JPH08255741A - Aligner - Google Patents

Abstract

(57) [Abstract] [Purpose] Even if illuminance fluctuations such as arc fluctuations occur in the exposure light source, the integrated exposure amount on the photosensitive substrate is kept within an appropriate range. [Structure] Illumination detection signal S1 obtained by splitting illumination light from mercury lamp 1 by beam splitter 31 and receiving it by integrator sensor 33, and amplifying its output signal,
The target illuminance signal S2 output from the setting unit 73 is supplied to the variable gain amplifier 77 in the power supply system 22, and the mercury lamp 1 is turned on by the power supply drive signal S3 output from the variable gain amplifier 77. In the fluctuation detection system 25, the illuminance detection signal S
1 and the light source drive signal S3 are compared, and when the difference becomes large, the gain switching signal S5 is set to the high level "1".
As a result, the gain of the variable gain amplifier 77 is switched to a higher one, and the light source drive signal S3 is oscillated.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an exposure apparatus used for manufacturing a semiconductor device, a liquid crystal display device or the like in a photolithography process, and in particular, for scanning a mask and a photosensitive substrate in synchronization. The present invention relates to a scanning exposure type exposure apparatus such as a so-called slit scan method or a step-and-scan method that sequentially exposes a pattern on a mask onto the substrate.

[0002]

2. Description of the Related Art An exposure apparatus used for transferring and exposing a pattern of a reticle as a mask onto each shot area of a wafer (or a glass plate etc.) coated with a photoresist when manufacturing a semiconductor device or the like. Generally, an illuminance control mechanism is provided to keep the exposure amount for each shot area on the wafer within an appropriate range. The illuminance control mechanism in such an exposure apparatus mainly has an illuminance distribution control mechanism for suppressing unevenness of the illuminance distribution in the illumination area of the reticle, and an integrated exposure amount for each shot area on the wafer to be an appropriate exposure amount. Exposure dose control mechanism for

In this regard, the conventional exposure apparatus is mainly arranged such that each shot area of the wafer is positioned at the exposure position by the step-and-repeat method, and then the reticle pattern is set in a stationary state via the projection optical system. A batch exposure type projection exposure apparatus (stepper or the like) for transferring to a shot area has been used. In the collective exposure method, first, the illuminance distribution control is performed by superimposing light fluxes from a large number of light source images using an optical integrator (fly eye lens or the like) provided in the illumination optical system.
Further, in the batch exposure method, since each shot area is exposed in a static state, the integrated exposure amount for each shot area is calculated by dividing the light flux for monitoring obtained by branching the illumination light for exposure into the actual exposure time. It is calculated by multiplying a signal obtained by continuously receiving light inside and integrating the photoelectric conversion signal of the monitoring light flux by a predetermined coefficient which is experimentally obtained in advance.

Therefore, the exposure amount control mechanism for the projection exposure apparatus of the batch exposure system includes a photoelectric detector (integrator sensor) that receives the light flux for monitoring, and an integrating means that integrates the detection signal of the integrator sensor. It can be easily configured by the control means for controlling the illuminance of the illumination light or the exposure time so that the difference between the integration result by the integration means and the target value becomes small.

Further, in order to improve the resolution and the depth of focus for a fine periodic pattern, for example, a modified light source method (for example, a special light source method) in which the aperture stop of the illumination system is formed of a plurality of apertures eccentric with respect to the optical axis. Kaihei 4-225358), or an annular illumination method in which the shape of the illumination system aperture stop is annular. Even when the shape of the aperture of the illumination system aperture stop changes in this way, by arranging the light receiving surface of the integrator sensor on the detection surface that is substantially conjugate with the wafer surface, The illuminance of can be accurately monitored. Therefore, for example, by controlling the exposure time so that the value obtained by integrating the detection signal of the integrator sensor converges to a predetermined target value, the integrated exposure amount in each shot area of the wafer can be easily adjusted to an appropriate value. Can fit in range.

On the other hand, recently, a single chip pattern such as a semiconductor element tends to be large in size, and in a projection exposure apparatus, a large area for efficiently exposing a pattern of a larger area onto a wafer. Is required. In order to achieve such a large area, it is particularly necessary to keep the distortion within a predetermined amount or less on the entire surface. Therefore, in order to reduce the distortion over the entire wide exposure area, after stepping each shot area on the wafer to the scanning start position, the reticle and the wafer are synchronously scanned with respect to the projection optical system, so that the reticle on the reticle is scanned. An exposure apparatus of a so-called step-and-scan method, which sequentially exposes each pattern on each shot area on the wafer, has been attracting attention. This step-and-scan type projection exposure apparatus uses a conventional projection optical system of the same size to synchronously scan a reticle and a wafer, thereby sequentially exposing the reticle pattern onto the entire surface of the wafer. This is an extension of the slit scan type projection exposure apparatus (aligner, etc.).

Among the illuminance control mechanisms of the projection exposure apparatus of the scanning exposure system such as the slit scan system or the step-and-scan system, the illuminance distribution control system is an optical integrator as in the case of the collective exposure system. in use. However, when a fly-eye lens is used as an optical integrator, the incident surface of each lens element of the final stage fly-eye lens is conjugated with the pattern surface of the reticle. Further, in the scanning exposure method, the illumination area on the reticle is an elongated rectangular or arcuate area (hereinafter referred to as “slit-shaped illumination area”). The cross-sectional shape of each lens element forming the lens is preferably an elongated rectangle that is substantially similar to the slit-shaped illumination area.

On the other hand, as the exposure amount control mechanism for the scanning exposure system, it is difficult to directly apply the exposure amount control mechanism for the collective exposure system. In the scanning exposure method, since each shot area of the wafer is scanned with respect to a slit-shaped exposure field shorter than the length of these shot areas, the cumulative exposure amount in each shot area is controlled by It is executed so that the integrated exposure amount is constant at all points on the wafer. If the integrated exposure amount at each point on the wafer is different, unevenness of the integrated exposure amount occurs in each shot area, which is similar to the uneven illuminance in the illumination area in the batch exposure type exposure apparatus. There will be an error.

Further, as one method for controlling the integrated exposure amount in the batch exposure method, the exposure time is controlled by opening and closing a shutter, for example, but in the scanning exposure method, the exposure is continuously performed, so that the wafer is exposed. The integrated exposure amount at each point above cannot be controlled by opening and closing the shutter. Therefore, in the scanning exposure method, for example, the reticle and the wafer are each scanned at a predetermined constant speed to control the integrated exposure amount. With such a method of controlling the scanning speed, it is difficult to finely adjust the integrated exposure amount with time. Therefore, in the scanning exposure method, it is further necessary to perform illuminance control so that the illuminance is continuously maintained temporally during the exposure of each shot area. In this regard, in the case of the batch exposure method, as a control method for keeping the illuminance constant, the illuminance of the illumination light is constantly monitored, and the result is fed back to the power source of the exposure light source, and the power source for the exposure is used. A constant illuminance control method for controlling electric power supplied to a light source is known.

[0010]

However, even if the constant illuminance control method used in the batch exposure method is directly applied to the exposure apparatus of the scanning exposure method, the integrated exposure amount at each point on the wafer is within an appropriate range. Can be difficult to fit in. That is, for example, when a discharge lamp that performs arc discharge such as a mercury lamp is used as a light source for exposure, illuminance variation of illumination light called so-called arc fluctuation occurs due to the state of gas convection in the discharge lamp. Sometimes. The arc fluctuation is a fluctuation of the illuminance of, for example, about 30 Hz, and when the arc fluctuation occurs, the correspondence between the power supplied to the discharge lamp and the illuminance of the illumination light is destroyed, so that the illuminance actually measured. Even if you perform the conventional constant illuminance control that controls the power supplied to the discharge lamp based on
In some cases, the integrated exposure amount on the wafer was out of the proper range.

Therefore, when the exposure is performed by the scanning exposure method, the integrated exposure amount partially deviates from the proper range in the shot area where the arc fluctuation occurs, which results in the yield of the finally manufactured semiconductor elements and the like. Get worse 1
Was one factor. Further, as an exposure light source, for example, an excimer laser light (KrF excimer laser or ArF) is used.
Even if a laser beam such as an excimer laser or the like, or a harmonic wave of a YAG laser is used, if the output of the light source fluctuates to a relatively large degree for some reason, the integrated exposure amount may partially deviate from the proper range. . Therefore, even when a laser light source is used, a mechanism for suppressing fluctuation of illuminance is required.

In view of the above problems, the present invention can keep the integrated exposure amount on the photosensitive substrate within an appropriate range even when the illuminance of the illumination light output from the exposure light source fluctuates. It is an object of the present invention to provide a scanning exposure type exposure apparatus that can be used.

[0013]

An exposure apparatus according to the present invention is, for example, as shown in FIGS. 1 and 7, an exposure light source (1) for generating exposure illumination light, and a transfer pattern by the illumination light. Illumination optical system (6,8,9,16,34,3) for illuminating a predetermined illumination area on the formed mask (R)
8, 39), and scanning the photosensitive substrate (W) in the direction corresponding to the predetermined direction in synchronization with the scanning of the mask (R) in the predetermined direction with respect to the predetermined illumination area. In a scanning exposure type exposure apparatus that sequentially exposes a pattern of a mask (R) on a photosensitive substrate (W), an exposure amount measuring means (31 to 31) that continuously measures the exposure energy of illumination light.
33) and illuminance fluctuation detecting means (25, 33) for detecting fluctuations in illuminance of illumination light generated from the exposure light source (1).
And illuminance control means (20, 22) for controlling the output of the exposure light source (1) based on the output signals from the exposure amount measurement means (31-33). The fluctuation detection means (25, 33) changes the output of the exposure light source (1) at a frequency higher than a predetermined frequency when fluctuations in illuminance are detected.

In this case, an example of the illuminance fluctuation detection means is a photoelectric detection means (33) for detecting the quantity of illumination light.
And a comparison means (74) for detecting a difference between the output signal (S1) of the photoelectric detection means and the signal (S3) corresponding to the electric power supplied to the exposure light source (1). Another example of the illuminance fluctuation detection means is a photoelectric detection means (33) for detecting the light quantity of the illumination light, a low-pass filter means for extracting a low frequency component from the output signal of the photoelectric detection means, Is to have.

Further, the illuminance control means amplifies the variation of the output signal from the exposure amount measuring means (31 to 33) by a predetermined gain and controls the output of the exposure light source (1) (gain varying means ( 77), and the gain in the gain varying means is preferably controlled according to the fluctuation of the illuminance detected by the illuminance fluctuation detecting means. In these cases, an example of the exposure light source is a discharge lamp, and in this case, the fluctuation of the illuminance of the illumination light is caused by the fluctuation of the discharge of the discharge lamp (arc fluctuation).

[0016]

According to the present invention, the fluctuation of the illuminance of the illumination light is detected by the illuminance fluctuation detecting means (25, 33) when a certain shot area on the photosensitive substrate is exposed by the scanning exposure method. Then, the illuminance control means (20, 22) changes the output of the exposure light source (1), that is, the illuminance of the illumination light at a high frequency. In this case, since the integrated exposure amount at the point exposed when the fluctuation of the illuminance occurs on the shot area is the integration of the illuminance that fluctuates at the high frequency, the obtained integrated exposure amount is appropriate. It comes to fall within the range.

Further, the illuminance fluctuation detecting means uses the photoelectric detecting means (33) for detecting the light amount of the illumination light, the output signal (S1) of the photoelectric detecting means and the electric power supplied to the exposure light source (1). Comparing means (74) for detecting a difference from the corresponding signal (S3), the comparing means (7)
It can be determined that the fluctuation of the illuminance has occurred when the absolute value of the difference output from 4) exceeds a predetermined threshold value.

Further, the illuminance fluctuation detecting means includes a photoelectric detecting means (33) for detecting the light quantity of the illuminating light and a low pass filter means for extracting a low frequency component from the output signal of the photoelectric detecting means. If so, it may be determined that the fluctuation of the illuminance has occurred when the value of the output signal from the low-pass filter means exceeds a predetermined threshold value. Further, the illuminance control means includes a gain varying means (77) for controlling the output of the exposure light source (1) by amplifying the change amount of the output signal from the exposure amount measuring means (31 to 33) with a predetermined gain. When the gain in the gain varying means is controlled according to the fluctuation of the illuminance detected by the illuminance fluctuation detecting means, the gain varying means (77) provides the gain when the fluctuation of the illuminance is detected. To increase. As a result, the output of the exposure light source (1) changes at a high frequency.

[0019]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of an exposure apparatus according to the present invention will be described below with reference to the drawings. In this embodiment, the present invention is applied to a step-and-scan type projection exposure apparatus. FIG. 1 shows a projection exposure apparatus of this embodiment,
In FIG. 1, the illumination light from the mercury lamp 1 is condensed by the elliptical mirror 2. A shutter 4 which is opened and closed by a shutter control mechanism 5 is arranged in the vicinity of the converging point, and when the shutter 4 is in an open state, the illumination light is converted into a substantially parallel light flux via a mirror 3 and an input lens 6, Reach field stop 7. Immediately after the field stop 7, a light reduction plate 23 is arranged so that it can be freely taken in and out, and the light reduction plate 23 gradually changes the amount of illumination light passing through the field stop 7 within a predetermined range.
Alternatively, it can be changed continuously.

The light-reducing plate 23 is, for example, a plurality of reflective half mirrors arranged so as to be switchable, and the inclination of each half mirror with respect to the optical axis is set so that the overall transmittance becomes a predetermined transmittance. Has been done. Then, the dimming plate 2 is driven by the dimming plate driving mechanism 24 including the driving motor.
By moving 3 in steps, the amount of illumination light is adjusted. In this embodiment, it is the exposure amount control system 20 that controls the exposure amount for the wafer W, and the exposure amount control system 2
0 controls the operation of the dimming plate drive mechanism 24 and also controls the operation of the shutter control mechanism 5. Further, the exposure amount control system 20 controls the current supplied to the mercury lamp 1 via the power supply system 22 for the mercury lamp 1.

After passing through the aperture of the field stop 7, the dimming plate 2
The illumination light whose amount of light is adjusted by 3 enters the first fly-eye lens 9 of the two-stage fly-eye lens group via the first relay lens 8. Illumination light from the plurality of light source images by the first fly-eye lens 9 is emitted by the second relay lens 12
It is guided to the second fly-eye lens 14 via A. In this example, the light quantity diaphragm 10 is arranged near the exit surface of the first fly-eye lens 9, that is, the surface on which the light source image is formed.
The size of the opening can be adjusted to any size by the light amount diaphragm drive mechanism 11. The operation of the light amount diaphragm drive mechanism 11 is also controlled by the exposure amount control system 20. In this example, by adjusting the size of the aperture of the light amount diaphragm 10, the light amount of the illumination light traveling from the first fly-eye lens 9 to the second fly-eye lens 14 can be continuously adjusted.

FIG. 2A shows an example of the light quantity diaphragm 10. In FIG. 2A, the light quantity diaphragm 10 is composed of an iris diaphragm. In this case, for example, by moving a lever around the iris diaphragm, the size of the substantially circular opening of the iris diaphragm can be continuously adjusted as shown in FIG. Returning to FIG. 1, recently, by narrowing the numerical aperture (NA) of the illumination optical system, that is, by reducing the coherence factor (σ value), which is the ratio of the numerical aperture of the illumination optical system to the numerical aperture of the projection optical system, , Techniques for improving the depth of focus for a given pattern have been developed. When the σ value is reduced in this way, the illuminance of the illumination light that illuminates the reticle decreases. In this example, an adjusting mechanism for adjusting the size of the illumination area on the incident surface of the second fly-eye lens 14 is provided as a means for preventing such a decrease in illumination light intensity.

The adjusting mechanism is the second relay lens 12A.
And a second relay lens 12B having a refractive power larger than that of the second relay lens 12A and an exchange mechanism 13 for switching between the two second relay lenses 12A and 12B. The operation of the exchange mechanism 13 is controlled by a main control system 19 which controls the operation in a centralized manner. Then, when performing illumination with a normal σ value, one second relay lens 12A is provided between the first fly-eye lens 9 and the second fly-eye lens 14 via the exchange mechanism 13.
Is arranged, whereby almost the entire incident surface of the second fly-eye lens 14 is illuminated by the illumination light. On the other hand, when the illumination is performed with a small σ value (the numerical aperture of the illumination optical system is narrowed), the second fly-eye lens 9 and the second fly-eye lens 14 are connected to each other via the exchange mechanism 13. The two relay lens 12B is arranged so that the central portion of the incident surface of the second fly-eye lens 14 is partially illuminated with the illumination light. Therefore, when the σ value is reduced, the illuminance of the illumination light at the stage of the second fly-eye lens 14 increases, so that the illuminance of the illumination light on the reticle and the wafer is substantially constant regardless of the σ value. Maintained.

Although the adjusting mechanism of this embodiment is of a switching type, the adjusting mechanism is the same as that of the first fly-eye lens 9 and the second fly-eye lens 9.
The zoom lens system may be arranged between the fly-eye lens 14 and the zoom lens system, and a zooming mechanism for zooming the zoom lens system. By using the zoom lens system as described above, the size of the illumination visual field on the incident surface of the second fly-eye lens 14 can be continuously changed. Therefore, even when the σ value is continuously changed, there is an advantage that the illuminance on the reticle and the wafer can always be kept high.

Next, the second fly-eye lens 14 of this example
Are two lens bundles 14a and 1a each having a flat surface on one side in which the lens elements are closely arranged in a mosaic shape.
4b are arranged close to each other so that their flat portions face each other. Therefore, the second fly-eye lens 14 is hereinafter referred to as a "mosaic type fly-eye lens 14".

FIG. 3 (a) is a side view of the mosaic type fly's eye lens 14 of this example. In FIG. 3 (a),
Each flat surface portion FB along the optical axis AX1 of the illumination optical system
A mosaic fly-eye lens 14 is composed of two lens bundles 14a and 14b arranged so that FC and FC face each other at a distance δ. In this case, each lens element forming the first lens bundle 14a on the light source side has a refracting power on the incident surface FA side, and each lens element forming the second lens bundle 14b on the reticle side emits surface respectively. It has a refractive power on the FD side.

Further, the parallel light flux which is incident on the first lens bundle 14a from the light source side is the exit surface F of the second lens bundle 14b.
The refractive power of each lens element is such that the parallel light flux that is condensed on D and that is incident on the second lens bundle 14b from the reticle side is condensed on the incident surface FA of the first lens bundle 14a. Has been defined. That is, the second lens bundle 14b
The exit surface FD is the focal plane of the mosaic fly-eye lens 14, and a large number of light source images are formed on the exit surface FD. Therefore, the lens bundles 14a and 14b act as one fly-eye lens only when the two are combined. The two lens bundles 14a and 14b of the mosaic fly-eye lens 14 shown in FIGS.
The number of lens elements constituting the above is an example, and the number of lens elements is determined according to the required accuracy of the uniformity of the illuminance distribution that is actually required.

FIG. 3B is a front view showing the first lens bundle 14a taken along line AA of FIG. 3A, and FIG. 3C is taken along line CC of FIG. 3A. FIG. 4 is a front view showing the second lens bundle 14b, and in FIGS. 3A and 3C, the direction corresponding to the scanning direction of the reticle during scanning exposure of the projection exposure apparatus of this example is defined as X1 direction, and The direction corresponding to the non-scanning direction perpendicular to the scanning direction is the Y1 direction.

In this case, as shown in FIG. 3B, the first
The lens bundle 14a of No. 1 has the first row by arranging the lens elements 61 each having a slender rectangular cross-sectional shape with a width dx in the X1 direction and a width dy (dy> dx) in the Y1 direction in close contact with each other in the Y1 direction. 62A, 2nd row 62B, 3rd
A lens group of each row of the rows 62C, ... And an odd-numbered lens group of the first row 62A, the third row 62C, ... And an even-numbered second row 62B, the fourth row 62D ,. And are shifted in the Y1 direction by 1/2 of the width dy of the lens element.

In this example, as shown in FIG. 3A, the entrance surface of the mosaic fly-eye lens 14, that is, the first lens bundle 1
When the incident surface FA of 4a is conjugate with the pattern surface of the reticle, and the cross-sectional shape of the lens element 61 forming the first lens bundle 14a is similar to the slit-shaped illumination area on the reticle, the illumination efficiency is the highest. Becomes higher. Therefore, the width d in the X1 direction of the cross-sectional shape of the lens element 61
The value of the ratio of x to the width dy in the Y1 direction is set to be substantially equal to the value of the ratio of the width in the scanning direction of the slit-shaped illumination area on the reticle to the width in the non-scanning direction. Therefore, the cross section of the lens element 61 is Y corresponding to the non-scanning direction.
It is a rectangle elongated in one direction. As an example, dx:
dy is set to about 1: 3.

Further, as shown in FIG. 3C, the second lens bundle 14b has a width ex (= 2 · dx) in the X1 direction and a Y value.
By arranging the lens elements 65 having a substantially square cross section with a width ey (= dy / 2) in one direction in close contact with each other in the X1 direction, the first row 66A and the second row
A lens group of each row of the row 66B, the third row 66C, ... And an odd-numbered first row 66A, a third row 66C, ... and an even-numbered second row 66B, the fourth row 66D,
The lens group of ... is the width ex of the lens element in the X1 direction.
It is configured to be shifted by 1/2. Incidentally, when the cross-sectional shape of the lens element 61 of the first lens bundle 14a is about dx: dy = 1: 3, the cross-sectional shape of the lens element 65 of the second lens bundle 14b is ex: ey = 2: Since 1.5 = 4: 3, the lens element 65 has a substantially square cross section.

In such an arrangement, the center of a certain lens element of the first lens bundle 14a and the center of a certain lens element of the second lens bundle 14b are further aligned in the X1 direction and the Y1 direction. This allows
The centers 63 of all the lens elements 61 forming the first lens bundle 14a and the centers 67 of all the lens elements 65 forming the second lens bundle 14b are arranged at the same position in the X1 direction and the Y1 direction. ing.

Thus, the mosaic type fly-eye lens 1
The mosaic type fly-eye lens 14 of the present embodiment is a fly-eye lens of the second stage, and the operation and effect when 4 is divided into two lens bundles 14a and 14b will be described.
The individual light source images formed on the exit surface of the fly-eye lens in the first stage are the same as the light source images formed in the aperture of the light quantity diaphragm 10 on the exit face of the fly-eye lens 9 in the first stage. It is a statue. That is, each light source image formed on the exit surface of the mosaic fly-eye lens 14 is a large number of minute light source images uniformly distributed in, for example, a circular area.

Accordingly, the light source image formed by projecting the light source image formed on the exit surface of the mosaic type fly-eye lens 14 onto the end surface of the first lens bundle 14a as shown in FIG. A minute light source image is distributed in a circular area 64 centered on the center 63 of each lens element 61. The circular area 64 is the light amount diaphragm 1 shown in FIG.
It is similar to the shape of the 0 opening. However, since the cross-sectional shape of each lens element 61 of the first lens bundle 14a of this example is an elongated rectangular shape, when the aperture of the light quantity diaphragm 10 is set to be large, the circular area 63 forms an end surface of each lens element 61. It will stick out. Therefore, when a fly-eye lens in which lens elements having the same cross-sectional shape as the lens element 61 are bundled is used instead of the mosaic type fly-eye lens 14, vignetting of the light source image occurs on the exit surface and the illumination efficiency decreases. .

On the other hand, in this example, the first lens bundle 1
Immediately after 4a, as shown in FIG. 3C, a second lens bundle 14b composed of lens elements 65 each having a substantially square cross-sectional shape is arranged, and a circle having a center 67 of each lens element 65 as a center. The light source image is formed so as to be distributed in the area 64 of the. In this case, since the cross-sectional shape of the lens element 65 is close to a square, even when the aperture of the light quantity diaphragm 10 in FIG. 2 is set large, the circular region 64 is substantially within the cross section of the lens element 65. Therefore, the mosaic fly-eye lens 14
Vignetting of a large number of light source images formed on the exit surface of is reduced, and the illumination efficiency is improved. The illumination light from a large number of light source images formed on the exit surface of the mosaic fly-eye lens 14 illuminates the illumination light in a superimposed manner, resulting in extremely high uniformity of the illuminance distribution on the reticle and the wafer. .

Further, in FIG. 1, the second lens bundle 14b on the reticle side of the mosaic type fly-eye lens 14 is shifted in the direction perpendicular to the optical axis AX1 and the lens bundle 14b An adjusting mechanism 15 for adjusting the tilt angle (inclination angle) within a predetermined range is attached. In this example, the shift amount of the lens bundle 14b and the tilt angle are adjusted via the adjusting mechanism 15 to correct the shift amount of the telecentricity in the illumination optical system. For example, the main control system 19 controls the operation of the adjusting mechanism 15 when the mercury lamp 1 is replaced or when the illumination conditions are switched (switching between the normal illumination and the modified light source, etc.), so that the telecentricity is automatically achieved. Will be corrected.

In FIG. 1, an illumination system aperture stop plate 16 having a plurality of types of illumination system aperture stops is installed near the exit surface of the mosaic fly-eye lens 14. FIG. 4 shows the aperture stop plate 16 of the illumination system. In FIG. 4, the aperture stop 18A of a normal circular aperture and the coherence aperture of a small circular aperture are formed on the illumination system aperture stop plate 16 at substantially equal angular intervals. An aperture stop 18B for reducing the σ value that is a factor, a ring-shaped aperture stop 18C for annular illumination, and a modified aperture stop 18D in which a plurality of apertures are eccentrically arranged for the modified light source method are arranged. ing. By rotating the illumination system aperture stop plate 16, a desired aperture stop can be selected from the four aperture stops.

Returning to FIG. 1, the main control system 19 controls the rotation angle of the illumination system aperture diaphragm plate 16 via the illumination system diaphragm drive mechanism 17 composed of a drive motor. After being emitted from the mosaic fly-eye lens 14, the illumination system aperture stop plate 16
The illumination light IL that has passed through the aperture stop selected from the inside enters the beam splitter 31 having a transmittance of about 98%. Then, the illumination light transmitted through the beam splitter 31 passes through the first relay lens 34 and the two movable blades 3
Reach a movable blind (variable field stop) with 5A and 35B. Hereinafter, the movable blind will be referred to as “movable blinds 35A and 35B”. Movable blind 35A,
The arrangement surface of 35B is the Fourier transform surface of the exit surface of the mosaic fly-eye lens 14. That is, the arrangement surface of the movable blinds 35A and 35B is conjugate with the pattern formation surface of the reticle R described later,
A fixed blind 37 having a fixed opening shape is arranged near 35B.

The fixed blind 37 is, for example, a mechanical field stop that surrounds a rectangular opening with four knife edges, and the rectangular opening defines the shape of a slit-shaped illumination area on the reticle R. . That is, the illumination light IL limited by the movable blinds 35A and 35B and the fixed blind 37 is evenly distributed over the slit-shaped illumination area 41 on the reticle R via the second relay lens 38, the condenser lens 39, and the mirror 40. Illuminate with illuminance distribution.

In this case, the arranging surface of the fixed blind 37 is slightly defocused in the front or rear direction from the conjugate surface of the pattern forming surface of the reticle R, so that the illuminance of the contour portion of the slit-shaped illumination area 41 is increased. The distribution changes with a certain slope. In addition, the movable blinds 35A and 35B
Serves to prevent the slit-shaped illumination area from covering an area of the reticle R which should not be exposed at the start and end of the scanning exposure. Therefore, the movable blade 35A
And 35B are slide mechanisms 36A and 36B, respectively.
It is supported so that it can be opened and closed. The slide mechanisms 36A and 36B constitute a movable blind drive mechanism,
The operation of the movable blind drive mechanism is controlled by the stage control system 46.

The image of the pattern in the illumination area 41 on the reticle R is slit-shaped exposed on the wafer W through the projection optical system PL at a projection magnification β (β is, for example, 1/4 or 1/5). It is projected on the field 47. Here, the Z axis is taken parallel to the optical axis of the projection optical system PL, the X axis is taken parallel to the scanning direction of the reticle R and the wafer W at the time of scanning exposure in the plane perpendicular to the Z axis, and the axis perpendicular to the Z axis. The Y axis is taken in a direction (non-scanning direction) perpendicular to the X axis in the plane. In this example, Reticle R
Are held on the reticle base 43 via the scanning stage 42 which is slidable in the X direction, and the wafer W is held on the wafer stage 48 which scans the wafer W in the X direction and positions it in the Y direction. . The wafer stage 48 also incorporates a Z stage or the like for positioning the wafer W in the Z direction.

Scan stage 42 and wafer stage 48
A stage drive mechanism is constituted by the stage drive mechanism, and the operation of the stage drive mechanism is controlled by the stage control system 46.
During scanning exposure, the stage control system 46 controls the scanning stage 4
2 in synchronism with scanning the reticle R at a predetermined speed V _R with respect to the illumination region 41 + X direction (or -X direction) via the predetermined shot area on the wafer W via the wafer stage 48 The exposure field 47 is scanned in the −X direction (or + X direction) at the speed V _W (= β · V _R ). As a result, the pattern of the reticle R is successively transferred and exposed onto the shot area. Further, the stage control system 46 controls the positions of the movable blinds 35A and 35B via the slide mechanisms 36A and 36B during scanning exposure. A control method in this case will be described with reference to FIG.

First, immediately after the start of scanning exposure, as shown in FIG.
As shown in (a), the image 37R of the opening portion of the fixed blind 37 of FIG. 1 is exposed to the outside with respect to the light-shielding band 88 surrounding the pattern region 87 of the reticle R. Therefore, in order to avoid exposure to an unnecessary portion, the position of the movable blade 35B in FIG. 1 is moved so that one edge portion 35Ra of the image 35R of the movable blinds 35A and 35B is put in the light shielding band 88. After that, as shown in FIG. 6B, when the image 37R of the fixed blind 37 is within the pattern area 87 in the scanning direction, the movable blinds 35A and 35B.
Image 35R is set so as to surround the image 37R. Then, at the end of the scanning exposure, as shown in FIG. 6C, when the image 37R of the fixed blind 37 appears outside the light-shielding band 88, the position of the movable blade 35A in FIG. 1 is moved, Image 35R of movable blinds 35A and 35B
The other edge portion 35Rb of the above is put in the light-shielding band 88. By such an operation, the slit-shaped illumination region 41 on the reticle R is prevented from coming out of the light-shielding band 88, and the exposure of an unnecessary pattern on the wafer W is prevented.

Further, in FIG. 1, the wafer stage 48
An illuminance nonuniformity sensor 49 including a photoelectric detector having a light receiving surface at the same height as the exposure surface of the wafer W near the upper wafer W.
Is installed, and the detection signal output from the uneven illuminance sensor 49 is supplied to the main control system 19. Further, a reference mark plate 50 used when performing reticle alignment or the like is provided on the wafer stage 48.
The reference mark 50a having an opening pattern is formed on the reticle 0, and the alignment mark is also formed on the reticle R so as to correspond to the reference mark 50a. For example, when the reticle R is exchanged, the reference mark plate 50 is moved into the effective exposure field of the projection optical system PL, and the reference mark 50a of the reference mark plate 50 is illuminated by the light source 51 from the bottom side with the same wavelength band as the illumination light IL. Illuminate with light. Under this illumination light, the images of the reference mark 50a and the alignment mark on the reticle R are observed by the reticle alignment microscope 44 via the mirror 45 above the reticle R. Then, the reticle R is aligned with the fiducial mark plate 50 based on this observation result.

Further, a reference mark for focus calibration is also formed on the reference mark plate 50, and a detection system is arranged at the bottom of this reference mark. Figure 5
FIG. 5A shows the reference mark for focus calibration and the detection system, and in FIG.
A reference mark 50b having, for example, a cross-shaped opening pattern is formed in the light-shielding film on the reference mark plate 50, and the detection system 54 is arranged at the bottom of the reference mark 50b. Using the reference mark 50b, the projection optical system P is
The position of the L image plane is obtained. That is, the detection system 54
At the wafer stage 48 via the optical fiber 81.
Illumination light having the same wavelength band as the illumination light IL of FIG. 1 is guided into the interior of the reference mark 50, and the illumination light causes the reference mark 50 to pass through the collimator lens 82, the half mirror 83, and the condenser lens 84.
Illuminate b from the bottom side. The illumination light passing through the reference mark 50b forms an image of the reference mark 50b on the pattern forming surface of the reticle R via the projection optical system PL, and the reflected light from the pattern forming surface passes through the projection optical system PL. To return to the reference mark 50b. Then, the illumination light passing through the reference mark 50b passes through the condenser lens 84, the half mirror 83, and the condenser lens 85 in the detection system 54, and then the photoelectric detector 8 is detected.
It is incident on 6.

The detection signal (photoelectric conversion signal) S6 of the photoelectric detector 86 is supplied to the main control system 19 of FIG. In this case, when the Z stage in the wafer stage 48 is driven to change the position of the reference mark 50b in the Z direction,
As shown in (b), the detection signal S6 is the reference mark 50b.
Changes so that it has a peak when the Z coordinate of is coincident with the position of the image plane of the projection optical system PL. Therefore, the detection signal S
The position of the image plane of the projection optical system PL can be obtained from the change of 6, and thereafter the exposure surface of the wafer W is set at that position, so that the exposure is performed in a good state. Therefore, by using the reference mark 50b of the reference mark plate 50, the position of the image plane of the projection optical system PL is calibrated (focus calibration).

Returning to FIG. 1, the leaked light reflected by the beam splitter 31 having a transmittance of about 98% is collected by the condenser lens 3
An integrator sensor 3 consisting of a photoelectric detector through 2
It is condensed on the light receiving surface of No. 3. Integrator sensor 3
The light receiving surface of 3 is conjugate with the pattern forming surface of the reticle R and the exposure surface of the wafer W, and the integrator sensor 33
Detection signal (photoelectric conversion signal) is supplied to the exposure amount control system 20. The detection signal is also supplied to the power supply system 22 for the mercury lamp 1 through the exposure amount control system 20.

When the aperture stop plate 16 for illumination system is rotated to set the aperture stop 18C for annular illumination or the modified aperture stop 18D shown in FIG. 4 on the exit surface of the second-stage fly-eye lens 14, Since the light receiving surface of the integrator sensor 33 is conjugate with the exposure surface, the incident angle of the illumination light on the integrator sensor 33 becomes large, and a sensitivity error may occur due to the incident angle. In order to reduce such a sensitivity error, for example, a diffuser plate that diffuses a light flux is arranged immediately before the light receiving surface of the integrator sensor 33 (or a conjugate surface with the exposure surface), and the light flux diffused by this is integrated. Light may be received by the sensor 33.

A memory 21 is connected to the exposure amount control system 20, and a conversion coefficient for obtaining the exposure energy on the wafer W from the output signal of the integrator sensor 33 is stored in the memory 21. However, in this embodiment, the output signal of the integrator sensor 33 is calibrated using, for example, a predetermined reference illuminance meter, and a correction coefficient for correcting the output signal of the integrator sensor 33 is also stored in the memory 21 based on the calibration result. Remembered in.

The light-receiving surface of the integrator sensor 33 is arranged at a position conjugate with the pattern surface of the reticle, so that even when the shape of the illumination-system aperture stop is changed by rotating the illumination-system aperture stop plate 16, the integrator sensor 33 is rotated. Sensor 3
No error occurs in the detection signal of 3. However,
The light receiving surface of the integrator sensor 33 is connected to the projection optical system PL.
Fourier transform plane (pupil plane) of the reticle pattern at
It may be arranged on an observation surface that is substantially conjugate with, so that all the light fluxes passing through this observation surface can be received.

Further, in this example, the integrator sensor 33 is used for the beam splitter 31 having a transmittance of about 98%.
A condenser lens 52 and a wafer reflectivity monitor 53 including a photoelectric detector are installed on the opposite side to the light receiving surface of the wafer reflectivity monitor 53, which is substantially conjugate with the surface of the wafer W by the condenser lens 52. . In this case, of the illumination light that passes through the reticle R and is irradiated onto the wafer W via the projection optical system PL, the reflected light on the wafer W is the projection optical system PL,
The wafer reflectivity monitor 53 receives the light via the reticle R and the like, and the detection signal (photoelectric conversion signal) is supplied to the main control system 19. The main control system 19 passes through the projection optical system PL based on the light amount of the illumination light IL applied to the reticle R side and the light amount of the reflected light on the wafer W calculated from the detection signal of the wafer reflectance monitor 53. The amount of light (power) of the illumination light to be obtained is obtained. Further, the main control system 19 predicts the thermal expansion amount of the projection optical system PL based on the thermal energy obtained by multiplying the light amount thus obtained by the exposure time, and the projection based on this predicted thermal expansion amount. The amount of change in the image forming characteristics such as distortion of the optical system PL is obtained. And the main control system 1
Reference numeral 9 corrects the image formation characteristic of the projection optical system PL to the original state via an image formation characteristic correction mechanism (not shown) connected to the projection optical system PL.

Next, the exposure amount control mechanism in the illuminance control mechanism of this example will be described in detail. In this regard, in this example, the mercury lamp 1 is used as a light source for exposure, but in a discharge lamp such as a mercury lamp, fluctuation of illuminance at a relatively low frequency of about 30 Hz, which is called "arc fluctuation", occurs. Sometimes. Since such an arc fluctuation causes unevenness of the integrated exposure amount on the surface of the wafer W, the exposure amount control mechanism of this example also incorporates a mechanism for detecting and correcting the arc fluctuation.

FIG. 7 shows the main part of the exposure amount control mechanism of this embodiment. In FIG. 7, the exposure amount control is performed by a preamplifier 71, a digital / analog (D / A) converter 72, and a setting unit 73. A system 20 is constructed. Then, the detection signal from the integrator sensor 33 becomes the illuminance detection signal S1 corresponding to the illuminance of the illumination light via the preamplifier 71, and the illuminance detection signal S1 is D / at the predetermined high sampling frequency.
It is taken into the setting unit 73 via the A converter 72. The setting unit 73 is also supplied with information on the target integrated exposure amount for the wafer W from the main control system 19, and further, the setting unit 73
As described above, the memory 21 connected to the memory 21 stores the conversion coefficient and the like for obtaining the actual exposure amount (exposure energy per unit time) on the wafer W from the value of the illuminance detection signal S1. The setting unit 73 can recognize the exposure amount on the wafer W from the illuminance detection signal S1.

The setting unit 73 sets conditions for obtaining the target integrated exposure amount before the start of scanning exposure. Therefore, in FIG. 1, assuming that the output power of the mercury lamp 1 is p, the transmittance of the dimming plate 23 is q ₁ , and the transmittance of the light amount diaphragm 10 is q ₂ , it changes depending on the shape of the illumination system aperture diaphragm. The exposure amount e on the wafer W is expressed as follows by using the coefficient k.

E = k · p · q ₁ · q ₂ (1) Further, assuming that the width of the slit-shaped exposure region 47 on the wafer W in the scanning direction is D, the wafer stage 48 is moved in the X direction during scanning exposure. The cumulative exposure amount ΣE on the wafer W is as follows using the equation (1), where V _W is the scanning speed of.

ΣE = e · (D / V _W ) = k · p · q ₁ · q ₂ · (D / V _W ) (2) In this case, the width D of the exposure area 47 in the scanning direction is fixed. Therefore, in order to control the integrated exposure amount ΣE to a predetermined target integrated exposure amount ΣE ₀ , the output power p of the mercury lamp 1, the transmittance q _{1 at} the light reducing plate 23, and the transmittance q at the light amount diaphragm 10 are set. ₂ or the scanning speed V _W of the wafer stage 48, or a plurality of them may be simultaneously adjusted. Therefore, in FIG. 7, in order to make the integrated exposure amount ΣE converge to a predetermined target integrated exposure amount ΣE ₀ , the setting unit 73
Supplies the target illuminance signal S2 corresponding to the target output power of the mercury lamp 1 to the power supply system 22, sets the transmittance q ₁ of the dimming plate 23 via the dimming plate driving mechanism 24 of FIG. And set the transmittance q ₂ at the light amount diaphragm 10 via
The scanning speed V _W of the wafer stage 48 is set via the stage control system 46. In this case, assuming that the projection magnification of the projection optical system PL from the reticle R to the wafer W is β, the scanning speed V _R of the reticle stage 42 is −V _W / β.

Further, the setting unit 73 determines the actual exposure amount on the wafer W during scanning exposure based on the average value of a plurality of predetermined measurement values of the illuminance detection signal S1 sampled at high speed, for example. The value of the target illuminance signal S2 is corrected so that the calculated exposure amount becomes the target exposure amount. The mode in which the light emission power of the mercury lamp 1 is controlled so that the illuminance on the wafer W is constant based on the detection result of the integrator sensor 33 is called the constant illuminance control mode. In addition to this, there is a constant power control mode in which the power applied to the mercury lamp 1 is fixed to a constant value, but this constant power control mode is rarely used during actual scanning exposure.

In FIG. 7, a power supply system 22 is composed of a variable gain amplifier 77 and a power amplifier 78, and a target illuminance signal S2 and a preamplifier 71 are respectively provided in a non-inverting input section and an inverting input section of the variable gain amplifier 77. Is supplied with the illuminance detection signal S1. Variable gain amplifier 77
Then, the light source drive signal S3 is generated with the first gain or the second gain so that the illuminance detection signal S1 becomes the target illuminance signal S2. The first gain is set smaller than the second gain, and usually the smaller first gain is selected. When the arc fluctuation is detected as described later, the gain of the variable gain amplifier 77 is switched to the larger second gain. When the gain is increased in this way, a hunting phenomenon in which the light source drive signal S3 vibrates at a high frequency is likely to occur, but in the present example, hunting is actively caused to suppress the influence of the arc fluctuation. Light source drive signal S from variable gain amplifier 77
3 is amplified by the power amplifier 78, and the power amplifier 78
The mercury lamp 1 is turned on by the power (voltage) output from the mercury lamp 1. A photoelectric conversion signal corresponding to the light emission power of the mercury lamp 1 is output from the integrator sensor 33.

Further, the fluctuation detection system 25 of this example is composed of a subtractor 74, a window comparator 75, and a gain switching section 76. The light source drive signal S3 from the power supply system 22 is supplied to the subtraction side input section of the subtractor 74, and the illuminance detection signal S1 from the preamplifier 71 is also supplied to the addition side input section of the subtractor 74. Then, the difference signal (S1-S3) from the subtractor 74 is supplied to the window comparator 75, and the window comparator 75
Then, when the value of the difference signal (S1-S3) is between the predetermined negative value -SR1 and the positive value + SR1, the low level is "0", and the value of the difference signal (S1-S3) is-. An arc fluctuation detection signal S4 that becomes a high level "1" when it deviates from between SR1 and + SR1 is generated, and this arc fluctuation detection signal S4 is supplied to the gain switching unit 76. The gain switching unit 76 generates a gain switching signal S5 that remains at the high level “1” for a predetermined period when the arc fluctuation detection signal S4 becomes at the high level “1”, and outputs this gain switching signal S5 to the power system 22. Of the variable gain amplifier 77.

The variable gain amplifier 77 has a smaller first gain when the gain switching signal S5 is at a low level "0" and a larger first gain when the gain switching signal S5 is at a high level "1". It becomes a gain of 2.
That is, in this example, the light source drive signal S3 and the illuminance detection signal S1
When the absolute value of the difference between and becomes equal to or more than a predetermined value SR1, the arc fluctuation detection signal S4 is recognized as the occurrence of arc fluctuation and becomes a high level “1”, and the gain in the variable gain amplifier 77 is maintained for a predetermined period thereafter. The higher second gain is set.

Next, an example of the exposure amount control operation in the case of performing exposure by the scanning exposure method in this example will be described with reference to FIGS. First, in FIG. 1, the main control system 19 supplies information on the target integrated exposure amount on the wafer W to the exposure amount control system 20, and the exposure amount control system 20 responds thereto by the mercury lamp 1.
The light emission power in, the transmittance of the light attenuating plate 23, and the like are set. Then, the target illuminance signal S2 (see FIG. 7) for causing the mercury lamp 1 to emit light with the light emission power is supplied to the power supply system 22. After that, the exposure amount control system 20 opens the shutter 4 via the shutter control mechanism 5. However,
In this state, since the movable blinds 35A and 35B are closed, the wafer W is not exposed.

Thereafter, the main control system 19 starts scanning the reticle R and the wafer W via the stage control system 46,
After the scanning speeds of the reticle R and the wafer W reach the predetermined target scanning speeds, respectively, the movable blinds 35A, 35B are moved.
Is controlled to sequentially expose the pattern image of the reticle R onto the shot area of the exposure target on the wafer W. FIG.
At this time, when the mercury lamp 1 is driven with constant power,
That is, an example of changes in the illuminance detection signal S1 (see FIG. 7) obtained by amplifying the detection signal from the integrator sensor 33 when the variable gain amplifier 77 of FIG. 7 is not operated is shown. The illuminance detection signal S1 is represented by setting the level of the illuminance signal S2 (see FIG. 7) to 0. In this state, the illuminance detection signal S1 gradually drifts from the target value. In order to avoid that, the variable gain amplifier 77 is operated.

In this case, first, in the exposure amount control system 20, the illuminance on the wafer W is sequentially calculated from the average value of a plurality of signals obtained by sampling the illuminance detection signal S1 at successively higher frequencies, and the illuminance is calculated. Is out of the target illuminance, the level of the target illuminance signal S2 is corrected. As a result, the exposure is performed in the constant illuminance control mode. Further, in the initial state, it is assumed that the gain switching signal S5 is at the low level "0" while the fluctuation detection system 25 of FIG. 7 is not operating. At this time, the difference (S2-S1) between the target illuminance signal S2 and the illuminance detection signal S1 in the variable gain amplifier 77 of the power supply system 22 of FIG. 7 is as shown in FIG. 9, and the variable gain amplifier 77 has the difference. The light source drive signal S3 is generated with the smaller first gain so as to be zero. That is,
The light emission power of the mercury lamp 1 is controlled so that the difference becomes zero. Therefore, the difference (S2-S1) in FIG.
It can be regarded as the target value of the light source drive signal S3.

When such a smaller gain is used, so-called hunting in which the output signal oscillates at a high frequency does not occur, but due to the delay in the response speed of the variable gain amplifier 77, the gain shown in FIG. ), The phase and the amplitude of the light source drive signal S3 do not completely follow the difference (S2-S1). Figure 10 (a)
In, the dotted curve 89A represents the target value of the light source drive signal S3 (the same curve as in FIG. 9), and the solid curve 89B represents the actual change of the light source drive signal S3.

Further, in FIG. 10A, the curve 89A
At the upper positions A, B, and C, the arc fluctuation of the mercury lamp 1 occurs, and the difference (S2-S1) greatly deviates from the 0 level. In this case, when the gain of the variable gain amplifier 77 is small, the light source drive signal S3 does not follow the difference (S2-S1) as shown by the solid line 89B. The arc fluctuation is a component that largely fluctuates as compared with other noises, depending on the environment of the space (for example, convection) at the time of arc discharge. Although the frequency of this arc fluctuation is low, the absolute value of the illuminance fluctuation due to the arc fluctuation is large. Further, when the arc fluctuation occurs, the power supplied to the mercury lamp 1 (that is, the light source drive signal S3).
And the illuminance measured by the integrator sensor 33 change, the response speed in the servo system further deteriorates. Further, even when the response performance of the variable gain amplifier 77 with respect to other noises is optimized, the tracking performance with respect to arc fluctuation deteriorates. Therefore, if the variable gain amplifier 77 is operated at a low gain,
As shown in 0 (b), the drift of the DC component of the obtained illuminance detection signal S1 disappears, but the illuminance detection signal S1
Shows a large target illuminance signal S2 at the same positions A ', B', and C'as the arc fluctuation generation position of FIG.
It is out of (= 0).

In the scanning exposure type exposure apparatus, since the wafer W is scanned with respect to the slit-shaped exposure area 47 as described above, fluctuations in the illuminance at high frequencies are averaged during the scanning exposure and the integrated exposure amount. It is not uneven. However, when the error in the actual illuminance from the target illuminance (residual error) fluctuates greatly at low frequencies due to arc fluctuations, even if scanning exposure is performed, the integrated exposure amount is partially on the wafer W. It will be out of the proper range.

In order to avoid this, in this example, the fluctuation detection system 25 of FIG. 7 is operated during scanning exposure, and when arc fluctuation is detected, the gain switching signal S5 is set to a high level "1" so that the power supply system 22 is changed. Gain amplifier 7
The gain of 7 is switched to the higher second gain. When the fluctuation detection system 25 is operated in this way, for example, the arc fluctuation occurrence position A in FIG.
In B and C, the level of the light source drive signal S3 and the level of the illuminance detection signal S1 are greatly separated from each other in FIG. 7, the arc fluctuation detection signal S4 becomes a high level “1”, and the gain switching unit 76 outputs the gain switch. Since the signal S5 is at the high level "1" for a predetermined period, the gain of the variable gain amplifier 77 becomes high, and the oscillation becomes slightly oscillating. The gain in this case is optimized according to the characteristics of the arc fluctuation of the mercury lamp.

When the fluctuation detection system 25 is actually operated with respect to the illuminance detection signal S1 in FIG. 8, the gain switching signal S5 in FIG. 7 changes as shown by the rectangular waveform in FIG. 11 (a). . That is, in the rectangular waveform, it is recognized that the arc fluctuation has occurred in the large value portion 79a, the gain switching signal S5 becomes the high level "1", the gain of the variable gain amplifier 77 increases, and the small value portion 79b is normally operated. Is recognized, the gain switching signal S5 becomes low level "0", and the variable gain amplifier 77
The gain of has become small.

As a result, the illuminance detection signal S1 obtained by amplifying the signal output from the integrator sensor 33.
As shown in FIG. 11 (b), in the region where the arc is recognized to occur, it changes like oscillation. However, in the scanning exposure method, variations in illuminance at high frequencies are averaged, so that the unevenness of the integrated exposure amount on the wafer W is reduced, and the integrated exposure amount is within the proper range.

As described above, according to this example, since the gain in the variable gain amplifier 77 is increased when the arc fluctuation occurs, the illuminance fluctuates at a high frequency, and as a result, the unevenness of the integrated exposure amount after scanning exposure is reduced. There is. In the above embodiment, the light source drive signal S3 is used to detect the arc fluctuation.
Is compared with the illuminance detection signal S1, but other than that, for example, the illuminance detection signal S1 in FIG.
A signal S1 'as shown in A may be generated. And
It is checked whether or not the value of the signal S1 ′ has a gradient of a predetermined gradient or more and drifts in the + direction or − direction for a predetermined time or longer, and even if it is recognized that the arc fluctuation occurs when such a drift occurs. Good. The inclination of the signal S1 'is determined, for example, from the differential signal of the signal S1'. in this case,
The corresponding change ΔS3 of the light source drive signal S3 follows like a dotted curve 80B.

It is also possible to take out a low frequency component from the illuminance detection signal S1 via a low pass filter circuit and recognize that arc fluctuation occurs when the amplitude of this low frequency component exceeds a predetermined threshold value. Furthermore, the occurrence of arc fluctuation may be predicted by checking the temperature change of the discharge tube of the mercury lamp 1, and the gain of the variable gain amplifier 77 may be changed based on this prediction. As a method of coping with the arc fluctuation, other than the method of controlling the gain of the variable gain amplifier 77, another method such as adding a predetermined offset to the target illuminance signal S2 may be used.

Further, although the mercury lamp is used as the light source for exposure in the above-mentioned embodiments, the present invention is also applied to the case where a discharge lamp such as a xenon lamp is used as the light source for exposure. Further, even when a laser light source such as an excimer laser light source is used as the exposure light source, the present invention can be applied when a phenomenon occurs in which the correspondence relationship between the supplied power and the actual illuminance changes. As described above, the present invention is not limited to the above-described embodiments, and various configurations can be taken without departing from the gist of the present invention.

[0073]

According to the present invention, when the fluctuation of the illuminance is detected, the output of the exposure light source is primarily changed in, for example, a sine wave shape. Therefore, when using a discharge lamp as a light source for exposure, even if fluctuations in illuminance due to arc fluctuations occur, the average value of illuminance is maintained near the target illuminance. By averaging, the unevenness of the integrated exposure amount at each point on the photosensitive substrate is reduced, and there is an advantage that the integrated exposure amount can be easily kept within an appropriate range.

Further, the illuminance fluctuation detecting means compares the photoelectric detecting means for detecting the light quantity of the illuminating light and the difference between the output signal of the photoelectric detecting means and the signal corresponding to the electric power supplied to the exposure light source. And means for monitoring the difference, it is possible to accurately detect fluctuations in illuminance such as arc fluctuations. In addition, the illuminance fluctuation detection means
When it has photoelectric detection means for detecting the amount of illumination light and low-pass filter means for extracting low-frequency components from the output signal of the photoelectric detection means, by monitoring the output of the filter means, arc fluctuation It is possible to accurately detect fluctuations in illuminance.

Further, the illuminance control means has a gain varying means for controlling the output of the exposure light source by amplifying the variation amount of the output signal from the exposure amount measuring means with a predetermined gain, and the gain in the gain varying means. Is controlled according to the fluctuation of the illuminance detected by the illuminance fluctuation detecting means, the output of the exposure light source becomes vibrating by increasing the gain when the fluctuation of the illuminance occurs.

The present invention is particularly effective when the exposure light source is a discharge lamp and the fluctuation of the illuminance of the illumination light is an arc fluctuation caused by the fluctuation of the discharge of the discharge lamp. Even when the light source is a laser light source or the like, when the illuminance fluctuates, the present invention reduces unevenness in the integrated exposure amount after scanning exposure.

[Brief description of drawings]

FIG. 1 is a partially cutaway view showing a scanning exposure type projection exposure apparatus according to an embodiment of the present invention.

FIG. 2 is a diagram showing an example of a light amount diaphragm 10 used in an embodiment.

3A is an enlarged side view showing a mosaic type fly-eye lens (second fly-eye lens) 14 of FIG.
3B is a front view taken along the line BB of FIG. 3A, and FIG. 3C is a front view taken along the line CC of FIG. 3A.

4 is a view showing a plurality of illumination system aperture stops arranged on the illumination system aperture stop plate 16 of FIG.

5A is a diagram of a main part showing a mechanism for performing focus calibration, and FIG. 5B is a waveform diagram of a detection signal obtained by the mechanism of FIG. 5A.

FIG. 6 is a diagram which is provided for explaining the operation of the movable blinds 35A and 35B when performing scanning exposure in the embodiment.

FIG. 7 is a block diagram showing an exposure amount control mechanism of FIG.

FIG. 8 is a waveform diagram showing an example of an illuminance detection signal S1 obtained via an integrator sensor when the power supplied to the light source is constant.

9 is a waveform diagram showing a difference (S2-S1) between the target illuminance signal S2 and the illuminance detection signal S1 in the case of FIG.

10A shows the difference (S2-S1) and the light source drive signal S3 in FIG. 9 when the fluctuation detection system 25 is not operated.
And (b) is a waveform diagram showing an illuminance detection signal S1 obtained when the variable gain amplifier 77 as a servo system is driven without operating the fluctuation detection system 25.

11A is a diagram showing a gain switching state of the variable gain amplifier 77 when the fluctuation detection system 25 is operated, and FIG. 11B is a diagram showing how the fluctuation detection system 25 is operated to drive the variable gain amplifier 77. It is a waveform diagram which shows an example of the illuminance detection signal S1 at the time of doing.

[Explanation of symbols]

R Reticle PL Projection optical system W Wafer 1 Mercury lamp 4 Shutter 23 Light-reducing plate 9 First fly-eye lens 10 Light quantity diaphragm 12A, 12B Second relay lens 14 Mosaic type fly-eye lens (second fly-eye lens) 16 Illumination system aperture diaphragm Plate 19 Main control system 20 Exposure amount control system 22 Power supply system 25 Fluctuation detection system 33 Integrator sensor 37 Fixed blind 42 Reticle stage 48 Wafer stage

Claims (5)

[Claims] 1. An exposure light source for generating an illumination light for exposure, and an illumination optical system for illuminating a predetermined illumination area on a mask on which a transfer pattern is formed by the illumination light,
The pattern of the mask is sequentially exposed on the photosensitive substrate by scanning the photosensitive substrate in a direction corresponding to the predetermined direction in synchronization with the scanning of the mask in the predetermined direction with respect to the predetermined illumination area. In the scanning exposure type exposure apparatus, the exposure amount measuring means for continuously measuring the exposure energy of the illumination light, the illuminance fluctuation detecting means for detecting fluctuation of the illuminance of the illumination light generated from the exposure light source, An illuminance control means for controlling the output of the exposure light source based on an output signal from the quantity measuring means, wherein the illuminance control means is provided when the illuminance fluctuation is detected by the illuminance fluctuation detecting means An exposure apparatus, wherein the output of an exposure light source is changed at a frequency higher than a predetermined frequency. 2. The exposure apparatus according to claim 1, wherein the illuminance fluctuation detection means supplies photoelectric detection means for detecting a light amount of the illumination light, an output signal of the photoelectric detection means and the exposure light source. And a comparison unit that detects a difference from a signal corresponding to the generated electric power. 3. The exposure apparatus according to claim 1, wherein the illuminance fluctuation detection unit extracts a low frequency component from a photoelectric detection unit that detects the light amount of the illumination light and an output signal of the photoelectric detection unit. A low-pass filter means, and an exposure apparatus. 4. The exposure apparatus according to claim 1, 2, or 3, wherein the illuminance control unit amplifies a change amount of an output signal from the exposure amount measurement unit with a predetermined gain. An exposure apparatus comprising: a gain varying means for controlling an output of a light source, wherein a gain in the gain varying means is controlled according to a fluctuation of illuminance detected by the illuminance fluctuation detecting means. 5. The exposure apparatus according to claim 1, 2, 3, or 4, wherein the light source for exposure is a discharge lamp, and fluctuations in illuminance of the illumination light are caused by fluctuations in discharge of the discharge lamp. An exposure apparatus characterized by the above.