JPH05275312A - Aligner - Google Patents

Abstract

PURPOSE:To prevent diffusion of alignment light at a pellicle mounted on a reticle when an optical element for correcting color aberration is employed. CONSTITUTION:Two set of correction optical elements 1 having different correction amount of magnification color aberration with respect to an alignment light are arranged on the outside of a region 2, through which zero order diffraction light from a secondary light source passes, in the region 3 of the pupil surface of a projection optical system. Two set of correction optical elements are used properly depending on the size of the exposure region of a reticle.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an alignment apparatus suitable for application to an alignment system of a projection exposure apparatus having a projection optical system for transferring a pattern formed on a reticle (mask) onto a wafer, In particular, the present invention relates to an alignment apparatus that performs relative alignment between a reticle and a wafer by using alignment light in a wavelength band different from exposure light for transferring a reticle pattern onto a wafer.

[0002]

2. Description of the Related Art When a semiconductor element, a liquid crystal display element, or the like is manufactured by a photolithography technique, a projection exposure apparatus is used which transfers a reticle pattern onto a wafer via a projection optical system. In general, semiconductor devices are manufactured by forming a multi-layered circuit pattern on a wafer, so in the process of transferring a new reticle pattern to a wafer on which a pattern has already been formed, the pattern to be transferred this time has already been formed. It is necessary to accurately align the reticle and the wafer with each other.
Recently, as the fineness of the transferred pattern is further improved, it is the most important issue to improve the accuracy of the alignment.

Conventionally, regarding the alignment accuracy, a so-called TTL for detecting the alignment mark on the wafer through the projection optical system to align the reticle and the wafer.
(Through the lens) method is expected to achieve the highest accuracy in principle. Further, a so-called TTR (through the reticle) method, which is included in the TTL method and refers to a wafer through both a reticle and a projection optical system, is expected to achieve higher accuracy. Thus, in the method of detecting the alignment mark on the wafer via the projection optical system,
By using alignment light having a wavelength band different from that of the exposure light, consideration is given so that the resist applied on the wafer is not exposed.

However, when performing alignment based on alignment light in a wavelength band different from the exposure light, there is a problem that axial chromatic aberration and lateral chromatic aberration occur with respect to the alignment light by the projection optical system. Note that the chromatic aberration of magnification referred to in the present application means lateral chromatic aberration. This means off-axis light having the same wavelength as the exposure light that is imaged on the Gaussian image plane by passing through the projection optical system and the projection optics. Both the chief rays of the exposure light, which is imaged at the Gaussian image plane or before and after the Gaussian image plane by passing through the system, and the alignment light having a different wavelength, are between the intersecting positions when intersecting on the Gaussian image plane. It defines the deviation.

Then, the amount of lateral chromatic aberration (horizontal chromatic aberration amount) Δ
T is a value on the Gauss image plane from a crossing position where a chief ray in the off-axis light having the same wavelength as the exposure light imaged on the Gauss image plane by passing through the projection optical system intersects on the Gauss image plane. The distance to the optical axis position of the projection optical system is δ1, and the principal ray of the alignment light having a wavelength different from that of the Gaussian image plane or the exposure light imaged before and after passing through the projection optical system is the Gaussian image plane. When the distance from the crossing position intersecting with each other to the optical axis position of the projection optical system on the Gaussian image plane is δ2, it is defined by ΔT = | δ2-δ1 |.

The applicant of the present application has proposed in Japanese Patent Application No. 3-129563 an alignment apparatus in which axial chromatic aberration and lateral chromatic aberration are corrected by a projection optical system for alignment light. FIG. 7 shows an alignment apparatus disclosed in the Japanese Patent Application No. 3-129563. In FIG. 7, the pattern surface of the reticle 6 and the exposure surface of the wafer 9 are both side telecentric projection optics under the exposure light. System 21
Are arranged to be conjugate to. The projection optical system 21 is satisfactorily corrected for chromatic aberration with respect to exposure light such as excimer laser light. A reticle mark RM in the form of a diffraction grating for alignment is formed on the reticle 6, and a wafer mark 10 in the form of a diffraction grating for alignment is also formed on the wafer 9. The wafer 9 is adsorbed on a wafer stage (not shown) that can move in the X and Y directions in the horizontal plane, rotate in the horizontal plane, and move in the Z direction parallel to the optical axis of the projection optical system 21. An illumination optical system for exposure light (not shown) is arranged above the reticle 6.

In the alignment system shown in FIG. 7, reference numeral 31 denotes an alignment light source for generating alignment light having a wavelength band different from that of the exposure light. As the alignment light, for example, He-Ne laser light having a wavelength of 633 nm is used. .. The light beam emitted from the light source 31 is split into a first light beam 4a and a second light beam 4b by a semi-transmissive mirror 32, and the first light beam 4a passes through a first acousto-optic modulator 34a to a semi-transmissive mirror 35. The second light beam 4b faces the bending mirror 33.
And the semi-transmissive mirror 35 via the second acousto-optic modulator 34b.
Head to. The acousto-optic modulators 34a and 34b are driven by high-frequency signals of frequencies f1 and f2 (f2 = f1-Δf), respectively, and the frequencies of the light beams 4a and 4b passing through the acousto-optic modulators 34a and 34b differ by Δf. .. In this case, it is desirable that f1 >> Δf and f2 >> Δf, and the upper limit of Δf is determined by the responsiveness of alignment photoelectric detectors 51a to 51c described later.

Light fluxes 4a and 4 reflected by the semi-transmissive mirror 35
The light b is condensed by a condenser lens 36 on a reference standard diffraction grating 37 formed at a predetermined pitch in a direction parallel to the paper surface of FIG. Two beams 4 with relative frequency difference Δf
An interference fringe that flows on the diffraction grating 37 is formed by a and 4b, and the diffracted light that has passed through the diffraction grating 37 enters the photoelectric detector 38. The reference signal output from the photoelectric detector 38 becomes a sinusoidal AC signal (optical beat signal) corresponding to the cycle of the change in brightness of the interference fringes formed on the diffraction grating 37.

On the other hand, the two light beams 4a and 4b which have passed through the semi-transmissive mirror 35 are focused on the reticle mark RM provided outside the exposure area of the reticle 6 through the alignment objective lens 39 and the mirror 50. .. At this time, on the reticle mark RM, the difference Δf in frequency between the light beams 4a and 4b.
An interference fringe that changes so as to flow is formed. The reticle mark RM is formed by a diffraction grating formed at a predetermined pitch in a direction parallel to the paper surface of FIG.
A reticle window RW for transmitting alignment light is formed at a position adjacent to M.

The light beams 4a and 4b focused on the reticle 6 by the objective lens 39 of the alignment system cover the reticle 6 at a predetermined crossing angle so as to cover not only the reticle mark RM but also the reticle window RW at the same time. Illuminate from the direction. Then, one of the light beams 4a is reticle mark R.
When M is obliquely irradiated, the 0th order light is reflected in the direction (regular reflection direction) of the other light beam 4b in the opposite direction, and the + 1st order diffracted light is generated in the direction of the other light beam 4a in the opposite direction. Similarly, when the other light beam 4b obliquely irradiates the reticle mark RM, the 0th order light is reflected in a direction that follows the optical path of the one light beam 4a in the opposite direction, and the −1st order diffracted light in the direction that follows the optical path of the light beam 4b in the opposite direction. Occurs.

In this case, the light beams 4a and 4b incident on the reticle mark exist in a plane parallel to the paper surface of FIG. 7, and the incident angles of the light beams 4a and 4b with respect to the reticle mark are set to θ _R and −θ _R , respectively. If the pitch of the marks in the direction parallel to the paper surface of FIG. 7 is P _R and the wavelengths of the light beams 4a and 4b are λ (however, strictly speaking, they are slightly deviated from λ), sin2θ _R = λ / P _R The incident angle and the pitch are set so as to satisfy

Therefore, the + 1st-order light of one light beam 4a from the reticle mark RM and the 0th-order light of the other light beam 4b return to the semitransparent mirror 35 through the mirror 50 and the objective lens 39 in parallel with each other. Photoelectric detector 51a reflected by the transmission mirror 35 and arranged at a position conjugate with the pupil of the objective lens 39.
Incident on. Similarly, the 0th-order light of one light beam 4a from the reticle mark RM and the -1st-order light of the other light beam 4b return to the semi-transmissive mirror 35 through the mirror 50 and the objective lens 39 in parallel with each other, and this semi-transmissive light is transmitted. Photoelectric detector 51b reflected by the mirror 35 and arranged at a position conjugate with the pupil of the objective lens 39
Incident on. An optical beat signal corresponding to the position of the reticle mark RM is detected by these photoelectric detectors 51a and 51b.

Next, the light beams 4a and 4b obtained by illuminating the reticle window RW adjacent to the reticle mark RM from two directions at a predetermined crossing angle (2θ _R ) pass through the reticle window RW as they are, and are projected onto the projection optical system 21. Incident from off-axis. The projection optical system 21 is sufficiently chromatic aberration-corrected for the exposure light, but is not chromatic-aberration-corrected for the alignment light in the wavelength band different from the exposure light. Therefore, the projection optical system 2
A transparent member 5 is arranged on the pupil plane of No. 1, and X is a measurement direction passing through the center of the optical axis of the projection optical system 21 on the transparent member 5.
3 with different pitches along the direction
Phase type diffraction gratings 1a to 1c as individual correction optical elements
Are arranged. The diffraction grating 1c is arranged on the optical axis of the projection optical system 21, and the diffraction gratings 1a and 1b are arranged symmetrically with respect to the optical axis of the projection optical system 21. In addition, each of the diffraction gratings 1a to 1c includes the diffraction gratings 1b and 1b.
The diffraction gratings are arranged in the order of c and 1a along the X direction so that the pitches of the diffraction gratings become dense.

In FIG. 7, light beams 4a and 4b which are incident on the projection optical system 21 from off-axis and are incident on the pupil (incident pupil) of the projection optical system 21 are diffraction gratings 1b and 1 respectively.
The wafer mark 10 formed on the wafer 9 is deflected (diffracted) by a so as to be corrected by each of the correction angles θ1 and θ2, and is irradiated from two directions at a predetermined crossing angle. Interference fringes that change so as to flow at the frequency Δf of the difference between the two light fluxes are also formed on the wafer mark 10. The wafer mark 10 is composed of a diffraction grating formed at a predetermined pitch in the X direction, which is the measurement direction, on the street line outside each shot area.

By thus irradiating the wafer mark 10 with the light beams 4a and 4b at a predetermined intersection angle, the −1st-order light of one light beam 4a and the + 1st-order light of the other light beam 4b are exposed on the wafer 9. It occurs in the normal direction to the surface (direction parallel to the optical axis of the projection optical system 21). In this case, the wafer mark 1
The pitch of 0 is P _W , the wavelength of the alignment light is λ, and the luminous flux 4
When the crossing angle of a and 4b is 2θ _W , sin θ _W =
The pitch and the crossing angle are set so as to satisfy the relationship of λ / P _W.

The −1st-order light of the light beam 4a and the + 1st-order light of the light beam 4b generated in the normal direction of the wafer mark 10 travel in parallel with each other as beat interference light 4c on the optical path of the principal ray of the projection optical system 21, After being deflected (diffracted) by the correction angle θ3 by the diffraction grating 1c provided at the center of the pupil of the projection optical system 21, the objective is passed through the reticle window RW of the reticle 6, the mirror 50, the objective lens 39 and the semi-transmissive mirror 35 again. The light enters a photoelectric detector 51c arranged at a position conjugate with the pupil of the lens 39. The photoelectric detector 51c detects an optical beat signal corresponding to the position of the wafer mark 10. Reticle mark R output from the photoelectric detectors 51a and 51b described above.
The relative positional relationship between the reticle 6 and the wafer 9 can be accurately detected by the signal including the position information of M and the signal including the position information of the wafer mark 10 output from the photoelectric detector 51c.

This detection method is called a so-called optical heterodyne method, and if the positional deviation between the reticle 6 and the wafer 9 from the reference state is within 1 pitch of the reticle mark RM and within 1/2 pitch of the wafer mark 10. Even in the stationary state, the positional deviation amount can be accurately detected with high resolution. Therefore, it is suitable for use as a position signal when a position servo of the closed loop system is applied so as not to cause a minute position shift while the pattern of the reticle 6 is exposed on the resist of the wafer 9. In this detection method, after the reticle 6 or the wafer 9 is moved so that the phase of the optical beat signal from the reticle mark RM and the phase of the optical beat signal from the wafer mark 10 become predetermined values, alignment is completed. The servo lock can be continuously applied so that the reticle 6 and the wafer 9 do not move relative to each other at the alignment position.

As described above, in the alignment apparatus of FIG. 7, the phase type diffraction gratings 1a, 1b and 1c are arranged on the transparent member 5 provided at the pupil position of the projection lens 21.
The alignment light (light fluxes 4a, 4
Projection optical system 21 of b) and detection light (beat interference light 4c)
Axial chromatic aberration and lateral chromatic aberration due to are corrected. Incidentally, a correction optical element such as a deflection prism can be used instead of the phase type diffraction gratings 1a to 1c.

In the correction optical element such as the phase type diffraction grating (phase grating), the Japanese Patent Application No. 3-129563 is used.
As described in the publication, measures are taken so as not to affect the wavefront of the exposure light and, as a result, adversely affect the imaging performance of the projection optical system 21. For example, in the case of a phase grating, the etching depth of the phase grating is an integral multiple of the wavelength of the exposure light, or a thin film having a wavelength selection function of reflecting the exposure light on the phase grating and transmitting the alignment light. Have been devised such as forming by vapor deposition.
Furthermore, in order to further reduce the adverse effect of the correction optical element on the imaging performance, the correction optical element such as the phase grating on the pupil plane of the projection optical system should be prevented from hitting the 0th-order diffracted light of the illumination secondary light source. It is desirable to configure it, and an alignment device devised in such a way has been proposed.

[0020]

In the alignment apparatus according to the above-mentioned prior application of the present applicant, chromatic aberration in the magnification direction of the alignment light is corrected by the correction optical element, and a semiconductor element or the like to be transferred on the reticle. The alignment light to the wafer and the alignment light from the wafer side pass through a region outside (or inside) a so-called light-shielding band which is provided just outside the exposure region where the pattern exists and limits the exposure region. Is becoming

However, a frame that supports a pellicle, which is a thin film used to prevent adhesion of dust or the like to the exposure area of the reticle, that is, a pellicle frame may be attached to an area outside the light-shielding band of the reticle. FIG. 5 shows a reticle 6 to which such a pellicle frame is attached. In FIG. 5, a pellicle frame 41 is attached to the lower surface of the reticle 6, that is, the surface on the side of the projection optical system 21, and the inside of the pellicle frame 41. A pellicle 42 is attached to the lower surface of the reticle 6. In this case, the position of the reticle window, that is, the light beam 4a as the alignment light
And the position where 4b passes is, for example, the pellicle frame 41.
When the light beam 4a is away from, the incident light beam 4a directly enters the projection optical system 21, and the light beam 4b from the wafer mark passes through the reticle 1 as it is.

However, when the position of the reticle window approaches the pellicle frame 41 and the alignment light passing through the reticle window becomes a light beam 4a ', this light beam 4a' is blocked by the pellicle frame 41. That is, the conventional alignment apparatus has a disadvantage that the optical path of the alignment light is eclipsed by the pellicle frame 41 depending on the position of the reticle window. In particular, as shown in FIG. 7, the luminous fluxes 4a and 4b as alignment light are
When the telecentricity is broken by being obliquely ejected from the reticle 6 in the meridional plane that is a plane parallel to the paper surface of FIG. 7, the luminous fluxes 4a and 4b are distributed as much as the way of breaking the telecentricity. It is easily blocked by the pellicle frame 41.

In this regard, the amount of lateral chromatic aberration generated by the correction optical element, that is, the amount (distance) d of the alignment light on the reticle 6 in the meridional direction is made larger than in the case of FIG. As shown in FIG. 6, if the position of the reticle window for entering and emitting the alignment light is arranged outside the pellicle frame 41, the vignetting of the alignment light is temporarily eliminated. In FIG. 6, light fluxes 4a and 4b that pass outside the pellicle frame 41.
Indicates the alignment light passing through the reticle window corresponding to the maximum exposure area.

By the way, as a result of actual calculation, if the flying amount d determines the lattice constant of the correction optical element,
It is known that the alignment mark becomes a constant that hardly depends on the image height y of the alignment mark. Therefore, as described above, the reticle window corresponding to the maximum exposure area is skipped to the position outside the pellicle frame 41 by the skip amount d.
If the exposure area of the reticle to be used becomes smaller than this, the alignment light passing through the reticle window may be blocked by the pellicle frame 41 again as shown by a light beam 4a ′ in FIG.

Separately from this, when the skip amount d is set so that the alignment light passes just outside the light-shielding band that surrounds the maximum exposure area of the reticle 6, the following other inconvenience occurs. Generally, when a light having a long wavelength, such as a He-Ne laser beam, which is not exposed to a photosensitive material of a wafer, is passed through a projection optical system for the ultraviolet or vacuum ultraviolet region as an alignment light, the magnification is as long as the wavelength. The magnification is smaller, and this magnification is a constant that hardly depends on the image height. Therefore, the correction of the chromatic aberration of magnification by the correction optical element is performed by skipping the optical path of the alignment light in the direction in which the image height is large, but the difference (distance) between the exposure wavelength and the alignment light wavelength is The smaller the image height, the smaller the image height.

Therefore, the flying amount by the correction optical element is
If the alignment light of the maximum exposure area of the reticle 6 is set so as to pass just outside the shading zone of the maximum exposure area, when the same correction optical element is used for the alignment of a smaller exposure area, the alignment light is shaded. Passes away from. In this case, if the reticle window on which the alignment light enters and exits is formed wide in the meridional direction, the exposure area can be reduced to some extent within its width. However, there is a limit because the reticle window cannot be made large without limitation, and when the exposure area becomes small, design restrictions such as arranging the reticle window away from the light-shielding band are required, and confusion in actual use and There is a risk of making a mistake.

In view of the above point, the present invention provides an alignment apparatus using a correction optical element for correcting chromatic aberration with respect to alignment light of a projection optical system that projects a mask pattern onto a photosensitive substrate, in a predetermined area of the mask. Even if a member that blocks the alignment light is provided, it is an object of the present invention to prevent the vignetting of the alignment light by the member with a relatively simple configuration.

[0028]

The alignment apparatus according to the present invention is, for example, as shown in FIG. 2, an illumination optical system (not shown) for irradiating the mask (6) with exposure light from an illumination light source.
And a projection optical system (21) for transferring a predetermined pattern formed on the mask onto the photosensitive substrate (9) by exposure light from the illumination optical system. In an alignment apparatus for irradiating alignment marks (8b, 10) respectively formed on the photosensitive substrate with alignment light having a wavelength band different from the exposure light to perform relative alignment between the mask and the photosensitive substrate, The mask (6) and the photosensitive substrate (9)
And the alignment light (4aA, 4b) directed to the photosensitive substrate through the mask and the projection optical system.
A), the axial chromatic aberration generated by the projection optical system is canceled out, and the chromatic aberration of magnification caused by the projection optical system in the directions opposite to each other and different from each other are generated, thereby aligning the alignment light. A plurality of irradiation correction optical elements (1aA, 1bA and 1aB, 1bB in FIG. 1) for guiding the alignment marks (10) on the photosensitive substrate, and the mask and the photosensitive substrate, respectively, are arranged between the photosensitive elements. Substrate alignment mark (10)
To the alignment light (4 cA) directed to the mask via the projection optical system, the axial chromatic aberration generated by the projection optical system is canceled, and the lateral chromatic aberration caused by the projection optical system is opposite to each other. Further, it has a plurality of correction optical elements for detection (1cA and 1cB in FIG. 1) that generate different amounts of lateral chromatic aberration.

Further, the alignment apparatus according to the present invention is
Irradiation means for transmitting alignment light (4aA, 4bA) in the vicinity of the alignment mark (8b) on the mask side, and then irradiating the alignment light (4aA, 4bA) through any one of the plurality of correction optical elements for irradiation and its projection optical system. (11) and the alignment light (4 cA) irradiated on the alignment mark (10) of the photosensitive substrate (9) by any one of the correction optical elements for irradiation and reflected by the alignment mark (4 cA) for detecting a plurality of them. And a detection means (11) for receiving the alignment light from the alignment mark (8b) of the mask as well as receiving the light via any of the correction optical elements and the vicinity of the alignment mark of the mask.
The plurality of correction optical elements for irradiation and the plurality of correction optical elements for detection are selectively used according to the size of the pattern (7) of the mask (6).

In this case, the plurality of correction optical elements for irradiation and the plurality of correction optical elements for detection are arranged in parallel in the plane (3) near the pupil plane of the projection optical system (21). You may arrange.

[0031]

According to the present invention, the correction optical elements for irradiation (1aA in FIG. 1,
1bA and 1aB, 1bA), and a plurality of correction optical elements (1cA, 1cB in FIG. 1) for detection having different amounts of correction of lateral chromatic aberration are also provided. Therefore, when the exposure area of the mask (6) is large, the chromatic aberration of the alignment light incident through the window corresponding to the wide exposure area is corrected by, for example, the first correction optical element (1) for detection.
aA, 1bA) and the first detection correction optical element (1c
Correct using A). When the exposure area of the mask (6) is narrow, the chromatic aberration of the alignment light incident through the window corresponding to the narrow exposure area is corrected by, for example, the second detection correction optical element (1aB, 1bB). And the correction optical element (1cB) for the second detection.

In this case, even if a light blocking member for blocking the alignment light is provided in a predetermined region of the mask (6),
By selecting an appropriate correction optical element from the plurality of detection correction optical elements and the plurality of detection correction optical elements, the position of the window portion corresponding to the exposure area of the mask (6) is changed from the light shielding member. Can be placed in remote locations. Therefore, it is possible to prevent vignetting of alignment light due to the light shielding member with a relatively simple structure.

Further, the plurality of irradiation correction optical elements and the plurality of detection correction optical elements are arranged in parallel, and the surface (3) near the pupil plane of the projection optical system (21).
If it is arranged inside, the structure of the optical system is further simplified.

[0034]

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 projection exposure apparatus.
3 to 3, members corresponding to those in FIG. 6 are designated by the same reference numerals, and detailed description thereof will be omitted.

FIG. 1 shows the arrangement of a correction optical element such as a phase type diffraction grating formed on a transparent member arranged on the pupil plane of the projection optical system of this embodiment. A region 3 surrounded by the circumference is a region (pupil) within the pupil plane of the projection optical system and limited by the aperture stop. That is, when the numerical aperture of the projection optical system is NAp, the diameter of the region 3 is equal to the numerical aperture NAp in units of numerical aperture. The transparent member is composed of a circular parallel plane plate or the like which is slightly larger than the area 3.

Region 2 surrounded by a two-dot chain line inside region 3
Is an area irradiated with the 0th-order diffracted light (Fourier transform image) of the secondary illumination light source. When the numerical aperture on the exit side (reticle side) of the illumination optical system is NAi and the reduction magnification of the projection optical system is α, the diameter of the region 2 is equal to α · NAi in numerical aperture units. Reference numeral 1 denotes 24 correction optical elements collectively. In this example, these 24 correction optical elements 1 are arranged outside the region 2 through which the 0th order diffracted light of the illumination secondary light source passes.
This prevents strong exposure light from irradiating the correction optical element, and prevents deterioration of the imaging performance of the projection optical system.

These 24 correction optical elements 1 are arranged in the area 2
Of a first square of twelve correction optics arranged three by three along each side of the first square slightly larger than the square circumscribing to, and a second square larger than the first square. It is composed of twelve second sets of correction optical elements arranged three by three along each side. The first set of correction optical elements and the second set of correction optical elements differ from each other in the amount of lateral chromatic aberration generated with respect to the alignment light.

The first set of twelve correction optical elements arranged along each side of the first square includes a correction optical element at the center of one side of the square (for example, 1 cA) and an opposite side thereof. The correction optical elements on both sides (for example, 1aA and 1bA) are divided into three groups, and the three correction optical elements shown in the same figure belong to the same group. Similarly, the second set of twelve correction optics arranged along each side of the second square includes a correction optic at the center of one side of the square (eg, 1 cB) and correction on both sides of the opposite side. Optical elements (for example, 1aB and 1bB) are divided into three groups, and three correction optical elements shown in the same figure belong to the same group. In this way, the first set of correction optical elements is designated by the reference numeral A,
The second set of correction optical elements is designated by reference numeral B.

An example of the configuration at the time of alignment in this embodiment will be described with reference to FIG. Here, the alignment in the case of using the first set of 12 correction optical elements out of the 24 correction optical elements will be described. Therefore, in FIG. 2, for simplicity, the first set of 12 correction optical elements will be described. Show only. A transparent member 5 on which a first set of twelve correction optical elements is formed is arranged in a pupil plane region 3 of the projection optical system 21, and the projection optical system 21 causes the reticle 6 of the reticle 6 to be exposed under exposure light. The image of the pattern inside the rectangular exposure area 7 is transferred to the resist on the wafer 9. Adjacent to each of the four sides surrounding the exposure area 7 of the reticle 6 is shown in FIG.
As shown in FIG. 5, a reticle mark 8b made of a diffraction grating and an alignment mark 8 made of a window 8a are formed.

Returning to FIG. 2, one (four in total) alignment microscope 11 is prepared for each alignment mark adjacent to each side surrounding the exposure area 7. 1 of the 4 sides surrounding the exposure area 7 of the reticle 6
An example of the alignment operation when the alignment mark of the region C2 adjacent to the side is used will be described. The area C2
For the alignment mark of, the three correction optical elements 1aA to 1cA indicated by black circles on the transparent member 5 are used. Then, from the alignment microscope 11 to the reticle 6
The two alignment light beams 4aA and 4bA are emitted to the alignment mark in the upper region C2 at a predetermined crossing angle. The frequencies of the two light beams 4aA and 4bA are different from each other.

In this case, the light beams 4aA and 4bA transmitted through the window of the alignment mark in the area C2 are respectively corrected optical elements 1a on the transparent member 5 inside the projection optical system 21.
Aberration correction is performed upon incidence on A and 1bA. That is, the light beams 4aA and 4bA deflected (diffracted) by the correction optical elements 1aA and 1bA are applied to the diffraction grating wafer mark 10 in the vicinity of a predetermined shot area of the wafer 9. Then, beat interference light 4cA composed of two diffracted lights from the wafer mark 10 is incident on the correction optical element 1cA on the transparent member 5, and the beat interference light 4cA subjected to aberration correction by the correction optical element 1cA. Reticle 6
Of the alignment microscope 11 through the window of the region C2 of
Incident on. Therefore, the two correction optical elements 1aA and 1a
bA is a correction optical element for irradiation, one correction optical element 1c
A can be regarded as a correction optical element for detection. The detailed detection method of the heterodyne method is the same as the method described for the alignment device in FIG.

At this time, as shown in FIG. 2, it is assumed that the positional relationship between the reticle 6 and the wafer 9 is measured in the sagittal direction of the projection optical system 21 (direction of arrow A2). Then, the luminous fluxes 4aA and 4bA for alignment are shown in FIG.
As shown in (a), the projection optical system 21 is inclined in the meridional direction at an angle θr _1A . The angle θr _1A is
As shown in FIG. 1, the projection optical system 2 is displayed in numerical aperture units.
It is given by the following equation by the coordinate LA on the pupil plane (Fourier plane) of 1 (called the “pupil coordinate” at the distance from the center) and the reduction magnification α of the projection optical system 21. θr _1A = sin ⁻¹ (LA / α) (1)

Further, as shown in FIG. 3B, the light beams 4aA and 4bA for alignment are incident on the alignment mark of the reticle 6 while being inclined in the sagittal direction of the projection optical system 21 at angles θr ₂ in opposite directions. is doing. The tilt angle θr _{2 in} the sagittal direction is given by the following equation using the grating pitch p of the wafer mark 10 on the wafer 9 and the wavelength λ of the alignment light. θr ₂ × α = sin ⁻¹ (λ / p) (2)

Then, as shown in FIG. 1, two correction optical elements 1aA and 1bA for incident light expressed in a numerical aperture unit are used.
Let LD be the distance on the pupil plane of the following relationship. LD / 2 = sin (θr ₂ × α) (3)

The role of the second set of 12 correction optical elements is the same as that of the first set of correction optical elements,
The first group and the second group have different set values of the amount d of alignment light projection on the reticle 6 in the meridional direction. As shown in FIG. 1, the pupil coordinates of the second set of correction optical elements are LB.
And the distance on the pupil plane of the elements belonging to the same group of the second set of correction optical elements is LD. In this case, the tilt angle of the alignment light on the reticle with respect to the second set of correction optical elements 1aB and 1bB in the meridional direction is θ.
Assuming r _1B , the following equation is obtained from the correspondence with the equation (1). θr _1B = sin ⁻¹ (LB / α) (4) Therefore, the tilt angle in the meridional direction is properly used depending on which set of correction optical elements is used.

In this embodiment, it is assumed that two types of flying amounts of alignment light in the meridional direction are prepared, and there are the following methods for determining the two types of flying amounts. [First Determining Method] Generally, in the alignment marks corresponding to the maximum exposure area and the slightly smaller exposure area of the reticle 6, the interval between the light-shielding band and the pellicle frame is narrow, and the alignment light having the terencentricity in the meridional direction is broken. It is difficult to pass through. Therefore, the alignment marks corresponding to the maximum exposure area and the slightly smaller exposure area of the reticle 6 are largely shaken to the outside and arranged at appropriate positions outside the pellicle frame. The flying amount of the alignment light at this time is assumed to be dA, and this flying amount dA is realized by the first set of correction optical elements (correction optical elements 1aA, 1bA, etc. in FIG. 1).

As shown in FIG. 6, the flying amount dA
That is, that is, the first set of correction optical elements can correspond to δ from the alignment mark of the maximum exposure area of the reticle 6 until the alignment light 4a ′ comes into contact with the pellicle frame 41 due to the reduction of the exposure area. Only up to the alignment mark. In the alignment of the exposure area below this, another large skip amount dB (dA <dB) for setting the alignment light for the exposure area to be switched to a position where the alignment light for the previous maximum exposure area is passed is set. The skip amount dB is realized by the second set of correction optical elements (correction optical elements 1aB, 1bB in FIG. 1). The correction optical element of the second set is used until the alignment light is shifted inward by δ due to the reduction of the exposure area.

As the alignment mark on the reticle 6 in this case, the one shown in FIG. 4 can be considered. Figure 4
The alignment mark is composed of a reticle mark 8b made of a diffraction grating for detecting a reticle position and a window 8a through which alignment light for wafer mark detection enters and exits, but at least the window 8a has a meridional direction (m).
The width H of the direction) is set to δ or more. If such alignment marks are prepared on the reticle 6 and the correction optical elements of the first set and the second set are properly used, the light-shielding band adjacent to the pattern region, for example, can be used regardless of the size of the exposure region of the reticle 6. The alignment marks can always be placed in the same position, such as just outside.

[Second Determining Method] When the amount of collapse of the terencentricity of the alignment light in the meridional direction is small or no, there is no problem with the interference between the optical path of the alignment light and the pellicle frame. In such a case, it is desirable to make the alignment mark on the reticle 6 side between the light-shielding band outside the exposure area and the pellicle frame. This is because the area outside the pellicle frame is used to attach marks for aligning the reticle to the projection exposure apparatus, bar code marks for reticle management, and the like.

Therefore, in this method, the first flying amount dA is set so that the alignment light in the maximum exposure area of the reticle 6 passes just outside the light-shielding band. As mentioned above, if the same correction optical element is used for alignment of a smaller exposure area, the alignment light will pass away from the shading zone. Therefore, also in this method, although the meaning is different, as in the example shown in FIG. 4, the window portion 8a through which the alignment light enters and exits the reticle 6 is formed wide in the meridional direction (m direction). The reduction of the exposure area to some extent is dealt with by arranging the alignment mark (window portion 8a) of the same shape within the width of the window portion 8a in the meridional direction in contact with the light shielding band.

Then, below a certain exposure area, another second
By using the second set of correction optical elements having the skip amount dB, the skip amount dB is set so that the alignment light in the exposure area where this switching is performed can be skipped just outside the light-shielding band. To do. With such a device, even an alignment mark that is not so large can be used simply by always arranging it just outside the light-shielding band regardless of the size of the exposure area.

As described above, according to the present embodiment, even if the exposure area of the reticle is changed, for example, the optical path of the alignment light and the pellicle frame can be changed by properly using a plurality of sets of correction optical elements that respectively generate different magnification chromatic aberrations. You can prevent and from interfering. Alternatively, for example, an alignment mark which can be simply arranged outside the light-shielding band of the reticle can be used.

When the correction optical element is used so as not to interfere the optical path of the alignment light with the pellicle frame, the alignment mark on the reticle side can always be arranged outside the pellicle frame. Therefore,
The alignment microscopes arranged on the alignment marks can always observe the alignment marks in a state of being separated from each other. Therefore, the alignment microscope having the four-lens arrangement as shown in FIG. 2 cannot be arranged due to approach or collision when the shape or size of the exposure area is changed or when the alignment mark is moved. Has the advantage that it can be solved at the same time.
This advantage is the same even when there is no perifuru frame. Although the two irradiation beams and the one reflected beam pass through almost the same area on the reticle in the above-described embodiment, two irradiation beams may be emitted depending on the space on the reticle (excluding the circuit pattern area). The passing positions of the beam and one reflected beam may be positively separated. Further, although the case where the telecentricity of the alignment light is destroyed is described in the above-mentioned embodiment, the same applies when the telecentricity of the alignment light is not destroyed (when three spots are aligned on the pupil plane of the projection optical system). The effect of is obtained.

Although the number of sets of correction optical elements is two in the above embodiments, the number of sets of correction optical elements is increased when the switching amount of the alignment light is set to three stages or more. Just do it. Thereby, the range of the exposure area of the corresponding reticle can be widened or the alignment mark can be made smaller. As described above, the present invention is not limited to the above-described embodiments, and various configurations can be taken without departing from the gist of the present invention.

[0055]

According to the alignment apparatus of the present invention, it is possible to properly use a plurality of correction optical elements which respectively generate different magnification chromatic aberrations, and therefore a member for blocking the alignment light is provided in a predetermined region of the mask. However, it is possible to prevent the vignetting of the alignment light by the member with a relatively simple structure. Also, when a plurality of correction optical elements are arranged in the plane near the pupil plane of the projection optical system, the optical system configuration Is more simplified.

[Brief description of drawings]

FIG. 1 is a schematic diagram showing an arrangement of two sets of correction optical elements for alignment light in a pupil plane of a projection optical system in an embodiment of the present invention.

FIG. 2 is a perspective view showing a configuration example of an alignment system corresponding to the arrangement of the correction optical element in FIG.

3 (a) is an optical path diagram viewed from the direction of arrow A2 in FIG. 2,
(B) is an optical path diagram seen from the direction of the arrow B2 in FIG. 2, and (c) is an enlarged perspective view showing a region C2 of FIG. 2.

FIG. 4 is an enlarged plan view showing an example of an alignment mark on the reticle 6 side.

FIG. 5 is a cross-sectional view of a main part showing a state in which alignment light in which the telecentricity is lost in the meridional direction is eclipsed by the pellicle frame.

FIG. 6 is a cross-sectional view of a main part showing a case where alignment light is emitted to the outside of a pellicle frame.

FIG. 7 is a configuration diagram showing an alignment apparatus configured to correct chromatic aberration of the projection optical system with respect to alignment irradiation light and detection light by a correction optical element provided in a pupil plane of the projection optical system.

[Explanation of symbols]

1 24 correction optical elements 1aA to 1cA, 1aB to 1cB individual correction optical elements 2 area on the pupil plane of the projection optical system through which the 0th order diffracted light of the illumination secondary light source passes 3 area of the pupil surface of the projection optical system (pupil ) 4aA, 4bA Luminous flux for alignment 4cA Beat interference light 5 Transparent member such as parallel plane plate 6 Reticle 8 Reticle alignment mark 8a Window 8b Reticle mark 9 Wafer 10 Wafer mark 11 Alignment microscope 21 Projection optical system 41 Pellicle frame 42 Pellicle

Claims (2)

[Claims] 1. An illumination optical system for irradiating an exposure light from an illumination light source onto a mask and a projection optical system for transferring a predetermined pattern formed on the mask onto a photosensitive substrate by the exposure light from the illumination optical system. And a relative alignment between the mask and the photosensitive substrate by irradiating the alignment marks formed on the mask and the photosensitive substrate with alignment light having a wavelength band different from the exposure light, respectively. In an alignment device for performing various alignments, the alignment light, which is arranged between the mask and the photosensitive substrate and is directed to the photosensitive substrate through the mask and the projection optical system, is generated by the projection optical system. Along with canceling the axial chromatic aberration, the chromatic aberration of magnification caused by the projection optical system is opposite to the chromatic aberration of magnification in different directions. A plurality of irradiation correction optical elements for guiding the alignment light to the alignment marks of the photosensitive substrate, and the alignment marks of the photosensitive substrate, and the projection optical elements arranged between the mask and the photosensitive substrate. Axial chromatic aberration generated by the projection optical system with respect to the alignment light directed to the mask through the system is canceled, and chromatic aberration of magnification in directions opposite to the chromatic aberration of magnification by the projection optical system and different from each other. A plurality of correction optical elements for detection, and one of the plurality of correction optical elements for irradiation and the projection optical system after transmitting the alignment light in the vicinity of the alignment mark on the mask side. And an alignment means for irradiating the photosensitive substrate with an irradiating means for irradiating via Alignment light reflected by the alignment mark irradiated on the mask is received via any of the plurality of detection correction optical elements and the vicinity of the alignment mark of the mask, and the alignment light from the alignment mark of the mask is received. And a detection means for receiving light as well, wherein the plurality of correction optical elements for irradiation and the plurality of correction optical elements for detection are selectively used according to the size of the mask pattern. Characteristic alignment device. 2. The plurality of irradiation correction optical elements and the plurality of detection correction optical elements are arranged in parallel in a plane near a pupil plane of the projection optical system. The alignment apparatus according to claim 1.