WO2000059012A1

WO2000059012A1 - Exposure method and apparatus

Info

Publication number: WO2000059012A1
Application number: PCT/JP2000/001540
Authority: WO
Inventors: Makoto Kondo
Original assignee: Nikon Corporation
Priority date: 1999-03-26
Filing date: 2000-03-14
Publication date: 2000-10-05
Also published as: AU2944500A; KR20010112286A

Abstract

An exposing method in which the total exposure of the portion where four patterns are adjacent to one another is almost equal to that of the other portions when patterns are transferred two-dimensionally by image stitching. Projected images (30A to 30D) of the pattern of a reticle are so stitched as to overlap with one another in both X and Y directions on a wafer to expose the wafer. The distribution of exposure of the rectangular corner parts, i.e., the areas (stitching portions (31)) where four projected images (30A to 30D) are adjacent to one another is determined according to the characteristic which is the product of a first characteristic gradually decreasing in the X direction and a second characteristic gradually decreasing in the Y direction.

Description

Exposure method and apparatus

Technical field

The present invention relates to an exposure method and an exposure apparatus used in a lithographic process for manufacturing a device such as a semiconductor device, an imaging device (such as a CCD), a liquid crystal display device, or a thin-film magnetic head. Screen pattern

This method is suitable for use when exposing a large pattern by transferring (connecting exposure) while performing the patterning.

Background art

Conventional semiconductor integrated circuits are generally manufactured by repeating a process of transferring a pattern of one reticle as a mask to each shot area on a wafer as a substrate. On the other hand, recently, in order to manufacture a large-sized semiconductor integrated circuit device, the original pattern of one circuit pattern to be transferred is divided into a plurality of reticle patterns, and the reticle pattern An exposure method is used in which a pattern is transferred to one shot area on a wafer while screen joining is performed, that is, continuous exposure is performed. Bridge exposure is also called “angle of view synthesis”.

When performing bridge exposure using a projection exposure apparatus such as a stepper, multiple exposures may occur due to drawing errors of the reticle pattern, aberrations of the projection optical system, and positioning errors of the stage for positioning the reticle or wafer. The pattern may be cut at the joint (boundary) of the image of the reticle pattern. In order to prevent such cutting of a pattern, a method of performing exposure by overlapping a joint portion of an image of an adjacent pattern by a minute width has conventionally been developed. Has been issued.

When performing exposure while performing screen splicing as described above, double exposure was performed at the overlapped portion as a spliced portion. However, simply performing double exposure doubles the integrated exposure amount in that area, and depending on the characteristics of the photosensitive agent (resist) applied to the wafer, the joint of the pattern after development and etching, etc. Has a disadvantage that the line width of the image changes.

In order to avoid a change in the line width of the pattern at such a joint, for example, in an optical filter arranged on a surface conjugate to the reticle, the exposure light is applied to a region corresponding to the joint. There has been proposed a projection exposure apparatus provided with a dimming unit that linearly reduces the amount of transmitted light outward. In this case, for example, in the overlapping portion of the images of two patterns that are adjacent to each other in the one-dimensional direction, the distribution of the exposure amount in the case of double exposure is a distribution symmetrically inclined with respect to each other. The exposure amount coincides with the integrated exposure amount of the other parts. In other words, the extinction characteristic of the superimposed part at that time is defined by a one-dimensional function.

In this regard, as semiconductor integrated circuits have become larger in recent years, there has been an increasing need to expose two-dimensionally high-precision images of reticle patterns. However, in the state where the dimming characteristic of the superimposed portion is one-dimensionally reduced in a predetermined direction, when such two-dimensional screen splicing is performed, the corners of the images of four adjacent patterns overlap. In the region, the integrated exposure amount after the quadruple exposure is different from the integrated exposure amount in other portions, and there is a disadvantage that the line width of the circuit pattern formed at the corner changes.

In view of the above, in the present invention, when exposing a plurality of patterns while performing two-dimensional screen splicing, the integrated exposure amount in a portion where the four patterns are adjacent is different from the integrated exposure amount in other portions ( Exposure method) Its primary purpose is to provide law.

It is a second object of the present invention to provide an exposure apparatus capable of performing such an exposure method, and a method of manufacturing such an exposure apparatus.

Another object of the present invention is to provide a method of manufacturing a device or a mask using such an exposure method. Disclosure of the invention

The first exposure method according to the present invention is directed to an exposure method for exposing a pattern larger than each of the patterns onto the substrate by splicing and exposing a plurality of patterns on the substrate. 0 A to 30 D) are exposed in the first direction (X direction) and the second direction (Y direction), which intersect each other, so that some areas are overlapped. In the area (31) where the patterns are adjacent to each other, the corners of the four patterns are exposed while overlapping each other, and when exposing each of the four patterns, the exposure amount of the corners of the pattern is reduced. This is set based on a characteristic obtained by multiplying a first characteristic that gradually decreases outward along the first direction and a second characteristic that gradually decreases outward along the second direction.

According to the present invention, the dimming characteristics of the exposure amount at the corners of the four patterns are set to two-dimensional characteristics, respectively. It becomes almost the same as the integrated exposure amount in the portion. In addition, since the two-dimensional characteristics are obtained by multiplying the one-dimensional characteristics that intersect each other, for example, a neutral density filter having such neutral density characteristics can be easily manufactured. .

In this case, when exposing the corner of one of the four adjacent patterns, for example, as shown in FIG. 4, the width of the corner in the first and second directions is reduced. Let a and b be the vertices of the corners, respectively Let X and y be the coordinates that increase inside the corner along the first and second directions, respectively, and let the exposure at that corner be proportional to (xZa) · (y / b) It is desirable to set it to a value. By setting the characteristics of each of the other three corners to gradually rotate, the integrated exposure of the corner after the quadruple exposure exactly matches the integrated exposure of the other portions.

Next, a second exposure method according to the present invention is directed to an exposure method for exposing a pattern larger than each pattern onto the substrate by exposing a plurality of patterns onto the substrate. A plurality of patterns (32A to 32D) are spliced and exposed in a first direction (X direction) and a second direction (Y direction) crossing each other so that a part of each region overlaps. In the regions (33, 34) where the patterns are adjacent to each other, of the first pair of patterns (32A, 32D), the first pair of patterns (32A, 32D) are obliquely opposed. The corners of the rectangle are superimposed on each other and exposed, and for the second pair of patterns (32B, 32C), the corners (33, 34) of each triangle are replaced by the corners of the rectangle. Exposure is performed adjacent to the inside of the device.

According to the present invention, the corners of the three adjacent patterns (32A, 32B, 32D) are formed in one triangular area (33) of the four adjacent patterns. In the other triangular area (34), the corners of three adjacent patterns (32A, 32C, 32D) are overlapped and exposed. Since the exposure amount of each of the corners decreases outward with a predetermined characteristic, the integrated exposure amount becomes substantially equal to the other regions.

Next, a first exposure apparatus of the present invention is an exposure apparatus for transferring a pattern of a mask (R) onto a substrate (W). System (1-3, 6-8) and a field stop (4) arranged in the illumination optical system at a position substantially conjugate to the pattern surface of the mask to set an illumination area on the mask. ), A substrate stage (25) for positioning the substrate, and a surface near the pattern surface of the mask, or a surface near a conjugate surface with respect to the pattern surface, intersecting the pattern surface. The transmittance of at least one area corresponding to at least one corner of the pattern area having an outer shape substantially parallel to the first direction and the second direction to the illumination light for exposure to the outside along the first direction. And a dimming filter (55) set based on a characteristic obtained by multiplying a first characteristic that gradually decreases and a second characteristic that gradually decreases outward along the second direction. With this exposure apparatus, the first exposure method of the present invention can be used.

The second exposure apparatus according to the present invention is provided with a dimming filter (56) having the following characteristics instead of the dimming filter of the first exposure apparatus. . That is, the neutral density filter includes a pair of first and second diagonally opposed corners of a pattern area having an outer shape substantially parallel to the first and second directions intersecting with each other on the pattern surface. A first characteristic in which the transmittance for the illumination light for exposure in a region corresponding to the first pair of corners in the portion gradually decreases outward in the first direction, or in the second direction. Is set based on the second characteristic that gradually decreases outward along the distance, and the transmittance for the illumination light for exposure in the region corresponding to the second pair of corners is set in a direction opposite to the pair of corners. A characteristic obtained by adding a first characteristic that gradually decreases outward along the first direction and a second characteristic that gradually decreases outward along the second direction within a triangular area that extends outward along. This is set based on. With this second exposure apparatus, the above-described second exposure method can be used.

Next, in the third exposure method according to the present invention, the peripheral portion partially overlaps on the substrate. In an exposure method in which a pattern is transferred to at least two regions, the amount of illumination light applied to the pattern on the substrate is gradually reduced at least where the two regions overlap. In order to obtain at least one of the position information and the rotation information of the filter, at least one mark pattern provided in the dimming filter is detected. According to the present invention, the amount of illumination light, which is applied to the pattern by the dimming filter, on the substrate is gradually reduced at a portion where the at least two regions overlap, and thus the light is reduced at that portion. Is approximately equal to the other parts. Further, in the present invention, a mark pattern provided in the darkening filter is detected, and at least one of the position information and the rotation information of the darkening filter is obtained. Thereby, based on the obtained information, the relative positioning accuracy between the dimming filter and the pattern can be improved, and the line width accuracy of the device after the bridge exposure can be improved. .

In this case, it is desirable to adjust the relative relationship between the mask on which the pattern is formed and the dimming filter based on the obtained information.

In addition, it is desirable to adjust at least one of the position and the inclination of the extinction filter with respect to the optical axis direction in the optical system in which the extinction filter is arranged based on the obtained information. .

Further, the third exposure apparatus according to the present invention is an exposure apparatus that transfers a pattern to at least two regions where peripheral portions partially overlap each other on a substrate, wherein the substrate receives illumination light applied to the pattern. A dimming filter that gradually reduces the amount of light above at least where the two regions overlap, and a dimming filter that obtains at least one of positional information and rotation information of the dimming filter. And a detection device for detecting at least one mark pattern provided in the evening. With this third exposure apparatus, the third exposure method of the present invention can be used. In this case, it is desirable to further include an actuator for driving the neutral density filter in order to adjust at least one of the position and the rotation of the neutral density filter.

In addition, it is desirable that the detection device detects at least one of relative position information and relative rotation information between the darkening filter and a mask on which the pattern is formed.

Further, it is desirable that the dimming filter is arranged so as to be deviated from the pattern plane of the mask on which the pattern is formed or the conjugate plane thereof.

Next, the device manufacturing method according to the present invention includes a step of transferring a device pattern onto a photosensitive substrate using the exposure method of the present invention or using the exposure apparatus of the present invention.

Also, a method of manufacturing a mask according to the present invention is a method of manufacturing a mask using the exposure method of the present invention, wherein a step of transferring a plurality of mask patterns onto a mask substrate using the exposure method while performing screen splicing. Including. At this time, by transferring a plurality of mask patterns in a reduced scale, masks can be mass-produced with higher precision and higher throughput than a method in which a mask pattern is drawn directly on the mask substrate by using an electron beam drawing apparatus or the like. . Further, the mask according to the present invention uses a device pattern transferred while performing screen joining by the exposure method or exposure apparatus of the present invention as a mask pattern.

Next, a first method for manufacturing an exposure apparatus according to the present invention is a method for manufacturing an exposure apparatus for transferring a pattern of a mask onto a substrate, comprising: an illumination optical system for illuminating the mask; A field stop arranged at a position substantially conjugate to the pattern surface of the mask to set an illumination area on the mask, a substrate stage for positioning the substrate, Plane, or its conjugate plane or its pattern plane Illumination light for exposure of an area corresponding to at least one corner of a pattern area having an outer shape substantially parallel to the first direction and the second direction intersecting the pattern surface and arranged on a surface near the pattern surface Is set based on a characteristic obtained by multiplying the first characteristic, which gradually decreases outward along the first direction, with the second characteristic, which gradually decreases outward along the second direction. This is to assemble a file with a file in a predetermined positional relationship.

Further, a second method for manufacturing an exposure apparatus according to the present invention is a method for manufacturing an exposure apparatus for transferring a pattern of a mask onto a substrate, comprising: an illumination optical system for illuminating the mask; and the mask within the illumination optical system. A field stop which is arranged at a position substantially conjugate to the pattern surface of the mask to set an illumination area on the mask, a substrate stage for positioning the substrate, a surface near the pattern surface of the mask, Or, it is disposed on a conjugate plane with respect to the pattern plane or a plane in the vicinity of the conjugate plane, and opposes each other at a pattern area having an outer shape substantially parallel to the first direction and the second direction crossing each other on the pattern plane. Of the first and second pair of corners, the transmittance of the region corresponding to the first pair of corners to the illumination light for exposure gradually decreases outward along the first direction. Is set based on the first characteristic, or the second characteristic that gradually decreases outward along the second direction, and the transmittance for the illumination light for exposure in the area corresponding to the second pair of corners. Within the triangular area extending outward along the opposite direction of the pair of corners, the first characteristic gradually decreasing outward along the first direction and the outer characteristic along the second direction. And a dimming filter set based on the characteristic obtained by adding the gradually decreasing second characteristic to the second characteristic.

Further, the third method of manufacturing an exposure apparatus according to the present invention is a method of manufacturing an exposure apparatus for transferring a pattern to at least two regions where peripheral portions partially overlap each other on a substrate. That base of light A dimming filter that gradually reduces the amount of light on the plate at the portion where the at least two areas overlap, and a dimming filter to obtain at least one of the position information and rotation information of the dimming filter. A detection device for detecting at least one mark pattern provided in the optical filter is assembled in a predetermined positional relationship. BRIEF DESCRIPTION OF THE FIGURES

FIG. 1 is a schematic configuration diagram showing a projection exposure apparatus used in an example of an embodiment of the present invention. FIG. 2 is an enlarged view showing a configuration example of the reticle blind 4 in FIG. FIG. 3 is an enlarged perspective view showing the configuration of the movable stage of the positioning device 5 in FIG. FIG. 4 is a diagram showing a transmittance distribution of the density filter 55 in FIG. FIG. 5 is a view showing a projected image obtained by performing transfer while performing image splicing using the density filter 55 of FIG. FIG. 6 is an explanatory diagram of a method of removing an incomplete portion by using a reticle blind. FIG. 7 is a diagram showing a transmittance distribution of a density filter 56 according to another example of the embodiment of the present invention. FIG. 8 is a diagram showing a projection image obtained by performing transfer while performing screen splicing using the density filter 56 of FIG. FIG. 9 is a partially cut-away view showing a main part of an embodiment in which a dust-proof thin film is provided on the concentration filter 55. BEST MODE FOR CARRYING OUT THE INVENTION

Hereinafter, a first preferred embodiment of the present invention will be described with reference to the drawings.

FIG. 1 shows a schematic configuration of the projection exposure apparatus of this example. In FIG. 1, the illumination light (exposure light) IL for exposure emitted from the light source 1 at the time of exposure passes through a shutter (not shown). Input lens after being reflected by the mirror M1 Then, the light enters the illuminance uniforming optical system 2 including the optical integrator (fly-eye lens or rod lens) and the illuminance distribution is uniformed. An aperture stop (not shown) that determines the numerical aperture of the illumination light and, consequently, the coherence factor (σ value) is provided on the Fourier transform surface with respect to the pattern surface of the reticle R to be transferred in the illumination uniforming optical system 2. Are located.

The illumination light IL that has passed through the illumination uniforming optical system 2 passes through a relay lens 3 and enters a reticle blind 4 as a variable field stop. As an example, as shown in FIG. 2, the reticle blind 4 has four movable L-shaped light-shielding plates 4 1, 4 2 at four edges 4 1 Α, 4 1 Β, 4 2 A, 4 2. The illumination area (exposure angle of view) on the reticle R is determined by the variable aperture (hatched area) S surrounded by B.

In FIG. 1, the illuminating light IL that has passed through the reticle blind 4 passes through the density filter 55 and is provided with an illuminance distribution suitable for performing exposure (joint exposure) while connecting screens as described below. . In other words, the density filter 55 as the darkening filter has a transmittance distribution for making the integrated exposure amount of the joint portion at the time of the joint exposure equal to the integrated exposure amount of the other portions. (See below for details). The illumination light passing through the density filter 55 passes through a relay lens 6, a mirror 7 for bending the optical path, and a condenser lens 8, and illuminates the pattern surface (lower surface) of the reticle R on which the original pattern for transfer is formed. I do. Assuming that the surface conjugate to the pattern surface of the reticle R with respect to the relay lens 6 and the condenser lens 8 is a surface P1, the arrangement surface of the reticle blind 4 is close to the surface P1, and the filter forming surface of the density filter 55 is the surface P1. It is set at a position slightly deviated from 1 to the reticle R side. Due to the action of the density filter 55, the illumination light IL has an illuminance distribution that gradually decreases around the pattern area of the reticle R, and the pattern in the illumination area of the reticle R passes through the projection optical system PL. With a projection magnification of 3 (3 is 1 Z 4, 1 Z 5 etc.) The photoresist is projected onto the wafer W to which the photoresist has been applied.

The illumination optical system consists of a light source 1, a mirror Ml, an illuminance uniforming optical system 2, a relay lens 3, a reticle blind 4, a density filter 55, a relay lens 6, a mirror 7, a condenser lens 8, etc. I have. In this example, the i-line (wavelength 365 nm) of a mercury lamp is used as the illumination light IL. To increase the resolution, KrF (wavelength 248 ηm) is used as the illumination light IL. and a r F (wavelength 1 9 3 nm) excimer one laser light such as, F ₂ laser beam (wavelength 1 5 7 nm), ultraviolet light yo Ri short wavelength such as harmonics of a r ₂ laser or YAG laser It is desirable to use In the following, the Z axis is taken parallel to the optical axis of the projection optical system PL, the X axis is taken parallel to the plane of Fig. 1 in a plane perpendicular to the Z axis, and the Y axis is taken perpendicular to the plane of Fig. 1. .

At this time, the reticle R is held on the reticle stage 21, and the reticle stage 21 finely moves on the reticle base 22 in the X, Y, and rotation directions to position the reticle R. The position of the reticle stage 22 is measured by a laser interferometer incorporated in the reticle stage drive system 23, and the measured values and the control from the main control system 24 that supervise and control the operation of the entire apparatus The reticle stage drive system 23 controls the operation of the reticle stage 21 based on the information.

On the other hand, the wafer W is held on a wafer stage 25 via a wafer holder (not shown), and the wafer stage 25 steps on the wafer base 26 in the X and Y directions. The position of the wafer stage 25 in the XY plane is measured by the laser interferometer 27, and based on the measured values and the control information from the main control system 24, the wafer stage drive system 28 Control the operation of 25. Further, the wafer stage 25 adjusts the surface of the wafer W to the image plane of the projection optical system PL by an auto-force force method. Focus the light passing through the pinhole near wafer W on wafer stage 25 The illuminance sensor 63 for photoelectrically converting the light is fixed, and the detection signal of the illuminance sensor 63, which also functions as a mark detection system, is supplied to the main control system 24.

In addition, a reticle loader (not shown) for exchanging reticles on reticle stage 21 and a reticle library containing a plurality of reticles used for screen splicing are installed near reticle stage 21. The reticle R on the reticle stage 21 can be replaced with another reticle at high speed.

Then, at the time of exposure, when exposure of a reduced image of the pattern of the reticle R to a predetermined portion in one shot area on the wafer W is completed, the next shot on the wafer W is performed by the step movement of the wafer stage 25. The operation of moving the corresponding part of the area to the exposure area of the projection optical system PL and performing exposure is repeated in a step-and-repeat manner. After that, the reticle R is replaced with another reticle, and a reduced image of the pattern of the replaced reticle is exposed while screen shot is performed on each shot area on the wafer W, and thereafter, the reticle is replaced and connected. Exposure is repeated.

Instead of using a plurality of reticles, a large reticle is used as the reticle R, and a plurality of patterns sequentially selected by the reticle blind 4 from the pattern surface of this reticle are screen-connected while the screen is being screen-connected. May be transferred to each shot area.

The density filter 55 is arranged so as to be movable with six degrees of freedom on a base (not shown) (base 51 in FIG. 3) via the movable table 53 and the movable table 52. The system 24 is configured to be able to control the position and the inclination angle of the density filter 55 via the drive system 29. The movable table 52, 53, etc. constitutes a positioning device 5 for the density filter 55. Similarly, the shape of the opening of the reticle blind 4 can be set by the main control system 24 via a drive system (not shown). In the projection exposure apparatus of the present embodiment, when exposing to each shot area on the wafer W, the reduced images of the patterns of a plurality of reticles are exposed (joint exposure) while the screens are connected as described above. At this time, at the boundary between the reduced images of two adjacent patterns, a joint portion (connecting portion) of a predetermined width is overlapped and exposed, and in the area where the reduced images of the four patterns are adjacent, the four images are used. The overlapping portions of the corners of each corner of the reduced image are overlapped and exposed. This prevents the circuit pattern finally formed at the joint from being cut.

However, simply exposing a plurality of reduced images in a superimposed manner results in an increase in the integrated exposure amount compared to other portions. In this example, a reduced image of each reticle pattern is exposed using the density filter 55. At that time, the illuminance (and, consequently, the amount of exposure) at the periphery is reduced. Note that, at the time of such a joint exposure, depending on the reticle pattern, a joint portion thereof (for example, the first portion in one shot area exposed by the illumination light IL at the time of transfer of the first pattern, and the second pattern) The circuit pattern for the device does not necessarily exist in the one shot area exposed to the illumination light IL during transfer, and the connection part exists even if the circuit pattern exists. It may not be. Even in this case, the density fill 55 is effective for adjusting the integrated exposure amount to other areas. The following describes the method of supporting the concentration filter, the composition of the filter, and the method of use.

First, in Fig. 1, in this example, the illuminance distribution on the reticle R pattern surface is set by the transmittance distribution on the filter surface of the density filter 55, so that the filter surface is theoretically conjugate with the reticle R pattern surface. However, if there is a defect or foreign matter such as dust on the fill surface at this time, the defect or foreign matter may be transferred onto the wafer W together with the pattern of the reticle R. There is. Therefore, the filter surface of density filter Is located slightly away from the plane P1 toward the reticle side or the light source side (defocused position). However, in an environment where foreign matter on the fill surface can be reduced, the fill surface of the concentration fill 55 may be placed on the surface P1.

In this case, in order to improve the line width accuracy and the like of the device after the joint exposure, it is necessary to control the illuminance distribution of the illumination light IL at the joint with high accuracy, as well as the reticle R and the density filter. It is necessary to increase the positioning accuracy relative to 5. For example, in FIG. 5, when the projected image 30A and the projected image 30B adjacent in the X direction are overlapped and exposed at the joint 30AB, the width of the joint 30AB in the X direction is constant. The defocus amount for the surface P1 of the density filter 55 in FIG. 1 needs to be substantially equal in the periphery from the uniformity of the illuminance distribution.

Therefore, in order to set the relative positional relationship between the reticle R and the density filter 55 in a predetermined state, a positioning device 5 including movable tables 52 and 53 is used.

FIG. 3 shows an example of the configuration of a positioning device 5 for the density filter 55. In FIG. 3, the directions corresponding to the X, Y, and Z directions on the wafer stage 25 in FIG. X direction, Y direction, and Z direction. Then, a concentration filter 55 is provided so as to cover the opening 61 of the movable table 53, and a movable table 53 is provided so as to cover the opening of the movable table 52. Mounted on a base 51. In this case, the movable table 52 has three freedoms of translational movement in the X and Y directions with respect to the base 51 and rotation about the Z axis with respect to the base 51 by means of a three-axis drive motor 52A to 52C. The movable table 53 is moved in the Z direction with respect to the movable table 52 by three drive motors 53D, and is moved around the X axis and the Y axis. Fine adjustment of three degrees of freedom with rotation around It is arranged to be able to. The 6-axis drive motors 52A to 52C and 53D are provided with encoders for detecting the amount of movement or the rotation angle, respectively, and the detection results of these encoders are supplied to the drive system 29 in Fig. 1. ing.

In FIG. 1, the main control system 24 moves the illuminance sensor 63 to the exposure area of the projection optical system PL to start the irradiation of the illumination light IL for exposure, and then drives the wafer stage 25. The illuminance sensor 63 crosses the exposure area, and the detection signal of the illuminance sensor 63 is taken in accordance with the coordinates of the wafer stage 25, so that the position and rotation angle of the density filter 55 can be monitored. I do. At this time, for example, alignment marks are provided so as to correspond to both the density filter 55 and the reticle R, and the positions of the images of the alignment marks are also detected, whereby the density filter 55 is formed. The positional relationship (at least one of the positional relationship in the X direction, the positional relationship in the Y direction, and the relative rotation around the Z axis) between the projected image on the reticle R and the reticle R can be detected with high accuracy. . The main control system 24 controls the operation of the drive motors 52 A to 52 C and 53 D via the drive system 29 so that the detected positional relationship becomes a predetermined relationship. In this way, the position of the density filter 55 is determined.

At this time, the defocus amount at the corresponding position is obtained from the contrast of the images of the two alignment marks around the density filter 55, and the light of the density filter 55 is adjusted so that the defocus amount becomes equal. The position in the direction along the axis may be controlled. This makes it possible to adjust the position (defocus amount) and the one-dimensional tilt amount (rotation angle) of the density filter 55 in the Z direction. The illuminance sensor 63 may detect only the image of one alignment mark of the density fill 55 and adjust the defocus amount of the density fill 55, or the density fill 55 may be used. The illuminance sensor 63 detects the images of the alignment marks provided in at least three locations of In addition to the defocus amount of the fill 5 55, a two-dimensional tilt amount (rotation angle) may be adjusted. When detecting at least the position (defocus amount) and rotation angle in the Z direction of the position and rotation angle of the density filter 55, the density filter is used without using the alignment mark on the reticle R. Only the alignment mark 5 5 may be detected by the illuminance sensor 63 or the like. Further, a reference mark provided on reticle stage 21 may be used instead of the alignment mark on the reticle.

In FIG. 3, a manual drive micrometer head is used in place of the drive motors 52 A to 52 C and 53 D. The position may be adjusted.

Further, the density filter 55 may be replaced with a density filter having another transmittance distribution. When the density filter is exchanged in this manner, the movable table 53 may be pulled out of the movable table 52 using the handle 62 provided on the movable table 53.

Next, the transmittance distribution of the density filter 55 will be described. Fig. 4 (a) is a diagram showing the transmittance distribution of the fill area of the density fill area 55. In Fig. 4 (a), the directions corresponding to the X direction and Y direction in Fig. 1 are the X direction and the X direction, respectively. In the y direction. Also, the grid pattern formed in the fill area of the density fill 55 is a pattern drawn virtually to indicate the coordinates, and the transmittance in the fill area is actually 1 (100) %) And 0 (0%). That is, a large number of extremely fine dot patterns are formed in the filter portion so as to obtain a desired transmittance distribution by changing the size and density of each dot pattern depending on the position. The transmittance of 1 means the transmittance of the transparent substrate for the density filter 55 itself. Also generated from the dot pattern In consideration of the diffracted light and the optical characteristics of the illumination optical system (distortion, etc.), the size and density of the dot pattern are adjusted so that the desired illumination light distribution can be obtained on the reticle or wafer, and the light is transmitted. It is desirable to set a rate distribution.

In such a concentration filter, a light-shielding film such as chromium is formed on a transparent substrate, an electron beam resist is applied thereon, and a corresponding pattern is drawn thereon by an electron beam drawing apparatus. , Development, etching, resist stripping, and the like. Even if defects or continuous edges are formed in some regions during this manufacturing process, the defects are transferred to the wafer because the filter surface is defocused from the conjugate surface with the reticle R. It will not be done. Therefore, the defocus amount of the density filter 55 is determined by the drawing accuracy of the electron beam lithography apparatus at the time of manufacturing the density filter 55 and the tolerance for the error of the exposure dose on the wafer. Is also taken into account.

In the rectangular fill area of the density fill area 55 shown in Fig. 4 (a), overlapping exposure is performed during the splicing exposure. Joints at both ends in the X direction (overlapping area) 55a, 55b width Let a be the width of 55c and 55d at the joints (overlapping parts) at both ends in the a and y directions, and let b be the width in the X direction of the internal region surrounded by the joints 55a to 55d. a. , And the width in the y direction is b o. Also, assuming that the lower left vertex of the fill area of the rectangle is the origin of position X and position y, the range of the fill area in the X and y directions is as follows.

0≤x≤ 2 a + a o, 0≤ y≤ 2 b + b o

Then, assuming that the transmittance at the point P at the coordinates (X, y) in the filter area is T (x, y), the transmittance T (x, y) becomes the area (A i) (i = l ~ 9) Separately set to ΤΑ! Since the exposure Qi on the wafer is determined in proportion to the transmittance TAi, the transmittance TAi is replaced by the exposure Qi. It is also possible. In this case, 100% means the maximum exposure.

Area (A 1): 0≤x <a, 0≤y <b

T A! = 1 0 0 (x / a) · (y / b) [%] (1)

Area (A2): a≤x≤a + a. , 0≤y <b

TA ₂ = 1 0 0 (y / b) [%] (2)

Area (A3): a + a. X≤2a + a. , 0≤y

TA ₃ = 1 0 0 [1— {x-(a + a.) IZa] · (y / b) [%] (3) Area (A4): 0≤ x <a, b≤y≤ b + bo

TA ₄ = 1 0 0 (x / a) [%] (4)

Area (A5): a≤x≤a + a. , B≤y≤b + b.

Area (A6): a + a. X≤2a + a. , B≤y≤b + b.

TA ₆ = 1 0 0 [1-{x— (a + a.)} Za] [%] (6) Area (A 7): 0≤x <a, b + b. Y ^ 2b + b.

T AT = 1 0 0 (x / a)-[1 — {y-(b + b.) J / b] [%] (7) Area (A8): a≤x≤a + ao, b + b. Y≤2b + b.

TA8 = 1 0 0 [l— {y— (b + b.)} Zb] [%] (8) Area (A 9): a + ao <x≤ 2 a + ao, b + bo <y≤ 2 b + bo TA ₉ = 1 0 0 [1 — {x-(a + a.)} / a] · [1 — {y-

(b + b.)} Zb] [%] (9)

In the area outside of the phil evening, it is as follows.

T (x, y) = 0 [] (1 0)

In this case, the transmittance TA of the area (A 1), which is the lower left corner of the fill area, is one-dimensionally decreasing outward in the X direction (xZa) and linearly decreasing outward in the y direction. Multiplied by the original distribution (yZa) It is cloth. Also, the transmittances TA ₃ , TA? And TA _{9 in} the lower right, upper left, and upper right corners of the fill area are one-dimensionally reduced outward in the X direction and the y direction, respectively. Is multiplied by the one-dimensionally decreasing distribution to the outside. In addition, the transmittance T in the region along the line BB in Fig. 4 (a) is linear with respect to the position X as the position X changes from 0 to a as shown in Fig. 4 (b). From 0 to 1 (100%), similarly, the transmittance T in the region along the CC line in Fig. 4 (a) changes from 0 at position y as shown in Fig. 4 (c). As it changes to b, it changes linearly from 0 to 1 (100%) with respect to position y.

In this example, the pattern of the reticle R of FIG. 1 is illuminated through the density filter 55 having the transmittance distribution of FIG. Expose a part of the area. After that, the reticle on the reticle stage 21 is sequentially replaced with another reticle, and the wafer W is moved by a predetermined amount through the wafer stage 25, and then the pattern of the replaced reticle is filtered by the density filter 55. 4A, the reduced image of the pattern is exposed to another portion of the shot area on the wafer W, and the joint 55a of FIG. A region corresponding to ~ 55d (also referred to as a "joint") is overlaid and exposed. As described above, the reduced images of a plurality of reticle patterns are transferred onto the relevant shot area on the wafer W while the screens are connected in the X and Y directions. An almost uniform product exposure amount is provided over the entire area.

FIG. 5 shows a large projected image that is exposed to one shot area on the wafer W in FIG. 1 by the exposure for performing the screen splicing of the present example. The projected images 30 A, 30 B, 30 C, and 30 D of the rectangles become the joints 30 AB, 3 at the boundary in the X direction. 0 CD and the joint 30 AC, 30 BD at the boundary in the Y direction are exposed in a double overlapping manner. Furthermore, at the rectangular joint 31 where the four projected images 30A to 30D are adjacent, the corners of the rectangles of the four projected images 30A to 30D are superimposed four-fold. Exposed. At this time, if the rotation error between the density file 55 in FIG. 4A and the reticle on the reticle stage 21 in FIG. 1 cannot be removed by the positioning device 5, the rotation error is canceled. Thus, reticle stage 21 may be rotated, and the coordinate system of wafer stage 25 may be corrected by the rotation error, and wafer W may be obliquely moved stepwise based on the corrected coordinate system. As a result, it is possible to reduce the exposure error (dose error) at the joint.

When the exposure amount of each part of the projected image of FIG. 5 is evaluated, first, in the central region of the projected images 30 A to 30 D, which is denoted by reference signs A to D, the transmittance of the density filter 55 is 10. Since it is 0%, the exposure amount is 100%.

Hereinafter, it will be described that the integrated exposure amount of the joints 30 AB, 30 BD, 30 CD, 30 AC, and 31 becomes 100%. For simplicity of explanation, assume that the projection magnification from the density filter 55 in Fig. 4 (a) to wafer W in Fig. 5 is 1, the width in the X direction of the joints 30 AB, 30 CD is a, The width of the part 30 BD, 30 AC in the Y direction is b. If the point P 3 in FIG. 5 is taken as the origin of the coordinates (X, Y) and the coordinate of the point P is (X, Y) when the point P 3 is the origin, the projected images 30 A, 30 The coordinates (Χ.Λ, ΥΛ), (ΧΒ, ΥΒ), (Xc, Yc), (Χη, Yu) when the lower left point of B, 30C, 30D is the origin are as follows. It is.

(Χ _Λ , Υ _Λ ) = (X + (a + a.), Y) (1 1)

( _Xc , Yc) = (X + (a + a.), Y + (b + b.)) (13)

(Χ _υ , Υ ") = (Χ, Υ + ^ + ΐ3.)) (1 4) TAi (A), ΤΑ; (B), the values expressing the transmittance TA i on the coordinates (Χ _Λ , ΥΛ), (Χ, ΥΒ), (Xc, Yc) and (XD, YD), respectively. TA i (C) and TA i (D).

At this time, let AB (A) and AB (B) be the exposure amounts of the projected images 3OA and 30B at the joint 30AB, and let the integrated exposure amount obtained by adding these exposure amounts be AB. From Eqs. (6) and (4), they are as follows. However, the exposure amount in the areas denoted by the reference signs A to D is 100%.

AB (A) = T Ae (A)

= 100 [1 {X + (a ten a.) One (a + a.)} / A]

= 1 00 (1 X / a) [%]

AB (B) = TA ₄ (B) 100 (X / a) [%]

AB = AB (A) + AB (B)

= 1 00 [%] (1 5)

Similarly, exposure of the projected images 30B and 30D at the joint 30BD: B

D (B) and BD (D) and their integrated exposure amount BD are as follows.

BD (B) = TA ₂ (D) = 100 (Y / b) [%]

BD (D) = T As (B)

= 1 00 [1-{Y + (b + b.;) One (b + b.)} Zb]

= 1 00 (1-Y / b) [■%]

BD two BD (B) + BD (D)

= 1 00 [] (1 6)

Similarly, the exposure amounts CD (C) and CD (D) of the projected images 30C and 30D in the joint 30CD and the integrated exposure amount CD thereof are as follows. CD (C) = TA ₆ (C)

= 1 00 [1-{X + (a + a.) One + a. ) A] = 1 00 (1 XZ a) [%]

CD (D) = TA ₄ (D) = 100 (X / a) [%]

CD = CD (C) + CD (D)

= 1 00 [%] (1 7)

Similarly, the exposure amounts A C (A) and AC (C) of the projected images 3 OA and 30 C at the joint 30 AC and the integrated exposure amount AC are as follows.

AC (A) = TA ₂ (A) = 100 (Y / b) [%]

= 1 00 [1— {Y + (b + b.) One (b + b.)} Zb] = 100 (1-Y b) [%]

A C = A C (A) + A C (C)

= 1 00 [%] (1 8)

The exposure amounts of the four projected images 3OA, 30B, 30C and 30D at the joint 31 are ABCD (A), ABCD (B), ABCD (C) and ABCD (D). Assuming that the integrated exposure amount is ABCD, these are as follows from Eqs. (3), (1), (9), and (7).

ABCD (A) = TA ₃ (A)

= 1 00 [1— {X + (a + a.)-(A + a.)}

Za] · (Y / b)

= 1 00 (1 -X / a)

= 1 00 (X / a)

ABCD (C) = TA ₉ (C) = 1 0 0 [1 1 {X + (a + a.) 1 (a + a.)} / A] · [1 — {Y + (b + b.) — (B + b.)} Zb]

= 1 0 0 (1 -X / a)

= 1 0 0 (XZ a) '[l — {Y + (b + b.) —

(b + b.)} Zb]

= 1 0 0 (X / a) · (1-Y / b)

AB CD = AB CD (A) + AB CD (B) + AB CD (C) + A B CD (D)

= 1 0 0 {(1 -X / a) (Y / b) + (X / a) (Y / b) + (1 -X / a) (1-Y / b) + (X / a) ( 1-Y / b)}

= 1 0 0 [] (1 9)

From Equations (15) to (19), by using the density filter 55 shown in FIG. 4 (a), the four projected images 30A to 30D are connected to the rectangular joint 3 It can be seen that an integrated exposure amount of 100% can be uniformly obtained in all the regions after the bridge exposure including 1. Subsequent processes such as development, etching, and resist stripping form a circuit pattern having a uniform line width in each shot area on the wafer W. By repeating such a circuit pattern forming process, a large-area, high-performance semiconductor device can be mass-produced.

However, in FIG. 5, the peripheral area 11 is not subjected to the overlay exposure, so that the exposure amount gradually decreases outward as it is. Therefore, a region in the joint portion where multiple exposure is not performed is a region outside the pattern region, and the region may be shielded from light by the reticle blind 4 in FIG.

Figure 6 shows such a case where light is blocked by the reticle blind 6. However, as shown in FIG. 6, when the projected images of the four patterns are joined to expose an arbitrary exposure area on the wafer, the outermost area 11 lacks the other projected image to be superimposed. As a result, the amount of exposure is insufficient and the result is incomplete.

In this case, for example, in the reticle, light-shielding bands corresponding to the frame-shaped areas 9A to 9D in FIG. 6 are provided, and the light-shielding plates 41, 42 of the reticle blind 4 in FIG. The main control system 24 drives the reticle blind 4 via a drive unit (not shown) so that the shadow (image) of B, 42A, 42B (see Fig. 2) falls within the range of the light-shielding band. I do. As a result, the illumination light of the above-mentioned incomplete portion is shielded, so that the exposure amount on the wafer does not become uneven.

Also, in FIG. 1, since the fill surface of the density filter 55 exists near the plane P 1 conjugate with the pattern surface of the reticle R, the reticle blind 4 does not mechanically interfere with the density filter 55. Then, it is retracted from the plane P1 to a position shifted in the optical axis direction of the illumination optical system. However, in order to prevent the reticle blind 4 from deviating from the plane P1, which is conjugate with the pattern plane, a relay optical system that relays the plane P1 to another conjugate plane is arranged, and the conjugate plane is placed on the relay optics. Reticle blind 4 may be arranged.

In these cases, when the reticle blind 4 is defocused, the width of the light-shielding band provided on the reticle R depends on the amount of blur of the edge of the light-shielding plate of the reticle blind 4 due to this defocus, the control error of the light-shielding plate, and However, it is necessary to comprehensively consider the mechanical accuracy of the light shielding plate, the aberration of the optical system from the reticle blind 4 to the reticle R, and the distortion amount of the optical system.

In addition, the reticle blind 4 may be replaced with the reticle R pattern surface, for example. (Lower surface). Conversely, the density filter 55 may be arranged on the bottom surface of the pattern surface of the reticle R, and the reticle blind 4 may be arranged on a plane P1 conjugate with the pattern plane. Of course, the reticle blind 4 or the density filter 55 may be arranged on the opposite side of the reticle R from the pattern surface (illumination optical system side). When the projection optical system PL re-images the intermediate image of the reticle pattern on the wafer, the reticle blind 4 or the density filter 55 is formed from the predetermined surface where the intermediate image is formed in the projection optical system PL. The light amount distribution may be shifted, in short, as long as the light amount distribution in which the illumination light amount on the top gradually decreases outward.

Even when the filter surface of the density filter 55 is defocused by an appropriate amount with respect to the surface P1 as described above, if the cleanness of the surrounding environment is low, the filter surface Foreign matter, such as dust, having a size exceeding the allowable range may adhere to the wafer, and the foreign matter may be transferred onto the wafer W through the reticle R. In order to prevent this, it is desirable to provide a thin film (pellicle as a dust-proof film), such as cellulose, which does not affect optically, so as to protect the filler surface.

FIG. 9 shows an embodiment in which a dust-proof thin film 58 is installed via a rectangular metal frame (pellicle frame) 57 on the fill surface P2 of the concentration fill 55 in this way. In FIG. 9, the filter surface P 2 of the density filter 55 and the thin film 58 are arranged so as to sandwich the plane P 1 conjugate with the pattern surface of the reticle. The reticle blind 4 is arranged close to the thin film 58. As a result, no foreign matter adheres to the fill surface P2, and the image of the foreign matter adhered to the thin film 58 is defocused and projected on the reticle, and has no adverse effect. In addition, prepare a foreign substance inspection machine separately and adjust the allowable range for the fill surface P2 or thin film 58 as necessary. It is desirable to replace the density filter 55 with a density filter that does not have any other foreign matter if such foreign matter is found.

Note that, instead of the thin film 58, a glass plate that is permeable to the illumination light IL having a thickness about the distance between the thin film 58 and the fill surface P2 may be arranged. In this case, the glass plate may be fixed in close contact with the filling surface P2.

When reticle blind 4 is placed on a conjugate plane with a reticle different from that on plane P1, contrary to the arrangement in FIG. It may be arranged facing.

Further, a hole communicating with the outside air is formed in a part of the frame 57 so that the thin film 58 is not deformed by a pressure change. A chemical filter as well as a HEPA filter is provided in the hole to prevent ions and silicon-based organic substances from entering the fill surface P2.

Next, another example of the preferred embodiment of the present invention will be described with reference to FIGS. In this example as well, a projection exposure apparatus having the same configuration as the projection exposure apparatus of FIG. 1 is used to expose a projection image of a pattern of a plurality of reticles onto a wafer while performing screen joining. However, this example is different from the first embodiment in that the density filter 55 shown in FIG. 7A is used instead of the density filter 55 shown in FIG. 1 (FIG. 4A). Other than this, the configuration of the above-described embodiment including the modified examples can be applied. In the following, portions corresponding to FIGS. 4 and 5 in FIGS. 7 and 8 are denoted by the same reference numerals, and are obtained by performing exposure while performing the transmittance distribution of the density filter 56 and screen jointing. The exposure distribution will be described.

Fig. 7 (a) shows the fill area of the density fill 56 of this example. In Fig. 7 (a), the joints 56 a , 5 6b the width of a, the joints at both ends in the y direction 5 6c, 5 Let b be the width of 6 d, and let a be the width in the x direction of the internal area surrounded by the joints 56 a to 56 d. The width in the y direction b. And If the lower left vertex of the rectangular fill area is the origin of the position X and position y, the range of the fill area in the X and y directions is as follows.

0≤x 2 a + a. , 0≤ y≤ 2 b + b o

Then, assuming that the transmittance at the point P at the coordinates (X, y) in the filter area is T (x, y), the transmittance T (X, y) is expressed in the area (B i) (i = l ~ ll) Set separately to TB i. Also in this example, it is possible to replace the transmittance TB i with the exposure amount Q i.

Area (B 1): 0≤x <a, 0≤y <b, and (xZa) + (y / b)> 1

TB] = 1 0 0 {(x / a) + (y / b)-1} [%] (2 1) area (B 2): a≤x≤a + a. , 0≤y

TB ₂ = 100 (y / b) [%] (2 2) area (B 3): a + a. X≤2a + a. , 0≤y <b and bx + ay≤b (2 a + ao)

TB ₃ = 100 (y / b) [] (2 3) region (B 4): a + a. X≤2a + a. 0≤ y <b, and bx + ay> b (2 a + ao)

T B 4 = 1 0 0 [1-{x-(a + a.)} A] [%] (24) Area (B 5): 0≤ x≤ a, b≤y≤ b + b o

TB ₅ = 1 0 0 (x / a) [%] (2 5) area (B 6): a≤ x≤ a + ao, b≤ y≤ b + bo

TB _fi = 100 [%] (26) area (B7): a + a. X ≤ 2 a + ab ≤ y ≤ b + bo TB ₇ = 1 0 0 [1 1 {x— (a + a.) A] [%] (2 7) Area (B8): 0≤x, a, b + b. Y ^ 2 b + b. , And bx + ay≤ a (2 b + bo)

TB ₈ = 100 (x / a) [%] (28) Area (B 9): 0≤x <a, b + b. Y≤2 b + b. , And bx + ay <a (2 b + b.)

TB ₉ = 100 [l— {y—b + b. )} Zb] [%] (29) Area (BIO): a≤x≤a + a. , B + b. Y≤2 b + b.

T B i o = 1 00 [1 one {y—b ten! 3. )} C%] (30) Area (Bl l): a + a. <x ^ 2 a + a b + b. Y≤2 b + b o, and b x + a y≤3 a b + a bo + a b

T B!! = 1 0 0 [1-{x-(a + a.)} A-one {y— (b + b)} / b] [%] (3 1)

The other areas are as follows.

T (x, y) = 0 [%]

In this case, the transmittance TB, of the area (B 1), which is the corner of the lower left triangle of the filter area, has a distribution (x / a) that decreases one-dimensionally outward in the X direction, And the distribution (yZa) that decreases one-dimensionally. The transmittance T of this region (B1) along the DD line decreases linearly outward along the oblique position y ', as shown in Fig. 7 (d). Also, the transmittance TB ^ of the upper right corner of the filter region (B11) is set symmetrically with the transmittance TB !. Further, the triangular region (B 3) corners of the rectangular bottom right fill evening area adjacent divided into (B 4), the transmittance TB ₃ and TB ₄ corresponding to outside, respectively therewith y-direction One-dimensionally decreasing distribution and one-dimensionally decreasing distribution outward in the X direction. Similarly, the corner of the upper left rectangle of the filter area is also divided into adjacent triangular areas (B 8) and (B 9), and the corresponding transmittance TB 8 and TB ₉ are set symmetrically with the transmittances TB ₄ and TB ₃ . In addition, the transmittance T in the region along the line BB in Fig. 7 (a) changes linearly from 0 to 1 (100%) with respect to the position X as shown in Fig. 7 (b). The transmittance T in the region along the CC line in FIG. 7 (a) changes linearly from 0 to 1 (100%) with respect to the position y as shown in FIG. 7 (c).

In this example, the reticle on the reticle stage 21 shown in Fig. 1 is illuminated through the density filter 56 having the transmittance distribution shown in Fig. 7 (a), and a reduced image of the reticle pattern is formed while screen joining is performed. Is exposed to the shot area on the wafer W.

FIG. 8 shows a large projection image exposed to one shot area on the wafer W in FIG. 1 by the exposure for performing the screen splicing in this example. 3 A, 3 2 B, 32 C, 32 D force Joints at the boundary in the X direction 3 2 AB, 32 CD, and joints at the boundary in the Y direction 3 2 AC, Exposure is performed so that 32 BDs are superimposed twice. Furthermore, the rectangular joint where the four projected images 32A to 32D are adjacent to each other is divided into triangular joints 33 and 34 across the oblique boundary line 35, and the joint 33 projects at A part of the image 32A, 32B, 32D is exposed in a triple overlapped manner, and the projected image 32A, 3

A part of 2C and 32D is exposed in triple overlap.

When the exposure amount of each part of the projected image in FIG. 8 is evaluated, first, in the central areas of the projected images 32 A to 32 D, which are denoted by reference signs A to D, the transmittance of the density filter 56 is 10. Since it is 0%, the exposure amount is 100%. In addition, the joint 3 2 AB,

The integrated exposure amount of 32 BD, 32 CD, and 32 AC is 100% as in the embodiment of FIG. Hereinafter, it will be described that the integrated exposure amount of the joints 33 and 34 is also 100%. Similarly to FIG. 5, the point P 3 in FIG. 8 is set as the origin of the coordinates (X, Y), and the coordinates of the point P when the point P 3 is set as the origin is (X, Y). Y), the coordinates (ΧΛ, ΥΛ), (Χ, ΥΒ) (X Yc), with the lower left point of the projected images 32A, 32B, 32C, 32D as the origin. (Χο, Υη) are as follows.

(ΧΑ Y _A ) = (X + (a + a.) Y) (3 2)

(Xc Yc) = (X + (a + ao) Y + (b + b.)) (34) (XD YD) = (X, Y + (b + b)) (35) Transmittance TB above Let i be the coordinates (Χ _Λ , ΥΛ), (X, _s , YR) (Xc,

Yc) and (XD, YD) The values expressed above are denoted as TB i (A), T B i (B), T B i (C), and T B i (D), respectively.

At this time, the boundary 35 between the joints 33 and 34 is represented by (X / a) + (Y / b) = 1. Further, assuming that the exposure amounts of the projected images 32A, 32C, 32D at the joint 34 are ABCDl (A), ABCDl (0, ABCDl (D), these are as follows.

ABCDl (A) = T B 3 (A)

= 1 0 0 (Y / b)

ABCDl (C) = TB n (C)

= 1 0 0 [1 {X + (a + a (a + a.)} / A] 1 {Y + (b + b.) — (B + b.)} / B]

= 1 0 0 (1-XZ a-Y / b)

ABCDl (D) 2 TB ₈ (D)

= 1 0 0 (XZ a)

Therefore, the integrated exposure amount ABCDl at the joint portion 34 is as follows. ABCDl = ABCD1 (A) + ABCDl (0 + ABCDl (D)

= 1 0 0 {(Y / b) + (1 -X / a-Y / b) + (XZ a)

= 1 0 0 [%] (3 6) Similarly, assuming that the exposure amounts of the projected images 32A, 32B, and 32D at the joint 33 are ABCD2 (A), ABCD2 (B), and ABCD2 (D), these are as follows.

ABCD2 (A) = TB ₄ (A)

= 1 0 0 [1 — {X + (a + a.) One (a + a.)

= 1 0 0 (1 -X / a)

ABCD2 (B) = TB, (B)

= 1 0 0 (X / a + Y / b-1)

ABCD2 (D) = TB ₉ (D)

= 1 0 0 [1-{Y + (b + b.) One (b + b o)} Zb]

= 1 0 0 (1-Y / b)

Therefore, the integrated exposure amount ABCD2 at the joint 33 is as follows. ABCD2 = ABCD2 (A) + ABCD2 (B) + ABCD2 (D)

= 1 0 0 {(1 -X / a) + (X / a + Y / b-1) + (1-

= 1 0 0 [%] (3 7)

Therefore, in the embodiment of FIG. 8 as well, the same amount of exposure light as that of the other areas can be obtained in the rectangular areas (joints 33, 34) where the four projected images 32A to 32D are adjacent. Therefore, the uniformity of the exposure amount is maintained over the entire projected image.

In the above embodiment, the present invention is applied to the case of manufacturing a semiconductor device, a liquid crystal display, a plasma display, and the like by the bridge exposure method, but the present invention uses a peak reticle as a mask in the bridge exposure method. It can be applied to the case of manufacturing. In this case, the original pattern obtained by enlarging the reticle pattern is divided into a plurality of parts, and the divided original patterns are drawn on a plurality of master reticles. Then, the reduced images of the patterns of the master reticle are glass-connected by the density exposure method 55, 56 as in the above-described embodiment. Transfer onto a mask substrate such as a substrate. A light-shielding film such as chromium is formed on the mask substrate in advance, and a photoresist is applied thereon. Therefore, by performing development, etching, resist stripping, and the like after the bridge exposure, a single reticle is manufactured with high accuracy and high line width uniformity. In an electron beam exposure apparatus and an EUV exposure apparatus described later, a silicon wafer or the like is used as a mask substrate of a working reticle. Particularly in EUV exposure equipment, a reflective working reticle is used.

In the projection exposure apparatus according to the above-described embodiment, when short-wavelength ultraviolet light in the vacuum ultraviolet region such as ArF excimer laser light is used as the illumination light IL, the optical path of the illumination optical system is Gases with high transmittance, such as nitrogen gas (N ₂ ) and helium gas (H e), are purged. In this case, in FIG. 9, the space surrounded by the thin film 58, the frame 57, and the concentration filter 55 is filled with a gas having a high transmittance, or through an opening provided in the frame 57. It is desirable to flow the gas.

In FIG. 1, when a rod integrator is used as the illuminance uniforming optical system 2, a reticle blind 4 is arranged close to the exit surface, and a density filter 5 δ is close to the reticle blind 4. Can be provided. Alternatively, a fill surface of the density fill 55 may be arranged on a surface conjugate with the exit surface between the rod / integre and the reticle, or on a surface slightly shifted from this surface.

Further, it is desirable that the density filter 55 can be changed according to the size and shape of the pattern area on the reticle. At this time, in order to automatically perform the exchange, it is preferable that a plurality of density filters having different sizes be fixed to an evening lett plate on the positioning device 5 or the like.

The device for detecting the alignment mark of the density filter 55 (or 56) is not limited to the illuminance sensor 63. An optical system having at least a light receiving unit may be provided and used on the wafer stage 25 separately from 3, or a dedicated optical system may be incorporated in the illumination optical system. In addition, the above-mentioned illumination light IL may be used as the detection light of the alignment mark, or a light source different from the light source 1 may be prepared so that light having substantially the same wavelength as the illumination light IL is used. You may.

In the above embodiment, the present invention is applied to a batch exposure type projection exposure apparatus. However, the present invention can be similarly applied to a case where a connection exposure is performed by a proximity type exposure apparatus. it can. Further, the present invention can be applied to a case where a connecting exposure is performed by a scanning exposure type projection exposure apparatus such as a step-and-scan method. The present invention is also applicable to a case where a bridge exposure is performed by an EUV exposure apparatus that uses extreme ultraviolet light (EUV light) such as soft X-ray or X-ray having a wavelength of about 5 nm to 15 nm as an exposure beam. be able to. When EUV light is used, since there is almost no transmissive material, a concentration film (dimming filter) is used as a reflection film (for example, a multilayer of molybdenum and silicon) with a predetermined reflectance distribution on a reflective substrate. Alternatively, a reflective filter formed with a film or a multilayer film of molybdenum and beryllium) may be used.

In addition, a single-wavelength laser in the infrared or visible range oscillated from a DFB semiconductor laser or fiber laser as illumination light for exposure may be used, for example, by using Erbium (Er) (or both Erbium and Ytterbium (Yb)). ) May be amplified by a doped fiber amplifier, and a harmonic converted to a wavelength of ultraviolet light using a nonlinear optical crystal may be used. For example, if the oscillation wavelength of a single-wavelength laser is in the range of 1.544 to 1.553 m, the 8th harmonic in the range of 193 to 194 nm, that is, almost the same as the ArF excimer laser Assuming that ultraviolet light having the same wavelength is obtained and the oscillation wavelength is in the range of 1.57 to 1.58 m, the 10th harmonic in the range of 157 to 158 nm, that is, F ₂ Ultraviolet light having substantially the same wavelength as that of one laser is obtained.

In addition, an illumination optical system composed of an exposure light source and an illuminance uniforming optical system, and a projection optical system are incorporated in the main body of the exposure apparatus for optical adjustment, and a reticle stage and a wafer stage composed of many mechanical parts are mounted on the exposure apparatus. The projection exposure apparatus of the above-described embodiment can be mounted by attaching it to the main body, connecting wiring and piping, attaching the density filter 55 of the above-described embodiment, and performing overall adjustment (electrical adjustment, operation confirmation, etc.). Can be manufactured. It is desirable to manufacture the exposure apparatus in a clean room where the temperature, cleanliness, etc. are controlled.

The present invention is not limited to the above-described embodiment, and can take various configurations without departing from the gist of the present invention. In addition, all disclosures, including the specification, claims, drawings, and abstract, of Japanese Patent Application No. 11-83 178, filed March 26, 1999, are: Exactly quoted and incorporated here. Industrial applicability

According to the present invention, when a plurality of patterns are transferred while performing two-dimensional screen splicing, the integrated exposure amount in a portion where the four patterns are adjacent is substantially equal to the integrated exposure amount in other portions. There are advantages to doing the same. Therefore, a pattern of a large semiconductor device or the like can be exposed with high precision, and no defect occurs at a joint portion in a large device to be finally manufactured.

Further, according to the method of manufacturing a mask of the present invention, even a large-sized mask can be manufactured with high accuracy and high throughput without causing a defect even when a large-sized mask is formed by screen joining.

Claims

The scope of the claims

1. In an exposure method for exposing a pattern larger than each of the above patterns on the substrate by splicing and exposing a plurality of patterns on the substrate,

Exposure is performed by joining a plurality of patterns together in a first direction and a second direction that intersect with each other such that a part of each region overlaps,

In the area where the four patterns are adjacent to each other, the corners of the four patterns are overlapped with each other and exposed,

When each of the four patterns is exposed, a first characteristic in which the exposure amount of the corner of the pattern gradually decreases outward along the first direction, and gradually decreases outward along the second direction. An exposure method characterized in that the exposure method is set based on a characteristic obtained by multiplying the characteristic by a decreasing second characteristic.

2. When exposing the corner of one of the four adjacent patterns,

The width of the corner in the first and second directions is a and b, respectively, and the coordinates that increase inside the corner along the first and second directions with the vertex of the corner as the origin point. Let X and y be respectively

2. The exposure method according to claim 1, wherein the exposure amount at the corner is set to a value proportional to (xZ a) · (y / b).

3. An exposure method for exposing a pattern larger than each of the above patterns on the substrate by splicing and exposing a plurality of patterns on the substrate,

In the area where the four patterns are adjacent, the first and Among the second pair of patterns, the first pair of patterns are exposed by overlapping their rectangular corners with each other,

An exposure method comprising exposing the second pair of patterns such that corners of respective triangles are adjacent to each other inside the corners of the rectangle.

4. When exposing the second pair of patterns, a first characteristic in which the exposure amount of the corners of the triangle of the pattern is gradually reduced outward along the first direction, and the second direction. Is set based on the characteristic obtained by adding the second characteristic that gradually decreases outward along

4. The method according to claim 3, wherein, when exposing the first pair of patterns, the exposure amount of the corner portion of the patterns is set based on a characteristic of gradually decreasing one-dimensionally outward. Exposure method.

5. In an exposure apparatus that transfers a mask pattern onto a substrate,

An illumination optical system for illuminating the mask,

A field stop that is arranged in the illumination optical system at a position substantially conjugate to the pattern surface of the mask, and sets a field of illumination on the mask; and a substrate stage that positions the substrate.

The mask is arranged on a surface near the pattern surface of the mask, or on a common surface with respect to the pattern surface or on a surface near the pattern surface, and has an outer shape substantially parallel to the first direction and the second direction intersecting the pattern surface. A first characteristic in which the transmittance of the region corresponding to at least one corner of the pattern region having at least one corner to the illumination light for exposure gradually decreases outward in the first direction, and along the second direction. A dimming filter set based on a characteristic obtained by multiplying a second characteristic that gradually decreases outward,

An exposure apparatus comprising:

6. In an exposure apparatus for transferring a mask pattern onto a substrate,

An illumination optical system for illuminating the mask, A field stop that is arranged in the illumination optical system at a position substantially conjugate to the pattern surface of the mask, and sets a field of illumination on the mask; and a substrate stage that positions the substrate.

And a dimming filter disposed on a surface near the pattern surface of the mask or a common surface with respect to the pattern surface or a surface near the pattern surface. Among the first and second pair of corners that are obliquely opposed to each other in the pattern region having an outer shape substantially parallel to the intersecting first and second directions, Based on a first characteristic in which the transmittance for the illumination light for exposure in the corresponding region gradually decreases outward along the first direction, or based on a second characteristic that gradually decreases outward along the second direction. Set,

The transmittance for the illumination light for exposure in the region corresponding to the second pair of corners is set to be within the triangular region extending outward along the direction in which the pair of corners face each other. An exposure apparatus characterized in that it is set based on a characteristic obtained by adding a first characteristic that gradually decreases outward in one direction and a second characteristic that gradually decreases outward in a second direction.

7. a positioning member for positioning the neutral density filter in a plane perpendicular to the optical axis of the illumination light for exposure;

7. The exposure apparatus according to claim 5, further comprising: a dustproof film disposed at a predetermined distance from a filter surface of the neutral density filter.

8. The dimming filter is arranged close to the field stop in the illumination optical system so that the dust-proof film is set between the field stop and the field stop. 7. The exposure apparatus according to 7.

9. The dimming filter is disposed within the illumination optical system at a predetermined distance from a conjugate plane with respect to a pattern surface of the mask. 9. The exposure apparatus according to claim 8, wherein:

10. A mark detection system for detecting a mark on the mask and a mark on the darkening filter, and a driving mechanism for moving the darkening filter based on a detection result of the mark detection system. 7. The exposure apparatus according to claim 5, further comprising:

11. The exposure apparatus according to claim 10, wherein the drive mechanism moves the dimming filter in a plane perpendicular to an optical axis of the illumination optical system.

12. The exposure apparatus according to claim 10, wherein the neutral density filter is arranged at a predetermined distance from a conjugate plane with respect to a pattern surface of the mask in the illumination optical system.

13. The drive mechanism is capable of moving the dimming filter along the optical axis of the illumination optical system to adjust a positional relationship with the conjugate plane, and perpendicular to the optical axis. 13. The exposure apparatus according to claim 12, wherein the exposure apparatus can be inclined with respect to a plane.

14. An exposure method for transferring a pattern to at least two regions where peripheral portions partially overlap each other on a substrate,

In order to obtain at least one of the position information and the rotation information of the dimming filter, which gradually reduces the amount of illumination light irradiated on the pattern on the substrate at the portion where the at least two regions overlap. An exposure method comprising detecting at least one mark pattern provided in the darkening filter.

15. The exposure method according to claim 14, wherein a relative relationship between a mask on which the pattern is formed and the dimming filter is adjusted based on the obtained information.

16. Based on the obtained information, the position and inclination of the dimming filter in the optical axis direction in the optical system in which the dimming filter is arranged are reduced. 15. The exposure method according to claim 14, wherein one of them is adjusted.

17. An exposure apparatus for transferring a pattern to at least two regions where peripheral portions partially overlap each other on a substrate,

A dimming filter that gradually reduces the amount of illumination light applied to the pattern on the substrate at a portion where the at least two regions overlap; and at least one of position information and rotation information of the dimming filter. An exposure apparatus, comprising: a detection device for detecting at least one mark pattern provided in the darkening filter to obtain one of them.

18. The exposure according to claim 17, further comprising: an actuating unit for driving the dimming filter in order to adjust at least one of a position and a rotation of the dimming filter. apparatus.

19. The exposure according to claim 17, wherein the detection device detects at least one of relative position information and relative rotation information between the dimming filter and a mask on which the pattern is formed. apparatus.

20. The exposure apparatus according to claim 17, wherein the dimming filter is arranged so as to be shifted from a pattern surface of a mask on which the pattern is formed or a conjugate surface thereof.

21. The exposure apparatus according to any one of claims 5 to 13 and 17 to 20, wherein a device pattern is formed by performing screen joining of a plurality of masks and patterns. mask.

22. A device manufacturing method comprising a step of transferring a device pattern onto a photosensitive substrate using the exposure method according to any one of claims 1 to 4, and 14 to 16.

23. A device manufacturing method characterized by including a step of transferring a device pattern onto a photosensitive substrate using the exposure apparatus according to any one of claims 5 to 13 and 17 to 20. .

24. A method of manufacturing a mask using the exposure method according to any one of claims 1 to 4, and 14 to 16,

A method of manufacturing a mask, comprising a step of transferring a plurality of mask patterns onto a mask substrate by using the exposure method while performing screen connection.

25. In a method of manufacturing an exposure apparatus for transferring a mask pattern onto a substrate,

An illumination optical system for illuminating the mask,

The mask is arranged on a surface in the vicinity of the pattern surface of the mask, or on a common surface with respect to the pattern surface or on a surface in the vicinity thereof, and has an outer shape substantially parallel to the first direction and the second direction intersecting the pattern surface. A first characteristic that gradually reduces the transmittance of the region corresponding to at least one corner of the pattern region having the illumination light for exposure to the outside along the first direction, and the second characteristic along the second direction. A dimming filter set based on a characteristic obtained by multiplying a second characteristic that gradually decreases outward,

Are assembled in a predetermined positional relationship.

26. In a method of manufacturing an exposure apparatus for transferring a mask pattern onto a substrate,

An illumination optical system for illuminating the mask,

The mask is arranged on a surface in the vicinity of the pattern surface of the mask, or on a common surface with respect to the pattern surface or on a surface in the vicinity thereof, and intersects the pattern surfaces. Among the first and second pair of corners obliquely opposed to each other in the pattern area having an outer shape substantially parallel to the first direction and the second direction to be inserted, Based on the first characteristic in which the transmittance for the illumination light for exposure in the corresponding area gradually decreases outward in the first direction, or based on the second characteristic in which the transmittance gradually decreases outward along the second direction. The transmittance for the illumination light for exposure in the region corresponding to the second pair of corners is set within a triangular region extending outward along the direction in which the pair of corners oppose each other. A dimming filter set based on a characteristic obtained by adding a first characteristic that gradually decreases outward along the first direction and a second characteristic that gradually decreases outward along the second direction;

Are assembled in a predetermined positional relationship.

27. In a method for manufacturing an exposure apparatus for transferring a pattern to at least two regions where peripheral portions partially overlap each other on a substrate,

A dimming filter that gradually reduces the amount of illumination light applied to the pattern on the substrate at a portion where the at least two regions overlap; and at least one of position information and rotation information of the dimming filter. A detecting device for detecting at least one mark pattern provided in the darkening filter,

Are assembled in a predetermined positional relationship.