JP3316754B2 - Exposure method and apparatus, and element manufacturing method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a projection optical apparatus used for manufacturing, for example, a semiconductor integrated circuit or a liquid crystal display element by a lithography process, and more particularly to so-called stitching and slit scanning using a light source which emits pulsed light. The present invention relates to a projection optical apparatus suitable for being applied to a projection exposure apparatus that performs exposure by an exposure method.

[0002]

2. Description of the Related Art When a semiconductor device or a liquid crystal display device is manufactured by a lithography process, a pattern image of a photomask or a reticle (hereinafter, collectively referred to as a "reticle") is exposed through a projection optical system under exposure light. 2. Description of the Related Art A projection exposure apparatus that projects an image on a photosensitive substrate is used. As such an apparatus, for example, a projection exposure apparatus of a stitching and slit scan exposure system is known as disclosed in Japanese Patent Publication No. 46-34057. In the stitching and slit scan exposure method, the reticle and the photosensitive substrate are synchronously scanned in a predetermined first direction relative to an illumination area of a predetermined shape on the reticle, so that the first surface on the photosensitive substrate is scanned. Exposure is performed on the region in the column.

Thereafter, the reticle is replaced or the reticle is moved to a second position perpendicular to the first direction of the illumination area.
, The photosensitive substrate is laterally shifted (stitched) in a direction conjugate to the second direction of the illumination area. Then, the reticle and the photosensitive substrate are synchronously scanned in the first direction again with respect to the illumination region having the predetermined shape on the reticle, thereby exposing the second column region on the photosensitive substrate. Will be Thus, the reticle pattern can be exposed to an area on the photosensitive substrate that is wider than the exposure field of the projection optical system.

FIG. 10 (a) shows an illumination area on a reticle in a conventional stitching and slit scan exposure type projection exposure apparatus. In FIG. 10 (a), a regular hexagonal shape having a position A as a center is shown. Exposure light from the illumination optical system is radiated to the illumination area 1. Further, by scanning the reticle at a constant speed V in the −X direction with respect to the illumination area 1 at the position A, the illumination area 1 moves along the trajectory 2 on the reticle.
It relatively moves along A and reaches position B. By moving the reticle in the Y direction in this state, the illumination area 1
Moves relatively along the trajectory 2B on the reticle to reach the position C. After that, the reticle is scanned at a constant speed V in the X direction, so that the illumination area 1 moves along the trajectory 2 on the reticle.
It moves relatively along C.

FIG. 10B shows a region to be exposed on a wafer as a photosensitive substrate in a conventional stitching and slit scan exposure type projection exposure apparatus.
At 0 (b), a regular hexagonal exposure region 3 centered on the position AP is a region conjugate with the illumination region 1 at the position A on the reticle. The regular hexagonal exposure region 3 has two sides parallel to the Y direction. If the interval between two opposing vertices of the regular hexagonal exposure region 3 is R, and the interval between two opposing sides is W,
W = 3 ^1/2 R / 2. Further, by setting the projection magnification from the reticle to the wafer by the projection optical system as β, the wafer is scanned in the X direction at a constant speed β · V with respect to the exposure area 3 at the position AP, so that the exposure area 3 2AP
And relatively moves along the line to reach the position BP. In this state, by moving the wafer by the distance 3R / 4 in the −Y direction, the exposure area 3 relatively moves along the trajectory 2BP on the wafer and reaches the position CP. This operation is stitching. Thereafter, the wafer is moved at a constant speed β · in the −X direction.
By scanning with V, the exposure region 3 relatively moves along the trajectory 2CP on the wafer.

[0006] The exposure area 3 relatively moving along the locus 2AP and the exposure area 3 relatively moving along the locus 2CP are each a connection area having an isosceles triangle with a width of R / 4 in the Y direction. 4 are scanned so as to overlap. Therefore,
In the connection region 4, exposure is performed twice. The connection region 4 is provided in such a manner that the position exposed between the pattern exposed by the exposure region 3 relatively moving along the trajectory 2AP and the pattern exposed by the exposure region 3 relatively moved along the trajectory 2CP is determined. This is to prevent a shift from occurring. Further, by making the isosceles triangular regions of the regular hexagonal exposure region 3 overlap, the illuminance distribution on the wafer is made uniform as shown below.

Conventionally, a light source of continuous light such as a mercury lamp is generally used as a light source of the exposure light. Therefore, the exposure point P1 in the connection area 4 on the wafer is shifted to the exposure area relatively moving along the locus 2AP. 3 is continuously exposed in the region 5A, and is continuously exposed in the region 6A of the exposure region 3 relatively moving along the trajectory 2CP. Area 5A and Area 6
The total length of A in the X direction is equal to the width W of the exposure region 3.
Further, another exposure point P2 in the connection area 4 on the wafer is continuously exposed in the area 5B of the exposure area 3 relatively moving along the trajectory 2AP, and the other exposure point P2 in the exposure area 3 relatively moving along the trajectory 2CP. Continuous exposure is performed in the area 6B,
The sum of the lengths of B and the region 6B in the X direction is equal to W. In addition, at the exposure point P0 of the non-connection portion exposed only to the exposure region 3 relatively moving along the trajectory 2AP, the exposure is continuously performed in the region of the width W in the X direction. Therefore, when a continuous light source is used, which exposure point P0, P
The amount of exposure light to be irradiated is the same even in 1 and P2, and the illuminance distribution is uniform.

[0008]

Recently, in order to further improve the resolving power, a shorter wavelength of the exposure light is required. Among light sources that are currently in practical use, those having a relatively short wavelength are ArF excimer lasers (wavelength: 193 n).
m), an excimer laser such as a KrF excimer laser (wavelength: 248 nm) and a metal vapor laser. However, since the excimer laser light source and the metal vapor laser light source are of the pulse emission (pulse oscillation) type, their use requires different considerations from those of a continuous emission light source such as a mercury lamp.

FIG. 11A shows a case where a regular hexagonal exposure region 3 is illuminated with a pulse laser beam from a pulse laser light source. In FIG. This is an area inscribed in the outline of the circular exposure area 7 of the system. Further, it is assumed that two opposite sides of the exposure region 3 whose width W is parallel to the Y direction, and the wafer is relatively scanned in the X direction and the −X direction with respect to the exposure region 3. In this case, it is necessary to expose a plurality of pulses of the pulse laser light to each exposure point on the wafer in order to reduce the influence of the variation in energy of each pulse of the pulse laser light, the speckle, and the like. Therefore, it is assumed that the exposure point P0 exposed by the region having the width W in the X direction of the exposure region 3 is exposed m times (m is an integer of 1 or more) by the pulsed laser beam. This is defined as ΔL where the distance that the wafer is scanned in the X direction or the −X direction during one cycle T of pulsed light emission is ΔL.
Assuming that the scanning speed of the wafer is β · V, the following relationship may be satisfied.

## EQU1 ## W = m..DELTA.L = m.T..beta..V

FIG. 11A shows a case where m = 8, and the exposure point P0 is set to the exposure area 3 at the time when the pulse emission is performed.
The exposure point P0 is exposed to the pulse laser light seven times inside the exposure area 3 and twice exposed to the pulse laser light at the edge part. In this case, since the energy exposed at the edge portion is ２ of the energy exposed inside, the exposure point P0 is irradiated with energy for a total of eight pulses. Then, no matter where the exposure point P0 is in the X direction at the time of pulse emission, the exposure point P0 is irradiated with energy for a total of eight pulses.
In addition, among the exposure points through which the isosceles triangular area 3a on the right side of the exposure area 3 passes, the exposure points P1 to P1 shown in FIG.
The distance that P8 passes through the isosceles triangular region 3a is 8 · ΔL to 1 · ΔL. Therefore, when the exposure points P1 to P8 pass through the region 3a in the X direction, they are irradiated with energy for 8 to 1 pulses, respectively.

Then, after the wafer is stitched, and the wafer is scanned in the -X direction with respect to the exposure area 3, the exposure points P1 to P8 are substantially zero-pulse to 0 respectively.
The energy for 7 pulses is exposed. Therefore, at the exposure points P1 to P8, the energy corresponding to eight pulses is exposed similarly to the exposure point P0 by performing the slit scan exposure twice by stitching. However, if the intermediate exposure point between the exposure points P4 and P5 in FIG. 11A is the exposure point P9 in FIGS. 11B and 11C, two slit scan exposures are performed at the exposure point P9. Even if it is performed, there is a disadvantage that irradiation energy varies. That is, in the case shown in FIG. 11B, when the wafer is scanned in the X direction within the isosceles triangular area 3a on the right side of the exposure area 3, when the exposure point P9 is at the position 8, the pulse Light emission is performed, and when the wafer is scanned in the X direction within the left isosceles triangular area 3b of the exposure area 3 after the stitching, pulse emission is performed when the exposure point P9 is at the position 9. Therefore, the exposure point P9 is irradiated with energy for 9 pulses.

On the other hand, in the case shown in FIG. 11 (c), the wafer is placed in an isosceles triangular area 3a on the right side of the exposure area 3 by X.
When scanning in the direction, pulse emission is performed when the exposure point P9 is at the position 10, and when the wafer is scanned in the X direction within the left isosceles triangular area 3b of the exposure area 3 after stitching. Then, when the exposure point P9 is at the position 11, pulse emission is performed. Therefore, the exposure point P9 is irradiated with energy for seven pulses. Therefore, the exposure point P9 is irradiated with energy for 7 to 9 pulses depending on the timing of pulse emission. Therefore, there is a disadvantage that unevenness of the irradiation energy by the pulse laser beam, that is, uneven illuminance occurs at the connection portion 4 on the wafer.

In view of the above, the present invention provides an exposure method or an element manufacturing method in which exposure is performed by a stitching exposure method, wherein uneven illuminance of a connection portion exposed twice by stitching on a substrate is reduced. The purpose is to reduce. Further, the present invention
Easily adapts its stitching exposure method.
An object of the present invention is to provide a usable exposure apparatus.

[0014]

According to the first aspect of the present invention, there is provided
The light method uses a first pattern to form a first light on the substrate (28).
Exposing a region (40a) and, after said exposing, a first region
The second region (40) shifted in a predetermined direction (Y direction) with respect to
b) by the stitching method using the second pattern
In the exposure method for exposing, the first pattern is irradiated.
At one end of the illuminance distribution of the exposure light in the predetermined direction.
Forming an area (20a) where the illuminance gradually decreases,
Illuminance in the predetermined direction of exposure light applied to two patterns
At the other end of the cloth, an area (20) where the illuminance gradually decreases
b) and forming the first region on the substrate (28) with the first region.
The overlapping portion (40c) with the second region is defined as one of the illuminance distributions.
The area (20a) formed on the end side and the other end of the illuminance distribution
To expose twice with the region (20b) formed on the side
did.

In the exposure method according to the present invention, the exposure
In the illuminance distribution of light, a region where the illuminance gradually decreases (20a,
20b) to arrange the field stop in a defocused state
It is formed by. Further, the dew of claim 3
In the light method, the illuminance of the illuminance distribution of the exposure light gradually decreases.
In the lower regions (20a, 20b), the transmittance distribution changes.
By disposing the ND filter in the optical path of the exposure light.
Is formed.

Further, in the exposure method according to the present invention, the light
The degree distribution is such that the length in the predetermined direction is L and the illuminance is almost constant.
The area and the length in the predetermined direction on both sides thereof are M and the illuminance is
A pattern of the mask (19) having a region which is first reduced;
If the width of the region in the predetermined direction is LT, one or more
The condition of the following formula is satisfied using an integer n
You.

LT = n · L + (n−1) · M Further , according to the ninth aspect of the present invention, there is provided a device manufacturing method comprising the steps of:
It is used.

Further, in the exposure apparatus according to the present invention, the first
A first region (40a) on a substrate (28) using a pattern;
Is exposed, and after the exposure, a predetermined direction
The second region (40b) shifted in the direction (Y direction) is
Exposure system that exposes by stitching method using
Can move in the predetermined direction while holding the substrate
Substrate stage (29, 30) and its first pattern
One end of the illuminance distribution of the irradiated exposure light in a predetermined direction is
To form an area (20a) where the illuminance of the
A predetermined direction of the exposure light applied to the second pattern
At the other end of the illuminance distribution, where the illuminance gradually decreases
Optical members (13, 15) forming (20b),
The overlapping portion (40c) of the first region and the second region on the substrate is
A region (20a) formed at one end of the illuminance distribution and its
And the region (20b) formed on the other end side of the illuminance distribution of FIG.
Position of the substrate stage (29, 30) so as to be exposed twice
And a control system (25) for controlling the
Was.

In the exposure apparatus according to the present invention, the predetermined pattern
Exposing the substrate to form a pattern on the substrate (28)
In the exposure apparatus, the mask on which the first pattern is formed is
For holding the mask and the mask on which the second pattern is formed.
Stage and a substrate stage (2
9, 30) and running the substrate using the first pattern.
Mask stage and substrate stage
Move in synchronism with the
The mask stay to scan and expose the substrate
Control system that synchronizes the stage with its substrate stage
(25). In addition, claim 16
The element manufacturing method uses the exposure apparatus of the present invention.
You.

[0019]

According to the exposure method of the present invention, the first
The first region and the second pattern that are exposed using the pattern
The overlapping portion (40c) with the second region exposed by using
In areas (20a, 20b) where the illuminance of light gradually decreases
Since the exposure is performed twice, non-overlapping areas
It is possible to reduce illuminance unevenness as in the case of the section. Ma
According to the exposure method of claim 2 or 3,
Easily create regions where the illuminance gradually decreases in the illuminance distribution of light
can do. Further, according to the exposure method of claim 6,
By satisfying the above (Equation 2), the mask
Illumination on the board even when using the end of the turn area
Can be reduced. Further, according to the exposure apparatus of claim 10,
If so, the first region exposed using the first pattern and the second region
An overlapping portion (40) with the second region exposed using the pattern
c) is defined as a region where the illuminance of the exposure light gradually decreases (20a,
Since the exposure was performed twice in step 20b),
Can reduce uneven illuminance as well as non-overlapping parts.
Become. According to the exposure apparatus of claim 15, the present invention
Can be carried out.

[0020]

[0021]

[0022]

[0023]

[0024]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of a projection optical apparatus according to the present invention will be described below with reference to FIGS. In the present embodiment, the present invention is applied to a projection exposure apparatus of a stitching and slit scan exposure type provided with a pulse light emission type laser light source. FIG. 1 shows the overall configuration of the projection exposure apparatus of the present embodiment. In FIG. 1, a laser beam LB emitted from a pulse laser light source 12 such as an excimer laser light source includes a beam expander, an optical integrator, an aperture stop, The light enters the illumination optimization optical system 13 including a relay lens and the like. Lighting optimization optical system 13
Laser light IL as exposure light emitted from
Is reflected by the deflection mirror 14 and enters the field stop 15. Pulsed laser light I passing through the aperture of the field stop 15
L illuminates the reticle 19 with uniform illuminance via the relay lens 16, the deflection mirror 17, and the condenser lens 18. The arrangement surface of the field stop 15 is conjugate with the pattern formation surface of the reticle 19, and the shape of the slit-shaped illumination area 20 on the pattern formation surface of the reticle 19 is set by the opening of the field stop 15.

The reticle 19 is held on the reticle stage 21, and the reticle stage 21 is moved in the X direction (the left-right direction in the plane of FIG. 1) and the Y direction (the direction perpendicular to the plane of FIG. 1).
A reticle stage 21 and a movable mirror 22 are supported along a guide 23 so as to be movable in an XY plane and at a constant speed in the X direction. The reticle stage 21 is connected to a drive device 26 for performing movements in the X and Y directions and a minute rotation for yawing correction. Guide 23
The laser beam from the laser interferometer 24 fixed to the mirror is reflected by the moving mirror 22, and the position of the reticle 19 in the X and Y directions and the amount of yawing are constantly measured by the laser interferometer 24. It is supplied to the main control system 25. The main control system 25 includes the driving device 2
6, the operation of the reticle 19 is controlled, and the light emission operation of the pulse laser light source 12 is controlled via the laser light source control device 34.

The pulsed laser light IL that has passed through the pattern of the reticle 19 passes through the projection optical
The pattern image of the reticle 19 in the illumination area 20 is reduced at a predetermined projection magnification β and formed on the exposure area 20P on the wafer 28 conjugated with the illumination area 20. The wafer 28 is held on a micro-rotatable wafer holder 29, and the wafer holder 29 is fixed on a wafer stage 30. The wafer stage 30 positions the wafer 28 in a two-dimensional plane including the X direction and the Y direction.
It comprises a Y stage, a Z stage for positioning the wafer 28 in the Z direction parallel to the optical axis of the projection optical system 27, and the like. Laser interferometer 32 on wafer stage 30
The movable mirror 31 for reflecting the laser beam from the laser 28 is fixed, the laser interferometer 32 constantly measures the position and yawing amount of the wafer 28 in the XY plane, and this measurement data is supplied to the main control system 25. I have. The main control system 25
The operation of the wafer stage 30 is controlled via the driving device 33. Further, a storage device 25a is connected to the main control system 25.

FIG. 2A shows a rectangular slit-shaped illumination area 20 on the reticle 19, and the illumination area 20 is inscribed in the outline of a circular area conjugate with the maximum exposure field of the projection optical system 27, and has a longitudinal shape. The length in the Y direction is (L + 2
M), and the width in the X direction, which is the narrower direction, is D. By scanning the reticle 19 in the X direction with the width D, the pulse laser light in the illumination area 20 is changed to the reticle 1.
9 sequentially illuminates a wider pattern area. Further, as shown in FIG. 2B, the illuminance distribution S in the Y direction in the illumination area 20 is constant in the area of the central length L, and is substantially the same in the areas 20a and 20b of the length M on both sides. It falls to 0 linearly. That is, the illuminance distribution S in the Y direction perpendicular to the relative scanning direction in the illumination area 20 has a trapezoidal shape. As described above, in order to obtain a trapezoidal illumination distribution, the field stop 15 shown in FIG.
In the opening, the longitudinal direction may be defocused. Alternatively, a trapezoidal illuminance distribution can be obtained by disposing an ND filter plate or the like whose transmittance distribution changes linearly in the field stop 15 or the illumination optimizing optical system 13.

FIG. 3 shows the reticle 19 in FIG. 1. In FIG. 3, a pattern area 35 having a width in the Y direction LT is formed on the pattern forming surface of the reticle 19, and the pattern area 35 is to be transferred to a wafer. A circuit pattern is formed. A forbidden band 36 formed of a light-shielding portion having a width of M or more is formed outside the pattern region 35 in the Y direction.
In this example, the pattern area 35 is formed as the slit-shaped illumination area 2.
By scanning twice in the X direction at 0, the pattern in the pattern area 35 is transferred onto the wafer. Then, for example, the pattern in the substantially right half area 35a is transferred onto the wafer in the first scan, and the pattern in the substantially left half area 35b is transferred onto the wafer in the second scan.

At this time, the left end of the area 35a and the area 35
The right end of b is overlapped with the connecting portion 35c having the width M in the Y direction so that the connecting portion 35c is scanned by the region 20a or 20b of the illumination region 20 where the illuminance gradually decreases. Thereby, the illuminance distribution of the connection portion 35c becomes uniform, and the displacement of the transferred pattern is prevented.
Further, in order to make the illuminance in the pattern area 35 constant, the illumination area 20 is provided at an end of the pattern area 35 in the Y direction.
Is not generated by the regions 20a and 20b whose illuminance gradually decreases. The width in the Y direction of the area where the illuminance is constant in the illumination area 20 is L, and the illuminance gradually becomes 0.
The widths of the regions 20a and 20b in the Y direction are M
Therefore, the width LT of the pattern area 20 in the Y direction is as follows.

[Expression 3] LT = 2 · L + M

In general, the pattern region 35 is
By scanning n times in the X direction with the
In order not to generate a region illuminated only by the region 20a or 20b where the illuminance gradually decreases as the pattern is transferred onto the wafer, the width LT of the pattern region 35 in the Y direction is determined as follows. Just do it.

[Expression 4] LT = n · L + (n−1) · M

FIG. 4A shows a rectangular slit-shaped exposure area 20P on the wafer 28 of FIG.
Is conjugate with the illumination area 20 on the reticle 19 in FIG. In this case, since the projection magnification of the projection optical system 27 is β, the width of the exposure area 20P in the X direction is β · D, and the width in the Y direction is β · (L + 2M). Further, as shown in FIG. 4B, the width in the Y direction on both sides of the exposure area 20P is β.
In the M regions 20aP and 20bP, the illuminance S decreases almost linearly to zero. When performing slit scan exposure, the wafer 2 is exposed to the exposure area 20P.
8 is scanned in the X direction, the illuminance distribution in the Y direction perpendicular to the direction of the relative scanning of the exposure area 20P is also trapezoidal.

Next, the condition of the width β · D in the X direction, which is the direction of relative scanning of the exposure area 20P, will be described. In this case, the period of the pulse light emission of the pulse laser light source 12 in FIG.
Let ΔL be the distance scanned by X in the X direction.
The width β · D in the X direction of 0P is set to an integral multiple of the distance ΔL. When the scanning speed of the wafer 28 in the X direction is β · V, the distance ΔL is T · β · V. That is, the following equation is satisfied when m is an integer of 1 or more.

## EQU5 ## β · D = m · ΔL = m · T · β · V

FIG. 4A shows a case where β · D = 4 · ΔL. At this time, for example, the exposure point Q existing at the edge of the exposure area 20P at the time of the pulse emission
Numeral 0 is irradiated with three pulses of pulse laser light inside the exposure region 20P, and irradiated with two pulses of pulse laser light at the edge of the exposure region 20P. Further, assuming that the energy applied to the exposure point inside the exposure area 20P in one pulse emission is ΔE, the exposure point Q0 has 4 · ΔE (=
Energy of ΔE / 2 + 3 · ΔE + ΔE / 2) is applied. Further, for example, the exposure point Q on the wafer existing inside the edge portion of the exposure area 20P at the time of the pulse emission
The energy of 1 · 4 · ΔE is applied to the exposure point Q2 on the wafer existing outside the edge portion of the exposure area 20P when the pulse light emission is performed. As described above, according to this example, the same m-pulse pulse laser light is applied to all the exposure points scanned by the exposure area 20P on the wafer.
Therefore, the illuminance distribution becomes constant at the exposure point where the illuminance distribution in the exposure area 20P is scanned.

The regions 20a on both sides of the exposure region 20P
At the exposure point scanned once by P and 20bP, the irradiation energy is small even if the number of irradiation pulses is m pulses. However, as described later, in this example, the connection portion at the time of stitching is formed in the regions 20aP and 2aP.
Since the scanning is performed twice at 0 bP, the energy irradiated at each exposure point of the connection portion is m · ΔE, and the amount of energy irradiated at all the exposure points on the wafer becomes the same, and the illuminance unevenness is reduced. Is gone.

Next, an example of the stitching and slit scan exposure operations of this embodiment will be described. First, FIG.
, The slit-shaped illumination area 20 on the reticle 19
Is illuminated with the pulse laser beam IL, and the driving device 2
6 and the reticle 19 via the reticle stage 21
Scanning is performed at a constant speed V in the X direction. In synchronization with this, the wafer 28 is scanned in the X direction at a constant speed β · V via the driving device 33 and the wafer stage 30. As described above, when scanning the reticle 19 and the wafer 28, for example, when the predetermined alignment mark on the reticle 19 and the predetermined alignment mark on the wafer 28 match, the measurement value of the laser interferometer 24 and the laser The difference between the value obtained by multiplying the measurement value of the interferometer 32 by the projection magnification β is stored as a reference value, and the measurement value of the laser interferometer 24 and the value obtained by multiplying the measurement value of the laser interferometer 32 by the projection magnification β Of the driving device 26 so that the difference
And 33 are controlled. Therefore, the reticle 19 and the wafer 28 are always stationary with each other in a predetermined relationship,
Scanning is performed in the narrow direction with respect to the illumination area 20 and the exposure area 20P, respectively.

Thus, as shown in FIG. 3, on the reticle 19 side, the slit-shaped illumination area 20 relatively scans the area 35a on the right side of the pattern area 35 along the locus 37. Further, as shown in FIG. 5A, on the wafer 28 side, the slit-shaped exposure area 20P relatively scans the area 40a on the left side of the exposure area 40 corresponding to the pattern area 35 along the locus 37P.

Next, when the first slit scan exposure is completed, the reticle 19 is moved in the Y direction by stitching in FIG.
0 is moved to the upper left of the pattern area 35 along the locus 38.
Further, in FIG. 5A, the wafer 28 is moved in the −Y direction, so that the slit-shaped exposure area 20P is moved to the lower right of the exposure area 40 along the trajectory 38P. afterwards,
The reticle 19 is scanned at a speed V in the X direction, and the wafer 28 is scanned at a speed β
A second slit scan exposure is performed. As a result, as shown in FIG. 3, on the reticle 19 side, the slit-shaped illumination area 20 relatively scans the area 35b on the left side of the pattern area 35 along the trajectory 39. Further, as shown in FIG. 5A, on the wafer 28 side, the slit-shaped exposure area 20P relatively scans the area 40b on the right side of the exposure area 40 corresponding to the pattern area 35 along the trajectory 39P.

As shown in FIG. 3, in the first scan and the second scan, the connection area 35c at the center of the pattern area 35 of the reticle 19 in the Y direction is the illumination area 2
Exposure is performed so as to overlap with the left region 20a where the illuminance of 0 decreases and the right region 20b where the illuminance decreases. As a result, as shown in FIG.
In the connection portion 40c at the center of the exposed region 40 in the Y direction, the exposure is performed by overlapping the right region 20aP where the illuminance of the slit-shaped exposure region 20P decreases and the left region 20bP where the illuminance decreases. . For example, at the exposure point Q3 in the connection part 40c, the illuminance at the time of the first exposure is the illuminance SA in FIG. 5B, and the illuminance at the time of the second exposure is the illuminance SB. Since the illuminance distributions in the Y direction of the area 20aP and the area bP linearly decrease to 0 symmetrically with each other, the illuminance of the sum of the illuminance SA and the illuminance SB is
The illuminance of P becomes equal to the illuminance SC of a certain area.

As described above, the exposure area 20
At the exposure point scanned once by P, m pulsed laser light is irradiated. Therefore, at the exposure point Q3 in the connection part 40c, two scans of the exposure area 20P irradiate the same amount of energy as the exposure point of the non-connection part,
The illuminance distribution is uniform at all the exposure points on the wafer 28. Further, the number of pulses irradiated by two scans at the exposure point in the connection part 40c is twice as large as the exposure point at the non-connection part.
m. Therefore, in the connecting portion 40c, the influence of the energy variation and the speckle of each pulse of the pulse laser light is particularly reduced. Specifically, variations in illumination intensity due to variation in energy per pulse laser light pulses, the variation in the connection portion 40c is reduced to 1/2 ^1/2 of the variation in the non-connecting portion.

Next, in this embodiment, when the area 40a on the wafer 28 shown in FIG.
The relative positions of the reticle 19 and the wafer 28 are monitored by the laser interferometers 24 and 32, and the reticle 19 and the wafer 2 are monitored.
The relative displacement with respect to 8 is regarded as a mechanical control error.
It is stored in the storage device 25a of FIG. That is, if an image at an arbitrary exposure point on the wafer 28 is formed by irradiation of m pulses of pulsed laser light by the first scan, the relative positional shift amount in the X direction is synchronized with each pulse emission. Monitor. Assuming that the relative displacement in the X direction for each pulse is Δx _i and Σ represents the sum of 1 to m for the subscript i, the average displacement <Δx> can be calculated by the following calculation.

<Δx> = ΣΔx _i / m

Similarly, the relative displacement Δy _i between the reticle 19 and the wafer 28 in the Y direction perpendicular to the X direction, which is the direction of the relative scanning, and the rotational error between the reticle 19 and the wafer 28 are also stored in the storage device. 25a. Therefore, 1
When the storage capacity of the pieces of positional deviation amount [Delta] x _i and .DELTA.M, as the storage capacity of the storage device 25a, may be any multiple of the capacity of .DELTA.M × m pieces minute, for example, the relative positional deviation amount at the position in the vicinity By averaging the values, the number of relative displacements to be stored can be reduced. The relative position error and the rotation error stored in the storage device 25a as described above cause so-called “in-shot distortion” in the connection portion 40c on the wafer 28.

Next, when exposing the area 40b in FIG. 5A by the second scan, the main control system 25 adjusts the relative position error and the rotation error read from the storage device 25a so as to match. The operations of the reticle stage 21 and the wafer stage 30 are controlled via the driving devices 26 and 33. As a result, the overlay accuracy of the pattern at the connection portion 40c on the wafer 28 becomes high. Further, assuming that the positioning accuracy of the normal reticle stage 21 and the wafer stage 30 in the X and Y directions is △ x, △ y,
The overlay error at 0c is 2 ^1/2 Δx and 2 respectively.
^1/2 Δy. On the other hand, in the method of the present embodiment, the positional relationship at the time of exposing the next area 40b is corrected in accordance with the shot distortion at the time of exposing the first area 40a (the reticle 19 and the wafer are adjusted so as to have the same shot distortion). 28 is controlled), the overlay error is only △ x, △ y.

Next, a method of exposing the entire exposed surface of the wafer 28 will be described. First, when the method of FIG. 5A is applied, as shown in FIG. 6, the adjacent areas 40-1a and 40-1 are sequentially scanned by slit scan exposure.
b, 40-2a, 40-2b, ‥‥, 40-4b, 40
-4a is performed. However, the method of FIG.
The direction of (a) and the scanning direction are opposite. The circuit pattern of the pattern area 35 of the reticle 19 is transferred to two scanning areas, for example, the areas 40-1b and 40-1a. In this method, the scanning direction of the pattern area 35 of the reticle 19 in the conjugate image on the wafer 28 is reversed. Further, according to this scanning method, the pattern in the pattern area 35 can be transferred onto the wafer 28 in a short time, and there is an advantage that the pattern is not easily affected by expansion of the wafer 28 or the like.
On the other hand, there is a possibility that the accuracy of the connection portion may be deteriorated due to the characteristics in the scanning direction, and the trajectory 38 of the illumination area 20 in FIG.
It is necessary to move the reticle 19 corresponding to the above at a high speed.

Next, as shown in FIG. 7, first, for example, only the right half area 35a of the pattern area 35 of the reticle 19 is continuously exposed to the corresponding area on the wafer 28,
Next, there is a method of continuously exposing only the left half area 35b of the pattern area 35 to the corresponding area on the wafer 28. That is, in this method, first, as shown in FIG. 7A, the regions 40-1b, 40-2b,.
, 40-4b are exposed, and then, as shown in FIG. 7B, the regions 40-1a, 40-2a,. Exposure to 40-4a is performed. According to this method, the scanning direction of the slit-shaped exposure area 20P is the same in the two exposure areas (for example, the areas 40-1b and 40-1a) on the wafer 28 corresponding to the pattern area 35 of the reticle 19. . This may improve the overlay accuracy at the connection portion 40c.

Next, in the above embodiment, since a refraction optical system is used as the projection optical system 27, the illumination area on the reticle 19 is rectangular as shown in FIG.
It is 0. On the other hand, especially for shortening the wavelength of the exposure light, using a projection optical system constituted by a catadioptric optical system using a concave mirror or the like is advantageous in terms of aberration and the like. Further, since the concave mirror and the like have less aberration in a region away from the optical axis, when a catadioptric optical system is used, the slit-shaped illumination region on the reticle 19 becomes as shown in FIG. , An illumination area 41 having an arc shape. Even in this case,
Assuming that the width D of the illumination area 41 in the relative scanning direction is constant and the longitudinal direction perpendicular to the relative scanning direction of the illumination area 41 is the Y direction, the illuminance distribution of the illumination area 41 in the Y direction is as shown in FIG.
It has a trapezoidal shape as shown in FIG. That is, in the regions 41a and 41b at both ends in the Y direction of the illumination region 41, the illuminance distribution linearly falls to zero. By setting such an illuminance distribution, it is possible to reduce the illuminance unevenness of the connecting portion at the time of stitching as in the embodiment of FIG.

Next, another embodiment of the present invention will be described with reference to FIG. In the present embodiment, as described with reference to FIG. 11, the present invention is applied to the case where the wafer is scanned in the regular hexagonal exposure region 3. FIGS. 9A and 9B show the connection portion 4 on the wafer when stitching is performed in the present embodiment. In FIGS. Are uniform, but pulsed laser light is used as exposure light. Further, assuming that the relative scanning directions are the X direction and the −X direction, two opposite sides of the exposure region 3 with the interval W are parallel to the Y direction which is the stitching direction. Even in this case, at the time of slit scan exposure, assuming that the distance that the wafer moves in the X direction or the −X direction during one cycle of the pulse emission is ΔL, the interval W is expressed by the integer m of 1 or more. Is set as follows.

(7) W = m · ΔL

FIGS. 9A and 9B show the case of m = 8, and the energy of eight pulses is always applied to the exposure point P0 of the non-connection portion on the wafer by one scan. Is done. Further, in this embodiment, when the pulse laser light source emits a pulse, the wafer is scanned in the X direction with respect to the exposure region 3 and the wafer is scanned in the −X direction with respect to the exposure region 3. The positions of the wafers in the X direction are the same. For example, in FIG. 9A, when the exposure point P9 in the connection portion 4 on the wafer is scanned in the X direction with respect to the exposure area 3, that is, the exposure point P9 is a right isosceles triangle of the exposure area 3. When scanning the area 3a, the position in the X direction of the exposure point P9 when the pulse emission is performed is set as a position 8. Then, when the exposure point P9 is scanned in the −X direction with respect to the exposure area 3 by the second scanning, that is, when the exposure point P9 scans the left isosceles triangular area 3b of the exposure area 3 , Exposure point P when pulse emission is performed
Assuming that the position in the X direction 9 is position 42, this means that position 42 and position 8 match. In the case of FIG. 9A, since there are five positions 8 in the region 3a and three positions 42 in the region 3b, a total of eight pulses are obtained at the exposure point P9 by two slit scan exposures. Minutes of energy is applied.

FIG. 9B shows a case where the pulse emission timing is shifted by ΔL / 2 in the X direction as compared with the case of FIG. 9A. In FIG. 9B, when the exposure point P9 scans an isosceles triangular area 3a on the right side of the exposure area 3, the exposure point P9 when pulse emission is performed.
When the exposure point P9 scans the left isosceles triangular area 3b of the exposure area 3 by the second scanning with respect to the position 10 in the X direction, the exposure point P when the pulse emission is performed is performed.
9 so that the positions 43 in the X direction are equal. FIG.
In the case of (b), there are four positions 10 in the region 3a and four positions 43 in the region 3b.
9 is irradiated with a total of eight pulses of energy by two slit scan exposures. In general, according to the present example, at each exposure point in the connection part 4, the energy of eight pulses is irradiated similarly to the exposure point P0 of the non-connection part, and illuminance unevenness does not occur. In the above embodiment, the stitching operation using one reticle has been described.
Place multiple reticles on the same reticle stage,
The scanning exposure may be repeatedly performed while changing the reticle at the time of stitching.

As described above, the projection optical apparatus of the above embodiment
Generates pulsed exposure light as shown in FIGS. 1 and 2, for example.
A pulse light source (12) that emits light and a transfer
Illuminating the mask (19) with turns formed with uniform illuminance
Illumination optical system (13, 16, 18) and mask (1)
9) Set the illumination area (20) by the exposure light on the
Field stop (15) and putter for transfer of mask (19)
Optical system (2) for projecting an image of
7) and a predetermined direction of the illumination area (20) by the exposure light.
The mask (19) and the photosensitive substrate (28) are relatively
Relative scanning means (25, 26, 33)
Illuminated by the exposure light on the mask (19)
Exposes a transfer pattern image in a wider area than (20)
A projection optical device for exposing on a substrate (28),
The predetermined direction (X direction) of the illumination area (20) by the exposure light
Direction) relative to the second direction (Y direction)
Substrate moving means (30, 33) for moving the plate (28)
And the second one of the illumination area (20) by the exposure light
The illuminance that makes the illuminance distribution of the direction trapezoidal (for example, FIG. 2B)
Distribution setting means (13, 15);
8) to the second of the illumination area (20) by the exposure light.
While shifting in the direction, the illumination area (2
0) the photosensitive substrate (28) is relatively duplicated in the predetermined direction.
Scan several times and replace with mask (19) or this mask.
The image of the transferred pattern of the mask is transferred to the photosensitive substrate (28).
The top is exposed. In addition, illumination by the exposure light
The mask (1) relative to the predetermined direction of the region (20)
9) and when scanning the photosensitive substrate (28) synchronously.
Error of the relative position between the disk (19) and the photosensitive substrate (28).
A storage means (25a) for storing the difference is provided. Ma
For example, as shown in FIG.
The illumination area (20) whose illuminance distribution due to the exposure light is trapezoidal
In the second direction, the length of the area where the illuminance distribution is constant is L,
The illuminance distribution on the both sides of the trapezoidal area gradually decreases
On the mask (19), where M is the length of the
Of the transfer pattern formed on the substrate by the exposure light
If the width of the region (20) in the second direction is LT,
The following relationship is established by using an integer n of 1 or more.
I have.

LT = n · L + (n−1) · M and the second of the illumination area (20) by the exposure light
Mask movement to move the mask (19) relatively in the direction
Means (21, 26) are provided. Next, the above
An example of the projection optical apparatus includes, as shown in FIG.
Pulse light source (12) that emits light in a pulse
The mask (19) on which the pattern for transfer is formed is
An illumination optical system (13, 16, 18) for illuminating with illuminance;
The illumination area (20) on the disc (19) by the exposure light.
Field stop (15) to be set and for transfer of mask (19)
Light for projecting the image of the pattern on the photosensitive substrate (28)
Of the science system (27) and the illumination area (20) by the exposure light
The mask (19) and the photosensitive substrate (2) are relatively arranged in a predetermined direction.
8) synchronous scanning means (25, 26, 3)
3) illumination by the exposure light on the mask (19).
Image of transfer pattern in area wider than bright area (20)
A projection optical device for exposing the light onto a photosensitive substrate (28).
The predetermined area of the illumination area (20) by the exposure light.
Relative to a second direction (Y direction) intersecting the direction (X direction)
Moving means (30, 30) for moving the photosensitive substrate (28)
33) and when the pulse light source (12) emits pulse light
Of the photosensitive substrate (28) in a direction conjugate to the predetermined direction.
Light emission position storage means (25, 25) for detecting and storing the position
a) and the starting point of the pulse light emission of the pulse light source (12).
And a light emission control means (34) for controlling the photosensitive substrate (2).
8) to the second of the illumination area (20) by the exposure light.
While shifting in the direction, the illumination area (2
0) in the predetermined direction with respect to the mask (19) or this
The exchanged mask and the photosensitive substrate (28) are
The mask (19) or this mask and
The image of the pattern for transfer of the replaced mask is exposed on the photosensitive substrate.
(28) When exposing on the top, for example, as shown in FIG.
In each of its relative scans, a pulsed light source (1
2) The position of the photosensitive substrate (28) when the pulse emission is performed
The position in the conjugate direction (X direction) will be the same as the fixed direction
It is also a thing. According to such a projection optical device,
The illumination area (20) by the exposure light on the mask (19)
, The second direction intersecting the direction of relative scanning
The illumination distribution in the direction is trapezoidal. As a result, as shown in FIG.
Sea urchin, co in the illumination area on the photosensitive substrate (28) (20)
Of the useful exposure area (20P) in its second direction (Y direction)
The illuminance distribution is also trapezoidal. In this case, the exposure area (20
If the width in the relative scanning direction of P) is fixed, the exposure area
(20P) on the photosensitive substrate (28) relatively scanned by
Each exposure point arranged in the second direction is
It is irradiated with exposure light for a number of times. Also, by stitching
The exposure area (20P) on the photosensitive substrate (28)
When illuminating, the illuminance distribution gradually decreases as shown in FIG.
Regions (20aP, 20bP) to be overlapped.
This allows scanning twice, for example by stitching
In the exposure point Q3 existing at the connection portion (40c)
Of illuminance SA at the time of scanning and illuminance SB at the time of second scanning
However, the illuminance in the area where the illuminance distribution is constant in the trapezoidal illuminance distribution
It becomes equal to SC. Therefore, the connection on the photosensitive substrate (28)
Illuminance at an arbitrary exposure point on the portion (40c)
It becomes almost equal to the illuminance of the light spot, and the illuminance unevenness is reduced.
In the connection section (40c), a pulsed exposure light irradiation
Since the number of pulses is twice the number of pulses at the unconnected part,
Illumination unevenness and speckle shadow caused by variations in each
The sound is especially reduced. Note that the photosensitive substrate (28)
When performing the etching, stick the mask (19).
Replace mask (19) with another mask instead of performing
It may be replaced. The illumination area (2)
0) relative to the mask (19) and the sense in its predetermined direction.
When the optical substrate (28) is scanned synchronously, the mask (1) is used.
Stores the relative position error between 9) and the photosensitive substrate (28).
When the storage means (25a) is provided, for example, once
Mask (19) and exposure during eye slit scan exposure
The amount of displacement relative to the substrate (28) is stored.
And the second slit scan after stitching
At the time of exposure, the position of the mask (19) and the photosensitive substrate (28)
By adjusting the displacement amount to the stored displacement amount
Therefore, the overlay accuracy of the connection portion can be improved.
For example, as shown in FIG.
The illuminance distribution due to the exposure light of
In the second direction, the length of the region where the illuminance distribution is constant is L,
The illuminance distribution is such that the illuminance on both sides of the trapezoid
The mask (19), where M is the length of the region to be
The exposure light of the transfer pattern (35) formed thereon
Let LT be the width of the illumination area (20) in the second direction due to
I do. In this case, the illumination area by stitching
(20) is scanned n times on the mask (19),
(19) The pattern (35) on the photosensitive substrate (28)
If it is to be exposed, the connection of the illumination area (20)
In (35c), the area of length M where the illuminance distribution gradually decreases
Areas need to overlap. However, the pattern (3
5) To maintain the illuminance distribution at both ends at about the same
First, the illuminance distribution at both ends of the pattern (35) is
It is desirable that the area of the decreasing length M be shielded from light.
No. Thus, the width of the pattern (35) in its second direction
LT is as shown in (Equation 8). Also, depending on the exposure light
Mask relative to the second direction of the illuminated area (20)
Mask moving means (21, 26) for moving (19)
If provided, stitch also to mask (19)
Can be performed. Further, the above-described projection optical device
According to FIG. 9, for example, as shown in FIG.
Scanning on the photosensitive substrate (28) in the exposure area (3)
However, for any exposure point P9 in the connection area (4), 1
The position (8) of the pulse light emission of the exposure light at the time of the second scanning and two times
The position (42) of the pulse light emission of the exposure light at the time of scanning the eye is the same.
New In the case of FIG. 9A, the exposure point P9 has eight pulses.
Energy has been irradiated. 9 (b).
In this case, the timing of the pulse emission is different from the case of FIG.
Is shifted, but at the time of the first scanning with respect to the exposure point P9.
Of the pulse emission of the exposure light of (10) and the second scanning
(43) of the pulse light emission of the exposure light. Soshi
Therefore, in the case of FIG.
Energy is being irradiated. In other words, for stitching
In the connection portion (4) that is scanned twice more,
By setting the same position for pulse emission,
Energy is applied, and uneven illuminance is reduced.
It is. It should be noted that the present invention is not limited to the above-described embodiments, but can take various configurations without departing from the gist of the present invention.

[0050]

According to the present invention, when exposure is performed by a stitching exposure method using a pulsed light source, a connection portion on a substrate is exposed at the same illuminance as a non-connection portion by two exposures. Therefore, there is an advantage that illuminance unevenness of the connection portion on the substrate can be reduced.

[0051] Further, scan are scanned synchronously relatively mask and the photosensitive substrate in a predetermined direction of the illumination region by the exposure light
When the exposure is applied, a storage unit for storing an error of a relative position between the mask and the photosensitive substrate is provided,
By storing the relative position error in the second scan and performing position control in accordance with the relative error in the second scan at the connection, the overlay error at the connection is reduced. it can. In addition, the length of an area where the illuminance distribution of the exposure light on the mask is constant in the second direction of the trapezoidal illumination area in the second direction is L, and the illuminance on both sides of the trapezoidal area whose illumination distribution is gradually reduced. Assuming that the lengths of the regions are M, the width LT of the illumination region of the transfer pattern formed on the mask by the exposure light in the second direction is {n · L +
By setting (N−1) · M｝, the transfer pattern on the mask is always illuminated with uniform illuminance.

Then, the second of the illumination area by the exposure light
When a mask moving means for moving the mask relatively in the direction of is provided, stitching can also be performed on the mask side. Next, according to the second illumination optical device of the present invention, the position of the pulse emission at the time of the first scan is equal to the position of the pulse emission at the time of the second scan at the connection part. The illuminance at the exposure point is equal to the illuminance at each exposure point of the non-connection portion, and the illuminance unevenness is reduced.

[Brief description of the drawings]

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

2A is a plan view showing a slit-shaped illumination area on the reticle 19 in FIG. 1, and FIG. 2B is a distribution chart showing an illuminance distribution of the illumination area.

FIG. 3 is a plan view showing a reticle pattern of the embodiment.

FIG. 4A is a plan view showing a slit-shaped exposure region on a wafer according to the embodiment, and FIG. 4B is a distribution diagram showing an illuminance distribution of the exposure region.

FIG. 5A is a plan view showing a region to be exposed on a wafer,
(B) is a distribution diagram showing an illuminance distribution in the exposed region.

FIG. 6 is a plan view showing an example of a locus of slit scan exposure on a wafer in the embodiment.

FIG. 7 is a plan view showing another example of the locus of the slit scan exposure on the wafer in the embodiment.

8A is a plan view showing a modification of the illumination area on the reticle, and FIG. 8B is a distribution diagram showing an illuminance distribution of the modification of the illumination area.

FIG. 9A is an enlarged plan view showing an example of the relationship between the positions of pulsed light emission in another embodiment of the present invention, and FIG. 9B is another enlarged view showing the relationship between the positions of pulsed light emission in another embodiment of the present invention. It is an enlarged plan view which shows the example of.

FIG. 10A is a plan view showing a relative scanning state of an illumination area on a reticle when performing stitching and slit scan exposure with a projection exposure apparatus having a conventional continuous light source, and FIG. FIG. 11 is a plan view showing a state of relative scanning of an exposure area on a wafer corresponding to FIG.

FIG. 11 is a diagram for explaining illuminance unevenness on a photosensitive substrate when a pulsed light source is used when performing stitching and slit scan exposure in a regular hexagonal exposure region.

[Explanation of symbols]

 Reference Signs List 12 pulse laser light source 13 illumination optimization optical system 15 field stop 16 relay lens 18 condenser lens 19 reticle 20 rectangular slit-shaped illumination area on reticle 20P rectangular slit-shaped exposure area on wafer 21 reticle stage 23 guide 24, Reference Signs List 32 laser interferometer 25 main control system 26, 33 drive unit 28 wafer 30 wafer stage 34 laser light source control unit 35 pattern area 40 exposure area

Continuation of the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) H01L 21/027 G03F 7/20

Claims (17)

(57) [Claims] 1. A first region on a substrate is exposed using a first pattern, and after the exposure, a second region shifted in a predetermined direction with respect to the first region is stitched using a second pattern. An exposure method of exposing the first pattern on the substrate by irradiating the first pattern with exposure light having an area where the illuminance gradually decreases at one end of the illuminance distribution in the predetermined direction. And, after the exposure of the first area, irradiating the second pattern with exposure light in which an area where the illuminance gradually decreases is formed on the other end side of the illuminance distribution in the predetermined direction. By exposing the second region, an overlapping portion of the first region and the second region on the substrate is formed at a region formed at one end of the illuminance distribution and at another end of the illuminance distribution. The feature is that it is exposed twice with the Exposure method for. 2. The exposure method according to claim 1, wherein the area where the illuminance gradually decreases is formed by disposing a field stop that defines an illumination area of the exposure light in a defocused state. . 3. The exposure method according to claim 1, wherein the region where the illuminance gradually decreases is formed by disposing an ND filter having a transmittance distribution changing in an optical path of the exposure light. 4. The exposure method according to claim 1, wherein the first pattern and the second pattern are formed on different masks. 5. The exposure method according to claim 1, wherein the first pattern and the second pattern are formed on the same mask. 6. The illuminance distribution has a region in which the length in the predetermined direction is L and the illuminance is substantially constant, and regions on both sides thereof in which the length in the predetermined direction is M and the illuminance gradually decreases, 6. When the width of the pattern region of the mask in the predetermined direction is LT, the condition LT = nL + (n-1) M is satisfied using an integer n of 1 or more. Exposure method according to 1. 7. When information using the first pattern is performed, information on a position error between the mask and the substrate is obtained. When performing exposure using the second pattern, information on the position error is obtained. The exposure method according to any one of claims 1 to 6, wherein a positional relationship between the mask and the substrate is adjusted based on the relationship. 8. The exposure method according to claim 1, wherein the exposure light is pulsed. 9. An element manufacturing method using the exposure method according to claim 1. Description: 10. A first region on a substrate is exposed using a first pattern, and after the exposure, a second region shifted in a predetermined direction with respect to the first region is stitched using a second pattern. An exposure apparatus that performs exposure by a method, comprising: a substrate stage that holds the substrate and is movable in the predetermined direction; and an illuminance on one end of an illuminance distribution in the predetermined direction of exposure light applied to the first pattern. An optical member that forms an area where the illuminance gradually decreases, and an optical member that forms an area where the illuminance gradually decreases on the other end side of the illuminance distribution in the predetermined direction of the exposure light applied to the second pattern; The substrate such that an overlapping portion of the first region and the second region is exposed twice in a region formed on one end of the illuminance distribution and a region formed on the other end of the illuminance distribution. Controls stage position An exposure apparatus, comprising: 11. The exposure apparatus according to claim 10, wherein the optical member includes a field stop arranged in a defocused state. 12. The exposure apparatus according to claim 10, wherein the optical member includes an ND filter arranged in an optical path of the exposure light and having a transmittance distribution changed. 13. The exposure apparatus according to claim 10, wherein the first pattern and the second pattern are formed on different masks. 14. The apparatus according to claim 1, wherein the mask on which the first pattern is formed and the mask on which the second pattern is formed are mounted on the same mask stage.
4. The exposure apparatus according to 3. 15. An exposure apparatus for exposing a substrate to form a predetermined pattern on the substrate, comprising: a mask stage for holding a mask on which a first pattern is formed and a mask on which a second pattern is formed; A substrate stage for holding the substrate, and synchronously moving the mask stage and the substrate stage for scanning and exposing the substrate using the first pattern, and scanning the substrate using the second pattern An exposure apparatus, comprising: a control system that moves the mask stage and the substrate stage in synchronization with each other for exposure. 16. The apparatus according to claim 14, further comprising a first interferometer system for measuring position information of the mask stage and a second interferometer system for measuring position information of the substrate stage. Exposure equipment. 17. The method device fabrication using the exposure apparatus according to any one of claims 10 to 16.