JPH02133913A - Alignment apparatus - Google Patents

Abstract

PURPOSE:To execute a high-accuracy alignment operation by a method wherein a first mark formed on a mask and a second mark formed on a substrate are irradiated individually from two directions and the following are provided: a mark detection optical system to detect a first piece of optical information where a zero-order diffracted beam and a high-order diffracted beam reflected by the first mark have been made to interfere and to detect a second piece of optical information generated from the second mark; a photoelectric detection means of them. CONSTITUTION:A diffracted beam 105 from a mark WM on a wafer 4 is reflected by means of a beam splitter 20 and reaches a photoelectric detector 53 by means of a polarized-light beam splitter PBS; a photoelectric signal S1 is output. On the other hand, optical information (-LBR0, LBR+1) 104A from a mark RM and optical information (+LBR0, LBR-1) 104B reach photoelectric detectors 58A, 58B by means of the polarized-light beam splitter PBS; alternating-current signals S2, S3 at frequencies of interference beat signals are output individually. Then, a phase difference with reference to a reference signal is measured. The phase difference with reference to the reference signal is measured also regarding the signal S1; a displacement between a reticle 1 and the wafer 4 is determined.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to an alignment device for a projection exposure apparatus used for manufacturing semiconductor devices, etc., and particularly relates to a mask having an original pattern, and a mask having an original pattern. The present invention relates to a device for relatively aligning a substrate such as a semiconductor wafer onto which an image is to be transferred.

[Conventional technology]

This type of projection exposure equipment (stepper) uses a reticle (
A circuit pattern image is exposed onto a predetermined position on the wafer through a projection lens that reduces and projects the circuit pattern formed on the mask (115 or 1/10). At this time, the reticle is illuminated with monochromatic light from a mercury discharge lamp or the like to suppress the occurrence of chromatic aberration in the projection lens. For this reason, the projection lens has various aberrations corrected so as to have the best distortion characteristics near the wavelength of monochromatic illumination light for exposure (for example, i-line, g-line, etc.).

Furthermore, when manufacturing semiconductor devices, multiple layers of circuit patterns are created on the wafer surface, which requires mask work several to several times. The main work of this mask work is the work of precisely aligning the reticle to be newly overlaid and exposed with the circuit pattern area already formed on the wafer, that is, alignment. Many of the projection exposure apparatuses currently in practical use incorporate a device for optically automatically aligning a reticle and a wafer, and have achieved great results in semiconductor device manufacturing.

By the way, there are various methods for this automatic alignment device, and one of the methods expected to have the highest accuracy is called the through-the-reticle (TTR) method. This method simultaneously detects alignment marks around the circuit pattern area of the reticle and alignment marks formed around one shot area on the wafer using an alignment optical system (alignment objective lens, etc.) placed above the reticle. This method directly measures the deviation between both marks and moves the reticle or wafer slightly so that the amount of deviation becomes zero.

In this case, the alignment illumination light (
The scanning laser spot light (or uniform illumination light) has been set to have the same wavelength as the exposure illumination light or a wavelength close to it, taking into consideration the chromatic aberration of the projection lens. Therefore, when alignment illumination light illuminates a mark on a wafer for alignment, the resist layer in that area is exposed to light, and when it is subjected to various processes after development, the mark on the wafer is destroyed. A problem arose in that it could not be used for alignment during overlapping exposure of the next layer.

Therefore, as one solution to this problem, as shown in Figure 1O, TTR alignment is performed using illumination light in a wavelength range that is insensitive to the resist layer, and the chromatic aberration caused by the projection lens PL is reduced between the reticle R and the projection lens. A correction optical system O6 provided only in the alignment optical path between the reticle R and the projection lens PL, or a correction lens system 0. ! etc., the reticle R and wafer W can be corrected even in wavelength ranges other than exposure light.
A method has been proposed that maintains the conjugate with.

In FIG. 10, RM is the alignment mark of the reticle R (or just a transparent window), WM is the alignment mark of the wafer W, and ALg sends alignment illumination light and detects optical information from the marks RM and WM. This is an alignment optical system for

In this way, the method of providing some kind of chromatic aberration correction system between the reticle R and the projection lens PL is excellent in principle, but the overlapping position of the projected images shifts slightly between alignment and exposure. It is known to have serious drawbacks. This is mainly due to mechanical stability of the chromatic aberration correction system, setting errors during device manufacturing, temperature drift, etc.

Therefore, a method for dealing with chromatic aberration errors without providing any correction system between the reticle R and the projection lens PL is disclosed in, for example, Japanese Patent Laid-Open No. 153820/1983. In this Japanese Patent Application Laid-Open No. 63-153820, an alignment optical system including a bifocal element is arranged above the reticle, and a beam is imaged as a spot light on the reticle and a beam is imaged as a spot light on the wafer. is guided coaxially from the alignment laser light source to the alignment objective lens, and the imaging plane of the beam emitted from the objective lens is made into two focal points. Therefore, the beam of spot light to be imaged on the wafer is largely defocused on the reticle due to chromatic aberration.

Then, using the coaxiality of the two beams, the two beams (spot lights) are scanned by a scanner, and the optical information from the mark on the reticle and the optical information from the mark on the wafer are
Photoelectric detection is performed at the same time.

[Problem that the invention seeks to solve]

In any case, in the above-mentioned conventional techniques, since the mark on the wafer is scanned by a spot light of a laser beam, various problems associated with a spot scanning type alignment system remain unsolved.

In the spot scanning method, scattered light or diffracted light from a mark edge on a wafer is photoelectrically detected, so a peak waveform of a photoelectric signal is obtained only when a mark edge exists within a spot light irradiation area (for example, 2 to 4 μm width). That is, basically, the mark position is determined from a single peak waveform, and if the peak waveform is distorted, this directly affects alignment accuracy.

Therefore, the present invention provides a positioning method for a projection exposure apparatus that can perform more accurate alignment I under alignment illumination light of a different wavelength without performing mark detection using such a spot scanning method. The purpose is to obtain equipment.

Furthermore, it is an object of the present invention to provide an apparatus that employs an interference alignment method using a diffraction grating mark or the like in order to enable alignment with higher resolution, and can detect marks even during exposure operation.

A further object of the present invention is to provide a device that prevents the optical information from the alignment marks provided on the mask and the alignment marks provided on the photosensitive 2S plate from being mixed up.

[Means for solving problems]

In the present invention, when the mask and the photosensitive substrate are aligned, the area of the alignment mark formed on the mask and the area of the alignment mark formed on the photosensitive substrate are mutually aligned under the wavelength of the exposure illumination light. At the same time, each mark on the photosensitive substrate of the mask is arranged as a lattice pattern, etc., and the lattice constant (pitch, interval, etc.) of the image (lattice image) of the lattice mark of the mask by the projection optical system and the pattern on the photosensitive substrate are arranged so that they do not overlap. The lattice constants of the lattice marks are different from each other. Then, illumination light with the same polarization component is simultaneously irradiated to the mark on the mask from two directions, and the 0th-order diffracted light from the mark on the mask and the higher-order (
The interference light with the first-order) diffracted light (or scattered light) is photoelectrically detected.

(Function) In the present invention, the signal from the chopper/ticle mark and the signal from the wafer mark are not confused, and the reticle signal is structured so that only the diffracted light reflected by the reticle diffraction mark can be used. High precision alignment with few components can be achieved.

[Example] Next, a positioning device according to an example of the present invention will be described, but before that, the technology underlying the present invention will be described with reference to FIG. 1.

A reticle 1 having a predetermined circuit pattern and a diffraction grating mark for alignment is held on a two-dimensionally movable reticle stage 2. Each pattern on the reticle I is imaged onto the wafer 4 under exposure light by a projection lens 3 which is telecentric on both sides. However, this projection lens 3 has well-corrected chromatic aberration with respect to the exposure illumination light wavelength (g-line, i-line, etc.), and is arranged so that the reticle l and the wafer 4 are conjugate to each other with respect to the exposure wavelength. Ru. Further, a diffraction grating mark similar to the grating mark formed on the reticle l is also formed on the wafer 4. Now, the wafer 4 is attracted onto a stage 5 that moves two-dimensionally in a step-and-repeat manner, and when the transfer exposure of the reticle I to one shot area on the wafer 4 is completed, it is stepped to the next shot position. A part of the reticle stage 2 includes
A movable mirror 6 that reflects a laser beam from a laser beam interferometer (hereinafter referred to as an interferometer) 43 for detecting the position in the direction and rotation (θ) direction is fixed. This interferometer 43 has three laser beams for length measurement in order to independently detect positions in the X direction, X direction, and θ direction, but some of them are omitted here to simplify the explanation. .

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. O
It is set at about 1 μm. On the other hand, wafer stage 5
A movable mirror 7 that reflects a laser beam from an interferometer 45 for detecting the position of the wafer 4 in the X direction and the X direction within the horizontal plane is fixed to a part of the wafer 4 . This interferometer 45 is also
Although it has two length measuring laser beams to independently detect the position in the direction and the X direction, some of them are omitted here to simplify the explanation. Reticle stage 2 X
The drive motor 42 drives the wafer stage 5 in the X direction, the θ direction, and the θ direction, and the drive motor 46 moves the wafer stage 5 in two dimensions.

By the way, the illumination system for exposure consists of a mercury lamp 30 and an elliptical mirror 3.
1. Input lens group 32 including condenser lens, interference filter, etc., optical integrator (fly's eye lens)
33, a mirror 34, a main condenser lens 35, a dichroic mirror 22, and the like. The dichroic ink mirror 22 is obliquely installed above the reticle l at an angle of 45'', and reflects the exposure light from the condenser lens 35 vertically downward to uniformly illuminate the reticle l. 9 for wavelength
It has a reflectance of 0% or more, and a transmittance of 50% or more for the wavelength of the illumination light for alignment (the iI front light also has a long wavelength).

Next, the alignment system of this stepper will be explained. Irradiation light for alignment is emitted from a laser light source 10, passes through a radial grating 11 in which a transmission type reference diffraction grating is formed radially, and passes through a Fourier transform lens 13.
The light reaches a spatial filter 15 arranged on the Fourier plane (pupil plane of the alignment optical system). The radial grating 11 is configured to be rotatable by a motor 12 at a substantially constant speed. The laser light incident on this radial grating 11 is diffracted into 0th order light, ±1st order light, ±2nd order light, etc., and each spreads at a different diffraction angle, as shown in Figure 1. Then, 0th order light LB,,+1st order light+
Only LB and -1st order light -LB are shown. The principal rays of these 0th-order light and ±1st-order light become parallel due to the action of the lens system 13, and they are clearly separated and distributed on the spatial filter 15 arranged on the Fourier plane, and the 0th-order light L B o Only the first-order light is blocked, and the ±1st-order light is transmitted. spatial filter 1
After passing through 5, the ±1st-order light is reflected by beam splitter 14, passes through pupil relay system 17A, and enters beam splitter 14.
20 and enters the bifocal optical system 21. The bifocal optical system 21 is a combination of a birefringent material (quartz, calcite, etc.) 21b arranged conjugately with the pupil of the alignment system, that is, the pupil EP of the projection lens 3, and a telecentric objective lens 21a such as for a microscope. It provides different powers depending on the polarization components (P-polarized light and S-polarized light) of the ±1st-order light of the laser light. Here - the light'
It is assumed that R10 oscillates a laser beam of orthogonal linear polarization. Therefore, one polarized light (for example, P polarized light) exiting the bifocal optical system 21 forms an image at the focal point 26a in the space above the reticle 1, and the other polarized light (for example, S polarized light) coincides with the pattern surface on the lower surface of the reticle 1. The image is formed at a focal point 27a. Also 2
The other focal point of the focusing optical system 21, that is, the laser light source 10
The surfaces conjugate to the respective focal points 26a, 27a on the side coincide with the radial grating 11. Here, the distance between the two focal points 26a and 27b of the bifocal optical system 21 in the optical axis direction corresponds to the amount of chromatic aberration on the reticle 1 side of the projection lens 3 at the wavelength of the alignment laser beam. The focal plane 26a in this space becomes conjugate with the imaging plane 26b which coincides with the surface of the wafer 4 by the projection lens 3, and the focal plane 27
a (reticle pattern surface) becomes conjugate with the imaging surface 27b spatially spaced downward from the surface of the wafer 4 by the projection lens 3. The distance between the imaging planes 26b and 27b is the same as that of the projection lens 3.
This corresponds to the amount of chromatic aberration on the wafer 4 side. Here, the distance between the imaging planes 26b and 27b is 0, and the focal plane 26a and 2
If the interval distance of 7a is Dr% and the projection magnification of the projection lens 3 is 1/M (usually M is one of 1.2.5.5.10), then - static, = M"・The relationship D.The farther the wavelength of the alignment laser light is from the wavelength of the exposure light, the larger D is, depending on the aberration characteristics of the projection lens 3. The depth of focus of the projection lens is extremely shallow, about ±1 μm, and the distance is several 10 μm depending on the wavelength of the alignment illumination light.
It can reach up to about m. Note that it is desirable that the illumination light (laser light) for flymen 1- has a wavelength that has almost no sensitivity to the resist applied to the wafer 4;
In the present invention, this condition does not necessarily have to be met because the projection lens causes an extremely large aberration between the wavelength of the exposure light and the wavelength of the illumination light for alignment, and in particular, the optical information itself from the diffraction grating mark on the wafer 4 is affected. This is because large distortions will be added. Therefore, it is more important to give priority to determining the optimum illumination light for alignment in consideration of the aberration. Therefore, if the alignment illumination light irradiates the resist for a long time (for example, 1 minute or more), it will be exposed to light (fading will occur after development).
In some cases, it may be a wavelength with weak sensitivity such as

Now, the ±1st order light LB of the laser light for alignment, (S
The polarized light enters the diffraction grating mark portion of the reticle 1 at the focal plane 27a from two directions at angles formed by the +1st order light +LB and the 11th order light -LB and forms an image. Further, the ±1st order light LB, (P polarized light) from the focal plane 26a that has passed through the transparent part of the reticle 1 is transmitted to the wafer 4 at the focal plane 26b via the projection lens 3.
The +1st order light and the 11th order light enter the diffraction grating mark portion from two directions at angles formed by the +1st order light and form an image. The reflected and diffracted light from the diffraction grating mark on the reticle L passes through the dichroic mirror 22.2 and the focusing optical system 21 to the beam splitter 2.
After being reflected at 0, only the diffracted light that passes on the axis is filtered by a spatial filter 23 placed on the pupil conjugate plane (Fourier plane) through the pupil relay system 17B, and is further transmitted to the photoelectric detector 25 by the condensing lens 24. reach. In addition, the reflected diffracted light from the diffraction grating mark on the wafer 4 returns to the original optical path via the projection lens 3, passes through the transparent part of the reticle 1, and passes through the dichroic mirror 22, the focusing optical system 21, the beam splinter 20, and the pupil relay. System 17B, spatial filter 2
3, and passes through the condensing lens 24 to reach the photoelectric detector 25. The spatial filter 23 is arranged at a position conjugate with the pupil plane of the alignment optical system, that is, at a position substantially conjugate with the pupil (exit pupil) of the projection lens 3, and blocks specularly reflected light from the reticle 11 or the wafer 4. Alternatively, it is set so that only light diffracted perpendicularly to the diffraction grating of the wafer 4 (in the normal direction of the surface) passes through. In front of the photoelectric detector 25, there is a bifocal optical system 21, a pupil relay system 17B, and a lens 2.
An aperture plate 25° is provided which is disposed conjugately with the reticle 1 and the wafer 4 via the aperture plate 4.

Now, the photoelectric signal obtained from the photoelectric detector 25 includes ±1st order light ±LB that irradiates the reticle 1 or wafer 4 from two directions;
The interference fringes created by this are irradiated onto each diffraction grating mark so as to flow in the pitch direction, resulting in a sinusoidal alternating current signal with a frequency corresponding to the rotation speed of the radial grating 11. By the way, the ±1st-order light and the 0th-order light from the radial grating 11 are transmitted through the beam splinter 14 and used as a reference diffraction grating by the lens system (inverse Fourier transform lens) 16 that converts the pupil (free plane) into the image plane. Image formed on 18 (beams from two directions + LB, ,
-LB, intersect), this reference diffraction grating 18 is fixed on the apparatus. This diffraction grating 18 also has +1st order light +LB and 11st order light -LB at a predetermined angle.
incident from the direction. The photoelectric detector 19 is a reference diffraction grating 1
8, and outputs a sinusoidal photoelectric signal. This photoelectric signal has a frequency proportional to the rotation speed of the radial grating 11,
This becomes the reference beat signal. The phase detection system 40 inputs the photoelectric signal from the photoelectric detector 25 and the photoelectric signal from the photoelectric detector 19, and detects the phase difference in the waveforms of both signals. The detected phase difference (±180'') is 1/2 of the grating pitch of the diffraction grating marks formed on each of the reticle 11 and the wafer 4.
It uniquely corresponds to the amount of relative positional deviation within. Control system 4
1 controls a drive motor 42.46 based on information on the detected phase difference (positional deviation amount), position information from each of the interferometers 43.45 obtained via the servo system 44, and controls the reticle 1 and Relative positioning (alignment) of the wafer 4 is performed.

In the explanation of FIG. 1, two beams +LB are used.

Although L B I is shown to intersect in the plane of the paper, in reality they are tilted to each other in a plane perpendicular to a plane containing the axis AX of the projection lens 3.

In the above overall configuration, part of the alignment optical system,
In particular, the bifocal optical system 21 is movable to any position according to the arrangement of the alignment marks on the reticle 1, so that mark detection is possible regardless of the arrangement of the marks. Furthermore, since the exposure light and the alignment illumination light are separated by the guichroic mirror 22 provided obliquely above the reticle 1, mark detection is possible even during the exposure operation. This may occur due to some disturbance during exposure.
This means that even if the alignment state between the wafer 4 and the wafer 4 goes out of alignment, it can be immediately detected at that point. Furthermore, it also means that the positioning servo between the reticle stage 2 and the wafer stage 5 can be executed in a closed loop based on the phase difference information from the phase detection system 40 even during the exposure operation.

Therefore, the line width of the exposed resist pattern does not increase due to slight image blurring. Note that the light source of the exposure light may be replaced with an excimer laser light source other than the mercury lamp.

Next, using Figure 2, the detailed configuration of only the alignment system,
and the principle of alignment will be schematically explained. In FIG. 2, a guichroic mirror 22, a spatial filter 1
5. The beam splinter 14 and the pupil relay system 17A are omitted for the sake of simplicity, and the same members as those in FIG. 1 are given the same reference numerals. A laser beam (almost parallel beam) LB formed from a laser light source 1O is incident on the radial grating (frequency shifter) 11.

The polarization direction of this laser beam LB is separated into P-polarized light and S-polarized light by the bifocal optical system 21 and focused at 26a and 27a.
When the light is focused on the wafer 4, the light intensity (light amount) of the P-polarized light and the S-polarized light is adjusted to be a predetermined ratio. Increase the amount of light to 4. To do this, the bifocal element can be rotated around the optical axis, or a λ/2 plate can be inserted between the laser beam #lO and the radial grating 11 and it can be rotated around the optical axis. In other words, the light amount ratio between the polarized light reaching the reticle l and the polarized light reaching the wafer 4 can be adjusted to an optimum value. Now, the ±1st-order light LB, (parallel light flux) from the radial grating 11 is directed to the pupil plane of the bifocal optical system 21 which is a telesend link due to the action of the lens system 13, that is, ? ! The +1st-order light 1L B + enters the refractive substance 21b so as to be focused as a spot, and the +LBIP of P change and +L of S polarization are changed by the polarization components at the birefringence substance 21b.
B + =, and reaches the reticle l as a parallel light beam tilted by an angle determined by the diffraction angle with respect to the optical axis of the bifocal optical system 21. Similarly, the first-order light -LBl is separated into P-polarized light -LBIF and S-polarized light -LB,
+1st order light (+LBI) across the optical axis of the objective lens 21a
F, +L, B +s) and reaches the reticle l as a parallel beam of light at a symmetrical angle. Focal point 27 for P polarized light
Since a and the radial grating 11 are conjugate, the P-polarized first-order lights +LB2, -LB, , become almost parallel beams at the diffraction grating mark RM and intersect (image formation).
do. In FIG. 2, the lattice arrangement direction of the marks RM is in the left-right direction in the plane of the paper, and the inclination direction of each of the primary lights +LB+r and LBIF from the optical axis is also determined in the plane of the paper in FIG. A diffraction grating mark RM and a transparent window P0 are formed on the reticle l as shown in FIG.
The reticle l is irradiated with a size that covers , and. Third
The mark RM shown in figure (a) is in the X direction (lattice arrangement direction)
The diffraction grating mark WM on the wafer 4 also corresponds to this as shown in FIG. 3(b). Mark WM during alignment (or during exposure)
The window P0 of the reticle 1 is aligned with the position of the window P0 of the reticle 1. Now, the almost parallel S-polarized primary lights +LB, -L, B, emitted from the bifocal optical system 21.

is imaged (intersected) once at the focal point 26a in space, transmitted through the window P0 of the reticle 1, focused once as a spot light at the pupil EP of the projection lens 3, and then struck the diffraction grating mark WM on the wafer 4. Images are formed so that the light is incident from two different directions. This is because the focal point 26a (wafer surface) and the radial grating 11 are conjugate with respect to S-polarized light. Each of the nearly parallel S-polarized first-order lights LB, , -LB, , emitted from the projection lens 3 has a diffraction grating mark W.
The incident light is symmetrically inclined with respect to the lattice arrangement direction of M.

S-polarized primary light that reached wafer 4 +LB, l, -LB
Even if the angle +s is large, it does not exceed the numerical aperture of the projection lens 3 on the exit (wafer) side. Note that since the reticle 1 and the wafer 4 are arranged conjugately with respect to the radial grating 11, assuming that the laser beam LB is a parallel beam, each beam 1 LB, , -LB, □+L
B, , -LBll also becomes a parallel light beam.

Here, the behavior of the P-polarized primary lights +LB, . . . -LB, with respect to the mark RM on the reticle I will be described in detail with reference to FIG. FIG. 4 schematically shows the mark RM on the reticle 1, and assumes that P-polarized primary light 1L f3□ is incident on the mark RM at an angle θ. At this time, the specularly reflected light DIF of the primary light +LB, . . . on the reticle 1 is also reflected at an angle θ.

The fact that the light flux +LB, is incident at an angle θ means that the light flux LB,
, also means that the specularly reflected light DIP is incident on the reticle l at an angle θ in a direction opposite to that of the specularly reflected light DIP. Therefore, the grating pinch of the diffraction grating mark RM is P, the wavelength of the laser beam LB is λ,
Then, the pitch P and the angle θ are determined so as to satisfy the following equation (1), where n is an integer.

λ sin θ−−X n ・・・・・・ (
1) If this formula (1) is satisfied, the primary light + LB, , L
The diffracted light 104 of a specific order generated from the mark RM by the irradiation of B, .
It advances in the direction along the optical axis of the focusing optical system 21. Of course, other diffracted light 103 is also generated, but this is diffracted light 104.
go in a different direction.

By the way, the mark RM of reticle 1 receives light flux + from two directions.
L B 1F, -L B , , are irradiated so as to intersect with each other, and since both of the light beams are of the same polarization and emitted from the same laser light source 1o, two light beams +LB, , and -LB, , produce one-dimensional bright and dark fringes, so-called interference fringes. Assuming that the radial grating 11 is stopped, the step stripes are arranged at a predetermined pitch in the grating arrangement direction of the marks RM. The pitch of the interference fringes and the grating pinch of the mark RM are appropriately determined depending on the required detection resolution. Therefore, the diffracted light 104 from the mark RM is generated by the interference fringes irradiating the mark RM. Alternatively, by irradiating one of the light beams +LB+r, mark R
It is also thought that the diffracted light generated from M and the diffracted light generated from the mark RM due to irradiation with the other beam -LB, interfere with each other because they return along the same optical path (on the axis of the bifocal optical system 21). It will be done. In this way, mark RM has two different
When the light flux +LBIF, -LBIF is irradiated from the direction,
Interference fringes occur on the mark RM, but when the radial grating 11 is rotating, the interference fringes move (flow) in the direction of the grating arrangement of the mark RM.This is a radial grating. 11 primary light + LB
, , -LB, the dark field image of mark R on reticle L
This is because the image is formed on M. For this reason, the mark RM
By scanning the interference fringes (diffraction image projected by the bifocal optical system 21 of the radial grating 11, etc.) thereon, the diffracted light 104 periodically repeats changes in brightness and darkness. Therefore, the signal from the photoelectric detector 25 becomes a sinusoidal alternating current signal corresponding to the period of the change in brightness and darkness.

The above is exactly the same in the relationship between the diffraction grating mark WM on the wafer 4 and the S-polarized light beam +L B +-1-LB, and the diffracted light 105 is generated from the mark WM, which is It travels along the chief ray of the projection lens 3 and reaches the photoelectric detector 25 via the window P of the reticle 1. 2
The S-polarized light flux emitted from the focusing optical system 21 +LB, , -
LB, , are formed to intersect at the focal point 26a, but
The mark RM and window P of the reticle 1 are greatly defocused.

Now, it is assumed that the photoelectric detector 25 is placed conjugately with each of the marks RM and WM via the bifocal optical system 21, but in reality, as shown in FIG. An aperture plate 25' as shown in FIG. 3(C) is provided at a conjugate position, and the diffracted light 104, 105 transmitted through the apertures Ar and As of this mask member 25' is configured to be photoelectrically detected. Here, the aperture A is for taking out a diffraction image of the diffracted light 104 from the mark RM of the reticle 1, and the aperture A$ is for taking out the diffraction image of the diffracted light 105 from the mark WM of the wafer 1. Therefore, by separately providing the light-receiving surface of the photoelectric detector 25 after each aperture AP, As, the position of the reticle l can be detected by the mark RM, and the mark WM
It becomes possible to detect the position of the wafer 1 independently. still,
Aperture A has a P-polarized light flux of 10 LB,...LBI
An image of the mark RM on the reticle 1 illuminated by F is formed, but at the same time, the S-polarized light flux +LBI! , -LBl, reflected diffracted light also enters as background noise. For this reason, it is preferable to provide a polarizing plate for passing P-polarized light in aperture A, and a polarizing plate for passing S-polarized light in aperture A. In this way, each of the two photoelectric detectors 25 can detect light from the wafer and Crosstalk caused by mixing with light from the reticle is sufficiently reduced.

Here, we will analyze the photoelectric signal of the diffracted light 104 obtained through the aperture AP when the radial grating 11 is stopped.

If n-±1 is used in equation (1) above, the grating pitch P is the pitch of the reference grating of the radial grating 11 and the image formation through the lens system 13, pupil system 17A, and bifocal optical system 21. It is related to magnification.

(Similarly, the grating pitch of the mark WM on the wafer 4 is also
It is related to the grating pitch P of the mark RM and the imaging magnification of the projection lens 3. ) Now, the luminous flux ten L incident on the mark RM
The amplitude VR'' of the diffracted light generated by B, , is given by equation (2), and the luminous flux -LB,.

The amplitude VR- of the diffracted light generated by is given by (3) 2.

VR”=a-sin(φ+2 π-) −(2) VR-
= a'・s in (φ-27! -) - (3) In 22, P is the lattice pinch of mark RM, and X is mark R
This is the displacement amount of M in the lattice arrangement direction. Since the interference of these two diffracted lights VR" and VR- with each other is photoelectrically detected, the change in the photoelectric signal (amplitude of the diffracted light 104) is expressed as (4)
It is expressed as the formula.

VR"+VR-"=a"+a'snow+2-a-a'・C05(4tt
−) −(4) Here, a” + a” is the signal bias (
(DC component), and 2a·a′ is the amplitude component of the signal change.

As is clear from equation (4), the photoelectric signal changes sinusoidally when the radial grating 11 and the mark RM are relatively displaced in the grating arrangement direction. Each time the relative displacement 1tx becomes x=P/2 (half of the grating pinch), the signal amplitude changes by one cycle. On the other hand, the same is true for the diffracted light 105 from the mark WM on the wafer 4.
It is expressed as in equation (4). The alignment is then completed by moving the reticle 1 or the wafer 4 so that the phase relationship between these two photoelectric signals matches. However, as can be seen from equation 4), each signal is sinusoidal, and the detectable phase difference is within the range of ±180'', so reticle l and wafer 4 are pre-aligned with marks RM, W.
It is necessary to perform prealignment with an accuracy of I/2 or less of the grating pitch P of M. When the radial grating 11 is stopped in this way, the amplitude level of the obtained photoelectric signal changes sinusoidally only by moving the reticle 1 or the wafer 4.

By the way, when the radial grating 11 is rotating, the diffracted lights 104 and 105 become periodic (sinusoidal) brightness information, and the obtained photoelectric signal is sinusoidal even if the reticle 1 or wafer 4 is stationary. It becomes a wavy AC signal. Therefore, in this case, the photoelectric detector 1 shown in FIG.
Using the photoelectric signal (sine wave AC signal) from mark RM as a basic signal, the phase detection system 40 detects the phase difference φ between the diffracted light 104 from mark RM and the photoelectric signal (sine wave AC signal).

Similarly, the phase difference φ8 between the photoelectric signal of the diffracted light 105 from the mark WM and the basic signal is detected. Then, by finding the difference between the phase differences φ1 and φ, the amount of deviation between the reticle I and the wafer W in the X direction can be determined. This detection method is called the optical heterodyne method, and as long as the reticle 1 and the wafer 4 are within a positional error range of 1/2 of the grating pitch P, positional deviation can be detected with high resolution even when the reticle is stationary. 1
This is convenient for applying closed-loop position servo to prevent minute positional deviations from occurring while exposing the pattern on the resist on the wafer 4.

In this detection method, after alignment is completed by moving reticle 1 or wafer 4 so that φ and -φ1 become zero (or a predetermined value), reticle 1 and wafer 4 continue to move relative to each other at that alignment position. A servo lock can be applied to prevent this from happening.

In this embodiment, during step-and-repeat exposure, the movement of the wafer stage to each shot area on the wafer is performed based on the measured values of the interference system, and two light beams of 10L are
The mark WM is ±1/in the irradiation area of B,,,-LBls.
Once the positioning is performed with an accuracy of two pitches, the reticle stage or wafer stage can be servo-controlled based only on information from the phase detection system 40. At this time, the reticle stage and wafer stage are driven by a DC motor, and analog voltages corresponding to the phase differences φ and -φ1 are applied to the D/
It is also possible to generate this analog voltage using an A converter or the like and directly apply this analog voltage to the servo circuit of the DC motor as a deviation voltage. This servo is carried out until the end of exposure of the Shono HH region.

In this case, since the servo is not based on the measured value of the interferometer, it is possible to reduce minute fluctuations of the stage due to fluctuations in air density in the beam optical path of the interferometer. Therefore, when phase difference information that allows servo control is obtained from the phase detection system 40, the measured value of the interferometer on the wafer stage side is separated from the servo system on the wafer stage side, and the voltage applied to the morph on the wafer stage is changed. The voltage is set to zero, and the analog voltage described above is applied to the servo system on the reticle stage side.

In this way, during the exposure operation, the minute fluctuations that occur especially on the wafer stage side can be suppressed, making it possible to create a gentle drift-like slight movement, and by moving the reticle stage to follow at high speed, the reticle and wafer can be It is possible to keep the relative positional deviation of Therefore, even if the line width of the exposed pattern becomes thicker, there is no decrease in resolution, and extremely faithful transfer is achieved.

Here, the frequency of each photoelectric signal from the photoelectric detector 19.25 is compared with the rotation speed of the radial grating 11, and from the resolution of phase difference detection and the responsiveness of the photoelectric detector 19.25, it is 1KH2~ Approximately 100KHz is desirable.
Of course, if the photoelectric detectors 19 and 25 are of a high-speed response type, the signal frequency can be further increased, so that the mark position detection by phase difference detection can be performed with higher resolution.

In principle, the pitch of the interference fringes (dark-field image of the grating of the radial grating 11) that is larger than the mark RM is set to 1 degree l/2 of the pitch P of the mark RM, and the interference fringes that are formed after passing through the mark are also the same as the mark. The pitch is set to 1 degree 1/2 of the pitch of WM.

Furthermore, as is clear from FIG. 2, the beam ±LB (±LB, , ±t
, B IF) are all focused as spot light on the pupil conjugate plane (Fourier plane), so the diffracted light (interference beat signal) from mark RM (interference beat signal) lO4 and the diffracted light (
The interference beat signal) 105 is also focused as a spot light on the pupil plane EP and the pupil conjugate plane, and is focused on the object (reticle, wafer).
On the surface or the image surface conjugate thereto, the light beams are all approximately parallel.

In addition, two luminous fluxes +LB+s (+LB+r) and -LB+
s(LBIF) will have a frequency difference depending on the rotation speed of the radial grating 11. Therefore, instead of the radial grating 11, an acousto-optic modulator (AOM) can be used to generate two waves with a certain frequency difference, as disclosed in Japanese Patent Application Laid-Open No. 62-56818.
May produce two beams.

The technology underlying the present invention has been described above. Next, problems with this technology will be described with reference to FIG. 6(A). Figure 6 (A) shows the reticle 1, projection lens 3, and wafer 4.
This diagram schematically shows the relationship between mark R on reticle I and
The projection image of the pitch and grating axis of M onto the wafer 4 is
It is determined to be equal to the pitch and grating width of the marks WM above.

Now, in FIG. 6(A), mark RM of reticle 1
is the luminous flux +LB, , -LB, from two directions.

When irradiated by the mark RM, ±1st-order diffracted light LBR±1 (diffraction light 104) travels upward perpendicularly from the mark RM, and transmitted diffracted light (±1st-order diffracted light) LBw±1 travels downward perpendicularly to the mark RM. and occur at the same time. Transmitted diffraction light LB
Since w±1 will proceed along the principal ray of the projection lens 3, it will be reflected after reaching the wafer 4 perpendicularly, and will again pass through the mark RM part on the reticle l from top to top. It will be mixed with the original diffracted light LBR±1 from the mark RM. These diffracted lights LBR±1, LBw±,
Both are the same polarized light, and if they are mixed, a phase shift may occur during phase detection. Furthermore, Figure 6 (A)
LB, indicates the state of the zero-order light after the light flux -LB, , passes through the mark RM, LB,'' indicates the state of the zero-order light after the light flux +LB, passes through the mark, and these θ-order lights LB, , L.B.
After being reflected by the wafer 4, O' returns to the chrome surface C5 on the lower surface of the reticle 1. Furthermore, since the reticle 1 and the wafer 4 are in a largely defocused relationship under the alignment illumination light, the diffracted light LBw+ is not formed as an image on the plane of the mark RM (focal point 27a). Furthermore, the diffracted light 104 from the mark RM on the reticle 1 and the mark W on the wafer 4
Since the diffracted light 105 from M returns in the same optical axis direction, even if the aperture plate 25' was provided, there was a possibility that the optical signals of both would be mixed.

Therefore, in the embodiment of the present invention, a part of the configuration of the alignment system is changed as shown in FIG. Here, the same members as those in FIG. 1 are given the same reference numerals, and the guichroic mirror 22 is omitted.

In addition, in conjunction with this change in the apparatus, the pitch relationship between marks RM and marks WM was changed as shown in FIG. 7.

The mark WM on the wafer 4 remains as described above, and is set at exactly twice the pitch of the interference fringes formed on the wafer 4. On the other hand, the mark RM on reticle 1 is
In this embodiment, the pitch is set to be the same as the pitch of the interference fringes on the reticle l. That is, unlike the previous case, the pitch of the marks RM is 1/2 times larger to further widen the diffraction angle (higher frequency diffraction grating).

Therefore, the diffraction state when the pitch of the marks RM is reduced to 1/2 in this way will be explained with reference to FIG. 6(B).

Light flux +L that illuminates mark RM on reticle l from two directions
B, , -LB, , 0th-order light and ±1st-order light are generated from the mark RM in the reflection direction and the transmission direction, respectively. First, by irradiation with the light beam +LB, .

Similarly, by irradiation with the light flux -L81F, a light flux of 10LB, . . . , a 0th order light -LBR, and a +1st order diffracted light LBR, .

In this example, two luminous fluxes → -LBIP, LBtr
The incident angle of , the pitch of interference fringes on the reticle, and the MacRM
By appropriately determining the pitch, the ±1st-order diffracted light LBR±, generated from the mark RM, is determined to be in the same direction as the 0th-order light ±LBR. Therefore, the diffracted light traveling in the direction perpendicular to the mark RM is theoretically a 0.5th-order diffracted light, and the amount of this light is so small that it can be almost ignored compared to the first-order light.

On the other hand, the 0th-order light is also below the mark RM (on the transmission side).
±1st order light etc. are generated. First, the 0th-order light LB,, LB,'' that has passed through the mark RM behaves exactly the same as in the case of FIG. B W−+ is
It follows the same optical path as the luminous flux - the 0th order light LB of LBIP. Similarly, by irradiating the light flux -LB, , mark RM
The +1st-order diffracted light LBw, which travels downward, is the luminous flux +LBI
It travels along the same optical path as the zero-order light LBo' of F. The diffracted light generated downward perpendicularly to the mark RM is theoretically the 0.5th order diffracted light, and can be almost ignored. Also, mark RM
Since the second-order and higher-order diffracted light from the +/- first-order light is generated at a wider angle than the ±1st-order light, it can be ignored.

As is clear from the above, the ±1st-order diffracted light LBw, ..., LB w −+ that passes through the mark RM, is reflected by the wafer 4, and returns to the mark RM is the 0th-order light LB,
, LB, and degrees, the light is blocked by the chrome surfaces C3 on both sides of the mark RM due to the chromatic aberration of the projection lens 3, and does not return to the objective lens 21a, as explained in FIG. 6(A). There will be no.

Here, as shown in FIG. 7, the width in the X direction (pitch direction) of the window in the chrome surface C7 on which the mark RM is formed is WX, and as shown in FIG. 6(B), the width on the wafer 4 side is The amount of axial chromatic aberration is calculated as ΔL1 Two luminous fluxes on the wafer 4 side + LBIF (
+LB+s), LBIF (-LB, ,) When the incident angle of θ is the zero-order light LB, , LB, which becomes noise power
, ' and the ±1st-order light LBw,1, LBW-1, the following equation (5) may be satisfied.

WX≦2・ΔL・θ (5) On the other hand, the behavior of the 0th order light and ±1st order light from the mark WM on the wafer 4 is the same as the previous explanation, and from the mark WM The ±1st-order diffracted light 105 of
Defocus and pass through the transparent window next to objective lens 2.
1a.

Therefore, in this embodiment, the reflected diffracted light LBR at the mark RM
±, (104) and the reflected diffracted light 105 at the mark WM are condensed as subont light at positions laterally shifted from each other on the pupil plane or pupil conjugate plane of the system. That is, the diffracted light 105 is focused on the center on the pupil, and the diffracted light L B R
±1 will be located symmetrically around it.

Now, returning to the explanation of FIG. 5, the apparatus of this embodiment will be explained. In FIG. 5, the diffracted light (interference beat signal) 105 from the mark WM on the wafer 4 passes through the transparent part near the mark RM on the reticle, then travels backward along the optical axis of the objective lens 21a, causing birefringence. thing! (A member made into a concave-convex lens shape and bonded together) Passes through the beam splinter 21b, is reflected by the beam splinter 20, and passes through the pupil relay system 17B to the polarizing beam splinter P.
Reach BS. The diffracted light 105 is generated by irradiating the wafer 4 with an S-polarized light beam ±LB, and naturally preserves the S-polarized light. Therefore, 90 of the diffracted light 105
% or more is reflected by the polarization beam splinter PBS and reaches the photoelectric detection 2S53 via the lens system 50, aperture plate 51, and lens system 52. In FIG. 5, the broken line indicates the conjugate relationship of the pupils, and the aperture plate 51 between the lens systems 50 and 52
is conjugate with the mark WM on the reticle 1, that is, the wafer 4, and the aperture plate 51 has only an aperture AS through which only the image of the mark WM passes. The light receiving surface of the photoelectric detector 53 is arranged at the pupil conjugate, receives the focused spot light of the diffracted light 105, and outputs an alternating current photoelectric signal S corresponding to the intensity change (interference beat signal).

On the other hand, one optical information (-LBR,, LBH,,) 104A and the other optical information (+LBR,, LBR-,) 104B from the mark RM shown in FIG. 6(B) are mutually exclusive. It is an interference beat signal between 0th-order light and 1st-order light that have a frequency difference, and the transmitted light beams are +LB, , -LB,
, and reaches the polarizing beam splitter PBS via the bifocal optical system 21, beam splitter 20, and relay system 17B. As shown in Figure 6(B), the 0th order light -L
BRo and the +1st-order diffracted light LBH-+ have a frequency difference with each other, overlap each other in the same direction, and are the same polarized light (P-polarized light) component, causing interference, the 0th-order light +
The same applies to LBR and the first-order diffracted light L B H-. Therefore, more than 90% of the optical information 104A and 104B is transmitted through the polarizing beam splitter PBS, and the lens system 5
4, reaches photoelectric detectors 58A and 58B via mirror 55, aperture plate 56, and lens system 57. Lens system 54
The aperture plate 56 between and 57 has only an aperture AP that passes only the image of the mark RM on the reticle.
7a). Also two photoelectric detectors 58
The light receiving surfaces of A and 58B are arranged at pupil conjugate, and optical information 104
It receives each of the spot lights A and 104B and outputs AC signals St and Ss at the frequency of the interference beat signal, respectively.

These two signals St and Ss have the same signal properties, and either one may be sent to the phase detection system 40. However, in this embodiment, the optical information 104A, 104B is created by interference between the 0th-order diffracted light and the 1st-order diffracted light, so if the light intensity (light amount) of the 1st-order light and the 0th-order light differ greatly, it will be difficult to measure the phase difference. It is also possible that an offset occurs. Therefore, signals S2 and S.

After passing through an analog circuit that calculates the sum (or difference) of
It is preferable to measure the phase difference φ with the reference signal from the photoelectric detector 19. It is also possible to use a switching type so that one of the signals S2, S, or a combined signal is used.

Furthermore, the phase difference φ1 is also measured between the signal S from the photoelectric detector 53 and the reference signal, and finally the phase detection system 4
0, perform the calculation of φ, -φU, and reticle l and wafer 4.
All you have to do is find the positional shift.

Here, let us consider the phase relationship as in the case of FIG. Mark R of P-polarized primary light +LB, -LB,
The relationship with M is expressed by the above equation (1). Therefore, by setting the pitch P appropriately, the interference fringes generated on the reticle 1 by the luminous fluxes +LBIF and -LBIF are the ±1st-order lights (LBR,,, LBR-,) when the mark RM is irradiated, and the luminous flux +LBIF , , - Set to return again in the direction of LBIF. In other words, 1 generated by irradiation with luminous flux +LB, .
The secondary light LBR4I is returned in the direction of its luminous flux +LB, .

In this way, since the interference beat between the 0th order light (-LBR,, +LBR,) and the 1st order light (LBR-+, LBR-,) is photoelectrically detected, the reference signal (the output of the photoelectric detector 19 ) and the signal SZ (or Sl, or the composite signal) changes by exactly one period when the relative displacement IX of the mark RM shifts by a pitch P (same as the pitch of the interference fringes on the reticle). However, the pitch of mark RM is ±LB
Since the ±1st-order light (LBR±1) is returned in the same direction as , , , it has to be reduced to 1/2 compared to the previous method [Figure 6 (A)].As a result, , the phase difference between the reference signal and the signal S, the reference signal and the signal S2 (or Ss)
When the displacement (2) of the wafer 4 and the amount of displacement of the reticle 1 are both the same, the phase difference between the two changes by the same amount. This means that the phase difference φ and the phase difference φ8 can be directly subtracted, and no special conversion is required.

As described above, according to this embodiment, the reticle line and
Pl is the pitch of the space-like marks RM, and P is the pinch of the line-and-space marks WM on the wafer.
When the pitch of the interference fringes on the reticle due to the two light beams ±LB, is P2, and the magnification of the projection lens 3 is 1/M (M = 5 in the case of 115 reduction), the relationship of the following formula is satisfied. Established.

Pt = P, = 172 - M - P, (however, the duty of marks RM and WM of the diffraction grating is l:1) Therefore, in the pupil of the alignment system or in the vicinity, the diffracted light 105 from mark WM and the mark RM Diffracted light 104A, 104B
This enables phase difference detection with an extremely good S/N ratio.

Furthermore, as shown in Figures 1 and 2, when using a rotating radial grating II as a frequency shifter, even if there is some unevenness in its rotational speed, the two signals (reference (signal and measurement signal) Only the frequencies of the same signal fluctuate together, but the phase difference itself does not change.

This means that precise speed control of the motor 12 that rotates the radial grating 11 is not necessary.

Also, when configuring a frequency shifter using AOM,
Since the modulation wave (ultrasonic wave) to the AOM can be easily changed, after switching to several modulation frequencies (switching the frequency of the optical beat) and sequentially detecting the phase difference for the same mark, , one with a desired resolution may be selected. Alternatively, the optical beat frequency (the frequency difference between the two luminous fluxes +LB and -LB) may be changed according to the required mark position detection resolution.

In addition, the two luminous fluxes that illuminate the mark RM (WM) ±LB
, the incident angle of P (±LB1s) is the luminous flux +-LBLB,
The circle can be adjusted freely by changing the position within the pupil so that they both maintain the same distance from the pupil center. For this purpose, an optical block (parallel plane glass, wedge) is installed between the spatial filter 15 and the beam splitter 14 in FIG. prism, etc.) may be provided. Further, the photoelectric detector 53 in FIG. 5 may be placed immediately after the aperture AS of the aperture plate 51, and the photoelectric detector 58A,
Only one 58B may be placed immediately after the aperture plate 56.

Furthermore, even if a pattern (alignment mark, actual device pattern, etc.) exists at a position on the wafer 4 that is conjugate with the mark RM, according to this embodiment, the diffracted light 10 from the mark RM
4A, 104B, and optical information from wafer l (diffracted light 10
5, etc.), and can be detected completely separately from the photoelectric detector 58A.
, 58B do not detect the pattern on the wafer 4, so the mark RM position on the reticle 1 may be shared when replacing the next mark. In other words, in FIG. The mark WM may be provided to be located next to the mark RM, and the mark WM on the wafer during exposure of another layer may be located directly below the mark RM. In this case, photoelectric detectors 53.58A, 58B
must all be placed on or near the pupil conjugate plane.

As described above, the alignment system of this embodiment shown in FIG.
Only the position measurement in the dimensional direction is performed, but if marks RM for the X direction and mark RM for the X direction are provided on each of the two orthogonal sides around the circuit pattern area of the reticle 1, and an alignment system corresponding to each is placed. , two-dimensional alignment between the reticle 1 and the shot area on the wafer 4 is possible.

By the way, the alignment system of this embodiment does not require the provision of an optical system for correcting chromatic aberration between the reticle 1 and the projection lens 3. A major feature is that mark position detection, alignment, and servo operations can be performed continuously even during exposure operation by the action of the dichroic mirror 22 provided in the die-pie-die.
During alignment, when the size of the circuit pattern area (shot area on the wafer) of the reticle 1 changes, a part of the alignment system (in this case, the objective lens 21) changes accordingly.
A configuration in which the birefringent material 21b, etc.) is movable is adopted. Furthermore, since the mark RM on the reticle l is optically detected via a glass plate or a quartz plate as a reticle substrate, the upper surface of the reticle l ( Part of the light reflected on the glass surface) and the light generated from the mark RM on the pattern surface (bottom surface) (+LBR,, -LBH,, LBH12, L B R-
+) may cause interference, which will give a phase offset to the diffracted lights (optical beat signals) 104A and 104B.

Therefore, as shown in FIGS. 8 and 9, the phase offset due to the thickness of the reticle 1 is measured in advance so that the offset can be corrected by the phase detection system 40. Figure 8 (A) shows a case where a bifocal optical system is provided as a part of the conventional die-pie-gui alignment system, and the tip mirror M
and only the bifocal optical system 21 is shown. The mirror M and the bifocal optical system 21 are movable parallel to the reticle l according to the position of the mark RM as indicated by the arrow in the figure.

Figure 8 (B) schematically shows how the phase offset amount Δφ varies due to changes in the glass thickness of reticle 1. Even within the same reticle, the offset amount Δφ may vary slightly due to uneven thickness depending on the position. It's changing. The width of this slight fluctuation is the average absolute offset amount Δφ. Although it is often small compared to , it is often not negligible in alignment methods that use interference. Therefore, we measured the amount of phase offset at the actual alignment position and used that data at the 9th position.
The cent is placed in the correction data section 40D shown in the figure. In FIG. 9, phase difference detection units 40A and 40B are each photoelectric detector 19.
The phase difference φ1, φ between the signals S, , S, , S, (or the composite signal of S2 and 83) with respect to the reference signal S, from Power phase offset! (For example, Δφ.) is added or subtracted. Moreover, since this phase offset amount Δφ corresponds to the thickness of the glass above the alignment mark RM, it is the phase offset amount Δφ when the thickness is a certain Do. For the offset amount Δφ7 of various reticles, enter the glass thickness at the alignment position, Δφ, = K・Δφ. - It is also possible to obtain by approximate calculation of D, /D, (K is a constant).

In addition, to actually measure the offset amount Δφ7 at the alignment position, place the reference mark (grid-like) FM on the wafer stage 5 below the mark RM, and observe the reticle mark RM and the reference mark FM simultaneously with exposure light. Then, the deviation ΔX0 between the mark RM and the reference mark FM is precisely measured. In this case, the illumination with the exposure light may be performed by coaxial epi-illumination via the bifocal optical system 21, or the fiducial mark FM itself may be caused to emit light with the exposure light. The mark RM and the reference mark F are formed through the bifocal optical system 21 under illumination by exposure light.
When detecting M, the illumination light of the exposure light (or the imaging light from the mark) may be limited to one polarization component.

Next, the phase detection system 40 calculates the deviation ΔX1 between the mark RM and the reference mark FM in the same state with the correction data set to zero.
Find it with If ΔX0−Δx7 is calculated here, it corresponds to the phase offset amount Δφ7 to be actually measured.

Further, the same effect can be obtained even if the mark RM on the reticle l or the mark WM on the wafer 4 is one bar mark or two or three bar marks. In this case, the optical information from the bar mark is not diffracted light with uniform directionality, but
Although it is similar to a kind of scattered light, it also enables an alignment method that similarly utilizes interference, and has the advantage that the mark forming area can be made extremely small.

〔Effect of the invention〕

As described above, according to the present invention, since the reticle mark and the wafer mark are irradiated separately, apertures are provided at positions conjugate to the reticle and wafer positions existing in the light receiving system after division based on polarization characteristics, and both signals are can be completely divided. Also, by changing the pitch of the wafer mark and the pinch of the reticle mark, it is possible to capture signals at different pupil positions (the wafer is near the center and the reticle is at ±LB), so the inclination of the telecenter allows the mixing of both signals. Sometimes it is possible to separate.

Furthermore, the mark pitch and the light-shielding band CR are provided so that even if the light transmitted through the reticle mark is reflected back to the wafer, it will not be confused with the signal, so that it is possible to prevent the phase of the reticle signal from changing due to noise. This allows highly accurate alignment of the reticle and wafer.

[Brief explanation of drawings]

FIG. 1 is a diagram showing the configuration of a projection exposure apparatus using the technology underlying the embodiment of the present invention, FIG. 2 is a diagram schematically showing the configuration of the alignment system in FIG. 1, and FIG. 3 is a diagram showing the configuration of the alignment system in FIG. (a), (
b) is a plan view showing an example of a mark on a reticle and a mark on a wafer, FIG. 3(c) is a plan view showing the structure of an aperture plate in the alignment system, and FIG. A diagram showing how the interference beat signal) is generated,
Fig. 5 is a diagram showing the configuration of an alignment system according to an embodiment of the present invention, Fig. 6 (A) is a schematic diagram explaining a problem with the basic technology, and Fig. 6 (B) is a diagram showing a solution to the problem. A schematic diagram explaining the operation of this embodiment, FIG. 7 is a plan view showing the relationship between the reticle mark and wafer mark in this embodiment, and FIG. 8 shows an offset occurring during phase difference detection according to changes in alignment position. Figure 9 is a block diagram showing the configuration of the phase detection system that corrects the offset.
FIG. 10 is a diagram illustrating an example of an alignment method for a conventional projection exposure apparatus. (Explanation of symbols of main parts) 1, R... Reticle, 3, PL... Projection lens, 4, W... Wafer, 10... Laser light source, 11... Radial grating, 15 ... Spatial filters 17A, 17B... Relay system, 18... Reference grating, 19.25.53.58A, 58B... Photoelectric detector,
21... Bifocal optical system, 22... Dichroic ink mirror 30... Exposure light source, 40... Phase detection system, 25. 51.57...Aperture (anti, 104, 1
04A, 104B...Diffracted light from reticle mark, 105...Diffracted light from wafer mark, RM...Reticle mark, ~VM...Wafer mark, +-LB,, -LB...2 A beam of light that illuminates a mark from any direction.

Claims (5)

[Claims] (1) An apparatus that images and projects a pattern formed on a mask onto a photosensitive substrate via a projection optical system, the apparatus comprising a first mark formed on the mask and a second mark formed on the photosensitive substrate. A light source that outputs coherent illumination light in a wavelength range in which the projection optical system generates a predetermined amount of chromatic aberration; The illumination light is
an illumination light orienting means that orients the illumination light so as to irradiate the mark from two directions, and also orient the illumination light so that it irradiates the second mark from two directions via the projection optical system; A first method in which the reflected 0th-order diffracted light and the higher-order diffracted light reflected in approximately the same direction as the 0th-order diffracted light interfere with each other.
Of the optical information and the diffracted light generated from the second mark,
a mark detection optical system that detects second light information passing through the mask in a direction different from the first light information through the projection optical system; 1. A positioning device comprising: photoelectric detection means for separating and individually photoelectrically detecting two optical information. (2) The illumination light orienting means is arranged to illuminate the mask via the mark detection optical system with first and second illumination lights containing the same polarization components, and 2. The apparatus according to claim 1, wherein the two illumination lights are directed so as to pass through positions shifted by a predetermined interval in a pupil plane of the mark detection optical system. (3) The mark detection optical system includes a bifocal element having focal points at two positions corresponding to the amount of axial chromatic aberration of the projection optical system by different polarization components;
or a telecentric objective lens disposed near the first mark, first and second illumination lights that irradiate the first mark, and first and second illumination lights that irradiate the second mark via the projection optical system.
3. The apparatus according to claim 2, wherein the illumination light is separated by polarization by the bifocal element. (4) The convergence in the position measurement direction of the transparent area forming the first mark is W_x, the amount of axial chromatic aberration on the photosensitive substrate side of the projection optical system is ΔL, and the first illumination light and the second illumination light are measured. Claim 2, characterized in that the width W_x is determined so as to satisfy W_x≦2・ΔL・θ, where θ is an incident angle on the photosensitive substrate when tilted symmetrically in a direction. or the device according to paragraph 3. (5) Each of the first mark and the second mark
The pitch or width of a grating image when the first mark is projected onto the photosensitive substrate by the projection optical system, and 3. The apparatus according to claim 2, wherein the second marks have different pitches or widths.