter, a correction scanning method that can cope with expansion and contraction of the wafer will be described. The expansion coefficient of the wafer is β (β>
Let's say that it is expanded when it is 0), and consider the case where it is expanded. If the wafer expands as described above, the pattern shift will occur if the mask and the wafer are scanned in perfect synchronization. In order to prevent this, the zigzag scanning may be performed by setting the constant scanning speed and the constant width shift amount of the mask and the wafer to be completely the same, but different values by the expansion coefficient β.

Considering the case where the length of the mask in the constant velocity scanning direction is L _M and the distance is the scanning velocity v _M , and the constant velocity scanning is performed in t seconds, L _M = v _M · t. _Assuming that the expansion coefficient of the wafer is β, the length L _W of the region on the wafer corresponding to the distance L _M becomes long as L _W = (1 + β) · L _M = (1 + β) · v _M · t. ing. Therefore, if the scanning speed v _{W of the} wafer is v _W = (1 + β) · v _M with respect to the scanning speed v _{M of the} mask, the scanning of the corresponding region of the mask and the wafer can be performed during the moving time t. Will be over. That is, constant speed scanning can be performed while maintaining the correspondence between the mask pattern and the wafer.

The same applies to the constant width shift. When the mask is subjected to a constant width shift by Δ _M (= K _y + h), if the constant width shift amount of the wafer is set as Δ _W = (1 + β) · Δ _M , the pattern of the area corresponding to the mask and the wafer The deviation can be corrected. However, in order to correct this pattern shift in the constant width shift direction, it is necessary that the visual field K _{y in} the y direction of the MLA is appropriately small. This is because the field of view in the y direction is determined as K _y in each constant speed shift, so β · K _y (≈β ·) in the y direction around the field of view.
This is because the pattern shift of Δ _M ) occurs. Since the pattern shift amount has an allowable limit, it is necessary to make the field of view K _y small so as not to exceed the limit.

For example, the minimum line width of the pattern to be transferred is 3
μm, the allowable amount of pattern deviation is 10%, 0.3μ
If it is m or less, β · K _y <0.3 μm, and if β = 20 ppm, the upper limit of the visual field K _{y in} the y direction is determined as K _y <15 mm. In this example, K _y = 9.17 mm, so this condition is satisfied. Also in other cases, pay attention to the allowable limit of the pattern shift amount.
_y must be set. It should be noted that since the scanning is continuously performed in the x direction, a problem is unlikely to occur, but the visual field width K
It goes without saying that _x should not be too large.

As described above, the pattern area is set to 300 in this embodiment.
Although the size is set to mm × 200 mm, it is possible to expose a wider exposure area by increasing the constant velocity scan distance in the x direction and increasing the number of constant width shifts in the y direction. . Further, in order to reduce the time required for the entire zigzag scanning, it is preferable that the long side of the mask pattern is the scanning direction and the short side is the shift direction.

[0036]

As described above, according to the present invention, a high resolution large screen liquid crystal exposure apparatus can be constructed in a small size, a light weight and a low cost.

[Brief description of drawings]

FIG. 1 is a sectional view showing the principle of projection of a mask pattern according to the present invention.

FIG. 2 is a perspective view showing a zigzag scan according to the present invention.

FIG. 3 is a perspective view showing the entire exposure apparatus.

FIG. 4 is a plan view showing an imaging optical system.

FIG. 5 is a sectional view showing an imaging optical system.

FIG. 6 is a plan view showing an MLA of the image forming optical system.

FIG. 7 is a plan view showing an aperture stop array of the imaging optical system.

FIG. 8 is a plan view showing a field stop array of the imaging optical system and an illumination light stop array of the illumination optical system.

FIG. 9 is a sectional view showing an illumination optical system.

FIG. 10 is a plan view showing a method of zigzag scanning.

FIG. 11 is a sectional view showing a conventional technique.

FIG. 12 is a sectional view showing another conventional technique.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Mask 1a ... Pattern area 2 ... Wafer 2a ... Resist 1b, 2b ... Movable stage 1c, 2c ... Gap sensor 3 ... Inverted equal-magnification image for each area 4 ... Equal-magnification erect image 4a ... Partial pattern image 5a, 5b ... Aperture array 6 ... Light source 7 ... Illumination optical system 8 ... Imaging optical system 10 ... Previous stage MLA 11 ... Previous stage first flat plate MLA 12 ... Previous stage second flat plate MLA 10a, 11a, 1
2a ... Element lens 15 ... Rear stage MLA 16 ... Rear stage first flat plate MLA 17 ... Rear stage second flat plate MLA 15a, 16a, 1
7a ... Element lens 20 ... Front aperture stop array 21 ... Rear aperture stop array 20a, 21a ... Aperture stop element 22 ... Field stop array 22a, 22b ... Field stop element 23 ... Spacer 24 ... Metal holder 30 ... Beam expander 31 ... Fly Eye
Lens array 32 ... Space coherence diaphragm 33 ... Condenser lens 34 ... Illumination light diaphragm array 34a, 34b ... Illumination light diaphragm element 35 ... Illumination first flat plate MLA 36 ... Illumination second flat plate MLA

Claims (11)

[Claims] 1. A front micro-lens array is arranged behind a mask on which a pattern to be transferred to a wafer is recorded, whereby each pre-micro-lens array is rotated by 180 ° for each arrayed area. Forming an inverted equal-magnification partial pattern image, and arranging a rear microlens array having the same arrangement as that of the preceding microlens array behind the inverted equal-magnification partial pattern image for each area. An erecting equal-magnification image of the partial pattern is formed on the wafer by means of the mask and the wafer, and the front and rear micro.
The lens array is relatively scanned in the row direction of the pattern, and the mask and the wafer and the front and rear micro
A liquid crystal exposure apparatus that transfers the pattern onto a wafer by relatively shifting the lens array in the column direction of the pattern and repeating the scan. 2. The liquid crystal exposure apparatus according to claim 1, wherein each element lens of the front-stage and rear-stage micro lens arrays is configured to be telecentric on both sides. 3. The liquid crystal exposure apparatus according to claim 1, wherein an exposure field of the mask, which is temporarily imaged on the wafer, is illuminated with a uniform intensity. 4. A field stop array having the same arrangement as the front and rear micro lens arrays is arranged at an image forming position of the inverted equal-magnification partial pattern image for each area. LCD exposure equipment. 5. The field stop arrays are arranged so as to be continuous while overlapping in the column direction when viewed from the row direction, and the field stop arrays overlap in the column direction when viewed from the row direction. The field stop array and the shift amount are set such that the shift amount is set so as to be continuous while the integrated exposure amount due to the scan is uniform in the column direction. 4. The liquid crystal exposure apparatus according to item 4. 6. The liquid crystal exposure apparatus according to claim 4, wherein the shape of each field stop element of the field stop array is a regular hexagon having sides parallel to the column direction. 7. The area of the mask, which is conjugate with the individual field stop elements of the field stop array with respect to the preceding microlens array, is illuminated with a uniform intensity. The exposure apparatus for a liquid crystal described. 8. An illumination light diaphragm array having the same arrangement as the field diaphragm array is prepared, and the illumination light diaphragm array is illuminated with a uniform intensity. The liquid crystal according to claim 7, wherein an illumination micro lens array having the same arrangement as the front and rear micro lens arrays is arranged to illuminate a region of the mask which is conjugate with the field stop element. Exposure equipment. 9. A front gap sensor for detecting a distance between the mask and the front micro lens array, and a rear gap sensor for detecting a distance between the rear micro lens array and a wafer are provided. Based on the output of the rear gap sensor, the mask and front micro lens
9. The liquid crystal exposure apparatus according to claim 1, wherein the distance between the liquid crystal and the array and the distance between the latter micro lens array and the wafer are controlled to be constant. 10. The method according to claim 1, wherein the change in the length of the wafer is absorbed by adjusting the scan speed and the shift amount according to the change in the length of the wafer from the reference state. The exposure apparatus for liquid crystal according to any one of claims 1. 11. The liquid crystal exposure apparatus according to claim 1, wherein the long side of the pattern recorded on the mask is the scanning direction and the short side is the shift direction.