JPH0645228A - Projection aligner - Google Patents

Abstract

PURPOSE:To measure imaging characteristics such as a curve of image plane immediately before an actual exposure of a projection optical system with a good throughput without using a test reticle. CONSTITUTION:A projection aligner comprises a reference member 20 which is installed on a Z-stage 14 and in which two slight openings 21a and 21b having each different optical path length to a projection optical system PL are formed with the openings closed each other illumination systems 26 and 27 in which their slight openings 21a and 21b are illuminated from the side of the Z-stage 14 with a light beam having the same wavelength as an exposure beam and the transmission beam is emitted into a reticle through the projection optical system PL, and a photoelectric transducer for performing a photoelectric transduction of the beam which is reflected from the reticle and is transmitted in each slight openings 21a and 21b again through the projection optical system PL.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a projection exposure apparatus, and more particularly to a projection exposure apparatus equipped with a mechanism for measuring the image forming characteristics of a projection optical system for manufacturing semiconductor integrated circuits or large liquid crystal substrates.

[0002]

2. Description of the Related Art Techniques for measuring the image forming characteristics of a projection optical system and correcting the projection optical system so that the image forming state of a mask pattern is the best are disclosed in JP-A-60-26343 and JP-A-60-26343. 63-306626 or JP-A-1-27.
It is disclosed in Japanese Patent No. 3318.

Among them, the one disclosed in Japanese Patent Laid-Open No. 60-26343 discloses a test reticle having a minute line element and a transmitted light of the test reticle which is installed on a stage and passes through a minute slit. The characteristics of the projection optical system based on the change in the output signal from the light receiving element, which is provided when the stage is moved up and down on the Z axis (axis parallel to the optical axis of the projection optical system). , For example, the inclination of the image plane or the curvature of the image plane is obtained. Then, the image forming characteristic of the projection optical system is corrected based on the measurement result.

Further, the one disclosed in JP-A-63-306626 or JP-A-1-273318 has a reference member provided on the stage with a reference mark and a special reference mark. A test reticle is provided, and the reference member is illuminated from the lower surface thereof, and the luminous flux transmitted through each reference mark of the reference member and the test reticle is received to enable measurement of the imaging characteristics of the projection optical system. Then, a change in the image forming characteristic of the projection optical system due to a temperature rise due to the absorption of the exposure light is predicted in advance, and based on the result, the change in the image forming characteristic at the time of exposure is predicted and corrected. According to these devices, it is possible to correct, in real time, the image forming characteristics of the projection optical system due to thermal energy during exposure, such as variations in image plane inclination and image plane curvature.

[0005]

However, in the apparatus for detecting the focus position by the TTL (through-the-lens) method of this type to measure the imaging characteristics, a reticle (test reticle or test reticle having a specific mark) is used. An actual reticle with specific marks formed around the exposure area was required. Therefore, first, it is inevitable that the throughput is reduced when the imaging characteristic is measured using the test reticle for each measurement and the imaging characteristic is corrected. Furthermore, when a test reticle is used, it is not possible to measure the in-focus position for each part of the pattern of the actual reticle immediately before exposure, and it is not possible to accurately detect the imaging characteristics of the projection optical system immediately before actual exposure. could not.

Secondly, when the image forming characteristic of the projection optical system is detected by using an actual device reticle (actual reticle) without using the test reticle, the image is formed by using the marks formed around the exposure area. Although the surface condition is measured, it is unrealistic to arrange the marks in each part in the transfer area, which adversely affects the degree of integration of the circuit pattern. For this reason, the focus cannot be detected at the center of the exposure area, and it is difficult to measure the curvature of field or the tilt of the image plane of the projection optical system.

Further, it is necessary to previously perform focusing between the reticle and the optical system of the focus position detection system of the TTL system, and further, the stage is made to scan in the direction perpendicular to the optical axis, or the stage is temporarily moved in the optical axis direction. After moving, it is necessary to repeat scanning in the direction perpendicular to this optical axis. Therefore, it takes a long time to detect the focus, which causes a decrease in throughput.

In view of the above point, the present invention can measure the image forming characteristics such as the field curvature, the image surface inclination, and the average focal plane of the projection optical system immediately before the actual exposure with a high throughput without using a test reticle or the like. An object of the present invention is to provide a projection exposure apparatus capable of correcting the image forming characteristic of the projection optical system.

[0009]

A projection exposure apparatus according to the present invention includes a first illumination system (1, 1) for uniformly illuminating a mask (R) with exposure light (IL) as shown in FIGS. 1 and 2, for example.
6, 9a, 9b, 13), a projection optical system (PL) for forming an image of the pattern of the mask on the image plane, and the exposure surface of the photosensitive substrate (W) on the image plane of the projection optical system. In a projection exposure apparatus having a stage (14, 15) which holds the photosensitive substrate so as to be substantially aligned and is movable in the optical axis direction of the projection optical system and in a direction perpendicular to the optical axis, the stage (14 , 15) and a plurality of minute apertures (21a, 21b) that are arranged close to each other and have different optical path lengths to the projection optical system (PL), and the plurality of minute apertures. The stage (1) is exposed to the exposure light or light having a wavelength substantially equal to the exposure light.
4, 15) side, and a second illumination system (1, 23, 25, 26) for making transmitted light that has passed through the plurality of minute apertures enter the mask through the projection optical system. .

Further, according to the present invention, in the light reaching the mask (R) by the illumination by the second illumination system, the light reflected by the mask (R) is reflected by the projection optical system (P).
L) through the plurality of minute openings (21a, 2) again.
1b) photoelectric conversion means (28, 51) for individually photoelectrically converting the light transmitted therethrough, and the projection optics based on a plurality of photoelectric conversion signals (SA, SB) output from the photoelectric conversion means. Image plane state detection means (32A) for detecting the positions in the optical axis direction of the projection optical system at the positions of the plurality of minute apertures on the image plane of the system, and each detected by the image plane state detection means. Based on the position information, at least one of the focus position, the image plane tilt, and the field curvature of the projection optical system.
And an image forming characteristic calculating means (32B) for calculating one image forming characteristic.

In this case, the plurality of minute openings (21
a, 21b) includes a phase type diffraction grating and the like in addition to a pattern composed of a light transmitting portion and a light shielding portion. Also,
The projection optical system (P
An example of a plurality of minute openings having different optical path lengths with respect to L) is that the optical path length is changed by the step on the reference member (20), as shown in FIG. 2 (a), for example. Another example of a plurality of minute apertures formed in the reference member (20) and having different optical path lengths with respect to the projection optical system (PL) is, for example, as shown in FIG. The optical path length is changed by the phase shifter (54) provided on (20).

[0012]

According to the present invention, a plurality of minute apertures (21a, 21) provided on or near the plane substantially conjugate with the pattern surface of the mask (R) and the projection optical system (PL).
b) is illuminated with exposure light or light having a wavelength substantially equal to that of the exposure light from the side opposite to the projection optical system (PL), and the light transmitted through the plurality of minute openings reaches the mask via the projection optical system. Then, the light reflected on the back surface of the mask again passes through each of the plurality of minute openings through the projection optical system and individually enters the photoelectric conversion means (28, 51). With respect to the projection optical system, when the back surface of the mask and the plurality of minute openings are optically conjugate with each other, that is, when the minute openings are at the image formation position (focus position) of the pattern on the back surface of the mask by the projection optical system, the photoelectric conversion is performed. The image plane of the projection optical system is detected by using the fact that the amount of light incident on the conversion means becomes maximum (or minimum).

According to this principle, for example, the signal SA corresponding to the amount of light which is transmitted through one opening (21b) of the plurality of openings in FIG. 2A and is incident on the photoelectric conversion means and the other opening ( 21b), the signal SB corresponding to the amount of light entering the photoelectric conversion means corresponds to the optical path length difference with respect to the projection optical system in the optical axis direction (Z direction), as shown in FIG. 3A. Has a phase difference. This signal SA and signal S
If the difference from B is taken, the signal SC as shown in FIG. 3B is obtained. This signal SC becomes 0 when the intermediate plane between one opening (21b) and the other opening (21b) (hereinafter referred to as "aperture reference plane") and the image plane of the mask substantially coincide with each other. In the vicinity of the image plane, the relationship between the distance between the image plane and the aperture reference plane and the magnitude of the signal SC is almost linear.
By utilizing this, the image forming position at an arbitrary position within the exposure field of the projection optical system can be obtained without scanning the reference member (20) in the optical axis direction. The imaging characteristic is corrected based on this result.

When the reflectance of the back surface of the mask is different, the relationship between the distance between the image plane and the reference plane of the opening and the magnitude of the signal SC changes, but this changes in advance along the optical axis in the reference member (2).
This can be solved by scanning 0) and capturing the signal SC. In addition, a reference member (2
0) while maintaining the position in the optical axis direction of the stage (14, 1
This problem can be avoided by scanning 5) in the direction orthogonal to the optical axis.

As described above, in the present invention, it is not necessary to form a specific mark on the mask (R), and the projection optical system (P) is simply used with the actual mask (R) set.
Since the average image plane position, the image plane curvature and the image plane inclination can be obtained only by disposing the reference member (20) in the exposure region of L), the image forming characteristics of the projection optical system can be measured at high speed. Further, if the image forming characteristic correcting means corrects the variation of the image forming characteristic, it is possible to always maintain a good image forming characteristic.

Further, in the case where the plurality of minute apertures formed in the reference member (20) and having different optical path lengths with respect to the projection optical system have different optical path lengths due to the step on the reference member, The reference member (20) can be relatively thin. Further, the plurality of minute openings formed in the reference member (20) and having different optical path lengths with respect to the projection optical system are changed in optical path length by the phase shifter (54) provided on the reference member. In this case, it is easy to form a plurality of minute openings having different optical path lengths with respect to the projection optical system.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the projection exposure apparatus according to the present invention will be described below with reference to the drawings. FIG. 1 shows a schematic configuration of a projection exposure apparatus of this embodiment. In FIG. 1, an exposure illumination light source 1 such as an ultrahigh pressure mercury lamp or an excimer laser light source is a g-line, i-line or ultraviolet pulsed light. (For example, KrF
Illumination light IL having a wavelength (exposure wavelength) that sensitizes the resist layer such as excimer laser light is generated. The illumination light IL is a shutter 2 that closes and opens the optical path of the illumination light.
After passing through the beam splitter 4 that allows most (90% or more) of illumination light to pass therethrough, it reaches the secondary light source forming optical system 6 including an optical integrator (fly eye lens) and the like. At this time, the mirror 22 is in the normal state (during exposure)
In this case, the motor is retracted from the optical path, and it moves on the optical axis by a motor or air pressure only when measuring the image forming characteristics of the projection optical system PL as described later.

The shutter 2 is driven by the drive unit 3 so that the transmission and blocking of the illumination light can be controlled. Also,
Part of the illumination light reflected by the beam splitter 4 is P
The light enters the power monitor 5 including a photoelectric detector such as an IN photodiode. The power monitor 5 outputs the optical information PS obtained by photoelectrically converting the illumination light IL to the main control system 32,
This optical information PS serves as basic data for obtaining a variation amount of the image forming characteristic of the projection optical system PL during exposure (details will be described later).

The illumination light IL, which has been made uniform in luminous flux and reduced in speckles in the secondary light source forming optical system 6, is first reflected after being reflected by the beam splitter 7 having a high reflectance.
Relay lens 9a, variable reticle blind 10, second
It goes toward the mirror 12 through the relay lens 9b and the aperture stop. The illumination light IL reflected vertically downward by the mirror 12 is
The main condenser lens 13 illuminates the pattern area PA of the reticle R with a uniform illuminance. The surface of the variable reticle blind 10 and the reticle R are in a conjugate image forming relationship, and the variable reticle blind 10 is driven by the drive motor 11.
The illumination field of the reticle R can be arbitrarily set by opening and closing the movable blade constituting the above to change the opening position and shape. Further, in this embodiment, the reflected light generated from the wafer W by the irradiation of the illumination light IL passes through the beam splitter 7 and is incident on the reflection amount monitor 8 composed of a photoelectric detector. The reflection amount monitor 8 outputs the optical information RS obtained by photoelectrically detecting the reflected light to the main control system 32. The light information RS serves as basic data for obtaining a variation amount of the image forming characteristic of the projection optical system PL during exposure (details will be described later).

The reticle R is a reticle stage RST that is two-dimensionally movable in a plane perpendicular to the optical axis AX of the projection optical system PL.
The positioning is performed so that the center point of the pattern area PA is placed on the optical axis AX of the projection optical system PL.
Initialization of the reticle R is performed by finely moving the reticle stage RST based on a mark detection signal from a reticle alignment system RA that photoelectrically detects an alignment mark (not shown) around the reticle. Reticle R
Are used after being appropriately replaced by a reticle exchanger (not shown). Especially when performing high-mix low-volume production, replacement is frequently performed.

The illumination light IL which has passed through the pattern area PA of the reticle R is incident on both sides (or one side) of the telecentric projection optical system PL, which projects the projected image of the circuit pattern of the reticle R onto the surface. It is projected (imaged) on one shot area on the wafer W on which the resist layer is formed. The wafer W is placed on a leveling stage (not shown), and the leveling stage
It is provided on a Z stage 14 that can be finely moved in the optical axis direction (Z direction) of the projection optical system PL by a drive motor 17.
Further, the Z stage 14 is mounted on the XY stage 15 which can be two-dimensionally moved by the step-and-repeat method by the drive motor 18, and the XY stage 15 transfers and exposes the reticle R onto one shot area on the wafer W. When finished, the next shot area on the wafer is moved until it coincides with the exposure area of the projection optical system PL. The two-dimensional position of the XY stage 15 is constantly detected by the interferometer 19 with a resolution of, for example, about 0.01 μm, and a movable mirror 14m that reflects the laser beam from the interferometer 19 is fixed to the end of the Z stage 14. Has been done. The interferometer 19 and the movable mirror 14m are provided in pairs for position detection in the X and Y directions.

Further, an irradiation amount monitor 16 composed of a photoelectric detector is provided on the Z stage 14. The light receiving surface of the dose monitor 16 has substantially the same area as, for example, the image field of the projection optical system PL or the projection area of the reticle pattern, and the dose monitor 16 is set so that the light receiving surface substantially coincides with the surface position of the wafer W. It is provided on the Z stage 14. Information LS on the irradiation amount of the shot area on the wafer W obtained from the irradiation amount monitor 16 is sent to the main control system 32, and basic data for obtaining the variation amount of the image forming characteristic of the projection optical system PL during exposure. (Detailed later).

Further, a projection optical system P is mounted on the Z stage 14.
A reference member 20 used when measuring the image forming characteristics of L, for example, the focus position, the field curvature, and the image surface inclination is provided. FIG. 4 is a plan view of the reference member 20.
In, the circle with the radius L indicates the maximum exposure area (image field) IF of the projection optical system PL, and the square inscribed in this circle indicates the maximum square chip area CH when actually exposing the wafer. As shown in FIG. 4, a stepped minute opening 21 is provided in the center of the image field IF on the surface of the reference member 20.

FIG. 2A shows a stepped minute opening 21 of the reference member 20. As shown in FIG.
The minute opening 21 is configured such that the lower opening pattern portion 21a and the upper opening pattern portion 21b are adjacent to each other in the optical axis AX direction (Z direction) of the projection optical system PL. Z between the lower opening pattern portion 21a and the upper opening pattern portion 21b
The step in the direction is ΔZ (to the extent that the skirts of the signals SA and SB in FIG. 3A overlap). In addition, the reference member 20
The shutter member 51 is disposed at the bottom of the shutter member 51, and the shutter member 51 can selectively cover the regions 52a and 52b corresponding to the bottoms of the opening pattern portions 21a and 21b.

FIG. 5A shows the lower opening pattern portion 21a and the upper opening pattern portion 21b of the minute opening 21.
5B shows another example of the opening patterns of the lower opening pattern portion 21a and the upper opening pattern portion 21b. In FIGS. 5A and 5B, the dark portion is a light shielding portion and the bright portion is a light transmitting portion. That is, the pattern provided in the stepped minute opening 21 is, for example, a checkerboard pattern shown in FIG. 5A or a lattice pattern shown in FIG. 5B, and the height of the surface of the reference member 20 shown in FIG. It is desirable that the height and the height of the surface of the are almost the same.

As the shape of the opening pattern provided in the stepped minute opening portion 21, it is necessary to select a shape whose projected image does not completely overlap the circuit pattern on the reticle R. This is because when the projected image of the aperture pattern completely overlaps the aperture pattern, the position at which the maximum or minimum of the change in the amount of light reflected from the reticle R (which will be described later in detail) that passes through the aperture 21 is obtained is not always the focal point. This is because that is not always the case.

Here, by making the opening pattern a checkered pattern as shown in FIG. 5A, the influence of this circuit pattern can be reduced, and the influence of astigmatism of the projection optical system PL can be reduced. Can be avoided. Even in the lattice pattern shown in FIG. 5B, the influence of the circuit pattern can be reduced by inclining the longitudinal direction of the lattice with the sides of the circuit pattern by about 10 to 30 °. This is because there are many vertical and horizontal patterns (0 ° and 90 ° patterns) which are orthogonal to each other in the circuit pattern used for semiconductor manufacturing. Further, it is preferable that the shape of the aperture pattern is a plurality of parallel line shapes that are not parallel to any line of the circuit pattern, and if the parallel lines are made orthogonal to each other in order to avoid the influence of astigmatism of the projection optical system PL. Good. Further, it is desirable to set the area ratio of the light-shielding portion and the transmission portion of the opening pattern to 1: 1 so that the SN ratio becomes the best.

As a pattern of the opening pattern portions 21a and 21b of the minute opening portion 21 of the reference member 20, FIG.
As shown in (c), if a photoelectric detector that independently detects the transmitted light from the parallel lines in each direction by using parallel lines and space lines having a plurality of directions is provided in the exposure field. It is also possible to measure astigmatism at any point. In addition, as shown in FIG.
1 is used, the reference member 2 is used as shown in FIG.
It is also possible to use a transparent phase member 54 such as glass provided on a part of the pattern forming portions 53a and 53b above the pattern forming portion 53a. Since the optical path length is changed by the phase member 54, the same effect as the step can be realized. Since the glass has a large refractive index, the optical path length is extended, and the pattern forming portion 53
a is equivalent to being present on the surface 56 below the step ΔZ from the pattern forming portion 53b. Pattern forming portion 53b and surface 5
The surface intermediate 6 is coincident with the surface 55 of the reference member 20. Assuming that the refractive index of the transparent phase member 54 is n and the thickness above the pattern forming portion 53a is t, the step (optical path length difference) ΔZ.
Is given by ΔZ = (n−1) t.

Now, in FIG. 1, the minute opening 21 is illuminated from below (on the stage side) by the measurement illumination system LL. The measurement illumination system LL includes a mirror 22, a beam splitter 23, a condenser lens 24, an optical fiber 25, which is inserted into the optical path of the illumination light IL by a driving device (not shown).
It is composed of a condenser lens 26, a mirror 27, and a shutter member 51, and guides the illumination light IL into the XY stage 15 to selectively illuminate the adjacent aperture pattern of the minute aperture 21. The illumination light emitted from the minute opening 21 reaches the reticle R via the projection optical system PL, is reflected by the back surface (pattern surface) of the reticle R, and returns to the minute opening 21 via the projection optical system PL again. This reflected light has a small opening 2
After passing through 1, the light enters through a shutter member 51, a mirror 27, a condenser lens 24, an optical fiber 25, a condenser lens 26 and a beam splitter 23 to a photoelectric detector 28 composed of a photomultiplier, a photodiode or the like. Then, it is photoelectrically converted by the photoelectric detector 28 and becomes the detection signal SA or SB.

The detection signal SA is the detection signal of the light passing through the lower opening pattern portion 21a of FIG. 2A, and the detection signal SB is the detection signal of the light passing through the upper opening pattern portion 21b of FIG. 2A. It is a detection signal. Here, although not shown, the measurement illumination system LL may illuminate each stage of the stepped minute aperture 21 and the photoelectric detector 28 may receive reflected light separately. Further, even if the shutter member 51 as shown in FIG. 1 is not provided, the entire back surface of the reference member 20 is illuminated, and the signal from the photoelectric detector 28 is electrically masked to reflect the light from the aperture of the desired step. You may make it process only the signal corresponding to light. In this way, only one set of illumination system and photoelectric detector is needed,
The device can be simplified.

The illumination light IL is emitted from the light source 1 to the mirror 22.
It is supposed that the light source 1 is guided to the reference member 20 via the light source 1.
The reference member 20 may be illuminated by separately providing a light source that emits light having a wavelength substantially equal to that of the illumination light IL. On the other hand, in FIG. 1, 34 denotes an oblique incidence type detection optical system as a whole.
It is disclosed in Japanese Patent No. 361. The oblique-incidence detection optical system 34 detects the vertical position (Z direction) of the wafer surface with respect to, for example, the best designed image plane of the projection optical system PL, and focuses the wafer W and the projection optical system PL. The focus detection system 29 for detecting the state and the leveling detection system 30 for detecting the inclination of the wafer surface with respect to the designed image plane of the predetermined region on the wafer W are combined.

Focus detection system 29 and leveling detection system 3
Reference numeral 0 denotes light sources 29a and 30a that emit illumination light that is incident on the wafer surface from an oblique direction with respect to the optical axis AX, a half mirror 31a that combines two light fluxes, and a half mirror 31b that splits the reflected light on the wafer surface. And light receiving optical systems 29b and 30b for respectively receiving the divided light beams.
The illumination light emitted from the light source 29a is an image forming light flux that forms an image of a pinhole or a slit, and the light source 30a.
The illumination light emitted from is a parallel light flux. In this embodiment, the angle of the parallel flat glass (plane parallel) (not shown) provided inside the light receiving optical system 29b is adjusted in advance so that the best designed image plane becomes the zero reference, and the focus is adjusted. While the detection system 29 is calibrated,
When the surface of the wafer W coincides with the image plane, the light source 30a
The horizontal position detection system is calibrated so that the parallel light flux from the light source is focused on the center position of the four-division light receiving element (not shown) inside the light receiving optical system 30b.

Next, the structure of the correction means for correcting the image formation state will be described. In this embodiment, by driving the lens element of the projection optical system PL, the imaging characteristics (projection magnification, distortion, curvature of field, etc.) are corrected, and the imaging characteristics of the projection optical system PL. A part of the optical element is movable so as to be adjustable. As shown in FIG. 1, the lens elements 40 and 41 of the first group closest to the reticle R are the support members 4
The lens element 43 of the second group is fixed by the supporting member 44, and the lens element 45 of the third group is fixed by the supporting member 46. The lens elements below the lens element 45 are fixed to the lens barrel portion 47 of the projection optical system PL.

FIG. 6 is a view of the projection optical system PL seen from above (reticle side). The driving elements 48a to 48c are arranged at positions rotated by 120 °, respectively, and can be independently controlled by the driving element controller 33. Has become. Similarly, the driving elements 49a to 49c and 50a to 50c are also rotated by 120 ° and arranged independently by the driving element control unit 33. The drive elements 48a, 49a and 50a are arranged offset from each other by 40 °, and the drive elements 48b, 49b and 50b and the drive elements 48c, 49c and 50c are also arranged offset from each other by 40 °. In FIG. 1, for example, only the driving elements 48a and 48c among the driving elements 48a to 48c are shown.

In the following, the three driving elements 48a to 48c will be collectively referred to as a driving element group 48, and the driving element groups 49 and 5 will be described.
The same applies to 0. As the drive element groups 48 to 50, for example, electrostrictive elements or magnetostrictive elements are used, and the displacement amount of the drive elements according to the voltage or magnetic field applied to the drive elements is obtained in advance. Although not shown here, a capacitive displacement sensor, a differential transformer, and the like are provided as a position detection device near the drive element in consideration of the hysteresis of the drive element. As a result, the position of the drive element corresponding to the voltage or magnetic field applied to the drive element can be monitored, so that highly accurate drive is possible.

The projection optical system P in this embodiment shown in FIG.
The optical axis AX of L indicates the optical axis of the lens element fixed to the lens barrel portion 47 of the projection optical system PL.
Now, the support member 46 is the drive element 50a, 50 which can be expanded and contracted.
It is connected to the lens barrel portion 47 via b and 50c.
In addition, the support member 44 is configured to extend and retract the drive elements 49a, 49a.
The support member 42 is connected to the support member 46 via b and 49c, and the support member 42 is expandable and contractable.
It is connected to the support member 44 via b and 48c.

Here, in the present embodiment, the drive element controller 33
The lens elements 40, 41, the lens element 43, and the lens element 45 close to the reticle R can be independently moved by controlling the above-mentioned driving element by, and these lens elements have a magnification, a distortion, and an image plane. A lens element whose influences on the image forming characteristics such as curvature, image plane inclination, and focus position are greatly controlled as compared with other lens elements is selected. Further, in this embodiment, since the movable lens element is composed of three groups, it is possible to increase the movement range by moving the lens element while suppressing variations of other various aberrations. In addition, various shape distortions (trapezoidal, rhomboidal, barrel-shaped, pincushion-shaped, and other distortions) can be coped with, and fluctuations in the imaging characteristics of the projection optical system PL due to absorption of exposure light can be sufficiently coped with. It should be noted that the movement of the lens element is performed within a range in which the influence on other various aberrations of the projection optical system PL (for example, astigmatism, spherical aberration, etc.) can be ignored. Alternatively, as described above, a method may be adopted in which the field curvature is controlled and other aberrations (distortion, etc.) are also corrected by adjusting the distance between the lens elements.

With the above structure, the three peripheral points of the lens elements 40 and 41, the lens element 43 and the lens element 45 of the three groups are independently driven by the main control system 32 in the optical axis AX direction of the projection optical system PL. It can be moved by an amount according to the command. As a result, each of the lens elements 40, 41, 43, and 45 of the three groups can be translated substantially along the optical axis AX and can be arbitrarily tilted with respect to a plane substantially perpendicular to the optical axis AX. Is possible. The lens elements are supposed to be tilted with the optical axis AX as a virtual tilt reference. Also, the lens elements 40, 41, 43 and 4
5 is movable in a plane substantially perpendicular to the optical axis AX.

The main control system 32 not only controls the drive element control section 33, but also has an in-focus position detection section 3 inside as described later.
2A and an image forming characteristic detector 32B are provided, and the image forming characteristics of the projection optical system PL, such as the field curvature, the image surface inclination, and the average focus position, are obtained from them. Further, as will be described later, the stage is started by obtaining optical information from the power monitor 5, the reflection monitor 8, and the irradiation amount monitor 16, respectively, and calculating the variation amount of the imaging characteristic of the projection optical system PL during exposure. Controls the entire device including the above.

Next, a method of measuring the image forming characteristics of the projection optical system PL such as the field curvature, the image surface inclination, and the astigmatism in this embodiment will be described. In FIG. 1, while the reticle R is mounted on the reticle stage RST, while monitoring the stage position by the interferometer 19, the stepped minute opening 21 on the reference member 20 is set to an arbitrary position within the exposure field of the projection optical system. Drive motor 1 to drive XY stage 15 so that it comes to the position
Move at 8. At this time, the Z stage 14 is simultaneously driven by the drive motor 17 so that the surface of the reference member 20 is projected onto the projection optical system P.
It is set at the best focus position (best image plane) of L in design.

Here, the mirror 22 is moved into the optical path to guide the illumination light IL to the upper or lower opening pattern of the stepped minute opening 21, and the minute opening 21 and the projection optical system P.
The back surface (pattern surface) of the reticle R is illuminated via L. The reflected light reflected from the reticle R again passes through any one of the projection optical system PL and the minute aperture 21 and then the shutter member 51, the mirror 27, and the condenser lens 2.
6, incident on the photoelectric detector 28 via the optical fiber 25, the condenser lens 24, and the beam splitter 23. The photoelectric detector 28 can output a signal (voltage value) having a magnitude corresponding to the amount of light transmitted through the upper pattern of the stepped minute opening 21 and the amount of light transmitted through the lower pattern 21.

Next, in this example, the characteristics of the output signal of the photoelectric detector 28 are obtained by scanning the Z stage 14 in the Z direction in advance before measuring the imaging characteristics of the projection optical system. As a result, the characteristics shown in FIG. 3 are obtained. FIG. 3 (a) corresponds to FIG. 2 (a).
2 and the detection signal SA which is the output signal of the photoelectric detector 28 according to the light passing through the lower opening pattern portion 21a.
(A) shows a detection signal SB corresponding to light that has passed through the upper opening pattern portion 21b. The peak of the detection signal SA and the peak of the detection signal SB are the optical axis of the projection optical system PL (Z direction).
Are separated by ΔZ equal to the step. In addition, FIG.
Indicates a difference signal SC obtained by subtracting the detection signal SB from the detection signal SA. When this detection signal SC becomes 0, the minute opening 21 of the reference member 20 on the Z stage 14
An intermediate surface (aperture reference surface) between the upper and lower rows matches the image forming surface of the projection optical system PL. Then, for example, when the focus position of the projection optical system is measured during exposure, a pair of values of the detection signals SA and SB is obtained with the Z stage 14 stopped, and the value of the difference signal SC is calculated from these values. Ask for. From the value of the difference signal SC thus obtained, FIG.
Since the amount of deviation between the current position of the Z stage 14 and the in-focus position can be known from the characteristics of,
The position where the amount of deviation is added to the position in the direction is the current focus position of the projection optical system (the position on the best imaging plane). As a result, it is possible to quickly perform measurement without moving the Z stage 14. However, as another measurement method, the Z stage 14 may be driven so that the value of the difference signal SC becomes 0, and the position of the Z stage 14 may be detected.

Further, in FIG. 1, the difference signal SC is 0.
When the Z stage 14 is driven so as to maintain the above condition and the position thereof is read and the XY stage 15 is scanned within the range where the minute opening 21 is within the exposure field, the output signal of the focus detection system 29 is shown in FIG. The characteristics as shown are obtained, and the field curvature or field tilt can be measured. Field curvature,
When measuring the image plane tilt, a checkerboard pattern or the like, which is not easily affected by astigmatism, is suitable as the pattern of the minute aperture 21. However, by using parallel line patterns in several directions, It is also possible to measure the image plane in the sagittal direction and the image plane in the meridional direction, that is, astigmatism, at any position in the exposure field by the above method.

As the means for measuring the position of the Z stage, a focus detection system 29 of the oblique incidence type or a displacement sensor in the Z stage can be used. The focus detection system 29 is the reference member 20.
Since the height of the surface of the reference member 20 is detected, if the surface of the portion other than the minute opening portion 21 of the reference member 20 is made to coincide with the opening reference surface of the stepped minute opening portion 21, the origin of the focus detection system 29 can be determined. This is convenient when performing calibration.

When the surface of the reference member 20 other than the minute opening 21 does not coincide with the opening reference surface,
The distance between the surface and the opening reference plane may be measured in advance and stored as a constant of the apparatus. In this measurement, while scanning the minute opening portion 21 in the exposure field in the Z direction,
It is only necessary to monitor the output signal of the oblique incidence type focus detection system 29 and the difference signal SC and read the offset between the zero point of the output signal of the focus detection system 29 and the zero point of the difference signal SC.

The relationship as shown in FIG. 3A is obtained between the detection signals SA and SB corresponding to the opening pattern portions 21a and 21b of each stage and the position of the reference member 20 in the Z direction. Can be explained as follows. The minute opening 21 is the reticle R
When the rear surface of the reticle is in the defocus area of the projection optical system PL, the illumination light emitted from the minute opening 21 is defocused with an opening pattern (for example, a checkerboard pattern) on the back surface of the reticle R via the projection optical system PL. Make a statue. After that, the illumination light reflected on the back surface of the reticle R is again projected onto the projection optical system PL.
It is incident on the minute opening 21 via. Further, it is an image of the defocused aperture pattern. Therefore, when this reflected light passes through the aperture pattern, the defocused portion does not pass through the light-shielding portion of the aperture pattern, and the amount of its spectrum is reduced.

On the other hand, a position where the minute opening 21 is optically conjugate with the back surface of the reticle R with respect to the projection optical system PL, that is,
When it is on the focal plane (imaging plane) of the projection optical system PL, the illumination light emitted from the minute aperture 21 forms a sharp image on the back surface of the reticle R, so that when the reflected light passes through the aperture pattern. There is almost no reflected light that reaches the light-shielding portion of the aperture pattern (is eclipsed by the light-shielding portion), and the amount of light received by the photoelectric detector 28 is larger than that when defocusing is performed. As a result, as shown in FIG. 3A, the detection signal SA corresponding to the light transmitted through the upper aperture pattern portion 21b becomes maximum when the upper aperture pattern portion 21b coincides with the image plane, and the lower aperture opening 21a is detected. The detection signal SB corresponding to the amount of light that has passed through the pattern portion 21a is generated by the lower opening pattern portion 21.
It becomes a maximum when a coincides with the image plane. Since these signals are output almost symmetrically near the image formation position,
The difference signal SC in (b) becomes almost 0 when the aperture reference plane coincides with the image plane, and in this vicinity, the difference signal SC becomes
It changes substantially in proportion to the distance between the aperture reference plane and the image plane.

Here, it is also possible to obtain the best focus position by using a transparent reticle R having no circuit pattern or a reticle R provided with a light shielding portion on the entire surface of the circuit pattern area. It may be used to periodically check whether or not the best focus position is measured without being affected by the circuit pattern. By the way, in the case of a transparent reticle without a circuit pattern, the glass surface on the back surface of the reticle serves as a reflecting surface, and its reflectance is 4
%, But can be sufficiently detected by increasing the detection sensitivity.

The detection of the best focus position is performed by the main control system 32 of FIG.
This is performed by the in-focus position detection unit 32A. Since the focus position detection unit 32A outputs the difference signal SC of the detection signals corresponding to the upper opening and the lower opening in the reference member 20 as described above, the Z stage is located at the position where the difference signal SC crosses 0. If the position is controlled, the focus position (position in the Z direction) can be obtained from the output value of the focus detection system 29 at that time. If this processing is performed at any position within the exposure field, the best focus position at each point is obtained. For example, at the same time as the above operation, the stepped minute opening 21 may be scanned in the radiation direction within the exposure field.

Based on the above results, when the relationship between the image height (the distance from the optical axis AX to the periphery of the exposure area) and the best focus position is obtained, a graph as shown in FIG. 7 is obtained. Further, the graph of FIG. 8 is obtained by separating the data relating to the image plane inclination and the data relating to the field curvature from the graph of FIG. 7. Figure 7
8 shows the position in the Z direction, and the horizontal axis shows the image height. In FIG. 8, position Z ₁ is the maximum focus position when the image plane tilt component is removed from the data value shown in FIG. 7, and position Z ₂ is the minimum focus when the image plane curvature component is removed from the data value shown in FIG. The position and the position Z ₃ are the maximum focus positions when the field curvature component is removed from the data values shown in FIG. 7.

Then, in FIG. 7, the image plane inclination ΔZD is obtained. First, the positions Z ₂ and Z _{3 which} are basic data for obtaining the image plane inclination are obtained by the following equation. Z ₃ = f ₁ + (f ₃ −f ₂ ) / 2 Z ₂ = f ₁ − (f ₃ −f ₂ ) / 2 Then, using these positions Z ₂ and Z ₃ , the image plane inclination ΔZD
Is obtained from ΔZD = Z ₃ −Z ₂ = f ₃ −f ₂ and is represented by the straight line in the upper part of FIG. 8.

Here, the image plane inclination ΔZD is corrected by calculation, and the image plane curvature Δr is obtained by adding the focal positions at the four corners around the exposure field to the focal positions at the center of the exposure field. Here, there is a relationship of Z ₁ = (f ₂ + f ₃ ) / 2, and from this, the field curvature Δr becomes Δr = Z ₁ −f ₁ , which is represented by the lower curve in FIG. 8. Further, the average focus position f _A is obtained as f _A = f ₁ + Δr / 2.

Incidentally, the image plane inclination ΔZD and the image plane curvature Δ
The calculation for obtaining r and the average focus position f _A is performed by the image formation characteristic detection unit 32B in the main control system 32. Next, a method for correcting the field curvature, the field inclination, and the average focus position will be described.
Basically, as described above, the lens elements 40, 41, the lens element 43, and the lens element 45 of the three groups of the projection optical system are driven so as to be tilted with the optical axis direction or an axis perpendicular to the optical axis as the rotation axis. As a result, a desired image formation state is obtained. Here, for simplicity of explanation, only the curvature of field, the inclination of the field, and the average focus position will be described, but it is also possible to correct the projection magnification, distortion, and the like. For example, the change in the field curvature due to the driving of the lens elements 40 and 41 will be described with reference to FIG. Figure 9
The change of the field curvature shown in (1) is just an example and is not general, and the way of change actually varies depending on the configuration of the lens element. In the present embodiment, the field curvature changes as shown in FIG. 9 by moving the lens element 40 or 41 in the optical axis direction. Further, depending on the method of configuring the lens, the field curvature is changed by moving the lens elements 40, 41 and the lens element 42 in the direction perpendicular to the optical axis. Such lens elements 40, 41,
The relationship between the drive amount of the lens element 43 and the lens element 45 and the change amount of the field curvature is stored in advance in the memory in the main control system 32.

Driving the lens elements to correct the field curvature may worsen magnification, distortion, and other aberrations (eg, coma and astigmatism), but drives a plurality of lens elements. Therefore, desired field curvature correction can be performed while correcting aberration. Also, the aberration or the like may be corrected by changing the image forming characteristic of the projection optical system by changing the pressure of the air portion between the lens elements as disclosed in JP-A-60-78454. Lens element 40, 4
Although it is possible that the image plane moves up and down by driving 1, 43, and 45, if the output value of the focus detection system 29 is offset according to the amount of change, the wafer W will always have the best image. Since it is set on the surface, this effect can be prevented.

The correction of the image plane inclination is performed by the above-mentioned image forming characteristic detecting section 3
It is also possible to add the focus differences (image plane inclinations) at the four corners of the exposure field obtained from 2B to the horizontal position detection system 30 as offsets for calibration. That is, the image plane tilt is corrected by driving the leveling stage and tilting the wafer W while detecting the horizontal position of the wafer surface by the horizontal position detection system 30 to which the offset is applied.

As for the focus position, the average focus position f _A is obtained as described above, and the drive motor 17 controls the Z stage 14 so that the wafer W and the image plane are always coincident with each other based on the value. At this time, it is desirable to perform calibration so that the zero point reference of the focus detection system 29 becomes the average focus position f _A. In this way, when the Z stage 14 is controlled, the front surface of the wafer becomes optically conjugate with the back surface of the reticle R, and good image formation can be obtained. Further, although the simple average value is used in the above, the focus detection system 29 may be calibrated on a surface that minimizes the maximum deviation between the image plane and the wafer surface. As described above, by performing the exposure on the surface or the average focus position that minimizes the maximum deviation, even when the correction of the field curvature is not sufficient, the good exposure is performed as compared with the conventional method. . When the linearity guarantee range (in other words, the detectable range) of the focus detection system 29 is small, a position sensor such as an encoder is attached to the drive motor 17. Depending on the value of this position sensor, Z
The stage 14 is scanned to find the relationship between the output signal of the photoelectric detector 28 and the focus position (Z stage position) as shown in FIG. After that, while driving the herbing glass in the focus detection system 29 by a drive system (not shown), the focus position (encoder value) is associated with the output of the focus detection system 29, and finally the focus detection system 29 is detected. The output of the photoelectric detector 28 is obtained.

Here, in the case of correcting the curvature of field or the inclination of the field, if only half the exposure area is exposed, for example, FIG.
In the case where the curvature of field is 81 as shown in FIG. 7, the variable reticle blind 10 exposes the area left from the center line 80, and the surface 82 is adjusted so that the wafer W is better aligned with the image plane. A method of bringing the wafer W is also conceivable.

Also, regarding the image plane tilt and the average focus position, the relationship between the driving amount of the lens elements 40, 41, 43 and 45 and the change amount of the image plane tilt and the focus position is stored in advance in the memory in the main control system 32. However, it can be corrected by moving the lens elements 40, 41, 43 and 45 as described above. At this time, an average image forming surface is obtained based on the image surface curvature, the image surface inclination, and the average focus position measured by using the minute aperture portion 21, and the lens elements 40, 41, 42 are aligned with the image forming surface. It may be driven. As a result, the exposure can be started in a state where the image forming characteristic is corrected in advance.

Incidentally, the measurement of the image forming characteristic of the projection optical system as in the above-mentioned embodiment may be carried out every unit time or every predetermined number of wafers to be processed. Next, the operation of this embodiment will be described. (1) First, the curvature of field, the tilt of the image plane, and the average focus position that the projection optical system PL has in the initial state are set to the signals corresponding to the reflected light obtained from the minute aperture 21 as described above when the system is set up. By measuring the image forming characteristic based on the method, and by driving the lens elements 40, 41, 43, 45, or by the lens elements 40, 4
The correction may be performed so that the optimum imaging state and the in-focus position can be obtained by driving 1, 43, 45, the leveling stage, and the Z stage.

(2) Further, the focus position and the field curvature are changed by the change of the environment such as the atmospheric pressure and the temperature, the absorption of the illumination light of the projection optical system PL, or the deflection of the reticle R by its own weight and the deformation of the reticle R by the absorption of the illumination light. Even if the image plane tilt changes from the time of system up, those changes are relatively slow, or the focus position at regular measurement intervals (for example, every 10 wafer exchanges, every wafer exchange, every few shots, etc.) If the amount of change in the curvature of field or the inclination of the field is negligible, the imaging characteristics are measured based on the signal corresponding to the reflected light obtained from the minute opening 21 at the time of wafer replacement, for example. , (1) may be corrected.

(3) The imaging characteristics of the projection optical system PL (focal position, field curvature,
When the change of the image plane inclination) is fast, it is possible to cope with the change of the image forming characteristic by using the predictive control and the measurement using the above-mentioned minute aperture 21 in combination. The predictive control is, for example, JP-A-1-
Similarly to the one disclosed in Japanese Patent No. 273318, the imaging characteristics that vary depending on the amount of heat that the projection optical system PL receives from the exposure light are stored in advance in the memory, and the actual exposure is performed based on the values stored in this memory. The imaging characteristic is corrected in the same manner as (1). Here, for example, the correction by the predictive control is performed by driving the lens element.

The change characteristic of the image forming characteristic of the projection optical system PL is
Information about opening and closing of the shutter 2 of FIG. 1, outputs PS, RS, and L of the power monitor 5, the reflectance monitor 8, and the dose monitor 16.
The amount of heat is measured by using S, and the state variable of the imaging characteristic that changes depending on the amount of heat is stored in the memory of the main control system 32. The amount of heat is calculated based on the output value LS of the irradiation amount monitor 16 that detects the irradiation energy for each exposure condition and the open time of the shutter 2. Here, when the reflectance of the base of the projection optical system PL is zero, only the transmitted light of the projection optical system PL needs to be considered, but the heat quantity accumulated in the projection optical system PL by the reflectance of the base of the projection optical system PL. Therefore, when the reflectance is not zero, (transmitted light + reflected light) contributes to thermal energy. Therefore, in this case, the output value from the reflection monitor 8 is also used. Further, the power monitor 5 constantly monitors temporal changes such as fluctuations of the light source during exposure, and determines the amount of heat accumulated in the projection optical system PL in consideration of changes in its output value LS.

When the image forming state of the projection optical system PL fluctuates due to other factors such as atmospheric pressure change, data predicted in advance and stored (for example, image forming characteristic with respect to atmospheric pressure change). These changes are also added to the state variables based on the fluctuation data). Calculate the optimum correction amount for the average focus position, field curvature, and field tilt for this total value,
The lens elements 40, 41, 43, and 45 are driven by the lens element drive element groups 48 to 50 to perform correction.

In this way, the lens elements 40, 4 are exposed during the exposure of, for example, one wafer.
1, 43 and 45 are driven by predictive control. On the other hand, periodically, for example, every time the wafer is exchanged or every unit number of processed wafers (for example, every time the reticle is exchanged, every 10 wafers are processed), photoelectric detection corresponding to the reflected light from the reticle passing through the minute opening portion 21 is performed as described above. The signal from the device 28 is detected, and the focus position, the curvature of field, and the tilt of the field are calculated from the results. When the focus position, the field curvature, and the image surface inclination of the predictive control are corrected by the focus position, the image surface curvature, and the image surface inclination obtained from the photoelectric detector 28, the predictive control is periodically calibrated and the projection optical system. The imaging characteristics of PL are improved.

This calibration calculates the error between the prediction control and the calculation, corrects the value of the prediction control to the measured value,
This is performed by correcting the calculation formula of the predictive control. Further, when the predictive control is also used, a number setting means (a counter or the like) for setting, for example, how many wafers the measurement of the imaging characteristics of the projection optical system as described above should be provided. Is desirable. Generally, from the start of the exposure operation to the predetermined number of wafers, the difference (calculation error) between the image formation characteristic calculated by the calculation in the predictive control and the actual image formation characteristic obtained by the above measurement is small, The number of wafers set in the number setting means may be large, and here, for example, 1
It shall be set to 0 sheets.

As described above, when the number setting means is used,
After measuring and correcting the imaging characteristics before starting the exposure operation,
The wafers up to the tenth wafer are exposed while the imaging characteristics are corrected only by the predictive control, and the measurement and the correction are performed again when the exposure is completed. Then, this measurement result is compared with the imaging characteristic calculated by the calculation of the predictive control, and the difference is equal to or more than a predetermined allowable value (a value determined by the depth of focus of the projection optical system or the control accuracy of the imaging characteristic). In such a case, the set number set in the number setting means may be changed and, for example, measurement and correction may be performed for each wafer. With this method, the number of measurements can be reduced and throughput is improved. Furthermore, by determining the state variable based on the actually measured data and comparing it with the state variable in the predictive control, it is determined how many more wafers will be exposed and the difference will exceed the allowable value. Alternatively, the set number may be automatically changed to a smaller number (for example, for each wafer). As a result, the set number can be changed without the difference exceeding the allowable value, and if the image formation state is stable, the exposure can be performed only by predictive control without the difference exceeding the allowable value. Can be expected to improve.

It should be noted that, for example, an environment sensor for monitoring environmental changes (pressure, temperature, humidity, etc.) is provided, and during the predictive control, an abnormality such that the image forming characteristic of the projection optical system PL can greatly change can be detected. , The exposure operation is immediately stopped when an abnormality is detected, and the minute opening 2
It is also possible to perform the measurement using 1 and to correct the image forming characteristic of the projection optical system PL by driving the lens element.

When the change of the image forming characteristic is fast, it is easier than the above method to perform the measurement every time the wafer is exchanged without performing the predictive control, and perform the linear interpolation or the secondary interpolation based on the value at the time of the previous measurement or the measurement before the two. A certain degree of effect can be obtained by making a prediction in the manner of the next interpolation. Here, even when the reference member 20 is tilted, the tilt is measured in advance by changing the position of the XY stage by the oblique incidence type autofocus detection system, and an error caused by the tilt when measuring by the minute opening 21 is performed. The minutes can also be corrected.

In the above-described embodiments, the projection optical system P
The case where only one focus detection system 29 for measuring the optical axis center of L is provided has been described, but a plurality of focus detection systems may be provided so that each position in the exposure area can be measured.
At this time, if the plurality of focus detection systems 29 can measure each point of the reference member 20, even if the reference member 20 is tilted or the tilt is changed, measurement can be performed without error. Conversely, if the reference member 20 can be held strictly horizontally, the minute openings 21 having a plurality of opening patterns of the reference member 20 can be used for calibration of the plurality of focus detection systems 29.

Further, in this case, more detailed focusing can be performed by taking into consideration the unevenness of the image plane and the unevenness of the wafer W. For example, it is also possible to intentionally bend the image plane according to the warp of the wafer W and perform exposure. The focus detection system 29 and the leveling detection system 30 are not limited to the configuration of this embodiment. For example, the height position at each point (position in the Z direction) is obtained by irradiating an image of a slit or a pinhole on each of the four corners of the rectangular shot area on the wafer and the center thereof.
If the focus detection system is configured to detect the, the inclination amount of the wafer surface can be obtained from the height positions at five points without providing a horizontal position detection system.

Next, a modification of the pattern of the minute opening 21 will be described. In the above-described embodiments, the pattern of the minute opening portion 21 is described as the combination pattern of the transmitting portion and the light shielding portion as shown in FIGS. 2A and 2B, but the wavelength of the illumination light is 1
A pattern equivalent to that of FIG. 2 can be formed by using a phase grating pattern that is shifted by / 4 wavelength. This principle will be briefly described with reference to FIG. As shown in FIG. 11A, diffracted light is generated by illuminating the minute opening 21 made of the phase grating pattern from below. At this time, it is ± 1 compared to the 0th order light.
The phase of the next light is shifted by ¼ wavelength. These light rays are reflected by the back surface of the reticle R and return to the minute opening 21. When the minute opening 21 is on the focal plane of the projection optical system PL, the ± first-order light is diffracted again as shown in FIG.
The next ray returns along the same optical path as the 0th order ray. At this time, the phases of the light rays ± first-order light are further shifted by ¼ wavelength, and the phases are shifted by ½ wavelength from the zero-order light. Therefore, the light rays ± first-order light act to cancel the zero-order light. When the minute aperture 21 as the phase grating is not on the focal plane of the projection optical system PL, the phase difference does not become 1/2 wavelength, and thus the above-mentioned canceling condition is not satisfied. From the above, as shown in FIG. 12, the best focus is when the output of the photoelectric detector 28 corresponding to the reflected light passing through the minute opening 21 is minimum. The horizontal axis of FIG. 12 represents the output signal of the focus detection system 29 of FIG. 1, that is, the position in the Z direction. According to the method using the phase shift pattern as described above, since all the light rays pass through the minute opening 21 as compared with the above-described method, the SN ratio becomes high, and the best focus point can be easily found.

Further, when the image forming characteristic is corrected based on the measurement result (field curvature, image surface inclination, etc.), the herbing glass of the focus detection system 29 is driven by the above-described predictive control to roughly focus in advance. You may find the position.
In this case, it can be expected that the focus position is detected with high accuracy in the region where the difference signal SC has good linearity. The focus detection system 2
9 and the method of giving an offset at the time of calibration of the leveling detection system 30 may be either an optical method (such as driving a herbing glass) or an electrical method (such as electrically adding an offset amount to an output value). . Further, an average image forming surface of the projection optical system PL is obtained based on the measurement result (image surface curvature, image surface inclination, average focus position), and the focus detecting system 29 and the focus detecting system 29 are set so that this image forming surface serves as a zero point reference. It suffices to simply calibrate the leveling detection system 30. At this time, the average image plane detected as described above is evaluated, and the amount of change of this average image plane with respect to the best image plane (design value) is within a predetermined allowable value (focus detection system 29 or leveling). If it is within the detectable range of the detection system 30), only the calibration is performed, and if it exceeds the allowable value, the correction may be performed by the method such as the lens driving described above.

Here, in the above-described embodiment, the projection optical system PL
As an example of means for correcting the image forming characteristic of the projection optical system P
Although a mechanism for driving a part of the lens element forming L is used, the correction means applicable to the present invention is not limited to the above mechanism, and for example, a method of driving the reticle R, or a reticle R and projection optics. A field lens arranged between the lens and the system PL is two-dimensionally moved in a plane perpendicular to the optical axis of the projection optical system PL, or is tilted with respect to this plane, or the air pressure between specific lens elements is changed. It is also possible to adopt a method of making it happen.

Further, it is also possible to obtain the graph as shown in FIG. 7 and to obtain the spherical aberration from the asymmetry in the same manner as the above-mentioned measuring method, and it is also possible to make the same correction by driving the lens element as described above. is there. The spherical aberration can be corrected by inserting a parallel flat plate glass between the projection optical system PL and the wafer and changing its thickness. For changing the thickness, for example, a pair of wedge-shaped glasses may be slid, or a rotating plate provided with a plurality of kinds of glasses having different thicknesses may be rotated.

Further, in the above, the reference member 20 having the minute opening 21 is arranged on the Z stage 14, and the illumination light for the image forming characteristic of the projection optical system PL is irradiated from below the stage.
The reflected light from the back surface of the reticle R is received by the photoelectric detector 28 again through the projection optical system PL and the minute opening 21, but the illumination light for measuring the imaging characteristics is emitted as follows. You may do it.

That is, a reticle having an opening pattern is used, and the reference reflecting surface on the wafer stage is irradiated with measurement illumination light from above the reticle through the opening pattern and the projection optical system PL, and the light is reflected from the reference reflecting surface. The light is received by the photoelectric detector through the projection optical system PL and the opening pattern of the reticle. In this case, the photoelectric detector is arranged at a position almost conjugate with the reticle. Further, the imaging characteristics can be similarly obtained by providing the opening pattern at a plurality of positions of the actual exposure reticle without using the test reticle.

Although the reduction projection type exposure apparatus for manufacturing a semiconductor integrated circuit has been described above as an example, the present invention is another apparatus, for example, an exposure apparatus used for manufacturing a large-sized liquid crystal display element substrate. The same can be applied to.
Further, the apparatus disclosed in the above embodiments can be applied to a surface position detecting apparatus that detects the surface position of the position of the second substrate with respect to the first substrate.

[0078]

According to the present invention, the curvature of field, the inclination of the field or the average focus position can be obtained at high speed and with high accuracy. Therefore, by separately operating the image forming characteristic correcting means, the image forming characteristic of the projection optical system can be corrected without lowering the throughput of the apparatus. Further, even when the mask on which the actual circuit pattern is formed is set, the imaging characteristic of the projection optical system can be measured without lowering the throughput, and by combining with the predictive control of the imaging characteristic, the parameter for predictive control can be set. Calibration can also be performed, and the correction accuracy of the imaging characteristics is improved.

If a plurality of minute apertures formed on the reference member and having different optical path lengths with respect to the projection optical system have different optical path lengths due to the steps on the reference member, the reference member should be thin. it can. Further, when a plurality of minute apertures formed on the reference member and having different optical path lengths with respect to the projection optical system have their optical path lengths changed by a phase shifter provided on the reference member, the minute apertures are changed. The part can be easily formed.

[Brief description of drawings]

FIG. 1 is a configuration diagram including a partial cross-sectional view showing an entire embodiment of a projection exposure apparatus according to the present invention.

FIG. 2A is a minute opening 21 of the reference member 20 of FIG.
FIG. 3B is a cross-sectional view showing an example, and FIG. 3B is a cross-sectional view showing another example of the minute opening of the reference member 20.

3A is a waveform diagram showing the relationship between the Z-direction position of the reference member 20 of FIG. 1 and the detection signals SA and SB output from the photoelectric detector 28, and FIG. 3B is a detection signal SA and detection signal SA. 6 is a waveform diagram showing the relationship between the difference signal SC from the signal SB and the position of the reference member 20 in the Z direction. FIG.

FIG. 4 is a plan view showing a place where an opening pattern of a reference member 20 is installed in an exposure area of the projection optical system PL of FIG.

5A is a minute opening 21 of the reference member 20 of FIG.
Is a plan view showing an example of the pattern of the minute opening 21, a plan view showing another example of the pattern of the minute opening 21, and a plan view showing yet another example of the pattern of the minute opening 21. is there.

FIG. 6 is a cross-sectional view of the projection optical system PL of FIG. 1 viewed from the direction of the optical axis AX.

7 is a characteristic diagram showing the output signal of the focus detection system 29 with respect to the image height of the minute opening 21 of the reference member 20. FIG.

8 is a characteristic diagram showing the characteristic diagram of FIG. 7 divided into an image plane inclination component and an image plane curvature component.

FIG. 9 is an explanatory diagram for removing a field curvature.

FIG. 10 is an explanatory diagram when only half of the exposure area of the projection optical system PL is used.

11 is an explanatory diagram when a phase grating pattern is used as the minute openings 21 of the reference member 20 of FIG.

FIG. 12 shows a case where a phase grating pattern is used,
3 is a waveform diagram showing the relationship between the output signal of the photoelectric detector 28 and the output signal of the focus detection system 29 in FIG. 1. FIG.

[Explanation of symbols]

1 Light Source 2 Shutter 14 Z Stage 15 XY Stage 19 Interferometer 20 Reference Member 21 Minute Aperture 21a Lower Aperture Pattern Part 21b Upper Aperture Pattern Part 28 Photoelectric Detector 29 Focus Detection System 30 Leveling Detection System 32 Main Control System 32A Combination Focus position detector 32B Imaging characteristic detection unit 33 Drive element control system 40, 41, 43, 45 Lens element 48a, 48b, 49a, 49b, 50a, 50b Drive element R Reticle PL Projection optical system W Wafer 54 Phase member

Claims (3)

[Claims] 1. A first illumination system for uniformly illuminating a mask with exposure light, a projection optical system for forming an image of the pattern of the mask on an image forming surface, and a photosensitive surface for the image forming surface of the projection optical system. In a projection exposure apparatus having a stage that holds the photosensitive substrate so that the exposure surfaces of the substrates are substantially aligned and is movable in the optical axis direction of the projection optical system and in a direction perpendicular to the optical axis, A reference member that is disposed and has a plurality of minute apertures each having a different optical path length with respect to the projection optical system formed in close proximity; and the plurality of minute apertures with the exposure light or light having a wavelength substantially equal to the exposure light. A second illumination system that illuminates from the stage side and makes transmitted light that has passed through the plurality of minute openings enter the mask through the projection optical system; and the mask is illuminated by the second illumination system. Within the light that has reached Photoelectric conversion means for individually photoelectrically converting the light reflected by the mask and transmitted again through each of the plurality of minute openings through the projection optical system; and a plurality of photoelectric conversions output from the photoelectric conversion means. Image plane state detection means for detecting the positions of the projection optical system in the optical axis direction at the positions of the plurality of minute openings on the image plane of the projection optical system based on the signal; and the image plane state detection means. An image forming characteristic calculating means for calculating at least one image forming characteristic among the focus position, the image plane inclination and the image plane curvature of the projection optical system based on each position information detected by Projection exposure system. 2. The plurality of minute apertures formed on the reference member and having different optical path lengths with respect to the projection optical system are changed in optical path length by a step on the reference member. 1. The projection exposure apparatus according to 1. 3. A plurality of minute apertures formed on the reference member and having different optical path lengths with respect to the projection optical system are changed in optical path length by a phase shifter provided on the reference member. The projection exposure apparatus according to claim 1.