JP2814520B2 - Position detecting apparatus, position detecting method, and projection exposure apparatus - Google Patents

Description

The present invention relates to a position detection apparatus using a diffraction grating formed on an object such as a semiconductor element, and is used particularly when aligning an exposure apparatus for lithography. The present invention relates to an apparatus for detecting a relative positional relationship between a mask having an original pattern and a substrate onto which the original pattern is transferred.

[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 for only the KrFI xima 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 illumination light for alignment is different from the wavelength of the exposure light), optically detects a one-dimensional diffraction grating mark formed on the wafer or reticle, 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 for position detection using a diffraction grating mark, and have been put to practical use. 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
A heterodyne method for giving a constant frequency difference to the book, 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.

[Problems to be solved by the invention]

In the position detection system employing the above-described interference fringe alignment method, as a result of an experiment, it has become clear that the symmetry of the incident angles of the two beams that irradiate the diffraction grating from two directions is extremely important. If the symmetry of the incident angle is very small and unstable, the telecentricity is deteriorated, and the ratio of the detection error to the measurement resolution becomes a serious problem because the measurement resolution is originally high. The higher the measurement resolution, the higher the accuracy of the alignment.However, if the ratio of the detection error is large, the alignment accuracy is ultimately limited by the width of the error. The merit that was there is gone.

There are various factors that can degrade the symmetry of the incident angles of the two beams and destabilize them. Typically, there are a problem due to the device structure and a problem due to the object (grating mark) side at the time of position detection. Divided into

As a problem on the apparatus side, in an exposure apparatus or the like, it is necessary to have a structure in which an objective lens and a mirror of an alignment system can be moved in response to a change in a mark position, and a diffraction grating is required from a light source for producing two beams. Various optical elements are interposed in the optical path up to this point, and these also cause more or less manufacturing errors and assembly errors.

The problem at the time of measurement on the object side is mainly due to the symmetry of the incident angle, but within the area where the two beams intersect,
The partial surface of the diffraction grating to be measured is slightly displaced in the direction of the optical axis each time measurement is performed. This means that when sequentially measuring a plurality of diffraction grating marks on a wafer as in the alignment of a wafer or the like, a sine error occurs for each mark. Many of the causes are due to flatness (warpage, curvature, etc.) of the wafer surface.

Similar problems exist with conventional TTR alignment systems that focus the laser beam in a slit shape and relatively scan bar marks on the wafer to detect the scattered and diffracted light photoelectrically. However, the influence was small compared to the total detection accuracy in consideration of noise error, waveform distortion, and the like at the time of mark detection, and was hardly noticed. In the method of condensing a laser beam on a slit-shaped spot, the shape of the beam cross section is made elongated in the direction perpendicular to the slit longitudinal direction of the spot on the wafer within the pupil plane of the projection lens or alignment objective lens. Become. Therefore, in the detection direction of the slit-like mark (the width direction of the slit-like spot), the numerical aperture of the beam is larger than that in the direction orthogonal to the direction. In general, the larger the numerical aperture of the beam, the smaller the diameter (width) of the spot light and the higher the detection resolution. However, the width of the beam waist in the optical axis direction is correspondingly reduced. As a result, stability at the time of mark detection is eventually lost. Furthermore, in the method of condensing a beam on a spot, the wavefront of the beam changes in the direction of the optical axis with the center of the beam waist sandwiched. If scanning is performed while displacing in the direction, an original sharp photoelectric signal waveform cannot be obtained.

The present invention has been made in view of the above-described problems, and has as its object to stably maintain the symmetry of the incident angle of two beams in a two-beam interference type position detecting device used in the interference fringe alignment method.

[Means to solve the problem]

In the present invention, focusing on the fact that beams that irradiate the diffraction grating on the object from the two directions via the objective optical system are separated by a predetermined distance in the pupil plane of the objective optical system, two beams passing through the pupil plane The first optical member, which is displaced by the same amount in the grating arrangement direction in the pupil plane while keeping the distance substantially constant, is provided in the optical path between the light source for beam generation and the objective optical system. The first optical member adjusts the symmetry of the incident angle of the two beams to the object, and includes a gap adjusting means for displacing the gap between the object and the objective optical system in the optical axis direction, and a diffraction grating. Control means for adjusting the first optical member in accordance with a detection signal when photoelectrically detecting the generated diffracted light (interference light) via the objective optical system, whereby the symmetry of the two beams is stabilized. I tried to keep it.

(Operation)

In the present invention, the two-beam interference type can expand the effective detection range in the optical axis direction several times to several tens times or more as compared with the conventional imaging type or beam spot type position detecting device. Attention is paid to the point, that is, the point that the wavefront created by the intersection of the two beams within the effective detection range is substantially constant at each position in the optical axis direction.

This will be described with reference to FIG. Second
In the figure, two parallel beams LB ₁ and LB ₂ are projected so as to intersect on a virtual plane IP, and a one-dimensional diffraction grating is projected on the plane IP.
This figure shows how WP (duty is 1: 1) are arranged in parallel.
The wavelength of the _two beams LB ₁ and LB ₂ is λ, the pitch of the diffraction grating WP is P, the optical axis of the objective optical system that projects the _two beams LB ₁ and LB ₂ is AXa perpendicular to the plane IP, and the beam LB ₁ , .theta.a incident angle of LB ₂ a by respective interposing the optical axis AXa, When .theta.b, in order to generate the optical axis AXa parallel to interference light (diffracted light) BTL from the diffraction grating WP is satisfy the equation (1) Need to be

When this equation (1) is satisfied, a one-dimensional interference fringe IFw at a pitch of 1 / 2P is set in an arbitrary plane parallel to the plane IP in a spatial region where the _two beams LB ₁ and LB ₂ intersect. Is made.

Each bright and dark fringe of the interference fringe IFw is set to be parallel to each grating element (bar pattern) of the diffraction grating WP. What is important here is that the incident angles θa and θb
And the symmetry is maintained with respect to the optical axis AXa perpendicular to the optical axis AXa. In FIG. 2, it is shown as θa = θb, and interference fringes formed at two places, a space above and below the diffraction grating WP.
The phase of IFw in the pitch direction does not laterally shift, and the phase information is stored identically in any plane in the vertical direction. That is, the line lk connecting the same phase point between the interference fringes on the upper surface side and the lower surface side is perpendicular to the plane IP.

Here, if the symmetry of the incident angles θa and θb is lost, the line lk
Is tilted by Δθ so as to be exactly a bisector of the optical principal lines of the beams LB ₁ and LB ₂ . Therefore, when the diffraction grating WP is displaced by ΔZ in the vertical direction, interference fringes IF formed on the diffraction grating WP are generated.
The same phase point of w is relatively displaced by sin Δθ in the grating arrangement direction, and an error of sin Δθ occurs in the position detection of the diffraction grating WP.

In the above description, the homodyne method in which the _two beams LB ₁ and LB ₂ do not have the frequency difference Δf has been described. The importance of preserving the same phase point is exactly the same as described above. Here, the speed V is expressed by the equation (2), where the pitch of the interference fringes IFw is P '(= 1 / 2P).

V = Δf · P ′ (2) By the way, the symmetry of the incident angles θa and θb is broken, and the line lk
Is tilted by Δθ with respect to the optical axis AXa, a tilt also occurs in the direction in which the interference light BTL is generated, which may cause inconvenience during photoelectric detection. It returns to the first-order light of the high-order diffracted light generated from the diffraction grating WP than the irradiation of the beam LB ₁ is vertical, the primary light of the high-order diffracted light generated by irradiation of the beam LB ₂ to return to the vertical The interference light BTL is created by the
The coaxiality of these two primary lights is lost, and in extreme cases, the interference light BTL cannot be obtained.

Now, let the intersection angle of the _two beams LB ₁ and LB _{2 be} θ (θa + θ
b), assuming that the beam width (diameter) is D, two beams LB
₁ , the maximum width Z in the optical axis AXa (Z-axis) direction of the intersection space area of LB ₂
m is given by equation (3).

Further, if sin (θ / 2) = n · λ / P from the above equation (1), the maximum width Zm can be easily calculated.

As an example, beams LB ₁ and LB ₂ have a width of 100 μm and a wavelength of 632.8
nm He-Ne laser and the pitch P of the diffraction grating WP is 8 μm
(4 μm line and space), when n = 1, 2
The incident angle θ / 2 of the beams LB ₁ and LB ₂ is θ / 2 = sin ⁻¹ (0.6328 / 8) {4.5368}, and the maximum width Z
m is Becomes However, since the diffraction grating WP has a certain width in the grating pitch direction, and the interference fringes IFw need to be stably generated within the width, the actually effective range is the width Zl as shown in FIG. Limited. However, the width Zl is about 300 μm even when estimated under the above conditions, that is, 150 μm above and below the surface IP.
m was obtained. In addition, within this width Zl, the waveform of the interference fringe IFw is reproduced extremely accurately in any plane, and it can be seen that the wavefront does not change.

Here, comparing with the conventional spot scanning method, the effective beam waist storage distance (depth of focus) of the He-Ne laser which forms an image at 2 μm on the object is obtained by reducing the numerical aperture of the beam. Is about 8 μm, and the spot size on the object becomes about 50 μm in a state where ± 150 μm is defocused. Further, in the conventional spot light imaging, since the beam is condensed by the objective optical system and the projection lens, the shape of the wavefront changes sequentially in the optical axis direction.

In the present invention, focusing on the fact that the two-beam interference method has an extremely wide stable detection range (Zl) in the optical axis direction, the symmetry of the incident angles of the _two beams LB ₁ and LB ₂ is extremely enhanced. It is to detect with accuracy.

〔Example〕

Next, a configuration in which the position detecting device according to the embodiment of the present invention is incorporated in a projection type exposure apparatus (stepper) will be described.
FIG. 3, FIG. 4, FIG. 4, FIG. 5, FIG. 6, and FIG. 7 will be described, but the basic configuration is the same as that disclosed in JP-A-63-283129. It is.

In FIG. 1, a reticle 1 having a predetermined circuit pattern and a diffraction grating mark for alignment is held on a reticle stage 2 which can move two-dimensionally. Each pattern on the reticle 1 is imaged on the wafer 4 under exposure light by the projection lens 3 which is telecentric on both sides. However, the projection lens 3 is satisfactorily corrected for chromatic aberration with respect to the wavelength of illumination light for exposure (g-line, i-line, etc.), and the reticle 1 and the wafer 4 are arranged so as to be conjugate to each other with respect to the wavelength of exposure. You. Also, reticle 1 on wafer 4
A diffraction grating mark similar to the grating mark formed in the above is formed. When the wafer 4 moves two-dimensionally in the x and y directions in a step-and-repeat manner and is attracted onto a stage 5 that slightly moves in the z direction, when the transfer exposure of the reticle 1 to one shot area on the wafer 4 is completed. Is stepped to the next shot position. A part of the reticle stage 2 has x in the horizontal plane of the reticle 1.
A movable mirror 6 that reflects a laser beam from a laser light wave interferometer (hereinafter referred to as an interferometer) 43 for detecting the position in the direction, the y direction, and the rotation (θ) direction 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 for reflecting a laser beam from an interferometer 45 for detecting the position of the wafer 4 in the x and y directions in a 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 and the Z-axis movement of the wafer stage 5 are performed by a drive motor 46.

By the way, the illumination system for exposure is a mercury lamp 30, an elliptical mirror 3
1, input lens group 3 including condenser lens and interference filter
2, Optical integrator (fly-eye lens) 3
3, the mirror 34, the main condenser lens 35, the dichroic mirror 22, and the like. The dichroic mirror 22 is inclined at an angle of 45 ° above the reticle 1, reflects the exposure light from the condenser lens 35 vertically downward, and irradiates the reticle 1 uniformly. The dichroic mirror 22 has a reflectance of 90% or more with respect to the wavelength of the exposure light, and has a transmittance of 50% or more with respect to the wavelength of the illumination light for alignment (wavelength longer than the exposure light). .

Next, the alignment system of this stepper will be described. The irradiation light (beam) LB for alignment is a laser light source
The two beams emitted from 10 are incident on a two-beam frequency shifter 12 including two acousto-optic modulators (AOMs), and have different frequencies from each other and include linearly polarized beams orthogonal to each other.
Converted to LB ₁ and LB ₂ . The frequency difference corresponds to the frequency difference between the high-frequency signals SF ₁ and SF ₂ driving the two AOMs. Two parallel beams L from the frequency shifter 12
After B ₁ and LB ₂ are reflected by the beam splitter 14, pass through the pupil relay system 17 A, and pass through the beam splitter 20, the parallel flat glass 50 as the first optical member of the present invention
The light passes through the tiltable lens and enters the bifocal optical system 21. 2
The focusing optical system 21 integrates a birefringent substance (a lens such as quartz or calcite) 21b arranged conjugate with a pupil EP of the alignment system, that is, a pupil EP of the projection lens 3, and a telecentric objective lens 21a for a microscope or the like. The polarization components of the beams LB ₁ and LB ₂ (dichroic mirrors)
On the other hand, light having a plane of polarization parallel to the plane of the drawing is defined as P-polarized light, and light having a plane of polarization perpendicular thereto is defined as S-polarized light. )
Different powers depending on the power. Here, it is assumed that the laser light source 10 oscillates laser light of orthogonal linear polarization. For this reason, the beams LB ₁ and LB ₂ composed of one polarized light (for example, P-polarized light) component emitted from the bifocal optical system 21 are applied to the reticle 1
The beams LB ₁ and LB ₂ composed of the other polarized light (for example, s-polarized light) components are formed (crossed) at the focal point 27 a coincident with the pattern surface on the lower surface of the reticle 1. I do. The other focal point of the bifocal optical system 21, that is, the surface conjugate with each of the focal points 26a and 27a on the side of the laser light source 10, is
It is present in the beam frequency shifter 12. Here, the distance in the optical axis direction between the two focal points 26a and 27 of the bifocal optical system 21 corresponds to the amount of chromatic aberration on the reticle 1 side of the projection lens 3 at the wavelength of the alignment laser light. The focal plane 26a in this space is conjugated with the imaging plane 26b coincident with the surface of the wafer 4 by the projection lens 3, and the focal plane 27a (reticle pattern plane) is spatially below the surface of the wafer 4 by the projection lens 3. Becomes conjugate with the imaging plane 27b far away from With imaging surface 26b
The interval 27b corresponds to the amount of chromatic aberration on the wafer 4 side of the projection lens 3. Here, the distance between the imaging planes 26b and 27b is Dw,
Assuming that the distance between the focal planes 26a and 27a is Dr, and the projection magnification of the projection lens 3 is 1 / M (usually M is one of 1, 2.5, 5, and 10), generally, Dr = M ² -There is a relationship of Dw.
As the wavelength of the laser light for alignment becomes farther from the wavelength of the exposure light, Dw,
Dr grows up. The depth of focus of this type of projection lens is extremely shallow, about ± 1 μm, and the interval Dw may reach several tens μm depending on the wavelength of the alignment illumination light. The illumination light (laser light) for alignment is
Although it is desirable that the wavelength has little sensitivity to the resist applied to the resist, the condition is not necessarily satisfied in the present invention. This is 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 may cause a large distortion particularly to the optical information itself from the diffraction grating mark on the wafer 4. is there. Therefore, it is more important to give priority to determining the optimal illumination light for alignment in consideration of the aberration. Therefore, when the alignment illumination light irradiates the resist for a long time (for example, 1 minute or more), the wavelength may be weakly sensitive such that the resist is exposed (a thinning occurs after development). In this case, two beams L
A shutter that opens during alignment is provided in the light transmission path of B ₁ and LB ₂ .

The two P-polarized beams LB ₁ and LB ₂ of the laser beam for alignment are incident on the diffraction grating mark portion of the reticle 1 at the focal plane 27a from two directions at an angle formed by the beams LB ₁ and LB _2. And form an image. Also, two S-polarized beams LB ₁ and LB from the focal plane 26a transmitted through the transparent portion of the reticle 1.
Numeral ₂ enters the diffraction grating mark portion of the wafer 4 on the focal plane 26b through the projection lens 3 at two angles from the beams LB ₁ and LB ₂ to form an image. The reflected diffracted light (interference light BTL) from the diffraction grating mark of the reticle 1 is reflected by the beam splitter 20 via a dichroic mirror 22, a bifocal optical system 21, and a plane-parallel glass 50, and then a pupil relay system 17
Only the diffracted light (BTL) traveling on the axis is filtered by the spatial filter 23 arranged on the pupil conjugate plane (Fourier plane) through B, and furthermore, the photoelectric detector 25
Reach The diffracted light (interference light BTL) reflected from the diffraction grating mark on the wafer 4 returns to the original optical path via the projection lens 3 and passes through the transparent portion of the reticle 1, and passes through the dichroic mirror 22, the bifocal optical system 21, The light reaches the photoelectric detector (light receiving element) 25 through the flat glass 50, the beam splitter 20, the pupil relay system 17B, the spatial filter 23, and the condenser lens 24. The spatial filter 23 is disposed at a position substantially 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 or near the pupil, and specularly reflected light from the reticle 1 or the wafer 4 ( 0th order light)
It is set so that only light diffracted perpendicularly (in the direction of the normal to the plane) through the diffraction grating of the reticle 1 or the wafer 4 passes.
In front of the photoelectric detector 25, the reticle 1, the wafer 4 via the bifocal optical system 21, the pupil relay system 17B, and the lens 24.
And an aperture plate 25 'arranged conjugate with each of the above.

Now, the photoelectric signals DSr and DSw obtained from the photoelectric detector 25 are
Beam LB for irradiating reticle 1 or wafer 4 from two directions
₁ , because the interference fringes created by LB ₂ will be irradiated on each diffraction grating mark so as to flow in the pitch direction,
A high-frequency drive signal SF ₁ becomes a sinusoidal AC signal (beat frequency) having a frequency corresponding to the difference Δf between SF. By the way, the two beams LB ₁ and LB from the two-beam frequency shifter 12
₂ is transmitted through the beam splitter 14, and the pupil (Flier surface)
Lens system (inverse Fourier transform lens) for converting image to image plane 16
Forms (crosses) on the reference diffraction grating 18. The reference diffraction grating 18 is fixed on the device. In this diffraction grating 18, a beam LB ₁ and beam LB ₂ is incident from two directions at a predetermined intersection angle. The photoelectric detector (light receiving element) 19 receives diffracted light (or interference light) other than the zero-order light transmitted through the reference grating 18 and outputs a sinusoidal photoelectric signal DR. This photoelectric signal DR has a frequency proportional to the difference frequency between the _two beams LB ₁ and LB ₂ and serves as a reference beat signal.
The phase detection system 40 receives the reticle-side photoelectric signal DSr from the photoelectric detector 25, the wafer-side photoelectric signal DSw, and the photoelectric signal DR from the photoelectric detector 19, and receives both signals DSr, D based on the signal DR.
Detects the phase difference on the Sw waveform. Detected phase difference (±
(180 °) uniquely corresponds to the relative positional deviation within 1/2 of the grating pitch of the diffraction grating marks formed on the reticle 1 and the wafer 4, respectively. The main control system 41 controls the drive motors 42 and 46 based on information on the detected phase difference (position shift amount), position information from each of the interferometers 43 and 45 obtained via the servo system 44, The relative positioning (alignment) between the reticle 1 and the wafer 4 is performed. In the description of FIG. ₁ , the two beams LB ₁ and LB ₂ are shown to intersect in the plane of the paper. However, actually, the two beams LB ₁ and LB ₂ are mutually inclined in a plane perpendicular to the plane including the optical axis AX of the projection lens 3. . The parallel flat glass 50 is driven by the drive control system 52 so that the two beams L
B ₁ and LB ₂ are tilted and driven so as to move in the same direction in a plane including the optical axis (AXa) of the objective lens 21b.

In the above overall configuration, a part of the alignment optical system, particularly the bifocal optical system 21, can be moved to an arbitrary position in accordance with the arrangement of the alignment mark on the reticle 1, and the mark detection can be performed regardless of the mark arrangement. Is possible. More on that later. Furthermore, since the exposure light and the alignment illumination light are separated by the dichroic mirror 22 obliquely provided above the reticle 1, the mark can be detected even during the exposure operation. This means that even if the alignment between the reticle 1 and the wafer 4 is out of order due to some disturbance during the exposure, it can be detected immediately at that time. Further, it means that the positioning servo between the reticle stage 2 and the wafer stage 5 can be executed in a closed loop even during the exposure operation based on the phase difference information from the phase detection system 40. Therefore, the line width of the exposed resist pattern does not increase due to slight image blur. The light source of the exposure light may be replaced with an excimer laser light source other than the mercury lamp.

Next, the detailed configuration of only the alignment system and the principle of alignment will be schematically described with reference to FIG. Third
In the figure, the dichroic mirror 22 and the beam splitter 14 are omitted for simplification, and the same members as those in FIG. 1 are denoted by the same reference numerals. Third
The arrangement of the parallel plane glass 50 in the figure is a schematic one, and the arrangement in FIG. 1 is actually correct. Two beams L emitted from the two-beam frequency shifter 12 and incident on the pupil relay system 17A
B ₁ and LB ₂ become parallel light beams inside the pupil relay system 17A,
Fire after crossing once. At this time, the objective lens 21a
The principal rays LA ₁ , LA ₂ of the beams LB ₁ , LB ₂ emitted to the optical axis AXa, etc., are parallel to each other, and are formed as spots in the pupil plane to be focused light fluxes.

By the way, the polarization directions of the beams LB ₁ and LB ₂ are bifocal optical systems.
When the light is separated into P-polarized light and S-polarized light by 21 and condensed at the focal points 26a and 27a, the light intensity (light amount) of the P-polarized light and the S-polarized light
Is adjusted to have a predetermined ratio. Normally, the light reaching the wafer 4 has more loss, so the amount of light to the wafer 4 is increased. For that purpose, the dual focus element 21b
Or a λ / 2 plate inserted between the laser light source 10 and the two-beam frequency shifter 12 and rotated by a predetermined angle around the optical axis AXa. Should be adopted. That is, it is possible to adjust the light amount ratio between the polarized light component reaching the reticle 1 and the polarized light component reaching the wafer 4 to an optimum value. By the way, if the parallel plane glass 50 is arranged so as to be orthogonal to the optical axis AXa exactly, the two beams LB ₁ and LB ₂ generate the pupil of the telecentric bifocal optical system 21 by the action of the pupil relay system 17A. The beam LB ₁ is incident on the surface, that is, two points symmetrical with respect to the optical axis AXa in the birefringent material 21b so as to sandwich the optical axis AXa.
At this point, the beam is separated into a P-polarized beam LB _1P and an S-polarized beam LB _1S by the polarization component, and the optical axis AX of the bifocal optical system 21
A parallel light beam inclined by an angle determined by the spot position in the pupil plane with respect to a reaches the reticle 1. Similarly, the beam LB ₂ is separated into a P-polarized beam LB _2P and an S-polarized beam LB _2S, and the beam LB ₂ is sandwiched between the optical axis AXa of the objective lens 21a.
_1P and LB _1S , respectively, reach the reticle 1 as parallel light beams having a target angle. As for the P-polarized light, the focal point 27a and the intersection plane (conjugate image plane) IP 'inside the pupil relay system 17A are conjugate, so that the P-polarized beams LB _1P and LB _{2P become} almost parallel light beams at the diffraction grating mark RM. And cross (image). Third
In the figure, the grid arrangement direction of the marks RM is the horizontal direction in the plane of the paper, and the inclination direction of each of the beams LB _1P and LB _2P from the optical axis AXa is also determined in the plane of the paper in FIG. The reticle 1 has a diffraction grating mark RM and a transparent window as shown in FIG.
P ₀ is formed, and the beams LB _1P and LB _2P irradiate the reticle 1 with a size that covers the mark RM and the window P ₀ . The mark RM shown in FIG. 4A is used for detecting the position in the x direction (grid arrangement direction), and the diffraction grating mark WM on the wafer 4 is also used as shown in FIG. 4B. It corresponds to. Mark WM is determined so as to align the position of the window P ₀ of the reticle 1 during the alignment (or exposure). Now, the substantially parallel S-polarized beams LB _1S and LB _2S emitted from the bifocal optical system 21 form an image (intersect) once at a focal point 26a in space, and then pass through the window P ₀ of the reticle 1 to form a projection lens. The light is once focused as a spot light by the third pupil EP, and is imaged so as to be incident on the diffraction grating or WM of the wafer 4 from two different directions. This is the focus for S-polarized light
Intersection I between 26a (wafer conjugate plane) and pupil relay system 17A
This is because P ′ is conjugate. Each of the substantially parallel S-polarized beams LB _1S and LB _2S emitted from the projection lens 3 is incident symmetrically inclined with respect to the grating arrangement direction of the diffraction grating mark WM. Even if the angle θ formed by the principal rays of the S-polarized beams LB _1S and LB _2S reaching the wafer 4 is large, it does not exceed the numerical aperture of the projection lens 3 on the exit (wafer) side. Naomi relay system 17
Since the reticle 1 and the wafer 4 are respectively conjugated with respect to the intersection plane IP 'inside A, if the beams LB ₁ and LB ₂ passing through the intersection plane IP' are parallel light beams, the reticle and the wafer 4 Each of the beams LB _1S , LB _2S , LB _1P , and LB _2P becomes a parallel light beam.

As described above, the beam LB from two different directions on the mark RM
_{When 1P} and LB _2P are irradiated, interference fringes are generated on the mark RM, and the interference fringes move in the grid arrangement direction of the mark RM in accordance with the difference frequency between the _two beams LB ₁ and LB ₂ ( Flowing)
Will be. A diffracted light 104 is generated from the mark RM along the optical axis AXa of the bifocal optical system 21, but due to the movement of the interference fringe, the diffracted light 104 has a beat wavefront that periodically repeats a change in brightness. Therefore, the signal DSr from the photoelectric detector 25 is a sine-wave AC signal corresponding to the cycle of the change in brightness.

The above is exactly the same for the relationship between the diffraction grating mark WM on the wafer 4 and the S-polarized beams LB _1S and LB _{2S. A} diffracted light 105 having a beat wavefront is generated vertically from the mark WM, This travels along the principal ray of the projection lens 3,
It reaches the photoelectric detector 25 via a window portion P ₀ of the reticle 1.

Now, the photoelectric detector 25 is marked through the bifocal optical system 21.
Although it is assumed that the RM and the mark WM are arranged conjugate with each other, actually, as shown in FIG. 3, an aperture as shown in FIG. 4C is provided at a position conjugate with each of the marks RM and WM. 'provided, the aperture plate 25' plate 25 aperture a _P of, configured to detect separately the photoelectric diffracted light 104, 105 transmitted through the a _S. Here the aperture A _P is, for example, those taking out the diffraction image by the diffracted light 104 from the mark RM on the reticle 1, the aperture A _S diffracted light 105 from the mark WM on the wafer 1
To extract the diffraction image by So the photoelectric detector
25 each aperture A _P the receiving surface of, by providing separately after A _S, the position detecting wafer 4 by the position detection mark WM of the reticle 1 by mark RM is possible independently. Incidentally, the aperture A _P A P-polarized light beam LB _1P, but it is an image of the irradiated mark RM on the reticle 1 by LB _2P, simultaneously S-polarized light beam LB _1S, reflected diffracted light LB _2S as background noise Come in. Therefore the aperture A _P provided a polarizing plate through a P-polarized light, the aperture A _S
It is preferable to provide a polarizing plate for transmitting S-polarized light. In this case,
In each of the two photoelectric detectors 25, crosstalk in which light from the wafer and light from the reticle are mixed is sufficiently reduced.

Now, the diffracted lights 104 and 105 become periodic (sinusoidal) brightness information, and the obtained photoelectric signals DSr and DSw are sinusoidal AC signals even if the reticle 1 or the wafer 4 is stationary. Therefore, in this case, the phase difference φr between the photoelectric signal DSr of the diffracted light 104 from the mark RM and the photoelectric signal DSr is detected by the phase detection system 40 using the photoelectric signal DR from the photoelectric detector 19 shown in FIG. 1 as a reference signal. . Similarly, diffracted light 10 from mark WM
The phase difference φw between the photoelectric signal DSw of No. 5 and the reference signal DR is detected. Then, if the difference between the phase difference φr and θw is obtained, the amount of displacement between the reticle 1 and the wafer 4 in the x direction can be determined. This detection method is called a so-called optical heterodyne method. If the reticle 1 and the wafer 4 are within a position error range (pre-alignment accuracy) of 1/2 of the grating pitch P, the position is shifted with high resolution even in a stationary state. Can be detected. Therefore, it is convenient to apply a closed-loop position servo so that a minute positional deviation does not occur while exposing the pattern of the reticle 1 to the resist of the wafer 4. In this detection method, φr−
After completing the alignment by moving reticle 1 or wafer 4 so that φw becomes zero (or a predetermined value), servo lock is applied so that reticle 1 and wafer 4 do not move relative to each other at the alignment position. Can be.

In this embodiment, during the exposure of the step-and-repeat method, the movement of the wafer stage to each shot area on the wafer is performed based on the measurement value of the interferometer 45, and the irradiation area of the two beams LB _1S and LB _2S When the mark WM is positioned within ± 1/2 pitch accuracy, the reticle stage or the 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 the phase difference φ
An analog voltage corresponding to r-φw can be generated by a D / A converter or the like, and this analog voltage can be directly applied as a deviation voltage to a servo circuit of a DC motor. This servo is performed until the exposure of the shot area is completed.

With this configuration, since the servo is not performed in accordance with the measurement value of the interferometer, it is possible to reduce the minute fluctuation of the stage due to the fluctuation of the air density in the beam path of the interferometer. Therefore, when the phase difference information capable of 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 motor on the wafer stage is reduced. The voltage is set to zero, and the above-described analog voltage is applied to the servo system on the reticle stage side.

In this manner, during the exposure operation, minute fluctuations generated particularly on the wafer stage side are suppressed, and a gentle drift-like fine movement can be achieved. By moving the reticle stage at high speed, the reticle and the wafer can be moved. It is possible to keep the relative displacement substantially zero. Therefore, there is no increase in the line width of the exposed pattern and no reduction in resolution, and extremely faithful transfer is achieved.

Next, the configuration of the two-beam frequency shifter 12 will be described with reference to FIGS. 5, 6, and 7. FIG.

As shown in FIG. 5, the parallel beam LB (orthogonal linearly polarized light) from the laser light source 10 is reflected by the mirror 70, and the polarization beam splitter 71 converts the P-polarized component beam LB _P and the S-polarized component beam LB _S Divided into Beam LB _P is reflected by the mirror 102, is incident on the AOM (acousto-optic modulator) 73, the beam L
B _S is reflected by the polarization beam splitter 101 and enters the AOM 74. AOM73 is driven by the high-frequency signal SF ₁ of frequency f _1, and outputs a primary beam deflected by the diffraction angle determined by the frequency f ₁ as the beam LB _P. AOM74 has a frequency f ₂ (f ₂ = f ₁
−Δf) driven by the high-frequency signal SF ₂ and its frequency f
Only the diffraction angle determined by the ₂ outputs a 1-dimensional deflected as beam LB _S. 0 of incident beam (for each AOM)
Order light D ₀ is blocked by a slit or the like disposed in appropriate positions. The relationship between the drive frequencies f ₁ and f ₂ and the difference frequency Δf is desirably f ₁ ≫Δf and f ₂ ≫Δf, and the upper limit of Δf is determined by the responsiveness of the photoelectric detectors 19 and 25. Now,
Beam LB _S from AOM74 is incident on the polarization beam splitter 76 is reflected by the mirror 75, the beam LB _P from AOM73 is incident from a direction perpendicular thereto on the polarizing beam splitter 76.
Here, the polarizing beam splitter 76 has two beams LB _P and L
Instead of synthesizing B _S completely coaxially, they are synthesized parallel to each other at a certain distance. In the case of the present embodiment, this interval defines the intersection angle θ between the _two beams LB ₁ and LB ₂ that irradiate the reticle 1 and the wafer 4.

Two parallel beam LB _P, LB _S is then incident on the optical system shown in Figure 6. Beam LB _P (P polarization) and LB _S (S polarization)
Has a frequency difference of Δf, but in this embodiment, it is necessary to irradiate two P-polarized beams having a frequency on the reticle and irradiate two S-polarized beams having a frequency difference on the wafer. There is. That system of Figure 6 is one of S-polarized beam LB _S and one of P polarized light beam LB _P of the frequency f ₂ of the frequency f ₁
From these two beams, four beams LB _1P , LB _2P , LB _1S , and LB _2S are created.

Two beams LB _S, polarization direction as LB _P is shown in Figure 6
The P-polarized beam LB _1P of frequency f ₁ and the S-polarized beam LB _1S are coaxially combined by a 45 ° half-wave plate 117, polarizing beam splitters 118, 119, etc., to become a beam LB ₁ , and a frequency f ₂ (f ₁ −Δ
f) The P-polarized light beam LB _2P and the S-polarized light beam LB _2S are coaxially combined to form a beam LB ₂ and exit from the polarizing beam splitter 119.

As shown in FIG. 6, when the S-polarized beam LB _S (frequency f ₂₎ is incident on the half-wave plate 117, its polarization direction is rotated by 45 °.
It is divided into a polarized beam LB _2P and an S-polarized beam LB _2S . P-polarized beam LB _P passing through a half-wave plate 117 (frequency f
_{In the case of 1} ), the polarization direction is also rotated by 45 °, so that the beam splitter 118 outputs the P-polarized beams LB _1P and S
It is split into polarized beams LB _1S . Each beam is synthesized by the polarizing beam splitter 119 via right-angle prisms 120 and 121 having a metal reflecting surface, and again two parallel beams LB ₁ and LB
Inject as ₂ The two beams LB ₁ and LB ₂ are located symmetrically with respect to the optical axis AXa of the alignment optical system.

Next, the _two beams LB ₁ and LB ₂ are incident on an optical system as shown in FIG. 7, for example. This optical system has two beams LB ₁ and L
Including incident the B ₂ in the plane (the image plane conjugate) IPm the prism 122 to cross at a predetermined angle, and a lens 123 to match the surface IPm the front focal plane. The two parallel beams LB ₁ and LB ₂ that intersect at the plane IPm converge as spots SP ₁ and SP ₂ on the rear focal plane (pupil conjugate plane) of the lens 123, and then diverge, resulting in a pupil relay system.
It is incident on 17A. As shown in FIG. 3, a principal ray LA ₁ of the beam LB _1, when the principal ray of the beam LB ₂ and LA _2, the principal ray LA also between the lens 123 and the pupil relay system 17A in this embodiment ₁ and LA ₂ are parallel to the optical axis AXa.

Further, in order to the front focal plane of the pupil relay system 17A coincide with the spot (beam waist position) SP _1, SP ₂ surface,
The spots SP ₁ and SP ₂ shown in FIG. 7 are relayed by the pupil relay 17A to the pupil plane where the birefringent substance 21b is located as shown in FIG.

Next, the operation and operation of this embodiment will be described with reference to FIG. FIG. 8 shows a configuration in which a part of the alignment optical system is made movable to cope with a change in the mark position. The same members as those described in FIGS. 1 and 3 are denoted by the same reference numerals. Marks RM ₁ and RM ₂ on reticle 1
Are different in position depending on the reticle, and usually only one of them is formed in the same reticle.
The actual alignment optical system, an objective lens 21a, 2-focus optical system 21 by the birefringence material 21b is positioned horizontally, the mirror M ₁ that bends the optical axis AXa vertically is provided at the tip of the objective lens 21a. The mirror M ₁ , the objective lens 21a, and the birefringent substance 21b are fixed to the holding hardware 62 so as to move integrally in the direction of the horizontal optical axis AXa, and the hardware 62 is moved in the horizontal direction by the drive control system 60. The difference between FIGS. 1 and 3 is that the birefringent substance 21b and the parallel plane glass 50
In the provision of the mirror M ₂ and the relay system 17C. Relay system 17C is a pupil plane EP ₁ located birefringent material 21b, is intended to relay the plane EP ₂ in the vicinity of the parallel plane glass 50. Incidentally I
Pm ′ is an image conjugate plane, and the reticle 1, the wafer 4 and the reticle 2
It is conjugate with the first intersection plane IPm of the _two beams LB ₁ and LB ₂ respectively. The mirror M ₂ and the relay system 17C are fixed to the apparatus body, and the physical optical path length between the birefringent substance 21b and the relay system 17C changes according to the position of the mark RM, but the optical path between them is afocal. Therefore, the image conjugate plane IPm ′ formed inside the relay system 17C is always conjugate with the reticle 1 and the wafer 4.

Now, when the reticle is replaced and the mark changes from RM ₁ to the position of RM ₂ (optical axis AXa ′), the drive control system 60
Is moved horizontally to make the optical axis AXa coincide with AXa '.

In this case, the degree of coincidence does not need to be very strict, and it is sufficient that the two beams LB ₁ and LB ₂ cover the mark RM ₂ and the window P ₀ . Therefore, it is sufficient to control the feeding of the hardware 62 using an inexpensive position monitor system such as an encoder and a position sensor. In this way, when a new reticle is set and the observation position of the alignment optical system is set, a reference mark plate provided at substantially the same height as the wafer 4 on a part of the wafer stage 5 as shown in FIG. the FM, bring to the projection position of the mark RM ₂ of the reticle. The reference mark plate FM has a diffraction grating mark on the wafer 4
The same diffraction grating mark as WM is formed. Therefore, the mark is hereinafter referred to as a fiducial mark.

Diffraction grating element of the center of the reticle mark RM ₂ is 4 (a), the projection lens 3 as shown in (b)
Since it extends linearly in the y direction toward the optical axis AX of the
Just below the RM ₂ position to the (bottom of the window P _0).

In this state, the servo lock is applied by the interferometer 45, the motor 46, and the like so that at least the position of the wafer stage 5 in the x direction is fixed.

Next, a Z stage (not shown) included in the wafer stage 5 is moved up and down in the z (optical axis AX) direction within a predetermined range, and the signal DSw and the signal Dw detected by the phase detection system 40 at that time.
The change in the phase difference φw from R is sampled for each minute displacement of the Z stage in the z direction. This is performed by the main control system 41. At this time, although not shown in FIG. 9, the distance between the projection lens 3 and the reference mark plate FM in the z direction is monitored by a well-known oblique incident light type focus detection system. In FIG. 9, the two beams LB _1S and LB _2S are inclined on the reference mark plate FM with respect to the optical axis AXa of the alignment optical system in the plane of the paper, but are actually inclined in the plane perpendicular to the plane of the paper. doing. When the phase difference φw is monitored at each position of the Z stage in the z direction, the characteristic as shown in FIG. 10 is obtained.
VT ₁ is obtained. In FIG. 10, the horizontal axis represents the position of the reference mark plate FM in the z direction, and the vertical axis represents the phase difference φw. The position _{Z 1, Z 2, Z 3} , Z reference mark plate FM is the projection lens 3 in the order of ₄
Shall be approached. As described above with reference to FIG. 2, since the width Zl of the detection area in the z direction is 300 μm,
Spacing position Z ₁ and Z ₄ can be about 300 [mu] m. Further, in the case of the present embodiment, the phase difference φw can be continuously detected within the range of ± 180 °, so that the displacement in the x direction is correspondingly ± 1/1.
Position shift can be measured within the range of 2 pitches. The grid pitch of the fiducial mark is 8 μm, and the phase detection system 40
Assuming that the resolution is 0.2 °, the measurement resolution of the displacement is as high as 0.0044 μm. Actually, due to the influence of noise, the practical resolution is about 0.01 μm (0.4 ° in phase). Thus, at 300μm position Z ₁ and Z ₄ to temporarily, the difference Δφw the phase difference .phi.w ₁ and .phi.w ₄ position at that time is assumed (in lateral deviation of about 0.02 [mu] m) 0.8 ° was characteristic VT The inclination of ₁ , that is, 2 of two beams LB _1S and LB _2S
The inclination (the inclination of telecentricity in the plane including the optical axis AX of the projection lens 3) from the normal line of the principal ray of the two beams, which are equal lines, is
Assuming that the amount of lateral displacement corresponding to the difference Δφw is ΔX, ΔX / | Z ₁ −Z ₄ | = 0.02 / 300 ≒ 6.7 × 10 ⁻⁵ .

This figure shows two beams LB _1S and LB _2S for alignment.
(The parallelism of the optical axis AX and the optical axis AXa on the wafer side), and the smaller the smaller, the better the telecentricity. Also, the phase difference φ between the positions Z ₁ and Z ₄
Assuming that the w ₁ and .phi.w ₄ is changed just 180 ° (1/2 pitch), telecentricity, [Delta] X / | a ≒ 4/300 ≒ 1333 × 10 -5 | Z 1 -Z 4.

Once the telecentricity is determined as described above, the main control system
41 determines whether the telecentricity is within an allowable range.
If the telecentricity has deteriorated, the main control system 41 determines, based on the difference between the target telecentricity and the measured telecentricity,
The tilt amount of the parallel flat glass 50 is calculated, and a command for tilting the parallel flat glass 50 by the calculated amount is output to the drive control system 52.

The inclination of the parallel plane glass 50 is caused by the beams LB ₁ and LB _{2 formed on} each pupil plane in the alignment optical system and the projection lens 3.
While keeping the position of the spot light at a constant distance,
It is determined that the diffraction grating (or the interference fringes) is shifted by the same amount in the pitch direction.

With this adjustment, the incident angles θa and θa of the two beams LB _1S and LB _{2S to} the object on the wafer side of the projection lens 3 are obtained.
b matches very accurately, and the symmetry of the incident angle is maintained.

Next, a method for manually adjusting the inclination of the parallel plane glass 50 will be described with reference to FIG.

FIG. 11 shows a control rack 200 attached to the stepper shown in FIG. 1 and an operation console 210 including an input keyboard. The control rack 200 incorporates a CRT (cathode ray tube) 202 for displaying mark images observed by various alignment systems of the stepper. Therefore, the CRT display control system 204 is provided with a switching function, and a display signal (video signal or the like) 208 such as a mark image is normally applied to the CRT 202. When the inclination of the parallel flat glass 50 is adjusted, the signals DR and DSr are used. , And a signal DSw, and each waveform is
Switch to display. That is, switch the CRT202 to the oscilloscope function of two phenomena, the vertical axis is the signal, DR,
The voltages of DSr and DSw are displayed, and the horizontal axis indicates the real time. Then, for example, in a state where the reticle is aligned in advance by displaying the waveform WF ₂ waveform WF ₁ and signal DSw signal DSr the CRT 202.

Next, the joystick 212 of the console 210 is tilted back and forth, and the reference mark plate FM on the wafer stage 5 is moved up and down within a predetermined range (for example, ± 20 μm). When the joystick 212 is tilted back and forth, a command Saz for moving the Z stage up and down is sent from the console 210 to the rack 200,
The motor of the Z stage is controlled via the rack 200.

Then, during the vertical movement of the Z stage, the operator
It is checked whether or not the phase relationship between the _two waveforms WF ₁ and WF ₂ displayed in 02 is displaced in the time axis direction.

At this time, since the waveform WF ₁ does not displace in the time axis direction,
Waveform WF ₂ is displaced to the left and right by the vertical movement of the Z stage. Therefore, the operator adjusts the inclination of the parallel flat glass 50 by tilting the joystick 212 left and right. When the joystick 212 is tilted left and right, a command Sb corresponding to the tilt direction and the tilt amount of the parallel flat glass 50 is sent from the console 210 to the rack 200, and a predetermined drive signal 206 is sent to the drive control system 52 via the rack 200. Sent. Thus, if the phase relationship between the signals DSr and DSw does not change significantly during, for example, the vertical movement of ± 20 μm of the Z stage, the vertical movement range of the Z stage is further expanded (for example, ± 100 μm), and the adjustment is similarly performed. Do. By the above operation, the parallel flat glass 50
Is adjusted to the optimal one.

Conventionally, similar adjustments have been made in an alignment system that projects a beam spot on a wafer surface, but this kind of adjustment is called optical adjustment, which is performed by a specialized maintenance worker over time. According to the present embodiment, since the operation can be replaced with a very simple operation such as checking the phase relationship (may be a Lissajous waveform) of the beat signal waveform displayed on the CRT 202, the adjustment can be easily performed even by a normal stepper operator.

In this embodiment, the signal DR and the signal DS are used in the same manner as in the previous embodiment.
The same effect can be obtained by checking the phase relationship with w.

Further, as a modified example of the present embodiment, the telecentricity using the alignment servo using the phase detection system 40 shown in FIG. 1 using the amount of vertical movement of the Z stage and the measurement value of the interferometer 45 is used. A method for adjusting the degree will be described. In the conventional method, the phase difference between the two beat signal waveforms is checked, so there is not much problem with the manual operation. However, at the time of the automatic check, the phase difference is ± 180 due to the movement of the Z stage.
You may jump out of the range of ゜. In this case, the telecentricity measurement data (FIG. 10) becomes discontinuous beyond the boundary of ± 180 °.
(N = 1, 2,...) To give an offset and make it continuous. However, when the alignment servo using the phase detection system 40 is performed on the reference mark plate FM, the measurement of the telecentricity is based on the amount of vertical movement of the Z stage (measured value of a micro sense, a potentiometer, etc.) and the interferometer 45
It is possible only with the two measured values.

 Hereinafter, the operation will be briefly described.

First, the Z stage is positioned at the top or bottom. At this time, the two beams LB _1S and LB _2S are moved to the reference mark plate FM.
So that the signal DSw can be normally obtained. The phase difference Δφt between the two signals is obtained by the phase detection system 40.
Is obtained by either Δφt = DR−DSw or Δφt = DSr−DSw, and the wafer stage 5 is driven by the alignment servo so that the phase difference Δφt becomes zero. Thereafter, the servo lock of the wafer stage 5 continues to operate so that the phase difference Δφt becomes zero.

Next, the Z stage is moved to a predetermined detection range (for example, 200 μm).
Just move in one direction. Then, for example, the measurement value of the interferometer 45 is sampled at each fixed Z position (every 5 μm). This operation may be performed repeatedly by moving the Z stage upward and downward several times within a predetermined detection range.

The sampling data obtained in this way has the same tendency as in FIG. 10, but the only difference from FIG. 10 is that the vertical axis is the measured value of the interferometer 45. Subsequent adjustment of the parallel flat glass 50 may be either automatic adjustment or manual drive adjustment.

According to this method, the amount by which the reference mark plate FM laterally shifts along the interference fringe wavefront distributed in the z direction is measured. However, there is an advantage that the Z stage can be moved up and down over a wide range.

As described above, in the embodiment using the stepper of FIG. 1, only one set of the alignment system is shown, but by providing two to three sets of similar alignment systems, the reticle 1
Alignment of the wafer 4 and the wafer 4 in the x, y and rotation directions and measurement of a magnification error are possible. In this embodiment, since the phase is detected based on the reference signal DR, for example, when the first layer is exposed (first printed) on the wafer 4,
The phase difference between the signal DSr and the signal DR obtained from the mark RM of the reticle 1 is always constant during the exposure operation.
The reticle stage can be servo-locked.

Next, a configuration of a stepper according to a second embodiment of the present invention will be described with reference to FIGS. 12 (A) and 12 (B).

12 (A) and (B) show the pupil relay system shown in FIG.
Only the part after 17B is shown, and the other configuration is the same as FIG. In the present embodiment, at the position of the spatial filter 23 provided after the pupil relay system 17B, the divided light receiving elements 300R, 300W
Is placed directly on the pupil plane of the system and diffracted light (beat signal) 10
4, 105 was received. As shown in FIG. 12 (A), after the relay system 17B, a polarizing beam splitter 310 is provided obliquely, and since the diffracted light 105 from the wafer 4 is an S-polarized component, it is reflected here to the light receiving element 300W. Reach. On the other hand, since the diffracted light 104 from the reticle 1 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, for example, FIG.
300a is formed, the upper and lower sandwiching it (or lateral) to the surface 300b for receiving the 0-order light L _0, 300c are formed. These 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 the pupil plane of the system, that is, the pupil EP of the projection lens 3.

Therefore, the photoelectric signal of the light receiving surface 300a of the light receiving element 300R is a signal.
DSr, and the photoelectric signal of the light receiving surface 300a of the light receiving element 300W becomes a signal DSw, which is used for phase detection just like the previous embodiment.

As is clear from the optical path shown in FIG. 3, the diffracted lights 104 and 105 (and the zero-order light L ₀ ) all converge on a beautiful spot (beam waist) on the pupil plane of the system. 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, the numerical aperture (N.
A.) is considerably smaller (for example, 0.01 or less), so that the light receiving surface 300a is made to have an area slightly larger than the exact spot size of the diffracted lights 104 and 105 so that the light receiving element
It is not necessary to make the position setting accuracy in the optical axis direction of 300R, 300W so strict, and the device manufacturing becomes simple. This means that it is possible to cope with a shift in pupil conjugate due to a change in the position of the objective lens 21a during alignment. The parallel plane glass 50 has two beams LB.
₁ , since the light transmission path of the LB ₂ and the light reception path of the diffracted lights 104 and 105 coexist, the diffracted lights 104 and 105 and the light receiving surface are provided no matter how the parallel flat glass 50 is tilted. The relative positional relationship with 300a does not change.

Above, the first present invention, in the second embodiment, although plane-parallel glass 50 for telecentricity adjustment slightly light source side than the pupil EP ₁ of the objective lens 21a, the pupil EP ₁ and the objective lens 21a It is preferable that the pupil EP ₁ is arranged so as to coincide with the pupil EP ₁ or a plane conjugate with the pupil EP ₁ . For this reason,
As shown in FIGS. 1, 3, and 8, the birefringent substance (2
When the focusing element 21b is formed in a parallel plane shape as a whole, the element 21b may be tilted and used for adjusting the telecentricity.

The position of the parallel plane glass 50 is different between FIG. 1 and FIG. 3. However, when photoelectric detection is performed in the pupil space as shown in FIG. 12, the transmitted light beams LB ₁ and LB ₂ are diffracted. It is necessary to provide them in the common optical path of the light 104 and 105, and they may be provided in either case when photoelectrically detecting in the image space as shown in FIG.

Next, a third embodiment of the present invention will be described with reference to FIG. 13 together with FIG. 5 and FIG.

Although the parallel plane glass 50 is used for the adjustment of the telecentricity in each of the embodiments described above, the present embodiment enables the telecentricity to be adjusted with higher accuracy without performing mechanical driving. However, the adjustment range of the telecentricity is limited to a smaller range than when the parallel flat glass is used. Therefore, the telecentric adjustment according to the present embodiment is effective when used for fine adjustment after coarse adjustment with parallel plane glass.

Now, as shown in FIG. 5, the two beams LB _P and LB _S
Are the first-order diffracted lights emitted from the AOMs 73 and 74, respectively.
Two beams LB _P, LB _S is set so as to be positioned symmetrically across the normal optical axis AXa. The frequencies of the drive signals SF ₁ and SF ₂ of the two AOMs 73 and 74 are, for example, 80.0
It is set to 25 MHz and 80.000 MHz.

Since the frequency difference is as low as 25 KHz, the two AOM73, 74
Are the same. So two
Change the direction of the diffraction angles of AOM73 and 74 to the optical axis AXa, beams LB _P and LB _S
In the same direction within the plane including the three
When F _1, I'll frequencies SF ₂ is increased or decreased by the same value, the beam LB after being synthesized by the beam splitter 106 _P, while spacing of LB _S was approximately constant, a point symmetrical relationship with respect to the optical axis AXa Can be changed.

In order to effectively operate the present embodiment, convex lenses are provided between the AOM 73 and the beam splitter 106 in FIG. 5 and between the AOM 74 and the beam splitter 106, respectively. What is necessary is just to replace 122 with a convex lens. These convex lenses are all connected by a focal length relationship. Then, the diffraction points of AOM73 and AOM73 and the image conjugate plane IPm in FIG.
It is desirable to further conjugate. FIG. 13 (A)
This is a schematic representation of the above configuration, showing the optical path from the beam splitter 106 in FIG. 5 to the image conjugate plane IPm in FIG. 7, and new convex lenses 400a, 400b, and 402 are provided. Other configurations (particularly, FIG. 6) may be the same. Also Pe
Is a pupil conjugate plane, and the two beams LB _P and LB _S are convex lenses 40
The light is focused on the spot there by the action of 0a and 400b.
The beams LB ₁ and LB ₂ emitted from the convex lens 402 similarly become parallel light beams and intersect at the surface IPm.

FIG. 13 (B) is a block diagram schematically showing a configuration of a circuit for generating drive signals SF ₁ and SF ₂ . The drive signals SF ₁ and SF ₂ are generated by frequency synthesizers 410 and 412 using a crystal oscillator Cx. The output frequency is set by counters 414 and 416 for setting the frequency division ratio and the like, and the set values CN ₁ and CN ₂ are given from the main control system 41 (FIG. 1) and the like.
The set values CN ₁ and CN ₂ are determined so that the frequency difference between the signals SF ₁ and SF ₂ is 25 KHz.

Now, increase or decrease the set values CN ₁ and CN ₂ by the same amount, and
_1, varying about 10% the frequency of the SF ₂ with respect to the original value, the two beams LB _P, LB _S varies about 10% relative to the original diffraction angle when emitting the AOM73,74. Since the diffraction points of the AOMs 73 and 74 are almost image conjugate, the inclinations (incident angles) of the _two beams LB ₁ and LB ₂ reaching the reticle 1 or the wafer 4 change in the same direction by the same amount. That is, when the incident angle of the beam LB ₁ is increased by 10% from the original value,
Angle of incidence of the beam LB ₂ is reduced by 10% than the original value. As a result, only the telecentricity can be adjusted without changing the intersection angle of the _two beams LB ₁ and LB ₂ .

In this embodiment, since there is no mechanical movable part, the time required for adjustment is extremely short. Therefore, as shown in FIG. 8, when the reticle 1 is replaced and the position of the mark RM is changed, the time required for adjusting the telecentricity after the hardware 62 is set is reduced, and the operating rate of the apparatus can be increased.

As described above, in the present embodiment, the two AOMs 73 and 74 are the first of the present invention.
It has a function corresponding to an optical member. Thus AOM73, 7
When adjusting the telecentricity using 4, two beams LB ₁ ,
Since the telecentric correction is performed only on the light transmission path side of the LB ₂ , the light receiving path side, that is, the diffracted lights 104 and 105 slightly changes in position in the pupil plane. However, if the adjustment by the AOMs 73 and 74 is used only in a very small range, it is sufficient to determine the size of the light receiving surface 300a in FIG. Of course, as shown in FIG. 3, in a system in which an aperture plate 25 'is provided on the image conjugate plane and the beat signal is received by the photoelectric detector 25, the beams LB ₁ and LB ₂
It is not affected at all even if it is changed by Telesen AOM73, 74.

[Other modifications]

(1) The two beams LB ₁ and LB ₂ irradiating the diffraction grating mark RM or WM may have different polarization directions. For example beam LB ₁ to P polarized light, when the beam LB ₂ to S-polarized light, on the grating mark but interference fringe does not occur, light-receiving elements 19 and 25,
If a polarizing plate (analyzer) birefringent element or the like is inserted before 300, a beat signal can be obtained in the same manner.

(2) The laser light source 10 may be a Zeeman laser in order to obtain a frequency difference between the beams LB ₁ and LB ₂ . In this case, AO
M73 and M74 can be omitted.

(3) Even if the homodyne scheme to zero frequency difference between the beams LB ₁ and LB _2, likewise can the crossing angle adjustment and telecentricity adjustment.

(4) When the AOM is used, it is possible to easily switch between the homodyne method and the heterodyne method only by changing the frequency of the drive signal. Therefore, the alignment sequence for scanning the wafer stage 5 using the homodyne method and the wafer during the exposure using the heterodyne method The processing can be selectively switched at high speed between the stage servo lock and the servo lock.

(5) TTR, the alignment system of the TTL system, the two wavelengths of the beam LB ₁ and LB ₂ by approximating the wavelength of the exposure light, may be omitted bifocal element 21a.

In that case, the image conjugate plane IP in the light transmission path of the beams LB ₁ and LB ₂
An illumination field stop is arranged at m, and the irradiation area of the resist layer on the wafer is limited to the mark WM.

(6) When the diffraction angle of the beam drifts due to the heat generated by the AOMs 73 and 74, the generated heat is monitored by a temperature sensor, and the frequency of the drive signals SF ₁ and SF ₂ is automatically adjusted as the temperature rises to thereby adjust the beam LB ₁ , The symmetry of the incident angle of the LB ₂ may be kept constant.

〔The invention's effect〕

As described above, according to the present invention, in an alignment optical system or the like that irradiates two beams so as to intersect on a diffraction grating,
Since the first optical member for shifting the positions of the two beams on the pupil plane of the objective optical system (projection optical system) by the same amount in the grid array direction is provided, the telecentricity of the two beams can be easily adjusted.

Therefore, there is an effect that the accuracy of position detection using the diffraction grating and the interference fringes can be stably obtained while maintaining the advantage of the original high resolution.

Further, even if the interference fringes and the diffraction grating are relatively displaced relatively in the optical axis direction, an extremely high telecentricity is maintained, so that a special effect that no hindrance to position detection is obtained.

For this reason, the focusing (AF) operation at the time of alignment, which is performed by a conventional stepper or the like, can be omitted, and there is also an advantage that the throughput is improved in the sequence using the EGA method in combination. In a proximity exposure apparatus (for example, an X-ray stepper using SOR, etc.) in which the gap setting accuracy and the interval setting accuracy between the light source and the mask are relatively low, the mask and the wafer are almost mechanically set and immediately. Operations such as alignment and exposure can be performed, and similarly, improvement in sleep can be expected.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration of a stepper according to a first embodiment of the present invention, FIG. 2 is a principle diagram of the present invention for explaining the state of interference fringes and a diffraction grating, and FIG. 3 is shown in FIG. FIG. 4 is a diagram illustrating the main part of the optical system in further detail. FIG. 4 is a plan view showing the relationship between the diffraction grating mark and the aperture plate in FIG.
FIGS. 6, 6 and 7 show the detailed structure of the frequency shifter in FIG. 1, FIG. 8 shows the practical structure of the alignment system in FIG. 1, and FIG. FIG. 10 is a diagram illustrating a method of measuring telecentricity on the image plane side of a lens, FIG. 10 is a graph showing measured values of telecentricity, and FIG. 11 is a block diagram illustrating an example of a control system of a stepper suitable for telecentricity adjustment ,
12 (A) and 12 (B) are diagrams showing the structure of a light receiving system according to a second embodiment of the present invention, and FIG. 13 (A) is a diagram showing an optical system according to a third embodiment of the present invention. FIG. 13 (B) is a block diagram showing an example of a frequency shifter drive circuit used with the optical system of FIG. 13 (A). [Description of Signs of Main Parts] 1 reticle, 3 projection lens, 4 wafer 5 wafer stage, 10 laser light source, 12 frequency shifter, 17A, 17B, 17C relay system 18 Reference grid plate 19 Photodetector (light receiving element) 21a Objective lens 21b Birefringent substance 23 Spatial filter 25 Light receiving element 40 Phase detection system , 41… Main control system, 50… Parallel plane glass, WP, RM, WM… Diffraction grating mark, 104, 105… Interference beat light (diffraction light) from marks RM, WM, 73, 74… Acoustic-optic modulator (AOM), 200 Control rack, 202 CRT display, 210 Console, 300, 300R, 300W Photodetector, LB Laser beam, LB _P P-polarized beam LB _S …… S polarized light beam, LB ₁ …… Frequency f ₁ beam, LB ₂ …… Frequency f ₂ beam, EP, EP ₁ …… pupil plane, SF ₁ , SF ₂ …… AOM drive signal, DR …… Reference signal, DSw, DSr …… Measurement signal, LA ₁ , LA ₂ …… Main ray, AXa …… Alignment system optical axis, IFw …… Interference fringe, IP ', IPm… ... image conjugate plane, FM ... fiducial mark plate

──────────────────────────────────────────────────続 き Continuation of the front page (72) Inventor Koichiro Komatsu 1-6-3 Nishioi, Shinagawa-ku, Tokyo Nikon Oi Works Co., Ltd. (56) References JP-A-63-56917 (JP, A) JP-A-63-282603 (JP, A) JP-A-63-283129 (JP, A) JP-A-64-64319 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) G01B 11 / 00-11/30 H01L 21/30

Claims (16)

(57) [Claims] 1. A beam generating means for generating two beams for irradiating a diffraction grating formed on an object from two directions at a predetermined intersection angle, and disposed between the beam generating means and the object. An objective optical system that irradiates the object at a predetermined intersection angle by passing the two beams at a predetermined interval in a pupil plane, and receives a diffracted light from the diffraction grating through the objective optical system. A photoelectric detector that detects the position of the object based on a detection signal from the photoelectric detector, comprising: an optical unit that changes an incident angle of the two beams to the object. Position detecting device. 2. The optical unit is disposed in an optical path from a light source included in the beam generating unit to the object, and changes an incident angle of the two beams on the object; and the objective optics. Interval adjusting means for relatively changing the distance between the system and the object in the optical axis direction; and control means for adjusting the optical member based on a detection signal from the photoelectric detector. The device according to claim 1. 3. The control means adjusts the optical member such that a detection signal from the photoelectric detector maintains a predetermined state when the distance is changed within a predetermined range by the distance adjusting means. The device according to claim 2, characterized in that: 4. The beam generating means has a frequency modulator for giving a predetermined frequency difference between the two beams, and a reference signal generating unit for generating a reference signal according to the frequency difference; The apparatus according to claim 2, wherein the optical member is adjusted based on a phase relationship between a detection signal from the photoelectric detector and the reference signal. 5. The apparatus according to claim 4, wherein said control means adjusts said optical member such that a phase difference between a detection signal from said photoelectric detector and said reference signal is substantially constant. apparatus. 6. The distance adjusting means includes a coordinate detecting system for detecting a relative position change amount between the objective optical system and the object in a direction substantially orthogonal to an optical axis of the objective optical system; A correction system for inputting a displacement amount detected by the coordinate detection system when the interval is changed by the interval adjusting unit as a correction amount at the time of adjusting the optical member. The apparatus according to any one of claims 2 to 5, wherein 7. The apparatus according to claim 2, wherein the optical member changes the incident angle while keeping the intersection angle of the two beams substantially constant. 8. The optical device according to claim 2, wherein said optical means includes said optical member for changing a tilt of a principal ray of two beams which are bisectors of said two beams. The device according to item. 9. The apparatus according to claim 2, wherein said beam generating means has a frequency modulator for providing a predetermined frequency difference between said two beams, and uses said frequency generator as said optical member. An apparatus according to claim 1. 10. The apparatus according to claim 1, wherein said optical means includes said optical member for changing said angle of incidence while keeping the crossing angle of said two beams substantially constant. . 11. The optical means according to claim 2, wherein said two beams are two-dimensional.
The apparatus according to claim 1, further comprising the optical member that changes a tilt of a chief ray of the two beams that are bisectors. 12. A projection optical system for projecting a pattern formed on a mask onto a substrate, holding the substrate,
A substrate stage movable in an optical axis direction of the projection optical system and a direction orthogonal to the optical axis, and two beams for irradiating a diffraction grating provided on the substrate from two directions at a predetermined intersection angle. An object for irradiating the object at a predetermined intersection angle on the object by passing the two beams at a predetermined interval in a pupil plane, the object being disposed between the beam generating means and the object; An optical system, comprising a photoelectric detector that receives the diffracted light from the diffraction grating via the objective optical system, and a position detection device that detects the position of the object based on a detection signal from the photoelectric detector. Wherein the substrate stage has a reference plate on which a reference grating similar to a diffraction grating formed on the substrate is formed, and moves the reference plate in a direction parallel to the optical axis. So that the substrate Control means for controlling a stage; based on a detection signal by diffracted light from the reference grating,
A projection exposure apparatus comprising: an adjustment unit that adjusts the inclination of the two beams. 13. A projection optical system for projecting a pattern formed on a mask onto a substrate, holding the substrate,
A substrate stage movable in an optical axis direction of the projection optical system and a direction orthogonal to the optical axis, and two beams for irradiating a diffraction grating provided on the substrate from two directions at a predetermined intersection angle. An object for irradiating the object at a predetermined intersection angle on the object by passing the two beams at a predetermined interval in a pupil plane, the object being disposed between the beam generating means and the object; An optical system, comprising a photoelectric detector that receives the diffracted light from the diffraction grating via the objective optical system, and a position detection device that detects the position of the object based on a detection signal from the photoelectric detector. In the projection exposure apparatus, the substrate stage has a reference plate on which a reference grating similar to the diffraction grating formed on the substrate is formed, based on a detection signal based on diffracted light from the reference grating,
A first step of controlling a position of the substrate stage in a direction perpendicular to the optical axis and servo-controlling the substrate stage;
Controlling means for controlling the substrate stage to move the reference plate in a direction parallel to the optical axis under the servo control;
Control means; and adjusting means for adjusting the inclination of the two beams based on a detection signal based on diffracted light from a reference grating while the substrate stage is being controlled by the second control means. Projection exposure apparatus. 14. The apparatus according to claim 12, wherein said adjusting means adjusts an inclination of a principal ray of the two beams, which is a bisector of the two beams. 15. A diffraction grating formed on an object is irradiated with two beams from two directions through an objective optical system at predetermined crossing angles through an objective optical system, and receives a diffraction light from the diffraction grating from a photoelectric detector. A method of detecting a position of an object based on a detection signal, wherein the inclination of a principal ray of two beams, which is a bisector of the two beams, is changed. 16. The apparatus according to claim 1, wherein the inclination is adjusted based on a detection signal from the photoelectric detector when an interval between the objective optical system and the object in the optical axis direction is relatively changed. Item 16. The method according to Item 15.