JPH0875415A - Aligning device - Google Patents

Abstract

PURPOSE: To measure the positional relation between a reticle and a reticle stage and between the reticle and a wafer stage with high accuracy when the exposed light is pulse light. CONSTITUTION: The continuous illumination light AL from a halogen lamp 42 is cast on the reticle mark 10A of a reticle 1 via a light guide 44A, a field stop 46A, the second objective lens 48A, and the first objective lens 51A, and the reflected light from the reticle mark 10A forms a reticle mark image on a photoelectric microscope 22A via the first objective lens 51A and the second objective lens 48A. The same pulse illumination light as the exposed light from a wafer stage forms the images of the reference mark and reticle mark on a two-dimensional image pickup element 21A via the first objective lens 51A, a dichroic mirror 50A, and the second objective lens 54A.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention is suitable for application to an apparatus for aligning a reticle, which is provided in a projection exposure apparatus used for manufacturing a semiconductor element, a liquid crystal display element or the like in a lithographic process. Automatic alignment device.

[0002]

2. Description of the Related Art When manufacturing a semiconductor device or a liquid crystal display device, a reticle (or photomask, etc.) is illuminated with exposure light, and the pattern of the reticle is coated with a photoresist through a projection optical system. A projection exposure apparatus that projects an image onto (or a glass plate or the like) is used.
Conventionally, as the exposure light, the bright line (g line,
Light that is continuously emitted (e.g. i-line) has been mainly used. In such a projection exposure apparatus, it is necessary to accurately align the reticle and the wafer during exposure.
For that purpose, first, the reticle is aligned with the reticle stage by using the measurement result of the reticle alignment system.

Further, after completion of the reticle alignment, the position of the reticle is measured with the wafer stage on which the wafer is placed as a reference, and based on the measurement result, for example, the alignment on the wafer side is performed. In order to measure the positional relationship of the reticle with respect to the wafer stage, a so-called stage emission type reticle position detection system may be used.

FIG. 6 shows a main part of a projection exposure apparatus provided with a conventional stage light emitting type reticle alignment system and reticle position detection system. In FIG. 6, the reticle 1 is held on a reticle stage 2 and continuously. The reticle 1 is exposed under exposure light (for example, i-line and g-line) (not shown)
Pattern is projected and exposed on each shot area on the wafer 4 through the projection optical system 3. In this case, the projection optical system 3
The Z axis is taken in parallel with the optical axis AX of the above, and the X axis and the Y axis are taken in the directions parallel to and perpendicular to the paper surface of FIG. 6 in the plane perpendicular to the optical axis AX. Further, cross-shaped reticle marks 10A and 10B are formed at both ends of the pattern area of the reticle 1 in the X direction, and the reticle alignment system 1 is provided above the reticle marks 10A and 10B, respectively.
5A and 15B are arranged.

Further, the wafer 4 is placed on the wafer stage 6 via the wafer holder 5, and the wafer stage 6 moves in the X direction.
It is composed of an XY stage for positioning the wafer 4 in the Y plane, a Z stage for positioning the wafer 4 in the Z direction, a leveling stage for correcting the tilt angle of the wafer 4, and the like. The X coordinate and the Y coordinate of the wafer stage 6 are always measured by a laser interferometer (not shown). Then, the wafer holder 5 on the wafer stage 6
A light-transmissive stage substrate 7 is attached in the vicinity of, and in the light-shielding film on the upper surface of the stage substrate 7, as shown in FIG.
For example, a pair of reference marks 8A and 9A each having a rectangular opening are provided in two centrally symmetric regions along the X axis, and
A pair of reference marks 8B and 9B are formed.

Returning to FIG. 6, at the time of reticle alignment, the reference marks 8A, 9A and 8B, 9B are illuminated from the bottom side by the stage emission type illumination system, and the reference marks 8A and 9B are illuminated.
Illumination light that has passed through A and 8B is transmitted through the projection optical system 3 to the reticle marks 10A and 1A on the lower surface of the reticle 1, respectively.
Illuminate 0B. Further, depending on the specifications of the projection area of the projection optical system used, for example, inner reticle marks (which are referred to as reticle marks 11A and 11B) may be used. Therefore, the illumination light that has passed through the reference marks 9A and 9B illuminates the reticle marks 11A and 11B via the projection optical system 3, respectively.
That is, with the center of the stage substrate 7 aligned with the optical axis AX, the reference marks 8A and 8B are reticle marks 10A and 1A.
The reference marks 9A and 9B are substantially conjugate with 0B, and the reference marks 9A and 9B are substantially conjugate with the reticle marks 11A and 11B.

In the stage emission type illumination system, the illumination light continuously emitted in the same wavelength band as the exposure light guided by the light guide from the exposure light source (not shown) is emitted from the exit ends 12a and 12b of the light guide. It is injected into the wafer stage 6. The illumination light emitted from the emission end 12a is reflected vertically upward by the mirror 13A and then illuminates the reference marks 8A and 9A via the condenser lens 14A. Similarly, the illumination light emitted from the emission end 12b is reflected by the mirror 13
Reference mark 8B through B and condenser lens 14B
And 9B. At this time, when the center of the stage substrate 7 is aligned with the optical axis AX, since the illumination light has the same wavelength band as the exposure light, the images of the reference marks 8A, 8B, etc. respectively surround the reticle marks 10A, 10B, etc. Imaged,
Illumination of the reticle mark is performed by these images. The stage emission type illumination system is configured as described above.

FIG. 7 is a plan view of the reticle alignment system in FIG. 6, and in the reticle alignment system 15A on the right side of FIG. 7, the reticle mark 1 on the reticle 1 is shown.
The illumination light transmitted around 0A (or 11A) is reflected by the mirror 16A for optical path deflection, and then the first objective lens 1
The light enters the half mirror 20A through the 7A, the mirror 18A, and the second objective lens 19A. Then, the light reflected by the half mirror 20A is composed of a two-dimensional CCD or the like.
A composite image of the reference mark 8A and the reticle mark 10A is formed on the image pickup surface of the three-dimensional image pickup device 21A. From the two-dimensional image data from the two-dimensional image pickup device 21A, the displacement of the reticle mark 10A with the reference mark 8A as an index mark is detected by visual observation, for example. Further, by performing image processing on the two-dimensional image data, the positional deviation is detected as the positional deviation at the X coordinate and the Y coordinate on the wafer stage.

Further, the light transmitted through the half mirror 20A is oscillated slit 23A in the two-dimensional photoelectric microscope 22A.
An image of the reticle mark 10A that is transmitted and illuminated by illumination light that forms the image of the reference mark 8A as an illumination field (illumination field) is formed on the upper side. The Y-axis photoelectric microscope of the two-dimensional photoelectric microscope 22A includes a vibrating slit 23A that vibrates in a direction corresponding to the Y direction on the image plane of the reticle mark 10A, and a photomultiplier immediately after the vibrating slit 23A. It is composed of a photoelectric detector 24A. By synchronously rectifying the output signal of the photoelectric detector 24A with the drive signal of the vibration slit 23A, a signal corresponding to the amount of positional deviation of the reticle mark 10A in the Y direction with respect to a predetermined reference position fixed with respect to the reticle stage 2. Can be obtained. Further, although not shown in the drawing, the photoelectric microscope 22A for the X-axis includes a vibration slit vibrating in a direction corresponding to the X direction on the image plane of the reticle mark 10A and a photoelectric detector immediately after the vibration slit. The X-axis photoelectric microscope can detect the amount of positional deviation of the reticle mark 10A in the X direction from a predetermined reference position.

The reticle alignment system 15B on the left side also
The mirror 16B and the two-dimensional photoelectric microscope 22B are configured symmetrically with the right alignment system. The two-dimensional image pickup device 21B of the reticle alignment system 15B can detect the positional deviation amount of the reticle mark 10B by using the other reference mark 8B as an index mark, and the photoelectric microscope 22B can detect the reticle mark from a predetermined reference position. The amount of displacement of 10B in the X and Y directions can be detected.

As a result, the positional relationship of the reticle 1 with respect to the wafer stage 6 (two-dimensional positional deviation amount, by the processing of the imaging signals of the two two-dimensional imaging devices 21A and 21B,
And rotation angle). Furthermore, two photoelectric microscopes 2
Two reticle marks 10A, 1 by 2A and 22B
Since the two-dimensional displacement amount of 0B can be detected, the displacement amount of the reticle 1 with respect to the reticle stage 2 in the X and Y directions and the rotation angle can be obtained. The reticle alignment is completed by setting the positional deviation amounts in the X and Y directions and the rotation angle within the allowable ranges.

Next, for example, the reticle mark 1 inside the reticle 1 is selected according to the projection area of the projection optical system 3 to be used.
When using 1A and 11B, the first objective lens 17
A and the mirror 16A are integrally moved in the -X direction, and the first objective lens 17B and the mirror 16B are integrally moved in the + X direction to move the mirrors 16A and 16B.
Should be positioned above the reticle marks 11A and 11B, respectively.

At this time, as disclosed in Japanese Patent Application Laid-Open No. 57-142612, the lower surface (pattern surface) of the reticle 1 is focused on the first objective lenses 17A and 17B via the mirrors 16A and 16B, respectively. The mirrors 16A, 16B and the first objective lenses 17A, 1
Even if 7B and 7B move in the X direction along the reticle 1, the lower surface of the reticle 1 always matches the focal planes of the first objective lenses 17A and 17B. According to such a configuration, the first objective lens 17A (or 17B) and the second objective lens 19A
(Or 19B) is always a parallel system, and the conjugate relationship and magnification between the pattern surface of the reticle 1, the two-dimensional imaging device, and the vibration slit of the two-dimensional photoelectric microscope are maintained even by the movement. Therefore, the reticle marks 11A and 11A
Even when B and the reference marks 9A and 9B are used, this can be dealt with by the reticle alignment systems 15A and 15B.

[0014]

As described above, in the conventional reticle alignment system, the reference mark and the reticle mark are generated by the stage emission type illumination system using the illumination light continuously emitted in the same wavelength band as the exposure light. Since it was illuminated, the two-dimensional image pickup devices 21A and 21B, and the photoelectric microscope 2
The position was accurately detected in both 2A and 22B.

On the other hand, recently, in order to cope with further improvement in the degree of integration of semiconductor elements and the like, a projection exposure apparatus in which the exposure light has a shorter wavelength to improve the resolution has also been used. . As the exposure light having a short wavelength such as the far ultraviolet region at present, for example, KrF excimer laser light (wavelength 248n
m) or ArF excimer laser light (wavelength 193n
Excimer laser light such as m), metal vapor laser light, or pulsed light such as a harmonic of a YAG laser is mainly used.

However, when the reference mark and the reticle mark are illuminated by the stage emission method by using the pulsed illumination light branched from the pulsed exposure light, the detection signals are oscillated in the photoelectric microscopes 22A and 22B. When the synchronous rectification is performed with the drive signal of No. 2, a high-quality signal cannot be obtained, and the reticle alignment by the photoelectric microscopes 22A and 22B cannot be performed accurately.

In view of the above point, the present invention considers the positional relationship between the reticle (mask) and the reticle stage and the positional relationship between the reticle and the wafer stage even when using pulsed light as the exposure light. It is an object of the present invention to provide an alignment device that can measure with high accuracy.

[0018]

As shown in FIGS. 1 and 2, for example, an alignment apparatus according to the present invention includes a mask (1) on which a transfer pattern is formed by exposure light (IL) emitted in pulses. Illumination optical system (31, 32, 3)
4-39), a projection optical system (3) for projecting an image of the pattern of the mask (1) on the photosensitive substrate (4) under the exposure light, and holding the substrate (4). An alignment mark (10) formed on a mask (1) is provided in a projection exposure apparatus having a substrate stage (6) for positioning the substrate on a plane perpendicular to the optical axis (AX) of the projection optical system.
A) and an apparatus for detecting the position of the mask using the reference mark (8A) arranged on the substrate stage (6).

Then, according to the present invention, the first alignment illumination for illuminating the alignment mark (10A) on the mask (1) with continuous illumination light (AL) continuously emitted in a wavelength region different from the exposure light. The alignment mark (10A) and the reference mark (8A) on the substrate stage (6) are illuminated with the system (42A to 52A) and pulse illumination light (BL) obtained by branching the pulsed exposure light. Second
Alignment illumination system (33, 40, 12, 13A,
14A), continuous illumination light (AL) from the alignment mark (10A), and the alignment mark (10A)
Illumination light (BL) and reference mark (8A)
Illumination light (BL) from the projection optical system (3)
And a first objective optical system (52A, 51A) for condensing the light beam and a continuous illumination light (A
The image of the alignment mark (10A) is formed by the wavelength selection optical system (50A) that divides the light beam by L) and the light beam by the pulsed illumination light (BL) and the light beam by the continuous illumination light that is split by this wavelength selection optical system. The second objective optical system (48A) to be formed and the photoelectric detecting means (23A, 24A) are provided, and the image of the alignment mark by the continuous illumination light and the photoelectric detecting means are vibrated relatively to each other, and An image position detecting means (22A) for detecting the position of the image.

Furthermore, the present invention provides a wavelength selection optical system (50
Alignment mark (10A) and reference mark (8A) from the luminous flux of the pulsed illumination light divided by A)
A third objective optical system (53A, 54A) for forming an image of
Alignment mark (10A) with the pulsed illumination light
And an image pickup means (21) for picking up an image of the reference mark (8A).
A), and detecting the positional relationship of the mask (1) with respect to the image position detecting means (22A) based on the detection result of the image position detecting means (22A), and performing alignment based on the detection result of the imaging means (21A). Mark (10A) and reference mark (8
The amount of positional deviation from (A) is detected.

In this case, the first objective optical system (52A, 51
A) is movably arranged corresponding to the position of the alignment mark on the mask (1), and the wavelength selection optical system (50A)
And the second objective optical system (48A), or between the wavelength selection optical system (50A) and the third objective optical system (53A, 54A), a correction optical system (49A) is arranged. It is desirable to convert the light flux from the first objective optical system into a parallel light flux in cooperation with the first objective optical system.

Further, the first alignment illumination system includes an alignment mark (10) from above the mask (1).
(A) is epi-illuminated, the continuous illumination light transmitted through the mask (1) is removably inserted into and removed from the alignment mark (10) between the mask (1) and the projection optical system (3).
It is desirable to arrange a movable mirror (56A) that reflects on the A) side.

At this time, it is desirable that the field stop (46A) is arranged on the plane conjugate with the arrangement plane of the movable mirror (56A) in the first alignment illumination system.

[0024]

According to the present invention, the alignment mark (10A) on the mask (1) is illuminated by the second alignment illumination system using the pulse illumination light obtained by branching the exposure light.
And the reference mark (8A) on the substrate stage side is illuminated,
Since the images of the two marks are picked up by the image pickup means (21A), the positional relationship of the mask (1) with respect to the substrate stage (6) is measured based on the images obtained thereby. At this time, since the pulsed illumination light has the same wavelength as the exposure light, there is no chromatic aberration in the projection optical system (3), and the two marks are clearly imaged on the image pickup surface of the image pickup means (21A).

Further, the position of the image of the alignment mark is detected by relatively vibrating the image of the alignment mark (10A) and the photoelectric detecting device (23A, 24A), for example, an image position detecting device such as a photoelectric microscope ( 22A), since continuous illumination light is used, the position detection of the mask (1) with respect to the image position detection means is performed with high accuracy. Then the first
When the objective optical system (52A, 51A) is movably arranged corresponding to the position of the alignment mark on the mask (1), the alignment mark (11A) at a different position.
Is to be detected, the first objective optical system may be moved along the mask (1) accordingly. However, the exposure light that is pulsed is generally in the far-ultraviolet region, and since usable glass materials are limited, the light flux from the first objective optical system is commonly used for the pulse illumination light and the continuous illumination light. Is difficult. Therefore, for example, when the pulsed illumination light side is set to a parallel system, the wavelength selection optical system (50
A correction optical system (49A) is arranged between A) and the second objective optical system (48A), and this correction optical system works in conjunction with the first objective optical system to collimate the light flux from the first objective optical system. Convert to light flux. Thereby, even if the first objective optical system moves on the mask (1), the position can be accurately detected.

Further, the first alignment illumination system includes an alignment mark (10) from above the mask (1).
If the movable mirror (56A) is inserted between the alignment mark (10A) on the mask (1) and the projection optical system (3) in the case of epi-illuminating the movable mirror (56A), Alignment mark (1
0A) is transilluminated. Therefore, even if the reflectance of the alignment mark is low, the position can be accurately detected.

At this time, when the field stop (46A) is arranged on the plane conjugate with the arrangement plane of the movable mirror (56A) in the first alignment illumination system, whether or not there is a movable mirror. Alignment mark (10A)
The illuminance distribution at is similar.

[0028]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the alignment apparatus according to the present invention will be described below with reference to FIGS. In this embodiment, the present invention is applied to a reticle alignment system of a projection exposure apparatus that uses far-ultraviolet light that is pulsed as exposure light. Further, in FIGS. 1 and 2, parts corresponding to those in FIGS. 6 to 8 are designated by the same reference numerals, and detailed description thereof will be omitted.

FIG. 2 shows the projection exposure apparatus of this embodiment,
In FIG. 2, from the pulse light source 31 formed of, for example, a KrF excimer laser light source, the far ultraviolet region (for example, wavelength 24
Exposure light of 8.4 nm) is emitted in pulses. The exposure light emitted in a pulsed manner (pulse exposure light) has its cross-sectional shape enlarged by the beam expander 32 and enters the fly-eye lens 34. However, in this embodiment, a mirror 33 for switching the optical path is provided between the beam expander 32 and the fly-eye lens 34 in a state of being inclined by 45 ° with respect to the optical axis. Although retracted to the position 33A, the mirror 33 is set in the optical path of the exposure light during reticle alignment.

At the time of exposure, the pulse exposure light IL from the secondary light source formed on the focal plane on the rear side (reticle side) of the fly-eye lens 34 is reflected by the mirror 35, and then the first relay lens 36 and the variable field of view. Aperture (reticle blind) 3
The pattern area on the lower surface (pattern surface) of the reticle 1 is illuminated with a uniform illuminance distribution via the 7, the second relay lens 38, and the main condenser lens 39. Under the pulse exposure light IL, the image of the pattern of the reticle 1 is projected and exposed on each shot area of the wafer 4 on the wafer stage 6 via the projection optical system 3.

On the other hand, during reticle alignment, the mirror 33 is set in the optical path of the pulse exposure light, and the mirror 33
Light reflected by the light (hereinafter referred to as "pulse illumination light")
Are supplied to the incident end 12c of the light guide 12 via the beam expander 40 that reduces the beam diameter. Two emission ends 12a and 12b of the light guide 12 are inserted into the wafer stage 6, and the pulsed illumination light BL emitted from the emission ends 12a and 12b is reflected by the mirrors 13A and 13A.
The reference marks 8A, 9A and 8B, 9B, which are rectangular apertures formed on the upper surface of the stage substrate 7, are illuminated via B and the condenser lenses 14A, 14B, respectively. When the center of the stage substrate 7 is substantially aligned with the optical axis AX of the projection optical system 3, the reference marks 8A, 9A and 8B, 9 are provided.
The pulsed illumination light BL that has passed through B passes through the projection optical system 3 and the reticle marks 10 A, 1 on the pattern surface of the reticle 1.
An image of each fiducial mark is formed in the vicinity of 1A, 10B and 11B. In FIG. 2, reticle alignment systems 41A and 41B are arranged on the reticle marks 10A and 10B, respectively.

FIG. 1 is a plan view showing the reticle alignment system of the present embodiment. In the reticle alignment system 41A on the right side of FIG. 1, the reticle alignment system 41A shown in the right side of FIG. Continuous light in a wavelength range (for example, red or near-infrared light) different from the pulse exposure light IL due to the filter (hereinafter, referred to as “continuous illumination light”)
AL is selected. A light emitting diode or the like can be used instead of the halogen lamp 42A. The continuous illumination light AL, as shown by the dotted line, is a condenser lens 43A,
The light is projected onto the aperture of the field stop 46A via the light guide 44A and the condenser lens 46A.

The continuous illumination light AL that has passed through the aperture of the field stop 46A is reflected by the half mirror 47A, and then enters the dichroic mirror 50A through the second objective lens 48A and the correction lens 49A. Dichroic mirror 50
A reflects the pulsed illumination light BL to produce continuous illumination light AL
Has wavelength selectivity for transmitting light. Therefore, the continuous illumination light AL passes through the dichroic mirror 50A and then illuminates the reticle mark 10A via the first objective lens 51A and the mirror 52A.

In FIG. 2, movable mirrors 56A and 56B are arranged on the bottom surface side of the reticle stage 2, and the movable mirrors 56A and 56B are supported so as to be movable along the lower surface of the reticle 1. Reticle mark 10A and 1
When the reflectance of 0B is low, the reticle mark 1
Movable mirrors 56A and 56B are installed below 0A and 10B, respectively. At this time, the movable mirror 56A is installed on a surface conjugate with the arrangement surface of the field stop 46A in FIG. The reason for this will be described later in detail.

The continuous illumination light AL reflected by the reticle mark 10A and the rear reticle mark 10 reflected by the movable mirror 56A when the movable mirror 56A is installed.
The continuous illumination light AL that has passed the periphery of A and the pulsed illumination light BL that has passed the periphery of the reticle mark 10A are directed to the dichroic mirror 50A via the mirror 52A and the first objective lens 51A. Returning to FIG. 2, pulsed illumination light B
After being reflected by the dichroic mirror 50A, L is a two-dimensional C through the mirror 53A and the second objective lens 54A.
Images of the reference mark 8A and the reticle mark 10A are formed on the image pickup surface of the two-dimensional image pickup device 21A such as a CD.
On the other hand, the continuous illumination light AL is emitted from the dichroic mirror 50A.
After passing through, the correction lens 49A and the second objective lens 48
Two-dimensional photoelectric microscope 2 through A and half mirror 47A
An image of the reticle mark 10A is formed on the vibration slit 23A of 2A. Although not shown, the photoelectric microscope 22A also includes a vibration slit whose vibration direction is orthogonal to that of the vibration slit 23A, and a photoelectric detector immediately after that.

By processing the image pickup signal of the two-dimensional image pickup device 21A, the positional deviation amount of the reticle mark 10A with respect to the reference mark 8A in the X and Y directions is measured. Further, the two-dimensional photoelectric microscope 22A measures the amount of positional deviation of the reticle mark 10A with respect to the reference point (reticle stage 2) in the photoelectric microscope 22A in the X and Y directions. Further, the correction lens 49A to the mirror 53A
The movable optical system 55A is arranged so as to be movable in the X direction. For example, when detecting the position of the reticle mark 11A, the movable optical system 55A moves so that the mirror 52A is located above the reticle mark 11A.

At this time, the pulse illumination light BL emits a pulse in the far ultraviolet region (for example, a wavelength of 248.8 nm) KrF.
If the exposure light from the excimer laser light source is a branched light beam, the glass material that can be used for the pulsed illumination light BL is limited. Therefore, pulsed illumination light B
When the lower surface of the reticle 1 is aligned with the focal plane of the first objective lens 51A with respect to L, the first objective lens 51 below the reticle 1 for continuous illumination light AL having a wavelength longer than that.
The focus of A comes to be in focus. In this state, the first objective lens 51A and the second
Although it is a parallel system with the objective lens 54A, the first objective lens 51 does not receive the continuous illumination light AL which is the non-exposure light.
A divergent system is formed between A and the second objective lens 48A.

In order to avoid this, a correction lens 49A for correcting the focus position of the continuous illumination light AL of the first objective lens 51A on the lower surface of the reticle 1 is arranged on the optical path of the continuous illumination light AL transmitted through the dichroic mirror 50A. As a result, the first objective lens 51A and the correction lens 49A move integrally. Therefore, the correction lens 4
9A and the second objective lens 48A form a parallel system, and even if the movable optical system 55A moves in the X direction, as shown by the dotted line in FIG. 1, the lower surface of the reticle 1 and the vibration slit of the photoelectric microscope 22A are separated from each other. The conjugate relationship between them and the imaging magnification are maintained in the initial state, and accurate position measurement is performed even for reticle marks 11A having different positions.

The reticle alignment system 41 on the left side
B is also symmetrical to the alignment system on the right side and includes the halogen lamp 42B to the second objective lens 54A and the two-dimensional image pickup device 21.
B and a photoelectric microscope 22B. With the reticle alignment system 41B, the positional relationship between the reticle marks 10B and 11B with respect to the reticle stage 2,
And the reticle mark 10 for the reference marks 8B and 9B.
The positional relationship between B and 11B is measured.

Next, the operation of the movable mirror 56A shown in FIG. 2 will be described. In this regard, when epi-illumination is used for the continuous illumination light AL as in this embodiment, the reflectance of the reticle marks 10A and 11A always becomes a problem, but in this embodiment, the following is performed. Then, the problem caused by the reflectance was solved. Normal reticle mark 10A,
As 11A, a pattern obtained by etching a chrome film formed on a glass substrate and leaving the chrome film in the shape of a cross mark, for example, is used. In particular, a pattern having a normal reflectance for light incident from the upper surface (reflectance of 40% is used). There is a lower pattern compared to (about%).

For example, when the circuit pattern in the pattern area of the reticle 1 has such low reflection, the reflectance of only the reticle mark cannot be changed, so that the reticle mark also necessarily has low reflection. Even in such a case, in the conventional alignment system, since only the stage light emission is used, the reticle marks 10A and 11A are transmitted and illuminated, and the detection accuracy does not depend on the reflectance. However, the reticle alignment system 41A of this embodiment is
Since the epi-illumination is performed from above, when the reflectance of the reticle mark is low, the light intensity of the obtained image becomes poor, and in the worst case, the image of the reticle mark cannot be obtained. In such a case, in this embodiment, the movable mirror 56A is moved below the low-reflection reticle mark to reflect the epi-illumination light that has transmitted the reticle mark once, and to transmit the reticle mark more uniformly from below. Can be illuminated.

An illumination state in which a movable mirror 56A parallel to the reticle 1 is inserted below the reticle 1 only when necessary will be described with reference to FIGS. FIG. 3 is an optical path diagram showing a luminous flux of continuous illumination light AL having an illumination field (illumination field) of uniform illuminance distribution on the surface 57C, and a rectangular distribution 5
8C represents the illuminance distribution on the surface 57C centered on the optical axis 61. Further, the illumination area gradually expands as the distance from the surface 57C increases in the direction of the optical axis 61 (the negative direction is defined as) 57B and 57A, and similarly, from the surface 57C downward in the direction of the optical axis 61 (the + direction). (2) As the distance from the surfaces 57D and 57E increases, the illumination visual field gradually expands. Face 57
The illuminance distribution in A to 57E is the distribution on the right side 58A to
58E. From this, it can be seen that the illuminance uniform region (the region where the illuminance distribution is flat) decreases with increasing distance from the surface 57C in the + or − direction. A shaded area 62 indicates a uniform illuminance area.

Let φ _E be the diameter of the uniform illuminance region. Assuming an ideal case where the luminous flux is symmetrical above and below the illumination field on the surface 57C as shown in FIG. 3, and assuming that the half aperture angle θ of the continuous illumination light AL is small, the numerical aperture NA = sin θ≈θ Can be approximated.
When the diameter of the illumination field is φ and the distance from the illumination field in the vertical direction (Z direction) is d, the following relationship is established.

[0044]

## EQU1 ## φ _E = φ-2dθ By utilizing such a property, a low reflection reticle mark is dealt with. FIG. 4 shows a state in which the movable mirror 56A is installed below the reticle mark 10A. In FIG.
A movable mirror 56A is installed below the reticle 1 at a position separated by L in parallel with the reticle 1, and the continuous illumination light AL is provided.
The illumination field of the epi-illumination by is the mirror surface of the movable mirror 56A. Below the movable mirror 56A, the light flux reflected by the movable mirror 56A and the reticle 1M are indicated by a two-dot chain line in a mirror image. Movable mirror 56A for normal reticle 1
However, the illuminance distribution of the reticle mark 10A is like a distribution 60A where the periphery is blunt.

In this case, if the uniform illuminance area in the distribution 60A is larger than the field of view actually observed, it means that uniform illumination is performed. Further, when the low reflection reticle is used, the movable mirror 56A is inserted as shown in FIG. 4, and the reticle 1M having a mirror image is illuminated. That is, the reticle mark 10AM, which is a mirror image of the reticle mark 10A, is transmitted and illuminated.
At this time, the light flux shown by the solid line is the actual reticle mark 1
It is necessary to consider how the illuminance changes due to vignetting due to 0A.

FIG. 5 is a mirror image reticle mark 1 in FIG.
0AM and the reticle mark 10A are enlarged and shown in FIG. 5, where the line width of the reticle mark 10A is Δ, the illuminance on the lower surface of the reticle 1 when there is no vignetting, that is, when Δ = 0, is E ₀ And The actual illuminance E at that time is, as shown in FIG. 4, the distance between the movable mirror 56A and the reticle 1 being L, and the numerical aperture NA by the continuous illumination light AL.
Let sin θ ≈ θ, the following relationship holds.

[0047]

[Number 2] _{E ≒ E 0 {1-2Δ / (} πLθ)} Here, if setting each variable as deemed E ≒ E _0, illuminance unevenness caused by vignetting by the reticle mark 10A is no longer worth consideration Also, even in the transmissive illumination with the reflected light from the movable mirror 56A, the illuminance distribution like the distribution 60A can be obtained. At this time, almost no image of the reticle mark 10A is obtained by epi-illumination, so reticle alignment can be performed by an image of transmitted illumination by reflected light from the movable mirror 56A. Note that the state as considered E ≒ E _0, a state state falls below about several%, i.e. the 2Δ / (πLθ) becomes less than several% of _{E 0 (E-E 0)} .

As described above, in the present embodiment, the conventional reticle alignment can be performed on all the reticles by simply adding a simple mirror moving mechanism.
When using a normal reticle, the movable mirror 5
Since it is not necessary to move 6A and 56B, it is advantageous in terms of throughput. Although not shown, the illumination system of the continuous illumination light AL, which is non-exposure light, is moved to the movable mirrors 56A and 56B.
Alternatively, the reticle 1 may be installed below the reticle 1, and the illumination may be performed from below only when necessary. According to this, although the reticle having any reflectance can be supported, the reticle 1
There is a need for a space between the projection optical system 3 and the projection optical system 3, and the illumination system needs to be taken in and out even when the reticle is normally used.

Further, when the movable mirror is placed below the reticle, the illumination field is placed on the surface of the movable mirror, but the position may be slightly displaced if there is no problem in performance. Further, if there is no problem in performance, the actual visual field at the reticle mark position may slightly extend beyond the constant illuminance area. In addition, the movable mirrors 56A and 56B
Alternatively, a dichroic mirror that reflects the continuous illumination light AL and transmits the pulsed illumination light BL (exposure light) may be used.

As described above, the present invention is not limited to the above-described embodiments, and various configurations can be adopted without departing from the gist of the present invention.

[0051]

According to the present invention, continuous illumination light is used for the image position detecting means (photoelectric microscope or the like) which causes the image of the alignment mark and the photoelectric conversion detecting means to relatively vibrate even if the exposure light is pulsed. Therefore, the position measurement of the mask with respect to the image position detection means is performed with high accuracy. Furthermore, since the same pulsed illumination light as the exposure light is used in the imaging means, the positional relationship of the alignment mark with respect to the reference mark can be detected with high accuracy without being affected by the chromatic aberration of the projection optical system. Therefore, reticle alignment can be accurately performed based on the result.

Further, the first objective optical system is movably arranged corresponding to the position of the alignment mark on the mask, and it is provided between the wavelength selecting optical system and the second objective optical system or between the wavelength selecting optical system and the third objective optical system. A correction optical system is arranged between the objective optical system and
In the case where the correction optical system works in conjunction with the first objective optical system to convert the light beam from the first objective optical system into a parallel light beam,
A parallel system can be configured for the objective optical system and the third objective optical system. Therefore, even if the positions of the alignment marks at different positions on the mask are measured, the measurement can be performed with the same detection accuracy.

The first alignment illumination system illuminates the alignment mark from above the mask by epi-illumination, and the continuous illumination light transmitted through the mask is removably inserted between the mask and the projection optical system. When the movable mirror that reflects light is disposed on the alignment mark side, accurate position detection can be performed even when the reflectance of the alignment mark is low.

Furthermore, in the first alignment illumination system,
When the field stop is arranged on a plane conjugate with the arrangement plane of the movable mirror, there is an advantage that the position of the alignment mark can be detected with almost the same illuminance distribution regardless of whether the movable mirror is used or not. is there.

[Brief description of drawings]

FIG. 1 is a plan view showing a reticle alignment system of a projection exposure apparatus to which an embodiment of an alignment apparatus according to the present invention is applied.

FIG. 2 is a partially cutaway block diagram showing a projection exposure apparatus of the embodiment.

FIG. 3 is a diagram for explaining a change in illuminance distribution due to continuous illumination light.

FIG. 4 is an explanatory diagram when the movable mirror 56A illuminates the reticle 1 through transmission.

5 is an enlarged perspective view showing a relationship between a mirror image reticle mark in FIG. 4 and an actual reticle mark.

FIG. 6 is a partially cutaway block diagram showing a main part of a conventional projection exposure apparatus.

FIG. 7 is a plan view showing the reticle alignment system of FIG.

8 is an enlarged plan view showing a reference mark under the stage substrate 7 of FIG.

[Explanation of symbols]

 1 Reticle 3 Projection optical system 4 Wafer 6 Wafer stage 7 Stage substrate 8A, 8B, 9A, 9B Reference mark 10A, 10B, 11A, 11B Reticle mark 12 Light guide 31 Pulse light source 33 Movable mirror 41A, 41B Reticle alignment system 42A, 42B Halogen lamp 46A, 46B Field diaphragm 48A, 54A Second objective lens 49A Correction lens 50A Dichroic mirror 51A First objective lens 52A Mirror 55A, 55B Movable optical system

Claims (4)

[Claims] 1. An illumination optical system for illuminating a mask on which a transfer pattern is formed with exposure light emitted in pulses, and an image of the pattern of the mask is projected on a photosensitive substrate under the exposure light. A projection optical system having a projection optical system for holding the substrate, and a substrate stage for holding the substrate and positioning the substrate on a plane perpendicular to the optical axis of the projection optical system, the alignment being formed on the mask. In an apparatus for detecting the position of the mask using a mark and a reference mark arranged on the substrate stage, an alignment mark on the mask with continuous illumination light continuously emitted in a wavelength range different from the exposure light. A first alignment illumination system for illuminating the alignment mark; and the alignment mark on the substrate stage with pulsed illumination light obtained by branching the pulsed exposure light. A second alignment illumination system for illuminating a reference mark; the continuous illumination light from the alignment mark, the pulse illumination light from the alignment mark, and the pulse from the reference mark via the projection optical system A first objective optical system that condenses the illumination light; a wavelength selection optical system that divides the luminous flux condensed by the first objective optical system into a luminous flux of the continuous illumination light and a luminous flux of the pulsed illumination light; A second objective optical system that forms an image of the alignment mark from a light flux of the continuous illumination light split by a wavelength selection optical system; and a photoelectric detection unit, and an image of the alignment mark and the photoelectric detection by the continuous illumination light. Image position detection means for detecting the position of the image of the alignment mark by vibrating the means relative to each other; division by the wavelength selection optical system A third objective optical system that forms an image of the alignment mark and the reference mark from the light flux of the pulsed illumination light, and an imaging unit that captures the image of the alignment mark and the reference mark by the pulsed illumination light. Then, the positional relationship of the mask with respect to the image position detection means is detected based on the detection result of the image position detection means, and the positional deviation amount between the alignment mark and the reference mark is calculated based on the detection result of the imaging means. An alignment device characterized by detecting. 2. The first objective optical system is movably arranged corresponding to the position of an alignment mark on the mask, and between the wavelength selection optical system and the second objective optical system,
Alternatively, a correction optical system is arranged between the wavelength selection optical system and the third objective optical system, and the correction optical system interlocks with the first objective optical system to convert the light flux from the first objective optical system. The alignment device according to claim 1, wherein the alignment device converts the light into a parallel light beam. 3. The first alignment illumination system illuminates the alignment mark from above the mask by epi-illumination, and the mask is removably inserted between the mask and the projection optical system. The alignment apparatus according to claim 1, further comprising a movable mirror that reflects the transmitted continuous illumination light toward the alignment mark. 4. In the first alignment illumination system,
4. The alignment apparatus according to claim 3, wherein the field stop is arranged on a surface conjugate with the arrangement surface of the movable mirror.