JP3152313B2 - Projection exposure apparatus and pattern transfer method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a projection exposure apparatus having, for example, an alignment system for performing relative positioning between a mask and a substrate and projecting and exposing a pattern formed on the mask onto the substrate.

[0002]

2. Description of the Related Art In the manufacture of semiconductor integrated circuits and the like, a projection exposure apparatus for transferring a circuit pattern or the like formed on a mask or a reticle (hereinafter collectively referred to as a "mask") onto a photosensitive substrate is used. In this type of apparatus, a stepper for a semiconductor integrated circuit exposes an image of a circuit pattern formed on a mask to a plurality of transfer areas (shot areas) on a wafer one after another through a projection optical system. The wafer is placed on a wafer stage which can be moved two-dimensionally by step and repeat.
In this case, one shot area on the wafer and the projected image of the circuit pattern need to be two-dimensionally and accurately overlapped with, for example, an accuracy of about ± 0.2 μm or less. for that reason,
The circuit pattern area on the mask and each shot area on the wafer are directly or indirectly aligned by an alignment system.

As an alignment method, that is, an alignment method, various automated methods are adopted in most steppers. Among them, the TTL (through-the-lens) method and the TTR (through-the-reticle or through-the-mask) method are those that can obtain high accuracy. The TTL method is a method using an optical system (alignment system) for detecting an alignment mark on a wafer substantially only through a projection objective lens, and the TTL method is a method using a wafer through both a mask and a projection objective lens. This method uses an optical system (alignment system) for detecting the upper alignment mark and the alignment mark on the mask.

In the TTL method, since only the alignment mark on the wafer is substantially detected, the positional relationship between the detection center of the alignment system and the mask is accurately measured in advance, and the detected position of the wafer mark is determined based on the measured value. Need to be measured. On the other hand, in the TTR method, since the alignment marks on the wafer and the mask are simultaneously or directly detected, the alignment between the mask and the wafer (strictly, each shot area on the wafer) is directly performed.

[0005] Incidentally, semiconductor integrated circuits are becoming finer and finer, and an exposure apparatus for printing the pattern is required to have a higher resolution. To meet this requirement, the numerical aperture (NA) of the projection objective must be increased while shortening the wavelength of the light source. However,
As the exposure wavelength becomes shorter, the practically usable glass material for absorbing light is limited. When the exposure wavelength is shorter than 300 nm, only synthetic quartz and fluorite (calcium fluoride) can be practically used. Fluorite has a poor temperature characteristic and cannot be used in large quantities. Therefore, it is difficult to make a projection objective only with a refraction system. In addition, it is almost impossible to produce a projection objective having a large numerical aperture only by using a reflection system because of the difficulty in correcting aberrations.

For this reason, Japanese Patent Application Laid-Open No. 2-66510 proposes a projection objective comprising a catadioptric optical system. Then, even when the projection objective is composed of a catadioptric system, in order to maintain high alignment accuracy, alignment is performed via a catadioptric projection objective by the TTL system or the TTR system as described above. It is desired.

At the present stage, such an alignment system that performs alignment by the TTL system or the TTR system has not been put into practical use for a projection exposure apparatus using a projection objective lens composed of a catadioptric optical system. Therefore, with reference to FIG. 7, a case will be considered in which a conventional alignment system for a projection exposure apparatus using only a refraction optical system is directly applied to a projection exposure apparatus using a catadioptric optical system. In the present application, this is a conventional example.

FIG. 7 shows a projection exposure apparatus using a catadioptric optical system as a projection objective lens and virtually including a TTR type alignment system. In FIG. 7, reference numeral 1 denotes an illumination system for exposure. The exposure light IL emitted from the illumination system 1 is converted into a substantially parallel light beam by the main condenser lens 2 and enters the dichroic mirror 3. As the exposure light IL, for example, ultraviolet light having a wavelength of 250 nm is used. The dichroic mirror 3 has wavelength selectivity for reflecting the exposure light IL and transmitting the alignment light described below,
The exposure light IL reflected by the dichroic mirror 3 illuminates the mask 4. An alignment mark MA made of an amplitude or phase diffraction grating is formed on the mask 4. Further, the mask 4 has a mask stage 5 capable of parallel movement and minute rotation in a plane perpendicular to the optical axis of the projection objective lens.
Is placed on top.

The exposure light IL transmitted through the mask 4 is converged by the first lens group 6 and travels to the semi-transmission mirror 7, and the exposure light IL transmitted through the semi-transmission mirror 7 is a correction lens having a negative refractive power. The light enters the concave reflecting mirror 9 via the group 8. And
The exposure light IL reflected by the concave reflecting mirror 9 again enters the semi-transmissive mirror 7 via the correction lens group 8, and the semi-transmissive mirror 7
The exposure light IL reflected by the second lens group 10 is focused on the shot area of the wafer 11 on the wafer stage 12. Thereby, the circuit pattern on the mask 4 is transferred onto the wafer 11. In a region between the respective shot regions on the wafer 11 on which a photosensitive material such as a resist is applied, a wafer-side alignment mark WA made of an amplitude or phase diffraction grating is formed. The wafer stage 11 can move in parallel in a plane perpendicular to the optical axis of the projection objective lens.
A wafer, a θ stage capable of minute rotation in the plane, a Z stage movable in the direction of the optical axis, a leveling stage capable of tilting with respect to a plane perpendicular to the optical axis, and the like. And can be positioned freely.

Reference numeral 13 denotes an alignment system as a whole,
In the alignment system 13, a diffraction grating 14 is irradiated with a parallel laser beam LB0 from a laser light source (not shown). As the laser beam LB0, light in a wavelength band where photosensitivity to the photosensitive material on the wafer 11 is weak, for example, a near-infrared laser beam with a wavelength of about 633 nm is used. The laser beam LB
0 is + 1st-order diffracted light LB1 and -1 by the diffraction grating 14.
The light is divided into the next diffracted light LB2, and one of the diffracted lights LB1 further passes through the acousto-optic modulator AOM, so that the frequency is shifted by Δf. At this time, the diffraction grating 14 is of a phase type and the generation of zero-order light is suppressed, or only the ± first-order light is passed by a spatial filter (not shown). Diffracted lights LB1 whose frequencies are different from each other by Δf
And LB2 are a focusing lens 15 and a beam splitter 1
The light is converged once after each through 6. Thereafter, these diffracted lights LB1 and LB2 are converged by the converging lens 17 to become first illumination light LM1 and second illumination light LM2, respectively. These illumination lights LM1 and LM2 pass through the dichroic mirror 3 and irradiate the alignment mark MA on the mask 4 at a predetermined intersection angle.

In this case, the rear focal point of the focusing lens 15 and the front focal point of the focusing lens 17 are at the same point, and the illumination lights LM1 and LM2 incident on the alignment mark MA are parallel laser beams. Further, the reflected light of the first-order diffracted light (this is + 1st order) of the illumination light LM1 by the alignment mark MA and the reflected light of the -1st-order diffracted light by the alignment mark MA of the diffracted light LM2 are parallel to each other. The grating pitch of the alignment mark MA and the intersection angle of both beams are determined. As a result, from the alignment mark MA, reflected light LM0 in which the first-order diffracted light of the illumination light LM1 and the -1st-order diffracted light of the illumination light LM2 are mixed is returned to the dichroic mirror 3 side. Due to the interference between the first-order diffracted light and the -1st-order diffracted light, the intensity of the reflected light LM0 changes sinusoidally at the beat frequency Δf.

On the other hand, the alignment mark MA of the mask 4
Illumination lights LM1 and L1 that have passed through the transmission window near
M2 enters the concave reflecting mirror 9 via the first lens group 6, the semi-transmissive mirror 7, and the correcting lens group 8. The illumination light beams LM1 and LM2 reflected by the concave reflecting mirror 9 travel again to the semi-transmissive mirror 7 via the correction lens group 8, and are reflected by the semi-transmissive mirror 7, and are then given by the second lens group 10. The alignment mark WA on the wafer 11 is irradiated at the intersection angle. Since the pattern region of the mask 4 and the exposure surface of the wafer 11 can be considered to be substantially conjugate to the alignment light, the illumination light LM1 and LM2 on the wafer 11
Are also parallel laser beams. In this case, FIG.
As shown in the figure, the alignment mark MA on the mask 4 and the image of the alignment mark WA on the wafer 11 have the same lattice pitch direction and are vertically displaced from each other. The illumination light beams LM1 and LM2 that have passed through the laser beam irradiate the alignment mark WA on the wafer 11 side.

Then, the reflected light of the first-order diffracted light (let it be + 1st-order) of the illumination light LM1 by the alignment mark WA and the reflected light of the -1st-order diffracted light by the alignment mark WA of the illumination light LM2 become parallel. Thus, the grating pitch of the alignment mark WA and the intersection angle of both beams are determined. Thereby, from the alignment mark WA of the wafer 11, reflected light LW0 in which the first-order diffracted light of the illumination light LM1 and the -1st-order diffracted light of the illumination light LM2 are mixed,
The light is returned to the semi-transmissive mirror 7 via the second lens group 10. Due to the interference between the first-order diffracted light and the -1st-order diffracted light, the intensity of the reflected light LW0 changes sinusoidally at the beat frequency Δf. The reflected light LW0 is reflected by the semi-transmissive mirror 7 and enters the concave reflecting mirror 9 via the correcting lens group 8, and the reflected light LW0 reflected by the concave reflecting mirror 9 is
The light enters the mask 4 via the semi-transmissive mirror 7 and the first lens group 6.

The reflected light LW0 from the wafer 11 transmitted through the mask 4 passes through the dichroic mirror 3 and the converging lens 17 together with the reflected light LM0 from the mask 4 and is transmitted through the semi-transmissive mirror 1
6, the reflected light LW reflected by the semi-transmissive mirror 16
0 and LM0 are aperture plates 19
Is irradiated. The mask 4 reflects specularly reflected light and higher-order diffracted light out of the reflected light LM0, and the wafer 11 also reflects specularly reflected light other than the reflected light LW0.
Reflected light other than M0 and LW0 is blocked by the aperture plate 19. The reflected lights LW0 and LM0 pass through the central opening of the aperture plate 19,
0 irradiates the photoelectric sensor 21. At that time mask 4
And the photoelectric sensor 21 are in a conjugate relationship.

In the photoelectric sensor 21, two light receiving elements 21a and 21b independent of each other are arranged in parallel so as to correspond to the alignment mark MA on the mask 4 and the window 4a, and the light receiving elements 21a and 21b are In FIG. 7, they are arranged in a direction perpendicular to the paper surface, but are drawn in the paper surface direction in FIG. 7 for easy understanding. Then, the reflected lights LW0 and LW0
M0 is independently incident on, for example, the light receiving elements 21a and 21b. The light receiving elements 21a and 21b output detection signals each having a beat frequency Δf. In this case, the phases of the signals obtained from the light receiving elements 21a and 21b are different depending on the positional relationship in the pitch direction of the lattice between the alignment mark MA on the mask 4 side and the alignment mark WA on the wafer 11 side. By positioning the wafer stage 12 so as to set it to a predetermined value, highly accurate positioning can be performed.

Since the catadioptric optical system as shown in FIG. 7 has little chromatic aberration, in an exposure apparatus having such a projection objective lens, axial chromatic aberration with respect to the exposure light IL and the illumination lights LM1 and LM2 as alignment light. Does not matter much. However, if there is chromatic aberration,
For example, a bifocal optical system is arranged between the semi-transmissive mirror 16 and the converging lens 17 to illuminate the mask 4 with illumination lights LM1 and LM1.
A positional shift may occur between the intersection of M2 and the intersection of the illumination lights LM1 and LM2 that illuminate the wafer 11.

[0017]

However, as described above, even if the projection objective lens including the catadioptric optical system is simply adapted to the TTR type alignment, the alignment light (that is, the illumination light LM1 and LM2) is used. )
Illuminates the alignment mark WA of the wafer 4, the alignment light is transmitted and reflected once by the semi-transmissive mirror 7 once. Then, on the return path where the alignment light from the alignment mark WA on the wafer 11 returns to the mask 4 side, the alignment light is reflected and transmitted once more by the semi-transmissive mirror 7. Therefore, if the intensity when the alignment light is incident on the mask 4 is I ₀ , and the transmissivity of the semi-transmissive mirror 7 to the alignment light is T _A , the semi-transmissive mirror 7 will not move until the alignment light is detected. Since transmission and reflection are performed twice each, the photoelectric sensor 2
The detected light amount I _{1 at 1} is as follows. However,
Attenuation in the wafer 11, the semi-transmissive mirror 16, and the like will be ignored. ^{_{I 1 = T A 2 (1}} -T A) 2 I 0 ...... (1)

From the equation (1), the transmittance T of the semi-transmissive mirror 7 is obtained.
_A is a detected light intensity I ₁ is a maximum at 50%, the quantity of light detected at this time is found to be a I ₀ /16(=0.0625I _0). However, the alignment light has a different wavelength from the exposure light in order not to expose the resist or the like applied on the wafer. In addition, it is difficult to manufacture a semi-transmissive mirror 7 that can maintain a constant transmittance characteristic for each wavelength. Further, in order to maximize the amount of exposure light to the wafer 11, the transmittance of the semi-transmission mirror 7 to the exposure light IL is set to 50.
% Is preferable. Therefore, by setting the transmittance of the semi-transmissive mirror 7 with respect to the exposure light of 50%, the transmittance T _A of the semi-transmissive mirror 7 with respect to the alignment light is likely to deviate significantly from the 50%.

For example, if the transmissivity T _A of the semi-transmissive mirror 7 to the alignment light is 90%, the detected light amount I is
From equation (1), I ₁ = (0.9) ² (1-0.9) ² I ₀ = 0.0081I ₀ , which is about 12.96% of the maximum detected light amount. As a result, the detected light amount I is extremely reduced, and the SN ratio is deteriorated, making it difficult to realize high detection accuracy.
In the above, the TTR system using the objective lens of the catadioptric system has been described, but the same problem applies to the TTL system.

The present invention has been made in view of the above-mentioned problems, and has been made in consideration of a projection optical system including a catadioptric optical system, and a TTR.
It is a first object of the present invention to provide a projection exposure apparatus having a high-performance alignment system capable of increasing the amount of detected light even when performing alignment using a TTL system or a TTL system. Further, the present invention provides a projection exposure apparatus which can achieve a certain amount of detected light even when the transmittance of a semi-transmissive mirror in a projection optical system composed of a catadioptric optical system differs between exposure light and alignment light. The second purpose is to provide. Further, the present invention uses such a projection exposure apparatus.
Pattern that can perform high-precision alignment
A third object is to provide a transfer method.

[0021]

A projection exposure apparatus according to the present invention, as shown in FIG. 1, for example, projects a pattern image on a mask (4) illuminated with exposure light IL onto a substrate (11). The optical system, its mask (4) and its substrate (1)
In a projection exposure apparatus having an alignment system for performing relative alignment of the 1), in the projection optical system, the exposure light IL through the mask (4) the first direction D1 and the first direction An optical path dividing member (7) for dividing the substrate (11) in the second direction D2 different from that of the substrate (11) by reflecting the exposure light divided in the second direction D2 and again passing through the optical path dividing member (7). ) An optical path between the concave reflecting mirror (9) leading above, the mask (4) and the optical path dividing member (7), and an optical path between the optical path dividing member (7) and the concave reflecting mirror (9). An optical path or a refractive optical system (6, 8, 10) provided on at least one optical path among optical paths between the optical path dividing member (7) and the substrate (11) is arranged , and alignment thereof is performed. The system is connected to its substrate via its projection optics
(11) From the alignment mark WA formed on
Yusuke detection optical system for detecting in a first direction D1 alignment light through the optical path splitting member (7) and (26)
It is those that. Also, the pattern transfer method according to the present invention
Is a pattern that transfers the mask pattern onto the photosensitive substrate
A transfer method using the projection exposure apparatus of the present invention.
Illuminate a mask and project an image of that mask pattern.
The light is projected onto the photosensitive substrate via the shadow optical system.

[0022]

According to the present invention, the alignment light passing through the projection optical system composed of the catadioptric system is illuminated on the alignment mark WA on the substrate (11), and the light from the alignment mark WA is emitted from the projection optical system. It is detected in the first direction D1 only once through the middle optical path dividing member (7).
Therefore, in the configuration of FIG. 1, for example, the alignment light is transmitted twice through the optical path dividing member (7) and is reflected once by the optical path dividing member (7) until the alignment light is finally detected.
On the other hand, in the configuration of FIG. 4, for example, the alignment light is transmitted once through the optical path dividing member (7) and is reflected twice by the optical path dividing member (7) until it is finally detected.

Accordingly, in the prior art, the alignment light is reflected twice and transmitted twice by the optical path dividing member (7), whereas in the present invention, the alignment light is reflected or transmitted by the optical path dividing member (7). Is reduced by one. Therefore, the detected light amount of the alignment light is larger than that of the conventional example while the TTL method or the TTR method is substantially adopted. Further, even when the transmittance of the optical path dividing member (7) in the projection optical system differs between the exposure light IL and the alignment light, it is possible to secure a certain amount of detected light. Thereby, S
Since an improvement in the N ratio can be achieved, highly accurate alignment is guaranteed.

[0024]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of a projection exposure apparatus according to the present invention will be described below with reference to FIGS. This example is an improvement of the virtual conventional example of FIG.
The same reference numerals are given to the portions corresponding to and the detailed description thereof will be omitted. FIG. 1 shows the entire configuration of the projection exposure apparatus of the present embodiment. In FIG.
The transflective mirror 7, the correction lens group 8, the concave reflecting mirror 9, and the second lens group 10 constitute a projection objective.

Reference numeral 22 denotes an illumination optical system for exposure, 23 denotes an illumination optical system for alignment, and 24 denotes an objective lens for alignment.
Exposure light IL composed of ultraviolet light of nm is emitted, and a laser beam having a wavelength of, for example, 633 nm is emitted from alignment illumination optical system 23 as alignment light AL. A dichroic mirror 3A is provided diagonally on the right side of the mask 4, and the dichroic mirror 3A
Are provided with an exposure illumination optical system 22 in the transmission direction, and an alignment objective lens 24 and an alignment illumination optical system 23 in the reflection direction. The dichroic mirror 3A transmits the exposure light IL and outputs the alignment light AL.
Has wavelength selectivity to reflect light.

Exposure light IL from the exposure illumination optical system 22
Illuminates the pattern area of the mask 4 uniformly through the dichroic mirror 3A. Then, the exposure light IL transmitted through the mask 4 enters the concave reflecting mirror 9 via the first lens group 6, the semi-transmissive mirror 7, and the correction lens group 8. The exposure light reflected by the concave reflecting mirror 9 is directed again to the semi-transmissive mirror 7 via the correction lens group 8, and the exposure light reflected by the semi-transparent mirror 7 is finally transmitted by the second lens group 10. Wafer 11
Focused on top. Therefore, the image of the circuit pattern on the mask 4 is transferred onto the wafer 11 by the projection objective composed of the catadioptric optical system.

On the other hand, the alignment light AL emitted from the alignment illumination optical system 23 is focused by the alignment objective lens 11, and the alignment light AL is reflected by the dichroic mirror 3A and focused on the mask 4. As shown in FIG. 2A, a transmission window 4b for transmitting the alignment light AL is formed near the convergence point of the alignment light AL on the mask 4. The transmission window 4b is provided outside the pattern area PA of the mask 4.

Returning to FIG. 1, the transmission window 4b of the mask 4 will be described.
The alignment light transmitted through the upper part enters the concave reflecting mirror 9 via the first lens group 6, the semi-transmissive mirror 7, and the correcting lens group 8. The alignment light reflected by the concave reflecting mirror 9 is directed to the semi-transmissive mirror 7 via the correcting lens group 8, and after being reflected by the semi-transmissive mirror 7, is phase-shifted on the wafer 11 by the second lens group 10. It is focused on an alignment mark WA made of a diffraction grating. This alignment mark WA
As shown in FIG. 2B, the mask 4 is formed at a position on the wafer 11 corresponding to the transmission window 4b of the mask 4 in FIG.
This alignment mark WA has a certain shot area 27.
The alignment light transmitted through the transmission window 4b of the mask 4 is applied to the inside of the rectangular area 28A surrounding the alignment mark WA.

The alignment light reflected from the alignment mark WA on the side of the wafer 11 returns to the second lens group 10
And goes to the semi-transmissive mirror 7. The direction in which light from the first lens group is reflected with respect to the semi-transmissive mirror 7 is referred to as a first direction D1,
Assuming that the direction in which the light from the first lens group 6 is transmitted as it is is the second direction D2, the concave reflecting mirror 9 is disposed in the second direction D2. In this example, a condensing lens 25 and a photoelectric detector 26 composed of a two-dimensional charge-coupled imaging device (CCD) are sequentially arranged in the first direction D1 of the semi-transmissive mirror 7. Then, the alignment light reflected from the wafer 11 and transmitted through the semi-transmissive mirror 7 is focused on the photoelectric detector 26 by the condenser lens 25. Thus, an image of the alignment mark WA is formed on the light receiving surface of the photoelectric detector 26.

In this case, on the light receiving surface of the photoelectric detector 26,
As shown in FIG. 3A, an image WAI of the alignment mark WA and an image 4bI of the transmission window 4b of the mask 4 are formed. Therefore, the mask 4 and the wafer 1 as the center and between edge E ₁ and E ₂ at both ends of the image 4bI transmission window 4b of the mask 4 coincides with the center of the image WAI of alignment marks
The relative positioning of the mask 4 and the wafer 11 is achieved by relatively moving 1. Specifically, from a certain scanning line of the photoelectric detector 26, FIG.
As shown in (b), an imaging signal S composed of a sine wave signal and an edge signal is obtained. By processing this signal S, the relative position between the image 4bI of the transmission window 4b and the image WAI of the alignment mark is obtained. It is possible to obtain a simple positional relationship.

The photoelectric detector 26 may detect the alignment mark WA and the scattered light from the edge of the transmission window 4b, and may perform the alignment based on the detected relative positional relationship. Also, mask 4
A plurality of transmission windows are formed outside the transmission window 4b, and a plurality of alignment marks are also formed on the wafer 11 corresponding to the transmission windows 4b. By performing alignment for each of them, finally alignment in a two-dimensional plane is performed. Is completed.

Note that the projection objective lens (6 to 10) has a slight chromatic aberration (for example, axial chromatic aberration) with respect to the alignment light.
Is left, a beam splitter is provided between the condenser lens 25 and the photoelectric detector 26 so that the detection optical path is
The light may be divided, and separate photoelectric detectors may be arranged at the light-collecting position of the exposure light and the light-collecting position of the alignment light in the divided light path. Then, if one photoelectric detector detects the image of the alignment mark WA and the other photoelectric detector detects the image of the transmission window 4b, both sharp images can be detected.

Next, the received light amount of the alignment light in the photoelectric detector 26 of this embodiment is compared with the received light amount of the conventional photoelectric sensor 21 of the TTR system shown in FIG. First, FIG.
Projection objective lenses (6 to
10), the intensity I _EW of the exposure light IL reaching the wafer 11 via the mask 11 is the intensity of the exposure light when the mask 4 is illuminated.
_E, the transmittance of the semi-transmissive mirror 7 with respect to the exposure light when a T _E, the following relation is established. _{I EW = T E (1-} T E) I E ...... (2) , however, it is 0 <T _E <1. Here, the intensity of the exposure light on the wafer 11 is expressed by the transmittance T
_The maximum value is obtained when _E is 0.5 (50%), and the intensity I _EW of the exposure light at this time is I _{E / 4 (} = 0.25I _E ).

In the alignment system according to the present embodiment, the alignment light is illuminated on the alignment mark WA and is detected by the semi-transmissive mirror 7 in the projection objective until it is detected.
Is transmitted twice and reflected once by the semi-transmissive mirror 7. Therefore, assuming that the intensity of the alignment light is I ₀ , the transmittance of the semi-transmissive mirror 7 with respect to the alignment light is T _A , and the detected light amount is I ₁ , the detected light amount I ₁ in this embodiment is represented by the following equation. The relationship is established. I ₁ = T _A ² (1−T _A ) I ₀ (3)

Here, considering the maximum illumination efficiency of the exposure light, the transmittance of the semi-transmissive mirror 7 with respect to the exposure light IL is almost 1 /.
2 (50%) is desirable.
Therefore, it is assumed that the transmissivity of the semi-transmissive mirror 7 is the same for the exposure light and the alignment light, and the transmissivity of the semi-transmissive mirror 7 to the alignment light at this time is 1/2 (50%). Then, the quantity of light detected alignment system alignment light of the embodiment shown in FIG. 1, the equation (3), I _0/8
(0.1255I ₀ ), while T as shown in FIG.
From equation (1), the detected light amount of the alignment light of the TR type alignment system is I 0/16 ( ₌ 0.0625I ₀ ).
Becomes Accordingly, in the present example, the detected light amount is doubled while the TTR method is substantially adopted, as compared with the apparatus of FIG. 7, so that the SN ratio can be improved. It can be understood that it is performed.

As described above, the detected light amount in the TTR type alignment system shown in FIG. 7 is maximum when T = 1/2 (50%) according to the equation (1). Is I 0/16 ( ₌ 0.0625I ₀ ). On the other hand, the detection light amount in the alignment system of the embodiment of FIG. 1 is maximized when T _A = 2/3, the detection light amount at this time is 4I ₀ /27(≒0.1485I _0). Therefore, the theoretical maximum detected light quantity in this example is twice or more as compared with the conventional example.

As described above, since the alignment light has a different wavelength from the exposure light in order not to expose the resist on the wafer 11 during the alignment, the transmittance of the semi-transmissive mirror 7 is different from that of the exposure light. It is often different from light. Specifically, when the transmittance T _A with respect to the alignment light AL of semitransparent mirror 7 and 0.9 (90%), detected light intensity I ₁ of the alignment light from the equation (3) in the example of FIG. 1 is as follows Become. Meanwhile I ₁ = 0.081I _0, when the transmittance T _A is 0.9, the detection light quantity I ₁ of the alignment light in the conventional example of FIG. 7 is as follows. I1 = 0.0081I ₀ Therefore, the detected light amount in this example is ten times as large as that in the conventional example, and a great improvement in alignment is achieved.

At this stage, it is difficult to set the transmittance of the semi-transmission mirror 7 for each wavelength band to an arbitrary value. However, the transmittance of the future semitransparent mirror 7 consider the case where it can be set arbitrarily for each wavelength band, consider the set range of the transmittance T _A with respect to the alignment light AL of semitransparent mirror 7. In this case, to ensure alignment light amount necessary for detection, to obtain a higher alignment accuracy than the conventional example, the alignment system of the embodiment shown in FIG. 1, the maximum detected light intensity I _0/16 of the device of FIG. 7 ( = 0.0625I ₀ ). That is, it is sufficient that the following condition is satisfied from Expression (3). ^{_{T A 2 (1-T A}} ) I 0> I 0/16 ...... (4) maximum detected light intensity I _0/1 than the equation (4), the apparatus of FIG. 7
The preferred range of the transmittance T _A of the semi-transmissive mirror 7 that can secure a light amount larger than 6 (= 0.0625I ₀ ) is as follows. _{0.30 <T A <0.93 ...... (} 5)

As described above, even if the transmissivity T _A of the transflective mirror 7 for the alignment light is different from the transmissivity T _{E of the} exposure light, a transflective mirror 7 satisfying the range of the expression (5) is used. In addition, it is possible to secure a sufficient amount of alignment light for detection. In this embodiment, the transmittance T _E of the semi-transmissive mirror 7 for the exposure light when the illumination efficiency of the exposure light is maximized is 0.5, but the transmittance T _E of the beam splitter for the alignment light is obtained from Expression (5). _{A is} the transmittance T for the exposure light
_It can be understood that the detected light amount of the alignment light becomes particularly large when the value becomes higher than _E. As described above, according to this example, the number of times that the alignment light AL passes through the semi-transmissive mirror 7 is smaller than that of the conventional example by one time, so that the attenuation of the alignment light in the semi-transmissive mirror 7 is small and the alignment light is detected. There is an advantage that the amount of light increases.

Next, a modification of the embodiment of FIG. 1 will be described with reference to FIG. In FIG. 4, parts corresponding to those in FIG. 1 are denoted by the same reference numerals. In the modification of FIG. 4, FIG.
The positions of the concave reflecting mirror 9 and the photoelectric detector 26 are exchanged. That is, in FIG. 4, the direction in which the light from the first lens group 6 passes through the semi-transmissive mirror 7 is referred to as a first direction D1,
The direction in which the light is reflected to the second direction D2, the condenser lens 25 and the photoelectric detector 26 are arranged in the first direction D1, and the correction lens group 8 and the concave reflecting mirror 9 are arranged in the second direction D2. Deploy. The other configuration is the same as that of FIG. 1, and a detailed description thereof will be omitted.

By the way, the wafer 11 is passed through the projection objective (6 to 10) composed of the catadioptric optical system shown in FIG.
Intensity I _EW of exposure light reaches the top is the same as equation (2), the transmittance T _E of the concave reflecting mirror 7 becomes maximum when 1/2 (50%). At this time, the intensity I _EW of the exposure light on the wafer 11 is I _{E / 4 (} = 0.25I _E ).

Next, the relationship between the amount of detected alignment light AL and the transmittance of the semi-transmissive mirror 7 will be considered for the alignment system of the modification shown in FIG. 4 and the alignment system of the TTR system as shown in FIG. . In the alignment system of this modification, the alignment light is reflected twice by the semi-transmissive mirror 7 in the projection objective lens until the alignment light is illuminated on the alignment mark WA on the wafer 11 side and detected. Transmit once. Therefore, assuming that the intensity of the alignment light is I ₀ , the transmittance of the semi-transmissive mirror 7 for the alignment light is T _A , and the detected light amount is I ₁ , the detected light amount I ₁ in the present modified example is represented by the following equation. Relationship is established. I ₁ = (1−T _A ) ² T _A I ₀ (6)

From this equation (6), the detected light amount in the alignment system of the modification of FIG. 4 is T = 1/3 (≒ 0.33
3) becomes maximum when the detection light intensity at this time 4I _0/2
7 (≒ 0.1485I ₀ ). This detected light amount is more than twice the maximum detected light amount in the conventional example of FIG. If the transmissivity T _{A of the} transflective mirror 7 with respect to the alignment light is 0.1, the detected light amount I ₁ in the present modified example is 0.081I ₀ , whereas in the conventional example of FIG. The detected light amount I ₁ of the alignment light is 0.0081.
It becomes I ₀ . Therefore, in the present modified example, especially when the transmittance of the semi-transmissive mirror 7 to the alignment light is small, the detected light amount of the alignment light is about 10 times that of the conventional example.

Also in this modification, an optimum transmittance range for alignment light when the transmittance of each wavelength band of the semi-transmissive mirror 7 can be arbitrarily controlled is determined. To ensure alignment light amount necessary for detection in this modification, in order to obtain a high alignment accuracy, the maximum detected light intensity I _0/16 of the device of FIG. 7
(= 0.0625I ₀ ). That is, it is sufficient that the following condition is satisfied from the equation (6). ^{_{(1-T A) 2 T}} A I 0> I 0/16 ...... (7) Maximum detected light intensity I _0/16 than this equation, the apparatus of FIG. 7 (=
_A preferred range of the transmittance T _A of the semi-transmissive mirror 7 that can secure a light amount larger than 0.0625I ₀ ) is as follows. _{0.07 <T A <0.70 ...... (} 8)

As described above, even if the transmissivity T _A of the transflective mirror 7 for the alignment light is different from the transmissivity T _{E of the} exposure light, the transflective mirror 7 which satisfies the range of the expression (8) is used. An alignment light amount sufficient for detection can be secured. In particular, in the present modified example, from equation (8), the transmissivity T _E (= 0.5) of the semi-transmissive mirror 7 with respect to the exposure light when the illumination of the exposure light is maximized is smaller than the transmissivity TE with respect to the alignment light. This is advantageous when the transmittance T _A is lower.
As described above, in this modified example, the number of times that the alignment light has passed through the semi-transmissive mirror 7 is one less than that in the conventional example.
Since the attenuation due to transmission through the semi-transmissive mirror 7 is small, there is an advantage that the detected light amount of the alignment light can be finally increased.

Next, another embodiment of the present invention will be described with reference to FIGS. This embodiment is obtained by applying the alignment system of the conventional example shown in FIG. 7 to the embodiment shown in FIG. 1. In FIG. 5, parts corresponding to those shown in FIG. 1 and FIG. I do. FIG. 5 shows the configuration of the present example. In FIG.
Exposure light IL is transmitted through the dichroic mirror 3A to irradiate the pattern area of the wafer 4. The first illumination light LK1 and the second illumination light LM2, which have been converted into parallel laser beams by the focusing lens 17, respectively,
The light is reflected by the dichroic mirror 3A and enters the mask 4 at a predetermined intersection angle. The frequency of the first illumination light LM1 and the frequency of the second illumination light LM2 are different by Δf, but the two illumination lights are coherent. In the vicinity of the pattern area PA of the mask 4, as shown in FIG. 6A, a transmission window 4b for transmitting alignment light (that is, illumination light LM1, LM2) and a diffraction grating having an amplitude or phase pitch P1. The alignment mark MA is formed in parallel. Then, the illumination lights LM1 and LM2 reflected by the dichroic mirror 3A intersect in a region including the transmission window 4b on the mask 4 and the alignment mark MA.

Further, on the wafer 11 of this embodiment, FIG.
As shown in (b), an alignment mark WA and a flat reflection are formed in a region 29A conjugate to the transmission window b of the mask 4 and a region 30A conjugate to the alignment mark MA outside the shot region of the wafer 11 by a diffraction grating having a pitch Q1. Form a part.

In FIG. 5, transmitted light of the first-order diffracted light (hereinafter referred to as + 1st-order diffracted light) by one illumination light LM1 incident on the alignment mark MA on the mask 4 and the other illumination light LM2 incident on the mark MA The pitch P1 of the mark MA and the intersection angle of both beams are selected so that the transmitted light of the -1st-order diffracted light is parallel. The transmitted light parallel and mixed with each other is represented by transmitted light LM3. The intensity of the transmitted light LM3 changes sinusoidally at the beat frequency Δf. The transmitted light LM3 and the illumination lights LM1 and LM2 transmitted through the transmission window 4b on the mask 4 pass through the first lens group 6, the semi-transmission mirror 7, and the correction lens group 8, and enter the concave reflection mirror 9. The transmitted light LM3 and the illumination lights LM1 and LM2 reflected by the concave reflecting mirror 9 pass through the correcting lens group 8 again to reach the semi-transmitting mirror 7, and after being reflected by the semi-transmitting mirror 7, the second lens group At 10 it is focused on a wafer 11.

In this case, the transmitted light LM3 irradiates the flat area 30A on the wafer 11, and the illumination light LM1 and LM2
Is incident on the alignment mark WA on the wafer 11 at a predetermined intersection angle. Then, the reflected light of the first-order diffracted light (hereinafter referred to as + 1st-order diffracted light) of the one illumination light LM1 by the alignment mark WA and the reflected light of the -1st-order diffracted light by the mark WA of the other illumination light LM2 are parallel to each other. Then, the pitch Q1 of the mark WA and the intersection angle of both beams are selected. The reflected light mixed in parallel with each other is represented by reflected light LW0. The intensity of the reflected light LW0 also changes sinusoidally at the beat frequency Δf. Further, the transmitted light LM3 incident on the region 30A on the wafer 11 is reflected as it is.

The transmitted light LM3 and the reflected light LW0 thus reflected pass through the second lens group 10 in parallel to the semi-transmissive mirror 7
Then, the transmitted light LM3 and the reflected light LW0 transmitted through the semi-transmissive mirror 7 are focused near the opening at the center of the aperture plate 19, and then are incident on the light receiving elements 21a and 21b of the photoelectric sensor 21 by the relay lens 20. At that time, wafer 1
1 and the photoelectric sensor 21 are in a conjugate relationship. The light receiving elements 21a and 21b are actually shifted in a direction perpendicular to the paper surface of FIG. 5, for example, the transmitted light LM3 enters the light receiving element 21a, and the reflected light LW0 enters the light receiving element 21b. The 0-order reflected light or the high-order diffracted light other than the transmitted light LM3 and the reflected light LW0 is blocked by the aperture plate 19.

Then, one of the light receiving elements 21a is
, And the other light receiving element 21b detects a beat signal corresponding to the alignment mark WA of the wafer 11. In this detection method, the relative alignment between the mask 4 and the wafer 11 is performed by relatively moving the mask 4 and the wafer 11 such that the phase difference between the two beat signals becomes, for example, zero. . This alignment method is a so-called heterodyne type alignment, which is disclosed in, for example, Japanese Patent Application Laid-Open No. 2-133913 by the present applicant.

Also in the embodiment of FIG. 5, the alignment light (ie, the alignment light in the semi-transmissive mirror 7) is different from that of the conventional example of FIG.
Since the number of times of transmission of the transmitted light LM3 and the illumination light LM1, LM2) is one less, there is an advantage that the detection intensity of the alignment light can be increased. When the optical system shown in FIG. 5 is modified as shown in FIG. 4, the number of times of reflection of the alignment light by the semi-transmissive mirror 7 can be reduced by one, and similarly, the detection intensity of the alignment light is increased. be able to.

It should be noted that the present invention is not limited to the above-described embodiment, but can adopt various configurations without departing from the gist of the present invention.

[0054]

According to the present invention, the number of times of transmission or reflection of alignment light in the optical path dividing member in the projection optical system can be reduced by one, so that the amount of detected alignment light can be increased. There are advantages. Furthermore,
Even when the transmittance of the optical path splitting member is different between the exposure light and the alignment light, there is an advantage that a certain amount of detected light for the alignment light can be secured.

[Brief description of the drawings]

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

FIG. 2A is a plan view showing a pattern of a mask 4 of the embodiment, and FIG. 2B is a plan view of a main part showing a pattern of a wafer 11 of the embodiment.

FIG. 3A is a diagram illustrating an image observed in the embodiment, and FIG. 3B is a waveform diagram illustrating an imaging signal corresponding to the image.

FIG. 4 is a configuration diagram showing a modification of the embodiment of FIG. 1;

FIG. 5 is a configuration diagram including a partial cross-sectional view showing another embodiment of the present invention.

6A is an enlarged plan view of a main part showing a pattern of a mask 4 in the embodiment of FIG. 5, and FIG. 6B is an enlarged plan view of a main part showing a pattern of a wafer 11 in the embodiment of FIG. is there.

FIG. 7 is a configuration diagram including a partial cross-sectional view showing an example in which a conventional alignment system is virtually added to a projection exposure apparatus in which a projection objective is configured by a catadioptric optical system.

8 is an enlarged plan view showing a main part of a pattern of a mask 4 and a pattern on a wafer in the example of FIG. 7;

[Explanation of symbols]

 Reference Signs List 4 mask 6 first lens group 7 semi-transmissive mirror 8 correction lens group 9 concave reflecting mirror 10 second lens group 11 wafer 22 exposure illumination optical system 23 alignment illumination optical system 25 condenser lens 26 photoelectric detector IL exposure light AL Alignment light WA Wafer-side alignment mark MA Mask-side alignment mark LM1 First illumination light LM2 Second illumination light

Claims (4)

(57) [Claims] 1. A projection exposure apparatus comprising: a projection optical system for projecting an image of a pattern on a mask illuminated with exposure light onto a substrate; and an alignment system for performing relative positioning between the mask and the substrate. in, the projection optical system, the optical path splitting member for splitting the exposure light through the mask and in a second direction different from the first direction and the first direction, divided in the second direction exposure A concave reflecting mirror that reflects light and guides the light onto the substrate again through the optical path dividing member; and an optical path between the mask and the optical path dividing member, between the optical path dividing member and the concave reflecting mirror. a refractive optical system provided on at least one optical path of the light path between the light path or the optical path splitting member and the substrate, arranged respectively, the alignment system, the group via the projection optical system
Characterized in that it has an alignment illumination optical system for illuminating the alignment mark formed on the plate, and a detection optical system for detecting in said first direction alignment light through the optical path splitting member from the alignment mark Projection exposure apparatus. 2. The optical system according to claim 1, wherein the detection optical system is arranged in a direction in which the exposure light through the mask is reflected by the optical path dividing member with respect to the optical path dividing member, and the transmission of the optical path dividing member to the alignment light. 2. The projection exposure apparatus according to claim 1, wherein the following condition is satisfied when the ratio is T _A. 0.30 <T _A <0.93 3. The detection optical system is disposed in a direction in which exposure light passing through the mask passes through the optical path splitting member with respect to the optical path splitting member, and a transmittance of the optical path splitting member to the alignment light. when to the T _a, the projection exposure apparatus according to claim 1, characterized in that the following condition is satisfied. 0.07 <T _A <0.70 4. A pattern transfer method for transferring a mask pattern onto a photosensitive substrate, wherein the projection exposure apparatus according to claim 1 illuminates the mask, A pattern transfer method, wherein an image of a pattern is projected onto the photosensitive substrate via the projection optical system.