JPH06232026A - Aligner - Google Patents

Abstract

PURPOSE:To obtain an aligner mounted with a highly accurate and highly stable positioning device by substantially passing two coherent light rays having different vibration frequencies through the same optical path in a precision alignment system which has relatively high detection accuracy for measuring the absolute position of a first or second object. CONSTITUTION:Two beat signals are respectively obtained from two precision alignment systems 3 and 5. A phase comparator 6 measures the phase difference between the two beat signals and calculates the relative positional deviation between a wafer 9 and reticle 1. A control system 7 positions the wafer 9 by moving a stage 8 so that the positional deviation can become zero or a prescribed value. Thus two kinds of light rays having frequencies f+F1 and f+F2 are made incident to a diffraction grating 10 at different angles theta1 and theta2 against the wafer 9. In addition, the light rays are also made incident to another diffraction grating 2 at different angles theta1' and theta2' against the reticle 1. Therefore, the reflected and diffracted light rays of the light rays can be made to pass through the same optical path.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an exposure apparatus for transferring a pattern of a first object onto a second object, and more particularly to an alignment apparatus for measuring the relative positions of the first object and the second object and performing alignment. The present invention relates to an exposure apparatus having

[0002]

2. Description of the Related Art In recent years, semiconductor devices, especially integrated circuits, which have been the driving force of technological innovation, have been increasingly densified, and the size of fine patterns forming elements is about 0.5 μm or less.

Until the wafer becomes an integrated circuit, a large number of overlay exposures are performed. In order for it to operate properly as a semiconductor device, the alignment of each exposure is extremely important. The overlay accuracy is 0.
1 μm or less is required.

At present, many kinds of optical system configuration examples used in the alignment apparatus have been reported, but as a candidate for an optical system that achieves an overlay accuracy of 0.1 μm or less, a theoretically high resolution optical system is used. A configuration to which heterodyne interference is applied can be considered.

Although the positional deviation detection accuracy of the alignment apparatus using optical heterodyne interference is very high, the detection range is small when it is used as an alignment apparatus for an exposure apparatus due to its small detectable area. It is necessary to combine another wide position detecting device and a position adjusting device applying optical heterodyne interference as one position adjusting device.

A configuration example of a position detection device having a wide detection range mounted in a conventional exposure apparatus and a configuration example of a conventional alignment device using optical heterodyne interference will be described below.

For the sake of simplicity, a positioning device having a wide detection range will be abbreviated as a coarse alignment system, and a positioning device system using optical heterodyne interference will be abbreviated as a precision alignment system.

FIG. 25 shows an example of mounting a conventional alignment device on an exposure device. In FIG. 25, 200
Is an illumination optical system, 201 is a light source for projection, and the illumination light 20
Generate 8.

Reference numeral 206 is a reticle, 207 is a projection lens,
214 is a wafer, and the pattern on the reticle 206 is transferred onto the wafer 214 by the projection lens 207.

205 and 205 'are first alignment marks, 213 and 213' are second alignment marks,
215 and 215 ′ are X-recognition marks, Y-recognition marks,
Reference numerals 101 and 101 ′ are an X-coarse alignment system and a Y-coarse alignment system, which are used for reticle 206 and wafer
14 rough alignments are performed.

Reference numerals 204 and 204 'denote the first diffraction grating and 21.
1, 211 'is the second diffraction grating, and 102, 102' is X
-A precision alignment system and a Y-precision alignment system for more precise alignment.

Reference numerals 212 and 212 'denote the second diffraction grating 21.
Positional deviation information is carried by the interference light of the reflected and diffracted light from 1, 211 '.

Reference numeral 202 denotes a light source for rough alignment, which emits wide band light in a wavelength band where the resist has low sensitivity, or white light when resist exposure is not a problem.

A light source 203 for precision alignment generates two coherent light beams having slightly different frequencies.

The two coherent lights are linearly polarized lights whose polarization planes are orthogonal to each other and are emitted as one light beam. As such a light source, a Zeeman laser is actually used.

Reference numeral 209 denotes a stage control system, 210 denotes a stage, which is an X-alignment system and a Y-alignment system 10.
Based on the positional deviation information from 1, 101 ′, 102, and 102 ′, the stage 210 is driven and alignment is performed.

The operation of the position adjusting device having the above structure will be described below.

First, the illumination light 208 generated from the projection light source 201 is guided to the reticle 206 after being variously corrected by the illumination optical system 200.

Illumination light 2 from the pattern of the reticle 206
The transmitted scattered light of 08 is projected onto the wafer 214 by the projection lens 207.
The pattern of the reticle 206 is transferred onto the wafer 214 by focusing on the surface.

At this time, in order to transfer the pattern of the reticle 206 to a predetermined position on the wafer 214, the reticle 2
The wafer 214 is aligned with the wafer 06.

If the relative rotational component of the wafer 214 and the reticle 206 is removed by some method before the wafer 214 is transported to a position right below the projection lens 207, the alignment of the wafer 214 with respect to the reticle 216 will be performed. It can be performed only by the parallel movement of the X axis and the Y axis.

Therefore, in this conventional example, the exposure apparatus is equipped with two independent aligning devices for aligning the X-axis and the Y-axis in two directions orthogonal to each other.

In FIG. 25, the X-coarse alignment system 1
01 and the precision alignment system 2 perform alignment in the X-axis direction, and the Y-coarse alignment system 101 ′ and the precision alignment system 2 ′ perform alignment in the Y-axis direction.

Since the precision alignment systems 2 and 2'are alignment devices that apply optical heterodyne interference, X, Y
Another coarse alignment device 101 on each of the axes,
As described above, 101 'is required.

The alignment method of the coarse alignment systems 101 and 101 'is based on image recognition.
The first alignment marks 205 and 205 ′ that are engraved at sufficiently upper positions, the second alignment marks 213 and 213 ′ that are pre-engraved at predetermined positions on the wafer 214, and the coarse alignment system 101. ,
Identification marks 215, 215 'built into 101'
The three types of mark images are optically combined into one composite image, and the combined image is processed by the image recognition method to perform rough alignment in the X and Y axis directions.

After that, two diffraction gratings 20 orthogonal to each other in the grating direction are formed on the reticle 205 at positions sufficiently separated from each other.
4, 204 ′ is illuminated vertically from above with the dual-frequency light from the precision alignment light source 203, and is passed through an appropriate spatial filter and polarizing plate to block the 0th-order and higher-order diffracted light and select only ± 1st-order diffracted light. In addition, the projection lens 207 is used in such a state that the diffracted light is only linearly polarized light of one frequency.
Incident on.

Specifically, for example, when the frequencies of the two-frequency light from the light source 203 are F1 and F2, respectively,
The diffracted light that has passed through the diffraction gratings 205 and 205 'has two-frequency light F1 and F2, respectively. However, by utilizing the characteristic that the two-frequency light has different polarization planes depending on the frequency,
If you extract only one linearly polarized light using a polarizing plate, +1
It is possible that only the light of the frequency F1 exists in the diffracted light of the second order, and only the light of the frequency F2 exists in the diffracted light of the −1st order.

In the above-described state, the ± first-order diffracted lights that have passed through the projection lens 207 are orthogonal to each other in the lattice direction, which is pre-marked at a predetermined position on the wafer 214.
One spot is formed on the two diffraction gratings 211 and 211 ', and the reflected diffraction light 21 of the diffraction gratings 211 and 211' is formed.
The reference numerals 2, 212 'extend vertically upward along the same optical path.

The reflected diffracted lights 212 and 212 'become interference lights in the precision alignment systems 2 and 2', respectively, and become beat lights that repeatedly blink at the difference frequency of the two-frequency light emitted from the light source for precision alignment.

By measuring the phase change of the beat light and using the results, more precise alignment is performed.

The control system 209 is based on the position measurement values from the coarse alignment systems 1 and 1'and the fine alignment systems 2 and 2 '.
It is the operation of the conventional alignment apparatus that drives the stage 210 through the wafer to align the wafer 214 with the reticle 206.

[0032]

However, the above-mentioned conventional structure has the following problems.

First, the two-frequency light from the precision alignment light source 203 is +1 at the first diffraction gratings 204 and 204 '.
The light is separated into two lights of the 1st and -1st orders, and the separated ± 1st order diffracted light is divided into the first diffraction gratings 204 and 20.
From 4'to the second diffraction grating 211, 211 ', different paths will be passed.

Therefore, when each path undergoes independent optical path length fluctuations due to changes in the surrounding environment, it is possible to distinguish between the phase change of the beat light due to the optical path length fluctuations and the phase change due to the positional deviation between the reticle and the wafer. Therefore, there is a big problem that the alignment accuracy is significantly lowered.

Further, since the reticle, the diffraction grating on the wafer, and the alignment mark are located separately from each other,
Since the positions for performing the rough alignment and the precise alignment are different from each other, it is likely to be affected by disturbance such as expansion and contraction of the wafer surface and temperature change due to the process, and there is also a problem that the alignment accuracy is lowered.

Further, in actual use, in order to avoid the deterioration of the diffraction grating due to many processes going on, affecting the alignment accuracy, a new diffraction grating is re-corrected based on the old diffraction grating. There are several opportunities to update the diffraction grating.

However, in this case, the new diffraction grating is updated with sufficient accuracy with respect to the old diffraction grating, and even if the new diffraction grating is aligned with sufficient accuracy, the residual error of the alignment will occur with each repeated update. There is a case where they are added, which causes a problem that the overall alignment accuracy deteriorates.

When the spot size of the coherent light from the precision alignment light source 203 which is incident on the second diffraction gratings 211 and 211 'is larger than the area of the diffraction grating, irregular reflection light from other than the diffraction grating, especially rough surface. Another problem is that the speckle light generated when the coherent light is reflected by the reflecting surface is mixed with the beat light and the S / N of the detection signal is lowered, so that the alignment accuracy is deteriorated.

As described above, it is necessary to remove the deviation of the rotation angle between the reticle and the wafer with sufficient accuracy by some means before the wafer is conveyed to the position just below the projection lens. .

Therefore, in consideration of the fact that the current projection lens tends to have a higher NA, the distance between the lower surface of the projection lens and the wafer surface cannot be sufficiently taken into consideration. After that, the stage is moved in parallel with high accuracy and the wafer is set at the exposure position.

However, it is inevitable that the rotational angle of the reticle and the wafer will be displaced from each other due to the movement error of the stage during the movement, resulting in deterioration of the alignment accuracy.

Further, the projection magnification of the reticle pattern of the projection lens on the wafer changes due to environmental changes around the projection lens such as atmospheric pressure, temperature, and humidity. However, in the configuration of the conventional example, this change in projection magnification actually occurs. There was also a problem that it was impossible to know until overlay exposure was performed.

The present invention solves the problems of the conventional example described above, and provides an exposure apparatus equipped with a highly accurate and highly stable alignment device that is not affected by the environment around the exposure device, is not easily affected by the process. The purpose is to provide.

[0044]

To achieve this object, an exposure apparatus of the present invention is an exposure apparatus for projecting and exposing a pattern of a first object onto a second object, the exposure apparatus comprising: An alignment device for aligning the first object and / or the second object is provided, and the alignment device uses the two first coherent light beams having different frequencies.
From the optical signal obtained by illuminating at least one first mark portion of the object and / or the second object,
A precision alignment system having relatively high detection accuracy for measuring the relative position of the first object and the second object, or the absolute position of the first object and / or the second object, and in the precision alignment system, 2 with different frequencies
An exposure apparatus in which two coherent light beams pass through substantially the same optical path.

An exposure apparatus for projecting and exposing the pattern of the first object onto the second object, said exposure apparatus comprising:
An alignment device for aligning the first object and / or the second object is provided, and the alignment device uses two coherent light beams having different frequencies for the first object and / or the second object. From the optical signal obtained by illuminating at least one mark part, the first
A precision alignment system having relatively high detection accuracy for measuring the relative position of the object and the second object, or the relative position and absolute position of the first object and / or the second object,
The alignment device may be configured such that the two coherent light beams having different frequencies illuminate a part of at least one of the first mark portions of the first object and / or the second object. It is an exposure apparatus having a control means for controlling coherent light.

In this case, the specific control means is
Limit the flux of two coherent light beams with different frequencies,
The opening means is movable so as to intersect the optical axis of the light flux.

Further, an exposure apparatus for projecting and exposing the pattern of the first object onto the second object, comprising an alignment apparatus for aligning the first object and / or the second object, The alignment device performs relative alignment of the first object and the second object or absolute alignment of the first object and / or the second object, and a relatively low detection accuracy of a coarse alignment system, Relative alignment of the first and second objects, or the first
An optical system having a precision alignment system with relatively high detection accuracy for performing absolute alignment of an object and / or a second object, wherein the coarse alignment system and the precision alignment system have a common optical axis. It is an exposure apparatus provided.

In this case, the coarse alignment system is
At least a part of the rough alignment mark portion and the fine alignment mark portion of the first and / or second object,
By aligning the first object and / or the second object by performing image processing on the same screen, preferably, the rough alignment mark portion and the fine alignment mark portion of the first and / or second object are aligned. And are integrally formed.

An exposure apparatus for projecting and exposing the pattern of the first object onto the second object, wherein the first object and the second object are relatively aligned using moire fringes, or the first object is used. An exposure apparatus having an alignment device for performing absolute alignment of an object and / or a second object.

Then, in this case, the exposure apparatus preferably detects the projection magnification of the projection lens by detecting the shape of the moire fringes, and performs projection so that the projection magnification of the projection lens becomes substantially constant. It has means for controlling the projection magnification of the lens.

Further, the alignment device has a coarse alignment system having relatively low detection accuracy, and the moire fringes are caused by the first mark portion provided on the first object and / or the second object, and the rough object. It is formed by the second mark portion provided in the alignment system.

Alternatively, the rough alignment device has an image pickup means for picking up the light emitted from the first mark portion provided on the first object and / or the second object, and the second mark portion has the image pickup means. Is formed.

[0053]

According to the present invention, the two-frequency light can be passed through the same optical path by the above-mentioned structure, and the optical path length due to environmental changes around the device such as temperature, humidity, and air flow, which have been problems in the above-mentioned conventional example, can be achieved. It is possible to eliminate the deterioration of the alignment accuracy due to the change.

Further, since the diffraction grating and the alignment mark are adjacent to each other, coarse and precise alignment can be performed at the same place on the object to be measured, and by equipping the aperture, the scattered light is mixed into the signal. As a result, it is possible to provide an exposure apparatus having a highly accurate alignment device.

Further, by observing the moire pattern generated by the diffraction grating and the grating added to the recognition mark, the rotation angle and the parallel movement amount in one direction can be measured by one alignment device. Therefore, it is possible to measure the rotation angle during the exposure.

Further, since the projection magnification of the projection lens can be constantly monitored by observing the moire pattern, the exposure apparatus is provided with a means for keeping the magnification constant so that the exposure is highly stable and highly accurate and is not easily affected by the fluctuation of the surrounding environment. A device can be provided.

[0057]

【Example】

(Embodiment 1) Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a schematic diagram of the device configuration of an X-axis coarse and precise alignment system in the first embodiment of the present invention.

In FIG. 1, 1 is a reticle, 2 is a first diffraction grating, 3 is a reticle precision alignment system, 4 is a light source which generates two coherent light beams of frequencies f + F1 and f + F2, and 5 is a wafer precision alignment. The system, 6 is a phase comparator for detecting the phase difference of the beat signals from the reticle precision alignment system 3 and the wafer precision alignment system 5, 7 is a control system, and 8 is the phase difference detected by the phase comparator 6. A stage which is moved by a control system 7 to perform precise alignment between the reticle 1 and the wafer 9, 9 is a wafer, 10 is a second diffraction grating, 11 is a projection lens which transfers the pattern of the reticle 1 onto the wafer 9, Is a rough alignment system for performing rough alignment by an image recognition method, 13 is a first alignment mark, and 14 is a second alignment mark. That.

Although not shown in FIG. 1, a Y-axis precision alignment system having the same optical system configuration as that of the X-axis precision alignment system is arranged in the same manner as in the conventional example, and the wafer 9 from which miscalculation of rotation has been removed in advance. The parallel movement amount between the reticle 1 and the reticle 1 is measured by the two precision alignment systems.

With respect to the alignment apparatus configured as described above, the configuration and operation of the entire apparatus will be described in more detail with reference to FIG.

First, the rough alignment system 12 will be described. First, the alignment mark 1 on the reticle 1
3 and the alignment mark 14 imprinted at a predetermined position on the wafer 9 and the reference recognition mark inside the coarse alignment system 12 are optically combined.

Using the obtained composite image, the relative positional relationship between the reticle 1 and the wafer 9 is measured using an image recognition method, and rough alignment is performed. It should be noted that the reticle 1 and the wafer 9 can also be absolutely aligned with respect to a predetermined reference position by a similar image recognition method.

As described above, this coarse alignment system 12
The basic configuration of the optical system is the same as that of the conventional example.

Since the coarse alignment system 12 has to capture the alignment marks 13 and 14 within the same field of view, the positions of the alignment marks 13 and 14 on the reticle and the wafer are optical with respect to the projection lens 11. Must be in a conjugate relationship.

That is, the alignment mark 14 must be engraved at the position of the image of the alignment mark 13 by the projection lens 11.

Next, the wafer precision alignment system 5 will be described. Dual frequency light f + F emitted from the light source 4
1 and f + F2 are halved on the way into two-frequency light for the reticle precision alignment system 3 and two-frequency light for the wafer precision alignment system 5.

One of the separated two-frequency light is incident on the wafer 9 by the precision alignment system 5 for the wafer at angles θ1 and θ2 with respect to the normal line thereof.
It is incident on the projection lens 11.

The angles θ1 and θ2 are special angles, which will be described in detail below. Θ2 is the first-order reflection diffraction angle of the coherent light incident at θ1, and θ1 is the − of the coherent light incident at θ2. This is the primary reflection diffraction angle.

Thus, when the incident angle of the two-frequency light is selected, for example, if f + F1 is incident at the angle θ1, then
The first-order diffracted light of f + F1 is reflected at an angle θ2, and the −1st-order diffracted light of f + F2 incident at an angle θ2 is reflected at an angle θ1. Can be interfered with to generate beat light.

The reticle precision alignment system 3 also has the same optical system configuration as the wafer precision alignment system 5, and the other half of the two-frequency light from the light source 4 after two minutes is incident on the diffraction grating 2. The ± 1st-order reflected diffracted lights that are incident so as to form one spot at the angles θ1 ′ and θ2 ′ follow the same optical path as each other, but become interference light in the reticle precision alignment system 3. Generates beat light.

Here, the relationship between the incident angles θ1 ′ and θ2 ′ is the same as in the case of the wafer alignment system 5, where the first-order reflection diffraction angle of the ray incident on the diffraction grating 2 at the angle θ1 ′ is θ2 ′,
The first-order reflection diffraction angle of the light beam incident on the diffraction grating 2 at the angle θ2 ′ is θ1 ′.

As described above, the two beat signals from the two precision alignment systems 3 and 5 are obtained. The phase difference between the two beat signals is measured by the phase comparator 6, and the wafer 9 The first embodiment of the present invention is to calculate the relative positional deviation amount between the reticle 1 and the reticle 1 and move the stage 8 by the control system 7 so that the positional deviation amount becomes zero or a predetermined value. The operation of the alignment device in FIG. In this case, the absolute positions of the wafer 9 and the reticle 1 can also be obtained by comparing the phase of each beat signal with the predetermined phase of the reference signal.

Thus, the two-frequency light f + F1, f + F2 is incident on the diffraction grating 10 at different angles θ1 and θ2 with respect to the wafer 9, and different angles θ1 ′ and θ are different from the reticle 1.
The reflected diffracted light can be made to pass through the same optical path by being incident on the diffraction grating 2 at 2 '.

Therefore, in the present invention, it is possible to prevent the deterioration of the position measurement accuracy due to the surrounding environment change, which has been a problem in the above-mentioned conventional example.

Next, two coherent light beams f + F
The actual incident angles of 1, f + F2 to the diffraction grating will be described in detail below with reference to FIG.

FIG. 2 shows the path of diffracted light seen from a cross section perpendicular to the diffraction grating on the wafer and parallel to the normal to the diffraction grating plane.

A and B are lattice points of the diffraction grating, and the distance between the two points is P. LA is a light ray incident on the point A and LB is a light ray incident on the point B, both of which have an incident angle of θin and a diffraction angle of θout.

That is, in FIG. 2, one coherent light beam is incident on the diffraction grating at the angle θin, and the diffracted light is reflected at the angle θ.
It represents the state reflected by out.

Here, the diffraction conditions are ray LA and ray LB.
The optical path length difference (x1-x2) is equal to an integral multiple of the wavelength of the coherent light.

Therefore, when the first-order diffracted light of the ray with the incident angle θin returns at the angle θout, both angles θin, θ
The relationship of out is as shown in (Equation 1) below.

[0082]

[Equation 1]

On the other hand, when the −1st-order diffracted light of the light beam having the incident angle θout returns at the angle θin, the relationship between the two angles is as shown in (Equation 2), which coincides with (Equation 1).

[0084]

[Equation 2]

Therefore, if two light rays are incident on the diffraction grating on the wafer at an angle that satisfies the relationship of (Equation 1), the ± first-order diffracted light travels through the same optical path as the two incident light rays, You can see that it returns to the precision alignment system.

As an example, when a helium neon laser having a wavelength of 633 nm is used and θin = 8 ° and P = 8 μm, the above optical system can be realized by setting (θout = 12.6 °) from (Equation 1). You can

Next, the optical system configuration of the light source 4 and the precision alignment system will be described in detail with reference to the drawings.

First, the structure of the optical system of the light source 4 in the embodiment of the present invention will be described in detail with reference to FIG.

In FIG. 3, reference numeral 15 is a laser for generating linearly polarized light of a single frequency, 16 is a beam splitter, and 1 is a beam splitter.
Reference numerals 7 and 17 'are acousto-optic modulators.

Since the position of alignment on the wafer is extremely close to the exposure position, the oscillation wavelength of the laser 15 of the light source 4 must be different from the exposure wavelength.

Regarding the light source 4 configured as described above,
The configuration and operation will be described with reference to FIG.

First, the laser 15 emits linearly polarized light having a frequency f and a plane of polarization perpendicular to the plane of FIG. 3 (S polarization).

The emitted light beam is divided into two by the beam splitter 16, and the frequencies of the internal oscillators are f ′ + F1, respectively.
The frequency is modulated by the acousto-optic modulators 17 and 17 ′ of f ′ + F2, and the frequencies (f + f ′) + F1 and (f +
f ′) + F2 and the frequency difference between them is (F1-F2) S
It becomes polarized light.

The two modulated light rays are corrected into parallel light rays by a mirror or the like, and travel to a precision alignment system.

Here, the reason for making parallel rays close to each other is as follows.
In the embodiment of the present invention, this region is a portion where two light rays pass through independent optical paths, and is the only portion which is easily affected by fluctuations of air.

Therefore, in order to reduce such an influence, it is possible to reduce the influence of the two rays by as much as possible.
The two rays are brought close together.

Now, the effect of using the acousto-optic modulator will be described below. 2 with slightly different frequencies
A dual-frequency Zeeman laser may be used as a light source for generating two lights, but it is not possible to completely separate the dual-frequency light from the Zeeman laser into each frequency. There is a phenomenon that other frequencies are mixed.

With such mixing, the reflected diffracted light from the wafer produces beat light, and when measuring the displacement of the diffraction grating from its phase, a measured value that is considerably different from the actual displacement is obtained. The problem of measuring occurs.

Further, a slight amount of harmonic components is mixed in the two-frequency light from the two-frequency Zeeman laser, which is also one of the causes of returning a measured value different from the actual value.

As will be described in detail below, the phase of the beat light and the amount of movement of the diffraction grating have a linear relationship only when the reflected and diffracted light of the pure two-frequency light produces the beat light. If there is a gap, this linear relationship will break.

Therefore, the actual amount of positional deviation differs from the measured value, and the positioning accuracy deteriorates.

Therefore, if single-frequency lights having slightly different frequencies are independently generated by using an acousto-optic modulator, this mixing is completely eliminated, and the alignment accuracy is guaranteed theoretically.

Next, the reticle precision alignment system 3,
The optical system configuration of the wafer precision alignment system 5 will be described in detail with reference to FIG.

However, since the optical system configurations of the wafer precision alignment system 3 and the reticle precision alignment system 5 are exactly the same, both will be abbreviated as precision alignment system and will not be distinguished below.

FIG. 4 is a block diagram of the optical system of the precision alignment system in this embodiment. In FIG. 4, 18 is 1 / 2λ
A plate and 19 are polarization beam splitters.

Here, the two-frequency light f emitted from the light source 4
The directions of the polarization planes of + F1 and f + F2 are aligned with the polarization axis direction of the polarization beam splitter 19 by the ½λ plate 18 having the function of a polarizing plate rotatable with respect to the optical axis.

Reference numeral 20 is a 1/4 λ plate, 21 is an epi-optical system, 2
2 is a 1 / 2λ plate, 23 is a polarization beam splitter, 24 is a polarizing plate, 25 is a total reflection mirror, 26 is a beat photodetector,
52 is a diffraction grating.

The operation of the precision alignment system constructed as described above will be described in detail with reference to FIG.

First, two single-frequency lights f + from the light source 4
The polarization planes of F1 and f + F2 are P
It is corrected to polarized light.

The two P-polarized light rays pass through the polarization beam splitter 19 and are converted into circularly polarized light by the 1/4 λ plate 20.

The two circularly polarized lights are reflected by the epi-optical system 21.
Angle θi with respect to reticle or wafer plane
It is incident at n and θout.

Here, the angles θin and θout are the special angles described with reference to FIG. 2. As long as this incident angle is satisfied, the ± first-order diffracted light from the diffraction grating 52 passes through the same path again and is reflected by the incident optical system. Return to 21.

The returned ± 1st-order diffracted light passes through the ¼λ plate 20 again, and the circularly polarized light becomes S-polarized light there.

The two diffracted lights that have been S-polarized are reflected by the polarization beam splitter 19 and become light rays 27 and 28.

The S-polarized light beam 27 has its polarization plane rotated 90 ° by the ½λ plate 22 to become P-polarized light, and passes through the polarization beam splitter 23.

The other diffracted light 28 is the total reflection mirror 2
5 is reflected in the direction of the polarizing beam splitter 23 at
Since 8 is S-polarized light, it is totally reflected by the reflection surface of the polarization beam splitter 23, and thereafter shares an optical path with the light ray 27.

The P-polarized light ray 27 and the S-polarized light ray 28 are
Since it does not interfere as it is, the polarization planes are aligned by passing through the polarizing plate 24 placed so that the polarization axes form a specific angle with respect to the polarization planes of the light rays 27 and 28.

The angle of the polarization axis of the polarizing plate 24 is defined by the light rays 27,
It is determined corresponding to the amplitude of 28 and is selected so that the ratio of the AC component of the beat light is high.

For example, when the light beams 27 and 28 have the same amplitude, they may be installed at an angle of 45 °.

Now, after passing through the polarizing plate 24, 2
The two light rays 27 and 28 interfere with each other to generate a beat light 29, which is converted into an electric signal by the detector 26,
The phase of the signal is detected.

Here, the beat light is proportional to the amount shown by the following (Equation 3).

[0122]

[Equation 3]

Here, A1 and A2 are rays 27 and
The amplitude of 28, t is time, and θ is a phase constant, which is the phase difference between two light beams.

Further, Φ is a phase constant that changes depending on the position change of the diffraction grating, and there is a linear relationship represented by the following (Equation 4) with the movement amount Δx of the diffraction grating in the direction perpendicular to the grating. .

[0125]

[Equation 4]

Here, P is the grating constant of the diffraction grating 22. If the phase Φ of the beat light is measured, the movement amount Δx of the diffraction grating can be measured according to (Equation 4).

Actually, a brief description was given with reference to FIG. 1, but the actual alignment is, as shown in (Equation 5), the beat signals obtained from the reticle and the precision alignment system for wafers 3 and 5. Positioning is performed based on the difference between the phase Φ [reticle] and Φ [wafer].

[0128]

[Equation 5]

Finally, the reason why the coarse alignment system is required in addition to the precise alignment system will be specifically described.

As can be seen from (Equation 3) and (Equation 4), since the phase of the beat light changes by 2π or more when the movement amount Δx becomes P / 2 or more, the precision alignment system diffracts over P / 2. The amount of movement of the grid cannot be measured.

Therefore, as described with reference to FIG. 1, a coarse alignment system capable of capturing position information exceeding P / 2 is required.

As described above, in this embodiment, the two single frequency lights f + F1 and f + F2 from the light source 4 are converted into the diffraction grating 52.
On the other hand, when the light rays are incident at an angle satisfying (Equation 1), the two light rays pass through substantially the same optical path until the beat light is generated. Therefore, the measurement accuracy due to the fluctuation of air in the optical path, etc. The deterioration of can be eliminated.

Especially in the precision alignment system 5 for wafers, due to the structure of the exposure apparatus, two rays from the precision alignment system hit the diffraction grating and go through a very long optical path until the anti-diffracted light returns to the alignment system again. Therefore, the effect of the present invention is very high.

Further, in the light source 4, the generation of the two-frequency light is performed independently for each frequency by using the acousto-optic device, and the optical arrangement of the precision alignment system is mixed to maintain a small number of pairs of light or more. Therefore, the mixing of the two-frequency light is eliminated, and the reliability of the position measurement is improved.

(Embodiment 2) In the first embodiment, the case where the diffraction grating size of the wafer and the reticle is smaller than the laser beam spot size is not taken into consideration.

Then, in the second embodiment of the present invention described below, the aperture is arranged immediately before the precise alignment system so that the spot size of the laser beam is always smaller than the diffraction grating size.

The optical system configuration of the second embodiment will be described in detail below with reference to FIGS. 5 to 8.

The conventional example and the first example are so-called On Axis, where the position of alignment, that is, the position of the diffraction grating on the wafer and the exposure position substantially coincide with each other.
Since it is an optical system for alignment, the size of the diffraction grating cannot be increased so much.

Therefore, if the light emitted from the laser tube of the light source 4 is used as it is, the spot size on the wafer becomes much larger than that of the diffraction grating, and the area other than the diffraction grating is illuminated.

On the other hand, the wafer surface itself becomes a rough reflecting surface as it goes through many processes.

As shown in FIG. 8, the coherent light 51
When a rough reflecting surface is illuminated as the illumination area 54, the reflected light 57 and the diffracted light 5 depend on the roughness of the surface.
In addition to 5, the scattered light component called speckle light 56 comes to be mixed.

As shown in FIG. 8, the speckle light 56 is scattered light in which light spots are randomly scattered, and its spatial intensity distribution is considerably large.

Therefore, if the speckle light is mixed in the beat light measured by the precision alignment system in the conventional example and the first embodiment, the beat light intensity fluctuates greatly as the diffraction grating moves. Will get worse.

Therefore, in the second embodiment of the present invention, as shown in FIG.
As shown in, the beam diameter is narrowed by passing the laser beam through the aperture immediately before entering the precision alignment system so that the spot size is always smaller than that of the diffraction grating.

By doing so, it is possible to effectively prevent the speckle light from mixing in the beat light, and improve the alignment accuracy.

The configuration of the second embodiment of the present invention will be described in detail below with reference to FIG.

The difference between the second embodiment and the first embodiment is that
Since the other device configurations are the same except that the beam diameter reduction optical system is inserted between the light source 4 and the precision alignment system, only the beam diameter reduction optical system will be described below.

In FIG. 5, reference numerals 30 and 30 'denote beam diameter reduction optical systems, and other than that, the device configuration is the same as that of the first embodiment of FIG.

Since the beam diameter reduction optical systems 30 and 30 'installed at the above positions have the same optical system configuration, the beam diameter reduction optical system 30 will be described below without distinguishing both. .

Details of the device configuration of the beam reduction optical system 30 will be described with reference to FIG. FIG. 6 is a schematic configuration diagram of the beam reduction optical system 30.

In FIG. 6, 31 is the first optical system, and 32 is
Is an aperture, and 33 is a second optical system.

The two light beams from the light source 4 form one spot at the aperture 32 by the first optical system 31.

The aperture 32 has an opening, and the beam diameter is narrowed by the passage of light rays therethrough.

The squeezed light rays are again corrected into parallel light rays by the second optical system 33, and head for the precision alignment system.

Here, the size of the aperture of the aperture 32 is selected so that the spot size of the two light beams on the diffraction grating is smaller than the diffraction grating size.

The aperture 32 prevents the laser light from illuminating the outside of the diffraction grating, so that the proportion of speckle light mixed in the reflected diffracted light is reduced, and the S / N of the beat light measured by the precision alignment system is increased. The alignment accuracy is improved.

As described above, according to the present embodiment, by providing the aperture 32 for making the spot size of the two light beams on the diffraction grating smaller than the diffraction grating size, the process wafer It is possible to provide an alignment device that is extremely resistant to surface roughness.

The mounting example of the alignment apparatus of the second embodiment described with reference to FIG. 5 in the exposure apparatus has the same apparatus configuration as that of the first embodiment, but as shown in FIG. Also, by inserting the aperture 32 in the device configuration of the conventional example, the same effect as that of the second embodiment can be obtained.

In FIG. 7, the precision alignment light source 2 is shown.
The aperture 32 is added after 03, but other than that, it is the same as the device configuration of the conventional example in FIG.

By narrowing the light beam emitted from the precision alignment light source 203 by this aperture 32, it is possible to make the spot size smaller than the diffraction grating size on the wafer and improve the alignment accuracy.

(Third Embodiment) Next, a third embodiment of the present invention will be described in detail with reference to FIGS. 9 to 11.

In this embodiment, except for the beam reduction optical system corresponding to the beam diameter reduction optical systems 30 and 30 'of the second embodiment, the other device construction is the same as that of the second embodiment.

Hereinafter, the beam diameter reducing optical system of this embodiment will be described in detail on behalf of the beam diameter reducing optical system 300.

FIG. 9 is a block diagram of the beam diameter reduction optical system in the third embodiment of the present invention.

In FIG. 9, reference numeral 34 denotes an optical path selection control system, which allows the aperture 32 to be two-dimensionally moved in a direction parallel to the aperture plane.

The operation of the beam diameter reduction optical system 300 having the above-described device configuration will be described below with reference to FIGS. 10 to 11.

By moving the aperture 32,
Although the aperture of the aperture 32 also moves, as long as the aperture is within the incident ray spot on the aperture plane, the light beam 60 after being focused moves in parallel as the aperture 32 moves.

The light beam after being focused on the aperture 32 is X
FIG. 10 schematically shows the state of parallel translation about the axis and the Y axis.

Now, the two parallel rays incident on the precision alignment system move in parallel due to the movement of the aperture without changing the distance between the two rays. Therefore, the two rays 60 are different from each other on the wafer and the reticle. To form one spot.

When a plurality of diffraction gratings are present in close proximity to each other, a spot can be formed on the desired diffraction grating by moving the aperture 32, so that the desired position can be obtained without moving the diffraction grating. It will be possible to align with the diffraction grating of.

Specifically, FIG. 11 schematically shows that only a desired diffraction grating is illuminated by the parallel movement of light rays.

The two light beams 60 from the precision alignment optical system, which are focused by the aperture 32 and aligned by the parallel movement, illuminate a desired diffraction grating more accurately than the spot 62 when the aperture is not focused. can do.

Generally, a wafer undergoes many processes until it is completed as a semiconductor device such as an IC.

Therefore, the diffraction grating engraved on the wafer deteriorates as the process progresses, and the grating pattern is disturbed or broken.

When the deterioration of the grating pattern becomes large, there arises a problem that the intensity of the diffracted light decreases and the S / N of the signal detected by the beat light detector decreases.

In order to avoid such a problem, in general, a new diffraction grating is rewound near the old diffraction grating before the deterioration of the grating pattern becomes remarkable. Then, the alignment is performed based on the updated diffraction grating.

Since the new diffraction grating is imprinted at a predetermined position after the alignment with the old diffraction grating, the relative positional relationship between the old and new diffraction gratings does not completely meet the design value, and the alignment error is included. It becomes a value.

Therefore, if the diffraction grating is repeatedly updated as described above, the total alignment accuracy may not be expected even if the latest diffraction grating is sufficiently aligned.

In order to suppress the influence of the addition of the alignment error in updating the diffraction grating, the relative positional relationship between the old and new diffraction gratings is measured in advance every time the diffraction grating is updated, and the result is measured by the new diffraction grating. It will be necessary to include it in the alignment process in.

That is, the conventional method of measuring the positional relationship between the old and new diffraction gratings, in which the alignment apparatus is not equipped with a mechanism for selecting the diffraction grating for the alignment, first performs the alignment with the old diffraction grating and then performs a sufficient measurement. The relative position of the old and new diffraction gratings is read by introducing the new diffraction grating into the measurement area of the alignment device on the stage equipped with translational accuracy and length measuring means and performing alignment, and reading the stage movement amount by the stage length measuring means. The amount of deviation from the design value of the relationship is obtained.

However, as long as such a method is used, the movement error of the stage is always included in the measured value,
After all, we cannot expect improvement in the overall positioning accuracy.

That is, the specific position error between the old and new diffraction gratings must not exceed the grating constant P of the diffraction gratings, considering that the new diffraction gratings are stamped based on the old diffraction gratings. It is comparable to the measurement accuracy of the alignment system.

Therefore, in the present embodiment, the alignment device is provided with a diffraction grating selecting means.

Therefore, sufficient alignment is performed with the old diffraction grating, the stage is not moved in that state, and the new diffraction grating is introduced into the measurement region only by moving the aperture 32, and the positional deviation amount is set by the precision alignment system. If you measure
The value becomes a marking position error of the new diffraction grating based on the old diffraction grating, and the above-mentioned problem that the movement accuracy of the stage affects the measurement value between the diffraction gratings is eliminated.

As described above, according to the third embodiment, since the optical path selection control system 34 is provided, the target diffraction grating can be introduced into the measurement region without moving the wafer and reticle. When the diffraction grating is updated, the position of the new diffraction grating with respect to the old diffraction grating can be measured with high accuracy.

Therefore, it is possible to provide a positioning apparatus in which the total positioning accuracy is less deteriorated even if the diffraction grating is frequently updated.

(Embodiment 4) A fourth embodiment of the present invention will be described in detail below with reference to the drawings.

In FIG. 12, 35 is a half mirror,
The other device configuration is similar to that of the first embodiment.

As shown in FIG. 12, the half mirror 35.
Is arranged in the optical path between the light source 4 and the reticle precision alignment system 3, the rough alignment system optical path and the reticle precision alignment system optical path can be made to coincide with each other, and therefore the alignment mark and the diffraction grating are brought close to each other. In addition to the above, the coarse alignment system 12
In, it becomes possible to capture the alignment mark and the diffraction grating image in the same visual field.

The coarse alignment system 12 of the alignment apparatus configured as described above will be described in more detail below with reference to FIG.

FIG. 13 is a schematic configuration diagram of an optical system of the coarse alignment system. In FIG. 13, 36 is an alignment mark, 37, 40, 41, 43 are lenses, 38 is a mirror, 42 is a recognition mark, and 44 is an image receiving portion.

The operation of the coarse alignment system configured as described above will be described below.

The illumination light passes through the lens 40 and is introduced to the coarse alignment optical axis by the half mirror 39.

The illumination light illuminates the alignment mark 36 by the mirror 38 and the lens 37, and the scattered light from the alignment mark 36 forms an image on the surface of the recognition mark 42 by the lens 41.

Therefore, the image of the alignment mark 36 and the image of the recognition mark 42 are optically combined, and the combined image is
The image signal is converted by the lens 43 and the image receiving unit 44.

Here, as the image receiving unit 44, for example, CC
It is possible to use D (solid-state image sensor).

By processing the image signal of the image receiving section 44, the relative displacement between the alignment mark 36 and the recognition mark 42 is extracted, and the alignment mark 36 is aligned.

Next, FIG. 14 shows an example of a pattern in which the alignment mark 36 and the diffraction grating 2 usable in this embodiment are close to each other.

In FIG. 14, the linear alignment mark 36 and the linear diffraction grating 2 are integrated as one alignment mark.

There is a triangular pattern at the tip of the alignment mark portion, but this is for automatically recognizing the alignment mark by image recognition. The triangular portion is discriminated by the pattern recognition and other confusing patterns are detected. The purpose is not to catch.

Further, FIG. 15 shows an example of a recognition mark pattern usable in this embodiment. This recognition pattern 36
Is a four rectangular figure consisting of two sides having a predetermined width,
They are arranged so that their apex angles face each other. As an example, a transparent optical substrate having a surface plated with chrome can be used.

Now, the combined image of the recognition mark 42 of FIG. 15, the alignment mark 36 of FIG. 14, and the diffraction grating 2 is shown in FIG.
It is shown in.

In FIG. 16, the alignment mark 36
Is slightly shifted to the right with respect to the recognition mark 42.

The composite image photographed by the image receiving unit 44 is subjected to predetermined image processing, and as shown in FIG. 16, the gap G1 between the recognition mark 42 and the alignment mark 36,
Detect X2 and perform alignment from it.

The image pickup area 58 photographed by the image receiving section 44 is
As shown in FIG. 17, the size is such that the alignment mark 36 and the diffraction grating 2 enter.

In this embodiment, since the rough alignment system and the fine alignment system share the same optical path, the spot image 59 of the laser light is also photographed at the same time as shown in FIG.

As a result, it is possible to determine whether the diffraction grating pattern is broken or the laser beam forms a spot at a predetermined position on the diffraction grating, so that the probability of a measurement error in the precision alignment system can be reduced.

As described above, according to this embodiment, the diffraction grating 2 and the alignment mark 36 can be adjacent to each other by making the optical paths of the rough alignment system 12 and the fine alignment system 3 coincide with each other.

Therefore, the rough alignment and the fine alignment can be performed at the same position, and the influence of expansion and contraction of the wafer surface is eliminated, so that the reliability of the alignment accuracy is improved.

Also, the diffraction grating 2 and the alignment mark 3
6 can be simultaneously imaged by the coarse alignment system, and the grating pattern shape of the diffraction grating, the spot position of the laser beam from the precision alignment system, and the overlapping degree of the spots can be observed even during the alignment. It is possible to effectively prevent a measurement error in the precision alignment system 3.

The example of mounting the alignment apparatus of the second embodiment described with reference to FIG. 12 on the exposure apparatus had the same apparatus configuration as that of the first embodiment, but as shown in FIG. ,
In addition to the conventional device configuration, half mirrors 45, 46, 47,
By inserting 48, the same effect as the fourth embodiment can be obtained.

In the present embodiment, the description has been limited to the reticle precision alignment system 3, but the half mirror 35
By similarly arranging on the optical axis of the wafer precision alignment system 5, it is possible to perform the rough alignment system and the precision alignment at the same position for the wafer precision alignment system 5 as well.

(Fifth Embodiment) The fifth embodiment of the present invention will be described in detail below with reference to the drawings.

This embodiment is similar to the fourth embodiment in that the half mirror 35 is provided, but the structure of the coarse alignment system 12 is partly different.

The coarse alignment system 12 of the alignment apparatus of this embodiment will be described in more detail below with reference to FIG.

FIG. 19 is a schematic block diagram of the optical system of the rough alignment system. In FIG. 19, 36 is an alignment mark, 37, 40, 41, 43 are lenses, 38 is a mirror, 42 is a recognition mark, 44 is an image receiving portion, and 49 is a newly added light-shielding mark. You can move back and forth to block it.

The operation of the coarse alignment system configured as described above will be described below.

Since the optical path of the coarse alignment system and the optical path of the fine alignment system are matched by the half mirror 35,
Some of the two laser beams for the precision alignment system form one spot on the recognition mark 42.

The light-shielding mark 49 is formed by providing a light-shielding portion on a part of a transparent glass substrate by chrome plating or the like, and this mark can shield only a spot image of laser light.

Since the light-shielding mark 49 is movable back and forth, the image receiving unit 44 can capture both the spot image of the laser beam of the precision alignment system and the alignment mark image, or can capture only the alignment mark image. It is possible.

The advantage of providing the light-shielding mark 42 for performing the above operation will be described.

Since the laser light from the alignment mark 36 passes through many optical elements while forming a spot on the recognition mark 42, some stray light components due to irregular reflection are mixed.

Since the laser light has a high coherence, even if the intensity of stray light is low, an interference phenomenon may form a disordered light spot with high intensity on the recognition mark 42.

To specifically describe, in FIG. 20, a combined image of the alignment mark 36 and the recognition mark 42 in the image pickup area 63 captured by the image pickup section 44 when the light shielding mark 49 is not provided, on the diffraction grating 2. It is shown that the spot image 59 of the formed laser beam and the stray light image 65 due to irregular reflection of the laser beam are randomly distributed in the imaging region 63.

As shown in FIG. 20, when the light spots of the stray light 65 overlap the alignment mark 36 image, the length measurement by the image processing may not work properly.

For example, when alignment is performed by detecting the contours of the alignment mark 36 and the recognition mark 42 and measuring their mutual distances, the density of stray light may be erroneously detected. However, there is a problem in that the reliability of is reduced.

A shield mark 49 is newly provided to solve this problem. When performing rough alignment by image recognition, stray light is removed by introducing a light-shielding mark 49 into the coarse alignment system optical path and blocking the laser light from the alignment mark 36.

When observing the disorder of the diffraction grating pattern and the state of the laser beam spot 64, the light shielding mark 49 is removed from the optical path.

By moving the light-shielding mark 49 back and forth in this manner, it is possible to remove stray light 65 generated by the laser beam of the precision alignment system during operation of the rough alignment system. Therefore, the effect of the fourth embodiment can be obtained. In addition,
Further, the reliability can be improved.

As described above, the coarse alignment system according to the present embodiment can provide a highly reliable coarse alignment system that is not affected by stray light by providing the retractable shield mark 49.

(Embodiment 6) A sixth embodiment of the present invention will be described below with reference to the drawings.

The mounting example of the alignment device in the sixth embodiment is the same as the device configuration of the third embodiment shown in FIG. 12, and the optical system configuration of the coarse alignment system 12 is also the same as that of FIG. The pattern shape of the recognition mark 66 having the changed configuration will be described in detail.

FIG. 21 shows the pattern shape of the recognition mark 66 in the sixth embodiment. As shown in FIG. 21, a grid-like pattern is newly added adjacent to the cross-shaped recognition mark of the third embodiment.

The positions of the grid pattern are arranged so that when the diffraction grating images of the wafer and reticle and the recognition mark 66 are combined, the two grating images are exactly overlapped.

Further, the interval of the lattice fringes is selected so as to be as equal as possible to the lattice interval of the projection image of the alignment mark on the recognition mark 66 of the diffraction grating, but it is not necessarily required to be completely the same.

The recognition marks 66 may have the shape shown in FIG. 21 for both the X- and Y-coarse alignment systems, or one of them may have the shape shown in FIG. 21 and the other may have the conventional shape shown in FIG. Good.

The operation of the grid pattern of the recognition marks 42 added as described above will be described below.

FIG. 22 shows a composite image of the alignment mark 36 and the recognition mark 66 photographed by the image receiving unit 44, and by the overlapping of two lattices, that is, the lattice and the lattice-like pattern of the recognition mark 66, the lattice pattern of the recognition mark It is shown that moire fringes 50 are formed with an angle ξ with respect to the direction and a fringe spacing of d.

Here, as shown in detail in FIG. 23, the grating gap of the recognition mark is d1, the grating gap of the diffraction grating is d2, and the gratings of the recognition mark are overlapped at an angle θ. Fringe spacing d forming an angle ξ with respect to
Assuming that the moire fringes 50 are formed, the respective quantities have the following relationships (Equation 6) and (Equation 7).

[0240]

[Equation 6]

[0241]

[Equation 7]

In general, in order to align the position of an object having a two-dimensional spread, it is necessary to measure three kinds of amounts of parallel movement in two directions and a rotation angle.

However, in all the embodiments described so far, only two alignment devices are mounted, so that the parallel movement amount and the rotation angle cannot be measured by one measurement.

Therefore, to explain the wafer, the positions of a large number of diffraction gratings whose positional relationship is accurately known in advance at different points on the wafer are measured while moving the stage, and statistical processing is performed to measure the wafer and the like. Measure the amount of translation and the angle of rotation.

After the wafer is sufficiently aligned in this way, the stage is accurately translated in parallel by a laser interference system or the like, and exposure is sequentially performed.

At this time, particularly with respect to the wafer, even if the laser beam is frequently irradiated and the diffraction grating is stamped with sufficient accuracy at the beginning, the diffraction may occur due to the change of the wafer itself due to the expansion and contraction of the wafer after each step. Since the positional relationship of the lattice changes, the alignment accuracy decreases.

Furthermore, in such an exposure method, the movement error of the stage is always affected in addition to the change of the wafer itself.

First, since the position of the pre-marked diffraction grating is not known more accurately than the movement accuracy of the stage, the movement error of the stage always reduces the alignment accuracy at the time of statistical processing. Become.

The positioning accuracy for each exposure after the positioning is once performed is that the positioning is not performed thereafter.
Since it is determined only by the parallel movement of the stage, the influence of the stage error is large here.

In order to effectively avoid this problem, it is only necessary to measure the parallel movement amount and the rotation angle using only the diffraction grating in the exposure area and perform the alignment for each exposure regardless of the stage movement. .

In order to realize this, the structure of this embodiment is one in which a unit for measuring the rotation angle is incorporated in one aligning device.

The rotation angle in this embodiment is measured by measuring the moire fringe angle. As can be seen from (Equation 6), there is a close relationship between the angle ξ of the moire fringes and the angle θ of the two gratings, and regardless of the grating spacing of the diffraction grating image,
It can be seen that when θ = 0, ξ = 0.

That is, if the angle ξ of the moire fringes is measured by image recognition and the stage is rotated so that ξ = 0, the directions of the two gratings match, and the rotation error between the wafer and the reticle of the recognition mark 66 is There will be no.

Similarly, by using Moire fringes,
A rotation error between the reticle and the recognition mark 66 can be eliminated.

Therefore, as a result, the wafer and the reticle can be brought into a state with no relative rotation error. That is, the angle of the moire fringes 50 generated by the two gratings is measured by the coarse alignment system 12 by image recognition, and is zeroed to align the wafer, the reticle, and both of them.

As described above, according to the present embodiment, by observing the moire fringes 50 due to the addition of the grid pattern to the recognition mark 66 by image recognition, one alignment device moves in parallel in one direction. Since it is possible to measure both the rotation angle and the rotation angle, it is possible to perform the alignment that is not affected by the movement error of the stage and the change of the wafer itself by performing the measurement once for each of the two alignment devices.

Further, since it is possible to observe moire fringes at all times, there is an advantage that a high alignment accuracy is guaranteed even during exposure.

(Embodiment 7) Hereinafter, a seventh embodiment of the present invention will be described in detail with reference to the drawings.

The configuration of the apparatus in this embodiment is shown in FIG.
This is similar to the fourth embodiment shown by 2 and 13.

The contents are the same as in the sixth embodiment.
It is related to alignment by observing Moire fringes.

That is, in this embodiment, the difference lies in the method of generating moire fringes. In the sixth embodiment, a grating pattern is newly provided on the recognition mark 66, and moire fringes are generated by superimposing the diffraction grating image and the grating pattern.

On the other hand, in the present embodiment, the image receiving section 44 generates moire fringes.

The moire fringe generation method of this embodiment will be described below in detail with reference to the drawings.

FIG. 24 is a schematic representation of the array of pixels 61 on the surface of a CCD (solid-state image sensor) 60.

Since the CCD 60 has a structure in which small rectangular pixels are two-dimensionally spread, it is a kind of 2
It can be considered that a three-dimensional lattice (a grid lattice) is formed on the image receiving surface.

Therefore, the diffraction grating image on the image receiving surface is extended with the CCD by adjusting the enlargement magnification of the rough alignment system so that the lattice spacing is approximately the same as the spacing between the pixels 61 of the CCD 60. By combining with the diffraction grating image, it is possible to generate the moire fringes similar to those in the sixth embodiment.

The method of performing the alignment through the moire fringes is exactly the same as that of the sixth embodiment, so the description thereof will be omitted.

As described above, in this embodiment, the CCD
It is possible to easily generate moire fringes by using 60 in the image receiving unit and setting the expansion magnification of the rough alignment system so that the grating spacing of the diffraction grating is about the same as the pixel.

(Embodiment 8) The eighth embodiment of the present invention will be described below.

The device configuration of this embodiment is the same as that of the seventh embodiment. That is, the alignment apparatus is mounted on the exposure apparatus as shown in FIG. 12, and the rough alignment system 12 can be configured by either of the apparatus configurations of the sixth and seventh embodiments, and the diffraction grating image can be obtained. It suffices that the coarse alignment system 12 be provided with a mechanism for generating moire fringes by using.

In the sixth and seventh embodiments, it has been described that the rotation angle between the reticle and the wafer can be known by observing the moire fringes.

However, useful information other than the rotation angle can be obtained from the moire fringes. From the above (Equation 6) and (Equation 7), after adjusting the rotation angle so that the angle ξ of the moire fringes becomes zero, the moire fringes are generated in the direction parallel to the lattice of the coarse alignment system. It turns out that

From the observation of the moire fringe spacing in that state,
The relationship between the two lattice intervals d1 and d2 is obtained.

Since the grating interval d1 of the rough alignment system is known, the grating interval d2 of the diffraction grating image at the recognition mark position of the diffraction grating on the wafer is finally obtained.

Therefore, when the expansion magnification of the optical system constituting the coarse alignment system is unchanged, the projection magnification of the projection lens can be constantly monitored from the ratio of the grating spacing d2 to the actual size of the diffraction grating spacing on the wafer. It will be possible.

Even if the distance between the reticle and the projection lens does not change, the projection magnification of the projection lens depends on the temperature, humidity, atmospheric pressure fluctuations, and changes over time of the lens itself.
It changes every moment.

Since different patterns are repeatedly exposed on the wafer itself until it becomes a semiconductor device, if the projection magnification of the projection lens varies with each exposure, the exposure patterns partially overlap each other. There is a problem that some parts do not exist.

This problem can be avoided by measuring the moire fringe spacing for each exposure and adjusting the magnification of the projection lens so that the moire fringe spacing by the diffraction grating of the wafer is always constant.

As described above, in this embodiment, by observing the interval of moire fringes, even during exposure,
Further, the projection magnification of the projection lens can be constantly monitored during the exposure, and the magnification of the projection lens can be immediately adjusted according to the change.

[0280]

As described above, according to the present invention, the two laser beams from the precision alignment system pass through the same path, and the optical axes of the precision alignment system and the coarse alignment system coincide with each other. As a result, the diffraction grating and the alignment mark can be arranged adjacent to each other, the alignment device is provided with means for selecting only the diffraction grating to be measured from the plurality of diffraction gratings, and the coarse alignment system and the precision alignment system are provided. A structure that provides a light-shielding mechanism to prevent the laser light from entering from the device, provides the coarse alignment system with means for generating moire fringes from the diffraction grating image, and allows the moire fringes to be observed by image recognition. By adopting, the overall alignment accuracy is high without being affected by changes in the optical path length due to environmental changes, and the rotation angle of the wafer and reticle is constantly maintained. You can monitor the projection magnification of the projection lens and, in which can realize an exposure apparatus having an excellent alignment device for measuring the positional deviation amount of the wafer relative to the reticle.

[Brief description of drawings]

FIG. 1 is an overall configuration diagram of an alignment device for an exposure apparatus according to a first embodiment of the present invention.

FIG. 2 is an explanatory diagram of an optical path of a laser beam of a precision alignment system according to the first embodiment.

FIG. 3 is a configuration diagram of a light source optical system in the first embodiment.

FIG. 4 is a configuration diagram of a precision alignment optical system in the first embodiment.

FIG. 5 is an overall configuration diagram of an alignment device for an exposure apparatus according to the second embodiment.

FIG. 6 is a configuration diagram of a beam diameter reduction optical system in the second embodiment.

FIG. 7 is an overall configuration diagram of another alignment device according to the second embodiment.

FIG. 8 is an explanatory diagram of speckle light in the second embodiment.

FIG. 9 is a configuration diagram of a beam diameter reduction optical system in the third embodiment.

FIG. 10 is an operation explanatory diagram of an aperture in the third embodiment.

FIG. 11 is an explanatory view of the irradiation state of the diffraction grating in the third embodiment.

FIG. 12 is an overall configuration diagram of an alignment apparatus for an exposure apparatus according to the fourth embodiment.

FIG. 13 is a configuration diagram of a coarse alignment optical system in the same Example 4.

FIG. 14 is a shape diagram of an alignment mark and a diffraction grating in the fourth embodiment.

FIG. 15 is a shape diagram of a recognition mark according to the fourth embodiment.

FIG. 16 is a diagram showing a combined image of the recognition mark, the alignment mark, and the diffraction grating in the fourth embodiment.

FIG. 17 is a composite image of a recognition mark, an alignment mark, a diffraction grating, and a laser beam spot in the fourth embodiment.

FIG. 18 is an overall configuration diagram of another alignment device according to the fourth embodiment.

FIG. 19 is a configuration diagram of a coarse alignment optical system in the fifth example.

FIG. 20 is a schematic diagram illustrating the necessity of a light-shielding mark in the fifth embodiment.

FIG. 21 is a shape diagram of a recognition mark according to the sixth embodiment.

FIG. 22 is a diagram of moire fringe generation according to the sixth embodiment.

FIG. 23 is a detailed diagram of moire fringes according to the sixth embodiment.

FIG. 24 is a surface view of a CCD according to the seventh embodiment.

FIG. 25 is an overall configuration diagram of a conventional alignment device.

[Explanation of symbols]

 1 reticle 2 first diffraction grating 3 reticle precision alignment system 4 light source 5 wafer precision alignment system 6 phase comparator 7 control system 8 stage 9 wafer 10 second diffraction grating 11 projection lens 12 coarse alignment system 13 first Alignment mark 14 Second alignment mark 15 Laser 17 Acousto-optic modulator 17 'Acousto-optic modulator 18 1 / 2λ plate 19 Polarization beam splitter 20 1 / 4λ plate 21 Epi-optical system 22 1 / 2λ plate 23 Polarization beam splitter 24 Polarization Plate 25 Total reflection mirror 26 Beat light detector 27 Light beam 28 Light beam 29 Beat light 30 Beam diameter reduction optical system 30 'Beam diameter reduction optical system 31 Optical system 1 32 Aperture 33 Optical system 2 34 Optical path selection control system 35 Half mirror 36 Alignment mark 37 Lens 38 Mira 40 lens 41 lens 42 recognition mark 43 lens 44 image receiving part 45 half mirror 46 half mirror 47 half mirror 48 half mirror 49 shading mark 50 moire fringes 51 coherent light 52 diffraction grating 53 diffraction grating 54 illumination area 55 diffracted light 56 speckle light 57 Reflected light 58 Imaging area 59 Spot image 60 Light ray after passing through aperture 61 Diffraction grating 62 Illumination area of light ray not passing through aperture 63 Imaging area 64 Spot image 65 Stray light 66 Recognition pattern 101 X-coarse alignment system 101'Y-coarse alignment system 102 X-precision alignment system 102 'Y-precision alignment system 200 Illumination optical system 201 Projection light source 202 Coarse alignment light source 203 Precision alignment light source 204 First diffraction grating 204'th First diffraction mark 205 First alignment mark 205 ′ First alignment mark 206 Reticle 207 Projection lens 208 Illumination light 209 Control system 210 Stage 211 Second diffraction grating 211 ′ Second diffraction grating 212 Diffraction grating 213 Second Alignment mark 213 'Second alignment mark 214 Wafer 215 Recognition mark 215' Recognition mark 300 Beam reduction optical system

Front page continued (72) Hiroyuki Takeuchi, 3-10-1, Higashisanda, Tama-ku, Kawasaki, Kanagawa Prefecture Matsushita Giken Co., Ltd. (72) Shinichiro Aoki 3--10-1, Higashisanda, Tama-ku, Kawasaki, Kanagawa Prefecture No. Matsushita Giken Co., Ltd. (72) Inventor Yoshiyuki Sugiyama 3-10-1 Higashisanda, Tama-ku, Kawasaki City, Kanagawa Matsushita Giken Co., Ltd.

Claims (21)

[Claims] 1. An exposure apparatus for projecting and exposing a pattern of a first object onto a second object, the exposure apparatus performing alignment for aligning the first object and / or the second object. An alignment device, wherein the alignment device comprises an optical signal obtained by illuminating at least one first mark portion of the first object and / or the second object with two coherent light beams having different frequencies, The first
The precision alignment system has a relatively high detection accuracy for measuring the relative position of the object and the second object, or the absolute position of the first object and / or the second object. An exposure apparatus in which two different coherent light beams of different wavelengths pass through substantially the same optical path. 2. In the precision alignment system, an optical path of one of two coherent light beams having different frequencies, which is reflected after entering the first mark portion of the first object and / or the second object. , The other coherent light is substantially the same as the optical path of the other light that is incident on the first mark portion of the first object and / or the second object. The exposure apparatus according to claim 1, wherein the exposure apparatus passes through an optical path. 3. An exposure apparatus for projecting and exposing a pattern of a first object onto a second object, the exposure apparatus performing alignment for aligning the first object and / or the second object. An alignment device, the alignment device comprising: a first optical signal obtained by illuminating at least one mark portion of the first object and / or a second object with two coherent light beams having different frequencies; A precision alignment system having relatively high detection accuracy for measuring the relative position of the object and the second object, or the relative position and absolute position of the first object and / or the second object, and the alignment device, The two coherent light beams having different frequencies illuminate a part of at least one mark portion of the first mark portion of the first object and / or the second object so as to illuminate a part thereof. An exposure apparatus having control means for controlling coherent light. 4. The control means limits the illumination of two coherent light beams having different frequencies to a specific mark portion of at least one first mark portion of the first object and / or the second object. The exposure apparatus according to claim 3, wherein the exposure apparatus is controlled by the method. 5. The exposure according to claim 3, wherein the control means controls so that two coherent light beams having different frequencies illuminate the inside of the first mark portion of the first object and / or the second object. apparatus. 6. The exposure apparatus according to claim 4, wherein the control means is an opening means for limiting the luminous flux of two coherent light beams having different frequencies. 7. The exposure apparatus according to claim 6, wherein the control means is an opening means that restricts two light fluxes of coherent light beams having different frequencies and is movable so as to intersect the optical axis of the light fluxes. 8. The exposure apparatus according to claim 1, wherein the first mark portions of the first object and the second object are diffraction gratings imprinted on the first object and the second object. 9. The alignment device further includes a coarse alignment system having relatively low position detection accuracy, wherein the coarse alignment system includes a second mark on the first object and / or the second object. Part and the third mark part of the coarse alignment system are optically combined, and the first object and / or the second object is absolutely aligned by image recognition of the combined image, or the first object and / or the second object. The exposure apparatus according to claim 1, wherein the relative alignment of the objects is performed. 10. An exposure apparatus for projecting and exposing a pattern of a first object onto a second object, comprising an alignment device for aligning the first object and / or the second object, The alignment device performs relative alignment of the first object and the second object or absolute alignment of the first object and / or the second object, and a relatively low detection accuracy of a coarse alignment system, A precision alignment system having relatively high detection accuracy for performing relative alignment of the first object and the second object, or absolute alignment of the first object and / or the second object,
The coarse alignment system and the precise alignment system,
An exposure apparatus equipped with an optical system having a common optical axis. 11. The coarse alignment system captures at least a part of the coarse alignment mark portion and the fine alignment mark portion of the first and / or second object on the same screen to perform image processing to obtain a first object. 11. The exposure apparatus according to claim 10, which aligns the second object and / or the second object. 12. The exposure apparatus according to claim 11, wherein the rough alignment mark portion and the fine alignment mark portion of the first and / or second object are integrally formed. 13. The precision alignment system has a light source for emitting coherent light, and the coarse alignment system further comprises:
The exposure apparatus according to claim 11 or 12, further comprising an excluding unit that excludes at least a region of the precision alignment mark portion of the first and / or second object on which the coherent light is incident from a screen used for image processing. 14. Excluding means for excluding at least a region of the precision alignment mark portion of the first and / or second object on which the coherent light is incident from the screen intersects the optical axis of the light emitted from the region. 14. The exposure apparatus according to claim 13, wherein the exposure apparatus is a movable shielding unit. 15. The exposure apparatus according to claim 14, wherein the shielding means is a shielding mark formed on the transparent substrate. 16. An exposure apparatus for projecting and exposing a pattern of a first object onto a second object, wherein the first object and the second object are relatively aligned using moire fringes, or the first object. An exposure apparatus having an alignment device for performing absolute alignment of an object and / or a second object. 17. The exposure apparatus further detects the projection magnification of the projection lens by detecting the shape of the moire fringes, and adjusts the projection magnification of the projection lens so that the projection magnification of the projection lens becomes substantially constant. 17. A means for controlling, comprising:
The exposure apparatus described. 18. The alignment device has a coarse alignment system with relatively low detection accuracy, and the moire fringes are the first one.
The exposure apparatus according to claim 16 or 17, which is formed by a first mark portion provided on the object and / or a second object and a second mark portion provided on the rough alignment system. 19. The coarse alignment device further includes an image pickup means for picking up light emitted from a first mark portion provided on the first object and / or the second object, and the second mark portion is the above-mentioned image pickup means. 19. The exposure apparatus according to claim 18, which is formed on the imaging means. 20. The exposure apparatus according to claim 19, wherein the image pickup means is a CCD in which pixels are two-dimensionally arranged. 21. The first mark portion provided on the first object and / or the second object and the second mark portion provided on the rough alignment system both have a linear portion at least in part. 21. The exposure apparatus according to claim 20, wherein the linear portions of the first mark portion and the second mark portion extend in substantially the same direction.