JP2787698B2 - Alignment device and position detection device - Google Patents

Description

The present invention relates to a projection exposure apparatus for transferring a mask pattern onto a sensitive substrate (resist substrate), an alignment apparatus incorporated in a proximity exposure apparatus, and the like, and a position detection apparatus. In particular, the present invention relates to a method for detecting a diffraction grating alignment mark formed on a sensitive substrate.

[Conventional technology]

2. Description of the Related Art In recent years, a projection type exposure apparatus (stepper) has been frequently used as an apparatus for transferring a fine pattern such as a semiconductor element onto a semiconductor wafer with high resolution. Conventionally, in a stepper of this type, alignment between a reticle (synonymous with a mask) and one shot area on a wafer is performed by using a so-called alignment method. An apparatus having a TTR (through-the-reticle) type alignment system for simultaneously detecting an alignment mark formed around a region is known.

In this alignment method, both the mark on the reticle and the mark on the wafer are detected with high accuracy, the relative positional shift amount is obtained, and the reticle or wafer is finely moved so as to correct the shift amount. In general, in a projection type exposure apparatus, in order to form a reticle pattern on a wafer with high resolution, a projection optical system uses an illumination light for exposure (for example, a g-line having a wavelength of 436 nm, an i-line having a wavelength of 365 nm, or a wavelength of 248 nm).
At present, chromatic aberration is well corrected only for the KrF excimer laser beam. This means that in an alignment optical system that detects a reticle mark and a wafer mark via a projection optical system, the light for illuminating the mark is limited to the same wavelength as or very close to the wavelength of the exposure light. I do.

A resist layer is formed on the surface of the wafer in the exposure step, and a mark on the wafer is detected via the resist layer during alignment. In order to enable the formation of a pattern with higher resolution, it has been considered to adopt a multilayer resist structure or the like having a high absorption rate for exposure light and a low transmittance rate. In this case, the illumination light for alignment is attenuated before reaching the mark on the wafer, and the reflected light (specular reflection light, scattered light, diffracted light, etc.) from the mark is also attenuated. There is a problem that the mark is not recognized by the alignment optical system with a sufficient light amount, and the mark detection accuracy is reduced.

When the alignment illumination light is applied to the mark on the wafer for alignment, the resist layer at that part is naturally exposed, and when the film passes through various processes after development, the mark on the wafer is destroyed. As a result, although not essential, there arises a problem that it cannot be used for alignment at the time of overlay exposure of the next layer.

Therefore, for example, Japanese Patent Application Laid-Open No. 63-153820 discloses
Based on another wavelength alignment system of the TTR method (a method in which the irradiation light for alignment is different from the wavelength of the exposure light), one-dimensional diffraction grating marks formed on the wafer or reticle are optically detected, and the Japanese Patent Application Laid-Open No. 63-283129 proposes a method of detecting the position of a wafer or a reticle from pitch information with a high resolution (one-several to several tenths of a pitch).

Various methods have been proposed and practically used for position detection using a diffraction grating mark. In the method disclosed in Japanese Patent Application Laid-Open No. 63-283129, a diffraction grating mark is simultaneously irradiated with laser beams substantially parallel from two directions to form one-dimensional interference fringes. In this method, the position of the diffraction grating mark is specified by using the interference fringes, which is also called an interference fringe alignment method.

There are two more methods for such interference fringe alignment, and a laser beam (2
This method is a heterodyne method for giving a constant frequency difference to the present invention and a homodyne method without a frequency difference. In the homodyne method, a stationary interference fringe is formed parallel to the diffraction grating, and it is necessary to finely move the diffraction grating (object) in the pitch direction for position detection, and the position of the diffraction grating is obtained based on the interference fringe. On the other hand, in the heterodyne method, because of the frequency difference (beat frequency) between the two laser beams, the interference fringes flow at high speed at the beat frequency in the pitch direction, and the position of the diffraction grating is based on the interference fringes. It cannot be determined, but is determined solely on the basis of the temporal element (phase difference) associated with the high-speed movement of the interference fringes.

By the way, the TTR for performing the interference fringe alignment method
In an alignment system (alignment apparatus), as the mark position is changed, as described in, for example, Japanese Patent Application Laid-Open No. Sho 57-142612, the alignment objective lens and the mirror at the tip are illuminated at an afocal position. It is necessary to move in the axial direction. However, JP-A-57-1426
In Japanese Patent Publication No. 12, although the image conjugate of the alignment mark is maintained, the pupil conjugate of the system is not maintained, and in a system that passes a laser beam through an objective lens, such as an interference fringe alignment method, a light transmission system, Various disadvantages of the light receiving system occur.

Thus, a configuration of an alignment optical system that also takes into consideration the maintenance of pupil conjugate is disclosed in, for example, Japanese Patent Application Laid-Open No. 58-150924. Here, when the alignment objective moves in order to change the observation position, a part of the optical system for forming an image of the illumination light source on the entrance pupil of the objective is linked with the movement of the alignment objective in the optical axis direction. The moving method is shown.

[Problems to be solved by the invention]

However, in order to maintain the pupil conjugate in this way, the configuration in which the correction optical member moves in the illumination light transmission path must be coordinated with the objective lens, and the alignment optical system has a complicated structure. Would. In particular, when irradiating a diffraction grating mark for measurement with a light beam from two directions through an objective lens for interference fringe alignment, preserving the crossing angle of the beam and the symmetry of the incident angle is extremely important. It must be taken into account that these conditions will go out of order with the movement of, and the structure of the alignment optical system will be more complicated.

In addition, these conditions may be changed by the movement of the objective lens or the movement of the mirror at the tip, and this point must be sufficiently considered. The present invention has been made in view of the above points, and an object of the present invention is to provide an alignment apparatus and a position detection apparatus which can simplify the structure of an alignment optical system and can respond to a change in a mark position as in the related art. Aim.

[Means to solve the problem]

In the present invention, the alignment mark is made of a diffraction grating, and the objective optical system for alignment (preferably telecentric) moves in accordance with the change of the mark position. Is premised on an alignment apparatus having a photoelectric detector for photoelectrically detecting the light through an objective optical system.

A light source (a laser light source or the like) for irradiating the alignment mark with a light beam via the objective optical system, and a beam transmitting system for guiding the light beam from the light source to the objective optical system are provided. The beam transmitting system is configured such that the principal ray of the light beam from the light source passing through the pupil plane of the objective optical system is substantially parallel to the optical axis of the objective optical system, and a predetermined position on the principal ray in the pupil space. The light beam is converged at the set waist position. As a result, the light beam reaching the alignment mark is converted into a substantially parallel light beam. Further, when the 0th-order diffraction beam (specular reflection light) and the higher-order diffraction beam (for example, the 1st-order light) of the optical information generated from the alignment mark are guided to the photoelectric detector via the objective optical system, 0 Driving means for moving the objective optical system is provided within a range in which the second-order diffraction beam and the higher-order diffraction beam are separated from each other at, for example, the waist position of the light beam on the light-transmitting side or near a position substantially conjugate to the waist position. I made it.

Further, the position detecting device of the present invention is arranged in parallel in the pitch direction of the diffraction grating pattern with the first photoelectric detector interposed therebetween, and detects the two beams generated from the object different from the diffracted light. , A third photoelectric detector, the first photoelectric detector, a detector holding member for holding the second and third photoelectric detectors, and the second and third photoelectric detectors. Based on the received signal, the first photoelectric detector,
Adjusting means for adjusting the relative position of the diffracted light with respect to the optical path.

(Operation)

In the present invention, attention is paid to the fact that the size of an alignment mark handled by this type of alignment apparatus is extremely small (for example, 1 mm square or less on a reticle), and therefore, the diameter of a parallel light beam irradiating this mark is also small. Even if the light beam converges (or diverges) in the pupil space, it utilizes the fact that the numerical aperture (NA) of the beam itself can be made extremely small.

Therefore, the principle of the present invention will be described with reference to FIGS. 1 (A) and 1 (B). Figure 1 (A) is a telecentric objective lens OBJ in the alignment system, the arrangement of the mirror M _1,
The relationship between the transfer target area (shot area) SS ₁ and SS ₂ is shown.
Generally, the size of a shot area on a sensitive substrate or a mask (reticle) changes according to the size of a device to be manufactured. Shot areas SS ₁ of the center CC to the exposure center axis (optical axis) AX is assumed through the vertical and define an x-y coordinate system with the center CC as the origin, x-axis around the shot area SS _1, y-axis each diffraction grating-shaped mark RM ₁ is formed in the upper. Those on the x-axis of the marks RM _1, the grating elements are arranged in the y direction, is used to detect the position of the y direction of the shot area SS _1, marks on the y-axis is used to detect the position of the x-direction.

Now, the optical axis AXa of the objective lens OBJ has become a vertical on marks RM _1, are bent horizontally by the mirror M _1. While the objective lens OBJ and the mirror M ₁ is fixed to the positional relationship shown, it can be moved in the arrow A direction together. Mark R on the rear focal plane of the telecentric objective lens OBJ
Because it matches the surface of the M _1, the front side of the objective lens OBJ (the mirror M ₁ side) is in the afocal system. Here due to interference fringes alignment, irradiating two beams LB _1, LB ₂ to mark RM ₁ through the objective lens OBJ. Beam LB _1, L
B ₂ was supplied from the beam transmission system so as to be located in a point-symmetric relationship with respect to the optical axis AXa, and was collected as spot lights SP ₁ and SP ₂ at a beam waist position BW previously set on the apparatus. Thereafter, the light enters the objective lens OBJ with a small divergence angle. Then, beams LB ₁ and LB ₂ emitted from the objective lens OBJ
Is substantially parallel beam which intersects at a predetermined angle on marks RM _1. The lattice element array direction of the mark RM ₁ is assumed to be y-direction, each of the two principal rays of the beams LB _1, LB ₂ emerging from the objective lens OBJ, symmetrical with respect to the plane including the x-axis and the axis AX slope Located in.

In FIG. 1 (A), EP ₁ is a front focal plane (pupil plane) EP ₁ of the objective lens OBJ, and when the waist position BW and the pupil plane EP ₁ are largely displaced in the direction of the optical axis AXa as shown in FIG. The parallelism of the beams LB ₁ and LB ₂ reaching the mark RM ₁ is degraded according to the amount of the shift, resulting in convergent light or divergent light.

Now, the shot area SS ₁ is slightly larger than the shot area S.
When the change in S _2, mark RM ₁ even when the position is changed, must observe the mark RM _2. In this case, as the optical axis AXa of the objective lens OBJ is deviated from the mark RM ₁ to mark RM _2,
The mirror M ₁ and the objective lens OBJ are moved in the direction of arrow A.
As a result, the distance between the waist positions BW and pupil plane EP ₁ changes.

As will be apparent from the technique disclosed in JP-A 63-283129 JP presented above, BTL (interference beat light of first-order diffracted light each other) the diffracted light from the mark RM _1, or mark RM ₂ the objective lens OBJ Returns along the optical axis AXa, passes through the center of the pupil plane EP ₁ and the waist position BW, and reaches the photoelectric detector. The photoelectric detector needs to travel on the axis and receive only the diffracted light BTL.
Zero-order light (beams LB ₁ , LB ₂₎ other than the diffracted light BTL is applied to the photoelectric detector.
(Specularly reflected light component) of the light, it becomes noise, which is extremely inconvenient. Therefore, it is necessary to provide a photoelectric detector or a spatial filter in a space where the zero-order light and the diffracted light BTL are completely separated. Generally, the photoelectric detector is fixedly arranged on the device side,
Objective lens OBJ, the distance from the photoelectric detector (or spatial filter) to the pupil plane EP ₁ is changed by the movement of the mirror M _1. Therefore, when the observation position of the objective lens OBJ is changed in a large range, the diffracted light BTL and the zero-order light do not separate at the position of the photoelectric detector (or the spatial filter) due to defocusing but partially overlap. Sometimes.

In order to avoid the above disadvantages, the present invention provides an objective lens OB
The movement range of J was restricted under certain conditions.

This will be briefly described with reference to FIG. 1 (B). In FIG. 1 (B), the focal length of the objective lens OBJ is f, and the chief ray of the beam LB ₁ (LB ₂ is omitted) is LA.
₁ , LA ₂ , the incident angles of the principal rays LA ₁ and LA ₂ on the mark RM ₁ are θ
r, a mark RM ₁ in a plane including the principal rays LA ₁ and LA ₂ and the optical axis AXa.
Dr, the width of the beam LB ₁ (LB ₂ ) reaching the
NA of the beam LB ₁ beam LB ₁ towards the spacing waist positions BW and pupil plane EP ₁ of (LB ₂₎ ΔZ, the waist positions BW (LB ₂₎
Θb is the angle related to (numerical aperture) and the optical axis in pupil space
The distance from the optical axis AXa of the principal rays LA ₁ and LA _{2 that} are parallel to AXa
Da. As an example, a spatial filter that extracts only the diffracted light BTL is a waist position BW in FIG.
It is assumed that a plane almost conjugate to the above is created by a relay system and arranged on the conjugate plane.

Therefore, whether the 0 order light L ₀ on the spatial filter and the diffracted light BTL is to know whether or not separated, the zero-order light L ₀ at waist positions BW and diffracted light BTL is separated Just think.

First, to return the diffracted light (primary light) BTL to the axis,
It is necessary to satisfy the condition of Expression (1).

In the equation (1), n is the order (1, 2,...), And λ is the beam L
The wavelengths of B ₁ and LB ₂ , and Pr is the pitch of the diffraction grating of the mark RM ₁ . Further, the distance Da is represented by Expression (2).

Da ≒ f · sin θr (2) Further, the numerical aperture (sin θb) of the light transmission beam LB ₁ is the diffraction light
BTL, 0-order light for each well of L _0, if assumed to be substantially saved, the following equation (3) holds.

Dr / 2 = f · tanθb ...... (3) approximately equal to the width of the irradiation width Dr marked RM _1, the angle θr
Assuming that a small, diffracted light BTL, 0 order light L ₀ is made to the beam waist at the waist position BW and symmetrical plane BW 'across the pupil plane EP ₁ in geometrical optics. So West Position BW
The radius of the beam of diffracted light BTL, 0 order light L ₀ and when together in the DB in the distance between the waist position BW and surface BW 'is 2-
Since ΔZ is satisfied, the relationship of Expression (4) holds.

DB = 2 · ΔZ · tan θb (4) Finally, the condition for separating the zero-order light L ₀ and the diffracted light BTL is expressed by equation (5).

DaDB (5) As described above, the expression (6) is obtained by substituting the expression (3) into the tan θb of the expression (4) in order to collect the conditions of the expressions (1) to (5).

DB = ΔZ · (Dr / f) (6) On the other hand, by substituting equation (1) into equation (2), equation (7) is obtained.

Therefore, substituting equations (6) and (7) into equation (5) gives n
= 1, the condition of equation (8) is obtained.

As is apparent from the equation (8), when the incident angles θr of the _two beams LB ₁ and LB ₂ are set so as to return the diffracted light (first-order light) BTL to the axis, the diffracted light BTL and the zero-order shift amount from the waist position BW of the pupil plane EP ₁ for separating the light L ₀ delta
Z can be uniquely determined only by constants predetermined in design, λ, f, Dr, and Pr.

Incidentally, contrary to the case of FIG. 1 (B), so likewise there are allowable range of ΔZ may pupil EP ₁ is located on the left side of the waist position BW, spatial filter West position BW plane conjugate If you put eventually movable range of the arrow a direction between the objective lens OBJ and the mirror M ₁ is Only be obtained.

Here, a practical numerical example in the case of a 1/5 reduction projection exposure apparatus (stepper) will be described. Shot area SS ₁ , SS ₂
Is a pattern area on the reticle, and the grid pitch Pr of the marks RM ₁ and RM ₂ is 20 μm (4 μm on the wafer).
= 1,000μm (200μm on wafer), objective lens OBJ
The focal length f is 50 mm and the wavelength λ is 0.633 μ of a He-Ne laser.
Assuming that m, the range ΔZ is ΔZ79 mm from Expression (8). However, that there is diffraction development of the beam, the zero-order light L ₀ ,
The diffracted light BTL can be actually moved due to limitations such as the preservability of the numerical aperture (θb), the width Dr of the irradiation area does not always match the mark width, and the setting accuracy of the spatial filter. The range is smaller than the above theoretical value and is about 50 to 60 mm. However, in the case of a general reticle, the movement amount of the mark position is x in FIG.
A few tens of millimeters on the axis is sufficient, and it turns out that it is still a practical value.

As is clear from equation (8), the movable range increases in proportion to the square of f, so that the longer the focal length of the objective lens OBJ, the more advantageous.

The angle θb for determining the numerical aperture (sin θb) of the two beams LB ₁ and LB _{2 in} the pupil space is approximately given by the following equation, considering that the pupil plane EP ₁ and the wafer position BW coincide with each other. Is represented by

Then, an image space (marks RM ₁ , RM ₂ from the objective lens OBJ) generated by the deviation of the pupil plane EP ₁ from the waist position BW.
The deviation from the parallelism of the beams LB ₁ and LB ₂ during the period (2) is expressed by the following equation, where the numerical aperture of the beams LB ₁ and LB _{2 in} the image space is sin θi.

In the case of the numerical values exemplified above, Δ around the waist position BW
Assuming that Z = ± 25mm (50mm for the entire movable range),
The tan θb is extremely small, about 0.01, from the equation (9). Therefore, when ΔZ changes from 0 to ± 25 mm, tan θi becomes 0 to 0.0
It varies between 05. The change amount of tan θi is proportional to ΔZ,
Since it is inversely proportional to f, if f is lengthened, the change in tan θi can be practically ignored, and the beams reaching the marks RM ₁ and RM ₂
LB ₁ and LB ₂ may be regarded as almost parallel light beams. Therefore, as discussed above, the zero-order light L ₀ from the mark RM and the diffracted light BT
The condition may be set by giving priority to the separation in the pupil space from L. Also, even when the spatial filter is not exactly conjugated with the waist position BW, the condition of the above equation (8) is applied as it is. For example, assuming that the conjugate plane of the spatial filter coincides with the plane BW 'in FIG.
Since the zero-order light L ₀ and the diffracted light BTL have a beam waist at the surface BW ′, the objective lens OBJ (pupil plane EP ₁ ) can be moved by ± ΔZ around the state shown in FIG. 1B.

Further, in the position detecting device of the present invention, the adjusting means may control the first and the second signals based on the signals obtained by the second and third photoelectric detectors.
Since the relative position between the photoelectric detector and the optical path of the diffracted light is adjusted, the position of the diffracted light received by the first photoelectric detector can be easily adjusted.

〔Example〕

Next, a first embodiment of the present invention adopting a configuration equivalent to that of the projection exposure apparatus disclosed in JP-A-63-283129 will be described.
This will be described with reference to FIGS. 3, 3, 4 and 5. FIG. Second
In the figure, a reticle R having a predetermined circuit pattern area SS and a diffraction grating mark RM for alignment is held on a reticle stage 2 that can move two-dimensionally. Reticle R
Each pattern above is a telecentric projection lens PL on both sides
Thereby, an image is formed on the wafer W under the exposure light (first illumination light). However, this projection lens PL illumination light wavelength for exposure (g-line, i-line, KrF excimer laser, etc.) IL ₁ are satisfactorily chromatic aberration corrected for, conjugated with each other and the reticle R and the wafer W with respect to the wavelength for the exposure It is arranged so that it becomes. A diffraction grating mark WM similar to the grating mark formed on the reticle R is also formed on the wafer W.
Now, the wafer W is sucked onto the stage 5 which moves two-dimensionally in a step-and-repeat manner, and when the transfer exposure of the reticle R to one shot area SA on the wafer W is completed, the wafer W is stepped to the next shot position. A part of the reticle stage 2 has a laser light wave interference type length measuring device (hereinafter, referred to as an interferometer) 43 for detecting the positions of the reticle R in the x direction, the y direction, and the rotation (θ) direction in the horizontal plane.
The movable mirror 6 that reflects the laser beam from the camera is fixed. The interferometer 43 has three laser beams for length measurement for independently detecting the positions in the x, y, and θ directions, but some of them are not shown here for the sake of simplicity. . The movement stroke of the reticle stage 2 is several millimeters or less, and the detection resolution of the interferometer 43 is, for example, 0.
It is set to about 01 μm. On the other hand, the wafer stage 5
A movable mirror 7 that reflects the laser beam from the interferometer 45 for detecting the position of the wafer W in the x direction and the y direction in the horizontal plane is fixed to a part of the mirror. The interferometer 45 also has two laser beams for length measurement for independently detecting the positions in the x and y directions, but some of them are not shown here for simplicity. The reticle stage 2 is driven in the x, y, and θ directions by a drive motor 42,
The two-dimensional movement of the wafer stage 5 is performed by a drive motor 46.

By the way, the illumination system for exposure is a mercury lamp 30, an elliptical mirror
31, input lens group including condenser lens and interference filter
32, optical integrator (fly-eye lens) 3
3, mirror 34, relay lens group 36A, 36B, variable reticle blind 37, mirror 38, main condenser lens
39, a dichroic mirror 22, and the like.
The dichroic mirror 22 is 45 degrees obliquely above the reticle R, it is transmitted through vertically downwards exposure light IL ₁ from the condenser lens 39, uniformly irradiating the reticle R. This dichroic mirror 22 is 9 mm for the wavelength of the exposure light IL _1.
It has a transmittance of 0% or more, and has a reflectance of 50% or more with respect to the wavelength of the illumination light for alignment (longer wavelength than the image-forming exposure light). The reticle blind 37 functions as an illumination field stop that defines an illumination area on the reticle R, and includes a lens group.
The reticle R is conjugated to the reticle R by a condenser lens 36B.

Next, the alignment system of this embodiment will be described. The illumination light (second illumination light) LB for alignment is a laser light source 10
Drive signal S having a frequency difference Δf
After entering a frequency shifter 12 including two acousto-optic elements modulated by F ₁ and SF ₂ , a polarization beam splitter, etc., both polarization components (P polarization and S polarization) are included, and a frequency difference Δf is given. The two beams LB ₁ and LB ₂ are converted. The beams LB ₁ and LB ₂ have chief rays symmetrically decentered about the optical axis AXa. In FIG. 2, the two beams LB ₁ and LB ₂ that have emitted the frequency shifter 12 are the optical axes AXa Are located in a plane perpendicular to the plane of the drawing including

Two beams LB ₁ and LB ₂ emitted from the frequency shifter 12
A part of the beam is reflected by the beam splitter 14, enters the reference signal creation unit 16, and is transmitted through the beam splitter 14.
LB ₁ and LB ₂ pass through relay system 17A and beam splitter 20,
After being reflected by the mirror M _1, 2 focal element by birefringence material
The light is projected from a horizontal direction onto a dichroic mirror 22 via a 21B and a telecentric objective lens 21A, and is reflected therefrom to irradiate a predetermined area including a mark RM of the reticle R. Bifocal element 21B is plano-convex and plano-concave lens convex glass birefringent material consists of those bonded to the concave surface, is arranged in the pupil plane EP ₁ of the objective lens 21A. beam
If the chief ray of LB ₁ is LA ₁ and the chief ray of beam LB ₂ is LA ₂ ,
In this embodiment, the principal rays LA ₁ and LA ₂ are parallel to the optical axis AXa between the frequency shifter 12 and the relay system 17A and between the relay system 17A and the bifocal element 21B. The beams LB ₁ and LB ₂ are parallel beams inside the relay system 17A, and the principal ray L
A ₁ and LA ₂ are image conjugate planes (reticle R,
Or a plane conjugate with the wafer W) IP ′, and then spots SP ₁ , SP at the beam waist position BW by the relay system 17A.
₂ is formed. Therefore, the bifocal element 21B and the objective lens 21A
The beams LB ₁ and LB ₂ that have passed through each become substantially parallel light beams, and one polarized (for example, P-polarized) component is
Cross on the mark RM, and the other polarization component is
Cross at a. This surface 26a is conjugate to the wafer WM under the wavelength of the beam LB, and corresponds to the chromatic aberration of the projection lens PL. The rear focal plane of the objective lens 21A matches one of the plane 26a and the pattern plane of the reticle R under the wavelength of the alignment illumination light (beam LB).

When the beam LB ₁ is incident on the bifocal element 21B, different refractive indexes are given to the S-polarized light and the P-polarized light by the action of the birefringent substance.
The beam is split into a P-polarized beam LB _1P and an S-polarized beam LB _1S and emitted obliquely from the objective lens 21A. Similarly, beam LB ₂
By the action of the bifocal element 21B, the beam is split into a P-polarized beam LB _2P and an S-polarized beam LB _2S and is emitted obliquely from the objective lens 21A.

The chief rays of the beams LB _1S and LB _2S cross each other at the surface 26a, and the beams LB _1P and LB _2P cross each other at the surface of the mark RM of the reticle R. Optical axes AXa of beams LB ₁ (LB _1P , LB _1S ) and LB ₂ (LB _2P , LB _2S ) emitted from objective lens 21A
Is set symmetrically, and the inclination angle is uniquely determined by the relationship with the grating constant of the diffraction grating mark WM provided on the wafer W or the mark RM on the reticle R. LB _1S, LB _2S intersecting a plane 26a is transmitted through the transparent window P ₀ of the reticle R enters the projection lens PL, again crosses over the mark WM.

On the other hand, the inclinations of the beams LB _1P and LB _2P are also different from the window P _{0 of the} reticle R.
The beams LB _1P and LB _2P intersect on the mark RM uniquely determined by a diffraction grating mark RM provided in a part of the inside.

Thus, when the beams LB _1S and LB _2S are irradiated on the wafer mark WM from two directions, the frequency difference Δf between the beams LB _1S and LB _2S
For this reason, one-dimensional interference fringes are formed at a predetermined pitch on the diffraction grating mark WM, and the interference fringes flow at a speed corresponding to the frequency difference Δf in a direction orthogonal to the direction in which the fringes extend. Also at the mark RM in the window P _{0 of the} reticle R, the beam LB _1P
Similar interference fringes are created by LB _2P and flow with a frequency difference Δf.

Here, the pitch of the interference fringes on the reticle R by the beams LB _1P and LB _2P is determined so as to be exactly half the grating pitch of the mark RM in the y direction. Therefore, the first-order diffracted light generated from the mark RM by the irradiation of the beam LB _1P and the beam LB _1P
The first-order diffracted light generated from the mark RM by the irradiation of the LB _2P is coaxially synthesized, and the optical beat BTL changes in the amount of light at the frequency Δf.
And the objective lens 21A and the bifocal element 21 along the optical axis AXa.
B, and to reverse through the mirror M _1, the beam splitter 20
After being reflected by, it enters the measurement signal creation unit 24. Then, a signal DSr corresponding to the optical beat BTL from the mark RM is generated.

The beams LB _1S and LB _2S transmitted through the window P _{0 of the} reticle R
Passes through the projection lens PL and is condensed to a beam waist in the pupil EP, and then intersects as a parallel beam on the wafer W. This is because the surface 26a of the space above the reticle R becomes conjugate with the wafer W at the wavelengths of the beams LB _1S and LB _2S .

Then, a one-dimensional interference fringe is formed on the wafer mark WM, and flows with a frequency difference Δf. Also in this case, the pitch of the interference fringes in the y direction is set to exactly half the grid pitch of the mark WM. This means that the lattice pitch of the mark RM of the reticle R and the lattice pitch of the mark WM of the wafer W are
It means that they have a relationship of projection magnification.

From the wafer mark MW, the first-order diffracted light generated by the irradiation of the beam LB _1S and the first-order diffracted light generated by the irradiation of the beam LB _2S
The next-order diffracted light is coaxially combined and reversely travels through the center of the pupil EP of the projection lens PL. Since the two first-order diffracted lights have the same polarization component, they interfere with each other, become optical beats, pass through the transparent window of the reticle R, and travel along the optical axis AXa with the objective lens 21A, the bifocal element 21B, and the mirror. M ₁ , and incident on the measurement signal creation unit 24 via the beam splitter 20,
Then, a signal DSw corresponding to the optical beat BTL from the mark WM is generated.

The phase detection system 40 detects a phase difference between the reference signal DR created by the reference signal creation unit 16 and each of the signals DSr and DSw created by the measurement signal creation unit 24, and detects a difference between the mark RM and the mark WM. The relative displacement is measured with high accuracy within a range of ± 1/2 of the grating pitch.

The control system 41 controls the servo system 44, which drives both the interference systems 43 and 45, the drive motors 42 and 46, and the phase detection system 40 as a whole.

Incidentally, the objective lens 21A of the alignment system in Figure 2, bifocal element 21B, and the mirror M ₁ is integrally fixed to the holding hardware 62 that moves along the arrow A in the vertical direction by a drive control system 60 ing. Due to the movement of the hardware 62, the observation position of the objective lens 21A can be freely changed on the reticle R in the radial line passing through the center CC.

FIG. 2 shows only an alignment system for detecting one mark, but it is desirable to adopt a configuration as shown in FIG. 3, for example, in order to actually detect two or more marks. FIG. 3 is a perspective view showing a practical arrangement of the reticle R, the dichroic mirror 22, and the objective lenses 21x, 21y, 21θ of the three alignment systems. Here, the optical axis AX passes through the center CC of the reticle R, and the pattern area SS separated from the center CC in the x direction and the y direction, respectively.
The diffraction grating marks RM _x , RM _y , and RM _θ formed in the transparent window are provided at three locations in the periphery, and a frame-shaped light-shielding band LSB having a constant width surrounding the pattern area SS is provided outside each window. Is formed. Reticle alignment marks RS _x , RS _y , and RS _θ are formed at three positions outside the light-shielding band LSB.

Now, the objective lenses 21x and 21 of the three alignment systems
Mirrors M _1x , M _1y , and M _1θ are disposed integrally with the objective lens at the tips of y and 21θ, and each of the objective lenses 21x, 21y and 21θ is
The optical axes AX _ax , AX _ay , and AX _aθ are arranged in parallel with the XZ plane.

Then, the objective lens 2 for detecting the mark RM _x for the x direction
1x and the mirror M _1x move in parallel in the z direction (direction along the optical axis AX) and the x direction, and a set of an objective lens 21y and a mirror M _1y for detecting the mark RM _y for the y direction, and y the set of the objective lens 21θ a mirror M _1Shita for detecting the mark RM _theta for direction, configured to translate independently in the x and z directions. In FIG. 3, the objective lenses 21x, 21y, 21
Although only θ is shown, a bifocal element is also provided integrally, and the bifocal element and thereafter (the side of the alignment illumination light source) are afocal. In addition, 3
Two mirrors M _1x , M _1y and M _1θ are also dichroic mirrors 22
Are arranged so as not to enter the space on the lower surface of the. Thus, the mirror M is provided at the tip of each objective lens 21x, 21y, 21θ.
_{When 1x} , M _1y , and M _1θ are provided, the size and change of the pattern area SS of the reticle R, that is, each mark RM _x , RM _y ,
There is an advantage that the amount of movement of the observation position by each of the objective lenses 21x, 21y, and 21θ does not spatially interfere and can be relatively large, in accordance with the position change of the RM _θ with _{respect to} the reticle center CC.

Figure 4 shows the positional relationship between the mark WM on their mark RM and the wafer W, the window P ₀ is provided in a light-shielding band LSB of the reticle R, mark RM to about half the area of the window P ₀ is formed Is done. The remaining half of the region of the window P _0, the wafer mark WM is located. FIG. 4 shows a state in which the two marks RM and WM are aligned, and a line passing through the center of the marks RM and WM in the pitch direction is substantially directed to the center CC of the reticle R. Therefore,
The measurement direction of the marks WM and RM is a direction orthogonal to the line. The area of the window P ₀ the mark WM is located, although larger with respect to the measuring direction, which two beams LB _1S toward the wafer W, LB _2S is because spread in terms of the window P _0.

Next, referring to FIG. 5, the two-beam frequency shifter 12,
A series of configurations of the reference signal creation unit 16 and the measurement signal creation unit 24 will be described. Parallel beam LB from the laser light source 10 as shown in FIG. 5 (orthogonal linearly polarized light) is divided into a beam LB _S beam LB _P and S-polarized component of the P-polarized light component by the polarizing beam splitter 71. Beam LB _P is reflected by the mirror 72, is incident on the AOM (acousto-optic modulator) 73, the beam LB _S is reflected by the polarization beam splitter 71, is incident on AOM74. AOM73 is frequency
It is driven by a high-frequency signal SF ₁ of f _1, and outputs a primary light that is polarized by the diffraction angle determined by the frequency f ₁ as the beam LB _P. AOM74 is driven at the frequency f ₂ (f ₂ = f ₁ -.DELTA.f) of the high-frequency signal SF _2, and outputs a primary light that is polarized by the diffraction angle determined by the frequency f ₂ as beam LB _S. Each AOM
0-order light D ₀ is slit plate 77A disposed in a suitable position of the incident beam relative, it is shielded by 77B. Note that the relationship between the drive frequencies f ₁ and f ₂ and the difference frequency Δf is f ₁ >> Δ
It is desirable that f, f ₂ >> Δf, and the upper limit of Δf is determined by the responsiveness of the photoelectric detector that receives each optical beat.
Now, the beam LB _S from the AOM 74 is reflected by the mirror 75 and is incident on the polarization beam splitter 76 via the lens 78B.
Beam LB _P lens from a direction perpendicular to that of the 73 78A
Through the polarization beam splitter 76. Here the polarizing beam splitter 76, two beams LB _P, rather than safely coaxially synthesized principal ray LA _1, LA ₂ of LB _S, parallel to synthesize each other so as spaced by some amount. In the case of this embodiment, this interval is set to 2 for irradiating the reticle and the wafer.
This defines the intersection angle θ between the _two beams LB ₁ and LB ₂ .
The lens 78A, 78B are beam LB _P, to converge the beam waist of LB _S in a suitable surface.

Two beams LB _P, LB _S is having a frequency difference of Δf alone, in this embodiment, by irradiating the two P-polarized beam having a frequency difference on the reticle, having a frequency difference on the wafer 2
It is necessary to irradiate the book with an S-polarized beam. That is, from two and one S-polarized beam LB _S and one of P polarized light beam LB _P of the frequency f ₂ of the frequency f _1, 4 beams LB _1P, LB _2P, L
It is necessary to produce B _1S and LB _2S . The system for that is the fifth
Half-wave plate 117 in the figure, two polarization beam splitters 11
8, 119 and two mirrors 120, 121.

Two beams LB _S, LB _P is polarization direction 45 ° half-wave plate 11
7. The P-polarized beam LB _1P having the frequency f ₁ and the S-polarized beam LB _1S are coaxially combined by the polarizing beam splitters 118, 119, etc., to become the beam LB ₁ , and the P-polarized beam L having the frequency f ₂ (f ₁ −Δf).
B _2P and S polarized light beam LB _2S are combined coaxially emits a polarizing beam splitter 119 becomes beam LB _2.

As shown in FIG. 5, the 1/2 wave plate 117 S-polarized light beam LB _S (frequency f ₂₎ is incident, its polarization direction is rotated by 45 °. For this reason, the polarization beam splitter 118 splits the beam into a P-polarized beam LB _2P and an S-polarized beam LB _{2S vectorwise} . P-polarized beam through half-wave plate 117
Since the polarization direction of LB _P (frequency f ₁ ) is similarly rotated by 45 °, the beam splitter 118 splits the beam into a P-polarized beam LB _1P and an S-polarized beam LB _1S . The beams are combined by the polarization beam splitter 119 via right-angle prisms 120 and 121 having a metal reflecting surface, and are emitted again as two parallel beams LB ₁ and LB ₂ . The two beams LB ₁ and LB ₂ are located symmetrically with respect to the optical axis AXa of the alignment optical system. The system up to this point is the configuration of the two-beam frequency shifter 12.

Next, two beams LB ₁ (LB _1S , LB _1P ) and LB ₂ (L
B _2S , LB _2P ) are each divided into two by a beam splitter 14, one of which is a reference signal generation unit composed of a lens 160, a mirror 161, a pass-type reference grating plate 62, a spatial filter 163, and a light receiving element 19. It is incident on 16. Here, the front focal plane of the lens 160 is arranged so as to coincide with the rear focal plane of the lenses 78A and 78B.
B ₁ and LB ₂ become substantially parallel light beams and are irradiated at a predetermined intersection angle. For this reason, one-dimensional interference fringes flowing in the grating pitch direction are formed on the reference grating plate 62, and the optical beat generated on the axis is generated.
BTr (parallel light beam) passes through the aperture 163 to the light receiving element 19
Received. The zero-order light L ₀ ′ from the reference grating plate 162 is shielded by the aperture 163. Output signal D of light receiving element 19
R is an AC signal having a frequency equal to the frequency difference Δf,
Used as a reference signal. In the present embodiment, the relative alignment between the reticle and the wafer is performed by separately measuring the positional deviation of the reticle mark RM with respect to the reference lattice plate 162 and the positional deviation of the wafer mark WM with respect to the reference lattice plate 162.

On the other hand, two beams reflected by the beam splitter 14
Although LB ₁ and LB ₂ pass through the relay system 17A as shown in FIG. ₂ , the principal rays of the _two beams LB ₁ and LB ₂ intersect at the image conjugate plane IP ′ therebetween. And a reticle R is provided on the image conjugate plane IP '.
A field stop 18 for limiting an intersection illumination area (width Dr in FIG. 1) of the two beams LB ₁ and LB ₂ on (or the wafer W) is arranged. Each of the beams LB ₁ and LB ₂ having passed through the relay system 17A passes through the beam splitter 20, and is converged to the minimum beam diameter at the waist position BW in FIG.

Now, the reflected light from the mark RM of the reticle R or the wafer W mark WM returns to the beam splitter 20 via the objective lens 21A and is reflected there. 0 order light L ₀ of the reflected light is completely returned to the coaxial and the principal ray of the beam LB _1, LB _2, 1 principal rays of diffracted light BTL returns to the optical axis AXa coaxial alignment system. These zero-order light L ₀ and diffracted light BTL are:
The light is reflected by the beam splitter 20 and reaches the spatial filter 23 through the afocal expansion relay systems 240A and 240B. Spatial filter 23 cuts the 0 order light L _0, to extract diffracted light BTL, is arranged substantially conjugate with waist position BW. At the position of the spatial filter 23, as the diffracted light BTL, the one from the reticle mark RM and the one from the wafer mark WM coaxially exist in mutually complementary polarization states. Therefore, in order to avoid the following description from being complicated, the diffracted light from the reticle mark RM is set to 104, and the diffracted light from the wafer mark WM is set to 105. Afocal expansion relay system 2
40A and 240B are diffracted lights 104 and 10 reaching the spatial filter 23.
And the principal ray of 5, the space between it and the two respective principal rays of the 0 order light L _0, but is intended to facilitate the extraction of the diffracted light 104 and 105, there is no need to expand servant必. Extracted diffracted light 10
4 and 105 are reflected by a mirror 241 and pass through an imaging lens (inverse Fourier transform lens) 242 to reach a polarization beam splitter 243 where the diffracted light 105 of the S-polarized component is reflected and arranged on the image conjugate plane. The light is received by the light receiving element 25A via the field stop 245A and the condenser lens 244A. On the other hand, the P-polarized diffracted light 104 transmitted through the polarization beam splitter 243 reaches the light receiving element 25B via the field stop 245B and the condenser lens 244B arranged on the image conjugate plane. The field stop 245A is the mark WM in FIG.
245B with an aperture that matches the size and position of the
Has an opening corresponding to the size and position of the mark RM in FIG. Here, since the optical path from the imaging lens 242 to the relay system 240B and the optical path from the relay system 240A to the objective lens 21A are afocal systems, the objective lens 21A (gold 6
Even if 2) is moved to change the observation position, the conjugate relationship between the field stop 245 and the wafer mark WM and the conjugate relationship between the field stop 245B and the reticle mark RM are both maintained. The light receiving element 25A outputs a measurement signal DSw having a frequency of the frequency difference Δf, and the light receiving element 25B also outputs a measurement signal DSr having a frequency of the frequency difference Δf.

Next, the conjugate relationship between the beams LB ₁ and LB ₂ and the positional relationship of the beam waist in FIG. 5 will be described. First, the relay system 17A
Inside, the beams LB ₁ and LB ₂ are both parallel light beams, and the optical path from the relay system 17A to the lenses 78A and 78B or the lens 160
The optical path from to the lenses 78A and 78B is an afocal system, and the principal rays of the beams LB ₁ and LB ₂ are parallel to the optical axis AXa.
The waist position BW a conjugate position between the mirrors M ₁ and the beam splitter 20 in FIG. 2, the lens 78A, present in the appropriate position between the 78B and the relay system 17A (or lens 160), Therefore, the beams LB ₁ and LB ₂ fall within the minimum beam diameter.

Further, the lenses 78A, 78B and one of the lenses having the relay diameter 17A form a conjugate position with the field stop 18 on the AOM73, 74 side. In the present embodiment, the conjugate position is set to AOM73, 74.
And the diffraction center point. Therefore, since each diffraction center point of AOM73, 74 has the bifocal element 21B interposed,
It becomes conjugate with each of the reticle mark RM and the wafer mark WM.

Now, in this embodiment the diffracted light BTL (104 and 105) passing through the spatial filter 23 of FIG. 5, is set the objective lens 21A (fittings 62) moving stroke as 0-order light L ₀ is not mixed. The most preferable condition is that when the hardware 62 is located at the center of the movement stroke for observation,
To match the pupil plane EP ₁ of A in waist position BW, i.e. the pupil plane EP ₁ is to spatial filter 23 and conjugate. In this state, the beam waist of the diffracted light BTL (104, 105) is determined to pass through the center of the opening of the spatial filter 23.

Of course, with the hardware 62 is deviated from the center position in the stroke, but the pupil plane EP ₁ and the spatial filter 23 may be determined to be conjugated, if the spatial filter 23
Opening size, afocal magnifying relay system 240A, note the magnification or the like of 240B, so that there is no contamination of the zero-order light L ₀ in the full stroke. The specific condition setting is the same as the equation (8) described above with reference to FIG. 1 (B).

By the way, as shown in FIG. 3, when the number of alignment systems is two or more, a set of the beam splitter 14, the reference signal creation unit 16, the relay system 17A, the beam splitter 20, and the measurement signal creation unit 24 is What is necessary is just to provide corresponding to objective lens 21x, 21y, and 21 (theta). However, in that case, the reference grid plate 162
Is provided separately for each of the plurality of marks, and there is a possibility that uncertainty remains in the uniformity of the standard. At that time, three sets of systems after the relay system 17A in FIG. 5 are provided corresponding to each of the objective lenses 21x, 21y and 21θ, and these systems are further provided with three beams LB ₁ and LB ₂ from the beam splitter 14. The reference signal DR obtained by the single reference grating plate 162 may be used as a reference when measuring the marks RM _x , RM _y , RM _{θ, and the} like.

Further, in the case of the present embodiment, since the die-by-die alignment method is mainly employed, a phase detection system for directly comparing the phase difference between the measurement signals DSr and DSw may be provided. Is unnecessary.

As described above, in the first embodiment of the present invention, the diffracted light 104 from the reticle mark RM and the diffracted light 105 from the wafer mark WM
Although the polarization was different from each other, the interference fringes formed on the reticle mark RM were P-polarized (two beams LB
Limited to _1P interference and LB _2P), because the interference pattern created on the wafer mark WM is limited to S-polarized light (interference of two beams LB _1S and LB _2S). However, depending on the magnitude of the amount of axial chromatic aberration of the projection lens PL, etc., interference fringes due to both polarization components may be simultaneously applied to the reticle mark RM and the wafer mark WM, and the diffraction lights 104 and 105 Crosstalk occurs because polarization separation is difficult. This crosstalk gives distortion to the waveforms of the measurement signals DSw and DSr, and becomes an error when detecting a phase difference.

This will be described with reference to FIG. FIG. 6 shows the state of incidence of each beam near the reticle R.
The beams irradiating the mark RM are P-polarized beams LB _1P and LB _2P having a frequency difference Δf.
Are S-polarized beams LB _1S and LB _2S having a frequency difference of Δf. The surface 26a has a wavelength λ of the beam LB.
Is a wafer conjugate plane corresponding to the chromatic aberration generated in the projection lens PL, and the amount of axial chromatic aberration is ΔLk. The principal rays of the S-polarized beams LB _1S and LB _2S intersect in the plane 26a, and the principal rays of the P-polarized beams LB _1P and LB _2P intersect in the plane of the reticle mark RM. This operation is achieved by the bifocal element 21B shown in FIG.
It is due to.

Now, the S-polarized beams LB _1S , LB _{2S in} the intersection area of the P-polarized beams LB _1P and LB _2P on the reticle R or in the surface 26a
The size Dr of the intersection area is uniquely determined by the aperture size of the illumination field stop 18 in FIG. On the wafer W side, it may be considered that the wafer mark WM is arranged in the surface 26a, but the size Dr and the like are reduced by the magnification of the projection lens PL.

Beams LB _1S , LB intersecting on wafer W (or surface 26a)
_{Since 2S} has a constant beam diameter, the dimension of the three-dimensional region where the beams LB _1S and LB _2S intersect in the direction of the optical axis AXa 2.ΔZs
Can be considerably increased depending on the incident angle θr of the beams LB _1S and LB _2S . However, in the case of FIG. 6, an interference fringe of S-polarized light occurs in an arbitrary plane between a plane PSd separated from the plane 26a by ΔZs downward and a plane PSu separated from the plane 26a by ΔZs upward, Unfortunately, there is a reticle mark RM between the planes PSd and PSu. Then, on the reticle mark RM, in addition to the original interference fringes due to the P-polarized beams LB _1P and LB _2P , S-polarized interference fringes as noise components appear. For this reason, the diffracted light 104 from the reticle mark RM includes an S-polarized beat signal in addition to the original P-polarized beat signal. In the case of the light receiving system shown in FIG. 5, the original P-polarized beat signal is split toward the light-receiving element 25B at the polarizing beam splitter 243, but the S-polarized beat signal that becomes noise is reflected toward the field stop 245A. You. Field stop
The 245A is set to pass only the image of the wafer mark WM. However, in the process of achieving the alignment between the reticle R and the wafer W, the field stop 245A and the wafer mark WM do not always match exactly. Part of the reticle mark RM may enter the 245A window. Therefore, the light receiving element 25A, which should have received only the diffracted light 105 from the wafer mark WM, receives the S signal from the reticle mark RM.
The polarization beat signal is mixed as noise. Of course, even if the field stop 245A and the mark WM match exactly, the noise component of the S-polarized light is superimposed at a low level. This means that the light receiving element 25B and the field stop 245
The same applies to the B side.

Therefore, it is desirable to set a condition such that the mark RM of the reticle R does not exist between the surfaces PSu and PSd shown in FIG. For that, the beam LB _1P for reticle R irradiation
And a surface PPu where the LB _2P begin to intersect, it is brought close to the reticle R side of the relatively surface 26a. That is, the distance (the amount of chromatic aberration) ΔLk between the surface 26a and the reticle R in the direction of the optical axis AXa,
The relationship between the intermediate plane Pk between Su and PSd (the same plane as the plane 26a) and the distance ΔZs between the plane PSd may be set to ΔZs ≦ ΔLk.

In FIG. 6, the beam cross-sectional dimensions of the beams LB _1S , LB _2S , LB _1P , and LB _2P are the same, and the lateral dimension of the intersection area of the beams LB _1S and LB _2S on the surface Pk (surface 26a) is Dr. Then, θr
Is small, the following relationship holds.

Note that the incident angle θr is the same for both P-polarized light and S-polarized light on the reticle.

On the other hand, the incident angles θr of the beams LB _1S and LB _2S are determined by the above equation (1), and the wavelength λ is a value sufficiently smaller than the grating pitch (the value obtained by converting the pitch of the wafer mark WM to the reticle side) Pr. Yes, if the expression (11) is put together with n = 1, the following approximate expression is obtained.

In general, the chromatic aberration amount ΔLk of the projection lens PL increases as the wavelength λ increases, so that the further the wavelength λ is away from the wavelength of the exposure light, the more easily the condition of the expression (12) is satisfied. This is because the change rate of the chromatic aberration amount ΔLk may be larger than the change rate of the wavelength λ. In contrast, the dimensions
Dr and Pitch Pr have an optimal relationship to some extent.
Of course, Pr is uniquely determined by the condition of equation (1). Further, assuming that the minimum required number of diffraction grating elements is m for the wafer mark WM, an area dimension of about m · Pr is required in the measurement direction, and the dimension Dr of the intersection area is Dr> m · Pr There is a condition. Therefore, if h is a ratio of the size of the irradiation area to the size of the mark WM and is a constant of about 1 to 3, Dr = h
・ Can be m ・ Pr.

 Therefore, equation (12) can be transformed into the following equation.

The dimension Dr of the intersection area of the beams LB _1S and LB _{2S on} the surface 26a is also uniquely determined by the opening size of the irradiation field stop 18 in FIG.

Next, the configuration of the stepper according to the embodiment of FIG. 2 of the present invention will be described with reference to FIGS. 7 (A) and 7 (B). FIGS. 7A and 7B show only the relay system 240A and B shown in FIG. 5, and the other configuration is the same as FIG. In this embodiment, the divided light receiving elements 300R and 300W are directly arranged at the position of the spatial filter 23 provided after the relay systems 240A and 240B, and receive the diffracted lights (beat signals) 104 and 105 in the pupil space of the system. I did it. As shown in FIG. 7 (A), after the relay systems 240A and 240B, a polarizing beam splitter 310 is provided obliquely. Since the diffracted light 105 from the wafer W is an S-polarized component, it is reflected here to receive light. Reaches 300W. On the other hand, since the diffracted light 104 from the reticle R is a P-polarized component, it passes through the beam splitter 310 and reaches the light receiving element 300R. Each of the light receiving elements 300R and 300W has a light receiving surface as shown in FIG. 7B, for example, and a surface 30 for receiving the diffracted light 104 or 105 at the center.
0a is formed, and the 0th order light is vertically
Surface 300b for receiving L _0, 300c are formed. These three
The three light receiving surfaces 300a, 300b, and 300c are electrically insulated, and individually output photoelectric signals. Light receiving surface 30
0a, 300b, and 300c are arranged substantially conjugate with a predetermined position in the pupil space of the system, for example, the beam waist position BW. Therefore, the photoelectric signal of the light receiving surface 300a of the light receiving element 300R becomes the signal DSr, and the photoelectric signal of the light receiving surface 300a of the light receiving element 300W becomes the signal DSw.
And used for phase detection exactly as in the previous embodiment.

As is clear from (principle) shown in FIG. 1, all the diffracted lights 104 and 105 (and the zero-order light L ₀ ) are condensed into a minute spot (beam waist) in the pupil (in space) of the system. Would.
Therefore, the areas of the light receiving surfaces 300a, 300b, and 300c can be made extremely small, which is advantageous for improving the responsiveness as a light receiving element. Also, since the numerical aperture (NA) of the diffracted lights 104 and 105 in the pupil space is considerably small, the light receiving surface 300a is set to have a slightly larger area than the accurate spot size of the diffracted lights 104 and 105, so that the light receiving element 300R,
The position setting accuracy in the direction of the optical axis of 300 W does not need to be so strict, and the device manufacturing becomes simple.

Here, each beam LB ₁ , LB ₂ , at the beam waist position
BTL, and consider the spot diameter of L _0, minimum beam diameter r
Is approximately given by:

Here, f is the focal length of the objective lens 21A, and f = 100
Assuming that mm, λ = 0.63 μm, and Dr = 500 μm (on the reticle), r ≒ 0.15 mm.

The distance Da in a pupil space of the principal ray and the principal ray of the 0 order light L ₀ of the diffracted light 104 and 105 (see FIG. 1) is, Da ≒ f · sin [theta
r, λ = 0.63 μm, Pr = 20 μm (4 μm on the wafer
(m pitch), θr {1.8} from the equation (1), which is a small value of Da {3.15 mm.

Therefore, the light receiving surfaces 300a and 300b shown in FIG.
300c mutual positional relationship and of the diffracted light 104 (or 105), it is necessary to accurately determine the spatial position relationship of the zero-order light L ₀ and the light-receiving element 300R (300 W). In addition, the light receiving element 300R (300
Each light receiving surface in (W) is conjugated to the waist position BW, but may be located anywhere as long as the condition of Expression (8) is satisfied.

FIG. 8 (A) shows a structure of a light receiving system block according to a third embodiment of the present invention, which is suitable when a light receiving element is arranged in a pupil space. The light receiving element 300, the light receiving surface for receiving the light-receiving surface 300a and the 0 order light L ₀ for receiving the spot of the diffracted beam 104, 105
302A, 302B, 302C, and 302D are provided similarly to FIG. 7B. Wherein the two light receiving surface for receiving the 0-order light L ₀ is further have two divided light receiving surface 302A and 302B are electrically insulated by the number μm~ dozen μm separation zone, the light-receiving surface 302C and 302D
Are also electrically insulated by a separation band of several μm to several tens μm. The light receiving element 300 is fixed to a mount member 303, and the mount member 303 is mounted inside the frame 304 so as to be vertically movable. A leaf spring 307 is fixed inside the lower end of the frame 304, and urges the mount member 303 upward with a relatively large force. Two screws 306A and 306B are screwed into the upper end of the frame 304, and the tips of the screws abut the upper end of the mount member 303. Around the frame 304, mounting tabs 305A and 305B for fixing the light receiving system block to the mounting are provided.

The adjustment of the light receiving system block is performed as shown in FIG. 8 (B). In Figure 8 (B), K ₁ is a light receiving surface 302A
302B of indicates separation zone, K ₂ denotes a separation zone of the light receiving surface 302C and 302D. The distance between the separation bands K ₁ and K ₂ is determined so as to match as closely as possible the distance between the principal rays of the two zero-order lights L _{0 on} the pupil plane. Spot of the diffracted light 104 (105) for measurement are always located in the middle of the two zero-order light L _0. So during the adjustment, as separation zone K ₁ and two 0-order light L _0, the positional relationship between the K ₂ are symmetrical, two screws 306A, rotates the 306B, fine movement of the position of the light receiving element 300 Let it. At the time of this adjustment, a mirror member (chrome surface of the reticle or chrome surface of the reference mark plate FM) is arranged on the reticle surface where the _two beams LB ₁ and LB ₂ intersect or on the wafer surface. This will give you two zeros
Since the secondary light L ₀ is not affected (modulated) by the diffraction grating at all, the light receiving surfaces 302A and 302B and the light receiving surfaces 302C and 302D
Will be received simple amount of L _0, photoelectric signals from the light receiving surface becomes a DC component only (including ripple slight beat signal). Therefore, the light receiving surfaces 302A, 302
B, the level of the photoelectric signal from each of 302C, 302D is A,
Assuming B, C, and D, the light receiving element 300 is finely moved by a differential operation circuit or the like so that the following voltage value ΔV becomes zero.

ΔV = (A + C) − (B + D) As a result, as shown in FIG. 8B, the two zero-order L ₀ are symmetrically positioned with respect to the separation bands K ₁ and K ₂ , and as a result, two of the zero-order light center L _0, i.e., the spot center of the diffracted light 104 and 105 will be located just around the separation zone K _1, K _2, the spot center of the diffracted light 104 and 105 light-receiving surface 300a
Also coincides with the center of

FIG. 9 shows an example of a preamplifier circuit in a printed circuit board PCB fixed to the back surface of the mount member 303 or the frame 304. Each receiving surface is made of all the photodiodes (or photo transistor), an operational amplifier U _1, U _2, U ₃ is the resistance _{R 1, R 2, R 3} , they photodiode 300a, 302A to 302D of the current-voltage Works as a converter. Output signal J ₁ of the operational amplifier U ₁ is diffracted beam 104, or 105
Are the measurement signals DSr and DSw corresponding to.

The operational amplifier U ₂ is connected photodiode 302D,
The operational amplifier U ₃ is connected photodiode 302A, the remaining two photodiodes 302B, 302C are connected appropriately photodiode 302A, in parallel with each of 302D by switching the switch Sk. When the switch Sk is in the state shown, the photodiode 302B is connected in parallel with the photodiode 302D, the level of the output signal J ₂ of the operational amplifier U ₂ is (B + D)
Is equivalent to On the other hand, the photo diode 302C are connected in parallel with the photodiode 302A, the output signal J ₃ of the operational amplifier U ₃
Corresponds to (A + C). Operational amplifier U ₄ acts as a subtractor with four resistors R, the output signal J ₄ is a (J ₃ -J _2). Therefore in the state of the switch Sk is shown, by monitoring the level ([Delta] V) of the output signal J _4, can adjust the position of the light receiving element 300. After the position adjustment is completed, the switch Sk is switched from the position shown in the figure,
The photodiodes 302A and 302B are connected in parallel, and the photodiodes 302C and 302D are connected in parallel. This output signal J _2, J ₃ becomes one corresponding to the light amount of each 0th order light L _0. Note that the switch Sk may be a relay switch or an analog switch that can be switched in response to an external signal.
In addition, the power supply voltage ± Vc for each operational amplifier and the ground are supplied to the substrate PCB from the apparatus main body via connectors J ₅ , J ₆ , and J ₇ . Furthermore, the position adjustment of the light receiving element 300 with signals J ₄ from the circuit, it is desirable to conduct again after the telecentricity adjustment during device fabrication (symmetry adjustment of the angle of incidence θr of the beam LB _1, LB ₂₎ .

In addition, diffracted light 104 and 105 for the light receiving system, zero-order light
For adjusting the position of the L _0, a parallel plane glass tiltable in front of the light receiving element 300 is provided, a method of controlling the tilt amount of the parallel plane glass motor also conceivable. In this case, the diffracted light 104 (105) and the two zero-order lights are caused by the inclination of the parallel plane glass.
A total of three spots L ₀ are determined so as to be simultaneously displaced on the light receiving element 300 in the spot arrangement direction. Then the switch Sk in the state of FIG. 9, the output signal J ₄ by applying to the servo circuit of the motor for tilt control as a feedback signal (error signal), into two beams LB _1, the LB ₂ crossing regions The servo works at the moment when the mirror member is arranged, and the parallel flat glass stops at a predetermined inclination amount. And the output signal J ₄
After a certain period of time has passed since the level ΔV has become almost zero, the motor input terminal is forcibly dropped to zero (the motor is cut off) and the servo drive is stopped. With such a method, the position between the light receiving surface 300a and the diffracted light 104 (105) can be similarly adjusted. In the case where the parallel flat glass is tilted in the lattice arrangement direction instead of moving the light receiving element, the reticle R is replaced and the objective lens of the alignment system is replaced.
Each time the position of 21A is changed, the optimum light receiving position can be adjusted using the mirror portion of the reference mark plate FM on the wafer stage 5. In FIG. 9, the same adjustment can be performed only by the differential between the photodiodes 302A and 302D without the photodiodes 302C and 302B and the switch Sk.

Next, a fourth embodiment of the present invention will be described with reference to FIG. FIG. 10 shows a configuration of a stepper provided with a conventional TTR alignment system, and members having the same functions as those in FIG. 2 are denoted by the same reference numerals. Here, bifocal objective optical system 21 and the mirror M ₃ for alignment is moved one-dimensionally in the horizontal direction as indicated by the arrow A together. In FIG. 10, as shown in FIG. 3, alignment marks RM are formed at three locations around the reticle 1, and the same alignment system is provided for each mark position. Further, photoelectric detectors that receive diffracted light from the wafer mark WM and the reticle mark RM include light receiving elements 300R and 300W (FIG. 7 or FIG. 8) arranged in the pupil space.
Figure). Here, the mirror M ₃ are during exposure to about 45 ° rotation as shown by an arrow from the position shown, retracted so as not to shield the exposure illumination optical path.

In such a configuration the stepper, the time of alignment and exposure, since the mirror M ₃ is rotated, the position reproducibility of the mirror M ₃ is poor, the beam (L relative to the wafer W side
B ₁ , LB ₂ ) or the telecentricity of the optical axis AXa of the alignment system changes, and the reproducibility of the mark detection position may slightly change for each shot area on the wafer. Mirror
For loading and unloading the M ₃ at high speed, it is necessary to mirror M ₃ compact, lightweight, the mechanical position repeatability is naturally limited.

In this embodiment, telecentricity immediately before the alignment (mark detection) via a mirror M ₃ (beam L
(Symmetry of incident angles θr of B ₁ and LB ₂ ) can be automatically corrected. In this case, the alignment for each shot area on the wafer is to finely move the wafer W (wafer stage), and the reticle R is fixed after the initial setting. First, after fixing the reticle R, the reflecting surface of the fiducial mark plate FM is arranged below the projection lens PL, and the fiducial mark plate FM is moved up and down, so that a parallel plane disposed at or near the beam waist position BW. The telecentricity is adjusted by the glass 50. This parallel flat glass 50 is
The beam 52 rotates about an axis perpendicular to the plane including the beams LB ₁ and LB ₂ . After the adjustment, the position of the mark RM on the reticle R is
The phase difference φr between the reference signal DR and the measurement signal DSr is detected and stored.

Then immediately after the mirror M ₃ is about 45 ° rotated to align the sequential shot areas in the die-by-die method, firstly a reference signal DR by the phase detecting system 40 measurement signal DS
The phase difference φr ′ from r is checked, and the deviation from the initially stored phase difference φr is monitored. When the deviation is larger than a predetermined value, the parallel flat glass 50 is finely moved according to the amount of the deviation, the telecentricity is automatically adjusted, and then the measurement signal DSw corresponding to the waist mark WM is used. To perform alignment between the reticle 1 and the shot area.

In this way, without telecentricity on the west W of the optical axis AXa of the alignment optical system is affected by the position repeatability by rotation of the mirror M _3, there is an advantage is kept substantially constant. Incidentally, the position reproducibility of the mirror M ₃ say bad, after all reference gratings 162 and (FIG. 5) a positional relationship between the reticle mark RM that will by relatively small shifts in the grating arrangement direction Nothing else.

The above example can be realized because the position of the reticle R is fixed during the alignment and the exposure.
Is taken from the interferometer 43 (see FIG. 2), the movement amount of the mark RM in the measurement direction is calculated from the value, and the output value and
It is necessary to determine the driving amount of the parallel flat glass 50 based on the deviation amount corresponding to the phase difference φr stored in advance.
For example, the deviation amount corresponding to the previously stored phase difference φr is Δ
X _m , the reading of the interference system 43 when this phase difference φr is obtained is
And xr _1, '[Delta] X _m excursion amount corresponding to' the phase difference φr obtained based on the signal DSr when inserted into the optical path of the mirror M ₃ in order to align certain shot area, the interferometer at that time When the reading of 43 and Xr _2, error component [Delta] X _T of telecentricity by the mirror M ₃ are given by the following equation.

_{ΔX T = ΔX m -ΔX m '} - (Xr 1 -Xr 2) Thus conditions for a [Delta] X _T almost _{zero, ΔX m' = ΔX m '} - is _{(Xr 1 -Xr 2), ΔX} m The parallel flat glass 50 may be rotated until ′ becomes equal to ΔX _m − (Xr ₁ −Xr ₂ ). (Xr ₁
When the value of −Xr ₂ ) is equal to or more than ± 1/2 pitch of the mark RM, the measured phase difference φr is also treated as if the period is shifted by an integer multiple.

Although the embodiments of the present invention have been described above, the present invention can be implemented in exactly the same manner even in a proximity method in which a mask and a wafer are exposed close to about 10 to 300 μm. Also,
In the case of the system shown in FIG. 5, the spatial filter 23 may be moved in the optical axis direction along with the movement of the objective lens 21A. In that case, since the amount of movement of the spatial filter 23 is that proportional to the amount of movement of the objective lens 21A is desired, NA of the diffracted light BTL, 0 order light L ₀ in a pupil space as analyzed above is extremely small However, it is not always necessary to make a proportional relationship. Alternatively, a spatial filter that can be inserted and removed may be provided at a few places in the pupil space, and a spatial filter at an appropriate position may be selected according to a change in the position of the objective lens 21A. Further, the position of only the wafer mark WM may be detected without detecting the reticle mark RM. In this case, the objective lens 21A can be arranged by observing the wafer W from beneath the reticle R via the projection lens PL, or by making the reticle mark RM a simple transparent window.

FIG. 11 shows the configuration of an alignment system according to a fifth embodiment of the present invention. The same members as those shown in FIG. 2 or FIG. 10 are denoted by the same reference numerals. In this embodiment,
The focal length of the objective lens 21A is increased, a beam splitter 320 is provided between the objective lens 21A and the bifocal element 21B, and the reticle mark R
M or the diffracted light BTL from the wafer mark WM, the objective lens
I took it out right after 21A. Then, the diffracted light BTL branched by the beam splitter 320 is changed by the modified beam splitter 322 into a P-polarized diffracted light 104 from the reticle mark RM.
And the S-polarized diffracted light 105 from the wafer mark WM, and the light receiving elements 300R, 30R of FIG. 7 or FIG.
Received at 0W. Furthermore, the mirror M ₃ at the tip and the objective lens
21A, the bifocal element 21B, the beam splitters 320 and 322, and the light receiving elements 300R and 300W are integrally fixed to the holding hardware 62, and move as indicated by an arrow A with a change in the mark position. Here, the light receiving elements 300R, 300 W is disposed in substantially the same position as the pupil plane EP ₁ of the objective lens 21A. Further, the positional relationship between the optical axis AXa and the light receiving surface is such that the two beams LB ₁ and LB ₂ are positioned so as to sandwich the optical axis AXa in a direction perpendicular to the plane of FIG. And the light receiving elements 300R and 300W can be adjusted by finely moving them in a direction perpendicular to the plane of the paper and in a horizontal direction within the plane of the paper.

The mirror M ₃ of the tip may be rotated as Figure 10. Further, the system of this embodiment can be applied to the system shown in FIG. According to the present embodiment,
Since the light receiving elements 300R and 300W are arranged in the pupil space and are fixed to the hardware 62 integrally with the objective lens 21A, the positional relationship between the light receiving elements 300R and 300W and the optical axis AXa of the objective lens 21A is completely disturbed. It is possible to cope with a change in the observation position without any change.

FIG. 12 shows a configuration of a sixth embodiment based on the configuration of FIG. 11, in which the reference signal generating unit 16 shown in FIG. 2 is provided integrally in a holding hardware 62 of the objective lens 21A. It is. In the beam splitter 320, the two beams LB ₁ and LB ₂ that have passed through the bifocal element 21B are amplitude-divided, so that about half the amount of light is wasted. Therefore, as shown in FIG. 12, about half of the beams LB ₁ and LB ₂ reflected by the beam splitter 320 are
The light is guided to the lens 160, the mirror 161, the polarizing plate 165, the reference diffraction grating plate 162, and the like, and these are integrally housed in the hardware 62. Of course, the spatial filter 163 and the light receiving element 19 shown in FIG. However, a deflection plate 165 is provided in front of the reference grating 162, and the P-polarized beams LB _1P and LB _2P and the S-polarized beam LB
_2S, irradiating either the LB _2S reference grid 162.

〔The invention's effect〕

As described above, according to the present invention, when a part of the alignment optical system moves with a change in the mark position, it is not necessary to provide a special optical path length correction optical system for maintaining pupil conjugate as in the related art. The structure of the alignment system can be made extremely simple. For this reason, various error factors added at the time of mark detection are reduced, and the alignment accuracy can be improved.

Further, in the position detecting device of the present invention, the adjusting means may control the first and the second signals based on the signals obtained by the second and third photoelectric detectors.
Since the relative position between the photoelectric detector and the optical path of the diffracted light is adjusted, the position of the diffracted light received by the first photoelectric detector can be easily adjusted, and the diffracted light can be detected with high accuracy. can do.

[Brief description of the drawings]

FIG. 1 (A) is a perspective view illustrating the principle configuration of the present invention,
FIG. 1B is a diagram schematically showing the state of various types of diffracted light in the pupil space of the alignment system. FIG. 2 is a diagram showing the configuration of the projection exposure apparatus according to the first embodiment of the present invention. FIG. 4 is a perspective view showing a specific configuration near the objective lens in the alignment system in FIG. 2, FIG. 4 is a plan view showing an example of mark arrangement, and FIG. 5 is a configuration of the alignment system in FIG. FIG. 6 is a view for explaining the state of two beams irradiated near the reticle, and FIGS. 7A and 7B show the configuration of a light receiving system according to a second embodiment of the present invention. FIGS. 8A and 8B are diagrams showing a configuration of a light receiving system according to a third embodiment of the present invention. FIG. 9 is a circuit diagram showing a configuration of a preamplifier for signals of the respective light receiving elements. FIG. 10 is a perspective view showing a configuration of a projection exposure apparatus according to a fourth embodiment of the present invention, and FIG. 11 is an arc exposure apparatus according to a fifth embodiment of the present invention. Diagram showing the configuration of the liment system,
FIG. 12 is a diagram showing a configuration of an alignment system according to a sixth embodiment of the present invention. [The main part of the description of the code] R ...... _{reticle, RM, RM x, RM y} , RM θ ...... reticle mark, W ...... wafer, WM ...... wafer mark, PL ...... projection lens, EP, EP ₁ ... ... pupil, BW ...... beam waist position, beam of LB ₁ ...... frequency f _1, beam of LB ₂ ...... frequency f _2, BTL ...... diffracted light (primary _{light), L 0, L 0 '} ...... 0 Next light, AXa: Optical axis of alignment system, OBJ: Objective lens, DR: Reference signal, DSr, DSw: Measurement signal, 10: Laser light source, 12: Two-beam frequency shifter, 16: Refer to Signal generator, 21A ... objective lens, 21B ... bifocal element, 40 ... phase detection system, 62 ... holding hardware.

──────────────────────────────────────────────────続 き Continuation of front page (72) Inventor Koichiro Komatsu 1-6-3 Nishioi, Shinagawa-ku, Tokyo Nikon Oi Works Co., Ltd. (56) References JP-A-63-283129 (JP, A) 57-142612 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) H01L 21/027

Claims (7)

(57) [Claims] 1. A diffraction grating formed along with a transferred area on a sensitive substrate prior to transferring a pattern area formed on a mask to a transferred area on the sensitive substrate. Optically detecting the alignment mark, and a movable objective optical system corresponding to the distance of the alignment mark from the center of the transferred area; and Using a photoelectric detector that photoelectrically detects optical information,
In an alignment apparatus for detecting a position of the alignment mark, a light source for irradiating the alignment mark with a light beam through the objective optical system, and a principal ray of the light beam on an optical axis on a pupil plane of the objective optical system And a beam transmitting system that guides the light beam to a predetermined waist position in the pupil space on the principal ray and converts the light beam to the alignment mark into a substantially parallel light beam. And when a zero-order diffraction beam and a high-order diffraction beam generated from the alignment mark by the irradiation of the light beam enter the objective optical system, the zero-order diffraction beam and the high-order diffraction beam are separated by the photoelectric detector. And a drive unit for moving the objective optical system by limiting the range to be separated from each other on a predetermined surface on the light receiving system side including Raimento apparatus. 2. The light beam transmitting system according to claim 1, wherein the light beam transmitting systems are arranged substantially symmetrically with respect to the optical axis, have two principal rays separated by a predetermined distance in the pupil space, and converge at the waist position. When the objective optical system is positioned at a substantially middle point of the moving range by the driving unit, the pupil plane of the objective optical system 2. The alignment device according to claim 1, wherein the alignment device is set to be located near the position. 3. The photoelectric detector according to claim 1, wherein the photoelectric detector is disposed at the waist position or at a position substantially conjugate with the waist position, and has a light receiving surface for receiving only the high-order diffraction beam. Or the alignment apparatus according to claim 2. 4. A relay optical system for relaying the waist position in a direction branched from a light path of the light beam, and a conjugate plane of the waist position formed by the relay optical system, or A spatial filter that is arranged in the vicinity and that extracts only the higher-order diffraction beam, and a light-receiving element that receives the extracted higher-order diffraction beam, is provided in accordance with a position in a moving range of the objective optical system, 2. The apparatus according to claim 1, wherein a position of the spatial filter is provided so as to be continuously or discretely changeable in the optical axis direction.
Item 3. The alignment device according to Item 2 or 2. 5. A splitter for splitting a beam from a light source into two radiation beams, wherein said two radiation beams are incident, and said two beams are arranged on a diffraction grating pattern on an object to be located. An objective optical system that emits light so as to intersect at an angle, a pupil plane of the objective optical system or a space conjugate with the pupil plane, which is generated coaxially in the same direction from the diffraction grating pattern, A first photoelectric detector for receiving two diffracted lights traveling through a system, wherein the position detector detects the position of the object based on a signal from the first photoelectric detector; A second and a third photoelectric detector that are arranged in parallel in the pitch direction of the diffraction grating pattern with the photoelectric detector interposed therebetween and detect each of two beams generated from the object differently from the diffracted light; The first photoelectric test An output unit, a detector holding member that holds the second and third photoelectric detectors, and a first photoelectric detector based on signals obtained by the second and third photoelectric detectors. Adjusting means for adjusting the relative position of the diffracted light with respect to the optical path. 6. The position detecting apparatus according to claim 5, wherein said adjusting means includes a fine movement mechanism for accurately moving said detector holding member with respect to the optical path of said diffracted light. apparatus. 7. The apparatus according to claim 7, wherein said adjusting means includes a plane-parallel glass disposed on a front surface of said detector holding member to shift an optical path of said diffracted light with respect to said detector holding member. The position detecting device according to claim 5, characterized in that: